//  Script generated by Search Maker Pro:

sQ1=new Array();

sQ1[1]=new Array("../7337/0001.pdf","New Molecular Tools for Light and Electron Microscopy","","1 doi:10.1017/S143192761500080X Paper No. 0001 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 New Molecular Tools for Light and Electron Microscopy Erik A. Rodriguez1, John T. Ngo1, Sakina F. Palida2, Stephen R. Adams1, Mason R. Mackey3, Ranjan Ramachandra3, Mark H. Ellisman3, and Roger Y. Tsien1,4 2. Dept. Pharmacology, University of California San Diego, La Jolla CA, USA. Biomedical Sciences Graduate Program, University of California San Diego, La Jolla CA, USA. 3. Dept. Neurosciences and National Center for Microscopy and Imaging Research, University of California San Diego, La Jolla CA, USA. 4. Howard Hughes Medical Institute. 1. Fluorescent proteins (FPs) are invaluable tools for biology, enabling tracking of gene expression, cell fate, and genetically encoded fusion proteins for precise localization within a cell. Traditional FPs developed from jellyfish and coral are limited in wavelengths, consume O2, and produce a stoichiometric amount of H2O2 upon chromophore formation, thus requiring an aerobic environment tolerant of reactive oxygen species. Far-red/near-infrared FPs are desirable for imaging in living animals because less light is scattered or absorbed or reemitted by endogenous biomolecules. Previous near-infrared FPs were engineered from nonfluorescent phytochrome precursors and have had poor quantum yield (QY). We have developed a new class of FP by evolving an allophycocyanin -subunit from a cyanobacterium, Trichodesium erythraeum. Native allophycocyanin is a highly fluorescent hexamer composed of three + dimers and uses an auxiliary protein, known as a lyase, to incorporate phycocyanobilin (PCB). The new FP, named small Ultra-Red FP (smURFP), was engineered to bind biliverdin (BV), an endogenous heme metabolite ubiquitous to mammals, without an auxiliary lyase or autoxidation chemistry. It is a dimer of 15 kDa subunits or a tandem dimer of 32 kDa, and has excitation and emission maxima at 642 and 666 nm and the largest QY (0.18), BV incorporation rate, metabolic stability, and photostability of any BV binding FP so far. SmURFP is even more photostable than GFP or Cy5. Collaborations are currently underway to utilize smURFP for superresolution imaging. SmURFP expressed in HT1080 mouse xenografts show significant, visible fluorescence without exogenous BV, but provision of extra chromophore by various means increases the fluorescence yet further. Using smURFP and a phytochrome FP, a far-red/near-infrared fluorescent ubiquitination cell cycle indicator (FUCCI) was created, which should be suitable for monitoring cell cycle progression in intact mammals. The development of this new class of FP and far-red/near-infrared biosensors should dramatically increase our ability to image and monitor dynamics deep in tissues of living animals. Electron microscopy (EM) achieves the highest spatial resolution in protein localization and has long been the main technique to image cell structures with nanometer resolution. However, making specific molecules standout for EM is a challenge. Recently, powerful genetically-encoded tags have been introduced that allow specific proteins to be tracked by EM via genetic fusion, in a manner similar to how green fluorescent protein (GFP) is used to track proteins by light microscopy (LM). Tagged proteins are revealed by tag-mediated conversion of 3,3-diaminobenzidine (DAB) into a localized osmiophilic polymer that is readily distinguished under the electron microscope. Currently available EM tags precipitate DAB either enzymatically through peroxidase activity or via photo-generated singlet oxygen. While such tags are powerful tools for "painting" individual proteins, researchers lack analogous tools for marking biochemical processes, or non-proteinaceous molecular species for EM. To complement the existing EM tags, we describe "Click-EM," a new method for imaging nucleic acids, lipids, and glycans via bio-orthogonal ligation of photo-sensitizing dyes to functionalized metabolic analogs. These analogs Microsc. Microanal. 21 (Suppl 3), 2015 2 mimic the fates of their natural counterparts and can be used to track cellular metabolism. Analogs functionalized with azides and alkynes can be selectively ligated to chemical probes that do not react with endogenous (unlabeled) biomolecules. For detection, azide- and alkyne-functionalized analogs can be revealed by Cu(I)-catalyzed azide�alkyne cycloaddition (CuAAC), a reaction often referred to as "click chemistry," to appropriately functionalized dyes. These labeled structures, conjugated to a singlet oxygen-generating fluorescent dye (a "photosensitizer"), can be visualized first by fluorescence and subsequently by EM through photogeneration of singlet oxygen for DAB precipitation. Using these methods, we have imaged neuronal protein assemblies within the full context of cellular ultrastructure and have visualized DNA replication (see Fig. 1) and mRNA transcription at the nanometer scale. Distinguishing multiple biomolecules in EM is presently limited to attachment of different sizes of gold particles or quantum dots to specific antibodies, which poorly penetrate into the strongly fixed cells or tissue required for optimal cellular ultrastructure. Localized precipitation of DAB by antibody conjugates or genetically-encoded chimeras with photosensitizers and subsequent staining with osmium can overcome many of these constraints but is limited to a single protein or tracer "color". For "multi-color" EM, we have now synthesized "Ce-DAB" and "Pr-DAB", shorthand for conjugates of DAB with chelates of Ce or Pr. Ce-DAB is locally photooxidized with the first photosensitizer tracer, and then the polymer is quenched. Pr-DAB is either oxidized by a peroxidase-antibody conjugate to a second label, or photooxidized with a second targeted photosensitizer irradiated at much longer wavelengths than the first. The two deposited lanthanides can be detected by electron energy loss spectroscopy (EELS), selectively imaged with energy-filtered transmission EM, and overlaid as elemental maps on a conventional electron micrograph. Pancreatic cancer cells labeled with NBD-ceramide and an EpCAM antibody gave the expected Golgi and plasma membrane staining respectively. We detect sharing of a single synapse by two adjacent astrocytes in mouse brain slices and reveal the endosomal localization of cellpenetrating peptides in tissue culture cells, demonstrating high spatial resolution and selectivity. Supported by NIH grants GM86197, GM103412 and NS27177 and HHMI. Fig. 1. Click-EM imaging of newly replicated DNA in a dividing HeLa cell. Live cells were pulsed for 12 hr with 5ethynyl-2-deoxyuridine (EdU), a nucleoside analog readily incorporated into DNA during replication. After fixation with glutaraldehyde, alkyne-containing DNA was conjugated to azidodibromofluorescein (DBF) via CuAAC. Cells were incubated with DAB and illuminated for 5 min with blue light for singlet oxygen generation and photo-oxidation of DAB, which formed optically dense precipitates coincident with DBF fluorescence. Cells were stained with OsO4, embedded in resin, and thin sectioned for EM. Scale bar 2 m.");sQ1[2]=new Array("../7337/0003.pdf","Some Unexpected Difficulties in Microscope Operation in Microgravity","","3 doi:10.1017/S1431927615000811 Paper No. 0002 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Some Unexpected Difficulties in Microscope Operation in Microgravity Donald Pettit1 1. NASA, Johnson Space Center, Houston, TX The International Space Station (ISS) is a research laboratory in low earth orbit where the magnitude of gravitational forces are reduced by a factor of one million. Changing any other earthly parameter by this magnitude rapidly takes one into an experimental frontier, and this orbital environment is no exception. Many facilities typical of ground-based laboratories are onboard ISS: furnaces, centrifuges, freezers, incubators, plant growth and combustion chambers, with supplied resources of vacuum, inert gas, oxygen, liquid and forced air cooling, 28 and 120 volt DC power, and near-continuous real-time communication of data, voice and high definition video transmitted over distances approaching 50,000 miles (roundtrip to geostationary orbit). Microscope facilities include many state-of-the-art imaging techniques: transmission, reflection, brightfield, darkfield, epi illumination, phase contrast, differential interference contrast, fluorescent, confocal, polarization, and student educational instruments [1]. During the operation of these microscopes in low earth orbit, some unexpected difficulties unrelated to the undergoing research but directly resulting from their operation in microgravity can delay the progress of an experiment. Difficulties can stem from errant fluid behavior, residual gravity gradients, cosmic rays, and safety of flight. Subtle forces stemming from surface tension, liquid-solid contact angle, and static electric charge dominate fluid behavior in microgravity. These can conspire to give non-intuitive behaviors [2] resulting in possible operational delays or equipment maintenance. The precise placement of immersion oil on a slide using a pipette can be challenging (Fig. 1 left). Filling a sample chamber with bubble free liquid requires significant on-orbit practice. Flow induced charging of liquids, a small charge developed when a dielectric fluid (such as immersion oil) is forced through a small insulated capillary (such as a Teflon pipette), can result in subtle charge forces making the liquid misbehave [3]. These subtle forces under microgravity can interfere with the sample placement within the optical path (bubbles) or result in the fouling of optical surfaces (Fig. 1 center). The time necessary to learn the handling skills or to keep the instrument in operating order can cause delays in experimental progress. The magnitude of residual acceleration on ISS is near 1.2 E-6g, nominally referred to as microgravity, where g is the acceleration due to gravity on the surface of Earth. At this level of residual acceleration, sample motion is possible [4,5]. The direction of this residual acceleration in relation to the orientation of the experimental sample can cause unexpected fluid-particle-bubble motion within a sample chamber. Such motion might cause the intended subject to settle out of suspension or migrate outside the optical field of view over a period of a few hours. The flux of cosmic rays in low earth orbit causes camera CCD or CMOS detector arrays to degrade after periods of about one year. They produce images strewn with hot pixel "snow" that can compromise their scientific usefulness (Fig 1. right). By design, some instrument cameras were never meant to be replaced, and after years in orbit, can suffer significant image degradation. A maintenance plan including periodic camera replacement should be considered. When subjected to microgravity while living in a sealed thin shell surrounded by infinite vacuum, flight Microsc. Microanal. 21 (Suppl 3), 2015 4 crew safety becomes paramount. Standard means of conducting research on ISS may require seemingly prohibitive constraints when compared to similar ground-based research. Other than water (recovered from urine,) there are no cleaning solvents available on ISS due to their detrimental effects on the regenerative life support systems. The lack of solvents complicates cleaning of optics and other delicate surfaces especially from unintended fluid migration (Fig. 1 center). Microgravity promotes the possibility of inhalation or eye damage from small free floating objects with possible serious consequences. On ISS, the handling of small parts (screws and nuts) or shards created from accidental breakage of fragile glass components (cover slips and slides) presents a significant crew hazard. Laborious, time-consuming practices are often required for what would normally be a trivial operation on Earth. Sometimes a compromise must be made between the safety requirements needed for the best scientific practices versus non-optimum materials, substituted to ease the handling requirements, such as plastic cover slips and slides. [1] https://iss-science.jsc.nasa.gov/investigation_detail.cfm?investigationsid=541 [2] Pettit, D, "Exploring the Frontier: Science of Opportunity on the International Space Station", Proc. Am. Philo. Soc. vol. (153), No. 4, Dec. 2009, pp. 381-402. [3] Pettit, D, "Flow Induced Charging of Liquids in Reduced Gravity", Eng. Const. & Op. in Space, Space 96, S Johnson, Ed., ASCE Pub. Vol. I (1996), pp.545-551. [4] Alexander, I, and Lundquist, C, "Residual Motions Caused by Micro-Gravitational Accelerations," Jour. Astro. Sci., Vol (35), no. 2, 1987, pp 193-211. [5] Delombard, R, Kelly, E, Hrovat, K, Nelson, E, Pettit, D, "Motion of Air Bubbles in Water Subjected to Microgravity Accelerations", 43rd AIAA, Reno, Jan. (2005), AIAA-2005-0722. Figure 1. Examples of unexpected experimental difficulties on ISS: pipette tip (1.5mm diameter) showing migration of a water drop from the tip to the side during fluid operations on Exp. 6 (left) complicating its precise placement; front surfaced gold mirror (100mm wide) inadvertently contaminated with silicone oil that migrated from the experimental stage during Exp. 30 requiring about two hours of crew time with a three week schedule delay to clean (center); and a 450 by 450 pixel enlargement from a camera C-MOS detector dark frame image showing red-green-blue-white hot pixels from one year of cosmic ray damage during Exp. 42 (right).");sQ1[3]=new Array("../7337/0005.pdf","Microstructural Developments Leading to New Advanced High Strength Sheet","","5 doi:10.1017/S1431927615000823 Paper No. 0003 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microstructural Developments Leading to New Advanced High Strength Sheet Steels: A Historical Assessment of Critical Metallographic Observations David K. Matlock1, Larrin S. Thomas1, Mark D. Taylor1, Emmanuel De Moor1, and John G. Speer1 1. Advanced Steel Processing and Products Research Center, The George S. Ansell Department of Metallurgical and Materials Engineering, Colorado School of Mines, Golden, Colorado 80401 Critical needs for new high strength steels to enable design of lighter weight systems which incorporate components that operate at higher stress levels than previously specified have led to the development of multiple new steel grades in all product forms. Of particular interest here are the extensive recent developments that have been employed to produce and implement new advanced high strength sheet steel (AHSS) grades in affordable lighter-weight vehicles with significantly enhanced fuel efficiency and safety [1]. Multiple different alloying and processing strategies have been considered [1], and designations for different AHSS grades, characterized by specific tensile strength-ductility combinations, have evolved. "First generation" AHSS grades include DP (dual-phase), CP (complex phase) and TRIP (transformation induced plasticity) steels. Current emphasis is on enhancing properties of first generation AHSS [2] while developing new "third generation" AHSS grades including Q&P steels (quenched and partitioned), TBF (TRIP aided bainitic ferrite) steels, and others with the higher strength-ductility combinations required by vehicle designers [1]. Initially DP steels were envisioned to contain a two-phase microstructure of ferrite and martensite produced by quenching a low carbon steel from an intercritical annealing temperature. However, early research in the 1970's showed that DP steels contained complex combinations of other constituents including retained austenite and bainite. Many of the new AHSS grades have evolved from similar underlying principles of microstructural evolution in DP steels, and developments have relied on systematic microstructural analyses that have incorporated conventional light and electron microscopic techniques as well as many of the advanced techniques available today. In this paper microstructural observations from early research on DP steels are contrasted with current results on new AHSS grades to illustrate the importance that microstructural analyses have had to advance modern steels. Early DP steel research showed that the resulting microstructures were complex, sensitive to both alloy content and time-temperature processing history, and contained other constituents, including retained austenite, that contributed to the observed unique mechanical properties. Specialized etching techniques based on multi-step and color etching were employed to highlight specific microstructural constituents as illustrated in Figure 1 [3]. Figure 1 reveals that the final ferrite present at room temperature is comprised of intercritical ferrite (i.e. present at the annealing temperature) and epitaxial ferrite formed on cooling as carbon redistributed to martensite and other high-carbon transformation products. Modifications to classic DP alloys and thermal process histories now incorporate alternate cooling cycles leading to the development of TRIP, TBF, and Q&P steels, with microstructures designed to contain increased amounts of retained austenite, a critical constituent in newer AHSS grades [1]. Multiple conventional and advanced metallographic techniques have been employed to evaluate the refined microstructures, often containing sub-micron sized constituents. Determination of the amount, morphology, and distribution of retained austenite is one important outcome [4]. For a Q&P steel, Figures 2a and 2b compare secondary electron and EBSD images and show the distribution of retained austenite, information essential to assessing, and potentially modifying, imposed thermal cycles [5]. Microsc. Microanal. 21 (Suppl 3), 2015 6 Figures 2c and 2d illustrate, in a study on hydrogen embrittlement in TRIP steels, how application of a modified hydrogen microprint technique shows clear evidence of hydrogen in ferrite but essentially absent in the austenite containing constituent. This latter point is important for AHSS grades that may be exposed to hydrogen. AHSS advances continue, in part as a direct result of the application of modern analytical tools that are becoming routinely available and used to comprehensively assess the ultra-fine non-equilibrium microstructures currently being produced, often with important composition gradients within individual constituents. Recently applications involving atom probe tomography, nano secondary ion mass spectroscopy, and M�ssbauer spectroscopy [6] have provided valuable information to interpret the fine features associated with new AHSS products, and research combining all technologies leading to new steel products will continue [7]. [1] E. De Moor et al., AIST Trans., Iron and Steel Technology, 7, no. 11 (2010), p. 133. [2] M. D. Taylor et al., J. of Mater. Engr. and Performance, 23 (2014), p. 3685. [3] R.D. Lawson et al., Metallography, 13 (1980), p. 71. [4] G.A. Thomas et al., Microscopy and Microanalysis, 17 (2011), p. 368. [5] J.A. Ronevich et al., Metallography, Microstructure, and Analysis, 1 (2012), p. 79. [6] D.T. Pierce et al., Acta Materialia, in press, (2015) DOI: 10.1016/j.actamat.2015.01.024. [7] The authors acknowledge the support of the corporate sponsors of the Advanced Steel Processing and Products Research Center at the Colorado School of Mines. ICF (a) 10 �m ICF (c) (b) Figure 1. Intercritical ferrite (ICF) (grey) and epitaxial ferrite (white) revealed after cooling a 0.08C1.47Mn-0.34Si-0.053Nb from a 4 min anneal at 810 oC, 12 oC/s (a), 135 oC/s (b), and 1000 oC/s (c); (etched with picral followed by alkaline chromate staining) [3]. 1 �m (a) 1 �m (b) 1 �m (c) 1 �m (d) Figure 2. Example images of retained austenite in 0.19C-1.59Mn-1.63Si steel with different AHSS processing histories: Q&P steel where (a) and (b) are secondary electron and EBSD images (different areas of same sample), respectively. In (b) austenite = green, ferrite and martensite = red, carbide structures and martensite with reduced image quality appear dark [4]. ((a) and (c) 2 pct nital etch); TRIP steel where (c) is SEM image of as processed steel and (d) after hydrogen charging and preparation with the modified hydrogen microprint technique showing hydrogen concentrated in the ferrite constituent [5].");sQ1[4]=new Array("../7337/0007.pdf","The Biodynamic Microscope: Doppler Imaging inside Living 3D Biological Tissues.","","7 doi:10.1017/S1431927615000835 Paper No. 0004 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Biodynamic Microscope: Doppler Imaging inside Living 3D Biological Tissues. David Nolte1,3, John Turek2,3 and Ran An3. 1. 2. Dept. of Physics, Purdue University, West Lafayette, IN USA 47907. Dept. of Basic Medical Sciences, Purdue University, West Lafayette, IN USA 47907. 3. Animated Dynamics Inc. West Lafayette, IN USA 47906. Biodynamic imaging is a new form of microscopy that combines coherent laser imaging with digital holography to extract Doppler images up to 1 mm deep inside living tissue [1]. Intracellular biophysical processes that are captured by biodynamic imaging include exo- and endocytosis, mitochondrial activity, nuclear morphology, cytoskeletal remodeling, cellular integrity, cell proliferation and mitosis, as well as necrosis and induced apoptosis, among others. Biodynamic microscopy is label-free, relying on Doppler scattering inside living tissue to provide image contrast. Biodynamic microscopy is fully 3D, relying on low-coherence laser ranging techniques to extract three dimensional images. Biodynamic microscopy is non-invasive, using low intensity LED illumination. Because motion is ubiquitous in all living samples, biodynamic imaging is a general new form of microscopy that can be applied in any situation involving living 3D tissue or samples. The number of applications of biodynamic microscopy is growing rapidly. This presentation provides an overview of a wide range of recent studies using biodynamic imaging. These are: 1) 3D tissue growth monitoring; 2) oocyte and embryo assessment for IVF applications; 3) chemotherapy sensitivity assays for personalized cancer care; and 4) phenotypic profiling for pharmaceutical drug discovery and development. Three-dimensional Tissue Growth Monitoring: Three-dimensional tissue growth is a new market that is growing rapidly with the realization that conventional 2D cell culture fails to capture important threedimensional microenvironmental effects. Biodynamic imaging assesses differences in different approaches to 3D cell culture growth, such as bioreactors, hanging drops and nonadhesive plates. The use of biodynamic imaging to measure the response to applied pharmaceuticals of tissues, grown with these different methods, uncovered important differences that can relate to the downstream success rate for drug development. IVF Applications: The success of assisted reproductive technologies relies on accurate assessment of reproductive viability at successive stages of development for oocytes and embryos. Biodynamic imaging measures physiologically-relevant dynamics and may provide an objective approach to oocyte and embryo assessment. The changes in intracellular activity during cumulus-oocyte complex maturation, before and after in vitro fertilization, and the subsequent development of the zygote and blastocyst provide a new approach to the assessment of pre-implant candidates. Personalized Cancer Care: Intracellular motions encompass a diverse range of intracellular functions that provide fundamentally new biomarkers of the phenotypic response of living tissue to therapy. These dynamic biomarkers can be used to predict therapeutic efficacy for the treatment of disease in preclinical and clinical trials. A preclinical trial has been completed [2] for canine non-Hodgkins lymphoma using biodynamic imaging to predict patient outcome under doxorubicin therapy. The biodynamic assay took only 24 hours to complete, while the clinical outcome was often unknown for Microsc. Microanal. 21 (Suppl 3), 2015 8 one to several months. The therapeutic efficacy of doxorubicin was predicted by BDI with 90% accuracy. Based on this success, biodynamic imaging is entering human pilot trials in ovarian, pancreatic, and esophageal cancer. Biodynamic imaging is a radically different approach to the prediction of therapeutic efficacy for personalized medicine with higher accuracy and more biological relevance than standard cell-based assays. Drug Discovery: The existence of phenotypic differences in the drug responses of three-dimensional tissues relative to two-dimensional cell culture is a concern in high-content drug screening. The information content of biodynamic imaging for drug discovery applications is displayed through tissue dynamics spectroscopy (TDS), which captures and displays the changes in the Doppler signatures from intracellular constituents in response to applied compounds. Here we present the comparison of TDS against morphological image analysis of two-dimensional cell culture. There are significant 2D versus 3D phenotypic differences exhibited by 25% of the drugs/cell-lines which could relate to therapeutic efficacy and toxicity in 3D that are missed by 2D screens [3]. References: [1] D. D. Nolte, R. An, J. Turek, and K. Jeong, "Holographic tissue dynamics spectroscopy," Journal of Biomedical Optics, vol. 16, pp. 087004-13, Aug 2011. [2] M. R. Custead, J. J. Turek, R. An, D. D. Nolte, and M. O. Childress, "Use of biodynamic imaging to predict treatment outcome in a spontaneous canine model of non-Hodgkin's lymphoma," in preparation for submission to Cancer Research, 2015. [3] R. An, D. Merrill, L. Avramova, J. Sturgis, M. Tsiper, J. P. Robinson, J. Turek, and D. D. Nolte, "Phenotypic Profiling of Raf Inhibitors and Mitochondrial Toxicity in 3D Tissue Using Biodynamic Imaging," Journal Of Biomolecular Screening, vol. 19, pp. 526-537, Apr 2014. [4] The authors acknowledge funding from NSF1263753-CBET and NIH NIBIB 1R01EB016582-01. Motility Contrast Imaging (MCI) Cell Line Motility Images 10 M Taxol Drug Response Bioreactor Tumor Spheroid Tumor-Normal Margin Normal Tissue Tumor Motility = 0.8 Motility = 0.5 Active Porcine Oocyte Hanging Drop A2780 Mouse Tissue 200 microns MotilityContrastImaging.ai A2780 (Human Colon Cancer) Tumor Spheroid TDSIHD.ai hdrg.m Figure 1. (Left) Examples of biodynamic images of 3D tumor spheroids, with a three dimensional reconstruction, a tumor margin and an oocyte. (Right) Tissue response to Taxol depends on growth technique (bioreactor vs. hanging drop). Tissue dynamics spectroscopy (TDS) is used for hyperspectral imaging.");sQ1[5]=new Array("../7337/0009.pdf","Field-Portable Nano-Imaging: A New Tool for On-Demand Microscopy","","9 doi:10.1017/S1431927615000847 Paper No. 0005 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Field-Portable Nano-Imaging: A New Tool for On-Demand Microscopy Christopher S. Own, Matthew F. Murfitt, Lawrence S. Own Voxa, Seattle, WA USA Electron microscopy is widely regarded as a high-end laboratory science tool, where substantial resources are pooled to collect image data of exquisite quality. Electron microscopes (EM's) are uniquely able to produce detailed structural images that support discoveries from basic science to monitoring of industrial process. The strong scattering, large depth of focus, and unique blend of signals including elemental analysis are attractive in many applications. Being difficult to operate relative to many other common laboratory tools, EM's are traditionally housed in centers at universities and large research institutions where ample laboratory space, support staff, supplies, and skilled operators come together, or at industrial sites for organizations with research or quality control needs that justify the substantial cost. For those who do not have access to an on-site EM, many larger institutions and service centers accept samples sent in to be imaged, at great expense and often delay of weeks to months for complex analyses. The complexity, high cost, and significant maintenance associated with collecting EM image data has until now severely limited the fields in which EM can be realistically used [1]. This is exemplified in the number of EM instruments deployed (in the tens of thousands) to the number of deployed light microscopes (in the hundreds of millions) [2]. A new, smaller, more reliable and user-friendly personal EM -- the Mochii scanning electron microscope -has been developed by us at Voxa in Seattle, WA (Fig. 1). This low voltage microscope has features that bring accessible and on-demand EM imaging into fields and laboratories where EM was previously hindered by form factor, complexity, and cost. Among these features are small size and light weight (0.25m tall, light enough to carry in a suitcase); user-friendly native wireless tablet interface; multi- and distance-user capabilities (connection to unlimited client nodes), exceedingly low power consumption (by virtue of lowpower magnetic-electrostatic optics), and an integrated metal evaporator for easy sample preparation (only one pump-down cycle to image). We expect the cost to own and operate a Mochii microscope to be a fraction of the cost of typical EM's with similar imaging performance due to its low power consumption, simple design, and commoditized user-replaceable consumables. TM TM The improved tablet interface and reduced cost compared to existing benchtop systems significantly lowers the barrier of entry to EM imaging in fields where money, space, and/or operational expertise were limited. Learners and scientists in schools and community science centers can begin using EM imaging to explore scientific phenomena at below light-diffraction-limit resolutions for the first time (Fig. 2). We also expect that smaller research labs and new labs or startup companies can easily begin accessing image data without the enormous investment that would be required for a typical EM. Novel miniaturization features also open up access to EM imaging in scientific fields where samples would normally have to be stored and brought back from field work. With its exceedingly low power consumption, the Mochii can be operated in the field without AC power or a generator. Field forensic scientists can examine bullet casings, burn fragments, and other evidence on-site; shipboard oceanographers can examine flora in water samples while at sea (Fig. 3); and soil scientists can examine samples without packaging and shipping the samples (Fig. 4), thereby avoiding loss of critical structural data due to morphology changes during transit. Further, the Mochii microscope's remote viewing and operation capabilities allow field scientists to collaborate instantly with colleagues around the world, as remote viewing systems in radiology have permitted radiologists to do for over two decades [3]. TM TM Microsc. Microanal. 21 (Suppl 3), 2015 10 At the meeting, we will present images collected using Mochii under a variety of conditions, including images collected in the field, images collected by novice users, and images collected by remote operators. The additional use cases afforded by Mochii `s unique feature set represent a significant paradigm shift in microscopy instrumentation, taking the microscope out of the lab and into new environments, and opening high resolution high contrast imaging of to a wider audience and therefore broadening both scientific and citizen access to scientific phenomena at the microscale [4]. TM TM References: [1] Stahlberg H & Walz T, ACS Chemical Biology 3 (2008), p. 268-281. [2] Grand View Research, Microscopes Market Analysis (2014). [3] Thrall J, Radiology 243 (2007), p. 613-617. [4] This work was supported by Voxa. Figure 1. microscope MochiiTM field-portable personal electron Figure 2. Androlaelaps schaeferi, a tiny mite species that populates Madagascar hissing cockroaches, a common household pet. Figure 3. Holocene-era radiolaria collected from Sicily. Figure 4. Cross section of root of vetiver grass, a grass with high holding power that, when planted, can reduce soil erosion in loose soils.");sQ1[6]=new Array("../7337/0011.pdf","Automated, Programmable Processing of Specimens and Grids with the mPrepTM ASP-1000","","11 doi:10.1017/S1431927615000859 Paper No. 0006 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Automated, Programmable Processing of Specimens and Grids with the mPrepTM ASP-1000 Strader TE and Goodman SL Microscopy Innovations LLC, 213 Air Park Rd, Suite 101, Marshfield, WI, 54449, USA Electron microscopy (EM) specimen preparation and grid processing are tedious, time consuming and prone to error. For most life science applications and some materials applications, preparation for Transmission or Scanning EM requires each specimen to be sequentially processed with multiple fluids such as fixes, rinses, solvents, resins, stains, and labels. Similarly, TEM grids are fluid processed to stain sections, to deposit macromolecules and nanoparticles, for negative staining, and for immuno-labeling. In a typical life science TEM tissue preparation process there are 20-25 fluid exchanges for each specimen, and numerous grid fluid exchanges. Thus with several specimens and grids it is not uncommon for there to be well over 100 exchanges that can easily occupy most of a day. Yet, in most labs, these fluids are delivered manually due to reasons such as: 1. The number of samples does not warrant using automated sample immersion batch-type instruments, 2. Different specimens in the same study must be prepared with different reagents or protocols, and this can not be done using specimen immersion batch process instruments, 3. Reagents used are expensive and immersion batch processing requires several milliliters when only microliters are required for manual handling (such as for immuno-labeling tissues en bloc or grids), 4. The lab does not have an automated tissue processer, an automated grid stainer, and/or an automated immuno-labeling grid processor. This presentation introduces the mPrep ASP-1000 (micro-Preparation Automatic Specimen Processor). This entirely new class of automated instrument uses mPrep/sTM capsules for specimens and mPrep/gTM capsules for grids (Fig 1). As with all use of mPrep capsules, manual handling is nearly eliminated. Specimens are only directly touched when they are placed into mPrep/s capsules (and oriented when desired), while grids are only touched when inserted into mPrep/g capsules and removed for imaging. Since all fluid processing and storage occurs in a labeled mPrep capsule, sample identity is assured. The mPrep ASP-1000 incorporates intuitive programing to specify "Reagent, Time, Agitations, Repeats" for any desired number of steps (72 steps are available in the configuration shown). Capsules are simply mounted on the ASP pipette head (Fig 1) and reagents are placed in microtiter plates on the ASP tray. The selected program then sequentially delivers reagents by moving mPrep capsules to different microtiter well locations. Set-up and clean up are fast, and reagent consumption can be as low as 10 l per specimen step (typically 100 � 150 l), and 35 l per mPrep/g step for 2 grids. Each capsule may receive the same or different reagents as other capsules, thereby enabling simultaneous processing of titrations and controls as needed for immuno-labeling and other complex protocols [1]. The ASP-1000 is the first automated instrument that processes both specimens and TEM grids, and can perform essentially any protocol with minimal reagent consumption. Figure 2 shows high quality preservation and staining of porcine heart and skin prepared with Karnovsky and OsO4 fixation, acetone dehydration, epoxy embedding, and uranyl acetate and lead citrate grid staining. Laboratory efficiency is increased by enabling other tasks to be performed during computer-controlled, reproducible specimen and grid preparation. Process monitoring is intuitive with direct observation, while the ASP-1000 Microsc. Microanal. 21 (Suppl 3), 2015 12 software also enables sophisticated remote monitoring including notifications of procedure completion and other events via web browser, email or text message. Since any fluid can be delivered to specimens or grids within mPrep capsules, with any agitation or timing, only imagination limits the processing applications of the mPrep ASP-1000. Fig 1: a) The mPrep ASP-1000 processes specimens or grids mounted on the computer controlled pipette "head" (circled) that moves them between reagents in microtiter plates on the processor tray. b) Processor pipette head with 8 specimens entrapped in 8 labeled mPrep/s capsules (arrow points to one specimen). c) Head with 8 mPrep/g capsules containing 16 grids (arrow points to one grid). Fig 2: Porcine heart (a) and skin (b) prepared using mPrep/s tissue and mPrep/g grid processing. Reference [1] M McClain, High Throughput Multi Parameter TEM Chemical Processing Protocol Development with the mPrep-s Capsule System. Microsc. Microanal. 20 (Suppl 3) (2014) 1288-9.");sQ1[7]=new Array("../7337/0013.pdf","Correlative Microscopy based on Secondary Ion Mass Spectrometry for High-Resolution High-Sensitivity Nano-Analytics","","13 doi:10.1017/S1431927615000860 Paper No. 0007 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Correlative Microscopy based on Secondary Ion Mass Spectrometry for HighResolution High-Sensitivity Nano-Analytics T. Wirtz, D. Dowsett, S. Eswara Moorthy Advanced Instrumentation for Ion Nano-Analytics (AINA), Luxembourg Institute of Science and Technology, 41 rue du Brill, L-4422 Belvaux, Luxembourg Nano-analytical techniques and instruments providing both excellent spatial resolution and highsensitivity chemical information are of extreme importance in materials science and life sciences for investigations at the nanoscale. Due to the ever increasing complexity of devices and the continuously shrinking geometries in materials research, characterization tools and techniques are facing new challenges and need to anticipate future trends. Electron Microscopy, Helium Ion Microscopy and Scanning Probe Microscopy are commonly used for high-resolution imaging. However, these techniques all have the same important drawback: they provide no or only very limited chemical information. In electron microscopy, chemical information can be obtained by using techniques like EELS or EDS, but the sensitivity is limited. Moreover, these techniques do not permit to distinguish between isotopes, which is a major handicap today due to the increasing use of isotopic labeling, and have limitations in the low mass range. By contrast, Secondary Ion Mass Spectrometry (SIMS) is an extremely powerful technique for analyzing surfaces owing in particular to its excellent sensitivity, high dynamic range, very high mass resolution and ability to differentiate between isotopes. In order to get chemical information with a highest sensitivity and highest lateral resolution, we have investigated the feasibility of combining SIMS with Transmission Electron Microscopy, Scanning Probe Microscopy and Helium Ion Microscopy and developed three prototype instruments corresponding to these three combinations: TEM � SIMS : FEI Tecnai F20 equipped with a Ga+ FIB column and dedicated SIMS extraction optics, mass spectrometer and detectors (figure 1) HIM � SIMS : Zeiss ORION Helium Ion Microscope with dedicated SIMS extraction optics, mass spectrometer and detectors [1,2] SPM � SIMS : Cameca NanoSIMS 50 with integrated AFM/SPM [3-5] In order to reach good detection limits when probing very small voxels in imaging applications, the ionization probability of the sputtered atoms and molecules needs to be maximized. When using primary ion species such as Ga (as is the case on the integrated TEM � SIMS instrument) or noble gases (HIM � SIMS instrument), the intrinsic yields are low compared to the ones found in conventional SIMS. However, the yields may be drastically increased by using reactive gas flooding during analysis, namely O2 flooding for positive secondary ions and Cs flooding for negative secondary ions [1,2]. Our results show that both negative and positive ion yields obtained with Ga + bombardment are increased by up to 4 orders of magnitude when using such reactive gas flooding. This optimization of secondary ion yields leads to detection limits varying from 10-3 to 10-6 for a lateral resolution between 10 nm and 100 nm (figure 2). Microsc. Microanal. 21 (Suppl 3), 2015 14 Different possibilities arise from this in-situ instrument combination. A first approach consists of imaging the sample by SIMS to localize hot spots (e.g. a high concentration of a given element or a given isotope of interest), and then zooming onto this hot spot by TEM/HIM/SPM to identify the feature corresponding to this hot spot. A second approach consists of starting by TEM/HIM/SPM to identify zones or features of interest, and then to determine the chemical composition (or the isotopic ratios) of these features or zones by SIMS (see example in figure 3 for SPM - SIMS). It is important to note that ex-situ multi-technique combinations do not allow the same performances as such an approach is hampered by several limitations, including precise re-localization of analyzed zones after transferring the sample between the standalone instruments and artifacts due to surface oxidation and surface reorganization during sample transfer between the instruments. The results are very encouraging and the prospects of performing SIMS in combination with TEM, HIM and SPM are very interesting. The combination of high-resolution microscopy and highsensitivity chemical mapping on a single instrument represents a new level of correlative microscopy. References [1] T. Wirtz et al., Appl. Phys. Lett. 101 (2012) 041601 [2] L. Pillatsch et al., Appl. Surf. Sci. 282 (2013) 908 [3] T. Wirtz et al., Surf Interface Anal. 45 (1) (2013) 513-516 [4] T. Wirtz et al., Rev. Sci. Instrum. 83 (2012) 063702 [5] C. L. Nguyen et al., Appl. Surf. Sci. 265 (2013) 489-494 100 10-1 Detection limit without flooding Detection limit with flooding EDX FIB Detection limit 10-2 10-3 10-4 10-5 10-6 10-7 Ga+ beam Si- signal UY = 6x10-5 without flooding UY = 1.6x10-1 with flooding 1 10 100 SIMS 10-8 Lateral resolution (nm) Figure 2: Detection limit using a Ga+ FIB with and without Cs0 flooding vs. minimum feature size: example for the detection of Si-. Figure 1: Prototype of a combined TEM-SIMS instrument: modified Tecnai F20 equipped with a Ga+ gun and dedicated SIMS column Figure 3: Combined SIMS-SPM 3D reconstruction of a nickel-based superalloy: Al distribution (left), Cr distribution (right)");sQ1[8]=new Array("../7337/0015.pdf","Mastering the Multi-scale Challenge: A Modern Correlation Environment","","15 doi:10.1017/S1431927615000872 Paper No. 0008 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Mastering the Multi-scale Challenge: A Modern Correlation Environment Arno P. Merkle1, Lorenz Lechner1, Andy Steinbach1, Jeff Gelb1, Fabian Perez-Willard2, Michael W. Phaneuf3, David Unrau3 1 2 Carl Zeiss X-ray Microscopy, Inc. 4385 Hopyard Road, Pleasanton, CA, USA Carl Zeiss Microscopy, GmbH, Oberkochen, Germany 3 Fibics Incorporated, Ottawa, Ontario, Canada In most fields of study, it is imperative to understand the behavior of a system across several length scales in three dimensions in order to properly address the structural parameters that govern its performance. In order to characterize a system, be it the brain, a structural metal alloy, or a porous rock reservoir, multiple microscopic methods have evolved to specialize in capturing a relatively well defined window of scales, modalities or dimensions of information. Examples of this include medical-CT, confocal light microscopy, X-ray microscopy (XRM), FIB-SEM tomography, serial block face SEM, TEM tomography, atom probe tomography, and more. As these techniques have progressed individually, a clear challenge that has emerged has been how to intelligently and efficiently navigate to and acquire 3D volumes of interest (from centimeter to nanometer), and, subsequently, to fuse multi-scale and multimodality datasets in such a way that leaves the microscopist in control, as recently published in the context of a corrosion study [1]. Here we present the emergence of a major development enabling efficient correlation across modalities, length scales, and dimensions. ZEISS Atlas 5, released in 2015, is a modern, advanced SEM and FIB-SEM acquisition system combined with a modern correlative microscopy workflow environment which acts as the common hub and interface between experiments performed across multiple platforms (SEM, LM, FIB-SEM, XRM, etc.). Atlas 5 automates several advanced SEM and FIB-SEM acquisition tasks, but also provides a visualization environment to co-locate, calibrate and register multiple datasets from multiple instruments in one place. Going further, this environment extends beyond the conventional 2D correlation approach, by incorporating 3D datasets such as those obtained by XRM or FIB-SEM tomography. We review workflows that are enabled in the context of three examples, from Materials Research, Life Science and Geoscience. In Materials Science, XRM tomograms collected on an Al 7075 aluminum alloy (Figure 1) are used to locate inclusions and pores within the interior of the microstructure, which are then selected and examined at higher resolution by targeted FIB-SEM serial sectioning at defined regions of interest [4]. In Life Sciences, XRM presents a unique opportunity to bridge the length scales between light and electron microscopy (Figure 2), easing the `needle in a haystack' navigation problem for locating the same region of interest using multiple microscopes [5]. In Geosciences, we demonstrate the great 2D and 3D multi-scale challenge, and how Atlas 5 has been successfully used to enable the efficient study of such porous media systems. Microsc. Microanal. 21 (Suppl 3), 2015 16 References: [1] Burnett T., et. al. Correlative Tomography, Scientific Reports, 2014 [2] A. P. Merkle and J. Gelb, Ascent of 3D X-ray Microscopy in the Laboratory, Microscopy Today, 21 (2013), p. 10 [3] E Maire and P Withers, Quantitative X-ray tomography, International Materials Review, 59 (2014), p. 1 [4] A. P. Merkle et al., Automated Correlative Tomography Using XRM and FIB-SEM to Span Length Scales and Modalities in 3D Materials, Microscopy and Analysis, 28 (2014), p. S10-S13 [5] E. Bushong et al., X-Ray Microscopy as an Approach to Increasing Accuracy and Efficiency of Serial Block-Face Imaging for Correlated Light and Electron Microscopy of Biological Specimens, Microscopy and Microanalysis (2014). Figure 1: A 3D automated correlative workflow demonstrated on an Aluminum 7075 alloy to access information about inclusions, voids, precipitates and the Al matrix grain structure. In collaboration with N. Chawla and S. Singh of Arizona State University. Figure 2: XRM dataset of stained (for EM) mammalian brain tissue, used to navigate to specific subsurface volumes of interest quickly, thereby multiplying the efficiency of 3D EM techniques. In collaboration with NCMIR at the University of California San Diego.");sQ1[9]=new Array("../7337/0017.pdf","RISE Microscopy: Correlative Raman and SEM Imaging","","17 doi:10.1017/S1431927615000884 Paper No. 0009 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 RISE Microscopy: Correlative Raman and SEM Imaging Ute Schmidt1, Olaf Hollricher1, Wei Liu2, Edward L Principe3 1. 2. WITec GmbH, Ulm, Germany WITec Instruments, Knoxville, TN, USA 3. Tescan USA, Warrendale, PA, USA RISE Microscopy is a novel correlative microscopy technique which combines confocal Raman Imaging and Scanning Electron (RISE) Microscopy within one integrated microscope system (Fig. 1a). This unique combination provides advantages for the microscope user with regard to comprehensive sample characterization: electron microscopy is an excellent technique for visualizing the sample surface structures in the nanometer range; confocal Raman imaging is an established spectroscopic imaging method used for the detection of the chemical and molecular components of a sample with diffraction limited resolution. In contrast to existing combinations, where single Raman spectra are typically collected from few micrometer size areas, the RISE combination allows for the first time diffraction limited confocal Raman imaging on the same sample position as the SEM image was taken. It can also generate 3D-images and depth profiles to visualize the distribution of the molecular compounds within a sample volume. Both analytical methods are fully integrated into the RISE Microscope. Between the different measurements a precise translation stage automatically transfers the sample inside the microscope's vacuum chamber and re-positions it. The integrated RISE software carries out the required parameter adjustments and instrument alignments. The acquired results can then be correlated and the Raman and SEM images overlaid. The instrument as well as various examples for using this new possibility for correlative confocal Raman imaging with SEM will be presented. A first example of correlative Raman and SEM imaging is presented in Figure 1b-d showing the distribution of TiO2 nanoparticles. Fig. 1b shows the SEM image of TiO2 nanoparticles, revealing the presence of two different sizes of the particles. From the same sample area an array of 150x150 complete Raman spectra was acquired with an integration time per spectrum of 37 ms. From this array two distinct spectra were evaluated as shown in Fig. 1c. These two spectra correspond to the two polymorphic phases of TiO2: the rutile and anatase phases. The correlative Raman-SEM image (Fig. 1d) reveals that the large particles correspond to the rutile phase of TiO2, whereas the small particles consist of anatase TiO2, enabling to differentiate the fine crystallographic structure of the nanoparticles. Microsc. Microanal. 21 (Suppl 3), 2015 18 Figure 1. RISE microscope (a), SEM image of TiO2 nanoparticles (b), Raman spectra of rutile and anatase TiO2 (c), and correlative Raman-SEM image (d).");sQ1[10]=new Array("../7337/0019.pdf","The Surprising Dynamics of Electron Vortex Beams","","19 doi:10.1017/S1431927615000896 Paper No. 0010 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Surprising Dynamics of Electron Vortex Beams Stefan L�ffler1,2,3, Thomas Schachinger2,3, Michael St�ger-Pollach2, Peter Schattschneider3,4 1 2 Department of Materials Science and Engineering, McMaster University, Hamilton, Canada. University Service Centre for Transmission Electron Microscopy, Vienna University of Technology, Vienna, Austria. 3 Institute of Solid State Physics, Vienna University of Technology, Vienna, Austria. 4 Ecole Centrale Paris, Chaten�y-Malabry, France. Vortex structures were predicted for light optical beams in 1974 [1] and experimentally realized two decades later. Shortly after the discovery of vortex electrons [2], free electrons with quantized angular momentum could routinely be produced with the holographic mask technique [3]. Owing to their short wavelength, these matter waves can be focused to atomic size. Another novel aspect is their magnetic moment �B m quantized in multiples of the Bohr magneton, independent of the spin polarization. Both features make them extremely attractive as a nanoscale probe for solid state physics. The theory of electron vortex dynamics is well developed [4-7]. The most intriguing prediction is a quantized rotation of electron vortex beams under certain conditions � when they represent free electron Landau states � including zero- and cyclotron (double-Larmor)-rotation. It is caused by the LandauZeeman phase acquired over a wave packet's trajectory and is analogous to the image rotation of optical beams caused by a Berry phase. This is in obvious contradiction to standard electron optics that predicts Larmor image rotation in the magnetic lens field of a TEM. Recently, this contradiction could be unraveled by extending the theoretical model [6] to diffracting Laguerre Gaussian modes. It was shown that the peculiar rotational dynamics of electron vortices can be explained by a combination of slow Larmor- and fast Gouy rotations and that the Landau states naturally occur in the transition region in between the two extrema. This generalized description is confirmed by experimental data showing an extended set of peculiar rotations, including zero, cyclotron-, Larmor- and rapid Gouy-rotations all present in one single convergent electron vortex beam. Experiments were performed borrowing a method from visible light optics, namely obstructing half the vortex ring and observing the inclination of the shadow image of the edge when it propagates through the magnetic field of the lens. The rotation dynamics can be cast into a dimensionless universal form covering four orders of magnitude of angular frequency (Fig. 1). An important aspect of vortex beams is their propagation in matter. It can be shown that they are topologically protected when centered on an atomic column [8], and that they exchange orbital angular momentum with the medium (Fig. 2). Finally, we study the spin-orbit interaction of vortex electrons propagating in the magnetic field of the objective lens in a TEM. It turns out that such a beam, being post-selected on axis in the far field, acts as a perfect spin filter in the limit of infinitely small detectors. Increasing the detector size, the polarization Microsc. Microanal. 21 (Suppl 3), 2015 20 decreases rapidly, dropping below 10-4 for realistic setups. It has to be seen to which extent this figure of merit can be improved by optimizing parameters such as voltage and convergence angle. References: [1] J Nye and M Berry, Proceedings of the Royal Society of London, Series A 336 (1974), p. 165. [2] M Uchida and A Tonomura, Nature 464 (2010), p. 737. [3] J Verbeeck, H Tian and P Schattschneider, Nature 467 (2010), p. 301. [4] KY Bliokh et al, Physical Review Letters 99 (2007), p. 190404. [5] P Schattschneider and J Verbeeck, Ultramicroscopy 111 (2011), p. 1461. [6] KY Bliokh et al, Physical Review X 2 (2012), p. 041011. [7] P Schattschneider et al, Nature Communications 5 (2014), p. 4586. [8] A Lubk et al, Physical Review A 87 (2013), p. 033834. [9] The authors thank K. Bliokh and J. Verbeeck for fruitful discussions. The financial support of the Austrian Science Fund (I543-N20) is gratefully acknowledged. Fig. 1: Left: Experimental images of the vortex rotation dynamics. The scale bar is 20 nm. Right: Universal dimensionless rotation dynamics in a magnetic field along the z axis, scaled by the Rayleigh range zR. Angular frequency in units of R=v/zR, where v is the velocity of the electron. Fig. 2: Propagation of an m=1 vortex beam through a 20 nm thick Fe crystal in [0 0 1] orientation. The circles and squares mark atom positions in different layers of the sample. The curve shows the change of the expectation value of the orbital angular momentum of the vortex.");sQ1[11]=new Array("../7337/0021.pdf","Unveiling the OAM and Acceleration of Electron Beams","","21 doi:10.1017/S1431927615000902 Paper No. 0011 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Unveiling the OAM and Acceleration of Electron Beams Roy Shiloh1, Yuval Tsur1, Roei Remez1, Yossi Lereah1, Boris A. Malomed1, Vladlen Shvedov2, Cyril Hnatovsky2, Wieslaw Krolikowski2, and Ady Arie1. 1 Department of Physical Electronics, Fleischman Faculty of Engineering, Tel Aviv University, Tel Aviv 6997801, Israel 2 Laser Physics Centre, The Australian National University, Canberra ACT 0200, Australia Electron beams, specifically in a transmission electron microscope (TEM), are mainly used to investigate biological samples and materials. It was not until recently that investigation of special kinds of beams, namely vortex beams, has begun [1,2]. These beams are especially interesting because they carry orbital angular momentum (OAM) which may be coupled to the atomic wave-function, thus enabling probing of magnetic dichroism [3], for example. In light-optics these beams have long been known, and research into other types of beams, such as accelerating beams, is flourishing. Here we study the well known Airy beam [4,5] in the electron microscope � a shape-invariant, multi-lobed, nonspreading beam whose nodal trajectory follows a parabolic dependence, which has already been exploited in light-optics to overcome the diffraction limit implementing a "super-resolution" technique [6]. Where for the case of vortex beams the OAM property is of utmost importance, in this work * we develop a tool for easy measurement of the Airy's nodal trajectory coefficient, which is the defining property of the Airy beam, derive an elegant analytic model and verify it by fabrication of the relevant amplitude masks and consequent measurement and analysis. Our results agree completely with the proposed model, which is derived without approximations, and nicely relates light- to electron-optics via the geometric ray-tracing technique. -1 2 3 In the optics literature, the familiar form of the Airy beam is given by [0 ( - 2 /4 0 )], where 0 is the transverse length-scale, the de-Broglie k-vector and (, ) the transverse and longitudinal 2 3 coordinates, respectively. is the Airy function. The quantity 1/(4 0 ) is the nodal trajectory parameter, sometimes referred to as the acceleration of the Airy beam, due to the coordinate's parabolic dependence. The Airy beam is easily generated on the optical axis using, for example, the amplitude mask depicted in Fig.1f. This mask, a computer-generated hologram, recreates both the object and image of the encoded pattern in the Fourier (or diffraction) plane, which is why we observe two Airy-like patterns in Fig.1b,d. The nodal trajectory coefficient could be directly calculated by measuring the distance between two lobes and fitting it to the Airy function's zeros; this is difficult, however, since the Airy (a diffraction pattern) must be in focus for an accurate measurement, thus endangering the camera CCD. The accuracy also strongly depends on the resolution. A second method would be to take a focus series and follow the trajectory of the main lobe [5]. Instead, our method involves a cylindrical transformation, easily achieved in the TEM by using the stigmator lenses. It is interesting to note that the same astigmatic transformation is also useful for determining the orbital angular momentum of vortex beams. Mathematically, the cubic phase imposed by the mask on the astigmatic beam is Fouriertransformed, thus yielding the "astigmatic Airy" (see Fig.1), the curve of which is dependent on the nodal trajectory coefficient according to the following formula: 3 (2 0 )-1 (1) = (3 + )3/2 2 2 3 )-1 3/2 ( 0 (�2) = �(-3 + ) 2 * Accepted to Physical Review Letters, 2/2015. Microsc. Microanal. 21 (Suppl 3), 2015 22 3 where 2 = (2 0 )2 . These curves are marked in both Fig.1 and Fig.2, with , the coordinates in Fourier space, is a measure of the astigmatism and 0 the afore-mentioned length-scale. The resulting curves of the astigmatic Airy are dependent on the amount of astigmatism and the Airy acceleration, thus allowing flexible, accurate, and safe measurement of Airy beams with nearly-arbitrary acceleration and only one image taken. In conclusion, we show that the astigmatic transformation provides an easy method to reveal the OAM of vortex beams with integer OAM, and can be exploited to generate an astigmatic Airy and use it to measure the nodal trajectory from one image. While Airy beams have only recently been introduced to electron optics, research in light-optics yielded many interesting applications, the most promising of which could be super-resolution electron imaging. (c) Figure 1. TEM images of (a) Astigmatic Airy, (b) mildly Astigmatic Airy, both with matching curves (blue and magenta) derived from our analytic model. The double-sided pattern is a result of the amplitude nature of the computer-generated hologram: a reconstruction of the object and image. (c) an example of an on-axis amplitude Airy mask (diameter ~ 75um). [1] K. Y. Bliokh et al, Phys. Rev. Lett. 99 (2007), 190404. [2] B. J. McMorran et al, Science 331 (2011), 192. [3] J. Verbeeck et al, Nature 467 (2010), 301. [4] M. V. Berry, American Journal of Physics 47 (1979), 264. [5] N. Voloch-Bloch et al, Nature 494 (2013), 331. [6] S. Jia et al, Nat. Phot. 8 (2014), 302. [7] The work was supported by the Israel Science Foundation, grant no. 1310/13, by DIP, the GermanIsraeli Project cooperation and by the Australian Research Council. [8] Presentation of this work in M&M2015 is made possible by a travel grant from the Israeli Ministry of Science and Technology.");sQ1[12]=new Array("../7337/0023.pdf","An Orbital Angular Momentum Spectrometer for Electrons","","23 doi:10.1017/S1431927615000914 Paper No. 0012 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 An Orbital Angular Momentum Spectrometer for Electrons Tyler R. Harvey1, Vincenzo Grillo1,2,3, Benjamin J. McMorran1 1. 2. Department of Physics, University of Oregon, Eugene, OR 97405, USA CNR-Istituto Nanoscienze, Centro S3, Via G. Campi 213/a, I-41125 Modena, Italy 3 CNR-IMEM Parco Area delle Scienze 37/A, I-43124 Parma, Italy Measurement of magnetization with atomic resolution via dichroic electron energy loss spectroscopy (EELS) with vortex beams remains a hotly-discussed and as-yet unproven goal for electron microscopy [1-5]. A satisfactory explanation for the lack of successful demonstrations of this technique has not yet emerged. However, a number of recent predictions [2,5,6] for dichroism include one common oversight: post-selection of a final state without orbital angular momentum (OAM) is not automatically guaranteed in a real dichroic EELS experiment. Unlike dichroic absorption spectroscopy experiments with circularly polarized light, where the final photon state is a no-photon state which necessarily has zero angular momentum, electrons in a transmission electron microscope are not absorbed, and instead pass through a specimen. Incident electron vortices with ml OAM have a non-zero probability to scatter to an outgoing state with unchanged ml OAM. Dichroic electron energy loss spectroscopy, then, must involve some postselection for the component of the final state which has nonzero ml. There exist qualitative methods to post-select for more of this component of the final state [7], and semi-quantitative interferometric methods [8] that do not work for the inelastically scattered final states which are incoherent with respect to the initial state that one must measure in a dichroic EELS experiment. In order to accomplish dichroic EELS, a quantitative non-interferometric method to spatially separate OAM will be necessary. We propose a magnetic field-based mechanism for on-axis separation of electron OAM modes. Inside a magnetic lens field approximated by the Glaser model [9], with the addition of the lowest-order radial curvature, we can write the vector potential as (1) where B0 is the field strength at the center of the lens, R sets the longitudinal extent of the field, and = (x2+y2)1/2 is the radial position coordinate. The length parameter b defines the length scale of radial curvature. The Hamiltonian which results from the addition of the vector potential show in (1) includes, among other terms, two parabolic lensing terms: the standard Glaser lensing term, and additional term which depends on OAM. (2) Figures 1 and 2 show multislice-simulated [10] phase structure of vortices in the back focal plane of a lens which includes only the second, OAM-dependent lens term. The simulation parameters are chosen to exaggerate the effect for illustration purposes and are not physically reasonable. If we ignore the effect of higher-order aberrations produced by the vector potential in (1) and calculate the shift of the focal length of a thin magnetic lens due only to the OAM-dependent lensing, we see Microsc. Microanal. 21 (Suppl 3), 2015 24 (3) where f0 is the standard focal length of a thin magnetic lens [9] due to the first term in the Hamiltonian in (2). We expect that, under suitable conditions, this change of focal length can be used, in combination with a round aperture, to preferentially pass a single focused OAM mode. This spectrometer disperses by orbital angular momentum along the z-axis. Figure 3 shows simulated far-field intensities of the vortices shown in Figures 1 and 2 after interaction with the OAM-dependent lens. A major challenge in physical implementation of this mechanism will be to maximize the difference in focal length between modes that differ by one quantum of OAM. The unitless quantity is typically very small in standard magnetic lenses. To approach unity, one must either decrease the field strength such that the focal length becomes impractically large or decrease the curvature length scale below a reasonable size for machined lens components. If successfully implemented, the OAM measurement device we propose here will allow for near-perfect post-selection of OAM modes and therefore will enable far more reliable dichroic EELS experiments. References: [1] J. Verbeeck et al., Nature 467 (2010) p.301. [2] S. Lloyd et al., Phys. Rev. Lett. 108 (2012) p.074802. [3] P. Schattschneider et al., Phys. Rev. Lett. 110 (2013) p.189501 . [4] P. Schattschneider et al., Ultramicroscopy 136 (2014) p.81�85. [5] J. Rusz and S. Bhowmick, Phys. Rev. Lett. 111 (2013) p.105504. [6] J. Rusz et al., Phys. Rev. Lett. 113 (2014) p.145501. [7] K. Saitoh et al., Phys. Rev. Lett. 111 (2013) p.074801. [8] L. Clark et al., Phys. Rev. A 89 (2014) p.053818. [9] L. Reimer and H. Kohl, "Transmission Electron Microscopy: Physics of Image Formation", (Springer, New York), p.22. [10] V. Grillo et al., New Journal Phys 15 (2013) 093026. Figure 1: Simulated phase and intensity of +10 vortex subject to the OAM-dependent lens term with b=3 um, B0 = 0.01 T, R = 10 um, and beam energy E = 100 keV. Figure 2: Simulated phase and intensity of -10 vortex under the same conditions. Figure 3: +10 (small, teal) and -10 (large, red) vortices after propagation over 100m after passing through the spectrometer field. Scale applies to all images.");sQ1[13]=new Array("../7337/0025.pdf","Structured Electron Beam Illumination: A New Control Over the Electron Probe","","25 doi:10.1017/S1431927615000926 Paper No. 0013 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Structured Electron Beam Illumination: A New Control Over the Electron Probe Weird Probes and New Experiments Vincenzo Grillo,1,2,3 Jordan S Pierce3, Ebrahim Karimi,4, Tayler R Harvey3 , Roberto Balboni5, Gian Carlo Gazzadi1, Erfan Mafakheri6, Federico Venturi6, Benjamin J McMorran3 , Stefano Frabboni,1,6 and Robert W Boyd4,6 1. 2. CNR-Istituto Nanoscienze, Centro S3, Via G. Campi 213/a, I-41125 Modena, Italy CNR-IMEM Parco Area delle Scienze 37/A, I-43124 Parma, Italy 3. Department of Physics, University of Oregon, Eugene, 97403-1274 Oregon, USA 4. Department of Physics, University of Ottawa, 25 Templeton, Ottawa, Ontario, K1N 6N5 Canada 5. CNR-IMM Bologna, Via P. Gobetti 101, 40129 Bologna, Italy 6. Dipartimento FIM, Universit� di Modena e Reggio Emilia, Via G. Campi 213/a, I-41125 Modena, Italy 7. Institute of Optics, University of Rochester, Rochester, New York 14627, USA Since the introduction of electron vortex beams it has become increasingly clear the possibility to have an unprecedented control over the electron beams and to produce arbitrary electron beams [1][2][3]. Electron microscopy is basically rediscovering most of the progresses of optics. In this field, researchers have been using holograms based on liquid crystal technology to produce any kind of light beam shape. These beams are, for example, at the basis of some of the most interesting fundamental experiments of quantum mechanics [4]. It is well known that the equations for electron and photons, in paraxial approximation, are formally the same; however, what really paved the road to holographic electron optics has been the possibility to nanofabricate patterns on SiN membranes to induce a controlled phase plate [5][6]. For the sake of example we reported in fig 1 some noticeable electron beam shapes like a) Ince-Gauss and b) Helix Ince Gauss beams. c) A beam obtained by the combination of two Laguerre Gaussian beams with opposite topological charge +/- 3. All these beams are interesting because shape invariant upon propagation. Finally as an extreme example we demonstrate the possibility to shape the beam as the CNR institute logo. By the use of different phase-retrieval methods or measuring the hologram characteristics we were able to characterize the generated wavefunction and to evaluate the quality of the generated electron beam. We verified that our results are comparable in terms of quality with optical holograms. Beyond this basic analogy there are important added values in electron beams since 1) e-beams can reach atomic scale 2) electron possess charge. This motivate the use of complex beams and illumination structuring as new tools for material investigation. As a first example we will show how the new electron optical elements can be used to increase in different ways the STEM resolution. Fig 2a, for example, shows an holographic spherical aberration corrector. Fig 2b is the experimental intensity and the calculated intensity and phase ( as hue color) of a Bessel beam (in this case L=2) that can be also used for tomography and superresolution [3]. Fig 2c is the intensity and vortex phase of a beam with L=200 that can be used for its very large magnetic moment to couple with material magnetism. Therefore in this talk the state of the art of beam structuring and a series of present and future applications will be shown. Microsc. Microanal. 21 (Suppl 3), 2015 26 References: [1] J. Verbeeck, H. Tian, and P. Schattschneider, Nature 467, (2010) p 301. [2] B. J. McMorran, A. Agrawal, I. M. Anderson, A. A. Herzing, H. J. Lezec, J. J. McClelland, and J. Unguris, Science 331, 192 (2011). [3] V. Grillo, E. Karimi, G C Gazzadi, S Frabboni,M R. Dennis, and R W. Boyd Phys. Rev. X 4, 011013 (2014) [4] Alois Mair, Alipasha Vaziri, Gregor Weihs & Anton Zeilinger Nature 412 (2001) 313 [5] V. Grillo, G C Gazzadi, E. Karimi, E Mafakheri,R W. Boyd, and S Frabboni Applied Physics Letters 104, 043109 (2014) [6] T.R Harvey, J S Pierce1, A K Agrawal, P Ercius, M Linck and B J McMorran New Journal of Physics 16 (2014) 093039 a) b) c) d) Figure 1. a,b,c) Intensity of different shape invariant beams a) Ince-Gauss b) Helix Ince Gauss c) Laguerre Gauss ( a combination of states with orbital number=�3) . Additionally we show the ability on holographic shaping by producing a beam in the shape of the Logo of CNR. a) b) c) Figure 2. a) Hologram of a Spherical aberration corrector b) Tiled image of the experimental intensity (left) and calculated intensity/phase (right) for a Bessel beam L=2 c) Experimental image and calculated phase (after removing tilt effects) for a beam with Orbital angular momentum L=200. The color-scales in fig b,c are such that brightness is related to wavefunction intensity and hue indicates the phase.");sQ1[14]=new Array("../7337/0027.pdf","Direct Observation of Nano-porous Materials Using Low Voltage High Resolution SEM","","27 doi:10.1017/S1431927615000938 Paper No. 0014 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Direct Observation of Nano-porous Materials Using Low Voltage High Resolution SEM Shunsuke Asahina1, Yusuke Sakuda1 and Osamu Terasaki2 1. 2. JEOL Ltd., 3-1-2 Musashino, Akishima, Tokyo, 196-8558, Japan Dept of Materials & Environmental Chemistry, EXSELENT, Stockholm Univ, Stockholm, Sweden & Graduate School of EEWS, KAIST, Daejeon, Republic of Korea Understanding surface fine structures of nano-porous materials and their compositions is essential for controlling the growth of these materials and for the future material design and the utilization of their functions [1]. Technical developments in low acceleration voltage high resolution scanning electron microscopy (LVHR-SEM) have made it possible to acquire invaluable information about them even though the materials are electrically insulating and electron beam sensitive. In this report, we focused on observations with electron beam at ultra-low impact energy in order to acquire information from top surface of nano porous materials. We have developed a new objective lens called the super hybrid lens (SHL): a compound lens consisting of both magnetic and electrostatic lenses [2]. The SHL is capable of producing a small probe size even at low accelerating voltages (for example 0.7 nm at 1 keV); moreover, using beam deceleration mode allows imaging down to 10 eV with high spatial resolution [3]. Both SHL and beam deceleration mode are installed in a JEOL Field Emission SEM (FE-SEM) JSM-7800F Prime. The sample preparation is also important for LVHR-SEM observation. Especially, we have to avoid contamination, since sample contamination has a negative effect on imaging quality. It is particularly detrimental in low voltage imaging where only the top few nanometers of the structure are probed with the electron beam. The main sources of contamination are from the specimen itself (physically adsorbed gaseous species), the specimen holder and the microscope chamber. Therefore, we selected a highdensity conductive carbon stub for LVHR-SEM observations. This stub was first polished with sandpaper (#1200) and then with a filter paper to make the surface smooth. It was then cleaned by ultrasonication in alcohol and heated in vacuum. The samples described in this paper were prepared by one of two methods. With the first method, nano particles were dispersed in solution and then a drop was placed on the carbon stub. The stub was then heated in a vacuum oven for 10 min at 200�C and subsequently cooled to room temperature. The second preparation method was employed for dried particles, which were picked with a cotton ball and scattered on a carbon stub that was polished with filter paper and dried on a hot plate at 250�C. Mesoporous zeolite LTA [4] images using low landing energy are show at Fig.1. Fig.1 (a) shows low magnification of mesoporous zeolite LTA at 80 eV as landing energy; (b) and (c) show high magnification of mesoporous zeolite LTA at different landing energies such as 80 eV and 500 eV. All images clearly show top surface topographical information of mesoporous zeolite LTA. Especially, the image of 80 eV shows less edge effect due to small interaction volume. That is a great feature of LVHRSEM, since it can show fine edges and give high accuracy measurement of nano-porous materials. The images also demonstrate that higher landing voltage results in specimen damage due to electron beam irradiation (see Fig. 1 (c)), whereas lower landing voltage allows observation without any specimen damage (Fig.1 (b)). Microsc. Microanal. 21 (Suppl 3), 2015 28 We will also demonstrate the ability of LVHR-SEM to not only obtain high spatial resolution image with minimal electron beam damage, but also to acquire high spatial resolution chemical information by using Energy Dispersive X-ray Spectrometer (EDS) or Soft X-ray Emission Spectrometer (SXES) [6]. References: [1] S. Asahina, et al., ChemCatChem. 1038-1048, June 2011. [2] J. Frosien, et al., J. Vac. Sci. Technol. B7 (6). 1874, 1989. [3] S. Asahina, et al., Microscopy and Analysis.S12-S14, November 2012. [4] K. Cho, et al., Solid State Sciences. Vol. 13. (4). 750�756, April 2011 [5] M. Suga et al., Prog. Solid State Chem. 42, 1�21 (2014). [6] S.Asahina, et al., APL mat. 2, 113317 (2014). Acknowledgement: Mesoporous zeolite LTA crystals were kindly provided by Prof Ryong Ryoo (KAIST & IBS, Korea). a: Impact energy: 80 eV b: Impact energy: 80 eV c: Impact energy: 500 eV Fig. 1. Low landing energy images at mesoporous zeolite LTA");sQ1[15]=new Array("../7337/0029.pdf","Examination of Graphene in a Scanning Low Energy Electron Microscope.","","29 doi:10.1017/S143192761500094X Paper No. 0015 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Examination of Graphene in a Scanning Low Energy Electron Microscope. Ilona M�llerov�, Eliska Mikmekov� and Ludk Frank Department of Electron Microscopy, Institute of Scientific Instruments ASCR, v.v.i., Brno, Czech Republic. Although graphene has been available [1-2] and intensively studied [3-5] for a full decade, new methods are still required for its examination and diagnostics. Even checking the continuity of layers and the reliable counting of layers of graphene and other 2D crystals should be easier to perform. Scanning low energy electron microscope (SLEEM) equipped with a cathode lens [6] offers an innovative tool enabling one to see graphene samples at nanometer lateral resolution in both transmitted and reflected electrons and to count the number of layers. This diagnostics can be performed on freestanding graphene samples as well as on graphene grown on the surfaces of bulk substrates. The freestanding graphene samples were first examined in the standard vacuum high resolution SLEEM. Fig. 1 shows micrographs taken in the reflected electron (RE) as well as transmitted electron (TE) mode at several energies. The RE signal was composed of both secondary and backscattered electron emission, accelerated in the cathode lens field toward the detector. In the RE frames the maximum contrast between the graphene layers and lacey carbon appears at 1 keV and decreases toward higher and lower energies because of extending and shortening information depth, respectively. These images identify empty holes but do not reveal thicker islands of graphene. In the TE mode we do not see multilayer graphene islands above 100 eV. This fact underlines the suitability of very low energy electron microscopy for examination of 2D crystals. Interpretation challenges are presented by some details inverting their contrast more than once, see the arrow. These probably arise from contaminations that become charged. Usually used counting of graphene layers by Raman spectroscopy is faced by the issue of the low lateral resolution of light optical imaging. SLEEM provides much higher resolution, so it is worth checking its selectivity for the same purpose. Fig. 2 left shows that the contrast of individual graphene layers is preserved down to units of eV. Measurement of the transmissivity was calibrated between the zero signals on mesh rungs and the full signal in empty holes, see Fig. 2 and Table 1. The transmissivity of the graphene samples naturally depends on cleanness of the surface. We have established that while fast electrons decompose the adsorbed hydrocarbon molecules creating a carbonaceous contamination layer, below 50 eV electrons release these molecules and leave surface of the graphene atomically clean, see Fig. 3 [7]. References: [1] KS Novoselov et al, Science 306 (2004) p. 666. [3] JC Meyer et al, Nano Letters 8 (2008) p. 3582. [4] PY Huang et al, Nature 469 (2011) p. 389. [5] W Zhou et al, Microscopy and Microanalysis 18 (2012) p. 1342. [6] I M�llerov� and L Frank, Advances in Imaging and Electron Physics 128 (2003) p. 309. [7] The authors acknowledge funding from the Technology Agency of the Czech Republic (Competence center Electron microscopy, no: TE01020118) and from the MEYS of the Czech Republic (LO1212). Microsc. Microanal. 21 (Suppl 3), 2015 30 Figure 1. Freestanding graphene sample of 3 to 5 layers imaged in reflected and tramnsmitted electrons. Table 1: Total transmissivity measured on graphene samples for 40 eV incident electrons. No. of graphene layers Transmissivity (%) 1 11.9 2 9.0 3 6.5 4 4.9 5 3.5 6 2.6 7 2.0 Figure 2. Micrographs of a 3to5LG graphene sample taken in a UHV microscope (left). The measured energy dependence of transmissivity (right). Figure 3. Changes in properties of 1 LG due to prolonged bombardment with 30 eV electrons shown in the transmission (a) and reflection (b) modes. Quantitative development of 1LG transmissivity at 50 eV (c) and 100 eV (d) in dependence on vacuum conditions and the energy of electrons used for prolonged bombardment.");sQ1[16]=new Array("../7337/0031.pdf","Conductivity Contrast in SEM Images of Hydrogenated Graphene Grown on SiC","","31 doi:10.1017/S1431927615000951 Paper No. 0016 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Conductivity Contrast in SEM Images of Hydrogenated Graphene Grown on SiC Iwona Jozwik1, Jacek M.Baranowski1, Kacper Grodecki1, Pawel Dabrowski2 and Wlodzimierz Strupinski1 1. 2. Institute of Electronic Materials Technology, 133 Wolczynska Str., 01-919 Warsaw, Poland. Department of Solid States Physics, Faculty of Physics and Applied Informatics, University of Lodz, Pomorska 149/153, 90-236 Lodz, Poland Graphene consists of a planar single sheet of sp2-bonded carbon atoms arranged in a two-dimensional (2D) honeycomb lattice. One of methods of obtaining epitaxial graphene is growth on SiC(0001) by CVD (Chemical Vapour Deposition) [1]. In this growth mode the first carbon layer it is not a graphene one, it is attached to Si atoms of the substrate by sp3 bonds and it is called the buffer layer. The conversion of the buffer layer into graphene may be obtained by hydrogenation process. Hydrogen molecules introduced between the buffer layer and the SiC substrate, break most of the sp3 Si � C bonds and the buffer layer converts into graphene lattice [2]. Graphene grown by CVD technique on 4H-SiC(0001) substrates and hydrogenated at a high temperatures were subjected to SEM investigation with low energy (~0.2 keV) electrons using Auriga CrossBeam Workstation (Carl Zeiss) equipped with In-lens SE (true SE1) detector and Energy selective Backscattered electrons (EsB, low-loss BSE) detector, both positioned on the optical axis of the Gemini (TM) column. The main aim of the work was to determine the contrast origin of the specified areas of the graphene intercalated by H2. The results obtained by low-kV SEM were supported by LC-AFM (Local Conductivity AFM) technique. The presence of graphene has been confirmed by Raman spectroscopy measurements. Figure 1 presents the SEM images obtained in the In-Lens (a) and EsB (b) detectors in parallel. Both images show characteristic dark parts in the middle of the terraces, surrounded by brighter zones. The bright lines visible in the images are thermally driven cracks in the graphene (due to the high temperature during hydrogenation process) uncovering the SiC substrate. Assuming that the contrast in the SE image (In-lens) is mostly originating from the conductivity differences, and the image of low-loss BSEs is based on the compositional contrast, the reason of the unquestionable contrast of carbon layers may be explained as follows. It has been shown that the graphene buffer layer grown of SiC(0001) is not an uniform one and the area close to step edges are well saturated with carbon atoms [3]. On the other hand, terraces are not so well saturated with carbon atoms and characterize with a larger concentration of defects [3]. The hydrogenation of well saturated regions, close to step edges, will result in breaking Si � C sp3 bonds between buffer layer and substrate, and formation of Si � H bonds and the p-type conductive graphene layer. However, the defected graphene terrace regions, in addition to formation of Si � H bonds underneath of graphene layer, may have defects of a donor character, which will compensate p-type conductivity introduced by the Si � H bonds. Therefore, the central regions of terraces may have less conductive character and to be depleted of charge carriers. These subtle phase differences are clearly visible in the EsB images obtained in a very low energy regime, where the yield of low-loss BSEs is strongly influenced by the nature of molecular bonds, thus the presence of defects in graphene may create differences in scattering coefficient. The results have been confirmed by AFM measurements in Microsc. Microanal. 21 (Suppl 3), 2015 32 LC-AFM mode. Figure 2b presents AFM image showing local conductivity map of the same sample, with a high-conductivity regions of graphene along step edges and a low-conductivity on terraces. References: [1] C Riedl, C Coletti, T Iwasaki, AA Zakharov and U Starke, Phys. Rev. Lett. 103 (2009), p. 246804. [2] W Strupinski, K Grodecki, A Wysmolek, R Stepniewski, T Szkopek, PE Gaskell, A Gruneis, D Haberer, R Bozek, J Krupka and JM Baranowski, Nano Lett. 11 (2011), p. 1786. [3] W.Strupinski,K.Grodecki, P.Caban, I.Jozwik-Biala, J.M.Baranowski , Carbon 81, 63-72(2015) [4] This work was partially supported by the National Centre for Research and Development project GRAFTECH/NCBR/02/19/2012, the European Union Seventh Framework Programme under grant agreement no. 604391Graphene Flagship, the National Center for Research and Development under the GRAFTECH/NCBiR/12/14/2013 GRAF-MAG and the National Centre of Science under the UMO2012/07/N/ST3/03141 Preludium 4 grant. a) b) Figure 1. SEM images of CVD-grown graphene intercalated with H2: a) SE image, b) low-loss BSE image. a) b) Figure 2. AFM and LC-AFM images: a) topography b) map of local conductivity (dark regions are connected with a low and the bright ones with a high conductivity).");sQ1[17]=new Array("../7337/0033.pdf","Improved Low Voltage SEM Image Resolution Through the Use of Image Restoration Techniques","","33 doi:10.1017/S1431927615000963 Paper No. 0017 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Improved Low Voltage SEM Image Resolution Through the Use of Image Restoration Techniques Matthew D. Zotta1, Yudhishthir P. Kandel1, Andrew N. Caferra1 and Eric Lifshin1 1 College of Nanoscale Science and Engineering, SUNY Polytechnic Institute, Albany, NY, USA Low voltage SEM or LV-SEM (<5 keV) is particularly useful when observing insulators and beam sensitive materials, as well as for delineating fine surface structure [1]. Obtaining quality images at low beam energies can be challenging, however, because chromatic aberrations cause significant increases in the probe size. Since the probe size increases with the source energy spread, field emission guns (FEGs) or Schottky sources are preferred over thermionic sources. However, chromatic aberrations are still a problem because they lead to an increase in the probe size when the probe energy is decreased. To overcome this, manufacturers often use relatively high gun energies (>5 keV) and introduce a retarding field at the end of the column to lower the landing energy of the electrons. In this study, the beam retardation requirement is eliminated by working with a larger beam than the resolution needed and recovering the resolution by the method described by Lifshin et. al. [2] in which the point spread function (PSF) of the electron beam is determined and used to deconvolute a blurred image formed by a large probe. The PSF determination is based on a comparison between a reference image and an observed image acquired with the chosen operating conditions. In the previous work, the reference image was obtained with a high resolution microscope at a magnification approximating pixel level resolution. The blurred image was then obtained at the same beam energy and magnification, but with a larger probe size on a lower resolution microscope. In the current study, a single high resolution microscope is used where a reference image is obtained at high beam energy with pixel level resolution and used to establish the PSF for low beam energies. This requires knowledge of the materials in the reference standard and their respective secondary electron yields [3] so the image can be corrected for beam energy dependence. Furthermore, suitable adjustments for brightness and contrast were made. These corrections were used to calculate an approximate reference image at the lower beam energy and the subsequent PSF (Figure 3) from the low energy image. Figure 1 shows an Au-C Pella� image (c) calculated for 2 keV from a reference image taken at 20 keV using a TESCAN MIRA�. The measured image at 2 keV is given in (a) and the restored image in (b). The PSF obtained in this way was then applied to a Sn-C Pella� sample as shown in Figure 2 where (a) is the observed 2 keV image, (b) is the restored image and (c) is a high resolution 20 keV image taken of the same area for comparison of detail. The improvement in image resolution in the restored images can be estimated based on their contrast transfer functions (Figure 4) using the method developed by Joy [4]. References: 1. Michael, J.R., Scanning 33, 147-154 (2011). 2. Lifshin, E., Kandel, Y.P. and Moore, R.L., Microscopy and Microanalysis 20 (01), 78-89 (2014). 3. Lin, Y. and Joy, D. C., Surf. Interface Anal. 37, 895�900 (2005). 4. Joy, D.C., J Microsc 208 (Pt1), 24-34 (2002). The authors acknowledge the support of Mr. Jeffrey Moskin, President of Nanojehm for providing the resources that made this study possible as well as TESCAN for providing instrumental support. Microsc. Microanal. 21 (Suppl 3), 2015 34 a b c 250nm Figure 1. Au-C Pella sample: (a) 2 keV image, (b) restored 2 keV image, (c) 20 keV pixel level resolution image recalculated at 2 keV. � m a b c 250nm Figure 2. Sn-C Pella� sample: (a) 2 keV image, (b) the restored 2 keV image, (c) same area at 20 keV. m Au (nm) Sn (nm) Figure 3. The PSF determined from the 2 keV observed image relative to the 2 keV reference image. Figure 4. The contrast transfer functions for the Au and Sn 20 keV and 2 keV images and the 2 keV restorations.");sQ1[18]=new Array("../7337/0035.pdf","20-kV Diffractive Imaging of Graphene by using an SEM-based Dedicated","","35 doi:10.1017/S1431927615000975 Paper No. 0018 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 20-kV Diffractive Imaging of Graphene by using an SEM-based Dedicated Microscope Osamu Kamimura1, Takashi Dobashi1, Yosuke Maehara2, Ryo Kitaura3, Hisanori Shinohara3, and Kazutoshi Gohara2 1. 2. Central Research Laboratory, Hitachi, Ltd., Kokubunji-shi, Tokyo, Japan Division of Applied Physics, Faculty of Engineering, Hokkaido University, Sapporo, Japan 3. Department of Chemistry & Institute for Advanced Research, Nagoya University, Nagoya, Japan The optical microscope has reached a resolution finer than the wavelength of light; however, in the electron-microscopy field, decrease aberrations of lenses remains a challenge. Even with the help of aberration correctors and monochrometers, the resolution is limited to more than multiples of ten times the wavelength of the electron beam [1]. On the other hand, diffractive imaging, which is an imaging method using iterative phase retrieval from a diffraction pattern [2], can obtain high-resolution images without suffering any aberrations of the imaging lens. This method (using low acceleration voltage) have been applied by the authors to reach atomic resolution with a dedicated microscope based on a conventional scanning electron microscope (SEM) [3, 4], and this microscope resolved the atomic arrangement of a single-wall carbon nanotube at 30 kV [5]. In the present study, the atomic arrangement of multi-layer graphene in the case of an acceleration voltage of 20 kV was reconstructed, and the possibility of reconstruction of a non-periodic structure was investigated. Diffraction patterns were experimentally recorded by using a dedicated microscope based on in-lens type SEM (S-5500, Hitachi High-Technologies Corporation) with a film-loader system of a transmission electron microscope [4]. Graphene films were synthesized by CVD and transferred on a specimen stage with fine hole of about-75-nm diameter. A pattern of multi-layer graphene reconstructed from a diffraction pattern recorded by using an electron beam with an acceleration voltage of 20 kV (namely, wavelength of 0.00859 nm) is shown in Figure 1(a). Each bright spot ("atomic column" hearafter) shows the intensity of a single carbon atom or a few carbon atoms. This resolved atomic arrangement of graphene shows the resolution of this imaging is finer than 0.12 nm (14 times the wavelength, which is close to the results with aberration correctors). The atomic arrangement shown in Fig. 1(a) is composed of three different contrasts (low, medium, and high intensities) in each carbon atomic column. And the order of the contrast of the atomic column is regularly repeated. A model of the atomic arrangement of ABA-stacking tri-layer graphene is shown in Figure 1(b), and it shows three different forms of stacking carbon atoms (i.e., single atom, two atoms, and three atoms). Intensities of the pixels along lines #1 to #8 (open and close symbols), which correspond to each line in Fig. 1(a) and fitted Gaussian distribution are shown in Figure 1(c). Comparison the Gaussian distribution with the atomic arrangement model and a simulation of projected potential reveals that the graphene shown in Fig. 1(a) has tri-layer structure with ABA stacking. To investigate issues concerning reconstruction of a non-periodic structure such as grain boundaries or defects in crystals, diffraction patterns of defective graphene with various noises were simulated, and the atomic arrangement of the graphene was reconstructed from the simulated diffraction patterns. The defective structure could be reconstructed from a diffraction pattern with high signal-to-noise ratio, but only the periodic atomic arrangement of graphene was reconstructed from a low-signal-to-noise-ratio Microsc. Microanal. 21 (Suppl 3), 2015 36 diffraction pattern. It is thus concluded from this result that a defective atomic arrangement can be imaged by diffractive imaging and that high-signal-to-noise-ratio recording of diffraction patterns is essential to reconstruct a non-periodic structure. In summary, low-voltage diffractive imaging was shown to have the possibility of atomic-resolution imaging that can reconstruct not only periodic structures but also non-periodic structures. It is concluded that this method will open up new possibilities of application of electron microscopy. References: [1] R Erni et al, Phys. Rev. Lett. 102 (2009) p. 096101., OL Krivanek et al, Ultramicroscopy 110 (2010) p. 935., S Uno et al, Optik 116 (2005) p. 438., UA Kaiser et al, Ultramicroscopy 111 (2011) p. 1239., T Sasaki et al, Ultramicroscopy 145 (2014) p. 50. [2] D Sayer, Acta Crystallogr. A5 (1952) p. 843, RW Gerchberg and W.O. Saxton, Optik 35 (1972) p. 237., JR Fienup, Appl. Opt. 23 (1982) p. 2758., J Miao et al, Nature 400 (1999) p. 342., JM Zuo et al, Science 300 (2003) p. 1419. [3] O Kamimura et al, Appl. Phys. Lett. 92 (2008) p. 024106. [4] O Kamimura et al, Ultramicroscopy 110 (2010) p. 130. [5] O Kamimura et al, Appl. Phys. Lett. 98 (2011) p. 174103. [6] Part of this work was supported by the Japan Science and Technology Agency. (a) 1 2 3 4 5 6 7 8 12 11.5 (b) Single atom Two atoms Three atoms 0.5 nm : 1st layer (A stacking) : 2nd layer (B stacking) : 3rd layer (A stacking) (c) Int. along line 1 Int. along line 3 Int. along line 5 Int. along line 7 Gaussian fitting Int. along line 2 Int. along line 4 Int. along line 6 Int. along line 8 (Int.: Intensity) Intensity (arb. unit) 11 10.5 10 9.5 9 8.5 8 -0.05 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 Position (nm) Figure 1. (a) A pattern of graphene reconstructed from a diffraction pattern obtained at 20 kV. (b) Atomic-arrangement model of ABA stacking tri-layer graphene. (c) Intensity distributions of reconstructed pattern along eight lines and fitted Gaussian distribution with three peaks.");sQ1[19]=new Array("../7337/0037.pdf","Bridging the Gap between the Modeling Approach and the Experiment in Atom Probe Tomography","","37 doi:10.1017/S1431927615000987 Paper No. 0019 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Bridging the Gap between the Modeling Approach and the Experiment in Atom Probe Tomography Fran�ois Vurpillot1, Williams Lefebvre1, Lorenzo Rigutti1, Nicolas Rolland1, Florian Moyon1, Lorenzo Mancini1, Hocine Hideur1, Didier Blavette1 1. Groupe de Physique des Mat�riaux, UMR CNRS 6634, Universit� et INSA de Rouen, Saint Etienne du Rouvray, France Over the last twenty years, major breakthroughs in the development of Atom probe tomography (APT) make this nanoanalysis instrument an indispensable companion for the nano-scientist. For instance, complex nano-electronic device analyses are now routinely performed [1][2] with the introduction of laser pulsing [3][4], wide field of view atom probe and local electrode technology [7][8][9]. However, the standard tomographic reconstruction protocol remains very close to the original one developed at the early stage of the technique [10]. Even if attempts have been made to propose more evolved recipes, back projection onto a hemispherical specimen surface is still routinely used [8]. In a heterogeneous sample made up of several phases with different evaporation behaviors, the evaporation sequence is highly non uniform. It results in a non-hemispherical complex surface shape, which in turns gives rise to artefacts during reconstruction. Improving both the precision and the accuracy of Atom Probe Tomography reconstruction requires a correct understanding of this imaging process. In this aim, numerical modelling approaches have been developed for 15 years. The injected ingredients of these modelling tools are related to the basic physic of the field evaporation mechanism. The interplay between the sample nature and structure of the analyzed sample and the reconstructed image artefacts have pushed to gradually be more skeptical on the standard reconstruction protocol. The reconstruction artefacts have been first postulated and then demonstrated by numerical simulations, and they are now directly demonstrated experimentally by APT and TEM complementary analysis [2][8]. However, if great efforts have been made to understand the imaging process, at the moment, there are few theoretical frameworks to use the models in order to better reconstruct. The problem comes from first the difficulty to model the real structure of interest; the sample structure, the composition and its geometry are roughly known (this is the information sought by the nano-scientist). Second, the physical parameters injected in the model are also roughly known. Third, the model itself is complex, and time consuming. In order to bridge the gap between the modelling approach and the experimental reconstruction protocol, several parallel ways must be followed. First improvements in the physical understanding of the emission process must be done, in order to refine the physical constants that are required and injected in the computation. A brief review of the models used in this aim will be presented. The modelling approach must also be improved, to be faster, and simpler, giving access to more straightforward parameters for the experimental point of view of the APT user. A pseudo analytic model was developed in this aim (fig. 1). In the current work, on comparison with numerical simulations, we validate the model and we highlight the origin of some field evaporation features. Finally, this paper will show the advantages to use combined characterization techniques to improve our knowledge of the sample before, after or during the imaging process in APT. Characterization of the same object with other techniques such as TEM tomography, and micro-photoluminescence proved to be essential to make the link between the model and the reconstructed image. Microsc. Microanal. 21 (Suppl 3), 2015 38 References: [1] A. K. Kambham, et al., Ultramicroscopy, vol. 111, no. 6, pp. 535�539, May 2011. [2] A. Grenier, et al.," Ultramicroscopy, vol. 136, pp. 185�192, Jan. 2014. [3] B. Gault, et al." Rev. Sci. Instrum., vol. 77, no. 4, p. 043705, Apr. 2006. [4] J. H. Bunton, et al.," Microsc. Microanal., vol. 13, no. 06, pp. 418�427, 2007. [7] T. F. Kelly et al. Microsc. Microanal. Off. J. Microsc. Soc. Am. Microbeam Anal. Soc. Microsc. Soc. Can., vol. 10, no. 3, pp. 373�383, Jun. 2004. [8] D. J. Larson et al., Local Electrode Atom Probe Tomography: A User's Guide. New York: Springer, 2013. [9] M. K. Miller and R. G. Forbes, Atom-Probe Tomography: The Local Electrode Atom Probe. New York: Springer-Verlag New York Inc., 2014. [10] P. Bas et al.," Appl. Surf. Sci., vol. 87�88, pp. 298�304, Mar. 1995. (a) (b) (c) Figure 1. Comparison of the radial density of impacts in function of the analyzed depth obtained by simulating the evaporation of a layer B with high evaporation field (EB=1.2EA) deposited on a low evaporation field planar substrate A with (a) an atomistic simulation of the evaporation (b) a faster analytic simulation. (c) Schematic drawing of the tip modelled in both cases.");sQ1[20]=new Array("../7337/0039.pdf","Atom-Probe Tomography Measurements of Isotopic Ratios of High-field Materials with Corrections and Standardization: a Case Study of the 12C/13C of Meteoritic Nanodiamonds","","39 doi:10.1017/S1431927615000999 Paper No. 0020 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atom-Probe Tomography Measurements of Isotopic Ratios of High-field Materials with Corrections and Standardization: a Case Study of the 12C/13C of Meteoritic Nanodiamonds J. B. Lewis1, D. Isheim2, C. Floss1, T. L. Daulton1,3, D. N. Seidman2 1. 2. Laboratory for Space Sciences, Physics Department, Washington University, St. Louis, MO, USA. Center for Atom-Probe Tomography, and Dept. of Materials Science and Engineering, Northwestern University, Evanston, IL, USA. 3. Institute of Materials Science and Engineering, Washington University, St. Louis, MO, USA. Efforts are ongoing to measure the 12C/13C ratios of individual (~2.6 nm diameter) meteoritic nanodiamond (ND) grains using atom-probe tomography (APT) [1]. The origins of meteoritic NDs remain an enigma, which could be resolved by isotopic measurements of individual ND grains. Isotopic anomalies in trace elements, chiefly Xe in bulk meteoritic acid dissolution residues [2], suggest a Type II supernova origin for at least a subpopulation of the NDs. While the bulk measured C [3] and N [4] are consistent with formation in the Solar System, bulk measurements may not be diagnostic. Poorly-crystalline to amorphous carbonaceous phases, which can comprise a significant fraction of the residue [5], are potentially of different origin than the NDs and hence can have different C isotopic compositions. Measurements of NDs by APT is challenging because C, being a "high field" element, is known to yield biased isotopic ratios prior to correction [1, 6]. We outline an analysis procedure to calculate isotopic ratios for C, employed on an ensemble of detonation synthesized ND standards (DND) and meteoritic ND residue from the Allende DM separate (ADM). A full description of specimen preparation is given in [1, 6]. After focused ion-beam microscope liftout, APT of NDs embedded between sputter-deposited Pt layers was performed with a Cameca LEAP 4000X Si [1, 6], yielding 3D positions and time-of-flight data for ~57% of the atoms (i.e., collection yield) in needle-shaped sample "nanotips" of radius 20�100 nm. We use laser pulsing to reduce the probability of nanotip fracture; however, higher laser pulse fractions lead to more delayed evaporation events, which manifest themselves as peak-tails. Since isotopes of the same element generally have similar mass-to-charge-state (m/n) ratios, they are particularly susceptible to tail interferences. Peak shape is determined by the base temperature, length of the laserinduced thermal pulse, nanotip shape, and material properties. Unfortunately, no standardized peak-fitting method exists for APT. In the procedure we developed, the background is calculated by a linear fit to the baseline of the time-of-flight spectrum and subtracted. We then define the mass range about a given major peak as the full width at half maximum (FWHM). The bin size is set so that the defined range is five bins wide. For minor isotopes we use the same range width as used for the major isotope, rather than FWHM; e.g., 13C+ is fit with the same range width as 12C+. This is because for same-element, same-charge-state ions, none of the factors affecting peak shape should vary, and the shift in time-of-flight/amu over 1 amu is negligibly small compared to bin size. Therefore, ranges are better defined by the largest peaks. Finally, for 13C (+ & ++) the contribution from the tails of 12C (+ & ++) is estimated and subtracted. This procedure is largely based on [7]. Mean standard ratios are calculated along with standard error of the mean (M) and standard deviation. The standard data are compared to Allende data in Fig. 1. M is representative of the uncertainty in the determination of the mean value, while the standard deviation represents the scatter of the data. The standards reproduce the significant instrumental artifacts, noted in prior work, which lead to an underestimation of 12C/13C ratios [1, 6]. The mean Allende ratio (for NDs and poorly-crystalline to amorphous C residuals) is, within uncertainties, the same as that of the mean standard ratio. The scatter in the ADMs is not significantly greater than for the standards, consistent with isotopically normal grains. We conclude that little if any of the carbonaceous material we have analyzed from the Allende residue is presolar. Given our current sample size, the presence of nondiamond carbonaceous material, and remaining instrumental artifacts, this conclusion should not be applied too Microsc. Microanal. 21 (Suppl 3), 2015 40 broadly. The mean ratio of 12C/13C in charge state + is significantly greater than in charge state ++. There is no reason to believe the different charge states of C would have different isotopic ratios; therefore, we take this difference to be the result of an experimental artifact that increases the measured 12C+ and/or 13C++ as well as possibly decreasing the measured 12C++ and/or 13C+. (12CH)+ hydride interference would give the opposite effect. If 12C++ is preferentially affected by deadtime effects compared to 12C+, it could explain the discrepancy, as could a 12C2++ interference at 12 amu. Iterative proportional fitting of detected multiple isotope pairs [8], or Poisson statistics [9] can be used to generate a correction for signal loss due to detector deadtime, providing there are enough counts. To date, however, such methods have generated only small changes to the measured ratios. On average the deadtime correction to the 12C++ peak is several orders of magnitude greater than to the 12C+ peak. We intend to survey available APT ND data to determine if undercorrected deadtime signal loss is responsible for the low measured standard ratios, and to assess the effect of 12C2++ and (12CH)+ on C-ratio measurements. Correlated transmission electron microscopy and APT of nanotips will allow us to compare the crystal structure of the C residue with the reconstructed m/n data, and allow us to distinguish between ND and amorphous C phases. References: [1] Heck P. R. et al, Meteoritics and Planetary Science 49 (3) (2014), pp. 453�467. [2] Lewis R. S. et al, Nature 326 (1987), pp. 160�162. [3] Russell S. R. et al, Meteoritics and Planetary Science 31 (1996), pp. 343�355. [4] Marty B. et al, Science 332 (2011), pp. 1533�1536. [5] Stroud R. M. et al, Astrophysical Journal Letters 738 (2011), L27�L31. [6] Isheim D. et al, Microscopy and Microanalysis 19 (Suppl 2) (2013), CD974�CD975. [7] Hudson D. et al, Ultramicroscopy 111 (2011), pp. 480�486. [8] Lewis J. B. et al, 46th Lunar and Planetary Science Conference (2015), #1480. [9] Stephan T. et al, International Journal of Mass Spectrometry (2015, in press). [10] Coplen T. B. et al, Pure and Applied Chemistry 74 (10) (2002), pp. 1987�2017. [11] This work is supported by NASA grants NNX14AP15H (J.B.L.) and NNX13AF53G (C.F.). The atom-probe tomograph at the Northwestern University Center for Atom-Probe Tomography (NUCAPT) was acquired with support from NSF and ONR. Fig. 1 Mean C isotopic ratios for 24 terrestrial detonation ND standards and 21 Allende DM ND data sets. 18 of the Allende measurements had sufficient 13C++ counts for ratio calculations; each of these is plotted as a diamond symbol with �2 error bars. Small ovals represent �2� the standard error of the mean. Large ovals represent �2� the standard deviation of a datum from the mean. The diagonal line is the line of equal ratio for isotopes at the two different charge states. The horizontal and vertical lines are the expected Solar System isotopic 12C/13C ratio, ~89 [10].");sQ1[21]=new Array("../7337/0042.pdf","Data Quality Improvements in the Voltage-Pulsed LEAP 5000 R/XR","","41 doi:10.1017/S1431927615001002 Paper No. 0021 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Data Quality Improvements in the Voltage-Pulsed LEAP 5000 R/XR R.M. Ulfig, T.J. Prosa, D.R Lenz, and T.R. Payne CAMECA Instruments, Inc., Madison, WI USA. Performance advances in atom probe tomography (APT) in recent years have driven a dramatic increase in the number of APT articles published. One area of rapid research growth is analysis of fragile or insulating materials that require a laser-pulsed atom probe [1]. However, for multi-user facilities around the world, voltage pulsing is still used regularly (averaging ~40% of the experiments), and at some facilities voltage-pulsed analyses comprise greater than 60% of the experiments [2]. Local electrode atom probe (LEAP�) systems have provided substantial improvements in ease of use, speed, field of view (FOV), mass resolving power, imaging quality and sample preparation flexibility over previous generation TAP and 3DAP instruments. However, the voltage pulse performance has been somewhat limited by the maximum pulse rate (200 kHz) and pulse amplitude (1300-1700 V) in the earlier generation LEAP 3000 and 4000 systems. The pulse amplitude limitation restricted the ability to use the LEAP to its maximum specimen voltage and FOV because the necessary voltage pulse fraction (15-25% depending up on the material and run conditions) to assure a consistent measurement of composition through the entire experiment [3] could only be maintained over a fraction of the available specimen voltage. Figure 1 demonstrates situations where the pulse voltage is insufficient to maintain compositional accuracy [1]. Figure 2 illustrates the extended range of constant pulse fraction for the new LEAP 5000 R/XR systems over the previous systems (now 15% pulse fraction up to 15kV). Although maximum data collection rates increased dramatically with the introduction of the first LEAP system in 2003, data acquisition times can still be significant (several hours or more) for complex specimens with large regions of interest. To improve specimen throughput, the maximum pulse rate has increased in the LEAP 5000 X/XR by 250% to 500 kHz. In addition, adaptive pulse frequency controls (available in both laser and voltage modes) have been developed to allow selection of a constant mass range during the entire experiment. In this mode, the pulse frequency is automatically increased as the specimen voltage increases and ion flight times decrease, which further improves throughput by as much as a factor of two. An increased pulse rate can also result in improved signal-to-noise ratio (SNR) due to a higher duty cycle of pulsing, thus more signal relative to background per unit time if the background noise is uncorrelated with signal peaks. Some samples have long spectral tails following the main signal peak and these tails can extend into the TOF window associated with the next voltage pulse thus increasing the background noise level. Figure 3 demonstrates that, even on a log scale, mass spectra can appear to have a flat SNR by simple examination, but upon closer inspection can have long, exponential components. Fortunately, the application of a high pulse fraction can assure that evaporation only occurs in association with the very sharp peak of the voltage pulse, suppressing correlated evaporation. Figure 4 demonstrates that the SNR rise with increased frequency can be nearly eliminated when using a large pulse fraction. The advanced voltage pulsing capabilities of the LEAP 5000 R/XR systems allow simultaneous improved throughput and improved detection sensitivity for low concentration species [4]. Microsc. Microanal. 21 (Suppl 3), 2015 42 References: [1] D.J. Larson et al., "Local Electrode Atom Probe Tomography" (Springer, New York 2014). [2] Unpublished customer survey, CAMECA Instruments Inc., (2015). [3] Miller, M. K. & G.D.W. Smith, Vac. Sci. and Tech. 19, (1981), p. 57. [4] R.M. Ulfig et al, Microscopy and Microanalysis 19 S2, (2013), p. 986. [5] The authors would like to thank Montanuniversit�t Leoben, University of Sydney, Oxford, Alabama, and the University of California Santa Barbara for sharing their pulse mode statistics. Figure 1. Schematic relationship between temperature and evaporation field for a multi component sample. At any base temperature, a sufficient magnitude pulse is required all species to have a similar probability of evaporation. Figure 2. The extended range of operation for a constant pulse fraction for the LEAP 5000 to a maximum pulse voltage of up to 2350V (40% increase). Figure 3. Although the background noise for a Sb-doped silicon sample appears to be flat in time, an expanded vertical axis demonstrates that the noise is correlated with the main peaks and may contribute to noise in the adjacent pulse window. Figure 4. Higher voltage pulse fraction (30%) results in a significantly reduced noise floor, even with increasing pulse frequency. This enables higher sensitivity to low concentration species, down to ppm levels.");sQ1[22]=new Array("../7337/0043.pdf","Targeting Grain Boundaries for Structural and Chemical Analysis Using","","43 doi:10.1017/S1431927615001014 Paper No. 0022 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Targeting Grain Boundaries for Structural and Chemical Analysis Using Correlative EBSD, TEM and APT Adam Stokes1,2, Mowafak Al-Jassim2, Dave Diercks1, Brian Egaas2, and Brian Gorman1 1. 2. Colorado School of Mines, Material Science, Golden, Colorado, U.S.A National Renewable Energy Laboratory, Measurements and Characterization, Golden, Colorado, U.S.A Polycrystalline Cu(In,Ga)Se2 (CIGS) thin-film solar cells have achieved a record efficiency of 21.7%, making them the most efficient thin-film photovoltaic device [1]. Despite the excellent efficiencies demonstrated by the technology, the overall picture of the composition related to structure at surfaces and grain boundaries (GB's) still remains ambiguous. Great efforts have been devoted to nanoscale chemical characterization of Cu(In,Ga)Se2 thin film solar cells using atom probe tomography [2] but little correlative work has been done highlighting structure and chemistry. This contribution will discuss a novel technique used to target GB's and relate their structure to chemistry at the nanoscale. We used this technique to select from roughly 20 GB's with known misorientations, extracted from electron backscattered diffraction (EBSD) micrographs, and then chemically analyzed them in 3-D using atom probe tomography (APT). This technique may also be used for many different types of materials. The Cu(In,Ga)Se2 layer was grown on a Mo-coated glass using a modified three-stage process at the National Renewable Energy Laboratory. The samples were prepared for EBSD, APT, and transmission electron microscopy (TEM) analysis using an FEI Helios 600i DualBeam focused ion beam / scanning electron microscope (FIB/SEM). A cross-section of Cu(In,Ga)Se2 was prepared using the FIB, lifted out using an Omniprobe 200 nanomanipulator, and placed on a TEM grid (see Figure 1 left). A face of the sample was cleaned using a 2 kV ion beam energy to reduce damage and smooth the surface for EBSD analysis. An EBSD map was created with dimensions ~2.5 um x 2.5 um (see Figure 1 middle) on that same face of the sample. From the EBSD inverse-pole-figure (IPF) image (See Figure 1 right), a regionof-interest (ROI) was identified for APT analysis with desired GB characteristics. Next, the volume around the ROI was carefully FIB-milled, leaving a needle shaped volume containing the ROI (see Figure 2). A Philips CM200 TEM was used to capture the specimen's dimensions before and after APT using similar technique as Ref. [3], which allowed for more accurate 3D reconstructions. This was a very important step used to match the GB's found in the TEM image to the GB's found in the IPF from EBSD (See Figure 3) A final 2 kV cleaning was used to reduce the damage to approximately the outer 2 nm of the sample. APT data were collected using a LEAP 4000X Si instrument manufactured by Cameca Instruments, Inc. using laser energy of 5 pJ, a base temperature of 40K, a detection rate of 0.5%, and a laser pulse rate of 250kHz. Laser energy and base temperature were previously optimized to get equal evaporation rates of the constituent elements for a chemistry profile of the device that well represents the known stoichiometry of CIGS. Results obtained from this new technique will also be presented (not shown) as it pertains to Cu(In,Ga)Se2. Impurity segregation and change in matrix concentrations will be correlated to GB's misorientation [4] . Microsc. Microanal. 21 (Suppl 3), 2015 44 Figure 1. Left: Section of sample (blue box) sitting on TEM grid used for APT. Middle: SEM image of sample. White dashed line indicates the CIGS cross-section; the region of interest. Right: IPF of same cross-section. GB's with desired characteristics are chosen for APT analysis. Figure 2. SEM micrographs of FIB procedure. a) Sample on TEM grid as in Figure 1. b) Top view: Desired grain boundary is beneath the red ring. c-h) Micrographs showing step-by-step FIB procedure of milling material away material except region-of-interest for APT analysis. Figure 3. Image quality map from EBSD scan. TEM bright field image of atom probe tip taken from white dotted region that match GB's in both the SEM micrograph to the TEM bright field image. References: [1] P. Jackson et al, Phys. Status solidi- Rapid Res. Lett. Vol. 9999 (2014), p. n/a-n/a [2] O. Cojocaru-mir et al, IEEE Journal of Photovoltaics, Vol. 1 (2011), p. 207-212 [3] K. Thompson et al, Ultramicroscopy, vol. 107 (2007) p. 131�139 [4 ] This work was supported by the U.S. Department of Energy under Contract No. DE-AC36-08- GO28308 with the National Renewable Energy Laboratory. The atom probe used in this research is supported by NSF Award Number 1040456.");sQ1[23]=new Array("../7337/0045.pdf","Development of Quantitative Standards for Atom Probe Reconstruction Parameters for Analysis of Interfacial Chemistry","","45 doi:10.1017/S1431927615001026 Paper No. 0023 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Development of Quantitative Standards for Atom Probe Reconstruction Parameters for Analysis of Interfacial Chemistry S.R. Broderick, S. Dumpala, S.E. Young, K. Kaluskar, S. Srinivasan, and K. Rajan Department of Materials Science and Engineering and Institute for Combinatorial Discovery, Iowa State University, 2220 Hoover Hall, Iowa State University, Ames, USA. The present work is aimed at developing a standard for defining reconstruction parameters for optimal voxel and chemical thresholds. We develop quantitative techniques for the detection of sharp chemical interfaces from the APT outputs of 3D point cloud and 3D chemical interfaces of continuous geometry. The difficulty lies in mapping discrete data to a continuous output while minimizing the loss in chemical information. In the present work, we develop a novel approach, based on the principles of computational homology, to map the discrete 3D point cloud atomic data to the topology of a continuous chemical interface. The computational homology framework developed can be applied to APT data to get meaningful results and quantify sensitivity of the output on the data reconstruction parameters. To incorporate evaporation physics with chemistry and structure and to provide physical definition of uncertainty in spatially defining phases, Identifying nanoscale chemical features from atom probe tomography (APT) data routinely involves adjustment of voxel size as an input parameter, through visual supervision, making the final outcome user dependent, reliant on heuristic knowledge and potentially prone to error. This work utilizes Kernel density estimators to select an optimal voxel size in an unsupervised manner to perform feature selection, in particular targeting resolution of interfacial features and chemistries. The capability of this approach is demonstrated through analysis of the / ' interface in a Ni-Al-Cr superalloy. Further, feature extraction from APT data is usually performed by repeatedly delineating iso-concentration surfaces of a chemical component of the sample material at different values of concentration threshold, until the user visually determines a satisfactory result in line with prior knowledge. However, this approach allows for important features, buried within the sample, to be visually obscured by the high density and volume (~ 107 atoms) of APT data. This work provides a data driven methodology, free of user defined boundaries, classifying different phases such as precipitates by mapping the topology of the APT dataset using a concept from algebraic topology termed persistent simplicial homology. The demonstration of interfacial analyses provides a clear quantitative logic for defining reconstruction parameters of voxel size and chemical thresholds, and therefore provides a standard for analyses of interfaces [1]. References: [1] The authors acknowledge the support from Air Force Office of Scientific Research grants: FA955010-1-0256, FA9550-11-1-0158 and FA9550-12-1-0456; and NSF grants: ARI Program CMMI-09-389018 and PHY CDI-09-41576. Microsc. Microanal. 21 (Suppl 3), 2015 46 Figure 1. Definition of optimal voxel sizes for standardized definition of Ni-Al-Cr interface. (Left) Dependence of the normalized mean integrated square error (MISE) on the voxel size. The error is minimum for a voxel edge length of 1.6nm. (b) 3D mapping of atom positions based on voxel definitions. The effect of voxel edge length on capturing the precise interface between the / ' region of the Ni-AlCr sample is captured through this analysis. Figure 2. Concentration profiles for Al and Cr and variations in the interfacial width for different voxel sizes. From our approach, we measure the interfacial width (4.4 nm) while removing the inherent bias in the process. This further emphasizes the benefit of the approach, as the resulting changes in the concentration profiles are subtle, as compared with the more obvious variations in the 3D spatial mapping. We therefore through this approach capture free of bias the chemistry of the interface as measured through the profile which reflects a macroscopic measure of the phase chemistry, while the atom mapping demonstrates segregation of Cr to the interface which otherwise would not be identified.");sQ1[24]=new Array("../7337/0047.pdf","Ongoing Developments of the Cryo-SEM/STEM Technique","","47 doi:10.1017/S1431927615001038 Paper No. 0024 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Ongoing Developments of the Cryo-SEM/STEM Technique Alan C. Robins1 1. E. A. Fischione Instruments Inc. Export, PA The scanning electron microscope (SEM), with its wide variety of different detectors for imaging or analysis, proves to be an extremely versatile tool for studying the structure of bulk specimens. With the addition of a Scanning Transmission Electron Microscope (STEM) detector, the instrument can also be used for imaging thin sections, aqueous suspensions applied to a TEM-style EM grid, or electron transparent specimens. The entire instrument can be further equipped, for imaging frozen-hydrated bulk samples, with the addition of an on-column or remote cryogenic preparation chamber, a cryo-transfer system, and a nitrogen cooled SEM stage. Low temperature or cryo-SEM techniques have been widely used and adapted as the resolution of the scanning electron microscope has improved [1]. With the introduction of field emission gun technology, a finely focused probe produces sub-nanometer resolution even at low accelerating voltages. Specimens in solution are either prepared in bulk carriers, such as high pressure freezing planchets, or on EM grids. Cryo-preservation is executed through the use of high-pressure or plunge freezing apparatus. The value of preparing a bulk sample allows for the structural analysis of samples over many tens of microns. A specimen is typically frozen and then transferred to the preparation chamber where it is then fractured, etched and coated prior to examination on a cold stage in the SEM. Cryo-SEM imaging also benefits from sputtered or evaporated ultra fine grain or amorphous coatings. This is applied to the sample surface to impart conductivity and enhance the visualization of fine structure [2]. In some cases a shadowed primary metallic coating is used which helps enhance contrast while imaging the specimen in backscatter mode. The advantage of the metal coating is that it also reduces beam damage to the specimen. With the advancement of SEM instrumentation, sample cryo-preservation methods, and metal coating techniques, two branches of cryo-SEM were established. The first includes systems in which all specimen preparation and coating processes are performed remotely. Subsequently, the specimen is then cryo-transferred to the imaging position in the microscope. During transfer the specimen has to be cooled and shielded to stop ice contamination from developing on top of the coating. With in-lens FESEM the remote approach is the only option where a typical cryo transfer holder, used for transmission electron microscopy (TEM), is interfaced to a remote preparation chamber and a goniometer on the microscope [3]. The specimen is cooled in the preparation chamber and microscope by the same liquid nitrogen dewar. Secondly, on-column cryogenic preparation systems have been continually improved to keep up with SEM developments. Care has to be taken to ensure no loss of instrument resolution occurs by incorporating a vibration damped turbo pump. The preparation chamber is attached directly on the column were the sample is fractured, coated and then imaged on a cold stage in the SEM. The SEM stage is cooled to low temperatures, typically -170�C, either by continuous flow of cold nitrogen gas or conduction cooling from a dewar attached to the column. The different types of preparation techniques will be reviewed with associated application images from biology, food, pharmaceuticals and healthcare. Microsc. Microanal. 21 (Suppl 3), 2015 48 References: [1] P Echlin, Low-Temperature Microscopy and Analysis (Plenum Press, London) 1992, p. 349-411 [2] W Wergin, EF Erbe, and AC Robins, Scanning 15(Supplement III) (1993), p. 71. [3] RP Apkarian, KL Caran and KA Robinson, Microsc. Microanal. 5 (1999), p. 197-207 [4] ER Wright and VP Conticello, Advanced Drug Delivery Reviews 54 (2002), p. 1057-1073. Figure 1. Dilute solution of an elastin-mimetic block copolymer [4], at high magnification, imaged with Cryo-HRSEM (left) and Cryo-HRSTEM (right).");sQ1[25]=new Array("../7337/0049.pdf","Imaging Macromolecules and Viruses in a Hydrated State Using a Field Emission Scanning Electron Microscope (FESEM)","","49 doi:10.1017/S143192761500104X Paper No. 0025 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Imaging Macromolecules and Viruses in a Hydrated State Using a Field Emission Scanning Electron Microscope (FESEM) C.A. Ackerley1, C Nielsen 2 and G. Rodek3 Division of Pathology, Hospital for Sick Children, Toronto, Canada JEOL USA, Peabody, MA, USA 3 SPI Supplies, West Chester, PA, USA 2 1 A capsule capable of retaining liquid at one atmosphere with an electron lucent partition membrane was developed for the observation of cells and other materials in a hydrated state in the scanning electron microscope (SEM) was developed in the early 2000s [1]. Cells can be grown on the internal face of the electron lucent partition membrane and processed using a number of protocols including conventional electron microscopy fixation and staining procedures, cytochemistry and immunogold immunocytochemistry. The samples are then immersed in a media which dissipates electrons. The capsule is sealed and the cells imaged using backscatter electron (BE) imaging. More recently, we have significantly modified the capsule and have replaced the polymer partition window with a 20nm thick film of silicon oxide. Using a FESEM and a phosphor coated BE detector we have been able to obtain time lapse serial images of immunogold labelled live cells for up to 500 1.8 second scans at approximately 30 second intervals. Cells demonstrated at least 80% viability using an absorbed current of 10pA[2]. In a previous experiment, we had evaluated the sensitivity of three BE electron detectors and an in lens Wien based filter using NIH Image J and a SMART macro [3]. Although the BE detector with the phosphor scintillator was the most electron dose sensitive, resolution was impaired by its position in the column. Specimens were limited to a working distance of 8mm. Using the in lens Wien filter,the sample can be used at very short working increasing the resolution of the images. An added bonus to using this imaging mode is that the majority of the atomic contrast is generated from the surface of the object and electrons originating from within the sample contribute very little to the final product. The first hurdle to overcome in imaging hydrated small organic objects was developing a protocol for adhering the specimens to the inner face of the partition window. A capsule was redesigned where several �l of the solution containing either the macromolecules or the viruses was administered to the inner surface of the partition membrane. The vessel which contains the partition membrane was threaded and contained a relief hole to minimize pressure to the fragile partition membrane. The backing plate was then screwed into the vessel until no further fluid was coming out of the relief hole. The relief hole was then sealed with a small drop of cyanoacrylate glue. This results in a very thin film of solution containing the object to be studied in intimate contact with the partition membrane. We are currently using either 0.01% uranyl acetate, 0.02% ammonium molybdate or no contrasting additive in the specimen suspension. We have limited our studies to adenovirus, microtubules and glycogen. All three of the specimen types are easily imaged at near TEM resolutions when the specimen is suspended in solution with a contrasting additive (fig.1,2). Microsc. Microanal. 21 (Suppl 3), 2015 50 Unstained samples were very difficult to see however with some further image processing the structures were quite detectable. Chamber design is now being revisited. The most successful modification to date has been the inclusion of another silicon oxide membrane into the blanking plate eliminating the majority of the backscatter electrons contributed from the more robust standard backing plate. Using this design the success rate of eliminating the excess specimen fluid is quite difficult as either the top or bottom membranes are easily broken during the removal of the excess specimen fluid. When we have been successful at this, unstained specimen visibility increases several fold. Like our previous studies with live cells it is hoped that using this technique we will be able to serial image molecular interactions in these chambers. References: [1] S. Thiberge et al., Proc Natl Acad Sci USA 10(10) (2004) 3346. [2] C.A. Ackerley et al., Microsc and Microanal12(2) (2006) 428CD. [3] D.C. Joy, J Microsc (1) (2004) 24. [4] The authors acknowledge funding from the Hammond Foundation. Fig.1 In lens Wien filtered image of hydrated Fig.2 In lens Wien filtered image of hydrated adenovirus in a solution containing 0.01% adenovirus in a solution containing 0.02% uranyl acetate. Contrast has been reversed. ammonium molybdate. Contrast has been reversed.");sQ1[26]=new Array("../7337/0051.pdf","Field Emission In-Lens Scanning Electron Microscopy of the Nuclear Envelope and Other Cellular Structures.","","51 doi:10.1017/S1431927615001051 Paper No. 0026 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Field Emission In-Lens Scanning Electron Microscopy of the Nuclear Envelope and Other Cellular Structures. Martin W. Goldberg1, Jindiska Fiserov�2, Terence D. Allen3 School of Biological and Biomedical Sciences, Durham University, Science Laboratories, South Road Durham, DH1 3LE. UK 2. Institute of Molecular Genetics of the ASCR, v. v. i., Videnska 1083, 142 20 Prague 4, Czech Republic 3. The Institute of Cancer Studies, University of Manchester, UK Field emission in-lens scanning electron microscopes (FEISEMs) can achieve sub-nanometer resolution. This is the level of biological macromolecules and macromolecular complexes. Such instruments can therefore be used to study the surface structure of sub-cellular components, showing details of the organization of proteins and other macromolecules within these structures. Biological material, however, has a high water content and needs to be dried or frozen to be compatible with the vacuum within the microscope. It is electrically non conductive and therefore needs to be infiltrated, or coated, with metal to avoid problems with charge build up, causing imaging artifacts and instability. Most surfaces are enclosed within membrane bound compartments and are therefore inaccessible to surface imaging techniques such as FEISEM. Many of the membrane surfaces are fragile and can be damaged or altered during fixation drying and exposure to the intense beam of electrons in the microscope. Therefore to get useful information by FEISEM, care and consideration has to be given to the steps of sample preparation, and imaging parameters (e.g. current and voltage). Cryo methods have proved useful to tackle several of these problems. Cryo-fracturing/planing/sectioning is useful to expose surfaces of interest, whereas the use of a cryo-stage to image frozen hydrated samples within the microscope can prevent problems with fixation and drying artifacts. Work at room temperature on chromosome structure and banding used conductive staining involving thiocarbohydrazide and Osmium tetroxide [1], giving good structural detail without metal coating. However this results in a high level of backscatter electrons and hence SEIIs, creating noise. This proved unsuitable for imaging the nuclear membranes and nuclear pore complex (NPC) at the periphery of the nucleus overlying dense chromatin. Protocols [2] using tannic acid to stabilize filamentous components of the NPC (Fig. 1), such as the NPC basket (fishtrap), were an important development. However, development of high resolution metal coats [3], to provide signal solely from the surface, was pivotal to enable us to image fine structural detail of the NPC [4] and nuclear lamina [5]. Suitable coats can be obtained by sputter coating several refractory metals. However chromium is particularly useful because of its relatively low atomic number compared to gold. Gold nano-particles can be used to tag antibodies. A backscatter detector can be used to distinguish the gold particle from chromium coat, unequivocally identifying the epitope position. To image a structure such as the inner nuclear membrane requires the surface to be exposed. This can be achieved by isolating Xenopus oocyte nuclear envelopes. However, the inner nuclear membrane of most cells is tightly associated with the chromatin and other intra-nuclear structures. Therefore we developed simple fracturing methods [6] to remove the overlying material, exposing the inner membrane and allowing us, for instance, to show that plants possess a lamina-like structure and NPC baskets [7]. Alternatively samples can be frozen and fractured or sectioned. Such samples can be imaged directly using a cryo-SEM stage. Generally this 1. Microsc. Microanal. 21 (Suppl 3), 2015 52 requires samples to be freeze-etched, then metal coated at low temperature. Although routine in a dedicated cryo-SEM, this is challenging in an in-lens SEM where the coating and etching has to be done in a separate coating unit, necessitating transfer between this and SEM. A more straightforward alternative is to thaw the fractured/cryo-sectioned samples into fix, and dry, coat and image at room temperature. FEISEM can also be used to study the organization of cytoskeletal filaments, and clathrin coated pits and vesicles after detergent extraction and careful fixation (Fig.2). [8]. Fig. 1. FEISEM image of the cytoplasmic face of Xenopus oocyte nuclear envelope tilted 30 degrees. Fig. 2. Periphery of HaCaT cell after extraction with 0.5% Triton X100 showing actin filaments and clathrin coated vesicle. References [1] CJ Harrison et al, J Cell Sci 56 (1982), p409. [2] H Ris, Electron Microsc Soc Am Bull 21 (1991), p54 [3] RP Apkarian, Scanning Microsc 8 (1994)289. [4] MW Goldberg and TD Allen, J Mol Bio. 257 (1996), p848. [5] MW Goldberg et al, J Cell Sci 121 (2008), p215. [6] TD Allen et al, Nature Protocols 2 (2007), p1180. [7] J Fiserova, E Kiseleva and MW Goldberg, Plant J 59 (2009), p243. [8] Thanks to S Rutherford, G Bennion, C Richardson and H Grindley for technical assistance.");sQ1[27]=new Array("../7337/0053.pdf","Heterogeneity in the pre-40S ribosomal subunit reveals two distinct regions of","","53 doi:10.1017/S1431927615001063 Paper No. 0027 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Heterogeneity in the pre-40S ribosomal subunit reveals two distinct regions of variation at functionally important structural elements. Matthew C. Johnson1, Homa Ghalei2, Katrin Karbstein2, and M. Elizabeth Stroupe1 1. Department of Biological Science and Institute of Molecular Biophysics, Florida State University, Tallahassee, FL, USA. 2. Department of Cancer Biology, The Scripps Research Institute, Jupiter, FL, USA. Ribosomes represent a significant portion of the dry mass of all cells and actively dividing cells assemble about 2,000 ribosomes every minute [1]. This represents a massive investment by the cell in a complex RNA:protein assembly that will go on to translate mRNAs into protein, an essential process for all of life. Much is known about how a ribosome faithfully performs protein synthesis; less is understood about how they are faithfully assembled. Further, several high profile diseases like Diamond Blackfan Anemia (DBA), 5q- syndrome, and isolated congenital asplenia derive from defects in the erythropoetic lineage caused by haploinsufficiency of some ribosomal proteins (r-proteins) [2]. Ribosomes are composed of the small and large subunit, which are assembled independently in the nucleolus and nucleus before being joined after export to the cytoplasm for initiation of mRNA translation into protein (Figure 1). In eukaryotes, over 200 assembly factors (AFs) direct this maturation, chaperoning the structural transitions to ensure quality control as the subunits move through from the nucleolus to the cytoplasm [3]. The small subunit, 40S in yeast, of the ribosome directs mRNA decoding. In particular, the beak, (a structural element of the head), and the platform form the entrance and exit of the mRNA binding channel (Figure 2). As such, correct assembly of these elements is of particular importance and seven AFs have been associated specifically with beak and are critical for blocking premature translation initiation by the pre-40S ribosome [4]. Three AFs at the beak, Enp1/Ltv1/Rps3, have been shown to be critical in holding the beak in a position that sterically blocks subunit assembly [5]. Ltv1 release facilitates Rps3 repositioning that ultimately allows beak repositioning and subunit joining. Movement of the beak is related to nuclease events at the platform that finalize 18S rRNA maturation. How this happens remains undefined. We seek to use 3DEM to dissect the relationship between the beak and the platform. Recent advances in 3DEM imaging using a direct electron detector and computational image processing using 3D multivariate statistical analysis [6] have facilitated increasingly high-resolution (<4 �) structure determination of well-ordered complexes. The pre-40S ribosome is a heterogeneous specimen, which inherently limits the achievable resolution; nonetheless these same technical advances also allow precise analysis of structural heterogeneity. We have imaged pre-40 S ribosomes on a direct detector using a Titan Krios (FEI, Hillsboro, OR) microscope (Figure 3). Our analysis of the resulting 3D pre40S ribosome structures points to heterogeneity at the beak and the platform (Figure 3, inset), suggesting a structural and functional link between these two regions. Here, we present a method for localized 3D classification of heterogeneous specimens that vary at more than one spatially-distinct position on the pre-40S ribosome. Microsc. Microanal. 21 (Suppl 3), 2015 54 References: [1] J. R. Warner. Trends in Biochemical Sciences (1999) p. 437. [2] A. Bolze, et al. Science (2013 ) p. 976-978. N. Burwick, et al. Seminars in Hematology (2011) p. 136-143. [3] K. Karbstein. Trends in Cell Biology (2013) p.242. [4] B. Strunk, et al. Cell (2012) p. 111. [5] B. Strunk, et al. Science (2011) p. 1449. [6] M. M. Yusupov, et al. Science (2001) p. 883. [7] D. Lyumkis et al. JSB (2013) p. 377. S. H. W. Scheres JSB (2012) p. 519. [8] The authors acknowledge funding from the National Institutes of Health (R01-GM086451) and the National Science Foundation (MCB1149763). Figure 1. Model of the mature 70S ribosome [6]. Figure 2. Low-resolution model of the pre-40S ribosome with Enp1/Ltv1/Rps3 and showing the solvent exposed face [5]. Figure 3. Representative micrograph showing the high quality of the pre-40S ribosomes on FSU's DE20 detector (Direct Electron, San Diego, CA) and lower resolution of the map due to heterogeneity at the beak on the solvent exposed face (arrow).");sQ1[28]=new Array("../7337/0055.pdf","Determining a Sub-nanometer Resolution Structure of a Helicobacter pylori VacA","","55 doi:10.1017/S1431927615001075 Paper No. 0028 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Determining a Sub-nanometer Resolution Structure of a Helicobacter pylori VacA Toxin Oligomer Tasia M. Pyburn1, Nora J. Foegeding1, Timothy L. Cover2,3,4, and Melanie D. Ohi1 1 Department of Cell and Developmental Biology, Vanderbilt University School of Medicine, Nashville, Tennessee, USA 2 Department of Medicine, Vanderbilt University School of Medicine, Nashville, TN USA 3 Department of Pathology, Microbiology and Immunology, Vanderbilt University School of Medicine, Nashville, TN USA 4 Veterans Affairs Medical Center, Nashville, Tennessee, USA Helicobacter pylori is a gram negative bacterium that colonizes the human stomach. The presence of this bacterium can lead to the development of peptic ulcers, and in more severe cases, the development of gastric adenocarcinoma or lymphoma. One of the major virulence factors secreted by H. pylori is vacuolating cytotoxin A (VacA). VacA is secreted as an 88 kDa soluble monomer but transitions into a membrane protein when in contact with cells. VacA s1/i1/m1 causes the formation of large intracellular vacuoles and has been shown to induce various cellular effects, including T cell inhibition, mitochondrial dysfunction, and cell death. VacA is comprised of a 33kDa (p33) domain located at the N-terminus required for oligomerization and internalization and a C-terminal 55kDa (p55) domain required for receptor binding and oligomerization. Currently, the only high resolution structural information is a crystal structure of the p55 domain determined at 3.4 � resolution which showed that the p55 domain is comprised of three beta helices [1]. Negative stain electron microscopy (EM) analysis revealed that at a neutral pH, s1/i1/m1 VacA forms multiple oligomers including single-layer hexamers and heptamers, as well as double layer dodecamers and tetradecamers [2]. Unfortunately, these structures are at a ~15� resolution, too low to distinguish any secondary structural features. Also, no discernible information was obtained on the structure of the p33 pore-forming domain. However, the structures were able to provide a model for VacA oligomerization and the location of the p55 and p33 domains. Since s1/i1/m1 VacA forms multiple types of oligomers, it makes structural analyses difficult. In order to pursue sub-nanometer structural determination, it is necessary to obtain a more homogeneous sample. The laboratory of Dr. Timothy Cover provided a rung deletion mutant, 511-536, which importantly, retains its ability to vacuolate cells [3]. Figure 1A shows the region of the p55 domain deleted in this mutant. To determine whether the homogeneity of double-layer VacA oligomers could be improved by removing regions of the p55 domain involved in double-layer formation, VacA 511-536 was subjected to negative stain analysis. Single-particle electron microscopy was performed and determined that VacA 511-536 oligomerizes into one major dodecameric conformation (Fig. 1B and C, ~66%), with the remaining oligomers being hexamers, heptamers, or one conformation of tetradecamer (Fig. 1C). Using this mutant that forms more homogenous oligomers, we have found vitrified ice conditions that produce high contrast images (Fig. 1D). Over 600 images were taken using a FEI Tecnai F30 at 200kV using low-dose conditions and defocus values ranging from -1 to -5 �m on a Gatan 4Kx4K Ultrascan CCD. Class averages generated in SPIDER revealed that the particles were randomly oriented within the ice (Fig. 1E). Microsc. Microanal. 21 (Suppl 3), 2015 56 VacA forms anion-selective channels in planar lipid bilayers and in membranes of cells, including the plasma membrane and most likely also in endosomal and mitochondrial membranes [4-6]. Single channel analysis predicts VacA channels are composed of six subunits and mimic the action of endogenous sodium-chloride channels [5,6]. The p33 domain is postulated to form the anion-selective channel; however, atomic force microscopy (AFM) studies of VacA bound to mica-supported lipidbilayers were not able to definitively determine either the type(s) of oligomers bound to the lipid or the conformation of the central pore-forming region of VacA in this lipid environment [4-14]. Using single particle EM and cryo-electron tomography we have characterized the oligomerization state of VacA bound to lipid. References: [1] Gangwer, K. A., et al. Proc Natl Acad Sci 104, (2007), p. 16293-16298. [2] Chambers, M. G., et al. Journal of molecular biology 425, (2013), p. 524-535. [3] Ivie, S. E., et al. BMC microbiology 10, (2010), p. 60. [4] Cover, T. L. & Blanke, S. R. Nat Rev Microbiol, 3, (2005), p. 320-32. [5] Boquet, P. & Ricci, V. Trends Microbiol 20, (2012), p. 165-74. [6] Palframan, S. L., et al., Front Cell Infect Microbiol 2, (2012), p. 92. [7] Iwamoto, H., et al. FEBS Lett 450, (1999), p. 101-4. [8] Domanska, G., et al. PLoS Pathog 6, (2010) e1000878. [9] McClain, M. S., et al. J Biol Chem 278, (2003), p. 12101-8. [10] Torres, V. J., et al. J Biol Chem 280, (2005), p. 21107-14. [11] El-Bez, C., et al.. J Struct Biol 151, (2005), p. 215-28. [12] Czajkowsky, D. M., et al. Proc Natl Acad Sci U S A 96, (1999), p. 2001-6. [13] Adrian, M., et al. J Mol Biol, 318, (2002), p. 121-33 [14] Geisse, N. A., et al. Biochem J 381, (2004), p. 911-7. Figure 1. Analysis of VacA 511-536 reveals a shift in oligomeric type. A) 2.4 � crystal structure of s1/i1/m1 VacA p55. The rung deletion is highlighted yellow. scale bar=2.5nm B) Mutant forms one major dodecamer conformation. top-negative stain average of oligomer, scale bar=42nm, bottom-3D reconstruction of oligomer. `*' represents dodecamer type seen in panel B. C) Table showing percentages of each type of oligomer in both s1/i1/m1 and 511-536 VacA. Other=other dodecamer types not including type seen in panel B. D) Representative vitrified ice image of of VacA 511-536 collected on an FEI F30 electron microscope at 200kV. E) 80 class averages generated from 6478 particles. dash box=tetradecamer; solid box=dodecamer (different views); small dash box=hexamer. Side length of panels=36nm.");sQ1[29]=new Array("../7337/0057.pdf","Sar1 Forms Ordered Arrays That Can Facilitate Vesicle Scission in the COPII Pathway.","","57 doi:10.1017/S1431927615001087 Paper No. 0029 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Sar1 Forms Ordered Arrays That Can Facilitate Vesicle Scission in the COPII Pathway. Hanaa Hariri1, Alex Noble2, Arthur Omran3, and Scott M. Stagg1,2 1. 2. 3. Institute of Molecular Biophysics, Florida State University, Tallahassee, FL, USA Department of Physics, Florida State University, Tallahassee, FL, USA Department of Chemistry and Biochemistry, Florida State University, Tallahassee, FL, USA Secretion is a fundamental process in eukaryotes. It's disruption can lead to one of several diseases including chylomicron retention disease1, cranio-lenticulo-sutural dysplasia2, or congenital dyserythropoietic anemia3. Transport of secreted cargo such as membrane proteins or soluble secreted cargos is initiated at the endoplasmic reticulum (ER). There proteins are co-translationally translocated into the lumen of the ER. These cargo proteins are then concentrated at specific sites on the ER called exit sites, and are collected into a membrane-bound vesicle that is transported to the Golgi apparatus and on to the cell surface. The coat protein complex II (COPII) proteins Sar1, Sec23/24, and Sec13/31 facilitate the formation of transport vesicles at the ER exit sites4. Vesicle formation is completed when the vesicle detaches from the parent membrane in a process called scission. It has been shown that Sar1 is required for scission5 though the mechanism by which the bud neck is collapsed and fused remains unclear. We have shown that Sar1 can oligomerize on membranes and that it can vesiculate giant unilamellar vesicles (GUVs) in the absence of the other COPII proteins. It is clear that there is not enough Sar1 expressed in cells for it to facilitate vesiculation on its own; instead it works in concert with the other COPII proteins to form vesicles in the right place at the right time. We have hypothesized that the oligomerization of Sar1 plays two roles in COPII mediated vesiculation: 1) to initiate membrane curvature, and 2) to help facilitate scission. Here we show that Sar1 can transform membranes into a variety of stuctures including sheets, multibudded vesicles that are connected by narrow necks, and coated tubules with diameters ranging from 45 nm too 200 nm (Fig. 3). A low-resolution structure of the Sar1 lattice was determined by limiting the diameter of the tubules using galactosyl ceramide lipids and using iterative helical reconstruction. This revealed that the Sar1 lattice consists of arrays of dimers, but the resolution was too low to be able to determine the residues that mediate the intermolecular contacts in the lattice or to resolve how the N-terminal amphipathic is buried in the outer membrane leaflet to anchor Sar1 to the membrane. We are currently characterizing the mechanisms by which Sar1 deforms membrane by determining what residues mediate the formation of the Sar1 lattice and by determining how Sar1 inserts into the bilayer. We show that Sar1 forms ordered arrays on membranes with low curvature, and with more highly curved membranes, it only forms locally ordered arrays (Fig. 1), and we propose a mechanism by which Sar1 facilitates vesicle scission at COPII bud necks by polymerization of Sar1 dimers. Microsc. Microanal. 21 (Suppl 3), 2015 58 References: 1. Shoulders C. C., Stephens D. J. & Jones B. (2004). The intracellular transport of chylomicrons requires the small GTPase, Sar1b. Curr Opin Lipidol 15, 191-197. 2. Boyadjiev S. A., Fromme J. C., Ben J., Chong S. S., Nauta C., Hur D. J., Zhang G., Hamamoto S., Schekman R., Ravazzola M., Orci L. & Eyaid W. (2006). Cranio-lenticulo-sutural dysplasia is caused by a SEC23A mutation leading to abnormal endoplasmic-reticulum-to-Golgi trafficking. Nat Genet 38, 1192-1197. 3. Schwarz K., Iolascon A., Verissimo F., Trede N. S., Horsley W., Chen W., Paw B. H., Hopfner K. P., Holzmann K., Russo R., Esposito M. R., Spano D., De Falco L., Heinrich K., Joggerst B., Rojewski M. T., Perrotta S., Denecke J., Pannicke U., Delaunay J., Pepperkok R. & Heimpel H. (2009). Mutations affecting the secretory COPII coat component SEC23B cause congenital dyserythropoietic anemia type II. Nat Genet 41, 936-940. 4. D'Arcangelo J. G., Stahmer K. R. & Miller E. A. (2013). Vesicle-mediated export from the ER: COPII coat function and regulation. Biochim Biophys Acta 1833, 2464-2472. 5. Lee M. C., Orci L., Hamamoto S., Futai E., Ravazzola M. & Schekman R. (2005). Sar1p N-terminal helix initiates membrane curvature and completes the fission of a COPII vesicle. Cell 122, 605-617. Figure 1. Structure of the lattice of Sar1 dimers. Sar1 self assembles into a 2D lattice on membranes with low curvature. Class averages (yellow) indicated that the basic assembly unit of the Sar1 lattice is a dimer.");sQ1[30]=new Array("../7337/0059.pdf","Regulation of Cytoplasmic Dynein by Lis1","","59 doi:10.1017/S1431927615001099 Paper No. 0030 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Regulation of Cytoplasmic Dynein by Lis1 Katerina Toropova1*, Sirui Zou2*, Anthony J. Roberts2,3, Samara L. Reck-Peterson2# and Andres E. Leschziner1# 1 2 Department of Molecular and Cellular Biology, Harvard University, Cambridge, MA, 02138, USA Department of Cell Biology, Harvard Medical School, Boston, MA, 02115 3 Astbury Centre for Structural Molecular Biology, School of Molecular and Cellular Biology, Faculty of Biological Sciences, University of Leeds, Leeds, LS2 9JT, UK. *These authors contributed equally to this work; #Co-corresponding authors Regulation of cytoplasmic dynein's motor activity is essential for diverse eukaryotic functions, including cell division, intracellular transport and brain development. The dynein regulator Lis1 is known to keep dynein bound to microtubules; however, how this is accomplished mechanistically remains unknown. We used three-dimensional (3D) electron microscopy (EM) and single-molecule imaging to help establish this mechanism. The 3D structure of the dynein-Lis1 complex shows that binding of Lis1 to dynein's AAA+ ring sterically prevents dynein's main mechanical element, the "linker", from completing its normal conformational cycle. Single-molecule experiments show that eliminating this block by shortening the linker to a point where it can physically bypass Lis1 renders single dynein motors insensitive to regulation by Lis1. Our data reveal that Lis1 keeps dynein in a persistent microtubule-bound state by directly blocking the progression of its mechanochemical cycle. Cytoplasmic dynein 1 ("dynein" here) is a microtubule (MT)-based motor that moves towards the minus ends of MTs (generally towards the cell center). Dynein is a member of the AAA+ (ATPases Associated with various cellular Activities) superfamily [1] and its motor contains a ring of six concatenated AAA+ domains (termed AAA1-6). The dynein regulator Lis1 was first described as the gene mutated in patients with type-1 lissencephaly, a neurodevelopmental disease characterized by a smooth cerebral surface, cognitive defects, and seizures [2, 3]. Biochemical experiments suggest that Lis1 increases dynein's affinity for MTs [4-6]. Here, we set out to determine how Lis1 does this. To determine the mechanism of dynein regulation by Lis1 we used both cryo-negative stain EM to determine structures of dynein-Lis1 complexes and single-molecule fluorescence microscopy to test hypotheses derived from our structural studies. Our 3D EM structure showed that Lis1 binds at AAA4 in dynein and repositions dynein's main mechanical element, the linker domain (Figure 1) [7]. We hypothesized that Lis1 binding to dynein blocks the linker from reaching a key docking site at AAA5, an interaction that is required to release dynein from MTs. To test this idea, we made a mutation in dynein that we predicted would both allow Lis1 to bind dynein, as well as allow the linker to reach its AAA5 docking site. We confirmed that this was the case be determining EM structures of the dynein mutant bound to Lis1. Next we performed single-molecule experiments with this "short linker" mutant and found that the mutant was now insensitive to Lis1 binding. The short linker dynein no longer remained tightly bound to MTs in the presence of Lis1 and ATP (Figure 2) [7]. Together our results suggest that Lis1 keeps dynein in a persistent MT-bound state by directly blocking the progression of its mechanochemical cycle [7]. Microsc. Microanal. 21 (Suppl 3), 2015 60 Figure 1. Structure of the dynein-Lis1 complex. (A) Structural model of dynein's motor domain (PDB ID: 4AKG) docked into the EM map of dynein-Lis1 (resolution 21 �). The linker (grey arrow) is shifted away from its position in the crystal structure (purple arrow). (B) Structural model of dynein's motor domain docked into the EM map of dynein alone (resolution 15 �). Green circle: location of linker interaction with AAA5. Figure 2. Short linker dynein is insensitive to Lis1. Diagram of the single-molecule release assay (left). TMR-labelled (red star) dynein molecules release from MTs in the presence of ATP. Lis1 causes dynein to remain bound to MTs (middle), while the short linker dynein is insensitive to Lis1 (right). Scale bar = 5 seconds. References [1] Neuwald, A.F., et al., Genome Res, 1999. 9(1): p. 27-43. [2] Reiner, O., et al., Nature, 1993. 364(6439): p. 717-21. [3] Wynshaw-Boris, A., Clin Genet, 2007. 72(4): p. 296-304. [4] Huang, J., et al., Cell, 2012. 150(5): p. 975-86. [5] McKenney, R.J., et al., Cell, 2010. 141(2): p. 304-14. [6] Yamada, M., et al., EMBO J, 2008. 27(19): p. 2471-83. [7] Toropova, K., et al., Elife, 2014. 3.");sQ1[31]=new Array("../7337/0061.pdf","Microscopy, Microbiology and Regulations of Paper and Paperboard Utilized in Pharmaceutical Packaging","","61 doi:10.1017/S1431927615001105 Paper No. 0031 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microscopy, Microbiology and Regulations of Paper and Paperboard Utilized in Pharmaceutical Packaging Janet H. Woodward Strategic Marketing Group, Buckman, Memphis, USA The categories of pharmaceutical packaging are primary, secondary and tertiary containers. Primary packaging is in direct contact with the product. It provides protection from environment hazards (e.g., light, moisture, microbial contamination) during storage and handling without interacting with the drug (e.g., imparting odor or taste). For solid dosage forms, primary containers are glass, plastic, and blister packs. Secondary containers enclose one or more of the primary containers; they have no direct contact with the product. They serve several purposes, including additional protection during storage and handling, carry required labelling, and marketing purposes. Tertiary containers are for bulk storage, handling and shipping. Paper and board are commonly used for secondary and tertiary containers. They are also used in the lidding material of some "peel off-push through" blister packs. The type of furnish and chemicals used in the papermaking process for each type of container is dependent upon characteristics needed in the final packaging container. One widely used paper/board for secondary containers is made of virgin fibers that have been pulped via the kraft (sulfate) process and bleached by various chemicals (e.g., chlorine dioxide, hydrogen peroxide). Another popular board for secondary containers is made of kraft-processed unbleached virgin fiber. As interest in environmental sustainability has increased through the years, the use of 100% recycled fiber has also increased. The bleaching process removes color, making the final paper/board "white" in color while unbleached virgin paper/board is brown and recycled board is typically gray (Fig. 1). For pharmaceutical packaging, all of these are coated on one side with a high quality kaolinbased coating color to enhance printability. The boards are designed to have strength (rigidity), wear resistance, low moisture (water) adsorption, and be easily converted (e.g., fold, seal). Paper used for lidding material is made from bleached virgin fibers. The lid is a laminate of paper, polyester, and foil. The paper is clay-coated to enhance printability. One important characteristic of this paper is to have the appropriate elasticity (tensile) requirements for the conversion process. Common tertiary containers are corrugated boxes made from unbleached virgin, unbleached recycled fibers, or a combination of both. Strength properties of these boxes come from the design of a fluted sheet (corrugating medium) sandwiched between two flat sheets (liner). Corrugated boxes must be able to withstand stacking and various environments (e.g., high humidity) during storage and shipping. No matter the type of furnish, the paper or board-making process is not aseptic. Of the different furnishes, virgin (bleached or unbleached) typically has low microbial contamination rates as compared with recycled. However, process factors, such as long-term storage without sufficient agitation, can lead to anaerobic conditions that allow anaerobic bacteria to flourish. If left unchecked, some anaerobes produce odors, such as volatile fatty acids. These odors recirculate throughout the process. They are also fiber-substantive, imparting odors to the final paper/board. Improperly treated influent water adds to the overall contamination rate of the process. Organisms, such as fungi (Fig. 2a), filamentous bacteria Microsc. Microanal. 21 (Suppl 3), 2015 62 (Fig. 2b) and encapsulated bacteria (Fig. 2c) can readily form deposits, especially on the machines. This leads to holes and other defects as well as breaks (e.g., production downtime). Biocides are used throughout the process to minimize microbial growth. In the Unites States, all biocides are considered pesticides and thus are regulated under the Federal Insecticide, Fungicide, and Rodenticide Act and registered with the Environmental Protection Agency. Most biocidal actives are also "FDA-approved". This general statement refers to specific Code of Federal Regulations (CFR), 21 CFR � 176 (2014): Indirect Food Additives: Paper and Paperboard Components. The strictest allowance is 21 CFR � 176.170, components of paper and paperboard in contact with aqueous and fatty foods; 21 CFR � 176.180 is for components of paper and paperboard in contact with dry food. Section 21 CFR � 176.300 relates to actives used in the wet-end of the process. Although many mills do not produce foodcontact paper or board, they still require all of their chemicals to have FDA-approval. There is no specific FDA regulation regarding the use of biocides for paper or board made for pharmaceutical packaging, thus the producers follow the 21 CFR � 176 guidelines. Paper will continue to play an important role in the future of pharmaceutical packaging. In addition to being an integral component of various containers, uncoated and coated paper can offer solutions for tamper evident and security labels. Figure 1. Paper/board made from various furnishes: (a) bleached virgin fiber; (b) unbleached virgin fiber; and (c) recycled fiber. Figure 2. Problem-causing organisms in the papermaking process: (a) fungi - lactofuchsin stain; (b) filamentous bacteria � lactofuchsin stain; and (c) encapsulated bacteria � crystal violet stain.");sQ1[32]=new Array("../7337/0063.pdf","Microscopic Techniques for Sterility Assurance Support in the Medical Products Industry","","63 doi:10.1017/S1431927615001117 Paper No. 0032 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microscopic Techniques for Sterility Assurance Support in the Medical Products Industry Jason R. Mantei1, Mary Ann Murphy1, Laurie Stojanovic1, Laura Wahlen2, Mark Pasmore2, James P. DiOrio1. 1. 2. Baxter Healthcare Corporation, Particles and Interfaces � Microscopy, Round Lake, IL, USA. Baxter Healthcare Corporation, Sterility Assurance, Round Lake, IL, USA. Assurance of sterility in intravenous solutions, pharmaceutical drug products, and medical devices is of paramount importance for ensuring patient safety. Microbiologists use several analytical techniques to design, validate, and assure sterility of finished products. As medical devices and pharmaceuticals grow more complex, so have the challenges related to the microbial control analyses, including complex part geometries and variable surface properties. In these cases, direct visualization of a microorganism's interaction with a material, the progression of biofilm formation and growth, or the organization of inoculated spores with microscopic techniques is invaluable. Recent advances in highly sensitive camera technology have resulted in the creation of systems with large, light-tight chambers that can measure extremely low levels of luminescence or fluorescence emission. These systems, typically equipped with x-ray imaging and often used for live animal studies, are also extremely valuable for discerning the initial locations of microorganism attachment to surfaces and devices (Fig. 1). In some cases, a localization of mere tens of organisms can be detected, leading to a very accurate temporal assignment of biofilm colonization and growth events. A wide range of fluorescent stains specifically designed for bacteria have been developed and can be exploited in fluorescence and confocal microscopy. These stains allow for the study of bacterial viability, sporulation, biofilm organization, morphology, and many other aspects of interest (Fig. 2). Critical point drying of vegetative organisms and biofilms (and air drying of endospores, which are much more robust) allows for the use of scanning electron microscopy to extend the length scale available for study and view the details of individual microorganism and biofilm morphology (Fig. 3). Embedding in resins and sectioning for transmission electron microscopy yields the ultimate level of detail and resolution, providing information on the organization of biofilm extracellular matrix components, mechanisms of sterilization, and so on (Fig. 4). We employed a range of correlative microscopy techniques to study vaporous hydrogen peroxide sterilization of spores inoculated on surfaces and compare biofilms grown using a modified standard CDC reactor method with colony biofilms grown on filters on the surface of agar plates [1,2]. In the first case, microscopy revealed problems with the inoculation technique and was critical in an optimization study to find the best procedure for the particular materials in use. In the second case, fluorescence and electron microscopy were used to characterize and compare the morphological details of the two methods of biofilm growth. This study serves as an example highlighting the utility of microscopy in establishing and verifying model systems. Microsc. Microanal. 21 (Suppl 3), 2015 64 [1] JH Merritt et al, Current Protocols in Microbiology 1B.1 (2005), p. 1-17. [2] ASTM E2562-12: Standard Test Method for Quantification of Pseudomonas aeruginosa Biofilm Grown with High Shear and Continuous Flow using CDC Biofilm Reactor. 2012. Figure 1. Two-fold dilution series of luminescent Klebsiella pneuomoniae in duplicate, demonstrating a detection limit of approximately 10-20 microorganisms emitting measurable photoluminescence within one well of a 96-well plate. Figure 2. Membrane staining of individual bacteria in laser-scanning confocal microscopy. A wide variety of dyes allow for the study of viability, biofilm morphology, etc. Figure 3. Scanning electron microscopy of bacterial endospores, a useful model organism in testing worst-case scenario sterilization conditions. Figure 4. Transmission electron microscopy of bacteria, revealing high-resolution details of the layers of their membranes and inner organelles.");sQ1[33]=new Array("../7337/0065.pdf","Electron Microscopy as an Emerging Analytical Tool for Characterizing","","65 doi:10.1017/S1431927615001129 Paper No. 0033 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electron Microscopy as an Emerging Analytical Tool for Characterizing Biopharmaceuticals. Bridget Carragher, Anette Schneemann, Joyce J. Sung, Sean K. Mulligan, Jeffrey A. Speir, Kat On, Joel Quispe, Clinton S. Potter NanoImaging Services, Inc., San Diego, CA USA Characterization of nanoparticles and biologics is a critical step in the development of important new pharmaceutical products and biosimilars. Biologics pose unique characterization challenges that require an interdisciplinary approach in which several orthogonal methods are used to provide a complete picture. The physical characteristics of a biological product include properties such as the size, shape, morphology and aggregation state of the particles. These properties are often dependent on the specific environment of the particles and thus ideally must be assessed under conditions that reflect the final formulation of the pharmaceutical. Electron microscopy (EM) and in particular cryo-electron microscopy (cryoEM), has a unique advantage in that it provides a direct means of observing the individual particles in a sample, preserved in their natural hydrated state (cryoEM), simultaneously providing information on homogeneity, size distribution, titer, morphology, preservation state, flexibility, and aggregation state. For particles with a regular size and shape, particle averaging methods can provide 3D structural information, complementing X-ray crystallography analysis. We will demonstrate the use of EM as an analytical characterization tool by presenting a number of case studies as highlights. Specifically, we will discuss the characterization of Human Papilloma Virus (HPV) VLPs in GARDASIL�, including the structure of the VLPs alone, on adjuvants, and when interacting with neutralizing antibodies [1]. We will also show how TEM was used as a non-intrusive tool to understand the structure and function of Hepatitis B surface antigen (rHBsAg) VLPs, the active component in the HBV vaccine [2]. We will furthermore demonstrate how TEM can be used to provide supporting information for characterization of a biosimilar drug delivery nanoparticle, a recombinant tuberculosis vaccine antigen, interacting with a lipid-based adjuvant [3], and a bi-specific, tetravalent immunoglobulin G-like molecule [4]. References: [1] Zhao Q, et al. Characterization of virus-like particles in GARDASIL(R) by cryo transmission electron microscopy. Hum Vaccine Immunother. 10 (2013), 1. [2] Mulder AM, et al. Toolbox for non-intrusive structural and functional analysis of recombinant VLP based vaccines: a case study with hepatitis B vaccine. PloS One. 7 (2012), 4. [3] Fox CB, et al. Cryogenic transmission electron microscopy of recombinant tuberculosis vaccine antigen with anionic liposomes reveals formation of flattened liposomes. Int J Nanomedicine. 9 (2014); 1367. [4] Correia I, et al. The structure of dual-variable-domain immunoglobulin molecules alone and bound to antigen. MAbs. 5 (2013), 364. Microsc. Microanal. 21 (Suppl 3), 2015 66 Figure 1: CryoEM of Adenovirus (background) and 3D reconstruction using single particle methods (foreground). Figure 2: CryoEM of liposomal encapsulated doxorubicin. Figure 3: EM of Anti-HER2 mAb (background) and 2D class averages (foreground). Figure 4 : EM of H1N1 (influenza) vaccine.");sQ1[34]=new Array("../7337/0067.pdf","Imaging the Dynamic Release and Capture of Vesicle Membrane Proteins in Mammalian Cells","","67 doi:10.1017/S1431927615001130 Paper No. 0034 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Imaging the Dynamic Release and Capture of Vesicle Membrane Proteins in Mammalian Cells Justin W. Taraska, Kem A. Sochacki and Ben T. Larson National Heart Lung and Blood Institute, National Institutes of Health, Bethesda, MD USA Cells of the nervous system communicate with one another by releasing peptides, neurotransmitters, and small molecules by exocytosis. Triggered exocytosis in neurons and endocrine cells is a highly regulated process. Excitable cells go to great lengths to ensure that exocytosis occurs at precisely the right time and location and that the correct quantity of cargo is released during each exocytic burst. Once vesicle cargo such as lipids and proteins exits a vesicle and enters the plasma membrane it must be corralled and recaptured from the surface of the cell. A pathway proposed to be important for this process is clathrin-mediated endocytosis. Classically, clathrin, adapter proteins, and mechano-enzymes, have been suggested to assemble around vesicle material. This drives the formation of clathrin-coated pits which are able to internalize material into the cell. While many of the individual proteins important for this process have been identified, how these components capture vesicle material and assemble together to drive the formation of an endocytic vesicle is not well understood. To study the pathway of protein release and recapture from exocytic vesicles we imaged the postfusion dynamics of a single vesicle membrane protein, the vesicular acetylcholine transporter (VaChT), from individual exocytic vesicles in living neuroendocrine cells with total internal reflection fluorescence microscopy. We combine these measurements with super-resolution interferometric photoactivation localization microscopy (iPALM), transmission electron microscopy, and modeling to map the nanometer-scale topography and architecture of the plasma membrane structures responsible for the transporter's recapture [1]. We show that after exocytosis, VaChT rapidly diffuses into the plasma membrane but travels only a short distance before it is corralled over a dense network of membraneresident clathrin structures. We propose that the extreme density of these structures acts as a short-range diffusion trap. They quickly sequester diffusing materials and limit its spread across the membrane. This system could provide a means for clathrin-mediated endocytosis to quickly recycle exocytic proteins in highly excitable cells. Next, because the complete molecular identity of these resident clathrin-coated structures are unknown, we developed a novel high-throughput two-color total internal reflection fluorescence microscopy and automated image processing pipeline to map a library of 78 proteins proposed to have roles in membrane trafficking at these sites [2]. With this method we determine both the local and cellwide distributions of these proteins at the plasma membrane. These results provide a unified systemslevel spatial map of the protein landscape of individual endocytic sites and a new framework for the unbiased spatial analysis of cellular systems. With these data we identify a core group of proteins that associate with clathrin-coated structures (CCS) in PC12 cells. We also observe a broad range in the degree of associations. Furthermore, most proteins were found to be distributed amongst these structures and across the membrane in a well-mixed random pattern with no apparent underlying spatial organization. Our combined imaging results illustrate the ability of fluorescence microscopy and quantitative analysis to interrogate the underlying structure of a complex spatial system and screen for new Microsc. Microanal. 21 (Suppl 3), 2015 68 associations and patterns within the cell. The random distributions we observe for the proteins studied suggests an overall model for how the plasma membrane is organized in relation to endocytic structures in neuroendocrine cells. CCSs are very dense, from 2 to 11 times as dense as reported in other cell types. This dense network of CCSs may allow for robust rapid sorting and capture of cargo and a possible enhancement in signal transduction. The network map we generate can also be used as a reference for comparison between different cell types, states, and perturbations. Investigating individual functional units of interacting molecules in their cellular context will deepen our understanding of the selforganization, complexity, spatial organization, and control of biological systems. Figure 1. High resolution imaging of endocytic sites in neuroendocrine PC12 cells. On the left is a TIRF image of a living PC12 cell expressing a GFP-tagged protein that marks single endocytic sites. In the center is a transmission electron micrograph of a platinum replica of the inner membrane of a PC12 cell (Scale bar is 100 nm). Individual clathrin coated pits can be seen to stud the membrane. On the right is a super-resolution 3D iPALM image of labeled clathrin in the same cells (Scale bar is 5 microns). References: [1] Sochacki KA, Larson BT, Sengupta DC, Daniels MP, Shtengel G, Hess HF, Taraska JW, Nature Communications. 3 (2012), p. 1154. [2] Larson BT, Sochacki KA, Kindem JM, Taraska JW, Mol. Biol. Cell. 25(2014), p.2084-93 [3] The authors acknowledge funding from the Intramural Research Program of the National Heart Lung and Blood Institute, National Institutes of Health, USA.");sQ1[35]=new Array("../7337/0069.pdf","Focal Adhesion Kinase Anchoring Kinetics and Regulatory Interactions Quantified by Total Internal Reflection Fluorescence Microscopy","","69 doi:10.1017/S1431927615001142 Paper No. 0035 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Focal Adhesion Kinase Anchoring Kinetics and Regulatory Interactions Quantified by Total Internal Reflection Fluorescence Microscopy Taylor Zak1, Allen M. Samarel1,2 and Seth L. Robia1 Department of Cell and Molecular Physiology, Stritch School of Medicine, Loyola University Chicago, Maywood, IL USA 2 Department of Medicine, Stritch School of Medicine, Loyola University Chicago, Maywood, IL USA The ubiquitously expressed focal adhesion kinase (FAK) plays a central role in modulating vascular smooth muscle cell (VSMC) migration. FAK-Related Non-Kinase (FRNK), the autonomously expressed C-terminal domain of FAK, is a VSMC-specific protein whose expression is greatly increased in injured arteries, and serves to inhibit growth and migration.[1] We have previously shown that FRNK inhibits FAK-dependent VSMC signaling and migration in cell culture and in vivo. To investigate molecular mechanisms of FAK regulation by FRNK, we quantified intermolecular FRET with acceptor-selective photobleaching. We observed FRET from FAK to FAK and FRNK to FRNK (homo-oligomers) and from FAK to FRNK (heterooligomeric regulatory complex) (Fig. 1). For all complexes we observed very little FRET in the cytoplasm (Fig. 1), consistent with another group's previous studies of FAK-FAK binding [2]. To investigate the stoichiometry of these regulatory complexes, we performed progressive photobleaching FRET, which reveals whether the regulatory complex is a dimer or a higher order oligomer [3]. The data suggest a highly heterogeneous population of FAs, with differential FRET efficiency and stoichiometry observed for different FAs within the same VSMC. The data suggest that FAK forms predominantly dimers[2], but some FAs show evidence of higher order oligomers [4]. To investigate the kinetics of FAK/FRNK anchoring to FAs we performed single-molecule TIRF microscopy of YFP-FAK/FRNK in live cells (Fig. 2A). This image is the simple summation of all single molecule events in a series of 10,000 images, and panels B-D represent an enlargement of the 7.5 micron box shown in panel A. Fig. 2B reveals the typical diffraction-limit resolution obtained by conventional TIRF imaging. A single frame of the image series (Fig. 2C) reveals 4 discrete single molecules of FRNK visible in the region as Gaussian-shaped spots. Single molecule transients were fit with the ImageJ plugin QuickPALM to obtain super-resolution maps of all of the FRNK anchoring events (Fig. 2D). We quantified the fluorescence transients to observe single molecule binding/unbinding events and quantify the kinetics of these events. Single molecule fluorescence trajectories (Fig. 2E) were characterized by brief fluorescence bursts of uniform intensity flanked by periods of dark when no molecules were present. Histogram analysis (Fig. 2F) reveals two major Gaussian peaks corresponding to background ("0") and bright FRNK monomer binding events ("1X"). Events with approximately 2-fold brighter fluorescence intensity were only rarely observed ("2X"), suggesting FRNK anchors in FAs as a monomeric species. We obtained the average decay profile of a fluorescence burst (Fig. 2G) and performed dwell time analysis. A histogram of dwell times revealed a multiphasic distribution, with many short binding events (<50 ms) and another smaller population of longer dwell-time binding events (>200 ms) (Fig. 2H). Overall the data are compatible with the hypothesis that regulatory interactions of FAK and FRNK control the dynamics of FAK anchoring in focal adhesions, a key determinant of cell growth and migration. Future experiments will determine the role of structural mutations and phosphorylation in tuning FAK/FRNK anchoring kinetics. 1 Microsc. Microanal. 21 (Suppl 3), 2015 70 Fig. 1. FRET evidence of FAK-FAK and FRNK-FRNK homo-oligomers as well as a heterodimer species. FRET values were obtained from focal adhesions and very little FRET was observed in the cytoplasm. Fig. 2. Single molecule fluorescence studies of FRNK anchoring. A) TIRF microscopy sum image. Boxed region is 7.5 � 7.5 m B) Enlarged view of boxed region C) Image of single molecules of YFPFRNK observed in boxed region. D) Super-resolution map of FRNK anchoring events in boxed region E) Fluorescence bursts indicating single molecule anchoring events. F) A histogram of measured fluorescence values consistent with a single species. G) Avg. of 3000 fluorescence bursts. H) Dwelltime analysis of anchoring events reveals the persistence of FRNK at anchoring sites. References: [1] Koshman YE, Engman SJ, Kim T, Iyengar R, Henderson KK, Samarel AM. Role of FRNK tyrosine phosphorylation in vascular smooth muscle spreading and migration. Cardiovasc Res. 2010;85:571-81. [2] Brami-Cherrier K, Gervasi N, Arsenieva D, Walkiewicz K, Boutterin MC, Ortega A, et al. FAK dimerization controls its kinase-dependent functions at focal adhesions. EMBO J. 2014;33:356-70. [3] Kelly EM, Hou Z, Bossuyt J, Bers DM, Robia SL. Phospholamban oligomerization, quaternary structure, and sarco(endo)plasmic reticulum calcium ATPase binding measured by fluorescence resonance energy transfer in living cells. The Journal of biological chemistry. 2008;283:12202-11. [4] Goni GM, Epifano C, Boskovic J, Camacho-Artacho M, Zhou J, Bronowska A, et al. Phosphatidylinositol 4,5-bisphosphate triggers activation of focal adhesion kinase by inducing clustering and conformational changes. Proc Natl Acad Sci U S A. 2014;111:E3177-86. [5] The authors acknowledge funding from NIH (HL092321, HL062426).");sQ1[36]=new Array("../7337/0071.pdf","The Biogenesis of Endophilin B1 Containing Vesicles","","71 doi:10.1017/S1431927615001154 Paper No. 0036 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Biogenesis of Endophilin B1 Containing Vesicles Yan Chen1, Jinhui Li1, Joachim D. Mueller1, Barbara Barylko2, and Joseph P. Albanesi2 1. 2. Physics Department, University of Minnesota, 116 Church ST. S.E., Minneapolis, MN, 55455 Pharmacology Department, UT Southwestern Medical Center, Dallas, TX, 75239. Fluctuations of fluorescent light emerging from a small region of the sample were first considered more than four decades ago [1]. The technique is known as fluorescence correlation spectroscopy (FCS) and has been extended over the years to study a variety of processes. Here, we focus on a particular type of analysis that is sensitive to the stoichiometry of the complexes, namely brightness analysis. Brightness analysis and its related techniques are used to extract the concentration of the fluorescent complexes and their brightness ("the intensity of a single particle") from the distribution of photon counts. Brightness is a unique parameter that relies on the single molecule sensitivity of FFS and provides a direct quantification of the oligomeric status of populations of GFP-tagged proteins. Since EGFP is a monomer with distinct fluorescence intensity and recombinant 2x-EGFP elicits twice the intensity, the overall brightness of a GFP-tagged protein is a direct function of its oligomeric state (Fig. 1). Thus, brightness analysis is a robust parameter for determining the stoichiometry of protein oligomerizations. Our studies seek the utility of FFS in the characterization of small cytoplasmic complexes, such as interorganelle transport vesicles, which are below the limit of resolution of conventional live-cell imaging approaches. The mobility of intracellular vesicles is higher because the vesicles are in an aqueous environment, not on a membrane sheet. Whereas commonly used techniques, such as imaging, have limited ability to reveal the physical characteristics of intracellular vesicles, FFS is ideally suited to study such vesicles. The fluctuation amplitude increases with the decrease of sample concentration and increases with the increase of brightness per particle. Intracellular vesicles are at low concentration and each vesicle has multiple copies of a particular protein, which gives rise a high signal-to-noise ratio in fluctuation amplitude, a parameter that is absolutely crucial for quantitative fluorescence fluctuation spectroscopic measurements. The concentration of vesicles is determined by the brightness spike analysis [2]. The number of brightness spikes at a given data acquisition time, Spike Count Rate, is related to the sample concentration. Figure 2 displays the brightness spikes as a function of fluorescence sphere concentration. The Spike Count Rate was linear at low concentrations (below 100pm), but deviated from linearity at high concentrations. The non-linearity is due to the multiple events occurring very close in time at high concentrations. Nevertheless, even at high concentration, the Spike Count Rate still increases as a function of concentration. Spike Count Rate is therefore a suitable indicator of the sample concentration of rare but bright species. The brightness spike analysis was used to study endophilin B1 (EndoB1), a protein believed to be important for inducing membrane curvature. The Spike Count Rates of EndoB1 samples were not zero, but they were also not concentration dependent (Fig. 3). In fact, the Spike Count Rate of EndoB1 scattered at all the concentrations examined, suggesting that the incorporation of EndoB1 into a complex depends on another cellular process, perhaps under the control of limiting cellular cofactors. In such a case it is possible that the amount of these putative cofactors may vary from cell to cell, which will explain the scatter we observe in our data. We also performed a similar brightness spike analysis for Microsc. Microanal. 21 (Suppl 3), 2015 72 endophilin A2 (EndoA2), a soluble cytosolic dimeric protein. No brightness spikes were detected and the Spike Count Rate was zero at all EndoA2 concentrations examined (Fig. 3). The identity of vesicles is often revealed by the nature of their coat proteins. We employed heterospecies partition analysis, which allows detection and quantification of the co-existence of two species within the same complex [3]. We examined the co-mobility of EndoB1 and its putative coat protein by either cotransfecting EndoB1-GFP with Caveolin1-mCherry or with Clathrin-mCherry. The resulting heterospecies brightness vectors are displayed in figure 4. All the heterospecies vectors of EndoB1/Clathrin were scattered along the "green species only line", which indicated that EndoB1-GFP wasn't associated with Clathrin-mCherry on vesicles. The HSP vectors of EndoB1/Caveolin1, on the other hand, were scattered, but with many data in the "green-red co-mobile zone", suggesting that at least a subset of EndoB1 containing vesicles have Caveolin1 as a coat. In conclusion, we show that multiple copies of endophilin B1 (EndoB1) associate with vesicles coated with caveolin but not clathrin. Our data indicate that intracellular vesicle trafficking is a selective process, and overexpression alone doesn't guarantee that two proteins will co-exist on the same vesicles. By studying the coat, and the concentration of the vesicles, we are able to show the likelihood of the location and mechanisms of the biogenesis of EndoB1 vesicles. References: [1] D. Magde, E. L. Elson, and W. W. Webb, "Fluorescence correlation spectroscopy. II. An experimental realization," Biopolymers, vol. 13, no. 1, pp. 29�61, Jan. 1974. [2] J. Li, B. Barylko, J. Johnson, J. D. Mueller, J. P. Albanesi, and Y. Chen, "Molecular Brightness Analysis Reveals Phosphatidylinositol 4-Kinase II Association with Clathrin-Coated Vesicles in Living Cells," Biophys. J., vol. 103, no. 8, pp. 1657�1665, Oct. 2012. [3] B. Wu, Y. Chen, and J. D. M�ller, "Heterospecies partition analysis reveals binding curve and stoichiometry of protein interactions in living cells," Proc. Natl. Acad. Sci., vol. 107, no. 9, pp. 4117�4122, Mar. 2010. Fig. 1. A plot of 1xGFP or 2xGFP versus protein concentration. Each data point represents an independent live cell measurement. Fig. 2. Fluorescent sphere dilution experiment. Spike counting analysis was applied to fluorescent spheres at various concentrations. Two different cut-off brightness values are applied to the data. Fig. 4. Heterospecies partition analysis of EndoB1-GFP and clathrin-mCherry; EndoB1-GFP and caveolin-mCherry. Fig. 3. Spike counting analysis of EndoB1 and EndoA2. Brightness spikes analysis reveals that EndoB1 is enriched in higher order complexes.");sQ1[37]=new Array("../7337/0073.pdf","Ebola Virus Hemorrhagic Fever","","73 doi:10.1017/S1431927615001166 Paper No. 0037 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Ebola Virus Hemorrhagic Fever Cynthia S. Goldsmith Infectious Diseases Pathology Branch, Centers for Disease Control and Prevention (CDC), Atlanta, GA 30333. This past year has seen the largest Ebola virus outbreak on record, centered in West Africa in Guinea, Liberia, and Sierra Leone.1 Cases were first reported in Guinea on March 21, 2014; by March 30, cases were reported in Liberia and in May Sierra Leone also reported cases. At the time of this writing, within these three countries, there have been 22,444 cases, with 13,810 cases being laboratory-confirmed.2 There have been 8959 deaths, although the number may be higher due to under-reporting. Organizations from several countries have been working to defeat this disease, by treating the sick patients, by providing infrastructure, and by tracing and monitoring contacts of the patients. Tissue specimens were collected for electron microscopy in a previous Ebola virus outbreak in 1995. Overwhelming viremia was present, as evidenced by the presence of numerous Ebola virus particles in the blood vessels. In the liver, a large number of viral particles, both normal and aberrant, were present in the sinusoids. Large filamentous inclusions, indicative of viral replication, were observed in hepatocytes, and infected cells were particularly concentrated close to the portal tract (Fig 1). Proliferations of the plasma membranes of infected cells were commonly noted. In the lung, mature Ebola virus particles were observed in the interstitium and within the alveolar space (Fig 2), and viral inclusions were seen in alveolar macrophages and in endothelial cells. In the dermis of the skin, mature particles were observed in the connective tissue matrix, and inclusions were present in endothelial cells (Fig 3). The presence of mature virus particles in the dermis of the skin and in the alveolar space of the lung may have important implications in terms of person-to-person transmission. The involvement of endothelial cells may explain the increased vascular permeability and shock seen in patients. Electron microscopic studies should further our understanding of the pathology and pathogenesis of this devastating disease. References [1] CDC, MMWR 63(2014)548. [2] http://www.cdc.gov/vhf/ebola/outbreaks/2014-west-africa/case-counts.html Microsc. Microanal. 21 (Suppl 3), 2015 74 FIG 1. Large viral inclusions (arrowhead) in hepatocytes of liver from a patient who died of Ebola virus hemorrhagic fever. Mature virus particles (arrow) are within the sinusoid. FIG 2. Virions are found in the alveolar space of the lung. FIG 3. Ebola virus inclusion (arrowhead) in endothelial cell in the dermis of the skin. Virus particles (arrow) are seen within the connective tissue matrix. RBC, red blood cell. Bars, 1 �m.");sQ1[38]=new Array("../7337/0075.pdf","Microwave Assisted Rapid Diagnosis of Plant Virus Diseases by TEM","","75 doi:10.1017/S1431927615001178 Paper No. 0038 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microwave Assisted Rapid Diagnosis of Plant Virus Diseases by TEM Bernd Zechmann1 and G�nther Zellnig2 1. Baylor University, Center for Microscopy and Imaging, One Bear Place #97046, Waco, TX 767987046, USA. 2. Karl-Franzens University of Graz, Institute of Plant Sciences, Schuberstrasse 51, 8010 Graz, Austria. A rapid and unambiguous diagnosis of plant virus diseases is of great importance for agriculture. Investigations of ultrastructural changes induced by viruses within plants by using transmission electron microscopy (TEM) are often necessary to clearly identify the viral agent. Nevertheless, with conventional methods such investigations can take several days and are not suited for the rapid diagnosis of plant virus diseases [1]. Microwave assisted plant sample preparation can help to drastically reduce sample preparation time for TEM investigations with similar or even better ultrastructural results as observed after sample preparation with conventional methods [1,2]. This massive reduction of sample preparation time can be attributed to dielectric heating induced by microwave irradiation which causes a temperature rise inside the whole sample whereas conventional heating starts at the specimen surface. The increase in temperature can then enhance and accelerate the diffusion of reagents, protein crosslinking and the polymerization of the resin [3,4]. In this study microwave assisted sample preparation and negative staining were applied on two model virus-plant systems [Nicotiana tabacum plants infected with Tobacco mosaic virus (TMV), Cucurbita pepo plants infected with Zucchini Yellow mosaic virus (ZYMV)] to clearly diagnose the viral agent and ultrastructural alterations by TEM in less than one day. Sample preparation for ultrastructural and cytohistochemical investigations was performed with a commercially available microwave device. With the help of microwave irradiation sample preparation time was reduced to 136 min for ultrastructural investigations and to 89 min for cytohistochemical investigations. After cutting and contrasting of the samples ultrastructural investigations revealed typical features of TMV-disease such as virions accumulating in parallel layers within the cytosol of TMV-infected Nicotiana tabacum plants (Figure 1a) and of ZYMV disease like cylindrical inclusions in the cytosol of Cucurbita pepo plants (Figure 1b). Additionally, negative staining of the sap of TMV and ZYMV infected leaves revealed rod shaped viral particles with an average length and width of 280 and 17 nm for TMV (Figure 1c) and 707 and 12 nm for ZYMV (Figure 1d). These data were in accordance to the ultrastructural properties and size ranges for these virions described in the literature after samples had been prepared conventionally by chemical fixation and embedding at room temperature. [5,6,7,8]. For cytohistochemical investigations sections of TMV-infected leaves were treated with a primary antibody against TMV coat protein and a secondary gold conjugated antibody which took about 100 min. Gold particles bound to TMV coat protein could be clearly identified in less than 4 hours after the beginning of sample preparation in the areas of cells where TMV accumulated in parallel form (Figure 1e). Summing up, the results presented in this study clearly demonstrate that microwave assisted plant sample preparation for ultrastructural investigtations and cytohistochemical localization of viral coat protein together with negative staining methods enabled a clear diagnosis of plant virus diseases by TEM in less than one day. As these methods could also be applied in the fields of human and veterinary pathology they have a large potential for the rapid diagnosis of diseases in humans and animals. Microsc. Microanal. 21 (Suppl 3), 2015 76 References [1] B Zechmann and G Zellnig, J. Microsc. 233 (2009), p. 258. [2] T Kurth et al, Protist 163 (2012), p. 296. [3] ASY Leong and RT Sormunen, Micron 29 (1998), p. 397. [4] A De la Hoz et al, Chem. Soc. Rev. 34 (2005), p. 164. [5] A Gal-On, Mol. Plant Pathol. 8 (2007), p. 139. [6] ML Smith et al, Virology 348 (2006), p. 475. [7] AV Reunov et al, Biol. Bulletin 33 (2006), p. 409. [8] B Zechmann and G Zellnig, J. Virol. Meth. 162 (2009), p. 163. [9] The authors gratefully acknowledge funding from the Austrian Science Fund (P20619, P22988) Figure 1. Transmission electron micrographs showing (a) virions (*) in the cytosol of TMV-infected Nicotiana mesophyll leaf cells, (b) cylindrical inclusions appearing as scroll like structures (arrows) in the cytosol of ZYMV-infected Cucurbita mesophyll leaf cells, (c) TMV-particles after negative staining of the sap of TMV-infected Nicotiana leaves, (d) ZYMV-particles after negative staining of the sap of ZYMV-infected Cucurbita leaves, and (e) gold particles bound to TMV coat protein (arrowheads) in the cytosol of TMV-infected Nicotiana leaf cells. C=chloroplasts with and without starch (St), M=mitochondria, V=vacuole. Bars=1�m.");sQ1[39]=new Array("../7337/0077.pdf","Immunoelectron Microscopy Reveals Varied Surface Expression of Potential Vaccine Targets of Chlamydia trachomatis","","77 doi:10.1017/S143192761500118X Paper No. 0039 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Immunoelectron Microscopy Reveals Varied Surface Expression of Potential Vaccine Targets of Chlamydia trachomatis Ru-ching Hsia1,2, Johanna Sotiris1 and Patrik Bavoil3 1. Electron Microscopy Core Imaging Facility, University of Maryland Baltimore, Baltimore, USA Department of Neural & Pain Sciences, University of Maryland School of Dentistry, Baltimore, USA 3. Department of Microbial Pathogenesis, University of Maryland School of Dentistry, Baltimore, USA 2. Chlamydia trachomatis, an obligate intracellular bacterium, is a leading cause of sexually transmitted infection in the world. It is estimated that 6.8 % of sexually active young females aged 14-19 in the US may be infected with Chlamydia [1]. Previous studies have demonstrated that a nine-member C. trachomatis protein family, the polymorphic membrane proteins (Pmps), plays a critical role in pathogenesis and may be a potential vaccine target [2]. These proteins share specific structural motifs and are transported to the chlamydial surface via an autotransporter mechanism [2]. We report here on the differential subcellular location of two of these proteins, PmpD and PmpI, at different stages of chlamydial development using immuno electron microscopy. C. trachomatis-infected HeLa cells were fixed at mid or late developmental stages during a 48 hr infection cycle. Fixed monolayers were scraped off tissue culture vessels, enrobed in 2.5% low melting temperature agarose and trimmed into ~1 mm3 pieces. Agarose pieces containing infected cells were cryo-protected with 30% sucrose, attached to a cryo-ultramicrotome specimen pin and flash frozen in liquid nitrogen, or dehydrated and embedded in unicryl resin using the progressive lowering temperature (PLT) method [3]. Immunolabeling was performed using either thawed cryo ultrathin sections following the Tokuyasu method [4], or unicryl ultrathin sections on grids. At the mid developmental stage (18 hrs post infection), metabolically active chlamydial forms multiply within an inclusion as reticulate bodies (RBs). Immunogold particles labeling PmpI were distributed mostly on the RB membrane whereas the majority of gold particles labeling PmpD were located in the RB cytoplasm (Figure 1). Furthermore, a cluster of gold particles along one fraction of the RB membrane were occasionally observed both in PmpI and PmpD immunolabeling experiments. At the late developmental stage (45 hrs post infection), RBs transform into metabolically inert elementary bodies (EBs). The gold particles labeling PmpI were mostly found in the cytoplasm of EBs whereas those labeling PmpD were found on the outer surface of EB envelope (Figure 2). These results illustrate the differential spatial distribution of PmpI and PmpD at late developmental stages for C. trachomatis. This study demonstrates the capability of immuno electron microscopy to reveal the differential subcellular location of PmpD and PmpI at different stages of chlamydial development. The presence of PmpD at the external surface of infectious chlamydial EBs suggest PmpD may be a better vaccine target. Our results indicate that PmpI and PmpD may perform different functions at different developmental stages and imply that the secretion of these two autotransported proteins may be regulated at different developmental stages. Last, the unique clustered distribution of PmpI and PmpD gold labeling strongly suggests the existence of a supramolecular structure at a specific location of the chlamydial surface. Microsc. Microanal. 21 (Suppl 3), 2015 78 References: [1] CDC Morbidity and Mortality Weekly Report 60 (2011), p. 370. [2] D Crane et al, Proc Natl Acad Sci USA.103 (2006), p.1894. [3] P Gounon, Methods Mol Biol.117 (1999), p.111. [4] K Tokuyasu, Journal of Microscopy 143 (1986), p.139. [5] The authors acknowledge funding from the NIH Shared Instrument grant, Grant Number 1S10RR26870-1. Figure 1. Immunogold labeling of unicryl ultrathin sections of C. trachomatis-infected HeLa cells 18 hrs post infection. Sections were labeled using primary antibody against PmpI (A) and PmpD (B) and a secondary antibody conjugated with 10 nm gold particles. Yellow arrows indicate gold particles on the chlamydial surface. Red arrowheads indicate gold particles in the cytoplasm. The yellow line indicates clustered gold particles along the RB surface. Bar, 500 nm. Figure 2. Immunogold labeling of thawed cryo-ultrathin sections of C. trachomatis-infected Hela cells at 44 hrs post infection. Sections were labeled using primary antibody against PmpI (A) and PmpD (B) and a secondary antibody conjugated with 10 nm gold particles. Bar, 100 nm.");sQ1[40]=new Array("../7337/0079.pdf","EM Detection of Viruses in Organ Transplant Patients","","79 doi:10.1017/S1431927615001191 Paper No. 0040 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 EM Detection of Viruses in Organ Transplant Patients Sara E. Miller and David N. Howell Department of Pathology, Duke University, Durham, NC 27710 Due to immunosuppression, transplant patients are more susceptible to infections, particularly viruses. The strengths of EM detection include the facts that it is rapid, non-invasive or minimally invasive, does not require probes, and can detect many different viruses. A limitation is that is it not as sensitive as PCR, but this can be an advantage in instances where endogenous viruses in healthy individuals may be detected by exquisitely sensitive tests. Additionally, concentration methods, such as ultracentrifugation, ultrafiltration, or immunoaggregation can increase the chances of finding them by EM in fluids, and localization of focal pathology in tissues can be enhanced by semithin sections, on-slide embedment of H&E-stained sections, confocal microscopy, and laser capture. In renal transplant patients, the types of viruses that may be seen by negative staining in urine include polyomavirus, adenovirus, small round viruses (SRV), such as those in the enterovirus family, and herpesviruses, such as cytomegalovirus. In our small bowel transplant patient population, 8 of 13 pediatric patients have had episodes of viral gastroenteritis diagnosed by EM. Viruses observed in this group are adenovirus, rotavirus, and small round structured viruses (SRSV), such as astrovirus, calicivirus, and Norwalk virus. Patients with bone marrow/stem cell transplants are profoundly immunodeficient. Viruses observed in fluids from this population are the same as in other transplant patients, e.g., adenovirus and enteroviruses from several sites, adenovirus and polyomavirus from urine, and adenovirus and SRSV, from stool. EM of thin sections of biopsies occasionally may reveal viruses. Diagnostic pitfalls in the use of EM to detect viral pathogens can result from an absence of virus in the specimen submitted, because infections may exist in organs not transplanted but susceptible due to immunosuppression. Additionally, some viruses are not shed in effluents as they infect organs without contact with body cavities. Other types may remain cell-associated, i.e., stick to cell membranes, or incorporate viral DNA into the host genome and no longer produce recognizable virions. Incorporated "provirus" may also cause neoplasms, e.g., Epstein Barr virus (EBV) causes post transplant proliferative disease, a B-cell lymphoma, and Human herpes virus (HHV)-8 causes Kaposi sarcoma. Proper specimen preparation is important in detecting viruses. Useful procedures to enhance visualization in fluids include concentrating the specimen, using hydrophilic support films, and removing cell debris by low speed centrifugation. Tissue samples can be prepared by routine embedding procedures or by rapid methods, including use of small tissue slivers and high heat resin curing and/or by microwave techniques. Clues to virus identification in fluids include numerous small particles of similar size/shape, darker stain around the particle as opposed to a simple white spot, a fringe around the outside of amorphous-shaped particles, or filaments with a herringbone pattern [1, 2, 3, 4]. Viruses may be naked icosahedral particles (Fig. 1a, b) or enveloped with a pliable membrane (Fig. 1c). Nucleocapsids inside enveloped viruses may be spherical (like Fig. 1a, b), helical (Fig. 1c), complex, or nondescript. Confusing things that look like viruses in fluids include small round organelles, such as exosomes, or droplets, such as lipids. Filamentous things, such as mucoprotein, may resemble filamentous nucleocapsids of the myxoviruses. Microsc. Microanal. 21 (Suppl 3), 2015 80 Clues to virus identification in tissues include numerous small particles of similar size/shape, paracrystalline arrays (beware of glycogen), round particles in the nucleus (beware of nuclear pores), vesicles with regular-sized dark centers (beware of neurosecretory vesicles), and particles budding from cellular membranes [1, 2, 3, 4]. Other structures that look like viruses in tissues are round organelles such as clathrin-coated or Golgi vesicles. Filamentous organelles in tissue, e.g., microtubules, intermediate filaments, and tubuloreticular inclusions, may be confused with helical nucleocapsids. With some exceptions, DNA viruses are usually constructed in the nucleus (Fig. 2a), and RNA viruses are made in the cytoplasm (Fig. 2b). Enveloped viruses usually bud from cell membranes (Fig. 2b inset). In summary, all transplant patients are immunosuppressed and susceptible to virus infection. Diagnosis is imperative because the therapy for infection is diametrically opposed to that for organ rejection. Clues to virus identification include numerous sub-cellular particles of similar size/shape that differ from normal cellular organelles, but attention should be paid to possible look-alikes. [1] SE Miller. J Electr Microsc Tech 4 (now J Micros Res Tech) (1986) p 265. [2] FW Doane, N Anderson. Electron Microscopy in Diagnostic Virology, A Practical Guide and Atlas (Cambridge Univ Press, New York) (1987) 178 pp. [3] Palmer EL, Martin ML. Electron Microscopy in Viral Diagnosis (CRC Press, Boca Raton) (1988) 194 pp. [4] MA Hayat, SE Miller. Negative Staining (McGraw-Hill, New York) (1991) 253 pp. [5] CS Goldsmith, SE Miller. Clin Microbiol Rev 22 (2009) p 552. Fig.1. Negative stains. a. SRV (naked); b. adenovirus (naked); c. paramyxovirus (enveloped), spiked virion (right), helical nucleocapasids (left). Bar = 100 nm in a-c. Fig. 2. Thin sections. a. adenovirus, a DNA virus, in the nucleus (arrows), bar = 1 �m; inset, bar = 100 nm; b. paramyxovirus nucleocapsids, an RNA virus, in the cytoplasm (arrows), bar = 1 � m; inset, complete enveloped virions (fuzz around the outside corresponds to the spikes in Fig. 1c), bar = 100 nm.");sQ1[41]=new Array("../7337/0081.pdf","In-situ Topographical and Elemental Characterisation of Biological Soft Matter","","81 doi:10.1017/S1431927615001208 Paper No. 0041 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In-situ Topographical and Elemental Characterisation of Biological Soft Matter C.L. Collins1, C. McCarthy1, S.R. Burgess1 1 Oxford Instruments Nanoanalysis, Halifax Road, High Wycombe, HP12 3SE, UK Since the 17th century, light microscopy (LM) has been used to gain topographical and structural information about soft biological matter, however, it has limited spatial resolution and gives no information about the chemical composition of the structures. Fluorescence light microscopy (FLM) gives more detail about the antibody or protein responses of live cells but is localised to a specific chemical reaction. As a technique, it is offers some information about the organic matrix, but no wider chemical information about surrounding structures or the inorganic composition of the sample. Electron microscopy (EM) combined with Energy Dispersive Spectroscopy (EDS) offers biological analysts a way of simultaneously characterising their samples both structurally and chemically and at a much higher spatial resolution than can be achieved in either LM or FLM � limited only by the performance of the microscope. There is also no need to add any chemical markers or fluorophores. The introduction of large area EDS detectors has changed soft matter biological analysis. Large active area detectors (50 � 150 mm2) with excellent light element performance and greatly improved collection efficiencies [1,2] mean increased counts and improved prospects for elemental analysis in the EM. The increased speed of collection also opens the possibility for large scale elemental characterisation. This is important in biological analysis where samples are often inhomogeneous. Here, we present application examples of soft matter biological tissue analysed with Oxford Instruments X-MaxN 150mm2 SDD detectors. Figure 1 shows a human tissue retrieved from surgical biopsy from a patient who has had a metal-onmetal total hip replacement. Such prosthetics are known to physically degrade with use [3]. Wear nanoparticles are often found in the surrounding tissue, but until recently, there was no easy way to identify and characterise the different fragments. This tissue sample was prepared by embedding in a resin then sectioned into semithin sections (~1m thick) which were then laid onto a pure silicon wafer for analysis in the SEM. The sample was analysed using a Tescan Mira FEGSEM fitted with a multiple EDS detector system offering a combined active area of 600mm2. Data was collected at 10kV with an input count rate of 60kcps. The total collection time was 15 minutes. The results clearly show the chemical composition of the steel fragment as it lies embedded in the tissue. Figure 2 shows the importance of large scale characterisation. The sample is a whole wheat plant stem which was embedded in resin and planed down to the area of interest. A number of individual X-ray maps (66 in total) were collected and montaged together to create a single image (2.8mm in width) composed entirely of X-ray elemental information � there is no electron image included in this data. This removes topographical contrast. The widely varying distributions of the elements only become clear at low magnifications. The Cl and Os are artefacts from the sample preparation method used but offer structural information while the K and S are naturally present. Microsc. Microanal. 21 (Suppl 3), 2015 82 Conclusions: The introduction of new large area SDD EDS detectors offers a new way of collecting important insitu information about biological soft matter and tissues. Fast mapping allows large areas of sample to be analysed and viewed as a whole. The result is that truly informative chemical information data can now be collected on beam sensitive materials without compromising the sample or the data acquired. Figure 1. Human tissue biopsy from a metal hip replacement semi-thin section (1m thick) on silicon wafer. Maps were acquired at 10kV with an input count rate of 60kcps. Figure 2 Large area X-ray map cross section of a whole stem of a wheat plant composed of 66 individual x-ray maps collected at 25kV and montaged together to create a single image. Image width = 2.8mm. References: [1] S Burgess et al, Microscopy & Microanalysis (2011), p 1176-1177 [2] A Hyde et al, Microscopy & Microanalysis (2013), p 1312-1313 [3] J. Jacobs et al, J.Bone Joint Surg Am. 80 (10) (1998) p 1447-58. [4] Acknowledgements: Oxford Instruments are very grateful to Dr. Zhidao Xia from the Centre for Nanohealth, Swansea and Dr. Giorgio Perino from the Hospital for Special Surgery, New York for providing the tissue biopsy sample shown in Figure 1 and Jean Devonshire of Rothamsted Research for the wheat sample shown in Figure 2.");sQ1[42]=new Array("../7337/0083.pdf","Biomineralization at Interfaces Revealed with 4D Electron and Atom Probe Tomographies.","","83 doi:10.1017/S143192761500121X Paper No. 0042 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Biomineralization at Interfaces Revealed with 4D Electron and Atom Probe Tomographies. Xiaoyue Wang1, Brian Langelier1, Anders Palmquist2 and Kathryn Grandfield1 1. 2. Department of Materials Science and Engineering, McMaster University, Hamilton, Canada. Department of Biomaterials, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden. Biomineralization at engineered interfaces has huge implications not only for osseointegration of bone and dental implants in health sciences but also for biomimetic synthetic materials, which have wide applications in the energy and environmental sectors. Resolving the spatial and chemical structure of the interface of hierarchical biominerals, such as human bone interfacing to titanium, on the sub-nanometer scale has the potential to shed light on the mechanisms of biomineralization and structure-property relationships. Previous work has unveiled the structure of the bone�implant interface with single-tilt axis electron tomography, however the chemical structure remains unresolved until now [1]. In this work, on-axis rotation tomography, enabling a complete 360� sample rotation, was used to circumvent the "missing wedge" which is a main source of artifacts in conventional electron tomography [2]. Based on the needle-like sample geometry, correlative 4D electron energy loss spectroscopy (EELS) tomography from the same sample was acquired to offer a further understanding of nanoscale chemical structure at implant interfaces in three dimensions. Atom probe tomography (APT), combing sub-nanometer resolution with chemical sensitivity across the entire periodic table, supports and complements these electron tomographies to strengthen the whole 4D view (spatial plus chemical) of the human bone�implant interface. A nano-structured titanium implant and surrounding bone was retrieved from the human maxilla after 4 years in service [3]. Using focused ion beam (FIB), specimens from the bone-titanium interface were sharpened into approximately 200 nm diameter needle-shaped specimens for further investigation. In the scanning transmission electron microscope (STEM) (Titan 80-300, operated at 300kV), sharpened specimens were rotated through � 90�using an on-axis tomography holder and images were recorded every 2�on a high-angle annular dark-field (HAADF) detector which provides compositional contrast. After on-axis electron tomography, the needle-shaped sample was put back into the FIB and milled to 50 nm in diameter. Then, a series of STEM EELS spectrum images and correlative dark-field images were acquired between � 70� on a double-aberration corrected Titan 80-300 STEM. Chemical maps were extracted for elements Ca, C, Ti and O, consistent with the mineral and collagen phases of bone, titanium implant, and surface oxide, respectively and reconstructed by applying the alignment and reconstruction algorithm for images to each elemental map. Using visualization software, all elemental maps are merged to identify the organic and inorganic motifs at the interface. Taking advantage of the chemical sensitivity (down to 1 ppm) of laser-pulsed APT, trace elements and co-localization of ions within specific regions in bone structure and at the material surface were identified. We present for the first time 4D (spatial + chemical) tomographies of human bone interfacing to nanostructured titanium devices with correlative on-axis electron tomography, chemically sensitive EELS tomography and trace-element sensitive APT. These correlative advanced four-dimensional microscopies provide a foundation for understanding the structure and chemical nature of the interface of hierarchical biominerals, and could be extended to the study of biomineralization mechanism at other Microsc. Microanal. 21 (Suppl 3), 2015 84 biological interfaces. References: [1] K. Grandfield, S. Gustafsson and A. Palmquist. Nanoscale, 5 (2013), p. 4032. [2] P.A. Midgley and M. Weyland. Ultramicroscopy, 96 (2003), p. 413. [3] F.A. Shah et al. Nanomedicine-Nanotechnology Biology and Medicine, 10 (2014), p. 1729. [4] The authors acknowledge funding from the Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery Grant program and BIOMATCELL VINN Excellence Center of Biomaterials and Cell Therapy, Sweden. Microscopy was performed at the Canadian Centre for Electron Microscopy at McMaster University, a facility supported by NSERC and other government agencies. Figure 1. 4D EELS tomography and APT of human bone implant interface. (a) HAADF STEM image of needle-like sample. (b) A representative 2D EELS map and (c) corresponding 3D EELS tomogram, where red represents Carbon, green represents Calcium and white represents Titanium. (d) APT of a small fraction of the same sample.");sQ1[43]=new Array("../7337/0085.pdf","Magnetotactic Bacteria and Honey Bees: Model Systems for Characterising an Iron Oxide Mediated Magnetoreceptor.","","85 doi:10.1017/S1431927615001221 Paper No. 0043 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Magnetotactic Bacteria and Honey Bees: Model Systems for Characterising an Iron Oxide Mediated Magnetoreceptor. Jeremy Shaw1, Alastair Boyd1, Michael House2, Martin Saunders1, Falko Mathes3 and Boris Baer4. Centre for Microscopy, Characterisation and Analysis, The University of Western Australia, Perth, Australia. 2. Biomagnetics Group, The University of Western Australia, Perth, Australia. 3. School of Earth and Environment, The University of Western Australia, Perth, Australia. 4. Centre for Integrative Bee Research (CIBER), The University of Western Australia, Perth, Australia. Magnetic field perception (magnetoreception) has been described across a broad spectrum of animals, including insects, birds, reptiles and mammals. Despite extensive behavioural evidence demonstrating that these animals are able to sense magnetic fields, the cellular mechanisms involved in transducing a magnetic stimulus to a neuronal response remain a long standing mystery in biology. The magnetite hypothesis is one possible explanation for how a putative magnetoreceptor might function, and is based on neuronal activation by intracellular nanoparticles of magnetite (Fe3O4) [1]. This is expected to occur when torque is applied to the membrane anchored particles of magnetite in response to changes in Earth-strength magnetic fields. Magnetotactic bacteria provide good evidence for the magnetite hypothesis in animals as they are able to produce chains of magnetite nanoparticles (magnetosomes) in their cytosol (Fig. 1A) and use them for orientation in the Earth's magnetic field. Finding the cells responsible for magnetoreception has proven to be extremely challenging. The search has been hampered by the fact that only few cells are expected to harbour the sense and they could be located anywhere in the body. Additionally, iron is a widespread biological and environmental element, which can result in contamination of samples and misinterpretation of results [2 and 3]. For these reasons, finding an iron-based magnetoreceptor in situ within an organism or tissue using optical or electron microscopic methods is comparable to the classic needle-in-ahaystack problem. New methodological developments addressing these limitations are timely. The presence of magnetic particles is a key attribute of magnetoreceptive cells, which can be exploited experimentally in the search for the anatomical location of these cells. Bulk extraction methods that can separate inorganic particles from organic tissue are promising approaches for confirming the presence or absence of magnetite particles. We used such an approach and developed a novel procedure to extract and concentrate magnetosome particles from bacteria and prepare them for examination using a range of imaging and analytical platforms, including optical, X-ray and electron based microscopy. As a proof of concept, we demonstrate that these particles can be recovered from honey bee abdomen tissue that has been spiked with magnetotactic bacteria (Fig. 1B). Our data show that the magnetosome particles retain their mineral/crystallographic properties after the extraction process, as demonstrated by selected area electron diffraction patterns obtained from undigested and spiked particles (Fig. 1C and Table 1). Our technique was also able to extract iron oxide granules from the honey bee fat body (Fig. 1B). Although these granules are not believed to play a role in magnetoreception, our method can efficiently capture a range of iron materials, which can then be screened for the presence of candidate magnetoreceptor particles. The ultimate aim is to then locate and characterise these particulates in situ [4]. 1. Microsc. Microanal. 21 (Suppl 3), 2015 86 References: [1] J Kirschvink, M Walker and C Diebel, Current Opinion in Neurobiology 11 (2001) pp. 462-467. [2] C Treiber et al, Nature 484 (2012), pp. 367-370. [3] N Edelman et al, Proceedings of the National Academy of Sciences 112 (2015) pp. 262-267. [4] The authors acknowledge funding from the ARC (DE130101660), The University of Western Australia RDA scheme and the Australian Microscopy & Microanalysis Research Facility (AMMRF). Figure 1. A) Dark field STEM micrograph of a magnetotactic bacterium highlighting the chain of magnetite particles used for orientation (arrowhead). B) Bright-field TEM micrograph of iron granules (IG) extracted from honey bee abdomens using the digestion process. This sample has been spiked with bacterial magnetite particles to demonstrate that they can be recovered during the process and produce C) selected area electron diffraction patterns consistent with those generated from undigested controls. Half-circles denote the rings measured to produce the planar spacings presented in Table 1. Control d Spiked d 5.95 4.90 4.85 111 Magnetite hkl d 4.85 2.97 2.53 2.42 Maghemite hkl 110 111 210 211 220 212 310 311 d 5.9 4.81 3.73 3.40 2.95 2.78 2.64 2.51 2.41 Hematite hkl d 3.69 2.70 2.52 2.29 Goethite hkl d 4.98 4.17 3.38 2.69 2.58 2.52 2.49 2.45 2.30 2.25 Lepidocrocite hkl 200 d 6.20 3.28 2.97 2.42 2.36 - 020 110 120 012 3.02 3.00 220 210 101 104 110 130 021 101 040 111 200 121 2.58 2.57 311 2.45 2.25 2.45 2.14 222 222 410 111 006 Table 1. Measured (�2%) planar spacings of pre-processed (control) and processed (spiked) particles from magnetotactic bacteria, compared to known spacings of several biogenically relevant iron oxide species, values are in �.");sQ1[44]=new Array("../7337/0087.pdf","Development of a Bioinspired Stroma Model to Study the Role of Collagen Topology in Pancreatic Ductal Adenocarcinoma","","87 doi:10.1017/S1431927615001233 Paper No. 0044 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Development of a Bioinspired Stroma Model to Study the Role of Collagen Topology in Pancreatic Ductal Adenocarcinoma Cole Drifka,1,2 Agnes Loeffler,3 Kevin Eliceiri,1,2 W. John Kao1,2 1. 2. Department of Biomedical Engineering, University of Wisconsin, Madison, USA Laboratory for Optical and Computational Instrumentation, University of Wisconsin, Madison, USA 3. Department of Pathology, University of Wisconsin, Madison, WI, USA Pancreatic ductal adenocarcinoma (PDAC) is one of the deadliest human malignancies. It is currently the fourth-leading cancer killer in the U.S and projects to be second in cancer fatalities by 2030. The very poor prognosis for PDAC patients is largely attributed to a propensity for early and aggressive metastasis, late diagnosis, and resistance to current therapeutic regimens. Despite ongoing research, little significant progress in improving patient survival has occurred over the past decades. Traditionally, PDAC research and histopathological evaluation of tumor tissue has focused primarily on the genetics and behavior of transformed cells. There is growing evidence, however, that a dynamic interplay between cancer cells and the adjacent stroma influences tumor growth, angiogenesis, metastasis, and therapeutic resistance [1]. A mainstay tissue component of most tumor stromas, including PDAC, is the extracellular matrix protein collagen. Recently, a number of groups have shown that collagen fiber organization is uniquely altered in various cancer types and may carry biological and clinical significance [2,3]. To date, visualization of collagen-based changes has been greatly accelerated by Second Harmonic Generation (SHG) imaging, a laser scanning microscopy technique that can provide high-resolution, quantifiable images of discrete collagen fibers in intact tissues without the need for exogenous staining. For example, researchers have used SHG to identify unique collagen organizational patterns in breast cancer coined "tumor-associated collagen signatures" (TACS). One of these signatures, TACS-3, describes bundles of straightened, aligned collagen fibers oriented perpendicular to the tumor boundary [4]. Mechanistically, it is hypothesized that these fibers act as pathways that facilitate cancer cell migration away from the tumor and towards vasculature during the metastatic process. It has also been shown that the detection of TACS-3 in routine breast cancer histopathology slides can predict disease recurrence and patient survival [5]. Due to the emerging clinical significance of collagen reorganization, translational technologies have been developed to better detect and quantify changes [6]. Increased fibrillar collagen has long been clinically documented in PDAC, but its specific organization throughout the tumor has not been explored in the context of disease progression. Using SHG imaging and computational collagen fiber segmentation, we have demonstrated that robust collagen-based structural changes can be detected in the immediate vicinity of cancer cells in PDAC histopathology tissues (Figure 1). However, the underlying biological relevance of collagen remodeling in PDAC tumors and clinical manifestation remains unclear. Although histopathology is the gold standard for characterizing patient tissues, these samples are inherently 2D and static making it impossible to clearly elucidate what influence dynamic collagen changes have on cancer cells. Thus, a pairing between traditional histopathology and controllable 3D experimental in vitro models is needed to better study the role of collagen reorganization in PDAC. Microsc. Microanal. 21 (Suppl 3), 2015 88 Despite the emerging role of the stroma, experimental in vitro models that accurately recreate the heterogeneous PDAC microenvironment have been lacking. Realizing the importance of studying PDAC in the context of the stroma and the identifying key limitations of traditional 2D models, we previously developed a 3D in vitro microfluidic model of the PDAC microenvironment [7]. Microfluidic technology provides a unique platform to directly visualize live cell interactions with 3D collagen networks. Ongoing work is seeking to recapitulate key features of collagen topology at the PDAC cancer-stroma interface in viable in vitro tissue constructs by using SHG data obtained from clinical tissues as a blueprint. Individual fiber metrics (i.e. length, diameter) and bulk fiber metrics (i.e. concentration, alignment) are being assessed in the context of cancer and cancer-associated myofibroblasts. Experimental models that recreate collagen organization changes present in PDAC tissues are expected to provide mechanistic insight into the effect of collagen alterations on PDAC cell behavior during disease progression and the opportunity to assess novel therapeutics in a pathophysiologically relevant context. References: [1] M Erkan et al, Nat Rev Gastroenterol Hepatol 9 (2012), p. 454. [2] K Levental et al, Cell 139 (2009), p. 891. [3] O Nadiarnykh et al, BMC Cancer 10 (2010), p. 94. [4] P Provenzano et al, BMC Medicine 4 (2006), p. 38. [5] M Conklin et al, Am J Pathol 178 (2011), p. 1221. [6] J Bredfeldt et al, J Biomed Opt 19 (2014), p. 16007. [7] C Drifka et al, Lab Chip 13 (2013), p. 3965. Figure 1. Collagen fibers are elongated and aligned around PDAC ducts compared to normal ducts. Cells are green, SHG signal from fibrillar collagen is yellow. Scale = 100 �m.");sQ1[45]=new Array("../7337/0089.pdf","From Image Tiles to Web-Based Interactive Measurements in One Stop","","89 doi:10.1017/S1431927615001245 Paper No. 0045 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 From Image Tiles to Web-Based Interactive Measurements in One Stop Antoine Vandecreme1, Tim Blattner1, Michael Majurski1, Peter Bajcsy1, Keana Scott2, and John Henry J. Scott2 1. 2. Information Technology Lab, National Institute of Standards and Technology, Gaithersburg, MD. Material Measurement Lab, National Institute of Standards and Technology, Gaithersburg, MD. Several advanced microscopes have the capability to acquire overlapping image tiles automatically. Such automated acquisitions provide a way to image a large spatial coverage of a specimen and to take measurements at multiple physical length scales. However, there is a need to assist imaging scientists with computational solutions that convert raw image tiles to calibrated, stitched, and viewable Gigapixel images with interactive measurement tools. In this work, we addressed the problems of (1) designing image calibration and tile stitching algorithms, (2) prototyping a workflow of computational steps to form Giga-pixel images, and (3) enabling interactive viewing and measurements over Giga-pixel images. The goal of the work is to allow scientists to upload a set of image tiles and then receive a URL with a Google Map like interface for viewing large images together with web-enabled measurement tools. The challenges are primarily related to (a) automation of tile stitching, (b) applying automated algorithms to images larger than RAM, (c) minimizing execution times of pre-processing algorithms, and (d) managing on-demand image viewing and measurements over Giga-pixel images to enable interactivity of image explorations. The overall system is overviewed in Figure 1. Our approach to the automation of image tile stitching is derived from existing direct image alignment methods [1][2]. Specifically, the stitching algorithm is based on the Phase Correlation Image Alignment Method [3]. The computational workflow is constructed with a flat field calibration step, followed by tile stitching and multi-resolution image pyramid building. The multi-resolution image pyramid has been previously used in the context of "Virtual Microscopy" [4], and it allows for efficient transmission and viewing of large images. We made an extensive use of OpenSeadragon [5], a JavaScript library for pyramid-based visualization of images on the web. We added to OpenSeadragon a set of measurements tools such as scale bars, rulers, and image sub-setting options. These tools were designed as pluggable widgets which can be enabled based on image type and have been contributed to the open source OpenSeadragon project. One challenge was to build a pyramid out of stitched images (see Figure 2) which each is larger than the available system memory. This was solved by building a multi-resolution pyramid directly from input image tiles and positions generated by the image tile stitching software. Another challenge with on-demand stitching and visualization as a service is the execution time associated with each step of the processing pipeline. This was solved by parallelizing the algorithms to run either on hybrid CPU and GPU platforms (stitching) and on a computer cluster (calibration and pyramid building). The current capabilities are being tested on the NIST intranet. Disclaimer: Commercial products are identified in this document in order to specify the experimental procedure adequately. Such identification is not intended to imply recommendation or endorsement by NIST, nor is it intended to imply that the products identified are necessarily the best available for the purpose. Microsc. Microanal. 21 (Suppl 3), 2015 90 References: [1] R. Szeliski, "Image Alignment and Stitching: A Tutorial," Found. Trends� Comput. Graph. Vis., vol. 2, no. 1, pp. 1�104, 2006. [2] S. Preibisch, S. Saalfeld, and P. Tomancak, "Globally optimal stitching of tiled 3D microscopic image acquisitions.," Bioinformatics, vol. 25, no. 11, pp. 1463�5, Jun. 2009. [3] C. Kuglin and D. Hines, "The Phase Correlation Image Alignement Method," in Proceedings of the 1975 IEEE International Conference on Cybernetics and Society, 1975, pp. 163�165. [4] E. Romero, F. G�mez, and M. Iregui, "Virtual Microscopy in Medical Images: a Survey," Microscopy, no. 571, pp. 996�1006, 2007. [5] "Open Seadragon," Open Seadragon project, 2015. . Figure 1: Overview of the designed system. Figure 2: An illustration of multi-resolution image pyramid representation.");sQ1[46]=new Array("../7337/0091.pdf","Zen-like Simplicity is Hard Work, and Other Secrets I Learned from Peter Swann","","91 doi:10.1017/S1431927615001257 Paper No. 0046 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Zen-like Simplicity is Hard Work, and Other Secrets I Learned from Peter Swann Ondrej L. Krivanek1 1 Nion Co., 11511 NE 118th St., Kirkland, WA 98034, USA Peter Roland Swann was born in Williton in Somerset UK a few years before World War II. Wartime rationing was a fact of life when he was growing up, and there were other privations too. Their effect on Peter seems to have been to make him uncommonly resourceful and imaginative in transforming whatever life provided him with into wonderful new things. This talent served him particularly well when he finished his Ph. D. studies at Cambridge, UK, and joined the research staff at the US Steel Research Center in Monroeville PA, USA. Unraveling the mysteries of dislocations and other defects in metals by imaging them in the newly available electron microscopes was a hot topic back then. Peter developed a double-tilt sample holder for a Siemens Elmiskop I electron microscope [1], which allowed crystals to be imaged in the precise orientation needed for g.b contrast of dislocations (g = chosen reflection, b = Burgers vector). This was a major step on a career path that Peter honed to perfection over the years: developing new capabilities for electron microscopes, preferably in areas that had hardly been touched. The new holder also led to the founding of Gatan. Siemens liked the design so much that they approached Peter about supplying them with ten double-tilt holders. Peter's brother Rex, a mechanical engineer, was living nearby in Pittsburgh, and together they decided to start a business manufacturing the holders. They purchased a lathe and a milling machine at Sears, and sought legal help to incorporate the business. Peter was studying German at the time and thought that "getan", the past participle of "tun" (to do), which means "done, finished, completed" would be a great name for the business, with a nice ring and a good message. However, a typo in a legal document resulted in the name being spelled "Gatan". Peter and Rex thought that this name sounded good too. Hating unnecessary paperwork, they incorporated the company under the name Gatan on April 1, 1964. Peter became a Reader (and then Professor) at Imperial College London a few months after Gatan was founded. Spending his summers in Pittsburgh, he kept up a steady stream of designs for Gatan to manufacture, such as environmental chambers and heating holders for high-voltage electron microscopes. He also made major advances in corrosion science [2]. In 1976, he switched to working at Gatan full-time, and the range of new designs he introduced grew considerably. Keep It Small and Simple (KISS) was a key dictum of Peter's. My first collaboration with him, 16 years after Gatan was founded, involved redesigning a serial detection electron energy-loss spectrometer I had built at UC Berkeley. Peter did the mechanical design, and "KISS" very much stood out. Where I had used some 20 screws to hold the spectrometer's prism and detection chamber together, Peter used just 6. Where I did not worry much about the manufacturing steps that would need to be undertaken, Peter carefully optimized the design so that most parts could be made in a few steps on a lathe. Working with Peter was very inspirational, and I joined Gatan full-time (on April 1, 1985) after Peter established the Gatan R&D facility in California. This M&M session is focusing on environmental microscopy, to which Peter made major contributions. Microsc. Microanal. 21 (Suppl 3), 2015 92 Other products he designed or co-designed that enabled key progress to be made in various fields of research included cryo-transfer holders for cryo-microscopy, ion mills and mechanical polishing equipment for general specimen preparation, TV-rate cameras for EM imaging, energy-loss spectrometers for analysis, and an early focused ion beam (FIB) machining system (called PIMS � precision ion milling system) for precision sample preparation. Blaise Pascal's famous excuse: "I have made [this letter] longer than usual because I have not had the time to make it shorter" was not for Peter - he took whatever time was necessary for Zen-like purity of design. Long hours were needed to achieve the purity, and Peter once calculated that out of the 168 hours in each week, some 50 were needed for sleep, 20 for general upkeep, eating and recreation, and the rest � roughly 100 hours per week � could be spent on work. This was not popular with Peter's family and friends, but we all had the greatest respect for his work ethic and dedication. And we loved the amusing moments: Peter had a sharp wit and a wonderful sense of humor, and nobody ever got bored in his company. Peter's focus on the essential and great sense of humor made his scientific talks captivating and fun. His last talk, as far as I know, was given at Carnegie Mellon University in Pittsburgh in 2011. Fig. 1 shows the talk's last slide: a revolutionary new sample holder for a high voltage electron microscope. The surprised onlookers should have known better. Peter was always an original thinker/doer, and he excelled at finding the humor in any situation. Much like a Zen master would. References: [1] G.V. Patser and P.R. Swann, J. Sci Instr. 39 (1962) p. 58. [2] P.R. Swann in "Mats Science and Engineering" ed. M. McLean, (Maney Publishing, London, 2002) p. 53. Figure 1. A side-entry sample-heating holder designed by Peter Swann for a high voltage electron microscope equipped with his environmental chamber. The EM research group at Imperial College, London, surrounds the holder, Peter himself is clamped in it.");sQ1[47]=new Array("../7337/0093.pdf","Peter Swann's Key Role in the Past, Present and Future of Dynamic In-situ TEM","","93 doi:10.1017/S1431927615001269 Paper No. 0047 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Peter Swann's Key Role in the Past, Present and Future of Dynamic In-situ TEM Edward D Boyes1,2,3, Pratibha L Gai1,2,4 and Archie Howie1,5 York JEOL Nanocentre1 and Departments of Physics2, Electronics3 and Chemistry4, University of York, York, YO10 5DD, UK Cavendish Laboratory5, University of Cambridge, J J Thomson Avenue, Cambridge, CB3 0HE, UK The breadth of vision combined with meticulous attention to detail which characterized Peter Swann's contributions to electron microscopy can be detected in his PhD work on stacking faults. He studied the influence of stacking fault energy on the dislocation arrangements and plasticity in deformed copper alloys, becoming acutely aware that the standard stereo-viewing cartridge with a single-axis of tilt of only about 40 was woefully inadequate for systematic diffraction contrast imaging. Nevertheless one aspect of a broader vision was realized when his measurements of stacking fault energy in Hume Rothery alloys [1] showed that this depended mostly on the valence electron density. Not content with this bridge between mechanical and electronic properties, Peter boldly suggested that stacking faults could provide a link to chemistry through their role in the critical problem of stress corrosion cracking (SCC). The motivation to develop and imaginatively deploy instrumentation ranging from goniometers to full-blown microscopy systems for in-situ chemistry was therefore in place. After a few years at US Steel in Pittsburgh, where his first goniometer was developed, Peter returned to Imperial College and became involved in the UK HVEM program. The opportunity he found to design and build a variety of sophisticated specimen stages for this community together with the first incarnation of the Gatan company was later brilliantly described by him [2]. Peter's "pioneering work in promoting in-situ stage design and experimentation" was however gratefully acknowledged much more quickly in an influential account of the already long history of in-situ TEM [3]. This history goes in several directions, (a) better vacuum for deposition or sample modification under more controlled near-UHV conditions [4,5], (b) studies of reactive samples avoiding contact with the air by conventional sample loading, generally by in-situ preparation [6], (c) liquid environments to retain hydration [7], including of biomaterials, and to study structural transformation under stimuli, both chemical and electrical, (d) gas reaction studies [8], especially to characterize quantitatively key catalyst reaction sequences and to identify intermediate reaction species which may be metastable with respect to conditions of gas environment and/or temperature, and therefore not reliably accessible in discontinuous or ex-situ studies, but important to understanding reaction mechanisms and thereby to establish critical catalyst design criteria [9,10], and (e) irradiation studies using the e-beam or ions [11]. Peter Swann made revolutionary advances in many of these areas. In (d), his fully functional 4-aperture differentially pumped system and attendant custom-built hot stage [12] developed for the AEI EM7 HVEM most notably overcame the pressure limitation of earlier systems [8] and critically introduced useful specimen tilt. With this equipment he was able to follow technologically important processes such as the reduction of haematite to iron [13]. He did not take his eye off the ball of SCC but had to content himself with detailed study of the cracks formed after transfer to the HVEM from a separate reaction chamber [14]. Under (e) he studied some of the beam-induced knock-on irradiation processes occurring in the HVEM including the electron beam induced aqueous oxidation of silicon [15]. Microsc. Microanal. 21 (Suppl 3), 2015 94 Electron microscopy development by the main instrument manufacturers has targeted improvements towards atomic resolution and sensitivity of analysis, with static samples often created for microscopy. But the true value of the tools lies in addressing problems of significant societal importance where it is often a crucial challenge to be able to prepare a suitable specimen and to study its structure and behavior under realistic conditions. Gatan have played a major role here not only with the goniometers already mentioned but also in the development of ion thinning equipment and other specialized accessories for making thin samples from bulk material. Occasionally more extensive developments were pursued such as the UHV system for the Philips/FEI 300kV EM430 [5]. Peter's early ETEM system was developed further by Gai and Doole at Oxford leading to key classical g.b analyses of crystal defects and critical understanding of the atomic scale microstructural basis of catalyst properties, including activation, activity, selectivity, deactivation and recovery of complex oxide catalysts for commercial hydrocarbon oxidation [9,10]. At Oxford and DuPont [16] further progress was made in designing an integrated system to fit higher resolution modern instruments [10] and this is now used in commercial machines dedicated to the ETEM mode. The most recent equipment developed at York has 0.1nm image resolution in both ESTEM and ETEM and a full array of imaging and analytical methods [17,18,19]. The ESTEM uses redesigned Gatan hot stages and DENS MEMS technology. These both retain the full single atom sensitive performance of the core instrument with full gas modifications and pumping systems operational for fundamental reaction mechanism studies [19]. SCC continues to be a phenomenon of great technological importance particularly for the stainless steel components used in reactors and storage vessels. Although the influence of slip planes on the crack morphology is well demonstrated [14], the precise significance of the stacking fault energy in austenitic stainless steels is not clear. Nevertheless the stacking fault energy continues to be one of the input parameters in formulae simulating the corrosion sensitivity. The need to realize Peter Swann's dream of atomic resolution, in-situ SCC studies in boiling MgCl2 is therefore still here to challenge us. References [1] A Howie and P R Swann, Phil Mag, (1961) 6, p1215 [2] P R Swann in `Materials Science and Engineering; Its Nucleation and Growth' ed M McLean pub Woodhouse press, (2002) p53 [3] E P Butler and K F Hale, `Dynamic Experiments in the Electron Microscope', pub North Holland, 1981 [4] Y Kondo, K Takayanagi et al, Ultramicroscopy (1991) 35, p111 [5] D J Smith, P R Swann et al, Ultramicroscopy, (1993) 49, p26 [6] D A Goulden, Phil Mag, (1976) 33, p393 [7] D D Double, A Hellawell and S J Perry, Proc Roy Soc, (1978) A359, p435 [8] H Hashimoto, T Naikui, T Etoh and K Fujiwara, Jap J Appl Phys, (1968) 7, p946 [9] P L Gai, et al, Nature (1990) 348, p430; and J. Solid St. Chem. (1983) 49, p25. [10] P L Gai et al, Science (1995) 267, p661 [11] B L Eyre et al, Phil Mag (1971) 23, p439 [12] P R Swann and N Tighe, Jernkont. Ann, (1971) 155, p497 [13] P R Swann and N Tighe, Metall. Trans. B Process Metallurgy (1977) 8, p479 [14] G M Scamans and P R Swann, Corrosion Science (1978) 18, p983 [15] H M Flower and P R Swann, Corrosion Science (1977) 17, p305 [16] E D Boyes and P L Gai, Ultramicroscopy, (1997) 67, p219 [17] E D Boyes, M Ward, L Lari and P L Gai, Ann. Phys. (Berlin), (2013) 525, p423 [18] J Sagar, L Fleet, M Walsh, L Lari, E D Boyes, O Whear, T Huminiuc and A Hirohata, Applied Physics Letters, (2014) 105, 032401 [19] P L Gai, L Lari, MR Ward and E D Boyes, Chem. Phys. Lett. (2014) 592, p335");sQ1[48]=new Array("../7337/0095.pdf","In-Situ Tensile Deformation of Additively Manufactured Ti 6Al 4V","","95 doi:10.1017/S1431927615001270 Paper No. 0048 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In-Situ Tensile Deformation of Additively Manufactured Ti 6Al 4V John R. Porter1, Robert Wheeler1 and Michael Velez1 1. UES Inc., Dayton, USA. Ti 6Al 4V is the most mature alloy currently used in the manufacture of parts by the Electron Beam Melting (EBM) additive manufacturing (AM) process. EBM is a powder bed process in which parts are built up, layer by layer, by using an electron beam scanned over the powder bed in a raster that defines a given layer of the part. The electron beam melts a pool of material that resolidifies as the probe traverses the layer. Typically, each layer is between 50�m and 70�m thick. During resolidification, the underlying metal grows epitaxially such that mm long, [100]c oriented columnar grains can develop with the columnar grains growing parallel to the build direction. Such columnar grains can clearly be seen in Fig. 1A. After cooling from the build temperature, Ti 6Al 4V is an alpha/beta alloy that initially solidifies as cubic beta phase that mostly transforms on further cooling to hexagonal alpha phase. The well established (110)c//(0001)h orientation relationship results in a Widmanstatten microstructure with 12 variants of alpha phase within each prior beta columnar grain. The Widmanstatten microstructure is depicted in Figure 1B. While the AM as-built microstructure is well characterized and its formation well understood, the goal of an equiaxed prior beta structure has not been realized. The question then presents itself as to whether the prior beta grain boundaries represent planes of weakness. Here we address this by undertaking a series of in-situ micromechanical tests in the SEM. Microsamples were prepared that were oriented longitudinally within one prior beta grain, transversely within one prior beta grain and transversely across one prior beta boundary. Tests were conducted using a Microtest in-situ test rig. Samples were FIB cut from previously prepared electropolished TEM foils selected from regions of the foil that were ~10�m thick. After cutting the tensile bar outline, the samples were FIB trimmed to have flat and parallel opposing sides. The tensile gauge after preparation was approximately 100 �m x 40�m x 10 �m thick. Figure 3 shows a sequence of images taken in the unloaded condition and just prior to failure. Fracture is believed to have happened at a prior beta grain boundary. When tested parallel to the growth direction, the sample showed a very different fracture behavior. The sample necked and the supported load dropped. The test was interrupted prior to failure for further microstructural analysis. References: [1] [2] [3] [4] B. Saller et al., Microscopy and Microanalysis 16 (Suppl. 2), (2010) p. 702. J. R. Porter et al., Materials Science and Technology (MS&T) (2011) p. 1434. P. A. Shade et al., Acta Materiala 57 (2009) p. 4580. The authors acknowledge funding from the AFRL under contract FA8650-14-C-5021 Microsc. Microanal. 21 (Suppl 3), 2015 96 5000 �m Figure 1. (a) Tuning fork part grown by EBM revealing columnar prior beta grain structure. (b) Widmanstatten structure of the transformed alpha within the prior beta grains. 15 �m Figure 2. Sequence of images taken during micromechanical test just prior to failure. Transverse sample spanned a prior beta grain boundary. Figure 3. Sequence of images taken during micromechanical test just prior to yield. Build direction sample included a prior beta grain boundary parallel to the sample.");sQ1[49]=new Array("../7337/0097.pdf","Computer-Controlled In Situ Gas Reactions via a MEMS-based Closed-Cell System","","97 doi:10.1017/S1431927615001282 Paper No. 0049 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Computer-Controlled In Situ Gas Reactions via a MEMS-based Closed-Cell System L. F. Allard1, W. C. Bigelow2, Z. Wu1, S. H. Overbury1, K. A. Unocic1, M. Chi1, W. B. Carpenter3, F. S. Walden3, R. L. Thomas3, D. S. Gardiner3, B. W. Jacobs3, D. P. Nackashi3 and J. Damiano3 1. 2. Physical Sciences Directorate, Oak Ridge National Laboratory, Oak Ridge, TN, USA. Department of Materials Science & Engineering, University of Michigan, Ann Arbor, MI, USA. 3. Protochips, Inc., Raleigh, NC, USA. Closed-cell TEM specimen holders based on MEMS-fabricated heater devices allow atomic resolution to be obtained on e.g. catalyst materials, at elevated temperatures and pressures [1-3]. We have shown that resolution in STEM imaging mode is largely unaffected by the gas pressure and cell temperature [4]. Understanding the physical characteristics and behavior of the MEMS heater devices in different gas environments has enabled development of an in situ reaction cell that is easy to use, and generates more reliable data. The holder-based approach does not require a dedicated (S)TEM, and most existing microscopes are compatible with current holder designs. The Protochips AtmosphereTM Gas E-cell enables the ultra-stable heating performance of the AduroTM thin film heater devices, and incorporates a fully automated, computer-controlled gas delivery manifold with near-instantaneous closed-loop temperature control, independent of the gas pressure and composition. Figure 1 shows the holder tip, and a cross-section schematic showing the narrow gap (nominally 5 �m) between the heater membrane and an amorphous SiN window between which the gas is contained. The cell can be operated in both static and flowing gas conditions; careful selection of the supply and return capillary diameters ensures that the pressure in the reaction cell is essentially the same as the pressure measured in the chosen (of 3) supply tank. Figure 2 shows the gas manifold, and the computer GUI that is used to control the experiment. Several unique features are included, such as an automated Pump/Purge cycle to prepare the cell for the chosen reaction gas, automated shut down of the experiment, and most importantly, a running data record of the temperature and pressure that can be registered precisely to each time-stamped digital image recorded during the experiment. Examples of the operation of the Atmosphere system for studies of gas reactions on fine clusters, nanoparticles, and "bulk" alloy powders will be discussed; two examples are given here. Figure 3 shows the oxidation behavior of 22-atom clusters of Au (Au22) on titania (P25 powder). Figures 3a,b show the starting condition of the clusters at 100�C with "vacuum" (<1Torr) in the cell and then under 300Torr of O2 with the latter image recorded after only a few minutes at the higher pressure. Some minor cluster consolidation is seen e.g. between inset areas A and B. Temperature was increased in 50�C increments with additional imaging of the same sample areas, with a final set recorded at 400 �C as shown in Fig. 3c. Further cluster growth/consolidation and possible evaporation is suggested by areas such as insets B and C shown in Figs. 3d,e. Figure 3d was recorded at 10Mx, and even in the gas cell some discrete atoms can be discerned in the clusters. Figure 4 shows oxidation of a NiAl alloy particle prepared by crushing powder of 30-50 �m particle size. Initial imaging (Fig. 4a) at 300�C in a full atmosphere of air showed no discernible changes over time. Heating to 800 �C (Fig. 4b) caused the formation of a "frothy" coating that was shown by pre- and post-reaction EDS analysis (Figs. 4c,d) to be largely AlOx. Further details of both these experiments will be discussed [5]. References [1] LF Allard et al., Microsc. Microanal. 18 (2012), p. 656. [2] JF Creemer et al., Ultramicroscopy 108(9) (2008), p. 993. Microsc. Microanal. 21 (Suppl 3), 2015 98 [3] HL Xin, et al., Microsc Microanal 19 (2013), p. 1. [4] LF Allard, et al., Microsc. Microanal. 20 (Suppl 3) (2014), p. 1572. [5] Microscopy research at ORNL sponsored in part by the U.S. Dept. of Energy (DOE), Office of Energy Efficiency and Renewable Energy, Vehicle Technologies Program, as part of the Propulsion Materials Program, and (ZW and SO) by the DOE Office of Science, Basic Energy Sciences, under Award # ERKCC96. 1a 2 b 3a Fig. 1. a) Tip of gas-reactor holder; b) geometry of the gas cell. Fig. 2. Computer GUI and manifold, showing e.g. gas supply to manifold, and data area with running record of T and P for reaction. Fig. 3. a) Au22 clusters on P-25 titania in vacuum at 100�C; b) same T in 300Torr O2; c) after heating from 100�C to 400�C, note changes between e.g. inset areas B and C; d) and e) inset areas at higher magnification. Fig. 4. HAADF images and corresponding EDS showing oxidation of NiAl alloy particle for 8 min at 800�C in a full atmosphere of air. 4a c d b");sQ1[50]=new Array("../7337/0099.pdf","Approaching the "Lab in the Gap": First Results from a Versatile In-situ (S)TEM","","99 doi:10.1017/S1431927615001294 Paper No. 0050 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Approaching the "Lab in the Gap": First Results from a Versatile In-situ (S)TEM Felix B�rrnert1,2,4, Heiko M�ller3, Martin Linck3, Alexander Horst2, Angus I. Kirkland4, Bernd B�chner2, and Hannes Lichte1 1. 2. Speziallabor Triebenberg, Technische Universit�t Dresden, 01062 Dresden, Germany. IFW Dresden, PF 270116, 01171 Dresden, Germany. 3. CEOS GmbH, Englerstra�e 28, 69126 Heidelberg, Germany. 4. Department of Materials, University of Oxford, Parks Road, Oxford OX1 3PH, United Kingdom. In the past, the general development of microscopes was governed by the desire for ever higher resolution in the micrographs. This led to programs like the TEAM project resulting in as fantastic machines as the TEAM I microscope offering a resolving power of about half an �ngstr�m in both TEM and STEM mode [1]. In this course, virtually all means were taken to increase the resolving power even if it put severe limitations to other aspects of microscope aided research. As the most prominent examples in (S)TEM serve the millimeter-sized pole piece gap and the few-Tesla-scale magnetic lens field the sample resides in. A recent survey of the DOE in the future needs of microscopy research resulted in a significant change of direction; besides multi-dimensional and ultrafast microscopy the "lab in the gap" approach has been identified as the most promising development in future electron microscopy [2]. To this end, the key development is increasing the space around the sample for multimodal in-situ experiments. In this contribution, we present a fully operational prototype transmission electron microscope offering a sample chamber with a 70 mm "pole piece gap" that is accessible through five arbitrarily usable ports ISO-K DN63 from several sides, see Figure 1. Additionally, the sample space is completely magnetic field free. Conceivable configurations are, for example, a tube with a diameter of slightly more than 60 mm could be driven horizontally through the entire microscope column; with a differential pumping system enabling operation in gaseous environments. Another example is the possible combination of a liquid helium continuous flow cryostat providing long-time stable low temperatures and mobile electrical probers for in-situ transport measurements on structures found and controlled ad hoc in the microscope. To achieve a decent resolving power with this large sample space, the microscope is operated in a Cs corrected Lorentz mode for both conventional and scanning imaging mode. The resolving power of the instrument is about 1 nm in conventional imaging mode and better than 5 nm in scanning mode at 200 kV electron acceleration voltage [3]. An upgrade of the electron optics that enhances the microscope's resolving power in conventional imaging mode to ca. 0.5 nm is planned for the near future. It is noteworthy that all of the alterations to the original state of the microscope, a double Cs corrected JEOL JEM-2010F, are fully reversible. To aid navigation on sample positions that are not electron transparent, e. g. for placing mobile probers on pre-processed contact pads, we installed a secondary electron detector. For the mapping of electric and magnetic fields at the nanoscale via off-axis electron holography, an electrostatic biprism is available. For the design of new experiments to be performed inside the microscope, we built a testing rack with a copy of the microscope's new sample chamber. Also, the rack has the same outer shape as the microscope with its peripherals, thus, all necessary attachments can be checked for collisions and space Microsc. Microanal. 21 (Suppl 3), 2015 100 requirements. The main area of research the microscope is planned for is solid state physics. Examples are in-situ characterization and manipulation of unconventional superconductors, magnetic nanostructures, and topological phenomena. However, this does not exclude any other experiment conceivable for this microscope [4]. References: [1] P Ercius et al, Microsc. Microanal. 18 (2012), p. 676. [2] Report of the Basic Energy Sciences Workshop on the Future of Electron Scattering & Diffraction (2014), URL http://science.energy.gov/~/media/bes/pdf/reports/files/Future_of_Electron_Scattering.pdf. [3] F B�rrnert et al, Ultramicrosc. (2015), DOI 10.1016/j.ultramic.2014.11.011. [4] The authors acknowledge financial support from the European Union under the Seventh Framework Programme under a contract for an Integrated Infrastructure Initiative (Reference 312483--ESTEEM2). Figure 1. Microscope with new sample chamber, test rack with sample chamber, and sample chamber on test rack viewed along electron beam direction");sQ1[51]=new Array("../7337/0101.pdf","Growth Morphology and Defects in 2D Heterostructures and Interfaces","","101 doi:10.1017/S1431927615001300 Paper No. 0051 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Growth Morphology and Defects in 2D Heterostructures and Interfaces Joshua A. Robinson,1 Sarah Eichfeld,1 Yu-Chuan Lin,1 Ning Lu,2 Moon Kim2 1. Materials Science & Engineering; and Center for 2D and Layered Materials, The Pennsylvania State University Park, PA, 16802 2. Materials Science & Engineering, The University of Texas at Dallas, Dallas, TX Since the successful isolation of graphene, two-dimensional materials have rapidly moved to the forefront of "next generation" materials. The many applications range from enhancing the structural properties of composite material properties,1 to water filtration,2 biosensing,3 catalysis,4 photonics,5 and ultra-low power electronics.6 However, none of these applications will be possible without a concerted effort to develop techniques to understand how the 2D layers are grown and their material performance. We have developed processes to synthesize a variety of 2D materials, with an emphasis on "beyond graphene" layers.7�9 In the case of tungsten diselenide (WSe2), growth conditions, including temperature and total pressure, have a strong impact on the crystalline size, shape, and nucleation. Synthesis at high pressure results in a significant reduction in nucleation density and increase crystallite size, however, it also leads to the formation of particulates on the sample surface, which were subsequently identified as W-rich WSe2 via cross-sectional transmission electron microscopy (TEM). The presence of such particles provided evidence that the selenium-to-tungsten (Se:W) ratio in vapor phase can be critical to achieving stoichiometric WSe2. As a result, the Se:W ratio is a critical factor in controlling defect formation in WSe2. This is evident in Figure 1a-d, which consists of atomic force microscopy scans of the WSe2 surface, as a function of the Se:W ratio as it is increased from 170 to 14,000. The figure clearly demonstrates that domain size increases significantly as the Se:W ratio is increased, which is accompanied by a decrease in the density of W-rich WSe2 particulates. Figure 1e is the cross-sectional HRTEM image of MOCVD grown multilayer WSe2 directly on sapphire. HRTEM reveals some disorder at the WSe2/sapphire interface suggesting a reaction during growth, which is in agreement with previous works suggesting that sapphire (Al2O3) is not stable. In addition to synthesis of WSe2, the synthesis of van der Waals heterostructures are also of increasing importance in the advancement of the field. Figure 2a shows the cross-sectional HRTEM image of CVD grown monolayer MoS2 on bi-layer graphene and SiC substrate. The MoS2 layer appears to be "blind" to thickness variations in the underlying graphene when there are no defects in the top layer of the graphene. However, when the top graphene layer is interrupted, the MoS2 will also be discontinued, as indicated in the Figure 2a. Figure 2b shows HAADF image of monolayer MoS2/WSe2 grown tri-layer graphene and SiC substrate. The contrast in a HAADF-STEM image is approximately proportional to Z2, where Z is the atomic number. Using HAADF, the MoS2 and WSe2 layers were clearly distinguished, as shown in Figure 2c. There are existing monolayer WSe2 with brightest contrast and monolayer MoS2 with second brightest contrast. Finally, we note that defects and edge states in the base 2D layer lead to low energy nucleation sites, and therefore multilayer growth of a top layer is highly probable at these regions, as shown in in Figure 2c. This work was supported by the Center for Low Energy Systems Technology (LEAST), one of six centers supported by the STARnet phase of the Focus Center Research Program (FCRP), a Semiconductor Research Corporation (SRC) program sponsored by MARCO and DARPA; and the Defense Threat Reduction Agency (DTRA) under contract HDTRA1-14-1-0037. Microsc. Microanal. 21 (Suppl 3), 2015 102 References 1. Kuilla, T. et al. Recent advances in graphene based polymer composites. Prog. Polym. Sci. 35, 1350�1375 (2010). 2. Cohen-Tanugi, D. & Grossman, J. C. Water desalination across nanoporous graphene. Nano Lett. 12, 3602�8 (2012). 3. Sarkar, D. et al. MoS field-effect transistor for next-generation label-free biosensors. ACS Nano 8, 3992�4003 (2014). 4. Machado, B. F. & Serp, P. Graphene-based materials for catalysis. Catal. Sci. Technol. 2, 54 (2012). 5. Xia, F., Wang, H., Xiao, D., Dubey, M. & Ramasubramaniam, A. Two-dimensional material nanophotonics. Nat. Photonics 8, 899�907 (2014). 6. Fiori, G. et al. Electronics based on two-dimensional materials. Nat. Nanotechnol. 9, 768�779 (2014). 7. Lin, Y.-C. et al. Atomically Thin Heterostructures based on Single-Layer Tungsten Diselenide and Graphene. Nano Lett. (2014). doi:10.1021/nl503144a 8. Eichfeld, S. M., Eichfeld, C. M., Lin, Y.-C., Hossain, L. & Robinson, J. A. Rapid, non-destructive evaluation of ultrathin WSe2 using spectroscopic ellipsometry. APL Mater. 2, 092508 (2014). 9. Lin, Y.-C. et al. Direct Synthesis of Van der Waal Solids. ACS Nano (2014). Figure 1: (a-d) AFM of a WSe2 surface as a function of increasing selenium (Se) overpressure that shows significant increase in domain size, as well as improved stoichiometry (reduction in particles). (e) TEM provides evidence that the sapphire substrate may be unstable during synthesis. Figure 2: (a) TEM reveals that growth of 2D materials on graphene are highly sensitive to defects in the graphene. (b) TEM also provides evidence that multiple junction structures of MoS2/WSe2 on graphene can be directly grown, where (c) defects and edges on WSe2 leads to multilayer MoS2.");sQ1[52]=new Array("../7337/0103.pdf","Characterization of Layer Thickness and Orientation of 2D WSe2/MoS2 Heterostructures using EDS, EBSD and AFM","","103 doi:10.1017/S1431927615001312 Paper No. 0052 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of Layer Thickness and Orientation of 2D WSe2/MoS2 Heterostructures using EDS, EBSD and AFM Christian Lang1, Matthew Hiscock1, Kim Larsen1, Jonathan Moffat2, and Ravi Sundaram3 1. 2. Oxford Instruments Nanoanalysis, High Wycombe UK. Asylum Research, High Wycombe UK. 3. Oxford Instruments Plasma Technology, Yatton UK. 2D transition metal chalcogenides enable exciting new applications in electronic devices and show great promise to replace traditional silicon technology as functional building blocks [1]. However, in order to realize this potential there is a range of fabrication and integration challenges that have to be overcome and suitable, non-destructive characterization techniques are needed. Due to their high resolution, electron optical characterization in scanning electron microscopes (SEMs) and atomic force microscope is ideally suited. We show how a full structural and compositional characterization can be obtained by combining EDS, EBSD and AFM analysis. The number of layers present in a 2D material is critical to its performance. As figures 1 and 2 show, we can obtain data of sufficiently high quality to non-destructively measure the number of layers in 2D MoS2 and WSe2 as well as from heterostructures containing both materials by processing EDS data obtained in the SEM. Figure 1 shows an SEM image of a flake of MoS2 with two regions, one with two layers of MoS2 and one with one layer as verified by step height measurements in the AFM and by Raman spectroscopy. EDS spectra acquired from the two different regions show a clear difference in the peak height of the overlapping Mo L-lines and S-K lines. The difference can be quantified by processing the data in a special software designed to calculate the thickness of thin films on substrates (AZtec LayerProbe) [2]. For the calculation, a density of 5.06 g/cm3 was assumes for MoS2. The resulting values shown in figure 1 correspond well to a theoretical interlayer distance of 0.65nm. In order to test whether this method is also suitable for heterostructures of 2D materials, we obtained measurements from a sample where a flake of MoS2 had been transferred onto a flake of WSe2. In the region of where the two flakes overlap, Raman spectroscopy showed that while there is only a single layer of WSe2 present, MoS2 occurs in one layer and two layers. Figure 2 indicates the different regions of interest on the sample. As the W-M line overlaps closely with the Si-K line, the Se-K line was used for the layer thickness measurement. The results in figure 2b show that both the WSe2 layer and the MoS2 layer thickness can be accurately determined. We also show that Kelvin Probe force measurements (KPFM) can be used to image the contrast between different layer thicknesses in both single layers and heterostructures (figure 2c). Further work is necessary to determine whether the work function measured by KPFM can be quantified. In order to add crystallographic data revealing misalignment between flakes, we can use EBSD. IPF maps of an area that contains several flakes of exfoliated MoS2 clearly indicate significant misalignment between some of the flakes (figure 3). This may aid the understanding of the exfoliation process which is still widely used to produce 2D materials for research purposes. Our results indicate the great potential of SEM and AFM for the characterization of devices based on 2D Microsc. Microanal. 21 (Suppl 3), 2015 104 materials and indicate avenues of further work to establish them as means for failure analysis and production quality control. References: [1] S.Z. Butler et al., ACS Nano 7 (2013), p. 2898. [2] C. Lang et al., Microscopy and Microanalysis 19 (2013), p. 1872. Figure 1. A flake of MoS2 on an SiO2 substrate with EDS spectra and resulting layer thicknesses. Figure 2. (a) optical micrograph of a MoS2/WSe2 heterostructure. (b) EDS maps at 4kV of the overlap regions indicated by dotted lines and (c) KPFM images of the overlap regions. Figure 3. (a) Electron image and (b) IPF x map and (c) IPF z map of MoS2 flakes on SiO2 indicating only in-plane rotational misalignment of the flakes.");sQ1[53]=new Array("../7337/0105.pdf","Interfaces in Two-Dimensional Heterostructures of Transition Metal Dichalcogenides","","105 doi:10.1017/S1431927615001324 Paper No. 0053 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Interfaces in Two-Dimensional Heterostructures of Transition Metal Dichalcogenides Wu Zhou1, Junhao Lin2, 1, Yongji Gong3, 4, Xingli Wang5, Beng Kang Tay5, Jun Lou4, Zheng Liu5, Sokrates T. Pantelides2, 1, Pulickel M. Ajayan3,4 1. Materials Science & Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA 2. Department of Physics and Astronomy, Vanderbilt University, Nashville, TN 37235, USA 3. Department of Chemistry, Rice University, Houston, TX 77005, USA 4. Department of Materials Science & NanoEngineering, Rice University, Houston, TX 77005, USA 5. School of Materials Science and Engineering, School of Electrical and Electronic Engineering Nanyang Technological University, 639798, Singapore Two-dimensional (2D) transition-metal dichalcogenides (TMDs) are promising candidates for flexible nanoelectronics, with exceptional optical and electrical properties at monolayer thickness. Monolayers of different TMDs can be further combined to create van der Waals heterostructures, where multiple 2D layers are stacked vertically layer-by-layer, or stitched seamlessly in plane to form lateral heterojunctions. The coupling between the different 2D components provides unique opportunities for bandgap engineering and can create very unusual properties at the interface [1-4]. Revealing the atomic structure, including the stacking orientation, stacking order, and chemical inter-diffusion, is therefore important for understanding the novel properties generated by the heterostructure interfaces. Recently, we have demonstrated a simple one-step vapor phase growth of high quality heterostructures of WS2 and MoS2 [1]. High temperature growth yields predominantly vertically stacked bilayer heterostructures, while low temperature growth creates mostly lateral heterostructures of WS2 and MoS2 within the same monolayer. The atomic structure and electronic properties of the heterostructure interfaces are studied by aberration-corrected scanning transmission electron microscopy (STEM) annular dark field (ADF) imaging, electron energy-loss spectroscopy (EELS) at low voltage, and density functional calculations, STEM-ADF imaging reveals that the vertical heterostructures were obtained with WS2 epitaxially grown on top of the MoS2 monolayer, following the preferred 2H stacking (Figure 1). A small amount (~ 3%) of W substitution in the MoS2 layer and Mo substitution in the WS2 layer was observed in the sample, which should only have minimum effect on the properties of the MoS2 and WS2 monolayers at such low concentration. Photoluminescence (PL) analysis shows that the MoS2 and WS2 layers in the bilayer heterostructure, on one hand, behave as individual monolayers, and, on the other hand, generate a new direct band gap of WS2/MoS2 heterostructure via interlayer coupling owing to the clean interface. Atomically sharp interfaces were frequently observed in the lateral heterojunctions, with seamless connection and abrupt transition between the MoS2 and WS2 lattice within a single atomic row. Most of the abrupt lateral interfaces were achieved by lateral epitaxial growth of WS2 on fresh MoS2 edges along the zigzag direction, and sharp armchair interfaces were only occasionally observed. Lateral interfaces with large chemical inter-diffusion over a width of a Microsc. Microanal. 21 (Suppl 3), 2015 106 few hundred nanometers were also observed, presumably due to local fluctuations in the growth conditions. The different degrees of chemical inter-diffusion are most likely responsible for the observed inhomogeneous PL enhancement along the lateral interfaces. Besides the WS2-MoS2 system, results from WSe2/MoSe2 heterostructures will also be discussed, which provides insights into the growth mechanism and guidance for the growth of superlattice structures [5]. References: [1] Y Gong et al, Nature Materials 13 (2014), P. 1135-1142. [2] C Huang et al, Nature Materials 13 (2014), P. 1096-1101. [3] X Duan et al, Nature Nanotechnology 9 (2014), P. 1024-1030. [4] AK Geim & IV Grigorieva, Nature 499 (2013) P. 419-425. [5] This research was supported in part a Wigner Fellowship of Oak Ridge National Laboratory (WZ), by the Office of Science, Basic Energy Science, Materials Sciences and Engineering Division, U.S. DOE (WZ), by U.S. DOE grant DE-FG02-09ER46554 (JL, STP), and through a user project supported by ORNL's Center for Nanophase Materials Sciences (CNMS), which is sponsored by the Scientific User Facilities Division, Office of Basic Energy Sciences, U.S. DOE. YG and PMA were supported by Army Research Office MURI grant W911NF-11-1-0362. Figure 1. STEM-ADF imaging of vertical heterostructure of WS2/MoS2 at different magnifications. Figure A is shown in color scale where monolayer MoS2 is in blue, monolayer WS2 in green and WS2/MoS2 bilayer in orange. (D) is the structure model illustrating the 2H stacking [1]. Figure 2. STEM-ADF imaging of atomically sharp lateral interfaces between WS2 and MoS2 along the zigzag (A, B) and armchair (C) directions. Scale bars: 0.5 nm [1].");sQ1[54]=new Array("../7337/0107.pdf","Cross sectional STEM imaging and analysis of multilayered two dimensional crystal heterostructure devices","","107 doi:10.1017/S1431927615001336 Paper No. 0054 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Cross sectional STEM imaging and analysis of multilayered two dimensional crystal heterostructure devices Sarah J. Haigh1, Aidan P. Rooney1, Eric Prestat1, Fred. Withers2, O. Del Pozo Zamudio3, Artem Mishchenko2, Ali. Gholinia1, K. Watanabe4, T. Taniguchi4, A. I. Tartakovskii3, Andre K. Geim5, Konstantin. S. Novoselov2, 1. 2. School of Materials, University of Manchester, Manchester, M13 9PL, UK School of Physics and Astronomy, University of Manchester, Manchester, M13 9PL, UK 3. Department of Physics and Astronomy, University of Sheffield, Sheffield, S3 7RH, UK 4. National Institute for Materials Science, 1-1 Namiki, Tsukuba 305-0044, Japan. 5 Manchester Centre for Mesoscience and Nanotechnology, University of Manchester, Manchester, M13 9PL, UK The number of two dimensional crystals successfully isolated has expanded rapidly in recent years, providing a wide variety of interesting electronic properties.[1] By combining different two dimensional crystals within a Van der Waals heterostructure stack it is possible to produce devices with bespoke electronic bandstructure.[1] This is fast establishing a new generation of optical and electronic devices with advanced functionality.[2-6] Our Van der Waals heterostructures have been synthesized by sequential layering of different twodimensional crystals to form a stack on a silicon wafer substrate.[1] The crystals have been produced by mechanical exfoliation and transferred by deposition and subsequent removal of the polymeric support layers. Despite the inherently `dirty' nature of this processing we have found that the heterostructure stacks form large areas of pristine synthetic crystal, free from hydrocarbon contaminants at the atomic scale.[7] This has exciting implications for device performance and design. For example, the device shown in Figure 1 is a light-emitting diode (LED) composed of multiple quantum wells separated by h-BN tunneling layers.[8] Graphene electrodes above and below the stack allow charge carrier pairs to tunnel through the insulating h-BN layers into the MoS2 quantum wells, where they form excitons before undergoing radiative recombination, measured as light emission. Optical and electronic characterisation has shown electroluminescence quantum efficiencies of up to 5% are possible for these multiple quantum well devices, comparable with modern organic LEDs. Device performance depends critically on the width of the h-BN tunneling barriers as well as the thickness and uniformity of MoS2 layers within the device. These parameters are difficult or impossible to characterize within such a complex multilayer device. The only method by which this atomic scale structural detail can be obtained is cross sectional scanning transmission electron microscope (STEM) imaging and analysis. After the device has undergone electronic testing, a site specific cross section can be extracted from the active area of the device using the focused ion beam `lift out' method. STEM imaging of the device cross section reveals that it contains four atomically flat MoS2 monolayers. The composition of sequential device layers can be identified in elemental maps extracted from energy dispersive x-ray spectrum image data (Figure 1 c). The number of h-BN layers separating each MoS2 monolayer can be determined and the location of the top graphene electrode identified (indicated by dashed white line in Figure 1 c iii). Microsc. Microanal. 21 (Suppl 3), 2015 108 We have used a similar cross sectional STEM imaging and analysis approach to investigate a range of graphene - transition metal dichalcogenides (TMDCs) heterostructures at the atomic scale. We demonstrate that this technique can provide unprecedented insights into the structural nature and failure mechanisms of these complex devices.[4, 6-8] References [1] Geim, A.K. and I.V. Grigorieva,. Nature, 2013. 499(7459): p. 419-425. [2] Britnell, L., et al. Science, 2012. 335(6071): p. 947-950. [3] Britnell, L., et al. Science, 2013. 340(6138): p. 1311-1314. [4] Georgiou, T., et al. Nat Nano, 2013. 8(2): p. 100-103. [5] Yu, W.J., et al. Nat Nano, 2013. 8(12): p. 952-958. [6]Withers, F. et al. Nano Letters, 2014. 14(7): p. 3987-3992. [7]Haigh, S.J. et al. Nat Mater, 2012. 11(9): p. 764-767. [8] Withers, F. et al. Nat Mater, 2015. advance online publication. Figure 1 Cross sectional Imaging of MoS2 multilayer quantum well. (a) Bright field and (b) high angle annular dark field STEM images of the four layer MoS2 heterostructure cross-section. Boron nitride lattice fringes are clearly visible in both images, as are the position of the MoS2 monolayers. (c i-iii) Show elemental maps for Mo, S and N extracted from energy dispersive xray (EDX) spectrum image data. The top graphene electrode can be seen as a deficiency in the nitrogen EDX map (indicated by dashed white line in (ciii)). (d) and (e) show bright field and high angle annular dark field images at lower magnification. All scale bars are 4 nm.");sQ1[55]=new Array("../7337/0109.pdf","Measuring the Atomic and Electronic Structure of Black Phosphorus with STEM","","109 doi:10.1017/S1431927615001348 Paper No. 0055 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Measuring the Atomic and Electronic Structure of Black Phosphorus with STEM Ryan J. Wu1, Mehmet Topsakal1, Matt C. Robbins2, Nazila Haratipour2, Jong Seok. Jeong1, Renata M. M. Wentzcovich1, Steven J. Koester2, K. Andre Mkhoyan1 1 2 Department of Chemical Engineering and Material Science Department of Electrical and Computer Engineering University of Minnesota, Minneapolis, MN, USA Black phosphorus, a layered two-dimensional crystal with tunable electronic properties and high hole mobility, is quickly emerging as a promising candidate for future electronic and photonic devices [1]. Although theoretical studies using ab initio calculations have tried to predict its atomic and electronic structure [2, 3], uncertainty in its fundamental structural properties due to a lack of clear experimental evidence continues to stymie our full understanding and application of this novel material. In this work, aberration-corrected scanning transmission electron microscopy (STEM) is used to record annular dark-field (ADF) images of few-layer black phosphorus. Directly interpretable atomicresolution ADF-STEM images captured at the [001], [101], and [100] zone axes, as shown in Figure 1, provides a three-dimensional view of this layered material which allows all three lattice parameters to be measured. The ADF-STEM images also unambiguously identified its stacking order which differs from other possible arrangements discussed in literature [3]. Furthermore, STEM monochromated electron energy-loss spectroscopy (STEM-EELS) is used to measure the conduction band density of states (DOS) of black phosphorus. The core-loss EELS measured P L3-edge agrees well with density functional theory (DFT)-based calculations performed for the experimentally determined black phosphorus crystal, as shown in Figure 2a. Simultaneously collected low-loss EEL spectrum (Figure 2b) also shows energies of bulk and surface plasmons and changes with the number of layers. Finally, the effects of oxidation on both the atomic and electronic structure of black phosphorus are analyzed to explain observed device degradation. STEM-energy dispersive X-ray (STEM-EDX) maps show that black phosphorus transforms into amorphous (H3)PO3 during oxidation which may ultimately be responsible for the degradation of black phosphorus-based devices exposed to atmosphere over time. References: [1] L. Li et al., Nat Nano 9, (2014), p. 372-377 [2] J. Qiao et al., Nat. Commun 5, (2014).4475 [3] J. Dai and X.C. Zeng, J. Phys. Chem. Lett. 5, (2014). p.1289-1293 [4] This work was supported in part by C-SPIN, one of the six centers of STARnet, a Semiconductor Research Corporation program, sponsored by MARCO and DARPA; by the NSF award DMR-0819885; and by the Defense Threat Reduction Agency award HDTRA1-14-1-0042. Computational resources were partly provided by Blue Waters sustained-petascale computing project, supported by the NSF awards OCI-180725070 and ACI-1238993 and the state of Illinois. STEM analysis was carried out in the Characterization Facility of the University of Minnesota, which receives partial support from NSF through the MRSEC program. Microsc. Microanal. 21 (Suppl 3), 2015 110 Figure 1: Atomic-resolution ADF-STEM images of black phosphorus. a, ADF-STEM image of black phosphorus viewed along the [001] crystallographic direction, or top-down view. b, ADF-STEM image viewed along the [101] direction or 17� tilted off the [001] zone axis. c, ADF-STEM image captured at an edge of a black phosphorus flake showing multiple layers stacked together, or along [100] direction. Images in b, c and e have overlaid ball-stick atomic models to accentuate atomic columns. Figure 2. Electronic structure of black phosphorus. a, Monochromated EELS P L3 edge recorded with 0.25 eV energy resolution compared with calculated 3s +3d partial DOS. DOS data is broadened with a 0.25eV FWHM Gaussian function for better comparison. b, Low-loss EEL spectra from 4-layer and bulk black. Peaks I, II and III are mainly surface plasmon modes while peak IV is the bulk plasmon.");sQ1[56]=new Array("../7337/0111.pdf","Using In-Situ TEM to Characterize the Microstructure Evolution of Metallic Systems under External Solicitation","","111 doi:10.1017/S143192761500135X Paper No. 0056 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Using In-Situ TEM to Characterize the Microstructure Evolution of Metallic Systems under External Solicitation D. Kaoumi1, T. Gautier1, J. Adamson1, M. Kirk2 1. University of South Carolina, 300 Main St., SC, 29208, USA, djamelkaoumi@gmail.com 2. Argonne National Laboratory, Bldg 212, IL, 60439 USA, kirk@anl.gov Studying materials under external stimulus such as irradiation and/or mechanical stress can be difficult because of the lack of kinetics information, since usually samples are examined ex situ (e.g. after irradiation or after mechanical testing) so that only discrete snapshots of the process are available. Given the dynamic nature of the phenomena, direct in situ observation is often necessary to better understand the mechanisms, kinetics and driving forces of the processes involved. For this matter, using in situ Transmission Electron Microscopy (TEM) can be of great help[1]. Indeed, the spatial resolution of the TEM makes it an invaluable tool in which one can continuously track the real-time response of the microstructure to external stimuli, which can help discover and quantify the fundamental rate-limiting microscopic processes and mechanisms governing the macroscopic properties. In this presentation, two examples will be given which show how the technique can be used for nuclear engineering applications. (i) In-situ straining experiments in the TEM is applied to investigate deformation mechanisms in Ni-based alloys (Inconel 617 and Haynes 230) which are candidate materials for the heat exchanger in the GEN-IV Very High Temperature nuclear Reactor. In addition to showing dislocation dynamics under tensile strain, it also allows to follow crack propagation as it proceeds in the material. (ii) In-situ Ion-irradiation in the TEM has proven a very good tool for studying the basic mechanisms of radiation damage formation and evolution as a function of dose, dose rate, temperature and ion type. In-situ straining in a TEM: Foils of Ni-based Inconel 617 were deformed in situ in the TEM at 298K and 573K. The dynamic observations of the deformation processes are described especially in terms of dislocation motion and interaction, (e.g. dislocation (double) cross-slip and annihilation). The alloy presents dislocation slip bands in its as-received state (un-aged sheet), and several additional slip bands develop during the straining experiments. The movement of the dislocation through the slip bands, at the interfaces, and about the precipitates is observed in-situ as illustrated in Figure 1. The arrangement of moving dislocations, especially the propagation of pile-ups of bowed dislocations, is observed in detail. The dislocations moving in the slip bands are not the preexisting dislocations but originate from grain boundaries and other stress concentrators. Also, dynamic observations of crack initiation and growth mechanisms are obtained in the same experiments. Intragranular crack initiation and propagation is also observed in the same samples. The crack propagation in the un-aged alloy appears to occur mostly in a zigzag manner. Microcracks form at stress concentration sites in front of the major crack and then the major crack advances by microcrack linkage. Similar modes of fracture are observed at 300K, 573K[2]. The main observations will be reported and the possible mechanisms discussed. In-situ ion irradiation in a TEM: Ion-irradiation is a useful way to reach high level of radiation damage doses in a reasonable amount of time with a damage structure that can be similar to that achieved under neutron irradiation without the complications of radioactivity. When combined with Transmission Electron Microscopy it becomes a powerful tool to gain kinetics information on Microsc. Microanal. 21 (Suppl 3), 2015 112 irradiation induced phenomena because the microstructure evolution of the irradiated material can be followed in-situ as the damage proceeds. Using this technique the microstructure evolution under irradiation was studied in two model F/M steels (9Cr and 12Cr ) irradiated with 1 MeV Kr ions insitu at temperatures between 20K and 573K to doses as high as 10 dpa. During the early stages of irradiation of the two F/M steels, defect clusters appear to be rather uniformly distributed within grains, and the saturation density is quickly reached between 1.0 and 2.0 dpa. However, at doses as low as 3 dpa, self-ordered alignments of defect clusters are found in some grains. The regularly ordered arrays of small loops with spacing about 30-50 nm are observed in both F/M steels along <110> directions. Once the aligned structure is created, it is stable under further irradiation. The possible mechanisms for the "self-organization" of the clusters were investigated. These structures are thought to result from elastic interactions between defect clusters in the foil. The fact that such defect alignment is not observed at higher temperatures suggests that the relatively high density of defect clusters (at lower temperatures) and the resultant internal strains may be the main reason for the development of the aligned structure. The preferred crystallographic orientation of defect arrays may be driven by the minimization of elastic interaction energy between defect clusters [3]. The defect self-ordering process, its temperature dependence and the possible mechanisms are discussed. FIG. 1: TEM observations of Inconel 617 strained in-situ at 298K showing dislocation pile-ups and interaction with precipitate. References: [1] D. Kaoumi, A. Motta, R. Birtcher,Proceedings of the Workshop on the Use of In-Situ TEMIon Accelerator Techniques in the Study of Radiation Damage in Solids, The University of Salford, UK, 2009. [2] D. Kaoumi T. Gautier, M. Kirk, 19th International Symposium on Plasticity, 2013, Nassau, Bahamas. [3] D. Kaoumi, J. Adamson, Journal of Nuclear Materials, 448: p 233�238, 2014.");sQ1[57]=new Array("../7337/0113.pdf","In-Situ TEM He+ Implantation and Thermal Aging of Nanocrystalline Fe","","113 doi:10.1017/S1431927615001361 Paper No. 0057 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In-Situ TEM He+ Implantation and Thermal Aging of Nanocrystalline Fe B. Muntifering1,2, A. Dunn2,3 R. Dingreville2, J. Qu 1, and K. Hattar2 1. 2. Department of Civil and Environmental Engineering, Northwestern University, Evanston, IL, USA Sandia National Laboratory, Albuquerque, NM, USA 3. Woodruff School of Mechanical Engineering, Georgia Institute of Technology, Atlanta, Ga, USA A key aspect in predictively modeling the response of materials exposed to many radiation environments is understanding the role of light transmutation products. He in particular can result in the swelling and precipitation of bubbles, both of which can substantially deteriorate the mechanical properties [1]. In this study, in situ TEM characterization of nanocrystalline Fe samples implanted with 10 keV He+ is performed to understand and quantify the mechanisms underlying He diffusion and cavity nucleation under a wide temperature range Nanocrystalline free-standing Fe thin films were produced by pulse laser deposition and annealed in situ at 550�C in order to stabilize the grain structure for the subsequent experiments. Two types of in situ experiments were performed to study defect evolution and cavity formation: (1) He implantation at room temperature followed by annealing to 600�C, and (2) He implantation at elevated temperatures up to 600�C. In situ He implantations were carried out utilizing a JEOL 2100 TEM and a 10 kV Colutron ion accelerator that are part of the In situ Ion Irradiation TEM facility at Sandia [2]. The implantation occurred in a hummingbird heating stage tilted 40� towards the He beam operated at an average flux of approximately 1014 He+/cm2s to a total concentration of ~4% He at the end of range. Figures 1(a-d) illustrate bright field TEM images of He implanted Fe at (a) room temperature, (b) 200�C, (c) 400�C and (d) 600�C. It was observed that He implantation at room temperature and 200�C resulted in the formation of dislocation loops with a maximum size of approximately 3 nm. Implantation at 400�C also resulted in the formation of dislocation loops, but with much larger sizes (16 nm max). In contrast, at 600�C, first where He implantation occurred at room temperature followed by annealing to 600�C, and second where He implantation occurred directly at 600�C, Fresnel imaging revealed the presence of nanometer sized voids in both cases. Figure 2(a) shows an under-focus image of the roomtemperature implanted sample after annealing to 600�C where the cavities appear to be evenly distributed through the grains. In the under-focus image of the sample implanted directly at 600�C presented in Figure 2(b), the observed behavior is drastically different since cavities are seen only along grain boundaries. The experimental observations suggest different mechanisms are active during He+ implantation under sequential versus simultaneous He+ implantation and annealing. These results will be compared to other existing experimental observations and defect evolution models in order to provide insights into the various mechanisms contributing to the different behaviors. References: [1] G.S. Was in "Fundamentals of Radiation Materials Science", (Springer). [2] K. Hattar, et al. Nuclear Instruments and Methods in Physics Research B 338 (2014). [3] This work is funded by NEET through DE-NE0000678. Sandia National Laboratories is a multiprogram laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000. Microsc. Microanal. 21 (Suppl 3), 2015 114 Figure 1. He+ implanted Fe implanted at (a) room temperature (b) 200�C (c) 400�C (d) 600�C. Figure 2. Under-focus images of cavities at 600�C in samples implanted at (a) room-temperature and post-implantation annealed (b) 600�C.");sQ1[58]=new Array("../7337/0115.pdf","A New Method for Studying He Damage in Materials Demonstrated on Nanotwinned Cu Nanopillars","","115 doi:10.1017/S1431927615001373 Paper No. 0058 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A New Method for Studying He Damage in Materials Demonstrated on Nanotwinned Cu Nanopillars Zhang-Jie Wang 1, 2, Frances Allen 3, Zhi-Wei Shan 2 and Peter Hosemann1 1. 2. Department of Nuclear Engineering, University of California, Berkeley, California, 94720, USA Center for Advancing Materials Performance from the Nanoscale (CAMP-Nano) & Hysitron Applied Research Center in China (HARCC), State Key Laboratory for Mechanical Behavior of Materials, Xi'an Jiaotong University, Xi'an, 710049, China 3. Department of Materials Science and Engineering, University of California, Berkeley, California 94720, USA Ion implantation has been used for decades to investigate the response of materials to radiation damage. While displacement damage is caused due to an incoming particle, in high neutron energy environments gases are also produced due to transmutation or fission events. In particular fusion, fast reactor and spallation sources suffer from high He/dpa (displacements per atom) ratios. Understanding the effect of He in materials is a key aspect in these applications. It is known that interfaces are beneficial for radiation damage due to the fact that interfaces act as defect sinks. In the past He-denuded zones around grain boundaries have often been observed which prove that He can be managed by offering defect sinks [1]. Only a few studies have been performed investigating the effect of twin interfaces on He management [2]. The studies performed are rather limited in exploring different doses due to the fact that implanting the same grain with different doses and subsequent characterization of the resulting defects are challenging and rather time consuming, thus limiting systematic studies. Moreover, in order to obtain a direct comparison ideally one implants the same sample and same grain to different doses. Recently the Zeiss ORION NanoFab instrument was released which allows He and Ne ion beams in combination with a Ga ion source to quickly and efficiently manufacture nanostructures and direct He implantation [3]. In this work we utilize the combined Ga-He beam system to increase sample throughput and manufacture nanopillars with subsequent He implantation in a fast and efficient manner. Using this novel method one can manufacture nanopillars and implant these targeting the exact same grain in one session, thereby allowing for better and more accurate comparison and effect evaluation due to better controlled separate effect testing. In this work several nanopillar samples were manufactured in one grain of a nanotwinned Cu sample. Each pillar was implanted to a different dose using 25keV He ions generated by the ORION NanoFab instrument. The maximum dose was 1019 He+/cm2. Each pillar was then tested using a JEOL 2010 TEM equipped with a Hysitron PI95 nanomechanical testing system. Quantitative stress-strain data were collected in situ leading to a fundamental understanding of the deformation mechanism of the implanted pillars. It was found that the approach of directly implanting He into nanotwinned Cu nanopillars can easily be conducted using the ORION NanoFab. Figures 1a and b show two nanotwinned pillars cut from one grain using a Ga ion source and then implanted with different doses of He. In the case of samples implanted with a high dose (e.g. 1018 He+/cm2) He bubbles were formed in the material, as shown in the defocused bright field image in Figure 1c. Micro compression testing leads to the stress-strain curve shown in Figure 2. The strength of the nanopillars significantly depends on the dose of He. Microsc. Microanal. 21 (Suppl 3), 2015 116 Taking advantage of the accurate and controllable micro area He implantation technique described, different doses of He ions were implanted into Cu nanotwinned pillars manufactured in one grain. The influence of various He doses on the mechanical properties of nanotwinned Cu was systematically studied by in situ mechanical testing. This novel technique makes it feasible to evaluate He ion damage and its effect on small volume materials. References: [1] M. Zhernenkov et al., Trapping of implanted He at Cu/Nb interfaces measured by neutron reflectometry. Appl. Phys. Lett. 98, (2011), 241913. [2] K. Yu et al., Removal of stacking-fault tetrahedra by twin boundaries in nanotwinned metals. Nature Communications 4, (2013), 1377. [3] V. Veligura et al., Digging gold: keV He+ion interaction with Au, Beilstein J. Nanotechnol.,4, (2013), 453�460. [4] The Authors thank The China Scholarship Council for providing funding for the visiting scientist. This publication was made possible in part by NSF/DMR MRI DMR-1338139. This research is performed using funding received from the DOE Office of Nuclear Energy's Nuclear Energy University Programs and the Keck Foundation. The nanotwinned pillars were prepared and implanted at the Biomolecular Nanotechnology Center at the University of California, Berkeley, and the TEM in situ analysis was performed at The Molecular Foundry, which is supported by the Office of Science, Office of Basic Energy Sciences, of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231. Figure 1 nanotwinned pillars with different doses of He. Figure 2 Engineering stress versus engineering strain curves of nanotwinned Cu nanopillars implanted with two different doses of He.");sQ1[59]=new Array("../7337/0117.pdf","In-situ electron microscope observations and analysis of radiation damage in tungsten","","117 doi:10.1017/S1431927615001385 Paper No. 0059 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In-situ electron microscope observations and analysis of radiation damage in tungsten Xiaoou Yi1, 2, Michael L Jenkins3, Marquis A Kirk4 and Steven G Roberts1, 2 1. 2. Department of Materials, University of Oxford, Parks Road, OX1 3PH, Oxford, U.K. CCFE, Culham Science Centre, OX14 3DB, Abingdon, U.K. 3. Trinity College, University of Oxford, Broad Street, OX1 3BH, Oxford, U.K. 4. Nuclear Engineering Division, Argonne National Laboratory, IL 60439, Argonne, U.S.A. Tungsten is a prime candidate for building divertor components in fusion reactors. During its service life, these components may undergo up to 30-40 dpa of displacement damage per year, originated from the collision cascades of fusion neutrons. In this work, we investigated the production and evolution of radiation damage in tungsten with in-situ observations and analysis of 150 keV W+ ion irradiations, so as to mimic the effects of the average primary recoil energy of 14 MeV fusion neutrons [1]. TEM foils of pure tungsten (typically > 99.996 wt%, Plansee) were annealed (1673K, 20 h) and prepared from jet-electropolishing in a 0.5 wt.% NaOH aqueous solution. The in-situ irradiations were performed on the IVEM-Tandem facility at Argonne National Laboratory, with a focused beam of W+ ions directed to the specimen surface, at a high incident angle of ~75�. The irradiation conditions covered a wide temperature range from 30 K to 1073 K and a dose range from 1016 to 1018 W+/m2 (0.01 ~ 1.0 dpa) at a constant rate of ~ 6.25�1014 W+/m2s. We recorded the defect dynamics at 15 frames/s, and performed defect characterizations and analyses (population; size distribution; geometry; nature) following the methods described in Jenkins et al [2]. This comprehensive study of damage production and evolution in tungsten has led to several new findings. We have discovered that at doses 0.01 dpa, the first observable defects nucleated in tungsten were vacancy loops, predominantly of b = � <111>, and were formed within individual collision cascades. With the increase of temperature from stage I (30 K) to stage IV (1073 K), loops with b = � <111> gained increasing predominance over those with b = <100> and the analysis of defect size versus the frequency of occurrence suggested the engagement of strong elastic interactions among all radiationinduced defects in the cascade (more details in Mason et al) [3]. At doses beyond the overlap of cascades (> 0.01 dpa), radiation damage in tungsten evolved through the 1D migration of defect clusters, the elastic interactions and (typically) non-conservative reaction among these defect clusters, rendered by irradiation temperature and dose. Notably, a transition from random distribution to spatial ordering of loops has been observed in the z = <001> grains at doses > 0.4 dpa and T 773 K, as partially illustrated in Figure 1. We've also noticed that non z = <001> orientations may considerably lower the threshold of this ordering phenomenon. The highlights of damage analysis in tungsten are summarized in Figure 2. The rates of defect accumulation in Figure 2a indicated a rapid saturation for T < 773 K, whereas for T 773 K, the first signs of saturation did not occur until > 0.4 dpa. Post-irradiation analysis of defect size distributions (Figure 2b) and defect geometry (Figure 2c) at 1.0 dpa suggested that 773 K (stage III, migration of monovacancies) is a characteristic temperature in the radiation damage evolution of tungsten. Together with evidence found in defect dynamic behavior and the results of defect nature determination, we have found that temperature and dose tend to drive the damage microstructure in tungsten towards an increased proportion of interstitial � <111> loops, an increased degree of spatial ordering among them and facilitate their size increase through coalescence reactions. Microsc. Microanal. 21 (Suppl 3), 2015 118 References: [1] A. E. Sand, S. L. Dudarev and K. Nordlund, EPL. 103 (2013), 46003. [2] M. L. Jenkins in "Characterization of Radiation Damage by Transmission Electron Microscopy", 1st ed., (IOP Publishing, Bristol & Philadelphia) 27-69, 74-89, 110-128. [3] D. R. Mason, X. Yi and M. A. Kirk, J. Phys: Condens. Matter. 26 (2014), 375701. [4] The authors acknowledge funding from the EPSRC, UK (Grant No. EP/H018921/1) and the DOE Office of Nuclear Energy, USA (Contract No. DE-AC02-06CH11357, UChicago Argonne, LLC) for the in-situ electron microscopy work accomplished at ANL (IVEM-Tandem Facility). XY thanks Mr. Pete Baldo and Mr. Edward Ryan for their generous help with the irradiations, and CCFE for supporting a Junior Research Fellowship via St Edmund Hall, University of Oxford. Figure 1. Temperature and dose dependence of damage microstructures in W irradiated with 150 keV W+ at close to z = [001] orientations. Micrographs were recorded under weak-beam dark-field conditions: g = 200, 3-4g. a) b) c) Figure 2. Highlights of radiation damage analysis in W: a) the evolution of defect population over a dose range of 0.1 ~ 1.0 dpa as a function of temperature; b) defect size distributions and c) defect geometries at a dose of 1.0 dpa as a function of temperature.");sQ1[60]=new Array("../7337/0119.pdf","Convergent-Beam Electron Diffraction Study of Local Structural Fluctuations in Perovskite-Type Ferroelectrics","","119 doi:10.1017/S1431927615001397 Paper No. 0060 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Convergent-Beam Electron Diffraction Study of Local Structural Fluctuations in Perovskite-Type Ferroelectrics Kenji Tsuda Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, Sendai, Japan Perovskite-type ferroelectrics have been attracting special attention because of their intriguing physical properties as well as industrial importance. BaTiO3, a typical example of perovskite-type ferroelectrics, undergoes successive phase transformations from the cubic paraelectric phase to three ferroelectric phases: tetragonal, orthorhombic and rhombohedral ones. Coexistence of the displacive and orderdisorder characters in the phase transformations of BaTiO3 was pointed out from many experiments and theories [ex. 1, 2]. However, local structures related to the order-disorder character were observed neither in crystal structure analyses using neutron and X-ray diffraction nor by TEM observations. In this study, we applied the convergent-beam electron diffraction (CBED) method, which is a most powerful technique for local symmetry determination [3, 4], to examine nanometer-scale local structures of perovskite-type ferroelectrics. Using CBED, we observed that the orthorhombic and tetragonal phases of BaTiO3 have rhombohedral nanostructures [5]. We also found that the symmetry of the orthorhombic phase is formed as the average of two rhombohedral variants with different polarizations, and that of the tetragonal phase is formed as the average of four rhombohedral variants. These indicate an order-disorder character in their phase transformations. Similar rhombohedral local structures were observed in the ferroelectric orthorhombic phase of KNbO3 [6], while the local structure of the ferroelectric tetragonal phase of PbTiO3 was revealed to have the same tetragonal symmetry as the average structure [7]. To examine distributions of the nanostructures and local polarizations, we proposed a combined use of STEM and CBED methods (STEM-CBED method) [8]. Figure 1 shows a schematic diagram of the STEM-CBED method. CBED patterns are acquired pixel-by-pixel by scanning the convergent-beam electron probe with a sub-nanometer step. Accurate control of the incident beam direction and selecting specimen areas without local bending are necessary for correctly detecting symmetry breaking of CBED patterns. A similar method using cross-correlation coefficients was proposed by Kim et al.[9]. Using the STEM-CBED method, two-dimensional distributions of the rhombohedral nanostructures, or nanoscale fluctuations of polarization clusters, were successfully visualized in the tetragonal phase of BaTiO3 [8]. Figure 2 shows a STEM-CBED map and an energy-filtered CBED pattern of the orthorhombic phase of KNbO3 taken at the [001]PC incidence, where the scan step was set to 0.5 nm. From the acquired CBED patterns, intensity differences between the 1 -1 0 PC and -1 -1 0 PC reflections were mapped. Since these reflections are symmetrically equivalent in the orthorhombic phase, the map indicates breakdown of the orthorhombic symmetry and their nanometer-scale fluctuations. (The above indices of the direction and reflections are expressed by the pseudocubic (PC) axes.) The study was performed in collaboration with Emer. Prof. M. Tanaka and Mr. R. Sano (IMRAM, Tohoku University), and Mr. A. Yasuhara (JEOL Ltd.), and was supported by JSPS KAKENHI Grant Numbers 25287068 and 256630272. Prof. M. Terauchi is thanked for his support to this work. Microsc. Microanal. 21 (Suppl 3), 2015 120 References: [1] R. Pirc and R. Blinc, Phys. Rev. B 70 (2004) 134107. [2] G. V�lkel and K. A. M�ller, Phys. Rev. B 76 (2007) 094105. [3] M. Tanaka and K. Tsuda, J. Electron Microsc. 60(Suppl. 1) (2011) S245. [4] Y. Shi et al., Nature Mat., 12 (2013) p. 1024. [5] K. Tsuda, R. Sano and M. Tanaka, Phys. Rev. B 86 (2012) 214106. [6] K. Tsuda, R. Sano and M. Tanaka, Appl. Phys. Lett. 102 (2013) 051913. [7] K. Tsuda and M. Tanaka, Appl. Phys. Express 6 (2013) 101501. [8] K. Tsuda, A. Yasuhara and M. Tanaka, Appl. Phys. Lett. 102 (2013) 051913. [9] K. H. Kim, D. A. Payne and J. M. Zuo, J. Appl. Cryst. 46 (2013) p. 1331. Figure 1. Schematic diagram of the STEM-CBED method. Figure 2. (a) STEM-CBED map of the orthorhombic phase of KNbO3 and (b) energy-filtered CBED pattern taken at the [001]PC incidence (PC: pseudocubic). The STEM-CBED map shows the intensity difference between the 1 -1 0 PC and -1 -1 0 PC reflections, (I1 -1 0 PC � I-1-1 0 PC) / I-1-1 0 PC, which is colored with the attached color bar for a range of -60% to 60%.");sQ1[61]=new Array("../7337/0121.pdf","Chemical Bonding Effects in HAADF-STEM Imaging of Light-Element Ceramics","","121 doi:10.1017/S1431927615001403 Paper No. 0061 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Chemical Bonding Effects in HAADF-STEM Imaging of Light-Element Ceramics Michael Odlyzko,1 Jacob Held,1 K. Andre Mkhoyan1 1 Chemical Engineering and Materials Science, University of Minnesota, Minneapolis, MN 55455 Chemical bonding not only determines most of the useful properties of solids; it also alters the electrostatic potentials of bonded atoms, thereby also altering electron scattering from those same atoms. Despite this, it is standard to conduct multislice TEM image simulation [1,2] by modeling the electrostatic potential of a solid as that of a collection of unbonded neutral atoms, which is a computationally convenient approximation known as the "independent atom model" (IAM). IAM simulation has proven especially successful for modeling high-angle annular dark field scanning TEM (HAADF-STEM) imaging [3,4]. However, in two previous studies the authors found that charge redistribution due to chemical bonding can measurably affect HAADF-STEM imaging of polar crystals, because interatomic charge transfer alters probe channeling [5,6]. To test the theoretical prediction of bonding-dependent contrast, we have performed quantitative HAADF-STEM imaging of AlN and MgO single crystals 10-100 nm in thickness, complemented by bonding-inclusive multislice simulations using a model of atomic potentials that includes bonding contributions, termed the "bonded crystal model" (BCM). Imaging was performed using a FEI Titan G2 60-300 S/TEM operated at 200 kV, equipped with a CEOS DCOR probe corrector and a Fischione 3000 HAADF detector (using convergence semi-angles 25-31 mrad, HAADF inner semi-angles 55-68 mrad). The TEMSIM multislice package [7] was used to simulate imaging, with BCM inputs being generated by parameterizing projected atomic potentials calculated using Quantum Espresso [8]. By calibrating the HAADF detector and averaging cross-correlated image frames, high-quality quantitative HAADF-STEM images (Figure 1) were acquired; by carefully measuring thickness, orientation, and effective source size, these images are directly comparable to simulation, albeit without definitive evidence of bonding effects (Figure 2). Results will be discussed in full detail, with emphasis on the delicacy of matching experiment and simulation, either in absolute intensity or in relative contrast. Sensitivity of results to defocus, effective source distribution, and sample orientation will be considered, in addition to the challenges of thickness determination and detector characterization [9]. References [1] M.A. O'Keefe et al., Nature 274 (1978), p. 322. [2] E.J. Kirkland et al., Ultramicroscopy 23 (1987), p. 77. [3] J.M. LeBeau et al., Phys. Rev. Lett. 100 (2008), p. 206101. [4] C. Dwyer et al., Appl. Phys. Lett. 100 (2012), p. 191915. [5] M.L. Odlyzko and K.A. Mkhoyan, Microsc. Microanal. 19 S2 (2013), p. 602. [6] M.L. Odlyzko and K.A. Mkhoyan, Microsc. Microanal. 20 S3 (2014), p. 154. [7] E.J. Kirkland, Advanced Computing in Electron Microscopy, (Springer, New York, 2010). [8] P. Giannozzi et al., J. Phys. Cond. Matter 29 (2009), p. 395502. [9] This research was supported by NSF DMR-1006706 and NSF MRSEC under award DMR-0819885. Simulations were performed using Minnesota Supercomputing Institute resources. Drs. M. Cococcioni and B. Himmetoglu are thanked for their original density functional theory calculations of bonding charge density in AlN and MgO. Microsc. Microanal. 21 (Suppl 3), 2015 122 Figure 1. Cross-correlated quantitative HAADF images imaged with a 31 mrad aberration-corrected 200kV probe: (A) 20 nm thick <2110>-oriented AlN imaged with a 55-200 mrad detector, (B) 80 nm 110>-oriented AlN imaged with a 68-200 mrad detector, (C) 40 nm thick <110>-oriented MgO thick <2 imaged with a 68-200 mrad detector. Inset BCM simulations illustrate the close agreement between experiment and simulation. HAADF intensity is normalized to incident beam current and each scale bar is 0.2 nm in length. Figure 2. Quantitative linescan analysis of simulated vs. experimental data, same case as 1B above (80 nm thick <2110>-oriented AlN, 200 kV, 31 mrad probe, 68-200 mrad detector, gaussian defocus): (A) raw data, (B) background-subtracted data normalized to maximum. In this case, differences between experimental data and either simulation are greater than the differences between the two different simulation types, showing that any bonding effect is overshadowed by other factors (particularly effective source-size determination) even in the case of a sound experiment-simulation match.");sQ1[62]=new Array("../7337/0123.pdf","Enhancement of Oxygen Contrast in a STEM HAADF Image of Perovskite Oxide SrTiO3 Using Maximum Entropy Method","","123 doi:10.1017/S1431927615001415 Paper No. 0062 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Enhancement of Oxygen Contrast in a STEM HAADF Image of Perovskite Oxide SrTiO3 Using Maximum Entropy Method L. Xie1,2 and X. Q. Pan1,3 1. 2. College of Modern Engineering and Applied Science, Nanjing University, Nanjing China 210093. Department of Materials Science and Engineering, University of Michigan, Ann Arbor MI, USA 48105. 3. Department of Physics, University of California Irvine, Irvine CA, USA 92697. Functional oxides of perovskite structure are of great research interest both for their fundamental and novel physical properties, e.g. multiferroic, metal-insulator transition, two-dimensional electron gas and polar metal, as well as versatile industrial applications, for example, piezoelectric motors, actuators, ultrasound sensors, ferroelectric tunnel junction, etc.. Oxygen atoms, which constitute the cornerstone of perovskites, play an important role in the peculiar physical properties of these materials due to their strong orbital hybridization and interaction with cations. Therefore, a thorough understanding of oxygen atoms is essential for the understanding of structure-property relationships and the design of novel functional oxides and related devices. Owing to its small atomic number, it is once difficult to directly observe oxygen in transmission electron microscopy (TEM). Only recently, with the advance of aberration-corrected TEM, direct characterization of oxygen atoms at atomic scale becomes available in TEM, for instance, negative Cs imaging [1], annular bright-field imaging [2] and electron energy-loss spectroscopy [3]. However, there are only a few reports that discuss the observation of oxygen in high angle annular dark field (HAADF) images due to the relatively weak scattering factor of oxygen. In this paper, we show that the contrast of oxygen atoms in oxides of perovskite structure can be enhanced by deconvolution of the HAADF image using Maximum Entropy Method (MEM) [4]. Figure 1(a) displays a typical atomic-resolution HAADF image of SrTiO3, viewed along the [001] direction, in which Sr and Ti atoms are of bright contrast but O atoms cannot be observed by eye. The original image was then deconvoluted using the MEM algorithm with software package STEM_CELL [5]. The parameters used for deconvolution correspond to the experimental values of our aberrationcorrected STEM JEOL-3100 R05 equipped with a cold field-emission gun, with an accelerating voltage of V=300 kV, convergence semi-angle equals to 22 mrad, C3=0.5 m, f=-12.2 � and a source size about 0.7 �. As is shown in Figure 1(b) is the deconvoluted result, in which not only the Sr and Ti atoms can be seen clearly, but also the contrast of O atoms can be readily observed. In order to further understand the origin of the contrast of O atoms in the deconvoluted image, we compare the line profiles along one of the Ti-O atomic chains and the results are shown in Figure 1(c). It should be noted that the peaks corresponding to oxygen atomic columns actually already exist in the original HAADF image. So that it's not a surprise that the deconvoluted image could exploit such hidden information. A possible explanation for the occurrence of the oxygen contrast in the original HAADF image is that the local potential of the oxygen atom is rather shallow, and as a result, the Debye-Waller factor of oxygen is relatively larger than those of cations. It's worth pointing out that in addition to SrTiO3, we also observed oxygen contrast in the deconvoluted HAADF images for a series of perovskites, such as BaTiO3, BiFeO3 and GdScO3. Therefore, it's suggested that Z-contrast imaging combined with MEM Microsc. Microanal. 21 (Suppl 3), 2015 124 deconvolution provides a possible route for the atomic-resolution characterization of oxygen in a wide variety of functional oxides. Quantitative image simulation is ongoing to gain understanding of the underlying mechanism for the origin of oxygen contrast in HAADF images. References: [1] CL Jia, M Lentzen and K Urban, Science 299 (2003), p. 870. [2] E Okunishi et al, Microscopy and Microanalysis 15 (2009), p. 164. [3] SJ Pennycook and PD Nellist, "Scanning Transmission Electron Microscopy", (Springer, New York). [4] PD Nellist and SJ Pennycook, Journal of Microscopy 190 (1997), p. 159. [5] V Grillo, Microscopy and Microanalysis 17 (2011), p. 1292. Figure 1. (a) the original and (b) the deconvoluted Z-contrast image of SrTiO3 viewed along [001] zone axis. (c) line profiles of the yellow lines in (a) and (b). The faint oxygen contrast can barely be seen in the line profile of original image but can be seen clearly after the deconvolution.");sQ1[63]=new Array("../7337/0125.pdf","Cationic Ordering and Magnetic Properties of Re-Based Double Perovskite Oxides","","125 doi:10.1017/S1431927615001427 Paper No. 0063 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Cationic Ordering and Magnetic Properties of Re-Based Double Perovskite Oxides Si-Young Choi1, Minseok Choi1, Sung-Dae Kim1, Hyung Jeen Jeen2, and Young-Mok Rhyim1 1. 2. Korea Institute of Materials Science, Changwon 641-831, Korea Department of Physics, Pusan National University, Busan 609-735, Korea Sr2FeReO6 (SFRO) and Sr2CrReO6 (SCRO) are ferrimagnetic with tetragonal structure (space group, I4/m) and their magnetic properties are strongly influenced by Fe/Re and Cr/Fe cationic ordering [1]. Our aim is to fully understand the correlation between cationic ordering and magnetization in SFRO and SCRO. To this end, we performed a comprehensive study on cationic ordering and magnetic properties in SFRO and SCRO through a series of experiments by combining theoretical calculations. A variety of the samples were synthesized and characterized in terms of microstructure and magnetic properties by varying an amount of excessive Re during sample growth. SFRO and SCRO powders were prepared via conventional solid state reaction. Since Rhenium has the volatile nature, amount of Re-excess is controlled in order to create antisite defects. All the mixed powders were pressed into disks and contact between pellets and crucible is minimized in order to avoid Re loss. The pellets were calcined at 1000 C for 10 hr under Ar atmosphere (nominal purity 99.99 %). The prepared powders were sintered by spark plasma sintering (SPS) at 1150 C for 10 min under the 65 MPa in Ar atmosphere. Their microstructural and morphological properties were characterized, and distribution of antisite defects was checked by high-angle annular dark-field scanning transmission electron microscopy (HADDF-STEM). The magnetic properties were measured via vibrating sample magnetometer (VSM) driven up to a magnetic field of 9 T. The calculations were performed using the projector augmented-wave method and the Perdew-Burke-Ernzerhof (PBE)-GGA exchange-correlation functional with a Hubbard-U correction (GGA+U ) as implemented in the VASP code. For defect simulations, the 320-atom supercells were used. The wavefunctions were expanded in a plane-wave basis set with an energy cutoff of 400 eV, and integrations over the Brillouin zone were carried out using the 2�2�2 k-point mesh. The atomic coordinates were relaxed until the force acting on each atom was reduced to less than 0.02 eV/�. The series of experiments, the formation of antisite defects, which determines the net magnetization of SFRO and SCRO, strongly correlates with the amount of added Re (fig. 1). Antisite defects are clustered in a type of antiphase-boundary-like microstructure in SFRO whereas they are spatially well distributed in the whole SCRO samples, as shown in fig. 2. The cluster-type antisite defects in SFRO deteriorated net magnetization than that from the scattered case in SCRO. Through first-principles density-functional calculations, we suggest that the differences can be explained in terms of defect formation energy and binding energy, determined by interaction between antisite defects. Our calculations show that antisite defects in SFRO are likely to form and be clustered due to attractive interaction between them, but in SCRO they favor to be spatially scattered. Our findings provide the fundamental understanding of Rebased double perovskite oxides and pave the way to experimentalists to get better quality samples. Acknowledgement This work was supported by the Global Frontier R & D Program on Center for Hybrid Interface Materials (HIM) (GFHIM-2013M3A6B1078872). Reference Microsc. Microanal. 21 (Suppl 3), 2015 126 [1] B. Blamire, J. L. MacManus-Driscoll, N. D. Mathur and Z. H. Barber, Adv. Mater. 21 (2009), p. 3827. Figure 1. (a, c) Magnetization (M) - magnetic field (H) hysteresis loop and (b, d) correalated values of the magnetic saturation and the antisite defect concentration at 300 K. The amount of excess Re is (a, b) SFRO-xRe (0 mol% x 15 mol%) and (c, d) SCRO-xRe (0 mol% x 10 mol%). In the inset of (a) and (c), M-H loop measured at 4.2 K is presented for comparison. Figure 2. HADDF-STEM image of (a) defective pristine SFRO and (b) well-ordered SFRO by 15 mol% Re-excess; (c) pristine SCRO and (d) 10 mol% Re-excess SCRO. Note that (c) and (d) samples hold the significant amount of antisite defects, ~25%.");sQ1[64]=new Array("../7337/0127.pdf","High Performance Polymers in Additive Manufacturing Processes: Understanding Process, Structure and Property.","","127 doi:10.1017/S1431927615001439 Paper No. 0064 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 High Performance Polymers in Additive Manufacturing Processes: Understanding Process, Structure and Property. Manuel Garcia-Leiner1, Daniel P. Dennies2 and Atif Yardimci3 1. 2. Exponent, Inc. Polymer Science and Materials Chemistry, Bowie, MD, USA. Exponent, Inc. Materials and Corrosion Engineering, Irvine, CA, USA. 3. Exponent, Inc. Biomedical Engineering, Chicago, IL, USA. Additive Manufacturing (AM), otherwise known as three-dimensional (3D) printing, is a growing technology area comprised of a spectrum of processes that allow production of three-dimensional solid objects of virtually any shape from information obtained from a digital object. These days, AM processes are driving major innovations in multiple areas, such as engineering, manufacturing, art, education and medicine. In its broadest sense, AM processes use additive approaches where materials are applied in successive layers in order to produce a final part, differing from traditional subtractive manufacturing techniques that often rely on the removal of materials by methods such as cutting or milling. AM processes are not necessarily new. They were introduced commercially in the early 90's for the manufacture of complex metal parts and have almost a 30-year history for plastic objects, mainly due to prototyping efforts that drove the development of multiple commercial products using techniques from stereo-lithography to laser based powder based fusion processes. The global market size for AM products is expanding at a rapid pace. Even though the manufacturing costs for AM remain higher compared to conventional processes, significant reduction in efficiency and logistics in the coming years would make AM approaches attractive for specific cases, especially through reduction of tooling costs, design freedom and reduction in assembly requirements. High demanding applications such as medical, aerospace, oil and gas exploration, military and defense and semiconductor applications will benefit directly from the expected growth of AM processes. A growing number of polymeric resins are becoming available due to developments of new processes and technological advancements in AM. Specifically, high-performance thermoplastics are perhaps the most promising material candidates for the adoption of AM into high demanding engineering applications. Because of this, the fundamental understanding of the AM process physics, as well as the resulting structure of high-performance thermoplastics and its relation to their performance in critical environments is crucial for the development of new technologies and complex processing techniques. In this regard, we provide a series of examples where polymeric systems are used for the production of parts for high demanding applications using various AM processes. In particular, this work describes an in depth study of the morphological changes observed in selected high performance polymer resins when subjected to conditions typically observed in common AM processes, including powder bed fusion processes such as Selective Laser Sintering (SLS), as well as extrusion-based approaches such as Fused Deposition Modeling (FDM). In this study, we use selected poly(etherketoneketone) (PEKK) resins that due to their superior properties (including extremely high thermal properties and polymorphic crystalline nature, superior mechanical properties, chemical resistance and low flammability) represent a viable material choice for various high demanding engineering applications. We focus on the analysis of the crystal structure and Microsc. Microanal. 21 (Suppl 3), 2015 128 the development of polymorphism in PEKK systems (Figure 1) under conditions often encountered in SLS and FDM processes. We observed that parts produced through various processes and conditions will display different crystalline morphology that will lead to changes in their mechanical properties (i.e. tensile strength and elongation at break) (Figure 2). We conclude that the understanding and precise control over the morphological changes in these materials during AM processing is critical for their successful introduction into high demanding engineering applications. References: [1] [2] [3] [4] [5] [6] KH Gardner et al, Polymer 33 (1992), p. 2483-2495. ZD Cheng et al, Macromol. Chem Phys. 197 (1996), p. 185-213. EA Klop et al, Journal of Polymer Science: Part B: Polymer Physics 33 (1995), p. 315-326. F Muller et al, Unites States Patent 8,299,208 B2, (2012). C Bertelo et al, World Patent Application WO2012/047613 A1, (2012). C Bertelo et al, United States Patent Application US2013/0323416, (2013). Figure 1. Wide Angle X-ray Diffraction (WAXD) patterns for PEKK crystalline polymorphs (Form I and Form II). The presence of these polymorphs will depend on the specific AM process used. (a) (b) Figure 2. Influence of processing conditions on the crystal composition in PEKK parts produced by various AM processes. a) Powder bed fusion processes (SLS); b) Extrusion-based processes (FDM).");sQ1[65]=new Array("../7337/0129.pdf","Micro-scale X-ray Computed Tomography of Additively Manufactured Cellular Materials under Uniaxial Compression","","129 doi:10.1017/S1431927615001440 Paper No. 0065 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Micro-scale X-ray Computed Tomography of Additively Manufactured Cellular Materials under Uniaxial Compression Nikolaus L. Cordes1, Kevin Henderson1, Tyler Stannard2, Jason J. Williams2, Xianghui Xiao3, Mathew W. C. Robinson4, Tobias A. Schaedler5, Nikhilesh Chawla2, Brian M. Patterson1 Polymers and Coatings Group, Materials Science and Technology Division, Los Alamos National Laboratory, Los Alamos, NM USA Department of Materials Science and Engineering, School for Engineering of Matter, Transport, and Energy, Arizona State University, Tempe, AZ, USA 85287-6106 3 2 1 Advanced Photon Source, Argonne National Laboratory, 9700 South Cass Avenue, Argonne, IL, USA 60439-4837 Atomic Weapons Establishment, Aldermaston, Reading, Berkshire, RG7 4PR, UK HRL Laboratories LLC, 3011 Malibu Canyon Road, Malibu, CA, USA 90265 4 5 Additively manufactured (AM) materials are a current "hot topic" in materials science.[1] Current investigations of AM cellular materials, such as polymer foams, include the mechanical performance under compressive strain using 3D X-ray computed tomography (CT) at the microscale. This allows for an enhanced visualization of processes such as void collapse, brittle fracture, and increased surface area of AM materials while undergoing compressive strain. The results from this study will include interrupted in situ compression of AM polymer foams using laboratory-based CT as well as dynamic compression of AM polymer foams using synchrotronbased CT. Laboratory-based interrupted in situ compression of AM polymer foams was imaged using an Xradia (now Carl Zeiss X-ray Microscopy, Inc., Pleasanton, CA) MXCT micro-scale X-ray transmission microscope and a Deben (Suffolk, UK) 500 N tension/compression cell. Microscale CT of AM polymer foams under dynamic compression was carried out at Argonne National Laboratory's Advanced Photon Source (APS) using beamline 2-BM, which enabled data acquisition of 1 tomogram per second at a compressive strain rate of 10-2 s-1, resulting in 20 successive tomograms per sample. A variety of imaged samples were fabricated using different AM techniques. However, the majority of the presentation will focus on AM polymer foams printed using an Objet500 Connex 3D printer (Stratasys Ltd., Eden Prairie, MN). The material printed was a rubber-like elastomer. The presentation will also focus on a NiP microlattice: details of the fabrication of the NiP microlattice can be found in reference [2]. Figure 1 (Left) displays four volume renderings of reconstructed tomograms of an AM polymer foam, at four separate strains (0%, 15%, 29%, and 45%), as imaged under dynamic compression using synchrotron-based CT. The sample, fabricated to have parallel tubular pores, exhibits compressive behavior typical of an elastomeric foam (Fig. 1 (Right)) as described by Gibson and Ashby[3]. However, visibly absent from the stress-strain curve is a linear region at low strain, which indicates an absence of bending of the foam ligaments. Figure 2 (Left) displays four Microsc. Microanal. 21 (Suppl 3), 2015 130 volume renderings of reconstructed tomograms of a NiP microlattice, at four separate stages of compression. The corresponding stress-time curve (Fig. 2 (Right)) exhibits high stress (~1300 kPa) at a time of 25 s (corresponding to the 6th tomogram (Figure 2 (Left), top right), after which brittle fracture of the NiP ligaments occurs. We will also demonstrate how CT can accurately measure the change in the surface area of the AM polymer foams while undergoing compression, leading to a more accurate stress-strain curve, when compared to traditional methods. References [1] B. C. Gross et al., Anal. Chem., 86(7) (2014) 3240. [2] A. Torrents et al., Acta Mater., 60(8) (2012) 3511. [3] L. J. Gibson and M. F. Ashby, Cellular Solids: Structures and Properties, 2nd ed., The Press Syndicate of the University of Cambridge, 1997. Figure 1. (Left) Micro-scale X-ray CT tomograms (as volume renderings) of an AM polymer foam under uniaxial compressive strains. The scale bar is in micrometers. (Right) Stress-strain curve of the AM polymer foam shown in Figure 1. Data was acquired while the sample was undergoing dynamic uniaxial compression during CT imaging. Figure 2. (Left) Micro-scale X-ray CT tomograms (as volume renderings) of a NiP microlattice under uniaxial compressive strains, corresponding to the 1st (top left), 6th (top right), 8th (bottom left) and 13th (bottom right) tomograms. The scale bar is in micrometers. (Right) The stress-time curve of the NiP microlattice, indicating brittle fracture. Black squares correspond to tomograms presented in Fig. 2 (Left). Data was acquired while the sample was undergoing dynamic uniaxial compression during CT imaging.");sQ1[66]=new Array("../7337/0131.pdf","Recent Advancements in 3D X-ray Microscopes for Additive Manufacturing","","131 doi:10.1017/S1431927615001452 Paper No. 0066 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Recent Advancements in 3D X-ray Microscopes for Additive Manufacturing Leah Lavery1, William Harris1, Jeff Gelb1 and Arno Merkle1. 1. Carl Zeiss X-ray Microscopy, Pleasanton, CA, United States of America. Three-dimensional X-ray microscopy (XRM) is a powerful sub-surface imaging technique that reveals tomography of three-dimensional microstructure from a range of materials, non-destructively. The nondestructive nature of X-rays has made the technique widely appealing, with the potential for characterizing sample changes in "4D," delivering 3D microstructural information on physically the same sample over time, as a function of sequential processing conditions or experimental treatments. This has led to a new generation of functional studies with applications and is in a state of rapid expansion [1]. Recently, laboratory-based X-ray sources have been coupled with high resolution X-ray focusing and detection optics from synchrotron-based systems to acquire tomographic datasets with resolution down to 50 nm across a great span of sample dimensions [2]. This signifies an improvement of at least one order of magnitude in spatial resolution relative to the limits of `optic-free' laboratory computed tomography (CT) techniques. This talk will explore both the implementation of optics in nanoscale and sub-micron laboratory XRM architectures and review in detail several leading applications examples for the field of additive manufacturing. XRM tomographic data provides multiscale imaging and visualization for a wide variety of AM materials such as 3D characterization of scaffold implant materials, porosity quantification in sintered powder steels, even creation of accurate 3D-printed models in biomimetic studies [3-4]. Additive manufacturing (AM) techniques can produce complex 3D structures. The reliability and performance of the produced parts relies heavily on the resultant microstructure. Frequently the material performance can be quite sensitive to its discrete, complex pore structure, thereby motivating the need to investigate and understand the morphology of these materials in 3D at the appropriate pore scale. XRM provides a means to perform non-destructive 3D characterization of complex even anisotropic geometries. Beyond imaging, quantification of volumetric properties such as porosity can be calculated from the reconstructed dataset. This is highly dependent on the spatial resolution of the imaging system. Porosity is an important property because for materials destined for high stress applications should be fully-dense to avoid failure in service. XRM has been used to characterize porosity in metal additive process control studies [3]. Figure 1 shows a sintered powder steel cylinder. From the 3D reconstructed dataset (inset), a virtual 2D cross-section (left) reveals several regions of non-sintered volume. In addition to the non-sintered volume, several micron-size cracks were discovered. Through the incorporation of post-sample optical magnification detector technology, sub-micron XRM has recently extended the application scope of laboratory-based microCT by extending resolution and contrast. As illustrated in Figure 1, interior (local) tomography is routinely performed, as ZEISS Xradia Versa XRM (Figure 2) architecture has been designed to maintain sub-micron spatial resolution for a variety of working distances and sample sizes/geometries. System is capable of imaging at high resolution at large working distance, opening up opportunities for interior tomography of large samples and in situ experiments with large environmental cells. Furthermore, new contrast modes have emerged based on dual energy absorption or diffraction. Soft materials, such biocompatible photopolymers, consistently pose challenges in generating contrast Microsc. Microanal. 21 (Suppl 3), 2015 132 by several techniques, X-ray absorption included. We demonstrate the application of phase contrast techniques on such materials. The tunable contrast enhancement mechanism (propagation-based) within combines optimized absorption contrast, for enhanced imaging of soft materials (low-Z) such as scaffold or implant materials. Porosity is these materials is often engineered. [5] X-ray microscopes provide internal structural information critical to the evaluation of additive manufacturing materials as well as aid in the design and processing of next generation materials. The ability to visualize structures in 3D with good contrast and resolution even down to 50nm, particularly under operating conditions or elapsed time (4D), can help quickly determine why certain structures, synthesis methods, or operating conditions fail and why others succeed. This can help to bypass some of the empiricism and interpretation that can be associated with less direct structural characterization tools, expediting the iterative design process for the researcher and saving time in the design of future materials. Figure 1 From the 3D reconstructed dataset (inset), a virtual 2D cross-section (left) reveals several regions of non-sintered powder steel volume. In addition to the non-sintered volume, several micronsize cracks were discovered. Sample courtesy of NIST. Figure 2 Sub-micron XRM (ZEISS Xradia Versa) optical architecture. Magnification is achieved through a combination of geometric (sample, source, detector placement) and optical (post-sample, variable scintillator-lens-ccd coupling) methods. References: [1] A. Merkle and J. Gelb. Microscopy Today 21, (2013) pp. 10-15. [2] A. Tkachuk, et al., Z. Kristallogr. 222, (2007) pp. 650-655. [3] Slotwinski, J. et al. J. Res. Natl. Inst. Stan. 119, (2014) pp. 494-528. [4] Wen, L. et al. J. Exp Biol 217, (2014) pp. 1656�1666. [5] Sai, H. et al. Science 341, (2013) pp. 530�534.");sQ1[67]=new Array("../7337/0133.pdf","Characterization of Grain Structure and Porosity in Selective Laser Melted Cu-4.3Sn for Enhancing Electrical Connector Fabrication","","133 doi:10.1017/S1431927615001464 Paper No. 0067 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of Grain Structure and Porosity in Selective Laser Melted Cu4.3Sn for Enhancing Electrical Connector Fabrication Anthony P. Ventura1, C. Austin Wade1, Wojciech Z. Misiolek1, Masashi Watanabe1, Greg Pawlikowski2, Martin Bayes2 1. 2. Lehigh University, Materials Science and Engineering Department, Bethlehem, PA, USA TE Connectivity, Harrisburg, PA, USA Copper (Cu) has the highest electrical conductivity of any commercial metal; it is surpassed only by silver, which is prohibitively expensive for use in commercial applications. Worldwide, up to 65% of Cu and its alloys produced are destined for electrical applications [1]. Although Cu and its alloys have seen extensive use in the development of additive manufacturing (AM) technology, they often assume the role of a binding or infiltration phase rather than the primary material. Developments in selective laser melting (SLM) have allowed for the fabrication of relatively high density Cu alloy components with unique microstructures which opens up the possibility of using AM for the production of electronic components. However, the expansion of AM into electronic applications requires an understanding of microstructure development during fabrication of Cu alloys by SLM. Although grain structure has been characterized in some titanium, aluminum, and ferrous alloys, there has been limited work characterizing Cu alloys fabricated by SLM [2-4]. In this study the microstructure of a Cu-4.3Sn alloy fabricated by SLM was investigated using scanning electron microscopy (SEM) and aberrationcorrected scanning transmission electron microscopy (STEM). Pre-alloyed Cu-4.3Sn powder produced by air atomization was supplied by Ecka Granules with a maximum particle size of 63 microns. A 200 W EOSINT M 280 direct metal laser sintering unit at TE Connectivity was used for the fabrication of all samples. The laser parameters were optimized for a maximum attainable density of approximately 96% before 10 mm3 cubes were fabricated using a simple raster scan strategy rotated 30� between each build layer. Strip samples were also prepared to measure electrical conductivity via the 4 point probe technique. Metallography was performed using standard preparation techniques and samples were etched in Klemm's I for observation in polarized light optical microscopy. Metallographic samples were then re-polished and used for electron backscatter diffraction (EBSD) conducted on a Hitachi 4300 field emission SEM. TEM specimens were prepared by a Focused Ion-Beam (FIB) instrument on an FEI SCIOS dual-beam instrument and final thinning was performed on a Fischione 1040 Nanomill. TEM/STEM analysis was completed using a JEOL JEM-ARM200CF aberration-corrected STEM. The polarized optical micrographs in Figure 1 show weld pools characteristic to the SLM process. Micro-voiding and a unique substructure can be seen at the weld pool interfaces during backscattered electron SEM observation. Grains were found to be elongated along the build direction, often spanning multiple build layers. Figure 2 is EBSD inverse pole figure maps perpendicular to the build direction showing grain orientation in dense and porous regions. Additionally, grain texture was observed along the build height of the entire sample in the transverse direction. High-angle annular dark-field STEM imaging was used along with X-ray energy dispersive spectrometry to characterize the weld pool interfaces and identify second phase particles within the microstructure. Microsc. Microanal. 21 (Suppl 3), 2015 134 References [1] J.R. Davis in "Copper and Copper Alloys: ASM Specialty Handbook", ed. 1 (ASM International, Materials Park) p. 3. [2] L. Thijs et al., Acta Materialia 58 (2010), p. 3303. [3] L. Thijs et al., Acta Materialia 61 (2013), p. 1809. [4] T. Niendorf et al., Metallurgical and Materials Transactions 44B (2013), p. 794. [5] The authors would like to acknowledge the joint funding from the Research for Advanced Manufacturing in Pennsylvania program and TE Connectivity, Ltd. Build Direction Figure 1. Polarized micrographs of Cu4.3Sn etched in Klemm's I. Mounting material visible in bottom of left micrograph. Build Direction Figure 2. Inverse pole figure map of SLM Cu-4.3Sn at low magnification (left) and high magnification of a dense area (right). Black areas are porosity.");sQ1[68]=new Array("../7337/0135.pdf","Examination of 3D Manufactured Alloys and Materials for High-performance Drilling and Oilfield Service Tools","","135 doi:10.1017/S1431927615001476 Paper No. 0068 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Examination of 3D Manufactured Alloys and Materials for Highperformance Drilling and Oilfield Service Tools John N. Williard1, Somesh Mukherjee1, Jair Leal1 1. Baker Hughes, 9110 Grogan's Mill Road, The Woodlands, TX 77380 Additive manufacturing or 3D printing is a quickly maturing area that has shown rapid progress over the past few years. It is now possible to produce 3D printed objects directly from a computer aided design (CAD) file, with exceptionally high fidelity and precision. While the process is improving, there is no guarantee that a printed model is structurally sound. We have evaluated resulting microstructures of the produced alloys to show some of the abnormalities present. This presentation will cover examination by optical microscopy (OM), scanning electron microscopy (SEM) and other non-destructive analytical techniques such as color 3D laser scanning confocal microscopy (LSCM) ) as investigative tools for 3D alloy development. The results of these studies will assist in the development of materials for innovative tools as well as continuous improvements in manufacturing and drilling. We have examined the Ti6Al-4V alloy development using 3D printing technology. Figure 1 illustrates the prealloyed powder utilized in this study for laser sintering showing the sizes and distributions. As-received laser sintered and secondary processed (such as hot isostatic pressing) components, see Figure 2, are examined under microscope to evaluate the whole development process. The results show that, for as-laser sintered material, some areas of the samples had void regions of partially "melted" and structurally weak bonds. Porosity and inclusions are present in all direction of produced materials [1-4]. It also shows the directional grain of the as-laser sintered microstructure. The porosity is significantly reduced by hot isostatic pressing (HIP). HIP also produces the desired microstructure of the alloy. The details of our investigation will be discussed. OM requires the artful polishing and selective (chemical or electrochemical) etching of microstructures. This is illustrated in Figures 3-5 as progression from high porosity toward low porosity and high-quality etched alloy structure. SEM examination of alloy surface microstructures generally does not require any special specimen preparation except for cutting representative sections to fit within the imaging constraints (size limitations) in the SEM. Defect characterization of produced parts is another key factor in quality manufacturing. The laser scanning confocal microscope (LSM) provides non-destructive, non-contact high-accuracy measurements with a pinhole confocal optical system, enabling automated dimensional measurements. This system can be used to perform height, profile, area, 3D image, and surface topography measurements, see Figure 6. Microsc. Microanal. 21 (Suppl 3), 2015 136 Optical, SEM, and associated techniques along with non-contact metrology are valuable tools to gain information on the character and properties of materials being developed for high-performance drilling and oilfield service tools. References [1] Loeber.L., et al., SSF Conference Proceedings (2011) p. 1. [2] Hrabe N, et al., IFF Conference Proceedings 2012-79, 1045-1058. [3] Kenzari S., et al., Sci. Technol. Adv. Mater. 15 (2014), 024802. [4] Murr L.; Metallurgy of Additive Manufacturing (2014), 1074-1080. [5] The authors acknowledge Wes Bodenhamer for his many useful discussions and contributions to this work. Fig 1 Fig 2 Fig 3 Fig 4 Fig 5 Fig 6 FIG 1. Backscatter electron image of Ti-6Al-4V powder, note microdendritic structure. FIG. 2. Macro image of printed part with sections removed for optical and SEM examination. FIG 3. Optical image of unetched Ti6Al-4V material before HIP illustrating defects and sub-surface imperfections. FIG 4. Unetched HIP sample. Note material has less porosity. FIG 5. OM image of etched sample after HIP with fine microstructure. FIG 6. 3D image produced by Laser Scanning Confocal Microscope illustrating noncontact porosity void depth.");sQ1[69]=new Array("../7337/0137.pdf","AC-STEM Studies of Phase Transformation and Evolution of Li-Rich Layered Cathode Materials Induced by Battery Charge-Discharge Cycling and by Electron-beam Irradiation","","137 doi:10.1017/S1431927615001488 Paper No. 0069 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 AC-STEM Studies of Phase Transformation and Evolution of Li-Rich Layered Cathode Materials Induced by Battery Charge-Discharge Cycling and by Electron-beam Irradiation Ping Lu1, Pengfei Yan2, Eric Romero1, Erik David Spoerke1, Ji-Guang Zhang3, and Chong-Min Wang2 Sandia National Laboratories, P O Box 5800, Albuquerque, NM 87185, USA Environmental Molecular Sciences Laboratory, Pacific Northwest National Laboratory, 902 Battelle Boulevard, Richland, WA 99352, USA 3 Energy and Environmental Directorate, Pacific Northwest National Laboratory, 902 Battelle Boulevard, Richland, WA 99352, USA 2 1 Structural and chemical imaging of defects and crystal structures at the atomic-scale has become possible by recent technical advances in aberration-corrected scanning transmission electron microscopy (AC-STEM) and x-ray detector technology [1]. AC-STEM imaging using high-angle annular dark-field (HAADF) detector and chemical mapping using energy-dispersive x-ray spectroscopy (EDS) have been used in combination to resolve interfaces, defects and crystalline structures unambiguously [2]. The AC-STEM capability has also been recently used to study layered lithium and manganese-based materials, such as the lithium-rich, manganeserich (LMR) oxides (e.g., Li[LixMnyTM1-x-y]O2 (TM = transition metal, e.g., Ni, Co or Fe)), which are of great interest as cathode materials for secondary (rechargeable) lithium ion batteries [3]. These studies have largely attributed the loss of capacity and voltage in the layered oxides upon the charge-discharging cycling to the development of surface reconstructed layers (SRLs) on cathode particle surfaces. On the other hand, lithium-containing oxides are known to be prone to radiation damage by high-energy electron beams. In this study, we report a comparative ACSTEM study on the structural evolution of the surface layer on LMR Li[Li0.2Ni0.2Mn0.6]O2 (LNMO) nanoparticles (NPs) induced by high-energy electron-beam irradiation and by chargedischarge cycling [4]. Unless extreme caution in controlling electron exposure was taken during STEM imaging, the surfaces of LNMO NPs were always found to have the SRLs which resemble those reported due to charge-discharge cycling [3]. Fig.1 shows the phase transformation and evolution on the surfaces of LNMO due to electron-beam irradiation. The original state of the NPs without the SRL(Fig.1a) was quickly transformed to a thin, defect spinel SRL after exposed to a 200 kV electron beam for as little as 30 seconds under normal high-resolution STEM imaging conditions (with an equivalent electron beam current-density of about 18 A/cm2). Further electron irradiation produced a thicker layer of the spinel phase, ultimately creating a rock-salt layer at a higher electron exposure (Fig.1c). Fig.2 shows the surfaces of LNMO NPs after the electrochemical charge-discharge cycling. Both the defect spinel structure and the rock-salt structure are visible near the edge of the particle. Atomic-scale chemical mapping by EDS in STEM has also been used to study the electron-beam-induced SRL formation on LNMO. The result of the EDS study indicates the transformation is accomplished by migration of the transition metal ions to the Li sites without breaking down the lattice [4, 5]. The observations in this study enable a better understanding of the mechanism of forming the SRL and its structural evolution in the Li-rich layered oxides during the cycling, and also points to the special caution necessary when using TEM or STEM for the study of the lithium layered oxides and other similar oxides. Microsc. Microanal. 21 (Suppl 3), 2015 138 References: 1. Von Harrach, H.S., Dona, P., Freitag, B., Soltau, H., Niculae, A. & Rohde, M. (2009). Microsc. Microanal. 15 (suppl.2) 208-209. 2. P. Lu, J. Xiong, M. Van Benthem . & Q.X. Jia App. Phys. Lett. 102, 173111 (2013); P. Lu, E. Romero, S. Lee, J. L. MacManus-Driscoll & Q. X. Jia, Microsc. Microanal. 20, 17821792 (2014); P. Lu, L. Zhou, M.J. Kramer & D. J. Smith, Sci. Rep. 4, 3945-3949 (2014). 3. A. Boulineau, L. Simonin, J. �F. Colin, J.-F.; C. Bourbon, & S. Patoux, Nano Lett. 13, 3857 (2013); B. Xu, C.R. Fell, M. Chi, Y.S. Meng, Energ. Environ. Sci. 4, 2223 (2011); F. Lin, I.M. Markus, D. Nordlund, T.-C. Weng, M.D. Asta, H. L. Xin & M.M. Doeff, Nat. Commun. 5, 3529 (2014) 4. P. Lu, P. F. Yan, E. Romero, E. D. Spoerk, J.-G. Zhang and C.-M. Wang, Chemistry of Materials, accepted. (2015). 5. Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the US Department of Energy's National Nuclear Security Administration under contract DE-AC0494AL85000. a _ 200 001 b spinel c rock-salt (200) (001) 1nm spinel Fig.1. High-resolution STEM HAADF images showing the effect of electron-beam irradiation: (a) original state of LNMO NPs obtained without electron beam pre-exposure; (b) and (c) taken after electron-beam irradiation for 30 seconds and 200 seconds respectively under normal highresolution STEM viewing conditions (a current-density of about 18 A/cm2, at dwell-time of ~1 �s per pixel). Images are taken with the bulk in [010] direction (space group C2/m), as seen from FFT pattern in inset. The dash lines indicate the approximate positions separating the SRL from the bulk. Major lattice planes are marked for the bulk in (a). _ 200 _ (111) rock-salt 001 _ (111) (200) spinel (002) (001) 1nm Fig.2. High-resolution STEM HAADF image showing the SRLs formed due to the electrochemical charge-discharge cycling. The dash lines indicate the approximate positions separating the SRL from the bulk. Major lattice planes are marked in the respective phases. Images are taken with the bulk in [010] direction (space group C2/m).");sQ1[70]=new Array("../7337/0139.pdf","Structural Evolution During Electrochemical Cycling of Epitaxial LiCoO2 Films Studies by S/TEM","","139 doi:10.1017/S143192761500149X Paper No. 0070 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Structural Evolution During Electrochemical Cycling of Epitaxial LiCoO2 Films Studied by S/TEM H. Tan1, S. Takeuchi1, K. Kamala Bharathi1.2, I. Takeuchi2, and L.A. Bendersky1 Material Measurement Laboratory, National Institute of Standards and Technology, Gaithersburg, Maryland, USA 2 Department of Materials Science and Engineering, University of Maryland, College Park, Maryland, USA LiCoO2 (LCO) has been the most important and most studied positive electrode material for lithium-ion batteries; thus for this work we have selected LCO as a model material for studying electrochemical property of single orientation, binder-free cathodes in a form of epitaxial thin films [1]. In order to capture the effect of crystallographic orientation of a cathode/electrolyte interface and diffusion anisotropy, the different orientation films were obtained by deposition on single-crystal substrates of different orientations. SrTiO3 (STO) substrates with 111, 110, and 100 surfaces were used to induce 001, 110, and 104 out-of-plane orientation of LCO, respectively. In the course of this study it was realized that a layer of highly conductive SrRuO3 (SRO) between LCO and STO is essential to (a) remove a rectifying Schottky barrier between LCO and STO, (b) act as high-conductivity current collector, and (c) preserve the intended orientation of LCO films. Both SRO and LCO films were deposited sequentially by pulsed laser deposition (PLD) at 600 �C temperature of a substrate, 200 mTorr oxygen pressure with a KrF laser (wavelength 248 nm) using repetition rate 0 Hz and the laser energy 100 mJ per pulse. The analysis of the LCO/SRO/STO(111) cross-section sample prepared by FIB is shown in Fig. 1. The LCO surface is predominately flat, with crystallographic (001)r-LCO plane (r-LCO � rhombohedral layered structure), but with occasional faceted grooves. According to the SEM image, Fig. 1b, the surface consists of triangular islands with faceted surfaces identified as {104}r-LCO planes. From electron diffraction and HAADF/STEM imaging, Fig. 1c,d, the following orientation relationship is established: (111)STO//(111)SRO//(001)r-LCO, [110]STO//[110]SRO//[100]r-LCO. Similar analysis was performed for other two orientations, with conclusion that there is a single orientation relationship between LCO and STO, which can be understood though a common fcc oxygen sublattice for all three oxides. However a lower symmetry of the rhombohedral LCO phase introduces domains with different orientation of c-axis. Cyclic voltammograms of the films with different orientations were obtained by using a specially constructed three-electrode cell, with lithium metal as counter and reference electrodes and 1 mol dm-3 LiClO4/propylene carbonate as electrolyte. Example of a cyclic voltammogram for the LCO/SRO/STO(111) film is shown in Fig. 2a; the voltammogram shows the major redox charge/discharge peaks that are typical for the rhombohedral LCO and represent, according to literature [2], a series of phase transitions corresponding to the changes in Li content. In order to observe directly the phases and their spatial distribution, the electrochemical measurements were stopped at different stages of interest, e.g. at points C1-1 or D1-1 in Fig 2a, the film was cleaned from the electrolyte, and a TEM cross-section sample representing this stage was prepared by FIB (see schematic drawing in Fig.2b) and examined by TEM; example of the structure observed at C1-1 stage is shown in Fig. 2b. After extracting the TEM sample, the film was return to a glove box for continuing electrochemical testing until the next stage of interest. Structural changes, formation of defects and interfacial modification at different stages of lithiation/delithiation will be discussed in the paper in details. 1 Microsc. Microanal. 21 (Suppl 3), 2015 140 References: [1] S. Takeuchi, H. Tan, K. Kamala Bharathi, G.R. Stafford, J. M. Shin, S. Yasui, I. Takeuchi, L.A. Bendersky, to appear in ACS Applied Materials and Interfaces (2015). [2] J.N. Reimers and J.R. Dan, J. Electrochem. Soc., 139, (1992), 2091 Figure 1. Characterization of the as-deposited LCO on STO(111) film by (a) STEM overview of crosssection, (b) SEM of a surface, (c) electron diffraction and (d) high resolution HAADF-STEM of LCO/SRO interface. The {104} planes are 54.7� with respect to the horizontal STO(111) cut plane. Figure 2. a. Cycling voltammogram of LCO/SRO/STO(111) film showing stages at which TEM samples were extracted; (b) Example of STEM imaging of a sample at C1-1 stage showing structural changes from layered (area A) to spinel-like (area B) structure in a hear-surface region.");sQ1[71]=new Array("../7337/0141.pdf","Structural and Chemical Evolution of Li and Mn Rich Layered Oxide Cathode and Correlation with Capacity and Voltage Fading","","141 doi:10.1017/S1431927615001506 Paper No. 0071 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Structural and Chemical Evolution of Li and Mn Rich Layered Oxide Cathode and Correlation with Capacity and Voltage Fading Chong-Min Wang1, Pengfei Yan1, and Ji-Guang Zhang2 1 Environmental Molecular Sciences Laboratory, Pacific Northwest National Laboratory, 902 Battelle Boulevard, Richland, WA 99352, USA 2 Energy and Environmental Directorate, Pacific Northwest National Laboratory, 902 Battelle Boulevard, Richland, WA 99352, USA Lithium-rich, magnesium-rich (LMR) cathode materials have been regarded as one of the very promising cathodes for Li-ion battery applications. However, their practical application is still limited by several challenges, especially by their limited electrochemical stability and rate capability. In this work, we present recent progresses on the understanding of the structural and composition evolution of LMR cathode materials with emphasis being placed on the correlation between structural/chemical evolution and electrochemical properties. In particular, using Li[Li0.2Ni0.2Mn0.6]O2as a typical example, we clearly illustrate the structural characteristics of the pristine materials and their dependence on the materials processing history, cycling induced structural degradation/chemical partition and their correlation with degradation of electrochemical performance. The fundamental understanding obtained in this work may also guide the design and preparation of new cathode materials based on ternary system of transitional metal oxide. The advanced lithium ion batteries (LIBs), featured with high capacity, high operating voltage and high rate capability, are required for both portable electronic devices and electric vehicle applications. As one of the key components of a LIB, the cathode is a limiting factor for high energy density LIBs. Intense investigations have been conducted to search for the advanced cathode materials. Among various candidates, the lithium-and-manganese-rich (LMR) cathode materials has been regarded as a promising material which can deliver a much higher energy density than the traditional cathode materials such as LiMn2O4 spinel and LiCoO2. For example, the typical LMR cathode (Li[Li0.2Ni0.2Mn0.6]O2) can deliver a discharge capacity exceeding 250 mAh g-1 after initial charging. The pristine LMR cathode materials have a layered-layered structure, which can be written as (xLi2MO3(1-x)LiMO2, 0<x<1, M=Mn, Ni, Co, et al). As one of the most popular LMR cathode materials, Li[Li0.2Ni0.2Mn0.6]O2 (represented by 20N-LMR) can be expressed as (50% Li2MnO3 + 50% LiNi0.5Mn0.5O2). Upon cycling, the most discernable change of the 20N-LMR particles is the formation of a surface reconstruction layer (SRL) which was also observed in Ni-Mn-based cathodes [1, 2]. It has been generally accepted that cycling induced SRL have the following features: 1) Formation of oxygen vacancies; 2) Transition metal (TM) cations hopping into Li-sites; 3) TM cations being reduced to low valence state; and 4) Lattice structure transformation. Selective area electron diffraction (SAED) patterns confirmed the formation of the SRL (as shown in Figure 1 (a-d)), where extra spots are come from the newly developed SRL (highlighted by red and blue circles). Figure 1 (e-j) show that TM cations have occupied the octahedral sites in Li-slabs within the SRL. With increasing cycle numbers, the thickness of the SRL increases gradually. The SRL structure is actually an ordered structure, which means extra ordering was introduced during cycling as compared with pristine layered structure. Figure 1 (j) shows a snapshot of the SRL development, where there are three Microsc. Microanal. 21 (Suppl 3), 2015 142 distinguished structures from the outmost layer to inner bulk. By comparing experimental images with simulation structural models, Yan et al(Yan, et al., 2014) proposed the outmost layer are M3O4-type spinel and the middle transition layer matched best with a tetragonal structure with space group of I41 (the ICSD No. is 164994). For pristine 20N-LMR, samples synthesized by the CP and SG methods show Ni-rich layer on particle surface, which is detrimental to the electrochemical stability of the LMR cathode. Suppression on the Ni segregation gives rise to significantly enhanced electrochemical performances, as reflected by the largely improved cycling stability and mitigated voltage fade the sample synthesized by HA method. Systematic TEM investigations on the samples with different cycle numbers revealed that the structural degradation of cathode particle initiated from particle surface and propagated into inner bulk accompanied by a progressive chemical composition changes. Such structural and chemical change results in material's degradation as well as cell performance degradation. The better understanding on the correlation between the structure, especially the surface structure of cathode material and its electrochemical properties may also provide useful clues for improving the performances of other electrode materials used in rechargeable battery systems [4]. Figure 1. (a-b) [010] zone axis SAED patterns. Extra diffraction spots appeared in cycled samples, which are highlighted by red and blue circles. Red circles indicated the formation of (a) (b) (d) (c) ordered structure. Blue circles come from double diffraction. (e-h) High resolution STEM-HAADF images to show the cycling induced structure (e) (g) (f) (h) change on particle surfaces. Pristine [101] 45 cycled 45 cycled samples (e) shows homogeneous _ C2/m (101) structure from surface to bulk. Dashed C2/m lines in (f, g) highlight the thickness of the SRL. In (h), the whole areas were Spinel I41 transformed. (i) [101] zone axis STEM-HAADF image and its fast I41 Fourier transformation image. (j) [010] zone axis STEM-HAADF image (i) (j) to show spinel structure and I41 structure in a 45 cycled sample. The scale bars are 2 nm in (e-j) [3]. References [1] Gu, M., Belharouak, I., Zheng, J., Wu, H., Xiao, J., Genc, A., Amine, K., Thevuthasan, S., Baer, D. R., Zhang, J.-G., et al. (2012). ACS Nano; 7, 760-767. [2] Lin, F., Markus, I. M., Nordlund, D., Weng, T. C., Asta, M. D., Xin, H. L., Doeff, M. M. (2014). Nat. Commun.; 5, 3529. [3] Yan, P., Nie, A., Zheng, J., Zhou, Y., Lu, D., Zhang, X., Xu, R., Belharouak, I., Zu, X., Xiao, J., et al. (2015). Nano Lett.; 15, 514-522. [4] The work was conducted in the William R. Wiley Environmental Molecular Sciences Laboratory (EMSL), a national scientific user facility sponsored by DOE's Office of Biological and Environmental Research and located at PNNL. Pristine 5 cycled 45 cycled 100 cycled");sQ1[72]=new Array("../7337/0143.pdf","Multiscale Structural Architectures of Novel Sulfur Copolymer Composite Cathodes for High-Energy Density Li-S Batteries Studied by Analytical Multimode STEM Imaging and Tomography","","143 doi:10.1017/S1431927615001518 Paper No. 0072 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Multiscale Structural Architectures of Novel Sulfur Copolymer Composite Cathodes for High-Energy Density Li-S Batteries Studied by Analytical Multimode STEM Imaging and Tomography V.P. Oleshko1,2, A. Herzing1, J. Kim1, J. Schaefer1, C. Soles1, J.J. Griebel3, W.J. Chung3, A.G. Simmonds3, J. Pyun3 1 Material Measurement Laboratory, National Institute of Standards and Technology, Gaithersburg, MD 20899 2 Department of Materials Science and Engineering, University of Maryland, College Park, MD 20742 3 Department of Chemistry & Biochemistry, University of Arizona, Tucson, AZ 85721 Li-S rechargeable batteries are considered to be a promising light-weight, low-cost, and environmentally friendly candidate for next generation energy storage owing to high theoretical specific capacity of 1,672 mAh/g and high specific energy of 2,567 Wh/kg, which is 5 times that of current Li-ion technology. However, practical use of Li-S batteries remains limited because they suffer from gradual capacity fading caused by insulating properties of sulfur and polysulfide shuttle. Recently, poly(sulfur-random-(1,3diisopropenylbenzene) (poly(S-r-DIB)) copolymers have been introduced for their use as active materials in cathodes for Li-S batteries, and were found to be capable of realizing enhanced capacity retention (1005 mAh/g at 100 cycles) and a five-fold increase in lifetime (over 500 cycles) as compared to conventional sulfur-carbon cathodes [1, 2]. These materials are typically organized in a rough hierarchical 3D architecture which contains multiple components and is quite challenging to understand and characterize. Herein, we employ multimode analytical STEM imaging coupled with multivariate statistical analysis (MSA) and electron tomography (ET) to investigate the morphology and compositions of the composite cathodes created by combining the poly(S-r-DIB) copolymers with varying DIB content (0-50 % by mass) and Timcal Super C65 conductive carbons as well as their structural transformations under cycling in relation to the electrochemical performance, impedance, electrical conductivity, and physico-mechanical stability. Multimode STEM imaging was carried out by simultaneously acquiring HAADF, BF, low-angle and medium-angle ADF signals (FIG. 1a-c) which contain a wealth of information regarding the scattering characteristics of the various phases present and display strong contrast variations [3]. We subsequently analyzed these data with MSA in order to isolate statistically significant differences between spatial regions and produce an unbiased phase classification (FIG.1d). STEM ET data was also collected using small convergence angles of the incident beam down to 4 mrad in order to keep the entire sample in focus even in very thick specimen areas up to several m (FIG. 1e). Ideally, by recording multiple imaging and spectroscopic signals and analyzing correlations between the signal intensity and structure and elemental compositions, one can noninvasively identify and ultimately quantify all phases (components) in such multiscale structures with complex multifractal morphology. High-tilt-angle STEM ET indicates that aggregated 20 nm to 60 nm onion-like Super C65 carbon particles form conductive 3D percolation networks over the cathode as well as extended mesoscale and nanoscale pore structures (FIG. 1f, g). Driftcorrected XEDS/EELS STEM spectral imaging (SI) (FIG. 2) allowed us to evaluate uniformity of compositional distributions within the copolymers and local elemental variations at electroactive poly(Sr-DIB) � carbon interfaces. In addition, as a result of the improved physico-mechanical stability of the composite cathodes as compared to the conventional sulfur-carbon cathode, structural analyses of the poly(S-r-DIB) � carbon interfaces indicate usually intimate contacts around the carbon nanoparticles which also display a short-range ordering with graphitic-like outer shells and mixed sp2/sp3-bonding. Microsc. Microanal. 21 (Suppl 3), 2015 144 References: [1] W.J. Chung, et al, Nature Chemistry 5 (2013) 518. [2] A.G. Simmonds, et al, J. ACS Macro Lett. 3 (2014), 229. [3] V.P. Oleshko, et al, Nanoscale 6 (2014) 11756. BF MAADF HAADF Poly(S-r-DIB) Voids Embedding media Super C65 carbons 500 nm a HAADF b HAADF c d Poly(S-r-DIB) Super C65 carbons g e f Figure 1. (a-d) Multimode STEM of a thin section of a poly(S-r-DIB10-wt%) copolymer � carbon composite cathode: (a) BF, (b) medium-angle ADF, (d) HAADF, (d) Composite phase image obtained by MSA of a-c. (e) Cs probe-corrected HAADF STEM and (f, g) high-tilt-angle ET of the cathode powder, a single frame extracted from a tilt series; (g) the reconstructed and segmented 3D view of the cathode structure showing random percolation networks of conductive carbons and extended pores. HAADF CK b c SL2,3 500 nm 1 Poly(S-r-DIB) Super C65 carbons 200 nm a Figure 2. (a) HAADF STEM and (b, c) EDX/EEL STEM SI of a thin section of the S-r-DIB10 % copolymer�carbon composite cathode. (b) X-ray spectral line profile along orange line in (a). The point 1 marked by red cross in (a) corresponds to black line in (b). (c) overlaid C K- and S L2,3 EELS maps acquired in the area marked by red box in (a).");sQ1[73]=new Array("../7337/0145.pdf","Structural Characterization of Powders and Thin Films of Layered Li1.2Mn0.55Ni0.15Co0.1O2 Cathode Materials","","145 doi:10.1017/S143192761500152X Paper No. 0073 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Structural Characterization of Powders and Thin Films of Layered Li1.2Mn0.55Ni0.15Co0.1O2 Cathode Materials Aaron C. Johnston-Peck1, K. Kamala Bharathi1,2, Saya Takeuchi1, Igor Levin1, Andrew A. Herzing1 and Leonid A. Bendersky1 1. Material Measurement Laboratory, National Institute of Standards and Technology, Gaithersburg, MD, 20899, USA 2. Department of Materials Science and Engineering, University of Maryland, College Park, Maryland, USA Layered, lithium and manganese rich transition metal oxides are becoming the next generation high energy cathode material for lithium ion batteries with superior performance including high capacity (>200 mAh/g), low cost, and better thermal stability. A thorough atomic-level structural understanding of these materials is needed to understand and improve their electrochemical properties. To this end, we have studied a layer cathode material with a composition of Li1.2Mn0.55Ni0.15Co0.1O2 (HE5050, TODA Inc.) [1]. A commercially available powder can provide a common platform for study in a field where differences of compositions and processing methods can make comparisons between studies difficult. In addition to the powders, pulse laser deposition (PLD) was used to grow epitaxial thin films on singlecrystal SrTiO3 substrates. The targets for growth were made from the HE5050 powder. The samples were characterized using multiple techniques that probe over a variety of length scales. High angle annular dark field scanning transmission electron microscopy (HAADF STEM) was used to characterize the local structure. Selected area electron diffraction (SAED) provides structural information from entire particles or grains while X-ray diffraction (XRD) complements these localized measurements with a global measure. Careful examination of these materials is necessary due to their complex structure, which can consist of multiple phases based on different ordering of Li and transition phase can form which is comprised of a metals on the same oxygen sublattice. A trigonal repeating layer sequence of a Li-O-M-O-Li. Yet, depending on preparation conditions and precise composition, a monoclinic C2/m phase may form where lithium atoms are incorporated in the transition metal layer in an ordered manner. In this case the stacking sequence is the same however the transition metal atoms are periodically replaced by Li atoms. The results of structural characterization reported in the literature have been mixed. Some observe a composite type structure [2] where small regions of trigonal and monoclinic structures coexist, while others report a solid solution of a single monoclinic phase [3]. Structural characterization based on diffraction results alone can be quite difficult because the monoclinic phase is an ordered structure derived from the parent trigonal structure. Defects may further convolute interpretation of diffraction data. HAADF-STEM, while limited in its ability to sample large areas, can unambiguously differentiate between the trigonal and monoclinic phase. When used together we can begin to create a structural model to describe the sample with an increased degree of confidence. Characterization of the particles from the HE5050 powder suggests that the structure is primarily in the monoclinic phase; both diffraction and HAADF-STEM support this designation (Figure 1). No long range ordering of the trigonal phase was observed. However, at the surfaces of the particles a spinel phase was found to form along specific facets. Measurements by energy dispersive X-ray spectroscopy (EDS) indicate the spinel phase was enriched in Ni relative the particle (Figure 2). Along other facets the spinel phase did not form but there was still a local increase in the concentration of transition metal Microsc. Microanal. 21 (Suppl 3), 2015 146 atoms. The thin films were successfully grown in the monoclinic phase. It was noted that the quality of the underlayer could influence the structure of the as-grown films and in certain cases the trigonal and spinel phases were observed in addition to the monoclinic. Characterization of the film samples before and after electrochemical testing revealed a change in the structure of the samples as well as a migration of the elements revealed by EDS. References: [1] Certain commercial equipment and materials are identified in this paper in order to specify adequately the experimental procedure. In no case does such identification imply recommendations by the National Institute of Standards and Technology nor does it imply that the material or equipment identified is necessarily the best available for this purpose. The HE5050 powder and sintered pellet for PLD were kindly provided by B. Polzin and A. Jansen from Argonne National Lab. [2] J. Bare�o et al., Chem. Mater. 23 (2011), p.2039. [3] K.A. Jarvis et al., Chem. Mater. 23 (2011), p.3614. Figure 1. Experimental (blue) and calculated (red) X-ray diffraction profiles, the calculated profile is for the monoclinic phase. HAADF STEM images of the powder and thin film samples. A representative HAADF STEM image along [1-10] revealing ordering in the transition metal atom layer, indicating formation of the monoclinc phase. Figure 2. HAADF STEM image of a particle edge that has reconstructed to spinel. EDS map demonstrates the spinel phase is Ni-enriched (Ni = green, Mn = blue).");sQ1[74]=new Array("../7337/0147.pdf","Application of a Semi-in-Lens FE-SEM to the Crystallographic Analysis with the EBSD Technique","","147 doi:10.1017/S1431927615001531 Paper No. 0074 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Application of a Semi-in-Lens FE-SEM to the Crystallographic Analysis with the EBSD Technique Hiroyuki Ito1, Yoichiro Hashimoto1, Shuichi Takeuchi1, Masahiro Sasajima2, Hirofumi Sato2, Hirobumi Morita3 Application Development Department, Science Systems Design Division, Hitachi High-Technologies, 1040, Ichige, Hitachinaka, Ibaraki, 312-0033, Japan 2. Electron Microscope Systems Design 1st Department, Science Systems Design Division, Hitachi High-Technologies, 882, Ichige, Hitachinaka, Ibaraki, 312-8504, Japan 3. NanoAnalysis Department, Oxford Instruments, IS Building, 3-32-42, Higashi-Shinagawa, Shinagawa-ku, Tokyo, 140-0002, Japan An FE-SEM is a powerful tool by its high resolution and high performance and is utilized widely in various fields of study. The electron back-scattered diffraction (EBSD) using a FE-SEM is one of the useful methods for the determination of crystallographic characteristics such as direction, symmetry, etc. of the specimen. In general, an out-lens type SEM is good for EBSD because the objective lens field is kept within the lens structure and the magnetic field will not conflict with the EBSD signal. Recently, in the field of material sciences the demand of the EBSD analysis of fine crystals increases. To satisfy both the image quality and the EBSD performance, we examined the EBSD analysis using a semi-in-lens cold FEG-SEM. Figure 1 is a general view of the SU8240 (a) and a high resolution image of meso-porous silica (b). The SU8200 series has various signal detection system, in which various kinds of signals such as SE, high and low angle BSE are simultaneously taken at the wide ranges of acceleration voltage and utilized them for characterization of samples [1]. In this study, Si (001) sample was used for a fundamental experiment. The EBSD was obtained with an Aztec-HKL EBSD system equipped with the distortion correcting function of EBSD pattern (Magnetic Field Correction: MFC). We examined the stability of the cold FEG for EBSD analysis and the degree of distortion of EBSD patterns compared with the data from out-lens FE-SEM. Figure 2 shows the relationship between working distance and the distortion of the EBSD patterns. The value of the distortion of EBSD patterns without correction became small as the working distance become longer, resulting in 5 degrees at 25 mm working distance. With correction, the distortion was within 2 degrees at all working distances. This indicates that MFC is useful to obtain the precise EBSD pattern and high resolution images. Also, this is consistent with the notion of the semi-in-lens field no longer enveloping the sample at the longer working distances. And then, the MFC correction is available to obtain the precise EBSD patterns. In order to examine the stability of cold FEG for EBSD analysis, we acquired EBSD map of Si (001) for 3 hours. Then, we extracted monochrome graduation of each time from band contrast map, and plotted it. The band contrast is expressed the intensity of the EBSD pattern in monochrome graduation and has value from 0 to 255. If the probe current greatly drops, the band contrast will drop as well. As shown in Figure3, the change of image contrast with the time process was almost flat and stable. The EBSD pattern intensity was stable to within 6 % and the software was able to accurately index the pattern throughout the acquisition period. Figure 4 is an application of EBSD analysis of gold crystal by SU8240. Herein, fine grains were clearly 1. Microsc. Microanal. 21 (Suppl 3), 2015 148 captured with contour lines (a). As shown in IPF maps at z (b) and x (c) directions, the crystal directions grains smaller than 50 nm were clearly visualized. These results suggest that a semi-in-lens SEM is available to the EBSD analysis of fine grains with MFC system. References: [1] S Takeuchi et al, Microsc. Microanal. Vol.19(Suppl 2) (2013) p.p.1310-1311. (a) (b) Figure 1. General view of the SU8240 and high resolution image taken by the SU8200 (a)General view of the SU8240, (b) High resolution image of mesoporous silica The degree of distorted EBSD pattern (degree) 15 without MFC with MFC 10 Band Contrast (Arb.unit) 250 200 150 100 5 50 0 15 20 25 Working Distance (mm) 0 0 30 60 90 120 Time (min.) 150 180 Figure 2. The relationship between working distance and distorted EBSD pattern angle (a) (b) Figure 3. The change of band contrast (c) 111 200 nm 200 nm 200 nm 001 101 Figure 4. EBSD map of fine gold grain sample (a) Band contrast map, (b)IPF map(Z), (c)IPF map(X)");sQ1[75]=new Array("../7337/0149.pdf","Optimized Solutions for the Arrangement of Digital Imaging Detectors","","149 doi:10.1017/S1431927615001543 Paper No. 0075 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Optimized Solutions for the Arrangement of Digital Imaging Detectors T. Hashimoto1, J. F. Mancuso2, K. Nakano3, E. Nakazawa1, L. Blubaugh4 and B. L. Armbruster5 1 2 Hitachi High-Technologies Corporation, Ibaraki 312-8504, Japan Advanced Microscopy Techniques Corporation, 242 W. Cummings Park, Woburn MA, USA 3 Hitachi High-Technologies Corporation, Kanagawa 213-0012, Japan 4 Hitachi High Technologies America, Inc., 22610 Gateway Center Dr., Clarksburg MD USA 5 Hitachi High Technologies America, Inc., 5960 Inglewood Dr., Pleasanton CA, USA The Hitachi HT7700 120kV TEM is based on a product concept that allows all observation and image recording procedures to be done in daylight with fully digitized image acquisition devices [1]. A wideangle 1 megapixel camera monitors at 15 fps a fluorescent screen in an observation chamber positioned just below the projector lens. The live images are displayed in a window within Hitachi's system control GUI. A second window displays the image from the main image recording camera, imaging from a phosphor located in a traditional bottom-mount configuration. This camera can be an 8 megapixel CCD camera, XR81-B, or an optional high-sensitivity 4 megapixel scientific CMOS camera, XR401L-B. The bottom-mount camera is used for image recording and for automation functions [2], such as auto-focusing, stigmation and alignment, drift correction, stage translation and image acquisition for montaging, and sequential specimen tilting and image acquisition for electron tomography. Excellent high contrast at low magnification with a wide field of view (FOV) is also crucial to identify features of interest in correlative light/electron microscopy experiments [3]. AMT's XR16-DIR (Figure 1) was developed to address specific application fields such as pathology, histology, or anatomy that can require both a wider FOV than provided by conventional bottommounted cameras and finer image definition. The XR16-DIR camera exploits a 16M pixel CCD and a custom-made, finite conjugate lens to utilize a 41 by 62 mm2 area of the CCD scintillator, which is optimized for high contrast observation at 80kV. Figure 2 depicts the camera layout on the TEM column. To ensure a wider FOV, the XR16-DIR is mounted directly under the HT7700 viewing chamber, "Flange (1)" as shown in Figure 2. Calculated practical image magnifications for the camera are about 60% of nominal magnifications displayed on the HT7700 monitor. The estimated image size of the XR16-DIR camera for the HT7700 at 500 times nominal magnification is about 140 x 210 �m on the specimen. It is almost 3 x 3 times wider than the area taken with a standard XR81-B camera single frame image and equivalent to the FOV of a 3� x 4" sheet of photographic film. Due to the larger number of pixels, the frame rate without binning is slower than that of the XR81-B. Thus auto focusing must be done with binned sub-areas. Such operational conditions are necessary to obtain higher quality images with a reduced noise component. Finally, in addition to a greatly expanded FOV, viewers experience brighter images for a given set of column conditions. Usability improves for application fields such as renal pathology, as exemplified in Figure 3. References: [1] H. Tanaka et al., Proc. 67th Annual Meeting of the Japanese Society of Microscopy, 16Apm_I1-3 (2011) p.8 [2] T. Hashimoto et al., Proceedings M&M2012, Phoenix (2012) p. 1280. [3] W.G. Janssen, H.H. Hanson and B.L. Armbruster, Proceedings M&M2014, Hartford (2014) p. 1104. Microsc. Microanal. 21 (Suppl 3), 2015 150 Figure 1. XR16-DIR camera configured to mount to the HT7700 column at Flange (1). Figure 2. A FOV comparison of XR16-DIR and XR81-B cameras at different column positions. AMT XR16-DIR Specification Number of pixels.................3,248 � 4,864 pixels Frame rate.....................8 frame/s (4�4 binning) FOV range........41.44 � 62.06 mm (on scintillator) Pixel size on CCD.........................7.4 � 7.4 �m Pixel size on specimen at 10 k�....2.111 �2.111 nm (Estimated, 1.083 � 1.083 nm if camera is XR81-B) Coupling.....................................Optical lens Scintillator.............Phosphor optimized for 80 kV a Figure 3. a) Characteristic ultrastructural features of Lupus nephritis including subendothelial deposits and tuboreticular inclusions were imaged on the HT7700 by means of the XR16-DIR camera, b) a subarea of (a) enlarged 4 times. Main specifications for the XR16-DIR camera are included above.");sQ1[76]=new Array("../7337/0151.pdf","Cooling Temperature Control System for the Cross Section Polisher","","151 doi:10.1017/S1431927615001555 Paper No. 0076 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Cooling Temperature Control System for the Cross Section Polisher Shogo Kataoka1, Munehiro Kozuka1, Tsuyoshi Wakasa2, Koji Todoroki1, Toru Kasai1, Tsutomu Negishi1, Mituhide Matusita1, Hideo Nisioka1, Toshiaki Suzuki1, and Natasha Erdman3 1. 2. IB Business Unit, JEOL Ltd., 3-1-2 Musashino, Akishima, Tokyo 196-8558 JAPAN SASM Design, JEOL Ltd., 3-1-2 Musashino, Akishima, Tokyo 196-8558 JAPAN 3. JEOL USA, Inc., 11 Dearborn Road, Peabody, MA 01960, USA The Cross Section Polisher (CP) is the instrument to prepare a cross section of a specimen using broad Ar ion beam irradiation with the shielding plate. In the traditional mechanical polishing, experienced skill is required in order to reduce mechanical strain. On the contrary, the CP realized easy preparation of high quality cross sections of specimens without mechanical strain. However, heat-sensitive specimens are easily broken during ion beam irradiation using CP, because ion beam gives heat to the specimen. Then, we had developed a specimen cooling mechanism for the CP using Liquid Nitrogen (LN 2 ) cooling, but specimen damage such as interface peeling sometimes happened, especially for low glass-transition temperature (Tg) materials such as adhesive bond and molding material, due to excess cooling by LN 2 . To solve this problem, we have developed temperature control system for the cooling CP to prevent excess cooling. In this presentation, we report an effectiveness of this system. Figure 1 shows the temperature of the specimen holder with setting temperature of -20 �C. In this result, specimen cooling is controlled by the cooling temperature control system and temperature fluctuation is regulated within �1 �C. Figure 2 shows SEM images for the lead solder cross sections using the cooling temperature controlled CP. Figure 2(a) shows the milling result without specimen cooling. Under this milling condition, voids are generated at boundaries between lead and tin. Figure 2(b), (c) and (d) show milling results with specimen cooling and assigned temperatures are -20 �C, -50 �C, and -84 �C, respectively. Milling at -20 �C, some voids are generated (Figure 2(b)). Milling at -50 �C or less, any voids can't be observed (Figure 2(c, d)). These results suggest that specimen cooling during ion beam milling prevents sample damages, and the cooling temperature control system optimizes the cooling temperature realizing prevention of excess cooling. Microsc. Microanal. 21 (Suppl 3), 2015 152 30 20 10 Temp [] Temperature of Specimen Holder 0 -10 -20 -30 -40 -50 0 0:00 Assigned Temperature 10 0:10 20 0:20 30 0:30 Time [min] 40 0:40 50 0:50 60 1:00 Figure 1. Temperature of the specimen holder using the cooling temperature control system (a) (b) (c) (d) Figure 2. SEM images of cross section specimens of lead solder prepared using the cooling CP. The temperature is regulated by the cooling temperature control system. (a)Not cooling, (b)Cooling at -20 �C, (c)Cooling at -50 �C, (d)Cooling at -84 �C");sQ1[77]=new Array("../7337/0153.pdf","Correlative SEM and Raman imaging of hot spots on SERS substrate","","153 doi:10.1017/S1431927615001567 Paper No. 0077 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Correlative SEM and Raman imaging of hot spots on SERS substrate R. V�a1, M. Kocman1 and J. Jiruse 1 1 TESCAN Brno, s.r.o., Libusina trida 1, Brno, Czech Republic Surface Enhanced Raman Spectroscopy (SERS) is becoming more popular due to large signal amplification of the Raman signal and therefore for a quick chemical analysis. The enhancement of the signal is caused by amplification of the light by the excitation of localized surface plasmon resonances. This light concentration occurs usually in the gaps, crevices, or sharp features of plasmonic materials, which contain metals with nanoscale features. The spots which enhance the field are called ,hot spots`, and their size varies between a few and several hundred nm. We were able to visualize hot spots using a recently developed correlative Scanning Electron (SEM) and Confocal Raman (CRM) microscopy [1] we call RISE. Visualization of exactly the same area of the SERS substrate in SEM and Raman is very challenging in two stand-alone systems since the plasmonic nanostructures on substrates are usually periodically arranged and similar to each other. The unique configuration of the RISE system lies in the direct navigation between SEM image and corresponding area for Raman spectra acquisition. Figure 1 shows overlaid SEM and Raman images of mercaptopyridine on silver SERS substrate with corresponding spectra [2]. Different enhancement factor of the Raman signal of mercaptopyridine can be seen due to variance of the plasmonic nanostructures on the periodically ordered, plasma etched, and silver-coated polystyrene spheres. The RISE system comprises a high-resolution SEM and CRM integrated into the vacuum chamber. The CRM is equipped with a green excitation laser (532 nm) which provides a lateral resolution of 360 nm and depth resolution of 750 nm, respectively. Thus the resolution of CRM is comparable with standalone Raman instruments and it is not compromised by integration with the SEM. The electron microscope used here comprises the immersion magnetic optics with a resolution of 1.4 nm at 1 kV and 1 nm at 15 kV [3], however, other types of electron columns are possible as well. The high resolution at low electron acceleration voltages makes it suitable for beam-sensitive samples. References: [1] J Jiruse et al, Journal of Vac. Sci. Technol. B 32 (2014), 06FC03. [2] L Stolcov� et al, Proceedings of Progr. in Elmag. Res. Symp. (PIERS), Stockholm (2013), 426. [3] J Jiruse et al, Microsc. Microanal. 19 Suppl. 2 (2013), 1302. [4] The research leading to these results has received funding from the European Union 7th Framework Program [FP7/2007-2013] under grant agreement n�280566, project UnivSEM. Microsc. Microanal. 21 (Suppl 3), 2015 154 Figure 1: A detailed view of SERS substrate with correlative SEM-Raman imaging. Left: Overlaid SEM and Raman micrographs, Right: Corresponding Raman spectra showing characteristic peaks of mercaptopyridine. Various colors correspond to the different enhancement factor of the spectra caused by variance of the plasmonic structures on polystyrene spheres.");sQ1[78]=new Array("../7337/0155.pdf","Electron Diffraction Phase Analysis and Pattern Simulations Using the ICDD Powder Diffraction File (PDF-4+)","","155 doi:10.1017/S1431927615001579 Paper No. 0078 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electron Diffraction Phase Analysis and Pattern Simulations Using the ICDD Powder Diffraction File (PDF-4+) Stacy D. Gates1, Kai Zhong1, Tom Blanton1 and Justin Blanton1 1. International Centre for Diffraction Data, Newtown Square, PA, USA. The Powder Diffraction File TM (PDF�) is a comprehensive materials database that is the key source of reference powder diffraction data for identification and analysis of materials; from natural minerals and high-tech ceramics, to metals and alloys, and pharmaceuticals. New developments in X-ray analysis, and advances in scientific research, have dramatically influenced the range of data present in the PDF. ICDD's premier database, PDF-4+, contains d-spacing, intensity, and hkl data used for phase identification, as well as atomic coordinates for more than 223,000 phases. In addition to being a reference database used for the analysis of traditional single phase X-ray diffraction data, the PDF-4+ also supports electron diffraction (ED) based analyses. Often in materials characterization, X-ray, electron and neutron diffraction are used to complement one another. Given the complimentary character of these techniques, it seemed only logical to compile this data into one comprehensive database, the PDF-4+. As a result, a suite of electron diffraction based simulation tools have been designed. Research scientists often use electron diffraction as a method for obtaining higher diffraction intensities of a material when X-ray scattering is limited. The increased intensity of an electron beam allows the user to investigate smaller samples and utilize a more focused beam to determine the atomic arrangement of crystals. However, the high energy beam can often promote multiple scattering interactions; making structure solution by electron diffraction alone relatively difficult compared with X-ray diffraction structure analysis. In support of the diffraction community, the ICDD has developed tools that enable users to stipulate the type of diffraction they want to reference. The PDF electron diffraction tools currently consists of selected area electron diffraction (SAED) patterns, electron backscatter diffraction (EBSD) patterns, 2-D ring patterns, and 1-D powder diffraction patterns (Figure 1). In addition, several data analysis features have been included that allow the user to obtain zone axis, spatial orientation, and crystallinity information (Figure 2). The simulations are derived using calculations based on the atomic parameters, electron scattering factors, and/or X-ray scattering factors specified in the PDF entry of interest [1]. Each simulation is interactive and allows the user to perform instantaneous alterations to the pattern by adjusting dynamic parameters such as zone axes, camera length, electron voltage, etc. The PDF-4+ electron diffraction tools streamline comparisons and contrasts of Xray and electron diffraction data in one place. PDF-4+ offers a variety of algorithms and simulation options that allow users to analyze electron diffraction data. In many cases, this provides a unique capability to analyze the difficult material analysis problems. This presentation depicts the capabilities of the various electron diffraction simulation tools and describes the process of identifying phases from electron diffraction data in PDF-4+. Currently, PDF-4+ data mining capabilities provide extensive filtering options to enhance the identification process, and the search/match engine, SIeve+, enables users to carry out vital search/match analyses using ED data [2]. Microsc. Microanal. 21 (Suppl 3), 2015 156 Figure 1: Electron diffraction simulations of (a) an amorphous, (b) a polycrystalline (ring pattern), or (c) a single-crystal (SAED pattern) material showing the variation of ED patterns due to crystallinity. EBSD simulations in (0 0 1) orientation (d) and (1 1 0) orientation (e) are depicted, as well as a 1-D simulated powder pattern (f). Figure 2: Various displays encountered when performing zone axis auto-indexing. Screen (a) shows the user's imported data, (b) shows the location of the center of the pattern and the list of possible choices for zone indexing (box on the right), and (c) shows the simulated data points (based on the indexed zone axis; box on the right) compared to the user's original data. References [1] Reid, J. et al., Microscopy Today, January (2011), p. 38. [2] Kabekkodu, S. N. et al., Acta Crystallographica Section B: Structural Science B58 (2002), p. 333.");sQ1[79]=new Array("../7337/0157.pdf","Reel-to-Reel Electron Microscopy: Latency-Free Continuous Imaging of Large Sample Volumes","","157 doi:10.1017/S1431927615001580 Paper No. 0079 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Reel-to-Reel Electron Microscopy: Latency-Free Continuous Imaging of Large Sample Volumes Christopher S. Own1, Matthew F. Murfitt1, Lawrence S. Own1, Derrick Brittain2, Nuno da Costa2, R. Clay Reid2, David G. C. Hildebrand3, Brett Graham3, & Wei-Chung Allen Lee3. 1 Voxa, Seattle, WA USA 2 Allen Institute for Brain Sciences, Neural Coding group, Seattle, WA USA 3 Harvard Medical School, Department of Neurobiology, Boston, MA USA The electron microscope (EM) provides exquisitely detailed information about structural arrangements of matter through its high native resolution, contrast and wide variety of available signals. This enables broad application across numerous fields, including physics, materials science, medicine, and biology. In some fields, especially biology, there is increasing need for quantification at smaller length scales and simultaneous demand for structural data from larger volumes. While new digital automated tools make it possible in some cases to investigate most of a 3 mm transmission EM sample, they remain ineffective for significantly larger volumes, for example whole-genome patterned DNA [1], neural circuits [2], silicon wafers, and histological arrays. This is due in large part to operating cost � measured in both dollars and hours � arising from extensive sample preparation / handling and dependence on skilled operators [3]. For this reason, applicability of high-resolution EM beyond laboratory research is mostly limited to niche areas (for example, nephrology and ciliary dyskinesia in clinical medicine). Further, researchers and clinicians increasingly turn to methods that take ensemble measurements such as low-cost genetic sequencing and mass spectrometry despite the richness and spatial precision of information available from EM. Published EM results are generally cherry-picked from dozens to hundreds of images of painstakingly-prepared samples taken over weeks to months. What if EM's were optimized such that every image coming from the machine was scientifically significant and "publication-ready"? High throughput requires new ways of looking at EM imaging. The EM imaging process is a packet system, and throughput can be defined as the amount of useful data retrieved from the system during the packet divided by the time taken to do so. To be an honest throughput number, the time taken should include all aspects of the experiment, including sample preparation and loading, machine setup time (and any downtime experienced), as well as the time spent in the microscope itself. Generally the time spent actually examining the sample in the microscope is tiny compared to these other steps, and the time spent acquiring scientifically significant data is an even smaller fraction. Microscopy needs to significantly change to meet the throughput needs of modern biology: high resolution imaging may transition from a seldom-used relatively small part of a lengthy process to an always-on, always available part of a long term, continuous acquisition chain lasting weeks, months or even years. In this paradigm, collecting scientifically significant images becomes the majority of the process, rather than a small part of an arduous march dominated by sample preparation, sample handling, and experiment design. For a microscope to operate efficiently in this paradigm, it needs to offer robust and reliable performance over very long timescales. Such automation has begun with scanning EM [4] and work in EM-based gene sequencing [5,6]. Transmitted electron detection, however, has the distinct advantages of higher resolution, greater signal to noise, and is faster as area detectors are readily parallelizable. Throughput would then not be measured in megapixels per second, but rather terapixels per week, or even petapixels per year. The microscope needs native advanced, intelligent, and pre-emptive diagnostics able to provide notification of any problems or service required, if possible before they even occur. Once microscopes reach this level of reliability and sustainability, it Microsc. Microanal. 21 (Suppl 3), 2015 158 is likely that other problem domains will start to find applications for such a service. At Voxa, we have been producing new tools to provide solutions of this type, and report here on a novel tape-based in-column, reel-to-reel sample handling system for EM imaging produced by Voxa called GridStageTM (Fig. 1). Voxa's GridStageTM is derived from high-throughput pipeline elements developed by Voxa, and enables high-throughput EM imaging with maximum reproducibility. This pipeline was previously used to enhance the effective throughput on Hitachi and Nion stock instruments by 4-5 orders of magnitude in a novel EM-based DNA sequencer [5]. GridStageTM is compatible with a variety of tape-based sample carriers such as the chip-carrier tape shown in Fig. 2, with capacity exceeding 10,000 samples per reel, and will be transformative for applications that demand reduced sample handling, large volumes with high spatial information density, and consistency from sample preparation through to analysis. GridStagesTM are currently retrofitted to 120 kV JEOL 1200EX microscopes augmented with custom high-speed TEM camera arrays (TEMCAs) developed by coauthors at Harvard Medical School [2], and together are being used to acquire massive maps of neural circuits from serial sections of neural tissue. Previously, the prospects for mapping mammalian brain connectivity at the synaptic level spanned thousands to tens of thousands of years with conventional tools, an impractical endeavor. GridStageTM combined with TEMCA enables consistent high-speed automated operation with a theoretical bandwidth exceeding 100 megapixels per second, and advancement between samples in seconds. Image acquisition, sample rastering, and sample advance are completely automated and optimized; the system is accessed remotely via the web through a lightweight Python-based programming interface, enabling full monitoring of acquisition progress and datastreams from unlimited users. With GridStageTM and other throughput-enhancing tools by us and others [7], whole brain synaptic-resolution structural maps could potentially be in reach within a decade [8]. References: [1] Payne AC, et al, PLoS ONE 8 (2013), 1-11. [2] Bock D, et al, Nature 471 (2011), 177-182. [3] Peddle C & Collinson L, Micron 61 (2014), 9-19. [4] Denk W & Horstmann H, PLoS Biology 2 (2004), 1900-1909. [5] Own C, et al, Microscopy & Microanalysis 19 (2013), 208-209. [6] Bell DC, et al, Microscopy & Microanalysis 18 (2012), 1049-1053. [7] Marx V, Nature 503 (2013), 147-152. [8] This work was supported by Voxa, NIH grant R21 NS085320, and through resources provided by the Allen Institute for Brain Science. Drs. W. Ho and K. Hayworth are thanked for many useful discussions and contributions to this work. Figure 1. Section view of GridStageTM reel-to-reel sample handing system and high speed rastering cartridge installed on JEOL 1200EX microscope column. The exploded view shows a reel. Tape advances from the feed reel (left) to the takeup reel (right). Figure 2. a) Example tape used to carry microfabricated sample support chips, b) section of microfabricated sample support chip featuring <5 nm thick polymer support films suspended over 2 x 20 um slots, designed for linear polymer imaging.");sQ1[80]=new Array("../7337/0159.pdf","The New Zeiss GeminiSEM 500 Meets the Needs of Challenging Biological Applications.","","159 doi:10.1017/S1431927615001592 Paper No. 0080 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Isabel Angert1, Christina Berger1 , Martin Edelmann1 , Robert Kirmse1 , Kirk Czymmek2 , Alexander Thesen1 1. 2. The New Zeiss GeminiSEM 500 Meets the Needs of Challenging Biological Applications. Carl Zeiss Microscopy GmbH, Carl-Zeiss-Stra�e 22, 73447 Oberkochen, Germany Carl Zeiss Microscopy, LLC, 1 Zeiss Drive, 10594 Thornwood, United States of America Modern field emission scanning electron microscopy (FE-SEM) techniques are increasingly important for studying structural properties of cells and tissue. Typical applications include imaging of tissue by serial block face or array tomography methods (AT) [1,2]. In serial block face imaging a thin layer of the embedded sample is removed either mechanically or with a focused ion beam. Afterwards the fresh block face is imaged. In contrast for AT serial sections of the sample are produced in advance and mounted on a solid support (i.e. silicon wafer, glass). The individual sections are investigated in the FESEM afterwards. The imaging modes can vary from secondary electrons (SE) over back scattered electrons (BSE) detection to scanning transmission electron microscopy (STEM). Key performance parameters for biological imaging are optimum resolution and contrast, surface sensitivity, control of sample charging as well as the need of speeding up the data acquisition for high-throughput imaging. The new ZEISS GeminiSEM500 meets these challenges with an improved low kV performance offering optimal resolution across the entire voltage range and a new variable pressure (VP) concept without significant loss in resolution when using in-lens detection. In addition GeminiSEM 500 features a broad detector portfolio ranging from efficient parallel in-lens SE and energy selective backscattered (EsB) detection over backscatter detection on a five sector diode detector to STEM detection on a diode based detector with real annular design (aSTEM). Here we demonstrate a variety of application examples that highlight the benefits of the new system. The new low kV lens in the GeminiSEM 500 enables optimum resolution without biasing the stage (e.g 1.1 nm @ 1 kV, 1.2 nm @ 500V). Lower electron energies are particularly important for blockface imaging or nano-tomography to avoid beam damage of deeper layers that will be exposed next. Using the new Zeiss backscattered detector even at 1.5 kV enables the detailed structural investigation of biological tissue with optimal contrast (Fig. 1A). Even at 50 V surface structures can be resolved in great details as shown at the example of an uncoated moth wing (Fig. 1B). Using such extremely low energies even the most challenging non-conductive samples can be investigated without charge accumulation. At higher beam energies needed to resolve structures below the surface the new NanoVP mode from Zeiss comes into play. The pressure in the specimen chamber can be adjusted from 1 to about 150 Pa and a differential pumping aperture is inserted below the objective lens which reduces the scattering of the electrons in the gas allowing in-lens detection (Fig. 1C) without considerable loss of resolution and increasing VPSE detection efficiency. Transmission imaging mode in the scanning electron microscope (STEM) is a widely applied method for imaging transmissible samples often resulting in higher resolution than classical SE or BSE imaging mode [3]. The ZEISS aSTEM detector provides excellent contrast and dynamic range (Fig. 1D) and due to its design with concentric diode layout it allows bright-field (BF), annular dark-field (DF) and highangle annular dark-field (HAADF) detection. The detail that can be obtained in STEM mode is close to the limit in the amount of detail that can be expected from conventionally prepared, chemically fixed and plastic embedded samples. Microsc. Microanal. 21 (Suppl 3), 2015 160 References: [1] P Schneider et al. Bone 49 (2011) p. 304-311. [2] M Kuwajima et al., PLoS One 8(3): e59573. (2013) [3] KD Micheva and SJ Smith, Neuron 55 (2007) p. 25-36. A B C D Figure 1. A: Image of brain tissue with a backscattered detector at 1.5 kV primary energy with 10 nm pixel size, scale 1�m. B: Uncoated moth wing imaged at 50 eV with in-lens detection. This shows highest surface sensitivity and imaging of non-conductive samples, scale 1�m. C: Uncoated moth wing imaged at 20 keV with in-lens detection. With the nanoVP mode the pressure at the sample was set to 50 Pa, scale 1 �m. D: STEM image of brain tissue at 28 kV in BF mode, scale 200 nm.");sQ1[81]=new Array("../7337/0161.pdf","Effective New Plasma Cleaning Strategies for Scanning Electron Microscopes and FIBs","","161 doi:10.1017/S1431927615001609 Paper No. 0081 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Effective New Plasma Cleaning Strategies for Scanning Electron Microscopes and FIBs Ronald Vane1 and Michael Cable1 1 XEI Scientific, Redwood City, CA 94063 Remote plasma cleaning can be performed on SEMs and FIBs at pressures below 100 mTorr (13.3 Pa) during direct pumping with a TMP (Turbo Molecular Pump). Previous studies 1 showed that low chamber pressures increase the rate of cleaning and the distances at which cleaning was observed. If the chamber pressure can be brought down to < 25 mTorr and the flow of plasma cleaning gas into the chamber down to <25 sccm then using the Evactron� remote plasma cleaners results in flowing afterglow cleaning2. With 20 sccm of air flow rate, cleaning rates are as much as 100X faster than those above 400 mTorr, and there is the ability to clean larger volumes. Most TMPs can be operated without overheating, but active cooling may be needed on some TMP models at higher flow rates. Low pressures increase cleaning speeds by increasing mean free paths and reducing the recombination rates of the oxygen radicals by three body collisions. Achieving lower pressures by reducing flow rates below 10 sccm starves the plasma of the gas needed to produce cleaning radicals. Plasmas operating below 1 mTorr in the chamber can be achieved, but if the pressure is 10-5 Torr, the concentration of O radicals is 0.01 of that produced at 1.0 mTorr. This remarkable result plus new plasma ignition methods at low pressures allow more effective cleaning strategies for large instrument vacuum chambers. XEI Scientific has developed a new plasma ignition technology, "Pop" ignition, that allows Evactron plasma cleaning to be started directly from very high vacuum chamber pressure (<1 mTorr) and then to operate in the 1 to 20 mTorr range with full speed turbo pump operation. The needed pressure changes are part of the ignition process and the evacuation controls of the instrument are untouched. Cleaning is very fast at this pressure as illustrated in Figures 1 and 2. Figure is an RGA spectrum taken on a 48 liter cylindrical chamber that has been contaminated by pumping on petroleum jelly overnight. It shows a typical hydrocarbon mass spectrum with a series of peaks 14 amu apart. In figure 2 the chamber has been cleaned with 2 one minute plasma cleans and all of the hydrocarbon signature peaks have been removed leaving only greatly reduced water and air peaks. Before this new technology was developed, classic Evactron plasma ignition and cleaning was done at roughing pump pressures. This was done to assure no damage to electron and ion guns because the gun valves would be closed and to avoid pumping through diffusion pumps. Because plasma cleaning was slow at these pressures some users cleaned for long periods to assure cleaning of large or very contaminated chambers.With the new technology short cleaning times of under 5 minutes are often sufficient to remove all carbon. This allows overnight cleaning to be avoided, and it is discouraged because of extended exposure of the instrument interior to O radicals. The "pop" ignition also allows the plasma to be turned off after cleaning a short (1-5 minutes) time, allowing a quick return to base pressure with the turbo pump to remove reaction products, and then restarting the plasma to repeat the cleaning cycle. While at base pressure any hidden hydrocarbons can degas and redistribute within the chamber due to long free path molecular flow where they can be removed in the next plasma cycle. Cycle plasma cleaning is shown to be a very effective way to get the chamber pristine using this new plasma cleaning technology at high vacuum. Microsc. Microanal. 21 (Suppl 3), 2015 162 References: 1. C.A. Morgan and R. Vane " Carbon Contamination removal in larger chambers with low-power plasma cleaning" Proceedings of SPIE, vol 8324, 9 March 2012 \\, pages 83242F-83242F-8, XP055160546, ISSN: 0277-786X, DOI: 10.1117/12.917786 2. Ronald Vane et al. "Advancements in Decontamination of Vacuum Systems Using Plasma Cleaning" Micrcoscopy and Microanalysis 2104 poster 566 Fig. 1 RGA mass spectrum of evaporated petroleum jelly (Vaseline) contamination. Series of peaks at show hydrocarbons Fig. 2 RGA mass spectrum at base pressure after two 1 minute cleanings with the Evactron ES model at 20W and 2 mTorr pressure in Chamber. Remaining peaks are from air and water.");sQ1[82]=new Array("../7337/0163.pdf","Ease of Use Solution for Fast and Automated TEM-Lamella Preparation.","","163 doi:10.1017/S1431927615001610 Paper No. 0082 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Ease of Use Solution for Fast and Automated TEM-Lamella Preparation. Tobias Volkenandt1, Giuseppe Pavia1, Ingo Schulmeyer1 and Martin Kienle1 1. Carl Zeiss Microscopy GmbH, Oberkochen, Germany The main application for Focused Ion Beam (FIB) microscopes still is to gain information from regions underneath the sample surface. To access these regions it is sometimes sufficient to mill a trench and image the cross-section surface. But in other cases more sophisticated investigations in terms of resolution or chemical analysis are needed. Then it is necessary to extract a piece of the material for analysis in a TEM (Transmission Electron Microscope), STEM (Scanning Transmission Electron Microscope) or a SEM (Scanning Electron Microscope) with STEM detector [1]. The preparation of such a so called TEM lamella can be time consuming since wide trenches have to be milled into the sample material to the complete target depth. The closer the beam gets towards the lamella surface the smaller FIB currents have to be used to avoid amorphization, which elongates the overall milling time. Additional steps like the deposition of a protection layer or the milling of a cut-out add time as well. An operator performing the workflow by hand would waste a lot of valuable working time watching the microscope and waiting for one step to finish before executing the next. This can be avoided by using intelligent software, that once set up executes all necessary steps automatically. With its last generation of Crossbeam microscopes Zeiss introduced SmartFIB, a new software for FIB control. Besides all kinds of conventional shape milling, SmartFIB also comes with easy to use, workflow oriented wizards for cross-section and TEM lamella preparation. User interaction is only needed during parameter setup and choosing regions of interest. The actual milling is carried out afterwards in a row and unattended. This allows over night or weekend runs, providing the operator with a bunch of prepared samples to continue working on. The new TEM lamella wizard of SmartFIB consists of four main steps that are fully customizable to fit actual needs. The first step is defining the region of interest and dimensions of the TEM lamella. Secondly, the deposition of a protection layer by using a fitted Gas Injection System (GIS) can be set up. Thickness, margins and deposition material can be chosen. The third step is milling of the trenches. These can be divided into several milling steps, like coarse, medium and fine milling. Each of these millings steps is definable with respect to FIB current, proximity and overlap. Additionally the operator can decide if the milling shall take place in loops which drastically reduces redeposition of sputtered material. It is also possible to define an additional stage tilt to assure plane parallel milling at the surface of the lamella. The last main step is the cut-out. It can be configured regarding its dimensions and shape. Links can be set to remain on both sides or just one, providing compatibility for the lift-out with any micro manipulator configuration. The complete lamella configuration can be saved and imported again for another session. This drastically speeds up the setup. The milling can be started right away or transferred to a process list together with further milling jobs. These can be spread all over not only the current sample but all samples mounted on the microscope stage at that time. There are no restrictions regarding sample height or orientation. During the milling an advanced drift correction algorithm assures that the chosen regions of interest are relocated and that the lamella stays within the field of view during necessary stage tilting. Microsc. Microanal. 21 (Suppl 3), 2015 164 Figure 1 shows a screenshot of SmartFIB illustrating the variety of options in the TEM lamella wizard to fit every need. The setup of such a lamella does not take longer than three minutes and can be even sped up by using previously exported layouts and parameters. The milling of a corresponding lamella can take between 10 and 60 minutes, depending on the chosen parameters and sample material. During this time no user interaction is necessary. Afterwards the TEM lamella is ready for lift-out and further polishing as illustrated in Figure 2. The polishing can be supported by live thickness measurements based on electron backscatter contrast to assure precise polishing including end-point detection at a chosen lamella thickness [2]. We illustrated the capabilities of the new SmartFIB software by Zeiss. It allows fast setup of multiple milling sites through an easy to use but powerful wizard. All layouts can be added to a process list which is processed automatically without the need of user interactions including GIS operations and stage tilting. SmartFIB is available for the new Crossbeam 540 and Crossbeam 340 microscopes. References: [1] LA Giannuzzi and FA Stevie, Micron 30 (1999), p. 197 [2] R Salzer and L Lechner, Microscopy and Microanalysis 18 (2012), p. 654. Figure 1. Screenshot of SmartFIB showing the TEM lamella wizard. Figure 2. Exemplary result of a single TEM lamella milled by SmartFIB in silicon.");sQ1[83]=new Array("../7337/0165.pdf","Correlative Light and Electron Microscopy for Materials Science","","165 doi:10.1017/S1431927615001622 Paper No. 0083 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Correlative Light and Electron Microscopy for Materials Science Jessica L. Riesterer1 and Gregor Heiss1 1. FEI Company, Hillsboro, OR, USA. Although a powerful approach, correlative light and electron microscopy (CLEM) for materials and life sciences has been historically challenging because of low throughput. Providing full morphological information with respect to chemistry and topography is the strength of electron microscopy. Fluorescence microscopy excels at labeling components with unmatched sensitivity and specificity; however, it lacks contextual information and has limited resolution. By imaging the same sample with both imaging modalities it is possible to combine the advantages and overcome the limitations of the individual microscopy techniques.1 FEI has recently introduced new solutions to simplify the CLEM workflow, and in turn, increase the experimental throughput; CorrSightTM, a dedicated light microscope providing CLEM-specific functionality and automation of important workflow steps with MAPSTM, a software tool bridging the modalities to increase ease of use and provide time savings by automating the data collection process. When coupled with SEM, FIB, and additional imaging modalities,2 these tools address different correlative workflows helping to optimize efficiency and data quality across the full range of CLEM experiments. Ruby (Al2O3 with trace Cr) in fuchsite (K(Al,Cr)3Si3O10(OH)2)3 was imaged using both FEI's CorrSight and Helios NanoLab 660 DualBeamTM. Ruby has long been known to fluoresce4 due to the trace amounts of Cr in solid solution. Using the Helios' through-lens detector (TLD) in BSE-mode coupled with the MAPS software, extreme low-voltage high-resolution data was collected over large areas. The use of Cu tape on the specimen edges serves two purposes: enhances conductivity to reduce charge and serves as a rough fiducial marker to assist in correlation and navigation. Following BSE imaging, the fluorescent properties of specific regions were investigated using the CorrSight and correlated BSE data. In figures 1 and 2, red in the fluorescence image corresponds to blue light, while blue indicates signal collected via reflection. The CorrSight, used to obtain reflective polarized light microscopy images, was equipped with a 20/80 beam splitter and excitation and emission polarizer. The excitation light of 480 nm was linearly polarized and reflected to the sample surface via a beam-splitter. The reflected light from the surface was redirected to the emission polarizer through the same beam-splitter. However, the emission beam-splitter was cross-polarized with respect to the excitation polarizer to enhance the signal from bi-refringent properties of the crystal. Once the instrumentation is correlated using a simple two-point alignment in MAPS, fracture planes, inclusions, and secondary phases can be identified quickly and easily by comparing imaging modalities. Correlated images can be used to identify Cr-rich regions and ruby inclusions much faster than EDS mapping over large surface areas. Correlation also aids in narrowing the search for particular elementaldependent events. The demonstrated workflow is shown to be viable over large regions of interest, i.e. several millimeters, in an automated way, leading to significant time savings. Microsc. Microanal. 21 (Suppl 3), 2015 166 References: [1] M.A. Snyder, et al, Micropor. Mesopor. Mat. 76 (2004), p. 29 [2] B. Van Leer and R. Passey, 18th International Microscopy Conference, Prague, ISBN 978-80-2606720-7 (2014) p. 181 [3] http://www.mindat.org/min-1617.html [4] C. S. Venkateswaran, Proc. IISc. 2 (1935), p. 459. [ Figure 1. Three imaging modalities, visible-light, fluorescence, and SEM, can be easily correlated using the MAPS workflow. Black scale bar is equal to 6 mm. Figure 2. By overlaying (center) data from SEM (left) and LM (right), fine structure and second-phase inclusions can be quickly identified. Black scale bar is equal to 600 m.");sQ1[84]=new Array("../7337/0167.pdf","Setup and Practical Applications of a pnCCD Based XRF System","","167 doi:10.1017/S1431927615001634 Paper No. 0084 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Setup and Practical Applications of a pnCCD Based XRF System Jeffrey M. Davis1,* , Julia Schmidt2 , Martin Huth2 , Sebastian Ihle2 , Daniel Steigenh�fer2 , Peter Holl2 , Gerhard Lutz2 , Udo Weber1 , Adrian Niculae1 , Heike Soltau1 , Lothar Str�der2 1. 2. PNDetector GmbH, Otto-Hahn-Ring 6, 81739 M�nchen, Germany PNSensor GmbH, Otto-Hahn-Ring 6, 81739 M�nchen, Germany Micro-focused X-ray fluorescence (�XRF) analysis has proven very useful as a complementary X-ray imaging technique [1]. The technique requires a focused X-ray beam, a stage capable of moving in a raster pattern, and an X-ray spectrometer (normally a silicon drift detector, SDD). Unfortunately, images are built pixel-by-pixel, using the stages to bring the sample under the stationary beam. This means that the time required to create an image is limited by the speed of the stage, not the detector. High speed stages tend to have poor position reproducibility, and accurate stages tend to be slower and more expensive. To eliminate this limitation, a new method of X-ray imaging must be established. In optical experiments, the incoming light is focused with an optic onto an imaging device. This device records all of the information simultaneously, greatly improving the speed with which an image is produced. It preserves both the position and the energy information of each incoming photon, making it an ideal imaging device. Unfortunately, these imagers are not normally sensitive to X-rays. However, a new type of CCD, known as the pnCCD is capable of detecting both the position and energy of each X-ray event. The pnCCD was originally designed for use in X-ray telescopes [2]. More recently, these large, position sensitive, energy dispersive detectors have been used for experiments at various synchrotrons and free electron lasers [3]. Finally, the device has been applied to �XRF imaging, showcasing fast, large area imaging [4]. The pnCCD is capable of recording energy dispersive X-ray data at every pixel in a 264 x 264 pixel array. Each pixel in the array has a size of 48 �m, but sub-pixel resolution is possible down to a resolution of 10 �m. Much like in optical imaging, the optic is placed in front of the pnCCD, rather than in front of the X-ray source. As a spectrometer, the X-ray energy resolution at each pixel is equivalent to 150 eV at Mn K-alpha, and the pnCCD can count X-rays at a rate of over 200 000 counts/s. Ultimately, these properties reduce the time required to create X-ray images. Yet the question remains: How does one build an XRF based on a pnCCD spectrometer? The goal of this project was to create a laboratory scale �XRF using the pnCCD as the spectrometer. Figure 1 shows the setup of our system, which uses a 50 W X-ray source and two stepper motor stages. The X-ray source is aligned at grazing incidence, allowing for quasi-total reflection imaging. Once the hardware is setup, one must also process the data coming from the pnCCD. Although the pnCCD is perfectly capable of producing an X-ray Spectrum Image (XSI), the software and algorithms used to create it are different from the software used with an SDD based system. Due to the speed of the system, the pnCCD will produce data at a rate of 10 GB/min or more. These raw data are processed to create the XSI through a series of programs. Through these programs, phase analysis, quantitative analysis, noise reduction and sub-pixel resolution are all possible. An example X-ray image is shown as Figure 2, which demonstrates the high resolution imaging capabilities of the detector. The pnCCD �XRF system presents both hardware and software challenges, but the solution to these problems represents a significant improvement in �XRF technology. This work will discuss the challenges and opportunities of a pnCCD based �XRF system along with the practical applications of the system. Microsc. Microanal. 21 (Suppl 3), 2015 168 References: [1] Davis, J.M., Newbury, D.E., Ritchie, N.W.M., et al, Bridging the micro to macro map: A new application for milli-probe X-ray fluorescence, Microscopy and Microanalysis 17 (2011), p. 410-417. [2] Str�der, L., Briel, U., Dennerl, K., et al, The European photon imaging camera on XMM-Newton: The pn-CCD camera, Astronomy and Astrophysics, 365 (2001) L18 [3] Wiesemann, U., Thieme, J., Guttmann, P., et al, The new scanning transmission X-ray microscope at BESSY II, Proc. 6th International Conf. on X-ray Microscopy, American Institute of Physics, (2000) [4] Scharf, O., Ihle, S., Ordavo, I., et al, Compact pnCCD-based X-ray camera with high spatial and energy resolution: A color X-ray camera, Analytical Chemistry, 83 (2011), p. 2532:2538. Figure 1: A schematic drawing of the setup for a pnCCD based �XRF system. The X-ray source is unfocused and polychromatic, but the fluoresced X-rays from the sample are focused using a parallel beam, polycapillary optic. A stage allows for fine adjustment of the area imaged, but it is not used to create the images. Figure 2: An Fe X-ray intensity image produced using the pnCCD based �XRF system. The sample is an ordinary Fe-Ni meteorite measuring approximately 2 cm in length and 1.3 cm in height. The scale bar on the lower left portion of the image represents 3 mm.");sQ1[85]=new Array("../7337/0169.pdf","Electrical Probing and Current Imaging for Failure Analysis in the SEM/FIB.","","169 doi:10.1017/S1431927615001646 Paper No. 0085 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electrical Probing and Current Imaging for Failure Analysis in the SEM/FIB. Stephan Kleindiek1, Andreas Rummel1, Klaus Schock1, Gregor Renka1, Matthias Kemmler1 1 Kleindiek Nanotechnik, Reutlingen, Germany As the transistors and other circuits in semiconductor devices shrink to smaller and smaller sizes, failure analysis on these devices is becoming more and more challenging. There is a large need for highly precise nano-probers to quickly and reliably address contacts in the range of few tens of nanometers. Using piezo-driven micromanipulators, individual nano-scale components on semiconductor devices can be tested inside a SEM or FIB system. The SEM is primarily used for imaging the nano-scale contacts. Due to the multi-layer nature of modern semiconductor products, the FIB is used to reveal the layer of interest, deposit conductive, resistive, or insulating patches for modification of the circuit, as well as clean the sample to remove contamination or oxide layers in order to gain access to the structure of interest. A SEM/FIB system is also important for shaping (and re-shaping) as well as cleaning the probe tips, which are used to contact the nano-scale structures of interest. Next to actually contacting the sample, various tools are necessary to successfully characterize the structure of interest efficiently. Safely landing tips � without bending them � as well as ensuring that they are in low-ohmic contact with the substrate is important for reliable measurements. Further, methods such as EBIC and EBAC can be used to gather information on buried structures; this facilitates locating the area of interest. Finally, Current Imaging, a method for imaging the sample's current response to a biased needle that is swept over its surface, is described. In this work, we demonstrate the successful characterization of state of the art 22 nm transistors using the methods described above. Microsc. Microanal. 21 (Suppl 3), 2015 170 Figure 1. Four probes in contact with two transistors. Figure 2. Acquired curves from above transistors. Both transistors share the same gate (9 o'clock tip) as well as the same drain (6 o'clock tip). The n-MOS source is contacted by the tip at 3 o'clock, the pMOS source by the tip at 12 o'clock.");sQ1[86]=new Array("../7337/0171.pdf","Electrical Probing and Current Imaging for Failure Analysis in the SEM/FIB.","","171 doi:10.1017/S1431927615001658 Paper No. 0086 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Principal Component Analysis in EDS D.L. West ThermoFisher Scientific Inc., 81 Wyman Street, Waltham, MA 02451 Spectral imaging is a very powerful microanalysis tool for chemical phase identification in a variety of samples. However, a spectral imaging (SI) data set contains vary large amounts of information and traditional approaches to analyzing these data sets rely on element identification and X-ray mapping. These methods have the potential to misidentify or completely overlook minor phases. To efficiently analyze these large amounts of data, automated methods are needed. To be useful for routine analysis, these methods must make no assumptions about the sample chemistry, be fairly robust, work with low volume or sparse data sets and deal with noisy data. The microanalysis software program, COMPASS, automatically performs data reduction and analysis on SI data sets and conforms to these requirements [1, 2]. Compass is based on multivariate statistics and the application of multivariate statistics, multivariate analysis. Multivariate statistics is a form of statistics surrounding the simultaneous observation and analysis of more than one variable in a data set. Multivariate statistics is a subset of linear algebra which is the branch of mathematics concerning vector spaces and linear mappings between such spaces [3]. Within multivariate statistics there are a number of types of data analysis, one of which is principal component analysis (PCA). Principal component analysis creates a new set of orthogonal variables that contain the same information as the original data set but rotates the axes of variables to give a new set of orthogonal axes which are ordered such that they summarize decreasing proportions of the variation (in mathematics, orthogonality is the relation of two lines at right angles to one another, and the generalization of this relation into n dimensions; and to a variety of mathematical relations thought of as describing non-overlapping, uncorrelated, or independent objects of some kind). Principal component analysis was developed by Karl Pearson in 1901[4] to show how variables act as the decisive factors in the way a data set collectively varies. PCA is a powerful statistical technique to determine the relationship between variables within a data set and reduce the data to a smaller set that describes those variables which contribute to the overall influence of these variables to the data set. That is, PCA uses sophisticated mathematical principles to transform a number of possibly correlated variables into a smaller number of variables called principal components. References 1) This software is covered by U.S. patents 6,684,413 and 6,675,106. 2) P.G. Kotula, J.R. Michael and D.E. Newbury, (2007), Micros Microanal 13(Suppl 2), 1372-1373. 3) H. Anton (2005), Elementary Linear Algebra (Applications Version) (9th ed.), Wiley International 4) Pearson, K. (1901). "On Lines and Planes of Closest Fit to Systems of Points is Space". Philosophical Magazine Series 6 2 (11): 559�572. Microsc. Microanal. 21 (Suppl 3), 2015 172");sQ1[87]=new Array("../7337/0173.pdf","A Software for Lab Access and Chemical Process Logging in Shared Facilities","","173 doi:10.1017/S143192761500166X Paper No. 0087 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A Software for Lab Access and Chemical Process Logging in Shared Facilities Shuyou Li1 and Fernando Camino2 1 2 FOM Networks, Inc. One Northfield Plaza, Suite 300, Northfield, IL 60093-1214, USA Center for Functional Nanomaterials, Brookhaven National Laboratory, Upton, NY 11973, USA Research safety management, including safety training, documentation, and daily logging, is always a concern especially in multidisciplinary shared facilities. Two years ago in the Microscopy and Microanalysis meeting (Indianapolis, 2013), we presented a software system we designed for Northwestern University for university-level safety management. With that safety management system, Northwestern University shared facilities can make sure every facility user, no matter which department the user comes from, must pass the required safety training before they can use the resources in the laboratory. This safety management system was at instrument level, i.e. an instrument manager may define which safety clearances are required for each individual resource in the laboratory [1]. In this paper we present another system we developed for the Center for Functional Nanomaterials, Brookhaven National Laboratory. As an add-on to the BNL's Facility Online Manager (FOM�) system [2], this software provides the lab access and daily chemical process logging in the cleanroom. Figure 1 is a screenshot of the access control system. The access control station is usually located beside the main entrance of the laboratory, where every user must scan their ID cards to enter the lab. After scanning an ID card, the user's information is recorded in the main FOM� server and the laboratory can check this record or bill the user based on the time this user works inside the lab. Each user is also asked a random safety question every time they enter the lab. If a user fails any part of the safety questionnaire, the user will be banned from the lab and their name will be listed in the table of "Users who must see a staff member". The logging of the chemical processes is done on a second computer inside the lab, where a user should record all chemical usage in the lab, as shown in Figure 2. Users can also check out safety documents such as the SOP or MSDS of any chemical at any time while working in the lab. As soon as a user starts a chemical process, the record is shown in the first lab, where the lab manager or another user may check whenever needed. After doing a chemical process, the user must click the End Process button to finish using a chemical. Otherwise their name will be listed in the "Users who forgot to END chemical process". A lab manager can also see the list of current users in the lab, records of chemical usage. Lab managers can also manage users and resources and edit the tip of the week, recent changes, and safety questionnaire items on their office computers using the main FOM� system. References: [1] S. Li, S.V. Mallipeddi, S. Karlman, T. Moskal and V.P. Dravid (2013). Safety Management in Multidisciplinary Shared Facilities. Microscopy and Microanalysis, 19 (Suppl. 2), pp 1396-1397. [2] See more details about the Facility Online Manager� software at http://www.fomnetworks.com/. Microsc. Microanal. 21 (Suppl 3), 2015 174 Figure 1. A screenshot showing the access control and chemical processes log station in a cleanroom. Figure 2. A screenshot showing the chemical logging system in a cleanroom.");sQ1[88]=new Array("../7337/0175.pdf","Do-It-Yourself Database for Core Facility Management","","175 doi:10.1017/S1431927615001671 Paper No. 0088 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Do-It-Yourself Database for Core Facility Management Ru-ching Hsia1 1 Electron Microscopy Core Imaging Facility, University of Maryland Baltimore, Baltimore, USA. Core facility directors and staff who are often highly trained scientists face the unique challenge of managing the day-to-day operation of the facility [1]. A recent survey reported that core facility managers spend an average of 55 hours per month on administrative tasks [2]. Although several core facility management software packages are available commercially, these programs are often costly and may require the purchase of an annual maintenance contract at a substantial price. Nearly 50% of the core facilities that responded to the survey did not use an electronic management system specifically designed for core facility management. Here, I present a do-it-yourself database management system that can be custom-designed and implemented by core facility personnel without prior software programming training. The Electron Microscopy Core Imaging Facility (EMCIF) of the University of Maryland Baltimore is a mid size core facility that serves an active user base of ~100, performs 150 to 200 research projects annually and generates 60-100 GB of image data. The EMCIF management database uses Microsoft Office Access, a program included in the Microsoft Office Professional suite. The EMCIF day-to-day records are divided in several main tables such as client information, project information, project charges, project experiments, etc. Records in the tables are safely stored in a web-based server networked to all computers in the facility. Each facility client and project is given a unique identifier which is cross-referenced throughout the database via queries and defined relationships. Many command buttons for performing simple tasks such as opening and printing forms are already included in the Access program. Using these built-in command buttons, invoices, quotes and project summaries can be printed as reports, exported to other programs, or e-mailed to clients with a single click of the mouse. Figure 1 shows the switchboard that the EMCIF staff use on a daily basis to perform all tasks including conducting service projects, facility management, user communication, billing and purchasing (Figure 1A, B C and D). The EMCIF database was initially constructed in 2009 with five tables to manage service records. It has been expanded and updated regularly based on staff feedback and growth of the facility. The database now holds records for approximately 1000 projects and more than 300 clients; each record can be retrieved using custom-built search engines within seconds (Figure 1D). This database allows the EMCIF director and staff members to track down information concerning a given EMCIF project workflow, spending or income in real time. Data entry is simplified by specific entry forms designed for a given task with built-in drop-down menus to minimize typing (Figure 2). A significant advantage of this `DIY' database is its ease to adapt or expand in order to fit EMCIF and clients' changing needs, an option often not available in commercial software. Web-based surveys or registration forms can be seamlessly linked to the record table stored on the server, thus easing the task of collecting client information. Finally, it is also worth noting that many resources such as Access Database Template, tutorials and user forums are available to assist in the construction of a core facility management database that is tailored to the facility's needs. References: Microsc. Microanal. 21 (Suppl 3), 2015 176 [1] H Wallrabe, A Periasamy and M Elangovan, Microscopy Today 22 (2014), p.36. [2] iLabSolutions, The 2014 Core Facility Benchmarking Study. Figure 1. The switchboard of EMCIF database allows staff to keep track of all day-to-day tasks. The switchboard is divided into submenus (A), command buttons for data entry (B) and command buttons to generate reports and invoices (C). Search page with built-in search engine to retrieve or view data by keywords or specific criteria (D) are also accessible from the switchboard. Figure 2. Different entry forms are used for data entry of different tasks. Data for the same project can be entered simultaneously from different location and by different personnel. For instance, service charges for project O-15 are entered by EMCIF staff in project charge entry form (A). Scope time used by the client, Berry de Manding for project O-15 is entered directly from the scope PC to the database by using scope use entry form (B).");sQ1[89]=new Array("../7337/0177.pdf","High Performance Remote Electron Microscopy","","177 doi:10.1017/S1431927615001683 Paper No. 0089 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 High Performance Remote Electron Microscopy Daniel E. Huber1, Jonathan Orsborn1, Frank Scheltens1, Dave W. McComb1, Hamish L. Fraser1 1. Center for Electron Microscopy and Analysis, Materials Science and Engineering, The Ohio State University, Columbus, OH, U.S.A. Efforts to explore the feasibility and practicality of operating microscopy instruments from afar have existed for years in various forms [1-4]. The goal of these efforts is to enable the highest possible utilization of a scarce resource--powerful, expensive electron microscopes and related analytical instrumentation. Universities and research organizations require the use of electron microscopes, but may not have the facilities or the capital to acquire and maintain the modern, high-performance, corrected TEM instruments. Past efforts driving remote access sought to utilize users in different time zones to enable a Follow-the-sun work flow [4]. The focus of the current effort is to enable tools to expand local user base and build a critical mass of users by fostering high quality collaborations. Facilitating remote operation to teach and train large groups of students, simultaneously [5-7] helps to build this user base and expand exposure to characterization techniques. Whether an off-campus outreach effort, or a classroom demonstration, these educational efforts seek to build excitement for science, and show the wonderment of "seeing the unseen" to students of all ages. Often an understated benefit of remote operation is assisting staff and researchers in operation and maintenance of the instruments, thus allowing personnel to better utilize facility resources. This last benefit may be a key consideration for a facility deciding to allocate funding to enable such resources. Frequently, software-based tools such as VNC or Remote Desktop Connection are utilized to achieve remote operation of microscopes, thanks to their ubiquitous nature and low cost [8]. These tools can be invaluable for staff and researchers as they execute long-duration experiments and monitor tedious maintenance operations related to high-vacuum systems, etc. However, this approach has limitations; the high bandwidth and low latency video necessary for delicate operations, such as alignments, require real-time interaction and cannot be properly facilitated by current software solutions. Complicating matters even further, is the continuously changing nature of the software industry, with the eventual retirement of platforms, such as Microsoft Windows XPTM, and inevitable obsolescence of future platforms. Also, security issues not only present a prominent obstacle for current procedures related to remote operation of instruments, but also threaten future remote microscopy efforts. The Center for Electron Microscopy and Analysis (CEMAS), at The Ohio State University, is currently using several technologies for operation of TEM, SEM, and Dual-Beam FIBTM instruments. The facility is also evaluating new and different methods, while exploring multiple remote-operation scenarios. Accessing the facility across both private networks and the public Internet. These techniques have been constantly evolving, since first being implemented in 2007. Ever increasing network performance further enables these remote microscopy efforts and expands the viable range of practical application. A discussion of past lessons-learned and the current state-of-the-art will follow as we describe the implementation of high-performance remote TEM consoles across the state of Ohio using OARnet--the Ohio Academic Resource network. With the implementation of several remote consoles, CEMAS seeks to demonstrate production quality remote operation, act as a hub for research and a regional microscopy resource. Microsc. Microanal. 21 (Suppl 3), 2015 178 References: [1] N.J. Zaluzec, Teleconference Mag. 17 (1998) [2] G.Y. Fan, et al, Ultramicroscopy 52 (1993), p. 499-503 [3] K. Furuya, et al, Microscopy & Microanalysis 11 (2005), p. 68-69 [4] G.M. Brown, et al, Microcopy & Microanalysis 15 (2009), p. 1102-1103 [5] J.F. Mansfield, et al, Microscopy & Microanalysis 6 (2000), p. 31-41 [6] J.F. Mansfield, Microscopy & Microanalysis 14 (2008), p. 876-877 [7] T.C. Isabell, et al, Microscopy & Microanalysis 14 (2008), p. 872-873 [8] T.C. Isabell, et al, Microscopy & Microanalysis 17 (2011), p. 868-869");sQ1[90]=new Array("../7337/0179.pdf","Lock-n-Sync for Secure Data Storage in Shared Electron Microscopy Facilities","","179 doi:10.1017/S1431927615001695 Paper No. 0090 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Lock-n-Sync for Secure Data Storage in Shared Electron Microscopy Facilities Kuilin Li and Shuyou Li FOM Networks, Inc. One Northfield Plaza, Suite 300, Northfield, IL 60093-1214. Over the past two decades, nearly all electron microscopy facilities have finished the transition from analog imaging to digital imaging. Instead of working in the darkroom and struggling with stacks and stacks of negative films, more and more of our data is in the form of digital micrographs stored on our computers. While enjoying the convenience of digital imaging, we are facing new challenges of data storage and data security. In fear of computer viruses and internet hackers, a lot of EM managers have disconnected their instrument-control PCs from the internet. Since the USB flash drive is getting more and more popular, they have sealed USB ports to prevent flash drives from spreading viruses. In some laboratories, EM users are often forced to use CD or DVD discs to retrieve data from the lab after the experiments. Some other labs have come to the next level and implemented a local area network (LAN) to store EM data on a local file server. For the convenience of data retrieval and data backup, the experimental data, in most shared EM facilities, is transparent to all the facility users. Any user can peek or make copies of another user's data at any time. Some cautious users would remove their data from the shared file server as soon as the data is copied to their personal devices, then they lost the centralized back up of raw experimental data. This is apparently against the rule of thumb in data security. In the FDA's Code of Federal Regulations on Electronic Records (CFR 21 part 11), for Electronic Records, "...Such procedures and controls shall include... protection of records to enable their accurate and ready retrieval throughout the records retention period and limiting system access to authorized individuals..." [1]. In this work we demonstrated our best of practices to store, transmit, and secure the large data sets in a shared facility. Our procedures include a seamless integration of the FOM� online calendar with an access control software called Lock-n-Sync [2]. With this system, only authorized users are able to reserve time and use the microscope; only the user who booked time on the microscope may operate the instrument; and the operator may only view and store data in his/her own designated space. All user data collected during operation is automatically synchronized and backed up under the specific user's account on a shared server. The Lock-n-Sync software may also be used with any institutional level single-sign-on protocol, such as Active Directory, CAS, SAML, Shibboleth, and so on. A typical use case of the FOM� online calendar and the Lock-n-Sync software is depicted in Figure 1, which includes Step (a). A user reserves instrument time with the FOM� online calendar, typically one week before the experiment time. The instrument manager has the right to grant user's access levels, define reservation rules, and billing policies where applicable. Step (b). Before using the reserved instrument, a user must login to Lock-n-Sync system, which is installed on the instrument-control computer. The Lock-n-Sync system then authenticates the user using the FOM central database. Microsc. Microanal. 21 (Suppl 3), 2015 180 Step (c). During the experiment, the Lock-n-Sync software monitors all the data that this user collects on the microscope and automatically synchronizes it with the remote file server under this user's profile. Step (d). After the experiment, a user may access the data stored on the central data storage via the FOM� file browser. By default a user may only browse their own data. A user may also appoint another FOM� user as a collaborator and grant them access to some of their data. References: [1] FDA Code of Federal Regulations, Electronic Records; Electronic Signatures, 21 C.F.R. � 11.10 (2014), available at http://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfCFR/CFRSearch.cfm?CFRPart=11&showFR=1 [2] Lock-n-Sync is a patent-pending software by FOM Networks, Inc. For more information please contact Dr. Shuyou Li at FOM Networks, Inc. Figure 1. A typical use case of FOM� calendar and Lock-n-Sync system in an electron microscopy laboratory.");sQ1[91]=new Array("../7337/0181.pdf","Characterization of Crude Oil Emulsions by Cryo Electron Microscopies","","181 doi:10.1017/S1431927615001701 Paper No. 0091 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of Crude Oil Emulsions by Cryo Electron Microscopies Ali R. Behzad and Dalaver H. Anjum Imaging and Characterization Core Lab, King Abdullah University of Science & Technology (KAUST), Thuwal, Makkah 23955, Kingdom of Saudi Arabia (KSA) Water in oil forms emulsions during crude oil production and degrades the product quality as well as poses a technical challenge to oil industry. In order to achieve a specified product quality, the oil/water emulsion formation must be resolved. One of the most important features of a crude oil emulsion is its droplet size distribution (DSD). The DSDs determine the stability of emulsions and hence must be considered when choosing optimized treatment protocols [1]. There are various methods including nearinfrared light, optical microscopy, confocal laser scanning microscopy and nuclear magnetic resonance that are being used currently to characterize the DSDs of emulsions [2]. However, limited spatial resolution and/or lack of direct visualization of droplets diminish reliability and quality of measurements made by these techniques. Electron microscopy provides a method for the direct visualization of structures with highest spatial resolutions and therefore can be used in determining the DSDs of emulsion droplets more accurately. We used cryo scanning electron microscopy (cryoSEM) and cryo transmission electron microscopy (cryoTEM) to examine crude oil emulsions. A Nova NanoSEM (FEI Company) equipped with a cryo preparation chamber (Quorum Technology) was used for imaging the emulsions droplets in natural hydrated state. Prior to imaging, the emulsions were cryo-fixed, freeze fractured and then sublimed to reveal the internal structure of droplets. A Quanta 3D dual beam SEM (FEI Company) equipped a cryopreparation chamber was used to prepare cross sections of cryo-fixed emulsions free of artifacts for examining the droplets at low temperatures. The emulsions were also investigated with a cryoTEM instrument of model Titan G2 80-300CT S/TEM (FEI Company) to find the presence of deep submicron size emulsion droplets and to determine the nature of water and oil mixing in those droplets. This was achieved by freezing the emulsion-loaded QuantifoilTM copper grids comparatively slowly in order to have the water content solidified to crystalline ice as supposed to vitrified ice. The crystalline ice in the emulsion droplets will exhibit the diffraction or modulating image contrast. CryoSEM imaging of the freeze fractured plane of emulsion showed the presence of droplets that ranged from 200 to 5 �m in diameters (Fig. 1A). Freeze fracturing of emulsions followed by sublimation revealed a complex meshed-network of organic materials that formed the internal structure of droplets. CryoSEM analysis also indicated that the water was confined within this network. Due to rough fractured planes of emulsions after freeze fracturing, it was not possible to observe the droplets smaller than 5.0 �m in diameter. Emulsion cross sections prepared by cryo-FIB milling showed the presence of droplets that ranged from 10 to 1 �m in diameters (Fig. 1B). FIB-milled emulsion droplets after sublimation showed a meshed network of organic materials that was similar to that of larger ones (200-5 �m). Application of cryoTEM in bright-field TEM mode on emulsion revealed the presence of droplets that were smaller than 1.0 �m in diameter (Fig. 2). The contrast modulations in the droplets were result of electron diffraction by crystalline phase of water in the droplets. In summary, we demonstrated that the cryo electron microscopies are suitable to determine the DSDs of emulsions droplets. We were able to characterize droplets ranging from many tens of microns to submicron in diameters. The cryoSEM imaging after freeze-fracturing and cryoFIB milling showed the Microsc. Microanal. 21 (Suppl 3), 2015 182 presence of water and organic phases in the droplets. Finally, cryoTEM confirmed the water confinement in submicron size droplets. References: [1] S. Kokal, SPE Productions and Facilities, (2005), p6. [2] G. Van Dalen. Journal of Microscopy, 208(2002), p116. 1A 1B 300 �m 4 �m Fig. 1: CryoSEM images of emulsion droplets after freeze fracture (A) and FIB-milling (B) showing droplets ranging from 200 to 1 �m in diameters. 2A 3 1 2B water-rich phase 2 200 nm 400 nm Fig. 2: CryoTEM images of emulsion droplets in bright-field-TEM mode using 300 keV energy electron beam. (A) Submicron size droplets (labeled as "1" and "3") are suitable for TEM analysis. (B): A higher magnification of boxed area in A showing water-rich phase within the droplet (enclosed by "red" lines).");sQ1[92]=new Array("../7337/0183.pdf","Characterization of Crude Oil Emulsions by Cryo Electron Microscopies Polyhydroxyalkanoates in Stress Response of Bacteria","","183 doi:10.1017/S1431927615001713 Paper No. 0092 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Cryo-SEM and Raman Spectroscopy Study of the Involvement of Polyhydroxyalkanoates in Stress Response of Bacteria V. Krzyzanek1, K. Hrubanova1, O. Samek1, S. Obruca2, I. Marova2, S. Bernatova1, M. Siler1, P. Zemanek1 1. 2. Institute of Scientific Instruments of the ASCR, v.v.i., Brno, Czech Republic. Centre for Material Research, Brno University of Technology, Brno, Czech Republic. Many bacterial strains are capable of accumulation of very high amounts of polyhydroxyalkanoates (PHAs) in cytoplasm. Consequently, increasing content of the macromolecular material will strongly affect the overall material properties of cells and it is also likely that high content of PHAs inside the cells will also influence the rest of the cell volume � the cytoplasm. PHAs does not serve only as a carbon source in starvation, but presence of PHAs granules in cytoplasm represent additional advantage under stress conditions. It has an impact on metabolic activity of the cells and, moreover, it changes cells morphology as well as physico-chemical properties of the bacterial cytoplasm and cells. The main aim of our investigations is to utilize combination of modern methods, such as electron and Raman microscopy [1] in the complex study on changes in physico-chemical properties of bacterial cells and cell cytoplasm with respect to intracellular content of PHA. In particular, we aim to study the influence of PHAs presence in the cell on its mechanical properties, mobility of solutes in cytoplasm, and thermal behavior. For this behavior of cells with various contents of PHAs is compared and referenced to the PHA none-producing mutant. For this purpose, we have utilized following bacterial strains Cupriavidus necator H16 and also its PHA none-producing mutant C. necator PHB. . Further, we investigated Burholderia cepacia, Burholderia sacchari and Pseudomonas aeruginosa, which are capable of PHAs accumulation as well as biofilm formation. For cryo scanning electron microscopy (cryo-SEM) a thin piece of sample was quickly frozen in liquid nitrogen, moved into a vacuum chamber (ACE600, Leica Microsystems) where it was freeze-fractured and shortly sublimated at -95�C. In the next step, the sample was moved at high vacuum using a shuttle (VCT100, Leica Microsystems) into the SEM (Magellan, FEI) equipped with a cold stage and the fractured structure was observed with 1 keV electron beam at -135�C without any metal coating. During freeze-fractured experiments on bacteria containing polyhydroxybutyrate (PHB) we observed interesting morphological features which showed needle/mushroom-type deformations (Fig.1a). Moreover, for the pure copolymer of 3-hydroxybutyrate and 3-hydroxyvalerate (P(HB-co-HV)) sample similar features were also observed (Fig.1b). This behavior was observed previously by Sudesh et al. [2] and explained by the heat which is generated during fracturing process which is considered to be sufficient to introduce the needle/mushroom-type morphology for PHA granules in the cells. Raman micro-spectroscopic experiments with above mentioned cells were carried out using Renishaw system. Fig. 2 shows a typical Raman spectrum, here for Cupriavidus necator H16 cultured directly on a Petri dish. Raman spectroscopy seems to be very efficient tool for direct analysis of intracellular PHA content of bacteria and, furthermore, can be combined with SEM measurements to make it possible to measure nutrient dynamics and metabolism of the cells. Moreover, with this type of sensing one may detect variability of polymers contained within the cells which could help in elucidating different morphological features during the process of cryo-SEM technique. Microsc. Microanal. 21 (Suppl 3), 2015 184 We believe that our study will be of significant assistance to research group being involved in bacterial strains which accumulate PHB and P(HB-co-HV) and for investigations of this material properties under low temperatures. Our results are convincing enough to warrant more extensive investigations with larger sets of bacterial strains to evaluate combination of cryo-SEM and Raman spectroscopy. References: [1] O. Samek et al, Sensors 10 (2010), p. 8635. [2] K. Sudesh et al, Can J Microbiol 46 (2000), p. 304. [3] This work received support from the Ministry of Health, Ministry of Education, Youth and Sports of the Czech Republic (LO1212), together with the European Commission (ALISI No. CZ.1.05/2.1.00/01.0017) and the Grant Agency of the Czech Republic (GACR GP15-20645S, GA14-20012S). a b Figure 1. a). Cryo-SEM image of Cupriavidus necator H16. Needle-type deformations can be clearly seen within the fractured sample surface. b) Cryo-SEM image of pure sample of P(HB-co-HV). This sample shows needle-type deformation. Figure 2. Raman spectra of Cupriavidus necator H16. The most prominent contributions are collected in the Table.");sQ1[93]=new Array("../7337/0185.pdf","Comparison of Sample Preparation Methods for Analysis of Mucus-Secreting Colon Cancer Cells by Scanning Electron Microscopy","","185 doi:10.1017/S1431927615001725 Paper No. 0093 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Comparison of Sample Preparation Methods for Analysis of Mucus-Secreting Colon Cancer Cells by Scanning Electron Microscopy Xin Wang1,2, Reiner Bleher1,2, Vadim Backman3, Gajendra S. Shekhawat1,2, Vinayak P. Dravid1,2 1. Northwestern University Atomic and Nanoscale Characterization Experimental Center, (NUANCE), Northwestern University, Evanston, IL 60208, USA 2. Department of Materials Science and Engineering, Northwestern University, Evanston, IL 60208, USA 3. Department of Biomedical Engineering, Northwestern University, Evanston, IL 60208, USA Scanning electron microscopy (SEM) is a well-established technique for obtaining morphological information of biological samples with high spatial resolution. However, proper sample preparation is essential to avoid or minimize sample alterations, e.g. shrinkage, molecular structure collapse, and loss of soluble components during processing. Chemical fixation is most commonly used to achieve meaningful results for analyses of initially hydrated biological samples with structural and molecular integrity under the high vacuum condition required for SEM. In this study, we compared the cellular surface structure of mucus-secreting colon cancer cells prepared by critical point drying (CPD), by turbo freeze drying (TFD), or with highpressure freezing and cryo SEM, which is the current gold standard for the preservation of native sample morphology. In order to investigate the differences in appearance between results obtained by CPD, TFD, and cryo SEM, we have chosen mucus-secreting human colon cancer cells (CSK, shRNA transfected HT29 cells [1]) as a model system. Cells were seeded on gelatin-coated cover slips ( = 12 mm) in culture medium, grown for less than 24 hours, and washed three times in sterile D-PBS. For CPD, cells were fixed in glutaraldehyde (2.5%), formaldehyde (2%) and tannic acid (0.5%) in PBS for 30 min at room temperature. The cells were briefly rinsed three times with D-PBS and then three times with DI water. Cells were dehydrated through a series of ascending ethanol concentrations of 25%, 50%, 75%, 90% and 100% for 5 min each, followed by critical point drying (Samdri-795, Tousimis ). The TFD samples were chemically fixed and rinsed similarly as the CPD samples before plunge-freezing in liquid ethane (Vitrobot Mark III, FEI) and turbo freeze drying (K775X, Emitech) over a 6-hour period in a vacuum. CPD and TFD samples were coated with osmium (OPC60A, Filgen) and observed in a SEM (S4800-II, Hitachi) at an accelerating voltage of 5 kV. For cryo-SEM, cells were cultivated in a T-25 flask, trypsinized, diluted into 4 mL of culture medium and centrifuged at 100�g for 10 min to form a pellet. The supernatant was aspirated and the cells were resuspended in a small volume of PBS. 3 mm sample carriers with 150 �m recessions (Technotrade) were filled with the cell suspension, mounted, and high-pressure frozen (HPM100, Leica). The frozen samples were freeze-fractured at -120o C, etched for 6 min at -105o C and coated with platinum in a high vacuum cryo coater (ACE600, Leica). Samples were observed at -120o C in a SEM (S4800-II, Hitachi) equipped with a cryo stage at an accelerating voltage of 2 kV. SEM images of CPD and TFD cell surfaces showed overall comparable cell surface morphology with a dense brush of microvilli and few mucus conglomerations. The microvilli in the CPD samples appeared to be marginally smaller in diameter (d = 94.0 � 11.7 nm, n = 10), more numerous, and clustered in bundles, compared to the individually spaced microvilli in the TFD samples (d = 103.4 � 18 nm, n = 10). In both samples, the plasma membranes were well Microsc. Microanal. 21 (Suppl 3), 2015 186 preserved with no visible membrane damage, such as breakage or pitting. However, rapid freezing is used to provide superior ultrastructure and to preserve the cellular surface closer to the native state than any chemical fixation procedure [2]. In the cryo-SEM samples of highpressure frozen cells, only the upper regions of individual microvilli were visible, and the surface of the cells appeared to be coated with an even layer of mucus. The microvilli were slightly larger in diameter as in the TFD samples (d = 115.2 � 22 nm, n = 10). When cells were plungefrozen and freeze dried without prior chemical fixation, microvilli were not distinguishable at all (not shown). In summary, the high-pressure frozen cryo-SEM samples presented evenly distributed mucus on the cell surface as opposed to localized mucus conglomerations on CPD and TFD samples. Also, the mean diameters of microvilli were largest in the cryo-SEM samples and smallest in the CPD samples. References: [1] D Damania, H Subramanian, A Tiwari, Y Stypula, D Kunte, P Pradhan, H Roy and V Backman, Biophysical Journal (2010) 99: 989-996. [2] H Schatten, Micron (2011) 42: 175-185. [3] L Denault et al., Microscopy & Microanalysis 14 (Suppl 2), 2008. [4] Electron Probe Microscope studies in this work made use of EPIC (Electron Probe Instrumentation Center) facilities at NUANCE (Northwestern University's Atomic and Nanoscale Characterization Experimental Center), which has received support from the MRSEC program (NSF DMR-1121262) at the Materials Research Center, and the Nanoscale Science and Engineering Center (EEC-0118025/003), both programs of the National Science Foundation, the State of Illinois and Northwestern University. Figure 1. SEM images of human colon cancer cells (CSK) prepared by chemical fixation and CPD at magnifications of 5,000x (a) and 10,000x (b), TFD sample with pre-fixation by glutaraldehyde, formaldehyde, and tannic acid at 5,000x (c) and 10,000x (d), and a freeze-fractured cryo-SEM sample at 5,000x (e) and 10,000x (f).");sQ1[94]=new Array("../7337/0187.pdf","Monitoring of Multilayered Bacterial Biofilm Morphology by Cryo-SEM for Raman Spectroscopy Measurements","","187 doi:10.1017/S1431927615001737 Paper No. 0094 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Monitoring of Multilayered Bacterial Biofilm Morphology by Cryo-SEM for Raman Spectroscopy Measurements K. Hrubanova1,2, S. Bernatova1, O. Samek1, M. Sery1,, P. Zemanek1, J. Nebesarova3, F. Ruzicka4, V. Krzyzanek1 1. 2. Institute of Scientific Instruments of the ASCR, v.v.i., Brno, Czech Republic. Brno University of Technology, Institute of Physical Engineering, Brno, Czech Republic. 3. Biology center ASCR, Institute of Parasitology, Ceske Budejovice, Czech Republic. 4. Department of Microbiology, Faculty of Medicine, Masaryk University and St. Anne's Faculty Hospital, Brno, Czech Republic. Staphylococcus epidermidis has been recently recognized as an important cause of serious biofilm infections associated with implanted medical devices. In presented work the multi-layered biofilms formed by these microorganisms were observed by scanning electron microscope (SEM), in particular with using freeze-fracturing technique. The freeze-fracture technique consists of physically breaking apart (fracturing) a rapidly frozen biological sample; structural detail exposed by the fracture plane is then visualized by metal deposition. An optional step, involving vacuum sublimation of ice, may be carried out after fracturing. Freeze fracture is unique among electron microscopic techniques in providing planar views of the internal organization of membranes or biofilms. Deep etching of ultrarapidly frozen samples permits visualization of the surface structure of cells and their components. Our sample was fractured after rapid-freezing into liquid nitrogen, the air-water was removed by short freeze-drying at Alto 2500 chamber (Gatan), sputtered by 5 nm Pt-Pd and finally imaged at low temperature in the field emission SEM JSM 7401F (JEOL). When characterizing biofilms using spectroscopic techniques, and specifically Raman spectroscopy, a common approach is to analyze biofilm as a whole (cells embedded in extracellular matrix). Such spectra can be acquired point by-point at selected positions of individual colonies or using line-scan techniques such as e.g. Renishaw StreamLine. This approach collects data from many cells including the contribution from extracellular matrix. Therefore it is important to established contribution which translates to Raman spectra from extracellular matrix for reliable data presentation. This is crucial for understanding of the processes involved in cells embedded in the biofilm matrix. Such cells are well known to express phenotypes that differ from those of their planktonic counterparts. Moreover, they display specific properties including an increased resistance to chemical agent treatments. Raman spectroscopy employs a laser beam that is focused with a microscope objective in order to excite and collect Raman scattering from a small volume of the sample. A typical Raman spectrum from living microorganism contains a wealth of spectral peaks corresponding to unique interatomic vibrations in biomolecules e.g. nucleic acids, proteins, carbohydrates, and lipids [1]. In order to study how the extracellular matrix, that makes up the biofilm, translate into the Raman spectral "fingerprint" of the bacterial cells we have evaluated cryo-SEM image obtained from biofilmforming Staphylococcus epidermidis (Figure 1a). Using the detailed image (Figure 1b) we estimated that contribution of extracellular matrix is of about 40% of total volume. It means that when collecting Raman spectra we have not only contribution from bacterial cells of interest but also large contribution from the matrix. Thus, when analyzing cells embedded in extracellular matrix its contribution has to be taken into account to prevent misleading information about the cells. We show that combination of Microsc. Microanal. 21 (Suppl 3), 2015 188 Raman spectroscopy with cryo-SEM can provide deeper inside into the chemistry and composition of biofilms. Such studies involving influence of variations in the amount of extracellular material � which depends on cultivations conditions and bacterial strain under investigation � are currently under way in our laboratories, exploiting combination of cryo-SEM and Raman spectroscopy techniques. References: [1] Bernatov�, S. et al, Molecules 18 (2013), p. 13188. [2] This work received support from the Ministry of Health, Ministry of Education, Youth and Sports of the Czech Republic (LO1212), together with the European Commission and the Czech Science Foundation (ALISI No. CZ.1.05/2.1.00/01.0017) and the Grant Agency of the Czech Republic (GA14-20012S). a b Figure 1. (a) cryo-SEM image of a biofilm-forming bacterial colony obtained by fracturing of biofilm on a glass coverslip. In order to demonstrate the influence of biofilm on Raman spectra the beam spot size employed for Raman spectroscopy analysis is shown (the beam spot size is of about 3 x 20 �m). b) detail of the cells embedded in the extracellular matrix shows that the ratio of cells to extracellular matrix is about 60:40, respectively.");sQ1[95]=new Array("../7337/0189.pdf","Morphology of Organic Matrix of Human Enamel","","189 doi:10.1017/S1431927615001749 Paper No. 0095 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Morphology of Organic Matrix of Human Enamel V.M. Dusevich1, J. McGuire1, J.P. Gorski1, Y. Wang1, M.P. Walker1 1. School of Dentistry, University of Missouri � Kansas City, Kansas City, MO 64108 Adult human enamel is a highly mineralized tissue and contains just about 2% protein by volume, with the rest mostly of a hydroxyapatite [1]. Recently it was shown that protein distribution in enamel is not uniform: insoluble proteins concentrate in a band of 100-400 microns width which extends from dentin into enamel body and form an organic matrix [2]. Dentin � enamel junction (DEJ) has a complex organization and plays an important role in structural strength of a tooth; it transfers applied loads from enamel to dentin and arrest cracks initiating in enamel. Enamel organic matrix may strengthen DEJ by stabilizing inner layer of enamel and increasing its bonding to dentin [3]. However, properties and morphology of organic matrix are still insufficiently known. In this work morphology of organic matrix was studied by means of electron microscopy (SEM and TEM). Specimens were carefully demineralized (not to disturb delicate matrix), fixed with glutaraldehyde, dehydrated and some of them (for TEM study) were embedded in a resin. The important morphological feature of an organic matrix is that it represents a fine meshwork of small fibers of about 20 nm in diameter (Fig. 1 a, b). There is a strong indication that these are fibers of type VII collagen which can complex with other collagens and may contribute to the structural resilience of enamel and play a role in bonding enamel to dentin [4]. Crystals of hydroxyapatite in human enamel are long but narrow (20-80 nm in cross section); they are organized in parallel bundles, so called "enamel rods," of about 4-8 microns in diameter. Morphologically organic matrix is following the shape of these rods. Though organic matter is distributed throughout rod, most often it is concentrated in sheath regions (Fig. 1 c, d), and less frequently sheath regions are depleted of organic matter (Fig 1 e). Organic matrix has highly variable structure and changes its morphology even in the same tooth frequently. Figure 1f shows a region in which sheath regions are changing from enriched by organic matter (on the right) to depleted (on the left). Utilization of charge contrast in ESEM [5] provides another way of observation of organic matrix. Areas of enamel enriched by organic matter are darker. Figure 1g presents regions of enamel where organic matrix concentrated on rods boundaries, and Figure 1h shows regions where matrix was concentrated in rods bodies. References [1] [2] [3] [4] [5] [6] D.F. Poole, N.W. Johnson, Arch Oral Biol. 12(12) (1967) p. 1621-34. V. Dusevich et al, Arch Oral Biol, 57(12) (2012), p. 1585-94. J. McGuire et al, Connective Tissue Research, 55 (S1) (2014), p. 33-37. J. McGuire et al, Bone, 63(2014), p. 29-35. Griffin, B.J. (1997) Microscopy and Microanalysis, 3 (S2) (1997) 1197�1198. This research was supported by NIH/NIDCR grant R01DE021462. Microsc. Microanal. 21 (Suppl 3), 2015 190 Figure 1. Elements of an organic matrix of human enamel. a) SEM, bar 200 nm; b) TEM, bar 0.5 um; c) SEM, bar 10 um; d) TEM, bar 2 um; e) SEM, bar 2 um; f) TEM, bar 5 um; g) ESEM, bar 20 um; h) ESEM, bar 20 um");sQ1[96]=new Array("../7337/0191.pdf","Analysis of Glass Vial Interior Surfaces in Parenteral Drug Stability Studies","","191 doi:10.1017/S1431927615001750 Paper No. 0096 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Analysis of Glass Vial Interior Surfaces in Parenteral Drug Stability Studies R. A. Carlton GlaxoSmithKline, Collegeville, Pa, 19426 Most injectable pharmaceutical products (parenterals) are packaged in glass vials or syringes. Glass is an ideal packaging material due to its chemical compatibility with most parenteral products. There are, however, some parenteral products that can induce or accelerate glass corrosion processes [1]. Glass corrosion can result in the generation of glass related particulates in the drug product which is clearly undesirable. Various optical and electron microscopy techniques have been applied to this problem along with several other analytical methodologies [2]. There is an expectation by regulatory authorities that pharmaceutical manufacturers will understand the propensity of their product to cause corrosion and to monitor corrosion on accelerated stability studies [3]. Evaluation of incoming and treated (washing followed by depyrogenation) vials before filling with drug product is a key element of these glass corrosion evaluations. These initial studies serve as the standard against which progressive glass corrosion will be judged. Two of the more common microscopy techniques applied to glass vial interior surface studies are Differential Interference Contrast (DIC) optical microscopy and Scanning Electron Microscopy (SEM). DIC has the advantages of examination of the whole vial with minimal sample preparation whereas SEM has the advantage of high resolution images even though the glass vial must be sectioned for analysis. In practice, DIC is used to locate possible regions of corrosion or of glass imperfections in the vial followed by breaking or cutting the glass in these regions and mounting on SEM stubs for examination. For tubular glass vials, the neck and heel of the vial are the most vulnerable regions for corrosion due to the method of vial manufacture. The manufacturing process can result in various kinds of glass imperfections including pits and divots. These imperfections can be interpreted as glass corrosion in stability studies unless incoming vials are well characterized. Figures 1a and 1b show DIC images of the heel region of two different vial types before cleaning and filling. Pits and other interior glass imperfections can be seen clearly in the microscope, although difficult to photograph well. Figure 2a and 2b also show SEM images of imperfections from two different vial types before cleaning and filling. Figure 2A shows low level pitting while Figure 2B shows a more severe level of pitting and surface imperfections. In product stability trials, imperfections associated with Figure 2A often show little change over many (>24) months at accelerated temperature and humidity conditions. Imperfections associated with Figure 2B, however, often lead to evidence of progressive corrosion in only a few months stability storage. Without thorough examination of incoming vials, the evidence of progressive corrosion may implicate the drug product whereas the glass itself may be the main factor. Finally, it should be noted that microscopy is one part of a larger study of glass corrosion. The filled vials need to be examined for the presence of particles and the solution itself should be examined for dissolved glass elements (Si, etc.) by elemental analysis techniques such as inductively coupled plasma spectroscopy (ICP) before a judgment of progressive glass corrosion can be rendered. Microsc. Microanal. 21 (Suppl 3), 2015 192 REFERENCES [1] Iacocca R.G., Allgeier A., J Mater Sci 42, 2007 pp. 801-811. [2] Carlton R.A., Micros & Microanal 19, Suppl 2, 2013 pp 284-285. [3] United States Pharmacopeia, Chapters 1660 and 660, 2014 1A 1B Fig. 1 DIC Images of Initial, Unfilled Glass Vials Image Size = 720 x 500 �m 2A 2B Fig. 2 SEM Images of Initial, Unfilled Glass Vials");sQ1[97]=new Array("../7337/0193.pdf","ESEM Study of Magnetic Bubbles � Surface Morphological and Fluidic Aspects","","193 doi:10.1017/S1431927615001762 Paper No. 0097 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 ESEM Study of Magnetic Bubbles � Surface Morphological and Fluidic Aspects Zahava Barkay1, Yu Gao2, Chenjie Xu2 and Claus-Dieter Ohl3 1. 2. Wolfson Applied Materials Research Center, Tel-Aviv University, Tel-Aviv 6997801, Israel Division of Bioengineering, School of Chemical and Biomedical Engineering, Nanyang Technological University, Singapore 637457, Singapore 3. Division of Physics and Applied Physics, School of Physical and Mathematical Sciences, Nanyang Technological University, Singapore 637371, Singapore Liquid droplets encapsulated by hydrophobic powder, i.e. liquid marbles, were intensively studied over the last few decades due to their unique physical properties and applications [1]. Alternative structures of gas bubbles with nanoparticle shell have recently gained much attention for their compression properties and potential usage in a large variety of biological applications. The current work has focused on magnetic microbubbles with self-assembled nanoparticle shell for drug delivery applications. The preparation of similar magnetic bubbles was previously reported and their dynamics and manipulation in acoustic and magnetic fields was explored using optical microscopy [2]. The current study uses multiple modes of high resolution electron microscopy at wet and dry environmental conditions for characterization of innovative magnetic microbubbles with iron and silica self-assembled nanoparticle shell [2-3]. The results include: surface morphology and composition of bubble shell, characterization of bubble self-assembly, bubble stability under various environmental conditions, and in-situ dynamic imaging of bubble flow including triple line characterization. The imaging was carried out in a controlled temperature-pressure environment using Quanta 200FEG environmental scanning electron microscope (ESEM) as shown elsewhere [4]. Bubble saturated liquid droplets were examined in ESEM wet mode following a dehydration process for evaluating the bubble stability during liquid evaporation. The analysis of bubbles after dehydration was based on comparing the secondary electron (SE) mode, which provided surface morphology at various sample tilt angles, with the scanning transmission electron microscopy (STEM) mode [5]. The agglomeration of microbubbles (each of 4.2�0.9�m size) into chain-like shapes of tens micron length, provided indication for self-assembled bubble process in addition to the known self-assembly of shell particles (Figs. 1a-c). The existence of both silica and iron particles on the bubble surface was verified by EDS (energy dispersive spectroscopy) analysis. The size was 39�9nm and 340�80nm respectively for iron and silica shell particles and their relative surface arrangement was thus revealed (Figs. 1d-f). In-situ dynamic wet-mode ESEM imaging of bubble flow within the droplet showed that bubble selfassembly occurred already at the liquid state before dehydration. The bubbles were transported outward the droplet, which facilitated their self-assembly at the droplet edge as could be expected by the coffee ring effect [6]. Some bubbles were popping out of the droplet surface as shown at Fig. 2a, while most bubbles remained at the liquid volume and thus were revealed after liquid evaporation. The droplet triple line was examined showing bubble ejection out and air-pocket formation (Figs. 2b-c). The distortion of triple line shape and the inhomogeneous triple line pinning by bubbles were particularly significant at the droplet evaporation process. Previous ESEM research on liquid droplets on textured substrates showed similar air-pockets at the triple line [4]. However, the current study referred to droplets on smooth silicon wafer substrates and thus the distortion of triple line structure was attributed to the Microsc. Microanal. 21 (Suppl 3), 2015 194 intrinsic bubble-liquid properties. The described ESEM study of the magnetic microbubbles referred to two related aspects (morphological and fluidic), which should be further quantitatively studied and considered for high efficiency targeted drug delivery and for lab-on-chip device applications [7]. References: [1] P. G. de Gennes et al, "Capillarity and wetting phenomena", (Springer, NY) p.226. [2] X. Zhao et al, Phys. Rev. Lett. 102 (2009), p. 024501. [3] Y. Gao et al, J. Mater. Chem. B. in press (2015). [4] E. Bormashenko et al, Langmuir 29 (2013), p. 14163. [5] Z. Barkay et al, Micron 40 (2009), p. 480. [6] R. D. Deegan et al, Nature 389 (1977), p. 827. [7] The authors acknowledge the Materials and Nano centers at Tel-Aviv University for feasibility study at the ESEM laboratory. Figure 1. ESEM of self-assembled microbubbles: (a) bright-field STEM mode, (b) SE mode at zero tilt, (c) SE mode at 60 tilt ((a)-(c) scale bar 10�m), (d) a single bubble with silica particles indicated by an arrow (scale bar 2�m), (e) iron particles, (f) silica particles ((e)-(f) scale bar 500nm). Figure 2. ESEM "wet-mode" in-situ dynamic study: (a) microbubbles popping out of the liquid drop (indicated by arrows, scale bar 20�m), (b) a microbubble ejected (top arrow) and air-pockets at the triple line (two bottom arrows, scale bar 20�m), (c) the microbubble after ejection with mainly iron particles and some silica particles on top (indicated by an arrow, scale bar 2�m).");sQ1[98]=new Array("../7337/0195.pdf","Measurement of radiation-induced autophagic flux in human brain cancer cells","","195 doi:10.1017/S1431927615001774 Paper No. 0098 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Measurement of radiation-induced autophagic flux in human brain cancer cells Linda S. Yasui1, Megan Johnson1, Hannah Savage1, Abraham Baker1, Elle Baker1, Colton Burns1 1. Northern Illinois University, Department of Biological Sciences, DeKalb, IL, USA Radiation-induced autophagy mediates susceptibility to radiation [1]. However, autophagy is a multifaceted process. Herein, 2 aspects of autophagy, induction of autophagy or changes in autophagic flux, in radiation sensitivity were investigated. Relying on a systems-based approach and adding to a large, data dense morphome database [2, 3], we mined for complex cellular information on autophagy due to irradiation of human brain cancer (glioblastoma or GBM) cells. A quantitative stereology method is described here to chronicle these two important aspects of autophagy. These results add to our understanding of the role of autophagy in the efficacy of radiation therapy by providing a basis for the adaptive role for radiation-induced autophagy in GBM cells. Further, because GBM tumors are so resistant to radiotherapy, a novel strategy to radiosensitize GBM cells is urgently needed. Autophagy is a normal, basal housekeeping process in a cell. The main function of autophagy is to recycle damaged or old cellular constituents by encapsulating cargo within autophagosomes that are then trafficked to fuse with lysosomes, allowing lysosomal hydrolytic enzymes to catabolize the cellular material into simple macromolecular subunits. Induction of autophagy above basal levels provided some of the first insights into the complex role of this dynamic process in cell death and its role in radiation sensitization [4, 5]. To provide a proof-of-principle, morphomics data was collected and analyzed for human glioblastoma (U87) monolayer cultured cells irradiated with 0 or 10 Gy rays [2] and 3 days after treatment were collected and fixed for TEM (the most sensitive method to monitor autophagy [6]). Biased sampling methods were used to selectively acquire at least 30 cell profiles for each treatment. The number of autophagosomes was scored for each cell to reflect changes in induction of autophagy in the cell population and within each cell. Autophagic flux was estimated by measuring the area of the cytoplasm occupied by autophagosomes using ImageJ. The area measurement integrates both the number of autophagosomes and size of autophagosomes in the cell. Increased flux, a possible outcome of the area analysis, is revealed when a decreased number and area of autophagosomes is found because hydrolytic digestion of autophagosomes indicates completion of the autophagic process. In contrast, decreased autophagic flux is revealed by increased number or area of autophagosomes resulting from a blockade in autophagy before the final hydrolysis of autophagosomes. Finally, a third outcome of the analysis is the identification of cells that have cytoplasmic areas occupied by autophagosomes that are more than 1 standard deviation of the average area. These outliers focus our attention on those cells having the most dramatic change in radiation-induced autophagy where flux is decreased AND continued autophagosome formation is abnormal. Indeed, previous experiments identified a significant proportion of an irradiated population as outliers, having extremely large autophagosomes due to irradiation with 10 Gy rays or 2 Gy gadolinium neutron capture [2]. Image analysis results are shown in Fig 1. Fig. 1a shows 10 Gy irradiation induced autophagy in essentially 100% of the U87 cells (92% +/- 9%) compared to only 50% +/- 18% of the untreated control cells. Induction of autophagy by irradiation is further supported by our finding of a greater average number of autophagosomes per cell in the 10 Gy irradiated sample (2.48 +/- 1.33) compared to 0.73 +/- Microsc. Microanal. 21 (Suppl 3), 2015 196 0.83 for unirradiated cells (Fig. 1b). Area measurements exposed a significant decrease in radiationinduced autophagic flux with the finding of a significantly higher area for irradiated cells (0.59% +/0.86%) compared to the untreated cells (0.19% +/- 0.25%). The radiation-induced increase in average area certainly reflects the increased number of autophagosomes that are not degraded in the irradiated cells. However, a very large increase in area (21.67%) was identified by our criteria for an outlier from the 10 Gy irradiated U87 cell sample. In this outlier cell, the increased area resulted from measuring 5 extremely large autophagosome in the image. Although 5 autophagosomes is slightly greater than 1 standard deviation of the mean number of autophagosomes found in the irradiated cells, the large size of the autophagosomes (21.67% area of the cytoplasm versus 0.59% +/- 0.86%) clearly plays a more significant part in the extremely large area value. Combined with our previous observation of extraordinarily large, radiation-induced autophagosomes [2], these new results support our previous speculation that 10 Gy irradiation alters a final step for autophagy by promoting continued collection of more and more cargo into the autophagosome, resulting to extremely large autophagosomes. Quantitative examination of the image database reveals an increase in induction of autophagy by 10 Gy irradiation with a corresponding decrease in completion of autophagy (decreased flux). Radiationinduced autophagy results in unique, extremely large autophagosomes. Further support for radiationinduced aberrant cargo collection during the final stages of autophagy is pending assessment of more data collected at longer time points after irradiation. References: [1] JM Lucocq et al. Trends in Cell Biology 25 (2015) p. 59-64. [2] LS Yasui, K Owens. International Journal of Radiation Biology 88 (2012) p. 980-990. [3] CE Zois, MI Koukourakis. Autophagy 5 (2009) p. 442-450. [4] K Sharma et al. EXCLI Journal 13 (2014) p. 178-191. [5] N Mizushima, T. Yoshimori, B. Levine. Cell 140 (2010) p. 313-326 [6] SK Backues et al. Autophagy 10 (2014) p. 155-160. [7] The authors acknowledge support from a Department of Defense grant W81XH-10-1-017 a. b c. Figure 1. Morphomics of autophagy in U87 GBM cells. (a) Percentage of the population of 0 or 10 Gy irradiated cells that contained autophagosomes +/- 95% confidence intervals. (b) The average number of autophagosomes per cell profile +/- SD and (c) the average percent area of the cytoplasm occupied by autophagosomes +/- SD. ANOVA analysis, excluding outliers, showed the untreated cell populations were different from the 10 Gy irradiated population. Post hoc Tukey tests indicated that the untreated control were significantly different from the 10 Gy irradiated sample (at p = 0.05) for graphs (a) and (b).");sQ1[99]=new Array("../7337/0197.pdf","Sparse Feature Selection Identifies H2A.Z as a Novel Pattern-Specific Biomarker for Asymmetrically Self-Renewing Distributed Stem Cells","","197 doi:10.1017/S1431927615001786 Paper No. 0099 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Sparse Feature Selection Identifies H2A.Z as a Novel Pattern-Specific Biomarker for Asymmetrically Self-Renewing Distributed Stem Cells Yang Hoon Huh1, Minsoo Noh2, Frank R. Burden3, Jennifer C. Chen4, David A. Winkler3,5,6,*, and James L. Sherley7,*, Hee-Seok Kweon1,* 1. Division of Electron Microscopic Research, Korea Basic Science Institute, 169-148 Gwahak-ro, Yuseong-gu, Daejeon, Korea 2. College of Pharmacy, Seoul National University, Seoul, Republic of Korea 3. CSIRO Manufacturing Flagship, Clayton, Australia 4. The Senator Paul D. Wellstone Muscular Dystrophy Cooperative Research Center, University of Massachusetts Medical School, Worcester, MA, USA 5. Monash Institute of Pharmaceutical Sciences, Parkville, Australia 6. Latrobe Institute for Molecular Science, Bundoora, Australia 7. Asymmetrex, LLC Boston, MA, USA Despite the important role of distributed stem cell (DSCs) in homeostatic tissue cell renewal and tissue repair, and their use in currently available cell replacement therapies (e.g., hematopoietic stem cell transplantation), there are no effective means for accurate determination of their numbers for medical applications. Therefore, there is a long-standing unmet clinical need for biomarkers with high specificity for DSCs in tissues, or for use in diagnostic and therapeutic cell preparations (e.g., bone marrow). Previous searches for these illusive biomarkers relied on global gene expression profiles for DSCs that were based on comparisons of genes expressed in embryonic stem cells (ESCs) and genes expressed in cell populations enriched for adult stem cells (ASCs) [1, 2, 3], a sub-category of DSCs [4, 5]. These ASC-enriched populations also contained a significant fraction of non-stem committed progenitors and differentiating progeny cells that limited their utility for identifying genes whose expression was unique to DSCs, i.e., "stemness genes" [1, 2, 3]. To identify DSC such useful and specific biomarkers, we combined a novel sparse feature selection method with combinatorial molecular expression data focused on asymmetric self-renewal, is a conspicuous property of DSCs. The analysis identified reduced expression of the histone H2A variant H2A.Z as a superior molecular discriminator for DSC asymmetric self-renewal. Subsequent molecular expression studies showed H2A.Z to be a novel "pattern-specific biomarker" for asymmetrically self-renewing cells with sufficient specificity to count asymmetrically self-renewing DSCs in vitro and potentially in situ [6]. References: [1] NB Ivanova et al., Science 298 (2002), p. 601-604. [2] M Ramalho-Santos et al., Science 298 (2002), p. 597-600. [3] NO Fortunel et al., Science 302 (2003), p. 393. [4] JL Sherley, Breast Dis. 29 (2008), p. 37-46. [5] M Noh et al., Plos ONE 6 (2011), e22077. [6] This work was supported by NIH-NHGRI grant #PSO HG 003170, NIH-NIEHS grant #689471, and NIH-NIGMS Director's Pioneer Award #5DP1OD000805. JCC was supported by the Eunice Kennedy Shriver National Institute of Child Health & Human Development of the NIH (#U54HD060848). DAW also acknowledges support from the CSIRO Newton Turner Award for Exceptional Senior Scientists. HSK was supported by the KBSI grant #35700. Microsc. Microanal. 21 (Suppl 3), 2015 198 Figure 1. H2A.Z asymmetry is detected in prophase nuclei, identified by phosphorylated histone H3, for a variety of cultured cell types. p53-MEFs, example of H2A.Z asymmetry in genetically engineered mouse embryo fibroblasts (line Ind-8) under conditions that promote asymmetric self-renewal. HFSC, example of H2A.Z asymmetry in mouse hair follicle stem cells (strain 3C5) under conditions that promote asymmetric self-renewal. pc-MEF, example of cell with H2A.Z asymmetry detected in cultures of pre-crises MEFs. In cultures enriched for human skeletal muscle satellite stem cells: hSAT ASYM, example of cell with H2A.Z asymmetry, and hSAT SYM, example of cell with symmetric H2A.Z. Scale bar = 25 microns. Figure 2. Detection of mitotic cells with H2A.Z asymmetry in mouse hair follicles. Ten-micron thick, paraffin-embedded sections of adult mouse skin were evaluated by ISIF with antibodies specific for H2A.Z and phosphorylated histone H3 (pH3), a biomarker for mitotic cells. A and B, examples of mitotic hair follicle cells with H2A.Z asymmetry. C and D, examples of epidermal mitotic cells with symmetric H2A.Z. Top rows, low magnification images; scale bars = 50 microns. Bottom rows, 10X magnification images of the cell identified with arrows in upper images; scale bars = 5 microns.");sQ1[100]=new Array("../7337/0199.pdf","Novel Regulated Secretion Mechanism For a Nerve-Secreted Cell Signaling Molecule in C. elegans","","199 doi:10.1017/S1431927615001798 Paper No. 0100 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Novel Regulated Secretion Mechanism For a Nerve-Secreted Cell Signaling Molecule in C. elegans R.D. Schultz1,2, K. Beifuss1, S. Bageshwar1, E.A. Ellis3, and T.L. Gumienny1,2,4 1 Molecular and Cellular Medicine, Texas A&M University Health Science Center, College Station, TX, USA. 2 Interdisciplinary Program in Genetics, Texas A&M University, College Station, TX, USA 3 Microscopy and Imaging Center, Texas A&M University, College Station, TX, USA. 4 Department of Biology, Texas Woman's University, Denton, TX, USA. Bone morphogenetic proteins (BMPs) are cell-cell signaling molecules that are conserved from sea urchins to vertebrates [1]. The BMP family of ligands is dose-dependent and must undergo complex means of secretion, spatial regulation, and degradation to regulate processes such as extracellular growth and homeostasis. The receptors and signaling pathways that this protein family stimulates is well characterized, but the mechanisms controlling how much of the signal is released are unknown [2]. The free-living nematode C. elegans is an ideal system in which to study this cellular BMP regulatory process, with its conserved molecular pathways, ease of genetic manipulation, and simple, transparent body for visualizing fluorescently tagged transgenic proteins. The C. elegans BMP member DBL-1 is expressed in nerves and activates receptors on surrounding epidermal tissue [3]. Some nerve-secreted proteins can be secreted constitutively, without the help of vesicle transport, while other nerve-secreted proteins undergo regulated secretion, using synaptic or dense core secretory vesicles to be transported from the Golgi to the cell's plasma membrane [4]. Using a green fluorescent protein (GFP)-tagged DBL-1, we show that DBL-1 clusters in vesicle-sized punctae in neurons (Figure 1). This suggests that secretion of DBL-1 is regulated. A functional mammalian BMP4 expressed in C. elegans and visualized using a novel, microwave-based immunocytochemical procedure is also punctate and co-localizes with C. elegans DBL-1 (Figure 1). To determine the nature of these punctae, we examined for co-localization between either BMP4 or GFP-tagged DBL-1 and various vesicle components. While we did not find any evidence that DBL-1 is dependent on synaptic or dense core vesicle secretion, we discovered that BMP4 co-localizes at the neuronal plasma membrane with dynamin and caveolin (Figure 2). While caveolin is best known for its role in endocytosis and signal transduction, limited findings suggest caveolin transport from the Golgi to the plasma membrane can include cargo [5]. We found that DBL-1 not only co-localizes with caveolin, but loss of caveolin reduces DBL-1 levels and signaling. Using time-lapse microscopy, we found a sub-population of DBL1-positive punctae move in a manner consistent with microtubule-based transport (Figure 3). This directed transport of DBL-1 further supports a role for caveolin in DBL-1 secretion, as some caveolin vesicles can move on microtubules. Caveolin-associated vesicles have been shown to undergo "kiss-and-run" dynamics, where vesicles transiently fuse with the plasma membrane using dynamin, allowing a partial release of vesicle contents into the extracellular milieu [6]. We provide evidence that regulation of DBL-1 vesicle secretion may be provided through caveolin-based "kiss-and-run" dynamics, which could contribute to dose and spatial control of DBL-1 pathway signaling. We propose that C. elegans DBL-1/BMP signaling is regulated by a novel caveolin-dependent secretion mechanism that controls BMP release from neurons. Microsc. Microanal. 21 (Suppl 3), 2015 200 References [1] IL Blitz and KW Cho, Dev Dyn 238 (2009), p. 1321. [2] MC Ramel and CS Hill, FEBS letters 586 (2012), p. 1929. [3] TL Gumienny and C Savage-Dunn in "WormBook", ed. The C. elegans Research Community, doi/10.1895/wormbook.1.22.2. [4] S Houy et al, Front Endocrinol 4 (2013), p. 135. [5] RG Parton and K Simons, Nat Rev Mol Cell Biol 8 (2007), p. 185. [6] L Pelkmans and M Zerial, Nature 436 (2005), p. 128. [7] This work was supported by NIH 1R01GM097591-01 and by TAMHSC MCMD start-up funds. DBL-1:GFP BMP4 Merge Figure 1. Fluorescent imaging of GFP-tagged DBL-1 (green) with immunolabelled BMP4 (red) shows protein co-localization in motor neurons. Scale bar = 5 �m. BMP4 Caveolin Merge Figure 2. Fluorescent imaging of immunolabelled BMP4 (red) with caveolin (green) shows protein co-localization in motor neurons. Scale bar = 5 �m. t=0s t=1s t=3s t=4s t=5s t=6s Figure 3. Time-lapse images show active trafficking of GFP-tagged DBL-1 puncta (arrow) within a nerve cell. Scale bar = 100 �m.");sQ1[101]=new Array("../7337/0201.pdf","Mechanics of Interdigitating Morphogenesis in Pavement Cells","","201 doi:10.1017/S1431927615001804 Paper No. 0101 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Mechanics of Interdigitating Morphogenesis in Pavement Cells Amirhossein Jafari Bidhendi, Bara Altartouri and Anja Geitmann. Institut de Recherche en Biologie V�g�tale, University of Montreal, 4101 Rue Sherbrooke Est, Montreal, Qu�bec, Canada, H1X 2B2 Plant cells come in a wide variety of shapes, and each geometry is intimately related to the cell's respective function. How these sometimes complex cellular shapes are attained however is poorly understood. Unlike mammalian cells, the driving force for plant cell growth is provided by the internal turgor pressure, whereas spatio-temporal control of the process lies in the mechanics of the material forming the cellular envelope - the cell wall. Shaping of plant cells can therefore be regarded as a question of mechanics of the cell wall. Leaf pavement cells with their interdigitated protrusions and indents resembling a jig-saw puzzle make an ideal model to study the genesis of complex geometries in plant cells (Figure 1). Panteris and Galatis [1] speculated that a higher density of cellulose microfibrils in the neck regions of the undulating cell borders results in augmented stiffness of the cell wall thus locally restricting cell expansion. The deposition of cellulose is supposed to be regulated by the underlying microtubule cytoskeleton in the necks, while actin filaments are present mostly in lobe regions supposedly promoting wall expansion potentially through the delivery of soft pectin material to these sites. Despite this insightful hypothetical model to this date a mechanical validation has not been brought forward. A major challenge has been the visualization of cellulose microfibrils at sufficient resolution. In this project, we use confocal laser scanning microscopy and polarizing fluorescence microscopy in conjunction with in-silico mechanical modelling of the cell wall to understand the mechanical phenomena underlying the complex shaping of pavement cells. To visualize the spatial distribution of the main cell wall components in the leaf epidermis of Arabidopsis thaliana seedlings we used specific labels. Calcofluor white and Pontamine Fast Scarlet 4B (S4B) confirmed the presence of localized accumulation of cellulose microfibrils in the neck regions of the pavement cells (Figure 2a). Moreover, the polarized fluorescence microscopy of S4B shows characteristic spatial configuration of the microfibrils in the neck regions (Figure 2c). These results suggest that cellulose microfibrils locally reinforce the periclinal wall in those regions. Microfibrils also descend along the anticlinal walls in these regions (Figure 2b). Staining the seedlings with propidium iodide, a fluorescent probe with affinity for pectin, in particular for the weakly esterified variety, exhibits a more intense signal on the neck sides of cell undulations (Figure 3). As mechanical stiffness of pectin is known to increase through de-esterification, we suggest that pectin dynamics might be also involved in formation of pavement cell undulations. We used these experimental data to inform a mechanical model based on finite element methods that simulates the generation of lobed cell shapes (Figure 4). References: [1] E Panteris and B Galatis, New Phytol. 167 (2005), p. 721-32. Microsc. Microanal. 21 (Suppl 3), 2015 202 A B C Figure 1. a) Scanning electron micrograph of Arabidopsis leaf pavement cells. Scale bar = 30 �m b) Definition of a pair of lobe and neck in a wall undulation c) Rotated 3D reconstruction of 3D stack of confocal micrographs of a pavement cell. A B C Figure 2. Calcofluor white staining demonstrates a) localized fan-shaped cellulose bundles in necks b) Extension of cellulose bundles down the anticlinal walls in undulations. c) Polarized fluorescence of S4B in the pavement cells of Arabidopsis thaliana. Pseudocolor-coded orientation of cellulose microfibrils reveals subcellular variations in the periclinal wall of individual pavement cells. Scale bar = 10 �m. Figure 3. Propidium iodide staining shows presence of de-esterified pectin on the neck side of undulations. Scale bar = 10 �m. Figure 4. Finite element model of pavement cells generating undulating cell boundaries.");sQ1[102]=new Array("../7337/0203.pdf","Loss of STARD7 Expression Compromises Epithelial Barrier Integrity in Airway of the Shh+/-Cre/STARD7-/- Mice","","203 doi:10.1017/S1431927615001816 Paper No. 0102 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Loss of STARD7 Expression Compromises Epithelial Barrier Integrity in Airway of the Shh+/-Cre/STARD7-/- Mice C.-L. Na1, L. Yang1 and T.E. Weaver1 1. Cincinnati Children's Hospital Medical Center, Division of Pulmonary Biology, Cincinnati, OH, USA The steroidogenic acute regulatory protein-related lipid transfer (START) domain containing proteins are family of 15 proteins that regulate various biological functions including lipid transport, metabolism, and signaling. Our studies by immune blotting and immunocytochemistry showed that STARD7, one of the START proteins, is expressed in lung epithelial cells. Loss of single allele of STARD7 in global KO mice (STARD7+/-) is sufficient to induce airway hyperactivity to ovalbumin challenge, increased epithelial barrier permeability, and atopic dermatitis. It is not clear if hyperactivity to ovalbumin is directly related to loss of STARD7 functions in airway epithelial cells or if other cells contribute to this phenotype. In this study, we generated mice in which expression of STARD7 was selectively disrupted in lung epithelial cells. The loss of STARD7 expression compromised epithelial barrier integrity, providing further evidence that STARD7 is critical for maintaining airway cell homeostasis. Shh+/-Cre/STARD7-/- mice (STARD7 KO mice) were generated for selective disruption of STARD7 expression in lung epithelial cells. SEM studies using mouse lungs obtained from 4 month and 10 month old STARD7 KO mice (Figure 1) determined that they had shorter and fewer cilia and microvilli in small bronchi and bronchioles compared to age-matched controls. Fractures along luminal cell surface were detected in small bronchi and bronchioles and were more pronounced in 10 month old KO mice than in 4 month old KO mice. To determine if epithelial cell structure was disrupted in airway of the STARD7 KO mice, lung blocks of age-matched WT and KO mice were collected for TEM analysis. There was a significant reduction in the number and size of mitochondria in ciliated bronchiolar epithelial cells and Clara cells of the 4 month and 10 month old STARD7 KO mice (Figure 2). The decrease in mitochondrial population was often associated with formation of mitochondrial limiting membrane bound lipid inclusions, disintegration and degradation of mitochondrial cristae and matrix, suggesting that STARD7 is important for maintaining lipid homeostasis in mitochondria. In distal airway of the 4 and 10 month old STARD7 KO mice, perinuclear lipid inclusions were detected in ciliated bronchiolar epithelial cells and Clara cells. Concentric lamellar inclusions with or without degraded mitochondria and other cellular fragments were prominent in some Clara cells. Ciliated bronchiolar epithelial cells and Clara cells that had lipid accumulation and lipid inclusions often had leaky basolateral cell junctions and loose intercellular spaces, suggested that integrity of airway epithelial barrier was compromised. Taken together, our studies demonstrate that loss of STARD7 function in bronchiolar epithelial cells results in aberrant mitochondrial formation and compromised integrity in airway epithelial barrier. How STARD7 regulates mitochondria and epithelial barrier integrity in airway cells under asthmatic challenge will be determined in future studies. Microsc. Microanal. 21 (Suppl 3), 2015 204 Figure 1. Airway surface of (A) WT and (B) STARD7 KO mice surveyed by SEM. Fractures (arrow) were prominent along the airway surface of STARD7 KO mice. Scale bar is 5 �m. Figure 2. Ciliated bronchiolar epithelial cells and Clara Cells of the WT (A and C) and STARD7 KO mice (B and D). The number and size of mitochondria was significantly reduced in ciliated bronchiolar epithelial cells and Clara cells of the STARD7 KO mice (B and D) compared to WT (A and C). Lipid accumulation was pronounced and integrity of cell junctions (arrow) was compromised in airway epithelial cells of KO mice (B and D). MT: mitochondria. Scale bar is 1 �m.");sQ1[103]=new Array("../7337/0205.pdf","FIB/SEM Tomography of Wound Biofilm","","205 doi:10.1017/S1431927615001828 Paper No. 0103 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 FIB/SEM Tomography of Wound Biofilm Binbin Deng1,2, Kasturi Ganesh Barki1, Subhadip Ghatak1, Sashwati Roy1, David W. McComb2, Chandan K. Sen1 1. Davis Heart & Lung Research Institute and Comprehensive Wound Center, Department of Surgery, The Ohio State University Wexner Medical Center, Columbus, OH 43210 USA 2. Center for Electron Microscopy and Analysis, Department of Materials Science and Engineering, The Ohio State University, Columbus, OH 43212 USA Biofilm has a complex architecture and show high tolerance to host immune responses and antimicrobial agents [1]. Biofilm infection of chronic wounds often makes them very resistant to treatment, and can be a severe health problem in patients with compromised immune systems. Despite the extensive study of biofilm in the past 15 years [2], the pathogenic biofilm structures remain to be characterized. DualBeam FIB/SEM tomography is a suitable tool to investigate three-dimensional (3D) structures of biological materials at intermediate to high resolution. In order to understand the biofilm architecture in animal model and clinical chronic wound, we have studied a porcine pre-clinical wound model involving single-species and multi-species infection as well as chronic wound from a patient using FIB/SEM tomography. Domestic Yorkshire pigs were subjected to a full-thickness burn (2"x2"). A clinically relevant mixedspecies infection was established. Patient tissue was collected from one year old chronic wound. Pig biofilm and patient wound tissue were cut into 200-500�m thick sections. Samples were chemically fixed and en bloc stained. After dehydration and infiltration, tissue sections were embedded in durcupan resin and incubated at 60�C for 2 days [3]. The resin embedded tissue was trimmed by ultramicrotome and mounted on an SEM stub. FIB slice and view datasets were collected on a Helios Nanolab 600 DualBeam (FIB/SEM) (FEI, Hillsboro). Images were processed and visualized using MIPAR[4], ImageJ [5], IMOD [6], Chimera [7] and Avizo (FEI, Hillsboro) software packages. Biofilms formed by aggregated bacteria are encapsulated in extracellular polymeric substance (EPS). Our FIB/SEM tomography results revealed structures of pig biofilm (Figure 1) and patient wound (Figure 2). 3D structures demonstrated the non-uniform distribution of biofilm. Macrophages were found engulfing bacterium. Phagocytes were found squeezing through capillary walls to reach the infection sites. These data are providing unique insights that are improving our understanding of the process of biofilm wound healing. Microsc. Microanal. 21 (Suppl 3), 2015 206 References: [1] L Hall-Stoodley et al. "Bacterial biofilms: from the Natural environment to infectious diseases", Nature Reviews Microbiology. 2004; 2(2): p95-108 [2] T Bjarnsholt "The role of bacterial biofilms in chronic infections", APMIS suppl. 2013; 121(136): p1-51 [3] AJ Bushby et al. "Imaging three-dimensional tissue architectures by focus ion beam scanning electron microscopy", Nature Protocols 2011; 6(6): p845-858 [4] JM Sosa et al. "Development and application of MIPARTM: a novel software package for two- and three-dimensional microstructural characterization", Integrating Materials and Manufacturing Innovation 2014; 3:10 [5] CA Schneider et al. "NIH Image to ImageJ: 25 years of image analysis", Nature Methods 2012; 9(7): p671-675 [6] JR Kremer et al. "Computer visualization of three-dimensional image data using IMOD", Journal of Structural Biology 1996; 116(1): p71-76 [7] EF Pettersen et al. "UCSF Chimera - a visualization system for exploratory research and analysis", Journal of Computational Chemistry 2004; 25(13): p1605-1612 [8] The authors acknowledge funding from Ohio Third Frontier Program, The Institute of Materials Research and The Ohio State University" 10�m Figure 1. 3D volume views of pig biofilm obtained using FIB/SEM tomography Figure 2. 3D volume views patient wound obtained using FIB/SEM tomography");sQ1[104]=new Array("../7337/0207.pdf","Visualization of Plaque/Vessel Interfaces using FIB-SEM","","207 doi:10.1017/S143192761500183X Paper No. 0104 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Visualization of Plaque/Vessel Interfaces using FIB-SEM George Wetzel,1 Susan M. Lessner2 1. 2. Clemson University Electron Microscopy Facility, Anderson, SC University of South Carolina School of Medicine, Department of Cell Biology & Anatomy, Columbia, SC Delamination or tearing between tissue layers is a material failure mode observed in several clinically significant cardiovascular pathologies, particularly in arterial dissection. Delamination also occurs as a result of iatrogenic damage to vascular walls, as in the example of intimal flap formation during balloon angioplasty.1 To better understand vascular tissue delamination failure from a mechanical perspective, we have performed experimental plaque delamination studies in mouse models of atherosclerosis,2,3 which show that failure typically occurs at the luminal surface of the internal elastic lamina (IEL), an elastic membrane which separates the intimal and medial tissue layers in the aorta. The microstructural mechanism of failure at this interface has yet to be identified, but fibrous elements of the extracellular matrix, particularly collagen fibers, are believed to contribute to the observed adhesive strength.2 Characterizing the ultrastructural organization of matrix fibers at the plaque-IEL interface will provide critical information needed to define the interface geometry for computational simulations of tearing failure. Here we examine the use of FIB-SEM as a potential tool to help characterize the 3-dimensional ultrastructure of the plaque-IEL boundary. Methods Mice genetically deficient in apolipoprotein E (apoE KO) were used as an animal model of atherosclerosis. After being maintained on a high-fat diet for 12 months to induce advanced atherosclerotic plaque development throughout the descending aorta, the mice were euthanized by an overdose of anesthetic and the aortas were perfusion fixed with dilute Karnovsky's fixative at mean arterial pressure. Tissue samples were processed and stained using a modification of the OTO (osmium-thiocarbohydrazide-osmium) method, dehydrated in graded alcohols, and embedded in Durcupan ACM resin for electron microscopy. Fixed and embedded samples were sputter coated with approximately 10nm of platinum to form a conductive layer prior to ion beam milling. Samples were loaded into a Hitachi NB5000 nanoDUE'T Focused Ion Beam and Scanning Electron Microscope (FIB-SEM) system for milling and subsequent image capture. Using the Mill&Monitor� feature, serial slices were removed from the sample block using the ion beam, with an electron micrograph acquired after each slice. These images were stacked together to form a three-dimensional model. The specific region of interest was the interface between the vessel wall and plaque. After the region of interest had been located and oriented such that the interface would be visible to the SEM after sputtering, a box trench was cut into the sample using a high current, 40kV Ga+ beam. A tungsten cap was deposited to protect the region of interest, and the cross section face further polished with a lower energy ion beam to remove any artifacts. Serial slices spaced 18nm apart were removed from this face using a 0.07nA, 40kV ion beam. 396 iterations were done for a total sampling thickness of 7um. Milling time for each slice was set at 1min. Images of the sliced face were acquired at a magnification of 7kx using a 3kV, high probe current beam to minimize imaging depth and increase backscattered electron yield. Images were acquired Microsc. Microanal. 21 (Suppl 3), 2015 208 using the ExB in-column backscattered electron detector with a capture time of 160s for an increased signal to noise ratio. All images acquired using Mill&Monitor� were loaded into Avizo Ver.8 for 3D rendering. Orthoslice projections were viewed along the XY, ZY, and ZX planes. Conclusions FIB-SEM provides a potentially useful tool to uncover 3-dimensional ultrastructural organization of mouse aortic wall at the plaque-IEL interface. However, due to thermal drift, spatial resolution is best in the plane orthogonal to the direction of ion milling. Future work will focus on image analysis and 3-D reconstruction of the interface geometry. References 1. Honye, J, Mahon DJ, Jain A, White CJ, Ramee SR, Wallis JB, al-Zarka A, Tobis JM, Circulation, 85(3), 1012-25 (1992). 2. Wang Y, Johnson JA, Fulp A, Sutton MA, Lessner SM, J. Biomech., 46, 716�722 (2013). 3. Wang Y, Ning J, Johnson JA, Sutton MA, Lessner SM, J. Biomech., 44(13), 2439-2445 (2011). Figure 1. Backscattered electron (BSE) image of the milled sample face showing the plaque/vessel interface. Four 20s exposures were averaged together to reduce noise. Scale bar: 5um, divided into 10 increments of 500nm each; FC=foam cell; IEL=internal elastic lamina; SMC=smooth muscle cell; ECM=extracellular matrix FC IEL ECM SMC SMC IEL IEL ECM ECM Figure 2. 3D stack printed in reverse contrast to resemble TEM contrast. Model has been cropped to reveal details in the ECM adjacent to the IEL.");sQ1[105]=new Array("../7337/0209.pdf","Enhanced Large-Area TEM Analysis on Mitochondrial Alteration in Brain Neuronal Cell of Parkinson's Disease Model Mouse Using Montage Function Installed in JEM 1400-Plus System","","209 doi:10.1017/S1431927615001841 Paper No. 0105 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Enhanced Large-Area TEM Analysis on Mitochondrial Alteration in Brain Neuronal Cell of Parkinson's Disease Model Mouse Using Montage Function Installed in JEM 1400-Plus System Hyo-Jeong Kim1,2, , A Reum Je1, 2, , Mi Jeong Kim1, Jae-Kyung Hyun1, Hyung-Gun Kim2, Yang Hoon Huh1,*, Hee-Seok Kweon1,* 1. Division of Electron Microscopic Research, Korea Basic Science Institute, 169-148 Gwahangno, Yuseong-gu, Daejeon, Korea 2. Department of Pharmacology, College of Medicine, Dankook University, 119 Dandaero, Cheonan-si, Chungnam, Korea Despite the important role of transmission electron microscope (TEM) in providing tremendous ultrastructural information of cells, organelles, and molecules [1], one of the disadvantages of TEM is the difficulty associated with observing large-area with high resolution at low-magnification [2 and 3]. Although several analytical techniques and equipments including ultramicrotomy-combined serial block face scanning EM have been developed recently to overcome such problem, those are mostly concentrated in SEM applications [1, 2, 4, 5, and 6]. In this research, to acquire large-area TEM images with highresolution, we used the montage function in TEM Center, a software suite that controls JEM 1400-Plus TEM as described in Figure 1. Then, we compared the resolution of low mag. reference image (4,000X) acquired using normal imaging mode with that of 3X3 montage image (12,000X). The spatial resolution of montage image was improved without any image distortion during montage process (Figure 2). Next, using the montage function, we focused on the morphological alterations of the mitochondria in substantia nigra of hLRRK2 transgenic (G2019S) Parkinson's disease model mice. Normal mitochondria in cell body and axon of wild-type mice mostly contained well-conserved ultrastructure. They maintained dense matrix, compacted with thin and uniform cristae, and surrounded by clear inner and outer membrane. However, majority of mitochondria in cell body and axon of the substantia nigra of hLRRK2 transgenic mice were swollen, different from the structure found in wildtype mice. Also, the mitochondria have deteriorated structure, with the cristae severely disrupted through the entire mitochondrial lumen, and the inner and outer mitochondrial membranes were partially disappeared. This result clearly shows the complex structural alteration of mitochondria in brain neuronal cells of Parkinson's disease model mice, and demonstrates the contribution of montage-based large-area TEM image analysis system for the acquisition of high throughput statistical information for an accurate interpretation of various morpho-functional alterations of cellular organelles [7]. References: [1] K Miranda et al., Mol. Reprod. Dev. (2015), doi: 10.1002/mrd. 22455. [2] E Oho et al., Journal of Electron Microscopy 49 (2000), p.135-141. [3] Y Ikeda et al., Microsc. Microanal. 19 (2013), p.1330-1331. [4] C Blumer et al., Medical Image Analysis 19 (2015), p.87�97. [5] H Horstmann et al., Plos ONE 7 (2012), e35172. [6] LCS Medeiros et al., Plos ONE 7 (2012), e33445. [7] This research was supported by Korea Basic Science Institute (#E35700, T35444) and the Bio & Medical Technology Developmental Program of the NRF funded by the Ministry of Science, ICT & 390 Future Planning (#2013M3A9A9050076). Microsc. Microanal. 21 (Suppl 3), 2015 210 Figure 1. Outline of the experimental procedure for the enhanced large-area EM image acquisition using montage function in TEM Center of JEM 1400-Plus TEM. Simplified flowchart explaining the main experimental steps and conditions for the montage process is described. Figure 2. Improvement of the spatial resolution after montage process. The images of mouse substantia nigra without (A) or with (B) montage from same large-area are shown, respectively. A1'-B2': 2X zoomed image of A1-B2. A1"-B2": 4X zoomed image of A1'-B2'. A1"'-B2"': 4X zoomed image of A1"-B2".");sQ1[106]=new Array("../7337/0211.pdf","Renal Biopsy Diagnosis of IGA Nephropathy in Chin Refugees Living in the United States","","211 doi:10.1017/S1431927615001853 Paper No. 0106 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Renal Biopsy Diagnosis of IGA Nephropathy in Chin Refugees Living in the United States N. Rush, M. P. Goheen, C. L. Phillips Indiana University School of Medicine, Pathology, Indianapolis, USA IgA nephropathy (IgAN) is a common diagnosis in patients undergoing renal biopsy for hematuria and proteinuria, but histopathology may vary widely among affected patients. Worldwide, IgAN is the most common form of primary glomerulonephritis and is a leading cause of end-stage renal failure in patients presenting for renal replacement therapy1. In the United States, prevalence rates of IgA nephropathy are only 2 to 10 percent while in Asia prevalence rates range from 20 to 40%2. In Chin refugees relocated from Myanmar (formerly Burma) to a geographic area near our laboratory in the Midwestern United States (U.S.), IgAN was the most frequently encountered glomerulopathy detected in needle biopsy specimens. We report our biopsy experience with this small but unique cohort, many of whom have had little to no previous access to medical care. Our electronic database of surgical pathology specimens was searched for kidney biopsy samples originating from Chin patients who were born in Myanmar and immigrated to the Midwest region of the U.S. Needle cores from renal biopsy specimens were examined by bright-field (BM), immunofluorescent (IF), and/or electron (EM) microscopy using conventional techniques. Available clinical history that accompanied each specimen was reviewed. When sufficient tissue was available, specimens were scored by BM using criteria developed by The International IgA Nephropathy Study Group (i.e., Oxford criteria)3. Of 2790 renal biopsy specimens evaluated at our facility in the last six years, seven were obtained from Chin individuals who emigrated from Myanmar. All seven patients had been evaluated by a nephrologist. Six Chin patients were diagnosed with IgAN based on analysis by IF and EM: five females (ages 21 to 38 years) and one male (age 29 years). The 7th Chin patient, a male (age 38 years), had podocyte foot process effacement detected by EM in the absence of immune deposits by IF and was diagnosed with minimal change disease (MCD). Female patients were more likely to be proteinuric (5/5), hematuric (4/5), hypertensive (4/5), and pregnant (4/5), with initial serum creatinine levels that ranged from 0.5 to 1.12 mg/dL. Both males presented with hypertension, proteinuria, and acute renal failure. The patient with MCD had an initial creatinine of 1.2 mg/dL. Applying Oxford histologic criteria for scoring IgAN, two patients had one indicator of less favorable prognosis, one patient had three indicators, one patient had no morphologic indicators, and two specimens had insufficient tissue for analysis. A cohort of Myanmar's Chin minority has emigrated to the Midwestern U.S. where they presented with hypertension and abnormal urinalysis. Seven of these individuals underwent an ultrasound-guided renal biopsy procedure. Six were diagnosed with IgAN (Fig.1) and one with MCD (Fig. 2). While the majority of these patients had IgAN, their histomorphology and prognosis varied widely based on Microsc. Microanal. 21 (Suppl 3), 2015 212 Oxford criteria. Given the greater likelihood of encountering IgAN in the Chin population, nephrologists may initially decide to defer or delay a renal biopsy procedure, but these clinicians would not have access to tissue-based Oxford scores that might alter clinical management. Fig.1 Immune deposit (arrow) in the mesangium of a patient with IgA nephropathy. Fig. 2 Podocyte foot process effacement (arrow) in minimal change disease. References: 1. Ibels LS, Gy�ry AZ. Medicine (Baltimore). 1994 73(2):79-102. 2. Donadio JV, Grande JP.. N Engl J Med. 2002 347(10):738-48. 3. Coppo R, et al, The New Oxford Classification of IgA nephropathy 2010 (1) 241-8.");sQ1[107]=new Array("../7337/0213.pdf","Visualizing the Distribution and Stoichiometry of Growth Factor Receptors in Intact Cells in Liquid Phase with Correlative Fluorescence and Scanning Transmission Electron Microscopy","","213 doi:10.1017/S1431927615001865 Paper No. 0107 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Visualizing the Distribution and Stoichiometry of Growth Factor Receptors in Intact Cells in Liquid Phase with Correlative Fluorescence and Scanning Transmission Electron Microscopy Diana B. Peckys1, Justus Hermannsd�rfer1, Verena Tinnemann1, Ulrike Korf2 and Niels de Jonge1,3 1. 2 INM � Leibniz Institute for New Materials, 66123 Saarbr�cken, Germany Division of Molecular Genome Analysis, German Cancer Research Center, 69120 Heidelberg, Germany 3 Department of Physics, University of Saarland, 66123 Saarbr�cken, Germany The epidermal growth factor receptor (EGFR) family consists of four similar members that play important roles in various cellular processes such as cell proliferation and survival. Over-expression of these proteins, especially EGFR and HER2, is involved in many types of cancer. After activation by its ligand EGF, activated EGFR proteins can form homodimers, thus activating specific signaling pathways for cell growth. The HER2 is considered as an orphan receptor because it needs no ligand to form homodimers, thereby contributing significantly to uncontrolled intracellular signaling. Heterodimers are also formed in some cases. We present a method to study the distribution patterns of proteins in the plasma membranes of intact cells in liquid, which we successfully applied study HER2 and EGFR on SKBR3 breast cancer cells [1, 2]. The proteins were labeled with fluorescent quantum dots (QDs), exhibiting a core of an electron dense material. In order to detect the receptors without perturbing the native equilibrium, a short peptide was used for specific binding of the streptavidin-biotin coupled QDs [3]. The cells were fixed immediately after the labeling but kept in their native liquid state and staining was avoided. The overall cellular distributions were first evaluated using direct interference contrast (DIC) light microscopy, and fluorescence microscopy. These images provided an overview of the membrane topography, and the average amount of receptor in the membrane. Selected suitable regions were then studied at the nanoscale with environmental scanning electron microscopy (ESEM), while keeping the cells in their hydrated state (Fig. 1). The microscope was equipped with a scanning transmission electron microscopy (STEM) detector, so that strong contrast was obtained on the QDs (Fig. 2). Thousands of label positions were automatically detected and statistically analyzed via the calculation of the pair correlation function. The analysis results differentiated between distinct functional cellular regions. Conventional biochemical methods are capable of studying the distribution of the receptors in protein complexes from pooled cellular material but miss information about expression levels at the cellular level. In particular, cancer cells are known to exhibit significant differences in the expression level of the HER2 receptor between individual cells and even between distinct functional membrane regions, representing the characteristic feature of heterogeneity derived from the accumulation of various genetic alterations. Optical methods, on the other hand, do not exhibit sufficient spatial resolution to resolve the individual constituents of proteins complexes, and only indirect information is obtained about the proximity of proteins. A particular problem is the inevitable random occurrence of proteins in close proximity in relevant regions with a high protein density. Our method is capable of distinguishing between random occurring distances and preferred distances indicating the actual presence of dimers or higher order complexes, because the distances can be directly measured with 3 nm resolution. Microsc. Microanal. 21 (Suppl 3), 2015 214 References: [1] D.B. Peckys and N. de Jonge, Microsc. Microanal. 20 (2014), p. 346. [2] D.B. Peckys et al , Sci. Rep. 3 (2013), p. 2626-1. [3] D.B. Peckys, V. Bandmann and N. de Jonge, Meth. Cell Biol. 124 (2014), p. 305. [4] We thank E. Arzt for his support through INM. Research in part supported by the Leibniz Competition 2014. x,y scan H2O vapor Liquid H2O film Nucleus Cell QDs SiN membrane Scattered electrons towards ADF STEM detector Figure 1. A cell grown on a silicon nitride (SiN) membrane containing proteins labeled with QDs. A liquid water film is maintained over the cell, while the sample is surrounded by a water vapor environment. The ESEM electron beam scans the sample, and contrast is obtained using the annular dark field (ADF) STEM detector located underneath the sample. Figure 2. ESEM micrograph showing a membrane area of an intact SKBR3 breast cancer cell in liquid state. The white spots indicate positions of individual QD-labeled HER2 proteins. Many homodimers and large order clusters are visible. The image was recorded using an electron energy of 30 keV, at a vapor pressure of 740 pA and a temperature of 3.0� C. The image acquisition time amounted to 72 s. 500 nm");sQ1[108]=new Array("../7337/0215.pdf","Centrin Phosphorylation and Localization Dynamics during the Centriole Cycle","","215 doi:10.1017/S1431927615001877 Paper No. 0108 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Centrin Phosphorylation and Localization Dynamics during the Centriole Cycle Trace A. Christensen1, and Jeffrey L. Salisbury1 1. Department of Biochemistry and Molecular Biology, Mayo Clinic, Rochester, MN 55905 Centrin was first discovered as a major component of basal body-associated contractile flagellar roots in the algae, and subsequently was found to be a common protein of centrosomes [1-3]. Four centrin genes are expressed in mammals: Cetn-1 through Cetn-4 [4-7]. Centrin-2 and -3 are centriole proteins and recombinant GFP-centrin-2 and centrin-3 localize to centrioles and also to the pericentriolar material that surrounds the centrioles [2, 3, 8, 9]. Centrin is highly conserved in sequence and shows several striking structural features outside the four calcium-binding EF-hand domains. The carboxy-terminal half of centrin is highly conserved, and the short carboxy-terminal sequence - 167KKTSLY172 - is conserved among centrins associated with centrioles and basal bodies. This sequence is missing in human centrin-3 and in centrins of lower eukaryotes lacking centrioles or in centrins that localize to cellular structures other than centrioles. Three putative Mps1 phosphorylation sites that augment the canonical centriole assembly pathway have been identified on centrin-2 [10]. Earlier, phosphorylation of centrin-2 at a distinct carboxy-terminal phosphorylation site (-KKTS170PLY) by protein kinase A was shown to promote centriole separation [11]. Finally, abnormal centrin phosphorylation S170 is observed in human breast tumors that have aberrant centriole numbers [12], suggesting that phosphorylation of centrin must be tightly regulated. Here, we will present observations that show that in a mitogen-synchronized cell cycle, centrin abundance increases progressively and becomes phosphorylated at residue S170 prior to centriole duplication and DNA synthesis. Phospho-S170-centrin is asymmetrically localized to one of the preexisting centrioles at about the time that primary cilia are lost. The appearance of phospho-S170-centrin is temporally coincident with recruitment of robust SAS-6 staining to both pre-existing centrioles. A second peak of centrin S170 phosphorylation is seen at the onset of mitosis, preceding the late-mitotic ablation of centrin protein. Cells expressing conditional mutations at S170 (S>A170 and S>D170) in centrin-2 show altered frequency of primary cilia and a unique pattern of PCM-satellite material reminiscent of failed centriole assembly. The timing and localization of centrin phosphorylation and the opposing effects of conditional centrin mutations engineered to either block or mimic centrin phosphorylation suggest the intriguing possibility that centrin harbors a carboxy-terminal phosphorylation-sensitive alteration that corresponds to key events in the centriole cycle. References 1. JL Salisbury, et al., Striated flagellar roots: isolation and partial characterization of a calciummodulated contractile organelle. J Cell Biol, 99(1984) p. 962-70. 2. AT Baron, et al., Centrin is a component of the pericentriolar lattice. Biol Cell, 76(1992) p. 383-8. 3. JL Salisbury, et al., Centrin-2 is required for centriole duplication in mammalian cells. Curr Biol, 12(2002) p. 1287-92. Microsc. Microanal. 21 (Suppl 3), 2015 216 4. B Huang, et al., Molecular cloning of cDNA for caltractin, a basal body-associated Ca2+- binding protein: homology in its protein sequence with calmodulin and the yeast CDC31 gene product. J Cell Biol, 107(1988) p. 133-40. 5. R Errabolu, et al., Cloning of a cDNA encoding human centrin, an EF-hand protein of centrosomes and mitotic spindle poles. J Cell Sci, 107(1994) p. 9-16. 6. PE Hart and SM Wolniak, Molecular cloning of a centrin homolog from Marsilea vestita and evidence for its translational control during spermiogenesis. Biochem Cell Biol, 77(1999) p. 101-8. 7. S Middendorp, et al., Identification of a new mammalian centrin gene, more closely related to Saccharomyces cerevisiae CDC31 gene. Proc Natl Acad Sci U S A, 94(1997) p. 9141-6. 8. J Laoukili, et al., Differential expression and cellular distribution of centrin isoforms during human ciliated cell differentiation in vitro. J Cell Sci, 113(2000) p. 1355-64. 9. A Paoletti, et al., Most of centrin in animal cells is not centrosome-associated and centrosomal centrin is confined to the distal lumen of centrioles. J Cell Sci, 109(1996) p. 3089-102. 10. CH Yang, et al., Mps1 phosphorylation sites regulate the function of centrin 2 in centriole assembly. Mol Biol Cell, 21(2010) p. 4361-72. 11. W Lutz, et al., Phosphorylation of Centrin during the Cell Cycle and Its Role in Centriole Separation Preceding Centrosome Duplication. J. Biol. Chem., 276(2001) p. 20774-20780. 12. WL Lingle, et al., Centrosome hypertrophy in human breast tumors: implications for genomic stability and cell polarity. Proc Natl Acad Sci U S A, 95(1998) p. 2950-5. Figure Legend. a) Electron micrograph through a centrosome showing a primary cilium (pc), its centriole (C1), and two procentrioles (p1 and p2). b) Diagram illustrating the essential features shown in (a) with additional information from adjacent serial sections. c) Fluorescence microscopic localization of centrin in a pair of centrioles, showing a merged image of all channels, GFP-centrin2, phosphor-centrin at the carboxy-terminal sequence KKTSPLY, total centrin, and a model image with each of the channels adjusted to saturation.");sQ1[109]=new Array("../7337/0217.pdf","Immunofluorescence Microscopy As a Critical Confirmation Technique in Cell Cycle-Related Proteomics Studies","","217 doi:10.1017/S1431927615001889 Paper No. 0109 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Immunofluorescence Microscopy As a Critical Confirmation Technique in Cell Cycle-Related Proteomics Studies Silviu L. Faitar1 1. Department of Mathematics and Natural Sciences, D'Youville College, Buffalo, NY USA. The EVI5 gene was originally identified in a screen for integration sites of ectopic viruses that gave rise to tumors in mice [1]. The human EVI5 gene was isolated and characterized as a result of its location at the breakpoint junction of a t(1;10)(p21;q21) constitutional chromosome translocation in a patient with stage 4S neuroblastoma [2]. The product of the human EVI5 gene is an 810 amino acids long ubiquitously expressed protein. It carries two readily identifiable motifs [2,3], a Tre2/Bub2/Cdc16 (TBC) consensus sequence domain in the N-terminal region, which is known in other proteins to act as a GTPase activating domain (GAP), and an extensive coiled coil in the C-terminal region, which indicates possible protein-protein interaction sites. A conserved nuclear localization signal is also present in the N-terminal region. Previous immunocytochemical studies revealed that the Evi5 protein has a dynamic intracellular localization during the cell cycle, being associated with the mitotic spindle during metaphase, and migrating to the midzone and midbody in late mitosis and during cytokinesis [3]. Little is known however, about the function of the normal Evi5 protein or its possible association with neuroblastoma pathogenesis. To identify the putative metabolic pathways that involve Evi5, a functional proteomic approach was used in order to determine the identity of proteins that interact with it [4]. The TBC-containing N-terminal part and the coiled-coil rich C-terminal part of Evi5 were expressed as GST-fusion proteins, and GST pull-down experiments were performed using total protein extracts. The proteins that specifically bind to the two Evi5 fusion proteins were analyzed by a linear ion trap mass spectrometer (Thermo Electron LTQ) coupled to a nanochromatography system. A total of 8 proteins were found to interact with Evi5-N and 39 with Evi5-C. They included proteins involved in cytoskeleton reorganization, signal transduction, vesicular transport, membrane traffic and neuronal migration. Among these potential functional partners the highest matching scores were obtained for alpha�tubulin, annexin VI, beta-spectrin, ezrin, and Rab11. The mass spectrometry results were confirmed by western blotting and immunoprecipitation. However, these in vitro techniques do not demonstrate that these proteins actually interact in vivo. Immunoflorescence microscopy studies were necessary to confirm the co-localization of these proteins with Evi5. Furthermore, taking into account the dynamic localization of Evi5 during mitosis and cytokinesis, it is important to determine the specific phase of the cell cycle when these interactions may occur. This study used immonocytochemical stainings and immunoflorescence microscopy to test for these putative functional interactions. Human epithelial cells (Hep-2) immobilized on glass slides (Bio-Rad�Kallestad) were stained for Evi5, alpha-tubulin, annexinVI and Rab11, using commercially available antibodies. An antibody specific for Rab23 was also used as a negative control for the Evi5�Rab11 co-localization studies, knowing that in spite of the high homology between Rab11 and Rab23, the latter does not interact with Evi5 [4]. DyLight 594-conjugated AffiniPure goat anti-mouse IgG (H+L) and DyLight 488-conjugated AffiniPure donkey anti-rabbit IgG (H+L) (both Jackson ImmunoResearch) were used as secondary antibodies. Vectashield mounting medium with DAPI (Vector Laboratories Inc.) was used after staining, and the cells were examined using a Zeiss Axioskop 40 fluorescence microscope (Carl Zeiss Microscopy). Images were obtained using a 100 X Zeiss A-Plan oil immersion objective and Immersol Microsc. Microanal. 21 (Suppl 3), 2015 218 518F immersion oil. Image acquisition and processing was executed using the ZEN Imaging Software (Carl Zeiss Microscopy). Immunoflorescence microscopy experiments on interphase cells confirmed a strong Evi5 nuclear staining, as well as a relatively specific cytoplasmic staining that was largely localized to the centrosome. In double-staining experiments, Evi5 showed co-localization with alpha-tubulin at the centrosome, and co-localization with annexin VI in the nucleus (Fig.1). No significant co-localization was observed with either Rab11 or Rab23 in interphase. Both these proteins showed a punctuated cytoplasmic localization consistent with their role in vesicular transport and membrane trafficking (Fig.1). In mitotic cells, there was no observed co-localization between Evi5 and annexin VI. However, as the centrosome divides and forms the polar bodies, Evi5 is located at the spindle poles and colocalizes with the mitotic spindle tubulin (Fig.1). As telophase develops, Evi5 appears on the midzone between the two forming daughter cells and persists at the midbody to the terminal stages of cytokinesis, also showing co-localization with alpha-tubulin. Significantly, a strong Evi5�Rab11 co-localization is observed during the late stages of mitosis and especially during cytokinesis. This dynamic immunofluorescence pattern is not observed when anti-Rab23 antibodies were used instead of antiRab11 antibodies, suggesting a specific functional interaction between Evi5 and Rab11 during the furrow ingression and the vesicular transport associated with cytokinesis. The microscopy studies presented demonstrate the importance of immunoflorescence as a critical confirmation technique for protein-protein interaction proteomics data, especially when transitory shortterm interactions are investigated. Microscopy images could not only determine if the studied protein are localized in close vicinity to each other, but they also offer a timeline for these interactions, especially in relationship to the cell cycle. References: [1] X Liao, A Buchberg, N and N Copeland, J. Virol. 69, (1995) p. 7132. [2] T Roberts, O Chernova and J Cowell, Cancer Genet. Cytogenet. 100, (1998) p. 10. [3] S Faitar, K Sossey-Alaoui, T Ranalli and J Cowell, Experimental Cell Research 312, (2006) p. 2325. [4] J Dabbeekeh, S Faitar, C Dufresne and J Cowell, Oncogene 26, (2007) p. 2804. Figure 1. Immunofluorescence microscopy images showing co-localization of Evi5 (green) with annexin VI, alpha-tubulin (TUBA) and Rab11 (red). DNA is stained with DAPI (blue). Hep-2 cells are shown in interphase (left panel) and in various stages of cell division (center and right panels).");sQ1[110]=new Array("../7337/0219.pdf","Using Optical Tweezers to Quantify the Interaction Force of Dengue Virus with Host Cellular Receptors","","219 doi:10.1017/S1431927615001890 Paper No. 0110 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Using Optical Tweezers to Quantify the Interaction Force of Dengue Virus with Host Cellular Receptors Bo-Ying Tsai1, 2, Jyun-Yu Chen 1, 2, Arthur Chiou1,2,, Yueh-Hsin Ping 1, 2, 3, 4 1 2 Institute of Biophotonics, National Yang-Ming University, Taipei, Taiwan Biophotonics Interdisciplinary Research Center, National Yang-Ming University, Taipei, Taiwan 3 Infection and Immunity Research Center, National Yang-Ming University, Taipei, Taiwan 4 Department and Institute of Pharmacology, National Yang-Ming University, Taipei, Taiwan Optical tweezers (OT) is a powerful tool to manipulate a single particle in living cells. It is able to determine the interaction force of biological molecule(s) in an excellent resolvable range from 1 to 100 pN. Moreover, OT has been used to investigate the interaction between ligand and receptor [1] and to reveal the dynamics and properties of living cell elasticity, viscoelasticity and adhesion. Dengue virus (DENV) is one of the most widespread viral pathogens around the world. There is currently no anti-DENV drugs or vaccine to against DENV infection because little is known about the complex and highly dynamic process of dengue infection. Among DENV infectious process, receptor-binding is the first critical step contributes for successful infection. Elucidation of the receptor-binding details is helpful for developing anti-viral agents. Since force plays an important role in interaction dynamic and structure of biomolecule and the binding force between virus and receptor has not been extensively studied, we would like to measure the binding forces between ideally single virus particle and host cellular receptors by combining OT and a single-virus particle tracking approache that we had developed recently [2]. We have used OT to trap a single bead bound with DiI-labeled DENV particles (Figure 1). To determine the interaction force between DENV particle and host cellular receptor(s), a single cell ectopically expressed DC-SIGN was moved to touch the single virus-bound bead (Figure 2). The single virus binding forces is 3.4� 0.6 pN in the absence of the expression of DC-SIGN in THP-1 cells. In contrast, in the presence of DC-SIGN expression, the force is increased to 43.7� 8.1 pN. Herein, we not only have successfully elucidated the interaction force of DENV particle with a cellular receptor, but also demonstrate that OT is a real-time force measurement approach to determine the interaction forces between biological molecules in living cells. [1] Stangner, T. et al, ACS nano 7 (2013), 11388-11396. [2] Chu, L.W. et al, J Biomed Opt 19 (2014), 011018. Microsc. Microanal. 21 (Suppl 3), 2015 220 Figure 1. Design single quantity DiI-labeled DENV particles bound on polystyrene (PS) beads. (A) The size of single DiI-labeled DENV particle was measured using super-resolution microscopy. The scale bar: 40 nm. (B) The DIC and fluorescence images of single quantity DiI-labeled DENV particle bound on PS beads with epi-fluorescence microscopy. The scale bar: 1 m. (C) The DIC and fluorescence images of single quantity DiI-labeled DENV particle bound on PS beads under super-resolution microscopy. The scale bar: 200 nm. Figure 2. The measurement of force between virus bound on a bead and receptor on cell membrane surface by OT. (A) Schematic diagram of OT system. (B) Schematic diagram of the protocol to determine the binding force. (C) The binding force of DENV particle with DC-SIGN receptor *: p<0.05");sQ1[111]=new Array("../7337/0221.pdf","Application of FRET microscopy for visualizing the uncoating process of a single Dengue virus in living cells","","221 doi:10.1017/S1431927615001907 Paper No. 0111 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Application of FRET microscopy for visualizing the uncoating process of a single Dengue virus in living cells Li-Wei Chu1, 3, 4, Ya-Hsuan Lin1, 3, 4, Chau-Hwang Lee1, 3, 5, and Yueh-Hsin Ping1, 2, 3, 4 1. 2. Institute of Biophotonics, National Yang-Ming University, Taipei, Taiwan. Department and Institute of Pharmacology, National Yang-Ming University, Taipei, Taiwan. 3. Biophotonics interdisciplinary Research Center, National Yang-Ming University, Taipei, Taiwan. 4. Infection and Immunity Research Center, National Yang-Ming University, Taipei, Taiwan. 5. Research Center for Applied Sciences, Academia Sinica, Taipei, Taiwan. Single-virus tracking is a real-time imaging technique that can monitor successfully individual virus entry and trafficking behavior in live cells [1]. Conventional wide-field fluorescence microscopy for singlevirus tracking enables localization of fluorescently labeled molecules within the optical spatial resolution limits defined by the Rayleigh criterion, approximately 200 nanometers. However, the scale of virus uncoating involved virus-host membrane fusion is within dozens nanometer. It is smaller than optical resolution limitation and failed to detect by single-particle tracking. In order to resolve this limitation, the F�rster resonance energy transfer (FRET), when applied to optical microscopy, permits determination of the approach between two molecules within several nanometers [2]. It can be useful to determine the virus-host membrane fusion in viral uncoating events. In our previous study, using a single-virus particle tracking technology demonstrated that dengue virus (DENV) particles interacted directly with autophagosomes, and autophagy facilitated viral particle trafficking and promoted viral replication [3]. Herein, we established the FRET pair fluorophore labeling DENV particles using DiI as a FRET donor and DiD as a FRET acceptor. We hypothesized that the distances of DiI and DiD molecules in virus envelope are elevated during viral uncoating process, resulting in the decreasing of FRET efficiency (Figure 1A). The decreasing FRET signals can be considered as viral uncoating. To characterize the properties of DiI/DiD-labeled DENV, the fluorescent intensity and the FRET effect of DiI/DiD-labeled virus were measured by confocal microscopy and acceptor-photobleaching imaging (Figure1B&C). The FRET efficiency of DiI/DiD-labeled virus particles is about 0.75 (Figure 1C). After 30 min postinfection for the internalization of DENV, the DENV particles that colocalized with autophagosomes (green puncta) presented lower FRET efficiency in FRET accptor-bleaching imaging (Figure 2, white arrow), indicating that DENV uncoating occured in autophagosomes. In addition, we also measured the alteration of fluorescence intensity of individual DENV particles in living cells by single-particle tracking. In this expirement, the 532 nm laser was used to excite the FRET donor, DiI, and both DiI and DiD signal were recorded to caculate the emision ratio of DiD/DiI. The 488 nm laser was used to excite GFP-LC3 fusion protein, as a marker of autophagosome. The results showed that DENV particles co-transported with autophagosomes (green puncta) and presented the decreasing ratio of DiD/DiI suggesting that DENV uncoating is correlated the colocalization of viral particle with autophagosome (Figure 3). In conclusion, we utilized lipophilic DiI and DiD dye to generate FRET pair fluophores labeling DENV particles. This approach offers an opportunity to reveal DENV uncoating process in living cells at singlevirus level by FRET microscopy. Our results elucidated that autophagy facilitates the uncoating process of DENV. Furthermore, the combination of FRET and a single-virus tracking approaches not only provides more detail of virus-host interaction, but also can be further develop as a screening platform for developing anti-viral drugs. [1] B. Brandenburg, X. Zhuang, Virus trafficking - learning from single-virus tracking, Nat Rev Microsc. Microanal. 21 (Suppl 3), 2015 222 Microbiol 5 (2007) 197-208. [2] J. Zheng, Spectroscopy-based quantitative fluorescence resonance energy transfer analysis, Methods Mol Biol 337 (2006) 65-77. [3] L.W. Chu et al, Single-virus tracking approach to reveal the interaction of Dengue virus with autophagy during the early stage of infection, J Biomed Opt 19 (2014) 011018. Figure 1. (A) Scheme of DiI & DiD labeling as FRET pair for visualizing DENV uncoating. Characterization of DiI-DiD labeled DENV represented by (B) Confocal and (C) FRET Acceptor photoleaching imaging. (D) Quantification of FRET efficiency of DENV particles (n=50). Figure 2. DENV colocalized with autophagosomes and occurred to FRET signal decay (white arrow). When DENV didn't colocalize with autophagosomes, it kept high FRET signal (arrow head). Figure 3. FRET microscopy applied to single-virus tracking for visualizing individual DENV uncoating event in living cells. Scale bar: 2 m.");sQ1[112]=new Array("../7337/0223.pdf","The Observation of Saccharomyces cerevisiae Ultrastructure Changes under Proline Limitation","","223 doi:10.1017/S1431927615001919 Paper No. 0112 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Observation of Saccharomyces cerevisiae Ultrastructure Changes under Proline Limitation Han Chen1, Xinwen Liang2 and Donald Becker2 1. Microscopy Core Facility, Center of Biotechnology, University of Nebraska Lincoln, Lincoln, Nebraska, USA 2. Redox Biology Center and Department of Biochemistry, University of Nebraska Lincoln, Lincoln, Nebraska, USA Proline is an important amino acid, which involves in not only protein synthesis, but also in various stress response in many organisms [1]. It was suggested that increase of intracellular proline level protects plants or yeast cell against stress via improvement vacuole biogenesis since vacuole is an important compartment for cell survive under various stress [2]. Autophagy is a conserved cellular process that mediates protein degradation in lysosomes called vacuoles in yeast [3]. It plays an important role in resource energy when cells undergoing nutrient starvation. Nitrogen starvation is a common used assay to test whether or not cells bear functional autophagy to nutrient starvation. Here we report the proline biosynthesis gene PRO3 null mutant of Saccharomyces cerevisiae BY4741 is hyper sensitive to nitrogen starvation, the survive rate under nitrogen starvation is similar to that of atg8 mutant strain in which autophagy is deficient. However other amino acid starvation such as leucine, histidine doses not affect yeast cells survive under shortage of nitrogen source. The autophagy bodies were further examined using transmission electron microscope. While growing in minimal medium without nitrogen for 4 hours, similar to atg8 mutant, the yeast cells of pro3 strain showed significant fewer autophagy bodies compared with wild-type and pep4 mutant. Our data suggests that proline may play an important role in autophagy formation. Microsc. Microanal. 21 (Suppl 3), 2015 224 References: [1] JM Phang et al, Annu Rev Nutr 30(2010): 441-463. [2] K Matsuura and H Takagi, J Biosci Bioeng 100(2005): p538-544. [3] K Takeshige et al, J of Cell Biology 119(1992): p301-311 [4] J Mulholland and D Botstein, Methods in Enzymology 351(2002): p61-70 [5]The authors acknowledge funding from the National Institute of Health, Grant Number: GM079393. Figure 1. Ultrastructure of Saccharomyces cerevisiae BY 4741 cells grown in different medium. Saccharomyces cerevisiae BY 4741wild type, pep4, pro3, and atg8 mutants were growing at minimal synthetic minimal medium (SD) to log phase, then inoculated to minimal medium without nitrogen or without nitrogen and leucine, and with addition of 1mM PMSF for four hours, then subjected to transmission electron microscope photography [4]. Under nitrogen starvation (B,D-F) or nitrogen and leucine starvation (C) for 4 hours, wild type (B,C) and pep4 (D) mutant cells had accumulated multiple autophagic bodies with high-electron density in the vacuole, the bodies surrounded by a unit membrane, and the content of the bodies is morphologically indistinguishable from some cytosolic organelles. However, the autophagic bodies were not observed in pro3 (E) and atg8 (F) null mutant cells upon nitrogen starvation or wild type growing in SD medium (A). V: vacuole, Abs: autophagy bodies, M: mitochondria, N: nucleotide. Bars: 1�m.");sQ1[113]=new Array("../7337/0225.pdf","The Wisconsin Electron Microscopy Diagnostic Proficiency Program","","225 doi:10.1017/S1431927615001920 Paper No. 0113 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Wisconsin Electron Microscopy Diagnostic Proficiency Program Craig Radi1, Sara E. Miller2, Cynthia Goldsmith3, Kathy Toohey-Kurth 1,4 1 2 Wisconsin Veterinary Diagnostic Laboratory, Madison, WI Pathology EM, Duke University, Durham, NC 3 Infectious Disease Pathology Branch, Centers for Disease Control, Atlanta, GA 4 Pathobiological Sciences, School of Veterinary Medicine, University of Wisconsin, Madison, WI Negative stain imaging of specimens for virus identification is on the decline in some electron microscopy (EM) diagnostic laboratories. Our laboratory went from over 2000 negatively stained [1] samples to less than 100 per year. This dramatic decrease in samples can have a deleterious effect on keeping one's eye keen for spotting unknown pathogens that do come in for transmission electron microscopy (TEM) examination. In 2007, the Centers for Disease Control (CDC) and Duke University partnered in a program to improve the use of TEM laboratories in bioterrorism preparedness. Hosts Cynthia Goldsmith from the CDC and Sara Miller from the Department of Pathology at Duke sought to fill a gap in expertise. During this course, it was apparent that a number of laboratories did not have access to a variety of viruses or adequate diagnostic samples to improve identification skills. The Wisconsin Veterinary Diagnostic Laboratory (WVDL) has a comprehensive virology service and offered to provide additional practice samples. To this end, virologists at WVDL prepare and ship inactivated viruses to the participating laboratories. Each participant then has the opportunity to prepare and evaluate their own grid as well as make the diagnosis. This was the start of the "Wisconsin EM Diagnostic Proficiency Program" and is currently the only one in the United States. The laboratories that have participated show improved expertise in their abilities to prepare and identify unknown viruses grown in tissue culture [Figures 1, 2]. This program has been growing for the last 7 years with 17 laboratories from the mainland US, Hawaii, and Canada currently participating. PCR has become the standard in many diagnostic tests; however it can miss unsuspected pathogens if the correct primer is not chosen or the sequence is not available. If one simply suspects a viral pathogen, based on symptomatology and runs a PCR test for that pathogen, the "guess" may be incorrect, and the test result would be negative; yet, a virus could, nevertheless, be present. Additionally, some viruses do not grow or are fastidious in culture, making large-scale production of reagents difficult. Thus, commercial diagnostic primers and antibody reagents do not exist for all viral pathogens. In the case of electron microscopy, if a virus is present in sufficient numbers, and if the microscopist is experienced in viral morphology, the pathogen can be recognized and identified. This examination of "whatever is there" has been termed the "Open View" [2] by Hans Gelderblom, one of the giants in the field of virus diagnosis and whose program we have emulated here. In summary, one should make use of all the tools in their diagnostic arsenal, including EM, when it comes to detecting unknown viruses in samples. We encourage all diagnostic laboratories with EM facilities to participate in this proficiency program. Microsc. Microanal. 21 (Suppl 3), 2015 226 1. 2. Figure 1. Image of Herpesvirus on grids prepared by participating lab. Image by Sara Miller. Figure 2. Image of Rhabdovirus on grids prepared by participating lab. Image by Margaret Casey. References: [1] M.A. Hayat, S. E. Miller. Negative Staining. McGraw-Hill Pub. Co., 1990, 253 pp. . [2] P. R. Hazelton, HR Gelderblom. Emerg Infect Dis. Mar 2003; 9(3): 294�303. doi: 10.3201/eid0903.020327. PMCID: PMC2958539");sQ1[114]=new Array("../7337/0227.pdf","Wheat Leaf Rust Fungus: RIMAPS Analysis to Detect Germ-tube Orientation Pattern.","","227 doi:10.1017/S1431927615001932 Paper No. 0114 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Wheat Leaf Rust Fungus: RIMAPS Analysis to Detect Germ-tube Orientation Pattern. L.M. Setten1, N. Lendoiro2 and E.A. Favret3,4 1 2 Lab. Microscop�a y Capacidades Afines, CICVyA, INTA, Hurlingham, Argentina. Fac. Cs. Exactas, Qu�micas y Naturales, Universidad de Mor�n, Mor�n, Argentina. 3 Instituto de Suelos, CIRN, INTA. Hurlingham. Argentina. 4 CONICET. Buenos Aires. Argentina. Wheat is severely affected by a destructive disease, leaf rust, caused by fungus, Puccinia triticina E. All over the world, wheat leaf rust reduces crop yield up to 10%. When the uredospore reaches the leaf surface develops germ-tubes, which are directed towards the stomata. The germ-tube grows and elongates perpendicular to the long axis of the epidermal cells, this orientation is considered to be response to stimuli from the host (Fig. 1), such us physical and chemical features [1]. The topography of the leaf surface has been proposed by several investigators as a major factor that can influence the direction of the germ-tube growth. Our research work is based on determining which of these topographic parameters influence the development of the germ-tubes, by using artificial surfaces with known topographies. For that purpose it is necessary to quantify the angular orientation of the germ-tube pattern as a first step. The present paper shows how RIMAPS can be used as a tool to detect this orientation pattern. The RIMAPS technique consists basically of rotating the image using algorithms of commercial software and calculating the x-step of the two-dimensional Fourier transform for each y-line of the new image obtained after rotation. Averaged power spectra are obtained for each angular position. The corresponding maximum values are plotted as a function of rotation angle. The maxima of the RIMAPS spectrum indicate main angular directions of the topographic pattern [2]. For this study, two samples of an inert material, polystyrene (PS), were used. The surface of one sample was micro-structured with direct laser interference patterning technique mimicking the direction of the epidermal cells (Fig. 2), the other was smooth [3]. Both samples were inoculated with rust spore suspension. The surfaces were examined with a VP-SEM FEI Quanta 250. Afterwards, for the RIMAPS studies hand-made drawings from the SEM images were done in order to copy only the pattern of the germ-tubes observed on the PS surfaces (Fig. 3-4), eliminating any artifacts of the surface sample. The RIMAPS spectra are seen in figures 5 and 6. Figure 5 indicates two main maxima (4�, 142�). The direction of the micro-structure corresponds to 90�. Therefore, RIMAPS shows that the main orientation (4�) of the germ-tube pattern is almost perpendicular to the micro-structure. The other maximum (142�) needs to be properly explained with more studies. The RIMAPS spectrum of figure 6 determines several main maxima (18�, 75�, 112�, 147�). In this case the orientation of the germ-tube pattern has no preferential direction. Presently, other parameters of the spectrum are also been considered, such as the integral of the curve which increases with the number of germ-tubes directions, e. g., figure 5 has an integral of 118,5 and figure 6 an integral of 173,2. The difference observed between the RIMAPS spectra would support that the surface topography is one factor that influences the orientation of germ-tube and would be a method to quantify the orientation [4]. References [1] G Hu and FHJ Rijkenberg, Mycol. Res. 102 (1998), p. 391-399. [2] NO Fuentes and EA Favret, Journal of Microscopy 206 (2002), p. 72-83. [3] A Lasagni et al., Proc. of SPIE 8968 (2014), p. 1-9. [4] The authors acknowledge Andr�s Lasagni, Francisco Sacco and Lorena La Fuente for their collaboration. Microsc. Microanal. 21 (Suppl 3), 2015 228 Figure 1. SEM of a rust spore showing the orientation of germ-tube on the leaf wheat. Figure 2. SEM of a rust spore showing the orientation of germ-tube on the microstructured PS. Figure 3. Hand-made drawing copy of the germ-tubes on the microstructured PS. Figure 4. Hand-made drawing copy of the germ-tubes on the smooth PS. Figure 5. RIMAPS spectrum of Figure 3. Figure 6. RIMAPS spectrum of Figure 4.");sQ1[115]=new Array("../7337/0229.pdf","Ice Contamination Issues in the Visualization of the Ultrastructure of the Nuclear Envelope by Freeze-Fracture Technique","","229 doi:10.1017/S1431927615001944 Paper No. 0115 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Ice Contamination Issues in the Visualization of the Ultrastructure of the Nuclear Envelope by Freeze-Fracture Technique Nadezda Vaskovicova1, Vladislav Krzyzanek1 and Ales Patak1 1. Institute of Scientific Instruments of the ASCR, v.v.i., Brno, Czech Republic. The freeze-fracture technique and processing of the metal replicas of fractured structures is an excellent option for visualization of details in the nuclear membrane and for study of the ultrastructure of nuclear pores. In the replica the number of nuclear pores per unit area and the distribution of them can be evaluated, as well as the number of membrane proteins and their size [1], [2]. The process of replica preparation includes: freezing the biological sample, fracturing the sample under low temperature and in a high vacuum, shadowing fractured structures with metal (platinum) and stabilization of the metal by carbon deposit, and after melting the sample the residue of the biological material is removed by chemical agents. The replicas were fixed on cupper grids and observed in a transmission electron microscope. In these measurements replicas of HL-60 cells (human leukemic cells) were used. A holder with the samples was kept at the temperature -100�C in the vacuum chamber (Freeze Etching System BAF060, BAL-TEC). Time of etching: the ice contaminated sample was not etched, the sample without ice contamination was etched 15 min. The most critical part of a sample processing is freezing. Biological material in a culture medium or in a buffer can be frozen in cryogen but in most cases this leads to ice crystal production. The solid ice crystals occur in the exoplasmic face of the inner or outer nuclear membrane, figure 1A. This means that the ice contamination originates from the perinuclear space. Removal of ice crystals from a membrane could be processed by etching (ice sublimation). The ice is removed not only from the membrane of the organelles but also from the cell cytoplasm and the extracellular space. This leads to the exposure of a large surface of organelles and it can cause complications with replica production. Another option is to use cryoprotectives. The usual cryoprotectant used in the freeze-fracture technique is glycerol. Glycerol replaces a certain portion of water in the sample and prevents the formation of ice crystals. In addition, in the case of contamination by frost due to the transfer of the sample from a cryogen into a high vacuum chamber of Freeze Etching System, the ice crystals can be removed by sublimation from the sample surface without losing the sample material or exposing a large surface of organelles, figure 1B. The reason is that the cryoprotectant does not sublimate as fast as water. In order to evaluate pore distribution or size of pores it is necessary to detect individual pores in the membrane. In the recorded micrograph, the large and solid crystals can appear so close to a pore that it could be difficult to distinguish them from the edges of the pore or position in the membrane, figure 1A. Thus, the evaluation of protein distribution in the membrane and the exact number of them can be problematic. The number of proteins per unit area can be variable but there are some limits. With ice contamination the mean number of particles rapidly increases. The number of particles in the membrane contaminated by ice crystals is 490 � 40 per 1 �m2, and in the membrane without ice crystals 242 � 24 per 1 �m2. These results indicate that due to the contamination of the sample by ice crystals the number of particles in the membrane doubled. Microsc. Microanal. 21 (Suppl 3), 2015 230 In figure 2 the histogram shows the size of particles in the exoplasmic face of the nuclear membrane. The mean size of particles in the contaminated sample is 12 � 8 nm and in the membrane without ice contamination is 9 � 5 nm. The number of particles with size from 15 nm to 30 nm is higher in the contaminated membrane and there are some particles with size above 35 nm. In the ice-free membrane the largest particle has size of 32 nm. It can be concluded that the ice contamination influences the results of the evaluation. Ice crystals can be mistaken for membrane structures and complexes of proteins. On one hand using cryoprotectants and long etching can prevent mistakes in evaluation on the other hand their application may not be consistent with the aim of the experiment. References: [1] N. Vaskovicov� et al, J. Appl. Biomed. 11 (2013), p. 235. [2] A. Valigurov� et al, Frontiers in Zoology 10 (2013) 57. [3] The research was supported by MEYS CR (LO1212) and its infrastructure by MEYS CR and EC (CZ.1.05/2.1.00/01.0017). 200 nm 200 nm Figure 1. Exoplasmic face of a nuclear membrane: A) the ice contaminated membrane, B) the ice-free membrane. Black arrows show nuclear pores, white arrows show solid ice crystals in the membrane and small white unfilled arrows show particles in the membrane which could be small ice crystals. Figure 2. Histogram shows size of particles in the ice contaminated membrane (white boxes) and in membrane without ice contamination (black boxes). This shows that larger particles exist in the contaminated sample in contrast to the contamination-free sample.");sQ1[116]=new Array("../7337/0231.pdf","Imaging of Polyethylene Glycol Layers on Nanoparticles","","231 doi:10.1017/S1431927615001956 Paper No. 0116 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Imaging of Polyethylene Glycol Layers on Nanoparticles Sarah R. Anderson1, Mackensie C. Smith2, Jeffrey D. Clogston2, Anil K. Patri2, Scott E. McNeil2, and Ulrich Baxa1 Electron Microscopy Laboratory, and 2. Nanotechnology Characterization Laboratory, Cancer Research Technology Program, Leidos Biomedical Research, Inc., Frederick National Laboratory for Cancer Research, Frederick, MD 21702 The encapsulation of nanoparticles by adding polyethylene glycol (PEG) to their surface (PEGylation) is an important method to protect them from the immune system and from uptake by the reticuloendothelial system [1]. The addition of PEG creates nanoparticles with so-called "stealth" behavior that have significantly increased circulation times, reduced aggregation behavior, and less interaction with non-targeted serum and tissue proteins [2]. The PEG coating also increases solubility of nanoparticles in serum and buffer due to the long hydrophilic ethylene glycol repeats. The length of the PEG chain can be chosen according to application and the ends are usually modified for nanoparticle attachment on one end and often have a methoxy group on the other end. Characterizing the PEG coating on nanoparticles is an essential part of nanoparticle synthesis optimization and quality control. There are several bulk methods to check the presence of PEG on nanoparticles including UV-vis, DLS, and zeta potential to name a few. But imaging the PEG layer directly on a particle by particle basis in the electron microscope has been a challenge. The PEG layer is basically invisible under most conditions for drop-cast nanoparticles and for cryo-EM specimens. Here we show that negative stain does not enable visualization of PEG layers in spite of several claims in the literature. However, the commercial availability of PEG antibodies (e.g. from Life Diagnostics, Inc.) allows the application of traditional IEM methods for the routine visualization of PEG layers on nanoparticles. Colloidal gold nanoparticles, citrate-stabilized, and grafted with 10kDa and 20kDa mPEG have been imaged by traditional negative staining methods using uranyl acetate, uranyl formate, and methylamine tungstate (NanoW). Weak coronas have been observed around many of the particles including the nanoparticles without PEG. The appearance of the corona is strongly dependent on staining conditions and depth and usually appears dark (more stain around gold particle) but under some conditions appears bright (less stain around gold nanoparticle). However, we show the presence of such coronas in all samples including the gold nanoparticles without PEG (Figure 1). Such coronas have been used in the literature as evidence of the presence of PEG on nanoparticle surfaces. However, based on our experiments we conclude that negative staining is not a reliable method to verify presence and size of the PEG layer. Traditional solid phase IEM methods have been used to visualize the PEG layer of gold nanoparticles and metal oxide nanoparticles. Particles were applied to carbon film and imaged after a common IEM protocol. The IEM reaction was highly specific and had very little background (Figure 2). Care must be taken to separate any free PEG from the nanoparticles before application to the surface. 1. Microsc. Microanal. 21 (Suppl 3), 2015 232 In summary, we find IEM to be an easy, reliable, and reproducible method to visualize PEG layers on nanoparticles. Negative staining on the other hand was shown to not visualize the PEG layer and is at best considered highly unreliable. References [1] Jokerst JV, Lobovkina T, Zare RN, and Gambhir SS, Nanomed 6 (2011), p. 715 - 728. [2] van Vlerken LE, Vyas TK, and Amiji MM, Pharm. Res. 24 (2007), p. 1405 � 1414. [3] This project has been funded with Federal funds from the Frederick National Laboratory for Cancer Research, National Institutes of Health, under contract HHSN261200800001E. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products or organizations imply endorsement by the US Government. Figure 1. Negative staining with uranyl acetate of 60 nm gold nanoparticles: (A) without PEG. (B) coated with 10 kDa PEG. Bar = 500 nm in main panel and bar = 100 nm in insert. The presence of PEG on these particles has been verified with DLS and Zeta potential measurements. Figure 2. IEM of metal oxide particles using 10 nm gold labels: (A) antibody specific to PEG main chain (ethylene glycol moiety). (B) antibody specific to the methoxy end of PEG chain. (C) control = no primary antibody. Bar = 100 nm.");sQ1[117]=new Array("../7337/0233.pdf","Treatment of Influenza-Induced Acute Lung Injury with Iron Oxide Nanoparticles using an Ischemic-Reperfusion Model","","233 doi:10.1017/S1431927615001968 Paper No. 0117 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Treatment of Influenza-Induced Acute Lung Injury with Iron Oxide Nanoparticles using an Ischemic-Reperfusion Model K. Chew1, A. Waldron2, C. Queenan2, R. Pergolizzi1, and T. Hoffmann3 1. 2. Bergen County Academies, Stem Cell Research Lab, 200 Hackensack Avenue, Hackensack, NJ 07601 Bergen County Academies, Nano-Structural Imaging Lab, 200 Hackensack Avenue, Hackensack NJ 07601 3. Englewood Hospital and Medical Center, Surgical Research Laboratory, 350 Engle Street, Englewood NJ, 07631 Influenza is a highly contagious viral infection. To prevent complications such as pneumonia, a better treatment for lung inflammation must be found. One potentially effective solution may be to utilize iron oxide nanoparticles (ION). Nanoparticles have become increasingly used in medicine. Specifically, ION have been studied for their potential in influenza diagnosis [1]. Some studies have shown that ION assist in stabilizing inflammation in certain cell lines [2], while other studies have used ION to induce inflammation [3]. The interaction between ION and human monocyte-macrophages (HMMs) has also been studied [4]. HMMs are involved inflammatory signaling within a cell. It was found that ION were retained within HMM lysosomes, which prevented them from inducing inflammatory cytokines. In this study, it was hypothesized that ION stabilize inflammation. When the influenza virus enters a lung cell, the inflammatory process is initiated. As a result, the protein tumor necrosis factor alpha (TNF-) is secreted [5]. Additionally, nitric oxide acts as a mediator in the inflammatory process; it is consumed as inflammation increases. Therefore, cells studied in vitro that are stimulated with TNF- would produce lower levels of nitric oxide as compared to cells stimulated with ION [6]. It was also hypothesized that ION would be non-toxic to the proliferation of cells; cells treated with higher concentrations of ION would have increased cell viability than those treated with lower concentrations of ION. This hypothesis can also be studied in vivo. A super mesenteric artery (SMA) ischemia reperfusion procedure mimics the mechanism that causes TNF- secretion after interaction with influenza virus. Lipid peroxidation, overproduction of nitric oxide, and other complex mechanisms caused by SMA ischemia reperfusion results in lung injury [7]. In lung samples, the anatomical distance between each alveoli sac can be measured [8]; a large distance between each alveoli sac indicates the collapse of the sacs, leading to capillary obstruction, which is a mark of inflammation [9]. Additionally, the number of edema-filled sacs within a lung is a further indicator of inflammation [10]. Therefore, in vivo samples obtained after ION inhalation should indicate a reduced amount of lung injury, which can be determined by measuring the distance between alveoli sacs and calculating the number of edema filled sacs. For in vitro studies, application of ION to lung cells stimulated with TNF- showed increased levels of nitric oxide, indicating a decrease in inflammation. At high concentrations of ION, nitric oxide secretion nearly matched that present in cells without TNF- treatment. Additionally, cell viability in a 48-hour period was consistent for all concentrations of ION, suggesting that the nanoparticles were not cytotoxic. For in vivo studies, there was a statistically significant difference in the level of lung injury between samples that inhaled saline and samples that inhaled ION. Overall, ION show promise as an effective treatment for influenza-induced lung injury [11]. Microsc. Microanal. 21 (Suppl 3), 2015 234 References [1] T. Chou, et al., Journal of Nanobiotechnology 9 (2011) 1-13. [2] A. Jefferson, et al., Theranostics 3 (2013) 428. [3] A. Elder, et al., In Nanotechnology and the Environment: Applications and Implications Progress Review Workshop III 10 (2005) 105. [4] K. M�ller, et al., Biomaterials 28 (2007) 1629-1642. [5] I. Julkunen et al., Cytokine & growth factor reviews 12 (2001) 171-180. [6] J. Jiang, et al., Nitric Oxide 25 (2011) 275-281. [7] K. Liu, et al., World Journal of Gastroenterology 13 (2007) 299. [8] J. Warren, et al., Journal of Clinical Investigation 84 (1989) 1873. [9] M. McElroy, et al., European Respiratory Journal 24 (2004) 664-673. [10] L. Colletti, et al., Journal of Clinical Investigation 85 (1990) 1936. [11] The authors would like to acknowledge the administration of the Bergen County Technical Schools for their continued support of the research program. Impact of ION on Nitric Oxide Secretion in TNF-a Stimulated Epithelial Cells Nitric Oxide (mM) 1500 1000 500 0 * * * * * 0 5 10 20 50 100 200 Concentration of Iron Oxide Nanoparticles (�g/ml) 300 400 Figure 1: Nitric oxide levels in epithelial cell cultures treated with TNF- for 24 hours, followed by incubation with nanoparticles for 2 hours. Data is represented by mean � SD (n=3). Asterisk indicates statistical significance as compared to control (p< 0.005). Figure 2: Light micrographs of microthin sections of lung tissue stained with hematoxylin and eosin. Images A, C, and E acquired at 100X magnification. Images B, D, and F acquired at 400X magnification. Images A and B: normal lung tissue. Images C and D: lung tissue after ischemia reperfusion (IR) and inhaling saline. Images E and F: lung tissue after the IR and inhaling ION.");sQ1[118]=new Array("../7337/0235.pdf","Structure of Microtubule-Based Microtentacles","","235 doi:10.1017/S143192761500197X Paper No. 0118 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Structure of Microtubule-Based Microtentacles 1 2 Alison N. Killilea1, Roseann Csencsits1, Sam Kenny1,2, Ke Xu1,2, Kenneth H. Downing1 Life Science Division, Lawrence Berkeley National Laboratory, Berkeley, CA Department of Chemistry, University of California Berkeley, Berkeley, CA Microtentacles (McTNs) are thin projections found on circulating tumor cells (CTCs) that can be tens of micrometers long. They have been recognized as a potentially critical factor in metastasis due to their involvement in attachment of CTCs to secondary sites of metastasis. So far little structural characterization has been reported, other than that these projections are based on microtubules (MTs), unlike most other cell projections that are based on actin. Work published to date strongly suggests that stabilization of McTNs increases the risk of metastasis through enhanced reattachment of CTCs [1-3]. Stabilization occurs by a number of factors that often increase during progression of cancer, including increased expression of proteins such as vimentin and tau [2, 4]. Of particular clinical importance, Taxol stabilizes MTs and as a result stabilizes McTNs. Since Taxol is widely used in treatment of cancer, it is very important to understand the magnitude and structural basis for any increased risk of metastasis. In addition, developing a structural understanding for these effects would help to identify novel drug targets that could compensate for any added risk that treatment with drugs such as Taxol poses. With the goals of understanding McTN structure and the mechanisms by which stabilization of McTNs promote reattachment, we have used cryo-electron microscopy and tomography, scanning electron microscopy (SEM) and stochastic optical reconstruction microscopy (STORM) to characterize the structure and activity of these protrusions. SEM was used to follow the development of McTNs after drug treatment of MDA-MB-231 breast cancer cells with latrunculin (LA) and Taxol, with results shown in Fig. 1. A time course series showed formation of McTN within 3 minutes of drug addition (data not shown), and at 15 minutes (Figure 1b) McTNs longer than 10�m can be seen. We have studied their molecular composition by fluorescence labeling. New advances in super-resolution microscopy have allowed for lateral resolution of 10nm and axial resolution of 20nm, which is ideal for investigating the arrangement of 10nm intermediate filaments (IFs) and 25nm MTs inside McTNs. Using dual-objective STORM we imaged MTs and vimentin IFs in suspended MDA-MB-231 breast cancer cells treated with LA and Taxol. In general, the filaments inside the McTNs are arranged parallel to each other but appear spatially separated (Fig. 2). Resin embedding and sectioning were used to provide a view of structures within thicker parts of the McTNs near the cell body. MDA-MB-231 cells treated with LA and Taxol show structural details including the IFs, MTs and ribosomes (Fig. 3). The separation between MTs and IFs inside the McTN suggests that another protein may be linking these networks. Immunofluorescence experiments revealed the presence of plectin, a cytoskeletal linker protein, inside McTNs (data not shown). Cryo tomography has been used to obtain clear 3D maps of the MT distribution within thin parts of the McTNs (Fig. 4). So far we have seen no IFs in these regions, supporting the idea that these two components of the cytoskeleton are not tightly interconnected, even though there is evidence that vimentin helps to stabilize McTNs [4]. Microsc. Microanal. 21 (Suppl 3), 2015 236 Figure 1. McTN visualized by SEM. MDA-MB231 cells were suspended over ultra low attachment plates, treated with 5�M LA and 1.7 �M Taxol for 15 minutes. Cells were fixed in 2.5% glutaraldehyde followed by ethanol dehydration, critical point drying and gold-palladium sputter coating. Left: control; right: drug treated cell. Figure 2. STORM imaging of MT and IF networks in McTN. Glutaraldehyde-fixed cells were labeled with mouse DM1 antitubulin and chicken polyclonal anti-vimentin antibodies, washed, and AlexaFluor 647 and CF680 secondaries were applied. Figure 3. TEM visualization of internal McTN structure. Microtubules (MT) and intermediate filaments (IF) are visible, although they are often found in different regions of the sample. MDA-MB-231 cells were suspended over ultra low attachment plates and treated with 5�M LA and 1.7 �M Taxol for 30 minutes before application to polylysine treated, carbon-coated sapphire discs followed by fixation in 2% paraformaldehyde and 0.5% glutaraldehyde. Discs were high pressure frozen, freeze substituted and resin embedded. Figure 4. Cryo-electron tomography of McTN. (top) Section ~3 nm thick through 3D tomographic reconstruction obtained from tilt series of low-dose images of frozen-hydrated cell. Parts of several microtubules can be seen passing up and down through this section. (bottom) Model for distribution of MTs enclosed by membrane of McTN derived from full 3D map. References: [1] Balzer EM, et al, Oncogene 29 (2010), 6402. [2] Matrone et al, Oncogene 29 (2010), 3217. [3] Charpentier MS, et al, Cancer Res. 74 (2014), 1250. [4] Whipple RA, et al, Cancer Res. 68 (2008), 5678.");sQ1[119]=new Array("../7337/0237.pdf","Bartonella henselae Biofilm Detected on Catheter of Patient with Persistent Bartonellosis","","237 doi:10.1017/S1431927615001981 Paper No. 0119 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Bartonella henselae Biofilm Detected on Catheter of Patient with Persistent Bartonellosis Melissa Weber-Sanders1, Paulo ENF Velho2, Gislaine Vieira-Damiani2, Marilene Neves da Silva2, Vitor B. Pelegati3, Carlos Lenz Cesar3 and Marna Ericson1 1. Department of Dermatology, University of Minnesota Medical School, Minneapolis, MN USA. 2. Department of Medicin and Dermatology, UNICAMP Medical Sciences School, Campinas, Brazil 3. National Institute of Science and Technology on Photonics Applied to Cell Biology (INFABIC), UNICAMP Medical Sciences School, Campinas, Brazil Bartonella spp. can induce long-lasting bacteremia in mammals (1). Bartonella spp. are causative agents of cat-scratch disease, endocarditis, bacillary angiomatosis, bacillary peliosis, trench fever, neurocognitive dysfunction and regional pain syndrome (2). Chronic Bartonella spp. infection and subsequent manifestations of persistent infection may be the result of the bacteria growing in biofilms. In humans, at least four Bartonella species have been associated with infectious endocarditis, a biofilmassociated infectious disease (3). Biofilms are sessile communities of interacting unicellular organisms that adhere to a surface and to each other and create unique, complex structures, covered with a protective layer of extracellular polymeric substances (EPS). Biofilms and their inherent resistance to antimicrobial agents are often key players in many persistent and chronic bacterial infections (4, 5). We acquired skin biopsies from 5 patients with blood PCR-positive B. henselae. 60-100 micron-thick sections were immunostained with antibodies to B. henselae. Stained sections were imaged using both single- and multi-photon imaging to capture epifluorescence emission and second harmonic generation signal of non-stained collagen I and II (Fig 1). The crystalline triple-helix of native fibrillar collagen (I, II, III and V) generates a strong second harmonic generation (SHG) signal (6). In patients with longstanding bartonellosis, clusters of immuno-reactive B. henselae were found intimately associated with fibrillar dermal collagen (Fig 2). Additionally, we found a Bartonella spp. biofilm on a peripheral intravenous catheter line removed from a patient with persistent Bartonella (Fig 3). Further, B. henselae, cultured 14 or 21 days, on collagen I and II-coated coverslips revealed a B. henselae biofilm matrix (Fig 4). Identification and characterization of Bartonella spp. biofilms in vivo and in situ will lead to a better understanding of persistent Bartonellosis and provide direction in treatment modalities. 1) Breitschwerdt, E. B. et al (2010). "Bartonellosis: an emerging infectious disease of zoonotic importance to animals and human beings." J Vet Emerg Crit Care (San Antonio) 20(1): 8-30. 2) Harms, A. and C. Dehio (2012). "Intruders below the radar: molecular pathogenesis of Bartonella spp." Clin Microbiol Rev 25(1): 42-78. 3) Pulliainen, A T. and C. Dehio (2012). "Persistence of Bartonella spp. stealth pathogens: from subclinical infections to vasoproliferative tumor formation." FEMS Microbiol Rev 36(3): 563-599. 4) Dunny, G. et al. (2008). "Multicellular behavior in bacteria: communication, cooperation, competition and cheating." Bioessays 30(4): 296-298. 5) Oliveira, A. and L. Cunha Mde (2010). "Comparison of methods for the detection of biofilm production in coagulase-negative staphylococci." BMC Res Notes 3: 260. 6) Kukulski, et al. (2012). "Precise, correlated fluorescence microscopy and electron tomography of lowicryl sections using fluorescent fiducial markers." Methods Cell Biol 111: 235-257. Microsc. Microanal. 21 (Suppl 3), 2015 238 Figure 1: B. henselae immuno-reactivity associated with micro-vasculature in dermal layer of non-lesional skin of patient with PCR blood-positive for B. henselae, B. vinsonii, and B. spp. Fourmm punch biopsy from nonlesional calf skin was processed, anti-collagen type IV plus anti-goat secondary antibody conjugated to Cy3 (aqua) and anti-B. henselae made in mouse plus secondary antibody conjugated to Cy5. . 5 microns Figure 2: B.henselae intimately associated with collagen fibrils of dermis in non-lesional skin of patient with PCR blood-positive for B. henselae, B. vinsonii, and B. spp. Four-mm punch biopsy from non-lesional calf skin was processed, labeled with anti-B. henselae made in mouse plus secondary antibody conjugated to Cy5. Imaged with two-photon laser scanning microscopy, 60X oil objective and 5Z electronic zoom. Second harmonic generation signal produces robust image of non-stained collagen type I and II (magenta) and two-photon epifluorescent signal of Cy5 is also captured. (Zeiss multi-photon scope images courtesy of Lens Cezar, PhD, UNICAMP, Campinas, Brazil.) a b c 100 microns Figure 3. PIC line from patient with persistent Bartonella harbors Bartonella spp. biofilm. a: B. henselae-ir of pic line. b: Film-Tracer Sypro-Ruby stains PIC biofilm, c: SEM of same PIC line reveals meshwork of biofilm. a b Figure 4: B. henselae immunoreactivity of bacteria stain in B. henselae cultured cells. B. henselae was cultured for 7D, lightly fixed, and stained with combinations of anti-B. henselae made in mouse and anti-mouse conjugated to Cy5. All samples imaged using confocal microscopy with PlanApo 60x (oil, NA 1.42). Scale bars = 10 microns. a) Immunoreactive B. henselae made in mouse, anti-mouse conjugated to Cy5 and b) differential interference contrast .");sQ1[120]=new Array("../7337/0239.pdf","Low Angle Annular Dark Field Scanning Transmission Electron Microscopy is Sensitive to Oxidation State in CeO2 Nanoparticles","","239 doi:10.1017/S1431927615001993 Paper No. 0120 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Low Angle Annular Dark Field Scanning Transmission Electron Microscopy is Sensitive to Oxidation State in CeO2 Nanoparticles Aaron C. Johnston-Peck1, Jonathan P. Winterstein2, Alan D. Roberts4, Joseph S. DuChene4, Kun Qian4, Brendan C. Sweeny4, Wei David Wei4, Renu Sharma2, Eric A. Stach3, Andrew A. Herzing1 1. Material Measurement Laboratory, National Institute of Standards and Technology, Gaithersburg, MD, 20899, USA 2. Center for Nanoscale Science and Technology, National Institute of Standards Technology, Gaithersburg, MD 20899 USA 3. Center for Functional Nanomaterials, Brookhaven National Laboratory, Upton, NY 11793 USA 4. Department of Chemistry and Center for Nanostructured Electronic Materials, University of Florida, Gainesville, FL 32611 USA Ceria (CeO2), a rare-earth oxide, can easily capture and release oxygen. This favorable oxygen storage capacity has made it attractive for many applications including catalysis [1]. To study the oxidation and reduction process with high spatial resolution is nontrivial. Scanning transmission electron microscopy (STEM) coupled with electron energy loss spectroscopy (EELS) can be used [2]. However, acquiring atomically resolved EELS data involves expensive equipment, high electron doses and relatively long acquisition times. By comparison imaging techniques require less specialized equipment, lower electron doses, and shorter acquisition times. Low angle annular dark field (LAADF) STEM image contrast is sensitive to defects. As an extension of this concept and a novel application, LAADF STEM is presented as a technique sensitive to the oxidation state of the cerium ions in CeO2 nanoparticles. Areas with reduced Ce ions appear brighter by comparison to regions without. In high angle annular dark field (HAADF) STEM this relationship is reversed and regions with reduced Ce ions are comparatively darker. This contrast mechanism was verified through EELS, in-situ measurements, and multislice image simulations. Static displacements due to the increased Ce3+ ionic radius disrupt electron channeling and reduce the intensity of the higherorder Laue zones thereby reducing the HAADF signal. The distorted lattice generates additional diffuse scattering which increases the LAADF signal. This contrast mechanism is demonstrated in Figure 1 by comparing LAADF and HAADF images. LAADF STEM facilitates high spatial resolution measurements on time scales that would be unattainable for STEM EELS. With this approach the oxidation and reduction behavior of CeO2 nanoparticles were studied. Changes to the nanoparticles were stimulated locally and globally either by using the electron beam directly or by exposing the particles to oxidizing and reducing conditions in an environmental transmission electron microscope. Contrast changes associated with the oxidization and reduction of CeO2 was observed as well as chemical expansion of the lattice (Figure 2). This change in lattice parameter can be exploited to estimate the extent of reduction. This demonstrates that LAADF STEM imaging may open new avenues for in-situ experimentation to gain insight into dynamic processes [3]. References: [1] A. Trovarelli, Cat. Rev. - Sci. Eng. 38 (1996), p.439. Microsc. Microanal. 21 (Suppl 3), 2015 240 [2] S. Turner et al, Nanoscale 3 (2011), p.3385. [3] A.C.J.P. acknowledges support of the National Research Council Postdoctoral Research Associateship Program. W.D.W., J.S.D., K.Q., and B.C.S thank NSF for support under Grant DMR1352328-CAREER and the CCI Center for Nanostructured Electronic Materials (CHE-1038015). A.D.R. acknowledges the University of Florida Howard Hughes Medical Institute Intramural Award. The research carried out at the Center for Functional Nanomaterials (Brookhaven National Laboratory) was supported by the U.S. Department of Energy, Office of Basic Energy Sciences, under Contract No.DE-AC02-98CH10886. Figure 1. Low angle, medium angle, and high angle annular dark field STEM image from the edge of a truncated octahedral CeO2 particle. An increase in intensity can be observed at the periphery of the particle in the LAADF image while the decrease in intensity is observed in the HAADF image. Figure 2. Select frames from a LAADF STEM movie of a particle edge that was reduced by the electron beam (Frames 15 and 38) upon increasing the dose rate (higher magnification). By decreasing the dose rate (lower magnification) the particle recovers (Frames 62-195).");sQ1[121]=new Array("../7337/0241.pdf","Reduction of Electron Scattering Image Blur for Atmospheric Scanning Electron Microscopy","","241 doi:10.1017/S1431927615002007 Paper No. 0121 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Reduction of Electron Scattering Image Blur for Atmospheric Scanning Electron Microscopy Yusuke Ominami1, Kenji Nakahira2, Shinsuke Kawanishi1, Sukehiro Ito1 1 2 Hitachi High-Technologies Corporation, 882, Ichige, Hitachinka-shi, Ibaraki-ken, 312-8504, Japan Hitachi Ltd. Yokohama Research Laboratory, 292 Yokohama-shi, Kanagawa, 244-0817, Japan Recently, methods for observing samples under atmospheric pressure in a scanning electron microscope (SEM) have been reported by some investigators. We proposed a novel atmospheric SEM (ASEM) technique for observing samples which are present in ambient air conditions but are separated from the membrane [1]. In our system, the environment around the sample can be kept in ambient air conditions (Fig. 1(a)). While wet materials is clearly observed without direct sample membrane contact at an optimized distance, typical atmospheric SEM image taken in atmosphere is more blurred compared to conventional SEM image taken in vacuum condition. The reason why ASEM images looks like "blurred" is because electron beam is scattered by electron scattering region shown in Fig. 1(b). In order to reduce the electron scattering effect, some methods utilizing light element gas [2] or additional vacuum pump to reduce pressure [1] (104~105 Pa) have been developed. A typical atmospheric SEM image is shown in Fig. 1(c). Brightness of point B is brighter than that of point A, although the edge of number "9" is clear. The image gives us a consideration that the profile of electron beam arriving at sample is estimated as sum of scattered and un-scattered electrons beam. As a result, the image in Fig. 1(c) seems to be blurred. Based on the consideration, we develop an image enhancement algorism for ASEM (electron scattering corrector: ES-Corrector). By using this algorism, blurring created by scattered electrons in ASEM image can be improved after detection of SEM image. Figure 2 shows SEM images of Cu mesh (Fig. 2(a)(b)) taken in atmospheric pressure. Figure 2(c) and (d) are restored images using ES-Corrector. The images show great improvements in clarity and edge sharpness than the observed images. The microstructures on Cu mesh observed in Fig. 2(c) and (d) are compatible to those in SEM images taken in vacuum Fig. 2(e) and (f). Figure 3 shows SEM images of a filter paper (Fig. 3(a)), renal glomerulus without metal staining (Fig. 2(b)), a leaf surface of the Japanese radish(Fig. 3(c)), and blood cells fixed with 1% glutaraldehyde and immune-stained with gold particles (Fig. 3(d)) taken in atmospheric pressure at room temperature. Figure 3(a)-(h) is the original and restored images. The images show great improvements in clarity and edge sharpness than the observed images. It has been shown that the ES-Corrector algorism to reduce effect of scattered electrons from ASEM image can improve image quality. References [1] Y. Ominami et al., accepted in Microscopy (2014). [2] K. Nguyen, M. Holtz, and D. Muller, Microsc. Microanal. 19 (Suppl 2) (2013) Microsc. Microanal. 21 (Suppl 3), 2015 242 (a) Vacuum Pumps SEM Inner chamber Stage Sample Vacuum Membrane (b) Primary electrons (c) A detector Vacuum membrane B Electron scattering region Air Atmosphere 6m Sample Fig. 1 (a) Schematics of our ASEM. (b)Events of primary electrons. (c)A typical ASEM image. (a) x1,000 (c) (e) 20m (b) x4,000 (d) 20m (f) 20m 4m 4m 4m Fig. 2 SEM images of Cu mesh. Images (a)(b) are taken in atmospheric pressure. Images of (c)(d) are restored images using ES-Corrector. Images of (e)(f) are taken in vacuum condition. (a) x400 (b) x1,000 (c) x1,000 (d) x6,000 40�m (e) (f) 20�m (g) 40�m (h) 3�m 40�m 20�m 40�m 3�m Fig. 3. SEM images taken at 1 atm (a)(e) a filter paper, (b)(f) rat renal glomerulus(un-stained), (c)(g) a leaf surface of Japanese radish, (d)(h) blood cells fixed with 1% glutaraldehyde and immune-stained with gold particles. (e)(f)(g)(h) are images improved using the developed ES-Corrector.");sQ1[122]=new Array("../7337/0243.pdf","In-situ ETEM Observation of Propene Epoxidation by Gold Nanoparticles Supported on Anatase TiO2","","243 doi:10.1017/S1431927615002019 Paper No. 0122 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In-situ ETEM Observation of Propene Epoxidation by Gold Nanoparticles Supported on Anatase TiO2 Tadahiro Kawasaki1,2,3, Hiroki Murase4, Kaname Yoshida1, and Takayoshi Tanji2,3 1. 2. Nanostructures Research Laboratory, Japan Fine Ceramics Center, Nagoya, Japan EcoTopia Science Institute, Nagoya University, Nagoya, Japan 3. Global Research Center for Environment and Energy based on Nanomaterials Science (GREEN), Tsukuba, Japan 4. Department of Electrical Engineering and Computer Science, Nagoya University, Nagoya It has been well known that gold exhibits catalytic activity when it is in the form of fine particles having a size of less than several nanometers in diameter and is tightly supported on specific metal oxides [1]. One of the catalytic reactions the gold can perform is selective oxidation of propene (C3H6) to propene oxide (C3H6O), which is very important process in the chemical industry. Since the propene oxide (PO) has low vapor pressure (5�104 Pa @RT), the PO keep liquid form if surrounding pressure is controlled to be more than this value. Then we might be able to observe catalytic product PO in liquid form though it is almost impossible to visualize that in gas form. In this study, we have observed in-situ catalytic propene epoxidation on the gold nanoparticles supported on anatase TiO2 by windowed-type environmental TEM, and have tried to clarify active sites of the gold catalyst by revealing product PO distribution. Gold catalyst specimens were prepared by the deposition precipitation method, mixed with rod-shaped anatase TiO2 powder. Our windowed-type E-cell [3-5] consisted of C/SiN hybrid membranes of 10 nm in thickness [3] assembled in a co-axial-type gas-flow specimen holder [5], that both we originally developed, was inserted into a conventional 200 kV TEM (H-8000; Hitachi).The reactant gas consisted of C3H6 (30%), O2 (15%), traces of H2O (0.1%), and N2 (55%; for increasing pressure) was introduced at a pressure of ~5�104 Pa. Figure 1 shows TEM images taken at an interval of 5 minutes under gas atmosphere of only C3H6 at 6� 102 Pa. There is almost no difference between two images. This means that propene gas with e-beam irradiation doesn't cause contaminations on the specimen. We have confirmed that the other gas molecules listed above also have negligible effects on the specimen surface. Fig. 2 are TEM images picked up from among the movie observed in-situ during the propene epoxidation reaction. Before starting the reaction, as shown in (a), the surface of Au and TiO2 kept to be clean even in reactant gas of O2, H2O and N2 because of no propene. In contrast, by adding propene gas to a pressure of 5�104 Pa, the product molecules started to be accumulated at the perimeter of interface between Au and TiO2 substrate, as indicated by arrow-heads in Fig. 1(b) and (c). The reaction product disappeared in vacuum because surrounding pressure became lower than its vapor pressure (Fig. 1(d)). By increasing total gas pressure to 6�104 Pa, the product molecules were accumulated thicker and covered whole surface of the gold particle, as shown in Fig. 3(b), because higher pressure gas prevent the molecules from vaporization. These results indicate that the product should be considered as the PO molecules. They also directly proved that the active site where the catalytic reaction occurs is the perimeter of Au/TiO2 interface, which is consistent with a model previously proposed [2] and experimental results in chemistry [6]. Microsc. Microanal. 21 (Suppl 3), 2015 244 References: [1] M. Haruta, Catalysis Today 36 (1997), p. 153. [2] M. Haruta et al, J. Catal. 115 (1989), p. 301. [3] T. Kawasaki et al, AMTC Letters 4 (2014), p. 74. [4] T. Kawasaki et al, Rev. Sci. Inst. 80 (2009), p. 113701. [5] T. Kawasaki et al, Microsc. Microanal. 17(2) (2011), p. 534. [6] T. Fujitani et al, Angew. Chem. Int. Ed. 50 (2011), p. 10144. (a) (b) 20nm TiO2 Au Figure 1. TEM images taken under propene gas condition. No contaminations are found on the specimen surface in both (a) initial condition and (b) after 5 minutes. (a) Au TiO2 (b) (c) (d) 5nm Figure 2. In-situ TEM images observed during propene epoxidation reaction. (a) Surrounding gas were O2, H2O and N2 (not reacted), (b) Propene gas were introduced at a total pressure of ~5�104 Pa, (c) after 30 s from (b), and (d) in vacuum after gas evacuation. (a) (b) 5nm Figure 3. TEM images taken under higher gas pressure condition. (a) No propene gas (O2, H2O and N2), and (b) propene gas was added to a total pressure of 6�104 Pa.");sQ1[123]=new Array("../7337/0245.pdf","In situ observation of degradation of electrocatalysts in humidified air atmosphere using a cold FE 60-300 kV ETEM","","245 doi:10.1017/S1431927615002020 Paper No. 0123 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In situ observation of degradation of electrocatalysts in humidified air atmosphere using a cold FE 60-300 kV ETEM Manabu Shirai1, Hiroaki Matsumoto1, Isao Nagaoki2, ToshieYaguchi2, Takahiro Shimizu 3 and Takeo Kamino 3 Application Development Department, Science System Design Division, Hitachi High-Technologies Corporation, Hitachinaka, Ibaraki 312-0033, Japan 2. Electron Microscope Systems Design 2nd Department, Science System Design Division, Hitachi HighTechnologies, Hitachinaka, Ibaraki, 312-8504, Japan 3. Japan Automobile Research Institute, 2530 Karima, Tsukuba, 305-0822 Japan Polymer electrolyte fuel cells (PEFCs) attract a great attention as new generation power devices for automobile and cogeneration systems. Yet, the durability of electrocatalysts is one of the most critical issues that are delaying the commercialization of this technology. To improve the durability PEFCs, we focus on studying the degradation mechanism of the electrocatalysts in their real-time evolution during operation. For this purpose, in situ observation using an environmental transmission electron microscope (ETEM) is one of the most powerful techniques. We have developed a CFE-ETEM with capabilities of simultaneous SEM with STEM image observation and applied it to the study of Pt/CB electrocatalysts degradation mechanism [1]. Previous study revealed that the Pt particles on the carbon support were aggregated and the carbon support was depressed by the Pt particles under a dry air atmosphere at 200 �C. However, the mechanism of the particles penetration into the carbon support is an open question. In the present study, we have investigated the behavior of Pt particles and related corrosion of the carbon support of electrocatalysts in a highly humidified air atmosphere. In situ simultaneous SEM/STEM study of Pt/VC (Vulcan XC-72 carbon) electrocatalysts was performed using a Hitachi HF-3300 TEM combined with a newly developed humidified air supply system [2]. Figure 1 shows an external view of the humidified air system. This system consists of a humidifier, a duct hose and a built-in thermo-hygrometer. The humidified air from the humidifier is introduced into the specimen chamber of the microscope via a gas injection nozzle of a specimen holder. In situ observation in a humidified air atmosphere was carried out with the gas injection specimen heating holder [3]. Figure 2 shows an external view (a) and a schematic illustration of the specimen heating holder (b). The heating element is a spiral-shaped tungsten wire with 50 mm in diameter and the electrocatalysts specimens are directly mounted on the wire using a small paintbrush. We observed structural change of the Pt/VC electrocatalysts at 200 �C under the humidified air. The relative humidity and temperature of the introduced air was 99 % and 40 �C, respectively. Figure 3 shows a sequence of SEM and dark-field (DF) STEM images observed at 200 �C during the humidified air injection. After 60 seconds in 2.4�10-3 Pa, the Pt particles made a movement and the agglomeration of the Pt particles have occurred on the carbon support. After 120 seconds in 2.4�10 -3 Pa, the particles actively moved and the grain size increased significantly. After 600 seconds in 2.0�10 -2 Pa, the number of Pt particles in the SEM image reduced drastically. This result suggested that the Pt particles depressed and the particles inserted into the carbon support. These experiments demonstrated that this instrument allows for in situ observation under a highly humidified air atmosphere. Furthermore, it is evident that the simultaneous dynamic observation of SEM and STEM images is indispensable to analyze structural change of nanomaterials. 1. Microsc. Microanal. 21 (Suppl 3), 2015 246 References: [1] H. Matsumoto et al., Microscopy and Analysis 27(7), 13-18 (2013) [2] T. Yaguchi et al., J. Electron Microsc. 61, 199-206 (2012) [3] T. Kamino et al., J. Electron Microsc. 54(6), 497-503, (2005) [4]The authors acknowledge to Prof. K. Sasaki and associate Prof. A. Hayashi of Kyushu University for providing the electrocatalyst samples and giving their fruitful comments. (a) (b) Figure 1. External view of a humidified air supply system developed for the HF-3300 Cold-FE TEM. Figure 2. Gas injection specimen heating holder (a) and its schematic illustration (b). Figure 3. Sequence of SEM and DF-STEM images recorded at 200 �C in a humidified air atmosphere.");sQ1[124]=new Array("../7337/0247.pdf","XEDS and EELS in the TEM at Atmospheric Pressure and High Temperature.","","247 doi:10.1017/S1431927615002032 Paper No. 0124 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 XEDS and EELS in the TEM at Atmospheric Pressure and High Temperature. Eric Prestat1, Matthew Smith1, Arne Janssen1, Thomas J. A. Slater1, Pedro H. C. Camargo2, Matthew A. Kulzick3, M. Grace Burke1, Sarah J. Haigh1 and Nestor J. Zaluzec1,4 1. 2. School of Materials, University of Manchester, Manchester, M13 9PL, UK Instituto de Qu�mica, Universidade de S�o Paulo, S�o Paulo 05508-000, Brazil 3. BP Corporate Research Center, Naperville, IL 60563, USA 4. Electron Microscopy Center, NST Division, Argonne National Laboratory, Argonne, USA. Recent progress with environmental cell and microscope design has enabled in situ imaging studies within gaseous environments inside the (scanning) transmission electron microscope ((S)TEM) to become increasingly routine [1-2]. In contrast, complementary elemental information is more challenging to obtain in situ. Electron energy loss spectroscopy (EELS) has been reported inside an environmental TEM [3,4], although only at modest pressure (below ~30 mbar). Closed-cell design specimen holders, in which the specimen and the gaseous environment are sealed from the high vacuum of the TEM by two SiN windows, allow much higher pressures to be reached (up to ~1 bar). However, these environmental cells have two major drawbacks which limit their analytical capabilities: 1) X-ray energy dispersive spectroscopy (XEDS) is challenging, as the walls of the cell generally shadow the detectors, preventing the collection of characteristic X-rays; 2) EELS is challenging as the two 50 nm thick SiN windows cause multiple scattering, which limits signal-to-background ratio in the core-loss EEL spectra and thus significantly degrades performance [5]. Recent improvement in the detection efficiency of both XEDS and EELS allows spectrum images (SIs) to be acquired at high speed and low total electron dose. In this presentation, we demonstrate the improved analytical performance of a closed-cell holder and its application to study alloying and segregation phenomena in PdAg nanoparticles during an in situ experiment. We demonstrate that the use of a XEDS compatible Protochips Atmosphere holder in a FEI Titan ChemiSTEM allows elemental mapping to be performed at high spatial resolution, ambient pressure and high temperature using XEDS. Figure 1 compares XEDS and EELS maps acquired at 820 mbar H2 and 200�C. The specimen holder was tilted to 25� and only two of the four Super-X XEDS detectors were used. The probe current was set to 600 pA to perform elemental mapping with a pixel time of 20 ms and a pixel size of 5 �. The EELS "background" relative thickness t/ is ~ 0.8, due to the two 50 nm thick SiN windows - measured with a 200 kV incident electron beam, 81 mrad EELS collection angle and ~15 um gas channel height at 1 bar H2. This makes EELS analysis increasingly problematic for specimens with thicknesses greater than ~50 nm, while XEDS is still effective for much thicker specimens. As the nanoparticles used here have diameters of ~30 nm the Pd and Ag EELS elemental maps in Figure 1 show slightly better signal-tonoise ratio than the corresponding XEDS maps. However, even for these small nanoparticles the background increases significantly on the particles themselves leading to a decrease of the signal-tobackground ratio of the Si edge (figure 1 j). Figure 2 shows XEDS maps acquired at 910 mbar 5% O2/Ar and 350�C. The PdAg nanoparticles change morphology and begin alloying at temperatures above 300�C. To study this it is essential to perform spectrum imaging, as the very similar atomic number of Pd and Ag (46 and 47, respectively) prevents identification using STEM-HAADF imaging alone. Spectroscopic imaging requires higher total electron dose than standard imaging, therefore work is in progress to control beam-induced segregation. Microsc. Microanal. 21 (Suppl 3), 2015 248 References [1] S. Giorgio et al, Ultramicroscopy 106, (2006), p. 503 [2] J. F. Creemer et al, Ultramicroscopy 108, (2008), p. 993 [3] S. Chenna and P. Crozier, Microsc. Microanal. 17 (Suppl 2), (2011), p. 476 [4] P. Crozier and S. Chenna, Ultramicroscopy 111, (2011), p. 177 [5] N. J. Zaluzec et al, Microsc. Microanal. 20, (2014), p. 323 [6] Research supported by EPSRC Grants #EP/G035954/1 and EP/J021172/1, the DTR Agency Grant HDTRA1-12-1-003, the BP Innovation Fund and ICAM project at Manchester U.S. DoE, Office of Science, Contract No. DE-AC02-06CH11357 at Argonne National Laboratory Figure 1. a) Representative STEM-HAADF image showing PdAg nanoparticles in a 820 mbar H2 at 200�C. b) Pd, c) Ag and d) Pd/Ag composite maps obtained using XEDS. e) STEM-HAADF image acquired simultaneously than the SIs. f) Pd, g) Ag and h) Pd/Ag composite maps obtained using EELS. i) XED and j) EEL spectra of the positions marked in e). All spectra are integrated over 25 pixels. Figure 2. a) Representative STEM-HAADF image of the PdAg nanoparticles at 910 mbar and 350�C be) Spectrum imaging of elemental distribution b) STEM-HAADF c) Pd, d) Ag and e) Pd/Ag XEDS composite maps.");sQ1[125]=new Array("../7337/0249.pdf","A Controllable Environmental-Cell (EC) for Wet Environmental TEM (WETEM)","","249 doi:10.1017/S1431927615002044 Paper No. 0125 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A Controllable Environmental-Cell (EC) for Wet Environmental TEM (WETEM) Ryosuke Kagawa1, Yuma Kuwamura1, Wen-An Chiou2 and Hiroki Minoda1 1. 2. Department of Applied Physics, Tokyo University of Agriculture and Technology, Nakacho, Koganei,Tokyo 184-8588, Japan NISP Lab, NanoCenter, University of Maryland, College Park, MD 20742-2831, USA Water is an essential constituent of all biological materials as well as many non-biological materials. Not only the removal of water may result in undesirable morphological and structure change, the inability to sustain the hydrated conditions in the microscope also prevents the study of reactions which take place in aqueous environment. Much attention has thus been devoted to the development of wet environmental TEM (WETEM) in recent years due to the unique capability of imaging/investigating wet materials and resolving real-time dynamic environmental processes at micro- and nano-scales. Although a simple and cost-effectively constructed wet environmental-cell (EC) was able to examine materials in hydrated condition [1], a steady state of pressure in the EC cannot be maintained throughout the investigation. This paper presents a newly developed closed EC system in a TEM that allows for dynamic observation by controlling humidity and air circulating during the experiment. To ensure the EC maintains controllable moisture/humidity, a two-line sample holder for JEOL 2010 TEM was fabricated based on Fukami's wet EC holder which was used in JEOL 2000EX TEM [2]. The new EC system includes a small EC 0.4 mm in height and 2mm in diameter, and has a gas flow controlling system to control the flow rate of air/gas and the humidity of the EC. To ensure the high vacuum of TEM column and to prevent a rise in pressure (i.e., due to the leakage of the flowing gas in the column and eventually to the gun chamber during in-situ observation), the vacuum system was enhanced by adding: (1) a pumping system for controlling gas flow ((1) in Fig. 1); (2) a turbo molecular pump for additional column pumping ((2) in Fig. 1); and (3) a sputter ion pump to enhance gun chamber vacuum and evacuation independently ((3) in Fig. 1). This new gas flow system provides precise control of the humidity or amount of water in the EC, i.e., around the specimen. Smectite, one kind of clay minerals that is expandable in water, collected from different localities was purified to remove organic matter and other impurities from the original soil. Smectite particles were dispersed in deionized water and/or ethanol. A tiny drop of suspension was pipetted onto the EC for in-situ study. TEM micrograph reveals typical smectite aggregates and well-defined particle outlines when the sample was in dry state (Fig.2a). The blurred images and the diffused SAD patterns of smectite particles from WETEM observations indicate electron scattering by water molecules in the EC and thus in the specimen (Fig. 2b). While the morphological change (image) may not reflect the humidity change of specimen so precisely, electron diffraction patterns clearly reveal the d-spacing change when the concentration of in-take water vapor changed. A series of SAD patterns were taken at different stages while changing humidity in the EC (i.e., water content in the sample) (Fig. 3). These SAD patterns were obtained from the specimens during evacuation of the wet EC. Analysis of SAD patterns obtained from fully hydrated wet state (Fig. 3b) to dehydrated state (Fig. 3d) illustrates lattice contraction of (h k 0). The d-spacing reduction vs. time of evacuation plots clearly show that the d-spacing of a and b axes seem to shrink in the same trend as that of in c axis (Fig. 3e). The new EC vacuum system provides controllable and innovative EC environment for dynamic research. [1] Y. Kuwamura et.al., Microscopy and Microanalysis 19 supplement 2 (2013) p. 490-491. [2] A. Fukami et al., Proc. 45th EMSA, Baltimore, (1987), p. 142-143. Microsc. Microanal. 21 (Suppl 3), 2015 250 [3] Conventional TEM work performed at NISP Lab was partially supported by NSF-MRSEC (DMR 05-20471) and UMD. WETEM was carried out at TUAT. Fig. 1. Schematic diagrams (a and b) and photograph (c) show the new pressure control system and the new EC TEM holder (JEM2010). Vacuum pumps/controls - were added in the new system. Fig. 2. TEM micrographs and corresponding SAD patterns of smectite clay in dry state (a and a*) and in hydrated environment (b and b*). Note the blurred image and diffused SAD pattern of wet clay. Fig. 3. In-situ WETEM observation shows the change of d-spacing of the same clay particle (a) in different humidity during evacuation of the EC (i.e., dehydration, from (b) to (d)). The relative length (d-spacing) of crystal axes vs. evacuation time of the EC is plotted (e). Through the time the d-spacing change of a and b axes is coincided with that of c axis. The reduction of d-spacing was resulted from the loss of water molecular between tetrahedron and octahedron layers of clay.");sQ1[126]=new Array("../7337/0251.pdf","Exploring the Carbon Deposition Mechanism on Ni/Gd Ceria Catalysts","","251 doi:10.1017/S1431927615002056 Paper No. 0126 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Exploring the Carbon Deposition Mechanism on Ni/Gd Ceria Catalysts Ethan Lawrence and Peter Crozier School for the Engineering of Matter, Transport and Energy, Arizona State University, Tempe Arizona 85287-6106 Solid oxide fuel cells (SOFCs) can be operated using carbonaceous fuels by adding an internal reforming layer to the anode. Developing intermediate temperature devices (500�C - 750�C) would solve many issues that currently limit SOFC applications [1]. Ni/Gd co-doped ceria anode materials have shown promise for operating at intermediate temperatures and maintaining good catalytic performance [2]. Ni serves as the fuel reforming catalyst but may also result in carbon deposition onto the Ni metal surface [3]. Carbon deposition becomes a major problem as carbon can coat the surface and render the Ni inactive. Carbon can also dissolve into the bulk Ni metal and cause stress which may lead to fractures and failure [3]. Ceria has been shown to oxidize carbon on its surface through release of oxygen atoms from the lattice [2]. Therefore, Ni/Gd co-doped ceria may inhibit carbon deposition while maintaining good reforming activity. To investigate this process, we have performed reforming tests over a model Ni/Gd doped ceria catalyst used to reform methane. Ex situ and in situ environmental transmission electron microscopy (ETEM) is employed to investigate the atomic level processes that result in carbon formation. A 15wt% Gd, 8wt% Ni co-doped ceria reforming catalyst was prepared using a spray drying technique. During spray drying, an aqueous solution of Ce-nitrate, Gd-nitrate, and Ni-nitrate was sprayed into a reaction chamber where it comes in contact with a stream of air heated to approximately 350�C. The water evaporates rapidly producing a nanoparticle powder. The powder is then calcined at 500�C and 700�C for 4 hours to produce the reforming catalyst. XRD was performed on the fresh catalyst to confirm that GDC and NiO phases were present with average phase sizes of 17 nm and 25 nm, respectively. These particles were then loaded into an ISRI RIG-150 microreactor system through which CH4 and O2 gases were flowed while the temperature was ramped up to 900�C. The Varian GC-450 was used to analyze the exit gases. The NiO reduced to Ni around 730�C and activated the catalyst for partial oxidation of methane (POM). The exit gases were analyzed using a Varian GC-450 gas chromatography system. In order to study carbon deposition, temperature was ramped up to 800�C to ensure the catalyst was activated and then held at 675�C. To induce the methane decomposition reaction, the O2 flow rate was reduced. A high carbon loading sample and low carbon loading sample were created. The initial CH4:O2 flow rate was 32:16 cc/min for both samples. For the high carbon loading sample, the O2 flow rate was reduced in a stepwise manner over 3 hours to 2 cc/min. The low carbon loading sample had the O2 flow rate decreased directly to 4 cc/min and held for a half hour. Imaging and energy dispersive x-ray spectroscopy (EDS) were performed using a JEOL 2010F scanning transmission electron microscope (STEM) to investigate the morphology and structure of the material, as well as the location of carbon deposition on the catalyst. Typical fresh catalyst particles (Figure 1) show a GDC region with a NiO particle sitting on the surface. During the POM reaction, these NiO particles are reduced to metallic Ni which then acts as the reforming catalyst. In the low carbon loading sample, graphite formation (lattice spacing of 3.35�) was seen on Ni particles that were separated from the particle aggregates as seen in Figure 2. In areas where Ni particles were in intimate contact with GDC, small to no carbon formation was seen as in Figure 3. This suggests that the proximity of the Ni particles to the Microsc. Microanal. 21 (Suppl 3), 2015 252 GDC regions influences the carbon formation. In the high carbon loading sample, carbon was seen in the form of graphite and carbon nanotubes growing on Ni and GDC particles shown in Figure 4. In order to further study carbon deposition, in-situ TEM experiments will be performed under reactive gas conditions; results will be presented. Cermets made from these Ni/Gd doped ceria catalysts will also be studied. In situ ETEM will be performed to directly observe the structure of the cermet and the spatial distribution of carbon deposition under oxygen deficient and oxygen rich conditions [4]. References: [1] Brett, D. J. L. et al, Chemical Society reviews 37 (2008), p. 1568�78. [2] Zhou, Y. et al, Physical Chemistry Letters 9 (2010), p. 1447. [3] Atkinson, A. et al, Nature Materials 3 (2004), p. 17�27. [4] We gratefully acknowledge support of NSF grant DMR-1308085 and ASU's John M. Cowley Center for High Resolution Electron Microscopy. Figure 1. TEM image of fresh catalyst showing NiO on GDC particles. Figure 2. TEM image of low carbon loading catalyst where graphite has formed on a Ni particle. Figure 3. TEM image of low carbon loading catalyst showing Ni particles in close proximity to GDC regions. Figure 4. TEM image of high carbon loading catalyst where graphite and carbon nanotubes have formed.");sQ1[127]=new Array("../7337/0253.pdf","In-situ TEM Study of the Initial Oxidation Behavior of Zry-4","","253 doi:10.1017/S1431927615002068 Paper No. 0127 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In-situ TEM Study of the Initial Oxidation Behavior of Zry-4 Wayne Harlow1 and Mitra L. Taheri1 1 Drexel University Department of Materials Science and Engineering This research investigates the early stages of oxidation in Zircaloy-4 (Zry-4) using a Protochips environmental cell in a transmission electron microscope (TEM). The oxidation behavior was studied using bright field TEM to monitor the sample in an oxygen rich elevated temperature environment. Precession diffraction was used to identify the phases present in the sample before and after the in situ experiment. Following oxidation, the samples were then cross-sectioned to determine oxidation depth into the base metal. Zircaloy-4 is a zirconium-based alloy often used as nuclear fuel rod cladding because of its good corrosion resistance and low neutron cross-section[1,2]. However, despite Zry-4s good corrosion resistance, corrosion is still a limiting factor in fuel rod lifespan. This limits the burn up which can be achieved, especially as harsher environments, which increase corrosion, are being used to help increase burn up and decrease waste. In addition, for some proposed Generation IV reactors, which place high demands on cladding materials due to extreme conditions, Zirconium-based alloys are under consideration[3�5]. Extensive past work using ex-situ autoclave and reactor corroded samples has resulted in the long term corrosion behavior of Zry-4 being well characterized[4,6,7]. Thus, the longterm oxidation behavior is well understood, but the knowledge of the initial corrosion behavior is still lacking. The initial oxidation response is important, as the effects of microstructural features such as grain boundaries can be significant, and this knowledge will allow for better alloy design to resist corrosion. Using a ProtochipsTM in-situ gas cell holder, in which gas pressure and gas mixture as well as temperature can be controlled, we have oxidized FIB prepared samples of Zry-4 to study the initial oxidation response of this alloy. Focused Ion Beam (FIB) samples were prepared from the bulk Zry-4 sample to contain a random assortment of grain boundaries. These samples were placed onto Protochips E-Chips for use in the gas cell holder. As shown in figure 1, precession diffraction was used to characterize both the phases present in the pristine sample, as well as the boundary orientations. After characterization of the pristine sample, oxidation experiments are conducted at elevated temperature in an oxygen rich gas environment while the sample is observed using conventional TEM. Following oxidation, precession diffraction is again used to characterize the phases and orientations present in the sample. Finally, the sample is removed from the E-Chip and cross-sectioned to determine the depth of oxidation and study the oxide microstructure formed as seen in figure 2. In summary, in-situ TEM was used to study the initial corrosion behavior of Zry-4. We have observed the oxidation process initiating, and have cross-sectioned these samples to study the initial oxide structure. Precession diffraction shows that both tetragonal and monoclinic ZrO2 are present in the oxidized samples, although when cross-sectioned the sample show less tetragonal ZrO2 than the plan view imaging. The oxide formed has many small grains, as is expected for the oxide in this system[8]. Microsc. Microanal. 21 (Suppl 3), 2015 254 References: [1] C.L. Whitmarsh, Review of Zircaloy-2 and Zircaloy-4 Properties Relevant to N.S. Savannah Reactor Design, 1962. [2] B. Cox, J. Nucl. Mater. 336 (2005), p. 331. [3] DOE-NERAC, A Technology Roadmap for Generation IV Nuclear Energy Systems, 2002. [4] A.T. Motta et al, J. Nucl. Mater. 371 (2007), p. 61. [5] E.A. Marquis et al, Mater. Today 12 (2009), p. 30. [6] A. Yilmazbayhan et al, J. Nucl. Mater. 349 (2006), p. 265. [7] D. Fruchart, P. Convert, G. Lelievre, 347 (2002), p 288. [8] This research is funded by the Department of Energy's Nuclear Energy University Program (NEUP). The authors would like to thank Arthur Motta of Penn State University for providing the Zircaloy samples. Figure 1. Initial sample condition prior to oxidation. TEM image on left shows area from which precession diffraction was used to create a phase map (center) and orientation map (right). Figure 2. Cross section of TEM sample after oxidation. On left, TEM image showing the precession diffraction area, phase map in center showing sample oxidation, and orientation map on right showing many small oxide grains.");sQ1[128]=new Array("../7337/0255.pdf","Transmission Electron Microscopy of Crystallization of Lysozyme in a Solution","","255 doi:10.1017/S143192761500207X Paper No. 0128 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Transmission Electron Microscopy of Crystallization of Lysozyme in a Solution Tomoya Yamazaki1, Manabu Shirai2, Hiroaki Matsumoto2 and Yuki Kimura1 1. 2. Institute of Low Temperature Science, Hokkaido University, Sapporo, Japan Hitachi High-Technologies Corporation, Hitachinaka, Japan It is important to observe the crystallization process of proteins as the fundamental study of the material science and for the structure analysis of a protein molecule by X-ray diffraction pattern using a protein crystal. The useful method to study the crystallization process is in-situ microscope observation because it can directly visualize the real process. The observation of protein crystallization have been performed mainly using optical microscopy [1,2] and the micro-scale view of the process have been well understood. However, the protein crystallization process at nano-scale is still unclear because there is no in-situ observation of it. Recently, in-situ observations of the behavior of nano-particles and the crystallization processes of inorganic materials have been energetically performed by the transmission electron microscopy (TEM) combined with the liquid cell or an ionic liquid, and these mechanisms at nano-scale are partly demonstrated [3-5]. However, there are no observations of crystallization of proteins so far. We performed in-situ observation of the protein crystal for understanding its crystallization process using TEM with the liquid cell. We used the hen-egg white lysozyme as a protein sample. The lysozyme was crystallized using NaCl as a precipitant in a sodium acetate buffer solution at pH = 4.5. For the observation, we used a "Poseidon" liquid cell holder (Protochips Inc.) with two input and one output ports. The liquid cell consists of a pair of semiconductor-based plates with an amorphous silicon nitride window and 150 or 500-nm-thick spacer to form the flow path of the crystallization solution. We used two TEMs with LaB6 filament at an acceleration voltage of 200 kV (Hitachi H-8100) and with field-emission gun at an acceleration voltage of 300 kV (Hitachi HF-3300). We succeeded in observing the two crystalline phases of orthorhombic and tetragonal in addition to an amorphous phase of the lysozyme protein. Individual amorphous particles could not observe under optical microscopy because the size of the amorphous particles is only about 100-200 nm (Fig. 1), which is smaller than the spatial resolution of optical microscope. The growth rates of orthorhombic crystals and tetragonal crystals under TEM are consistent with those measured by optical microscopy under similar condition. Therefore, the electron-beam irradiation on the crystallization of lysozyme under TEM can be negligible in our experimental condition. We observed an orthorhombic crystal and amorphous particles in the same view. The orthorhombic lysozyme crystal grew continuously, whereas amorphous particles dissolved slowly. It clearly shows that the solubility of an orthorhombic crystal is lower than that of amorphous particles, i.e., the orthorhombic lysozyme crystal is more stable than the amorphous particles of lysozyme. As a conclusion, we succeeded in observing the nano-scale view of the protein crystallization using the TEMs and the liquid cell holder as a first step for understanding of the protein crystallization process [6]. Microsc. Microanal. 21 (Suppl 3), 2015 256 References [1] P. Dold et al., J. Cryst. Growth 293 (2006), 102. [2] A. E. S. Van Driessche et al., Cryst. Growth & Des. 7 (2007), 1980. [3] J. M. Yuk et al., Science 336 (2012), 61. [4] Y. Kimura et al., J. Am. Chem. Soc. 136 (2014), 1762. [5] M. H. Nielsen et al., Science 345 (2014), 1158. [6] The authors acknowledge supports from a Grant-in-Aid for Research Activity Start-up from KAKENHI (26887001) and for a grant for Young Scientists (A) from KAKENHI (24684033). Figure 1. A bright-field TEM image of a cluster of lysozyme amorphous particles in a solution. This image was recorded by a Hitachi HF-3300 with an acceleration voltage of 300 kV. The size of an individual particle is about 100-200 nm.");sQ1[129]=new Array("../7337/0257.pdf","Effect of Electron Beam on Nanoparticle Dynamics in Solution during in situ TEM Observation","","257 doi:10.1017/S1431927615002081 Paper No. 0129 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Effect of Electron Beam on Nanoparticle Dynamics in Solution during in situ TEM Observation Jingyu Lu1,2,3,4, Zainul Aabdin1,2,3,4, Utkarsh Anand3, and Utkur Mirsaidov1,2,3,4 1. 2. Department of Physics, National University of Singapore, 2 Science Drive 3, Singapore, 117551. Centre for Advanced 2D Materials and Graphene Research Centre, National University of Singapore, 6 Science Drive 2, Singapore 117546. 3. Centre for BioImaging Sciences, Department of Biological Sciences, National University of Singapore, Science Drive 4, Singapore, 117543. 4. Nanocore, 4 Engineering Drive 3, National University of Singapore, Singapore 117576. The recent development in in situ TEM enabled to visualize the nucleation, growth, coalescence and shape evolution of nanoparticles [1-4]. Some of these processes are driven by electron beam effects [5]. In this study we found that under intense electron beam nanoparticles aggregate together and form a large cluster that relaxes into a large spherical nanoparticle. While under low electron beam different attachment of nanoparticles results in nanorod structures. The observation was performed with the TEM JEOL 2010F operated at 200 kV, and dynamics were captured at 10 frames per second. The sample was made of 1 mM HAuCl4 and CTAB solution entrapped between two ~20 nm thick SiNx membranes in a liquid cell, more details can be found elsewhere[6]. Attachment of nanoparticles, observed in these liquid cells with TEM for low electron beam condition, is a common way to form nanorod structures as shown in Figure 1. When electron dose is less than 200 e/(A2.s), only small nanoparticles (radius<3 nm) exhibit evident motion (Figure 1(a)(b)). As the dose rate increases, larger nanoparticles start moving (Figure 1(c)-(f)). Note that the nanoparticles encircled in red ellipses in each frame merged together. Two large elongated nanoparticles, formed from individual particles, attach at their endpoints and new longer rod-like nanoparticle is formed. Once these rod-like structures were formed, the attachment of new nanoparticles takes place at the rod tips, thus the nanorod grows even longer, as shown in Figure 1(d)-(f). Similar phenomena are found in the synthesis of Pt3Fe nanorods [7] as well. At high electron doses, the consecutive attachment of 5 nanoparticles forms a big spherical nanoparticle (~11 nm in diameter) instead of a nanorods structure as shown in Figure 2. In each attachment case here, the smaller nanoparticle is fused into the larger nanoparticle; our observations reveal that the larger the two attaching nanoparticles, the longer it takes to form a spherical nanoparticle after the coalescence, (Figure 2(a)-(c) vs Figure 2(f)-(g)). Figure 2(i) shows the evolution of projected area of these nanoparticles. The projected area of each new nanoparticle formed by the attachment of two nanoparticles decreases steadily as they tend to become spherical in shape. On the whole, the total projected area was decreased by ~43% at the end of the process, indicating that the surface energy minimization contributes greatly to this shape relaxation [3]. Further study is on the way to uncover more details. Microsc. Microanal. 21 (Suppl 3), 2015 258 References: [1] H.-G Liao, D Zherebetskyy, H Xin, et al. Science, 345(2014): p. 916-919. [2] M H Nielsen, S Aloni, J De Yoreo. Science, 345(2014): p. 1158-1162. [3] Z Aabdin, J Lu, X Zhu, et al. Nano Letters, 14(2014): p. 6639-6643. [4] M J Williamson, R M. Tromp, P M Vereecken, et al, Nature Materials 2(2003): p. 532 � 536 [5] N de Jonge, and F M. Ross. Nature Nanotechnology, 6(2011): p. 695�704 [6] J Lu, Z Aabdin, L Duane, et al. Nano Letters, 14(2014): p. 2111-2115. [7] H.-G Liao, L Cui,S Whitelam, H Zheng. Science, 336(2012): p. 1011-1014. Figure 1. Formation of nanorod by consecutive attachment of nanoparticles. Dose rate in (a) is ~60 e/A2/s, and ~1480 e/A2/s for (b), and ~4000 e/A2/s for (c)-(f). The nanoparticles encircled in red ellipses in each frame merged into a single nanoparticle in the next figure. Figure 2. (a)-(h) Formation of a single spherical nanoparticle by consecutive attachment of small nanoparticles. Dose rate: ~16000 e/A2/s. (i) the evolution of projected area of these nanoparticles ("NP" is short for "nanoparticle"), and (j) the corresponding labelled number of nanoparticles.");sQ1[130]=new Array("../7337/0259.pdf","Role of Fluid-Mediated Interactions in Guiding Nanoparticle Assembly","","259 doi:10.1017/S1431927615002093 Paper No. 0130 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Role of Fluid-Mediated Interactions in Guiding Nanoparticle Assembly Utkur. Mirsaidov1,2,3* Guanhua Lin1,2,3, Duane Loh1,2,, Jingyu Lu1,2,3, Zainul Aabdin1,2,3 1. Centre for Advanced 2D Materials and Graphene Research Centre, Department of Physics , National University of Singapore, 6 Science Drive 2, Singapore 117546. 2. Center for BioImaging Sciences, Department of Biological Sciences, National University of Singapore, Singapore. 3. Nanocore, 4 Engineering Drive 3, National University of Singapore, Singapore 117576, The assembly process of nanostructures from nanoparticles in solution is fundamental for "bottom-up" fabrication of functional materials and devices. The collective behaviour of these nanoparticle assemblies can give rise to new optoelectronic (1), electrochemical (2) and magnetic properties (3) different from the bulk or individual nanoparticles. Using dynamic in situ TEM imaging (3-6) in liquids, we show the nanoparticle-nanoparticle interaction in thin fluid layer. We extend on our recent study (7) of nanoparticle bonding to show the effect of attractive hydration and depletion forces (in the case of small molecules in solution) which arise due to a single to double layers of water molecules separating these interacting nanoparticles. Using a statistical method, we probe the strength of both short-range and long-range inter-particle interactions and empirically derive the nanoparticle-nanoparticle interaction potential. Furthermore, we will show that interaction of nanoparticles with liquid-liquid interfaces in the case of phase separated nandroplets in solution can be used for self-assembly. Our observation of the assembly dynamics of nanoparticles at fluid-fluid interface around the dispersed fluid-like ethylenediaminetetraacetic acid (EDTA) nanodroplets in solution reveal that particles bind to interface irreversible until the interface is fully populated. Our findings reveal that nanoscale capillary force is the main driver of nanoparticle arrangement and organization at fluid-fluid interfaces. We find that nanoparticles assemble into rings around dispersed nanodroplets in solution through: (i) direct attachment to the fluid-fluid interface, (ii) insertion of new nanoparticles between interface-bound nanoparticles, and (iii) coalescence of assemblies (Fig. 1). Our findings show that the effect of hydration and capillary forces on assembly is one of many interactions that need to be understood for assembly of hierarchical nanostructures from nanoparticles serving as individual building blocks. References: [1] K. L. Kelly, E. Coronado, L. L. Zhao, G. C. Schatz, The Journal of Physical Chemistry B 107 (2002), p. 668. [2] Y. Yamada et al., Nature Chemistry 3 (2011), p. 372. [3] G. Singh et al., Self-assembly of magnetite nanocubes into helical superstructures. Science 345 (2014), p. 1149. [4] M. J. Williamson, R. M. Tromp, P. M. Vereecken, R. Hull, F. M. Ross, Nature Materials 2 (2003), p. 532. [5] H. Zheng, R. Smith, Y. Jun, C. Kisielowski, U. Dahmen, A. P. Alavisatos, Science 324 (2009), p. 1309. [6] U. Mirsaidov, H. Zheng, D. Bhattacharya, Y. Casana, P. Matsudaira, Proc. Natl. Acad. Sci. U.S.A. 109 (2012), p. 7187. [7] Z. Aabdin, J. Lu, X. Zhu, U. Anand, D. Loh, H. Su, U. Mirsaidov, Nano Letters 14 (2014), p. 6639. Microsc. Microanal. 21 (Suppl 3), 2015 260 [8] This work was supported by the Singapore National Research Foundation's Competitive research program funding (NRF-CRP9-2011-04). Figure 1. Graphene nanochannel platform. (A) Schematic illustration of binding dynamics of Pt nanoparticles to a nanodroplet by direct attachment (yellow arrows) to the circumference and by insertion (red arrow) between two closely spaced nanoparticles that are bound to nanodroplet circumference (blue arrow) in solution. (B) TEM image time series of binding dynamics of Pt nanoparticles to a nanodroplet circumference by direct attachment in solution. (C) TEM image and corresponding segmentation of Pt nanoparticles (yellow) assembled around the EDTA nanodroplet (blue) in water.");sQ1[131]=new Array("../7337/0261.pdf","The Two Dimensional Nanoplate Dynamics Revealed by in situ Liquid Cell TEM","","261 doi:10.1017/S143192761500210X Paper No. 0131 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Two Dimensional Nanoplate Dynamics Revealed by in situ Liquid Cell TEM Xuezeng Tian1, Jingyu Lu2, Zainul Aabdin2, Utkarsh Anand2, Utkur M. Mirsaidov2 and Haimei Zheng1,3 1. 2. Berkeley Education Alliance for Research in Singapore, 138602, Singapore Center for BioImaging Sciences, Department of Biological Sciences, National University of Singapore, Science Drive 4, Singapore 117543 3. Materials Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, CA 94720 The interface between solid and liquid is an intriguing topic in many fields of studies. However, such a heterogeneous interface is not easy to probe using conventional methods, especially when the length at nanometre length scales. Recent progress in in situ TEM enables the study of physical and chemical processes at interfaces. A liquid cell usually consists of a thin liquid layer sandwiched between two electron transparent silicon nitride membranes. This allows the real time observation of interfacial phenomena with sub-nanometre resolutions. For example, the diffusivity of gold nanoparticles is damped due to interaction with the membrane [1][2]; the movement of a few nanometres water droplet is greatly influenced by the interfacial forces[3]. In this work, we use liquid cell in TEM to study conventional materials with unexpected new features. We'll show peculiar dynamics of AgCl nanoplates, which result from the interplay of surface tension and Coulomb repulsive force, whereas we use electron beam to control the balance between these two competing phenomenon. The AgCl nanoplates were made by dissolving silver nanoparticles in KCl solution. Electron beam irradiation accelerates this dissolving process. All the AgCl in our experiments take a rock salt structure with lattice constant of 5.55�. We find that the AgCl particles prefer cubic shapes, with all the facets in {002} planes. The liquid we use in the experiment is 25mM KCl solution. To our best knowledge, when AgCl is immersed in KCl solution, the surface will preferentially be adsorbed by chloride ion due to the strong bonding of silver ions to chloride ions[4]. Thus the first adsorption layer of AgCl is dominated by negatively charged chloride ions whereas outer adsorption layers being balanced by positive potassium ions. Such an adsorption configuration results in both vertical and lateral potential differences. The lateral potential causes the Coulomb repulsive force which tends to expand the nanoparticles. However, the surface tension tends to contract the particles to lower its internal energy. There is a balance between the surface tension and Coulomb repulsion. It's interesting to note that electron dose could mediate this balance and lead to completely different dynamics. Under low electron dose, the surface tension of AgCl nanocubes will be dominating. Thus the nanocubes are relaxed into spherical shape. Increasing the dose will change the adsorption equilibrium, which leads to larger Coulomb repulsion which leads to expansion of nanocubes into nanoplates. Further increasing the dose causes the balance between surface tension and Coulomb repulsion will be broken. Consecutively, AgCl nanoplates split up into smaller pieces. Fig. 1 shows a typical splitting event. During this splitting event, sub-splitting always happens due to thinner nanoplate thickness. It's worth noting that all the splitting events happen at the corners of the nanoplates. Combining real-time observation and Finite Element Method (FEM) simulation, we were able to quantify the Coulomb repulsive forces on the nanoplates, as shown in Fig. 2. Here, the electric field in the middle of the nanoplate is the smallest, and largest at the edges and corners. The magnitude of the Coulomb repulsive forces is also largest at the edges and corners, as shown by the arrows. This simulation offers a potential reason for the preferential splitting of the nanoplate at the corners. We further observed liquid-like features on the crystalline AgCl nanoplates. Microsc. Microanal. 21 (Suppl 3), 2015 262 As shown in Fig. 3, a single crystalline AgCl nanoplate shrinks due to surface tension. The nanoplate maintains crystallinity while simultaneously behaves like a viscous liquid. We further estimated its selfdiffusivity by studying this surface tension induced shrinking, which is much larger than normal solids. This work indicates that both the physical and chemical properties of nanosize materials are influenced by the interfacial confinement. Such a platform is ideal in exploring various interfacial phenomenon. References: [1] H. Zheng et al, Science 324 (2009), 1309. [2] J. Lu et al, Nano Lett. 14 (2014), 2111. [3] U. Mirsaidov et al, PNAS 109 (2012), 7187. [4] K. Temsamani and K. Cheng, Sensor Actuat B 76 (2001), 551. Figure 1. A typical nanoplate splitting process. Insets are their segmented 2bit figures, respectively. The electron dose during this process was kept 13 e/(�2�s) Figure 2. Simulated electric field (color) and Coulomb repulsive force (arrows). Figure 3. A single crystalline AgCl nanoplate shrinks due to surface tension.");sQ1[132]=new Array("../7337/0263.pdf","Dynamic studies of solution-based reactions using operando TEM","","263 doi:10.1017/S1431927615002111 Paper No. 0132 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Dynamic studies of solution-based reactions using operando TEM Kun He1, 2, Yifei Yuan3, Yu-Peng Lu1, Tolou Shokuhfar3, and Reza Shahbazian-Yassar2, 3 School of Materials Science and Engineering, Shandong University, Ji'nan, China. Department of Materials Science and Engineering, Michigan Technological University, 1400 Townsend Drive, Houghton, Michigan, United States. 3. Department of Mechanical Engineering-Engineering Mechanics, Michigan Technological University, 1400 Townsend Drive, Houghton, Michigan, United States. 1. 2. In-situ investigation of chemical reactions in liquid is attracting increasing interest in recent years, due to the unprecedented capabilities provided by the fast development of in-situ liquid-cell for scanning transmission electron microscopy ((S)TEM) technologies[1]. Using liquid-cell in (S) TEM, as shown in Figure 1.a), the solution-based structure and morphology evolutions can be studied. In this research, we use sandwiched silicon nitride liquid cell to study the effect of different electron dose rates and different liquid thicknesses on the e--beam induced nucleation and growth of copper salts in solution. In the experiment, we observe that copper salts can only be generated in the e--beam exposure area, and its pre-nucleation time decreases as the dose rate increases over a threshold value. By studying the growth rate after nucleation, the different growth mechanisms of particles at thick and thin liquid-cell are revealed. In the thick liquid area (Figure 1.b)), the growth is mainly controlled by diffusion rate compared with the surface reaction-controlled growth when the liquid area is thin (Figure 1.c)). For thick liquid, the flower shaped amorphous precursor particles can be observed and identified by the diffraction patterns (DP). According to the different contrast of amorphous and crystalline phases in liquid, some crystals can be directly viewed to form in the center or at the edge of the amorphous particles and then expand. It proves that the amorphous phase exits during all the initial stages of the crystallization process. For the thin liquid cell, however, no amorphous phase was found. But some triangle, prismoid, rectangle and other regular shaped particles are formed in thin liquid. By changing dose rate and thickness of liquid, the morphology of new generated particles can be controlled. The dose rate and exposure time are directly related to pH variation and the radical products such as hydrogen, oxygen, and hydrated electrons, which are produced by e--beam in aqueous solution[2]. Such findings demonstrate that the thickness of liquid cell largely affects the nucleation and growth of Cu salts, implying an underlying relationship between electron dose rate and phase transition that still needs to be explored. This research indicate how to use the liquid cell to study radiolysis effects on materials. Microsc. Microanal. 21 (Suppl 3), 2015 264 References: [1] X. Chen, C. Li, H. Cao, Nanoscale (2015). [2] N.M. Schneider, M.M. Norton, B.J. Mendel, J.M. Grogan, F.M. Ross, H.H. Bau, The Journal of Physical Chemistry C 118 (2014) 22373-22382. Figure 1. a) Schematic of the liquid holder in the TEM. A liquid cell is enclosed between two electron transparent windows made of silicon nitride. b-c) Different morphology of the particles formed inside the cell with different liquid thickness(b thick c thin).");sQ1[133]=new Array("../7337/0265.pdf","Precision In Situ Control of Local Liquid Chemistry via Electron Irradiation","","265 doi:10.1017/S1431927615002123 Paper No. 0133 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Precision In Situ Control of Local Liquid Chemistry via Electron Irradiation C. Wang1, T. Shokuhfar 1,2, and R.F. Klie1 Dept. of Physics, University of Illinois at Chicago, Chicago, IL, USA 60607 Dept. of Mechanical Engineering, Dept. of Biomedical Engineering, Michigan Technological University, Houghton, MI, USA 49931 2. 1. In situ liquid electron microscopy has allowed direct observations of biological [1], electrochemical [2], physical [3], or chemical processes [4] near or at atomic resolution. Previous studies have focus on reporting various sample behavior under different electron microscope conditions. In this report, we discuss the possibility of utilizing electron irradiation to precisely control the liquid chemical reactivity by controlling the electron induced radical concentration. Since space-time resolution of the real-time observation is limited by electron dose rate, for higher resolution a higher electron dose rate is always desired. This, however contradicts the need to suppress either particular, or all types of beam effects caused by electron-induced radical species under high absorbed dose rate. By taking advantage of the nonlinear effects in the electron water interaction, we were able to suppress the steady state concentration of electron beam induced species at high electron dose rate, allowing high resolution electron microscopy and spectroscopy while minimizing the damage to the sample. On the other hand, by changing the liquid cell design, we were able to suppress the formation of hydrogen gas bubble while maintaining a high free radical concentration, allowing applications taking advantage of these radicals while avoiding phase change of the liquid environment. The electron beam induced radical species are often undesired as they can cause physical or chemical change to the sample. Figure 1 shows a dose rate series experiment conducted on graphene sandwiched ferritin molecules in STEM mode. We find that ferritin protein denatures into different types of subunits under different electron dose rate while maintaining an overall intact structure below a threshold of 6 e-/�2/s. Previous studies have also shown that many other beam effects, such as formation of bubble [1] or the reduction of metal nanoparticles [4], can be prevented by maintaining the electron dose rate at a level below the reaction threshold, which, as suggested by electron water interaction simulation [5], can be explained by a low steady state concentration of electron generated radical species. However, in real time observations this low dose rate approach imposes a theoretical limit on the attainable spacetime resolution, which is inversely proportional to the square of the electron dose rate. We introduce an electron pulse imaging approach to suppress the generated radical concentration under a time averaged high dose rate. Combined with simulation and experiment, we find that by turning off the electron pulse before the generated radicals reach threshold concentration, as well as allowing enough time for the sample to relax before the subsequent pulse, it is possible to maintain a below-threshold radical concentration at a higher time-averaged electron dose rate, which improves the space-time resolution for real-time observations free of beam effects. Using electron induced hydrogen bubble formation as an indicator, we showed that by optimizing the probe current, pulse time and relaxation time combination, the time averaged threshold electron dose rate to reach the saturation condition for the dissolved hydrogen gas is significantly increased. Simulation confirmed an overall lower concentration of all radical species produced by this method, which allows atomic resolution imaging free of bubble formation and other beam effects, as we have qualitatively shown in our previous paper [1]. Microsc. Microanal. 21 (Suppl 3), 2015 266 On the other hand, electron induced radical can be beneficial when an increased chemical activity in the liquid is desired. Previous experiments [1, 4] and simulations [5] both suggested that a higher electron dose rate in electron microscopy is analogous to a higher concentration of reacting agent in conventional liquid chemistry, resulting a more reactive environment for applications such as in situ nanoparticle synthesis. Although the existence of the steady state provides a convenient way of precision control of liquid chemistry in terms of concentration of radical species via control of electron dose rate, the extend of this approach remain limited due to bubble formation under moderate dose rate. To take advantage of the increased beam induced chemical activity, bubble formation needs to be suppressed while maintaining a high overall radical concentration, which also means a high dissolved hydrogen gas concentration. We find that increasing the saturation threshold will significantly increases the bubble formation threshold dose rate, allowing an electron-induced chemically reactive liquid environment free of gas bubble condensation. References: [1] Wang, C., et al, Advanced Materials 26.21 (2014), 3410-3414. [2] Sacci, R. L., et al, Chemical Communications 50.17 (2014), 2104-2107. [3] Chen, Q., et al, Faraday discussions 175 (2014), 203-214. [4] Woehl, T. J., et al, ACS nano 6.10 (2012), 8599-8610. [5] Schneider, N. M., et al, The Journal of Physical Chemistry C 118.38 (2014) 22373-22382. [6] This research was supported by Michigan Technological University and Research Resource Center, University of Illinois at Chicago. Figure 1: ABF images of sandwiched ferritin molecules at different area averaged dose rates. The area averaged dose rate used to acquire each image is gradually increased with the value of each image being (A): 1.4 e-/�2/s (~0.25�); (B): 2.2 e-/�2/s (~0.36�) [1]; (C): 6 e-/�2/s (1�); (D): 23 e-/�2/s (~4�); (E): 91 e-/�2/s (~16�); (F): 143 e-/�2/s (~25�). The number on the lower right corner of each image is the dose rate multiple compared to the threshold dose rate in (B). The denaturation process of ferritin protein became visible at a dose rate of 6 e-/�2/s as shown in image (B). The diameters of the sub-units in Figure 5B-E are measured to be between 0.4 nm to 1.6 nm. We identified these structure as polypeptides, sub-units of the ferritin protein shell. As shown in image (F), the protein is further broken down into biological structures with a width of approximately 1.3 �, which we identified as amino acids at a dose rate of 206 e-/�2/s (~36 times of the denaturation initiating dose rate 6 e-/�2/s).");sQ1[134]=new Array("../7337/0267.pdf","Probing Nanoparticle Dynamics in 200 nm Thick Liquid Layers at Millisecond Time Resolution","","267 doi:10.1017/S1431927615002135 Paper No. 0134 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Probing Nanoparticle Dynamics in 200 nm Thick Liquid Layers at Millisecond Time Resolution See Wee Chee1,2, Duane Loh1,2, Utkur Mirsaidov1,2,3 and Paul Matsudaira1 1. Center for BioImaging Sciences, Department of Biological Sciences, National University of Singapore, Singapore 117557 2. Centre for Advanced 2D Materials and Graphene Research Centre, Department of Physics, National University of Singapore, Singapore 117546 3. Nanocore, Faculty of Engineering, National University of Singapore, Singapore 117576 When imaging specimens in liquids using liquid cell TEM, there is a trade-off between spatial resolution and liquid layer thickness [1]. In TEM mode, the imaging resolution in a liquid cell with 1 micron thick liquid degrades to about 12 nm for objects near the top SiN window, due to scattering of electrons by the liquid [1]. While higher resolution can be obtained with thin liquid films (less than 100 nm thick), the desire for sharper images has to be balanced against keeping the liquid layer thick enough such that processes observed are representative of bulk systems. For example, nanoparticle diffusion observed in thin liquid layers with liquid cell TEM had been found to deviate from the Stokes-Einstein relation [2]. Maintaining a thick liquid layer is also crucial for imaging biological specimens, where, so far, in situ imaging is mainly accomplished using STEM [3]. However, STEM imaging cannot match the temporal resolution of TEM imaging for capturing the motion of dynamic objects where state-of-the-art direct detection cameras are capable of recording whole frames in milliseconds. Here, we used the motion of Au nanoparticles as a model system to probe the achievable spatial and temporal resolution in TEM imaging of thicker liquid layers. Au nanoparticles (~20 nm in size) are dispersed in water and sandwiched between 30 nm thick SiN windows in a Hummingbird Scientific liquid flow holder. The holder is loaded into a JEOL 2200FS TEM with an Omega filter, where zero loss imaging (with 20 eV to 40 eV energy slits) was used to mitigate the resolution loss from imaging through the liquid layer (see Figure 1 for a comparison of filtered and unfiltered images). We estimate a typical liquid layer thickness of ~200 nm at the window edge and ~800 nm at the window center from electron energy loss spectroscopy (using inelastic mean free paths reported in [4]) for our experiments. Live movies were recorded on a Direct Electron DE-12 camera system at 100 frames per second. Figure 2 shows the trajectory of a single particle near the window edge exhibiting L�vy flight where the motion blur indicates that the nanoparticle made at least two jumps in the 40 ms time frame. The motion of a pair of connected nanoparticles is described in Figure 3 where it can be seen that during rotation, one of particles moves significantly more than the other. It also shows that our ability to resolve the shape of the nanoparticles is mainly limited by motion blurring. Particle tracking analysis of the motion of the nanoparticles will be presented. The study of dynamics at millisecond time resolution requires finding an optimal balance between having sufficient signal to noise and minimizing beam induced effects. The nanoparticle dynamics presented here were captured with a dose rate of more than 125 e/(�2 s) (a lower bound calculated from the beam current post-specimen), which can cause significant beam damage (higher dose rates led to dissolution of the nanoparticles). This dose rate will be untenable for imaging soft materials and biological specimens. Strategies for imaging at minimal doses and to retrieve structural information from noisy images will be discussed. Microsc. Microanal. 21 (Suppl 3), 2015 268 References: [1] N. de Jonge and F.M. Ross, Nat. Nanotechnol.6 (2011) p. 695 [2] H.M. Zheng et al., Nano Lett. 9 (2009) p. 2460, J.Y. Lu et al. Nano Lett. 14 (2014) p. 2111. [3] N. de Jonge et al., PNAS 16 (2009) p. 2159, J.E. Evans et al. Micron 43 (2012) p. 1085, Woehl et al. Sci. Rep. 4 (2014), article 6854. [4] K.L. Jungjohann et al. Microsc. Microanal. 18 (2012) p. 621. [5] The authors acknowledge funding from Singapore National Research Foundation's Competitive research program funding (NRF-CRP9-2011-04) and thank Ms. S.F. Tan and Dr. C.A. Nijhuis for their assistance in providing the nanoparticles. Figure 1. Bright field TEM images of Au nanoparticles in water imaged (left) without energy filtering and (right) with a 20 eV energy filter inserted Figure 2. 40 ms sequence of a nanoparticle near the window edge in L�vy flight (40 eV energy slit). The motion blur indicates that the nanoparticle made at least two jumps. Scale bar represents 20 nm. Figure 3. 90 ms sequence of a two nanoparticle pair rotating (20 eV energy slit). The images show that one nanoparticles moves significantly more than the other. Scale bar represents 20 nm.");sQ1[135]=new Array("../7337/0269.pdf","Bonding Pathways of Gold Nanocrystals in Solution","","269 doi:10.1017/S1431927615002147 Paper No. 0135 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Bonding Pathways of Gold Nanocrystals in Solution Zainul Aabdin1,2,3,4, Jingyu Lu1,2,3,4, Utkarsh Anand1,2,4, N. Duane Loh1,3, Utkur Mirsaidov1,2,3,4 1. 2. Department of Physics, National University of Singapore, 2 Science Drive 3, Singapore, 117551. Graphene Research Center, National University of Singapore, 6 Science Drive 2, Singapore, 117546. 3. Center for Bioimaging Sciences and Department of Biological Sciences, National University of Singapore, 14 Science Drive 4, Singapore, 117543. 4. NUSNNI-Nanocore, 5A Engineering Drive 1, National University of Singapore, Singapore 117411. The bonding between nanocrystals in solution is one of the most important crystal growth pathways to the bottom-up assembly of nanocrystals into hierarchical structures for the fabrication of nanoscale devices and nanomaterials. The main challenge, however, is to understand and avoid defect formation at the bonding interfaces between nanocrystals. It is believed that any misalignment results in defect formation at bonding interface, and defect-free coherent bonding is only possible when bonding lattice planes are perfectly oriented. Recent studies employing in situ imaging showed that the dynamics of nanocrystal bonding in solution is rich. For the first time the pre-alignment of nanocrystals prior to attachment, post-bonding dipole-induced realignment of nanocrystals, and annealing of line dislocations in nanocrystals were observed [1-3]. However, two important key questions to nanocrystal bonding remain unresolved: 1) what are the necessary conditions for nanocrystal attachment in solution that will lead to defect-free coherent bonding? and 2) what is the likelihood of such coherent bonding? Here, we resolved the mechanisms underpinning these phenomena using liquid-cells developed for time-resolved electron microscopy [4]. Our study is the first to give a detailed description of two distinct bonding pathways which shows how attachment geometry of the system affects the bonding of nanocrystals with rotational and translational degrees of freedom [5]. Figure 1A shows the coherent bonding of two nanocrystals initially separated by ~1 nm, and their (111) lattice planes misaligned by 10� (at t = 6.9 s) prior to contact. As the nanocrystals approach each other, their lattice misalignment decreases to 9� upon contact (not shown here), then visibly realigns into coherent bonding (Fig. 1A: t = 25.5 s and 48.2 s). Here, the bonding was coherent because the common (111) lattice planes reflections of the nanocrystals merged into a single reflection with no visible defects (inset t = 48.2 s). Expectedly, defect-mediated bonding occurs when two nanocrystals meet with their (111) lattice planes initially misaligned by a large angle, for example by about 32 o in Figure 1B. Here the nanocrystals do not realign upon bonding and a visible defect forms at their merging interface (t = 6.9 s, black arrow). Post bonding images and splitted Fourier reflections show that the newly formed nanoparticle is not a single crystal (Fig. 1B: t = 11.2 s). The coherent and defect-mediated bonding sequences captured in Figure 1 are remarkable because upon contact the nanocrystals can reorient themselves to a limited extent, either to achieve coherent bonding when the two nanocrystals are initially slightly misaligned, or rotate to a mutual configuration that allows defects to form. The experimental results in Figure 2A show that there is a critical misalignment angle (critical ~ 15�) that separates two pathways for bonding: below this critical angle gold nanocrystals can realign for defectfree coherent bonding, and above which the defect are formed at the interface which mediate the bonding (Fig. 2B). Our results have significant implication for nanocrystal growth, and bottom-up design of hierarchical nanostructures [6]. Microsc. Microanal. 21 (Suppl 3), 2015 270 [1] R. L. Penn and J.F. Banfield, Science, 281 (1998), p.969. [2] Li. Dongsheng et al., Science, 336 (2012), p.1014. [3] H. G. Liao et al., Science, 336 (2012), p.1011. [4] U. M. Mirsaidov et al., P. Proc. Natl. Acad. Sci. U.S.A., 109 (2012), p.7187. [5] Z. Aabdin et al., Nano Letters, 14 (2014), p.6639. [6] This work was supported by the Singapore National Research Foundation's Competitive research program funding (NRFCRP9-2011-04). The authors acknowledge the support of the Electron Microscopy Facility at Center for Bioimaging Science, National University of Singapore. Figure 1. Time resolved images of (A) coherent and (B) non-coherent bonding between two gold nanocrystals (P and Q). Coherent bonding shares a common (111) lattice planes that are misaligned by only ~ 9� at contact yields defect free nanocrystal PQ, whereas, for non-coherent bonding (111) lattice planes are misaligned by 32� at contact yields a nanocrystal PQ with defect at the bonding interface. Lattices fringes along (111) plane with 2.4� spacing are marked by dashed parallel lines with corresponding reflections circled in the Fourier transform of the image as shown in the inset Figure 2. (A) Histogram of experimental nanocrystal bonding events at different orientation angles () measured between the common (111) lattice planes at contact as shown in schematic diagram (B). At 15� (critical), the bonding between two gold nanocrystals results in defect-free single nanocrystals, whereas, for angles between 20� and 35� clear persistent defects at the merging interface are observed.");sQ1[136]=new Array("../7337/0271.pdf","Electron Beam Induced Domain Motion in Ferroelectric RKTP Observed By Transmission Electron Microscopy","","271 doi:10.1017/S1431927615002159 Paper No. 0136 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electron Beam Induced Domain Motion in Ferroelectric RKTP Observed By Transmission Electron Microscopy James L. Hart1, Miryam Arredondo2, Mitra L. Taheri1 1. 2. Department of Materials Science and Engineering, Drexel University, Philadelphia PA, US School of Mathematics and Physics, Queens University, Belfast Northern Ireland, UK Utilization of ferroelectrics for practical applications requires precise control of domain structure, thus ferroelectric domain behavior is a widely researched area. In-situ transmission electron microscopy (TEM) is an increasingly popular tool for studying domain dynamics at high spatial and temporal resolution [1]�[3]. In this technique domain behavior is studied as a function of an applied electric field, however, the TEM's electron beam alone can induce ferroelectric domain motion independent of the applied field. Here we present electron beam induced domain motion in the uniaxial ferroelectric Rb doped KTiOPO4 (RKTP), using a JEOL 2100 LaB6 TEM operated at 200 keV. This process is dependent on both the sample geometry and electron beam conditions. By manipulating the electron beam we can control the direction of domain propagation, and by using a condensed probe we can locally nucleate domains. In TEM a negligible amount of beam electrons are trapped within the sample volume. Charging occurs due to Auger and secondary electron emission after inelastic scattering interactions [4]. For a conducting sample, charging is alleviated through compensating electric currents, but for an insulating material such as RKTP, large positive charge can accumulate beneath the electron beam leading to strong internal fields. RKTP samples were prepared using a focused ion beam (FIB). All images shown here are DF images of 001 type reflections (Figure 1A). Initial domain morphology is shown in Figure 1B. Pre-existing domains are present, extending across the length of the sample with domain walls parallel to the c-axis (ferroelectric axis). When the sample is uniformly illuminated under the electron beam, the entire sample develops a positive charge. Because these samples were prepared in the FIB, there is a layer of conducting platinum on the top edge of the lift-outs. This metallic contact locally alleviates charging, leading to non-uniform charge build-up and a resultant internal electric field pointing towards the metal contact. This field acts to shrink pre-existing c- domains (Figures 2A and 2B). Conversely, if the specimen is non-uniformly illuminated so that only the top half of the sample is under the beam, only the top section positively charges, leading to an electric field pointing away from the top metal contact. Under such conditions we see nucleation and propagation of domains from the bottom of the sample up (Figure 2C). When the beam is condensed to a fine probe and placed in a mono-domain area (Figure 3A), only the illuminated region charges. The resultant electric field points radially away from the illuminated area. Because RKTP is a uniaxial ferroelectric, there is a single direction where the field is aligned with the ferroelectric axis allowing for domain nucleation. Under condensed beam conditions, we see domain nucleation at the perimeter of the electron beam (Figure 3B). This is the expected result as the beam perimeter has the highest charge density gradient and thus the strongest electric field [4]. This work is important for two reasons. Firstly, it shows that TEM observation of ferroelectrics can induce substantial domain motion. Such effects must be considered when analyzing results from in-situ Microsc. Microanal. 21 (Suppl 3), 2015 272 TEM biasing experiments to ensure observed domain motion is a function of the applied field and not an artifact of beam induced fields. Secondly, this work demonstrates a possible mechanism for sub-micron scale ferroelectric control mediated through an electron beam. As shown here, a TEM beam can locally nucleate domains within the sample bulk or at an interface, and a TEM beam can dictate the direction of domain propagation based on beam location and sample geometry. References: [1] [2] [3] [4] C. R. Winkler et al, Micron 43 (2012), p. 1121�2226 C. R. Winkler et al, J. Appl. Phys 112 (2012) C. R. Winkler et al, Nano Letters 14 (2014), p. 3617�3622 J. Cazaux, Ultramicroscopy 60 (1995), p. 411�425 Figure 1. A. Two beam condition used to obtain DF images. B. Dark field TEM image of RKTP sample showing initial domain configuration. Initial configuration Domain motion Nucleated domains Figure 2. A. DF TEM image of initial RKTP domain. B. Downwards propagation of domain due to uniform sample illumination and charge dissipation through top metal contact. C. Nucleation and propagation of domains from the bottom of the sample up due to non-uniform illumination. Illuminated region Nucleated domains Figure 3. A. Image of monodomain area. Circle shows illuminated area. B. Domains which were locally nucleated due to the condensed beam. Nucleation occurred at perimeter of beam, shown in 3A.");sQ1[137]=new Array("../7337/0273.pdf","In Situ Analysis of the Fracture Behavior of Nanocrystalline Copper Using Precession-Assisted Crystal Orientation Mapping.","","273 doi:10.1017/S1431927615002160 Paper No. 0137 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In Situ Analysis of the Fracture Behavior of Nanocrystalline Copper Using Precession-Assisted Crystal Orientation Mapping. Paul Rottmann1, Kevin Hemker1 1. Johns Hopkins University, Department of Materials Science and Engineering, Baltimore, MD, USA In recent years nanocrystalline metals with average grain sizes of less than 100 nm have been introduced into a growing number of applications. Nanocrystalline materials differ greatly in many respects from their coarse-grained counterparts. They can be fabricated through vapor-phase deposition (e.g. sputtering, evaporation) or electrodeposition to produce complex parts for small-scale applications, and they are used extensively in micro electromechanical systems (MEMS) and microprocessors. Nanocrystalline materials possess many unique properties, such as fatigue limit [1], deformation mechanisms [2], and magnetic properties [3]. Understanding the physical properties of these materials is imperative if they are to be used as critical components of many applications. This study focuses on the fracture of nanocrystalline copper when subjected to tensile loading. Freestanding copper thin films were fabricated by electron beam vapor deposition at a rate of 1.5�/s to a final thickness of 65 nm. After being freed from the silicon wafer with an acetone lift-off procedure the films were attached to a straining grid. The as-deposited grain size was 28.7 � 14.4 nm. Each film was annealed at 300�C under vacuum (10-6 Torr) for 1 hour, producing a final grain size of 62.2 � 51.2 nm. This resulted in a microstructure with equiaxed grains containing numerous twin boundaries, and the majority of grains were larger than the thickness of the film. The films were characterized using precession-assisted crystal orientation mapping (PACOM) [4]. Utilizing PACOM is advantageous as compared to conventional TEM microscopy, because it allows for full mapping of the crystal orientation at each point. This produces high-resolution maps in which information about grain size, grain boundary character, and grain boundary morphology is readily available and quantifiable. For these samples, the experiments were conducted by straining each film using a Gatan single-tilt straining holder (Model 654) to strain the film until a crack began to propagate across the gage section. Once a crack began to move the straining was stopped and a PACOM scan taken. The scans for this study were taken with a step size of 3.6 nm with a nominal probe size of 2.2 nm. After each scan the specimen was strained further to propagate the crack by a few hundred nanometers, and a subsequent PACOM scan was taken. This process was repeated to produce maps of a given area as it is subject to strain. Figure 1 shows results from one such experiment. An initial scan, before crack propagation, and the final scan, taken after the crack had propagated 2.4 �m through the sample are shown in Figures 1a and 1b, respectively. The color in each of these pictures represents the out-of-plane orientation at each point (see inverse pole figure legend in Figure 1), and the grayscale indicates the image quality, or confidence, in the selected orientation. The eventual crack path is indicated by a dashed line in Figure 1a. As the film was strained, a number of changes in the local microstructure were observed around the crack tip. Figure 1c shows the evolution of a large grain (outlined by a white box in Figure 1a) as the crack front proceeds. Initially, the grain has a single orientation with a twin toward the right side of the grain. As the crack begins to propagate through the grain, new twin boundaries are formed (marked by dashed lines in Figures 1b and 1c) in the grain. The crack precedes transgranularly across this large grain, not following any of the twin boundaries. By analyzing the crack path, it was determined that the crack primarily propagated intergranularly, along high-angle grain boundaries, (Figure 1d) with the notable Microsc. Microanal. 21 (Suppl 3), 2015 274 exception that 3 twin boundaries were never found in the crack path even though 3 boundaries comprised 15.3 percent of all boundaries. This indicates that 3 boundaries in nanocrystalline copper are more resistant to fracture than other grain boundaries with random misorientations. References: [1] T. Hanlon, Y.-N. Kwon, S. Suresh; Scripta Materialia; 49 (2003) 675 [2] T.J. Rupert, D.S. Gianola, Y. Gan, K.J. Hemker, Science 326 (2009) 1686 [3] G. Herzer; Journal of Magnetism and Magnetic Materials; 112 (1992) 258 [4] E.F. Rauch et al., Microsc. Microscopy and Analysis, 93 (2008), S5. This work was supported by the U.S. Department of Energy, Office of Science, Basic Energy Sciences, under Award # DEFG0207ER46437. Marc Legros (CEMES-CNRS) is thanked for his many useful discussions and contributions to this work. (a) (c) 50 nm (i) 0.6 0.5 (ii) (iii) (d) Length Fraction 0.4 Average 0.3 0.2 0.1 0 Crack Path (b) 100 nm 100 nm Through grain 3 2�-20� 20�-40� 40�+ Interface Type Figure 1. Orientation map of specimen area prior to (a) and after (b) crack propagation. Color corresponds to crystal orientation and grayscale indicates image quality. Eventual crack path is marked by a dashed line. (c) Magnified images of a grain as the crack propagates through it. Specimen is further strained from images (i) � (iii). Twin boundaries nucleated during straining are marked with a dashed line. (d) Graph comparing the average grain boundary misorientation of the specimen to the path followed by the crack.");sQ1[138]=new Array("../7337/0275.pdf","Observation of the Potential Distribution in GaN-Based Devices by a Scanning Electron Microscope","","275 doi:10.1017/S1431927615002172 Paper No. 0138 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Observation of the Potential Distribution in GaN-Based Devices by a Scanning Electron Microscope T. Karumi1 and S. Tanaka2 1. 2. Dept. Electrical Eng., Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8603, Japan EcoTopia Science Institute, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8603, Japan For semiconducting devices, the precise control of their electro-static potential distributions is very important for device operations. For GaN-based light emitting diodes (LEDs), quantum wells (QWs) are usually used in the active region, and the wells are under strained state generating piezoelectric fields. These piezoelectric fields isolate electron-hole pairs, and this leads to a reduction of the luminescence efficiency. Therefore, understanding of the potential distribution is essential for the development of LEDs with improved emission efficiency. Electron holography has been used for this purpose [1-4], however, sample preparations are not easy. In recent years, mapping of the potential distribution using a scanning electron microscope (SEM) has been reported [5][6] for semiconductors such as Si, GaAs and InP. Sample preparation for SEM is easy and quick. But, there are no such studies on GaN-based devices, to our knowledge. In this study, we observed two types of GaN-based devices by SEM to see if there is a condition that the contrast matches the potential distribution of the devices. The first device we studied is GaN p-n junction (p, n ~5�1017 cm-3) grown on c-sapphire by metal organic vapor phase epitaxy (MOVPE). The device was cut, and polished from the cross-section to a flat surface. The cross-section was observed by SEM. Figure 1(a) shows an SEM image taken at 3 kV. The p-region appears bright and the n-region appears dark. The image intensity changes at the position of p-n junction, for which we used electron beam induced current (EBIC) technique to determine the p-n junction position. Figures 1(b) - 1(d) compare SEM line profiles across the p-n junction (shown in red) with calculated potential distributions (black) for three different values of p and n concentrations. In the calculation, p and n concentrations are assumed to be the same. Flat parts of SEM line profile in p and n regions were fitted with the potential profile, and the change at the junction is compared. It can be seen that the intensity profile across the p-n junction matches the potential distribution with p, n ~8�1017 cm-3 very well. The SEM observations were carried out for several accelerating voltages. But, best result was obtained at 3 kV. For lower accelerating voltages, the image seemed to reflect the surface potential. On the other hand, higher accelerating voltages resulted in blurred images. The second sample is a green light emitting diode with a multiple quantum well (MQW) structure in the active region. The MQW structure consists of five 2.5-nm-thick InGaN wells separated by 10-nm-thick GaN barriers and is sandwiched by p- and n-GaN. These layers were grown on c-sapphire by MOVPE. The sample was obliquely polished from the surface (~10�) to improve the lateral resolution. Figure 2(a) shows an SEM image of the MQW structure taken at 3 kV. The wells appear as dark bands. Figure 2(b) shows a line profile across the p-n junction. We think the difference of the SEM signal intensity between p and n region corresponds to the built-in potential, and we assume the SEM signal intensity from each barrier layer varies with the piezoelectric field of the neighboring well. Based on these assumptions, piezoelectric fields of each well are calculated and the result is shown in Figure 3. Overall direction of the piezoelectric field is from the surface side to the substrate side. The strength decreases from the surface side to the substrate side. These features match with our previous analysis using electron Microsc. Microanal. 21 (Suppl 3), 2015 276 holography [4]. But, the absolute strength is smaller than the previous one. This may be due to a possible relaxation of the strain in wells because of the oblique structure of the sample. References: [1] D. Cherns et al, Solid State Commun. 111, 281 (1999). [2] J. Cai and F. A. Ponce, J. Appl. Phys. 91, 9856 (2002). [3] Z. H. Wu et al, Appl. Phys. Lett. 91, 041915 (2007). [4] M. Deguch et al, J. Electronic Materials 39, 815 (2010). [5] C. P. Sealy et al, J. Electron Microscopy 49, 311 (2000). [6] B. Kaestner et al, Appl. Phys. Lett. 84, 2109 (2004). [7] We thank Professor H. Amano (Nagoya University) for providing the samples. Fig.1 SEM image (a) and comparison of intensity profile across the junction with calculated potential profiles. The line profile was measured at the boxed region in (a). Dopant concentration: (b) 3�1017 cm-3, (c) 8�1017 cm-3, (d) 3�1018 cm-3. Fig.2 SEM image (a) and intensity profile across the junction (b). b1 to b5 indicates the position of barriers. Fig.3 Estimated piezoelectric field strength. QWs are numbered from the p-side (surface side).");sQ1[139]=new Array("../7337/0277.pdf","Denickelification and Dezincification of Copper Alloys in Water Environments","","277 doi:10.1017/S1431927615002184 Paper No. 0139 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Denickelification and Dezincification of Copper Alloys in Water Environments Jun-Fu Liu, Mel J. Esmacher and David Kotwica GE Power & Water, Water & Process Technologies, 9669 Grogans Mill Road, The Woodlands, TX 77380 Due to their high thermal conductivity, excellent ductility, and high resistance against corrosion, copper alloys are widely used in heat exchanger, tube, pipe fittings, and other components in water environments. However, de-alloying corrosion of copper alloys can occur in water environments, in which one or more elemental constituents are preferentially dissociated [1]. This presentation reports on denickelification and dezincification cases involving copper alloys. To examine the failure of the copper alloy parts, optical microscopy and scanning electron microscopy (SEM) were used to acquire the surface morphology. Energy dispersive spectroscopy (EDS) was utilized to analyze the deposit and corrosion product composition. Case 1 (Figure 1a) is a failed copper-nickel tube section (10% Ni, 1% Fe, 0.55% Mg, balance Cu) from a heat exchanger. City make-up water circulating on the tube side was heated by a hot gas stream on the shell side of the exchanger. The time in service was 20 years. Purple deposits are observed at some locations on the internal surface (Figure 1b). SEM images in Figure 1(c, d) and EDS analysis in Figure 1e show that the purple deposits consist of Cu micro-crystals. Case 2 (Figure 2a) is a silicon bronze pump impeller (7.3% Zn, 4.3% Si, balance Cu) from a booster pump. The time in service in potable water was 2 years. In addition to erosion corrosion (Figure 2a) due to excessive flow velocity and cavitation, surface dezincification of the bronze alloy resulted in localized red deposit area (Figure 2b). The SEM images in Figure 2(c, d) show that the area has a micro-porous surface texture. It is 99% Cu from the EDS analysis result. The de-alloying process is consistent with the mechanism of simultaneous dissolution of Ni (or Zn) and Cu, followed by re-deposition of Cu. Ni (or Zn) � Ni2+ (or Zn2+) + 2e- (Ni or Zn oxidation) Cu � Cu2+ + 2e- (Cu oxidation) Later Cu2+ ions were reduced to metallic Cu due to a higher tendency of Cu2+ than Ni2+ or Zn2+ ions to plate out of solution. Cu2+ + 2e- � Cu (Cu reduction) References: [1] S. Selvaraj et al, Corrosion Review 21 (2003), p. 41. Microsc. Microanal. 21 (Suppl 3), 2015 278 a b c 200x d 1000x e Figure 1. Photographs of (a) Ni-Cu tube and (b) purple deposit on the inside surface, and SEM images (c, d) of Cu micro-crystals of the purple deposit, and (e) EDS result. a b c 100x d 1000x Figure 2. Photographs of (a) a silicon bronze pump impeller and (b) red deposit on the inside surface, and SEM images (c, d) of red Cu deposit area.");sQ1[140]=new Array("../7337/0279.pdf","Application of EFTEM and XEDS Elemental Mapping to Characterization of Nanometer Devices in Semiconductor Wafer-Foundries","","279 doi:10.1017/S1431927615002196 Paper No. 0140 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Application of EFTEM and XEDS Elemental Mapping to Characterization of Nanometer Devices in Semiconductor Wafer-Foundries Wayne W. Zhao1, Michael Gribelyuk1, and Jeremy D. Russell1 Engineering Analysis - Physical, Technology Development & Yield Engineering, GLOBALFOUNDRIES, Malta, New York, USA. Precise characterization of the elemental distribution in the nanometer semiconductor device with high spatial resolution is an essential element of the Physical Failure Analysis (PFA) Laboratory to help develop and manufacture next generation of devices [1][2][3]. Nowadays state-of-the-art analytical TEM instruments possess the x-ray energy dispersive spectroscopy (XEDS) and/or the electron energy loss spectroscopy (EELS) capabilities to fulfill this goal. The acquisition of XEDS spectra on modern instruments is relatively straightforward and does not require an extensive training of the TEM analyst. It has high sensitivity to species with medium and high atomic number [4][5]. These features made it attractive for application in PFA. However, XEDS may not be the method of choice for the following reasons. First, it is not very sensitive to light elements, (e.g., B, C) [4][5], which are typically found in semiconductor devices. Second, peaks of a number of elements which are used in device structures overlap in the XEDS spectrum. Examples include Cu (K 1) vs. Ta (L 1), P (K 1, 1) vs. Pt (M 1, 1) or Zr (L , 1), etc. This makes it challenging to directly interpret results without further processing. Third, spurious x-ray peaks originating from excitations of components inside the electron microscope can also contaminate the observed XEDS spectrum. To answer this, EELS should be used as a complimentary method to provide the full elemental analysis. It is more sensitive to light atomic weight elements [4][5]. In the STEM configuration only the region of the sample under the electron probe contributes to the EELS spectrum. Therefore EELS data provide results with higher spatial resolution as compared to XEDS. The energy resolution of the EELS spectrum in non-aberrated instruments is E<1eV, which makes it possible to detect individual edges of most elements used in semiconductor structures without having to deconvolute the edges. However, the analyzed samples need to be thin, i.e. less than 50nm, and have surfaces free of redeposition. Moreover, subtraction of the spectrum background, processing and interpretation of the results requires an in-depth understanding of the analytical TEM configuration and electron diffraction physics. Since no single method can provide a full solution of the problem, analysts in the PFA Lab are challenged to find the optimum plan to analyze the problem in the shortest time. 1 1. Presented here are examples of how we combine both methods to quickly provide engineering solutions for semiconductor process integration on an SRAM device. Fig. 1 displays a TEM, an overlay and individual EFTEM elemental maps, which revealed details of elemental compositions inside the bridging portion between the dummy gates. Fig. 2 is a STEM on the same structure, and XEDS counterparts acquired by a Super-X with Bruker Esprit analysis software. Even with the primitive version of EELS by EFTEM mapping, details of the defect can still be delineated (arrowed). In contrast, the XEDS elemental mapping yields more ambiguous results, due to its intrinsic disadvantage of spatial resolution relative to EELS. However, EFTEM can only process one element at a time and maybe subjected to artefacts caused by the background subtraction problems. For features with only a few light elements present in the structure with high content (see the example), or in case when the structure is not sensitive to electron beam damage, EFTEM is very useful for applications demanding a fast turnaround with medium spatial resolution. For complex device structures consisting of both medium and light Microsc. Microanal. 21 (Suppl 3), 2015 280 elements, or if high spatial resolution is essential, we suggest to use XEDS, and then complement results by STEM-EELS mapping. References: [1] W. Zhao, et al., Microscopy & Microanalysis, Vol. 20 (Supplement 3), (2014), pp.362~363. [2] W. Zhao, M. Gribelyuk, et al., Proc. 38th International Symposium for Testing and Failure Analysis, (2012), pp. 347~355. [3] W. Zhao, Symp. Proc. the Material Research Society, Fall Meeting, (2002), Vol. 738, pp. G7.15.1~6. [4] R. Leapman and J. Hunt, Microscopy, Microanalysis, Microstructure, Vol. 2, (1991) pp. 231-244. [5] H. Harrach, et al., Microscopy & Microanalysis, Vol. 16 (Supplement 2), (2010), pp.1312~1313.). [6] The authors wish to thank and acknowledge Derrick Franks for preparing this TEM sample. 1a 2a 1b 2b 1c 1d 1e 2c 2d 2e Figure 1. (a) TEM image; (b) an overlay of EFTEM maps; and individual elemental map by EFTEM, (c) N (red); (d) O (cyan); and (e) Si (yellow). Figure 2. (a) HAADF-STEM image; (b) an overlay of XEDS maps, and individual elemental map by XEDS, (c) N (red); (d) O (cyan); and (e) Si (yellow).");sQ1[141]=new Array("../7337/0281.pdf","Isolating the Photocatalytic Degradation of Methylene Blue Dye on TiO2 Surface","","281 doi:10.1017/S1431927615002202 Paper No. 0141 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Isolating the Photocatalytic Degradation of Methylene Blue Dye on TiO2 Surface Simon Fowler1, Emilio Molina2, Esteban Rodriguez-Ariza3, Sydney Quinton-Cox4, and Jun Jiao1,2 1. 2. Department of Physics, Portland State University, Portland OR, USA Department of Mechanical Engineering, Portland State University, Portland OR, USA 3. Department of Chemistry, Portland State University, Portland OR, USA 4 Department of Bioengineering, Oregon State University, Corvallis OR, USA Photocatalytic degradation of organic molecules has a broad applicability and has been studied with particular intensity for the purpose of water and air purification [1]. In this process, light is absorbed by a semiconducting material and converted to electron-hole pairs, which in turn react with water molecules to form hydroxyl radicals (�OH). These highly oxidative molecules go on to degrade organic molecules through what is known as an Advanced Oxidation Process (AOP) [2]. However, the result is usually achieved through diverse pathways, and isolation of the individual mechanisms is difficult. In particular, this experiment is designed to isolate the degradation of an organic marker at the surface of the catalyst material, as opposed to degradation away from the catalyst surface (i.e. within the aqueous medium). This has been achieved by designing an experiment to observe the relative mass of methylene blue (MB) on the catalyst surface during a photodegradation reaction. The TiO2 film was deposited onto glass beads in order to immobilize the catalyst material on a visibly transparent substrate, allowing measurement of dye concentration via visible light spectroscopy. In preparation for this experiment a thin film of nanocrystalline TiO2 was deposited onto the surface of borosilicate beads of 1 mm diameter. This was accomplished by mixing the glass beads into a solution of water, TiO2 powder, polyacrylic acid, and tetraethylorthosilicate. The beads were then removed from the solution, tumble dried, and finally heat treated in open air at 600C for 8 hours. SEM micrographs of one bead from this batch are shown in Figure 1. Next the beads were mixed in a solution of water and MB for several minutes and then removed. It was determined that 12.8 mg of MB had adsorbed to the TiO2 surface by measuring the concentration of MB in the water before and after the addition of the beads using a Shimadzu UV-3600 Spectrophotometer. In order to measure the relative amount of dye on the surface of the beads, the beads were packed into a quartz cuvette and the beads' absorption of visible light was measured and compared to a reference spectrum collected from the beads before they were soaked in the MB solution. The dyed beads were then packed into a small photoreactor and illuminated with light from a Newport solar spectrum simulator, model 69907, calibrated to AM 1.5 conditions as shown in Figure 2. After 2 min of illumination the relative absorbance was measured again. This process of illumination and measurement was then repeated two more times. The amount of remaining dye as a function of time was calculated from the relative height of the 572nm absorption peak above the reference. This data was plotted along with the raw data in Figure 3. This experiment presents a novel approach to for quantifying photocatalytic degradation on a catalyst surface: in this case the overall result was a degradation of 92% of the MB after 6 min of illumination. This result was 5 times faster Microsc. Microanal. 21 (Suppl 3), 2015 282 than a control experiment using the same process, but without the addition of catalytic beads to the reaction chamber. References: [1] Lazar, M. A., Varghese, S. & Nair, S. S., Catalysts 2, 572�601, 2012 [2] R. Andreozzi, et.al., Catalysis Today, vol. 53, no. 1, pp. 51�59, Oct. 1999. [3] The authors would like to thank the Research Experience for Undergraduates, Louis Stokes Alliances for Minority Participation, and Undergraduate Research Mentorship Program. Figure 1. SEM micrographs of coated beads showing the uniformity and macrostructure of the coating. Removal of a small area of the coating with a razor reveals a 4 �m thickness of the TiO2 layer. Figure 2. Photoreactor designed and used for these lab-scale tests of photodegradation reactions. Figure 3. a) Raw data obtained from UV-vis spectroscopy of dyed beads; b) relative mass of MB on TiO2 surface calculated from absorption spectra (control experiment plotted as dotted line).");sQ1[142]=new Array("../7337/0283.pdf","Influence of Hydrogen on Fracture Mode API X60 Steel","","283 doi:10.1017/S1431927615002214 Paper No. 0142 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Influence of Hydrogen on Fracture Mode API X60 Steel G. Gonz�lez-Mancera1, Cort�s-Su�rez V.2, Mariano-Alberto M.2 1. Depto. Ing. Metal�rgica Fac. Qu�mica, Universidad Nacional Aut�noma de M�xico, Ciudad Universitaria, M�xico, D.F., 04510 2. Depto. Materiales, Universidad Aut�noma Metropolitana-Azcapotzalco, Av. San Pablo 180, M�xico D.F., 02200. One of the applications of High Strength Low Alloy (HSLA) steel is its use in the manufacture of pipelines for transporting petroleum and its derivatives products [1]. These hydrocarbons produce hydrogen from hydrogen sulfide decomposition. Thus, hydrogen generated in the steel diffuses to regions such as grain boundaries, nonmetallic inclusions and other crystalline defects, and results in an embrittling effect, causing premature failure of the material [2]. The manufacture of HSLA steel involves getting a variety of microstructures which are produced from the transformation of austenite to phases such as ferrite, pearlite, and intermediate stages like bainite, martensite and acicular ferrite [3]. Intermediate stages can be obtained when the steel is subjected to high temperatures for long periods of time; conditions that promote the steel undergoes aging. Also, during welding, the melting zone shows a mix of microstructures. Each microstructure has a characteristic resistance to hydrogen embrittlement. Purpose of this paper is to determine the hydrogen effect on the fracture mode. Therefore, the microstructure and the fracture mode were compared between a control API X60 steel annealed at 650 �C and steel containing hydrogen, and aged at 315 �C and 650 �C. Hydrogen was introduced in these steels through cathode charged by applying a current density of 50 mA/cm2. After hydrogen charging, the specimens were tested in tension at a deformation rate of 10-4/s. And finally, SEM study was performed, to reveal the microstructure, and fracture surface of tested specimens. Results in Figure 1 (a) exhibits equiaxed microstructure with grains of polygonal ferrite corresponding to the annealing treatment. In contrast, the steel aged at 315 �C exhibited a mixture of steel microstructures of acicular ferrite, bainite and martensite as observed in Figure 1 (b). The steel treated at 650 �C exhibited grains of polygonal ferrite and acicular ferrite (Figure 1 c). In addition, this study reavealed three different fracture modes in APIX60 steel. Annealed steel presents fracture mechanism caused by coalescence of pores (Figure 2 a), indicating that the fracture is ductile. Steel aged at 315 �C has intergranular fracture facets and cleavage (Figure 2 b), which is characteristic of brittle fracture. In contrast, the specimen aged at 650 �C with dissolved hydrogen, has a mixed fracture mode (Figure 2 c). In this study, we determined that steel APIX60 aged causes fracture changes from ductile to brittle. In particular, the steel aged at 315 �C has the least resistance to hydrogen damage. References: [1] Pickering, F.B., Microalloying 75 (1977), p. 9 [2] Ren, X.C., et al.,Metallurgical and Materials Transactions A 39,(2008), p. 87-97 Microsc. Microanal. 21 (Suppl 3), 2015 284 [3] A.S. Sinha, Physical Metallurgy Handbook, Mc Graw Hill 15 (2002) a b b c Figure 1. SEM images from API X60 steel pieces. 1 (a) Control sample annealed at 650 �C. 1 (b) Specimen aged to 315 �C shows mixture of microstructures consisting of acicular ferrite, bainite and martensite from. 1 (c) Steel aged at 650 �C reveals microstructure consisting of acicular ferrite and polygonal ferrite grains a b b c Figure 2. SEM micrographs, of the fracture mode from API X60 steel samples. 2 (a) Annealed sample at 650 �C shows coalescence of pores and ductile fracture. 2 (b) Steel aged at 315 �C presents brittle fracture. 2 (c) Specimen aged at 650 �C displays a mixed mode fracture.");sQ1[143]=new Array("../7337/0285.pdf","Electron Energy-Loss Spectroscopy (EELS) Study of NbOx Film for Resistive Memory Applications","","285 doi:10.1017/S1431927615002226 Paper No. 0143 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electron Energy-Loss Spectroscopy (EELS) Study of NbOx Film for Resistive Memory Applications Jiaming Zhang , Kate Norris , Katy Samuels , Ning Ge , Max Zhang , Joonsuk Park , Robert 1 1 1 1 Sinclair2, Gary Gibson , J. Joshua Yang , Zhiyong Li , R. Stanley Williams 1. 2. 1 1 1 1 1 2 Hewlett Packard Labs, 1501 Page Mill Rd, Palo Alto, CA 94304 USA Department of Materials Science and Engineering, Stanford University, Stanford, CA 94305 USA Niobium dioxide (NbO2) is one of the Mott insulators that exhibit current-controlled negative differential resistance, also referred as threshold switching, when used in two-terminal devices. This phenomenon is caused by a reversible insulator-to-metal phase transition, which is proposed to induce a conductive channel in the device that bridges the two electrodes [1][2]. As the amount and local distribution of oxygen vacancy play important roles during the switching, characterizing the composition with high spatial resolution at the atomic level is required for understanding the working mechanism and potential failure. Previous EELS study has been conducted on metallic Nb and stable Nb oxides (NbO, NbO2, Nb2O5) in terms of energy-loss near-edge structures (ELNES) of all relevant Nb edges and O-K edge for fingerprints of Nb in different formal oxidation states [3][4]. In this study, we provide a quantitative study on the amorphous NbOx thin-films by ion beam sputtering for resistive switching applications. The EELS quantification in NbOx film can provide the variation of local composition and chemical states across the film, which helps to understand the device behavior. NbO2 film was deposited by RF magnetron sputtering from an NbO2 target. The sample substrate was pumped down to below 1E-6 torr and was then heated up in vacuum to 450 �C. 15nm NbO2 blanket film was deposited in Ar while substrate was maintained at 450 �C. Subsequently, top electrode in the stack of 3nm Ti4O7, 6nm TiN, 5nm Pt, and 20nm Cr was deposited over the NbO2 film through a shadow mask. Cross-sectional samples were prepared using dual beam FEI Helios system. Blanket NbO2 and Nb2O5 films, and Nb2O5 crystal powders are also studied as reference samples. The TEM and EELS characterization was done using a FEI Tecnai TF-30 equipped with Gatan GIF Quantum system. The 2D EELS Spectra Image (SI) with O-K (532eV) and Nb-L edges (2371eV) were collected in STEM mode at 300keV with convergence semi-angle 8.7 mrad and collection semi-angle of 19 mrad. The quantification procedure was done by subtracting power law background and integrating cross-section based on Hartree-Slater model from Gatan GMS software. Figure 1a shows the relative O composition in the NbO2 and Nb2O5 blanket films (~200 nm thick) deposited on a Si/SiO2 substrate. Nb2O5 crystalline powder was used to correct the instrument factor for EELS quantification as shown in Table I. The average of measured value in Nb2O5 powder is 60.8% and has been corrected to be 71.4% after multiply an instrument factor of 1.174. The line profiles in the films show the average of 58.7 O% in NbO2 and 63.4% in Nb2O5 after the correction, with higher O composition on the surface due to expose to the air after deposition. The lower O composition in the film than the desired is due to the preferential evaporation of oxygen. Figure 2 shows the NbO2 resistive switching device with TiN bottom electron. The color elemental mapping was processed from STEM/EELS SI. The line profiles of elements Nb, Cr, O, Pt, and Ti were analyzed and consistent with the nominal thickness. The NbO2 layer shows similar O deficiency Microsc. Microanal. 21 (Suppl 3), 2015 286 compared with reference blanked Nb oxide films. The top surface of NbO2 shows higher O content is due to adjacent Ti4O7 layer with higher O content. In conclusion, we have done quantitative EELS analysis on the O content in the sputtered NbO2 and Nb2O5. These results are important reference to study the dynamic materials evolution during heating and biasing with the advanced in situ TEM techniques. Preliminary in situ TEM study shows that there is an amorphous-to-crystalline transition at 550�C in the NbO2 film. References: [1] F. A. Chudnovskii, L. L. Odynets, A. L. Pergament and G. B. Stefanovich, Journal of Solid State Chemistry 122 (1), 95-99 (1996). [2] M. D. Pickett, G. Medeiros-Ribeiro and R. S. Williams, Nat Mater 12 (2), 114-117 (2013). [3] D. Bach, R. Schneider and D. Gerthsen, Microscopy and Microanalysis 15 (06), 524-538 (2009). [4] D. Bach, R. Schneider, D. Gerthsen, J. Verbeeck and W. Sigle, Microscopy and Microanalysis 15 (06), 505-523 (2009). Table I. EELS quantification results for NbO2, Nb2O5 film, and Nb2O5 standard powders from Sigma-Aldrich. FIG. 1 Relative oxygen composition across NbO2 and Nb2O5 film with one cross-sectional STEM DF image of reference film in the inset. FIG. 2 (a) Cross-sectional STEM Z-contrast image of device TE/NbO2/BE; (b) elemental color mapping from EELS spectrum image, (c) line profile across the device, (d) relative oxygen composition across NbO2 layer.");sQ1[144]=new Array("../7337/0287.pdf","Microscopic Origin of Strength and Microhardness of Titanium Alloy at Elevated Temperature","","287 doi:10.1017/S1431927615002238 Paper No. 0144 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microscopic Origin of Strength and Microhardness of Titanium Alloy at Elevated Temperature M. Islam1, C. Fermin1 and H. Aglan1 1. Tuskegee University, Tuskegee, AL 36088 USA Titanium alloys are promising materials for use in extreme environments especially in nuclear fusion and fission reactor systems, biomedical applications, aerospace and high temperature industrial applications [1]-[2]. In this study the influence of environmental temperatures up to 900oC on the mechanical properties of dual phase Ti-6Al-4V alloy was investigated. The origin of strength at room temperature and lack of ductility after elevated temperature exposure were identified microscopically. The titanium alloy used in this study was cold rolled Ti-6Al-4V (Grade 5) sheet of thickness 0.81 mm. The specimens were heated in a vacuum furnace at temperatures of 450oC and 900oC. A ramp rate of 5 C/min, hold time for 2 hours, and then cool down to room temperature at a rate of 1 C/min were used. Tensile tests were carried out at room temperature using an MTS 810 Servo-hydraulic Machine at a cross head speed of 0.02 mm/s. The fracture surfaces were examined with a Hitachi S-3400N SEM. Specimens were polished and etched in a metallographic Keller's reagent composed of 2.5% HNO3, 1.5% HCl, 1% HF, and 95% distilled water for optical microstructure studies. The ultimate tensile stress, yield stress, elongation, and microhardness as functions of environmental temperature were evaluated. The tensile strength and elongation were not affected after the 450oC exposure. After 900oC, the tensile strength and elongation dropped by 25% and 88%, respectively. The microhardness was increased by44%, from HRC 36 to 52, from their room temperature values. The microstructure of the Ti-6Al-4V alloy showed a randomly distributed equiaxed structure, composed of alpha grains around 5-20 m and surrounded by intergranular phase with bands of 0.5-1 m width and 2-5 �m elongated structure. The alpha stabilizer (6% Al) produced coherent ordered 2 phase (Ti3Al) to strengthen the structure (Figure 2a black arrow). The stabilizer (4% V)was dispersed in the phase structure (Figure 1a and 2a red arrows). The volume fraction of the phase was around 35 % at room temperature and after 450oC exposure. After 900oC exposure for 2 hours, partial phase transformation of HCP to BCC occurred and the phase transformed to primary alpha grains (Ti3Al). Some acicular grains of 5-10 m in length and 0.05-1 m in width were observed and the volume fraction was reduced by 10-15%. The material became brittle because the formation of Ti3Al increased (Figure 3a white area). This was a major contributor to the ductility loss which is supported by Wang et al.[3]. Optical and SEM micrographs at different temperatures are given in Figures 1-3 that reveal ductile dimple features were dominant. The dimple size became larger and deeper with increasing the environmental temperature. The average dimple size was between 3-6 m for both room temperature and 450oC. After 900oC exposure for two hours, the dimples were significantly larger (30-40 m). References [1] M Victoria et al., Structural materials for fusion reactors, Nuclear Fusion, 41(8), (2001) p. 1047. [2] I Gurrappa, Characterization of titanium alloy Ti-6Al-4V for chemical, marine and industrial applications. Materials Characterization, 51(2), (2003)p. 131-139. [3] SH Wang et al., Tensile properties of LBW welds in Ti�6Al�4V alloy at evaluated temperatures below 450 C. Materials Letters, 57(12), (2003) p. 1815-1823. [4] Supported by NSF; Partnerships for International Research and Education (PIRE) program. Microsc. Microanal. 21 (Suppl 3), 2015 288 Figure 1. Optical (1a) and SEM (1b) micrographs of the Ti-6Al-4V metal at room temperature. Alpha phase (black arrow) and beta phase (red arrow), Figure 1a. Dimple structure (yellow discontinuous circle) and pulled up strip features (yellow arrow), Figure 1b. Figure 2. Optical (2a) and SEM (2b) micrographs after 450o C exposure. Few large dimples (yellow discontinuous circle), Figure 2b. Figure 3. Optical (3a) and SEM (3b) micrographs after 900oC exposure. Acicular grains (bottom red arrow) and primary grains of 10-15 m size (top red arrow), Figure 3a. Large dimples (red discontinuous circle), voids (yellow continuous circles), and brittle like fracture facets (red arrow), Figure 3b. An 88% reduction in elongation is associated with these fracture features.");sQ1[145]=new Array("../7337/0289.pdf","Microstructural Properties of Piezoelectric 0.95(Na0.05Bi0.05TiO3)-0.05(BaTiO3) Thin Films on SrTiO3 (001) Substrates","","289 doi:10.1017/S143192761500224X Paper No. 0145 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microstructural Properties of Piezoelectric 0.95(Na0.05Bi0.05TiO3)0.05(BaTiO3) Thin Films on SrTiO3 (001) Substrates Xiao-Wei Jin1, Lu Lu1, Shao-Bo Mi1,, Sheng Cheng1, Ming Liu1, Chun-Lin Jia1,2 1 International Center for Dielectric Research, The School of Electronic and Information Engineering, Xi'an Jiaotong University, Xi'an 710049, China. 2 Peter Gr�nberg Institute and Ernst Ruska Center for Microscopy and Spectroscopy with Electrons, Forschungszentrum J�lich, D-52425 J�lich, Germany. Lead-free piezoelectric materials of sodium bismuth titanate (NBT)-based compounds, e.g. (1x)(Na0.5Bi0.5TiO3)-x(BaTiO3) (NBT-BT) solid solutions, have been extensively studied due to their excellent dielectric and piezoelectric properties [1-3]. Nevertheless, the strong deviation of the electrical properties from the bulk has been observed in the NBT-BT films, e.g. high leakage current density [4]. It is known that the microstructure, structural defects and chemical inhomogeneity can strongly affect the physical properties of materials and thus affect the performance of the related devices. In spite of the extensive studies on physical properties and thin film growth [5], the structural properties of the NBTBT films at the atomic scale still remain unclear. Thin films of 0.95NBT-0.05BT were deposited on SrTiO3 (001) substrates at 850�C using the highpressure sputtering system. The mixed ambient of Ar and O2 at the ratio of 1:1 was used with a flowing pressure of 0.2 mbar during the film deposition. Cross-sectional specimens for (scanning) transmission electron microscopy ((S)TEM) investigations were prepared by both argon ion milling method and focused ion beam (FIB) technique (FEI Dualbeam Helios NanoLab 600i). To explore the microstructural properties of 0.95NBT-0.05BT, atomic-resolution high-angle annular dark-field (HAADF) and annularbright field (ABF) imaging were performed on a JEOL ARM200F with a probe aberration corrector, operated at 200 kV. Figure 1(a) shows a low-magnification bright-field (BF) TEM image of 0.95NBT-0.05BT films on SrTiO3 (001) substrates. The contrast difference between 0.95NBT-0.05BT and SrTiO3 is visible, which allows us to determine the 0.95NBT-0.05BT/SrTiO3 interface. Contrast variations can also seen in the image, which is due to the existence of planar defects in the 0.95NBT-0.05BT films. In most cases, the planar faults show the habit plane parallel to the {100} plane of 0.95NBT-0.05BT and either terminate within films or penetrate the whole films, as indicated by a red and a white arrow, respectively. A selected-area electron diffraction (SAED) pattern is inserted in Figure 1(a), which was recorded from the films and part of the SrTiO3 substrate. Based on the zone axis SAED pattern, the crystallographic orientation relationship can be determined as [010]NBT-BT//[010]SrTiO3 and (001) NBT-BT//(001)SrTiO3. In addition, we noted that diffuse scattering streaks around reflection spots from the 0.95NBT-0.05BT films appear in the SAED pattern, as indicated by a pair of horizontal arrows. The presence of these diffuse streaks results from a high density of planar defects parallel to the film growth direction. An atomic-resolution HAADF-STEM image of a typical planar fault in the 0.95NBT-0.05BT films is displayed in Figure 1(b), which is recorded along the [010] zone axis of 0.95NBT-0.05BT. The planar defect consists of edge-sharing oxygen octahedra with Bi atoms at the center of the space formed by Electronic address: sbmi1973@gmail.com Microsc. Microanal. 21 (Suppl 3), 2015 290 octahedral network. In Figure 1(b), the structure of the planar defect has a repeat period three times as large as that of 0.95NBT-0.05BT along the film growth direction. This is consistent with the diffuse streaks appeared in the SAED pattern. It should be mentioned that this planar defect is identical to one of planar defects observed in the 0.95NBT-0.05BT films grown on Ni substrates buffered by a NiO layer [6]. The structural model of the planar defect is displayed in Figure 1(c). In fact, an ordered intergrowth of the planar defect and perovskite-type 0.95NBT-0.05BT phase has been observed in the 0.95NBT0.05BT/SrTiO3 system. Regarding the thickness of 0.95NBT-0.05BT between two adjacent planar defects and relative displacement of adjacent planar defects along the film growth direction, a number of ordered sequences characteristic of new structures could be locally obtained. The formation of the planar defect in the films occurs during the film growth since no planar defects have been observed in the 0.95NBT-0.05BT bulk materials [6]. Considering the difference in properties between thin films and bulk materials, the observed planar defect in the films can play an important role in modifying the film properties [7]. References: [1] H Wang et al, Mater. Des. 31 (2010) p. 4403. [2] Y Guo et al, Chem. Mater. 23 (2011) p. 219. [3] Y Kitanaka et al, Phys. Rev. B 89 (2014) p. 104104. [4] MM Hejazi, E Taghaddos and A Safari, J. Mater. Sci. 48 (2013) p. 3511. [5] SK Acharya et al, J. Alloy Compd. 603 (2014) p. 248. [6] SB Mi et al, submitted for publication. [7] The work was supported by the National Basic Research Program of China (No. 2015CB654903) and the National Natural Science Foundation of China (Nos. 51471169 and 51390472). Figure 1. (a) A low-magnification BF-TEM image of 0.95NBT-0.05BT thin films on SrTiO3 (001) substrates. The inset is the corresponding SAED pattern. (b) An atomic-resolution HAADF image of a typical planar defect projected along the [010] zone axis of 0.95NBT-0.05BT. Vertical green arrows denote the Bi-rich double planes, and horizontal yellow arrows mark the Bi-deficient Na/Bi(Ba) columns. (c) The model of the planar defects. The corner-sharing oxygen octahedra in the normal regions are displayed in white, while those with edge sharing ones in the fault area in green. Bi-atom columns in the defects with different height levels are in red and blue, respectively. The Na/Bi (Ba) columns in the normal regions are presented in black, and the Bi-deficient columns in yellow.");sQ1[146]=new Array("../7337/0291.pdf","TEM Study of Nano Crystals in Li2O-Al2O3-SiO2 Glass-Ceramics","","291 doi:10.1017/S1431927615002251 Paper No. 0146 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 TEM Study of Nano Crystals in Li2O-Al2O3-SiO2 Glass-Ceramics Sz-Chian Liou and Wen-An Chiou NISP Lab, NanoCenter, University of Maryland, College Park, MD 20742-2831, U.S.A Glass-ceramics have an amorphous glassy matrix and one or more crystalline phases that produced by controlled crystallization process (of base glass). These unique properties of glass-ceramics depend on their chemical composition and microstructure, including the shape, size and volume fraction of the crystalline phases in glassy matrix [1]. The LixO-AlxOx-SiO2 (LAS) system is an important commercial glass-ceramic due to its extremely low thermal expansion coefficient [1, 2]. The main crystalline phases in LAS system are metastable solid solution of high-quartz or keatite depending on the crystallization temperature [2]. Crystal structures of high-quartz and keatites phases have non-cubic symmetry, thus, anisotropic thermal expansion induces thermal residual stresses due to the thermal and elastic mismatch between the crystalline phase and amorphous glassy matrix [2-3]. Characterization of the structure, shape and volume fraction of crystalline phases in LAS is important to understanding the glass-ceramics system. This paper presents a TEM study of the microstructure (size, shape, distribution and volume fraction) of crystalline phase in LAS glass-ceramic. To prevent potential amorphization, the TEM sample was prepared by crushing a small piece of laboratory fabricated LAS glass-ceramic sample with ethanol using a pestle and mortar. A small drop of the fine suspension (after a brief ultrasonication) was pipetted onto a holey carbon coated Cu grid. TEM study was carried out using JEOL 2100FEG S/TEM equipped with Gatan Tridiem 863 EELS/GIF system. Electron tomography was performed using JEOL 2100 LaB6 TEM with a high tilt sample holder. Images for 3D reconstruction were obtained from 134 images (�67�) with one image per degree. Tomographic reconstruction and visualization were performed using TEMography software (System in Frontier, Inc.). TEM/STEM micrographs, BF/DF/HAADF as shown in Fig. 1, of the LAS glass-ceramic reveal nano-size crystals in glassy matrix. Nano-crystals of various morphologies were observed: rod-like (6~9 nm L x 2~3 nm W) (arrows in Figs. 1a and 2a); round/spherical (~6 nm in diameter, open arrows in Figs. 1a and 2b) and irregular shapes. The SAD pattern (Fig. 2c) with sharp and clear diffraction rings overlapping on a broad diffused band indicated poly-nanocrystallites "submerged" in amorphous matrix. The crystalline phase, verified by SAD and EELS (Fig. 2d), is -quartz LiAlSi2O6 (P6222). A threshold energy of ~8.5 eV with sharply increasing intensity to the maximum of ~24 eV was observed in the low-loss region (Fig. 2d). These correspond to the energy band gap and volume plasmon (collective excitation of valence electrons) of LAS glass-ceramics, respectively. The energy band gap of 8.5 eV indicates not only an electrical insulator, but also optical transparency similar to other SiO2-base glasses. HRTEM clearly revealed the lattice image of quartz LiAlSi2O6 nano-crystals in a glassy matrix (Figs. 2a and b). The -quartz LiAlSi2O6 phase, however, was destroyed quickly under electron beam irradiation, and further study of atomic arrangement became impossible. Estimated volume fraction of nano-size -quartz crystals in LAS was ~ 10% using TEM micrographs (total area of nano-size -quartz crystals/total observation area). TEM images are 2-D images projection from 3-D objects (Figs. 3 a and b). Thus, the volume fraction of objects in the image will be underestimated using TEM images. Using reconstructed 3D images via TEMography software, homogeneous distribution of nano -quartz crystals in LAS glass-ceramics was observed (Figs. 3 c and d). The rod-like, spherical and irregular morphologies were clearly shown in the 3D display. Assuming the morphology of nano -quartz crystals are spherical with ~10 nm average diameter, and the calculated total number of nano -quartz crystals is ~1500, a ~30% volume fraction of -quartz crystals in glassy matrix can be estimated (the total volume of nano -quartz crystals 1500 � ~500 nm3/the total reconstruction volume of 140nm � 175 nm � 115 nm is ~ 30%). Comparison of volume fraction of nano -quartz crystals in LAS using different approaches is in progress. Microsc. Microanal. 21 (Suppl 3), 2015 292 [1] E. D. Zanotto, Am. Ceram. Soc. Bull., 89, 19 (2010). [2] G. M�ller in "Low Thermal Expansion Glass Ceramics", ed. H Bach and D Krause, Springer (2005). [3] F. C. Serbena and E. D. Zanotto, J. Non-Crystal. Solid, 358, 975 (2012). [4] TEM work was partially supported by NSF-MRSEC (DMR 05-20471) and UMD. Fig. 1. TEM image (a), STEM BF (b) image and HAADF image (c) of LAS glass-ceramic shows rod-like (arrows) and round/spherical nanocrystals (open arrows). Fig. 2. HRTEM images depict rod-like (a) and round/spherical (b) nano -quartz crystals in glassy matrix. (c) SAD pattern obtained from area in Fig. 1(a). (d) EELS spectrum of LAS glass-ceramics. Fig. 3. A series of TEM images of a LAS glass-ceramic particle obtained by tilting around the long axis of the particle, e.g., image at -50� (a) and +50� (b). (c) 3D reconstruction image of nano -quartz crystals in LAS glass-ceramic. (d) Enlarged volume from the boxed area in (c).");sQ1[147]=new Array("../7337/0293.pdf","A New Method to Characterize Non-oxide Thin Film Uniformity at Device Level using Electron Energy Loss Spectroscopy","","293 doi:10.1017/S1431927615002263 Paper No. 0147 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A New Method to Characterize Non-oxide Thin Film Uniformity at Device Level using Electron Energy Loss Spectroscopy Zhi-Peng Li, Lifan Chen, Jianxin Fang, Haifeng Wang, Liang Hong, Xin Jiang, and Yuankai Zheng Western Digital Corporation, Magnetic Heads Operations, 44100 Osgood Road, Fremont, California, USA High quality thin films become critically important as the dimensions in hard disk magnetic recording heads and media continue to shrink down to nanometer/angstrom. It is increasingly challenging to characterize film properties at the device level using conventional methodology developed for large area full films, such as x-ray diffraction, during fabrication process due to limited spatial resolution and sensitivity. Transmission electron microscopy (TEM) with electron energy loss spectroscopy (EELS) has the unique advantage in providing angstrom level spatial resolution in location specific analysis. EELS technique has been well known for characterizing chemical bonding and valence state of transition metal oxide materials by studying the white line ratio and extended energy loss fine structure (EXELFS). This paper describes an empirical observation of crystalline uniformity through the correlation of EELS white line ratio and EXELFS. This approach demonstrates new EELS application to determine nonoxide thin film crystalline uniformity in magnetic reader structures. Iridium Manganese (IrMn) is a typical antiferromagnetic film used as reader element of the recording head and its crystalline orientation and texture have significant impact on the magnetic performance. Fig. 1 shows the STEM images of two types of IrMn films in complete magnetic reader structure for comparison. Other films above and below are the result of complete film deposition but not the focus of this study, therefore not labeled. The crystalline differences of the two IrMn films are not easily visible by images only, although nanodiffraction analysis (Fig. 2) clearly illustrates that there are more randomness in grain orientations in IrMn film 1 (Fig. 2a). TEM cross-sections of these two films are made with almost identical thickness. EELS spectra are collected with same conditions, e.g., beam current, dispersion, probe size, aperture size, pixel time and resolution, in order for the spectroscopic analysis to be comparable. Six different EELS are collected for each film at random locations, as shown in Fig. 3 and two insets. By normalizing all Mn L3 peaks, the large ratio variation in Mn L2 peaks can be easily visualized in IrMn thin film 1 (Fig. 3a, denoted as the arrow). The sharp Mn L2 and L3 peaks are known as the white lines in transition metal EELS spectra, arising from the core electrons excited into well-defined empty states [1]. Therefore, the white lines intensity ratio change can be used as a fingerprint to interpret the local element environment. However, in this study, there is no oxidation occurred in either film growth or sample preparation processes. The large Mn L2/L3 peaks ratio variation hence correlates well to the high degree of random orientation of IrMn grains of thin film 1. Moreover, since EXELFS is originated from those ejected electrons not filling the empty states but escaping outside the atom (different from those electrons of white lines) [2], a large variation in the EXELFS in IrMn thin film 1 (Fig. 3a, denoted as the dashed ellipse) is expected. The smaller EXELFS variation in Fig. 3b can be explained by more uniform grain orientations in IrMn thin film 2, consistent with nano-diffraction results. This new approach has been applied to many IrMn systems at various different times and all shows consistent results. We believe this new methodology has Microsc. Microanal. 21 (Suppl 3), 2015 294 potential to be an effective alternative to nano-diffraction, and its feasibility of applying to other materials and improved quantification are promising. References: [1] D B Williams and C B Carter, "Transmission Electron Microscopy", 2nd Edition (Springer, New York) p.744. [2] R F Egerton, "Electron Energy-Loss Spectroscopy in the Electron Microscope", 2nd Edition (Plenum Press, New York and London) p. 238. Figure 1. STEM dark field images of two magnetic thin film stacks. Figure 2. Nano-diffraction patterns of (a) IrMn thin film 1 showing more randomness, and (b) IrMn thin film 2 showing higher degree of texture. Figure 3. STEM EELS spectra comparison for (a) IrMn thin film 1 and (b) IrMn thin film 2.");sQ1[148]=new Array("../7337/0295.pdf","The Influence of pH Buffer as Agent Reaction Moderator in the Growth of CdS/ZnS Films by CBD Technique for Solar Cell Applications","","295 doi:10.1017/S1431927615002275 Paper No. 0148 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Influence of pH Buffer as Agent Reaction Moderator in the Growth of CdS/ZnS Films by CBD Technique for Solar Cell Applications F. V�zquez-Monroy1, A. Garc�a-Barrientos2, J.L. Bernal3, J. Plaza-Castillo4, H. Romero-Trejo2 and R. Ram�rez-Bon5 1. 2. Renewable Energy Department, Universidad Tecnol�gica Tula-Tepeji, Hidalgo, M�xico. Electronics Department, Universidad Aut�noma del Estado de Hidalgo, Hidalgo, M�xico. 3. Mechanics Department, Universidad Polit�cnica de Pachuca, Hidalgo, M�xico. 4. Physics Department, Universidad del Atl�ntico, Barranquilla, Colombia 5. CINVESTAV-IPN Unidad Quer�taro, Quer�taro, M�xico. Thin films of II�VI semiconductors are important for their applications in solid-state solar cells, optical coatings, optoelectronic devices, and light emitting diodes [1]. Cadmium sulphide (CdS) thin films have been extensively investigated as an n-type buffer layer to form thin film heterojunction solar cells with p-CdTe absorber layers. The buffer layer affects the electrical properties of the junction and protects it from chemical reactions. From the electronic point of view, the CdS layer optimizes the band alignment of the device [1,2] and builds a sufficiently wide depletion width that minimizes tunneling and establishes a higher contact potential that allows higher open circuit voltage [2]. Recently, particular attention of the research has been focused on the heterostructures involving CdS/ZnS multilayers [3�6]. Because of its band gap, it could be an excellent window layer in CdTe thin film solar cells. Since Chemical Bath Deposition (CBD) is known to produce solar cells over a large area at a low cost and low temperature, efforts have been made by many research groups over the world [5-7]. The effect of deposition parameters of CdS/ZnS thin films developed by CBD technique was investigated in this paper, principally, the influence of pH control of the reaction solution on the structural and optical properties of chemically deposited CdS/ZnS films. Different films thicknesses of CdS/ZnS were deposited onto a glass substrate. The structural surface morphology of as-deposited CdS/ZnS thin films was characterized by SEM, XRD, profilometer, and ultraviolet�visible spectroscopy. The physical conditions were kept identical while growing the samples. The investigation of the effect of the synthesis method on the change the ammonium hydroxide by buffer pH-11.4 contributed in increases the growth kinetics, resulting in thicker films. The CdS/ZnS thin films were fabricated by CBD technique on a glass substrate for different deposition times (15, 30, 45 and 60 minutes) at a bath temperature of 90 �C. The diffractogram of an as-deposited CBD-CdS/ZnS sample is shown in figure 1(a). The typical XRD pattern shows lines that correspond to the reflection mixture (002) of the greenockita (hexagonal) and (111) of the hawleyite (cubic) peak (1), showing in general that the preferential orientation of the film is along the (002) and (111) direction, other XRD pattern corresponds to the reflection of (220) of the hawleyite (cubic) peak (2), and (201) of the hawleyite (cubic) peak (3). We can also observe that these peaks are quite broad, which is indicative of nanosize particle. The transmittance at the high-energy region extends up to 300 nm as is shown in the figure 1(b). The SEM photos (see figure 1(c)) show the surfaces of CdS/ZnS films grown at 60 minutes deposition time and to different solution pH, A:11.4, B:11.8, C:12.6 and D:13, with thickness from 123, 102, 89 and 55 nm, and with the resistivity from 5.12, 4.18, 3.9, and 2.4 x 105 -cm, respectively. Based on the optical transmission measurements, the square of absorption coefficient (2) is plotted as a function of phonon energy (h) in figure 1 (d). At pH=11.4, Eg equals 2.74 eV and at pH=11.8, Eg equals 2.7 eV, these values are pretty similar of the literature [5]. Finally, these studies Microsc. Microanal. 21 (Suppl 3), 2015 296 show that the pH contributes noticeably to the growth and to the structure of deposited CdS/ZnS multilayer films. This may be interpreted by the decrease of the film thickness. From these studies, we are able to optimize the process in order to produce the layer suitable for optical window in solar cells. [1] Jongmin Kim, et al., Appl. Phys. Lett., vol. 102, no. 18, 183901, 2013. [2] M.A. Contreras, Thin Solid Films, vol. 204, 403-404 pp., 2002. [3] T. Ben Nasr, et al., Thin Solid Films, vol. 500, 4-8 pp., 2006. [4] A. Kariper, et al., Chalcogenide Letters, vol. 9, no. 1, 27-40 pp., 2012. [5] Isaiah O. Oladeji, et al., Thin Solid Films, vol. 474, 77-83 pp., 2005. [6] F. V�zquez-Monroy, et al., Microsc. Microanal., vol. 20, 1948-1949 pp., 2014 [7] Arreola-Jard�n G., et al., Thin Solid Films, vol. 519, 517-520 pp., 2010 The authors acknowledge funding from the CONACYT-Mexico, research projects grant numbers 169062 and 204419. (a) (b) (c) (d) Figure 1. X-ray spectrum of a typical CBD-CdS/ZnS sample (a). The transmittance at the high-energy region extends up to 300 nm (b). SEM photos of the samples surfaces of Cd/ZnS films grown at 60 minutes time deposition and to different solution pH, A:11.4, B:11.8, C:12.6 and D:13 (c). 2 versus h plot of CdS/ZnS films (d).");sQ1[149]=new Array("../7337/0297.pdf","Pressure Effect on the Deposition in the a-Si:H Films by PECVD Process for Solar Cell Applications","","297 doi:10.1017/S1431927615002287 Paper No. 0149 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Pressure Effect on the Deposition in the a-Si:H Films by PECVD Process for Solar Cell Applications J. Plaza-Castillo1, A. Garc�a-Barrientos2,3, M. Moreno-Moreno4, K.Y.Vizca�no1, J.A. Hoyo-Monta�o3 and G. Valencia-Palomo3 1. 2. Department of Physics, Universidad del Atl�ntico, Barranquilla, Colombia. Department of Electronics, Autonomous University of Hidalgo State, Hidalgo, M�xico. 3. Department of Electronics, Instituto Tecnol�gico de Hermosillo, Sonora, M�xico. 4. Department of Electronics, INAOE, Puebla, Mexico. The amorphous silicon (a-Si) is a material which has had a great acceptance in the microelectronic industry field due to its low cost in comparison with the one of crystalline silicon (c�Si). This material has a random network in its atomic structure, since its atoms are not located to either a specific angles or distance. In 1969, Chittik et al. [1] added hydrogen to the amorphous silicon finding a beneficial effect, since it saturated the defects of the network. This finding was key for the development of the amorphous semiconductors. Thus, W. Spear and P. LeComber [2] showed that silicon has semiconducting properties when together with a dopant gas such as phosphine and diborane. The hydrogenated amorphous silicon (a-Si:H) appears as a promising material in the photovoltaic industry due to its high absorption coefficient and low manufacturing cost [3,4]. Therefore, the optical and electrical properties of a-Si:H films, such as transmittance, absorption coefficient, conductivity, activation energy and thickness are very important. These properties can be optimized by the deposition process parameters, such as power, frequency mode, argon flow rate, temperature and principally the pressure deposition [57]. This parameter has influence in the transmittance, absorption coefficient and conductivity because is proportional to deposition rate and stress (compressive). In this work, the a-Si:H films were fabricated by the Plasma Enhance Chemical Vapor Deposition (PECVD) process at low frequency with a substrate temperature of 300 oC, varying the flow of hydrogen and dopant gases. In this way, implementing the PECVD technique, thin films have been doped with PH3 (n-type) and with B2H6 (p-type). The procedure was repeated with different values of flow of PH3 and H2 for the n-type films and B2H6 and H2 for ptype ones. To investigate the effects of pressure on the deposition in the a-Si:H films, all experimental parameters for various samples were kept constant except for the deposition pressure. The values of RF power, substrate temperature, and deposition time were 10 W/cm2, 300�C, and 30 minutes, respectively. On the other hand, the deposition pressure was varied to from 725 to 2500 mTorr in order to investigate the effects of this parameter on the films structure. In each experiment, the films were deposited both on glass as well as silicon substrates. The characterization of samples and the evaluation of the process were done by the measurements of the absorption coefficients, the conductivities, the activation energies and of the thickness of the films. In the Figures 1(a) and (b), we show the AFM images for the n-type and p-type, respectively, with different deposition pressure values. For both cases, for low deposition pressure values the nanoclusters do not appear, however these appear when the deposition pressure increase, this study reports the analysis of Si nanoparticles of approximately 1.5 nm in size, see Figure 1(f). A graph of absorption coefficient of aSi:H layer as a function of wavelength is shown in Figure 1(c) and (d), for n and p type, respectively. Note how the absorption coefficient of the a-Si:H drops near its band gap. In this sense, the Figure 1(e) presents the band gap as a function of deposition pressure, we can see the same band gap value, in both type films, for deposition pressure at 2000 mTorr. And due to high absorption coefficient, only 1 �m Microsc. Microanal. 21 (Suppl 3), 2015 298 thick a-Si:H layer is required to absorb photons of energy higher that the band gap energy. This study suggest to use the deposition pressure from 1500 to 2500 mTorr to obtain a-Si:H films with high performance for solar cell applications. [1] R. C. Chittick, et al., J. Electrochem. Soc., vol. 116, no. 77, 1969. [2] W. Spear et al., Solid State Comm., vol.17, no. 9, 1193 pp., 1975. [3] Jun Ma, et al., Solar Energy Materials and Solar Cells, vol. 123, 228�232 pp., 2014. [4] K. Ding, et al., Solar Energy Materials and Solar Cells, vol. 95, 3318-3327 pp., 2011. [5] V. P. Oleshko, et al., Microsc. Microanal., vol. 12, Suppl. S02, 634-635 pp., 2006. [6] J. Woerdenweber., et al., Solar Energy Materials and Solar Cells, vol. 95, no. 10, 2011. [7] Jeong C., et al., J. Nanosci. Nanotechnol., vol. 7, No. 11, 4169-73 pp., 2007. The authors acknowledge funding from the CONACYT-Mexico, grant numbers 169062 and 204419. (a) (b) (c) (d) (e) (f) Figure 1. AFM image for the n-type (a) and p-type (b) a-Si:H films. Absorption coefficient of a-Si:H ntype (c) and p-type (d) as a function of wavelength. The band gap as a function of deposition pressure for a-Si:H films (e). SEM photo of the sample at 2000 mTorr, where nano-crystals appear (f).");sQ1[150]=new Array("../7337/0299.pdf","Direct Evidence of Basal-Bonded MoS2 Cluster on Al2O3 Support","","299 doi:10.1017/S1431927615002299 Paper No. 0150 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Direct Evidence of Basal-Bonded MoS2 Cluster on Al2O3 Support C. Angeles-Chavez, J. A. Toledo-Antonio and M. A. Cortes-Jacome Instituto Mexicano del Petroleo, Gerencia de Desarrollo de Materiales y Productos Qu�micos, Eje Central Lazaro Cardenas Norte 152, C.P. 07730 M�xico, Distrito Federal, MEXICO Transmission electron microscopy (TEM) plays an important role in the characterization of heterogeneous catalysts, in particular on the MoS2-based hydrodesulphurization (HDS) catalysts. Information about their structure, chemical composition, crystallite size and dispersion on the support is required in the new formulations to improve the catalytic performance in the HDS reactions. MoS2-based/Al2O3 HDS catalysts have been widely studied by TEM. The results reported have shown that the MoS2 clusters are bonded to the Al2O3 support surface in two different morphologies and orientations [1,2]. These have been identified as basal-bonded MoS2 layers and edge-bonded MoS2 layers. However, many dispersion studies of the MoS2 phase on Al2O3 support only have taken into account the edge-bonded layers than the basalbonded layers because these are easier to observe by high-resolution transmission electron microscopy (HR-TEM). Therefore, the information obtained is still incomplete and conventional HR-TEM cannot give information about basal-bonded if it is thinner than the support. The development of new detectors as the annular dark field detector in the scanning transmission electron microscopy (STEM) can help to give this information. In this work, we present evidence of both morphology and orientations. Samples of MoS2/Al2O3 were prepared by conventional impregnation methods using Al2O3 and Mo and P salts. After impregnation, the samples were sulfide with H2S/H2. Powders were dispersed in isopropanol and an aliquot was deposited on a copper grid with a film Lacey/carbon. The characterization was performed in a transmission electron microscope JEM2200FS. TEM results at low magnification showed the typical morphology of the gamma-alumina, wrinkled nanosheets, see figure 1. At higher magnification can be appreciated in the HRTEM image of figure 2a, the edge-bonded MoS2 layers showing their different lengths and stacking number. There is not clear evidence of basal-bonded MoS2 layer clusters. This can be due to the thickness of the sample in the analyzed region. Figure 2b shows a thin area of the support. Chemical analysis performed from in this area were detected the Mo, S and P characteristic peaks, in addition to the peaks corresponding to the Al2O3. The detection of Mo, S and P in this region strongly suggest the presence of the MoS2 phase on the Al2O3. However, these were not observed a different focus. Only some lattice line was revealed by the change of focus. This result suggests that the thickness of the support is bigger than the basal-bonded MoS2 layers, absence of edge-bonded layers in this region or the Mo, S and P sintered with the support. In order to give light to these questions, a STEM study in the sample was performed and annular dark field images were obtained. The intensity observed in the image is related to the atomic number. Heavier elements produce brighter contrast than the lighter elements. The annular dark field images show the edge-bonded MoS2 layers Microsc. Microanal. 21 (Suppl 3), 2015 300 and polymorph plates of basal-bonded MoS2. Both cluster types are brighter in the image. HR-STEM of a basal-bonded MoS2 cluster shows the hexagonal arrangement of the Mo atoms in the [0001] basal plane. Therefore, a direct evidence of the basal-bonded MoS2 structure on the Al2O3 support was obtained with a STEM analysis. References: [1] Y. Sakashita and T. Yoneda, J. catal.,185, 1999, 487-495. [2] H. Shimada, Catal. Today, 86, 2003, 17-29. [3] The authors acknowledge financial support to IMP through project D.00447. 50 nm Figure 1. Bright field TEM image showing the wrinkled nanosheet of the gamma-alumina. 5 nm 2 nm (a) (b) Figure 2. HR-TEM images showing a) edge-based MoS2 structures on the Al2O3 surface and b) some lattice lines in a thin region of the Al2O3 sheet.");sQ1[151]=new Array("../7337/0301.pdf","Observations on Orientation Relationships between Rutile and Brookite","","301 doi:10.1017/S1431927615002305 Paper No. 0151 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Observations on Orientation Relationships between Rutile and Brookite M. Josefina Arellano-Jimenez1, Matthew T. Janish2, Weyshla A. Rodriguez-Rodriquez2, Bonnie B. McKenzie3, Joseph R. Michael3 and C. Barry Carter2,4 1 2 3 Department of Physics and Astronomy. UTSA. One UTSA Circle, San Antonio, TX 78249 Department of Materials Science and Engineering, University of Connecticut Sandia National Laboratories, Albuquerque, NM 87185-1405 4 Department of Chemical and Biomolecular Engineering, University of Connecticut TiO2 exists in nature in three different forms, namely rutile, anatase and brookite; it can also be prepared in the laboratory in each of these forms although the first two are by far the most common [1]. The transformation of anatase to rutile is well documented but that for brookite to rutile is not, in part because brookite is not often observed. It is common to find brookite and anatase together, where anatase is the major phase. These two phases could be easily taken for just one due to very similar diffraction patterns when analyzed by XRD. Anatase and brookite coexist at a consistent fraction until 600 �C, after which the fraction of brookite will decrease. Above 1000 �C both phases completely transform into rutile. In this context, understanding the mechanisms of the transformation process, the effect of impurities, and their relationship with the crystal structure in TiO2 polymorphs becomes relevant for new developments. Several sections of one naturally-grown single crystal of TiO2 were analyzed. Cross sections of the sample were cut and polished for SEM analysis. The materials have been examined in a Zeiss SEM equipped with EBSD and a JEOL TEM2010F equipment with a NanoMegas system. Figure 1 shows a region of a natural crystal of Brookite that has partially transformed to Rutile. A small grain of SiO2 has been trapped at the interface between the two materials. Interesting features in this image include the presence of a large number of pores in one phase and the different contrast in recorded by the backscatter detector even though both materials are TiO2. The orientation relationships have been examined using both EBSD and precession diffraction. References: [1] C.B. Carter and M.G. Norton, Ceramic Materials: Science and Engineering, 2�d ed., Springer 2013. [2] MJA-J acknowledges UTSA for use of the Microscopy Facility. MTJ and WAR-R acknowledge the Department of Education's GAANN fellowship program for research support and the Institute of Materials Science at the University of Connecticut for access to the JEOL 6335F FESEM and the FEI Strata 400S DualBeam FESEM FIB. Erica Pehmoeller, Jessica Bogart and Alice Kilgo are thanked for their assistance in specimen preparation. BBK and JRM acknowledge Sandia National Laboratories for support and access to facilities; Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-AC04- 94AL85000. Microsc. Microanal. 21 (Suppl 3), 2015 302 Figure 1. Natural crystal of brookite. Partial transformation from brookite to rutile and presence of SiO2 are observed. Some interesting features such as differences in porosity and BSE contrast between rutile and brookite polymorphs are also present.");sQ1[152]=new Array("../7337/0303.pdf","The Use of Electron Microscopy Techniques to Determine the Oxidation Mechanism SiAlON Ceramics Containing Yb203+Sm203+CaO as Sintering Additives","","303 doi:10.1017/S1431927615002317 Paper No. 0152 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Use of Electron Microscopy Techniques to Determine the Oxidation Mechanism SiAlON Ceramics Containing Yb203+Sm203+CaO as Sintering Additives Sinem Baskut, Servet Turan Faculty of Engineering and Architecture, Department of Materials Science and Engineering, Anadolu University, Eskisehir, Turkey SiAlON ceramics are used in advanced technology areas including mechanic, chemistry, metallurgy, optic, automotive and defense industry which requires particularly high temperature resistance due to their excellent physical, chemical and mechanical properties. However, SiAlON ceramics are thermodynamically instable in an oxidizing environment due to the sintering additives used to improve densification, fracture toughness and hardness [1]. The aim of this study is to investigate the oxidation mechanism of "Yb203+Sm203+CaO" doped SiAlON ceramics at high temperatures by using microscopy techniques. For this purpose, samples were sintered in gas pressure sintering furnace. After sintering, the samples were polished and exposed to oxidation in high temperature furnace at 1400 �C for 72 hours. Then, the backscatter electron images, EDX analysis (Oxford Instruments, INCA ENERGY) and EBSD analysis (Oxford Instruments, INCA HKL) were taken at scanning electron microscope (Zeiss, SUPRA 50 VP) from both the surface and cross sections of oxidized samples without coating in variable pressure mode. XRD result not given here indicated that Yb2Si2O7 (keivite) and crystoballite were formed at the surface of oxidized sample. Also, oxide layer formation was detected by the help of backscatter electron image which is given in Figure 1 (a). In the oxide layer newly formed rod like, light grey phase and dark grey phase was found as a mixture. To examine the formation of the oxide layer and the microstructural modifications occur in the main material during the oxidation, the backscatter electron image was taken from cross section of the sample. As shown in the Figure 1 (e) three different regions formed in the main material during oxidation. The region A in main material is not affected by oxidation. Modifications occurred in region B by the help of diffusion of sintering additives through grain boundaries to the surface. Region C is the most affected zone from oxidation and includes silica layer which is located in the interface of SiAlON and oxide layer [2]. Backscatter cross section image (Figure 1 (c)) shows that rod like light grey crystals dispersed at the surface whereas black spherical shaped crystals located at oxide-nitride interface. EDX analysis given in Figure 1 (f) were obtained from three different phases observed in the oxide layer. When EDX analysis were jointly evaluated with XRD results it can be deduced that rod like light grey crystals (1) are Yb2Si2O7, spherical like black crystals (2) are crystoballite and dark grey phase (3) is dopant rich glassy phase. Crystal orientation maps obtained from the surface and cross section of oxidized sample are shown in Figure 1 (b) and (d). In the orientation maps Yb2Si2O7 crystals are distributed in black color glassy phase. Also, color differences in orientation maps showed that the Yb2Si2O7 crystals were not grown in one direction by controlled manner. Microsc. Microanal. 21 (Suppl 3), 2015 304 References: [1] J. Yu, H. Du, R. Shuba, I. Chen, `Dopant-dependent oxidation behavior of -SiAlON ceramics', Journal of Material Science, 39 (2004), 4855-4860 [2] L.O. Nordberg, M. Nygren, P.O. Kall, Z. Shen, `Stability and Oxidation Properties of RE�SiAlON Ceramics (RE = Y, Nd, Sm, Yb)', Journal of Material Science, 81 (1998) 1461�70 a) b) c) d) e) f) 1 C B 2 A 3 Figure 1. Back scattered images (a, c, e) and EBSD orientation maps (b, d) are given along with EDX analysis (f) from different regions on (e).");sQ1[153]=new Array("../7337/0305.pdf","Microstructural Characteristics of (Na0.5K0.5)NbO3 Ceramics with Additives: Transmission Electron Microscopy Study","","305 doi:10.1017/S1431927615002329 Paper No. 0153 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microstructural Characteristics of (Na0.5K0.5)NbO3 Ceramics with Additives: Transmission Electron Microscopy Study Young Heon Kim1,2, Hyun Ryu1, Hwack-Joo Lee1, Yang-Koo Cho1, Sahn Nahm3, and Sang Jung Ahn1,2 1. Korea Research Institute of Standards and Science, 267 Gajeong-Ro, Yuseong-Gu, Daejeon 305-340, Republic of Korea. 2. University of Science & Technology, 217 Gajeong-Ro, Yuseong-Gu, Daejeon 305-350, Republic of Korea. 3. Department of Materials Science and Engineering, Korea University, 1-5 Ka, Anam-Dong, SungbukKu, Seoul 136-701, Rep. of Korea. Na0.5K0.5NbO3 (NKN) has attracted much attention as an alternative to Pb(Zr1-xTix)O3 (PZT) ceramics and lead-free piezoelectric materials because of its high piezoelectric properties and a high Curie temperature (Tc) [1, 2]. However, the high sintering temperature for NKN is an obstacle for realizing multilayer devices because they require low driving force, miniaturization, and hybridization. Specifically, the NKN ceramics have to be sintered at around 900 C because the melting point of silver (Ag), commonly used as an electrode in multilayer devices, is 961 C and the Na2O evaporates during the sintering process at temperatures above 1000 C [3, 4]. One of the solutions proposed to these problems is to add a small amount of ceramic compounds whose melting points are lower than 900 C as the sintering additives and to promote the low temperature sintering via liquid phase. A few research groups have reported that the NKN ceramics with additives, e.g. CuO and ZnO, had been successfully sintered at around 900 C without any degradation of the piezoelectric properties [5-7]. They suggested that the interaction of ZnO and CuO would make the formation of the liquid phase, thus enhance the densifications of the matrix by the liquid phase sintering. We have investigated the microstructures in NKN ceramics with additives (CuO, ZnO and the mixture) with transmission electron microscopy (TEM). As a new microstructural constituent, different types of pockets have been observed at the grain boundaries of NKN and also inside the NKN matrix. The characteristics of pockets were investigated from a microstructural point of view using various TEM techniques. The analysis of chemical elements in the pocket was carried out with energy dispersive spectroscopy (EDS) and electron energy loss spectroscopy (EELS) analysis. We found a CuO pocket as a new microstructural constituent when the CuO was introduced as an additive [8]. When the same methodologies were employed in order to understand the behaviors of the ZnO and CuO pockets when CuO and ZnO were added to the NKN ceramics (NKNCZ), two types of pockets, composed of CuO and ZnO as a dominant component, were observed in the microstructure as new microstructure constituents (Fig. 1). When the additive content of ZnO was increased to 3.0 mol%, there are interactions between CuO and ZnO and both elements are found in the compound pocket. The sintering kinetics was enhanced by the presence of both additives. In Figs. 2(b) and 2(c), the phase present in the pocket was identified as ZnO [hexagonal, P63mc (186), a = 3.250 � , c = 5.207 � ; JCPDS number 36-1451], and the -- diffraction pattern was indexed along the zone axis of [2110]. The pockets were melted partially or completely by the interactions with element Na in the matrix which has formed a eutectic compound whose melting point is lower than the sintering temperature. The reaction starts at the interfaces between the pocket and matrix and the kinetics depends not only on the size of the pocket but also on the Microsc. Microanal. 21 (Suppl 3), 2015 306 environments where the pockets are located [9]. References: [1] G. Shirane, R. Newnham, and R. Pepinsky: Phys. Rev. 96 (1954) 581. [2] J. R�del, W. Jo, K.T.P. Seifert, E-M. Anton, T. Granzow, and D. Damjanovic: J. Am. Ceram. Soc. 92 (2009) 1153. [3] H. Y. Park, C. W. Ahn, H. C. Song, J. H. Lee, S. Nahm, K. Uchino, H. G. Lee, and H. J. Lee: Appl. Phys. Lett. 89 (2006) 062906. [4] Y. Zhen and J.-F. Li: J. Am. Ceram. Soc. 89 [12] (2006) 3669. [5] H.-Y. Park, J. Y. Choi, M. K. Choi, K.- H. Cho, S. Nahm, H.-G. Lee, and H.-W. Kang: J. Am. Ceram. Soc. 91 (2008) 2374. [6] E. Li, H. Kakemoto, S. Wada, and T. Tsurumi: J. Am. Ceram. Soc. 90[6] (2007) 1787. [7] A.C. Caballero, J.F. Fern�ndez, C. Moure, P. Dur�n, and Y.M. Chiang: J. Am.Ceram. Soc. 81 (1998) 939. [8] Y.H. Kim, H. Ryu, H.J. Lee, Y.K. Cho, and S. Nahm: Jpn. J. Appl. Phys. 51 (2012) 035602. [9] Y.H. Kim, H. Ryu, Y.K. Cho, H.J. Lee, and S. Nahm: Jpn. J. Appl. Phys. 52 (2013) 031501. [10] This research was supported by Nano-Material Technology Development Program (NextGeneration Nano Fundamental Technology Development Program) through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (grant number 2011-0030233). Figure 1. (a) and (c) BFTEM images NKNCZ ceramic sintered at 920 C for one hour with 1.5 mol% CuO and 3.0 mol% ZnO showing the compound pocket. (b) and (d) Local elemental mappings by the EDS analysis, Cu-K and Zn-K, in the liquid pockets in (a) and (c). Figure 2. (a) BFTEM image taken under multi-beam diffraction condition. (b) HRTEM image of the pocket. (c) Selected area electron diffraction (SAED) pattern taken from the pocket. (d) Local elemental mappings by the EDS analysis, Zn-K.");sQ1[154]=new Array("../7337/0307.pdf","Synthesis of Mesopores of Zirconia by Using CTAC as Template","","307 doi:10.1017/S1431927615002330 Paper No. 0154 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Synthesis of Mesopores of Zirconia by Using CTAC as Template A. Medina-Flores1a, E. Borjas-Garc�a1b, Brenda Quezadas1c, L. B�jar1d, C. Aguilar2, J. L. Bernal3. Instituto de Investigaciones Metal�rgicas, 1b Instituto de F�sica y Matem�ticas, 1c Facultad de Ciencias F�sico y Matem�ticas, 1d Facultad de Ingenier�a Mec�nica, Universidad Michoacana de San Nicol�s de Hidalgo, Morelia, Michoac�n, M�xico, C.P. 58000 2 Depertamento de Ingenier�a Metal�rgica y Materiales. Universidad T�cnica Federico Santa Mar�a. Av. Espa�a 1680, Valpara�so, Chile. 3 Automotive Mechanics Department. Universidad Polit�cnica de Pachuca. Zempoala, Hidalgo. M�xico A mesoporous material is a material, which contains pores in a range between 2 and 50 nanometers [1]. This type of material can be used as catalyst for bulky reactant molecules due their large pore size. The first publication about synthesis of mesoporous materials by using a surfactant template was in 1992 by Mobil scientists [2]. After that, several methods have been developed for the synthesis of mesoporous metal oxides [1]. However, the synthesis of mesoporous zirconia by using Cetyltrimetylammonium chloride (CTAC) as template has not investigated. The mesopores of zirconia was prepared by using zirconium oxide chloride octahydrate (Sigma-Aldrich, purity 99.5%), tetramethylammonium hydroxide solution TMAOH (Sigma-Aldrich, 25 wt% in water) and cetyltrimethylammonium chloride solution, CTAC (Aldrich, 25 wt% in water) as source. In a first step, 1.611 g of ZrOCl2*8H2O was dissolved in 7.2 g of destilled water. Then, 3.2 g of CTAC solution was added to the Zr-solution. For the second step, 10.94 g of TMAOH solution was added slowly (drop by drop) to Zr-surfactant solution and stirred. The final suspension was stirred and heat in a hot stir plate at 90 �C to get a material with a molar ratio of ZrOCl2*8H2O:CTAB:TMAOH:H2O equal to 1:0.5:6:75. The gel obtained was aged in a polypropylene bottle at 80 �C for 1 day. After that, the hydrothermal treatment, the sample was washed with distilled water, centrifuged and dried at 80 �C for 1 day. The powder X-ray diffraction patterns were collected with a siemens D5000 X-ray diffractometer equipped with graphite monochromatized high�intensity CuK (=1.54178 �). The Bragg angle 2 ranges from 1� to 8� at a scanning rate of 0.02�/s. The surface morphology images of the samples were analyzed by using a scanning electron microscopy FEG-SEM JEOL JSM 7600. Figure 1 shows the XRD pattern of zirconia synthesized. This figure shows a characteristic peak about the presence of mesopores in the material around 1.04 degrees. Figure 2a shows an image of mesoporous zirconia. Figure 2b shows the EDS spectrum of the mesopores. The results shows that the optimal hydrothermal treatment temperature for mesoporous zirconia synthesis is at 80 �C because more temperature of treatment could collapse mesopores formation in zirconia and it was developed an easy procedure for the synthesis, which can be, applied at different materials. References [1] D.W. Bruce, et al., Volume 1 (2010), p. 1. [2] C. T. Kresge et al, Nature Volume 359 (1992) p. 710. [3] The authors acknowledge funding from H. Consejo T�cnico of Institute of Physics and Mathematics, and Consejo Nacional de Ciencia y Tecnolog�a (CONACyT), M�xico. 1a Microsc. Microanal. 21 (Suppl 3), 2015 308 Figure 1. XRD pattern of mesoporous zirconia Figure 2. a) SEM image of mesoporous zirconia, b) EDS spectrum of the mesoporous of zirconia.");sQ1[155]=new Array("../7337/0309.pdf","Effect Zirconia's Phases in the Production of Hydroxyapatite","","309 doi:10.1017/S1431927615002342 Paper No. 0155 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Effect Zirconia's Phases in the Production of Hydroxyapatite Kabir A. Jurado Villalobos, Armando Reyes Rojas, Mar�a del Carmen Arag�n Duarte, Hilda E. Esparza-Ponce* Departamento de F�sica, Centro de Investigaci�n en Materiales Avanzados S.C. (CIMAV), Miguel de Cervantes 120, Complejo Industrial, C.P. 31109, Chihuahua, M�xico. The different tested biomaterials as potential bone grafts which include: ceramics, polymers, metals, composites and hybrid materials. The low bonding strength of plasma sprayed hydroxyapatite (HA) coating has been a point of potential weakness in its application as biomedical prosthesis. There is a continuous interest on controlling the mechanical properties [1]. Zirconia is a good biomaterial, has good mechanical properties and can be used as coating on alloys and titanium for production of apatite. The present work was focused on the evaluation of addition effect of monoclinic, cubic and zirconia partial stabilized phases as thin film in the formation of apatite. The Thin films were deposited by sputtering RF on titanium and stainless steel substrates. The surface of thin films was immersed in Simulated Body Fluid (SBF) The SBF was prepared mixing sodium chloride, sodium bicarbonate, potassium chloride, calcium chloride, dibasic potassium phosphate, and magnesium chloride in deionized water [2-3]. The microstructural analyses of the samples were carried out by AFM and SEM. The crystalline state and the identification of phases were verified by X-ray diffraction (XRD) analysis using a diffractometer PAnalytical. Diffraction patterns were acquired from 20 to 100� 2 using a Scanning speed of 0.2�/min with KCu radiation and grazing incidence mode. Figure 1 show the typical growth of cubic thin films on stainless steel and titanium. Figure No.2 shows the thin film morphology by AFM a) monoclinic ,c) PSZ and e)cubic phase onto stainless steel and figure 2 b) monoclinic, d) PSZ and f) cubic phase on Titanium. The XRD data indicates the crystallization of apatite on all phases which was identified by the JCPDS card as 00-086-1201.Figure 3 and 4 shown the morphology of hydroxyapatite by SEM on thin films with monoclinic, PSZ and cubic phase onto stainless steel and Titanium after one and 21 days of immersion, respectively. Finally, although in all phases hydroxyapatite is formed in monoclinic phase has a further development and preferably on titanium substrates. References: [1] K.A. Khora,*, L. Fua, V.J.P. Lima, P. Cheang Materials Science and Engineering A276 (2000) 160 �166. [2] Kokubo T, Takadama H. Biomaterials. 2006 May;27(15):2907-15. Epub 2006 Jan 31. [3] A. B. Mart�nez-Valencia, G. Carbajal-De la Torre, A. Duarte Moller, H. E. Esparza-Ponce and M. A. Espinosa-Medina, International Journal of the Physical Sciences Vol. 6(29), pp. 66816691, 16 November, 2011. The authors acknowledge funding from the Mexican Government (CONACyT) through project No.24463 CB-2005-01-51478. Microsc. Microanal. 21 (Suppl 3), 2015 310 Figure 1. Thin films deposited by sputtering RF onto a) stainless steel and b) Titanium. Figure 2. Morphology by AFM of thin films as deposited by sputtering RF a) monoclinic ,c)PSZ and e)cubic phase on stainless steel and b) monoclinic, d) PSZ and f) cubic phase on Titanium. Figure 3. Morphology of hydroxyapatite by SEM growth on thin films a) monoclinic, c) PSZ and e)cubic phase on stainless steel and b) monoclinic, d) PSZ and f) cubic phase on Titanium formed after one day of immersion. Figure 4. Morphology of hydroxyapatite by SEM growth on thin films a) monoclinic, c) PSZ and e)cubic phase on stainless steel and b) monoclinic, d) PSZ and f) cubic phase on Titanium formed after 21 days of immersion.");sQ1[156]=new Array("../7337/0311.pdf","Closing �Loops Producing Added-Value Products as a Cost-Reduction Strategy in the Operation of Biorefineries","","311 doi:10.1017/S1431927615002354 Paper No. 0156 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Closing �Loops Producing Added-Value Products as a Cost-Reduction Strategy in the Operation of Biorefineries William Kerstein1, Elena Ten2, Luisa Dempere1,3, and Wilfred Vermerris4 1 Materials Science and Engineering, University of Florida, Gainesville, FL 32611 2 Agronomy Department, University of Florida, Gainesville, FL 32610 3 Major Analytical Instrumentation Center, University of Florida, Gainesville, FL 32611 4 Microbiology Department, University of Florida, Gainesville, FL 32610 Multiple studies are being conducted with the purpose of exploring the use of waste residues generated during the bioethanol processing of sugar cane bagasse as raw materials for addedvalue products, primarily Lignin. Lignin is next to cellulose the most abundant biopolymer on earth. Lignin coats cellulose micro-fibrils and other components in the cell wall preventing collapse and providing support to the cell walls in biomass. Being a renewable energy source, biomass is a common feedstock for biopolymers, paper and biofuel production. Lignin left over from the production of biofuels is a significant portion of biomass (up to 25%), and it is often burned to generate heat for the distillation of alcoholic fuels from the fermentation liquor. With increasing demand of sustainable energy sources, we are exploring the recovery and modification of Lignin to reduce costs of biofuel production. In this study we worked with pre-treated lignin provided by the University of Florida Ethanol Pilot Plant and The Stan Mayfield Biorefinery. Lignin is isolated from biomass using a two-step hydrolysis in phosphoric acid followed by enzymatic hydrolysis that removes polysaccharides. This process solubilizes approximately eighty to ninety percent of the hemicellulose, leaving the cellulose and lignin largely intact as polymers. Our group produces lignin-based nanotubes using a sacrificial template of alumina membranes. Isolated Lignin is covalently linked to the inner walls of activated alumina membranes and then layers of dehydrogenation polymer are added onto this base layer via a peroxidase-catalyzed reaction. The thickness of the polymer layer deposited within the pores is selected by using phenolic monomers displaying different reactivity. Thus, we synthesize nanotubes with a wall thickness of approximately 15 nm and nanowires with a nominal diameter of 200 nm. These novel nanostructures are flexible and can be bio-functionalized easily and specifically. Unfortunately, the cost of the alumina membranes is the limiting step in the upscaling of the synthesis of lignin nanotubes. Thus, the investigation of the generation of secondary products from the membranes processing waste was pursued as a strategy to reduce costs. Closed loops systems are based in the concept of "waste equals food" where by-products or waste of a specific process are used as raw or initial materials for other products. The waste solution generated in the dissolution of the membranes to release the nanotubes was investigated as a precursor mixture for the synthesis of Zeolites than can be used in some of the catalyzed stages of the bioethanol production. Figures 1 and 2 show the images of the synthesized Zeolites. References: [1] Robinson, H. (2011). AFI AlPO4-5. Verified syntheses of zeolithic materials. Elsevier. Microsc. Microanal. 21 (Suppl 3), 2015 312 Fig. 1 Secondary electron image of Zeolite crystals; magnification 3,000X. Fig. 2 Secondary electron image of Zeolite crystals; magnification 200X.");sQ1[157]=new Array("../7337/0313.pdf","TEM / HREM observations of thin foils obtained by SEM-FIB for the study of the phase transitions in chromium slag vitrification","","313 doi:10.1017/S1431927615002366 Paper No. 0157 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 TEM / HREM observations of thin foils obtained by SEM-FIB for the study of the phase transitions in chromium slag vitrification S. Ballesteros1 and J. Ma. Rinc�n2 1. Centro de Ense�anza T�cnica y Superior, CETYS Universidad Campus Tijuana. Av. CETYS Universidad No.4 Fracc. El Lago, B.C., M�xico, C.P. 22550 2. Lab/Grupo de Materiales V�treos y Cer�micos, IETcc, CSIC, c/Serrano Galvache 4, Madrid-28033, Spain A glass - ceramic obtained by controlled vitrification of slag with high levels of Cr+6 has been investigated under TEM transmission and analyzed in HREM (image high resolution mode). The partially crystalized areas were identified and other regions fully devitrified by nucleation and controlled crystalline growth. The mineralogical composition of the considered phases have shown, that these materials after their respective processing, contain pyroxenes in augite form, and crystallization of (Mg, Fe) (FeAlCr)2O4 spinels embedded into a residual vitreous phase that are produced from the composition of the vitrification within the region of MgO-CaO-Al2O3-SiO2 composition. A sample has been selected according to its optimum characteristics from physical, mechanic and microstructural point of view, in order to elucidate in detail the vitrification aspects that have not been able to be detected with SEM/EDS[1]. The preparation of the samples were held at a double beam high resolution electronic scanning microscope (SEM/ FIB) Nova-200 Nanolab with a magnifying capacity from 30X up to 1 200 000X and with a resolution until 1.1 nm., which has a Focalized Ion Beam column (FIB) that allows selection and separation by ionic thinning of interest area within a range from micrometric to nanometric dimensions. Simultaneously, it is possible to perform the polishing (thinning of the sample) in a controlled manner until a thickness of less than 100 nm [2-4]. For the observations with TEM and HREM Philips equipment -TECNAI TYPE CM-30 operating at an acceleration voltage of 300 kV used with a Field Emission filament. This TEM / HREM has also an EDS (Energy Dispersive Spectrometer) with an ultra-fine window that allows the light detection (light weight) elements. The high resolution images (HREM) obtention, has been achieved using a contrast objective with an aperture of 20 m, which corresponds point by point to a resolution of about 0.2 nm. In the figure 1a has shown a dispersion that looks like inmiscibity drops with diferent sizes and contrast of precipitate drops which are between the range of 5-20 nm diameter[5]. Now, when such "nanodrops" are observed in greater magnification, it is prove that they have some orderings, depicting atomic planes. Therefore, "nanodrops" are dispersed in the residual vitreous matrix that correspond to either nuclei or nanocristal particles (Figure 1b). The figure 2a shows areas without a visible order, that should correspond to residual glass (amorphous). In the figure 2b it can be observed a pseudo-rectangular crystal with 250 nm side, fringer image and a 8 nm separation between them (8nm would correspond to the projection of a spinel cubic crystal). We can also observe in the lower part of the micrography, contrast lines caused by the stress fields that this type of crystaline phase generates in the residual vitreus phase. The figure 2c (higher magnification), clearly indicates the interphase between the disordered (vitreous) and the crystaline AB2O4 (Mg, Fe) (FeAlCr)2O4 spinel area "decorated" in the edge by high contrast Microsc. Microanal. 21 (Suppl 3), 2015 314 chromium atoms. In the figure 2d depicts loops of contrast emerging from corners due to the high stress fields produced between the spinel crystal and the residual glassy phase. References [1] Ballesteros S., Parga J.R., Delgado A, Cano J. and Rinc�n J. Ma. Electron microscopy study of vitrified materials from chromite contaminated soil and confinated mud waste with chrome III & VI". XXIV Congress of the Spanish Microscopy Society XLIV annual meeting of the portuguese society for microscopy.16 -19th July 2009 Segovia, Spain Pp. 289 -290. [2] Ballesteros S., Parga J.R. y Rinc�n J. Ma. "Transmission Electron Microscopy (TEM) throughout Focussed Ion Beam (FIB) from vitrified chromium wastes". Journal Advanced Research Technology (JART). Ed.Universidad Aut�noma de M�xico (UNAM). Vol. 9, No. 2 August 2011. Pp. 242 -248. [3] Lucille A. Giannuzzi, RemcoGeurts and Jan Rignalda 2KeV Ga+ FIB "Milling for reducting Amorphous Damage in Silicon", Microsc. and Microanal. Vol.11 (Suppl S02 2005, 828 -829 [4] Ballesteros S and Rinc�n J. Ma. "Microstructure and microanalysis of porous glass obtained by sintering of vitreous powders from cr-ni sludge. aug. 2011, Proceedings of Microscopy and Microanalysis 2011, Microscopy Society of America. 7 -11 August. Nashville Tennessee USA. Pp. 1904-1905. [5] Rinc�n J.Ma. y Dur�n A. "Separaci�n de Fases en Vidrios" BSECV, 11 (1972) 2, Ed. SECV 1982 en pp.111-125. [6] J.Ma.Rinc�n and M.Romero. " Characterization Techniques of Glasses and Ceramics" Ed. Springer Verlag, 64 -80., 1999 Figure 1. Micrographs obtained by TEM and HREM in a glass � ceramic from Cr+6 slag waste after thinning by SEM-FIB. Figure 2. TEM/HREM observations in glass � ceramic obtained from a high chromium contrast metallurgical slag.");sQ1[158]=new Array("../7337/0315.pdf","An Improved TEM Sample Preparation Technique for Heavily Deformed Ceramic Materials having Extensive Sub-Surface Cracking","","315 315 doi:10.1017/S1431927615002378 Paper No. 0158 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 An Improved TEM Sample Preparation Technique for Heavily Deformed Ceramic Materials having Extensive Sub-Surface Cracking J. J. Swab1, C. V. Weiss Brennan1, S. D. Walck2, J. C. LaSalvia1, and K. D. Behler3 1. U.S. Army Research Laboratory, RDRL-WMM-E, Aberdeen Proving Ground, MD 21005, USA TKC Global, RDRL-WMM-C, Aberdeen Proving Ground, MD 21005, USA 3. TKC Global, RDRL-WMM-E, Aberdeen Proving Ground, MD 21005, USA 2. Understanding the deformation mechanisms in ceramic materials is crucial for optimizing and implementing next-generation ceramic materials in body and vehicle armor systems. Transmission electron microscopy (TEM) is a powerful tool for studying the deformation mechanisms of impacted materials. Indentation allows the study of deformation of microstructural features as a function of distance and depth from the indent. In the original development of the TEM sample preparation technique for observing the structure under indents in SiC used for ceramic armor, Brennan et al. [1] used a multi-step process to preserve the heavily cracked regions under the indent by Tripod polishing, FIB, and vacuum impregnation. Figure 1 outlines the steps of the process developed in that study. In solving the sample preparation problem of heavily cracked ceramic materials, the researchers found several results that are critical for success: 1) vacuum infiltration of a low viscosity epoxy on an exposed cross sectional surface is absolutely required, 2) vacuum infiltration from the top of the surface will not infiltrate the crack structure below the surface but does preserve any spallation from the indentation, 3) mechanical polishing, no matter how fine a grit size is used, closes the paths for the vacuum infiltration and ion milling is required to expose the cracks for infiltration, 4) the FIB lift-out technique cannot be used on the as-indented surface primarily because of re-deposition of amorphous material into exposed cracks, and 5) the epoxy penetration from one surface can be as great as 8 to 10 �m. To minimize ion milling times in the FIB, Brennan et al. aligned the long axis of the Knoop indents along the line of the indents of H-bar samples thinned by Tripod polishing. However, this is not the most favorable orientation to compare the microstructure to fracture mechanics models. Because of the time required to bring the very hard ceramic materials to a thickness of ~50 �m and the two FIB thinning steps, this multi-step technique was a very labor intensive and costly process. The goal of this work was to develop an efficient and less demanding sample preparation technique for heavily cracked ceramics that preserves the true microstructure for TEM imaging and analysis. Silicon carbide (SiC) and boron carbide (B4C) were indented using a Knoop indenter with a Tukon 2100B hardness tester. Samples were cut and polished, and then indented with a linear array of constant load indentations. Similar to the previous study, the array of indents were aligned to a surface with care in order to minimize the distance of material to be removed but not affect the indentation fracture mechanics. In the new method, Tripod polishing was eliminated by using a Leica TIC-3X ion milling machine to create the cross sections below the surface by masking approximately half of the indents along the array. After milling, the sample was vacuum infiltrated as in the earlier study. Instead of the final sample having the H-bar configuration, our new procedure uses an in-situ lift-out in the FIB. By removing epoxy on either side of the indent, the surface of the cross section formed by the TIC-3X can be located. This allows the final TEM sample to be made ~2 �m from the TIC-3X prepared surface. Figure 2 shows a low magnification image of a B4C sample with cross section of four indents visible after milling with the Leica TIC-3X, but prior to infiltration. The insert shows a higher magnification of the indent labeled "2". Figure 3 shows an SEM image of a vacuum infiltrated sample tilted to an angle Microsc. Microanal. 21 (Suppl 3), 2015 Microsc. Microanal. 21 (Suppl 3), 2015 316 316 of 45� with one FIB lift-out sample with the long axis of the Knoop indents oriented perpendicular to the cross sectional plane. Viewing at 45� allows the indent on the surface to be seen even with the epoxy present. Figure 1. A schematic of the original sample prep process, including mechanical polishing (a), ion milling to open up the cracks (b), epoxy infiltration (c), and final ion polishing (d). o Figure 2. Composite SEM image showing the cross section of indented B4C prepared by Leica TIC-3X prior to vacuum infiltration. Indents are indicated. The inset shows a higher magnification image of "2". A B Figure 2. A) 45� tilted SEM image of TIC-3X prepared sample after infiltration and first FIB lift-out area prepared. B) SEM image of indent after epoxy on the cross section surface was removed. [1] C.V Brennan, S.D Walck, and J. J. Swab, Microscopy and Microanalysis 20(6) (2014), p. 1646-53.");sQ1[159]=new Array("../7337/0317.pdf","Microtomy on Heat-Treated Electro-Spun TiO2 Fibers","","317 doi:10.1017/S143192761500238X Paper No. 0159 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microtomy on Heat-Treated Electro-Spun TiO2 Fibers M. Josefina Arellano-Jim�nez1, Matthew Janish2, Paul Kotula3, Weyshla Rodriguez4, and C. Barry Carter2,3-5 1 2 Dept of Physics and Astronomy. UTSA. One UTSA Circle, San Antonio, TX 78249 Dept of Chemical & Biomolecular Engng, U. of Connecticut, 191 Auditorium Rd, Storrs, CT 06269 3 Sandia National Laboratory, Materials Characterization Dept, MS 0886, Albuquerque, NM 87185 4 Dept of Materials Science & Engineering, U. of Connecticut, 191 Auditorium Rd, Storrs, CT 06269 5 Institute of Materials Science, U. of Connecticut, 97 North Eagleville Road, Storrs, CT 06269 The use of microtomy for sample preparation has been widely applied to the study of soft tissue in the biological area, but has been also successfully applied in materials science. The implementation of microtomy has been a valuable tool to prepare samples for the study of many different materials such as polymers, carbon fibers, ceramics and metallic alloys [1]. It is a widely used method to prepare polymers or composite materials. Samples prepared by this method can be analyzed using several techniques including VLM, SEM and TEM, which allows a more extensive characterization of the material. The fabrication of ceramic with nanometric dimensions has been identified as a promising approach to enhance their catalytic activity or selectivity. Several methods have been implemented to prepare ceramic materials with specific structures because the effect of size and morphology on the properties of materials is fundamental when seeking for new applications [1]. The possibility of producing nanotubes or nanowires composed entirely of TiO2 has been studied because of its many applications in such diverse fields as dye-sensitized fuel cells and catalysts. Electrospinning has been used in recent years for synthesizing nanofibers of both organic and composite materials; membranes fabricated by this method exhibit high surface area and small pore size. The shape and distribution of the ceramic component within the fiber is a critical factor in the behavior of the final material. It then becomes important to have a suitable method to produce samples where the distribution and orientation of materials can be analyzed using different techniques. The use of microtome sections allows the analysis of the bulk structure in large samples. As shown in the present study, the microtome can provide excellent results in producing samples from electrospun ceramic fibers. The process involves the spinning of a polymer solution, or polymer-precursor, to obtain fibers. Final oxide fibers result after heat treatment of the resulting mat. In this work, polymer and polymerTiO2 precursor were used to study the morphology and composition of TiO2 nanofibers. TiO2/polymer composite and TiO2 nanofibers have been prepared using an approach reported before [2]. A vertical set-up was used for the electrospinning process in which a syringe pump was used to direct a controlled flow of the solution through a capillary tube to a stainless-steel needle that was connected to a high-voltage supply. For this study, the nanofibers were collected on an aluminum foil wrapped around a rotating drum. The drum was electrically grounded so as to generate an electric field between the tip of the needle and the foil, which were separated by a distance of ~17 cm. Mats of TiO2/polymer composite nanofibers were collected on aluminum foil. The mats were allowed to stand overnight under ambient conditions. Part of the mat was removed from the foil and heat treated on a hot plate. Samples were embedded and ultramicrotomed for TEM observation. This procedure allowed the observation of cross-section and several slices of the fiber. Figure 1b shows a low-mag image where several sections of fibers are exposed. The TEM images in Figure 1 show the geometrical shape of the TiO2 nanostructures. After heat treatment, fibers show a circular cross section, in the range of 50 Microsc. Microanal. 21 (Suppl 3), 2015 318 to 150 nm. Differences in contrast suggest the fibers are polycrystalline with variations in particle size. Selected 80-nm-thick sections were used for the TEM analysis. This combination of microtomy and microscopy techniques allowed the examination of the same area in the micro- and nano-scale. The STEM image in Figure 1 shows the characteristic morphology of the fibers after heat treatment. The material appears as a mat formed by cylindrical fibers. The TEM image in Figure 1 clearly shows the geometrical shape of the TiO2 nanostructures. The synthesized fibers have a uniform width in the range of 50 to 150 nm. The contrast observed in the images confirms that the material is crystalline; the DP corresponds to polycrystalline anatase [3]. References: [1] Carter, C.B. & Norton, M.G., 2013, Ceramic Materials: Science & Engineering 2nd Ed (Springer, NY). [2] A. Suresh, M. J. Arellano-Jim�nez, J. McCutcheon and C. B. Carter. `Polycrystalline TiO2 Produced by Electrospinning' Microscopy and Microanalysis 2012 19 (suppl. 2) 1392-3 [3] Part of this work was performed at Sandia National Laboratory in the Materials Characterization Department. Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the US Department of Energy under contract DEAC04-94AL85000. The authors acknowledge useful discussions with Nathan Martin. Fig. 1. TEM images of materials produced by electrospinning. Microtomed sections allow the observation of fibers in all different directions such as cross and longitudinal sections. TiO2 nanofibers are approximately 50-150 nm wide. c) SADP confirms the material is polycrystalline anatase. Scale bar 500 nm and 200 nm.");sQ1[160]=new Array("../7337/0319.pdf","From Flat Membranes to Fabrication of Hollow Spun Fibers of Polyethersulfone/ Polyvinylpyrillidone (PES/PVP) - Correlative Scanning Electron Microscopy (FESEM), 3-D X ray Microscopy and Ultrafiltration Properties","","319 doi:10.1017/S1431927615002391 Paper No. 0160 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 From Flat Membranes to Fabrication of Hollow Spun Fibers of Polyethersulfone/ Polyvinylpyrillidone (PES/PVP) - Correlative Scanning Electron Microscopy (FESEM), 3-D X ray Microscopy and Ultrafiltration Properties Pooja Bajaj1, Albin Berzinis1, Rachel Halbfinger, Carl Strom1, Lars Peters2, Matthias Wessling2, Naomi Kotwal3 Sabic, 1 Noryl Avenue, Selkirk, New York 12158 USA, 2 RWTH Aachen University, Aachen, Germany, 3 Carl Zeiss X-ray Microscopy Inc. USA In a previous report, we presented the application of low voltage field emission scanning electron microscopy to characterize the structure and morphology of lab scale flat membranes cast using Polyethersulfone/polyvinylpyrrolidone (PES/PVP) (Figure 1a, b) [1 and all references within]. Morphology was insightful in tuning the compositional variables and optimizing the design space for translation into hollow spun fibers. In this work, morphological comparison of flat membranes with the hollow spun fibers and their performance properties will be presented. For (PES/PVP) based hollow fiber spinning, the dope solution containing PES (BASF E6020P) at 14 weight% and polyvinylpyrrolidone (PVP) K90 at 2 weight %, PVP K30 at 5 weight %, were blended into a homogenous solution. The blending of low and high K-value PVP is shown to be important in controlling the surface pore structure and tendency towards macro void formation using phase inversion process from n-methylpyrrolidone (NMP) into water [2]. Fiber spinning was performed by dry-wet immersion precipitation using a bore solution of 70 wt% deionized water and 30 wt% NMP [2]. Dope solution along with the bore liquid were simultaneously pumped through a double orifice spinneret and after passing the air gap, immersed into the water coagulation bath. Figure 1c visualizes the fiber spinning process. The fibers were post treated by washing in 70�C pure water for 3h and air dried. Some fibers were immersed for 24 into a mixture of water/glycerol (80wt%/ 20wt %) prior to the drying step. Lab scale hollow fiber membrane modules were prepared from these as spun and dried, and glycerol post treated and dried fibers and tested for the clean water flux and molecular weight cut off measurements. Non-destructive 3-D pore distribution was visualized using Carl Zeiss Xradia 510 Versa and Xradia 810 Ultra X ray microscopy using Zernike phase contrast mechanism to benefit for the otherwise low attenuating membrane material [3]. FESEM characterization of hollow spun fibers of PES/PVP reveal complimentary morphology as seen for flat membranes. Figure 1d shows a low magnification cross section morphology of a hollow fiber membrane with an open, interconnected pore structure highlighting a dense nanoporous skin layer, nano-to-microporous sublayer and macro voids which penetrate the outer membrane wall (Figure1e). Interestingly, the dense inner membrane surface shows pore collapse on as-dried fibers (Figure 1f) that could be reversed when fibers were immersed in glycerol/water prior to drying (Figure 1g). The high resolution images highlight the effect of glycerol in revealing the open nano channels on the selective layer, oriented in the fiber draw direction (Figure 1g). This surface morphology findings of the hollow fibers provide visual understanding and correlation on failing fibers (as spun and dried) compared to functional performance of fibers (glycerol treated and dried) as monitored using clean water flux and molecular weight cut off experiments (Table 1). X ray microscopy (XRM) reveals complimentary pore distribution in 3-D as seen for the cross sectional morphology for flat membranes (figure 2 a, b, c, d). The cross section was visualized at 65nm voxel resolutions using Xradia 510 Versa (Figure 2 e, g, h). The top dense layer was then scanned at high resolution of 32nm voxel to resolve less than 100nm features (Figure 2 f).In summary, combining complimentary techniques opens the opportunity to not 1 Microsc. Microanal. 21 (Suppl 3), 2015 320 only understand the composition variables, but also understand the spinning process attributes that can be tweaked towards robust material characterization and promote new product developments. Future work is focused on computing quantitative metrics based on 3D (e.g. porosity, pore size distribution, etc.) and combining for 4D dynamic (in-situ) studies. References: [1] P. Bajaj et al, Microsc. Microanal. 20 (2014), p. 2100. [2] L. Gambro et al, (2013), PCT/EP2013/054263 WO2013/131848A1 [3] A. P. Merkle and J. Gelb, Microsc. Today, 21 (2013), p. 10. (a) Tank with polymer solution T=35�C Bore fluid 1,56 3,10ml/min ml/min gear pump spinneret 40 cm (c) (d) (e) Air gap pulling wheel speed 9,12 m/min humidity 50% 100 �m T=65�C Rinsing bath Phase inversion by solvent/non-solvent exchange Coagulation bath 20 �m (b) T=30�C 2 �m (f) (g) 100 nm 100 nm 100 nm Figure 1. (a) SEM cross sectional and (b) surface morphology of flat membranes of PES/PVP. (c) Fiber spinning process to generate hollow membrane fibers of PES/PVP. (d), (e) SEM cross-sectional and (f) surface morphology of as spun hollow fiber membrane of PES/PVP (g) role of additive, glycerol on opening the porosity of the inner selective layer of the PES/PVP hollow fiber. Table 1. Performance properties of hollow spun fibers of PES/PVP, as spun and dried and after immersion in glycerol before drying. Hollow Fiber PES/PVP as-spun PES/PVP with glycerol before drying (a) (b) CWF [L/(h*m�*bar)] 9 ~500 (e) (f) MWCO (Dextrans as probe) ~ 30 - 40 kDa (much breakage) ~ 50 kDa (c) (d) (g) (h) Figure 2. (a) X ray microscopy (XRM) based 3-D viewer. (b), (c), (d) virtual cross sections taken at the orthogonal planes XY, YZ and XZ. (e),(g), (h) High resolution 3D rendered volume that can be combined for further segmentation, thresholding, binarization and quantitative analysis for 3-D porosity and interconnectedness of pore channels at 65 nm voxel resolution using XRM Versa 510 and (f) 32 nm voxel resolution using XRM Ultra 810.");sQ1[161]=new Array("../7337/0321.pdf","Surface Coating Effect on Si Nanowires Anodes for Lithium Ion Batteries","","321 doi:10.1017/S1431927615002408 Paper No. 0161 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Surface Coating Effect on Si Nanowires Anodes for Lithium Ion Batteries Langli Luo1, Pengfei Yan1, Ji-Guang Zhang2, Chunmei Ban3 and Chongmin Wang1 1 Environmental Molecular Sciences Laboratory, Pacific Northwest National Laboratory, 902 Battelle Boulevard, Richland, WA 99352, USA 2 Energy and Environmental Directorate, Pacific Northwest National Laboratory, 902 Battelle Boulevard, Richland, WA 99352, USA 3 National Renewable Energy Laboratory, 1617 Cole Boulevard, Golden, CO, 80401, USA Silicon based materials holds promise for next generation negative electrode for high-capacity Li-ion batteries, yet tremendous research effort have been made for tackling the chemo-mechanical failure that is associated with the intrinsic large volume change of Si during lithiation and delithiation process. Surface modification of Si nanostructures have successfully alleviated this problem and improved the cyclability of Si based anodes. Successful surface modifications are expected to provide both good protection and conduction between Si nanostructures without sacrificing the electrochemical performance. However, how these surface modifications will affect the lithiation and delithiation behavior of the Si nanostructures and whey they improve or deteriorate the performance of electrodes are largely unknown. The emerging in-situ transmission electron microscopy (TEM) techniques with localized electrical measurement capabilities provide a practical platform for investigating electrochemical reactions in Liion battery materials by building a full or half "nano-cell" inside the TEM specimen chamber [2]. Such real-time observations of dynamic composition and microstructural evolution in the electrochemical reaction have provided many novel clues to understand the lithiation/de-lithiation mechanisms at nano or atomic-scale for several novel anode materials. Herein, we report a comparative study of surface coatings effect on the lithiation and delithiation behavior and kinetics of Si nanowires and nanoparticles. Fig. 1A and B show time-resolved TEM images in top and bottom panel depict the typical lithiation process for alucone and Al2O3 coated Si nanowires with same diameter, respectively. The alucone coated Si nanowire has a clear V shaped lithiation profile comparing with symmetrical linear lithiation file of Al2O3 coated Si nanowire as shown in the corresponding schematic drawings. Al2O3 coating with a dense structure but poor Li ion conductivity leads to a symmetrical lithiation profile along the axial direction of the Si nanowires and a slower lithiation rate, while alucone coating with a less dense structure and fast Li ion transport yields to a typical V shape lithiation profile along the axial direction of the Si nanowire and high rate lithiation. The lithiation depth vs. time for two type of Si nanowires is illustrated in Fig. 1C that a much faster initial lithiation rate is found for alucone coated Si nanowire. Fig. 2 illustrates the risk for pulverization of Si nanostructures or damage to functional surface coatings through vigorous alloying reaction. Upon biasing, the no.4 nanowire lithiated with two distinct manners on upper and lower part of the nanowire. The upper part connected to no. 2 nanowire with a larger diameter alloyed aggressively to form amorphous Li xSi causing pulverization of half of the nanowire and Al 2O3 coating is also destroyed. On the contrary, the lower part of the nanowire connected to no. 3 with a smaller diameter Si was lithiated smoothly and the surface coating was preserved during lithiation. This variation Microsc. Microanal. 21 (Suppl 3), 2015 322 in lithiation behavior in one Si nanowire is largely attributed to the diffusion paths of Li ions. Once the reaction barrier is reached, large chemical potential drives the alloying reaction fast enough to induce large volume expansion that cannot be accommodated by the materials system leading to pulverization. Reference: [1] Arico, A. S., Bruce, P., Scrosati, B., Tarascon, J.-M. & van Schalkwijk, W. Nat Mater 4 (2005), 366. [2] Huang, J. Y. Zhong, L., Wang, C. M. et al. Science 330 (2010), 1515. Figure 1 A) and B) Time-resolved TEM images show the development of lithiation profiles of alucone and Al 2O3 coated Si nanowires and schematic Li diffusion paths through the Si NWs which determines the lithiation behavior and kinetics; C) Average thickness of lithiated vs. time for alucone (black square) and Al 2O3 (red dot) coated Si nanowires. Figure 2. (A-C) Time sequential TEM images of failure of Al2O3 coating on Si nanowire during lithiation process.");sQ1[162]=new Array("../7337/0323.pdf","EELS Investigations of Aging Mechanisms in LiFePO4 Cathodes Resulting From Prolonged Electrochemical Cycling","","323 doi:10.1017/S143192761500241X Paper No. 0162 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 EELS Investigations of Aging Mechanisms in LiFePO4 Cathodes Resulting From Prolonged Electrochemical Cycling Samartha Channagiri1, Frank Scheltens1, Nicholas Warner2, Marcello Canova2, Yann Guezennec2 and David W. McComb1 1 Center for Electron Microscopy and Analysis, Department of Material Science and Engineering, The Ohio State University, Columbus, OH, U.S.A 2 Center for Automotive Research, The Ohio State University, Columbus, OH, U.S.A Lithium iron phosphate (LiFePO4) has considerable potential as a cathode in batteries for automotive applications due to its high rate capability, reasonable energy density and environmentally benign nature [1]. However, performance degradation seen after thousands of cycles at high charging-rates (C-Rates) has been a point of major concern [2]. Studies of the aging mechanism suggest that phases (LiFePO4/FePO4) formed in the cathode during discharge influence the aging profile [3]. Previously, we used electron energy loss spectroscopy (EELS) to demonstrate the use of Li-K edge for identifying lithium in the sample with a potential for quantification [4]. This required a modified procedure for focused ion beam (FIB) milling to minimize ion beam damage during sample preparation [5]. We reduced beam dosage in the electron microscope to prevent knock-on damage to the lithium in the sample. Lithium content, combined with information of the oxidation state of iron, can be used to identify the phases formed upon intercalation. The study was able to identify fine variations in the Li-K edge structure, and hence, phase composition in nanoparticles inside the LiFePO4 composite electrode. We have produced a series of aged cells and performed the above mentioned EELS analysis on cathode samples extracted from unaged and aged cells. Aged cells were produced for this analysis utilizing capacity retention, cycling temperature and battery size factor (B.S.F) as metrics. For a given B.S.F and temperature, the battery was cycled until it retained 80% or 90% of its capacity. The charging profile used for this purpose was the charge depletion profile prescribed by the USABC for Plug-In Hybrid Electric Vehicles (PHEV's) [6]. A total of eight samples were aged (Table 1). The cells were unwrapped and cathode material extracted at specific sites for further analysis. Electron transparent thin foils were prepared from these materials by FIB milling [5]. We then performed EELS in the low energy-loss regime (E<90eV) at an energy resolution of 1eV and dispersion of 0.1eV /channel in an FEI Tecnai TF20 on both unaged and aged (aged at B.S.F = 1, T = 35�C upto 90% capacity retention) samples. The EELS results obtained at a nanoparticle level are shown in fig.1, with corresponding lower magnification STEM�HAADF images of the microstructures of the respective samples. Both unaged and aged cells have similar microstructures with a large particle size distribution making it challenging to identify directly structural variations introduced by aging. However, EELS performed in the low loss region exhibit subtle variation in the Li K-edge (fig.1) energyloss near-edge structure (ELNES). In particular, the intensity and separation of the peaks in ELNES varies within a given sample and between samples. The Fe M2,3-edge peak (fig.1) retains its position on the energy axis in all cases, with subtle variations in the shape of its pre-peak in both samples. The correlation between EELS and microstructure is providing new insights into the complex aging mechanism in these electrodes that is in part responsible for the capacity loss with extended cycling. Microsc. Microanal. 21 (Suppl 3), 2015 324 References: [1] Padhi, A. K. Phospho-olivines as Positive-Electrode Materials for Rechargeable Lithium Batteries. J. Electrochem. Soc. 144, (1997)1188 - 1193 [2] Vetter, J. et al. Ageing mechanisms in lithium-ion batteries. J. Power Sources 147, (2005) 269 - 281 [3] Channagiri, S. A. et al. Porosity and phase fraction evolution with aging in lithium iron phosphate battery cathodes J. Power Sources 243, (2013) 750 - 757 [4] Channagiri, S. A. et al. Spatially resolved characterization of phases in LiFePO4 battery cathodes using low loss electron energy-loss spectroscopy. Microscopy & Microanalysis 20, S3 (2014) 432 - 433 [5] Gilchrist, J.B. et al. Uncovering buried structures and interfaces in molecular photovoltaics Adv. Func. Mat. 24, (2014) 6473 � 6483 [6] U.S. Department of energy battery test manual for Plug-In Hybrid Vehicles, March 2008 [7] This work is supported by the Department of Energy, project: "U.S.-China Clean Energy Research Center for Clean Vehicles (CERC-CV)" under Award Number DE-PI0000012 Sample Temperature (�C) 1 2 3 4 5 6 7 8 55�C 55�C 55�C 55�C 35�C 35�C 35�C 35�C Capacity retention 90% 90% 80% 80% 90% 90% 80% 80% Battery size factor (B.S.F) 1 2 1 2 1 2 1 2 Table 1: Table of aged cells produced for microstructure analysis. Cell highlighted in yellow is used as `aged' cell in the current EELS analysis. Fig.1: STEM-HAADF microstructure images (top) and EELS spectra (bottom) of unaged cell (left) and aged cells (right).");sQ1[163]=new Array("../7337/0325.pdf","Contrasting Reaction Modality between Electrochemical Sodiation and Lithiation in NiO Conversion Electrode Materials","","325 doi:10.1017/S1431927615002421 Paper No. 0163 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Contrasting Reaction Modality between Electrochemical Sodiation and Lithiation in NiO Conversion Electrode Materials Kai He1, Feng Lin2, Eric A. Stach1, Yifei Mo3, Huolin L. Xin1 and Dong Su1 1. 2. Center for Functional Nanomaterials, Brookhaven National Laboratory, Upton, NY 11973 Environmental Energy Technologies Division, Lawrence Berkeley National Laboratory, Berkeley, CA, 94720 3. Department of Materials Science and Engineering, University of Maryland, College Park, MD, 20742 The diverse requirements of energy storage devices in different applications call for advancements not only in lithium ion chemistry but also in new battery chemistries. Specifically, renewed interest in sodium ion batteries arises from their potential to provide a sustainable and low-cost solution for largescale energy storage such as electric vehicles and power grids [1]. Although the lithium ion battery electrodes can possibly adapt to the sodium ion technology, the dissimilarities between lithium and sodium characteristics may strongly affect the electrochemical processes and the overall battery performance [2, 3]. Prior studies on nickel oxide (NiO) as a conversion electrode material for lithium ion batteries have uncovered the lithiation mechanism through the heterogeneous surface-to-bulk reaction pathways [4]. Therefore, identifying the dynamic sodiation process in the same electrode material and comparing it with lithiation to understand the underlying reaction mechanism is of fundamental importance. Here, we present the real-time observation of the sodiation process in NiO nanosheets using in-situ transmission electron microscopy (TEM). The in-situ nano-battery was set up in a dry cell configuration within TEM (Figure 1a) and further correlated to ex-situ electron and synchrotron X-ray spectroscopies of coin cells. We have integrated the in situ TEM imaging, diffraction, spectroscopy, and tomography (Figure 1a�e), to draw a full picture of the entire sodiation reaction at all relevant length scales. The structural and morphological evolutions during both sodiation and lithiation were in-situ tracked, as shown in Figure 1a. In contrast to the heterogeneous lithiation, in which the lithium conversion starts from the surface of pristine NiO and penetrates into the interior redox sites via the lithiation "fingers", the sodiation proceeds conformally from surface redox sites toward the inner bulk, but still leaving most of interior region unreacted after 960s. The reduced Ni nanoparticles precipitate from the NiO lattice and coherently attach on the surface, as revealed by the aberration-corrected scanning transmission electron microscopy (STEM) image in Figure 1b. The phase transformation of NiO + Na Ni + Na2O has been confirmed by the in-situ electron diffraction (Figure 1c), while the degree of the phase conversion is only in a limited amount. The electron energy-loss spectroscopy (EELS) mapping of Ni valance states (Figure 1d) indicates that only surface redox sites were reacted with sodium to form Ni nanoparticles, which can also be justified by three-dimensional visualization using electron tomography (Figure 1e). All these microscopic results consistently show a surface-dominant sodiation reaction modality distinct from the lithiation, which is also reproduced and explained by the molecular dynamics calculation. In summary, the fundamental understanding obtained by this in-situ methodology has contrasted the difference in reaction mechanisms between sodium and lithium chemistries and may provide valuable implications to future advances of sodium ion batteries [5]. Microsc. Microanal. 21 (Suppl 3), 2015 326 References: [1] M D Slater et al, Adv. Func. Mater. 23 (2013) 947. [2] Y Wen, K He et al, Nature Commun. 5, (2014) 4403. [3] K He et al, ACS Nano 8, (2014) 7251. [4] K He et al, Nano Lett. 15, (2015) 1437. [5] The authors acknowledge funding support from U.S. Department of Energy, Office of Basic Energy Sciences, under Contract Number DE-AC02-98CH10886 and DE-SC-00112704. Figure 1. (a) Schematics of in-situ dry cell setup and time-sequenced TEM/STEM image series showing structural evolution during in-situ sodiation and lithiation, respectively. (b) High-resolution STEM image of sodiated sample showing Ni nanoparticles coherently attached on the surface of NiO. The FFT showing the shrinkage of Ni lattice as reduced from NiO. (c) In-situ electron diffraction patterns before and after sodiation. (d) STEM-EELS mapping of Ni valance states and (e) 3D electron tomography showing reduced metallic Ni nanoparticles on the surface of NiO.");sQ1[164]=new Array("../7337/0327.pdf","STEM-HAADF Imaging Study of Spinel Cubic Li4Ti5O12 Nanocrystals","","327 doi:10.1017/S1431927615002433 Paper No. 0164 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 STEM-HAADF Imaging Study of Spinel Cubic Li4Ti5O12 Nanocrystals Xiangcheng Sun1, Kai Sun2, and Bo Cui1 1 2 Department of Electrical and Computer Engineering, University of Waterloo, Ontario, Canada Department of Materials Science and Engineering, University of Michigan, Ann Arbor, MI, USA Spinel cubic Li4Ti5O12 (LTO) has attracted great attention as a novel anode material for high performance Li-ion battery [1]. The LTO anode can provide a theoretical capacity of 175 mAh/g with negligible volume change during charging and discharging [1]. However, the high-current rate performance of pure LTO anode is seriously hindered by its low electrical conductivity [2]. Carboncoating has been approved to be an effective method to improve the electrical conductivity of LTO anode material, carbon-coated LTO nanocrystals (C-LTO) have been investigated [3, 4]. It is critically important to understand the influence of the carbon-coating effects on high-rate performance [5]. In this study, the C-LTO anode nanocrystal with optimized thin coating layers were successfully synthesized. In the present work, crystalline phase and coating structure were characterized by TEM imaging and high resolution TEM imaging using a JEOL 2010F FEG transmission electron microscopy. Atomic resolution structure of high-angle-annular dark-field (HAADF)-STEM images were obtained with collection semi-angles from 50 to 180 mrad in spherical aberration-corrected (Cs-corrected) scanning transmission electron microscopy of JEOL 2100F STEM/TEM operated at 200 kV. TEM images and HRTEM images in Figure 1 (a) confirmed that the C-LTO nanocrystal have a welldefined spinel nanocrystal structure, and exhibited the spherical particle shape with amorphous carboncoating and uniform thickness of less than 5 nm. Figure 1 (b) revealed that the well-resolved lattice fringes have an interplanar distance of 0.48 nm, corresponding to the d-spacing of the facets of the spinel LTO structure. Cs-corrected STEM technique allows us to obtain atomic-resolution HAADF images and count the number of atoms in an atomic column of a crystal [6]. The typical high-resolution STEM-HAADF images at the atomic scale are showed in Figure 2 (a). The titanium atom sites (16d), lithium atom sites (16d) and oxygen atom site (32e) can be clearly visualized in the enlarge atomic resolution HAADF image in Figure 2 (b), which is consistent with the atomic occupancies in schematic lattice view of spinel LTO along the [110] direction, as showed in LTO [110] projection model in Figure 4 (c). Bscially, such like the [110] projection of spinel LTO is most suitable for observing Li, O, and Ti atoms directly, because separate columns of these atoms are aligned in this [110] direction. It is worth noting that the pure spinel LTO phase was synthesized and further confirmed. The improved high-current rate performance can be ascribed to high phase purity and the enhanced intrinsic electronic conductivity resulting from the synergistic effect of nanocrystals, carbon-coating and uniform thin thickness. References: [1] T. Ohzuku, etal, J. Electrochem. Soc., 142 (1995) p.1431. [2] H.Q. Li, H. S. Zhou, Chem. Comm., 48 (2012) p.1201. [3] Z. Q. Zhu etal, J. Mater. Chem. A, 1 (2013) p.9484. [4] H.Q. Li, and H. S. Zhou, Chem. Comm., 48 (2012) p.1201. [5] H.G. Jung etal, Energy Environ. Sci., 4 (2011) p.1345. [6] X, Lu etal, Adv. Mater., 24 (2012) p.3233. Microsc. Microanal. 21 (Suppl 3), 2015 328 Figure 1. TEM images (a) and HRTEM images (b) of the C-LTO nanocrystal. The well-crystallized structure and an amorphous carbon layer covering (~4.5 nm) is clearly observed. Figure 2. (a) High-resolution STEM-HAADF images at atomic scale, (b) The enlarged HAADF panel showed atomic structure of LTO crystal plane, (c) Schematic lattice view of spinel LTO along the [110] direction, corresponding to the 16d, 32e, and 8a sites in the LTO atomic lattice.");sQ1[165]=new Array("../7337/0329.pdf","Dynamic Observation of Tunnel-driven Lithiation Process in Single Crystalline -MnO2 Nanowires","","329 doi:10.1017/S1431927615002445 Paper No. 0165 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Dynamic Observation of Tunnel-driven Lithiation Process in Single Crystalline MnO2 Nanowires Yifei Yuan1, Anmin Nie2, Gregory Odegard2, Kun He1,3, Dehua Zhou4, Sunand Santhanagopalan5, Dennis Desheng Meng5, Robert Klie6, Christopher Johnson4, Reza Shahbazian-Yassar2 Department of Materials Science and Engineering, Michigan Technological University, 1400 Townsend Dive, Houghton, Michigan 49931, United States 2. Department of Mechanical Engineering-Engineering Mechanics, Michigan Technological University, 1400 Townsend Dive, Houghton, Michigan 49931, United States 3. Department of Materials Science and Engineering, Shandong University, 17923, Jingshi Road, China. 4. Chemical Science and Engineering Division, Argonne National Laboratory, 9700 S. Cass Avenue, Argonne, Illinois 60439, United States 5. Department of Mechanical and Aerospace Engineering, University of Texas at Arlington, 500 West 1st Street, Arlington, Texas 76019, United States 6. Department of Physics, University of Illinois at Chicago, 845 West Taylor Street, Chicago, IL 60607, United State Manganese dioxide (MnO2) possess various allotropic forms such as -, - and -phases, which are constructed by combination of octahedral [MnO6] building blocks to exhibit various tunneled structures. These unique structures are believed to account for the various characteristics of MnO2 when it is employed as electrode material in lithium (ion) batteries [1-2]. There is, however, lack of direct proof demonstrating the role of tunneled structure during electrochemical lithiation/delithiation of MnO2. In this work, by applying aberration-corrected scanning transmission electron microscopy (ACSTEM) to single -MnO2 nanowire along both axial and radial directions, the tunneled structure is clearly shown and characterized in Figure 1. The -MnO2 nanowire is single crystalline and grows along [001] direction. Cross-sectional ACSTEM shows that the nanowire has a squared cross section and 2�2 tunnels align parallelly along its growth direction [001], matching very well with simulated crystal structure. An open cell design inside TEM for dynamic observation of MnO2's lithiation/delithiation process is also demonstrated. It is found that upon lithiation, the -MnO2 nanowire shows different orientation-sensitive morphologies. That is, -MnO2 unit cell expands asynchronously along [100] and [010] directions, resulting in macroscopic difference under [010] and [100] zone axes observations. Electron Energy Loss Spectroscopy further confirms such an asynchronous expansion property via quantification of Mn valence during lithiation. DFT simulation demonstrates that the asynchronous essential originates from the specific Li-occupancy sequence at Wyckoff 8h sites inside -MnO2's 2�2 tunnels. Following the theory, the predicted morphology of one partially lithiated nanowire and the experimental observation are shown in Figure 2, where both match very well. These findings provide fundamental understanding for stepwise potential variation during the discharge of Li/-MnO2 batteries as well as the origin for low practical capacity and fast capacity fading of -MnO2 as an intercalated electrode. 1. Microsc. Microanal. 21 (Suppl 3), 2015 330 Figure 1 (a,b) TEM image of a single -MnO2 nanowire and its SAED along <010>; (c,d) HAADF image of the same -MnO2 nanowire and the atomic model structure along <010> direction; yellow spots indicate Mn atoms, while pink represents K and red indicates O atoms. (e) TEM image of an MnO2 nanowire cross section and the corresponding SAED along [001] direction; (f) [001] atomic resolution HAADF showing K (pink) atomic columns and Mn (yellow) columns forming 1�1 and 2�2 tunnels; scale bars in c and f are both 1 nm; two inserts: the green line indicates EDS scan to distinguish Mn and K columns; the model shows [001] atomic configuration. Figure 2 Left: 3D model showing the asynchronous expansion of a partially lithiated K+-stabilized MnO2 nanowire; Right: Schematic [010] and [100] zone axes observation under TEM and the experimental TEM results. The color code (black, red, blue and green) indicates the corresponding zones observed in-situ. References: [1] Chen K. et al, The Journal of Physical Chemistry C 117 (2013), p.10770. [2] Xun Wang, Y. L., Journal of the American Chemical Society 124 (2002), p.2.");sQ1[166]=new Array("../7337/0331.pdf","EELS Analysis Of Lithiation/Delithiation Reactions In LiFePO4","","331 doi:10.1017/S1431927615002457 Paper No. 0166 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 EELS Analysis Of Lithiation/Delithiation Reactions In LiFePO4 J. Schneider-Haefner1, D. Su2, Y. Wang3, J. Fang3, F. Omenya3, F. N. Chernova3, and F. Cosandey1 1 2 Department of Materials Science & Engineering, Rutgers University, Piscataway, NJ 08854 Center for Functional Nanomaterials, Brookhaven National Laboratory, Upton, NY 11973 3 Department of Chemistry, Binghamton University, Binghamton, New York 13902 LiFePO4 has emerged as an important cathode material for Li-ion batteries because of it stability and high rate capabilities. It is now well established that lithiation-delithiation occurs via a twophase reaction. At high charge/discharge rates, the process of nucleation and growth of a two phase reaction is too slow and a non-equilibrium single phase reaction has been proposed followed by relaxation into LiFePO4 and FePO4 end product phases [1]. In this study, we studied reaction mechanisms and determined the spatial distribution of lithiated/delithiated phases by STEM/EELS spectrum imaging. LiFePO4 particles from partially charged or discharge electrodes were observed with a cold cathode field emission Hitachi HD2700C STEM and Gatan Enfina EELS spectrometer. The energy resolution of the combined STEM/EELS system was 0.5eV. The energy was calibrated with respect to the main O-K peak at 539 eV. Typical EELS spectrum for LiFePO4 and FePO4 are shown in Fig.1a and 1b respectively. A characteristic feature of delithiated FePO4 phase is the presence of an oxygen pre-peak marked by an arrow in Fig.1b. The existence of this O prepeak has been attributed to a transition from O 1s to 2p hybridized state with Fe 3d [2]. In addition the change in Fe valence state from LiFe2+PO4 to Fe3+PO4 is accompanied with a shift to higher energy of Fe-L3 peak position of about 1.5 eV. In this study we have quantified the existence of these two lithiated and delithiated phases from the shift in Fe-L3 peak energy, Fe L3/L2 peak intensity ratio and from quantification of normalized pre-O peak intensity. Measurements made from about 50 particles reveal two clusters of data with average Fe-L3 peak energy of 708.2 eV and 709.8 eV with O pre-peak intensity ratio of 0.037 and 0.16 respectively. These two data clusters correspond to the lithiated LiFePO4 and delithiated FePO4 phases. The spectrum images of the lithiated LiFePO4 and delithiated FePO4 expressed as the normalized O pre-peak intensity are shown in Fig.2a and 2b respectively revealing uniform lithiation throughout the particles, i.e. the particles are either fully lithiated or fully delithiated in accordance with the non-equilibrium solid solution transformation path followed by relaxation. An ADF-STEM image taken from an area with many particles and the corresponding phase distribution map are shown in Fig.3a and 3b respectively, revealing a non-uniform distribution of phases with agglomeration of fully lithiated and delithiated regions that include many particles. References [1] [2] [3] F. Omenya et al. Adv. Energy Mater. 4 (2014) 1401204 (9pp) M.K. Kinyanjui et al. J. Phys. Condens.Matter, 22 (2010) 275501 (8pp) Supported by NECCES a DOE-BES-EFRC funded center under Grant DE-SC0001294. Microsc. Microanal. 21 (Suppl 3), 2015 332 O-K Fe-L O-K Fe-L a b Fig. 1. EELS spectra of (a) fully lithiated LiFePO4 and (b) fully delithiated FePO4 showing characteristic O-K prepeak marked by an arrow. a b Fig.2. Normalized oxygen pre peak intensity map for (a) LiFePO4 and (b) FePO4 a b Fig.3. (a) ADF-STEM image of 50% delithiated LiFePO4 and (b) corresponding normalized oxygen pre peak intensity map.");sQ1[167]=new Array("../7337/0333.pdf","Atomic Observation of Phase Transformation from Spinel to Rock Salt in Lithium Manganese Oxide","","333 doi:10.1017/S1431927615002469 Paper No. 0167 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic Observation of Phase Transformation from Spinel to Rock Salt in Lithium Manganese Oxide Peng Gao1, Ryo Ishikawa1, Eita Tochigi1, Akihito Kumamoto1, Naoya Shibata1, Yuichi Ikuhara1 1 Institute of Engineering Innovation, School of Engineering, The University of Tokyo, Tokyo 113-8656, Japan Solid-state phase transformation are ubiquitous and of critical importance to numerous technological applications, such as inorganic ceramics, alloy and energy conversion and storage. These transformation processes usually involve only local atomic displacements without adding/reducing any atoms or longrange atom migration. For example, in perovskite ferroelectrics the oxygen octahedral shift and/or rotation results in a first order phase transition-polarization switching [1]. Investigating the dynamic processes of phase transformation in solids is critical to gain insights to understanding the fundamental physical properties of functional materials and benefiting industrial technologies. Here we show an atomic-scale observation of phase transformation from spinel to rock salt in lithium manganese oxide by using aberration corrected scanning transmission electron microscopy (STEM). Viewing along the [110] direction of spinel LiMn2O4 (space group Fd3m), the Li, O and Mn are well separated from each other (Fig. 1a). A `diamond' configuration consists of eight Mn columns in which the Mn density at Mn1 column is twice that of the Mn2 columns. The O atoms in the O1 columns are slightly misaligned and lean towards Mn1 sites (Fig.1a) while the two Li columns are separated by ~0.2 nm at the center of diamond. In our experiments, scanning the electron probe over a thin edge region of oxygen deficient LiMn2O4-x (LMO) particle will drive the spinel structure convert into rock salt (Fig.1a) via local atom migration without any mass loss which is evidenced by electron energy loss spectroscopy (EELS) in Fig.1b. Successive fast-scanned STEM frames are recorded to track this dynamic process. During the phase transformation, the intensity of Mn1 sites in the high angle annular dark-field (HAADF, Z-contrast) image is significantly reduced (35%) while the intensity at Mn3 columns gradually becomes brighter (in Fig. 1c), indicating Mn atoms migrate from Mn1 to Mn3 atomic sites. At 58 s in Fig.1c, the intensity between Mn1 and Mn2 columns is close because those columns have the same Mn density in the rock salt structure. Moreover we observe the anions shift during phase transformation. Simultaneously recorded sequential annular bright-field (ABF) images are shown in Fig. 1d. With the spinel structure, the O1 columns in the ABF image are not resolvable because the contrast of Mn1 columns is too intense and O1 columns are too close to Mn1. Once the structure has transformed into rock salt, the intensity of Mn1 is reduced significantly as half of Mn has been replaced by lithium. Furthermore, in the rock salt structure the O1 columns shift always from the Mn1 and are aligned perfectly and hence appear distinguishable in the ABF image as indicated by the arrows in Fig. 1d. Based on these observations, the phase transformations can be summarized as two step reactions: 1) tetrahedrally coordinated Li migrates to octahedral sites; 2) Mn migrates from one octahedral site to another. The detailed changes in composition, occupancy and electronic structures of both heavy and light atomic columns determined by STEM imaging (HAADF and ABF) and EELS, lead to useful insights into solid-state phase transformations and will bring beneficial technological solutions to a variety of applications such as the lithium ion batteries [2]. Microsc. Microanal. 21 (Suppl 3), 2015 334 References: [1] P.Gao et al, Nature Communications 2 (2011) 591. [2] The authors gratefully acknowledge the financial support through a Grant-in-Aid for Scientific Research on Innovative Areas "Nano Informatics" (Grant No. 25106003) from Japan Society for the Promotion of Science (JSPS), and "Nanotechnology Platform" (Project No. 12024046) from the Ministry of Education, Culture, Sports, Science and Technology in Japan (MEXT). Figure 1. (a) A perspective view of spinel LiMn2O4 along [110] direction and rock salt LiMnO2. Green: Li. Purple: Mn. Red: O. In the spinel phase, the atom density of Mn1 columns is twice of Mn2 columns. In the rock salt phase, the atom densities of Mn1, Mn2 and Mn3 sites are the same. Li site is unoccupied. (b) Intensities of simultaneously recorded Li-K and Mn-M edges, and simultaneously recorded O-K and Mn-M are plotted as a function of time showing no atom loss during phase transformation. Selected HAADF-STEM (c) and ABF-STEM (d) continuous images showing the phase transition from oxygen deficient spinel to lithium deficient rock salt in LiMn2O4-x (x denotes oxygen vacancies). Commercial LiMn2O4 nanoparticles were annealed in N2 at 700�C to form oxygen deficient LiMn2O4-x. In the STEM images with `TIFF' format, the intensity at each pixel has been automatically normalized between 0 and 255, the precise comparison in intensity is calculated from the original DM format. HAADF and ABF images are recorded simultaneously at 300 kV. The interval between frames is ~ 2 s. Each frame is 1024 by 1024 pixels in size. A post drift correction was applied to align the frames. Theses selected frames are unit-cell averaged from a region of 5 by 5 unit cells to reduce the noise. The arrows in (d) indicating the O1 sites in rock salt are resolvable from Mn1 sites contrasting to the pristine spinel structure.");sQ1[168]=new Array("../7337/0335.pdf","Microscopic Characterization of Electrodeposited Mg Layers for Battery Application","doi:10.1017/S1431927615002470","335 doi:10.1017/S1431927615002470 Paper No. 0168 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microscopic Characterization of Electrodeposited Mg Layers for Battery Application Mukesh Bachhav1, Emily Nelson2, Adam Crowe2, Bart Bartlett2, Nathan Hahn3, Kevin Zavadil3, PengWei Chu1, Emmanuelle A. Marquis1 1 Department of Materials Science and Engineering, University of Michigan, Ann Arbor 48109. MI, USA 2 Department of Chemistry, University of Michigan, Ann Arbor 48109-1055, MI, USA 3 Advanced Materials Laboratory, Sandia National Laboratories, Albuquerque, NM, 87185 USA Magnesium has been considered as a strong candidate for metal air battery applications due its low cost, environmentally benignity, high theoretical specific charge capacity (2.205 Ah/g), and high theoretical energy density (3.8 Ah/cm3) [1, 2]. However, one of the current technical limitations is the understanding of the interactions taking place at the electrolyte/Mg interface. In this study, atom probe tomography (APT) and transmission electron microscopy (TEM) have been successfully applied to the characterization of electrodeposited Mg layers to understand the functionality of electrolytes in Mg deposition. Mg layers were deposited in a series of electrolytes using a standard 3 electrode cell with working electrode as Au coated-Si wafer, counter electrode as Mg0, and reference electrode as Mg0. The compositions of the electrodeposited Mg layers were investigated using atom probe tomography to understand the role of electrolyte functionality in the deposition of Mg. The three electrolytes used were (PhMgCl)4-Al(OPh) (APCC) [3], 2MgCl2:AlCl3 (MACC) [4], both in tetrahydrofuran, and Mg (CF3SO2NSO2CF3)2 (TFSI) in diglyme to explore the effect of electrolyte chemistry. The morphology, structure and chemistry of the deposited Mg layers will be discussed. The deposited films are almost pure Mg (>98 at%) with impurities that include O, C, Al, Cl, MgHX, MgOX. The amount of impurity is slightly lower for MACC electrolyte, while the other two electrolytes yield significant of hydroxide and impurity incorporation. Compositions of Mg layers obtained from the three electrolytes are shown in Table 1. Efficient Mg deposition/dissolution using a Grignard-based electrolyte is commonly ascribed to the absence of a surface film, which would impact anode performance. The nucleation and growth mechanisms of Mg grains during deposition are important mechanisms to understand to ensure longterm stability of rechargeable Mg batteries. To investigate the interactions with possible surface films, analyses were carried out on Mg layers, for which deposition had been interrupted for some controlled amount of time [5] to let the surface equilibrate with the electrolyte. A TEM cross-section image (Fig. 1a) illustrates the two layers of Mg formed during the interrupted deposition using MACC. APT analysis indicates the presence of C, Al, Al, and Mg oxides/hydrides at the interface (Fig. 1b). Our understanding of the role of the surface film in governing the re-nucleation of Mg onto itself will be discussed in this presentation. References: [1] T. Zhang, Z. Tao, J. Chen, Materials Horizons, 1 (2014) 196-206. [2] F. Cheng, J. Chen, Chemical Society Reviews, 41 (2012) 2172-2192. [3] E.G. Nelson, J.W. Kampf, B.M. Bartlett, Chemical Communications, 50 (2014) 5193-5195. Microsc. Microanal. 21 (Suppl 3), 2015 336 [4] R.E. Doe, R. Han, J. Hwang, A.J. Gmitter, I. Shterenberg, H.D. Yoo, N. Pour, D. Aurbach, Chemical Communications, 50 (2014) 243-245. [5] Nathan T. Hahn, P.G. Kotula, David Wetzel, Marvin Malone, Ralph G. Nuzzo, and Kevin R. Zavadil, Submitted to J Phys Chem C [6] This work was supported as part of the Joint Center for Energy Storage Research an Energy Innovation Hub funded by the U. S Department of Energy, Office of Science, Basic Energy Sciences Table 1: Measured compositions (in at.%) of electrodeposited Mg films deposited using APCC(PhMgCl)4-Al(OPh)3, MACC-2MgCl2:AlCl3, and TFSI-Mg(CF3SO2NSO2CF3)2 Elements/ions TFSI MACC Mg 98.2�0.3 99.6�0.2 C, O, Cl, MgHx, MgOX 1.8�0.3 0.4�0.2 APCC 98.3�0.3 1.6�0.2 (a) Top Mg layer Mg MgOx, MgHx, Al, C, Cl, O 20 nm (b) (c) 2.0 MgHx C O MgOx Cl Interface 1.5 Concentration Targeted APT specimen Interface 1.0 0.5 Bottom Mg layer 100 nm 0.0 0 10 20 30 Distance (nm) 40 50 Figure 1: TEM image showing two distinct Mg layers marked with site targeted by APT. (b) 3-D reconstruction showing interface and distribution of Mg and impurities. (c) 1D distribution profile taken along interface from top layer to bottom layer");sQ1[169]=new Array("../7337/0337.pdf","New Developments in Automated Particle Analysis in the Electron Microscope � from Micro to Nano","","337 doi:10.1017/S1431927615002482 Paper No. 0169 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 New Developments in Automated Particle Analysis in the Electron Microscope � from Micro to Nano Christian Lang1, Matthew Hiscock1, James Holland1, Susumu Yamaguchi2, David Joyce,3 and Georgia Vatougia3 1. 2. Oxford Instruments Nanoanalysis, High Wycombe UK Oxford Instruments KK, Tokyo, Japan. 3. Mantis Deposition Ltd., Thame, UK As EDS detectors have become faster, computers more powerful and data storage space cheaper, EDS analysis has become more image centric. X-ray mapping is now used routinely and can be combined with automated stage movements and stitching of maps, so that large sample areas can be covered at high resolution. These very large data sets contain a wealth of information which so far has often not been fully utilized. A new Particle Analysis software (AZtecFeature) enables the automated processing of these large data sets in order to identify features of interest in the sample and measure their morphology and composition. As in conventional particle analysis, the electron image is used to identify the location of features. Different to conventional particle analysis, the new particle analysis software has the option to link a set of electron images to a set of EDS maps of the same area, from which it extracts the compositional information. As these maps can cover the whole sample area, the new particle analysis system can store a virtual sample (figure 1a). This virtual sample can be analyzed offline (figure 1b), enabling the user to adjust threshold settings and also to investigate features which span more than one field of view. Storing and working with virtual samples enables their sharing amongst research groups over the internet, avoiding the need for transportation of rare or precious samples. In order to increase data throughput, the new system can acquire data from multiple EDS detectors. In addition to increasing data throughput, we show how multiple detectors can also be used to minimize shadowing of smaller particles by larger ones if the detectors are mounted on opposite ports on the SEM. Figure 2a shows an X-ray map acquired with one EDS detector overlaid over an electron image of a large particle with several smaller particles surrounding it. The EDS detector is positioned on the left hand side of the image, which results in shadowing of some of the smaller particles to the right hand side of the image. Figure 2b shows the same particles, now mapped with two detectors mounted on opposing ports; shadowing is clearly eliminated. Developments in both the electron microscope and EDS detector hardware enable the acquisition of compositional data from particle sizes down to a few nanometers. Very high solid angle detectors on either a TEM or an Ultra-High Resolution SEM such as the Hitachi SU9000 enable the elemental mapping of particles down to 2nm in size (figure 3a inset). While the X-ray map clearly shows small particles on the scale of a few nanometers, measuring their size accurately is difficult due to noise in the map. Using automated particle analysis we can correlate the electron image with an EDS map of the area. This way, the particle size distribution can be measured from the less noisy, higher resolution electron image (figure 3). The particle size distribution plotted as a function of the equivalent circular diameter in figure 3 shows that particles below 5nm are clearly resolved. The location of the particles can be correlated to the EDS map shown in the inset. As every pixel in the electron image is correlated to a full X-ray spectrum in the EDS map, the software can reconstruct the X-ray spectrum for each particle in order to confirm the particle composition. Microsc. Microanal. 21 (Suppl 3), 2015 338 Figure 1. (a) shows a large area map of wear particles containing 12 fields of view. (b) shows the extraction and classification of particle data from the maps in progress. Figure 2. (a) one EDS detector on the left side of the images leads to shadowing in the area circle. (b) shows the same area now with two EDS detectors mounted left and right. Figure 3. Electron image of small Molybdenum particles down to 2nm size (Mo-L map in the inset). Particle have been detected in the electron image (colored). The particle size distribution is shown in the histogram (right).");sQ1[170]=new Array("../7337/0339.pdf","STEM-EDXS System for Atomic-Sensitivity Elemental Mapping","","339 doi:10.1017/S1431927615002494 Paper No. 0170 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 STEM-EDXS System for Atomic-Sensitivity Elemental Mapping T.C. Lovejoy1, R.M. Stroud2, N.D. Bassim2, G.J. Corbin1, N. Dellby1, W. Hahn3, P. Hrncirik1, M. Falke3, A. Kaeppel3, M. Rohde3, and O.L. Krivanek1 Nion Co., 11511 NE 118th St. Kirkland, WA 98034, USA 2. Materials Sci. and Tech. Division, Naval Research Laboratory, Washington, DC 20375, USA 3. Bruker Nano GmbH, Am Studio 2D, 12489 Berlin, Germany The Nion UltraSTEM aberration-corrected cold field emission scanning transmission electron microscope (AC-CFE-STEM) can focus a beam current of about 0.2 nA into an atom-sized (~1.5 � large) electron probe at a primary energy of 60 keV (and about 1 nA at 200 keV). Using such an electron probe with an ultra-thin window silicon drift detector (SDD) of 30 mm2 (0.1 sr solid angle) in a Mark I system at 60 keV allowed energy dispersive X-ray spectroscopy (EDXS) to identify individual atoms while tracking their occasional hops to neighboring lattice sites [1]. Here we report on the design and performance of a Mark II STEM-EDXS system with about 10x higher efficiency, which allows single impurity atoms to be identified even when they are very mobile. The new system uses a window-less SDD that is oval rather than round, and has an area of 100 mm2. Optimizing the shape of the objective lens polepiece and of the detector front end has allowed the detector to be brought to just 10.5 mm from the sample. Approximating the detector solid angle as 100/10.52 gives 0.91 sr, accounting for the planar geometry of the detector gives 0.7 sr. Making the detector window-less provides a further significant gain in quantum efficiency, especially for lowenergy X-ray emission lines. The detector is mounted on a Nion UltraSTEM200 set up for voltages ranging from 40 to 200kV, and operated at 60 kV for this work. To prevent sample contamination due to mobile hydrocarbons, or etching due to water vapor permeating through single O-rings and condensing on the sample, the microscope's column uses only metal seals and is bakeable to 140 C. The detector itself cannot be baked above 100 C, and it therefore needs to be removed from the column before baking. This is accomplished by having a gate valve that allows the detector to be isolated and removed without venting the column (Fig. 1). The internal (non-bakeable) volume of the detector is pumped with an ion pump and is maintained in the 10-7 torr range. The vacuum next to the sample is kept in the low 10-9 to high 10-10 torr range by a separate ion pump. When the detector is retracted, the two vacuums are separated by the gate valve; when the gate valve is opened and a motor drives the detector in, an O-ring sealing on the detector tube maintains the separation (Fig. 1). With the detector in the inserted position and cooled, the resultant samplelevel vacuum is high and clean enough for room temperature 60 keV imaging of graphene to be possible without significant contamination or etching. To reduce the impact of spurious system peaks without comprising system performance or the geometric solid angle with a collimator, the objective lens polepieces have been covered by thin copper caps, and Be-Cu gimbals are used in the double-tilt sample holder. For a sample mounted on a Cu grid and clamped with a Be-Cu spring washer, there is primarily just one type of system peaks: Cu, whose real concentration can be readily characterized by EELS. Figure 2 shows a spectrum from a 54nm thick NiO calibration sample on a Molybdenum grid acquired at 60 keV with ~440 pA. Peaks from Ni, O, Moly, and Cu dominate; minor system peaks of Ti, Si, and Al remain. 1. Microsc. Microanal. 21 (Suppl 3), 2015 340 Figure 3 shows an example of detecting a mobile single atom of Ca on few-layer amorphous carbon, using tracking methodology as in [1], but with a significantly lower beam current (46pA), to mitigate the atom's greater mobility, and shorter time. There were 22 Ca K-line counts in 35s of tracking, during which the beam only spent about 1 sec close enough to the Ca atom to ionize it. At 35 s, the atom left the field of view of the tracking window. Similar count rates were recorded for other types of single atoms present in the sample, including silicon and sulphur. This demonstrates that the new system allows EDXS analysis to be carried out with single atom sensitivity for many elements in the periodic table, on real-world samples that change under the electron beam. See [2] for a practical application to heteroatom mapping in nanodiamond of extra-terrestrial origin. References 1. T.C. Lovejoy et al., Appl. Phys. Lett. 100 154101 (2012) 2. R.M. Stroud et al., (this meeting) Fig. 1 Schematic drawing of 100 mm2 SDD chip 10.5 mm from sample showing copper polecaps over the polepieces, isolating Oring between the baked-ultra-clean sample vacuum (UHV) and non-baked detector vacuum (HV), isolating gate vale for when detector is retracted and approximate takeoff angles to detector extremes. Fig. 2 Spectrum from NiO on Mo grid showing primarily Cu system peaks. Fig. 3 (left) overview dark-field image of few-layer amorphous carbon showing several different species of isolated impurity atoms. The highlighted atom was tracked using the window shown (inset) to produce an X-ray spectrum (right) with 22 Ca K-line counts in 35 s (60 kV, 46 pA).");sQ1[171]=new Array("../7337/0341.pdf","Aberration Correction System Using Segmented Detector in STEM","","341 doi:10.1017/S1431927615002500 Paper No. 0171 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Aberration Correction System Using Segmented Detector in STEM Y. Kohno1, H. Sawada1 and N. Shibata2 1. 2. JEOL Ltd., 3-1-2 Musashino, Akishima, Tokyo 196-8558, Japan Institute of Engineering Innovation, School of Engineering, the University of Tokyo, Tokyo 116-0013, Japan Precise aberration measurement is essential for automated aberration correction systems, since it is one of determining factors in the performance of the automatic correction. Several aberration measurement methods for STEM are now in use, such as analyzing the Ronchigrams from an amorphous specimen, the probe form estimated from images of particles and so on. We cannot, however, observe the target specimen images immediately after the automatic aberration correction using these existing methods, since we have to change the illuminating apertures and exchange the specimen or the field of view. Moreover, we have to finally tune the aberrations (defocus, astigmatism or coma) manually during these procedures. To avoid these processes, we need to estimate the aberrations from the images of the target specimens. There is a method of aberration measurements with the target specimen using image shifts of the BF STEM images detected with different azimuth and radial detection angles. This method is basically equivalent with defocus correction using an image wobbler, which is used in BF TEM, if we apply the reciprocal theory. If the geometrical aberrations of the illuminating system exist, the electron beam at the specimen surface does not focus in one point and have blurred distribution depending on the different incident angles and aberrations, resulting in the image shifts in the BF STEM image. For example, the simple defocus results in the circular image shifts with the same radial angle and various azimuth angles, and two fold astigmatism results in the ellipsoidal image shifts with the same radial angle and various azimuth angles. In this way, we can estimate the aberrations from the shifts of the BF STEM images. This method with a single detector was proposed years ago [1], but is not commonly used. This is partly due to the long image acquisition time of multiple images having different azimuth and radial detection angles and the measurement errors arise from the image drift during the acquisition. These are unavoidable for normal STEM detection systems, which controls the detection angle by deflectors in the image forming system and acquire only one image at a time. With a segmented detector, where each detector has different azimuth and radial detection angles, a set of images with different incident angles can be acquired simultaneously without image drifts. We developed a new automatic aberration correction system using a segmented detector, which is composed of 16 multiple STEM detectors and is able to detect 16 images with different detection angles simultaneously [2]. The segmented detector is attached to the JEM-ARM300F with the STEM corrector system. The aberration correction module, embedded in the detector control system, calculates aberration coefficients from the relative image shifts of the acquired images. The module also communicates with the STEM corrector system through an ethernet, and the system controls the aberrations. This configuration realizes the automatic aberration correction on the target specimen using the segmented detector. Figure 1 shows the Ronchigrams before and after the automatic aberration correction processes. We intentionally introduced defocus, 2-fold astigmatism, coma and 3-fold astigmatism before the processes. Microsc. Microanal. 21 (Suppl 3), 2015 342 After the correction of these aberrations, the residual 6-fold astigmatism clearly appears in the Ronchigram. In this case, we use 8 BF STEM images acquired simultaneously, which are enough to calculate up to 2nd order aberration coefficients. One loop takes about 10 seconds, which includes acquiring images, calculating aberrations and giving feedback to the corrector system. Figure 2 shows the dark field STEM images of gold particles on carbon thin film before and after the processes. These images show that the system can cope with large aberrations and that the measurement accuracy is high enough for high resolution imaging. References [1] O. L. Krivanek et al, Ultramicroscopy. 78 (1999) 1 [2] N. Shibata et al, J. Electron Microsc. 59 (2010) 473 [3] We thank Prof. Y. Ikuhara, the University of Tokyo for his continuous support. This development was supported by SENTAN, JST. Figure 1. The Ronchigrams before (left) and after (right) the automatic aberration correction processes. Defocus, 2-fold astigmatisms, coma and 3-fold astigmatisms are intentionally introduced in the left Ronchigram. These are corrected in the right one. 20nm 20nm Figure 2. The dark field STEM images of gold nanoparticles on a carbon thin film. The left image was acquired before the processes and the right one is acquired after them.");sQ1[172]=new Array("../7337/0343.pdf","From Scintillator-based Detector to Direct Electron Detector: High Performance of Next Generation of Camera for In-situ TEM Testing and TEM Imaging","","343 doi:10.1017/S1431927615002512 Paper No. 0172 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 From Scintillator-based Detector to Direct Electron Detector: High Performance of Next Generation of Camera for In-situ TEM Testing and TEM Imaging Hua Guo1, Liang Jin1, Penghan Lu2, Zhangjie Wang2,Zhiwei Shan2, Benjamin Bammes1, Michael Spilman1, and Robert Bilhorn1 1. 2. Direct Electron, LP, San Diego, CA, USA. Center for Advancing Materials Performance from the Nanoscale (CAMP-Nano), Hysitron Applied Research Center in China (HARCC), XJTU-Hitachi High-Tech Research & Development Center (XHRDC), State Key Laboratory for Mechanical Behavior of Materials, Xi'an Jiaotong University, Xi'an 710049, China Transmission electron microscopy (TEM) is a powerful technique for characterizing materials at super high spatial resolution. In the past decades, many TEM related techniques like EDX, EELS, STEM, Cscorrection and HRTEM have been well developed [1]. As a result, both chemical and structural information has been obtained at atomic resolution for many important specimen types, which makes transmission electron microscopes indispensable scientific instruments in the modern laboratory. Recently, in-situ TEM testing techniques have attracted intense research attention, especially in the materials science community. For instance, the propagation of defects has been observed directly in compression tests of Ni nanopillars [2]; atomic rearrangement has been recorded when straining Au nanostructures [3]; a new deformation mode has been found in small amorphous alloy samples in tensile tests [4]; and even crystal growth processes and chemical reactions have been captured in liquid environments [5]. Such in-situ testing experiments provide important direct evidence about materials deformation mechanisms and chemical reactions, but also introduce new instrumentation challenges in the TEM. These include undesired dynamic specimen processes (e.g., drift, beam-induced motion, charging, radiation damage, etc.), insufficient field of view, insufficient frame rates, and inefficient electron detectors [6]. With the goal of overcoming many of these obstacles for the life sciences, Direct Electron (San Diego, CA, USA) introduced the first large-format Direct Detection Device (DDD�) in 2008, as the culmination of academic and industrial partnerships working through six generations of sensor development beginning in 2001 [7]. Recently, Direct Electron released its new 20-megapixel "DE-20" camera system, representing the ninth generation of DDD development and providing the largest field-of-view of any available TEM direct detection camera [8]. Direct Electron has now released new tools for in-situ TEM that enable high-speed recording of large format, very high quality "movies" of dynamic processes. Here, we show results from the DE-12 DDD camera with the new in-situ tools clearly observing the twining process in a ~200 nm diameter Cu nanopillar in a quantitative compression test. Figure 1 shows five snapshots from a video stream recorded with the camera at a frame rate of 57 frames per second. Fig. 1 (a) is a TEM bright field image of the pristine ~500 nm high Cu single crystal nanopillar. The primary twin, going through the whole specimen, started emerging in Fig. 1(b) at 8.25 sec, and formed completely in the subsequent 0.14 sec as shown in in Fig. 1(c). Afterwards, the twin grew gradually until it ruptured, which took about 0.39 sec as shown between Figs. 1(d,e). A FIB fabrication-induced damaged layer could be distinguished easily with a thickness of about 7 nm from the robust specimen. The layer was found to be flexible, peeling off at the surface where the primary twining happened as shown in Fig. 2, which is an enlarged view of the framed area in Fig. 1(e). Mechanical adhesion between Microsc. Microanal. 21 (Suppl 3), 2015 344 the damaged layer and the nanopillar appears not to be strong at all. In this regard, it is reasonable to ignore the influence of the damaged layer for this quantitative mechanical test. The very good signal-to-noise ratio (SNR) and wide dynamic range of individual frames taken from the movie, as illustrated by the observation of the FIB damage layer, illustrate the impressive capabilities of this new in-situ tool. References [1] D.B. Williams, and C.B. Carter, "Transmission Electron Microscopy: A Textbook for Materials", 2nd Ed. (Plenum Press, New York). [2] Z.W. Shan et al, Nature Materials, 7 (2007) 115. [3] H. Zheng et al, Nature Communications 1 (2010) 144. [4] H. Guo et al, Nature Materials, 6 (2007) 735. [5] H.G. Liao et al, Science, 345 (2014) 916. [6] R.M. Glaeser, and R.J. Hall, Biophys J 100 (2011) 2331. [7] L. Jin and R. Bilhorn, Microsc Microanal 16 (2010) 854-855. [8] B.E. Bammes et al, Microscopy & Microanalsis Conference Proceedings (2012) LB-36. Figure 1. Series of TEM bright field images of twining process in Cu nanopillar from video stream recorded at a frame rate of 57 frames per second. Note that the time scale is not linear. Figure 2. TEM bright field image of damaged layer peeled off from the surface, where twining happened as framed in Fig. 1(e). Instrumentation: DE-12 DDD camera system, Hitachi H-9500 E-TEM, Hysitron PI 95 in-situ TEM holder.");sQ1[173]=new Array("../7337/0345.pdf","Imaging Contrast with Multiple Ion Beams","","345 doi:10.1017/S1431927615002524 Paper No. 0173 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Imaging Contrast with Multiple Ion Beams Huimeng Wu, Sybren Sijbrandij, Shawn McVey and John Notte Ion Microscopy Innovation Center, Carl Zeiss Microscopy LLC, One Corporation Way, Peabody MA 01960, USA A commercial Ga-FIB/SEM system can directly image samples using an ion beam or e-beam before, after or during milling/depositing process. This capability provides important feedback for process control. Because of the limited spatial resolution and Ga contamination of the Ga+ ion beam, the e-beam is often considered as the primary imaging tool. But ion beam imaging also provides important information about the samples. Orion Nanofab integrates He+, Ne+ and Ga+ focused ion beams on one single platform. He+ and Ne+ ion beams are based on the gas field ion source (GFIS) technology. The images generated by He+/Ne+ ion beams are sub-nm in resolution capturing intricate details of the samples[1]. Nanofab also provides an optional state of art Ga-FIB. With this unique configuration, Orion Nanofab provides a great platform to study ion beam imaging with a variety of ion species. In this study, we investigate three imaging modes: secondary electron (SE), secondary ion (SI) and backscattered ion (BSI), using three ion beams: He+, Ne+ and Ga+. The most common imaging mode of the e-beam and ion beam is via SE. Compared with the SE image of e-beam, ion beam imaging is more sensitive to surface topography, easier charge neutralization with a flood gun, better for passive voltage contrast, and stronger grain orientation contrast due to ion-channeling effects. Other imaging modes via SI and BSI can provide additional information to better characterize the sample and eliminate the need for further analysis. SI images provide material contrast because of the different secondary ion yield of different materials, particularly sensitive to the presence of oxides and carbides. This property enables the studies of corrosion or grain boundary segregation in metallic systems[2]. Also the SI mode allows the electron flood gun to operate in continuous emission mode without multiplex imaging. BSI images provide less surface specification, but the atomic number "Z" contrast. Because BSI has the similar energy as the incident ion beam energy, the signal can pass through several layers. The schematic illustration in Fig. 1a shows three ion beams on a Orion Nanofab and the ion beam/sample interactions. When an ion beam impinges on a sample surface, it emits secondary particles, including low energy secondary electrons, secondary ions, backscattered ions and others. An experimental measurement was carried out to compare the BSI signal and SI signal for He+ and Ne+ ion beams on ten chosen target elements (Fig. 1b). The results show that the Ne+ induced SI signal is more than one order of magnitude larger than He+, as is expected from SRIM-calculations of sputter yield[3]. For a He+ ion beam, the BSI signal is about an order of magnitude larger than the SI signal. For a Ne+ ion beam, the BSI signal intensity is approximately equal to the SI signal. From this measurement, a He+ ion beam is suitable for SE and BSI images, but the SI signal is relatively low. A Ne+ ion beam is useful for acquiring all imaging with SE, SI and BSI signal. Two samples were tested to show that the combination of SE, BSI and SI images can provide complementary information about the samples. Fig. 2a and 2b are tilted He+ ion beam images of carbon thin film on coper grid. The SE image (Fig. 2a) shows the surface topography of the thin carbon film and the SE signal from the materials beneath the carbon film can't penetrate through the film. The BSI image (Fig. 2b) shows better contrast of the underlying material than the carbon film because of the Microsc. Microanal. 21 (Suppl 3), 2015 346 carbon low Z. Fig. 2c and 2d are Ne+ ion beam images of the oxide patterns with Al pads. The SE image (Fig. 2c) shows the very strong voltage contrast. Two Al pads on the bottom corner of the image are grounded and very bright. The two pads in the center of the image are floating, appearing almost black. All oxide dielectric patterns are very dark because of charging. While in the SI image (Fig. 2d), all oxide dielectric patterns show good contrast and all four metal Al pads are dark because of the low SI yield. No flood gun was applied during the imaging of either samples. References: [1] BW Ward, JA Notte and NP Economou, Vac. Sci. B. 24 (2006), p. 2871. [2] A Laquerre and MW Phaneuf, Microscopy & Microanalysis, 14 (2008), p. 620. [3] http://www.srim.org/ a b Figure 1. a) Schematic illustration of three ion beams on a) Orion Nanofab and the ion beam/sample interaction. b) Comparison of BSI and SI signal for 25 keV He+ and Ne+ ion beam. Figure 2. Tilted He+ ion beam a) SE and b) BSI images of carbon thin film on coper grid. Ne+ ion beam c) SE and d) SI images of the oxide patterns with Al pads.");sQ1[174]=new Array("../7337/0347.pdf","The Ultra-stable Scanning Transmission Electron Holography Microscope (STEHM)","","347 doi:10.1017/S1431927615002536 Paper No. 0174 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Ultra-stable Scanning Transmission Electron Holography Microscope (STEHM) Rodney A. Herring1 and David Hoyle2 1. 2. CAMTEC, MENG, University of Victoria, British Columbia V8W 2Y2 Canada Hitachi High-Technology, Ltd., 89 Rexdale Blvd., Toronto, M9W 6A4 Canada The Scanning Transmission Electron Holography Microscope (STEHM), a Hitachi HF 3300v equipped with the first Cs + coma TEM aberration corrector, i.e., an aplanatic TEM referred to as the b-corr, and the first Cs + Cc STEM aberration corrector referred to as the sccorr has been installed at the University of Victoria, Canada. The STEHM is an ultra-high resolution, ultra-stable electron microscope maintaining excellent atomic resolution stability for recording times of 120 s, Fig. 1, an unprecedented length of time, maintaining spatial frequencies of 70 pm within its lattice images. Other than the inherent stability of its electron optical system that can produce ~35 pm Fourier intensities for optimal imaging conditions, the residual stability of the STEHM is achieved by having an environmentally stable laboratory, as reported earlier [1], and a structurally stable specimen as reported in this abstract where a stable specimen for measuring the spatial resolution of the electron microscope has not yet been found. This is readily seen in Fig. 1 in the change in the crystal orientation of gold nanoparticles, now well known, during electron beam observation results in a change in contrast of the lattice fringes where some fringes recorded for 120 s are stronger in contrast than the 2 s image. The changes in the Au crystal's structure are verifiable in their Fourier transforms by the appearance and disappearance of Fourier peak intensities. Within the amorphous carbon substrate, specimen instabilities also exist. Although it is very difficult to follow the change in the amorphous structure by imaging, its periodic structures are manifested in its diffraction pattern consisting of speckle intensities. Self-interference of the speckle intensities can be used to follow the stability of the amorphous structure by using either a wavefront splitting method or an amplitude splitting method [2, 3]. The amplitude splitting method involves self-interfering each speckle intensity, which is carried by the direct beam and the gold crystal's Bragg diffracted beams, using an electron biprism as described fully elsewhere [4]. Monitoring the contrast of the a-carbon's interference fringes, Fig. 2, enables the stability of the a-carbon structure to be followed during electron beam irradiation. This interference method also enables the absolute phase of the amorphous specimen to be measured necessary for determining its structure [3, 4]. Fig. 2 shows the change in fringe contrast within the amorphous speckle intensities due to its changing structure, which causes decoherence of the speckle intensities. Thus, both the gold nanocrystals and its a-carbon support film, which are commonly used for measuring the spatial resolution of electron microscopes, are unstable, which will lead to an under-estimation of the true spatial resolution of the microscope. The long recording time shown possible using the STEHM enables very low dose imaging necessary for electron beam sensitive specimens. The STEHM's measured stability may be dependent more on the specimen instabilities, which produce decoherence of the electron beams, rather than the microscope instabilities. 1. Herring, R.A., David Hoyle, Yoshifumi Taniguchi and Max Haider, "The Ultra-Stable Scanning Transmission Electron Holography Microscope" Microscopy & Microanalysis, (Indiannapolis, August 4-8) 19 (Suppl 2) (2013) 302/303. Microsc. Microanal. 21 (Suppl 3), 2015 348 2. Herring R A, Saitoh K, N, Tanaka N, and Tanji T (2010) Coherent Electron Interference from Amorphous TEM Specimens. J. Electron Microscopy 59: 321. 3. Herring R A, Saitoh K, N, Tanji T, and Tanaka N (2012) Electron interference from an amorphous thin film on a crystal transmission electron microscopy specimen. J. Electron Microscopy 61: 17. 4. These proceedings. Funding from NSERC, CFI and BCKDF are gratefully acknowledged. a b c d Figure 1 � TEM images and their corresponding Fourier transforms of gold crystals on an amorphous carbon substrate taken for 2 s in a and b and for 120 s in c and d. Contrast differences are due to specimen instabilities. Recording lattice images at 120 s enables high-resolution low dose imaging. Figure 2 � Decoherence of interference fringes within the amorphous C intensities due to specimen instabilities. Images are recorded within a few seconds of each other from panels a, b, c. White circle is a reference location.");sQ1[175]=new Array("../7337/0349.pdf","Opportunities for Low-Voltage TEM/STEM.","","349 doi:10.1017/S1431927615002548 Paper No. 0175 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Opportunities for Low-Voltage TEM/STEM. R.F. Egerton 1. Physics Department, University of Alberta, Edmonton, Canada T6G 2E1. The main effect of reducing the TEM accelerating voltage V0 is to increase the fraction of electrons that are scattered elastically and inelastically. As this fraction approaches 1, plural scattering becomes excessive and is generally deleterious in TEM images and energy-loss spectra. Therefore low-V0 TEM is attractive only for very thin specimens [1]. Fortunately there are nanotechnology samples that fulfill this requirement, such as graphene and carbon nanotubes. For such a specimen, the increased elasticscattering power at low V0 increases the image contrast (in bright-field imaging) or the signal (for darkfield STEM). If the specimen is electrically conducting, the predominant mechanism of radiation damage is likely to be knock-on displacement, which can often be minimized or even eliminated by using an accelerating voltage below some threshold value. Unfortunately, reducing V0 increases the electron wavelength and increases the angular spread that must be focused by the objective lens, for a given diffraction-limited resolution: dd = 0.6/. If V0 is reduced below 40 kV, atomic resolution requires correction of both spherical and chromatic aberration, as illustrated in Table 1. It is tempting to imagine a 10 � 30kV STEM for nanotechnology specimens, similar to that used by Crewe et al [2], but with aberration correction. If equipped with a high-resolution monochromator, such an instrument could perform vibrational-mode EELS [3] with a spatial resolution determined by delocalization and radiation damage [4]. If provided with ultrahigh vacuum (UHV) and specimen cleaning devices, Auger spectroscopy and atomic-scale secondary-electron imaging [5] might also be possible. For beam-sensitive materials that damage by radiolysis, there is no useful accelerating-voltage threshold and the only way to minimize radiation damage is to defocus the incident electron beam and record a signal from a large area of specimen. This of course defeats the usual goal of microscopy, which is to obtain good spatial resolution. Although both the (elastic or inelastic) signal and radiolysis damage are inversely proportional to accelerating voltage, it is misleading to say that the signal/damage ratio is independent of V0: factors such as specimen thickness and imaging mode should be taken into account. The dose-limited spatial resolution is DLR ~ D1/2/C where C = image-contrast ratio, D = tolerable dose and ~ means "proportional to". D ~ V0 for radiolysis damage, while C ~ t/V0 for bright-field scattering contrast and a small sample thickness t, giving DLR ~ V01/2/t. Therefore low V0 is advantageous for very thin specimens, although DLR > 1 nm for typical D; see Fig. 1a. With increasing thickness, the resolution improves until plural scattering predominates, shown by the upward curvatures in Fig. 1a. For phase contrast, C ~ t/V01/2 at small t, making DLR independent of accelerating voltage [1]. Electrostatic charging is a further problem with poorly conducting thin specimens, which charge positively due to the emission of secondary and Auger electrons. In some materials, the emission current is compensated by a conduction current when the positive potential reaches some modest potential; see Fig. 1b. However, this voltage may be sufficient to cause dielectric breakdown (due to the local electric field) or even a "Coulomb explosion", which in practice means the emission of positive ions and hole drilling in oxides [6,7]. In more insulating specimens, the potential may be many thousands of volts, sufficient to deflect the incident beam or to interfere with electron focusing. These effects are likely to Microsc. Microanal. 21 (Suppl 3), 2015 350 be more troublesome at low accelerating voltage because the primary electrons are then more easily deflected and because the secondary-electron yield increases with decreasing V0 [8,10]. References: [1] RF Egerton, Ultramicroscopy 145 (2014), p. 85. [2] AV Crewe et al, J. Appl. Phys. 39 (1968) p. 5861. [3] OL Krivanek et al, Nature 514 (2014), p. 209. [4] RF Egerton, Microsc. Microanal. 20 (2014), p. 658. [5] H Inada et al, Ultramicroscopy 111 (2011), p. 865. [6] SD Berger et al, Phil. Mag. B 55 (1987), p. 341. [7] J Cazaux, Ultramicroscopy 60 (1995), p. 411. [8] L Reimer, "Scanning Electron Microscopy", (Springer, Berlin) p.154. [9] FJ Pijper and P Kruit, Phys. Rev. B 44 (1991), p. 9192. [10] The author gratefully acknowledges funding from the Natural Sciences and Engineering Research Council of Canada, and encouragement from his colleagues Ute Kaiser, David Bell and Jimmy Liu. kinetic energy E0 = eV0 30 keV 10 keV (mrad) 4.2 74 ds (nm) = 0.5 Cs 3 0.074 400 dc (nm) = 0.5 Cc (E/E0) 0.14 7.4 Table 1. Angular spread required for a diffraction limit of dd = 0.1 nm, with estimates of the loss of resolution due to spherical (ds) and chromatic (dc) aberration, taking Cs = Cc = 2 mm and E = 1eV. (a) (b) Figure 2. (a) Dose-limited resolution and contrast ratio C for bright-field TEM imaging (5-mrad objective aperture) at different accelerating voltages, calculated for a boundary in an amorphous specimen where the mean atomic number changes by 10%. (b) Solid and dashed curves: total (secondary + Auger) yield for amorphous carbon as a function of surface potential [9]. Straight lines: compensating conduction current (Ic) divided by incident-beam current (Ib), for two values of Ib (1 pA and 10 pA). The intersection of these lines with the yield curve determines the value of the positive surface potential within the irradiated area.");sQ1[176]=new Array("../7337/0351.pdf","Development of a Monochromated andAberration-Corrected Low-Voltage (S)TEM","","351 doi:10.1017/S143192761500255X Paper No. 0176 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Development of a Monochromated and Aberration-Corrected Low-Voltage (S)TEM Masaki Mukai1, Shigeyuki Morishita1, Atsushi Kimura1, Akihiro Ikeda1, Kazunori Somehara1, Hidetaka Sawada1, Luiz H. G. Tizei2, Yung-Chang Lin2, Koji Kimoto3 and Kazu Suenaga2 1. 2. JEOL Ltd., 3-1-2 Musashino, Akishima, Tokyo 196-8558, Japan Nanotube Research Center, National Institute of Advanced Industrial Science and Technology (AIST), Central 5, 1-1-1 Higashi, Tsukuba, Ibaraki 305-8565, Japan 3. National Institute for Materials Science (NIMS), 1-1, Namiki, Tsukuba, Ibaraki, 305-0044, Japan Low-voltage scanning/transmission electron microscope (STEM/TEM) equipped with delta-type aberration correctors [1] was developed under a project "Triple-C phase-1" to observe the atomic structure of carbon materials with less knock-on damage. This microscope enables us to observe and analyze defected structures of graphene edge by using electron energy-loss spectroscopy (EELS) at an atomic scale [2]. However this microscope was equipped with a cold field emission gun to obtain high brightness therefore its energy resolution remains to be approximately 0.3 eV. To study the electronic structures of materials in detail, we have developed a monochromator working at 15 - 60 kV for a lowvoltage aberration-corrected microscope under a project "Triple-C phase-2", whose targeted energy resolution is better than 25 meV. The developed microscope, which is based on JEM-ARM200F, is equipped with a high resolution EELS (Quantum-ERS optimized in low-voltage, GATAN Inc.) and delta-type aberration correctors for STEM and TEM. The optical system for the monochromator is chosen to be a double Wien-filter. The Wienfilter arranged between the extraction anode of Schottky source and the accelerator, as we developed in the previous design [3]. After the monochromator, the electron probe is achromatic and the energy spread is controlled by the width of the slit, independently on the probe size at the specimen. In addition, the setting of the monochromator and the electron trajectories inside the monochromator are independent of the change of the accelerating voltage, since the monochromator is located before the accelerator and the potential along the optical axis inside the monochromator is kept constant. The ultimate energy resolution with acquisition time of 2 ms was obtained to be 14 meV with a 0.1mwidth slit at 30 kV as shown in Fig.1 (a). Figure 1(b) shows the low loss spectrum from Hexagonal BN obtained with an energy resolution of 22 meV and an acquisition time of 300 ms at 30 kV. This spectrum, which was measured with a probe size of about 1 nm, a beam current of about 10 pA, a convergence semi-angle of 15 mrad and a collection semi-angle of 30 mrad, showed a sharp peak corresponding to an optical phonon at about 170 meV. These results suggest that the microscope enables us to analyze materials with very high energy resolution 25 meV in nanometer scale, owing to a large scattering cross-section and small specimen damage by using lower voltage electrons. Figure 2 shows a monochromated and aberration-corrected TEM image of single-layered graphene obtained with an energy spread of 172 meV at 60 kV. This image implies that the monochromated electron source in aberration-corrected TEM provides clearly resolved C-C bonds and the enhancement of spatial resolution arising from a small chromatic aberration in TEM at low accelerating voltage. This work is supported by Japan Science and Technology agency, Research Acceleration Program. References [1] H. Sawada, et al.: J. Electron. Microsc. 58 (2009) 341. [2] K. Suenaga and M. Koshino, Nature 468 (2010) 1088. [3] M. Mukai, et al.: Ultramicroscopy 140 (2014) 37. Microsc. Microanal. 21 (Suppl 3), 2015 352 Fig. 1. (a) Intensity profile of the zero-loss peak with the energy resolution of 14 meV, recorded using a slit of 0.1m wide and an acquisition time of 2 ms at 30 kV. (b) Low loss spectrum and its magnified spectrum from hexagonal BN obtained with an energy resolution of 22 meV and an acquisition time of 300 ms at 30 kV, showing a signal at 170 meV as indicated by an arrow. Fig. 2. TEM image of single-layered graphene obtained with an energy spread of 172 meV at 60 kV by the monochromated and aberration-corrected microscope.");sQ1[177]=new Array("../7337/0353.pdf","Demonstration of 40kV TEM Diffraction of Graphite for Structural Analysis","","353 doi:10.1017/S1431927615002561 Paper No. 0177 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Demonstration of 40kV TEM Diffraction of Graphite for Structural Analysis 1 1,2 Andrew Barnum and Jun Jiao 1. Physics Department, Portland State University, Portland, OR 97201, USA 2. Mechanical and Materials Engineering, Portland State University, Portland, OR 97207, USA Rapid structural assessment of newly synthesized nanomaterials is of tantamount importance both to exploratory researchers and industries seeking integration with consumer products. Most nanomaterials are susceptible to radiation damage during observation in the transmission electron microscope (TEM), causing propagation of dislocations, altering crystal phases, and destruction of the lattice structure through amorphization. Lowering the observation voltage in the TEM can result in less damage, though at the expense of decreased resolution, for many nanomaterials [1]. For microscopes not equipped with aberration correctors, such as the Tecnai F20 used here, an information space roughly equivalent to 200kV operation can be achieved at 40kV through diffraction, greatly increasing the length of time for observation and the certainty of the observed structural details. To ensure stable observation, nanographite was dissolved in a 0.75% formvar solution in chloroform, based on known dissolution rates [2]. Two drops from a Pasteur pipette were placed in rapid succession onto the surface of a beaker filled with distilled water and the chloroform allowed to dry � water tension naturally pulled the drying formvar into an electron transparent film. The films were pulled off and treated according to standard methods afterward [3]. During prolonged observation, the films proved highly susceptible to heating damage evaporation of 3nm of amorphous carbon onto the beam exit side alleviated the problem. Observation of these samples was carried out first at 200kV, where the 0.34nm graphite lattice was easily observed. During prolonged observation, the expected degradation of the crystal structure occurred. In diffraction mode, this degradation was readily apparent as a continuous loss in intensity within the diffraction rings in less than a minute. Repeating the observation at 40kV allowed an equal quantity of primary diffraction reflections to be observed, but with observation times exceeding 20 minutes. Intensity integration of the diffraction rings was accomplished by first applying a logarithmic function to the pixel values of the images, followed by background subtraction [4]. The indexed intensity plots, presented alongside the diffraction patterns in Figures 1a and 1b, demonstrate that little information is lost in reciprocal space during the transition to a 40kV, despite the inability to image lattice fringes in real space. One surprising benefit of lower voltage was an increase in the ability to resolve small peaks. Due to a decrease in camera saturation for otherwise equivalent gun and beam settings, the (100) and the much smaller (111) peak are well separated at 40kV at 200kV, deconvolution is required to fully separate and index the (111) contribution. The method presented here for 40kV diffractive observation promises to provide excellent structural resolution for many beam sensitive materials. Though requiring both careful alignment of the microscope and delicate sample preparation, this method permits increased observation times for carbonbased compounds, ranging from the inorganic, such as those described above, Microsc. Microanal. 21 (Suppl 3), 2015 354 as well as organic crystals all are especially well suited to analysis at this low voltage, without accompanying concerns for radiationrelated structural changes. References: [1] StogerPollach, M. "A Short Note on How to Convert a Conventional Analytical TEM into an Analytical Low Voltage TEM." Ultramicroscopy , 2014, 9497. [2] arcus, Yizhak. "Solubility of C60 Fullerene." M Journal of Physical Chemistry B 13 , no. (2001): 24992506. [3] http://www.2spi.com/catalog/grids/laceycarbonformvar.html. Accessed Jan 15th, 2015. [4] Vainshtein, B. K. Structure Analysis by Electron Diffraction . Burlington: Elsevier Science, 1964. Figure 1: (a) Nanographite diffraction pattern acquired at 200kV. (b) Nanographite diffraction pattern acquired at 40kV. Both patterns have been contrast inverted for ease of examination.");sQ1[178]=new Array("../7337/0355.pdf","Strategies for high-resolution imaging of radiation-sensitive materials in an aberration-corrected transmission electron microscope","","355 doi:10.1017/S1431927615002573 Paper No. 0178 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Strategies for high-resolution imaging of radiation-sensitive materials in an aberration-corrected transmission electron microscope Ute A. Kaiser Ulm University, Group of Electron Microscopy of Materials Science, Ulm, Germany. The practical realisation of aberration correction enabled low-voltage and high-resolution transmission electron microscopy which is relevant for structural, chemical and electronic analysis of such electron radiation-sensitive materials, which are suffering predominantly on displacements of atoms by elastic binary collisions between the beam electrons and the sample atoms, the so called knock-on damage. One main aim of the SALVE (sub angstrom low-voltage electron microscopy) project [1] is to explore the possibilities of high-resolution imaging of the pristine structure of a biological object and of those present and future materials, which are of low dimensionality or/and bridging the former frontier between materials and life sciences using a 20-80kV-optimised FEI microscope platform with spherical and chromatic aberration correction. In the first part of the talk we will briefly outline the current state of the SALVE microscope and show theoretical predictions for phase contrast imaging using our corrector with positive and slightly tunable fifth-order spherical aberration (C5), which is beneficial for high phase contrast if combined with a suitable small negative spherical aberration of third order (C3) and positive defocus (C1) as then the linear and non-linear contrast contributions often simply add up. In the second part of the talk we report on structural properties of low-dimensional materials obtained at 80kV acceleration voltage. In particular we investigate basic crystallographic defects such as zerodimensional defect [2,3], one-dimensional defect [3,4], extended defects [3] and determine � via density functional theory calculations � their properties. We briefly present our numerical post-processing method for removing the effect of anti-symmetric residual aberrations in high-resolution transmission electron microscopy (HRTEM) images of weakly scattering 2D-objects [5].We show moreover that now even the atomic structure of an amorphous phase can be unravelled in direct space just from the analysis of high-resolution TEM images [6,7]. We demonstrate that multiple channels of damage production have to be identified as not only knock-on damage plays an important role but also damage via excitations of the electronic system of the sample, i.e., ionization or radiolysis, chemical etching by radicals produced from impurities on the sample surfaces, and heating via deposition of energy from the electron beam can be responsible if the atomic structure of a sample is degraded. As results need to be confirmed by image calculation, for high-resolution TEM images, the contribution of inelastic scattering must be taken into consideration [8]. For energy-filtered images of low-Z materials at low voltages, the contributions of elastic and inelastic scattering to the image intensity cannot be separated from each other because the inelastic scattering amplitudes are influenced by elastic scattering, and vice versa [9]. All the damage channels can be suppressed by simply limiting the total electron doses on the samples. On the other hand, limited electron doses result, however, inevitably to worse signal to noise ratios, which imposes a lower bound on the electron dose with which the features of interest in the studied sample can still be discerned. Here we show an quantitative approach for estimating the visibility of objects in TEM images with limited doses [10], to sandwich beam-sensitive objects in-between two graphene layers [11,12], and how to get graphene clean [13]. We demonstrate that lowering the energy of the electrons down to 20kV prevents various metal clusters and molecule inside CNTs from electronbeam-stimulated damage [14]. The exchange of the isotopes of the molecules (deuterium instead of Microsc. Microanal. 21 (Suppl 3), 2015 356 hydrogen) is another strategy that enhances the stability against knock-on collisions independent on the voltage [15]. We discuss in more detail the discovery of new structures such as 2D square ice [16] and crystalline AuC [17]. In the third part we show 40kV EELS experiments for two-dimensional materials with exceptionally low background noise and explore the nature of electronic excitations by inspecting the low losses in the 0 � 20 eV range. In fact, information can be obtained, under controlled illumination conditions and sample orientation, to the onset of optical transitions. In addition to the energy of electronic excitations, information on the momentum transfer can be obtained in dependence of the crystallographic direction. We determine the dispersion behaviour for and + plasmons in free-standing single-layer graphene and multilayers as benchmark experiments [18] confirming earlier calculations. [1] U. Kaiser, J. Biskupek, J.C. Meyer, J. Leschner, L. Lechner, H. Rose, M. St�ger-Pollach, A.N. Khlobystov, P. Hartel, H. M�ller, M. Haider, S. Eyhusen and G. Benner, "Ultramicroscopy, 111 (2011) 1239 [2] J. Kotakoski, J. Meyer, S. Kurasch, D. Santos-Cottin, U. Kaiser and A. Krasheninnikov, Physical Review B, 83 (2011) 235420 [3] O. Lehtinen, H-P Komsa, A. Pulkin, M. B. Whitwick, M.-W Chen, T. Lehnert, M. Mohn, O. V. Yazyev, A. Kis, A. V. Krasheninnikov, U.Kaiser, ACSNano accepted (2015) [4] O. Lehtinen, S. Kurasch, A.V. Krasheninnikov and U. Kaiser, Nature Communications 4 (2013) 3098 [5] O. Lehtinen, D. Geiger, Z. Lee, M. B. Whitwick, M.-W. Chen, A. Kis, and U. Kaiser Ultramicroscopy (2015) 10.1016/j.ultramic.2014.09.010 [6] P. Y. Huang, S. Kurasch, A. Srivastava, V. Skakalova, J. Kotakoski, A. V. Krasheninnikov, R. Hovden, Q. Mao, J. C. Meyer, J. Smet, D. A. Muller, and U. Kaiser Nano Lett. 12(2) (2012)1081 [7] P. Y. Huang, S. Kurasch, J.S. Alden, A. Shekhawat, A.A. Alemi, P. L. McEuen, J.P. Sethna, U. Kaiser, D.A. Muller, Science 342 (2013) 224 [8] Z. Lee, J.C. Meyer, H. Rose and U. Kaiser, Ultramicroscopy 112 (2012) 39 [9] Z. Lee, H. Rose, R. Hambach, P. Wachsmuth and U. Kaiser, Ultramicroscopy 134 (2013) 102 [10] Z. Lee, H. Rose, O. Lehtinen, J. Biskupek, U. Kaiser, Ultramicroscopy 145 (2014) 3 [11] G. Algara-Siller, S. Kurasch, M. Sedighi, O. Lehtinen and U. Kaiser, Appl. Phys. Lett. 103 (2013) 203107 [12] O. Lehtinen, I.L. Tsai, R. Jalil, R.R. Nair, J. Keinonen, U. Kaiser, I.V. Grigorieva, Nanoscale 6, (2014) 6659 [13] G. Algara-Siller, O. Lehtinen, A. Turchanin, U. Kaiser, Appl. Phys. Lett. 104 (2014) 153115 [14] T. W. Chamberlain, T. Zoberbier, J. Biskupek, A. Botos, U. Kaiser, A. N. Khlobystov, Chemical Science 3 (2012) 1919 [15] T. W. Chamberlain, J. Biskupek, S. T. Skowron, P. A. Bayliss, E. Bichoutskaia, U. Kaiser, A. N. Khlobystov, Small 11 (2014) 510 [16] G. Algara-Siller, O. Lehtinen, F.C. Wang, R. R. Nair, U. Kaiser, H. A. Wu, I. V. Grigorieva, A. K. Geim, Nature (2015) accepted [17] B. Westenfelder, J. Biskupek, J. Meyer, S. Kurasch, X. Lin, F. Scholz, A. Gro�, U. Kaiser, Scientific Reports (2015) accepted [18] P. Wachsmuth, R. Hambach, M.K. Kinyanjui, M. Guzzo, G. Benner, U. Kaiser, Phys. Rev. B B 88 (2013) 075433 [19] Fruitful cooperation within the SALVE project and financial support by the DFG (German Research Foundation) and by the Ministry of Science, Research, and the Arts (MWK) of Baden-W�rttemberg are gratefully acknowledged.");sQ1[179]=new Array("../7337/0357.pdf","Investigation into Solute Stabilizing Effects in Nanocrystalline Materials: An Atom Probe Characterization Study","","357 doi:10.1017/S1431927615002585 Paper No. 0179 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Investigation into Solute Stabilizing Effects in Nanocrystalline Materials: An Atom Probe Characterization Study Gregory B. Thompson1, Monica Kapoor1, Tyler Kaub1, Brad Boyce2, Blythe Clarke2, Kris Darling3, Peter Felfer 4, and Julie Cairney 4 1 2 The University of Alabama, Tuscaloosa, AL 35487 USA Sandia National Laboratories, Albuquerque, NM 87123 USA 3 Army Research Laboratory, Adelphi, MD 20783 USA 4 The University of Sydney, Sydney, NSW 2006 AUS Over the past few years, there has been a concerted interest in understanding how solute segregation to grain boundaries stabilizes nanocrystalline materials against grain growth and stress effects [1-3]. To elucidate this behavior, the ability to quantitatively probe the chemistry of the grain boundary is essential. Atom probe tomography is ideally suited to achieve this level of atomic scale chemical analysis. This talk will address how atom probe is providing insights into solute segregation that leads to a variety of different nanocrystalline stabilization conditions. The first case study involves the influence of adatom mobility during the growth of a thin film. Thin film stress evolves from an initial compressive state created as adatoms nucleate and form embryonic islands to a tensile state as these islands coalescence. For post-coalescence films, the stress can retain the tensile state for low mobility adataoms or become compressive for higher mobility adatoms. To date, the governing mechanisms for post-coalescence compressive stress evolution has not been well understood. Chason et al. [4] have proposed that excess adatoms in the grain boundaries create this post-coalescence compressive stress condition; to date, there has been little direct experimental evidence characterizing these excess atoms in the grain boundaries. A series of segregating alloy films, with different enthalpies of segregation, have been sputtered deposited. For a weakly segregating system, such as Cu(Ni), we have found that a small amount of solute is sufficient to dramatically increase the compressive stress, Figure 1, but whose effects are limited once grain boundary compositional saturation has occurred. Using interfacial excess atom probe reconstructions refined by Felfer et al. [5], we have quantified solute segregation at these boundaries to be max ~ 0.9 atoms/nm2 which will be elaborated on in how this intrinsic segregation regulates the overall film stress states. In a synergistic study, we are investigating how solutes in similar boundaries can stabilize the grains themselves from growth with increasing temperature. Cu-Nb is an anti-segregating system which has been used to address these thermal stability effects as an extreme case experimental example. A series of `bulk' ball milled Cu(Nb)X alloys, where X varied from zero to ten at.%, have been processed and annealed. The nanocrystalline grains were shown to be stabilized up to 400oC, where upon evident clustering for the low solute concentrations was noted, Figure 2. Increasing the temperature resulted in phase separation for the higher alloy concentrations. These atom probe results are discussed in terms of the present thermodynamic [1] and mechanical stability [6] concepts of solute segregated stabilizing effects. Microsc. Microanal. 21 (Suppl 3), 2015 358 References: [1] H. A. Murdoch, C. A. Schuh. Stability of binary nanocrystalline alloys against grain growth and phase separation. Acta Materialia 61 (2013) 2121-2132. [2] F. Tang, D.S. Gianola, M.P. Moody, K.J. Hemker, J.M. Cairney, "Observations of grain boundary impurities in nanocrystalline aluminum and their influence on microstructural stability and mechanical behavior," Acta Materialia 60 (2012), 1038 [3] B. Fu, G. B. Thompson. Compositional dependent thin film stress states. Journal of Applied Physics 108 (2010) 043506-1-6. [4] E. Chason, B. Sheldon, et al. Origin of compressive residual stress in polycrystalline thin films. Physical Review Letters 88 (2002) 156103-1-4. [5] P. Felfer, A. Ceguerra, S. Ringer, J. Cairney. Applying computational geometry techniques for advanced feature analysis in atom probe data. Ultramicroscopy 132 (2013) 100-106 [6] Darling K. A. Darling, B. K. VanLeeuwen, et al. Stabilized nanocrystalline iron-based alloys: Guiding efforts in alloy selection. Materials Science and Engineering: A 528 (2011) 4365-4371. [7] GBT and TK acknowledge the AR0-W911NF1310436 grant; MK, BB and BC recognize The U.S. Department of Energy, Office of Science, Basic Energy Sciences, Materials Sciences and Engineering Division for support. Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000 Figure 1: (a) In situ growth stress plot of Cu(Ni) films at various Ni solute content levels. Note that at 5 at.% Ni the compressive stress was the largest and is contributed to the segregation of Ni to the grain boundaries. As the Ni content increased, atom probe revealed that the grain boundaries were solute saturated and the Ni was incorporated into the Cu matrix grains. (b) Atom map of Cu-5at.%Ni Figure 2: Atom map of Cu-1at.%Nb annealed at 400oC revealing Nb clustering at the grain boundaries.");sQ1[180]=new Array("../7337/0359.pdf","Precipitation in Highly Supersaturated Al-Sc-V, Al-Sc-Nb, and Al-Sc-Ta Alloys During Isochronal Aging","","359 doi:10.1017/S1431927615002597 Paper No. 0180 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Precipitation in Highly Supersaturated Al-Sc-V, Al-Sc-Nb, and Al-Sc-Ta Alloys During Isochronal Aging Keith E. Knipling1 1. U. S. Naval Research Laboratory, Multifunctional Materials Branch, Washington, DC 20375 Conventionally solidified Al-Sc alloys, strengthened by nanoscale Al3Sc (L12 structure) precipitates, have excellent coarsening and creep resistance to 300 �C [1,2], and can be improved up to 400 �C with ternary additions of the neighboring Group 4 elements, Ti [3] Zr [4,5] or Hf [6]. These ternary solutes have a much smaller diffusivity than Sc [1], resulting in Al3Sc1-xMx (M = Ti, Zr, or Hf) precipitates with a Sc-rich core enveloped in a Zr, Ti, or Hf-enriched shell. These slower-diffusing atoms limit coarsening and, since they substitute for Sc in the precipitates, can also reduce the relatively high cost of Sc additions. The Group 5 elements in the periodic table (M = V, Nb, or Ta) may also be beneficial alloying additions to Al-Sc alloys. They each form an Al3M trialuminide and also exhibit some solubility in Al3Sc [7], and are anticipated to be much slower diffusers than Ti, Zr, or Hf [1], potentially providing better thermal stability than Al-Sc-Ti, Al-Sc-Zr, or Al-Sc-Hf alloys. This study investigates the nanostructures and compositions of Al3Sc1-xMx precipitates formed in highly supersaturated Al-Sc-V, Al-Sc-Nb, and Al-ScTa alloys during isochronal aging from 200 to 500 �C. The Al-0.4Sc-0.4V, Al-0.4Sc-0.4Nb, and Al-0.4Sc-0.4Ta (all compositions are in at.%) alloys were first dilution cast from binary master alloys using non-consumable electrode arc-melting in a high-purity argon atmosphere. After arc-melting, a ~7 g portion of each ingot was melt-spun by induction melting in a quartz crucible under a 34 kPa helium atmosphere and ejecting the melt through a 1.0 mm orifice onto the perimeter of a copper wheel rotating at a surface speed of 20 m s-1. The ribbons produced were typically 3�5 mm wide, 50 �m thick, and several centimeters long. Prior to aging, the ribbons were encapsulated in quartz ampoules that had been evacuated and backfilled with high-purity argon to ~34 kPa. The encapsulated alloys were aged isochronally in 25�C increments lasting 3 h each, starting at 200�C and terminating at 500�C. The quartz ampoules were water-quenched between each aging increment. Since the solute concentrations in these alloys far exceed their equilibrium maximum solid solubilities, the specimens were not solution treated prior to precipitation aging. The microstructures that formed upon aging were investigated by atom-probe tomography (APT) with a Cameca LEAP 4000x Si. Specimens for APT were prepared using standard lift-out and focused ion beam (FIB) milling procedures [8] in an FEI Nova 600 NanoLab DualBeamTM SEM/FIB. An atom-probe reconstruction of the Al-Sc-V alloy aged to 500 �C is displayed in Figure 1(a), with each Sc atom represented as a blue pixel, V atoms with red pixels, and Si atoms with gray pixels (Al atoms are not shown for clarity). The presence of Si is a byproduct of melting each ingot in a quartz crucible during melt spinning. The APT analysis shown partially intercepted three ~12�15 nm diameter Al3Sc1xVx precipitates. While virtually all of the Sc segregates to the precipitates, a significant amount of V remains in -Al solid-solution, as evidenced by the predominance of red V atoms in the matrix surrounding the precipitates. This is conveyed quantitatively in Figure 1(b), which is a proxigram [9] displaying average solute concentration profiles in the -Al matrix and Al3Sc1-xVx precipitates with respect to a 2.5 at.% Sc isoconcentration surface delineating the two phases. Sc, V, and Si segregate to Microsc. Microanal. 21 (Suppl 3), 2015 360 the Al3Sc1-xVx precipitates. As in prior APT studies on Al-Sc-Ti, Al-Sc-Zr, and Al-Sc-Hf alloys [3�6], the V atoms constitute a small fraction of the precipitate and are segregated at the -Al/Al3Sc1-xVx precipitate interface, suggesting that V is much slower diffuser than Sc. The -Al solid-solution contains 0.0145(5) at.% Sc, 0.675(3) at.% V, and 0.356(2) at.% Si. While the solubility of V in -Al is not known below 600 �C, the equilibrium maximum solid solubility is 0.33 at.% V at 662.1 �C [1], which is less than half the concentration of V measured in Figure 1(b). Thus, the -Al solid-solution is metastably supersaturated in V, indicating that equilibrium has not yet been reached after aging to 500 �C. This is consistent with the expected extraordinarily sluggish diffusion kinetics of the Group 5 solutes in Al. References: [1] K Knipling et al, International Journal of Materials Research 97 (2006), p. 246. [2] J Royset and N Ryum, International Materials Reviews 50 (2005), p. 19. [3] M van Dalen et al, Acta Materialia 53 (2005), p. 4225 [4] C Fuller et al, Acta Materialia 51 (2003), p. 4803. [5] C Fuller and D Seidman, Acta Materialia 53 (2005), p. 5415. [6] H Hallem et al, Materials Science and Engineering A 421 (2006), p. 154. [7] Y Harada and D Dunand, Materials Science and Engineering A 329 (2002), p. 686. [8] K Thompson et al, Ultramicroscopy 107 (2007), p. 131. [9] O Hellman et al, Microscopy and Microanalysis 6 (2000), p. 437. [10] This work was funded by the Naval Research Laboratory under the auspices of the Office of Naval Research. Dr. Nhon Q. Vo (nanoAl) is thanked for his assistance with thermal aging of the alloys. Figure 1. (a) APT reconstructions of the Al-Sc-V alloy isochronally aged to 500 �C, containing three ~12�15 nm diameter Al3Sc1-xVx precipitates. (b) Proxigram displaying the distributions of Sc, V, and Si atoms in the Al3Sc1-xVx precipitates and the -Al solid-solution.");sQ1[181]=new Array("../7337/0361.pdf","Co-deformation of crystalline-amorphous nanolaminates","","361 doi:10.1017/S1431927615002603 Paper No. 0181 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Co-deformation of crystalline-amorphous nanolaminates Wei Guo1,2, Jiahao Yao3, Eric A. J�gle1, Pyuck-Pa Choi1, Dierk Raabe1 1. Department of Microstructural Physics and Alloy Design, Max Planck Institut f�r Eisenforschung, D�sseldorf, Germany 2. Microscopy Center for Nanophase Materials Science, Oak Ridge National Laboratory, Oak Ridge, USA. 3. Shenyang National Laboratory for Materials Science, Institute of Metal Research, Chinese Academy of Sciences, Shenyang, China Deformation of ductile crystalline-amorphous nanolaminates is not clearly understood due to the complex interplay of interface mechanics, shear banding and deformation-driven chemical mixing. In this work, we synthesized and indented model nanolaminates consisting of nanocrystalline Cu and amorphous CuZr (Fig1a-c). In order to study both, nanostructural and atomic-scale chemical deformation effects at the same specimen position we performed a joint analysis by transmission electron microscopy (TEM) and Atom Probe Tomography (APT). For the correlative TEM-APT analysis, tip-shaped specimens containing deformation shear bands were prepared by site-specific preparation, mounted onto an electro-polished bisected TEM Cu grid, and thinned by annular FIB milling. APT was performed with a local electrode atom probe (LEAP 3000X HR). Samples were analyzed at a base temperature of 60 K, applying 532 nm wavelength 10 ps laser pulses of 0.4 nJ with at a repetition rate of 250 kHz. The datasets were reconstructed and analyzed using the software IVAS 3.6.6 (CAMECA Instruments). Fig. 1 reveal zones with a large and very sharp, abruptly sheared Cu layer offset >30nm. The layer is not completely disconnected from the sheared region though. Instead, it is heavily deformed at a 70.5� tilt angle to the interface plane, which corresponds to the angle of the {111} planes in fcc Cu. In the shear band region, the Cu layers are subjected to a huge true shear strain of 4.0 � 0.2, as calculated from the Cu layer displacement. The observations indicate dislocation slip on {111} planes in the Cu layer. The large displacement (>30nm) of the sheared and hence fragmented Cu layer shown in Fig. 1e suggests that large numbers of Cu dislocations accommodate the shear inside the crystalline phase, extending it across the Cu/CuZr interface into the initially amorphous layer. We have probed several compositional profiles across the Cu enriched region inside the shear band using small sampling volumes (size: 5�5�15 nm3). This analysis reveals that deformation-induced mixing in the sheared regions of the Cu layer is caused by the shear band running through it: In this mixing zone the Cu concentration varies from 64.6 at.% to 95.4 at.% within a narrow shear layer thickness range of only 2~3 nm. Regions outside of the shear band have the same interface width and chemical concentration profiles as the as-deposited layers, i.e. they do not undergo deformation-driven mixing. Fig. 2 shows TEM obtained from the sheared offset region of shear bands in the CuZr/Cu multilayer specimen. NBD patterns taken along the shear band penetrating the initially amorphous CuZr layer show both, an inner halo ring (amorphous material) as well as some weak discrete diffraction spots Microsc. Microanal. 21 (Suppl 3), 2015 362 (crystalline material) outside of the inner halo ring (see small arrows, NBD spots 3,4,6). This indicated that in the shear banded Cu-enriched zones the initially amorphous CuZr forms an amorphous plus crystalline nanocomposite. The present observation can explain serveral co-deformation and shear�induced mixing phenomena: (i) Local thinning of the Cu phase can be attributed to the high density of gliding dislocations. (ii) Cu atoms are dragged across the hetero-interface when dislocations release their shear step into the amorphous CuZr phase. The associated Cu enrichment can lead to local crystallization. (iii) Besides such dislocation driven solute mixing, nm-sized portions of crystalline Cu are in compact form displaced into the amorphous CuZr phase. References: [1] W Guo et al, Phys. Rev. Lett. 113 (2014) 035501. [2] W Guo et al, Mater. Sci. Eng A. 628 (2015) 269-280. [3] The authors acknowledge fruitful discussions with Dr. Daniel Haley and Dr. Michael Herbig. Figure 1. Correlative TEM/APT analysis of a shear band in amorphous CuZr / nanocrystalline Cu nanolaminates. Figure 2. TEM observation of cross-phase shear bands in 100 nm CuZr / 10 nm Cu nanolaminates.");sQ1[182]=new Array("../7337/0363.pdf","An Experimental and Simulation Studies of a High Strain-Rate Deformation Shear Band in a High-Nickel Steel","","363 doi:10.1017/S1431927615002615 Paper No. 0182 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 An Experimental and Simulation Studies of a High Strain-Rate Deformation Shear Band in a High-Nickel Steel Sung-Il Baik,1,2 Dieter Isheim,1,2 Divya Jain,1 R. K. Gupta, 3 K.S. Kumar,3 David N Seidman1,2,* Department of Materials Science and Engineering, Northwestern University, Evanston, IL 60208, USA. 2. Northwestern University Center for Atom Probe Tomography (NUCAPT), 2220 Campus Drive, Evanston, IL 60208, USA 3 School of Engineering, Brown University, Providence, RI 02912, USA A high Ni-steel was developed for both high-strength and -toughness for cryogenic temperature applications [1]. High-Ni steels are used for severe working conditions, such as blast-resistant naval ship hulls or decks and liquefied natural-gas (LNG) tanks. Even though this particular high-Ni steel possesses excellent static properties, a thermoplastic instability in the form of localized shear bands is a major concern for high-speed deformation conditions. The adiabatic shear-band (ASB) is a specific type of deformation in a narrow zone that occurs at a high strain-rate with a significant localized temperature increase [2]. There isn't sufficient time for thermal diffusion from a local volume, resulting in a significant temperature increase in a localized zone. This results in the suppression of work hardening and localized softening occurs, which can cause material failures. Atomic-scale observations indicate grain segmentation, growth and solute redistribution at ASBs, which are caused by a severe deformation and local temperature increase on a short time-scale [3]. The interaction between solute atoms with dislocations and grain boundaries (GBs) is the source of the instability of mechanical properties. The temperature increase causes segregation of solute atoms at dislocations and GBs, and the supersaturation of solute may be sufficiently high for precipitates to nucleate. The current research explores the ferrite-austenite phase transformation and atomic-scale redistribution of solute atoms at an ASB. The ASB is analyzed by correlative experiments: electron backscatter diffraction (EBSD), transmission electron microscopy (TEM), atom-probe tomography (APT), and the thermodynamic and kinetic calculations. The material studied has a nominal composition of 0.32C-0.04N-0.61Mn-0.41Si-0.13Cu-0.64Cr-0.64Mo-0.08V-0.005Nb-9.75Ni- Bal. Fe (at.%). The dynamic compression deformation tests with strain rates of 2800 s-1, employing a split-Hopkinson confirmed that this alloy forms ASBs. Fig.1 displays the results of correlative experiments employing SEM, TEM, EBSD, and 3-D APT to study a specific region of interest of a shear band, failure area or crack tip in the high-Ni steel. The width of the shear band in front of the crack tip is about 10 m as observed by SEM, Fig.1(a). The crack has propagated along the ASB. The local temperature increase, with a hot-spot region developing in the ASB, facilitates nucleation of voids and ductile fracture occurs within the ASB. Fig.1(b,c) display a plan-view bright-field (BF) TEM image and a high-resolution EBSD map in the ASB region after using the dual-beam FIB microscope lift-out preparation technique to target the rectangular region indicated in Fig.1(a) for analysis. These TEM and EBSD analyses illustrate the change of grain shape across the ASB: from elongated lamellar grains to equiaxed grains with high misorientation angle as the center of the shear band is approached. This transition from elongated lamellar grains to equiaxed small grains is indicated by a white or blue dotted line in Fig. 1(b) or (c), respectively. The breakdown of deformed elongated grains into smaller grains or subgrains is an important process in ASB formation. Fig. 1(d,e) is a correlative analysis by TEM and 3-D APT of equiaxed small grains in the core of an ASB. TEM and 1. Microsc. Microanal. 21 (Suppl 3), 2015 364 3-D APT analyses indicate the redistribution of solute atoms during the ASB deformation. The analyses intersect the dark grains in the TEM image, which are enriched in carbon (black dots in (e)) and other austenite stabilizing elements. Vanadium-rich precipitates are also observed in the ferrite matrix in the core of ASB. The size and the density of V-rich nitride precipitate are 5-7 nm diameter and 2.2x1023 m-3 respectively. The thermodynamic and kinetic behavior of VN(C) precipitation was studied by austenite/carbonitride equilibrium model [4]. The precipitation kinetic is accelerated by up to two orders of magnitude faster in a severely deformed conditions during dynamic compression deformation. References [1] G.R. Brophy, A.J. Miller, T Am Soc Metal, 41 (1949) 1185-1203. [2] P. Wang, K.S. Kumar, Mat Sci Eng a-Struct, 519 (2009) 184-197. [3] R.W. Armstrong, S.M. Walley, Int Mater Rev, 53 (2008) 105-128. [4] P. Maugis, M. Goune, Acta materialia, 53 (2005) 3359-3367. [5] This research is funded by the Office of Naval Research under grants N00014-12-1-0425, N000141110681, Dr. W. Mullins, grant manager. The atom-probe tomography measurements were performed at the Northwestern University Center for Atom-Probe Tomography (NUCAPT). The LEAP 4000X-Si tomograph was purchased and upgraded with funding from NSF-MRI (DMR-0420532) and ONR-DURIP (N00014-0400798, N00014- 0610539, N00014-0910781). Fig. 1: The transition of the grain structure from elongated lath grains on a mesoscale and the formation of austenite phase and carbonitride on an atomic scale are found in this correlative study using (a) SEM, (b, d) TEM, (c) EBSD, and (e) APT analyses. Equiaxed small grains and VN-type precipitates are formed during the temperature increase in the course of the ASB formation during high strain-rate deformation. The precipitate behavior is also understood using thermodynamic and kinetic calculation for a steel [4].");sQ1[183]=new Array("../7337/0365.pdf","Characterization of Element Partitioning at the Austenite/Ferrite Interface of as Cast CF-3 and CF-8 Duplex Stainless Steels","","365 doi:10.1017/S1431927615002627 Paper No. 0183 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of Element Partitioning at the Austenite/Ferrite Interface of as Cast CF-3 and CF-8 Duplex Stainless Steels R. Prakash Kolli1, Sarah Mburu1, Daniel E. Perea2, Jia Liu2, Samuel C. Schwarm1, Arielle Eaton2, Sreeramamurthy Ankem1 1. 2. Department of Materials Science and Engineering, University of Maryland, College Park, USA. Environmental and Molecular Sciences Laboratory, Pacific Northwest National Laboratory, Richland, USA. Cast duplex stainless steels are used in cooling water piping of light water reactors (LWRs) due to their combination of strength, high ductility, high impact toughness, corrosion resistance, castability and weldability. This combination is due to the presence of the face-centered cubic (f.c.c.) �austenite and body-centered cubic (b.c.c.) �ferrite phases in the duplex microstructure. The ferrite phase is present in volume fractions less than approximately 20% in grades used in LWR power plants. These alloys experience thermal aging embrittlement during operational service due to spinodal decomposition of the �ferrite phase into chromium-rich (Cr-rich) '�martensite and iron-rich (Fe-rich) �ferrite phases, and possibly nucleation and growth of intermetallic G-phase precipitates or carbides [1,2]. Microstructural evolution and concomitant changes in macroscopic (bulk) mechanical properties lead to increased hardness, a loss of ductility, and a reduction in impact toughness. Interest in extending the operational life of these power plants to 80 years requires examining the complex phase decomposition and corresponding mechanical property changes of these stainless steels by employing accelerated isothermal aging at elevated temperatures relative to the operational temperatures. This requires detailed compositional characterization of the as cast stainless steels in order to provide a baseline reference to quantify the temporally evolving concentration profiles and phase decomposition at the different temperatures. Additionally, the �austenite/�ferrite heterophase interface must be characterized, as possible heterogeneous nucleation and growth of intermetallic precipitates or carbides at this location will significantly influence the local elemental concentration profiles and bulk mechanical properties. Thus, we have characterized the statically cast CF�3, [Fe, 0.02 C, 1.07 Mn, 0.98 Si, 19.69 Cr, 8.4 Ni, 0.28 Mo (wt.%)], and CF�8, [Fe, 0.06 C, 0.99 Mn, 0.97 Si, 19.85 Cr, 8.3 Ni, 0.35 Mo (wt.%)] duplex stainless steel microstructures at two different length scales employing state-of-the-art atom-probe tomography (APT) and energy-dispersive x-ray spectroscopy (EDS). We perform initial characterization of the concentration profiles with the proximity histogram [3] relative to a Ni isoconcentration surface, Figure 1, to illustrate element enhancement in the constituent phases. We illustrate a procedure to quantify the phase compositions from the proximity , where is histogram. The preliminary elemental partitioning ratio of each element i, the concentration of element i in phase j, at the �austenite/�ferrite heterophase interface is calculated. Additionally, we quantify the phase compositions employing elemental maps of each phase from the EDS spectra, Figure 2, and also calculate the quantity . Our initial results indicate that the �ferrite phase is enriched in Cr, whereas the �austenite phase is enriched in Fe and nickel (Ni). Additionally, the high spectral and spatial resolution of the APT technique illustrates that the �ferrite phase is enriched in silicon (Si) and molybdenum (Mo), whereas the �austenite phase is enriched in manganese (Mn). We also illustrate that the carbon (C), boron (B), and phosphorus (P) are enhanced at the � austenite/�ferrite heterophase interface. In addition, we discuss local enhancement and depletion of Fe and Cr at the interface, which is possibly due to diffusion gradients or to local magnification effects. Microsc. Microanal. 21 (Suppl 3), 2015 366 References: [1] HM Chung and TR Leax, Mater. Sci. Tech., 6 (1990), p. 249. [2] F Danoix and P Auger, Mater. Charact., 44 (2000), p. 177. [3] OC Hellman, JA Vandenbroucke, J R�sing, D Isheim and DN Seidman, Microsc. Microanal., 6 (2000), p. 437. [4] The authors acknowledge funding from the Department of Energy, Nuclear Energy University Program, Dr. Jeremy Busby, technical monitor. Figure 1. The �austenite/�ferrite heterophase interface delineated by a 4.5 at.% Ni isoconcentration surface in (a) an atom probe tomography (APT) reconstruction, and (b) the corresponding proximity histogram concentration profiles for Fe (pink), Cr (red), and Ni (green) in a CF-8 as cast duplex stainless error bars are based steel. The ca. 60 � 60 � 60 nm3 reconstruction contains ~5.0 million ions. The on counting statistics. Figure 2. Energy-dispersive x-ray spectroscopy (EDS) elemental maps for (a) Cr (red), (b) Ni (pink), and (c) Fe (teal) of the CF-3 as cast duplex stainless steel illustrating Cr enrichment in the �ferrite phase, and Fe and Ni enrichment in the �austenite phase.");sQ1[184]=new Array("../7337/0367.pdf","Designing Analytical Instrument Control Systems for Longevity and Maximum Upward Compatibility","","367 doi:10.1017/S1431927615002639 Paper No. 0184 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Designing Analytical Instrument Control Systems for Longevity and Maximum Upward Compatibility Edward Principe1, Anthony D. Owens2, William Mershon3 1. 2. Tescan USA, Product Manager FE-SEM and FIB-SEM Products, Warrendale, Pa. USA Tescan USA, Director of Technology, Warrendale, Pa. USA 3. Tescan USA, Applications Manager, Warrendale, Pa. USA Avoiding premature obsolescence of software and hardware which control and communicate with analytical instrument platforms is among the many challenges of instrumentation and facility managers. We will describe design and implementation factors to avoid obsolescence, maximum longevity, upgradeability and reparability. The most common upgrade of SEMs today is the replacement of controlling PCs running legacy operating systems due to security and supportability issues. An architecture that allows all command and data to pass between the SEM hardware and the controlling PC over a small number of standard interfaces is desirable. The use of Ethernet/TCP/IP for command and data as well as the use of USB for all other interfaces makes it possible to use PCs that are not modified by the inclusion of any special hardware. A robust remote control system that functions over Ethernet makes it easier to guarantee future operation with partners such as EDS analyzers. To maintain maximum compatibility with current and future Microsoft WindowsTM operating systems, application and instrument control software should make minimal use (if any) of OS-specific database structures such as the Windows registry. Alternative structures for storing information required by instrument control and data acquisition software include the use file types that are more likely to be compatible with current and future operating systems (e.g. ASCII text files or XML files). Instrument control and data acquisition systems should also be designed to minimize or eliminate the need for software drivers (e.g. instrument designs should avoid hardware components that require the use of drivers). Since drivers are typically OS-specific software elements, the use of such hardware components necessitates writing new software drivers to support this hardware when a new OS is employed. In practice, instrument manufacturers often resist investing the time and money to develop new software/drivers that would be required to support older generation tools. The design of the SEM software user interface can also affect the longevity and ease of upgrading. Most of the changes in software that are required by the need to migrate to newer operating systems come from the behavior of user interfaces. By using the most simple tools available and resisting reliance current fads, it is easier, faster and cheaper to migrate an SEM control application to a newer version of an operating system. The trade-off is not being able to make the interface look new and modern. The benefit is that an interface on WindowsTM 8 (or a future WindowsTM OS) can have very much the same functionality as the same interface on WindowsTM 95 did. Eluding software and hardware "dead-ends" requires the adoption of standard hardware interfaces and application programmer interfaces. While this can be a difficult task, instrument builders should strive to select and implement communications and data transfer interfaces & protocols that are likely to be supported in the future as computing platforms evolve. We believe that TCP/IP interfaces will not only Microsc. Microanal. 21 (Suppl 3), 2015 368 fulfill that requirement, but also provide additional benefits with respect to system longevity, upgradeability, and reparability: TCP/IP will continue as a standard well into the future even as communications/data transfer hardware evolves. TCP/IP provides a high speed interface suitable not only for instrument control and query, but for data acquisition and transfer as well, eliminating the need for 3 rd party or proprietary image acquisition and/or image data storage (`frame-store') hardware. In practice, such hardware has a limited useful lifetime because if inevitably relies on specific computer bus or hardware interface technologies (e.g. ISA, PCI, Firewire, etc.), as well as commonly requiring OS-specific drivers to function. The use of TCP/IP (in essence a pure software interface/protocol) for all instrument control, query, and data acquisition/flow, lends itself to much easier design and implementation of remote clients that can allow user access to instruments over networks as well as remote diagnostic sessions over networks (be these local or wide area networks). The combined implementation of TCP/IP and a software API (SEM server process) allows for much more functional, efficient, upgradeable (and bi-directional) interfaces with integrated OEM components (e.g. EDS,WDS, EBSD systems), and minimizes the need for special hardware interfaces that would be more difficult to support in the future. Finally, we will discuss considerations in electronics design that not only affect instrument longevity, but just as important, affect the time and resource investment (and, therefore, instrument builder's likelihood) to repair or replace electronics that incorporate obsoleted components. Longevity in electronics design is brought about by simplicity. Isolating functions allows the creation of physical modules that have lower parts count, are more compact, and have little or no dependency on the individual performance of other modules. The use of simple low level communications such as I2C, which have been in wide use for over 30 years, make it easy to consider a module a black box instead of a very specific integrated circuit. This allows a single module to be replaced in the field without the need to adjust any other components. The use of simple, commonly available components maximizes the probability that these components will remain available for use or repair well after they have been declared obsolete by their original manufacturer. It also increases the probability that a newer device can be substituted for the original device. If a module needs to be redesigned, then the scope of work is limited making it less costly for the manufacturer to produce, more likely for customers to purchase, and therefore more likely that both will invest in upgrade of older equipment. The use of digital signal processors and field programmable gate arrays (DSP/FPGA) allow for the creation of a software-defined SEM. While this constrains a manufacturer to the use of a product line from a single vendor of processors, backward compatibility of software is a primary concern in the design of future DSPs. This makes it more likely that today's development can be recreated in tomorrow's hardware.");sQ1[185]=new Array("../7337/0369.pdf","Now What am I Going to do.....This is Going to be a Large Data Set.","","369 doi:10.1017/S1431927615002640 Paper No. 0185 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Now What am I Going to do.....This is Going to be a Large Data Set. C. A. Brantner1 1. GW Center for Nanofabrication and Imaging, The George Washington University, Washington, D.C., USA. Many of us have gone digital in our microscopy labs. Our instruments now come with sensitive, sophisticated digital cameras. We can record enormous amounts of data from any given experiment. What are the things that we have to know and do to ensure that we obtain useful data? The instruments/tools that we are using as well as the cameras have many settings that need to be optimized for the collection of data that can be mined for quantification, large area imaging or 3D processing. Our fluorescence/light microscopes, electron microscopes, flow cytometers, x-ray spectrometers, etc. all produce kilobits, megabits, even gigabits and terabits of image and metadata files. Once the data has been collected, now there are questions about how to transmit the files to storage or to an off-site collaborator. Speed and efficiency are required to move large data faithfully. In some cases the policies of the institutions where we work do not have the provisions or the capacities to handle transfer of large data sets from your instruments. It takes collaboration with IT departments and administration to set up the hardware and policies to use your large data sets. Storage is another topic that we know to ask questions about, but what are the questions that are most important to ask? There may be a gigabit of data from today's experiment, which you have the capacity to store, but what of the experiment tomorrow or the next day or from others in the lab? Now there will soon be more data than the hard disc of the computer can handle. What is the next step? An external hard drive? A shared drive? SFTP sites? A cloud account? Large data sets need to be stored in close proximity to the acquisition device to improve the quality control of the transfer. Is an electron microscope facility/core responsible for storing data that is generated there? What policies do you have in place? Data security can be another topic that causes much confusion and anxiety. Again there is the need to work with our institutions' information security people to understand what needs to be done with data as it relates to policy of access. Is encryption needed for the transmission or the storage of the data? Passwords? Firewalls? If so, how will you need to accomplish that? How do you display the data so that you can extract meaningful trends from all of the gigabits that you are now storing? There are many software packages written to handle this. What are the pros and cons of such packages? How can you share data with an off-site collaborator? There are web-based solutions for making data widely available. These have been used for crowd-sourcing that some researchers are using to analyze data. Our speakers in this symposium will tackle these questions and more before the end of the day. References. [1] Ideas from 2014 FOM FIG symposia brain-storming session. [2] Ideas from 2012 NIST Big Data Workshop via discussion with Dr. Anastas Popratiloff. Microsc. Microanal. 21 (Suppl 3), 2015 370");sQ1[186]=new Array("../7337/0371.pdf","Networked Data Storage and Analysis for the Wisconsin Regional Materials Network","","371 doi:10.1017/S1431927615002652 Paper No. 0186 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Networked Data Storage and Analysis for the Wisconsin Regional Materials Network Jon J. McCarthy1, Paul M. Voyles2, Julie A. Last3, Davis R. Bittner4, Ian W. Sadkovich4 1. Wisconsin Materials Institute, University of Wisconsin, Madison WI, USA Materials Science and Engineering Department, University of Wisconsin, Madison WI, USA 3. Materials Science Center, University of Wisconsin, Madison WI, USA 4. Computer-Aided Engineering Department, University of Wisconsin, Madison WI, USA 2. Modern materials characterization instruments produce data in prodigious quantities which require sophisticated, often proprietary software to analyze. Making data accessible and providing access to specialized software is a significant challenge for central materials characterization facilities like the University of Wisconsin-Madison's Materials Science Center (MSC). The traditional MSC strategy for data is to leave it on the computers attached to this instruments for an unspecified amount of time, then delete it and trust that the users have made copies. The strategy for software is for MSC users to install software on their own computers when that is (rarely) possible, or, in the more common case of limited available licenses, for them to access software either on computers attached to instruments or on offline analysis computers maintained by the MSC in the Materials Science and Engineering building. Use of instrument computers for analysis is terribly inefficient, since it blocks the instrument for data acquisition, and the offline analysis computers are overused, out of date, and virus-ridden. The result is fewer people using the instruments at all and only some users taking best advantage of the available analysis tools. Our approach to overcoming these obstacles is to use a networked data analysis (NDA) system, briefly described here. The NDA system provides delivery over the Internet of data analysis applications with access to a central data repository. Figure 1. shows the basic system configuration. Users acquire their data on the instruments, either in person or by Internet-enabled remote access. Data is copied off the instrument computers to a centralized data store. The data store is connected to an analysis software server. Users can connect to the analysis software server from any Windows or Mac computer using any Internet connection faster than 1 Mbit/sec to perform analysis. Raw data and Analysis results are stored on the server, and can be copied to the users' computers over the Internet, or be left in the data store. Analysis software is installed on backend Dell servers running the application virtualization product Citrix XenApp [1]. Remote users are able to use the tools without the struggle of installing and configuring the software. In addition, the servers running the software are collocated with and have high-speed connections to the data to be analyzed. XenApp is currently used to by UW College of Engineering (CoE) Computer Aided Engineering (CAE) organization to deliver access to 18 different engineering applications for educational use by CoE undergraduates, graduate students, faculty, and staff. The XenApp client runs on the user's computer, but the application runs on a CAE-managed Windows server. XenApp delivery is legal for many of the end user license agreements (EULAs) for data analysis applications. User access to the analysis/data server is controlled by a configurable user group which can include University of Wisconsin engineering students, faculty, or staff, researchers from other disciplines, or from outside UW. Outside users cannot access other UW resources like journal subscriptions or email, or other computers on the CoE network. One XenAPP server is capable of supporting between 10 and 50 simultaneous users for typical engineering applications using 24 cores Microsc. Microanal. 21 (Suppl 3), 2015 372 on two processors and 256 MB of RAM. The initial data store consists of 4 TB of high-speed disk storage with redundant off-site backup. Initial storage was provisioned based on one year of data from tomographic (S)TEM, serial section reconstruction from a FIB, a spectral imaging XPS, and several XRD instruments. A typical tomogram is 3 GB of raw data and an additional 3 GB of analysis files, and up to 5 tomograms can be acquired per day. Estimating 100 days of heavy tomography use in a year yields 3 TB of data. XPS and XRD data generation is estimated at 1 TB per year. The system now has 10 XenApp licenses for concurrent users, and 200 licenses for CoE faculty, staff, and student users. Analysis software includes ImageJ, NIH IMOD, XPS Thermo Scientific Avantage package and several diffraction packages from Bruker [2,3,4]. References: [1] XenAPP, www.citrix.com/products/xenapp [2] C.A. Schneider, W.S. Rasband, and K.W. Eliceiri, Nature methods 9 (2012), p. 671. [3] J.R. Kremer, D.N. Kisielowski, and J. R. McIntosh, Journal of Structural Biology 116 (1996), p. 71. [4] The authors acknowledge The Wisconsin Materials Institute for financial support, and Thermo Fisher Scientific and Bruker for software license arrangements. Figure 1. Schematic of the networked data analysis system. Data is acquired on the instruments, then moved to a central data store, connected to the analysis software server. Results can be kept on the data store and copied to the user's computers, as can raw data. 2a 2b Figure 2. 2a) A serial section of a MOS FET collected on a Zeiss 1540 Cross Beam. 2b) 3-D volume view of the Gate section of this FET. Both images from data uploaded to the NDA system and processed with ImageJ2 via XenApp.");sQ1[187]=new Array("../7337/0373.pdf","Strategies for Managing Information Technology (IT) in Microscopy Facilities","","373 doi:10.1017/S1431927615002664 Paper No. 0187 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Strategies for Managing Information Technology (IT) in Microscopy Facilities John Henry J. Scott1 1. National Institute of Standards and Technology, Office of Data and Informatics, Gaithersburg, MD Microscopy and analytical facility managers are faced with an unusually broad and difficult set of challenges, including unrealistic cost-recovery models, a diverse and unevenly-trained population of sponsors and users, a heterogeneous collection of instruments, and the constantly changing landscape of instrument manufacturers. Among the most difficult of these challenges is keeping up with rapid changes in the information technology (IT) sector. Most managers of core imaging facilities or central analytical laboratories are not IT professionals, and they have minimal training in IT infrastructure, advanced networking, digital storage solutions, and IT security practices. Yet their job requires them to find creative solutions to difficult IT problems while operating within tight budget constraints. One of many such problems facing facility managers is the end of support for Microsoft Windows XP [1,2]. This talk is focused on providing useful advice, specific concrete solutions, and general strategies for dealing with IT challenges in the context of shared-use microscopy facilities embedded within both small and large parent organizations. While many of these strategies may be valuable for federal research facilities and relatively well-funded corporate research centers, the target audience is facility directors and staff at smaller laboratories and universities. Practical solutions are favored over theoretical content, the strategies have been chosen because they are effective even when they are not elegant, and the emphasis is on free and open source software and widely available tools instead of expensive commercial platforms. When hardware is required, consideration is given to tight budget constraints and older, obsolete computers are re-purposed when possible. Keeping the IT resources within a facility running smoothly is a high priority for most managers, and several worked examples are included that demonstrate how investing a small amount of time learning a suite of powerful software tools can pay dividends. The Sysinternals [3] suite of tools is covered, including the use of process explorer and process monitor to troubleshoot Microsoft DCOM communication problems between an Energy Dispersive X-ray (EDS) spectrometer computer and an SEM microscope control PC, and it is used to identify hidden file permission problems preventing apparently unrelated actions in light microscope image data analysis. Sysinternals AutoRuns is presented as one of several tools for enumerating the more than two dozen Windows autostart locations; such a tool is invaluable in discovering and identifying adware/spyware and maintaining system performance. The utility of SpeedFan [4], ostensibly a tool for controlling the speed of PC cooling fans, is displayed in a worked example prompted by the diagnosis of intermittent failures on a field emission TEM's control PC motherboard. By inspecting onboard temperature sensors while simultaneously detecting an imminent boot drive hardware failure via the hard drive's SMART interface [5], the hidden root cause of communications faults was revealed and costly downtime avoided. The free AutoIt [6] automation solution is introduced, including simple BASIC-like scripts and automation of graphical interfaces to simplify recurring management tasks on analytical instrument computers. Network attached storage (NAS) solutions, file sync and share services, digital data curation strategies, and other best practices for facility tuning will also be discussed briefly. Microsc. Microanal. 21 (Suppl 3), 2015 374 Finally, several topics in IT security will be addressed from the perspective of the facility manager. Industry best practices and pressure from an organization's Chief Information Officer (CIO) often lead to strict policies designed to improve IT hygiene, minimize data spills, limit the spread of malware, and guard against the rapidly growing threat of ransomware. Justified in part by the ubiquity of infected removable media such as flash drives, these policies are often at odds with the research needs of facility users and the efficient operation of the facility itself, with the facility manager and staff caught in the middle. This talk will provide strategies and options for balancing these needs, including an analysis and explanation of dedicated Research Equipment Networks (RENs) designed to isolate and protect instrument computers running obsolete and legacy operating systems such as Windows NT, while still permitting sufficient network-based data flow to allow end users to retrieve their images and analytical results. Figure 1 shows a typical REN configuration, based around a modest web-based dedicated firewall appliance that any facility manager can build from a discarded Pentium computer, a few $30 surplus network cards, and free and open source tools such as IPCop, pfSense, or m0n0wall [7]. References: [1] Any mention of commercial or free open source products is for information purposes only, and does not imply recommendation or endorsement by NIST. [2] Microsoft's extended support for Windows XP ended on April 8, 2014, and Security Essentials virus definitions and updates for the Malicious Software Removal Tool (MSRT) for XP will be discontinued on July 14, 2015. http://www.microsoft.com/en-us/windows/enterprise/end-of-support.aspx [3] Although originally developed as an independent project, Sysinternals is now available for free from Microsoft TechNet at: https://technet.microsoft.com/en-us/sysinternals/bb545021.aspx [4] SpeedFan is available at http://www.almico.com/speedfan.php [5] SMART, Self-Monitoring, Analysis and Reporting Technology, supported by most hard disk drive and solid state drive manufacturers. [6] AutoIt is available from: https://www.autoitscript.com/site/autoit/ [7] IPCop: http://www.ipcop.org/, pfSense: https://www.pfsense.org/, m0n0wall: http://m0n0.ch/wall/ Figure 1. Network diagram showing an example Research Equipment Network (REN) used to isolate instrument PCs running obsolete operating systems such as Windows XP or Windows 2000.");sQ1[188]=new Array("../7337/0375.pdf","Cryo-EM Methods for the Structural Analysis of Biomimetic Materials based on Peptides and Proteins","","375 doi:10.1017/S1431927615002676 Paper No. 0188 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Cryo-EM Methods for the Structural Analysis of Biomimetic Materials based on Peptides and Proteins Vincent P. Conticello1 1. Department of Chemistry, Emory University, Atlanta, GA, USA. Molecular self-assembly is a fundamental principle of life, with cells having mastered this process to encode incredible diversity of function. Sequence-specific biological macromolecules (i.e., proteins and nucleic acids) interact with very high selectivity within the nanometer to micrometer size regime to create the complex cellular machinery that performs the physico-chemical functions associated with metabolism, signal transduction, replication, and differentiation in living systems. These complex biological machines arise from self-assembly of structurally complementary biomolecules (protomers) on the basis of structural features programmed into polypeptide and polynucleotide sequences at the molecular level. As a consequence of the near-absolute control of macromolecular architecture that results from such sequence specificity, biological structural platforms may have advantages for the creation of well-defined supramolecular assemblies in comparison to synthetic systems, at least at the current state of development for the latter. Thus, the conceptual design of synthetic nano-scale systems can derive significant information from structural investigations of biologically derived supramolecular assemblies and, conversely, biological structural motifs present an attractive target for the synthesis of artificial nano-scale systems on the basis of relationships between sequence and supramolecular structure that have been established for native biological assemblies. Sequence-specific biological materials represent conceptual and structural prototypes for the design of artificial smart materials and will continue to be at the forefront of efforts to develop materials for specialized applications beyond those observed in the native biological context. However, structurally ordered supramolecular materials on the nanometer length-scale are the most challenging to rationally construct and the most difficult to structurally analyze. Cryo-EM (HRSEM and TEM) methods have provided the means to understand the mechanism of interaction between structural elements that guide the formation of higher order structure and, ultimately, functional properties within these biomimetic materials. Structural information from complementary methods (solid-state NMR measurements, small-angle X-ray scattering, and X-ray fiber diffraction) can place structural constraints on the EM analysis, which can assist in the development of structural models facilitate an understanding of the synthetic assemblies. In addition, recent advances in cryo-EM analysis have permitted, in combination with molecular modelling, direct access to near atomic resolution structural information for naturally occurring protein assemblies [1]. Several examples of cryo-EM structural analysis of selfassembled protein- and peptide-based materials are described, which, in combination with complementary methods, allowed the construction of structural models on different length scales that have provided insight into the mechanism of self-assembly. The first example focuses on the creation of synthetic protein-based materials that mimic the selfassembly behavior and elastomeric mechanical response of native elastin. Protein engineering was employed to create synthetic elastin derivatives that comprised blocks of different sequence that selfassembled reversibly from aqueous solutions into protein nanoparticles [2] and nano-textured hydrogels [3]. Cryo-HRSEM of the resultant materials indicated that the nano-scale structural features recapitulated those of native elastic tissue (Figure 1). Chemical sequence control afforded non-native Microsc. Microanal. 21 (Suppl 3), 2015 376 sequences that mimicked native elastin assembly, but could be tailored for specific applications in controlled release and tissue engineering. The second example describes the application of cryo-electron microscopy and helical reconstruction to the structural analysis of helical assemblies derived from short, synthetic peptide sequences [4]. The structures of two assemblies were determined at near atomic-level resolution (Figure 2). Surprisingly, despite very small changes in peptide sequence, profound structural differences were observed between assemblies derived from the respective peptides. Mutagenesis studies indicated that substitutions corresponding to one or two amino acids were sufficient to interconvert the structures of two assemblies. These studies suggest that cryo-EM structural analysis may provide answers to fundamental questions of acute significance to biomaterials science: from understanding of the functional roles of protein assemblies in biology to shedding light on the robustness of protein quaternary structure in sequence space. References: [1] X Li, P Mooney, S Zheng, CR Booth, MB Braunfeld, S Gubbens, DA Agard, Y Cheng, Nat. Methods 10 (2013), p. 584. [2] TAT Lee, A Cooper, RP Apkarian, VP Conticello, Adv Mater 12 (2000), p. 1105. [3] ER Wright, RA McMillan, A Cooper, RP Apkarian, VP Conticello, Adv Funct Mater 2 (2002), p. 149. [4] EH Egelman, C Xu, F DiMaio, E Magnotti, C Modlin, X Yu, ER Wright, D Baker, VP Conticello, Structure 23 (2015), p. 280. [5] The author acknowledges funding from the NIH, NSF, and DOE. Elizabeth Wright, Ed Egelman, and Rob Apkarian are acknowledged for essential contributions to the EM analysis in this research. Figure 1. A. Schematic representation of an elastin-mimetic hydrogel. B. Cryo-HRSEM image of the corresponding hydrogel. Figure 2. A. Cryo-EM image of a helical assembly derived from a synthetic coiled-coil peptide. B. Cross-section of the atomic resolution structure of the corresponding assembly indicating helix-helix interaction that are critical for its stability.");sQ1[189]=new Array("../7337/0377.pdf","Cryo-Planing Vs Freeze Fracture: Sample Preparation for Cryo-HRSEM","","377 doi:10.1017/S1431927615002688 Paper No. 0189 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Cryo-Planing Vs Freeze Fracture: Sample Preparation for Cryo-HRSEM Irene Y.-T. Chang1, Derk Joester1 1. Materials Science and Engineering, Northwestern University, Evanston, IL 60208, USA Cryo-SEM is particularly efficient at revealing the ultrastructure of biological systems in a near-tonative state and at nanometer resolution. Freeze fracture is conveniently used to prepare interior surfaces of frozen-hydrated samples, but the random nature of fracture does not ensure the passage of the fracture plane through the regions of interest [1, 2]. The resultant surface is rough, which can make interpreting delicate structural features in the sample difficult. To address these issues, we developed the cryo triple ion gun milling (CryoTIGM) technique. High pressure-frozen samples were trimmed using a custombuilt cryo-saw to expose a sample edge, which was brought into contact with a milling mask. Three broad Ar+ beams were aimed at the sample edge, and removed materials above the mask to create a cross-section in the sample at the level of the mask. In this manner, large areas in frozen-hydrated samples were cryo-planed in only a few hours. Cryo-planed samples were subsequently freeze-etched and coated with Pt to increase contrast. We evaluated sample preparation by CryoTIGM against freeze fracture for three biological systems, yeast cell suspensions, mouse liver biopsies, and suspensions of whole sea urchin embryos. The fracture plane frequently occurs between the leaflets of lipid bilayer membranes in freeze-fractured samples, such as the plasma membrane and the nuclear membrane. While this process reveals intra-membranous structures (e.g. nuclear pore complexes), structures inside organelles, for instance the cristae of the inner membrane of mitochondria, cannot be accessed reliably (Fig. 1A, C). In contrast, surfaces cryo-planed using CryoTIGM are very smooth and membranous compartments are always revealed in cross section (Fig. 1B, D). Moreover, morphological information of membranous structures in the out-of-plane direction becomes available. For example, nuclear pores appear as discontinuities in the nuclear membrane in ion-milled samples (Fig. 1D). Freeze fractured sample surfaces are expectedly rough and provide topographical information about the 3D arrangement. However, this at times complicates interpretation. This shortcoming is exemplified in a comparison of freeze-fractured and CryoTIGM-prepared sea urchin embryo samples (Fig. 2). While both techniques show the ectoderm, the enveloping hyaline layer, and the overall packing of ectodermal cells (Fig. 2A, B), tight contacts among neighboring ectodermal cells, suggesting the presence of intercellular junctions, can be identified only after CryoTIGM (Fig. 2B) Membrane tethers extended from ectodermal cells to the hyaline layer are now visible. The hyaline layer is resolved to consist of two layers, with the space between them occupied by some kind of vesicles or granules. In turns of image resolution and contrast, however, we observe both techniques capable of providing cellular and organellar details at high resolutions and with good contrast. Fundamental components, such as the nucleus, mitochondria, the endoplasmic reticulum, and the Golgi apparatus, can be visualized via both methods. In conclusion, cryo-planing of frozen-hydrated samples by CryoTIGM is a convenient way to prepare very large, smooth surfaces for subsequent analysis by cryo-HRSEM (and possibly other cryogenic imaging techniques). The ultrastructure of cells and tissues is well preserved and can be imaged at high resolution and with good contrast. Given the large area and high quality of the surfaces, the relatively fast sample preparation, and the excellent preservation of ultrastructural features, CryoTIGM is a Microsc. Microanal. 21 (Suppl 3), 2015 378 valuable addition to the toolbox that includes freeze-fracture, cryo-FIB-SEM, and cryo-EM. References: [1] Studer, Humbel and Chiquet, Histochemistry and Cell Biology 130 (2008), p. 877. [2] Gilkey and Staehelin, Journal of Electron Microscopy Technique 3 (1986), p. 177. [3] The authors acknowledge funding from the NSF Major Research Instrumentation program (NSF MRI-1229693), the Northwestern University Materials Research Center (DMR-1121262), and the NSF Biomaterials program (DMR-1106208). Figure 1. Cryo-SEM images of yeast cell suspensions prepared by freeze fracture (A, C) or CryoTIGM (B, D). cw: cell wall; m: mitochondria; mi: membrane invagination; mm: milling marks; n: nucleus; np: nuclear pore; sd: surface deposits; v: vacuole. Scale bars represent 1.5 �m in A, B and 500 nm in C, D. Figure 2. Cryo-SEM images of sea urchin embryos prepared by freeze fracture (A) and CryoTIGM (B), showing part of the thickened vegetal plate. hl: hyaline layer; hv: vesicles within the hyaline layer; ic: intercellular contacts; n: nucleus; t: tethers; v: cytoplasmic vesicle. Scale bars represent 5 �m.");sQ1[190]=new Array("../7337/0379.pdf","Serial Block Face-Scanning Electron Microscopy of the Developing Rat Brain Exposed to Ketamine Reveals Changes in Mitochondrial Ultrastructure.","","379 doi:10.1017/S143192761500269X Paper No. 0190 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Serial Block Face-Scanning Electron Microscopy of the Developing Rat Brain Exposed to Ketamine Reveals Changes in Mitochondrial Ultrastructure. Trisha Eustaquio1, Cheng Wang2, Christopher K. Dugard1, Nysia I. George2, Fang Liu2 , William Slikker Jr.2, Merle G. Paule2, Paul Howard2and Angel Paredes2 1. U.S. Food and Drug Administration, Center for Devices and Radiological Health, Silver Spring, Maryland, USA 2. U.S. Food and Drug Administration, National Center for Toxicological Research, Jefferson, Arkansas, USA In the brains of experimental animals there is accumulating evidence that prolonged exposure to ketamine, a N-methyl-D-aspartate (NMDA) receptor antagonist, causes widespread apoptotic cell death [1]. Understanding the long term impact of ketamine exposure on human health and in particular on child development is critical as ketamine is commonly used as a general anesthetic on young children undergoing certain medical procedures such as surgery. It is generally accepted that mitochondria play a pivotal role in apoptosis, a vital and necessary biological pathway used by many organisms for programed cell death. This role may be inadvertently triggered by ketamine causing neurotoxicity by inducing inappropriate apoptosis in affected neurons. Ketamine-induced alterations in mitochondria may be associated with intracellular signals that cause effected neurons to initiate apoptosis. In this study, we examined the brain damage accompanying ketamine exposure (20 mg/kg bw, subcutaneously every 2 hr for 12 hr; 12 hr recovery) in the frontal cortex of Sprague-Dawley rats (postnatal day 7). Initially, conventional transmission electron microscopy (TEM) was used to examine the overall mitochondrial structure in 2D in order to follow and grade the changes that occur within mitochondria after treatment versus control. Using a grading scale based on cristae conformation, there is a clear difference (P<0.05) in the mitochondria of ketamine-treated versus control animals. In order to relate images obtained by conventional TEM to the overall structural effect this drug has on brain tissue, serial block face scanning electron microscopy (SBF-SEM) was used to examine the 3D structure of 10-20 �m3 volumes of frontal cortex [2]. Segmentation of individual mitochondria from both untreated control and ketamine treated tissue reconstructed from SBF-SEM data allowed us to extract and compare individual mitochondria from the 3D volumes. We found that mitochondria in healthy control tissue were mostly elongated with many forming networks with neighboring mitochondria (Figure 1, left), while the mitochondria in tissue from ketamine-treated rats were spherical and discrete, present in tightly packed groups, lacking interconnectivity (Figure 1, right). Both TEM and SBF-SEM support the hypothesis that ketamine damages mitochondria and this effect is associated with the neurotoxicity observed following ketamine treatment. The information in these materials is not a formal dissemination of information by the FDA and does not represent agency position or policy. [1] J Liu, Q Liu, J Li, C Baek, K Han, U Athiraman, and S Soriano, Anesthesiology 117(1) (2012), p.64. [2] W Denk, and H Horstmann, PLoS biology (2004) Microsc. Microanal. 21 (Suppl 3), 2015 380 Figure 1. On the left, segmented volume of a neuron extracted from the frontal cortex of an untreated rat brain. On the right, segmented volume of a neuron extracted from the frontal cortex of a ketamine treated (20 mg/kg bw, subcutaneously every 2 hr for 12 hr; 12 hr recovery) rat brain. The mitochondria were segmented in arbitrary colors in order to enhance spatial contrast. The scale box represents 5:1:1 �m (x:y:z) shown in yellow in the top right corner.");sQ1[191]=new Array("../7337/0381.pdf","Architecture of Pancreatic Islets Imaged by Serial Block Face SEM","","381 doi:10.1017/S1431927615002706 Paper No. 0191 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Architecture of Pancreatic Islets Imaged by Serial Block Face SEM R.D. Leapman1, M.A. Aronova1, A. Shomorony1, C.R. Pfeifer1, J.D. Hoyne1, G.N. Calco1, B.C. Kuo1, G. Zhang1, H. Xu2, C. Tao2, A.L. Notkins2 1 2 National Institute of Biomedical Imaging and Bioengineering, NIH, Bethesda, MD 20892; National Institute of Dental and Craniofacial Research, NIH, Bethesda, MD 20892 The serial block-face scanning electron microscope (SBF-SEM) comprises an automated ultramicrotome built into the specimen chamber, a high-brightness field-emission electron gun, and a high-efficiency backscattered electron detector, enabling high-resolution 3D imaging of heavyatom stained and embedded tissues [1]. The technique provides ultrastructural images of tissue volumes as large as 107 �m3, with a spatial resolution of 10-nm in the plane of the block face and ~25-nm in the cutting direction. The technique complements conventional transmission electron tomography and scanning transmission electron tomography [2]. We have applied SBF-SEM to image entire mouse islets of Langerhans, which are microscopic endocrine organs distributed throughout the pancreas whose main function is to regulate blood glucose levels. Our data provide quantitative information about the distributions of glucagonsecreting alpha cells and insulin-secreting beta cells, as well as the nuclear, mitochondrial and total cell volumes, and the numbers of secretory granules in each cell type (Fig. 1) [3]. By segmenting randomly selected areas, we have determined the cellular volume fraction of the dense cores composed of crystalline insulin in beta cell secretory granules, as illustrated in Fig. 2, from which it was found that 13.2% �2.7% (std. dev.) of the granule-rich image areas consisted of dense-cores. By considering the fraction of the beta cell volume excluded by the nucleus and mitochondria and the dry density of crystalline insulin, it was estimated that beta cells contained 0.045 g � 0.010 g (std. dev.) of insulin per gram of cells [4]. Thus, for a typical beta cell volume of 930 �m3, we can estimate that each cell contains approximately 42 pg of insulin, which is consistent with biochemical measurements reported in the literature. We have also determined the organization of the microvasculature including the blood vessels and surrounding pericapillary space, which accounted for 9.0 � 0.3% of the islet's volume. Each beta cell was found to be in contact with the pericapillary space, with a mean contact area 9 � 5% of the cell's surface area. The role of the pericapillary space as a zone of mediation between blood and endocrine tissue cells fulfills the islet's crucial need for efficient perfusion [5]. References [1] W. Denk and H. Horstmann, PLoS Biology 2 (2004) p. 1900. [2] A.A. Sousa et al, J. Struct. Biol. 174 (2011) p. 107. [3] C.R. Pfeifer et al, J. Struct. Biol. 189 (2015) p. 44. [4] A. Shomorony et al, J. Microsc. (2015) in press. [5] This research was supported by the intramural programs of the National Institute of Biomedical Imaging and Bioengineering, and the National Institute of Dental and Craniofacial Research, NIH. Microsc. Microanal. 21 (Suppl 3), 2015 382 Fig. 1. Full 3-D surface rendering of typical alpha and beta cells: (A�C) Different views of alpha cell displaying plasma membrane (purple), nucleus (green), and mitochondria (pink). This cell had total volume of 570�14 �m3 and nuclear volume of 126�3 �m3. Bar = 5 �m. (D�F) Different views of beta cell displaying plasma membrane (pink), nucleus (green), and mitochondria (red). This cell had total volume of 1,010�30 �m3 and nuclear volume of 115�3 �m3. Bar = 5 �m. Adapted from reference [3]. Fig. 2. Determination of total volume of secretory granule dense-cores in beta cell. Sub-images of size 1.5 �m x 1.5 �m within beta cell are manually segmented. Adapted from reference [4]. Fig. 3. Islet's microvasculature: (A) rendered blood vessel (pink) and surrounding pericapillary space (blue). Bar = 20 �m; (B) Different layers of blood vessel are shown including representative ring of beta cells (green) surrounding pericapillary space (blue), endothelial-cell lining of capillary (pink), erythrocytes (red) and endothelial nuclei (yellow) inside capillary. Bar = 20 �m. Adapted from reference [3].");sQ1[192]=new Array("../7337/0383.pdf","Correlating Cryo-Electron Microscopy Methods for Structural Studies of Bacteria.","","383 doi:10.1017/S1431927615002718 Paper No. 0192 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Correlating Cryo-Electron Microscopy Methods for Structural Studies of Bacteria. Elizabeth R. Wright1,2, Hong Yi2, Jeannette V. Taylor2, Ricardo C. Guerrero-Ferreira1 and Cheri M. Hampton1 1. Emory University School of Medicine, Division of Pediatric Infectious Disease, Children's Healthcare of Atlanta, Atlanta, Georgia 2. Robert P. Apkarian Integrated Electron Microscopy Core, Emory University, Atlanta, Georgia Vibrio vulnificus, a halotrophic Gram-negative bacterium, which lives in coastal waters, is associated with shellfish in natural environments and is an opportunistic pathogen of humans. A number of studies have identified many of the virulence factors associated with V. vulnificus [1]. Some of the virulence factors identified include external structures of the bacteria such as the capsular polysaccharide (CPS), flagellum, and pili. However, limited efforts have been aimed at determining the ultrastructure of these factors and the arrangement of their individual components. In order to address how ultrastructure of the bacteria influences pathogenesis, we probed the architecture of the cells, the flagella, the pili, and the CPS with high-resolution scanning electron microscopy (HRSEM), conventional transmission electron microscopy (TEM) of cryo-preserved specimens, and cryo-electron tomography (cryo-ET). We used HRSEM methods to examine the structure of the bacterial outer membrane, appendages, and secreted outer membrane vesicles (OMVs). V. vulnificus cultures were fixed and then dehydrated with increasing concentrations of ethanol. Samples were dried in a Polaron critical point drying (CPD) apparatus. The specimens were mounted onto stubs and a thin film of chromium was deposited with a Denton DV-602 magnetron sputter coater. Specimens were imaged in ISI DS-130F FE-SEM. We used a combination of cryo-preservation methods and conventional TEM imaging to assess the presence or absence of the bacterial capsular polysaccharide (CPS) layer and whether secreted OMVs remained associated with the cell. V. vulnificus strains were cultured overnight either on agar plates or in liquid media. Bacteria grown on plates were either used as individual colonies or gently scraped together to produce a larger volume of material. The individual colonies or the pooled sample were applied to Balzers planchets and high pressure frozen (HPF) with a Balzers HPM 010. Alternatively, the pooled sample was pipetted into Leica copper capillary tubes, the tube ends were crimped, and subsequently frozen in liquid propane by self-pressurized rapid freezing (SPRF). The frozen hydrated samples were freeze substituted using a Leica EM AFS2 automatic freeze substitution system. Resin-infiltrated samples were sectioned, post-stained and imaged on a JEOL JEM-1400 120 kV LaB6-TEM. Liquid cultures of V. vulnificus strains were aliquoted onto glow-discharged Quantifoil carbon grids and then plunge-frozen in liquid ethane with a FEI Mark III Vitrobot. Cryo-EM and cryo-ET data collection was performed with a JEOL JEM-2200FS 200 kV FEG-TEM equipped with the Zernike phase plate airlock system, an in-column energy filter (slit width 20 eV), a cryo-transfer specimen holder (Model 914, Gatan), a DE-20 direct electron detection device (Direct Electron, LP), and a 4k x 4k Gatan Ultrascan CCD camera. Tilt series were acquired automatically with 2� tilt increments from -65� to +65� by using Serial EM [2]. Reconstructions were generated with IMOD [3]. HRSEM studies resolved the overall cell structure (Figure 1). Rapid cryo-preservation techniques combined with TEM imaging resolved the presence of the CPS extending from the cell surface (Figure 2). In addition, OMVs were observed within the CPS matrix. Cryo-ET data provided information Microsc. Microanal. 21 (Suppl 3), 2015 384 regarding the three-dimensional (3D) arrangement of released OMVs (Figure 3). Studies are underway to improve bacterial preservation and imaging methods of the V. vulnificus CPS as well as OMV order. References: [1] PA Gulig, KL Bourdage, and AM Starks, J. Microbiology 43 (2005), p. 118-131. [2] DN Mastronarde, J. Structural Biology 152(1) (2005), p. 36-51. [3] JR Kremer, DN Mastronarde, and JR McIntosh, J. Structural Biology 116(1) (1996), p. 71-76. [4] This research was supported by funds from Emory University, Children's Healthcare of Atlanta, the Emory Center for AIDS Research, the Georgia Research Alliance, Human Frontiers Science Program, National Institutes of Health (1R01GM104540 and 1R21AI101775), and the National Science Foundation (0923395) to E.R.W, and S10 RR025679 to P.W.S. Cryo-EM/ET data was collected at the Emory University Robert P. Apkarian Integrated Electron Microscopy Core. Figure 1. FE-SEM images of a V. vulnificus cells. Scale bar, 667 nm. Figure 2. Conventional TEM images of SPRF preserved V. vulnificus cell. Scale bar, 200 nm. Figure 3. Segmented 3D reconstruction of a frozen-hydrated V. vulnificus cell.");sQ1[193]=new Array("../7337/0385.pdf","Placing Molecules in a Cellular Context Using Light, Eelectron and X-Ray Microscopy","","385 doi:10.1017/S143192761500272X Paper No. 0193 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Placing Molecules in a Cellular Context Using Light, Eelectron and X-Ray Microscopy Lucy M. Collinson Electron Microscopy Unit, Francis Crick Institute, London, UK Fluorescence microscopy is a powerful tool for localising proteins within biological samples. However, information is limited to the distribution of the tagged protein, telling us little about the ultrastructure of the surrounding cells and tissues, which may be intimately involved in the biological process under study. Electron microscopy overcomes the resolution limitation inherent in light microscopy and can reveal the ultrastructure of cells and tissues. However, protein localisation tends to be complex and is often dependent on the availability of `EM-friendly' antibodies. Correlative light and electron microscopy (CLEM) combines the benefits of fluorescence and electron imaging, revealing protein localisation against the backdrop of cellular architecture. In this talk, I will introduce several ways in which we are developing 3D CLEM for application to biological samples. To image rare events in cells, tissues and whole model organisms, we developed Correlative Light and Volume EM (CLVEM), which combines correlative workflows with microscopes that automatically collect large stacks of high resolution images. The Serial Block Face Scanning Electron Microscope (SBF SEM) consists of a miniaturised ultramicrotome mounted inside the SEM chamber, which cuts resinembedded samples using a diamond knife prior to imaging the exposed surface with the electron beam. In this way, thousands of images can be automatically collected through the volume of the sample [1]. We used Correlative Light and SBF SEM to analyse the reformation of the nuclear envelope during mitosis in lipid-depleted mammalian cells, using the GFP-C1 construct to detect diacylglycerol in cell membranes [2]. Fast automated data acquisition allowed us to move towards high throughput quantitative CLEM at a rate of one whole cell per day. During this development work, it became clear that several technical challenges associated with CLEM are exaggerated when working in 3D. Firstly, the accuracy of the overlay between light and electron images (which is critical to successful localisation of molecules to cellular structures) becomes more difficult as sample size increases. Secondly, with data acquisition becoming more automated, the bottleneck in the workflow becomes the data analysis step. Thirdly, artifacts associated with chemical fixation, heavy metal staining, dehydration and resin embedding are exacerbated when imaging in three dimensions, and this can critically affect the outcome of high resolution imaging experiments in some biological systems. We are addressing these challenges as follows... To increase protein localisation precision, we developed an `In-Resin Fluorescence' (IRF) protocol that preserves the activity of fluorescent proteins in resin-embedded cells and tissues [3]. Once cut into ultrathin sections, out-of-plane fluorescence is removed resulting in `super-resolution' light microscopy in the axial direction, which increases the accuracy of the LM-EM overlays. Localisation precision is further increased by imaging the IRF sections in vacuo in the next generation of commercial integrated light and electron microscopes (ILEM, Fig.1): the SECOM ILSEM (DELMIC) and the iCorrTM ILTEM (FEI). We are seeking to further improve accuracy by developing integrated super resolution light and electron microscopes. To facilitate and expedite data analysis, we collaborate with computer vision scientists to design algorithms to automatically detect and select subcellular organelles in EM images, including the nuclear envelope, the endoplasmic reticulum and vesicles. This work is continuing with the ambitious aim of automating the process of recognition and 3D model generation for all cell organelles. Microsc. Microanal. 21 (Suppl 3), 2015 386 Finally, I will show how we are moving into the world of the native state cell, using synchrotron-hosted soft X-ray microscopes to image whole vitrified cells, without the need for chemical fixatives, stains or sectioning [4,5,6]. Our aim is to capture the most volatile cellular structures and elucidate their function. I will use the example of autophagy, a critical cellular process in cell health and disease, to illustrate the potential of this correlative cryo-fluorescence and cryo-soft X-ray workflow to answer otherwise intractable biological questions [7]. 1 2 Figure Legends. Figure 1 shows the localisation of diacylglycerol (top) in mammalian cell membranes (bottom) using GFP-C1 as a probe, imaged in an integrated light and scanning electron microscope [3]. Figure 2 shows a whole vitrified HeLa cell, imaged using a correlative cryo-fluorescence (left) and cryosoft X-ray tomography (right) workflow [4,5,6]. References [1] Peddie, C.J. & Collinson, L.M. (2014) Micron. 61,9-19. [2] Domart, M-C et al. (2012) PLoS One. 7,e51150. [3] Peddie, C.J. et al. (2014) Ultramicroscopy.143,3-14. [4] Carzaniga, R. et al. (2013) Protoplasma.251,449-58. [5] Duke, E. et al. (2014) Ultramicroscopy.143,77-87. [6] Duke, E. et al. (2014) J.Microscopy.255,65-70. [7] The author acknowledges funding from Cancer Research UK and from the MRC, BBSRC and EPSRC under grant award MR/K01580X/1. The research leading to these results has received funding from the European Community's Seventh Framework Programme (FP7/2007-2013) under BioStruct-X (grant agreement N�283570).");sQ1[194]=new Array("../7337/0387.pdf","Important steps in a Correlative Light Electron Microscopy Experiment","","387 doi:10.1017/S1431927615002731 Paper No. 0194 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Important steps in a Correlative Light Electron Microscopy Experiment Paul Verkade Wolfson Bioimaging Facility, Schools of Biochemistry and Physiology & Pharmacology, Medical Sciences Building, University Walk, University of Bristol, Bristol, UK E-mail: p.verkade@bristol.ac.uk Correlative Light Electron Microscopy combines the strengths of light and electron microscopy in one experiment and the sum of such an experiment should provide more data / insight than each technique alone (1 + 1 = 3). There are many ways to perform a CLEM experiment and a variety of microscopy modalities can be combined. The choice of these instruments and the experimental approach should primarily depend on the scientific question to be answered. Any CLEM experiment can usually be divided in 3 parts; probes, processing, and analysis. I will discuss 3 processing techniques based on light microscopy in conjunction with Transmission Electron Microscopy, each with its specific application and its advantages and challenges. 1. The first is based on the use of coverslips with an embossed finder pattern. Importantly, it allows for live cell imaging and captures an event of interest using chemical fixation (Figure 1) [1]. This is possibly one of the technically easiest CLEM techniques but retracing the same object in the light and electron microscope remains a challenge. Hence we have focussed on developing algorithms that allow confident retracing and identification of the same structure observed in the light microscope to be found in the electron microscope. 2. A second uses the Tokuyasu cryo immuno labelling technique to trace back objects of interest (Figure 2) [2]. This allows for relatively high immuno labelling efficiencies but is almost impossible in combination with live cell imaging. 3. The third is based on cryo-fixation to obtain best possible preservation of ultrastructure [3,4]. This allows us to capture events that would be lost because of chemical fixation, e.g. membrane tubules. It allows for live cell imaging but immuno labelling options are limited. REFERENCES. [1] Hodgson L, D. Nam, J. Mantell, A. Achim, and P. Verkade, Meth. Cell Biol. 124 (2014), p. 1. [2] Hodgson L, J. Tavar�, and P. Verkade, Protoplasma 251 (2014) p. 403. [3] Brown, E., J., Van Weering, T. Sharp, J. Mantell, and P. Verkade, Meth. Cell Biol. 111 (2012), p. 175. [4] P. Verkade, J. Microsc. (Oxford) 230, p. 317. Microsc. Microanal. 21 (Suppl 3), 2015 388 Figure 1. CLEM allows the identification and subsequent high-resolution analysis of 1 special cell amongst 100s. In the top left row, a dividing cell (1 in >100) is identified based on its DNA staining (blue). With the aid of embossed coverslips (for full details see [1]) the dividing cell can be traced back in the EM and studied at higher resolution. Figure 2. CLEM using Tokuyasu cryo immuno gold labelling, an excellent way to zoom into specific structures with high labelling efficiency. Modified from [2].");sQ1[195]=new Array("../7337/0389.pdf","Correlating Microscopies From Differing Imaging Modalities: From Experimental Design to Alignment and Overlay of Images.","","389 doi:10.1017/S1431927615002743 Paper No. 0195 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Correlating Microscopies From Differing Imaging Modalities: From Experimental Design to Alignment and Overlay of Images. Douglas R. Keene1 and Sara F. Tufa2 1,2. Shriners Hospital for Children, Micro-Imaging Center, Portland, Oregon USA. Correlated imaging is the process of imaging a single specimen with two (or more) complementary modalities, then registering and overlaying the images to create a composite view. The higher resolution image is colorized using the lower resolution image; for example micro-CT images may be used to colorize LM images [1]; laser scanning confocal (LSC) images may be use to colorize TEM images [2,3]. The lower resolution image is made somewhat transparent, allowing detail in the underlying image to be visible and assisting in the registration of the two images. If correlating images collected by TEM and LSC, the resulting composite image would demonstrate specific ultrastructural features in the high-resolution TEM field colorized by the aligned confocal image. Automated image registration may be facilitated by a variety of sophisticated computer programs and hardware platforms (ie. FEI Corrsight) which are utilized by high-throughput laboratories. This abstract is meant for the more occasional user wishing to align images manually. FIJI is a public domain image processing program developed at the National Institutes of Health. It is available to anyone as a free download and performs marvelously well for the purpose of image registration. Overlaying images is accomplished with ease using FIJI [4]. Sims and Hardin [1] presented a method to overlay LSC images of a GFP construct in C. elegans embryos with TEM images collected from the same ultrathin section and aligned using Adobe Photoshop [2]. We used a similar method to localize mutant cartilage oligomeric protein in human chondrocytes [3]. Unlike the high fluorescence intensity described in C. elegans, the fluorescence intensity in our GFP expressing chondrocytes lacked sufficient brightness to image in ultrathin sections. This required us to correlate confocal images collected from one-micron sections with TEM images collected from the next serial ultrathin section. Given that the sections are taken from close but still different heights ("Z") within the block face and also that there are differing sectioning artefacts present in each section, alignment is not straight forward. Here one image needs to be non-uniformly and nonlinearly stretched and compressed (skewed) during overlay. FIJI allows registration of grossly misaligned images with far greater ease than Photoshop. The "turboreg" plugin in FIJI initially aligns images based on four operator-selected points of reference. Then, if the two images do not register perfectly, many different points of reference may be selected to skew one image so that it aligns perfectly with the other. Step by step keystrokes to perform these alignments have recently been published [4]. The choice of method used to prepare cultures or tissues must balance resolution with the preservation of antigenicity or reactivity of fluorescent constructs. Sims and Hardin [2] found that the GFP construct in C. elegans tolerated mild glutaraldehyde and embedding in HM20; we found that YFP constructs tolerated higher concentrations of glutaraldehyde and embedding in LRWhite; however more recent formulations of LRWhite do not give optimal results. Even so, LRWhite is still our embedding media of choice if the goal is to preserve antigens on the surface of ultrathin sections; here we use the Molecular Probes Alexa 488/10nm secondary antibody conjugate which carries both Alexa 488 and also a 10nmgold particulate (Figure 1). Microsc. Microanal. 21 (Suppl 3), 2015 390 SyBr Green is a reagent useful for correlative imaging localizing nucleic acids. SyBr was developed to replace ethidium bromide as a stain for DNA/RNA in electrophoretic gels; however we use this reagent as an enbloc tissue stain to localize nucleic acids in structures as small as viruses. Fluorescent emission of SyBr endures fixation in 1.5% glutaraldehyde and 1% Osmium Tetroxide and embedding in Epon 812 (Figure 2). Figure 1. GFP expression could not be detected by LSC within ultrathin or one-micron sections of GFP expressing tenocytes. Therefore antibody to GFP was applied to the surface of LRWhite sections followed by Molecular Probes Alexa 488/10nm-gold, resulting in strong labeling specific to GFP expressing cells (A). Higher magnification demonstrates that all tenocytes within the tendon are labeled (B). Higher resolution immunolabeling of cell extensions is afforded by -10nm gold particles (C). Figure 2. Cells infected with hepatitis virus were fixed in 1% glutaraldehyde, and then exposed to Life Technology SyBr (1:1000 in Tris-HCl). Heterochromatin (h) and nucleoli label intensely as does viroplasm within double membrane vesicles (dm) (A,B). The presence of viruses within fluorescing regions (v) (inset, A) were confirmed by higher magnification TEM (C). References: [1] G Sengle, S Tufa, L Sakai, M Zulliger, D Keene (2013). J Histochem Cytochem. 61:263-71 [2] P Sims & J Hardin, J. D. (2005). Microscopy and Microanalysis, 11(Suppl. 2), 6�7. [3] D Keene, S Tufa, G Lunstrum, P Holden, W. Horton (2008). Microscopy and Microanal. 14:342-8. [4] D Keene, S Tufa, M Wong, N Smith, L Sakai and W Horton. (2014). In Correlative Light and Electron Microscopy II, eds. T Muller-Reichert and P Verkade, New York, pp. 391-417. [5] The authors acknowledge generous funding from the Shriners Hospitals for Children.");sQ1[196]=new Array("../7337/0391.pdf","Simultaneously Measuring Red Blood Cell Flux in vivo for a Large Number of Retinal Capillary Vessels Using Optical Coherence Tomography","","391 doi:10.1017/S1431927615002755 Paper No. 0196 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Simultaneously Measuring Red Blood Cell Flux in vivo for a Large Number of Retinal Capillary Vessels Using Optical Coherence Tomography Ruikang K. Wang1,2, and Zhongwei Zhi1 1. 2 Department of Bioengineering, University of Washington, Seattle, WA 98195, USA . Department of Ophthalmology, University of Washington, Seattle, WA 98195, USA In this presentation, we will present an application of optical coherence tomography (OCT) imaging technique for measurement of red blood cell (RBC) flux within multiple capillary of retina in mice in vivo. The technique would be also useful in the monitoring of changes in RBC flux as a function of pharmacological treatment doses for diseases such as diabetes, infectious and inflammatory diseases and cancer, where monitoring vascular transport could be essential. Ocular microcirculation plays an important role for maintaining a normal vision and abnormal ocular blood flow was shown to be related to many ocular diseases such as age-related macular degeneration, diabetic retinopathy, and glaucoma. Capillaries are the smallest unit of circulatory system, within which, red blood cells (RBCs) are squeezing through one by one. Capillary RBC flux, i.e. the number of RBC flow through the capillary per unit time, is an important capillary flow parameter. It is critical to develop a noninvasive imaging technique to visualize the retinal capillary network and quantify the RBC characteristics for understanding and early diagnosis of ocular diseases that have a vascular component in their pathogenesis. Optical coherence tomography (OCT) is a non-contact and non-invasive imaging modality and has been widely used for human eye imaging and blood flow measurement. By using a repeated B-scan mode, the fluctuating OCT signal has been shown to be associated with the RBC passing through the capillary in the mouse brain [1]. Ren et. al. also applied OCT to measure the red blood cell velocities of cerebrovascular networks by detecting the Doppler phase transients induced by the passage of RBCs through the capillary [2]. However, both of these works were done on cerebrovascular network, and no work has yet been done on retinal capillary flux measurement using OCT. In our work, we developed a technique that used OCT imaging [3] for measurement of RBC flux within multiple capillary of retina in the mice and rat. A fast speed spectral-domain OCT imaging system at 820nm with a line scan rate of 240 kHz was developed to image mouse retina. By acquiring repeated Bscans at an ultrafast 720 fps frame rate, we were able to detect the fluctuating OCT signal intensity caused by the passage of single RBC through the capillary (Fig.2A-C). RBC flux of each single capillary was obtained by counting the number of peaks in the OCT intensity profile after applying Gaussian filtering (Fig.2 D). Multiple capillaries in one cross-section were isolated and their flux was measured automatically with this algorithm (Fig.2 E). This developed technique was applied to measure the RBC flux of capillaries within retina of BTBR ob/ob mice, as well as wild type (WT) controls. The results (Fig. 3) indicate that the ob/ob mice retina have a lower average RBC flux than WT mice, which may provide a way for early diagnosis of Diabetic retinopathy. In summary, we demonstrated the capability of OCT imaging technique for the measurement of RBC flux within multiple capillary of retina at the same time in mice. Fast frame rate (720 fps in this application) repeated B-scans at one cross-section enable us to observe fluctuation in OCT intensity as RBC passing through the capillary. The flux of each single capillary was estimated by counting the number of peaks per unit time in the signal profile after applying a Gaussian filtering algorithm. The Microsc. Microanal. 21 (Suppl 3), 2015 392 average capillary flux within retinas of BTBR ob/ob mice were measured to statistically lower than its WT controls. This indicates that capillary RBC flux can be an important potential maker for the early detection of diabetic retinopathy as well as other ocular diseases. Fig. 1. (A) Depth-resolved angiography of moue retina. (B-D) Microvasculature within three different layers of retina after segmentation, where R1, R2 and R3 represent NFL/GCL, IPL, and OPL, respectively. (E) Cross-sectional blood flow image located at the white line in A. (F) Blood vessel through time obtained with repeated B-scans. The horizontal line artifacts are caused by animal breathing. Fig. 2. (A-B) Two cross-sectional blood flow frames captured at different times showing the fluctuating capillary signal. (C) Smoothed intensity profile at the two capillary regions labeled as 1 and 2. (D) The intensity profile after Gaussian fitting, and capillary flux was calculated by detecting the number of peaks. (E) Color coded capillary flux image and (F) histogram of the flux for all capillaries in the crosssection. Fig. 3. Comparison of the averaged capillary RBC flux between BTBR OB (N=6) and WT (N=6) mice. Note, * OB group is significantly lower than WT group (Two-sample t-test; p < 0.05). References: 1. Lee, J., et al., Journal of Cerebral Blood Flow and Metabolism, 2013. 33(11): 1707-1710. 2. Ren, H.G., et al. Applied Physics Letters, 2012. 100(23). 3. Zhi, Z., et al., Investigative Ophthalmology & Visual Science, 2014. 55(2), 1024�1030.");sQ1[197]=new Array("../7337/0393.pdf","Efficient and Documented Preparation of Pharmaceutical Particles for Correlative Microscopy Analyses using mPrepTM Capsule Processing","","393 doi:10.1017/S1431927615002767 Paper No. 0197 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Efficient and Documented Preparation of Pharmaceutical Particles for Correlative Microscopy Analyses using mPrepTM Capsule Processing Steven L. Goodman Microscopy Innovations LLC, 213 Air Park Rd, Suite 101, Marshfield, WI, 54449, USA. Pharmaceutics utilize many types of nanometer, micrometer and larger particles that require analyses to determine morphology, uniformity, and chemistry. This, in turn, requires several analytical instruments and multiple specimen preparations. Two additional challenges in the pharma lab are that analyses always seem to be required immediately, and that specimen prep must be documented and reproducible to meet GLP (Good Lab Practice) and similar quality standards. In a prior report we described how documentation and correlative microscopy are valuable for obtaining evidence for intellectual property litigation on tablet formulations [1]. In the present study, an efficient, rapid and GLP-level correlative approach was used to prepare and examine formulations of two types of pharmaceutical particles: siRNA (small interfering RNA) nanoparticles, and sub-millimeter hydrogel drug delivery particles. Two nanoparticle siRNA drug formulations were prepared for transmission electron microscopy (TEM) to assess morphology and uniformity. Formulations were prepared with and without a positive stain, and with and without uranyl acetate negative staining, for a total of 8 preparations. Four grids of each preparation were simultaneously prepared using 16 mPrep/gTM (grid) capsules, an 8-channel 200 �l pipettor and a 96-well microtiter plate: First, 2 Formvar-filmed grids were inserted into the mPrep/g capsules and then 2 mPrep/g capsules were stacked onto each pipettor channel to provide 4 grids/channel (Fig 1). The nanoparticle suspensions and other reagents were put into the microtiter plate with each reagent in one well, and each reagent step in a single row. Simultaneous preparation was achieved by first pipetting in 80 �l of the nanoparticle suspensions from the first microtiter row into the stacked mPrep/g capsules. The nanoparticles were then adsorbed onto the filmed-grids for 30 minutes. The suspensions were then emptied from the pipettor, and the pipetter was moved to the next row for a water rinse, then to the next row for the positive stain (stain in half the microtiter wells and water in the other half) thereby staining half of the grids. After staining, the reagent was emptied from the pipettor and the pipettor was moved sequentially to the remaining rows for a water rinse, negative stain (half the capsules), and water rinse again. The particles were then examined with an FEI Tecnai T-12. Hydrogel micro-particles were prepared and analyzed to determine chemistry and composition in the dry state from thin cross-sections. The particles were prepared by entrapping each formulation into 8 mPrep/sTM (specimen) capsules (Fig 2a). These mPrep/s capsules were then attached to an 8-channel pipettor and the particles were then simultaneously rinsed in 100% acetone followed by epoxy infiltration and embedment, in the same manner as shown for reagent delivery in Fig 1a. Epoxy-filled capsules were then ejected (without emptying) into a bottom-sealing mPrep/benchTM microtiter plate, and cured at 60 C (Fig 2b). After curing the mPrep/s capsules were mounted in a microtome chuck and faced thorough the capsule for sectioning (Fig 2c). Dry sections on coverslips were examined with infrared (FTIR) spectroscopy for chemical composition, and sections on SEM stubs were examined for elemental composition with energy dispersive spectroscopy (EDS) (Figs 2d-e). Additional particles were examined with SEM (Fig 2f) and in the hydrated state by light microscopy. Multichannel template-guided pipetting with mPrep capsules provided rapid, simultaneous preparation Microsc. Microanal. 21 (Suppl 3), 2015 394 of 16 TEM grids from 8 nanoparticle preparations, and 8 micro-particle specimen blocks. All were prepared with ease and experimental reproducibility. Since specimens were always labeled throughout the easily followed template-guided protocol, this also provided GLP-level documentation. a a b b 200#nm# 200#nm# Fig 1: a) Two mPrep/g capsules on each pipettor channel with sample preparation by pipetting reagents from each microtiter row guided by underlying template. b) TEM of siRNA nanoparticle preparations. a a b b c c screen screen d d e e f f Fig 2: a) Dry hydrogel micro-particles entrapped in mPrep/s capsule (arrow) by removable adjustable screen. b) mPrep/s blocks containing particles after epoxy curing in bottom sealing mPrep/bench microtiter plate. c) mPrep/s capsule with embedded particles in microtome chuck. Several visible crosssectioned particles (arrow). Note capsule trimmed away for sectioning (double arrow). FTIR (d) and SEM-EDS maps (e) from particle cross-sections. f) Whole particle imaged with FE-SEM. [1] SL Goodman, JJ Edwards, Microsc Microanal, 20 (suppl 3) (2014) 1416-7. !! 200 �m 200 �m");sQ1[198]=new Array("../7337/0395.pdf","Characterization of Glass Delamination by TEM: Results from a New Sample Preparation Technique","","395 doi:10.1017/S1431927615002779 Paper No. 0198 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of Glass Delamination by TEM: Results from a New Sample Preparation Technique E.F. Schumacher1, H.M. Talesky1 and K.J. Diebold1 1 McCrone Associates, Inc., Westmont, Illinois 60559 USA Transmission electron microscopy with energy dispersive X-ray spectrometry (TEM/EDS) has been shown to offer advantages over scanning electron microscopy (SEM) with EDS for analysis of very thin particulate resulting from delamination of glass vials and syringes used in the pharmaceutical industry [1]. Delamination occurs when injectable solutions and suspensions react with glass packaging, resulting in contamination of drug products with glass flakes, coatings, and residues formed from glass dissolution or reaction with the drug to form a secondary product. Particulate is typically isolated from liquid products by filtration, followed by examination using light microscopy. The filtered particulate may then be analyzed using SEM/EDS, a technique recommended for use in glass container screening studies as outlined in USP <1660> [2, 3]. Examination of glass surfaces and isolated particulate using light microscopy and SEM may be sufficient to confirm that glass delamination has occurred. However, elemental analysis can be critical in determining whether particulate in a liquid product is glass delamination, glass coating, dissolved glass residue or a secondary product. Removal of extremely thin samples from filters for transfer to SEM substrates can be difficult. An alternative is to mount the entire filter on a substrate. Either preparation technique typically results in EDS spectra dominated by a signal from the substrate or the carbon filter, and containing much smaller peaks for major particulate components. Peaks for minor or trace elements may be entirely absent, thereby lessening the value of EDS for discrimination between glass and other materials that may be present. Though the TEM is ideal for characterization of these very thin samples, transfer of flakes and residues from filters to TEM grids can be challenging. A method has been developed whereby filtration of liquid products directly onto holey carbon-coated TEM grids captures a sufficient amount of representative material to identify particulate isolated from liquid products. Results were found to be largely consistent with those obtained by SEM/EDS analysis of concurrently filtered products mounted on SEM substrates. More complete information about particulate elemental composition was obtained by TEM analysis, and in two cases, glass delamination flakes were captured on TEM grids but were not found on SEM substrates prepared from the same filtrations. Little to no background particulate was found when as-received grids and blanks grids prepared by filtration of particle-free water were examined. Characteristics of particulate isolated from an iron sucrose product were consistent with results of earlier TEM studies in which grids were prepared by direct transfer of the material rather than by filtration [1]. Delamination flakes and other types of contaminant particles filtered directly onto grids were unambiguously identified using TEM/EDS. TEM/EDS characterization of materials associated with glass delamination can verify composition in samples too thin for SEM/EDS analysis. Detection of smaller and thinner flakes may also aid in earlier detection of glass delamination. Though SEM is of value for analysis of bulk glass and some Microsc. Microanal. 21 (Suppl 3), 2015 396 types of associated particulate, this work demonstrates the benefit of TEM for unambiguous characterization of thin particulate and residues associated with glass delamination processes. References: [1] Schumacher, et al., Proc. Microscopy and Microanalysis 2012. [2] Haines, D., et al., Contract Pharma, published online June 2013. [3] General Chapter <1660> Evaluation of the Inner Surface Durability of Glass Containers, available online from the U.S. Pharmacopeial Convention. Si O Ca Na Mg C Cu K Ca Cu Cu Al Cu Ca K K Ca 8 Cu 10 12 14 16 18 20 keV 0 2 4 6 Full Scale 10275 cts Cursor: -0.200 (0 cts) Fig. 1. TEM EDS spectrum from glass delamination flake (left), and SEM EDS spectrum from similar flake (center). SEM spectrum Y-axis expanded to show peaks from glass elements (right). Fig. 2. TEM images of delamination flakes captured on holey carbon grids by filtration. Dark spots in image on right were found to be rich in phosphorus and rare earth elements.");sQ1[199]=new Array("../7337/0397.pdf","Prospects for High Resolution Analytical Electron Microscopy of Organic Crystalline Particles.","","397 doi:10.1017/S1431927615002780 Paper No. 0199 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Prospects for High Resolution Analytical Electron Microscopy of Organic Crystalline Particles. James Cattle1, Mark S'ari1, Neil Wilkinson2, Nicole Hondow1, Andy Brown1 and Rik Brydson1. 1. Institute for Materials Research, School of Chemical and Process Engineering, University of Leeds, Leeds LS2 9JT, United Kingdom. 2. Gatan UK, 25 Nuffield Way, Abingdon, Oxon. OX14 1RL, United Kingdom. Polymorphism is the ability of a compound to crystallise with different packing arrangements and is critically important in several industrial sectors as many of the solid-state properties of a compound are dependent on the crystal form. Recently, TEM has been applied to the study of polymorphism in organic materials such as pharmaceuticals, where polymorphs are extremely common, and has been shown to have advantages over more routinely used analytical techniques [1]. Using a combination of imaging and diffraction each of the crystals in a sample can be analysed one-by-one during TEM analysis and different polymorphs of a compound can be distinguished, or new polymorphs of a compound identified, on the basis of crystal morphology and/or by indexing electron diffraction patterns. A further advantage is the ability to solve crystal structures of beam sensitive organic compounds using a single crystallite with sub-micron sized dimensions, by combining electron diffraction and crystal structure prediction, where standard single crystal or powder X-ray approaches would not be applicable [2]. In addition to electron diffraction, chemical analysis at the single particle level using electron energy loss spectroscopy (EELS) is beneficial not only to identify light element stoichiometries in organic crystals, but also to identify polymorphic variants using electron energy loss near edge structure (ELNES). Direct high resolution diffraction and even phase contrast imaging could allow identification of particle distributions within complex particulate product mixtures, as well as defects within individual particles and surface facets arising as a result of powder processing or tableting requirements. This could affect overall product properties, such as the dissolution of active pharmaceutical ingredients. A major issue preventing such analysis in organic crystalline materials is electron beam induced damage. Hence, prior to any investigation, a detailed study of both the critical fluence (aka "dose") and fluence rates is required in order to identify the safe limits for analysis by electron diffraction, imaging and also spectroscopy. Here we present results for a model organic compound, Theophylline, which is a xanthine derivative with the chemical formula C7H8N4O2 used as a treatment for asthma and bears a structural and pharmacological similarity to caffeine (C8H10N4O2). Theophylline crystals were prepared by cooling a saturated solution in nitromethane. The nitromethane dispersion was drop cast onto either holey or continuous carbon coated (or graphene) TEM grids and analysed in a field emission TEM operated at 200 kV. Crystals were predominantly plate-like and oriented with the <100> axis parallel to the electron beam. Beam damage (presumed to be by a radiolytic mechanism) was monitored by observation of the diffraction pattern. Interestingly, progressive amorphisation by the electron beam proceeded via progressive loss of short range order (fading of the outer diffraction spots first) as observed by Glaeser [3] possibly due to breaking of hydrogen bonds and small molecular rearrangements during the initial stages of damage. Figure 1 shows plots of the intensity in the (011) diffraction spot of theophylline (polymorphic form II) as a function of increasing fluence, for a fluence rate of ca. 2 x 10-4 A/cm2. Five Microsc. Microanal. 21 (Suppl 3), 2015 398 different cases are shown (details in figure caption) and it is clear that using a graphene support (and liquid nitrogen sample cooling) gave the largest critical fluence. Such analyses for a particular material can help devise appropriate measurement protocols. At the higher fluence rates required for phase contrast imaging using conventional CCD cameras the critical fluence decreases significantly. Figure 2 shows a high resolution phase contrast image of theophylline recorded using a direct electron detection camera (Gatan K2 Summit) and a total fluence of ca. 200 electrons/nm2 at a relatively low fluence rate of 8 x 10-5 A/cm2. (002) lattice fringes are visible in the raw image and more clearly in the Fourier filtered image (figure 2(b)). EELS spectra (shown in reference [4]) were collected in diffraction mode with a collection semi-angle of 30 mrads (and typically a 20 second acquisition time) from a crystalline area of ca. 0.8 micron diameter recorded with an electron fluence rate of ca. 4 x 10-5 A/cm2 and a total electron fluence during spectral acquisition of typically 1500-2000 electrons/nm2. EELS showed the presence of carbon, nitrogen and oxygen in the sample and quantification gave a [C]/[N] atom ratio of 1.7 (1.75 is expected for theophylline) and a [C]/[O] ratio of 2.6 (3.5 is expected for theophylline). References: [1] M D Eddleston et al., J. Pharm. Sci. 99 (2010), p. 4072. [2] M D Eddleston et al., Chem. Eur. J 19 (2013), p. 7874. [3] R M Glaeser, J. Ultrastructure Res. 36 (1971), p. 466. [4] R Brydson et al., J. Phys.: Conf. Ser. 522 (2014), 012060. Figure 1. Graph of the decay in theophylline (011) diffraction spot intensity as a function of electron fluence for various accelerating voltages, sample support films and sample cooling. Figure 2. (a) Bright field TEM image of theophylline crystal with inset FFT from bend contour showing 0.424 nm (002) spots; (b) inset is the masked FFT with spots intensified by x10 and its inverse FFT showing lattice fringes.");sQ1[200]=new Array("../7337/0399.pdf","Quick Observation of Tissues in Solution by Atmospheric Scanning Electron Microscopy (ASEM)","","399 doi:10.1017/S1431927615002792 Paper No. 0200 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quick Observation of Tissues in Solution by Atmospheric Scanning Electron Microscopy (ASEM) Nassirhadjy Memtily, 1, 2, 3 Mari Sato2, Tomoko Okada2, Tatsuhiko Ebihara 2, Kazuhiro Mio1, 2, Mitsuo Suga 4, Hiditoshi Nishiyama4, Chikara Sato1, 2 Graduate School of Comprehensive Human Sciences, University of Tsukuba. 1-1-1 Tennodai, Tsukuba, Ibaraki Prefecture 305-0006, Japan. 2. Biomedical Research Institute, National Institute of Advanced Industrial Science and Technology (AIST), Higashi 1-1-1, Tsukuba, Ibaraki 305-8566, Japan. 3. Traditional Uyghur Medicine Institute of Xinjiang Medical University, 393 Xinyi Rd, Urumqi, Xinjiang Uyghur Autonomous Region, 830011 China. 4. Advanced Technology Division, JEOL Ltd., Musashino 3-1-2, Akishima, Tokyo 196-8558, Japan. Correspondence should be addressed to Chikara Sato (ti-sato@aist.go.jp). The structure-function relationships of tissues are key to understand their mechanisms, which should be precisely studied. Optical microscopy (OM) has been developed in various ways and has led to great discoveries and advances in biology and medicine [1, 2]. The tissue samples are prepared by paraffin embedding or quickly frozen. Cryo-thin-sectioning is used for intra-operative cancer diagnosis mainly based on cellular nucleus size, although it is difficult and the whole procedure takes about 15-30 minutes for each sample [3]. Standard electron microscopy (EM) has subnanometer or nanometer resolution, but samples must be observed in vacuum. For biological samples, transmission electron microscopy (TEM) observation, requires time-consuming tissue preparation including fixation, dehydration, embedding, ultra-thin sectioning and staining. On the other hand, in the atmospheric scanning electron microscope (ASEM), a 2 � 3 m-thick layer of the sample can be observed at high resolution without dehydration [3-5]. The applicability of ASEM for the quick observation of cancer-metastasized tissue and normal tissue in liquid was studied. ASEM was successfully used for the observation of cerebellum at high resolution, and a quick staining method for intra-operative cancer diagnosis was further developed. Excised mouse tissue slabs (cerebrum, cerebellum, spinal cord, optic nerve fiber, kidney, liver, esophagus, stomach, intestine, cardiac, tongue, ear skin and skeletal muscle) were fixed with 4% paraformaldehyde (PFA) and 1% glutaraldehyde (GA), perforated with Triton X-100, stained with phosphotungstic acid (PTA) for 3 hrs, and imaged immersed in 10 mg/ml glucose solution by ASEM at 30 kV acceleration voltage [6]. In cerebellum, three distinct cortical layers were observed (the molecular layer, the Purkinji cell layer and the granular layer) as well as cerebellar white matter (Fig. 1B). Neurons were connected by delicate systematic networks (Fig. 1C and D). At a higher magnification, nuclei including brightly stained patches, presumably heterochromatin and nucleoli were observed (Fig. 1D). These results are consistent with those obtained by hematoxylin-eosin (HE) staining and OM (Fig. 1A). Control lung tissue and lung tissue metastasized by breast cancer were also fixed, quickly stained with uranyl acetate (UA) for just 15 minutes without Triton X-100 treatment, and observed by ASEM. The nuclei located close to the surface of tissue slabs were clearly observed, and the size difference between normal and cancer cells was clear (Fig. 2). These results suggest the potential of the quick method for intraoperative cancer diagnosis. The results were comparable with those obtained when the tissue slabs were treated with Triton X-100 and stained with TI-Blue and PTA (data not shown). 1. Microsc. Microanal. 21 (Suppl 3), 2015 400 398 References [1] C Golgi, Bollettino dellaSociet� Medico-Chirurgica di Pavia, 13 (1898), p. 1-14. [2] SW Hell, Science, 316 (2007), p.1153-1158. [3] C Sato, JEOL news, 43 (2011), p.13-20. [4] H Nishiyama et al, J Struct Biol, 172 (2010), p. 191-202. [5] Y Maruyama et al, J Struct Biol, 180 (2012), p. 259-270. [6] N Memtily et al, International J Oncology, in press. Figure 1. (A) OM of an HE-stained horizontal thin-section of cerebellum, and ASEM images of a PTA-stained cerebellum slab (B-D). (B) Three clearly different layers - the molecular layer, Purkinji cell layer, the granular layer - are visible as well as cerebellar white matter. (C, D) Higher magnification image of the white rectangle in the preceeding panel. Scale bar 50 m in A and B panel, 10 m in C, D. Figure 2. Quick observation of lung tissue (A, B) and lung metastasized by breast cancer cells (C, D) by ASEM. Fixed lung tissue slabs without perforation were quickly stained with UA for 15 minutes, and observed by ASEM. (A, B) Normal lung. A normal thin wall structure with alveoli (a) and alveolar ducts (ad) was observed. The cell nuclei (arrow) located at the surface of the slab were brightly imaged. (C, D) Lung metastasized by breast cancer cells. Lung structures look fainter in the metastasized area (top right). Instead, the region was occupied by cells with unusually large nuclei (arrow). Scale bar: 50 m in A, C; 10 m in B, D.");sQ1[201]=new Array("../7337/0401.pdf","Elemental Analysis of Rat's Femoral Neck with Experimental Diabetes by means of Scanning Electron Microscopy","","401 doi:10.1017/S1431927615002809 Paper No. 0201 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Elemental Analysis of Rat's Femoral Neck with Experimental Diabetes by means of Scanning Electron Microscopy Gabriel Herrera-P�rez1a, X�chitl Sof�a Ram�rez-G�mez2a, Rafael Vargas-Bernal1, Esmeralda Rodr�guezMiranda3. Departamento de Ingenier�a en Materiales, Instituto Tecnol�gico Superior de Irapuato, M�xico. Departamento de Enfermer�a Cl�nica, Divisi�n de Ciencias de la Salud e Ingenier�a, Universidad de Guanajuato Campus Celaya, M�xico. 3 Departamento de Medicina y Nutrici�n, Divisi�n de Ciencias de la Salud, Universidad de Guanajuato Campus Le�n, M�xico. 2 1 Diabetes mellitus (DM) is a group of metabolic disorders, whose main characteristic is the hyperglycemia. The duration of the hyperglycemia as well as its severity, are the most important factors that establish its presence by intermediate or prolonged periods. Although, pathophysiological changes that accompany the DM are found in all body, there are specific clinical manifestations such as gradual vision loss, renal disease, and susceptibility to cardiovascular diseases. In addition, the skeleton is affected to a significant level, where a major complication is osteoporotic bone fracture. Both patients with type 1 (T1DM) and type 2 (T2DM) diabetes mellitus have a high risk to experience fracture (Vestergaard et al 2009), when its condition is compared to health individuals. Reported data suggest that patients with T1DM show a decrease in bone mineral density (BMD), which is attributed to the decrease of the bone formation during growth in children and adolescents (Hamann et al 2012). In contrast, adults with T2DM, BMD values are normal or slightly elevated; however, the risk of a fracture is high (Janghorbani et al 2007). This advises that these patients may have a poor bone quality, which cannot be estimated in conventional densitometry studies. In this paper, we use a model of experimental diabetes (ED) to evaluate macroscopically the femur; representing the part of the skeleton, in which fractures caused by the weaking of the skeleton occur more frequently. In this study, female Wistar rats were used, with 8 to 12 weeks old, and weighing 200 to 250 g. The slaughter and handling of animals was made considering the specifications outlined in the Official Mexican Standard NOM-062-ZOO-1999. Before the induction of ED, the rats were fasted for 4 hours, after this time, alloxan monohydrate (Sigma-Aldrich) was administrated intraperitoneally (180 mg/kg). Animals with blood glucose values > 150 mg/dL were considered diabetic. Rats were maintained under these conditions for 12 months before sacrifice. Prior to sacrifice, body weight and blood glucose level were recorded. The rats were sacrificed by cervical dislocation and immediately skeleton dissection was performed. To prevent damage to the bone by a chemical treatment, the bones were cleaned using beetles, Coleoptera order and Dermestidae family. Our results show that the glucose level increased significantly from 53 � 8 in the group for non-diabetic rats to values of *234 � 39 for the group of diabetic. *p < 0.05 student's t-test, n = 3 , two weeks after induction and kept under these conditions for 12 months. This glucose level in rats corresponds to the clinical condition of non-controlled diabetic patients. Femur macroscopic parameters such as weight, length, and diameter of the diaphysis decreased 40%, 10%, ~ 7% respectively; while the diameter of the upper and lower epiphysis increased 2%, 10%. Analysis by Scanning Electron Microscopy (SEM) shows that the group of rats with ED increased femoral neck diameter 30% resulting in widening and deformation of the head of the bone (see Figure 1A and 1B). a Authors with the contribution of first author. Microsc. Microanal. 21 (Suppl 3), 2015 402 (A) (B) Figure 1. Images of SEM for Non-Diabetic (A) and Diabetic (B) Rats 25x, Electron Source Gun, 15.0 KV, Spot Magnitude 4.0, Pressure 0.8 Torr. Clinical studies have shown that the femoral neck is the site in the body, where fractures caused by osteoporosis occur more frequently (Kosy et al 2013). Elemental analysis of samples by Energy Dispersive Spectroscopy (EDS-SEM) provides that the content of K is similar in both cases ~ 0.40 % (At-g). The content of Ca, Na, P, and O decreased in the diabetic group (Table 1), while for the case of N y Mg had an increase, all these differences were statistically significant (*p < 0.05 student's t-test). Table 1. Elemental analysis of the surface of the femoral neck. Group Non-Diabetic Diabetic Elemental Analysis EDS (% At-g) C 31.99 1.15 37.48* 3.18 N 6.87 0.18 7.76* 0.40 O 44.38 0.59 41.21* 2.84 Na 0.43 0.05 0.33* 0.05 Mg 0.18 0.03 0.21* 0.02 Ca 9.19 0.65 6.32* 1.18 P 4.95 0.40 3.90* 0.16 Others 2.13 N/A 0.80 N/A Ca/P 1.86 1.62 The stoichiometric ratio Ca/P in the hydroxiapatite (HA) is 1.667, however, it has been reported that an optimum value of this ratio in the femur of healty female Wistar rat is 1.664, and ~ 2.1 if they are rats with 8 to 12 weeks of age (Hern�ndez-Urbiola et. al. 2012). In this paper, the ratio of Ca/P of 1.856 was found in the HA for the group of non-diabetic rats and a value of 1.622 for the diabetic rats, both groups had 48 weeks of age when elemental analysis was performed. Although, the onset and development of osteoporosis is commonly asymptomatic; by reducing the weight, length and diameter of the diaphysis of the bone, does conclude that the experimental group with DM also developed secondary osteoporosis, that is, an effect caused by some other pathology and/or lifestyle and not due to age. Acknowledgement: This project was supported by CONCYTEG for 7th State Research Summer 2014. Hamann C, Kirschner S, G�nther KP, Hofbauer LC. 2012; 8(5): 297�305. Hern�ndez-Urbina M, Mineral Content and Physicochemical Properties in Female Rats Bone During Growing Stage, Absorption Atomic Spectroscopy 2012; 11: 201-220. Janghorbani M, Van Dam RM, Willett WC, Hu FB. Am J Epidemiol 2007; 166(5): 495�505. Kosy JD, Blackshaw R, Swart M, Fordyce A, Lofthouse R.A. J. Orthop Traumatol. 2013 Sep; 14(3):165-70. NORMA Oficial Mexicana NOM-062-ZOO-1999, Especificaciones t�cnicas para producci�n, cuidado y uso de los animales de laboratorio. http://www2.inecc.gob.mx/publicaciones/gacetas/220/062.html Vestergaard P, Rejnmark L, Mosekilde L. Calcif Tissue Int 2009; 84(1): 45�55.");sQ1[202]=new Array("../7337/0403.pdf","MacCallum's Triangle � Is It Rheumatic? Is It Traumatic? Or Is It Both?","","403 doi:10.1017/S1431927615002810 Paper No. 0202 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 MacCallum's Triangle � Is It Rheumatic? Is It Traumatic? Or Is It Both? S. Siew Division of Human Pathology, College of Osteopathic Medicine, Michigan State University, East Lansing, MI 48824-1316 Rheumatic carditis is a pancarditis, however, the parietal endocardium is affected to the least obvious degree. In 1924, MacCallum [1] described, in acute rheumatic fever, involvement of the endocardium of the posterior wall of the left atrium, above the base of the posterior mitral cusp, associated with numerous Aschoff nodules in the subendocardium. This was called MacCallum's Patch or Triangle, because of its triangular distribution (Fig 1). The etiology was thought to be rheumatic. Today, this lesion is considered to be traumatic in origin, caused by the abnormally directed regurgitant stream of mitral regurgitation impinging upon the posterior wall of the left atrium, resulting in the formation of a systolic pocket. Our investigation of the parietal endocardium in acute rheumatic carditis, demonstrated the presence of acute parietal endocarditis, with a reaction similar to that present in the valves [2]. A separate examination of the left atrial endocardium above the base of the posterior mitral cusp, showed the presence of prominence and palisading of the endocardial cells (Fig. 1), with irregular outpouchingprotrusion of the endocardial surface. Our investigation of the parietal endocardium in acute rheumatic carditis, demonstrated the presence of acute parietal endocarditis, with a reaction similar to that present in the valves [2]. A separate examination of the left atrial endocardium above the base of the posterior mitral cusp, showed the presence of prominence and palisading of the endocardial cells (Fig. 1), with irregular outpouching-protrusion of the endocardial surface. There were raised, irregular edematous nodules of the endocardium (Fig. 2), with an infiltration of elongated histiocytes, fibroblasts and nongranular cells (Fig. 3) and macrophages. There was evidence of neovascularization and the formation of collagen fibrils. An increase, was noted, in the smooth muscle components of subendocardium. An endocardial nodule (Fig. 4) showed organization of fibrinoid necrosis. This finding supports the postulate that the endocardial fibrosis results from the organization of the acute rheumatic process. A similar reaction was present in the mitral valve (Fig. 5). The cusp was edematous, with an infiltration of fibroblasts, elongated histiocytes and nongranular cells. Neovascularization was present with periarteriolar fibrosis and a nongranular cell infiltration (Fig. 6). We propose that the pathogenesis of MacCallum's Triangle is initiated by the rheumatic process, which undergoes organization and fibrosis. The localization of the lesion to the posterior wall of the left atrium is determined by the traumatic impingement upon it by the regurgitant stream of mitral regurgitation. References [1] MacCallum, WG (1924): Rheumatic lesions of the left auricle of the heart. Bull Johns Hopkins Hosp.,: 35, 329 [2] Siew, S (2008): Rheumatic Heart Disease Revisited: Acute Parietal Endocarditis. Another Piece of the Unsolved Puzzle. Microsc. Microanal. 1552 Microsc. Microanal. 21 (Suppl 3), 2015 404 Fig. 1: Graphic depiction of MacCallum's Triangle. Spreading vegetations of subacute bacterial endocarditis Fig. 2: Raised, irregular edematous nodules on the endocardium (original magnification 100x) Fig. 3: Endocardial nodule infiltration of elongated histiocytes and fibroblasts (original magnification 400x) Fig. 4: Fibrinoid necrosis in endocardial nodule (original magnification 200x) Fig. 5: Mitral valve edematous cusp (EC); papillary muscle (PM); chordae tendinae (CT) (original magnification 20x) Fig. 6: Mitral valve � vascularization (V) (original magnification 200x)");sQ1[203]=new Array("../7337/0405.pdf","Nanoscale 3D Refractive Indices Mapping on Native Cheek Cells by Axial Scanning Transmission Electron Tomography.","","405 doi:10.1017/S1431927615002822 Paper No. 0203 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Nanoscale 3D Refractive Indices Mapping on Native Cheek Cells by Axial Scanning Transmission Electron Tomography. Yue Li1, Jinsong Wu3,, Reiner Bleher3, Vinayak Dravid3, and Vadim Backman2 1. 2. Applied Physics Program, Northwestern University, Evanston, Illinois 60208, USA. Department of Biomedical Engineering, Northwestern University, Evanston, Illinois 60208, USA. 3. Department of Materials Science and Engineering, Northwestern University, Evanston, Illinois 60208, USA. Scanning transmission electron microscopy (STEM) has been engaged in the study of ultrastructure of biological samples both qualitatively and quantitatively since the 1970s[1] .Equipped with secondary electron (SE) detector, axial (TE) detector, and annular dark field (ZE) detector, STEM is able to image with secondary electron contrast, phase contrast, as well as mass-thickness contrast of the same region easily. In particular, high angle annular dark field (HAADF) detector collects electrons been scattered above a pre-defined large angle. The contrast given by those high-angle scattered electrons, which are described by Rutherford's scattering model, is proportional to mass. Thus nano-scale, quantitative 2D projection mass mapping of biological complex can be done by calibrating HAADF images with molecules with similar chemical composition and known mass or density[2]. However, 2D projection data is notoriously hard to interpret, if not impossible. To tackle the problem, 3D tomography information is needed. It has been proven that 3D tomography of a whole human erythrocyte infected with the malaria parasite Plasmodium falciparum can be reconstructed by tilting each of the thick serial cryo-sections (around 1�m)with 8-10 nmresolution using STEM axial detector[3]. Field carcinogenesis, also known as cancer field effect, has two most distinguishable characteristics: (a) It occurs prior to histological changes; (b) It's not restricted in areas that later develop into malignant tumors[4]. In clinical study, it is optical imaging and spectrometry that are widely applied, instead of time consuming and skill demanding electron microscopy. Subramanian et al. [5] in Backman's lab in Northwestern have successfully developed a fast, non-invasive method named Partial Wave Spectrometry (PWS) to detect field carcinogenesis in colon cancer and lung cancer. The resolution of PWS is around 50 to 100 nm, which is far beyond the 200 nm diffraction limit of light. To validate PWS, we need a 3D refractive index map of the entire cell. For biological samples, the local refractive index is linear to the local density [6]. So by measuring mass distribution and topographic reconstruction, we can calculate refractive index accordingly. To achieve ultrahigh resolution, cells are usually sectioned to less than 300 nm thick for STEM, and even thinner for convention TEM However, in order to keep cells in the most native state, neither sectioning nor staining are acceptable. Maintaining the quality of information in native samples requires longer pixel dwell time thus require longer electron exposure, but radiation damage becomes significant with more time under the beam. There is a trade-off. For the purpose of field carcinogenesis detection, getting information from an intact cell dominates. Despite its importance, the micro-scale refractive index map of an entire cell in its native state has not been investigated. In this study, we utilized the axial STEM imaging and tomography reconstruct a 3D refractive index map of a cheek cell in its native state. The cheeks cells are fixed by glutaraldehyde and formaldehyde then critical point dried to keep structure. Figure 1 shows TE-STEM images at -60�, 0�, and 60�. Since Microsc. Microanal. 21 (Suppl 3), 2015 406 the intensity in TE mode for thick samples obeys Bill-Lambert Law [7], we then calibrate the intensity with mass using polystyrene beads. The alignment of STEM images is done by patch tracking algorithm and SIRT algorithm embedded in IMOD. After reconstruction, the intensity is converted to mass then to refractive index. Figure 2 shows the 3D structure of the nucleus and organelles after volume rendering and visualizing by Amira5.5. All STEM images were acquired using Hitachi HD-2300A STEM at room temperature with no significant beam damage occurred. Obtained mass distribution and refractive index map will be presented. References: [1]Engel, A. and C. Colliex, Application of scanning transmission electron microscopy to the study of biological structure. Current opinion in biotechnology, 1993. 4(4): p. 403-411. [2] Pennycook, S., Z-contrast transmission electron microscopy: direct atomic imaging of materials. Annual review of materials science, 1992. 22(1): p. 171-195. [3] Hohmann-Marriott, M.F., et al., Nanoscale 3D cellular imaging by axial scanning transmission electron tomography. Nature methods, 2009. 6(10): p. 729-731. [4] Backman, V. and H.K. Roy, Advances in biophotonics detection of field carcinogenesis for colon cancer risk stratification. Journal of Cancer, 2013. 4(3): p. 251. [5] Subramanian, H., et al., Nanoscale cellular changes in field carcinogenesis detected by partial wave spectroscopy. Cancer research, 2009. 69(13): p. 5357-5363. [6] Barer, R. and S. Joseph, Refractometry of living cells part I. Basic principles. Quarterly Journal of Microscopical Science, 1954. 3(32): p. 399-423. [7] Thomas, D., et al., Mass analysis of biological macromolecular complexes by STEM. Biology of the Cell, 1994. 80(23): p. 181-192. Figure 1.Axial-STEM images of an entire critical point dried cheek cell at (a) -60�, (b) 0� and (c) 60�. Figure 2. 3D reconstruction of the entire cheek cell. (a) virtual 2D slice in the middle of the cell, (b) segmentation of nucleus and organelles by mass, and (c) volume rendering of nucleus and organelles");sQ1[204]=new Array("../7337/0407.pdf","Is Electron Microscopy Relevant Anymore in Diagnosing Disease?","","407 doi:10.1017/S1431927615002834 Paper No. 0204 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Is Electron Microscopy Relevant Anymore in Diagnosing Disease? W.T Gunning Department of Pathology, University of Toledo, Toledo, USA Ultrastructural pathology has lost favor during the past decade for a variety of reasons, the most obvious due to the advent of antibody utilization to identify specific epitopes to characterize disease states, especially in terms of neoplasm diagnosis. The most significant use of electron microscopy (EM) for diagnostic purposes occurred in the 1960's through the 1970's during the very time that the ultrastructural anatomy of the cell was being described. The use of EM in pathology remained strong through the 1990's but a number of pressures began to take hold in the 90's that ultimately caused widespread closure of EM facilities in community tertiary care hospitals and continued into the 21st century with many academic hospitals shutting the doors of their EM facilities. The common criticisms of using EM for diagnostic purposes for tissue samples included "It costs too much." and "It takes too long." and other underlying issues also included a lack of both professional and technical expertise. Immunohistochemical techniques (IHC) have become a mainstay of diagnostic protocols when evaluating tumor differentiation/etiology and immunofluorescence, using direct labeling of trapped immunoglobulins in kidney biopsies, is an essential tool to diagnose renal disease. The question at hand: Is electron microscopy relevant anymore in the practice of pathology? The rationale to criticize the use of EM in pathology is a fallacy! It is true that microscopes and ancillary equipment are expensive, however these instruments are functional for many more years than most instruments utilized in clinical labs. Electron microscopy techniques have been developed that allow tissue sections to be available within 24 hours or less of sample submission. A "secret" never discussed is that utilization of 3-5 antibodies for diagnosis is more than the cost of EM. In fact, some protocols employ as many as 8-10 different antibodies, dependent upon tumor type, and it is well known that antibody specificity and sensitivity may be questionable in some instances. The utilization of the "brown" stains has been of significant benefit without question, as most hospital laboratories can employ IHC reliably for ensuring a correct diagnosis of many cancers. Is electron microscopy relevant anymore in the practice of pathology? The answer is unequivocally: YES! The utilization of EM in renal pathology is essential; once waning in use for early transplant rejection, it is now very important to aid in diagnosis of humoral rejection by evaluating interstitial capillary basement membranes for evidence of structural alteration. Its use in routine work-up of patients presenting with nephrosis or nephritis remains essential for accurate diagnosis of renal disease. Muscle disease presents a variety of conditions that can be readily diagnosed using ultrastructural evaluation. Clinical history must guide interpretation of muscle biopsies and EM may be essential to correctly diagnose inclusion body myositis from polymyositis for which treatment differs. Metabolic disorders also present a challenge and EM serves a role to guide potential molecular investigations to determine genetic mutations. Bleeding diatheses may be extremely difficult to diagnose without the use of EM. Microsc. Microanal. 21 (Suppl 3), 2015 408 Yes, electron microscopy is still relevant in the practice of pathology; it isn't utilized as often as in years past but in a number of situations, a diagnostic dilemma may be easily resolved by employing ultrastructural evaluation. This presentation will describe some cases of the "wrong diagnosis" and demonstrate ultrastructural features that "saved the day" to make an accurate interpretation of disease. Examples of ultrastructural pathology will also be discussed to underscore the relevance of EM in the practice of pathology. 1. 3. 5. 2. 4. 6. Figure 1. Glomerulus from a diabetic patient with light microscopy features that were misdiagnosed as diabetic nodular sclerosis (K-W diabetic nephropathy). Figure 2. Ultrastructural morphology of a glomerulus as seen in Fig. 1, demonstrated significant accumulations of immunotactoids; the diabetic patient had a monoclonal gammopathy. Figure 3. Nucleated red blood cell from a bone marrow biopsy is characteristic of congenital erythropoeitic anemia, Type 1 (CDA-1). Figure 4. Branched glycogen accumulation in cardiac tissue, consistent with a glycogen storage disease. Figure 5. Accumulation of glycogen in hepatocytes representing a different glycogen storage disease. Figure 6. This image was used to make a diagnosis of pancreatic carcinoma whereas the light microscopic diagnosis, utilizing a number of antibodies reactions against the tumor cells, was misdiagnosed as a primary peritoneal carcinoma arising from the ovary.");sQ1[205]=new Array("../7337/0409.pdf","Imaging the Alphavirus Exit Pathway","","409 doi:10.1017/S1431927615002846 Paper No. 0205 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Imaging the Alphavirus Exit Pathway Maria Guadalupe Martinez1, Erik-Lee Snapp2, 3, Geoffrey S. Perumal4, Frank P. Macaluso4 and Margaret Kielian1 1 2 Dept of Cell Biology, Albert Einstein College of Medicine, Bronx, NY 10461, USA Dept of Anatomy and Structural Biology, Albert Einstein College of Medicine, Bronx, NY 10461, USA 3 Gruss Lipper Biophotonics Center, Albert Einstein College of Medicine, Bronx, NY 10461, USA 4 Analytical Imaging Facility, Albert Einstein College of Medicine, Bronx, NY 10461, USA Alphaviruses include important and widely distributed human pathogens such as Chikungunya virus and the encephalitic alphaviruses. No vaccines or antiviral therapies are available for most of these viruses. Alphaviruses are small-enveloped RNA viruses and contain an internal nucleocapsid and an external lattice of the viral E2 and E1 transmembrane proteins. Alphavirus bud from the plasma membrane (PM) but the process and dynamics of alphavirus assembly and budding are poorly understood [1]. Budding of the alphavirus particle requires both the capsid protein and the envelope proteins [2] and involves a one-to-one interaction of the cytoplasmic domain of E2 with a hydrophobic pocket on the capsid protein. Mutations in this critical region of E2 block E2-Cp interaction and inhibit budding [3]. Unlike the case for many less ordered enveloped viruses, structural and biochemical studies indicate that host proteins are strictly excluded from the mature alphavirus envelope [1]. Despite the fact that the structures and interactions of the alphavirus capsid and envelope proteins have been extensively characterized, many fundamental questions about alphavirus budding remain. Earlier electron microscopy data suggest that alphavirus budding takes place at localized PM sites. However, it is not clear how these sites are formed or specialized, which viral proteins are involved in the process of cell proteins exclusion from nascent particles, and what functional roles are played by host and viral proteins during assembly and budding. Such questions are amenable to study using imaging methods. We generated Sindbis viruses (SINV) with fluorescent protein labels on the E2 envelope protein and exploited them to characterize virus assembly and budding in living cells. These viruses will be referred to as WT-GFP and WT-mCherry. We then used these viruses to study budding using a combination of total internal reflection fluorescence microscopy (TIRFM), confocal microscopy, and correlated light and electron microscopy (CLEM). Our studies demonstrated that during virus infection E2 became enriched in localized patches on the PM and in filopodia-like extensions. These E2-labeled patches and extensions contained all of the viral structural proteins. We used CLEM methods to further characterize the E2 PM patches and extensions, and determine if they were sites of virus particle production. Representative SEM images of cells transfected with WT-mCherry or Y400K-mCherry RNA or mocktransfected are shown in Figure 1. The WT-mCherry-infected cells showed abundant particles across the cell surface, along the cell borders, and on plasma membrane extensions (Fig. 1A-B). Correlation between the fluorescence and SEM images of WT-mCherry-infected cells showed that the fluorescent particles observed on short extensions corresponded to nascent virus particles (Fig. 1C). The size of the particles is consistent with that of alphaviruses (~ 70nM in diameter). In contrast, SEM of cells infected with the non-budding mutant or of the uninfected control cells showed a smooth surface without particles (Fig. 1E-J). Thus, using CLEM we established that the patches and extensions co-localize with virus budding structures, while light microscopy studies showed that they exclude a freely diffusing PM marker protein. Exclusion of this PM marker required the interaction of the E2 protein with the capsid protein, a critical step in virus budding, and was associated with the immobilization of the envelope proteins on Microsc. Microanal. 21 (Suppl 3), 2015 410 the cell surface. Virus infection induced at least two different types of extensions: tubulin negative extensions that are shorter, exclude the PM marker and are completely occupied with viral particles when observed by CLEM, and longer extensions that are positive for tubulin staining, appear to mediate virus particle transfer and whose number is reduced in the non-budding mutant SINV. Together our data support a model in which alphavirus budding occurs at specialized sites of virus-induced reorganization of the PM and cytoskeleton. References: [1] Kuhn RJ (2007) Togaviridae: The Viruses and Their Replication. In: Knipe DM, Howley PM, editors. Fields Virology. Fifth ed. Philadelphia, PA: Lippincott, Williams and Wilkins. pp. 1001-1022. [2] Garoff H, Hewson R, Opstelten D-JE (1998) Virus maturation by budding. Microbiol and MolBiolRev 62: 1171-1190. [3] Zhao H, Lindqvist B, Garoff H, von Bonsdorff C-H, Liljestr�m P (1994) A tyrosine-based motif in the cytoplasmic domain of the alphavirus envelope protein is essential for budding. EMBO J 13: 42044211. [4] The authors acknowledge funding to M.K. from the National Institute of General Medical Sciences (GM-057454) and by Cancer Center Core support grant NIH/NCI P30-CA13330. Figure 1. Correlation of scanning electron microscopy (SEM) and fluorescence microscopy shows that E2 fluorescent extensions correlate to budding particles at the cell surface. Vero cells were transfected with WTmCherry (A), Y400K-mCherry RNA (E) or mock transfected (I) and incubated at 37�C for 6 h. Transfection media was replaced and after 3h cells were light fixed. Samples were imaged using a Zeiss AxioObserver microscope with "shuffle and find" software. Cells were then processed for SEM and imaged using a Zeiss Supra 40 Field Emission SEM. In the left panels the SEM image of a cell infected with WT-mCherry (A), Y400KmCherry (E) or mock infected (I) are shown (6000X). On the right panels (B,F,J) an inset of the region demarcated with the dashed black box overlaying with the fluorescence of the E2-mCherry proteins is shown (10000X) (bar= 1m).");sQ1[206]=new Array("../7337/0411.pdf","Localization and Number of Au Nanoparticles in Optically Indexed Cells by FIB Tomography.","","411 doi:10.1017/S1431927615002858 Paper No. 0206 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Localization and Number of Au Nanoparticles in Optically Indexed Cells by FIB Tomography. Aric W. Sanders1, Kavita M. Jeerage1, Alexandra E. Curtin1 and Ann N. Chiaramonti1 1. National Institute of Standards and Technology, Boulder, Colorado 80305, USA. Gold nanoparticles (GNPs) are gaining importance as therapeutic chemical delivery vehicles, medical diagnostic tools, and phototherapeutic and contrast enhancement agents. GNPs are uniquely suited for these biological uses because of their chemical stability, novel optical properties, and broad potential for functionalization. Additionally, each of these beneficial properties is further enhanced by the ability to manufacture GNPs in an almost endless combination of sizes and shapes. This versatility has allowed researchers to access and modify biological processes inside of a large variety of cells [1] and the observation of innocuous uptake of citrate stabilized GNPs [2]. To describe the effect of GNPs, characterization of affected cells and tissues is required from the macroscopic to nanoscopic level. In particular, the location of cells in the tissue or culture of interest and then the mapping of the number and spatial distribution of the GNPs inside of those cells is required, and frequently requires multiple imaging techniques [3]. We achieve the large scale mapping of mammalian stem cells using reflection optical microscopy and then explore the location and number of the nanoparticles inside these cells after exposure to 60 nm GNPs using focused ion beam � scanning electron (FIB-SEM) based tomography. To map the exterior of the cells at the size scale of a culture, here ~ 60,000 cells grown in vitro as a monolayer, to that of a single cell, we employ reflection optical microscopy. To access such different length scales, we stitch many optical fields of view together at low and high magnification. Figure 1 is a mosaic of 8 x 10 images stitched together using a 5x objective. In addition to the lateral mosaic we have also used several images at different objective-sample distances to extend the depth of field. Although reflection optical microscopy is not common in the biological sciences, it provides several distinct advantages for correlative imaging by optical and electron microscopy. First, sample preparation does not involve index matching fluid for optical microscopy, which removes a cleaning step in preparation for electron microscopy or the need for an environmental chamber. Second the sample viewing conditions closely mimic those of the FIB-SEM, allowing for the biological tissue or culture of interest to be viewed on the same specimen holder in the same orientation, making cross-tool registration easier. Finally, using a specialized optical microscope we can image large areas of the tissue or cell culture quickly and then create an index of specific cells of interest to be further investigated in FIB-SEM tomography (See figure1). Once the specimen has been mapped with optical microscopy, a small amount of conductive metal is deposited and the cells to be investigated are chosen. Individual cells are then milled using the ion beam and imaged using the electron beam creating a series of images of the interior of the cell. In order to enhance contrast and retain the external structure of the cell, cells have been fixed in formaldehyde, and desiccated by sequential rinses in a mixture of ethanol and phosphate-buffered saline, in which the concentration of ethanol increases. This preparation technique, while not highlighting intercellular structure, has been found to preserve the contrast between cells and nanoparticles, allowing for fast tomography that clearly resolves the location and number of nanoparticles inside of cells (see figure 2). Microsc. Microanal. 21 (Suppl 3), 2015 412 References: [1] S. Rana, A. Bajaj, R. Mount, VM Rotello Advanced Drug Delivery Reviews 64 (2012), p 200-216 [2] KM Jeerage, TL Oreskovic, AE Curtin, AW Sanders, RK Schwindt, Toxicology in Vitro 29 (1) (2015), p 187-194 [3] AW Sanders, KM Jeerage, CL Schwartz, AE Curtin, AN Chiaramonti, Microscopy and Microanalysis 20 (2014), p 976-977 Figure 1. Optical Imaging Based Maps of Cells of Interest. By mosaicking optical micrographs at different magnifications, cells of interest can be located in the in vitro culture and then sectioned to give a tomographic representation identifying nanoparticle location inside of the cell. Figure 2. SEM and FIB Tomography to Count Nanoparticles Inside of Cells. The repeated sectioning and imaging of a cell using FIB-SEM techniques creates a 3D representation of the cell, which allows for the determination of nanoparticle location and number. It is observed that nanoparticles inside of stem cells enter and reside in the cells as clusters, a single tomographic slice with a highlighted cluster is shown above (Bottom).");sQ1[207]=new Array("../7337/0413.pdf","Swept Confocally-Aligned Planar Excitation (SCAPE) Microscopy for High Speed Volumetric Imaging in Behaving Animals","","413 doi:10.1017/S143192761500286X Paper No. 0207 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Swept Confocally-Aligned Planar Excitation (SCAPE) Microscopy for High Speed Volumetric Imaging in Behaving Animals Elizabeth M. C. Hillman1,2, Matthew B. Bouchard1, Venkatakaushik Voleti1, Wenze Li1, C�sar S. Mendes3, Clay Lacefield4, Vanessa George5, Wesley B. Grueber6, Richard S. Mann3, Randy M. Bruno2,4, Kimara Targoff5. 1 2 Laboratory for Functional Optical Imaging, Departments of Biomedical Engineering and Radiology* Kavli Institute for Brain Science* 3 Mann Lab, Department of Biochemistry and Molecular Biophysics* 4 Bruno Lab, Department of Neuroscience* 5 ZFIN Lab, Department of Pediatrics** 5 Department of Physiology and Cellular Biophysics, Department of Neuroscience** *Columbia University, **Columbia University Medical Center, New York, New York 10027/32, USA, New contrast agents and genetic techniques have delivered complex organisms and in-vivo tissues painted with a palette of dynamically changing fluorescence [1]. These tissues are inherently threedimensional, yet methods to image volumes of tissue at the microscopic scale face many problems that make capturing high-speed dynamics a major challenge. Laser scanning confocal and two-photon imaging are generally limited by their need to serially visit single points within the volume. As a result, these modalities are almost always used to image a single plane rather than a volume of tissue if high acquisition speeds are required, causing them to miss information beyond the plane, and to be highly sensitive to sample movement [2]. Moreover, when 3D scanning is used, the standard method is to physically translate the objective lens up and down to sample different planes, a process that slows acquisition, increases the risk of mechanical failure and can disturb the living sample being imaged. Light-sheet microscopy benefits from parallel imaging of an entire plane of the sample in a single camera frame, achieving optical sectioning through the use of a thin sheet of light for illumination [3]. However, typical implementations of this approach use two orthogonally-aligned objective lenses, one to generate the light sheet and one to image it. This restricts the shape and size of the sample that can be imaged, while also imposing the need to translate the sample or the objective assembly relative to one another in order to generate a 3D image. In practice, this need for relative translation limits volumetric imaging speed and imposes the need to physically mount, hold and sometimes translate the sample [3]. Swept Confocally-Aligned Planar Excitation (SCAPE) microscopy is a new approach to fast 3D microscopy that overcomes many of the limitations of the techniques described above [4]. Firstly, SCAPE uses light-sheet illumination, but delivers it to the sample at an oblique angle through the same objective lens as is used to detect the resultant fluorescence, overcoming the need for two objective lenses (see Figure 1). Secondly, SCAPE sweeps this light sheet laterally (along x') using a scan mirror, as in confocal line-scanning microscopy, in order to sample a full volume of the tissue as a series of y'z' planes. Finally, SCAPE reflects the emitted fluorescence light off the same scan mirror (in our first prototype, an adjacent facet of a polygonal mirror mounted on a galvanometer motor), causing the light to be de-scanned as in confocal microscopy. The result is an oblique image plane that is stationary, and corresponds to an image of the tissue that is being optically sectioned by the sweeping oblique light sheet - the detection plane stays aligned with light sheet as the sheet moves. Image rotation optics map this stationary image plane onto a free-running high-speed camera (Andor Zyla-5.5-CL10) which can grab 2,560 x 80 (y'-z') planes at 2,400 fps. Since one sweep of the mirror corresponds to acquisition of an entire volume, there are no physical limits to acquisition speed except camera frame rate and signal to Microsc. Microanal. 21 (Suppl 3), 2015 414 noise: so a 200 or 100 (x') voxel wide, 80 layer deep volume can be acquired at 12 or 24 volumes per second respectively. Importantly, this volumetric imaging is achieved with no physical translation of the objective lens or sample, making SCAPE capable of imaging almost any un-mounted sample in an enface imaging geometry, with no concern for physical motion of components interfering with the sample or limiting imaging speed. SCAPE is currently implemented with a 488 nm DPSS CW laser source. c a galvo-mounted polygon mirror b objective lens de ed ct te l ig ht Drosophila larva (mCherry / GCaMP6) t t = 0 ms =0 t = 200 ms t = 400 ms 100 �m he et lig ht s d z = -38 �m z = -75 �m In-vivo mouse brain (FITC-dx IV) cylindrical lens scan lens collimated tube laser input lens intermediate oblique image plane y' z' scan angle (x') oblique image plane (optional) DV2 image splitter 2D detector (Andor Zyla sCMOS camera) x z objective lens sample volume excitation emission emission filter SCAPE two-photon Figure 1. SCAPE prototype and results. A polygon mirror is swept back and forth at the volume rate. This scanning causes the oblique sheet of light (blue) formed in the y'-z' plane of sample to sweep back and forth along x'. The light emerging from the sample (green) is descanned from the same moving mirror, causing the oblique image produced to remain stationary, but to correspond to the moving plane illuminated in the sample. Image rotation optics and a (removable) dual-color image splitter project this image onto the face of a fast camera. (c-d) Examples of SCAPE microscopy in living organisms. c) shows a time-sequence of rendered SCAPE data acquired on a freely moving first instar Drosophila melanogaster (fruit fly) larva expressing mCherry and GCaMP6 in its muscles (lexO-GcaMP6f;mhclexA,lexO-mCherry::CAAX) at 10 volumes per second (VPS) during a `head-lift'. d) Shows a comparison between SCAPE and two-photon microscopy of vasculature in living mouse brain. SCAPE images of neural activity in superficial dendrites have been recorded at 10 VPS in awake, behaving mice. See [4] for more details and results. Acknowledgments / conflict of interest: Kavli Foundation, NIH (NINDS) R21NS053684, R01 NS076628 and R01NS063226, NSF CAREER 0954796, DoD MURI W911NF-12�1-0594, HFSP, Wallace H. Coulter Foundation, NSF BioIGERT (OTASI), NSF GRFP and NDSEG. Conflict of interest: A patent on this technology (US20120140240) is held by the Trustees of Columbia. 1. Akerboom, J., et al., Genetically encoded calcium indicators for multi-color neural activity imaging and combination with optogenetics. Frontiers in Molecular Neuroscience, 2013. 6. 2. Cotton, R.J., et al., Three-dimensional mapping of microcircuit correlation structure. Front Neural Circuits, 2013. 7: p. 151. 3. Ahrens, M.B., et al., Whole-brain functional imaging at cellular resolution using light-sheet microscopy. Nature Methods, 2013. 10(5): p. 413-20. 4. Bouchard, M.B., et al., Swept confocally-aligned planar excitation (SCAPE) microscopy for highspeed volumetric imaging of behaving organisms. Nature Photonics, 2015. 9(2): p. 113-119.");sQ1[208]=new Array("../7337/0415.pdf","Correlative large volume imaging across scales","","415 doi:10.1017/S1431927615002871 Paper No. 0208 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Correlative large volume imaging across scales Rasmus R. Schr�der1,3, Holger Blank2, Andreas Schertel2, Marlene Thaler2, Alexander Orchowski2, and Irene Wacker3 1 2 Cryo EM, CellNetworks, BioQuant, Universit�tsklinikum Heidelberg, Heidelberg, Germany Carl Zeiss Microscopy GmbH, Oberkochen, Germany 3 Cryo EM, Centre for Advanced Materials, Universit�t Heidelberg, Heidelberg, Germany For correlative imaging of large volumes integrated workflows allowing shuttling of samples between different imaging modalities would be ideal. X-ray microscopy (XRM) is one modality, which in principle can bridge the gap between macroscopic and microscopic world. Here we have started to assess how XRM, light microscopy (LM) and electron microscopy (EM) might be integrated to gain comprehensive information about biological samples extended in 3D. In CLEM, correlated LM and EM, fluorescence LM is commonly used to define a certain functional state, which is then put into structural context using EM. That this is also possible for XRM has been shown for single cells grown on TEM grids which were plunge frozen and analyzed by cryo-XRM [1]. For tissue samples such an approach is not possible because samples thicker than 200 micrometers cannot be vitrified. In that case a different workflow is required, starting with a chemical fixation. In LM and XRM samples may be then imaged directly in aqueous medium, XRM being able to yield voxel sizes down to 350 nm. However, if ultrastructural resolution in the range of few nanometers is required XRM needs to be complemented by EM. There are several options to achieve that for large volumes [2] based on SEM imaging such as serial blockface or focussed ion beam scanning electron microscopy (SBFSEM or FIBSEM) or array tomography (AT). SBFSEM is optimal for connectomics where whole brains need to be imaged at high resolution. In other areas of cell and developmental biology the biggest part of the sample surface being imaged may not be interesting for the question being addressed, only a minute part of it may contain the target structure. So the question of how to identify this target is of eminent importance. Here we employ XRM to identify a rare event such as the formation of an immunological synapse (IMS) or a rare structure such as the neuromuscular junction (NMJ) within a large volume. We are using the new solution ZEISS Atlas 5 (Carl Zeiss Microscopy GmbH) which offers a sample centric correlative environment fusing all available 2D and 3D data of the sample from various modalities. It contains modules for automated SEM large area imaging and targeted crossbeam (XB) nanotomography. Based on a large volume XRM dataset it thus permits an efficient approach for nanometer scale analysis of identified buried features of interest using Crossbeam technology. Figure 1 illustrates typical workflows for large volume samples such as muscle tissue or cell pellets. Important is the definition of a reference framework, which is inherent to the samples of interest. This "sample coordinate system" needs to be mapped seamlessly from one imaging technique to the next, which then allows for a fast and accurate navigation of the sample in 3D. [1] [2] [3] Hagen et al., J Struct Biol 177 (2012) 193-201; Chicon et al., J Struct Biol 177 (2012) 202-211. I Wacker, RR Schr�der, J Microscopy 252 (2013) 93-99. The authors acknowledge C Bartels and L Veith for technical support, C Grabher for samples, BMBF for NanoCombine grant FKZ 13N11401 and MorphiQuant grant FKZ 13GW0044. Microsc. Microanal. 21 (Suppl 3), 2015 416");sQ1[209]=new Array("../7337/0417.pdf","Latest Developments for In Situ Measurements of Gas�Solid Interactions","","417 doi:10.1017/S1431927615002883 Paper No. 0209 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Latest Developments for In Situ Measurements of Gas�Solid Interactions Renu Sharma,1 Pin Ann Lin, 1,2 Matthieu Picher, 1,2 Zahra Hussaini1 1. Center for Nanoscale Science and Technology, National Institute of Standards and Technology, Gaithersburg, MD 20899-6203; 2. Institute for Research in Electronics and Applied Physics, University of Maryland, College park, MD 20740 The power of in situ observations at elevated temperatures to elucidate the details of gas-solid interactions was realized during the very early years of the advent of transmission electron microscope [1]. Of several designs developed between 1950 and 1970, the underlying approach of using a differential pumping system, as proposed by Swann and Tighe [2], has been further developed and is now widely used in commercially-available environmental transmission or scanning transmission electron microscopes (ETEM or ESTEM). In recent years ESTEM, has been successfully employed to reveal and understand the structural and chemical changes occurring in nanoparticles under reactive environments [3,4]. However, quantitative measurements of reaction rates and chemical changes are precluded by (a) the nanoscale region used for atomic imaging, (b) uncertainties in the actual temperature of the region under observation, and (c) the large amount of data to be analyzed. Here we present various technical and analytical methods we have developed to address these issues. First, we have incorporated a free-standing, broad-band light-focusing system in the ESTEM to excite and collect vibrational and optical spectra from TEM samples under reactive environments.[5] We have used it collect Raman spectra during the in situ growth of single-walled carbon nanotubes (SWCNTs) from 80 �m2 area ( 10 �m diameter laser beam) while collecting atomic scale images from a 200 nm2 area at 625 �C in 0.005 Pa of C2H2. Figure 1 shows the growth kinetics as measured (a) from the change in intensity of the G band of the SWCNTs in the Raman spectra and (b) of an individual tube from high resolution images extracted from a video sequence. It is important to note the overall shape of the growth curve and growth regimes are comparable for the two measurements, confirming that atomicscale data for this reaction is representative of the entire sample. The shifts in the Raman peaks with increasing temperature were also used to calibrate the sample temperature. For a membrane heating chip, we find that the temperature drops as we move away from the central hole (Fig. 2a). Moreover, the temperature drop due to introduction of N2 was also measured as a function of pressure for different temperatures (Fig. 2b). In order to obtain quantitative information from the large number of images generated during in situ measurements, we have developed a scheme that utilizes a combination of home-built and publicallyavailable algorithms for image drift correction, noise reduction and template matching to accurately locate the position of atomic columns. The latter employs a Delaunay triangulation to connect each point to its nearest neighbors, after which, the average nearest neighbor distance for each point is calculated. For each triangle, a value of the local spacing is calculated as the average of the average nearest neighbor distances of its three points to determine the structure. Fig.3 shows the fluctuation between the Co and Co2C area fraction during the SWCNT growth. A detailed description of the techniques and their applications will be presented. References: [1] Butler and Hale, (1981) Practical Methods in electron microscopy, Elsevier/North Holland Co. [2] Swann and Tighe, Proc. 5th Eur. Reg. Cong., p. 436 Microsc. Microanal. 21 (Suppl 3), 2015 418 [3] Sharma, R., J. Mat. Res. 20 (2005), p. 1695. [4] Hansen et al, Science 294 (2001), p. 1508. [5] Picher et al, Ultramicroscopy 150 (2015), p. 10. Figure 1. A: SWCNT growth rate measured from the evolution of G band in Raman signal. B: linear growth rate of an individual SWCNT measured from atomicresolution video data. Note the similarity in the growth curves. Figure 2. a) Temperature variations measured from the Raman shifts away from the center of a membrane heating chip. b) Temperature drop with increasing N2 gas pressure at different temperatures. Note the advent of the drop is dependent on both the pressure and initial temperature. Figure 3. Fluctuation in the size of Co (marked blue) and Co2C (marked violet) areas during the growth of SWCNT as measured using the image analysis method described in the text. The time is given at the right hand corner. These modulations can be directly related to the growth rate of SWCNT");sQ1[210]=new Array("../7337/0419.pdf","Electron Microscopy Advances for Studies of Catalysis at Atomic-Resolution and at Ambient Pressure Levels","","419 doi:10.1017/S1431927615002895 Paper No. 0210 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electron Microscopy Advances for Studies of Catalysis at Atomic-Resolution and at Ambient Pressure Levels C. F. Elkj�r1,2, S. B. Vendelbo1,3, H. Falsig1, I. Puspitasari3, P. Dona4, L. Mele4, B. Morana5, R. van Rijn6, B. J. Nelissen7, J. F. Creemer5, Ib Chorkendorff2, P. J. Kooyman3, S. Helveg1 1. 2. Haldor Topsoe A/S, Nym�llevej 55, DK-2800 Kgs. Lyngby, Denmark CINF, Technical University of Denmark, Fysikvej building 307, 2800 Kgs. Lyngby, Denmark 3. ChemE, Delft University of Technology, Julianalaan 136, 2628 BL Delft, The Netherlands 4. FEI Company, Acthtseweg Noord 5, 5651 GG Eindhoven, The Netherlands 5. DIMES-ECTM, Delft University of Technology, P.O. Box 5053, 2600 GB Delft, The Netherlands 6. Leiden Probe Microscopy BV, Niels Bohrweg 2, 2333 CA Leiden, The Netherlands 7. Albemarle Catalyst Company BV, P.O. Box 37650, 1030 BE Amsterdam, The Netherlands The size, shape and surface structure of nanoparticles affect their catalytic properties in ways that are difficult to predict. Because the nanoparticles tend to adapt a geometrical, compositional and electron structure dictated by the actual gaseous environment, it is important to characterize surface structure and dynamics under exposure to reaction conditions as close as possible to technological relevant conditions. Here we present recent advances of a micro-electro-mechanical system (MEMS) device enabling the first observations of nanoparticles catalyzing gas-phase reactions at ambient pressure levels by simultaneous high-resolution transmission electron microscopy (HRTEM), mass spectrometry (MS) and reaction calorimetry (CAL) [1-3]. The MEMS device contains a narrow gas flow channel extending just 4.5 �m along the electron beam direction. Herein, 15-nm-thin electron-transparent windows are placed, enabling HRTEM at ambient pressures. The windows are distributed between windings of a fourelectrode coil that allows heating and temperature-control of the reactor channel (Fig. 1). The nanoreactor is used for studying Pt nanoparticles (Fig. 1) under the oscillatory oxidation of CO. During exposure to CO-O2-He gas mixtures at 1 bar total pressure, aberration-corrected HRTEM reveals that the nanoparticles undergo reversible and periodic changes in their shape while the MS and CAL data shows synchronous oscillations in global CO conversion (Fig. 2). These non-invasive observations were ensured by conducting HRTEM at low electron dose-rates [3-5]. Specifically, as the reaction rate increases, the nanoparticle shape changes from a rounded towards more facetted shape and subsequently, as the reaction rate decreases, the shape reverts back towards the more rounded shape. Moreover, HRTEM reveals that the more facetted nanoparticles are terminated by extended (111) planes and that the more rounded nanoparticles are terminated by (111) planes and a higher abundance of step sites. The experimental observations of the oscillatory CO oxidation reaction are consistently explained by a time-dependent model that incorporates density functional theory and microkinetic modeling of CO oxidation on the (111) planes and step sites of Pt as well as a description of convection and diffusion in the gas flow channel. Specifically, the model demonstrates for the first time that a dynamic and reversible refacetting of Pt nanoparticles represents a mechanism for reaction oscillations. These findings therefore demonstrate that the nanoreactor is a beneficial complement to the multitude of in situ and operando spectroscopic techniques in heterogeneous catalysis and nanoparticle research, because gas-surface phenomena are uncovered at meaningful conditions at the atomic-scale. Microsc. Microanal. 21 (Suppl 3), 2015 420 References: [1] J.F. Creemer et al, Proceedings: IEEE MEMS (2011), p. 1103-1106. [2] S.B. Vendelbo et al, Ultramicroscopy 133 (2013), p. 72�79 [3] S.B. Vendelbo et al. Nature Materials 13 (2014), p. 884�890. [4] J.R. Jinschek, S. Helveg, Micron 43 (2012), p. 1156-1168. [5] S. Helveg, C.F. Kisielowski, J.R. Jinschek, P. Specth, G. Yuan, H. Frei, Micron 68 (2015) 176-185. Figure 1. CO oxidation catalyzed by Pt nanoparticles in the nanoreactor. (A) Optical micrograph of the nanoreactor channel with the reaction zone defined by the heater coil and electron-transparent windows for HRTEM. (B-C) Electron micrographs of an electron transparent window and of Pt nanoparticles dispersed on the window. (D) Low dose-rate HRTEM of a Pt nanoparticle in 1 bar CO:O2:He = 3:24:55. Figure 2. Oscillatory oxidation of CO catalyzed by Pt nanoparticles in the nanoreactor. Simultaneous, time-resolved (A) mass spectrometry of the CO, O2 and CO2 pressures, (B) reaction power and (C) HRTEM of a Pt nanoparticle. Reaction conditions: 1.0 bar of CO:O2:He at 3%:42%:55% and temperature is 659 K. Adapted from [3].");sQ1[211]=new Array("../7337/0421.pdf","Atomic-scale Dynamics in Catalysts for Sulfur Chemistry.","","421 doi:10.1017/S1431927615002901 Paper No. 0211 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic-scale Dynamics in Catalysts for Sulfur Chemistry. L.P. Hansenl, F. Cavalca1, C.C. Appel1, M. Brorson1, P. Beato1, J. Hyldtoft1, K.A. Christensen1, S. Helveg1 1 Haldor Topsoe A/S, Nymollevej 55, DK-2800 Kgs. Lyngby, Denmark In recent years, new opportunities for catalysis research have opened up due to a remarkable progress in transmission electron microscopy (TEM). Advances of differentially pumped vacuum systems and of closed gas cells, in combination with heating devices, enable observations of catalysts during the exposure to gasses at elevated pressure and temperature. Such in situ observations are of vital importance because catalysis is a surface phenomenon and because surfaces tend to restructure to adapt to the surrounding gas environment; strictly speaking, the catalytic active state is only present during catalysis. Information about surface structures and dynamics under conditions mimicking those encountered during catalysis is therefore very important to further improve the understanding of structure sensitive functionality and properties of catalysts. The different types of instrumentation provide complementary possibilities for in situ TEM experiments. In particular, the differentially pumped vacuum system is advantageous to exploit in situ TEM at the ultimate atomic resolution and sensitivity [1-2]. The differentially pumped vacuum system contains the gas phase as an integrated part of the microscope vacuum system. Thus, experiments involving gasses that strongly interact with the exposed surfaces are challenging, as cross-contamination of experiments and corrosion of electron optical elements, such as the objective lens or electron source, may occur. For those reasons, high-resolution TEM of catalysts in a sulfur-containing gas environment has not previously been emphasized despite their pivotal role in reduction of sulfur emissions that are harmful to environment and human health. To remedy this situation, we have dedicated a differentially pumped TEM to experiments with sulfur-containing gasses and the first in situ observations of catalysts involved in sulfur emission abatement will be presented. Specifically, the first in situ TEM experiments with sulfur-containing gasses focused on hydrotreating catalysts used for, e.g., the removal of S impurities in oil distillates [3]. The hydrotreating catalysts are activated by sulfidation of a supported MoO3 precursor into MoS2 nanocrystals that serve as the catalytic active phase. In the microscope, the gaseous sulfidation of a MoO3 precursor, atomically dispersed on a MgAl2O4 support, was monitored by means of time-resolved high-resolution TEM (Figure 1). The TEM images show that single-layer MoS2 nanocrystals form preferentially and that multi-layer MoS2 nanocrystals form later in the sulfidation process. Moreover, the single-layer MoS2 nanocrystals are observed to grow from steps and along the support surface, and the multi-layer MoS2 nanocrystals form layer-by-layer by the growth of additional MoS2 layers onto already formed single-layer MoS2 nanocrystals. As the catalytic active sites for S removal are associated with the MoS2 edges, the mechanistic insight explains how edge sites with better accessibility for more refractory S-containing compounds in the oil distillates can be obtained. This improved understanding of the formation of MoS2 edges helps tuning the catalyst activity. The second example concerns SO2 emission reduction from, e.g., steel industries. The industrial catalyst oxidizes SO2 to SO3 over vanadium oxides mixed with alkali metal pyrosulfates dispersed on a porous Microsc. Microanal. 21 (Suppl 3), 2015 422 silica carrier. The oxidation is followed by hydration of SO3 and condensation into sulfuric acid. In operation, the vanadia is dissolved in the molten pyrosulfate and forms a liquid film that is only present during the harsh reaction conditions. Detailed information about the catalytic active liquid species and their dynamic behavior in the pore system during SO2 oxidation has remained inaccessible. However, in situ TEM and Raman spectroscopy allows us to retrieve unprecedented new insight into the dynamic behavior and chemical functionality of SO2 oxidation catalysts in the active state [4]. References: [1] J.R. Jinschek and S. Helveg Micron 43 (2012), p. 1156. [2] S. Helveg, C.F. Kisielowski, J.R. Jinschek, P. Specht, G. Yuan and H. Frei, Micron 68 (2015), p. 176. [3] L.P. Hansen, E. Johnson, M. Brorson, S. Helveg, J. Phys. Chem. C. 118 (2014), p. 22768. [4] K. Christensen, F. Cavalca, P. Beato, S. Helveg, Sulphur Jan/Feb (2015). Figure 1. Time-resolved in situ high-resolution TEM image series of the growth of MoS2 nanocrystals consisting of (a) 1-layer (b) 2-layer, and (c) 3-layer MoS2 slabs. Growth conditions: 0.8 mbar of H2S:H2 =1:9 at 690�C. The images represent times of (a) 122 min, 138 min, 152 min (b) 225 min, 243 min, 257 min, and (c) 178 min, 192 min, 208 min, relative to the time t=0 min of reaching the temperature 690 �C. Sketches are included to guide the eye in identifying the MoS 2 slabs. The frame sizes are 17.3 nm x 7.9 nm. Adapted from [3].");sQ1[212]=new Array("../7337/0423.pdf","In Situ Observation of Ag Nanoparticle Catalyzed Oxidation of Carbon Nanotubes in an Aberration-corrected Environmental TEM","","423 doi:10.1017/S1431927615002913 Paper No. 0212 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In Situ Observation of Ag Nanoparticle Catalyzed Oxidation of Carbon Nanotubes in an Aberration-corrected Environmental TEM Yonghai Yue1,2, Datong Yuchi3, Jia Xu3, Lin Guo2 and Jingyue Liu1 1. 2. Department of Physics, Arizona State University, Tempe, Arizona 85287, USA School of Chemistry and Environment, Beihang University, Beijing 100191, PR China 3. School for Engineering of Matter, Transport and Energy, Arizona State University, Tempe, Arizona 85287, USA Carbon nanotubes (CNTs) [1], with their unique physicochemical properties, have attracted broad attention due to their potential applications [2]. A recent in situ study on the oxidation of different types of CNTs has revealed that the initiation temperature of CNT oxidation depends on the wall thickness of the CNTs [3]. Single- and double-wall CNTs can oxidize at lower temperatures and multi-wall CNTs can be stable in oxygen even at temperatures as high as 520 �C [3]. With the use of catalyst nanoparticles to facilitate the oxidation processes the initiation temperature and the degree of carbon oxidation can be drastically changed [4]. Understanding the behavior of catalyzed oxidation processes is critical to developing better catalysts for combustion of particulates including oxidation of soot particles [5]. We report here the in situ investigation of Ag catalyzed oxidation of multi-wall CNTs inside an aberration-corrected environmental TEM (AC-ETEM) with the goal of probing the nature of the catalyzed oxidation processes and thus developing better understanding of combustion catalysts. The CNTs, which were prepared via a CVD method, were suspended in ethyl alcohol and a drop of the solution was then transferred to a holey-carbon coated TEM grid. The FEI Titan G2 AC-ETEM, operating at 80 KV, was used to carry out the in situ oxidation experiment. The electron dose rate was kept low enough so that the electron beam exerts minimum influence on the integrity of the experimental results. A Gatan Inconel heating holder was used and the images were recorded by a standard CCD camera. To clean the CNTs, the TEM grid was heated to 450�C in 2 mbar O2 for 45 min inside the ETEM. Repeated experiments confirmed that such treatment did not oxidize the multi-wall CNTs. The CNTs were then loaded with AgNO3 aqueous solution and was then reduced inside the ETEM in 2 mbar H2 at 250�C for 2 hours, resulting in the formation of crystalline Ag nanoparticles. The electron beam was blanked during the sample treatment processes. Figure 1 shows AC-ETEM images of the CNTs after the sample treatment processes. Ag nanoparticles, with an average diameter of approximately 5 nm, decorated on the surfaces of the CNTs. After this initial quick examination the electron beam was blanked and the sample was oxidized in 2 mbar O2 at 250�C for 25 minutes. To avoid any potential ionization effects, the oxygen inside the ETEM chamber was completely purged after the oxidation reaction and before the electron beam was un-blanked [3]. The same sample region, before and after the oxidation reaction, was carefully examined and images were recorded. Fig. 1b shows the same area as that of Fig. 1a but after the oxidation reaction. The CNTs were clearly oxidized wherever they were in contact with the Ag nanoparticles. Figure 2 shows another pair of images of a single CNT before and after the oxidation reaction. Based on analysis of these and many other similar images we concluded that the Ag nanoparticles clearly catalyzed the oxidation of multi-wall CNTs at temperatures as low as 250�C, in contrast to the high temperature oxidation of CNTs without Ag nanoparticles [3]. Furthermore, the oxidation did not proceed along the length of, but across, the CNTs, a different process than that reported for the non-catalyzed oxidation of CNTs [3]. AC-ETEM Microsc. Microanal. 21 (Suppl 3), 2015 424 proves to be a powerful tool for studying the oxidation processes of carbon particulates including CNTs and soot particles [6]. References: [1] S. Iijima, Nature 354 (1991) p.56. [2] M. S. Dresselhaus in "Carbon Nanotubes: Synthesis, Structure, Properties, and Applications", ed. M. S. Dresselhaus, (Springer, New York) p.1. [3] A. L. Koh et al. ACS Nano 7 (2013), p.2566. [4] A. R. Leino et al. Carbon 57 (2013), p.99. [5] B. R. Stanmore and J. F. B. P Gilot, Carbon 39 (2001), p.2247. [6] This work was supported by the start-up fund of the College of Liberal Arts and Sciences of Arizona State University. The authors acknowledge the use of facilities in the John M. Cowley Center for High Resolution Electron Microscopy at Arizona State University. YY acknowledges China Scholarship Council (201406025060), National Natural Science Foundation of China (51301011) and Research Fund for the Doctoral Program of Higher Education of China (20131102120053). Figure 1. ETEM Images of multi-wall CNTs before (a) and after (b) oxidation at 250�C for 25 minutes, clearly demonstrating the Ag nanoparticle catalyzed oxidation of the multi-wall CNTs. Figure 2. ETEM images of a multi-wall CNT before (a) and after (b) oxidation at 250�C for 25 minutes, clearly revealing that the carbon oxidation proceeded across the CNT, not along its length.");sQ1[213]=new Array("../7337/0425.pdf","In Situ Environmental Transmission Electron Microscopy of Ice Nucleation","","425 doi:10.1017/S1431927615002925 Paper No. 0213 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In Situ Environmental Transmission Electron Microscopy of Ice Nucleation Lyle M. Gordon, Libor Kovarik Environmental Molecular Spectroscopy Laboratory, Pacific Northwest National Laboratory, 3335 Innovation Blvd. Richland, WA 99354 Heterogeneous ice nucleation on atmospheric aerosols results in cloud formation and represents one of the largest uncertainties in predicting climate. Elucidating the fundamental physicochemical and materials properties that result in highly active ice nuclei requires detailed observation of the ice nucleation process of model substrates in situ. Environmental transmission electron microscopy (TEM) represents a suitable tool for high-resolution analysis of nucleation in situ. Unlike a conventional TEM, gas can be introduced to the sample to simulate the reaction conditions of interest. Amorphous ice is frequently utilized by transmission electron microscopy (TEM) to study the structure of biological molecules frozen in a vitreous ice layer at liquid nitrogen temperature. Recently, Kobayashi et al. achieved atomically resolved image of ex situ and in situ nucleated ice nanocrystals using TEM [1-2]. These experiments demonstrate the possibility of conducting in situ ice nucleation in the TEM, however, the work presented was limited to cryogenic (< 100K) temperatures and high vacuum conditions in a conventional TEM. In situ investigation of atmospheric nucleation on natural aerosols requires simulation of atmospherically relevant conditions, specifically temperatures between 200-273K and water vapor partial pressures in the range of 10-3 � 10 mbar. I will present results from initial experiments performed in an FEI Titan ETEM. Experiments were performed with a custom-built gas mixer and a commercially available liquid nitrogen cooled cryogenic TEM holder equipped with a resistive heater and temperature controller. To study atmospheric heterogeneous ice nucleation processes, model aerosol nanoparticles were dispersed on an amorphous carbon substrate. The substrate was subsequently cooled and water vapor was introduced to the chamber. Multiple images were acquired throughout the nucleation process (Figure 1). Nucleation was frequently observed within high curvature regions between the nanoparticle and the substrate likely due to the Kelvin effect. This observation could be evidence that depositional ice nucleation (vapor-solid) is in fact a two-step process where liquid water condenses in high-curvature nanopores below bulk water saturation and then freezes by homogenous or heterogeneous nucleation [1-4]. References: [1] Kobayashi, K. et al, Physical Review Letter 106 (2011), p. 1. [2] Kobayashi, K. et al, Chemical Physics Letters 106 (2012), p. 9. [3] Marcolli, C., Chemical Physics Letters 14 (2014), p. 2071. [4] The research was performed using EMSL, a DOE Office of Science User Facility sponsored by the Office of Biological and Environmental Research and located at Pacific Northwest National Laboratory. Microsc. Microanal. 21 (Suppl 3), 2015 426 Figure 1. Selected frames from a movie of in situ nucleation of ice crystals on hematite nanoparticles supported on an amorphous carbon film. Substrate temperature: ~200K, H2O partial pressure during growth: ~10-2 mbar.");sQ1[214]=new Array("../7337/0427.pdf","STEM and EELS Investigation on Black Phosphorus at Atomic Resolution","","427 doi:10.1017/S1431927615002937 Paper No. 0214 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 STEM and EELS Investigation on Black Phosphorus at Atomic Resolution Giuseppe Nicotra1, Antonio Massimiliano Mio1, Anna Cupolillo2, Jin Hu3, Jiang Wei3, Zhiqiang Mao3, Ioannis.Deretzis1, Antonio Politano2, and Corrado Spinella1 1 2 CNR-IMM, Strada VIII, 5, 95121 Catania, Italy Dipartimento di Fisica, Universit� della Calabria, 87036 Cosenza, Rende, Italy 3 Department of Physics, Tulane University, New Orleans LA 70118 USA Recently, black phosphorus has attracted a huge attention for the discovery of unusual properties with high potential for microelectronics [1-4]. Black phosphorus is the thermodynamically more stable phase of phosphorus, at ambient pressure and temperature. Black phosphorus is a narrow-gap semiconductor with orthorhombic structure. It is a layered material, with each layer forming a puckered surface due to sp3 hybridization. The three bonds take up all three valence electrons of phosphorus, so unlike graphene, a monolayer black phosphorus (termed "phosphorene") is a semiconductor with a direct band gap of~ 1 eV at the point of the first Brillouin zone. For few-layer phosphorene, interlayer interactions reduce the band gap for each layer added, and eventually reach ~ 0.3 eV for bulk black phosphorus. The direct gap also moves to the Z point as a consequence. The presence of an energy gap is crucial for the field-effect transistor (FET) application of two dimensional (2D) materials, while for example graphene does not have an energy gap. Moreover, the thickness-dependent direct band gap may lead to potential applications in optoelectronics, especially for THz applications. Plasmons in black phosphorus have not been experimentally studied yet. Only a plasmon satellite at 20 eV has been identified in 2s and 2p core levels of P in X-ray photoemission spectroscopy (XPS) [5]. Moreover, STEM and EELS experimental end theoretical study are missing yet. For the first time, the direct atomic structure and electronic structure of the black phosphorus were studied and the results presented in cross-section. All the STEM and atomic EELS measurements were performed at 60kV and 200keV by state-of-the-art aberration-corrected microscope installation at Beyond-Nano sub-�ngstom Lab, in Catania, Sicily, Italy. This consists of a probe corrected STEM microscope equipped with a C-FEG and a fully loaded GIF Quantum ER as EELS spectrometer. This particular installation is capable to deliver a probe size of 0.68 � at 200kV, and 1.1 � at 60kV. Low and core-loss EELS spectra were nearly simultaneously acquired using the DualEELS capability. Figures 1 a) shows the simulated side view of the structure of bulk black phosphorus. Figures 1b) shows the experimental direct atomic structure acquired in STEM HAADF mode at 200keV of beam acceleration voltage with an illumination angle of 33mrad and a collection angle of 20mrad. Figures 2 a) show the plasmon peaked at 20eV and Figures b) shows the P L2.3 at 132eV and the P L1 at 189eV, both acquired with an illumination angle of 33mrad and a collection angle of 10mrad. Ab-initio simulation through FEFF of the EELS spectra will be also presented. References [1] Xia F, Wang H, Jia Y. Rediscovering black phosphorus as an anisotropic layered material for optoelectronics and electronics. Nat Commun. 2014;5:4458. [2] Buscema M, Groenendijk DJ, Blanter SI, Steele GA, Van Der Zant HSJ, Castellanos-Gomez A. Fast Microsc. Microanal. 21 (Suppl 3), 2015 428 and broadband photoresponse of few-layer black phosphorus field-effect transistors. Nano Lett. 2014;14(6):3347-52. [3] Fei R, Yang L. Strain-engineering the anisotropic electrical conductance of few-layer black phosphorus. Nano Lett. 2014;14(5):2884-9. [4] Han X, Morgan Stewart H, Shevlin SA, Catlow CRA, Guo ZX. Strain and orientation modulated bandgaps and effective masses of phosphorene nanoribbons. Nano Lett. 2014;14(8):4607-14. [5] Asahina H, Morita A. Band structure and optical properties of black phosphorus. J Phys C. 1984;17(11):1839-52. [6] This work was performed at Beyondnano CNR-IMM, which is supported by the Italian Ministry of Education and Research (MIUR) under project Beyond-Nano (PON a3_00363). Figure 1. a) Side view of the structure of bulk black phosphorus. b) cross-sectional view (HAADF STEM image) of bulk black phosphorus clearly showing the armchair arrangement of the P atoms. Figure 2. a) Plasmon and b)P L spectrum of black phosphorus");sQ1[215]=new Array("../7337/0429.pdf","Atomic Resolved Phase Map of MoS2 Monolayer Sheet Retrieved by Spherical Aberration Corrected Transport of Intensity Equation","","429 doi:10.1017/S1431927615002949 Paper No. 0215 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic Resolved Phase Map of MoS2 Monolayer Sheet Retrieved by Spherical Aberration Corrected Transport of Intensity Equation Xiaobin Zhang1 and Yoshifumi Oshima2 1. 2. Quantum Nanoelectronics Research Centre, Tokyo Institute of Technology, Tokyo, Japan School of Materials Science, JAIST, Nomi, Ishikawa, Japan An atomic sheet like graphene has attracted much interest for its unique physical and chemical properties. Using a spherical aberration corrected microscope, not only atomic defects [1] but also chemical bonding at the edge [2] have been measured. We think that retrieving a phase map, which represents the potential map, is also important in order to understand chemical reaction or charge distribution. Transport of intensity equation (TIE) is a convenient method for retrieving the phase map, since it does not require any special microscope attachments or a vacuum region. In this study, we retrieved atomic resolved the phase map of MoS2 monolayer sheet by TIE. In a TIE phase map, the defocus difference between two TEM images should be reduced in order to improve spatial resolution [3]. Aberration correction electron microscopy is thought to solve such a problem, because it has a high spatial resolution and show visible contrast variation by changing the sign or amount of defocus around in-focus condition. By using an aberration correction electron microscope, R005, having 50pm resolution, a MoS2 monolayer sheet was observed at an accelerating voltage of 300kV. The current density was kept as low as possible during observation. Through-focus series of high-resolution TEM images were obtained at 2 nm step in the range from -20 nm (under-focus) to 20 nm (over-focus). Atomic resolved phase maps were retrieved from two TEM images of -8 nm and 8 nm defocuses as shown in Fig. 1(a) and (b), as an example. An atomic-resolved TIE phase map was obtained by applying high pass filter to the original map as shown in Fig.1 (c), to remove low frequency noise. In Fig.1 (c), molybdenum (Mo) and sulfur (S) atomic columns seemed to correspond to the right and left peaks of the dumbbell curve, since the right peak is of slightly higher contrast than the left one. In order to confirm this statistically, 25 different phase profiles of the dumbbells were investigated. During analysis, the background was subtracted from the raw phase profile, since the background was modulated due to low frequency noise and so on. After subtraction, the left-side and right-side peaks of the dumbbell and the valley between both peaks were measured. The phase differences between the left-side peak and valley, and between the right-side peak and valley were obtained to be 0.045 and 0.058 radian on average, respectively. Phase map of MoS2 monolayer sheet were obtained by multi-slice simulation. It showed sharp peaks corresponding to Mo (0.435 rad.) and 2S, consisting of two sulfur, (0.459 rad.) atomic columns. But, actually, spatial resolution is reduced depending on defocus difference in the TIE phase map. Several phase maps were obtained by applying low pass filter of 20, 13.33, 10, 6.67, 5 nm-1 to the original phase map. And, Mo column, 2S column and the valley between them were measured. The phase differences between Mo (2S) column and valley were estimated to be 0.21 (0.20) (rad.), 0.12 (0.10), 0.062 (0.05), 0.015 (0.01) and 0.002 (0), respectively, for low pass filter of 20, 13.33, 10, 6.67, 5 nm-1. Since the experimental phase differences were 0.045 and 0.058 radian, we determined that the left-side peak correspond to 2S column and the right-side, to Mo column. This determination is in agreement with the Microsc. Microanal. 21 (Suppl 3), 2015 430 phase map of Fig.1 (c), showing the right peak is slightly higher contrast than the left one. And also, the spatial resolution was determined to be about 0.1 nm in the present phase map. In conclusion, we demonstrated that atomic resolved phase map of MoS2 monolayer sheet was retrieved by transport of intensity equation (TIE) using spherical aberration corrected transmission electron microscopy. The phase shifts by Mo and S atomic potential was confirmed to be measured quantitatively. The authors thank Prof. Takayanagi, Dr. Mitome and Dr. Ishizuka for constructive comments and fruitful discussions. This work was supported by the Japan Science and Technology Agency (JST) under the CREST project. References: [1] A Hashimoto et al., Nature 430 (2004) p. 870. [2] K Suenaga et al., Nature 468 (2010) p. 1088. [3] K Ishizuka, J. Electron Microscopy 54 (2005) p. 191. Figure 1. A TEM image of MoS2 atomic sheet (a) at under-defocus of 8nm and (b) at over-defocus of 8nm. (c) A TIE retrieved phase map, (d) Structure model of MoS2 atomic sheet and (e) simulated phase map, which was blurred by low pass filter.");sQ1[216]=new Array("../7337/0431.pdf","Aberration Corrected High Angle Annular Dark Field (HAADF) Scanning Transmission Electron Microscopy (STEM) and In Situ Transmission Electron Microscopy (TEM) Study of Transition Metal Dichalcogenides (TMDs)","","431 doi:10.1017/S1431927615002950 Paper No. 0216 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Aberration Corrected High Angle Annular Dark Field (HAADF) Scanning Transmission Electron Microscopy (STEM) and In Situ Transmission Electron Microscopy (TEM) Study of Transition Metal Dichalcogenides (TMDs) Jinguo Wang, Ning Lu, Juan Pablo Oviedo, Xin Peng, Guoda Lian, Moon J. Kim Department of Materials Science and Engineering, The University of Texas at Dallas, Richardson, Texas 75080, USA It is well documented that the dimensionality of materials plays a crucial role in determining their fundamental properties, in addition to the composition and arrangement of atoms. The most widely studied two-dimensional (2D) material to date is graphene, which exhibits exotic condensed-matter phenomena that are absent in bulk graphite [1]. Research in graphene and the methodology of preparing ultrathin layers has led to the exploration of other 2D materials [2]. In particular, single layers of transition metal dichalcogenides (TMDs) with lamellar structures similar to that of graphite have drawn significant attention because of their tunable bandgaps and abundance [2]. TMDs exhibit diverse properties that depend on their composition, including semiconductors (e.g., MoS2, WS2), semimetals (e.g., WTe2, TiSe2), metals (e.g., NbS2, VSe2), or superconductors (e.g., NbSe2, TaS2). The properties of TMDs also strongly depend on the crystalline structure and the number and stacking sequence of layers in their crystals and thin films [2]. Although electrical, magnetic, and mechanical studies of 2D materials as new material systems have been established, there is a lack of direct atom-by-atom visualization, limiting our understanding of these highly exciting material systems. Here, we present our recent studies on the characterization of 2D layered materials by means of Scanning Transmission Electron Microscopy (STEM), in particular via the techniques of High Angle Annular Dark Field (HAADF) imaging and in situ Transmission Electron Microscopy (TEM). The location and nature of individual atoms, defects, and layer by layer shearing of 2D crystals will be discussed. We have identified the atomic nature of single layer MoS2, 2H stacked TMDs (MoS2, MoSe2, WSe2, and MoTe2), 1T stacked TMDs (SnS2, SnSe2, HfSe2, and HfS2), as well as distorted 1T stacked TMDs (WTe2) by HAADF STEM imaging. Figure 1 presents HAADF STEM images of the (a) plane view and (b) cross section of 1T stacked HfSe2. In particular, we have successfully imaged distorted 1T stacked TMDs (WTe2) from three major crystalline orientations and determined the atomic species and their position in this crystal. Combined with first principle calculations, we investigated the characteristic atomic and electronic properties of the material. The interface and defects of various heterostructures of TMDs such as chemical vapor deposition (CVD) grown MoS2 on Gr/SiC, molecular beam epitaxy (MBE) grown HfSe2 on MoS2, molecular beam epitaxy (MBE) grown HfSe on SiO2, molecular beam epitaxy (MBE) grown SnSe on MoS2, and molecular beam epitaxy (MBE) grown Bi2Se3/MoSe2/SnSex on sapphire have been investigated by HAADF STEM at the atomic scale. These results demonstrate the feasibility and significant potential of fabricating 2D material based heterostructures with tunable band alignments for a variety of nanoelectronic and optoelectronic applications. Microsc. Microanal. 21 (Suppl 3), 2015 432 Furthermore, we have studied the coexistence of the orthorhombic crystal structure of SnSe and the hexagonal close-packed structure of SnSe2 in a molecular beam epitaxy (MBE) grown tin selenide film. We found that both the defect structure and the local composition fluctuation contributed to the phase transition between SnSe2 and SnSe. This provides an in-depth understanding on the design and synthesis of the SnSe2 heterostructures. Lastly, we have studied the layer by layer shear exfoliation of a MoS2 monolayer in a high electric field environment by using an in situ transmission electron microscopy nanoprobing technique. Assisted by first principles calculations based on density functional theory, we have demonstrated the ability to extract the zero-load shear strength value of few atomic layers of MoS2 with the capability of high resolution structural characterization of the sheared interface. Our findings can be used to understand and increase the efficiency of high shear mixing for the industrial production of monolayers from 2D materials, as well as a means to extend our current knowledge of tribology in nanoscale interfaces [3]. References: [1] A. H., Castro Neto, F. Guinea, N. M. R. Peres, K. S. Novoselov & A. K. Geim, Rev. Mod. Phys. 81, 109�162 (2009). [2] Q. H. Wang, K. Kalantar-Zadeh, A. Kis, J. N. Coleman & M. S. Strano, Nature Nanotech. 7, 699� 712 (2012). [3] J. P. Oviedo , S. KC , N. Lu , J.G. Wang , K. J. Cho , R. M. Wallace , and M. J. Kim, ACS Nano, 2015, Article ASAP. [4] This work is supported in part by the SWAN Center, a SRC center sponsored by the Nanoelectronics Research Initiative and NIST; by the Center for Low Energy Systems Technology (LEAST), one of six centers of STARnet, a Semiconductor Research Corporation program sponsored by MARCO and DARPA; and by the Louis Beecherl, Jr. endowment funds. HfSe2 plane view HfSe2 MoS2 Figure 1. HAADF STEM of the (a) plane view and (b) cross section of 1T stacked HfSe2 on MoS2");sQ1[217]=new Array("../7337/0433.pdf","Defect Dynamics in 2D Transition Metal Dichalcogenide Monolayers","","433 doi:10.1017/S1431927615002962 Paper No. 0217 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Defect Dynamics in 2D Transition Metal Dichalcogenide Monolayers Junhao Lin1,2, Yuyang Zhang1, Sokrates T. Pantelides1,2, Wu Zhou2 1. 2. Department of Physics and Astronomy, Vanderbilt University, Nashville, TN 37235, USA Materials Science & Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA Transition-metal dichalcogenide (TMDC) monolayers are promising atom-thin semiconductors for flexible nanoelectronic and optoelectronic applications [1, 2]. Intrinsic defects, such as vacancies, dislocations and boundaries, are known to have profound influence on the electronic and optical properties of these materials. For instance, chalcogen vacancies are the most common point defects in TMDC monolayers, and largely contribute to the n-type conductivity in CVD-grown MoS2 monolayers [3]; 60� grain boundaries in semiconducting TMDC monolayers (Fig.1A) have been shown to be metallic and quench the local photoluminescence [4,5]. In order to fully develop the potential of TMDC monolayer, it is useful to recognize and understand the dynamical behavior of structural defects in these monolayers. Here we show that the evolution of defects in TMDC monolayers not only can be directly monitored and explained at the atomic scale, but can also be utilized to create new nanostructures with desired properties. The experiments were performed on an aberration-corrected Nion UltraSTEM-100, operated at 100 kV. We first use a focused electron beam to generate and excite Se vacancies in the monolayer MoSe2, and monitor their dynamics through sequential atomic-scale annular dark field (ADF) imaging. We find that Se vacancies are first randomly created by mild ionization damage and then preferentially agglomerate into line defects under the energy transferred from the electron beam. Successive evolution of these line defects induces nucleation of distinct triangular inversion domains within the MoSe2 layer and generates conductive 60� grain boundaries within the semiconducting matrix (Fig.1B). Using density functional theory (DFT) calculations, we explain the driving force of the formation mechanism: the aggregation of chalcogen vacancies induces large lattice shrinkage, which causes the nearby metal atoms to undergo displacements and subsequently nucleates inversion domains in order to release the lattice strain. Migration of the grain boundaries can be further activated by deformation of the peripheral lattice, giving rise to the growth of the inversion domain [6]. Extensive irradiation generates both metal and chalcogen vacancies and activates their complex dynamics. Prolonged illumination of a small region of the TMDC monolayer with an intense focused electron probe can eventually lead to formation of small holes. By selectively patterning the holes, we recently reported the direct fabrication of ultrathin MX (M=Mo, W; X=S, Se) nanowires, including their ramified junctions, connecting designated points within a semiconducting TMDC monolayer (Fig. 2) [7]. These nanowires are intrinsically metallic, thus serving as promising candidates for ultrasmall interconnects within the 2D integrated circuit. Moreover, we show that these nanowires maintain extreme structural flexibility, as they remain conducting when being twisted or kinked. We will show that the chemical composition of such nanowires can be tuned by successfully fabricating intermixed MoS1-xSex and Mo1-xWxS nanowires. The concentration of S and Se in the MoS1-xSex nanowire is controlled by the acceleration voltage during the electron-beam sculpting. The transport properties of these intermixed nanowires are further studied by first-principles calculations [8,9]. Microsc. Microanal. 21 (Suppl 3), 2015 434 References: [1] Wang, Q.H., et al, Nature Nanotechnology 7 (2012), p. 699. [2] Radisavljevic, B., et al, ACS Nano 5 (2011) p. 9934. [3] Zhou, W., et al, Nano Letters 13 (2013) p. 92651. [4] van der Zande, A.M., et al, Nature Materials 12 (2013), p. 554. [5] Najmaei, S., et al., ACS Nano 8 (2014), p. 7930. [6] Lin. J.H., et al., under review [7] Lin. J.H. et al., Nature Nanotechnology 9 (2014), p. 432. [8] Lin. J.H., et al., in preparation [9] This research was supported in part by U.S. DOE grant DE-FG02-09ER46554 (JL, STP), by the Office of Science, Basic Energy Science, Materials Sciences and Engineering Division, U.S. Department of Energy (WZ), and through a user project supported by ORNL's Center for Nanophase Materials Sciences (CNMS), which is sponsored by the Scientific User Facilities Division, Office of Basic Energy Sciences, U.S. DOE. This research used resources of the National Energy Research Scientific Computing Center, which is supported by the Office of Science of the US Department of Energy under Contract No.DE-AC02-05CH11231. Figure 1. Nucleation process of the inversion domain in monolayer MoSe2. (A) Atomic resolution ADF image of an intrinsic 60� grain boundary in CVD-grown monolayer MoSe2. (B) Sequential ADF images showing the nucleation process of an inversion domain in monolayer MoSe2 induced by Se vacancies. Figure 2. In-situ fabrication of nanowires from a TMDC monolayer. Sequential ADF images showing the fabrication process of an individual MoSe nanowire entirely within the semiconducting MoSe2 monolayer. A Y-junction is shown on the right.");sQ1[218]=new Array("../7337/0435.pdf","Individual Mo Dopant Atoms in WS2 Monolayers: Atomic Structure and Induced Strain","","435 doi:10.1017/S1431927615002974 Paper No. 0218 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Individual Mo Dopant Atoms in WS2 Monolayers: Atomic Structure and Induced Strain Amin Azizi1, Bin Jiang2, Zhong Lin3, Ana Laura Elias3, Mauricio Terrones3, and Nasim Alem1 1. Department of Materials Science and Engineering, Materials Research Institute, and Center for Two Dimensional and Layered Materials, The Pennsylvania State University, University Park, USA 2. FEI Company, 5350 NE Dawson Creek Drive, Hillsboro, Oregon 97124, United States 3. Department of Physics, and Center for Two Dimensional and Layered Materials, The Pennsylvania State University, University Park, USA Two-dimensional (2D) materials, including graphene, hexagonal boron nitride and transition-metal dichalcogenides (TMDs), offer a wide range of chemical, optical and electronic properties (1�4). 2D semiconducting TMDs, i.e. MoS2 and WS2, show layer dependent properties (5). For instance, they experience an indirect to direct band gap transition when thinned down to a single-layer (5), and also exhibit increased catalytic activity when their thickness is decreased to a monolayer (6). WS2 monolayers, in particular, have shown a great potential for applications in optoelectronic devices (7) and hydrogen evolution reaction (HER) (8). The unique properties of WS2 can be further manipulated through substitutional doping. Doping has been used to tune chemical and physical properties of bulk materials, particularly in the semiconductors industry. Dopants can play a more significant role in tailoring the properties of 2D crystals as they are at the surface of the crystal. Importantly, dopants in these monolayer structures can be directly visualized at the atomic-scale using advanced transmission electron microscopes. Understanding how individual dopants interact with the host lattices is the key to modify their physical and chemical properties for future catalysis and device applications. In this work, we use aberration-corrected scanning transmission electron microscopy (STEM) to image the atomic structure of molybdenum (Mo)-doped WS2 monolayers. Through strain field mapping, we also demonstrate the local strain induced by individual Mo dopants in the host WS2 lattice. Single-layers of WXMo1-XS2 were grown at 800�C using a chemical vapor deposition (CVD) method (9). The samples grown on Si/SiO2 substrates were transferred to gold (Au) quantifoil TEM grids using a poly(methyl methacrylate) (PMMA)-assisted technique (9). An FEI Titan3 60-300 S/TEM at 80 kV was used for STEM imaging. Figure 1a indicates the structural model of a single-layer of WXMo1-XS2, in which tungsten (W) and Mo atoms share the metal atom sites and are sandwiched by sulfur (S) atoms. Individual Mo dopant atoms were distinguished from the W atoms in the host lattice through Z-contrast imaging using high angle annular dark field (HAADF) STEM. Figure 1b shows the HAADF-STEM image of a single-layer of WXMo1-XS2, in which two individual Mo atoms can be easily recognized. As the intensity of HAADF-STEM images depends on the atomic number, a higher contrast can be observed for W atoms while Mo atoms are slightly dimmer. The intensity profile obtained from the region in the yellow box in Figure 1c confirms a lower intensity for the Mo atom compared to the W atoms. Substitutional dopants can locally introduce strain in the host lattice, which can modulate the chemical and physical characteristics of the crystal. For example, the induced strain in these structures can be beneficial for catalysis due to the change in the free energy of chemisorption on the surface (8). In this study, we applied geometric phase analysis (GPA) (10) to map the strain fields at the dopant sites. We used a symmetric strain matrix to obtain strain fields. The shear strain (xy) map presented in Figure 1d, and the superimposed image of the HAADF-STEM micrograph and its xy strain map (Figure 1e) clearly confirm the presence of considerable local strain at the Mo atom sites. Such strain fields associated with Microsc. Microanal. 21 (Suppl 3), 2015 436 the dopants can result in unique chemical, optical and electronic properties in 2D crystals and open up opportunities for novel future applications. References: 1. X. Li et al., Science 324, 1312�1314 (2009). 2. C. R. Dean et al., Nat. Nanotechnol. 5, 722�6 (2010). 3. S. Z. Butler et al., ACS Nano 7, 2898�2926 (2013). 4. A. Azizi et al., Nat. Commun. 5, 4867 (2014). 5. Q. H. Wang, K. Kalantar-Zadeh, A. Kis, J. N. Coleman, M. S. Strano, Nat. Nanotechnol. 7, 699�712 (2012). 6. Y. Yu et al., Nano Lett. 14, 1�6 (2014). 7. H. R. Guti�rrez et al., Nano Lett. 13, 3447�3454 (2013). 8. D. Voiry et al., Nat. Mater. 12, 850�5 (2013). 9. Z. Lin et al., APL Mater. 2, 092514 (2014). 10. M. J. H�tch, E. Snoeck, R. Kilaas, Ultramicroscopy 74, 131�146 (1998). Figure 1. (a) The structural model of a single-layer of WXMo1-XS2, (b) atomic-resolution HAADFSTEM image of a single-layer of WXMo1-XS2, indicating W, Mo and S atoms, and (c) the intensity profile obtained from the yellow box region of the HAADF-STEM image. (d) The shear strain (xy) map of the monolayered WXMo1-XS2 (e) and the superimposed image of the atomic-resolution HAADFSTEM image and its xy strain map, showing local strain at the dopant sites. The strain color scale (indicated in Figure 1d) ranges from �1 (black) to +1 (white).");sQ1[219]=new Array("../7337/0437.pdf","TEM with in situ Ion Irradiation of Nuclear Materials: The IVEM-Tandem User Facility","","437 doi:10.1017/S1431927615002986 Paper No. 0219 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 TEM with in situ Ion Irradiation of Nuclear Materials: The IVEM-Tandem User Facility Meimei Li1, Marquis A. Kirk1, Peter M. Baldo2 and Edward A. Ryan1 1. 2. Nuclear Engineering Division, Argonne National Laboratory, Lemont, IL, USA Materials Science Division, Argonne National Laboratory, Lemont, IL, USA The IVEM-Tandem User Facility at the Argonne National Laboratory (ANL) is a world-leading research facility for in situ TEM study of ion irradiation damage and ion implantation near the atomic resolution. The IVEM-Tandem Facility interfaces a 500 kV ion implanter to a 300 kV Hitachi H9000NAR transmission electron microscope. This combination allows experiments conducted with simultaneous ion irradiation/implantation and electron microscopy at temperatures ranging from 20 � 1300 K. With superior electron brightness of a LaB6 filament, the H-9000 microscope is ideally suited for imaging irradiation-induced defects using diffraction contrast, a well-established technique in most cases superior to other imaging techniques. This permits effective real time observation of nanoscale defect formation and evolution during irradiation. Coupled with well-controlled irradiation conditions (constant specimen orientation and area, specimen temperature, ion type, ion energy, dose rate, dose, and applied strain), it provides key information on structure and defect dynamics of a material in an irradiation environment. The IVEM-Tandem Facility was commissioned in 1995 in the Materials Science Division of ANL, and was part of Argonne's Electron Microscopy Center, a National User Facility supported by the Department of Energy (DOE) Office of Science, Basic Energy Science. It was recently transitioned to the Nuclear Engineering Division of ANL, currently supported by the DOE Office of Nuclear Energy. The IVEM-Tandem serves more than 20 active user groups from universities, national laboratories, and nuclear industry in the U.S. and abroad, conducting research in areas of advanced alloys, accident tolerant materials and fuels, radiation-resistant fuel cladding, storage materials for spent fuels, and validation and verification of computer modeling and simulations. A facility upgrade is being considered to best address the critical problems of fission and fusion energy materials. TEM with in situ ion irradiation has made tremendous contributions to the understanding of defect production, accumulation and evolution under irradiation. Many of fundamental questions of radiation damage, e.g. single cascades, cascade � cascade, or cascade � subcascade interactions, defect production and annihilation rates of visible defect clusters, defect and dislocation interactions, etc. have been answered. A recent new direction is to predict neutron damage in bulk through in situ ion irradiation studies of thin films and coordinated computer modeling so that neutron irradiation effects can be evaluate by easily controlled ion irradiation experiments. For example, TEM with in situ ion irradiation experiments was conducted on molybdenum with 1 MeV Kr ions at the IVEM-Tandem Facility in combination with rate theory modeling. The in situ ion irradiation experiments were explicitly designed to compare with neutron irradiation data of the identical material reported in a previous study [1]. A spatially-dependent cluster dynamic model was developed to explicitly simulate the damage by 1 MeV Kr ion irradiation in a Mo thin film with temporal and spatial dependence of defect distribution. Both in situ ion irradiation and neutron irradiation produced the same defect structure (i.e. dislocation loops) in Mo irradiated to low-doses at 80C. This allowed direct, quantitative comparisons of defect number Microsc. Microanal. 21 (Suppl 3), 2015 438 densities and size distributions. In situ ion irradiation experiments took advantage of thin foil specimens, and used the thickness of a thin foil as an important variable in describing the defect depth distribution by three-dimensional diffraction contrast electron tomography. This additional spatial dimension improves the rate theory based cluster dynamic model for defect reaction kinetics and enabled direct comparison with experiments of defect structures at different foil depths and times (doses) [2,3]. The experimentally-validated damage model of ion-irradiated thin films was then used to predict neutron damage in bulk materials and validated by the neutron irradiation data. This study demonstrates a promising new direction in understanding and predicting neutron damage in bulk with in situ ion irradiation of thin films closely coupled with computer modeling using the IVEM-Tandem Facility. References: [1] Meimei Li, M. Eldrup, T. S. Byun, N. Hashimoto, L. L. Snead, S. J. Zinkle, J. Nucl. Mater. 376 (2008) 11. [2] Meimei Li, M.A. Kirk, P.M. Baldo, Donghua Xu, and B. D. Wirth, Phil Mag. 92 (2012) 2048. [3] D. Xu, B. D. Wirth, M. Li, and M. Kirk, Acta Mater. 60 (2012) 4286. [4] The electron microscopy with in situ ion irradiation was accomplished at the Argonne National Laboratory at the IVEM-Tandem Facility, a user facility funded by the Department of Energy Office of Nuclear Energy, operated under Contract No. DE-AC02-06CH11357 by UChicago Argonne, LLC.");sQ1[220]=new Array("../7337/0439.pdf","Effect of Grain Boundary Structure on Defect Absorption and Denuded Zone Formation in Irradiated Nanocrystalline Iron","","439 doi:10.1017/S1431927615002998 Paper No. 0220 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Effect of Grain Boundary Structure on Defect Absorption and Denuded Zone Formation in Irradiated Nanocrystalline Iron O. El-Atwani1, k. Hattar2, G. Vetterick1, M.L. Taheri1 1. 2. Drexel University, Department of Materials Science & Engineering, Philadelphia, PA USA. Department of Radiation Solid Interactions, Sandia National Laboratories, NM, United States The performance of nuclear materials in extreme environments poses important fundamental questions about the behavior of condensed matter under far-from-equilibrium conditions.[1] Nuclear materials are exposed to high heat flux and irradiation that alter their microstructure, mechanical properties, and performance.[2] To mitigate possible damage, the use of UltraFine (UF) and NanoCrystalline (NC) metals has been proposed, due to their high grain boundary densities that thus act as high defect and particle sinks,[3] and improve mechanical properties (strength and ductility). [4] While recent research has focused on the effect of grain boundary character [5] on the radiation tolerance, fundamental questions about the role of non-equilibrium grain boundaries in defect absorption in comparison with equilibrium grain boundaries have yet to be answered. Understanding the effect of the non-equilibrium boundary local strain in irregular defect absorption and denuded zone formation could be crucial in engineering polycrystalline materials of higher radiation tolerance that can sustain the severe environments of future nuclear reactor conditions. In this work, in-situ irradiation was performed in a transmission electron microscope (TEM) on freestanding nanocrystalline Fe samples of less than 100 nm grain size. To form the nanocrystalline samples, Fe films was sputter-deposited on NaCl substrates [6]. TEM micrographs of the samples after annealing showed both equilibrium and non-equilibrium (high local strain manifested by strong extinction bands) boundaries adjacent to each other (Figure.1), thus enabling the comparison between both boundary types. The in-situ TEM irradiation experiments were performed using the i3TEM facility in the Department of Radiation Solid Interactions at Sandia National Laboratories. Crystallographic orientation microscopy (ACOM) was also performed via NanoMEGAS ASTAR precession diffraction (eg. Figure.2). Effect of defect absorption on grain boundary structure is studied through the in-situ irradiation/ACOM experiments. The findings provide fundamental aspects to be considered in engineering high radiation tolerant nanomaterials. Microsc. Microanal. 21 (Suppl 3), 2015 440 References: [1] J. Hemminger et al., "Directing Matter and Energy: Five Challenges for Science and the Imagination." Department of Energy's Office of Science (http://science. energy. gov/~/media/bes/pdf/reports/files/gc_rpt. pdf) (2007). [2] A. Makhankov et al., J Nucl Mater, 290 (2001) 1117 [3] T.D. Shen et al., Appl Phys Lett, 90 (2007) 263115 [4] Y. Wang et al., Nature, 419 (2002) 912 [5] C.M. Barr et al., Acta Materialia, 67 (2014) 145 [6] G. Vetterick et al., J. Appl. Phys, 116 (2014) 233503 [7] M.L.Taheri, G. Vetterick, and O. El-Atwani gratefully acknowledge funding from the United States Department of Energy, Basic Energy Sciences under the Early Career program through contract DESC0008274. K. Hattar acknowledges the Division of Materials Science and Engineering, Office of Basic Energy Sciences, U.S. Department of Energy. Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000. Figure 1. Bright field TEM image of nanocrystalline iron film annealed at 500 �C. A B Figure 2. Microstructure of an annealed iron film. (A) Crystallographic orientation map and (B) the corresponding TEM bright field image.");sQ1[221]=new Array("../7337/0441.pdf","Correlative and dynamic in situ S/TEM characterization of heavily irradiated pyrochlores and fluorites.","","441 doi:10.1017/S1431927615003001 Paper No. 0221 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Correlative and dynamic in situ S/TEM characterization of heavily irradiated pyrochlores and fluorites. T.G. Holesinger1, S. Dey4, J.A. Aguiar3, P.A. Papin2, J.A. Valdez2, Y. Wang, B.P. Uberuaga2, and R.H.R. Castro4 1. 2. Materials Physics and Applications Div., Los Alamos National Laboratory, Los Alamos, NM, 80465 Materials Science and Technology Div., Los Alamos National Laboratory, Los Alamos, NM, 80465 3. Microscopy and Imaging Group, National Renewable Energy Laboratory, Golden, CO, 80401 4. Dept. of Chemical Engineering and Materials Science, & NEAT ORU, University of Davis, Davis, CA 95616. Irradiation of materials with light and heavy ions is a ubiquitous method for altering materials to examine their structural and functional processes. For example, materials for applications in highradiation environments, such as nuclear reactors or radioactive waste, require extremely high resistance to damage accumulation, amorphization, and/or volume swelling. The level to which these materials can tolerate damage is often probed with ion irradiation.[1] It can also be used to induce anion and cation disorder within the materials. This disorder can be connected to the defect kinetics within the material and their effects on diffusion and conductivity. For example, it has been observed that the oxygen mobility increases in pyrochlores with an increase in the crystalline disorder.[2] In this talk, we will examine heavy-ion irradiated (Kr+, 400 keV, room temperature) Ga2Ti2O7 and Ga2Zr2O7 pryochlores as well as nanostructured yttria-stablized zirconia (YSZ).[3] In the case of the pyrochlores, the differences in the resulting microstructures between the Ga2Ti2O7 (amorphous) and Ga2Zr2O7 (fluorite) and the stability of these structures and associated defects were determined before and after subsequent annealing. The nanostructured YSZ (10 at.%) was characterized for the observed grain growth and crack structures in the irradiated region and the stability of this structure upon further heating. Conventional (S)TEM was used to characterize the materials before and after irradiation to track the structural stability in each material. In situ (S)TEM was used to examine the evolution of microstructures at elevated temperatures within an atmospheric gas cell, hot-stage holder. Of particular interest was the stability of the materials and defects within the Kr implanted region (Figure 1), and any transient behavior as the materials are raised in temperature. Use of the in situ atmospheric holder allows for the exploration of dynamic, atomistic-scale processes (Figure 2) at temperature and under gas flow. Examination of the specimen surrounded by an oxidizing atmosphere within the (S)TEM alleviates specimen changes that may result from oxygen loss through heating and holding at temperature in vacuum. The combination of these two approaches will help to understand the dynamic processes regarding damage accumulation and defect stability/kinetics in ion irradiated ceramics.[4] References: [1] K. E. Sickafus, L. Minervini, R. W. Grimes, J. A. Valdez, M. Ishimaru, F. Li, K. J. McClellan and T. Hartmann, Science 2000, 289, 748-751. [2] P. J. Wilde and C. R. A. Catlow, Solid State Ionics 1998, 112, 173-183. [3] S. Dey, J. W. Drazin, Y. Wang, J. A. Valdez, T. G. Holesinger, B. P. Uberuaga and R. H. R. Castro, Sci. Rep. 2015, 5. Microsc. Microanal. 21 (Suppl 3), 2015 442 [4] This work was supported by the U.S. Department of Energy, Office of Science, Basic Energy Sciences, Materials Sciences and Engineering Division and the UC Lab Fees Research Program 12-LF239032. Figure 1. STEM spectral imaging of the Kr distribution and the local microstructure in nanocrystalline YSZ. Grain growth in the irradiated area has occurred as a result of the irradiation process. The stability and transient behavior of the microstructure and Kr distribution will be determined from in situ annealing experiments under gas flow in the S/TEM. Figure 2. High resolution imaging of in situ alloying of nanoparticles (a) under gas flow at 500C in the in situ gas holder demonstrates the ability to image dynamic processes in the S/TEM at the atomic level. The nm-sized defects induced during Kr irradiation (b) will be monitored as a function of time and temperature to track defect movement, interactions, and coalescence.");sQ1[222]=new Array("../7337/0443.pdf","In Situ Probing of the Evolution of Irradiation-induced Defects in Copper","","443 doi:10.1017/S1431927615003013 Paper No. 0222 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In Situ Probing of the Evolution of Irradiation-induced Defects in Copper N. Li1, K. Hattar2, A. Misra3 1. 2. Los Alamos National Laboratory, Los Alamos, NM 87545, USA Sandia National Laboratories, Albuquerque, NM 87185, USA 3. University of Michigan, Ann Arbor, MI 48109, USA Any attempt to produce nuclear materials with improved mechanical properties depends heavily on understanding the root causes of radiation damage, which are individual and clustered vacancies and interstitials (self or external) produced during collision cascades between energetic particles and target atoms [1,2]. The subsequent diffusion and clustering of these defects, along with the associated transport of impurities, will dramatically change the microstructure and lead to accelerated degradation of properties during irradiation. In order to effectively suppress irradiation-induced damage, significant research is being conducted to synthesize materials containing stable sinks for trapping and recombining irradiation-induced point defects [3,4]. However, experimental evidence through in situ studies is seldom provided [5] due to both the tediousness of identifying boundary types and the difficulty of directly performing experiments observing the interaction of boundaries with those defects. In order to unravel the evolution of irradiation damage with the dose, we performed in situ ion irradiation experiments in a transmission electron microscope (TEM). The present study will focus on a common grain boundary in Cu, the 3 {112}||{112} incoherent twin boundary (ITB). Through in situ Cu3+ ion irradiation at room temperature in a TEM, we have investigated the evolution of defect clusters as a function of the radiation dose at different distances from the 3 {112} ITB in Cu. During irradiation, defect clusters evolve through 4 stages: (i) incubation, (ii) non-interaction, (iii) interaction and (iv) saturation; and the corresponding density was observed to initially increase with irradiation dose and then approach saturation (in Fig. 1). No denuded zone is observed along the 3 {112} ITB and the configuration of defects at the boundary displays as truncated SFTs (in Fig. 2). Several defect evolution models have been proposed to explain the observed phenomena [6]. In summary, current study emphasis will be placed on the defect evolution as a function of the radiation dose at different distance to the boundary. These results will provide a useful basis for analyzing the influence of interface sink strength on the reduction of radiation-induced defects [7]. [1] M.J. Demkowicz, P. Bellon, and B.D. Wirth, MRS Bull. 35 (2010), p. 992. [2] B.D. Wirth, Science 318 (2007), p. 923. [3] T. Allen et al, MRS Bull. 34 (2009), p. 20. [4] Y. Guerin, G.S. Was, and S.J. Zinkle, MRS Bull. 34 (2009), p. 10. [5] A.H. King and D.A. Smith, Philos. Mag. A 42 (1980), p. 495. [6] N. Li, K. Hattar, and A. Misra, Journal of Nuclear Materials, 439 (2013), p. 185. [7] This work is supported by the Center for Materials at Irradiation and Mechanical Extremes (CMIME), an Energy Frontier Research Center (EFRC) under Award No. 2008LANL1026 by the DOE, Office of Science, Office of Basic Energy Sciences. Microsc. Microanal. 21 (Suppl 3), 2015 444 Figure 1. (a) Cross-sectional TEM Cu foils tilted to two-beam conditions with Cu 1 1 spots strongly excited. (b�f) A sequence of TEM video images recording the microstructure evolution of pure Cu under 3 MeV Cu3+ irradiation at room temperature. Figure 2. (a�c) The distribution of radiation-induced defects as a function of the distance from the boundary at three different doses.");sQ1[223]=new Array("../7337/0445.pdf","The Role of Computed X-ray Tomography in a Metallurgical Analysis","","445 doi:10.1017/S1431927615003025 Paper No. 0223 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Role of Computed X-ray Tomography in a Metallurgical Analysis Noah Budiansky, Joel Forman, and Ockert Van Der Schijff Exponent, Natick, MA., USA Metallographic examination has been an integral part of a failure analysis for nearly the past century. As metallurgists we are always interested in the structure-property relationship for a material that is the subject of a failure investigation. In many instances, indicators of why a component failed can be hidden and very subtle such as a manufacturing defect, a crack, the microstructure, or some other metallographic feature. When a failure analysis is being conducted we often rely on cross-sectional analysis of an identified feature or in suspect areas when features are not visible. The use of computed X-ray tomography (CT) in conjunction with metallographic analysis can greatly benefit a failure investigation resulting in a more comprehensive picture than what metallographic analysis can provide alone. Computed X-ray tomography is a radiographic technique where a series of X-ray images are reconstructed into a three-dimensional model of a component that has information regarding its internal features. Differences in material density show up as grey scale contrast in an X-ray image and can be utilized to visualize the internal structure or features. For example, materials with higher density absorb more X-rays than lighter materials resulting in less transmitted X-rays being detected by a detector. During a failure investigation, the use of CT as one of the initial steps, allows a non-destructive overview of the entire structure of the component and can be used to identify internal features of interest that can later be investigated by metallurgical analysis. For example, the depth of graphitic corrosion penetration of a grey cast iron water pipe may only be superficial in certain locations, but may reach full penetration in other locations. CT can be employed to identify locations of near through wall penetration, which can subsequently be sectioned and prepared for metallographic analysis showing the internal microstructure and confirming that the mode of failure is graphitic corrosion (Figure 1). A second example is a carbon steel heat exchanger tube/tube sheet weld area where leaks occurred (Figure 2). On the exterior surface, a large feature was observed where steam likely escaped resulting in erosion and enlargement of the larger initial penetration. CT analysis as a first step in the failure analysis indicated that a crack-like feature could be observed at the root of the weld only in a few isolated locations adjacent to the large eroded area. Cross sectioning through this location showed that the crack was branched and further microscopic and elemental analysis indicated that the crack was filled with corrosion product with evidence of sodium. The investigation concluded that the likely cause of failure was caustic cracking initiating at the root of the weld induced by built up caustic ions from the feed water (Figure 2). Without the use of CT prior to sectioning and metallographic analysis the isolated locations of cracking may never have been identified and the failure analysis may not have been able to identify the failure mode. Microsc. Microanal. 21 (Suppl 3), 2015 446 Similar failure analyses could indicate the directionality of cracking indicating whether cracking initiated from the internal surface or the outside surface of a pipe allowing the identification of the environmental cracking agent (e.g., a heating solution versus product). This work illustrates the complimentary benefits of CT and metallographic analysis in a failure analysis through a series of failure analysis examples. Figure 1. Overall photograph, CT cross section, and optical microscope image of graphitic corrosion in grey cast iron water pipe. Dark areas indicate the location of graphitic corrosion. Figure 2. CT cross section through a three dimension projection and optical microscope image of a roughly polished cross section of a welded carbon steel heat exchanger tube/tube sheet. Arrows indicate the location of observed cracking.");sQ1[224]=new Array("../7337/0447.pdf","Characterizing Failure in Commercial Li-Ion Batteries with 4D X-Ray Microscopy","","447 doi:10.1017/S1431927615003037 Paper No. 0224 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterizing Failure in Commercial Li-Ion Batteries with 4D X-Ray Microscopy Jeff Gelb1,2, Paul Shearing3, Donal Finegan3, and Dan Brett3 1. 2. San Jose State University, Department of Chemical and Materials Engineering, San Jose, CA, USA Carl Zeiss X-Ray Microscopy, Inc., Pleasanton, CA, USA 3. University College London, Department of Chemical Engineering, London, UK Modern devices are increasingly reliant on portable energy storage solutions. With the growing popularity of hybrid and fully-electric vehicles, demands for high-capacity and long-life batteries are also increasing. Understanding how the batteries perform and how long they last is critical to making a successful electric vehicle, to ensure the safety and satisfaction of the consumer. In spite of this, failure modes of Li-ion batteries remain poorly understood. It is well-established that batteries decline in charge capacity with an increased number of charge cycles, but the precise nature of the degradation as well as the cause of their ultimate failure remains largely uncharacterized. Here, we present the results of using a direct, non-destructive imaging method to reveal the structural component of failure modes of commercial Li-ion batteries [1]. Two commercially-sourced Panasonic NCR 18650B Li-ion battery cells were imaged using a Zeiss Xradia 520 Versa X-ray microscopy (XRM) system. The cells were initially surveyed with a ~20 mm field of view, in order to show the 3D structure of the jelly roll and for bulk feature identification, as shown in Figure 1. One cell was then deconstructed in an inert environment and a small piece of each electrode was imaged with ~200 nm resolution on a Zeiss Xradia 810 Ultra XRM. The higher resolution results, shown in Figure 2, revealed several features within the active layer, including cracks and voids within the particles. These cracks and voids represent a significant deviation from the idealized, structurally homogeneous models that are often incorporated into battery cell design, representing a unique finding about the real microstructure of the active particles. A second cell from the same supplier was then imaged with the same parameters as before, but several such volumes were virtually stitched together to show the internal geometry of the entire cell. From this result, a smaller region was imaged with ~1.8�m resolution without opening the package, utilizing the non-destructive nature of X-ray imaging. The cell was then charge cycled 100 times at 0.5C and the capacity fade was recorded. This same cell was then re-imaged with the same parameters as before, both with the large field of view and high-resolution imaging modes. The resulting 3D slices showed the development of many cracks within the cathode layer, which are suspected to contribute to the ultimate failure of the battery cell. Building on the results from previous studies, we aim to explore the changing microstructure of these complex materials and how this affects cell performance and lifetime [2]. This experiment represents the first observation in this effort, and future studies will further explore the relationship between microstructure, performance, and failure in greater detail. References: [1] A P Merkle and J Gelb, Microscopy Today (2013) pp. 10-15. [2] D S Eastwood et. al., Advanced Energy Materials 4 (2013) p. 1-7. Microsc. Microanal. 21 (Suppl 3), 2015 448 Figure 1. Multi-planar view of a commercial 18650 Li-ion battery, showing the different layers of jelly roll. Figure 2. Virtual slice through a 3D volume of the cathode layer from a deconstructed commercial 18650 Li-ion battery as manufactured, imaged with 150 nm resolution. Several defects in the active particles are observed, including internal voids and cracks that extend into the pore network.");sQ1[225]=new Array("../7337/0449.pdf","X-Ray Microcomputed Tomography for Analysis of Retrieved Medical Devices.","","449 doi:10.1017/S1431927615003049 Paper No. 0225 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 X-Ray Microcomputed Tomography for Analysis of Retrieved Medical Devices. Judd S Day1,2, Daniel W Macdonald2 and Steven M Kurtz1,2 1. 2. Exponent, Inc., Philadelphia, PA, USA. Implant Research Center, Drexel University, Philadelphia, PA, USA. X-ray microcomputed tomography (microCT) is a non-destructive method that uses x-rays to reconstruct a virtual three-dimensional stack of cross-sections. The resulting image data can be used to assess material geometry, density distribution, and the presence of cracks or voids, among other properties. In this presentation, we will present case studies illustrating the use of microCT for the non-destructive analysis of retrieved medical devices. The first case study involves investigation of suture material from a patient who had died following a carotid endarterectomy procedure. Carotid endarterectomy is a procedure that is performed to remove plaque from the carotid artery (i.e. the artery that provides blood to the brain). After removing plaque from the artery, the vessel is repaired by closing with a suture. Suture material was excised from the patient's carotid artery post mortem and analyzed using a combination of light microscopy, scanning electron microscopy and microCT to investigate the cause of failure. A "pigtail" suture end was observed by microscopy in the vicinity of the failed repair, consistent with slipping of a suboptimal surgical knot (Figure 1). Using scanning electron microscopy, it was determined that there was no evidence of failure of the suture ends due to voids or imperfections of the suture material. Further investigation of the retrieved suture material using microCT confirmed that there were no unaccounted suture ends, and that there were multiple knot throws constructed using suboptimal technique (Figure 2). The second case study involves the use of microCT to estimate material loss (wear) and damage from retrieved polyethylene orthopaedic bearing surfaces including total hips, knees, elbows, shoulders and spinal discs. Polyethylene is frequently used as an orthopaedic bearing, but is known to wear resulting in particles that can evoke an adverse response in the surrounding tissues. Evaluation of retrieved orthopaedic devices can include quantification of polyethylene creep and damage and examination of subsurface cracks (Figure 3). Using wear data from total elbow and total disc replacements, we have been able to verify anatomical loading conditions which were then used to develop in vitro validation tests of these devices. Using microCT we have also been able to estimate the material loss from highly cross linked polyethylene knee components. Highly crosslinked polyethylene is resistant to wear in the anatomical environment and its clinical use is increasing. However, there is little clinical data quantifying the volumetric wear in vivo. We found that retrieved knee bearing surfaces may contain embedded cement debris (Figure 4). Volumetric wear could be quantified using microCT and wear maps created to characterize the locations of component wear (Figure 5). This data will be useful for further understanding of the biomechanics of clinical failure. We also used microCT to demonstrate a substantial amount of bearing surface penetration and damage was due to material creep and could be recovered by remelting the material. We have demonstrated that microCT provides a versatile nondestructive tool for examining retrieved medical devices. Microsc. Microanal. 21 (Suppl 3), 2015 450 Figure 1. Pigtail suture end, consistent with slippage of the surgical knot. C To Knot 1 To Knot 2 D B A B CCW CW A Figure 2. Identification of individual knot strands from microcomputed tomography data. Once the individual knot strands were identified, it was clear that there were a number of slip knots present in the knot structure. It was also clear that knot two comprised a convoluted structure and that all suture ends were accounted for. D C Figure 3. Example of methodology for estimating the volumetric wear of a polyethylene total disc component. Figure 4. Total knee component with embedded third Figure 5. Penetration map for total knee body cement debris. component..");sQ1[226]=new Array("../7337/0451.pdf","MicroStamping for Improved Speckle Patterns to Enable Digital Image Correlation","","451 doi:10.1017/S1431927615003050 Paper No. 0226 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 MicroStamping for Improved Speckle Patterns to Enable Digital Image Correlation Andrew H. Cannon1, Jacob D. Hochhalter2, Alberto W. Mello3, Geoffrey F. Bomarito2, and Michael D. Sangid3 1. 2. 1900 Engineering, LLC, Clemson, SC, USA. NASA Langley, Durability, Damage Tolerance, and Reliability Branch, Hampton, VA, USA. 3. Purdue University, School of Aeronautics and Astronautics, West Lafayette, IN, USA. Surface strain measurements using image correlation require a pattern to be applied to the surface of the object being measured. Lithography, the most widely used method for repeatable patterning is expensive, requiring dedicated technical staff and significant infrastructure. Lithography is time consuming, often requiring several days for each patterning application, which limits throughput. An innovative method has been developed and tested whereby repeatable patterns for image correlation are applied without dedicated technical staff or special infrastructure and can be completed in a few minutes rather than days. This new method is more amenable to application of patterns to complex surface geometries and larger surface areas. The new micro stamping method allows for higher contrast patterning materials, which improves the accuracy of strain measurements using image correlation. Accurate surface strain measurements using image correlation are dependent on the application of a high-contrast pattern to the surface of the object being measured. Error in the strain measurement is dependent on the particular pattern applied, and repeatability of the pattern on various surfaces is ideal. Micro texture stamping is a repeatable, high throughput, high-resolution, low cost, parallel patterning method in which a stamp surface pattern is replicated into a material by mechanical contact. Details of the flexible micro textured stamps produced by 1900 Engineering have been published [1-2]. The stamps were fabricated with a 10 �m base-element size. The electron-beam lithography (EBL) process for generating the stamp master took 57 hours to complete a 12.7 mm x 12.7 mm area using an e-beam resist [3]. Without the stamping procedure, EBL would need to be repeated for each subsequent specimen to be patterned; however, after fabricating the stamp, the pattern application took approximately 10 minutes per subsequent specimen. A diagram for the stamp usage is in Figure 1. The procedure for application of the stamp to create a speckle pattern is: (1) Sonicate the specimen in acetone, then methanol, and dry; (2) With a fine liner, apply MCC primer on clean specimen; (3) Let stand for 15 s and gently apply compressed air from the top; (4) Bake for 3 min at 115 �C on a hot plate, then remove the specimen. (5) Let the hot plate cool to 60 �C, (6) Place two specimens side by side (to allow level stamping � the procedure can be applied to one specimen); (7) Apply Shipley 1805 photo resist on one specimen; (8) Wait 20 s; (9) Apply the Shipley in the same specimen; (10) Align the stamp by touching first the dummy specimen and then let the stamp lay down over the specimen to be stamped. (11) Adjust the hot plate for 115 �C; (12) Place a piece of cook paper over the stamp, to allow a non-stick surface for the weight; (13) Apply weight (~4 psi); (14) Bake for 8 minutes; (15) Remove the weight and the specimen from the hot plate; (16) Carefully peel the stamp off the specimen; (17) Check the gauge section on the microscope for the patterns. The resulting speckle pattern for the 10 �m pattern is shown in Figure 2a. The accuracy of measured strain results improved as compared to the patterns applied using EBL, because an optically-opaque material was used with the stamping procedure to create a higher-contrast Microsc. Microanal. 21 (Suppl 3), 2015 452 pattern. Results of the surface strain measurement using the stamped pattern for image correlation are shown in Figure 2b for deformation with the microstructure elucidated by electron backscatter diffraction [4]. After usage, the stamps are easily maintained by solvent cleaning to remove any pattern material residue from its surface. It is expected that these stamps can be reused approximately 100 times before losing some of the fine detail in the pattern. The stamps can be created with a wide range of sizes, resolution of features, and patterns. This technique has been used to create patterns with 2 �m base-element size and a fractal-motivated pattern that contains 10 �m and 1 �m features, which allows for strain measurements at multi-scales with the same speckle pattern. References: [1] AH Cannon and WP King, Journal of Micromechanics and Microengineering 19 (2009), p. 1-6. [2] AH Cannon, MC Maguire, and JD Hochhalter, US Patent Application 62116742 (2015). [3] VK Gupta, SA Willard, JD Hochhalter, and SW Smith, ASTM Materials Performance and Characterization 4 (2014), p 1-27. [4] W Abuziad, MD Sangid, J Carroll, H Sehitoglu and J Lambros, Journal of the Mechanics and Physics of Solids 60 (2012), p. 1201-1220. [5] JDH and GFB would like to thank funding from the NARI Seedling Project. MDS and AWM would like to thank support from the Office of Naval Research, N00014-14-1-0544 and DARPA, N66001-141-4041. The authors acknowledge WW, TM, and MM for stimulating conversations about this topic. Figure 1. Procedure for applying the microstamp to create the reference speckle pattern. Figure 2. (a) 10 �m speckle pattern produced by the microstamping procedure. (b) 7050-T7451 Al (in the L-T orientation), displaying axial strain after an average strain of 2.45%.");sQ1[227]=new Array("../7337/0453.pdf","Microstructural Characterization and Image Analysis in Ex-Service Ethylene Pyrolysis Tubes","","453 doi:10.1017/S1431927615003062 Paper No. 0227 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microstructural Characterization and Image Analysis in Ex-Service Ethylene Pyrolysis Tubes A.C. McLeod1, C.M. Bishop1, K.J Stevens2 and M.V. Kral1 1. 2. University of Canterbury, Christchurch, New Zealand. Quest Integrity NZL Ltd., Wellington, New Zealand. Ethylene is used extensively in the production of plastics, cabling, and automotive products. It is typically produced by the thermal cracking (pyrolysis) of more complex hydrocarbons, such as naphtha or ethane, at 950 to 1100�C inside HP alloy tubes (25%Cr-35%Ni) within an ethylene pyrolysis furnace. The main contributor to the failure of ethylene pyrolysis tubes is carburization of the tubes, which causes an increase in internal volume, a loss in weldability and a reduced ability to withstand thermal cycles [1-4]. Knowing the level of carburization of a tube in-situ can assist in remaining life estimates based on Finite Element Modeling, thermography, and fracture mechanics. Due to the changes in the microstructure and magnetic properties of the tubes over their service life, the level of carburization can be detected non-destructively using eddy current probes [3] and a tube crawler system [5]. However, the eddy current system requires calibration on ex-service tubes that have had their microstructure, mechanical properties and magnetic response characterized. The growth of chromium carbides and other chromium-rich phases due to carbon ingress results in a depletion of chromium in the matrix. In the as-cast condition the matrix is paramagnetic, but as carburization progresses it tends towards the binary Fe-Ni system and becomes progressively more ferromagnetic [3, 4, 6-8]. The change in bulk permeability of the tube is measured by the eddy current system. As such, the volume fractions of the phases that contribute to the matrix chromium depletion and the distribution of these phases across the tube wall are of interest. The volume fractions and distributions are being characterized by way of imaging the tube microstructure along radial wall profiles, and using a combination of two programs, ilastik and FIJI, for the segmentation and processing of the images. ilastik is a free software package that enables automated classification and segmentation of batches of images [9]. Initially, BSE (backscatter electron) images were proposed for image analysis, due to their high resolution, wide range of grayscale values for differentiating phases, and relatively short collection time. However, one major feature of BSE images of ex-service tubes proved to be detrimental to the success of automated segmentation using ilastik. A "watermark" effect is often visible in the austenitic matrix of highly carburized tubes when imaging in BSE mode, causing significant gradients in grayscale. Another phase often present in the microstructure of ex-service tubes, the -carbide, also exhibits gradients in grayscale across the phase regions. As a result, automated segmentation of the BSE images of highly carburized tubes in ilastik proved to be impossible, as ilastik is unable to distinguish the -carbide from regions of the matrix that exhibit this "watermark" effect. Despite testing numerous variations in the mechanical polishing parameters, the "watermark" effect was unable to be minimized. It is possible that it is a feature of the microstructure, for example a strain effect due to the internal stresses created during carburization, as opposed to a polishing artifact that can be removed. In the present work, EDS (energy dispersive spectroscopy) maps with SEI image overlays are being utilized as the input for image analysis as opposed to BSE images. Using EDS maps ensures that Microsc. Microanal. 21 (Suppl 3), 2015 454 polishing artifacts and/or strain variations in the matrix do not cause issues with segmentation in ilastik, while the SEI image overlay allows clear demarcation of the edges of the phase regions and the necessary contrast between phases to be achieved. In particular, as the -carbide (chromium-nickelniobium silicide) has a different composition to the austenitic matrix (iron-chromium-nickel), the two phases have two clearly distinguishable colors in EDS maps, allowing ilastik to segment these phase regions with high accuracy, as shown in Figure 1. Another significant benefit of EDS maps is the ability to analyze the chromium content of the matrix. As the matrix chromium content is the link between the microstructure of the tube and its magnetic response, the data collected via EDS mapping will be instrumental in comparing the microstructure with the eddy current NDT response of the tubes [10]. Figure 1: (a) EDS map of mid-wall region of a highly carburized ex-service HP alloy ethylene pyrolysis tubes, (b) ilastik segmentation of map shown in (a). [1] D. Jakobi, R. Gommans, Materials and Corrosion, 54 (2003) 881-886. [2] E. Lang, J. Norton, Commission of the European Communities, Physical Science, PETTEN, EUR 10566 EN, 1986. [3] K.J. Stevens, W.J. Trompetter, Journal of Physics D (Applied Physics), 37 (2004) 501-509. [4] K.J. Stevens, et al., Journal of Physics D (Applied Physics), 34 (2001) 814-822. [5] Quest Integrity Group. Ethylene Pyrolysis Tube Inspection System. Available from: http://www.questintegrity.com/technology/Ethylene-Pyrolysis-Tube-Inspection-System. [6] K.J. Stevens, et al., Current Applied Physics, 4 (2004) 304-307. [7] K.J. Stevens, et al., Journal of Physics D (Applied Physics), 36 (2003) 164-168. [8] I.C. da Silva, et al., NDT & E International, 39 (2006) 569-577. [9] C. Sommer, et al., 2011 8th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI'11, March 30, 2011 - April 2, 2011, IEEE Computer Society, Chicago, IL, United states, 2011, pp. 230-233. [10] This work has been supported by Quest Integrity NZL Limited, and the Ministry of Science and Innovation of New Zealand under contract QINZ1001. Thanks to Andy Saunders-Tack and Charles Thomas of Quest Integrity NZL Ltd for their continued support.");sQ1[228]=new Array("../7337/0455.pdf","MIPARTM: 2D and 3D Microstructural Characterization Software Designed for Materials Scientists, by Materials Scientists","","455 doi:10.1017/S1431927615003074 Paper No. 0228 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 MIPARTM: 2D and 3D Microstructural Characterization Software Designed for Materials Scientists, by Materials Scientists J.M. Sosa1, D.E. Huber1, B.A. Welk1 H.L. Fraser1 1. Center for the Accelerated Maturation of Materials, Department of Materials Science and Engineering, The Ohio State University, 1305 Kinnear Rd., Columbus, OH 43212 Stereology, the science of estimating three-dimensional quantities from two-dimensionally acquired measurements, has historically been the sole technique for microstructural quantification [1]. Over the last decade and a half, 3D characterization has begun to replace stereology with direct-3D quantification. As data acquisition techniques continue to advance, the need for more materials science-orientated analytical 2D and 3D software has become evident. This led to the development of a comprehensive software suite known as MIPARTM (Materials Image Processing and Automated Reconstruction) [2]. MIPAR was written and developed within MATLABTM, but is deployable as a standalone cross-platform application. MATLAB's powerful 2D and 3D processing libraries have greatly contributed to and accelerated MIPAR's development. MIPAR is more an application environment than a single program. With a total of five applications, it was designed to handle all post-acquisition stages of 3D characterization: alignment, pre-processing, segmentation, visualization, and quantification, as well as provide a powerful platform for materials science-oriented 2D image analysis. Images of MIPAR's three most commonly used applications are shown in Figure 1. While direct-3D quantification offers several advantages over stereology such as the absence of sectioning variation and superior quantification of complex shapes, it is not without limitations. With good reason, the representative volume or area element (RVE or RAE) has been an increasingly popular topic of study and discussion [3]. However, a definition of the target quantification precision is often left out of RVE/RAE-related discussions. Therefore, this paper will present the use of MIPAR, together with statistical tools such as random sampling and bootstrapping, to establish quantitative relationships between sampled volume/area size and measurement precision for a variety of microstructural metrics. Two such relationships are shown in Figure 2. In addition to exploring the influence of sectioning variation on various stereological metrics, direct-3D quantification can either validate or invalidate stereological assumptions. A common stereological metric is the mean linear intercept. Measured from a series of random lines placed within a segmented microstructure, the mean linear intercept has been employed to estimate three-dimensional quantities such as the mean diameter of spheroidal precipitates and mean width of plate-like features [4]. In both cases, the constitutive equations rely of several assumptions regarding the shape and size distribution of the intercepted features. For features in + titanium microstructures, MIPAR has been used to explore the validity of these assumptions and determine the sensitivity of stereological quantification to deviations from such assumptions. The efficacy of these characterization efforts was critically dependent on segmentation. Therefore, a strong focus has been placed on developing a method of objectively quantifying segmentation quality. This method, reliant on the similarity metric of mutual information, has been integrated into MIPAR's Image Processor and examples of its application will be presented. Microsc. Microanal. 21 (Suppl 3), 2015 456 References: [1] [2] [3] [4] J.C. Russ in "Practical Stereology", (Kluwer Academic Pub, Denmark). J.M. Sosa, D.E. Huber, B. Welk, H.L. Fraser, Integr. Mater. Manuf. Innov. 3 (2014), p. 18 Z. Shan, A.M. Gokhale, Computational Materials Science 24 (2002), p. 361�379. P.C. Collins, B. Welk, T. Searles, J. Tiley, J.C. Russ, H.L. Fraser, Mater. Sci. Eng. A. 508 (2009), p. 174�182. Figure 1. Images of three applications used for 3D characterization in MIPARTM where (a) reveals the Image Processor, (b) the Batch Processor, and (c) the 3D Toolbox Figure 2. A plot which reveals quantified relationships between mean intercept uncertainty and sampled volume (blue), and between mean intercept uncertainty and sampled area (red). Error bounds for each relationship are shown as dotted lines.");sQ1[229]=new Array("../7337/0457.pdf","Industrial Application of Thixomet Image Analyzer for Quantitative Description of Steel and Alloys Microstructure","","457 doi:10.1017/S1431927615003086 Paper No. 0229 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Industrial Application of Thixomet Image Analyzer for Quantitative Description of Steel and Alloys Microstructure Alexander A. Kazakov1, Daniil Kiselev1 1. St.Petersburg State Polytechnical University/Metallurgical Technologies Department, St.Petersburg, Russian Federation There are numerous examples of practical use in industry of the Thixomet image analyzer for quantitative description of the microstructure of steels and alloys for various purposes. In 1996, the very first Thixomet image analyzer unit was installed at the Alumax technical center (Golden, CO) for the quantification of the structure of semi-solid materials [1]. Along with the evaluation of the distribution of porosity, eutectic content (continuous and occluded) and silicon content across the billet, a special grain shape factor was calculated in studies of anodized specimens examined in polarized light. The identification and integration of the dendrite fragments, revealed in a planar image, into one structurally isolated object was possible as a result of this measurement. Thixomet software performs a total quantitative description of 2D SSM structure, as well as direct visualization and measurement of 3D structure by processing data obtained using serial sections. Since 1997 the Thixomet has been used by many Russian turbine blade plants for quality estimation of the nickel-based superalloy charge billets [2]. The total impurity rate is calculated with regard to the contribution of each impurity type: oxide films, slag globules, nitride clusters. Their recognition is based on the grayscale level of the impurity, their morphology and the character of mutual arrangement. The rate for nitride clusters is 10 times higher than for slag globules and oxide films in the same dimensional group since nitrides and their clusters are more difficult to remove from melts than slag globules or oxide films, which are easily assimilated by the refractory lining of the crucible during remelting or by ceramic filter during pouring. The decision to use the charge billet in the production of turbine blades is based upon the metallurgical quality evaluation results. Billets with an impurity rate above a defined critical value are withdrawn from production. A motorized hardware-software complex, the "Thixomet SmartDrive", was developed and installed in dozens of enterprises and companies to provide an objective quantitative estimation of all types of structural inhomogeneity in modern pipeline steels, such as microstructural banding, general anisotropy, blocks of bainite with lath morphology and centerline segregation [3-5]. Microstructural banding is determined according to GOST 5640-68, a chart method based upon the principle of increasing the number of bands of the second phase, taking into account their continuity and degree of ferrite grain elongation. The assigned grade is based on stereological parameters that have been created using a directed secant method with the assistance of the automatic image analyzer. It should be noted that comparisons between standard chart images and those of the specimen were conducted at �100, according to GOST 5640-68, while analysis of elongated ferrite grains required �500. Such an analysis is possible only by sequential examination of the specimen at two magnifications. In classic metallography, it is impossible to observe the same wide field of view provided by �100 at �500. Therefore, classic metallographic investigations are always a compromise between the examination area and the resolution. But, if we apply modern methods of quantitative metallography using the Thixomet� image analyzer, we can simultaneously evaluate both banding and ferrite grains elongation making measurements only at �500, while the required area corresponding to the size of the field viewed at �100 can be subsequently "stitched" together electronically from adjacent fields of view. When a motorized microscope stage moves to the next Microsc. Microanal. 21 (Suppl 3), 2015 458 field of view, the previous field is stitched precisely, "pixel to pixel" to the field that was captured just before. This way a high resolution "panoramic image" of any desired large area can be created. The standard deviation of the fraction of a second phase on secants that are parallel to the rolling direction devided by the standard deviation of the fraction of second phase on secants that are perpendicular to the rolling direction will unambiguously characterize structural banding without depending on the morphology of particles of the second phase. Modern pipeline steels are manufactured using thermomechanical treatment technology which includes accelerated cooling after hot rolling producing plate with an almost 100% bainitic structure free from microstructural banding. However, anisotropy of the bainitic structure can be observed in such steels. The stereological methods that have been described above for the evaluation of ferritepearlite and ferrite-bainite microstructure are not applicable for the description of complicated bainitic morphologies. Hence, a method for evaluation of bainite anisotropy based on texture analysis of images has been developed for this case. Using the methods of texture analysis, we have no need to select any constituent of microstructure; therefore, we can evaluate microstructure with a complicated morphology and increase evaluation objectivity. The ratio determined via this method characterizes unambiguously the general anisotropy of the microstructure of bainitic steels. The results of the investigations revealed the blocks of bainite with lath morphology among different morphological forms elongated along the rolling direction. These bainitic blocks with a lath morphology are the main contributor to general anisotropy and decrease the essential mechanical properties of the plate in the transverse direction. The technique of color etching with consequent analysis in polarized light has been developed and covered by a Russian Federation patent; it promotes perfect selection of the bainite blocks with the lath morphology and facilitates measurement of their volumetric content and the length of longitudinal inter-phase boundaries which mainly determine the mechanical properties of high strength pipeline steels. The parameters of centerline segregation are decisive in the formation of the mechanical and corrosion-resistant metal properties. The GB/T 13298 technique assigns classes for hot-rolled plate structure based upon research results of its central area at a magnification of �200. Additionally, the structure is evaluated at �500 magnification to assess the non-metallic inclusions that are decorating the band. Such inclusions or a wide single band can be the basis for an assignment of an additional 0.5 class penalty. Class 1 defines a structure with slightly visible discontinuous bands in the field of view; class 2 is assigned to the cases when the number of such bands is not more than 3; class 3 is for the structure with more than 3 bands; class 4 means 3 bands located close to each other uniformly. The image of centerline segregation is a periodic signal and for evaluation of its parameters we propose a spectral analysis using the Fourier transform. In order to see the image of microstructure of the center area at magnification �200, the average values of the grayscale level are calculated on the secants parallel to the rolling direction (M(y)). Deviations of these values reflect the presence of dark bands in the image. The spectrum of the (M(y)) function is calculated using the discrete Fourier transform, established via the expert evaluation method so that the amplitude of harmonics in the frequency range between 0 and 0.05 m-1 describs the centerline segregation degree in accordance with standard charts in optimal way. References: [1] A.A. Kazakov and N.H. Luong, Materials Characterization 46 (2001). No. 2-3, p. 155 [2] A.A. Kazakov and D.V. Kiselev, CIS Iron and Steel Review, No. 1�2 (2007), p. 40. [3] A.A. Kazakov et al., CIS Iron and Steel Review, 2012, p. 4. [4] A.A. Kazakov et al., Microscopy & Microanalysis, 17 (2011), p. 1024. [5] A.A. Kazakov et al., Microscopy & Microanalysis, 17 (2011), p. 1744.");sQ1[230]=new Array("../7337/0459.pdf","Quantitative image analysis of white etching areas in SAE52100 bearing steel","","459 doi:10.1017/S1431927615003098 Paper No. 0230 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantitative image analysis of white etching areas in SAE52100 bearing steel Philippe T. Pinard1, Mehmet �zel2, Moritz Plo�3 and Silvia Richter1 1 Central Facility for Electron Microscopy, RWTH Aachen University, Aachen, Germany Institute for Machine Elements and Machine Design, RWTH Aachen University, Aachen, Germany 3 Institute for Materials Applications in Mechanical Eng, RWTH Aachen University, Aachen, Germany 2 SAE 52100 bearing steel is a common steel grade for several types of bearings used in various applications such as automotive and wind turbines. It consists of a bainitic or martensitic matrix with 0.1 to 5 �m dia. spherical carbides. This material suffers from an unexpected failure mechanism where the bearing abruptly fails as early as 10% of its expected fatigue life [1]. The failure is associated with the formation of white etching areas (WEA) and white etching cracks (WEC) (named after their appearance after Nital etching) within the first millimeter below the contact surface. Different theories exist to explain these microstructural changes: hydrogen induced, carbide dissolution, initiation at inclusions, etc. To better understand the role of the operating conditions (loading, frequency, temperature, type of lubricants, etc.) on the WEA formation, a project was started with the aims of (1) reproducing bearing failures seen in the field within a laboratory-controlled environment, (2) correlating the observed microstructure to the operating conditions and ultimately (3) explaining the failure mechanism. For the characterization part, it was clear from the beginning that many samples and large areas would need to be observed. The established strategy was to section the failed bearing where pitting or scuffing appears on the contact surface, locate cracks and WEAs using a light optical microscope after polishing and Nital etching, slightly re-polish the surface to remove the etching, and insert the sample in a scanning electron microscope for high resolution images and chemical information. As WEAs can be quite small, the acquisition of electron images was required to ensure an unbiased identification. To facilitate the acquisition, a program was developed to first acquire low magnification images over a large area, detect cracks and then automatically acquire high magnification images along the cracks. A mosaic image resulting from this acquisition is shown in Figure 1. The next step was to identify the WEAs from the surrounding matrix and cracks using an image analysis routine. The goal is to statistically compare the WEAs found in different cross-sections based on potential criteria such as their size, shape, quantity, distance from the surface, etc. The developed image analysis routine utilized the fact that WEAs contain very small grains [2] and therefore appear brighter in a backscattered electron image due to lower electron channelling. The image analysis methodology goes as follows (Figure 2): thresholding to identify the cracks (low intensity pixels), morphological dilation of the identified cracks to remove edge effects (topographical effects), removal of the cracks from the original image, median filter to remove noise, smoothing to remove intensity fluctuations in the matrix, thresholding of WEAs using a random-walk algorithm and finally removal of false positive WEAs based on the criterion that WEAs should be close to a crack. All image processing procedures were written in Python with the scikit-image library [3]. Chemical information of the WEAs and their surroundings was obtained using a field emission electron microprobe (JEOL JXA8530F). This instrument was also used for the image acquisition. As indicated in [4], quantification of the carbon concentration in SAE52100 is challenging due to the high concentration of chromium. As preliminary results, a lower carbon concentration was observed in the WEAs than in the surrounding matrix (Figure 3). Further analysis of the composition as well as examples of failed bearings under different operating conditions will be presented. Microsc. Microanal. 21 (Suppl 3), 2015 460 [1] Evans et al, Tribology International 65 (2013), pp. 146-160. [2] Kang et al, Scripta Materialia 69 (2013), pp. 630-633. [3] van der Walt et al, PeerJ 2:e453 (2014) [4] Pinard and Richter, IOP Conf. Series: Materials Science and Engineering 55 (2014), p. 012016. Figure 1: Result of the automated acquisition of high resolution images along the cracks. Figure 2: Examples of the image analysis routine used to automatically detect cracks (red) and white etching areas (blue). (a) (b) Figure 3: Qualitative carbon (a) and chromium (b) mapping of a white etching area.");sQ1[231]=new Array("../7337/0461.pdf","Novel Materials and Structures Fabricated by Electron Beam Melting","","461 doi:10.1017/S1431927615003104 Paper No. 0231 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Novel Materials and Structures Fabricated by Electron Beam Melting Harvey West1, Ola Harrysson1, Tim Horn1, Denis Cormier2, Ron Aman2, and Denis Marcellin-Little3 1. 2. Edward P. Fitts Dept. of Industrial and Systems Engineering, N.C. State University, Raleigh, NC. Dept. of Industrial and Systems Engineering, Rochester Institute of Technology, Rochester, NY. 3. Dept. of Clinical Sciences, College of Veterinary Medicine, N.C. State University, Raleigh, NC. Researchers in the Center for Additive Manufacturing and Logistics (CAMAL) in the Edward P. Fitts Department of Industrial and Systems Engineering at North Carolina State University have been active in the area of additive manufacturing since 2000, and have gained an international reputation for their research and educational efforts. While, historically, the first rapid prototyping machines were created to produce parts from thermoset or thermoplastic polymers, the additive manufacturing of metal components has become the most important research topic for a wide variety of industries including aerospace, medicine, power industry, and the military. NCSU was the first user of Electron Beam Melting (EBM) technology in the world when, in 2003, the world's first EBM machine was acquired from Arcam AB (M�lndal, Sweden). This technology allows for production of fully dense metal parts using an electron beam to selectively fuse layers of metal powder. The Arcam EBM machine was originally designed to process powders of H13 tool steel for the direct production of injection molding dies. Figure 1a shows the typical as-produced martensitic microstructure of this material [1]. Electron beam melting also allows the fabrication of complex shapes from materials that do not possess sufficient ductility for forging, or are difficult to cast. One of the first materials targeted was GRCop-84. This copper alloy gains its high temperature strength and creep resistance from a fine dispersion of Cr2Nb particulates, and can only be processed by powder metallurgy techniques since casting results in coarsening of the second phase. However, as can be seen in Figure 1b, when processed by EBM, the microstructure retains the size and distribution of the dispersoids [2]. Another high temperature material investigated for processing by EBM was the intermetallic -TiAl [3]. Avio Aero is planning to use EBM technology instead of casting to fabricate turbine blades from this lightweight alloy. Other alloys that have been successfully processed at NCSU are Inconel 625 and 718, niobium, nitinol, high purity copper, and aluminum alloys 2024, 6061, and 7075. Electron beam melting can also be used to consolidate non-metallics such as the metal oxides present in lunar regolith simulant. The obvious difficulty with this material is the dissipation of the beam current through the material to ground, but by adjusting the beam conditions, fusing of the material is possible as shown in Figure 2. At the suggestion of Dr. Harrysson, in 2003, Arcam began developing the machine control parameters for processing Ti6Al4V powder. With this new material available, the EBM system could be used to fabricate components for the medical and aerospace industries. Figure 3a shows the as-produced microstructure of this alloy [4]. One main advantage of the additive manufacturing process is that it allows the fabrication of structures that could not be produced conventionally. Figure 3b shows a custom transdermal osseointegrated implant fabricated for a canine patient in the College of Veterinary Medicine. Instead of making fully solid parts, incorporating mesh structures into implants can reduce the overall stiffness, and encourage bone ingrowth. Mesh structures can also be used to reduce the weight of aerospace components. However, the surface roughness of the mesh struts (Figure 3c), while appropriate for biological applications, may contribute to poor fatigue performance, and strategies to improve surface finish of as-processed or post-processed parts are being investigated. Microsc. Microanal. 21 (Suppl 3), 2015 462 Additional research at CAMAL involves the design of alloys specifically tailored for additive processes. For instance, the addition of elements to promote grain refinement could reduce the anisotropy that accompanies the growth of grains along the build direction (Figures 4a and 4b) [5]. Also, characterizing the effect of powder morphology and size distribution on the properties of parts built by additive manufacturing is essential for developing new materials or material systems (Figure 5). [1] D Cormier et al, Rapid Prototyping Journal 10 (2004) p. 35. [2] T Mahale, Ph.D. dissertation, N.C. State University (2009). [3] D Cormier et al, Research Letters in Materials Science 2007 (2007) p. 1. [4] D Cormier et al, Proceedings of the 15th Solid Freeform Fabrication Symposium (2004). [5] T Horn et al, Proceedings of Materials Science & Technology (2014). a 500 �m b 50 �m 500 �m Figure 1. Microstructures of (a) H13 tool steel, (b) GRCop-84 Figure 2. Lunar regolith simulant a 200 �m b c 1000 �m Figure 3. Ti6Al4V (a) microstructure, (b) custom implant, (c) as-fabricated mesh strut a 1000 �m b 1000 �m 50 �m Figure 4. (a)100% Ti6Al4V, (b) Ti6Al4V+1%B Figure 5. Powder examination");sQ1[232]=new Array("../7337/0463.pdf","Microstructural Analysis of 3D-Printed Alloy 718.","","463 doi:10.1017/S1431927615003116 Paper No. 0232 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microstructural Analysis of 3D-Printed Alloy 718. G. Bertali1, Y. Wang1, J. J. H. Lim1, F. Scenini1, C.J. Long2, P.D. Freyer3, and M. G. Burke1 1. 2 Material Performance Centre, The University of Manchester, Manchester (UK) Westinghouse Electric Company LLC, Hopkins, SC 29061 (USA) 3 Westinghouse Electric Company LLC, Pittsburgh, PA 15235 (USA) Three-dimensional printing (3DP), as an emerging additive manufacturing process in the last decade, is a promising technique for high-speed low-cost fabrication of complex parts compared with traditional methods. Although it is possible to 3D print complex alloys such as Ni-Cr-Fe alloys and stainless steels, the effect of this manufacturing process on the material microstructure remains a topic of research. The microstructural complexity of Ni-Cr-Fe-Nb-Ti-Al alloy (Alloy 718) especially in terms of second phase precipitation is well-known and has been extensively characterized and studied in the past [1-3]. The main precipitates that have been identified in Alloy 718 are '' (Ni3(Nb,Ti)), (Ni3Nb), ' (Ni3(Al,Ti)), MC, M23C6, M7C3, M6C and Laves. 3D-printed Alloy 718 microstructure is expected to be characterized by the presence of the same precipitates. However, the extent and distributions of these precipitates might be different from the conventionally-produced wrought alloy. These possible microstructural variations may affect macroscopic properties of the alloy such as elevated temperature resistance, creep resistance and strength. Therefore, a detailed microstructural characterization of the 3D printed Alloy 718 is needed in order to assess any differences from the conventionally-produced alloy. In this study, a combination of field-emission gun (FEG) scanning electron microscopy (SEM), brightfield (BF) and dark-field (DF) transmission electron microscopy (TEM), and analytical electron microscopy (AEM) techniques have been used to study the microstructure developed in 3D-printed Alloy 718 with emphasis on the type and extent of precipitation. The as-polished Alloy 718 was examined in a Zeiss MERLIN FEG-SEM with a GEMINI II column using both secondary electron (SE) and backscattered electron (BSE) modes. The microstructure was characterized by a layered and elongated grain structure (Fig. 1(a)) and by numerous subgrains (Fig. 1(b)). High resolution energy-selected backscattered electron analysis highlighted three different precipitate morphologies: coarse globular and plate-like intergranular brightly-imaging precipitates and fine acicular intragranular brightly-imaging precipitates. TEM analysis confirmed that the intragranular precipitates were " with a size of ~100 nm (Fig. 2). Discrete needle-like precipitates preferentially precipitated at high-angle grain boundaries. High resolution scanning transmission electron microscope (STEM) energy dispersive x-ray (EDX) microanalysis, using the FEI Titan G2 80-200 aberration corrected S/TEM equipped with super EDX, revealed marked micro-chemical variations in the fine precipitates. The disc-shaped " was composed of Ni, Nb and Ti, with a small amount of Al, and served as a nucleation site for subsequent precipitation of very fine" precipitates, which were enriched in Al (Fig. 4). These results will be discussed and compared with conventionally thermally-treated Alloy 718. References: [1] M.G. Burke et al. J. de Physique (1989) p. C8 395-400. [2] Moukrane Dehmas et al. in Advances in Materials Science and Engineering (2011). [3] D.F. Paulonis et al. in Trans. ASM (1969) p. 611. Microsc. Microanal. 21 (Suppl 3), 2015 464 a b Fig. 1: (a,b) BSE images of the 3D-printed Alloy 718 microstructure and the subgrain structure. Fig. 2: BSE image showing the brightly-imaging precipitates (" and ) with different morphologies. Fig. 3: TEM images of (a) the subgrain structure; (b) [001] BF and (c) DF images of the intragranular ". Fig. 4: HAADF STEM image and corresponding Titan "ChemiSTEM" SDD EDX spectrum images for Ti, Nb, Al, Ni of a coarse " precipitate with smaller discrete ' precipitates that have nucleated at the "/matrix interface.");sQ1[233]=new Array("../7337/0465.pdf","Characterization of Nickel Based Superalloys Processed Through Direct Metal Laser Sintering Technique of Additive Manufacturing","","465 doi:10.1017/S1431927615003128 Paper No. 0233 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of Nickel Based Superalloys Processed Through Direct Metal Laser Sintering Technique of Additive Manufacturing Yaakov Idell1, Carelyn Campbell1, Lyle Levine1, Fan Zhang1, G. Olson2, D. Snyder2 1. Materials Science and Engineering Division, National Institute of Technology and Standards, Gaithersburg, MD, USA 2. QuesTek Innovations LLC, Evanston, IL, USA Ni-based superalloys are widely used for their excellent mechanical properties, particularly in the field of high temperature applications, through the control of and ' phase ratios [1]. Due to the advances in technology, additive manufacturing of Ni-based superalloys will allow direct production of complex shaped components based on 3-D computer aided drawings. The aerospace industry is interested in exploiting this technology to reduce time and cost for production of complex parts; however, the effects resulting from repeated cycles of rapid heating, melting, cooling, and solidification on the microstructure-property relationships are not well understood. Traditionally, microstructural studies of Ni-based superalloys are limited to the transmission electron microscope (TEM) due to the small precipitate phases (', ", and phases); however, this characterization technique will not provide statistically significant data sets. The most extensively used X-ray technique for statistically significant characterization of bulk microstructure is small angle X-ray scattering (SAXS). In SAXS, the elastic scattering from low scatter angles within a material can reveal information regarding parameters such as the size, shape, volume, and information regarding scatters for both ordered or partially ordered scattering through small changes in the electron density induced by the long-range elastic dilation field, density changes in the dislocation core, and the nonlinear part of the elastic field [2]. Information delivered using SAXS can resolve feature sizes from 1 nm to 100 nm, which is not a useful technique when dealing with precipitates in the ultrafine and submicron grain size regime. For resolving microstructures in these size regimes, ultra-small angle X-ray scattering (USAXS) will need to be utilized as the smaller scatter angle is sensitive to microstructures as large as 30 �m through a BonseHart double crystal configuration with a silicon monochromator, attenuators, filters, and a variable slit [3]. Here we report and discuss the evolution of the microstructure-property relationships of Allvac 718Plus, a Ni-based superalloy, that have been built using direct metal laser sintering and its response to annealing at 1066 �C. In-situ annealing experiments using a combined USAXS/SAXS setup at the Advanced Photon Source in Argonne National Laboratory is shown in Figure 1a and 1b were performed, which is sensitive to feature sizes over a three decade range from just below 1 nm to just above 30 �m [4]. Complementary ex-situ experiments for the as-built and annealed microstructures are characterized through scanning and transmission electron microscopy, electron back scattering diffraction, X-ray diffraction, and synchrotron microbeam diffraction to determine the residual stress distributions, porosity, phase fractions, and compositional differences in an effort to develop the microstructureproperty relationships. These results are compared with multicomponent diffusion simulations and FEM simulations that predict the phase fraction, composition, and residual stress as functions of time and temperature. Figure 1c shows an example of 1D-collimated USAXS/SAXS during an in-situ annealing experiment from room temperature to 1066 �C. The 718Plus as built morphology shows an microstructure with Microsc. Microanal. 21 (Suppl 3), 2015 466 significantly elongated grains (aspect ratio of 6.5) with a dendritic solidification structure and possible precipitates present in Nb-rich interdendritic regions, a strong {200} cube texture, and tensile strength of 1270 MPa. Scheil solidification simulation and thermodynamic modeling indicate that regions of high compositional imbalance, such as the observed high concentrations of Nb in interdendritic regions, can lead to the formation of the undesirable carbides and phase. These results are similar to a previous study of Inconel 718, which is a Ni-based superalloy with a similar but slightly different composition than 718Plus, as-built by the additive manufacturing technique of electron beam melting [5]. Additionally, the processing technique introduces significant amounts of local tensile residual stress with local variations up to 1 GPa. Upon annealing, the microstructure is altered to an equi-axed morphology with no subgrains or dendritic structure, precipitates along grain boundaries, and local tensile residual stress, and a tensile strength of 720 MPa. By observing the size, shape, and volume evolution of the precipitates through the use of the USAXS/SAXS instrument, it is evident that the precipitates are growing with time. The significantly increased local Nb concentration from the DMLS processing segregation in turn causes precipitate dissolution temperature to increase as well [6] allowing precipitate growth. The USAXS/SAXS data will aid in selecting an optimal solutionizing heat treatment time to balancing stress relief, homogenized composition, and an equi-axed non-dendritic microstructure while minimizing precipitate and matrix grain size. [1] D.M. Shah, D.N. Duhl, Superalloys, (1984) 105-114. [2] L. Levine, G. Long, Diffuse Scattering and the Fundamental Properties of Materials, (2009) 345. [3] U. Bonse, M. Hart, Applied Physics Letters, 7 (1965) 238-240. [4] J. Ilavsky, et al, Journal of Applied Crystallography, 42 (2009) 469-479. [5] A. Strondl, et al, Materials Science and Technology, 27.5 (2011) 876-883. [6] R.E. Schafrik, et al, TMS, (2001) 1-11. Figure 1: Schematic of the APS USXS instrument in (a) one-dimensional collimated or USAXS imaging configuration, and (b) two-dimensional collimated configuration; (c) example of statistically relevant USAXS/SAXS data set illustrating microstructural changes during in-situ annealing experiment.");sQ1[234]=new Array("../7337/0467.pdf","Comparison of Additive Manufactured and Conventional 316L Stainless Steels","","467 doi:10.1017/S143192761500313X Paper No. 0234 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 J. J. H. Lim1, A. R. C. Malheiros1, G. Bertali1, C. J. Long2, P. D. Freyer3 and M. G. Burke1 1. Material Performance Centre, School of Materials, University of Manchester, Manchester (UK) 2 Westinghouse Electric Company LLC, Hopkins, SC 29061 (USA) 3 Westinghouse Electric Company LLC, Pittsburgh, PA 15235 (USA) Additive manufacturing (AM) technology provides a high degree of design freedom and is capable of producing near-net-shape sophisticated components with tailored properties [1, 2]. These unique properties provide savings in terms of raw material costs and a simplified manufacturing chain process. Thus, the prospect of microstructural design through AM is extremely attractive for manufacturing structural components for nuclear power plants. However, with this level of control, process-induced imperfections, such as porosity [3], residual stress [4] and preferential grain growth orientation [5], need to be addressed before AM can be used for `real' engineering applications. In order to overcome these imperfections, it is crucial to understand the connection between microstructure, processing and mechanical properties of AM materials. Stainless steels are commonly used in the AM process to manufacture near-fully dense components from 3D CAD models. The tribology and corrosion properties of stainless steel make it important in many engineering applications, including the nuclear energy sector. However, little information has been reported about the microstructure of AM stainless steels. In this work, the microstructures of an AM AISI Type 316L stainless steel and a conventional thermomechanically-processed Type 316L stainless steel were investigated. These 316L stainless steels produced via different processing routes were characterized by electron microscopy techniques, which provides microstructural information from the nanometer to micrometer level. Electron backscattered diffraction (EBSD) analysis, as shown in Figure 1(a), indicates a high degree of anisotropy in the microstructure of the AM 316L stainless steel. Large columnar grains with a strong <101> texture parallel to the deposition/melting direction (Rolling Direction/Inverse Pole Figure-X direction; RD/IPF-X) were also observed, as shown in Figure 1(b). A high level of strain within the grains is clearly visible in Figure 1(c). The scanning electron microscopy (SEM) analysis revealed a very small volume of micro-porosity present in the AM 316L stainless steel. Figure 2(a) shows the scanning transmission electron images obtained from the AM 316L stainless steel. Numerous dislocations were observed in each grain, as shown in Figure 2(b). Also, numerous TiMn oxides, which had not been observed in the conventional 316L stainless steel, were observed to decorate the grain boundaries and dislocation lines. Initial grain boundary analyses indicate that no segregation was observed at either high- or low-angle grain boundaries. The comparison of the AM 316L stainless steel and conventional wrought stainless steel will be further discussed in the presentation. References: [1] Murr, L. E., et al. (2012). J. of Mat. Sci. & Tech. 28(1): 1-14. [2] NIST, US Department of Commerce, May 2013 [3] Frazier, W. E. (2014). J. of Mat. Eng. & Per. 23(6): 1917-1928. [4] Wu, A., et al. (2014). Met. & Mat. Trans. A 45(13): 6260-6270. [5] Niendorf, T., et al. (2013). Met. & Mat.Trans. B 44(4): 794-796. Comparison of Additive Manufactured and Conventional 316L Stainless Steels Microsc. Microanal. 21 (Suppl 3), 2015 468 (a) (b) (c) Figure 1: EBSD grains orientation maps showing (a) grain orientation at direction of IPF-Z, (b) grain orientation along the deposition/melting direction, i.e. IPF-X or RD, & (c) higher magnification of grain orientation along the direction of IPF Z (a) (b) Figure 2: (a) High-angle annular dark field (HAADF) micrograph showing the columnar grains of AM 316L stainless steel, (b) Medium-angle annular dark field (MAADF) micrograph showing high number density of dislocations present in a grain of AM 316L stainless steel. The line features with brighter contrast are dislocations. Ti-rich oxides that appear in darker contrast are clearly visible and are decorated along grain boundary or associated with dislocation lines.");sQ1[235]=new Array("../7337/0469.pdf","Imaging the Rapid Solidification of Metallic Alloys in the TEM","","469 doi:10.1017/S1431927615003141 Paper No. 0235 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Imaging the Rapid Solidification of Metallic Alloys in the TEM John D. Roehling1, Aur�lien Perron1, Jean-Luc Fattebert2, Daniel R. Coughlin3, Paul J. Gibbs3, John W. Gibbs3, Seth D. Imhoff3, Damien Tourret3, J. Kevin Baldwin4, Amy J. Clarke3, Patrice E.A. Turchi1, and Joseph T. McKeown1 Materials Science Division, Lawrence Livermore National Laboratory, Livermore, CA, USA 2 Center for Applied Scientific Computing, Lawrence Livermore National Laboratory, Livermore, CA, USA 3 Materials Science and Technology Division, Los Alamos National Laboratory, Los Alamos, NM, USA 4 Center for Integrated Nanotechnologies, Los Alamos National Laboratory, Los Alamos, NM, USA The macroscopic properties of a metal solidified from a liquid melt are strongly dependent on the final microstructure, which in turn is the result of the solidification conditions. With the growing popularity of laser-based additive manufacturing (AM), there is an increasing need to understand the microstructures that result from rapid solidification processes. Rapidly solidified alloy microstructures are typically far from equilibrium and therefore traditional thermodynamic approaches used to predict structure and composition (i.e., phase diagrams) must be extended to describe these deviations from equilibrium and ensuing metastable states. This work highlights progress toward corroborating predictive (phase-field) modeling capabilities [1] with in situ experimental observations [2] in order to better understand the non-equilibrium structures produced during rapid solidification following laser melting. Identification of solidification modes and associated velocities using ex situ analyses after laser melting and rapid solidification usually represent averages for the complete liquidsolid transformation and fail to detect changes in velocity that are expected when transitions in the solidification mode occur. Models predicting the solidification mode can potentially bridge the gap, but direct experimental measurements of the kinetics and growth modes are needed to verify the modeling results. To directly observe rapid alloy solidification under laser-induced conditions that apply to AM processing, the dynamic transmission electron microscope (DTEM) at Lawrence Livermore National Lab (LLNL) was used in Movie Mode [3] (Figure 1). The Movie Mode capability at LLNL allows for single solidification events to be monitored in situ. The quasi-instantaneous velocity can be measured (Figure 2) over a given time domain, which would be inaccessible with lower temporal resolution techniques such as conventional TEM, and related directly to the observed solid-liquid interface evolution in the time-resolved images. Additionally, the ability to alter the power of the sample drive laser allows for a range of solidification 1 Microsc. Microanal. 21 (Suppl 3), 2015 470 velocities to be accessed. The resulting solidification structures are analyzed with grain orientation mapping and elemental analysis (EELS and EDS). Finally, the data are compared against phase-field modeling of solidification under different conditions. References [1] J.-L. Fattebert, M.E. Wickett, P.E.A. Turchi, Acta Materialia 62 (2014) 89. [2] J.T. McKeown et. al, Acta Materialia 65 (2014) 56. [3] T. LaGrange, B.W. Reed, D.J. Masiel, MRS Bulletin 40 (2015) 22. [4] This work was performed under the auspices of the U.S. Department of Energy (DOE) by Lawrence Livermore National Laboratory under Contract DE-AC5207NA27344 and Los Alamos National Laboratory (LANL), operated by Los Alamos National Security, LLC, under Contract DE-AC52-06NA25396. Work at LLNL was funded by the Laboratory Directed Research and Development Program at LLNL under project tracking code 15-ERD-006. AJC, DRC, PJG, JWG, SDI, and DT were supported by AJC's Early Career award from the U.S. DOE, Office of Basic Energy Sciences, Division of Materials Sciences and Engineering. Work at LANL was performed, in part, at the Center for Integrated Nanotechnologies, an Office of Science User Facility operated for the U.S. DOE Office of Science. Figure 2: Measured solidification velocities from the DTEM Movie Mode acquisition shown in Figure 1. Figure 1: Dynamic TEM Movie Mode images of an Al-Si film on amorphous SiNx, microseconds after induced laser melting at the edge of the solidifying melt pool.");sQ1[236]=new Array("../7337/0471.pdf","Advanced Electron Microscopy for Energy Related Materials","","471 doi:10.1017/S1431927615003153 Paper No. 0236 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Advanced Electron Microscopy for Energy Related Materials Ryo Ishikawa1, Naoya Shibata1,2 and Yuichi Ikuhara1,3,4 1. 2. Institute of Engineering Innovation, University of Tokyo, Bunkyo, Tokyo, 113-8656, Japan. Japan Science and Technology Agency, PRESTO, Saitama 332�0012, Japan 3. WPI Research Center, AIMR, Tohoku University, Sendai 980-8577, Japan 4. Nanostructures Research Laboratory, Japan Fine Ceramics Center, Atsuta, Nagoya 456�8587, Japan Recent progress on electron optics promotes the capability of scanning transmission electron microscopy (STEM) and it becomes versatile tool for solving materials problems at atomic scale. The most popular ADF imaging mode has a strong sensitivity of the atomic number (Z-contrast) and one can see even single heavy dopants, but it is insensitive to the light elements. On the other hand, recently developed annular bright-field (ABF) imaging is capable to directly see oxygen, lithium and even hydrogen atoms within thin crystals [1, 2], and now ABF imaging is routinely used for the determination of atomic structures of energy related materials. Fig. 1 shows the typical ADF and ABF STEM images obtained from Li2TMO3 in which oxygen and lithium atoms are basically invisible in ADF but can see these atoms in ABF image. In this presentation, we show the basic image formation of ABF STEM and some applications of ABF/ADF for light element materials such as lithium ion battery [3] or photocatalytic materials [4]. Manganese-rich lithium oxide (Li1+xTM1-xO2, TM: transition metal) has actively investigated as cathode materials of lithium ion battery. On the basis of X-ray or neutron diffraction, it was revealed that the material consists of two phases such as rhombohedral and monoclinic structures, but the spatial distribution of two phases is still under controversial. The typical grain size is a few micrometers and we often found two phases in single grains through conventional electron diffraction, suggesting that the size of two phases (or domains) is relatively small. So it requires atomic-resolution observation to identify whether interlayer growth or two-phase separation. In the bulk regions of Li2TMO3, it is found several stacking faults along c-axis or a few layer inter growth of monoclinic-type structures, which gives not new spots but diffuse streaks in diffraction pattern. After the heavy investigation, we found two-phase separation and it forms hetero-interface along c-axis, which is directly observed by ABF/ADF STEM imaging. The other example is photocatalytic water splitting oxynitride of LaTiO2N. The photocatalytic reaction occurs on the surface and hence it is important to understand the surface atomic structure. Fig. 2(a) shows ABF STEM image obtained from the surface atomic structure of LaTiO2N after annealing in the air. One can see the reconstructed atomic structure on the first top-surface layer. When the specimen is quickly treated with aqua regia (15 seconds), it removes only surface reconstructed regions. This quick treatment significantly improves the photocatalytic activity for both the H2 and O2 evolution reaction, which is firstly realized by ABF/ADF STEM imaging. Details and further examples will be shown in this presentation. Microsc. Microanal. 21 (Suppl 3), 2015 472 References [1] SD Findlay et al, Appl. Phys. Lett. 95 (2009) 191913. [2] R Ishikawa et al, Nature Mater. 10 (2010) 278. [3] H Yu et al, Angew. Chem. Int. Ed. 52 (2013) 5969. [4] M Matsukawa et al, Nano Lett. 52 (2014) 5969. [5] The authors acknowledge our collaborators of Dr. H. Yu, Prof. H.S. Zhou in AIST, and Mr. M. Matsukawa, Dr. T. Hisatomi, Dr. Y. Moriya and Prof. K. Domen in Univ. Tokyo. The work was supported by the Funding Program for World-Leading Innovation R&D on Science and Technology (FIRST Program). Figure 1. (a), (b) Atomic resolution ADF and ABF STEM images obtained from Li2TMO3 viewed along the [110] direction, where TM is purple, oxygen is red and lithium is green. The simulated images are overlaid on the right top, respectively. (a) Vacuum (001) (b) Vacuum Figure 2. Atomic resolution ABF STEM images obtained from LaTiO2N viewed along the [100] direction: (a) as-annealed and (b) treated with aqua regia for 15 seconds. The high-magnified image is overlaid on (a). The scale bar is 5 �.");sQ1[237]=new Array("../7337/0473.pdf","Charge-Discharge Cycling Induced Structural and Chemical Evolution of Li2MnO3 Cathode for Li-ion Batteries","","473 doi:10.1017/S1431927615003165 Paper No. 0237 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Charge-Discharge Cycling Induced Structural and Chemical Evolution of Li2MnO3 Cathode for Li-ion Batteries Pengfei Yan1, Jianming Zheng2, Ji-Guang Zhang2, and Chong-Min Wang1 1 Environmental Molecular Sciences Laboratory, Pacific Northwest National Laboratory, 902 Battelle Boulevard, Richland, WA 99352, USA 2 Energy and Environmental Directorate, Pacific Northwest National Laboratory, 902 Battelle Boulevard, Richland, WA 99352, USA Capacity and voltage fading of lithium-and-manganese-rich (LMR) cathodes is a major challenge for the application of this category of material, which is believed to be associated with the structural and chemical evolution of the materials.[1-3] As one of the most promising cathode materials for next generation lithium ion batteries, LMR cathode materials have the expression xLiMO2�(1-x)[Li2MnO3] (M=Mn, Ni, Co, 0x<1). As the parent component of LMR, Li2MnO3 cathode materials also attracted much attention for many reasons. First, Li2MnO3 itself has a very high theoretical capacity. Second, it can enhance our fundamental understanding of the electrochemistry of Mn4+-containing cathode materials. Thirdly, we can obtain the knowledge necessary for design new LMR cathode materials. Li2MnO3 cathode was initially believed electrochemical inert because all the Mn cations are Mn4+ in this material until Kalyani et al. demonstrated that Li2MnO3 could be electrochemically activated in 1999.[4] It has been claimed that prolonged cycling could transform Li2MnO3 to LiMn2O4-spinel.[5, 6] However, recently X-ray absorption spectroscopy results from cycled sample are not consistent with the LiMn2O4-spinel structure, instead the P3 structure was proposed.[7] In a chemically delithiated Li2MnO3 sample, Wang et al. observed the formation of MnO2 phase.[8] Up to now, it is still not clear which structure is formed after charge-discharge cycles and how the structure evolves during the cycling of Li2MnO3 cathode. In this work, we report the detailed structural and chemical evolutions of Li2MnO3 cathode captured by using aberration corrected scanning/transmission electron microscope (S/TEM) after certain numbers of charge-discharge cycling of the batteries. It is found that structural degradation occurs from the very first cycle and is spatially initiated from the surface of the particle and propagates towards the inner bulk as cyclic number increase, featuring the formation of the surface phase transformation layer and gradual thickening of this layer. The structure degradation is found to follow a sequential phase transformation: monoclinic C2/m tetragonal I41 cubic spinel (as shown in Figure 1), which is consistently supported by the decreasing lattice formation energy based on DFT calculations. For the first time, high spatial resolution quantitative chemical analysis reveals that 20% oxygen in the surface phase transformation layer is removed and such newly developed surface layer is a Li-depleted layer with reduced Mn cations (Figure 2). This work demonstrates a direct correlation between structural degradation and cell's electrochemical degradation. In a more general term, since Li2MnO3 cathode is the parent compound for LMR cathode, this work will enhance our understanding on the degradation mechanism of LMR cathode materials during cycling. References: [1] F. Lin, I. M. Markus, D. Nordlund, T. C. Weng, M. D. Asta, H. L. Xin, M. M. Doeff, Nature Commun. 5 (2014), 3529. Microsc. Microanal. 21 (Suppl 3), 2015 474 [2] M. Gu, I. Belharouak, J. Zheng, H. Wu, J. Xiao, A. Genc, K. Amine, S. Thevuthasan, D. R. Baer, J.-G. Zhang, N. D. Browning, J. Liu, C. Wang, ACS Nano 7 (2012), 760. [3] P. Yan, A. Nie, J. Zheng, Y. Zhou, D. Lu, X. Zhang, R. Xu, I. Belharouak, X. Zu, J. Xiao, K. Amine, J. Liu, F. Gao, R. Shahbazian-Yassar, J. G. Zhang, C. M. Wang, Nano Lett. 15 (2015), 514. [4] P. Kalyani, S. Chitra, T. Mohan, S. Gopukumar, J. Power Sources 80 (1999), 103. [5] A. D. Robertson, P. G. Bruce, Chem. Mater. 15 (2003),, 1984. [6] J. Reed, G. Ceder, A. Van der Ven, Electrochem. Solid-State Lett. 4 (2001), A78. [7] J. Rana, M. Stan, R. Kloepsch, J. Li, G. Schumacher, E. Welter, I. Zizak, J. Banhart, M. Winter, Adv. Energy Mater. 4 (2014), 1300998. [8] R. Wang, X. Q. He, L. H. He, F. W. Wang, R. J. Xiao, L. Gu, H. Li, L. Q. Chen, Adv. Energy Mater. 3 (2013), 1358. Figure 1. (a-d) [100]m zone SAED patterns form different samples. Red arrows in (a) highlight the streaks due to the formation of stacking faults and lamellar domains in C2/m structure . Dashed red circles in (b-d) highlight extra diffraction spots due to structure transformation from C2/m to spinel. (e-h) TEM images to show particle morphology evolution from pristine sample to 45-cycles sample. (i-k) STEM-HAADF images show lattice structure change after 10 cycles. Figure 2. (a, b) STEM-EELS mapping of a 10-cycles sample. The mapping area is highlighted by dashed red frame in (a). (c-e) STEM-EELS analysis of 3 positions (1, 2 and 3, shown in (c)) of a 10-cycles sample. The spectra are normalized using Mn-M edge and Mn-L3 edge in (d) and (e), respectively. In (d), the depressed Li-K edge in positions 1 and 2 indicates less Li content at the two positions. In (e), the depressed Mn-L2 edge was shown for the 10-cycles sample as compared with pristine sample.");sQ1[238]=new Array("../7337/0475.pdf","Lithium Ordering in Next-Generation High-Voltage Lithium-Rich Layered Oxide Cathode Battery Materials","","475 doi:10.1017/S1431927615003177 Paper No. 0238 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Lithium Ordering in Next-Generation High-Voltage Lithium-Rich Layered Oxide Cathode Battery Materials Michael Behr1, Michael Lowe1, Britt Vanchura1, Fu Zhou1, Wenjuan Liu1, Jia Liu1 1 The Dow Chemical Company, Midland, MI, USA Lithium-rich layered oxide cathode materials, with general formula Li1+MO2 (M = transition metal or combination of metals), represent the next generation of high-voltage, high-capacity Li-ion battery cathode materials for plug-in electric vehicle applications. These materials offer up to ~90% increased lithium capacity, increased safety, and lower cost potential compared to incumbent LiMO2 positive electrode active materials such as LiCoO2 or Li[NixMnyCo1-x-y]O2. In this study, transmission electron microscopy (TEM), electron diffraction (ED), x-ray energydispersive spectroscopy (XEDS) and x-ray diffraction (XRD) techniques were used to characterize these structurally-complex materials both in their pristine as-synthesized state, and after extensive electrochemical cycling. As synthesized, both ED and XRD reveal the overall monoclinic C2/m symmetry, indicating the presence of lithium ordering in transition-metal layers, similar to its parent Li2MnO3 crystal structure. However, it was found that this lithium ordering is not uniformly present throughout the sample. Convergent-beam electron diffraction (CBED) obtained from numerous nanometer-size regions across single nanoparticles reveal the presence of domains with trigonal R-3m symmetry, indicating disordered transition metal layers similar to the LiCoO2 structure. Additional high-resolution imaging confirms that ordered lithium-rich domains are inter-grown with Li-poor domains, over length scales ~1-3 nm, and randomly distributed throughout the sample. Upon electrochemical cycling, conversion from a layered crystal structure to spinel-like structure was captured as a function of number of cycles using both dark-field imaging and CBED. Cathode nanoparticles were synthesized from Ni, Mo, and Co metal and lithium carbonate precursors. Precursor powders were mixed together using ball-milling processes, and then heat-treated at 850 � 1050 �C for ~10 hours to convert to ~100-200 nm particles Li1+ [NixMnyCo1-x-y]O2. Figure 1 shows the inter-grown C2/m and R-3m structures present within a single nanoparticle of assynthesized Li1+ [NixMnyCo1-x-y]O2. Nanometer size regions exhibiting either monoclinic or trigonal character were observed by CBED patterns collected from the nanocrystal. Specificially, CBED patterns were collected from regions spaced ~10 nm apart along the indicated lines from a [001]oriented nanoparticle. The relative intensities of the unique C2/m (blue) and R-3m (black) diffraction spots were evaluated for each diffraction pattern, and their ratio (IC2/m/IR-3m) is plotted as a function of probe position across the nanoparticle. As is evident from this plot, nanometer size regions of R-3m and C2/m symmetry are present and distributed randomly throughout the nanocrystal. Blue arrows indicate regions with high C2/m diffraction spot intensity, and thus indicate high levels of Li content and ordering. The variation in intensity ratio further indicates that lithium content and ordering varies continuously throughout the nanoparticle, consistent with a solid-solution of these two phases. This finding is consistent with previous findings that the stochastic distribution of C2/m and R-3m local structures is determine both by the Li/Me ratio, and (Ni+Co)/Mn ratio [1,2]. Figure 2 shows images and diffraction patterns of materials cycled for 2 and 200 cycles. Upon Microsc. Microanal. 21 (Suppl 3), 2015 476 electrochemical cycling, diffraction patterns reveal the appearance of new diffraction spots, whose spacings and symmetries are consistent with that of Fd3m spinel. After two cycles, dark-field images show that nanoparticle surfaces exhibit spinel-like structure, yet the layered R-3m and C2/m structures remain throughout the bulk. However, after 200 cycles, only R-3m and spinel-like diffraction patterns are observed, indicating the loss of Li ordering, and migration of transition metal ions into the Li layers. In addition, new nm-size crystalline facets are observed, indicating migration and recrystallization. References [1] M. Gu et al, Nano Letters 12 (2012), p. 5186. [2] J. Bareno et al, Chemistry of Materials 23 (2011), p. 2039. Quantification scheme A B C2/m B = [001] R-3m C2/m [001] R-3m [211] A B Figure 1. (upper left) Bright-field TEM image of a single cathode nanoparticles with representative C2/m and R-3m CBED patterns obtained along the indicated lines. (upper right) The quantification scheme used to evaluate the degree of C2/m character from each CBED pattern. (bottom) Intensity ratio IC2/m/IR-3m plotted as a function of 2 cycles 200 cycles Faceting C2/m [013] R-3m [100] Spinel [011] Figure 2. (Left) Dark-field image of nanoparticle after 2 cycles produced with weak spinel diffraction spot indicated (simulated spinel and R-3m pattern shown in inset) shows spinel phase distributed on the surface of the nanoparticles. (Right) HRTEM image of nanoparticles surface after 200 cycles shows formation of new facets.");sQ1[239]=new Array("../7337/0477.pdf","Comparison of Co3O4 and CoO Nanoparticles as Anodes for Lithium-ion Batteries","","477 doi:10.1017/S1431927615003189 Paper No. 0239 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Comparison of Co3O4 and CoO Nanoparticles as Anodes for Lithium-ion Batteries Jing Charlotte Li1,2, Kai He1, Eric A. Stach1, and Dong Su1,2 1. 2. Center for Functional Nanomaterials, Brookhaven National Laboratory, Upton, NY 11973, USA Department of Materials Science and Engineering, Stony Brook University, Stony Brook, NY 11794, USA The development of portable electronic devices demands improvements in rechargeable batteries that will lead to high energy and power density, low cost and long cycle life. Among of the available kinds of batteries, lithium-ion batteries have attracted extensive attention because of their high energy density. Because the graphite anodes conventionally used in commercial lithium-ion batteries have a limited specific capacity (372mAh g-1), other anode materials are being investigated, e.g. transition metal oxides which have much higher specific capacity. Cobalt based oxide materials have attracted significant attention due to their high theoretical capacity (890mAh g-1) and good capacity retention[1,2]. While the electrochemical performance of Co3O4 and CoO has been widely studied, the reaction mechanisms that occur during lithiation in these materials remain unclear. Early studies suggested that the lithiation of Co3O4 is a two-step reaction (first proceeding by intercalation then by conversion), other studies have shown that decomposition occurs at the initiation of the process [3]. Because Co3O4 and CoO have both different cycling performance and different crystal structures, an understanding of their reaction mechanisms and kinetics and the relation of these to their structures is of fundamental interest. Here, we report investigation of the lithiation process of Co3O4 and CoO nanoparticles using aberration-corrected transmission electron microscopy (TEM) techniques. We have studied samples that have been cycled in coin cells, and correlated these with real time studies using an in-situ electrochemical dry cell approach (Figure 1a). As shown in the scanning TEM (STEM) images of Figures 1b and c, we observed different lithium induced changes in the structure of Co3O4 and CoO respectively. In Figure 1b, from left to right, significant contrast changes occur in Co3O4 at the beginning of lithiation, with the formation of fine-sized nanoparticles during further lithiation. In contrast, only fine-sized nanoparticles form during the lithiation of CoO (Figure 1c). Electron diffraction patterns (Figure 2) allow us to track the phase transitions that occur during lithiation directly. As shown in Figure 2a, an intermediate rock-salt phase (LiCoO2) is generated when Li+ ions first enter the structure, and the system decomposes into metallic Co and the anti-fluorite structure Li2O during further lithiation. In contrast, CoO is reduced to metallic Co and Li2O directly (Figure 2b). We found that during lithiation, the irreversible incorporation of Li+ pushed the Co2+ ions that were present in the 8a site to neighboring 16c sites during the initial insertion into the spinel Co3O4 structure. At the same time, Co2+ ions at the 16d sites maintain the general framework along with oxygen expansion, which corresponds to the intermediate phase (LiCoO2). With further lithiation, the Co2+ ions originally present at the 16d sites had been removed from these sites, leading to the formation of the anti-fluorite Li2O structure. In the case of CoO (which has the rock-salt structure), the incorporation of Li+ pushed the Co2+ ions out of all of the tetrahedral sites. Electron energy-loss spectroscopy (EELS) and high resolution TEM (not shown) confirmed these results in the ex situ cycled samples. Understanding the difference in the reaction mechanisms between Co3O4 and CoO provides needed insight into the differences observed in their performance [4]. References: Microsc. Microanal. 21 (Suppl 3), 2015 478 [1] M.M. Thackeray, S. D. Backer., K.T. Adendorffand J.B.Goodenough, Solid Stale Ionics 17, (1985) 174-181. [2] P. Poizot, S.Laruelle, S.Grugeon, L.Dupont and J-M. Tarascon, Nature 407, (2000) 496-499 [3] V. Pralong, J.B. Leriche, B. Beaudoin, E. Naudin, M. Morcrette, J.-M. Tarascon. Solid State Ionics 166(3-4), (2004) 295-305. [4] Research carried out at the Center for Functional Nanomaterials, Brookhaven National Laboratory, which is supported by the U.S. Department of Energy, Office of Basic Energy Sciences, under Contract No. DE-SC0012704." Figure 1. (a) Schematic illustration of in-situ TEM dry cell. Time sequence of STEM images showing the structural evolution of (b)Co3O4 and (c) CoO during in-situ lithiation. Figure 2. Selected area diffraction patterns and corresponding structure model showing the phase evolution of (a) Co3O4 and (b) CoO.");sQ1[240]=new Array("../7337/0479.pdf","The Evolution of Project NANO: A Program that Enables Students to Explore in Real Time Several Crosscutting Concepts of the Next Generation Science Standards","","479 doi:10.1017/S1431927615003190 Paper No. 0240 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Evolution of Project NANO: A Program that Enables Students to Explore in Real Time Several Crosscutting Concepts of the Next Generation Science Standards Sherry L. Cady1, Mikel Blok2, Keith Grosse3, and Jennifer Wells4 1 Pacific Northwest National Laboratory, Richland, WA USA Beaverton High School, Beaverton, OR USA 3 Lake Oswego High School, Lake Oswego, OR USA 4 Center for Science Education, Portland State University, Portland, OR USA 2 The teacher professional development program known as Project NANO (Nanoscience and Nanotechnology Outreach) has more than 75 teachers in 8 different school districts that have given more than 7500 students in Portland area high school and middle school classrooms hands-on experience exploring the submicroscopic world with a research-grade scanning electron microscope (SEM). These students experience the same type of excitement and joy that every career microscopist has felt when they discover something new or unexpected with their microscope of choice. It's hard to imagine a more direct means to stimulate every student's sense of wonder about how and why natural and man-made materials function as they do than to allow students to use research-grade instruments to conduct their own inquiries and make their own discoveries. Project NANO students' explore several crosscutting concepts of the Next Generation Science Standards (NGSS) in real time with technology that illustrates how structure links with function, reveals hidden sub-microscopic patterns, and demonstrates when scale, proportion and quantity can affect a system's structure or performance. Project NANO students can deepen their understanding of the natural sciences and nanoscale concepts through inquiry projects. While it's easy to understand how Project NANO students could increase their awareness of nanoscience by over 60% and become aware of more STEM careers that involve using an SEM, the program has also increased rigor and the use of inquiry in the classroom. Project NANO workshops have evolved i) via the results of pre- and post-assessments, which demonstrate how directed and inquiry-based authentic research experiences impact student's attitudes toward (nano)science, and ii) through the use of resource folios, which support teachers' efforts to develop and scaffold lesson plans that incorporate the SEM and stereoscopes into authentic research experiences that reinforce their curriculum and reflect their unique teaching styles. Workshops: Project NANO operates via a collaboration between high school teachers, educational researchers, university faculty and administrators, school districts and administrators, and funding partners that include a philanthropic donor and electron microscope manufacturers and representatives. Our teacher training workshops (one-week summer professional development courses for novice and veteran Project NANO teachers) enable in-service teachers to go back into their classroom with a well crafted unit that fits their needs. Middle school teachers in particular develop cross-curricular lessons as part of a forensics science unit that involves biology, chemistry, physics, and geology. Veteran Project NANO teachers refine their initial lesson plans by taking the more advanced workshop, which has led to highly vetted lesson plans for each of the natural sciences. We have modified our workshops to provide all participants with more time to learn how the SEM works, become proficient with the instruments and image analysis with freeware like Image J, and prepare written and oral lesson plans and presentations. As part of the workshops, the teachers themselves work through an inquiry process as they collect data and learn to use the SEM. Teachers build their lesson plans and vet them with other in-service teachers from different schools, pre-service teachers still involved in University Education studies, relatively new Microsc. Microanal. 21 (Suppl 3), 2015 480 (1st -3rd yr) teachers, and with veteran (20-35 yrs) teachers via the workshops. This multi-tiered training format fosters and facilities new networks among teachers that enables them to develop a new community of practice among teachers in the Portland Metro Region. Each group brings value to the process as they share their experiences that resulted from use of the technology, the development of classroom pedagogy savvy, and by encouraging each other to teach with research-grade technology in the classroom. Veteran Project NANO teachers provide wise counsel for those early career teachers who are still developing strategies and grasping the depth of ideas they can share with their students about nanoscience. As a result of our integrated, multi-level workshop classrooms, Project NANO instructors are developing a growing collection of excellent next generation nanoscience/nanotechnology lessons. Pre- and Post-Assessments: Project NANO teachers assign an on-line survey that asks students to selfevaluate their own experience and level of proficiency related to using microscopes and their interest in using the SEM to conduct research or as a possible career option. We found statistically significant (16% on average) gains between the pre- and the post-survey results after just one exposure to a Project NANO lesson. The highest gains were linked to questions related to students' sense of proficiency in using both the SEM and a stereomicroscope (average 25% increase) as well as their interest in pursuing future nanoscale science related learning experiences (21% increase). As Project NANO has evolved over the past 6 years, its network of middle school and high school teachers has grown. In this way, some students can even be exposed to Project NANO in 7th, 8th and each year of high school. We hypothesize that the positive effect of Project NANO on student learning and awareness of possible STEM careers will increase even more as more students explore nanoscience concepts with the use of research-grade nanotechnology in sequential years in a variety of science classes. Resource Folios: The use of Resource Folios has also informed program adaptations in subsequent years. We found that i) when the summer workshop teachers scored lower on their pre- and post-surveys and units of instruction, most often because those teachers had less experience working with nanoscale concepts, their average class grades were lower; ii) teachers who were the most experienced working with nanoscale concepts scored the highest on the pre- and post-surveys and exhibited the highest level of confidence working with the SEM during the summer workshops � their classes also had the highest learning outcomes; iii) the teachers who expressed the highest level of anxiety throughout the summer workshop planning period, classroom observations, and during the focus group interviews also reported the lowest average student grades; and iv) when student grades were triangulated with classroom observation and focus group data, we found that teachers with the largest class sizes struggled to engage all of the students consistently throughout the entire inquiry process. We have found that summer workshops and academic year coaching positively supports teachers by enabling them to set their own well-reasoned learning goals, which also included extending their efforts to support student research in science competitions. Students are thrilled to work with leading edge technology and teachers are excited to improve student learning using technology. Project NANO has evolved in ways that "move the needle" by breaking new ground in teaching and learning in areas that cover cross-cutting concepts of the NGSS and with regard to the involvement and use of advanced technology and nanoscale concepts in science inquiry and authentic research experiences. Support for Project NANO has come from the M.J. Murdock Charitable Trust, Portland State University, Nanoscience Instruments, Phenom-World., FEI Hillsboro, and SLC from the Environmental Molecular Sciences Laboratory, a DOE Office of Science User Facility sponsored by the Office of Biological and Environmental Research and located at PNNL.");sQ1[241]=new Array("../7337/0481.pdf","Electron Microscopy Education Outreach for Secondary and Professional Education","","481 doi:10.1017/S1431927615003207 Paper No. 0241 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electron Microscopy Education Outreach for Secondary and Professional Education Greg Baty1, Barbara Miner.2, E. Koehler2, Zhiqiang Chen1 1. 2. Center for Electron Microscopy and Nanofabrication, Portland State University, Portland, OR 97221 Saturday Academy, Portland, OR 97203 In addition to university level Electron Microscopy courses with practical laboratory components, Center for Electron Microscopy and Nanofabrication (CEMN) offers outreach to professionals and high school students. Through our education efforts, students from junior / high school, university and industry have received intensive training in electron microscopy and related techniques. Instrument vendors provide introductory instrument training when a microscope is purchased. Over time, staff attrition often affects a company's technical expertise. Professionals may not have access or time to take a traditional university course when on-the-job training is insufficient. Many universities have professional education outreach that helps industry maintain expertise. Lehigh is perhaps, the most well known for their highly structured microscopy short courses [1]. CEMN offers an `Engineering Certificate in Applied Electron Microscopy' (ECAEM). The goal of the ECAEM program is to increase the competence of new and practicing microscopists, improving their ability to solve engineering related problems. To achieve this goal we provide a series of applicationbased short course lectures with intensive hands on labs, primarily to engineers and technicians in industry. To improve participant's skills, specimens are selected that have sufficient complexity and artifacts to make interpretation of results difficult enough that students will question results. A portion of the labs are reserved for analyzing samples provided by the participants. Often participant samples are not optimal, so we help students find better preparation and analytical conditions. We find personal connection to the samples improves student engagement and helps professionals acquire new skills that immediately benefit the student and company. Positive results can be demonstrated by students identifying specific improvements that increase throughput, accuracy or improved analytical capabilities after taking the SEM course. Partnering with Saturday Academy, we offer education outreach classes for high school students. Microscopy of common objects is an excellent introduction to scientific discovery. Students make discoveries deepening both their knowledge and curiosity. Project Micro [2] has proven methodologies for reaching large numbers of young students at low cost using optical microscopes and hand lenses. High school students are ready to be fascinated by the capabilities and complexities of SEM with compositional analysis. Generally high schools do not have the resources and expertise to operate or sustain a high school electron microscopy lab. However, there are notable exceptions such as Bergen County Academies NISL [3] and Project Nano [4]. Optimally outreach occurs in a university research lab, but table-top electron microscopes are an alternative that can reach more students. We have found four critical components for high impact secondary education outreach: personal connection to the samples, contributing to important real world problem solving, personal ownership Microsc. Microanal. 21 (Suppl 3), 2015 482 and presentation of results, and discussion of careers with passionate professionals. Personal connection to samples: Through a few iterations of microscopy classes, we have found that students are highly motivated by samples that they have a connection with (e.g., hair from their own pets). We compared the results from two summer classes: one class studied pollens that were selected by the instructor to be `photogenic' and the other class brought samples from home. All students were excited by doing hands-on microscopy. However, student engagement with the data from samples they had selected was much higher. Students willingly did extra research to gain more understanding of details seen in the images and created long lists of follow-on work. Contribution to important and real world problem solving: As the students measure their own particles, we present data about EPA particulate matter classifications and the known health risks. We show a map of EPA- particle collection sites around the Portland area. For composition, we present data on how lead is detrimental to neurological development and a brief history of lead in paint and gas. This is presented as a scientific success, how scientists tracked down the roots of the problem, policy was changed and tested blood levels of lead in children dropped dramatically. Students connected their personal samples to the larger community and to the real world problem of lead exposure, increasing student engagement. Individual presentation of results: The ability for each student to summarize his or her findings at the end of the class is a key element of learning. Presentation of results encourages students to take ownership of data and learning. As each student describes their "top 3" particles, they are entered by the instructor into the database of Portland particles. Students leave the class knowing that their work contributed to a public forum which increased their engagement and personal ownership. Discussion of careers with professionals: It is important to have both female and male experts tell their personal career stories. Students can more easily imagine themselves as a scientist or engineer if they meet one. This type of intensive outreach can influence students at a time they are thinking of educational and career options. Figure 1 shows some of the free-form feedback of student's comments about the impact of our program. References [1] C. Lyman Microscopy and Microanalysis, 19 (Suppl. 2) (2013), 294-295. [2] http://www.microscopy.org/education/projectMICRO/index.cfm [3] http://research.bergen.org/index.php/nsil-lab-tour [4] http://www.microscopy.org/MandM/2010/cady.pdf [5] The authors thank Saturday Academy http://www.saturdayacademy.org/, Center for Electron Microscopy and Nanofabrication htpp://www.pdx.edu/cemn, for enabling these outreach activities.");sQ1[242]=new Array("../7337/0483.pdf","Direct Observations of Photoexcitation Induced Dynamics of Charge Density Wave and Charge-Orbital Ordered State Using 2.8 MeV Ultrafast Electron Diffraction","","483 doi:10.1017/S1431927615003219 Paper No. 0242 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Direct Observations of Photoexcitation Induced Dynamics of Charge Density Wave and Charge-Orbital Ordered State Using 2.8 MeV Ultrafast Electron Diffraction Yimei Zhu1, P. Zhu1,2, L.Wu1, J. Li1, T. Konstantinova1, J. Cao3, Y. Hidaka1, J.P. Hill1 and X.J. Wang1,4 1. 2. Brookhaven National Laboratory, Upton, NY 11973 USA Shanghai Jiao Tong University, Shanghai 200240 China 3 Florida State University, Tallahassee, FL 32310 USA 4 SLAC National Accelerator Laboratory, Menlo Park, CA 94025 USA The recent development of ultrafast imaging and diffraction opened a new frontier for studying structural dynamics at nanoscales. It has become one of the key directions for future electron microscopy. Ultrafast electron microscopy combines the superior spatial resolution of conventional electron microscope with short electron pulses enabling the detection of electronic and atomic motion on their natural time and lengths scales. Two paths have been pursued by modifying commercial TEMs. One is the multi-shot UEM [1] utilizing the single electron limit (one or few electrons per pulse) that circumvents the repulsive space charge problem but requires multimillion-shot accumulation to get a decent image thus suitable for repeatable dynamic events. The other is the single-shot DTEM [2] to achieve nm-s resolution. In parallel home-built ultrafast electron diffraction (UED) instruments have also been developed, especially the RF-based UED using accelerator technology that enabled to developing x-ray free electron lasers. Among the state-of-the-art electron sources, only photocathode RF guns are capable of generating >107 electrons per pulse, multi-MeV energy and sub ps-long electron beams. The distinct merit of photocathode RF guns is the very high acceleration gradient (>100MV/m), over two-order of magnitude higher than 50KeV DC-guns [3]. It enables to minimize considerably the space charge effects, which can be scaled as inversely proportion to the product of velocity square and energy cube of the electrons, thus yielding much higher current density and beam brightness as recently demonstrated at Brookhaven [4,5]. Here we present two case studies on photoinduced structural dynamics of the charge-density wave (CDW) state in 2H-TaSe2 [4] and the charge-orbital ordered state in bi-layered LaSr2Mn2O7 at 77K using MeV-UED. By simultaneously tracking both the melting of the periodic lattice distortion (PLD) associated with the CDW and the lattice heating, following an impulsive photoexcitation, the separate contributions of electronic excitation and lattice thermalization to the melting process are disentangled in the time domain. Distinct time-constants, reflecting the corresponding individual dynamics of the electronic and lattice systems, are observed. Our results demonstrate, for the first time in 2H-TaSe2, that the PLD is first suppressed promptly by the electronic excitation and scattering, and then subsequently by lattice thermalization through electron-phonon interaction, on a much longer time scale, which leads to the final, full melting of the PLD. For LaSr2Mn2O7 our time-resolved electron diffraction study reveals the dynamic path of atoms and ions and indicates that the relative intensity change before and after the photoexcitation of the charge-order (CO) superlattice peaks mostly results from the orbital order (OO). Among various phonon modes Jahn-Teller distortion dominates the OO process and the CO plays a minimal role during the electronic excitation. Our study shed light on the various and often intertwined degrees of freedom and their responses to external perturbations in strongly correlated electron systems [6]. References: [1] D.J. Flannigan et al, PNAS, 107 9933 (2010). [2] J.S. Kim et al, Science 321 1472 (2008). [3] M. Eichberger et al, Nature 468, 799 (2010) [4] P. Zhu et al, Appl. Phys. Lett., 103 071914 (2013). [5] P. Zhu et al, "Femtosecond Time-resolved MeV Electron Diffraction", submitted. [6] The authors would like to thank S.W. Cheong, J. Geck, S. Pjerov, T. Ritschel, H.Berger, Y.Shen, and R.Tobey for providing samples and assistance. Work supported by the U.S. DOE, under contract No. DE-SC0012704. Microsc. Microanal. 21 (Suppl 3), 2015 484 Fig. 1. Schematic of the relativistic MeV-UED (ultrafast electron diffraction) set-up at Brookhaven national Laboratory. The entire system is ~4.5 m long. UV photons from a Ti-sapphire laser are used to generate electrons in the RF gun. A solenoid magnet focuses the beam onto the detector screen 4 m downstream of the sample. NearIR pulses from the same Ti-sapphire laser are used to optically pump the sample. The instrument has a cryogenic capability and can operates at 2-4MeV with 106 electron per pulse to achieve 120fs temporal resolution. 77K 200 shots 77K 100 shots 1 110 Fig. 2 Photoexcitation induced MeV UED diffraction data from: (left panel) Charge density wave (CDW) state in 2H-TaSe2; (right panel) Charge ordered state (marked by the red circle) and the orbital ordered state (marked by the green circle) in LaSr2Mn2O7. (center panel): Temporal evolution of the CDW peak (a) and the Bragg peak and diffuse scattering intensity (b) in 2H-TaSe2. For clarity, the change of diffuse scattering intensity is multiplied by 3. The pump was a 795 nm optical pulse at a fluence of 1.4mJ/cm2. The solid lines with given time constants are fits to the experimental data using an exponential function.");sQ1[243]=new Array("../7337/0485.pdf","Pump-Probe X-ray Holographic Imaging of Dynamic Magnetization Processes Down to the Femtosecond Timescale","","485 doi:10.1017/S1431927615003220 Paper No. 0243 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Pump-Probe X-ray Holographic Imaging of Dynamic Magnetization Processes Down to the Femtosecond Timescale Stefan Eisebitt1,2,3 1. 2. Institut f�r Optik und Atomare Physik, Technische Universit�t Berlin, 10623 Berlin, Germany Div. of Synchrotron Radiation Research, Lund University, SE-221 00 Lund, Sweden 3. Joint Research Group Functional Nanomaterials, Helmholtz Zentrum Berlin, 14109 Berlin, Germany X-ray holography as an approach for nanoscale imaging is experiencing a boost facilitated by the possibility to realize Fourier transform holography geometries via lithographic masks [1] in conjunction the increasing availability of x-ray sources with appreciable coherent photon flux, namely: accelerator based free-electron lasers and 3rd generation synchrotron sources as well as laser driven sources such as based on high harmonic generation (HHG). All these sources are intrinsically pulsed with pulse durations ranging from picoseconds to attoseconds. With the ability to excite resonant electronic transitions from core to valence levels and the resulting detailed spectroscopic information which can be used as a contrast mechanism for imaging, time resolved spectro-holographic studies with resolution down to nanometers and femtoseconds have become possible. In particular, studies of magnetization dynamics on the nanoscale are feasible via x-ray holography exploiting x-ray magnetic circular dichroism (XMCD) contrast [1,2]. In fact, so far the only spatially resolved magnetization maps recorded at free-electron x-ray lasers were recorded via holography, exploiting the high coherent photon number per pulse for fs single shot imaging [3]. Here, on behalf of the authors in Refs [4] and [5], I report on recent advances in combining x-ray holography with XMCD contrast for the study of dynamic magnetization rearrangements on the nanoscale with temporal resolution in the ps and fs regime via repetitive pump-probe experiments. The experiments are carried out via mask-based Fourier Transform x-ray holography (FTH) in transmission as described in Refs.[1,6] Magnetic field induced dynamic in a thin magnetic film with perpendicular anisotropy patterned in a 550 nm disk was studied at BESSY II in single bunch mode using soft x-rays at the Co L3 edge for XMCD contrast [4]. The magnetic sample and holographic mask were monolithically integrated, together with a micro coil allowing to apply a pulsed magnetic field via a pulsed current.[7] The rigid integration of the sample with the critical imaging components (i.e. object aperture and reference aperture) as well as the micro coil leads to insensitivity against vibration and drift, resulting in the very high real space tracking accuracy of about 3 nm shown below. The multilayer film with composition Pt(2)/[Co68B32(0.4)/Pt(0.7)]30/Pt(1.3) (thickness in nm) was developed for low pinning center density [8] to facilitate the observation of deterministic domain wall motion not dominated by stochastic movements from one pinning site to the next. In Fig. 1(left) we show a SEM micrograph of the sample structure. A static magnetic field was applied via an in vacuum quadrupole magnet in order to bring the magnetic domain configuration within the disk to a point where a bubble domain exists. Such a bubble has the topology of a skyrmion (with skyrmion number 1) and the study of its dynamics when displaced from its equilibrium position was the goal of this work. The displacement from the equilibrium position was accomplished via a bipolar magnetic field pulses, which reproducibly nucleate and subsequently annihilate an additional magnetic domain. The dipolar interactions with this domain lead to the skyrmion displacement. In Fig. 1(right) we present the skyrmion trajectory, recorded with 50 ps (sigma) temporal resolution. Note that while the spatial resolution in the hologram reconstructions is 40 nm, the tracking accuracy is about 3 nm (depending on data point), reflecting the accuracy with which we can Microsc. Microanal. 21 (Suppl 3), 2015 486 determine the center of magnetization of the skyrmion based on the repetitive pump-probe measurements with good statistics. This is possible due to the extraordinary stability and drift insensitivity of monolithic FTH holography. The analysis of the trajectory allowed us to conclude that the skyrmion has to be described with an inertial mass, and to determine the lower limit of this mass [4]. Results from a pump probe FTH experiment in the context of ultrafast optical demagnetization are shown in Fig. 2. [5] Here, an infrared laser pulse was used to excite a perpendicular anisotropy multilayer film with composition Pd(2) [Co(0.4)Pd(0.2)]20 Al(3). The optical excitation was localized within the elliptical object hole (defining the field of view) via suitable reflection from the elliptical aperture wall and interference. This region is clearly visible at a pump-probe delay of 600 fs at the top of the elliptical field of view, as the magnetic domain contrast, representing up/down magnetized regions, is washed out due to transient demagnetization. At different delays, a spatio-temporal evolution of this region is observed. These non-local changes in the demagnetization region are consistent with mechanisms including ultrafast transport of spin-polarized electrons during the demagnetization process. This experiment was carried out at the DiProI endstation at the FERMI free-electron laser at the Co M edge. Note that despite the large wavelength of 20.8 nm, high resolution imaging is possible [2], which is of particular interest for use with coherent laboratory sources in the near future. Figure 1. Left: Sample for FTH after magnetic field pump. Right: Skyrmion trajectory determined from pump-probe FTH. Time scale ranges from zero (blue ) to 14 ns (red). [4] Figure 2. FTH imaging of optically induced demagnetization. Left: unpumped sample, elliptical field of view. Black and white contrast indicates magnetization pointing up/down. Right: local pump at top (red arrow), imaged at 600 fs delay [5] References: [1] S. Eisebitt et al., Nature 432, 885 (2004). [2] S. Schaffert et al., New J. Phys. 15, 093042 (2013). [3] T.H. Wang et al. Phys. Rev. Lett. 108, 267403 (2012). [4] F. B�ttner et al, Nature Physics, advance online publication DOI: 10.1038/nphys3234 (2015) [5] C. von Korff Schmising et al., Phys. Rev. Lett. 112, 217203 (2014) [6] W.F. Schlotter et al., Appl. Phys. Lett. 89 (2006). [7] F. B�ttner et al., Opt. Express 21, 30563 (2013). [8] F. B�ttner et al., Phys. Rev. B 87 (2013).");sQ1[244]=new Array("../7337/0487.pdf","Probing Ultrafast Carrier Dynamics by Laser-Combined STM","","487 doi:10.1017/S1431927615003232 Paper No. 0244 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Probing Ultrafast Carrier Dynamics by Laser-Combined STM Hidemi Shigekawa*, Shoji Yoshida and Osamu Takeuchi Faculty of pure and applied sciences, University of Tsukuba, Tsukuba, 305-8573, Japan With the progress of science and technology, we are now facing limitations of current technologies that cannot be overcome by the simple expansion of existing engineering concepts. Thus, the introduction of new concepts based on new ideas, as well as the discovery of new physical properties and their applications, is a key factor in breaking this deadlock. However, for example, the technology node of semiconductor devices has become as small as 20-30 nm, and the fluctuations in the spatial distribution of dopants, which are introduced to control the properties of devices, and in the nanostructured interface between dissimilar materials directly affect the device functions obtained. Charge transfer in composite materials governing, for example, phase transitions, photoelectric conversion and catalysis is generally over the picoseond-to-subpicosecond range. Phonon-related phenomena are much faster. Namely, with the miniaturization of functional devices consisting of composite materials, quantum dynamics, which has been analyzed by techniques providing spatially and/or temporally averaged information, does not provide a sufficiently detailed description for the analysis and design of macroscopic functions. Therefore, the development of a method for exploring the ultrafast transient dynamics in small organized structures with high spatial resolution is a key factor for further advances in current science and technology. Scanning tunneling microscopy (STM) is one of the most promising techniques for the analysis of such properties because of its high spatial resolution. However, since its temporal resolution is low, the development of STM and related techniques with high-timeresolution has been an attractive target since its invention [1]. We have been working on lasercombined STM [1-8 and references therein], and recently we have succeeded in developing a new microscopy technique by combining STM with ultrashort-pulse laser technology, which enables the visualization of ultrafast carrier dynamics up to the femtosecond range and even at the single-atomic level [5, 7, Fig. 1]. The combination of STM with optical technology has advantages to enable the analysis of photo-induced dynamics on the nanoscale as well as the realization of ultrafast timeresolved microscopy. In OPP-STM, a non-equilibrium carrier distribution is generated using ultrashort laser pulses and its relaxation processes are probed by STM using the OPP method realized in STM. Furthermore, with the development of a new modulation technique of circularly polarized light, detection of spin dynamics has been realized [9]. We have succeeded in observing spin dynamics in D'yakonov-Perel mechanism. The relaxation of spins optically oriented in single quantum wells formed by GaAs/AlGaAs was observed independently (Fig. 2). We have also succeeded in observing spin precession dynamics (Fig. 3). The spin precession in GaAs under magnetic field was successfully probed using tunneling current by the time-resolved STM. Because the nanoscale properties of materials differ from bulk structures due to various factors, the successive development of novel microscopy techniques based on new ideas and the complementary use of these microscopy techniques will greatly aid the further development of nanoscale science and technology [10]. * http://dora.bk.tsukuba.ac.jp/, e-mail: hidemi@ims.tsukuba.ac.jp Microsc. Microanal. 21 (Suppl 3), 2015 488 References: [1] Y. Terada, S. Yoshida, O. Takeuchi, and H. Shigekawa: J. Phys.: Condensed Matter 22 (2010) 264008. [2] Y. Terada, S. Yoshida, and H. Shigekawa: Advances in Optical Technology 2011 (2011) 510186. [3] Y. Terada, S. Yoshida, O. Takeuchi, and H. Shigekawa: Nature Photonics 4 (2010) 869. [4] S. Yoshida, Y. Terada, R. Oshima, O. Takeuchi, and H. Shigekawa: Nanoscale 4 (2012) 757. [5] S. Yoshida, Y. Terada, M. Yokota, O. Takeuchi, H. Oigawa, and H. Shigekawa: The European Physical Journal Special Topics 222 (2013) 1161-1175. [6] S. Yoshida, M. Yokota, O. Takeuchi, and H. Shigekawa: Appl. Phys. Exp. 6, (2013) 016601. [7] S. Yoshida, O. Takeuchi, H. Oigawa, and H. Shigekawa: Appl. Phys. Exp. 6, (2013) 032401. [8] M. Yokota, S. Yoshida, Y. Mera, O. Takeuchi, H. Oigawa, and H. Shigekawa, Nanoscale 5 (2013) 9170. [9] S. Yoshida, Y. Aizawa, Z. Wang, R. Oshima, Y. Mera, E. Matsuyama, H. Oigawa, O. Takeuchi, and H. Shigekawa: Nature Nanotechnology 9 (2014) 588. [10] H. Shigekawa, S. Yoshida and O. Takeuchi: Nature Photonics 8 (2014) 815. Fig. 1 Fig. 2 Figure 1. Time-resolved spectra of hole capture process at single Mn/Fe atoms deposited on a GaAs(110) surface [5, 7]. Figure 2. Time-resolved spectra of spin dynamics in single GaAs/AlGaAs quantum wells [9]. Fig. 3 Figure 3. (a) Precession of optically oriented spins in GaAs measured with the newly developed optical pump-probe STM, and (b) local g factors obtained from the full series of the data [9].");sQ1[245]=new Array("../7337/0489.pdf","Advanced Analytical Electron Microscopy: New Perspectives on Real Materials","","489 doi:10.1017/S1431927615003244 Paper No. 0245 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Advanced Analytical Electron Microscopy: New Perspectives on Real Materials M. G. Burke1, S. J. Haigh1, J. J. H. Lim1, and A. Janssen1 1. Material Performance Centre and the Electron Microscopy Centre, School of Materials, The University of Manchester, Manchester (UK) The recent advances in XEDS afforded by new Si drift detector (SDD) geometries and new analytical transmission electron microscope platforms have enabled major progress in the characterisation of both conventional metals/alloys and complex materials with nanoscale features that have a dramatic effect on material performance. In particular, detailed microchemical analyses that had previously only been possible using techniques such as atom probe field-ion microscopy (for compositional and structural information) or 3D atom probe analysis (compositional data) are now very amenable to advanced Analytical Electron Microscopy (AEM) analysis. The increased solid angle for X-ray energy dispersive spectroscopy (XEDS) data collection has resulted in greatly reduced data acquisition times for spectrum imaging, enabling the detection of nm-scale segregation and precipitation. Thus, these new systems can be readily applied to address "conventional" materials issues such as precipitation behaviour, embrittlement and environmentally-assisted fracture. To demonstrate the improved capabilities for materials research using STEM-XEDS in new advanced AEMs, we have examined both conventional precipitation-hardened Ni-base Alloy 625, and a neutron-irradiated low alloy steel weld. An aberration-corrected FEI Titan G2 80-200 S/TEM with Super X (ChemiSTEMTM) and X-FEG operated at 200 kV was used in this study. All XED spectrum images were acquired with a beam current of 1 nA; all samples were conventional electropolished thin-foils. Alloy 625, with a nominal composition (wt.%) of 20 Cr � 10 Mo � 4 Fe � 3.5 Nb � 0.3 Ti � 0.3 Al � 0.05 C � bal Ni, is generally used in the aged (high strength) condition. During ageing in the temperature range ~650 � 800�C, DO22ordered " (Ni3(NbTi)) precipitates form, leading to the increased strength of the alloy. These " precipitates are disc-shaped and form so that the disc is parallel to {100} in the fcc matrix. The HAADF and XED spectrum images (acquired for 27 min) in Figure 1 clearly shows the 3 variants of " precipitates formed during ageing at 650�C for 100 h. Using HR-STEM, it was possible study the coherent (001) interface of the precipitate and visualize the pure Ni and mixed Ni-Nb planes in the ", as shown in Figure 2. The XED spectra - both sum spectrum and discrete spot analyses in the precipitates - demonstrated that there was detectable Al in the ". Such data are useful for alloy design/optimization. The improved solid angle of the Super X configuration also enabled the direct visualization of irradiation-induced Mn-Ni-enriched clusters in neutron-irradiated low alloy steel welds. These features are of great interest because they are associated with the irradiation-induced hardening and embrittlement that can degrade the toughness of the reactor pressure vessel of pressurized water reactors (PWRs). 3D-Atom Probe analysis is generally used to analyze such irradiation-induced features. The Titan "ChemiSTEM" SDD XED spectrum images for Ni and Mn in a 1.6 Ni � 1.4 Mn weld clearly show the Ni and Mn-enriched solute clusters formed during neutron irradiation. These examples highlight the wide applicability of the improved XEDS/microscope system, and will be discussed in this presentation. References: [1]The authors thank Matt Smith for technical assistance and support of the Titan G2 80-200 S/TEM provided by HM Government (UK). Microsc. Microanal. 21 (Suppl 3), 2015 490 a b c Fig. 1: TEM images of (a) the subgrain structure; (b) [001] BF and (c) DF images of the intragranular ". Fig. 1: (a) HAADF image and complementary (b) Nb and (c) Ti XED spectrum images of " precipitates in aged Alloy 625. All 3 variants of the disk-shaped DO22-ordered precipitates are visible in this [001]-oriented specimen. a b c Fig. 2: (a) HR-STEM image of " precipitate with {001} (disc) face parallel to {100}matrix. (b) HAADF and (c) complementary XED spectrum image of Ni and Nb showing the pure Ni layers in the ordered " precipitate. a b c Fig. 3: (a) HAADF and XED spectrum images for (b) Ni and (c) Mn showing the presence of the soluteenriched clusters formed during neutron irradiation. Preferential solute clustering/segregation also occurred at grain boundaries and dislocations.");sQ1[246]=new Array("../7337/0491.pdf","Grain Boundaries across Length Scales; Correlating SEM, Aberration-Corrected TEM Orientation Imaging and Nanospectroscopy","","491 doi:10.1017/S1431927615003256 Paper No. 0246 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Grain Boundaries across Length Scales; Correlating SEM, AberrationCorrected TEM Orientation Imaging and Nanospectroscopy W.J. Bowman1, A. Darbal2, M. Kelly3, G.S. Rohrer3, C.H. Hernandez1, K. McGuinness1, P.A. Crozier1 1. School for the Engineering of Matter, Transport and Energy, Arizona State University, Tempe Arizona 85287-6106, USA 2. AppFive LLC, Tempe, Arizona, USA 3. Materials Research Science and Engineering Center, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA 15213, USA Grain boundaries significantly influence the behavior of a huge range of materials; however, their small size (~1 nm structural width) and significant variety make them difficult to characterize, fully describe, and ultimately optimize. Statistical methods such as stereological analysis coupled with SEM orientation imaging via electron back scatter diffraction (EBSD)-- which typically samples >50,000 boundaries, are powerful techniques for quantifying microscopic grain boundary parameters such as the crystallographic plane distribution and misorientation texture [1]. Conversely, TEM and STEM atomic resolution imaging and spectroscopies have proven invaluable in elucidating nanoscale boundary properties including local atomic structure and elemental composition [2]. However, correlating such micro- and nano-scale observations has been challenging due to the difficulty of unambiguously determining the statistical relevance of grain boundaries observed in the TEM. Recently-commercialized precession-enhanced nanobeam diffraction (PEND) instrumentation and techniques have made possible routine orientation imaging in the aberrationcorrected TEM (AC-TEM) [3]. This mesoscale technique facilitates the location and identification of statistically relevant grain boundaries in a TEM specimen that can then be targeted for in-depth analysis via high resolution imaging and/or nanospectroscopy. This capability bridges the gap between microscopic SEM measurements and nanoscopic and ACSTEM characterizations to offer a more complete and contiguous understanding of grain boundary properties across a considerable range of length scales. Here, we employ a suite of techniques including SEM EBSD, AC-TEM PEND and ACSTEM EELS and EDX to correlate data sets acquired across micron, meso and nanometer length scales from ceria-based oxygen-conducting ceramics. We present findings and results based on stereological analysis performed to characterize microscopic grain boundary parameters derived from SEM EBSD data (fig. (a)) such as plane distribution and misorientation texture. We describe the use of these parameters, such as boundary misorientation angle distribution (e.g. fig. (b)), to relate data acquired from SEM and AC-TEM orientation imaging techniques. Figure (c) presents a representative orientation image acquired with an AC-TEM equipped with an ASTAR PEND system. We discuss correlation of this mesoscale boundary characterization with SEM EBSD results. Furthermore, we present image and nanospectroscopic data acquired from the PEND specimens using an AC-STEM equipped with EELS and EDX. Figure (d) shows ADF AC-STEM images of boundaries selected for analysis based on PEND results. The images presented here demonstrate the diversity in atomic structure found in the observed boundaries. AC-STEM EELS spectra acquired on and off the boundaries in fig. (d), at the locations highlighted, are presented in fig. (e). Energy-loss spectra show variation in cation concentrations Microsc. Microanal. 21 (Suppl 3), 2015 492 with distance from the boundary core, as well as changes in near-edge fine structure, indicating spatial dependence of the cation oxidation states likely stemming from local variations in oxygen non-stoichiometry. References and acknowledgements [1] L. Helmick et al. Int. J. Appl. Ceram. Technol., 8 [5] 1218-1228 (2011) [2] S.J. Pennycook & P.D. Nellist. Scanning Transmission Electron Microscopy, Springer [3] A.D. Darbal et al. Microsc. Microanal. 19, 111-119 (2013) [4] C.A.H. and K.M. wish to thank the Fulton Undergraduate Research Initiative at ASU for generous financial support throughout this work. W.J.B. would like to acknowledge the National Science Foundation's Graduate Research Fellowship (DGE-1211230) for continued financial support. Finally, we gratefully acknowledge support of NSF grant DMR-1308085 and ASU's John M. Cowley Center for High Resolution Electron Microscopy. Fig. a. SEM EBSD inverse pole figure of Ca-doped ceria. Fig. b. Misorientation angle distribution determined from SEM EBSD. Fig. c. AC-TEM PEND inverse pole figure of Gd/Pr-doped ceria. Fig. d. AC-STEM ADF images of grain boundaries selected from figure (d). Fig. e. AC-STEM EELS spectra acquired on and off of the grain boundaries in figure (e).");sQ1[247]=new Array("../7337/0493.pdf","Super-X EDS Characterization of Chemical Segregation within a Superlattice Extrinsic Stacking Fault of a Ni- based Superalloy","","493 doi:10.1017/S1431927615003268 Paper No. 0247 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Super-X EDS Characterization of Chemical Segregation within a Superlattice Extrinsic Stacking Fault of a Ni- based Superalloy Robert E. A. Williams1,2, Tim Smith1, Bryan D. Esser1, Nikolas Antolin1, Wolfgang Windl1, David W. McComb1,2, Michael J. Mills1, Hamish L. Fraser1 1 2 Department of Materials Science and Engineering, The Ohio State University, 477 Watts Hall, 2041 College Road, Columbus, OH 43210, USA Center for Electron Microscopy and Analysis, The Ohio State University, 1305 Kinnear Road, Columbus, OH 43212, USA Superalloys are essential materials for high temperature applications in aerospace and energy production. Improving the temperature capability of disk alloys by a modest 25�C could translate into approximately a 1% increase in aircraft engine efficiency, resulting in significant cost savings as well as benefit environmental impact by reducing carbon emissions. While superalloys are prime candidates for an ICME approach to accelerated alloy development, at present this approach is severely hampered by the lack of quantitative models that connect alloying effects to deformation mechanisms. This linkage is critical for developing quantitative, physics-based deformation models, but has been limited severely by the complexity of the alloys (typically 10 components or more). One aspect, critical to a successful model is accurate determination of the degree of segregation to defects, such as faults, in order to enable the identification of the compositional configuration of such defects. The examples presented in this work consist of stacking faults created by superpartial dislocation shearing of the ' precipitates during high temperature (760�C) creep in an advanced, polycrystalline Ni disk superalloy. Under the deformation conditions studied, conventional TEM defect analysis determined this as the primary mode of precipitate shearing and it is expected to be a critical, ratelimiting process. Recent analysis of these faults using a Super-X system without a probe corrector has revealed for the first time that there appears to be a chemical signature associated with these faults, as evidenced in Fig. 1(a)[1]. In this case, involving a two-layer, superlattice extrinsic stacking fault (SESF), Al appears to be deficient at the fault while the concentrations of Ti and Ta (and Cr, not shown) are enhanced at the fault. The information exhibited in Fig. 1(a) appears to indicate elemental segregation at the fault, however, the compositional details are not resolved on an atomic scale, which is necessary to develop a detailed understanding of the configuration of these defects for inclusion in a predictive model of deformation. The results shown in Fig. 1(b) were obtained using a DCOR probe corrected STEM and demonstrate remarkably the improved resolution for XEDS mapping; made possible with 5th order probe aberration correction and the increased collection efficiency and solid angle of the Super-X collection system of a FEI Titan ThemisTM. Features that are now characterizable for the first time in these [110] zone axis maps are: (a) discriminating the predominantly Ni (001) planes from the "mixed" Ni and Al planes, which also contain Ta and Ti in the "bulk"; (b) a slight reduction in the Ni content at the two-layer fault, occurring primarily on "mixed" (001) plane; (c) Ta and Ti replaces Ni and Al on the mixed planes; (d) a rectangular grid of columns with high Ta concentration corresponds to the high intensity columns in the HAADF images. These structural and chemical characteristics are consistent with the phase, which can form in superalloys and is deleterious in bulk form[2]. Preliminary first principles DFT calculations Microsc. Microanal. 21 (Suppl 3), 2015 494 indicate that a single unit cell width of phase having a composition estimated directly from the EDS data has a significantly lower energy than an SESF in the absence of segregation. The work to be presented demonstrates that, while indications of elemental segregation may be gleaned from experiments performed on an uncorrected STEM equipped with Super-X EDS collection, the atomic-scale, elemental information provided by 5th order probe aberration corrected STEM equipped with Super-X EDS has provided direct chemical information related to SESF's for the first time and provided crucial information towards more accurate modeling of segregation phenomena. Post processing and quantification of data will also be discussed, as this is a critical aspect of apparent, atomic-resolution XEDS data. Refereneces: [1] G. B. Viswanathan, et al., Scripta Materialia 94 (2015): p. 5-8. [2] E. J. Pickering, et al., Acta Materialia, 60(6-7), (2012): 2757�2769.");sQ1[248]=new Array("../7337/0495.pdf","Atomic-scale Dual-EELS/EDX Spectroscopy Applied to Rare-earth Oxide Superlattices","","495 doi:10.1017/S143192761500327X Paper No. 0248 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic-scale Dual-EELS/EDX Spectroscopy Applied to Rare-earth Oxide Superlattices P.J. Phillips1 , P. Longo2 , E. Okunishi3 , R.F. Klie1 Department of Physics, University of Illinois at Chicago, Chicago IL 60607 Gatan, INC, Pleasanton, CA 94588 3 EM Application Group, JEOL Ltd., 3-1-2 Musashino, Tokyo 198558, Japan 2 1 Rare-earth nickelates are known to display complex electronic and magnetic behaviors owed to a very localized and sensitive Ni-site atomic and electronic structure. Specifically, in the wake of recent predictions that nickelates can achieve an electronic structure which mimics that of the high-temperature cuprate superconductors, much research has been focused on manipulating the energetic ordering of Ni d orbitals and 2D conduction in these heterostructures [1,2]. Toward this goal, the present work focuses on the experimental characterization of thin film nickelate superlattice structures, while highlighting critical results which can only be attained by employing the most recent and advanced spectrometers, detector designs, and software. Specifically, the superlattice in question consists of alternating layers of LaTiO3 and LaNiO3 sandwiched between a dull insulator, LaAlO3 . Using aberration-corrected scanning transmission electron microscopy (STEM)-based methods, properties such as interfacial sharpness, electron transfer, O presence, and local electronic structure can be probed at the atomic scale, and will be discussed at length. As both energy dispersive X-ray (EDX) and electronic energy loss (EEL) spectroscopies are required, a JEOL JEM-ARM300CF (operated at 160 kV), equipped with dual large-angle EDX silicon drift detectors (SDDs) and a Quantum Gatan imaging filter (GIF) was used; of note is the attainable 0.35 eV energy and 100 pm spatial resolution in spectroscopy mode. Simultaneous acquisition of both signals allows one to bypass the difficult high-energy EELS edges (La, Ni, and Al in this case), favoring a higher energy dispersion, thereby relying on EDX to identify the remaining elements. The higher energy dispersion also allows for detailed fine structure analysis of relevant energy loss edges (Ti L and O K). All of these elements are critical to the above analysis. A subset of experimental results is presented in Figure 1 below. The repeated superlattice structure is barely visible in the annular dark field (ADF) STEM image, although with the simultaneous EELS/EDX acquisition of the Ni, Ti, and Al signals, the layers are easily identifiable. The color map was recorded on a similar area, and contains only EELS signals from Ni, Ti, Al, and La, demonstrating the large energy range possible with the Quantum GIF. Detailed EELS fine structure analysis will also be discussed in detail, specifically with regards to the O K and Ti L edges, as these can provide fingerprints to identify the local electronic structure, on a layer-by-layer (atomic) scale. The focus of the talk will remain not only on the aforementioned properties, but will also include details and parameters of the acquisitions to facilitate future characterization at this level. References [1] D.P. Kumah, A.S. Disa, J.H. Ngai, H. Chen, A. Malashevich, J.W. Reiner, S. Ismail-Beigi, F.J. Walker, C.H. Ahn, Adv. Mater. 26 (2014) 1935�1940. [2] H. Chen, D.P. Kumah, A.S. Disa, F.J Walker, C.H. Ahn, S. Ismail-Beigi, Phys. Rev. Lett. 110 (2013) 186402. Microsc. Microanal. 21 (Suppl 3), 2015 496 Figure 1: ADF STEM image spanning approximately five superlattice repeats in the horizontal (growth) direction, with various spectroscopic maps shown. The grayscale images resulted from a simultaneous EELS/EDX acquisition; with both the Ni and Al signals coming from EDX, and the Ti coming from EELS. The color map is from a similar region, but contains only EELS data, including Ni, Ti, Al, and La.");sQ1[249]=new Array("../7337/0497.pdf","Oxygen Intercalation and Doping-Induced 2D High-Tc Superconductivity at the CaCuO2/SrTiO3 Interface","","497 doi:10.1017/S1431927615003281 Paper No. 0249 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Oxygen Intercalation and Doping-Induced 2D High-Tc Superconductivity at the CaCuO2/SrTiO3 Interface Claudia Cantoni1, Daniele Di Castro2 and Giuseppe Balestrino2 1. Materials Science & Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN 378316056, USA 2. Dipartimento di Ingegneria Civile e Ingegneria Informatica, Universit� di Roma Tor Vergata, Via del Politecnico 1, I-00133 Roma, Italy Energy consumption in high-performance computing is nowadays a serious issue, with data centers consuming more than 2% of the world energy and growing. As semiconductor technology is reaching the critical integration density, supercoducting devices with high switching speed and lossless interconnects are regarded as a possible answer to this problem. At the same time, the interface between complex oxides (e.g., the LAO/STO interface) is emerging as a very promising electronic system because it hosts many fascinating electronic phenomena that are absent in its original constituents. Superconductivity is one of such phenomena but until now only with a very low critical temperature (Tc 100 mK). Here, by means of aberration-corrected STEM and EELS, we demonstrate doping-induced high-Tc superconductivity with Tc = 50 K at a single interface plane between the insulating oxides CaCuO2 (CCO) and SrTiO3 (STO). Our results show that it might be possible to isolate a single CuO2 plane with high-Tc superconductivity for realization of 2D superconducting field effect devices operating at a temperature much higher than the LAO/STO interface. [(CaCuO2)n/(SrTiO3)m]N superlattices made by N repetitions of the (CCO)/(STO) interface (n and m being the number of unit cells of CCO and STO, respectively), and STO/CCO/STO trilayers were grown by PLD using different oxidizing conditions on NdO terminated, (110) single crystal NdGaO3 (NGO). The superlattices grown in highly oxidizing conditions showed Tc = 50 K [1]. Earlier studies had suggested a superconductivity mainly confined at the interface between CCO and STO [2]. Moreover, R(T) measurements on NGO/CCO/STO and NGO/STO/CCO samples had shown that only the NGO/CCO/STO sample, having the interface stacking CuO2-Ca-TiO2-SrO hosts superconductivity. By analogy with cuprate superconductors, we surmise that superconductivity arises as extra oxygen ions are incorporated in the interface Ca plane, acting as apical oxygen for Cu and providing holes to the CuO 2 planes. The fact that only in the CuO2-Ca-TiO2-SrO interface the Ca plane can host extra oxygen can be explained considering that 1) CaTiO3 is a stable phase, and 2) in the non-superconducting interface TiO2-SrO-CuO2-Ca, the SrO plane is already stoichiometrically full of oxygen ions; therefore, no doping can occur. Here we prove this hypothesis by using annular bright field (ABF) imaging, a technique for imaging light elements, and atomically resolved EELS at the O-K edge. Figure 1a shows an ABF image of the CCO/STO interface with CuO2-Ca-TiO2-SrO stacking. The image has been inverted to display the atomic columns as bright on a dark background. The position of the interface was identified by simultaneously acquiring the high angular annual dark field (HAADF) image, in which the intensity depends nearly on Z2 and thus can be used to identify cation positions. In addition, spectrum images were taken near the interface and the Ti-L2,3, Ca-L2,3, and Cu-L2,3 integrated intensities plotted as shown in Fig. 1b.The upper arrows in Fig. 1a indicate the interfacial CaOx plane, where O columns are clearly seen between the brighter spots indicating the Ca columns. This is also seen in the intensity profile for this plane shown in Fig. 1c (upper panel). The lower panel of Fig. 1c shows the intensity profile taken on Microsc. Microanal. 21 (Suppl 3), 2015 498 the second Ca plane from the interface. For this plane, both image and line profile show a dramatic reduction of O. The subsequent planes as well as the Ca plane at the STO/CCO interface with stacking TiO2-SrO-CuO2-Ca, do not show any detectable amount of O. As documented by x-ray absorption spectroscopy (XAS), the doping holes in cuprate superconductors show a specific signature in the O-K edge. Upon doping the parent compound, a pre-edge feature (peak A) emerges in the O-K at around 529 eV, while the prepeak around 532 eV (peak B) decreases in amplitude. Peak A is associated to the low-energy quasi-particle band called Zhang-Rice singlet: a locally bound d9 copper 3d hole hybridized with a doped ligand hole distributed on the planar oxygen 2p orbitals. Peak B instead, is associated with the upper Hubbard band and results from a 3d9->1s3d10 transition of the undoped material [3]. Figure 1d compares the EELS at the O-K edge acquired within the same spectrum image across the trilayer sketched in the inset. The curves are spectra well within the STO (green, dashed line), at the second Ca plane from the interface (blue, solid line), and well within the CCO layer (red, dotted line). We find that peak A (the doped hole concentration) decays with a characteristic length of 1-2 u.c. from the interface and is absent at the non-superconducting STO/CCO interface, for which ABF images do not show intercalated extra O [4]. References: [1] D. Di Castro, et al., Supercond. Sci. Technol. 27, (2014) 044016. [2] D. Di Castro, et al., Phys. Rev. B 86, (2012) 134524. [3] C. T. Chen, Phys. Rev. Lett. 68, (1992) 2543. [4] Research supported by the U.S. Department of Energy, Office of Science, Basic Energy Sciences, Materials Sciences and Engineering Division. Figure 1. Imaging and spectroscopy of the CCO/STO interface. a) Inverted ABF image. b) Ti, Ca, and Cu EELS maps. c) Intensity profiles of the CaOx plane at the location of the arrows in a). d) O-K edge of STO (dashed line), interface CaOx plane (solid line), and bulk CCO (dotted line).");sQ1[250]=new Array("../7337/0499.pdf","Mapping Magnetic Properties of Materials At Atomic Spatial Resolution","","499 doi:10.1017/S1431927615003293 Paper No. 0250 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Mapping Magnetic Properties of Materials At Atomic Spatial Resolution Jan Rusz1, Juan-Carlos Idrobo2, Alexander Edstr�m1, Jakob Spiegelberg1, and Somnath Bhowmick3 1. Department of Physics and Astronomy, Uppsala University, Uppsala, Sweden Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, TN, USA 3. Indian Institute of Technology, Kanpur, India 2. Quantitative characterization of magnetic materials at atomic resolution is one of important challenges for the aberration-corrected scanning transmission electron microscopy (STEM). The ability to measure spin and orbital magnetic moment with atomic spatial resolution at surfaces, across interfaces, or nearby defects, would provide unprecedented new insights into the magnetism at the shortest scale. A candidate experimental technique to achieve this is called electron magnetic circular dichroism (EMCD; [1]). EMCD is analogous to the well established x-ray magnetic circular dichroism, which allows to quantify spin and orbital magnetic moments from spectra measured at core level edges of the magnetic elements. In recent years, several research groups worldwide are investing substantial effort into high spatial resolution EMCD. This has been largely motivated by the introduction of electron vortex beams (EVBs) [2,3,4] into the world of transmission electron microscopy (TEM). It has been suggested [3] that EVBs will allow measurement of EMCD directly at the transmitted beam. If confirmed, this would bring an enormous improvement of signal to noise ratio compared to the classical EMCD, where spectra are acquired in between the Bragg spots, using electron beams with small convergence angles [5]. However, it was shown recently [6], that to acquire EMCD at the transmitted beam requires that the size of the EVB must be small enough to allow for atomic resolution. Expressed in terms of convergence angles and resulting convergent beam electron diffraction (CBED) pattern, the convergence angle must be large enough so that the disks in the CBED pattern overlap. This is analogous to the elastic scattering case, where an interference of overlapping CBED disks is a necessary condition for atomic resolution STEM images [7]. An interesting corollary has been deduced on the basis of arguments of the coherence and symmetries present in the diffraction patterns: a nonzero EMCD should be detectable in the direction of the transmitted beam not only for vortex beams, but also for beams distorted by aberrations of suitable symmetry. For example, for tetragonal or cubic crystals one can utilize a nonzero antisymmetric fourfold astigmatism, i.e., C3,4b [6]. Similar symmetry considerations also provided a qualitative explanation, why electron vortex beams formed with spiral aperture do not lead to observable EMCD signal, despite that they can be rather easily focused to an atomic size, with full-width half-maximum sizes of 3 � or less [8]. With a clean vortex probe of sufficiently small size it should be possible to characterize magnetic moments at column-by-column basis. An example is shown in Fig. 1 summarizing simulated EMCD spectrum images for antiferromagnet LaMnAsO with a checkerboard pattern of magnetic ordering [9]. Simulations show that with a vortex of diameter 1.5 � one should be able to resolve differences of the order of 3-4% at the Mn-L2,3 edges, providing an EMCD signal from an antiferromagnetic compound � a feature that is out of reach in X-ray based techniques. Notice the doughnut shaped spots with a minimum at the position of the Mn atomic columns, when the probe is a vortex beam. Microsc. Microanal. 21 (Suppl 3), 2015 500 Experimental realization of atomic resolution EMCD has not been published yet, however the first measurements with aberrated probes show very promising results [9]. In the analysis of experimental data one faces a problem of low signal to noise ratio � individual spectra in atomic resolution spectrum images measured at energy losses of several hundreds of electron-Volts are rather noisy. In such case the extraction of EMCD signal will profit from advanced statistical data processing techniques, such as blind source separation methods, image registration and unwarping or machine learning approaches. References: [1] P. Schattschneider et al., Nature 441, 486 (2006). [2] M. Uchida, and A. Tonomura, Nature 464, 737 (2010). [3] J. Verbeeck, H. Tian, and P. Schattschneider, Nature 467, 301 (2010). [4] B. McMorran et al., Science 331, 192 (2011). [5] H. Lidbaum et al., Phys. Rev. Lett. 102, 037201 (2009). [6] J. Rusz, J. C. Idrobo, and S. Bhowmick, Phys. Rev. Lett. 113, 145501 (2014). [7] S. J. Pennycook, and P. D. Nellist, Scanning Transmission Electron Microscopy: Imaging and Analysis (Springer, New York, 2011), Chap. 2. [8] D. Pohl et al., Ultramic. 150, 15 (2015). [9] J. C. Idrobo et al., submitted. [10] This research was supported by the Swedish Research Council and STINT (J.R., A.E, J.S., S.B), and by the Center for Nanophase Materials Sciences (CNMS), which is sponsored at Oak Ridge National Laboratory by the Scientific User Facilities Division, Office of Basic Energy Sciences, U.S. Department of Energy (J.C.I.). Figure 1. Simulated Mn-L3 edge STEM spectrum images (nh) and pure magnetic signal component (sz) for an antiferromagnet LaMnAsO, assuming a probe with OAM=0 (left), +1 (middle) and -1(right) with corrected aberrations, 100keV acceleration voltage, 35mrad convergence angle and 45mrad collection angle. Sample thicknesses of 10nm, 15nm or 20nm have been considered.");sQ1[251]=new Array("../7337/0501.pdf","At-Focus Observations of High Quality Electron Vortex Beams Created from Ferromagnetic Rods","","501 doi:10.1017/S143192761500330X Paper No. 0251 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 At-Focus Observations of High Quality Electron Vortex Beams Created from Ferromagnetic Rods Arthur M. Blackburn1, James C. Loudon2, Rodney Herring3, Ales Hrabec4, David Hoyle5 1 2 Hitachi Cambridge Laboratory, Hitachi Europe Ltd., Cambridge, U. K. Department of Materials Science and Metallurgy, University of Cambridge, Cambridge, U. K. 3 Department of Mechanical Engineering, University of Victoria, Victoria, BC, Canada. 4 School of Physics and Astronomy, University of Leeds, Leeds, U.K. 5 Hitachi High-Technologies Canada, Toronto, Canada. Electron vortex beams carry orbital angular momentum and have use in electron microscopy for characterizing the magnetic and chiral properties of materials [1, 2]. For these applications it is preferable to have a single isolated vortex beam to simplify the interpretation of scanning probe measurements. However, the first-used fork-like grating masks for producing vortex beams [1] produce multiple beams off-axis and thus require an additional selection aperture, ideally accompanied with additional lenses or a monochromator, to produce a single beam for scanning [3]. To avoid this requirement for additional hardware and to produce brighter vortex beams, thus increasing the collected dichroic signal strength, alternative approaches to producing vortex beams on-axis are being investigated, including: continuous thin-film phase plates; spiral-like apertures; carefully tuned electron microscope aberrations in combination with an annular aperture; and using a narrow ferromagnetic rod, magnetized along its long axis, partially inserted into the beam path [4, 5]. The ferromagnetic rod based approach (Fig. 1(a)) offers the advantage of producing a beam-energy independent, single electron vortex beam on-axis with high intensity. However, a potential perceived disadvantage that is experimentally addressed here is the quality of the vortex beam at focus which has not been clear in earlier observations [4], where the apparent quality of the produced beam may have been limited by the imperfect spiral-like nature of the produced phase-shift or other artefacts such as charging and dispersion in structures surrounding the rod. Other evidence of vortex beams produced by ferromagnetic rods has used defocussed probe observations, with this condition alleviating the obscuring of the beam point spread function by the finite coherence of the illumination, or equivalently by convolution with the electron source image, while also allowing beam rotation observation [5]. In this work, cantilevered ferromagnetic rods were formed by sputter deposition of 10 nm Co50Fe50 with Ta adhesion and capping layers, upon one side of a series of micro-machined cantilevers which were subsequently coated in Au (Fig. 1(b)). Following magnetization in an external magnetic field, electron holograms were collected in the region of the rod-tip to determine the form of the electron phase-shift produced by the stray magnetic field and the flux-contained in the rod. From this we found that though not all the rods had the desired single domain arrangement and related helical phase shift, which could likely be remedied by improved fabrication processes, the majority of rods behaved in the desired manner and proved able to produce vortex electron beams. With the sample rods in a standard transmission electron microscope sample holder, a 10 �m diameter selected area aperture was centered upon a single rod-end, employing conventional low magnification imaging (i.e. with the objective lens off) and images were collected of the focused beam, occurring about the imaging-lens diffraction plane. Images of the beam at its minimum diameter `waist', for rods Microsc. Microanal. 21 (Suppl 3), 2015 502 determined from electron holography to contain a magnetic flux nh/e with n near 2, 3 and 4 (Figs. 2(a � c)), show the vortex nature of the beam, with a clear central minimum through focus. The average radial intensity profile of these beams (Fig. 2(d)) shows fair agreement with a simple aberration-free model, once finite electron source size effects are included. Adding further aberrations would enable a better model fit. The source size, or the related finite coherence of the electron illumination over the region of the rod, manifest in these observations, need not dominate in higher resolution probes when greater source demagnification can readily be employed with high-brightness electron sources. Indeed, the observations presented here, which directly show that it is possible to achieve high quality vortex beams at focus with a ferromagnetic rod, indicate that it is worthwhile to investigate such arrangements to further the application of electron vortex beams for characterizing and understanding magnetic and chiral material behavior. [1] J Verbeeck, H Tian, and P Schattschneider, Nature, 467 (2010), p. 301. [2] A Asenjo-Garcia and F J G de Abajo, Physical Review Letters, 113 (2014), p. 066102. [3] O L Krivanek, J Rusz, J C Idrobo, et al., Microscopy and Microanalysis, 20 (2014), p 832. [4] A M Blackburn, J C Loudon, Ultramicroscopy, 136 (2014), p. 127. [5] A B�ch�, R Van Boxem, G Van Tendeloo, et al., Nature Physics, 10 (2014) p. 26. Figure 1. Schematic arrangement of (a) the ferromagnetic rod and aperture used to create an electron vortex beam, and (b) the rod and its cross-sectional composition. Figure 2. At focus images of 300 keV vortex beams produced from ferromagnetic rods containing a magnetic flux of ~nh/e with n = 2, 3, and 4 (a, b, and c) and the (d) average radial intensity profiles (solid-lines) and simple aberration free modelled intensities (dashed-lines) with source convolution.");sQ1[252]=new Array("../7337/0503.pdf","Electron holograms encoding amplitude and phase for the generation of arbitrary wavefunctions","","503 doi:10.1017/S1431927615003311 Paper No. 0252 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electron holograms encoding amplitude and phase for the generation of arbitrary wavefunctions Vincenzo Grillo,1,2,3 Ebrahim Karimi,4 Roberto Balboni5, Gian Carlo Gazzadi,1 Federico Venturi6, Stefano Frabboni,1, 6 Jordan S Pierce3, Benjamin J McMorran3 and Robert W Boyd4,7 1. 2. CNR-Istituto Nanoscienze, Centro S3, Via G. Campi 213/a, I-41125 Modena, Italy CNR-IMEM Parco Area delle Scienze 37/A, I-43124 Parma, Italy 3. Department of Physics, University of Oregon, Eugene, 97403-1274 Oregon, USA 4. Department of Physics, University of Ottawa, 25 Templeton, Ottawa, Ontario, K1N 6N5 Canada 5. CNR-IMM Bologna, Via P. Gobetti 101, 40129 Bologna, Italy 6. Dipartimento FIM, Universit� di Modena e Reggio Emilia, Via G. Campi 213/a, I-41125 Modena, Italy 7. Institute of Optics, University of Rochester, Rochester, New York 14627, USA Since the introduction of vortex beams it has become clear that it is possible to structure the electron beams and produce arbitrary electron phase profiles [1][2]. We also demonstrated that the control over the phase in some cases like Bessel beams is sufficient to shape the overall amplitude profile for example by controlling the zeros of the electron wavefunction [3]. However in many cases the control of both amplitude and phase is demanded to engineer more structural properties of the electronic beam. We report for the first time the electron holograms for phase and amplitude manipulation. The hologram we present is based on a modification of blazed phase holograms where the groove depth is modulated to produce the desired amplitude effect. The same philosophy can be applied to any initial groove profile. Calculations based on Fourier-Taylor expansion demonstrate the appropriate form of the transmission function has to be T = exp(iM ( x, y ) Mod ( F ( x, y ) + 2x,2 )) with 1 M ( x, y ) = 1 + sinc -1 ( A) , F ( x, y ) = P - M where A and P are the aimed phase and amplitude distribution at the hologram plane, Mod(a,b) is the remainder of division of a by b and sinc-1 is the inverse of the function sinx/x. Fig 1 a,b show one example of FIB nanofabricated hologram encoding phase and amplitude information for a non-trivial beam. The aimed pattern calculated with the equation above is shown in fig 1a while the thickness map of the SiN membrane after FIB patterning is shown in fig 1b. The thickness measurement have been carried on by EFTEM in a JEOL 2010 equipped with LaB6 emitter and operated at 200 keV. The aimed beam is given by the superposition of two Laguerre Gauss beams. If we indicate Ll, p the Laguerre-Gauss with OAM quantum number and radial number p, the aimed beam is = 22 (L-3, 0 + L8, 2 ) This pattern has been chosen as example of complicated amplitude and phase modulation. The hologram has been introduced in a FEI-Tecnai F20T operated at 200 kV, in the sample position and the diffraction has been observed using low-mag diffraction mode. Fig 1c shows the comparison between the experimentally generated intensity of the beam (left) and the Microsc. Microanal. 21 (Suppl 3), 2015 504 aimed beam (right). In the simulated beam the intensity is proportional to image brightness and the hue is proportional to the phase. It is worth remarking that the new holographic scheme, if compared to the original blazed hologram [4], presents a larger flexibility: however while phase-only Blazed holograms permits, at least in principle, to obtain a situation with only one diffracted beam, the present phase-amplitude hologram cannot eliminate residual intensity on all diffracted orders and in particular on the 0th order . An example in both experiment (above) and theory (Below) of +/-1 and 0th diffraction order is shown in fig 1d. References: [1] J. Verbeeck, H. Tian, and P. Schattschneider, Nature 467, (2010) p 301. [2] B. J. McMorran, A. Agrawal, I. M. Anderson, A. A. Herzing, H. J. Lezec, J. J. McClelland, and J. Unguris, Science 331, 192 (2011). [3] V. Grillo, E. Karimi, G C Gazzadi, S Frabboni,M R. Dennis, and R W. Boyd Phys. Rev. X 4, 011013 (2014) [4] V. Grillo, G C Gazzadi, E. Karimi, E Mafakheri,R W. Boyd, and S Frabboni Applied Physics Letters 104, 043109 (2014) 4 �m a) b) d) c) Figure 1. a) Aimed hologram pattern characterized by amplitude modulated blazed groove profile b) Experimental thickness map of the generated hologram c) comparison between experimental (left) and theoretical (right) beam. In the latter the phase has been highlighted by hue colorscale where hue is the phase. d) comparison of the full diffraction in experiment (above ) and simulations (below).");sQ1[253]=new Array("../7337/0505.pdf","Electron Vortex Beams for Magnetic Measurements on Ferromagnetic Samples via STEM","","505 doi:10.1017/S1431927615003323 Paper No. 0253 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electron Vortex Beams for Magnetic Measurements on Ferromagnetic Samples via STEM Darius Pohl1, Jan Rusz3, Sebastian Schneider1,2, Peter Tiemeijer4 and Bernd Rellinghaus1 1 IFW Dresden, P.O. Box 270116, D-01171 Dresden, Germany 2 TU Dresden, Institute for Solid State Physics, D-01062 Dresden, Germany 3 Uppsala University, Department of Physics and Astronomy, SE-752 37 Uppsala, Sweden 4 FEI Company, PO Box 8066, 5600 KA Eindhoven, The Neverlands X-ray magnetic circular dichroism is a well-established method to study element specific magnetic properties of a material, while electron magnetic circular dichroism (EMCD), which is the electron wave analogue to XMCD, is scarcely used today. Recently discovered electron vortex beams, which carry quantized orbital angular momenta (OAM) L, promise to also reveal magnetic signals [1]. Since electron beams can be easily focused down to sub-nanometer diameters, this novel technique provides the possibility to quantitatively determine local magnetic properties with unrivalled lateral resolution. In order to generate the spiralling wave front of an electron vortex beam with an azimuthally growing phase shift of up to 2 and a phase singularity in its axial centre, specially designed apertures are needed [2]. Dichroic signals on the L2 and L3 edge are expected to be of the order of 5% [3,4]. For this purpose, we have successfully implemented two types of apertures (spiral- and fork-type, see Fig. 1) into the condenser lens system of a FEI Titan3 80-300 transmission electron microscope (TEM) equipped with an image spherical aberration corrector. The spiral aperture allows for the generation of focused electron vortex beams with user-selectable OAM that can be used as probes in scanning TEM (STEM). Since for such spiral apertures, the different OAM are dispersed along the beam direction, tuning the focal length of the condenser lens allows for a selection of the OAM. In the case of the fork-type aperture, the EVB are dispersed in the x-y plane. First investigations were focused on probing the presence of an EMCD signal with such vortex beams on different thin ferromagnetic Ni [5] and Fe films and on hard magnetic L10 ordered FePt nanocubes. Here, the diameter of the L = 0 probe (= full with at half maximum, FWHM) is 0.14 nm, whereas the |L| = 1 probes have diameters of roughly 0.3 nm. However, first experiments did not reveal any differences in the absorption edges in the electronenergy-loss spectra (EELS) generated by vortex beams with different OAM (cf. Fig. 2). In order to understand the lack of any dichroic signal when using spiral apertures, the generation and propagation of the vortex wave function and the spatial distribution of the OAM were simulated for the given experimental setup. The simulations reveal that although the OAM is largely localized (in all three dimensions) symmetrically around the geometrical focal points, the superposition of the selected vortex state (e.g., L = +1) with contributions from adjacent vortex states (e.g., L = 0 and L = -1) results in a suppression of the total OAM [5]. We have recently devised an escape route to this problem by blocking any partial beams that carry other but the desired OAM prior to the interaction of the beam with the ferromagnetic sample. This is achieved by using a special condensor aperture in combination with a fork-type aperture to select a single partial beam with the chosen OAM. This approach allows to generate atom-sized EVB with probe diameters of 0.3 nm and a well-defined OAM. Although this discretization results in an increased signal-to-noise ratio, this novel technique allows for atomic resolution EELS Microsc. Microanal. 21 (Suppl 3), 2015 506 measurements. First experiments using this new optical setup show very promising EMCD results on ferromagnetic FePt nanoparticles. [1] J. Verbeeck et al., Nature 467 (2010), p. 301-304. [2] J. Verbeeck et al., Ultramicroscopy 113 (2012), p. 83-87. [3] P. Schattschneider et al., Ultramicroscopy 136 (2014), p. 81-85. [4] J. Rusz and S. Bhowmick, Phys. Rev. Lett. 111 (2013), 105504. [5] D. Pohl et al., Ultramicroscopy 150 (2015), 16-22. Fig 1: Generation of electron vortex beams. a, Scanning electron microscope image of a spiral-type aperture. b, Scanning electron microscope image of a fork-type aperture. c, Image of the electron beam that is generated by the fork-type aperture in b. Yellow circles mark the area of the inserted additional aperture that is used for the generation of single EVB. Fig 2. Magnetic measurements using a spiral-type aperture at FePt nanocubes. a, Scanning dark-field image of a FePt nanocube acquired with the L=0 beam of the spiral-type aperture. Inset show FFT with marked superstructure reflections of L10-ordered FePt. b, EEL spectra acquired with an EVB scanned over the yellow marked area in b. No significant EMCD signal (= difference between the two spectra) is observed.");sQ1[254]=new Array("../7337/0507.pdf","Low Voltage SEM and Correlative Microscopy to Analyze Delicate Biological Material","","507 doi:10.1017/S1431927615003335 Paper No. 0254 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Low Voltage SEM and Correlative Microscopy to Analyze Delicate Biological Material Heide Schatten University of Missouri, Department of Veterinary Pathobiology, Columbia, MO 65211, USA Advances in correlative microscopy have made it possible to compare and analyze samples on light and electron microscopy levels. For delicate isolated biological material the capabilities provided by low voltage field emission scanning electron microscopy (LVFESEM) have particularly been helpful to allow comparison analysis of structures such as the isolated mitotic apparatus [1-9]. Our research has been focused on microtubule organization and microtubule functions in cells during early reproduction in which the sperm triggers assembly of microtubule formations including the sperm aster, zygote aster, and the resulting mitotic apparatus. Centrosome organization and dynamics, and the interactions with centrioles are further revealed by LVFESEM and correlated with immunofluorescence images. While conventional TEM has not revealed sufficient detail of centrosome structure, the FESEM has given us the first close insights into microtubule-centrosome interactions in the sperm aster, and its dynamic reorganization and remodeling into the mitotic apparatus during the first embryonic cell in which centrosomes play critical roles, as will be discussed in the presentation. References [1] Schatten, H. (2008). High-Resolution, Low Voltage, Field-Emission Scanning Electron Microscopy (HRLVFESEM) Applications for Cell Biology and Specimen Preparation Protocols. In: Biological Low-Voltage Scanning Electron Microscopy. Edited by H. Schatten and J. Pawley. Springer, New York, Heidelberg, Berlin. [2] Schatten, H. (2008). The mammalian centrosome and its functional significance. Histochem. Cell Biol. 129:667-686. [3] Schatten, H., and Sun, Q-Y. (2009). The functional significance of centrosomes in mammalian meiosis, fertilization, development, nuclear transfer, and stem cell differentiation. Environ Mol Mutagen 50(8): 620-636. [4] Schatten, H., and Sun, Q-Y. (2010). The role of centrosomes in fertilization, cell division and establishment of asymmetry during embryo development. Seminars in Cell and Developmental Biology 21:174-184. [5] Schatten, H., and Sun, Q-Y. (2011). New insights into the role of centrosomes in mammalian fertilisation and implications for ART. Reproduction 142:793�801. [6] Schatten, H., Rawe, V.Y., and Sun, Q-Y. (2012). Cytoskeletal architecture of human oocytes with focus on centrosomes and their significant role in fertilization. In: Practical Manual of In Vitro Fertilization: Advanced Methods and Novel Devices, edited by Zsolt Peter Nagy, Alex C. Varghese, and Ashok Agarwal. Humana Press (Springer Science+Business Media, New York, USA). Microsc. Microanal. 21 (Suppl 3), 2015 508 [7] Schatten H, Sun Q-Y. (2012). Nuclear-centrosome relationships during fertilization, cell division, embryo development, and in somatic cell nuclear transfer (SCNT) embryos. In: The Centrosome; edited by Heide Schatten, Springer Science and Business Media, LLC (July 2012) [8] Schatten, H. and Sun, Q-Y. (2013). The role of the sperm centrosome in reproductive fitness. In: Paternal Influences on Human Reproductive Success, edited by Dr Douglas Carrell, published by Oxford University Press, 2013. [9] Schatten, H. and Sun, Q-Y. (2014). "Posttranslationally modified tubulins and other cytoskeletal proteins: Their role in gametogenesis, oocyte maturation, fertilization and preimplantation embryo development. In Posttranslational Protein Modifications in the Reproductive System; Edited by Peter Sutovsky; published by Springer Science and Business Media Figure 1: Left: LVFESEM of aster from isolated mitotic spindle in which microtubules emanate from a central core. Right: Isolated centrosomal material Modified from [1].");sQ1[255]=new Array("../7337/0509.pdf","Structural Biology at The Single Particle Level: Imaging Tobacco Mosaic Virus","","509 doi:10.1017/S1431927615003347 Paper No. 0255 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Structural Biology at The Single Particle Level: Imaging Tobacco Mosaic Virus by Low-Energy Electron Holography Jean-Nicolas Longchamp1, Tatiana Latychevskaia1, Conrad Escher1 and Hans-Werner Fink1 1 Physics Department, University of Zurich, Switzerland A milestone for structural biology would definitely be attained if methods and tools were available, that do away with averaging over an ensemble of molecules and thereby enable structural biology on a truly single molecule level. To obtain atomic resolution information about the structure of any individual biological molecule, entirely new concepts and technologies are needed. One approach of this kind is associated with the recent X-ray free electron laser (XFEL) projects. Unfortunately, there are now strong indications that also for XFEL experiments averaging over a large number of molecules will be inevitable in order to obtain images with a sufficiently high signal-to-noise ratio to enable numerical reconstruction of the diffraction pattern with atomic resolution [1,2]. Our approach to structural biology at the single molecule level is motivated by the experimental evidences that electrons with a kinetic energy in the range of 50-250eV are harmless to biomolecules [3,4]. Even after exposing fragile molecules like DNA or proteins to a total electron dose of 106e-/�2, more than six orders of magnitude higher than the critical dose in transmission electron microscopy, no radiation damage could be observed [3]. This, combined with the fact that the de Broglie wavelengths associated with this energy range are between 0.8 and 1.7�, makes low-energy electron microscopy a candidate for structural biology at the single molecule level. In our low-energy electron holographic setup (Figure 1), inspired by Gabor's original idea of inline holography [5,6], a sharp tungsten tip acts as source for a divergent beam of highly coherent electrons [6]. The electron field emitter is brought as close as 100nm to the sample with the help of a 3-axis nanopositioner. Part of the emitted electron wave is elastically scattered off the object and hence is called the object wave, while the un-scattered part of the wave represents the reference wave. At a distant detector, the hologram, i.e. the pattern resulting from the interference of these two waves is recorded. A hologram contains the phase information of the object wave, and the object structure can thus be reconstructed unambiguously. In low-energy electron holography, a lens-less technique not suffering from lens aberrations, the resolution limit is given by the deBroglie wavelengths () and by the numerical aperture (NA) of the detector system. With being as small as 0.7� and NA=0.54 in our instrument, atomic resolution shall eventually be possible [7]. Here, we will report nanometer resolution imaging by means of low-energy electron holography of individual tobacco mosaic viruses (TMV) deposited onto ultraclean freestanding graphene [8]. We will show that structural details arising from the helical structure of the viruses can be revealed and that the agreement between our images and an atomic model of TMV available from the protein database is remarkable (Figure 2). We will also describe our on-going efforts towards 2� resolution by means of low-energy electron coherent diffraction imaging (CDI). By implementing a CDI experimental scheme for low-energy electrons, we could already image a 210nm freestanding region of graphene at 2.3� resolution, revealing more than half a million of graphene unit cells at once [9]. Finally, we will describe in details the procedure to prepare ultraclean freestanding graphene by the Pt-metal catalysis method; a method enabling the application of single-layer graphene as template for electron microscopy [10]. Microsc. Microanal. 21 (Suppl 3), 2015 510 References: [1] J.W. Miao, H.N. Chapman, J. Kirz, D. Sayre, K.O. Hodgson, Taking X-ray diffraction to the limit: Macromolecular structures from femtosecond X-ray pulses and diffraction microscopy of cells with synchrotron radiation, Annu. Rev. Biophys. Biomol. Struct. 33 (2004) 157�176. H.N. Chapman, P. Fromme, A. Barty, T.A. White, R.A. Kirian, A. Aquila, et al., Femtosecond X-ray protein nanocrystallography, Nature. 470 (2011) 73�77. M. Germann, T. Latychevskaia, C. Escher, H.-W. Fink, Nondestructive imaging of individual biomolecules, Phys. Rev. Lett. 104 (2010) 95501. J.-N. Longchamp, T. Latychevskaia, C. Escher, H.-W. Fink, Non-destructive imaging of an individual protein, Appl. Phys. Lett. 101 (2012) 93701. D. Gabor, A new microscopic principle, Nature. 161 (1948) 777�778. H.-W. Fink, W. Stocker, H. Schmid, Holography with low-energy electrons, Phys. Rev. Lett. 65 (1990) 1204�1206. T. Latychevskaia, J.-N. Longchamp, C. Escher, H.-W. Fink, Holography and coherent diffraction with low-energy electrons: A route towards structural biology at the single molecule level, Ultramicroscopy. (2014). J.-N. Longchamp, T. Latychevskaia, C. Escher, H.-W. Fink, Structural biology at the single particle level: imaging tobacco mosaic virus by low-energy electron holography, (2014). J.-N. Longchamp, T. Latychevskaia, C. Escher, H.-W. Fink, Graphene Unit Cell Imaging by Holographic Coherent Diffraction, Phys. Rev. Lett. 110 (2013) 255501. J.-N. Longchamp, C. Escher, H.-W. Fink, Ultraclean freestanding graphene by platinum-metal catalysis, J. Vac. Sci. Technol. B Microelectron. Nanom. Struct. 31 (2013) 020605. [2] [3] [4] [5] [6] [7] [8] [9] [10] Figure 1: Scheme of the experimental setup of low-energy electron holography. The source-sample distance amounts to typically 100-1000nm, which leads to kinetic electron energies in the range of 50-250eV. Figure 2: a High-magnification hologram of TMV recorded at 80eV. b Reconstruction of the shape of TMV from a the red arrows mark the presence of details arising from the helical structure of TMV. c An atomic model for TMV is superimposed on the image presented in b.");sQ1[256]=new Array("../7337/0511.pdf","Helium Ion Microscopy of Plant Tissues and Mammalian Cells","","511 doi:10.1017/S1431927615003359 Paper No. 0256 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Helium Ion Microscopy of Plant Tissues and Mammalian Cells Andrew F. Loftus1, Matthew S. Joens1, Sarah E. Dunn1, Michael W. Adams1, Chuong Huynh2, Bernhard Goetze2 and James A.J. Fitzpatrick1 1 Waitt Advanced Biophotonics Center, Salk Institute for Biological Studies, 10010 N. Torrey Pines Road, La Jolla, CA 92037 2 Ion Microscopy Innovation Center, Carl Zeiss Microscopy, 1 Corporation Way, Peabody, MA 01960 From plants to animals, the helium ion microscope (HIM) facilitates ultrastructural investigations of biological systems at both high contrast and resolution without the need for heavy metal coatings [1]. In this study, we describe not only the advantages of HIM over the low-voltage scanning electron microscopy (LV-SEM) of plant tissues, but will also demonstrate advances in preserving sub-nanometer membrane structures of in vitro cell cultures. In principle, the operation of a HIM is similar to SEM with the exception that the ion beam has no crossovers. Briefly, a metal tip, forming a three-atom apex is held at a bias in the presence of helium gas, creating a localized ion field. Positively charged ions are subsequently extracted, accelerated, shaped and scanned in the form of a beam across the sample with the resultant secondary electrons collected. The higher mass helium ions give rise to a shorter de Broglie wavelength when compared to the SEM, effectively resulting in an order of magnitude smaller probe size. This higher mass also yields deeper beam penetration prior to scattering which causes less beam-induced damage and enhancement of SE1 secondary electron signal. The HIM has an added benefit in that charging is negated using an electron flood gun interlaced with the scanning ion beam. Compared to SEM, HIM provides the greatest benefit in charge compensation for uncoated samples. Traditionally SEM samples are imaged with a heavy metal coating for conduction and increasing ionization yield, or imaged at low accelerating voltage to reduce charging artifacts. The imaging of uncoated biological samples can be limited in LVSEM by charging as exhibited in the side-by-side evaluation of Arabidopsis thaliana with the HIM (Figure 1). While decreasing the field of view (FOV) in the HIM does not result in any evident loss of resolution, the LV-SEM is plagued by image distortions resulting from charge accumulation. However, both metal coating and low-voltage imaging result in the obstruction of surface features, or a decrease in material contrast and topography at higher resolutions. With the clear advantages of HIM imaging demonstrated in plant tissues, the surface ultrastructure of in vitro HeLa cell cultures were investigated due to their ubiquity in biological studies. Previous efforts resulted in a wide range of cell membrane irregularities resulting in the loss topological information [2]. Subsequent efforts have centered on attempting to overcome the issue of membrane degradation utilizing a dual approach involving fixation and labeling methods. Briefly, HeLa cells were cultured on coverglass slips, fixed using standard aldehyde treatments, and critical point dried (CPD) as described previously [2]. Some samples were supplemented with 1% Tannic Acid (TA) and/or osmium fixed prior to the CPD step to enhance membrane integrity. In tandem, additional HeLa cells were incubated with cholera toxin subunit B (CTxB) to investigate whether CTxB binding to the saturated lipids present in raft structures might act as a stabilization agent. Briefly, CTxB samples were incubated with CTxBFITC, and were either aldehyde fixed and osmium treated prior to CPD or were immunolabeled with a rabbit anti-cholera primary antibody, followed by a AlexaFluor-594 goat anti-rabbit secondary antibody, aldehyde fixed, osmium treated, and CPD. From the high magnification HIM images (Figure 2), there are clear variations in the effectiveness of each fixation approach to preserve the cellular membrane. Microsc. Microanal. 21 (Suppl 3), 2015 512 Osmium fixation prior to CPD allows for a more stable membrane even in the absence of other fixation agents. The use of tannic acid, which is believed to stabilize the unsaturated lipids during fixation with gultaraldehyde [3], resulted in no improvement in the stability of the membranes (Figure 2C). The CTxB-primary-secondary antibody (Figure 2E) samples showed improvement, particularly in the retention of saturated lipids found in caveolar raft structures (Figure 2E) as compared to CTxB with no cross-linking antibody present (Figure 2D). This work showcases the application of HIM as compared to LV-SEM in its capability to visualize the sub-nanometer ultrastructure of biological systems with both high-resolution and reduced charging artifacts. These results clearly show the enhancement of lipid membrane stability via the use of enhanced fixation and labeling approaches. In addition, the approach of using CTxB protein, known to bind to ganglioside lipids found primarily in cellular rafts along with crosslinking antibodies seems to allow for the retention of those raft components on the cell surface (Figure 2F). We propose a hypothesis of a CTxB protein / antibody network that serves to stabilize the cellular membrane in saturated lipid raft regions where the proteins have associated. References: [1] Bell, D. C. Microscopy and Microanalysis, (2009), 15(2), p.147-153. [2] MS Joens et al. Scientific Reports, (2013), 3, p. 3514. [3] Hayat, M.A. "Stains and cytochemical methods", Plenum Press (New York) p.336-339. [4] The authors acknowledge financial support from the Waitt Foundation, W.M. Keck Foundation, NCI (CA014195) and NINDS (NS072031). Figure 1. Comparison of SEM and HIM in Arabidopsis thaliana. Increasing magnification series of uncoated and dried A. thaliana sepal cuticle in the SEM at 3 KeV (a, b) and HIM (c, d). The difference in resolution is apparent in panels (b) and (d) and the difference in depth of field is apparent at low SEM magnifications as indicated by the black arrowheads in (a). Scale bars: (a,c) 5�m (b,d) 250nm. Figure 2. HIM images of HeLa cells with various fixation conditions and schematic of CTxB binding. (A) Aldehyde fixed with CPD. (B) Aldehyde fixed, osmium fixed, with CPD. (C) Fixed as in (B) with 1% TA supplemented. (D) Incubated with CTxB and fixed similar to (B). (E) Incubated with CTxB, primary antibody and secondary antibody and fixed similar to (B). (F) Schematic of the binding of CTxB to lipid rafts and antibodies to the CTxB. Black arrows in (B-D) indicate membrane vacancies considered to be caveolar raft structures eliminated of saturated lipids. Scale Bar = 500nm.");sQ1[257]=new Array("../7337/0513.pdf","Imaging of Carbon Nanotubes Embedded in Polymer Composites via Low Energy Scanning Electron Microscopy and Scanning Lithium Ion Microscopy","","513 doi:10.1017/S1431927615003360 Paper No. 0257 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Imaging of Carbon Nanotubes Embedded in Polymer Composites via Low Energy Scanning Electron Microscopy and Scanning Lithium Ion Microscopy Minhua Zhao1,2, Kevin A. Twedt1,3, Jabez J. McClelland1, and J. Alexander Liddle1 1 Center for Nanoscale Science and Technology, National Institute of Standards and Technology, Gaithersburg, MD 20899, USA 2 Department of Materials Science and Engineering, University of Maryland, College Park, MD 20742, USA 3 Maryland NanoCenter, University of Maryland, College Park, MD 20742, USA Understanding the dispersion of carbon nanotubes (CNTs) embedded in a polymer matrix is critical for constructing the structure-property relationships that can be used to improve the performance of CNT composites. SEM imaging is, in principle, an effective way to determine the CNT dispersion. However, the limited conductivity of the polymer matrix can lead to excessive negative charge buildup in the polymer. To avoid this, we have developed approaches to create optimum imaging conditions by generating weak positive charging using both low-energy scanning electron microscopy (SEM) and scanning lithium ion microscopy (SLIM).1 Positive charging on the polymer surface can be created in SEM in two ways, as shown in Figure 1. First, we use low-voltage SEM with electron landing energies between the first (E1) and second (E2) crossover energies (Fig 1. b). The contrast of embedded CNTs is bright relative to the polymer matrix at incident beam energy of 350 V, a typical case for positive charging2. Second, if a conducting substrate is present, when the electron beam range is greater than the composite film thickness, we can use a penetrating beam to create a positive charging at region III with landing energy E > E3, which is defined as a third crossover energy.2,3 The fact that we see similar contrast at both 3 kV and 350 V is consistent with the existence of E3 for our samples, considering the electron beam range in epoxy at 3 kV accelerating voltage is comparable to the film thickness (t=100 nm). Although positive charging by a penetrating beam has been reported in both experiment4 and simulation5 before, the use of higher energy beams has been neglected by experimentalists and theorists. The advantages of a penetrating beam over a low-voltage beam in SEM imaging include improved spatial resolution due to reduced lens aberrations at higher accelerating voltages and deeper subsurface imaging depth due to the greater beam penetration.2 We also compare SEM imaging to SLIM imaging, where positive surface charging occurs for all ion beam energies. Figure 2 shows the same location of a 1 % CNT-epoxy film imaged by SLIM and SEM. Similar bright contrast for the embedded CNTs is observed by both techniques, which further suggests the existence of weak positive charging at the polymer surface. In addition, the application of these techniques is not limited to imaging of CNTs embedded in polymer composites. They are generally applicable to conducting nanostructures embedded in a dielectric matrix, such as graphene polymer composites or integrated circuit conductors covered by a dielectric layer. References: [1] KA Twedt, L Chen, and J J McClelland, Ultramicroscopy 142 (2014), p. 24�31. [2] MH Zhao et al, Nanotechnology 26 (2015), 085703. [3] L Reimer et al, Optik 92 (1992), p. 14-22. [4] W Liu, J. Ingino and R.F. Pease, Journal of Vacuum Science and Technology B, 13 (1995), p. 1979. [5] WQ Li and HB Zhang, Micron, 41 (2010), p.416-422. Microsc. Microanal. 21 (Suppl 3), 2015 514 [6] Dr. Minhua Zhao and Dr. Kevin Twedt acknowledge support under the Cooperative Research Program in Nanoscience and Technology between the University of Maryland and NIST. Microtomeprepared CNT-epoxy films are courtesy of Dr. Chelsea S. Davis at NIST. Figure 1 (a) SEM imaging of microtome-prepared 100 nm thick 1% CNT-epoxy film on conductive substrate by beam accelerating voltage of 350 V, 1 kV and 3 kV respectively. Horizontal field of view for all images is 270 �m. (b) Schematic curve of total secondary electron yield () vs. electron landing energy (E). Inset: Schematic of a SEM beam interaction with a dielectric thin film attached to a conductive substrate. Figure 2: Same location imaging of microtome-prepared 100 nm thick 1% CNT-epoxy film on conductive substrate by SLIM (5 keV landing energy) and SEM (3 keV landing energy) respectively. Secondary electron signal is collected for both images. Horizontal field of view is 30.4 �m.");sQ1[258]=new Array("../7337/0515.pdf","Probing the Organic/Inorganic Interface of the Ferritin Protein using Atom Probe Tomography","","515 doi:10.1017/S1431927615003372 Paper No. 0258 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Probing the Organic/Inorganic Interface of the Ferritin Protein using Atom Probe Tomography Daniel E. Perea1, Jia Liu1, Jonah Bartrand1, Quinten Dicken1, Nigel Browning2, Theva Thevuthasan1,3, and James E. Evans1 1. Environmental Molecular Sciences Laboratory, Pacific Northwest National Laboratory, Richland WA 99352, USA. 2. Fundamental & Computational Sciences, Pacific Northwest National Laboratory, Richland, WA 99352, USA. 3. Qatar Environment and Energy Research Institute, Qatar Foundation, Doha, Qatar The ability to image biointerfaces over nanometer to micrometer length scales is fundamental to correlating biological composition and structure to physiological function, and is aided by a multimodal approach using advanced complementary microscopic and spectroscopic characterization techniques. The regular application of atom probe tomography (APT) to soft biological materials is lacking in large part due to difficulties in specimen preparation and inabilities to yield meaningful tomographic reconstructions that produce atomic scale compositional distributions as no other technique currently can. Here we describe advancement in the atomic-scale tomographic analysis of biological materials using APT that is facilitated by an advanced focused ion beam based approach. A novel specimen preparation strategy is used in the analysis of horse spleen ferritin protein embedded in an organic polymer resin which provides chemical contrast to distinguish the inorganic-organic interface of the ferrihydrite mineral core and protein shell of the ferritin protein, as well as the organic-organic interface between the ferritin protein shell and embedding resin. Given the complex mass spectra that result from the pulsed-laser assisted field evaporation of organic polymer compounds using APT [1], the detection of multiple species having overlapping mass/charge ratio (m/z) can complicate or prevent a definitive identification of the nitrogen and iron species of interest. A systematic analysis of the pure resin at varying laser energies indicates that at 450 pJ laser energy, the contribution of organic species at 14 Da (i.e. CH2+) is eliminated to allow detection of N+ from resin specimens containing the ferritin protein. Individual ferritin cores surrounded by carbon are observed in the ferritin embedded resin specimen (Fig. 1a). Using proximity histogram analysis, we are able to map the relative composition profiles of Fe, C, N, and P. Results are corroborated with the analysis of similarly prepared specimens of Fe2O3 nanoparticles embedded in the same resin. The results demonstrate a viable application of APT analysis to study the complex biological organic/inorganic interfaces with an extension of the specimen preparation technique to further enhance the study of organic and inorganic nanomaterials relevant to energy and the environment. In addition, progress in the analysis of cryo-prepared specimens will be discussed. References: [1] [2] TJ Prosa, SK Keeney & TF Kelly, J. Microsc. 237(2010), 155-167. The research was performed at the Environmental Molecular Sciences Laboratory; a national scientific user facility sponsored by the Department of Energy's Office of Biological and Environmental Research located at Pacific Northwest National Laboratory, and was supported by the Chemical Imaging Initiative Laboratory Research & Development program. Microsc. Microanal. 21 (Suppl 3), 2015 516 Figure 1: a) 3D compositional map of Fe from the iron-rich ferrihydrite mineral core of a ferritin molecule. Purple mesh surface is a 15% Fe isoconcentration surface. 2D composition maps of b) carbon and c) iron.");sQ1[259]=new Array("../7337/0517.pdf","Methods in Creating, Transferring, & Measuring Cryogenic Samples for APT","","517 doi:10.1017/S1431927615003384 Paper No. 0259 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Methods in Creating, Transferring, & Measuring Cryogenic Samples for APT S.S.A. Gerstl, R. Wepf Scientific Center of Optical and Electron Microscopy (ScopeM), ETH Zurich, Zurich, Switzerland In its infancy, atom probe microscopy began as an analytical instrument primarily for metals and later semiconductors. As laser pulsing technology rapidly improved and computational capabilities were integrated, expediting many functions, analysis of additional materials classes were enabled, such as oxides, ceramics, and minerals [1]. All of these materials have something in common when prepared for APT analysis: they are dry, dense, and shaped into a specimen geometry with a sharp needle point with the region of interest (ROI) in the top 100s of nm of the apex. In this paper, we discuss the potential and various fields of application where cryogenic-preparation and cryogenic-transfer of specimens into an atom probe is favorable compared to conventional room temperature methods that are standardized. Cryo-fixation of specimens involves methods long developed for cryo�electron microscopy capabilities [2]. The methods enable freezing in and arresting dynamic events of various types of states of highly energetic materials. Such freezing events can controllably take place in �sec to msec. This is more commonly known as rapid quenching (such as splat quenching) in metals, where relatively little attention is required after the solidified state is achieved. Thermal and atomic transport within the metals with respect to the cooling surface is of greatest concern, hence the resulting sizes of such materials are designed to be very thin (on order of several microns). Specimens containing water however need special attention, particularly after the solidified state has been achieved. Care must be taken to avoid ice crystal formation, as observed on the shank of a typical APT specimen in fig. 1, which can damage the ROI structure or impair the electrical conductivity of the ideally vitreous frozen sample. Developments from the life-science fields linking rapid freezing (Plunge and High Pressure Freezing) and vacuum-cryo-transfer with cryo�FIB/SEM processing have been adapted to our LEAP system, such that a frozen specimen in the FIB, fig. 1(a) can be transferred, with little contamination, to the LEAP, analyzed, and returned for imaging, fig. 1(b). The extensively tested hardware additions enable cryotransfer of highly dynamic metals, environmentally sensitive or corrosive materials, and originally fluid based ROIs. The various benefits of cryo-enabled LEAP for materials sciences will initially be addressed, whereby dynamically changing alloys, such as those suffering from natural aging or preferential/rapid corrosion, can be analyzed in their native, `frozen-in' or unaltered states not having been exposed to thermal or environmental stimuli. Several pathways will then be discussed and shown regarding plunge freezing nanoparticles in liquid films on prefabricated tips, or in plasma FIB-fabricated vessels (fig. 2), or high pressure frozen bio-organic matter with the cryo-preparation of tips via cryo-ultramicrotomy and cryoFIB milling. Microsc. Microanal. 21 (Suppl 3), 2015 518 References: [1] D.Larson, et al, Local Electrode Atom Probe Tomography, (Springer, New York), 2013 p.201. [2] P.Echlin Low-Temperature Microscopy and Analysis (Plenum Press, New York) 1992. [3] the authors would like to acknowledge the ongoing hi-level support from the EMEZ/ScopeM team. (a) (b) Figure 1. (a) A W needle plunge frozen and imaged via cryo-SEM showing ice crystal contamination prior to cryo-FIB final preparation. (b) The same needle having been transferred to the LEAP, analyzed, and returned to the cryo-FIB. Note changes near the apex where the field was applied and little to no change in contamination of the shank during transfers between instruments. Figure 2. One example `microcup', geometrically designed by plasma FIB capabilities for cryogenic preparations of fluid based ROIs.");sQ1[260]=new Array("../7337/0519.pdf","Atom Probe of Apatites � from Single Crystals to Interphases in Tooth Enamel","","519 doi:10.1017/S1431927615003396 Paper No. 0260 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atom Probe of Apatites � from Single Crystals to Interphases in Tooth Enamel Lyle M. Gordon1, Karen A. Derocher1, Michael J. Cohen1, Derk Joester1. 1 Materials Science and Engineering, Northwestern University, Evanston, IL, USA. Tooth enamel is the hardest tissue in vertebrates. Optimized to withstand the forces of mastication, it is composed of hydroxylapatite nanowires, thousands of which are bundled into rods that are organized in a three-dimensional weave. The outstanding fracture resistance of enamel and its long fatigue life are the consequence of this hierarchical architecture. Tooth enamel is also the target of the most prevalent infectious disease in humans: dental caries. It is an infectious disease that has extremely high morbidity, with 60-90% of children and nearly 100% of adults worldwide having or having had caries.[1] Caries, simply put, is the destruction of tooth biominerals by dissolution and commonly begins with the demineralization of enamel by acids produced in plaque biofilms. It has long been known that the susceptibility of enamel to acid dissolution is greatly dependent on the presence of magnesium, carbonate, and fluoride ions. A major bottleneck in understanding formation, structural evolution, and degradation of enamel under normal conditions and during tooth decay has been that imaging the distribution of these impurities in enamel has remained a great challenge. Others and we have recently shown that UV laser-pulsed atom-probe tomography (APT), in combination with correlative techniques, enables unprecedented insight into the nano-structure and chemistry of apatitic biominerals.[2-4] This required developing a work flow for sample preparation, data collection, and data analysis tailored for apatites (Ca5(PO4)3X, X=F, Cl, OH), wide band gap (~5.3 eV) insulators[5] that had never before been investigated by atom probe. We approached this by first performing a comprehensive analysis of synthetic and geologic single crystals of apatite end members.[2] This was the basis for a second step, in which we analysed biogenic nanocomposites, including cortical bone and dentin.[2] Our approach culminated in the recent discovery of amorphous interphases in tooth enamel and their importance in controlling mechanical properties and resistance to acid corrosion.[6,7] We report here on the optimization of sample preparation conditions using FIB, the dependence of spectral quality and stoichiometry on atom probe operational parameters (Fig. 1), an analysis of multihit events in single-crystalline apatites (Fig. 2), and the differentiation of organic and inorganic carbon in enamel samples (Fig. 3). Finally, we will discuss preliminary results from atom probe of human enamel. References: [1] World Health Organization Media Centre, http://www.who.int/mediacentre/factsheets/fs318/en/. [2] L. M. Gordon, L. Tran, D. Joester, ACS nano 6, (2012). [3] J. Karlsson, G. Sundell, M. Thuvander, M. Andersson, Nano Letters 14, (2014). [4] E. A. Marquis, M. Bachhav, Y. Chen, Y. Dong, L. M. Gordon, D. Joester, A. McFarland, Current Opinion in Solid State and Materials Science 17, (2014). [5] P. Rulis, L. Ouyang, W. Y. Ching, Physical Review B 70, (2004). [6] L. M. Gordon, D. Joester, Frontiers in Physiology 6, (2015). [7] L. M. Gordon, M. J. Cohen, K. W. MacRenaris, J. D. Pasteris, T. Seda, D. Joester, Science 347, (2015). [9] The authors acknowledge funding from the National Science Foundation (NSF DMR-0805313, DMR-1106208, and DMR-1341391), the Northwestern University Materials Research Center (NSF- Microsc. Microanal. 21 (Suppl 3), 2015 520 MRSEC DMR-1121262), the International Institute for Nanotechnology, the Institute for Sustainability and Energy at Northwestern (ISEN), the Petroleum Research Fund of the ACS. LMG was supported in part by the Canadian National Sciences and Engineering Research Council. MJC was supported in part by NIH predoctoral Biotechnology Training Grant T32GM008449. Figure 1. Optimization of stoichiometry (A, C), mass resolving power (B,D), single hit frequency (E,G), and charge state ratio (F,H) as a function of laser pulse frequency (A,B,E,F) and laser power (C,D,G,H). Figure 2. Double hit correlation histograms of a synthetic single crystal of hydroxylapatite. A. Correlation histogram with simulated decay tracks indicated below the diagonal (red arcs). B. Close up of (A), showing Figure 3. 3D Reconstructions of inner mouse enamel. A. Mg segregation at grain boundaries between hydroxylapatite nanowires. B. Nitrogen-containing ions in the intergranular precipitate. C,D. Carboncontaining ions originating from both organic and inorganic carbon.");sQ1[261]=new Array("../7337/0521.pdf","3D Atomic Scale Analysis of CMOS type structures for 14 nm UTBB-SOI technology","","521 doi:10.1017/S1431927615003402 Paper No. 0261 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 3D Atomic Scale Analysis of CMOS type structures for 14 nm UTBB-SOI technology Robert Estivill1,2,3, Adeline Grenier2, Tony Printemps2, Marc Juhel1, Magali Gregoire1, Pierre Caubet1, Didier Blavette3. 1. 2. STMicroelectronics, 850 rue Jean Monnet, 38926 Crolles, France. Univ. Grenoble Alpes, F-38000 Grenoble, France CEA, LETI, MINATEC Campus, F-38054 Grenoble, France. 3. Groupe de Physique des Mat�riaux � GPM UMR CNRS 6634, Universit� de Rouen, France. As the dimensions of microelectronic devices are progressively reduced new architectures and materials are being introduced to try and meet ever stricter performance criteria. The use of high-k dielectrics (hafnium based oxides) can reduce leakage current leading to better electrical properties. The coupling of these dielectrics with metallic gate materials (TiN) has led to structures of greater complexity in CMOS (complementary metal oxide semiconductor) devices. Due to current dimensions (a few nanometres) atom probe tomography (APT) is one of the very few techniques which can give 3D chemical information at this scale [1-2]. But due to the insulating and high evaporation field nature of these materials analysis is often difficult, with very low analysis yields, or even impossible [3]. We here show that using simplified test structures it is possible to overcome these limitations and perform complete analysis of a finished gate stack, including source/drain regions. Although simplified, these structures are representative of current nanometre scale transistors, and give the opportunity to study the chemical composition of the final device. Analysis of both n and p-type 14/20nm structures have been performed by APT on both SOI and bulk silicon substrates, before and after a spike anneal. The samples have followed a conventional gate first UTBB-SOI (ultra-thin body buried silicon-onoxide) 14nm deposition flow up to silicide deposition. At this point manufacture has been halted and a 150 nm nickel layer deposited to facilitate specimen preparation by focused ion beam. By preparing the tips in different parts of the test zones, it is possible to obtain information about the gate stack or source/drain areas. The samples have then been analysed under pulsed UV laser until the SOI interface or Si substrate in the case of bulk samples. The APT reconstructions have been calibrated using transmission electron microscopy (TEM) images. To improve the reconstructions, correlation between the APT element mapping and TEM tomography volumes has been made, thus reducing artefacts [4]. Once the structures were correctly reconstructed, localised chemical data could be extracted. Of particular interest is the dopant localisation after deposition and anneal. All dopants in the structures (arsenic, phosphorous and boron) have been mapped. Of these, the most interesting is boron, due to the impossibility to obtain this information via other characterisation methods. We show that the boron distribution follows an irregular profile, with a concentration maxima situated between the channel and epitaxial source/drain. Due to the fact that this distribution is already present before anneal, we attribute it to inhomogeneous boron incorporation during epitaxy of the source/drain regions. This has been compared with time-of-flight secondary ion mass spectroscopy measurements made on the same test structures and larger planar zones. There is a good agreement between the profiles, although a smaller amount of boron is detected by APT. This can be explained by a saturation of the APT detector and can be accounted for in post-treatment of the data.5 Microsc. Microanal. 21 (Suppl 3), 2015 522 In conclusion, using simplified test structures and careful sample preparation it is possible to analyse a complete high-k metal gate device. Using complementary techniques on the same sample it is possible to obtain a better reconstruction and a more complete and quantitative picture of dopant distribution. An irregular boron distribution can thus be clearly observed. References [1] Thomas F. et al.,, Rev Sci Instrum 78, (2007), 031101. [2] F. Panciera, et al., Appl Phys Lett 100, (2012), 201909. [3] S. Jin et al.,. J Vac Sci Technol B 29, (2011) 061203. [4] A. Grenier et al., Ultramicroscopy 136, (2014), 185. [5] This study has been performed at the nanocharacterisation platform (PFNC) of the Minatec Campus and ST Microelectonics, Crolles. The author would like to acknowledge a CIFRE (ANRT) scholarship. Figure 1. (a) Schematic representation of test structures, showing the extended nature of the gate-stacks. (b) TEM micrograph of p-type stack on SOI substrate Figure 2. (a) Atom probe tip prepared by FIB and reconstructions of both p-type (b) and n-type (c) on bulk silicon substrate. Figure 3. Comparison of Boron concentration profiles extracted from source/drain area by APT and ToF-SIMS after spike anneal.");sQ1[262]=new Array("../7337/0523.pdf","Microscopy of Chemical and Mechanical Heterogeneities in Lithium Cobalt Oxide","","523 doi:10.1017/S1431927615003414 Paper No. 0262 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microscopy of Chemical and Mechanical Heterogeneities in Lithium Cobalt Oxide David R. Diercks1, Matthew Musselman1, Mukesh Kumar1, Brian P. Gorman1, Corinne E. Packard1,2 1. 2. Colorado School of Mines, Department of Metallurgical and Materials Engineering, Golden, CO USA National Renewable Energy Laboratory, Golden, CO USA Lithium ion batteries (LIB) are used in numerous portable electronic devices primarily due to their light weight and relatively high energy densities. During cycling, they are exposed to both chemical and mechanical gradients. The repetitive action of these can lead to irrecoverable loss of capacity. Both intra- and inter-granular failures arising from Li intercalation induced stresses have been observed [1,2]. These may result from both anisotropies in the strain field as well as anisotropies in the crystal properties [3-5]. Particles that are fractured may become separated from the primary electrical percolation pathway, meaning that this material can no longer participate in the charging and discharging cycles [6,7]. Additionally, solid electrolyte interphases, which bind Li, may be formed at the freshly created surfaces [8]. Lithium cobalt oxide is among the most common LIB cathode materials with a relatively modest volume change during cycling [9]. A transmission electron microscopy (TEM) study of this material has shown a preferred plane for dislocation formation [10] and nanomechanical measurements have indicated orientation-dependent values [11]. However, there have been relatively few studies regarding the nanoscale chemistry and fracture of LiCoO2, particularly for cycled material. A more complete understanding of the nature and interaction of the chemical and mechanical effects could lead to LIB designs and materials that have greater resistance to capacity fade. In the present work, LiCoO2 particles were taken from a commercial cell that was charged at a 1C rate to 4.2 V and discharged at a 3C rate to 2.75 V for 1000 cycles resulting in 72% of its original capacity. These were mounted and polished to expose the cross-sections of the particles. Nanoindentation of the particles followed by scanning electron microscopy analysis revealed fracturing of many of the particles, preventing determination of the mechanical response of the particles. The fractures, while occurring in the regions around the indentations, did not always originate from the regions of highest stress (Figure 1a). Focused ion beam cross-sectioning revealed that the cracks often propagated through entire particles as shown in Figure 1b. TEM diffraction analyses indicated both intra- and inter-granular fracture along {001} planes. Inter-granular fracture terminated at grain boundaries. Atom probe tomography (APT) performed on a Cameca LEAP 4000X Si instrument was used to characterization the nanoscale compositions. Uncycled specimens were studied to establish analysis conditions which produced the expected stoichiometry. From these assessments, base temperatures of 35 � 40 K and laser energies of 1 � 2 pJ were used for composition analyses of the cycled material. TEM imaging of the specimens before and after APT analysis (Figure 2a) was used to assist in generating the reconstructions [12]. From the APT analyses, some cycled particles were found to have the stoichiometric Li composition while others were found to be deficient in Li. Analyses across grain boundaries indicated differences in Li concentration on either side of the boundaries. In those cases, one side was found to be enriched in Li, suggesting poor transport into the adjacent grain. The above results indicate that both mechanical and chemical mechanisms played a role in the observed LIB capacity fade [13]. References: Microsc. Microanal. 21 (Suppl 3), 2015 524 [1] M. Wohlfahrt-Mehrens, C. Vogler and J. Garche, J. Power Sources, 127 (2004), p. 58. [2] D. Y. Wang et al., J. Power Sources, 140 (2005), p. 125. [3] S. Yamakawa et al., J. of Power Sources, 223 (2013), p. 199. [4] V. Malav� et al., Electrochimica Acta, 130 (2014), p. 707. [5] D.-W. Chung et al., J. of the Electrochem. Soc., 158 (2011), p. A1083. [6] X. Zhang, W. Shyy and A. Marie Sastry, J. of the Electrochem. Soc., 154 (2007), p. A910. [7] Y.-T. Cheng and M. W. Verbrugge, J. of the Electrochem. Soc., 157 (2010), p. A508. [8] R. Deshpande et al., J. of the Electrochem. Soc., 159 (2012), p. A1730. [9] J. N. Reimers and J. R. Dahn, J. of the Electrochem. Soc., 139 (1992), p. 2091. [10] H. Gabrisch, R. Yazami and B. Fultz, Electrochem. and Solid-State Lett., 5 (2002), p. A111. [11] M. Qu et al., Adv. Energy Mater., 2 (2012), p. 940. [12] B. P. Gorman et al., Microscopy Today, 16 (2008), p. 42. [13] D. R. Diercks et al., J. Electrochem. Soc., 161 (2014), p. F3039. [14] This work was principally funded by Seed Grants from the Renewable Energy Materials Research Science and Engineering Center, which received its funding from NSF, DMR-0820518. The atom probe used in this research was supported by NSF Award Number 1040456. Figure 1. (a) LiCoO2 particle showing a crack across the indented region. (b) Cross-section of the particle from (a) showing that the crack propagated through the entire depth of the particle. Figure 2. (a) Overlay of TEM images before and after APT analysis indicating the volume of material removed. (b) APT reconstruction of the region indicated by the box in (a). The dark blue surface shows a region enriched in Co (deficient in Li).");sQ1[263]=new Array("../7337/0525.pdf","Considerations for Physical Facility Design and Management of a State-of-the-Art Electron Microscopy and Analysis Laboratory","","525 doi:10.1017/S1431927615003426 Paper No. 0263 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Considerations for Physical Facility Design and Management of a State-of-the-Art Electron Microscopy and Analysis Laboratory Daniel E. Huber1, Jonathan Orsborn1, Henk O. Colijn1, Cameron Begg1, Michael J. Mills1, David W. McComb1, Hamish L. Fraser1 1. Center for Electron Microscopy and Analysis, Materials Science and Engineering, The Ohio State University, Columbus, OH, U.S.A. Modern electron microscopes demand a physical environment to enable them to achieve or even surpass the manufacturers performance specifications. This can prove to be formidable challenge to facility managers and researchers. Factors that commonly influence the performance of electron microscopes include, but are not limited to temperature fluctuations, air currents and pressure changes, electromagnetic fields, vibrational and acoustic disturbances. However, with proper considerations given to design and implementation of the physical microscopy facility, the influence of all of these factors can be mitigated. The Center for Electron Microscopy and Analysis (CEMAS), at The Ohio State University, opened in September 2013, was designed with the goal of optimizing the environment for state of the art microscopy. The following paper will address common issues relevant to facility mangers. Then describe how these issues are addressed at CEMAS. Firstly, we will address environmental concerns, temperature stability, electromagnetic fields, vibrational stability and grounding. However, these issues can be tied closely to the second effort related to building automation and physical plant design. Finally, a facility needs to address the possible loss of utility power and minimize the possibility of down instrument time. With these critical issues addressed the obvious problem becomes managing the volume of data produced in a large multi user facility. A brief discussion on the implementation of virtualized environment for computing and data storage capabilities at CEMAS. Where the design of a facility wide network can address security, connectivity and the management of computing resources and related workflows. CEMAS has also adopted Facility Online Manager to handle, scheduling and billing activities. The close coupling of the scheduling and billing activities increases facility efficiency by allowing instrument managers to focus on instrumentation, research and training. The result of these efforts in designing an optimal environment can be seen in the following Figure 1. This figure shows the installation of an FEI Titan 80-300 in a purpose built instrument room, where operators of the microscope is performed from a joining control room. The operator controls are simply transmitted from the instrument room to the control room. This configuration isolates the instrument from the operator and promotes high environmental stability. These levels of environmental stability are a complement to one with low EMI. To achieve a site with very low EMI carful characterization of the site of was necessary. Field survey work prior to construction concluded that nearby power lines, shown being demolished in Figure 2, induced current in building steel and would negatively affect instrument performance. The relocation effort significantly reduced levels of EMI observed within the facility and has greatly contributed to the success of the facility. Microsc. Microanal. 21 (Suppl 3), 2015 526 Figure 1. FEI Titan 80-300 installed at CEMAS at The Ohio State University. Figure 2. Relocation of power lines adjacent to new CEMAS, performed by local utility (Left). Magnetic fields from power lines inducing current in building steel as evidenced by field survey work (Right).");sQ1[264]=new Array("../7337/0527.pdf","The Use of Online Tools in Microscopy and Microanalysis Core Facilities.","","527 doi:10.1017/S1431927615003438 Paper No. 0264 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Use of Online Tools in Microscopy and Microanalysis Core Facilities. Miles Apperley1, Jenny Whiting1, Bronwen Cribb2 and Anna Ceguerra3 1. Australian Microscopy and Microanalysis Research Facility, Madsen Building F09, The University of Sydney, NSW, 2006, Australia. 2. Center for Microscopy and Microanalysis, The University of Queensland, Brisbane, QLD, 4072, Australia 3. Australian Center for Microscopy and Microanalysis and the School of Aerospace, Mechanical and Mechatronic Engineering, Madsen Building F09, The University of Sydney, NSW, 2006, Australia. The Australian Microscopy and Microanalysis Research Facility (AMMRF) is a national grid of equipment, instrumentation and expertise in microscopy and microanalysis that provides nanostructural characterisation capability and services, from widely used optical, electron, X-ray and ion-beam techniques to world-leading flagship platforms. This collaborative facility, comprising a distributed network of microscopy and microanalysis core facilities spread over fourteen institutions, manages a fully supported user experience to more than 3,000 researchers annually. Once a researcher has identified the facility thy need to work with, each core facility is responsible for supporting the users through a common research experience. This experience or work flow comprises the stages of defining the scientific question and project formulation, identifying techniques, registering a project and a meeting to discuss the project, plan training and arrange access to instrumentation. Later stages include data acquisition and analysis, data management and ultimately an outcome such as a research publication, grant application or invention registration or patent. With more than 3,000 users annually there is a strong need to adopt processes that enable users to be productive as well as to enable efficient management by facility staff at all stages of the research experience. The Technique Finder (TF) is a web application that enables prospective facility users to identify the techniques most suited to their research, based on a researcher-centric approach and terminology as opposed to instrument-oriented jargon. Specifically, it offers two areas, one for biological scientists and another for researchers in physical sciences, which allow them to identify techniques based on research dimensions in corresponding fields. In addition, it offers a term search based on a comprehensive term index created for each technique including all the directly and indirectly linked information available in the application. The techniques themselves display a full description with sample examples, key reviews and potential for case studies and links. Locations and contact details to each of the AMMRF facilities invite users to get started immediately. To meet the training needs of microscopy core facilities, the AMMRF has developed MyScope: Training for Advanced Research [2]. MyScope is an online suite of education tools for teaching and learning in the area of microscopy and microanalysis. It comprises a range of modules including scanning electron microscopy; transmission electron microscopy; scanning probe and atomic force microscopy; microanalysis; confocal microscopy; and X-ray diffraction techniques. The modules provide a novel advancement in online training. They contain a number of components including: an interactive questionnaire to allow the user to assess their knowledge, guide choices and tailor the learning environment for flexible learning; also, tailoring capability for academics and trainers; self guided tutorials with videos, animations and glossary to prepare students with knowledge and specialist Microsc. Microanal. 21 (Suppl 3), 2015 528 language; virtual instrument platforms to practice use of instrumentation; and online competency testing to demonstrate readiness for hands-on experience [3]. Global interest in MyScope since it was launched in 2011 has grown markedly with more than 150,000 visits from over 20 countries in 2014. The Characterisaton Virtual Laboratory (CVL) is a cloud-based data analysis and visualisation platform that integrates existing analysis tools and techniques with a network of specialised cloud-based computing systems and data-storage facilities. Currently the CVL hosts tools for analysis of data from cryo-EM, X-ray microCT and atom probe platforms. Specifically the Atom Probe Workbench (APW) enables that enables the atom probe research community to access and create valuable tools, accelerating the research process. Features of the workbench includes an atom probe data management system that performs automatic ingestion of data & meta-data directly from instruments, processing-modules for accessing reconstruction and analysis techniques, a visualisation engine for exploring data and support for the preparation of publication-quality images . References: [1] S P Ringer and M H Apperley, Networking strategies of the microscopy community for improved utilization of advanced instruments: (1) The Australian Microscopy and Microanalysis Research Facility (AMMRF), C. R. Physique 15 (2014) 269�275. [2] B Cribb et al, Advanced microscopic characterisation through integrated learning tools, Microscopy and Microanalysis, Volume 17, Issue S2, July 2011, pp 870 � 871 [3] B Cribb et al, MyScope: a national approach to education in advanced microscopic characterization through integrated learning tools, Office for Learning and Teaching, Australian Government Department of Education, ISBN: 978-1-925092-03-5");sQ1[265]=new Array("../7337/0529.pdf","Evolution of Data Acquisition, Storage and Analysis in a Multi-user Facility","","529 doi:10.1017/S143192761500344X Paper No. 0265 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Evolution of Data Acquisition, Storage and Analysis in a Multi-user Facility Nestor J. Zaluzec Electron Microscopy Center, Nanoscience and Technology Division, Argonne National Laboratory, Argonne, Il, USA When the principle output of electron optical instruments consisted of images recorded on film, data storage and archiving was a relatively simple task. Each image was logged in an appropriate paper notebook and the negative or positive print filed in cabinet. Forty years ago I kept track of some 10,000 negative recorded during my thesis research using that method. Although that sounds like a daunting task, that was only for the images I recorded for my own research it did not include information for other users of the EM facility at the University of Illinois Materials Research Lab that also used our core instruments. Even more importantly, back in the 70's we were just in the early stages of developing both x-ray and electron loss spectroscopy in the instruments. To organize that data, which also needed archiving it was necessary to develop a scheme to do a similar task, however, the challenge was formatting and storage protocols as something as simple as used for film did not exist. Now that forty years have passed, and technology has changed several questions are relevant: 1.) can that data still be accessed, 2.) do we have software that can access archival data, 3.) how do we preserve current data and the associated meta-data which documents today's experiments, and 4.) how do we overcome the issue of obsolescence of modern computing accessory hardware on instruments which far out live the life time of a typical operating system. Obsolescence of peripheral computing hardware is one of the biggest challenges. Modern IT staff principally worry about keeping all systems up to date with the latest operating systems and off the shelf "office" software and stopping spam/virus attacks. They rarely if ever deal with scientific resources, those problems are relegated to the core facility staff. As the lifetime of major instruments can exceed a decade, I have found it is essential that core facilities maintain backup computational hardware and software resources to replace items which simply fail with age. Older hardware and software can be difficult to find and without it some instruments can become dead weight. Because custom software and hardware are integrated for controlling our complex, computationally mediated instruments it is not simply just the task of upgrading the hardware and software to keep up with the OS updates, we must insure that something as trivial as a $3K computer and a few hundred dollars of software takes down a multithousand/million dollar research too. Movement of data to servers which are not only backed up but more importantly connected using secure communications protocols is essential. Users should never be allowed to directly connect portable storage devices to core instruments, but rather data on core instrument should be transferred to remote servers and users required to download their data from therein. Using secure protocols similar to SFTP [1] or Globus tools sets [2] for these jobs should be standard procedure. Obsolescence of data formats is even more significance. Data (images, diffraction patterns, etc...) should never be stored in compressed image format (JPG, GIF) instead full raw data or at the least full bit depth files without compression should be used. Microsc. Microanal. 21 (Suppl 3), 2015 530 Manufacturer's formats are frequently used as substitutes, which is fine as long as software platforms are also archived that can read and process that data. If a software platform is about to be retired then translator program should be procured or developed to allow access to the data from the archive. Spectroscopic data should be handled no differently, and translators or alternative formats such as MSA or HMSA Spectral Image file formats should be considered [3-5]. Translators into public domain file formats also simplifies sharing of data with colleagues when commercial programs using proprietary formats are used to acquire data. Record keeping also needs to shift along with modern data needs. While it is sometimes possible to incorporate meta-data into the image or spectroscopic files which we are now storing, this really does not fully document our experiments. The traditional paper notebook which once followed me into the laboratory has been replaced by an electronic notebook. In it I can fully annotate an experiment, using not only handwritten notes, but in addition (and with much better readability) typed text as well as photographs of experimental setups and/or screen captures. The added ability to convert and upload the pages of an eNotebook into an archivable PDF file and store it concurrently with the data files it documents is not only a logical but necessary protocol. The last challenge is a simple monetary one. Namely whose responsibility is it to store and archive data. It the past, when we principally used film, all data was stored in the facility. Today that is not practical. Maintaining a reasonably sized file server of ~ 10-20 Tbytes is neither expensive nor difficult and that is a task which can be readily delegated to an IT group. The approach which is most reasonable to take, in a multi-user core facility, is that the individual user is responsible for the long-term archiving of their data sets. Temporary storage of data for a fixed period of time, the length of which can be determined by each local facility, on the facility servers is reasonable. Users are expected to use secure protocols to move data from the temporary storage facility to their own systems for off-line analysis and record keeping. By the way, as I am careful, as well as a packrat as many of my colleagues at ANL will attest to, I can still read and process the images and data, which I recorded back in the late 70's, but then again I also wrote some of the original code to store and process that data. References: [1] STFP http://en.wikipedia.org/wiki/SSH_File_Transfer_Protocol [2] Globus https://www.globus.org [3] RF Egerton, et al, Proc. of the EMSA, San Francisco Press, (1991) p. 526. [4] International Organization for Standardization, standard ISO 22029:2003. [5] A Torpy et al, Microscopy and Microanalysis 19 S2 (2013), p. 830 [6] Work supported by the U.S. DoE, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357 at the Electron Microscopy Center, Nanoscience and Technology Division of Argonne National Laboratory");sQ1[266]=new Array("../7337/0531.pdf","Detection of GFP-labeled Proteins by Electron Microscopy","","531 doi:10.1017/S1431927615003451 Paper No. 0266 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Detection of GFP-labeled Proteins by Electron Microscopy Robert G. Parton1,2 , Thomas Hall1 and Nicholas Ariotti1 1 2 Institute for Molecular Bioscience Centre for Microscopy and Microanalysis, The University of Queensland, Brisbane, Queensland 4072, Australia Reliable and quantifiable high-resolution protein localization is critical for understanding protein function. Green fluorescent protein (GFP) has revolutionized the detection of proteins by light microscopy. Similarly, new genetic tags for electron microscopy (EM) [1] have great potential for EM visualization of proteins in cells and, potentially, in whole organisms. We have sought to develop systems which allow EM detection of GFP-tagged proteins. We have pioneered the use of a processing method involving fast freezing, freeze substitution, and resin embedding that preserves GFP fluorescence and facilitates correlative light and electron microscopy (CLEM) in cells and tissues [2,3,4]. We show that in the absence of primary chemical fixation, excellent ultra-structure, preservation of GFP fluorescence, immunogold labelling and electron tomography can be obtained using a single technique involving high-pressure freezing and embedding in Lowicryl resins at low temperature. This method has been used to great effect to correlate dynamic membrane trafficking events in living cells with electron microscopic 3D ultrastructure of the same events [5]. This processing scheme has been further modified to involve ultra-rapid freeze substitution providing a simple fast method to correlate GFP fluorescence in sections with EM ultrastructure (Figs. 1 & 2). More recently we have developed a system for detection of GFP-tagged proteins in cells and tissues. This method significantly reduces the time required for the determination of subcellular protein distribution. We demonstrate that this technique can allow localization of numerous GFP-tagged proteins of unknown cellular distribution at high-resolution within a few days and is compatible with both electron tomography and serial blockface scanning electron microscopy. We also demonstrate that we can readily apply this method to the localization of GFP-tagged proteins to EM-resolution in vivo. These new methods can be used to complement light microscopic studies as a simple way to provide ultrastructural information on protein localization in a high throughput manner. References [1] J.D. Martell et al, Nat Biotechnol 30 (2012), p. 1143. [2] S.J. Nixon et al, (2009) Traffic 10 (2009), p. 131. [3] N.L. Schieber et al, (2010) 96 (2010), p. 425. [4] E.C. Thomas et al, PloS One 7 (2012), p. e51096. [5] W. Kukulski et al, Cell 150 (2012), p. 508. Microsc. Microanal. 21 (Suppl 3), 2015 532 1 2 Figure Legends. Figure 1 shows GFP fluorescence in the muscle of a 3-day zebrafish embryo expressing a plasma membrane protein, caveolin-3, fused to GFP. GFP fluorescence is associated with the muscle surface and the T-tubule system. Figure 2 shows a Lowicryl section from the same zebrafish embryo after fast freezing, rapid freeze substitution, and embedding in resin at low temperature. Note the excellent preservation of the GFP fluorescent signal after EM processing. Bars 10�m.");sQ1[267]=new Array("../7337/0533.pdf","Systematic characterization of fluorophore behavior in the presence of electron microscopy sample preparation reagents.","","533 doi:10.1017/S1431927615003463 Paper No. 0267 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Systematic characterization of fluorophore behavior in the presence of electron microscopy sample preparation reagents. Haiyan Li1, Danielle M. Jorgens1, Lei Wang2, Robert M. Stongin2, Joe W. Gray1,3,4 and Summer L. Gibbs1,3,4 1. 2. Department of Biomedical Engineering, Oregon Health and Science University, Portland, OR 97201. Department of Chemistry, Portland State University, Portland, OR 97201. 3. Knight Cancer Institute, Oregon Health and Science University, Portland, OR 97201. 4. OHSU Center for Spatial Systems Biomedicine, Oregon Health and Science University, Portland, OR 97201. Correlative light and electron microscopy (CLEM) combines the strengths of multicolor fluorescence labeling utilizing light microscopy (LM) with the near angstrom resolution of electron microscopy (EM) making it an ideal tool to interrogate biology at the nanoscale. However, combining these two microscopy techniques is not a trivial task, as sample preparation for fluorescence LM is typically completed using hydrated samples that are thick by comparison to the samples used for EM. Additionally, sample preparation for EM involves dehydration, staining with heavy metal contrast, and embedding in plastic-like resin, which are typically thought to be incompatible with the low levels of fluorescent staining desirable for molecularly targeted applications. In an effort to expand and advance efforts toward direct CLEM experimentation, we sought to holistically characterize commonly used fluorophores as substrates for correlative microscopy. In the current work, the sensitivity of four commonly used fluorophores including Atto 488, BODIPY FL, Cy3B, and Alexa Fluor 647 that represent three different fluorophore scaffolds, rhodamine, BODIPY, and cyanine, were thoroughly tested for their ability to withstand EM sample preparation. Each fluorophore was characterized in the presence of the typical chemicals used for EM sample preparation in such a way as to mimic both a benchtop environment (conventional) and also cryopreservation methods (high pressure freezing and freeze substitution). Fluorophores were characterized systematically through the addition of single EM staining reagents to determine the percentage of fluorescence decrease cased by each staining reagent before completing the entire EM sample preparation on fluorescently stained samples. We spectrally characterized fluorophores in-solution and conjugated to bovine serum albumin (BSA) in the presence of heavy metal stains including osmium tetroxide and uranyl acetate at concentrations ranging from 0.03 � 4%, as well as two common resins used for CLEM samples including Lowicryl HM20 and London Resin (LR) White over an eight hour time period. Some in-solution fluorescence decrease was seen in organic solvents, although no time dependence was observed pointing to solvent dependent fluorescence intensity rather than a time dependent interaction between the fluorophore and the solvent. Interestingly, LR White had a more detrimental affect over time on fluorescence intensity than Lowicryl HM20, a result that was utilized during our preparation of samples for CLEM studies. The same fluorophores were spectrally characterized conjugated to bovine serum albumin (BSA) to more closely mimic the environment within cells in the presence of varying concentration of heavy metals over time. Microsc. Microanal. 21 (Suppl 3), 2015 534 To further elucidate the relationship between protein fluorophore binding and the effect of heavy metal staining, wheat germ agglutinin (WGA) was conjugated to each fluorophore and used to label cell membranes both in suspension as well as adhered upon cover glass. In both samples significantly more fluorescence was seen at all concentrations of uranyl acetate staining as compared to osmium tetroxide staining, an effect previously reported (1, 2). Cell membrane visualization using CLEM was completed as a proof of concept for future work that will utilize the fluorophore preservation strategies we have determined through this study. References [1] Peddie CJ, et al. (2014) Correlative and integrated light and electron microscopy of in-resin GFP fluorescence, used to localise diacylglycerol in mammalian cells. Ultramicroscopy 143:3-14. doi:10.1016/j.ultramic.2014.02.001 PMID:24637200 [2] Biel SS, Kawaschinski K, Wittern KP, Hintze U, & Wepf R (2003) From tissue to cellular ultrastructure: closing the gap between micro- and nanostructural imaging. Journal of microscopy 212(Pt 1):91-99. doi:10.1046/j.1365-2818.2003.01227.x PMID:14516366");sQ1[268]=new Array("../7337/0535.pdf","Studying Membrane Trafficking in Toxoplasma gondii Using Correlative Light and Electron Microscopy (CLEM)","","535 doi:10.1017/S1431927615003475 Paper No. 0268 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Studying Membrane Trafficking in Toxoplasma gondii Using Correlative Light and Electron Microscopy (CLEM) Ru-ching Hsia1,2, John Strong1, Julia Romano3 and Isabelle Coppens3 1. 2. Electron Microscopy Core Imaging Facility, University of Maryland Baltimore, Baltimore , USA Department of Neural and Pain Sciences, University of Maryland Dental School, Baltimore, USA 3. Department of Molecular Microbiology and Immunology, Johns Hopkins University, Baltimore, USA Toxoplasma gondii is an obligate intracellular parasite that can cause fatal encephalitis in immunocompromised individuals. During infection, T. gondii multiplies in the cytoplasm of mammalian cells within a self-made membrane-bound compartment � the parasitophorous vacuole (PV) from which it acquires nutrients from the mammalian host. The PV membrane has a unique lipid and protein composition and does not fuse with endocytic or exocytic organelles in the host cytoplasm. Previous studies have shown that Toxoplasma acquires cholesterol and sphingolipids from their mammalian host [1]. It has been suggested that Toxoplasma targets several mammalian Rab GTPases, intercepts cellular traffic mediated by Rab recycling vesicles and Rab vesicles from the secretory pathway, and acquires nutrients such as lipids by macro-endocytosis of nutrient-filled vesicles into the PV. In this study, we examined the ultrastructure of host GFP-Rab vesicles located near and within the PV using correlative light and electron microscopy (CLEM). HeLa cells transfected with GFP-tagged Rab GTPase and infected with T. gondii (expressing RFP) were grown on 10 mm glass coverslips with a locator grid pattern generated with a carbon coater. Cells containing GFP-tagged Rab vesicles and RFP-tagged Toxoplasma were identified by fluorescence microscopy and their locations were recorded according to the locator pattern. Serial optical section images of several cells of interest were collected using a Nikon Eclipse E800 microscope. After light microscopy (LM) imaging, cells were fixed, dehydrated, and embedded in epoxy resin directly on the coverslips. The cells of interest were located in the resin under the dissecting microscope. Serial ultrathin sections were collected on slot grids and imaged in a transmission electron microscope (TEM) at 80 keV (Tecnai T12, FEI). Photoshop CS2 and Image J were used to analyze the LM and EM images. A Toxoplasma-infected HeLa cell expressing GFP-Rab was selected for further CLEM analysis (Figure 1a). Two T. gondii PVs (PV-a and PV-b) were present in the cell with punctuate GFP-Rab signal (white arrows) closely associated with or trapped within the PVs (Figure 2). We identified five GFP-Rab positive targets and further analyzed their ultrastructure and temporal and spatial relationship with the parasites and PV by EM. Approximately 60 Serial sections (70 nm thickness) were collected and analyzed by TEM to identify the specific sections corresponding to the LM optical sections. A group of 5 serial sections correlating to one of the GFP-Rab positive area is presented in Figure 3. We describe here a CLEM workflow appropriate for the study of membrane trafficking in Toxoplasma gondii-infected cells. Our results confirmed that GFP-Rab positive LM signals observed adjacent to the Toxoplasma PV consist of double membrane vesicles. These vesicles are most likely Rab-GFP positive vesicles enclosed within the Toxoplasma PV membrane confirming our hypothesis that host nutrientfilled vesicles are taken up by the parasite via a process similar to phagocytosis. Microsc. Microanal. 21 (Suppl 3), 2015 536 References: [1] Coppens, I., Sinai, A.P., Joiner, K.A. J. Cell Biol 149 (2000), p. 167. Figure 1. Localization of T. gondii-infected Hela cell that also expresses GFP-Rab by light microscopy. Bright field LM images showing a cell of interest and the locator grid pattern at 200X (A) magnification. Fluorescence images of one cell (white arrow) located in E4 region containing two Toxoplasma PVs was chosen for CLEM analysis (B and C). One optical section LM image is shown in D. Figure 2. Correlation of the TEM image (A) with one of the optical section LM image of the same T. gondii infected cell. White arrows indicate host Rab GTPase vesicles trapped near or within PV (B). Overlay of LM and EM images reveals the ultrastructure of the corresponding LM signals (C) Figure 3. TEM images confirm that the ultrastructure of the vesicles correlate to GFP-Rab signal in PV-b by LM. Five serial sections (A) illustrate the temporal and spatial relationship of the vesicles, the parasite and the PV. GFP-Rab signal in LM is overlayed on the TEM image (B) to reveal the vesicle ultrastructure and the neighboring parasite (C).");sQ1[269]=new Array("../7337/0537.pdf","New CLEM Method to Reveal Ultrastructural Reorganization in the Host Cell during Coronavirus Infection","","537 doi:10.1017/S1431927615003487 Paper No. 0269 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 New CLEM Method to Reveal Ultrastructural Reorganization in the Host Cell during Coronavirus Infection Gabriella Kiss1, Cedric Bouchet-Marquis1, Lee Pullan1, Doug Keene2, Benjamin W. Neuman3 1 2 FEI Company, Hillsboro, OR, USA Shriners Hospital, Micro-Imaging Center, Portland, OR, USA 3 University of Reading, UK Correlative light and electron microscopy (CLEM) combines localization data from fluorescent microscopy (FM) with ultra-structural information from electron microscopy (EM). Recent improvements on fluorescent dyes and proteins enabled wider applications of this approach [1][2]. One of the main difficulties in CLEM is that often time, sample processing for EM imaging quenches the FM signal. Despite those difficulties, some newer labeling methods enable to perform CLEM experiments with a high level of success. The study of mouse hepatitis virus replication in murine cells described below demonstrates the most recent improvement made in this field. Mouse hepatitis virus (MHV) belongs to the Coronavirus family. Alongside other positive-strand RNA viruses, MHV induce many subcellular changes in the host cell during infection [3]. These subcellular reorganizations mainly affect lipid structures and leads to further compartmentalization. This compartmentalization has been shown to be very important for the viral replication both functionally and strategically [4]. During this process, MHV reshapes the endoplasmic reticulum (ER) membranes into double membrane vesicles (DMVs). These DMVs contain the viral RNA to protect it from the host cell antiviral machinery. After the structural proteins are also transported to the DMVs, viral budding can occur. DMVs are thought to form a network with actively growing and maturing in the infected cell according to the stage of infection, thus 3 dimensional (3D) structural studies are necessary to fully understand their organization. Previous studies revealed many similarities among +RNA viruses which can help to better understand viral infection mechanistic [5][6][4][7][8][9][10][11]. There is still an ongoing need to find new ways to study these structural changes during infection in the host cell. Here we demonstrate using SYBR-Gold (Life Technologies) nucleic acid stain before EM sample processing to locate the viral factories (e.g. DMVs) in the infected cells directly. We have compared two EM sample processing methods, using LR White resin without OsO4 and a Standard EM method using EPON hard resin with 1% OsO4 to compare the effect of OsO4 on the FM signal. OsO4 is known to quench the FM signal, albeit necessary for ultrastructural preservation and EM contrast [12]. As expected the OsO4 weakened the FM signal in the Standard processing method, however SYBR-Gold retained enough signal for detection, even at single DMV level. Additionally the EM contrast in the samples was good to enable tilt series data collection on regions with DMVs and budded virus particles. Microsc. Microanal. 21 (Suppl 3), 2015 538 Figure 1. Fluorescent (FM) and electron microscopy (EM) images of the samples processed using LRW no OsO4 or the Standard method with OsO4 treatment. The FM signal is still visible from individual DMVs containing the viral RNA in both processing method. The quality of the EM images are also suitable for further 3D ultrastructural studies of the DMV network. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] H. Xiong, et al, Nat. Commun., vol. 5, p. 3992, Jan. 2014. M. G. Paez-Segala, et al, Nat. Methods, Jan. 2015. K. Knoops, et al, J. Virol., vol. 86, no. 5, pp. 2474�87, Mar. 2012. B. G. Kopek, et al, PLoS Biol., vol. 5, no. 9, p. e220, Sep. 2007. L. K. Gillespie, et al, J. Virol., vol. 84, no. 20, pp. 10438�47, Oct. 2010. K. Knoops, et al, PLoS Biol., vol. 6, no. 9, p. e226, Sep. 2008. R. W. A. L. Limpens, et al, MBio, vol. 2, no. 5, Jan. 2011. P. Monaghan, et al, J. Gen. Virol., vol. 85, no. Pt 4, pp. 933�46, Apr. 2004. A. Schlegel, et al, J. Virol., vol. 70, no. 10, pp. 6576�88, Oct. 1996. E. J. Snijder, et al, J. Virol., vol. 80, no. 12, pp. 5927�40, Jun. 2006. S. Welsch, et al, Cell Host Microbe, vol. 5, no. 4, pp. 365�75, Apr. 2009. S. Watanabe, et al, Nat. Methods, vol. 8, no. 1, pp. 80�4, Jan. 2011.");sQ1[270]=new Array("../7337/0539.pdf","On the road to large volumes in LM and SEM: New tools for Array Tomography","","539 doi:10.1017/S1431927615003499 Paper No. 0270 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 On the road to large volumes in LM and SEM: New tools for Array Tomography Irene Wacker1,2, Waldemar Spomer2,3, Andreas Hofmann2,3, Ulrich Gengenbach2,3, Marlene Thaler4, Len Ness5, Pat Brey5, and Rasmus R. Schr�der2,6 Cryo EM, Centre for Advanced Materials, Universit�t Heidelberg, Heidelberg, Germany HEiKA, Heidelberg Karlsruhe Research Partnership, Heidelberg, Karlsruhe, Germany 3 Institute for Applied Computer Science, Karlsruhe Institute of Technology, Karlsruhe, Germany 4 Carl Zeiss Microscopy GmbH, Oberkochen, Germany 5 RMC Boeckeler, Tucson, Arizona, USA 6 Cryo EM, CellNetworks, BioQuant, Universit�tsklinikum Heidelberg, Heidelberg, Germany 2 1 To reconstruct large volumes at ultrastructural resolution a number of methods are available [1]. Serial blockface or focussed ion beam scanning electron microscopy (SBFSEM or FIBSEM) are inherently destructive: After many alternating cycles of blockface imaging and subsequent removal of a thin material layer to expose a new blockface the sample will be gone. Array tomography (AT) [2] on the other hand offers the possibility of imaging at different resolutions and with different techniques, such as (fluorescence) light microscopy (LM) and SEM. It therefore has the greatest potential for correlative imaging. When aiming at large volumes, a high degree of automation during both, sample preparation and imaging procedure is desirable. One way to automate the generation of the thousands of serial sections necessary for brain mapping was introduced as ATUMtome [3] and is now commercially available for early adopters (RMC Boeckeler). Because the sections are collected on carbon-coated Kapton tape epifluorescence LM is possible in principle. However, super-resolution LM methods [4] do not work on such non-transparent tape. To overcome this limitation we developed a novel device enabling reliable collection of long ribbons of serial sections (cf Spomer et al., this conference) onto a super-resolution LM-compatible solid substrate such as ITO-coated glass coverslips or onto silicon wafers for SEM imaging only. Samples can be prepared either by classical aldehyde-based chemical fixation or by cryofixation and freeze substitution, both followed by epoxide embedding. For preservation of fluorescence signals or for post-embedding immuno-labeling, hydrophilic resins such as e.g. lowicryls are advisable. Blocks are trimmed and coated with a thin layer of glue on the leading and trailing edges to stabilize the ribbons. Substrates are hydrophilized by glow discharge and immersed into the boat of a Jumbo knife (Diatome) using the novel substrate holder. After cutting ribbons of ultrathin (down to 50nm) or semithin (up to 3�m) sections, several ribbons are attached to the substrate (Fig. 1A) and gently lifted from the water. Fig. 1B shows the same four ribbons on a silicon wafer after lift-out and drying. Similarly, ITO-coated glass coverslips can be used as substrates (Fig. 1C). Sections may be either stained with fluorescent dyes or antibodies for LM or with uranylacetate and lead citrate for SEM imaging. With sections up to 200nm thickness high resolution SEM imaging is possible without further coating, using either secondary or back-scattered electrons. To image large numbers of sections automation of the imaging process is necessary. The newly released product ZEISS Atlas 5 Array Tomography (Carl Zeiss Microscopy GmbH) allows repeated imaging of the sample at increasing resolution. Usually an overview of the whole substrate (Fig. 2A) is recorded first, at a pixel size of several 100nm. Then regions of interest (ROI), for example individual sections or parts thereof (Fig. 2B, C) are recorded with intermediate pixel sizes. Finally Microsc. Microanal. 21 (Suppl 3), 2015 540 single cells or cell groups (Fig. 2D) can be selected and imaged automatically at pixel sizes down to 2-5 nm (Fig. 2C-D) and over large numbers of sections to reconstruct whole cells. [1] [2] [3] [4] [5] I Wacker, RR Schr�der, J Microscopy 252 (2013), p93-99. KD Micheva, DJ Smith, Neuron 55 (2007), p25-36. Erratum in: Neuron 55 (2007), p. 824. KJ Hayworth et al., Front Neural Circuits 8 (2014), article 68, p1-18. S Nanguneri, et al., PLoS One 7 (2012), e38098. The authors acknowledge C Bartels and L Veith for technical support, C Grabher for samples, BMBF for NanoCombine grant FKZ 13N11401 and MorphiQuant grant FKZ 13GW0044, and HEiKA for initial funding of the substrate holder development. Fig. 1 Production of arrays of serial sections: A) Four ribbons floating in knife boat, attached to silicon wafer by their lower ends. B) The same ribbons (175 consecutive sections in total, about 40 sections per ribbon) dried onto wafer after lift-out. C) Six ribbons (about 200 sections in total) deposited between the fiducials (L-marks) on an ITO-coated glass coverslip. Scale bars 10 mm Fig. 2 Automatic imaging of arrays using ZEISS Atlas 5 Array Tomography solution: A) Overlay of ribbon images (SEM) on imported image of whole substrate (digital camera) used for initial navigation, B) individual sections, cell pellets (blue outline) imaged at 50nm pixel size, C) detail of B) with ROI (blue rectangle) imaged at 5nm pixel size, D) Golgi region from cell marked by red circle in C)");sQ1[271]=new Array("../7337/0541.pdf","Modeling Protein Structure in Macromolecular Assemblies at Near Atomic Resolutions","","541 doi:10.1017/S1431927615003505 Paper No. 0271 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Modeling Protein Structure in Macromolecular Assemblies at Near Atomic Resolutions Matthew L. Baker1, Muyuan Chen1,2, Philip R. Baldwin1, Steven J. Ludtke1, Wah Chiu1 National Center for Macromolecular Imaging, Verna and Marrs McLean Department of Biochemistry and Molecular Biology, Baylor College of Medicine, Houston, TX USA 2. Program in Structural and Computational Biology, Baylor College of Medicine, Houston, TX USA In nearly every cellular process, macromolecular machines play critical roles. As such, understanding the structure of these complexes is critical in preventing disease and developing efficacious treatments. However, structural studies of such large complexes, typically by electron cryo-microscopy or X-ray crystallography are often difficult and result in structures with non-atomic resolutions. Limited resolvability and noise in the density map can complicate direct interpretation, and as such, model construction at near-atomic resolutions is generally not automated and often results in only C only models[1]. Previously, we developed the Pathwalking protocol, which semi-automatically enumerates putative configurations of protein models from a density map[2,3]. Pathwaking is based on the Traveling Salesman Problem (TSP), in which possible cyclical paths (i.e. protein fold) are calculated through a density map without using any sequence or structure constraints. A TSP solver is used to find a path through a set of pseudoatoms by optimizing the spatial distance between the pseudoatoms such that they are representative of C-C distances in consecutive amino acids in the protein structure. Here, the only required inputs are a density map better than 6� resolution and the number of amino acids in the protein of interest, which is used for the initial seeding of pseudoatoms in the density map. In the initial testing of Pathwalking, reasonable first-approach models, models with the correct overall folds, were derived directly from the density map with limited user intervention. However, Pathwalking was not designed to directly consider protein chemistry or density map constraint, and as such, nonprotein like connections were sometimes observed and required the user to correct. Building on the success of our first implementation of Pathwalking and the rapidly growing number of near atomic resolution structures, we developed an enhanced version of our original protocol capable of producing more accurate models with reduced user interaction. In the new version of Pathwalking, all of the interactive steps in the original version, including identifying secondary structures assignment, pseudoatom placement and path evaluation, have been optimized for nearly automated usage. Additionally, Pathwalking improvements in the implementation of our TSP-based search now allow for modeling multiple chains in a density map simultaneously. In testing of our new Pathwalking protocol, we have not only improved the ease of use but have also increased the accuracy of our models. In our benchmark of 20 authentic density maps between 3.2� and 7� resolution, Pathwalking models averaged ~69% structural overlap, 2.35� RMSD and ~67% correctly registered Cs when compared to the known structure. Errors in modeling were generally restricted to register shifts and improperly placed pseudoatoms due to map noise/resolution, though these errors generally did not affect the overall model topology. Three examples of Pathwalking on cryo-EM and X-ray crystallographic density maps from 34� resolution are in Figure 1. 1. Microsc. Microanal. 21 (Suppl 3), 2015 542 References: [1] Baker ML, Zhang J, Ludtke SJ, Chiu W. Cryo-EM of macromolecular assemblies at near-atomic resolution. Nat Protoc. 2010 Sep;5(10):1697-708. PMCID: PMC3107675. [2] Baker MR, Rees I, Ludtke SJ, Chiu W, Baker ML. Constructing and validating initial C models from subnanometer resolution density maps with pathwalking. Structure. 2012 Mar 7;20(3):450-63. PMCID: PMC3307788. [3] Baker ML, Baker MR, Hryc CF, Ju T, Chiu W. Gorgon and pathwalking: macromolecular modeling tools for subnanometer resolution density maps. Biopolymers. 2012 Sep;97(9):655-68 PMCID: PMC3899894. [4] Wang Z, Hryc CF, Bammes B, Afonine PV, Jakana J, Chen DH, Liu X, Baker ML, Kao C, Ludtke SJ, Schmid MF, Adams PD, Chiu W. An atomic model of brome mosaic virus using direct electron detection and real-space optimization. Nat Commun. 2014 Sep 4;5:4808. PMCID: PMC4155512. [5] This work was supported by grants from the National Institutes of Health (P41GM103832, R01GM079429, R21GM100229), and National Science Foundation (DBI-1356306). Density map Pathwalking Model BMV (cryo-EM) Structural overlap: 80% RMSD: 2.05� Topology Score: 1.0 Correctly registered Cs: 80% DNA ligase IV:C (X-ray) Structural overlap: 62% RMSD: 2.61� Topology Score: 1.0 Correctly registered Cs: 80% 80S Ribosome (cryo-EM) Structural overlap: 95% RMSD: 1.94� Topology Score: 0.83 Correctly registered Cs: 82% Figure 1. Pathwalking at near atomic resolutions. Shown are two examples of the new Pathwalking protocol. In the left panels, the density maps for the structural protein from Brome Mosiac Virus [4] (top row, EMDB ID:6000), subunit C from the DNA ligase IV complex (PDB ID: 1Z56, middle row) and the 80S ribosome (chains C,I and M only) (bottom row, EMDB ID:2566). In the middle panel, the Pathwalking model is shown overlaid on the density map. Model quality statistics are shown in the right panel. For the 80S ribosome example, a portion of the map was segmented and contained density for only 3 chains. Pathwalking results for the 80S ribosome data are averaged over all three subunits.");sQ1[272]=new Array("../7337/0543.pdf","IP3R1 � Assessing Map Interpretability at Near Atomic Resolution","","543 doi:10.1017/S1431927615003517 Paper No. 0272 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 IP3R1 � Assessing Map Interpretability at Near Atomic Resolution Steven J. Ludtke2, Guizhen Fan1, Matthew L. Baker2, Zhao Wang2, Mariah R. Baker1, Stephen Murray2, Pavel A. Synyagovsky1, Wah Chiu2, Irina I. Serysheva1 1. Department of Biochemistry and Molecular Biology, Structural Biology Imaging Center, University of Texas at Houston Medical School, Houston, TX, USA. 2. National Center for Macromolecular Imaging, Verna and Marrs McLean Department of Biochemistry and Molecular Biology, Baylor College of Medicine, Houston, TX , USA. Once resolutions beyond ~6 � have been achieved in single particle reconstruction it becomes possible to begin interpreting maps in terms of the protein sequence. At the low resolution extreme, C-alpha tracing and homology modeling can be applied. As resolution increases towards 4 �, beta strand separation becomes clear, and larger sidechain densities begin to appear. However, simply citing the resolution of a map, regardless of criteria, does not provide sufficient information to assess the appropriate level of interpretability. It is always possible to filter maps such that, for example, sidechain densities appear, even when such densities are likely due primarily to noise. If the same data were refined independently in two different software packages or with two somewhat different starting maps, the final maps will never be identical near the resolution limit, since the 0.143 FSC threshold represents a signal to noise ratio of only 0.33. That is, as we approach the limiting resolution of our map, we know that an increasing fraction of the observed features are due to noise, and yet this process is gradual enough that establishing a specific hard limit on interpretation is not feasible. IP3R1 is a ubiquitous ion channel located in the ER, responsible for the release of Ca2+ and involved in a broad array of cellular processes. We have solved the structure of IP3R1 to a gold-standard resolution of 4.7 �, permitting us to unambiguously trace ~85% of the primary sequence through the map, and through a combination of de-novo modeling, homology modeling, and crystal structures of fragments, constructed an all-atom model. However, this raises the question of whether at this resolution it is appropriate to even consider an all-atom model, which could be easily over-interpreted. To help assess the level of interpretability of our map and model, we performed a number of assessments and validations. First, we performed completely independent refinements of the particle data in both EMAN2.1 [1] and Relion [2]. The resulting structures were virtually identical, with the EMAN map resolving some peripheral density better than Relion, and Relion showing more sidechain detail in a few helices near the core of the structure. After performing real-space refinement of the allatom model against the two maps using PHENIX, an RMSD of only 0.8 � was achieved. While clearly the specific atomic coordinates were heavily influenced by molecular mechanics to achieve this value, it does indicate that no significant disagreements exist between the maps. Applying Resmap [3], we see that the apparent level of detail is significantly higher in some regions of the map than in others (Fig 1). This `resolution' varies from 3.6 � in the transmembrane domain to 6.5 � in parts of the cytoplasmic domain. This clearly indicates some level of variability is present in the underlying data. This could be due to actual structural flexibility and/or the presence of minor components of the other two highly homologous isoforms of IP3R in a portion of the particles. Clearly, though, the simple statement that we have "4.7 � resolution" is insufficient as an assessment of the map. Microsc. Microanal. 21 (Suppl 3), 2015 544 The logical next step is to ask how well the atomistic model represents the map density. After converting the model to a density map we can compute a map vs. model FSC curve. We can also compute an FSC curve between the EMAN2.1 and RELION1.3 maps. As shown in Fig 2, if we interpret these curves with the 0.25 threshold mandated by use of all (rather than half) of the data in the maps, we see agreement to only ~5.9 � resolution, on average. We can also compute a localized version of the FSC, and again observe that self consistency is much improved in similar areas to those observed by Resmap. If we take the most conservative position, and considered the map to be only 5.9 � resolution, it would be ridiculous to claim any sort of significant sidechain interpretation. However, this is an overly pessimistic view. As a simple example, the presence of apparent phenylalanine residues at exactly the location where they would be expected to act as a gate blocking the closed channel would seem to reaffirm that the registration of the primary sequence with our density map is correct, at least in that region. Similar observations can be made elsewhere in the map. Clearly map interpretability is a complex issue, not easily encompassed by any single number, or even by any single procedure. We will consider these issues, and various strategies for quantifying them, in the context of IP3R1 and other recently solved structures. References: [1] G Tang et al., J Struct Biol 157 (2007), p.38-46. [2] SH Scheres, J Struct Biol 180 (2012), p.519-30. [3] A Kucukelbir, FJ Sigworth and HD Tagare, Nat Methods 11 (2014), p.63-5. [4] This work was supported by grants from the National Institutes of Health (R01GM072804, R21AR063255, S10OD016279, P41GM103832, R01GM079429, R01GM080139, and R21GM100229), Department of Defense (R038318-I), the American Heart Association (14RNT1980029), the Muscular Dystrophy Association (295138) and National Science Foundation (DBI-1356306). Figure 1. Final reconstruction sliced in half, colored by Resmap value from 4-6 �. Figure 2. Map vs. model (after real-space refinement), EMAN vs. RELION and gold standard FSC");sQ1[273]=new Array("../7337/0545.pdf","Exploiting the Susceptibility of HIV-1 Nucleocapsid Protein to Radiation Damage in Tomo-Bubblegram Imaging.","","545 doi:10.1017/S1431927615003529 Paper No. 0273 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Exploiting the Susceptibility of HIV-1 Nucleocapsid Protein to Radiation Damage in Tomo-Bubblegram Imaging. Juan Fontana1, Kellie A. Jurado2, Naiqian Cheng1, Alan Engelman2, and Alasdair C. Steven1. 1 Laboratory of Structural Biology Research, National Institute of Arthritis, Musculoskeletal and Skin Diseases, National Institutes of Health, Bethesda, MD 20892, USA 2 Department of Cancer Immunology and AIDS, Dana-Farber Cancer Institute, Boston, MA 02215, USA During HIV-1 maturation, a process required for viral infectivity, the capsid protein CA assembles into a conical shell that is thought to contain the viral ribonucleoprotein (vRNP) complex comprised of the viral RNA (vRNA) and nucleocapsid protein NC. Maturation is inhibited by different classes of antiHIV drugs, including allosteric integrase inhibitors (ALLINIs) [1]. To further understand the maturation process and the mode of action of these drugs, we imaged HIV-1 virions produced in the presence of ALLINIs by cryo-electron tomography (cryo-ET) and "tomo-bubblegram" imaging. The latter is a novel labeling technique that extends "bubblegram imaging" [2, 3] into three dimensions. In the HIV-1 system, we found that it exploits the exceptional susceptibility of NC to radiation damage. Bubblegram imaging takes advantage of the fact that vitrified protein specimens subjected to relatively high levels of electron irradiation generate bubbles of hydrogen gas that are readily visible. Moreover, proteins in complex with DNA bubble relatively early because the DNA impedes the diffusion of radiation products, accelerating the formation of bubbles. Hypothesizing that some component(s) of HIV may have distinctive bubbling behavior, we visualized HIV-1 virions in this way, and found that bubbles started to appear after a cumulative dose of ~160-200 e-/�2. We then extended the approach to tomobubblegrams. First, a regular tilt-series was collected and a tomogram was calculated (total dose of ~75 e-/�2). Second, an untilted dose-series was performed, stopping after the first bubbles began to develop (another ~70 e-/�2). Third, after waiting ~2 h to allow radiation products to dissipate, a second tilt series was recorded (~70 e-/�2), allowing the visualization of the bubbles in the "tomo-bubblegram". Imaged by cryo-ET (Figure 1), most wild-type virions have conical cores with internal density that is thought to be the vRNP. In contrast, ALLINI-treated virions have few such cores and instead contain "eccentric condensates", spheroidal aggregates that are located outside of the capsid and close to the viral envelope [1]. When an ALLINI-treated virion contains both an eccentric condensate and a conical core, the core is relatively empty (cf. Figures 1A & B). In virions visualized by bubblegram imaging, the bubbles co-project with core interiors and eccentric condensates, and the two structures have the same bubbling threshold (Figure 2). Tomo-bubblegram imaging confirms that the bubbles specifically label the contents of filled cores and the eccentric condensates (Figures 3A & B). To further clarify which virion component is bubbling, we examined immature virions, which have a thick-walled shell in which the domains of the polyproteins Gag and Gag-Pol are radially ordered. Here, the bubbles appear in the inner part of that shell (Figure 3C). The interpretation of NC as the bubbling component was further supported by analyses of Gag/-minus particles, which do not incorporate vRNA, (Figure 3D); of GagLeuZip particles, in which the NC domain is replaced by a leucine zipper (Figure 3E); and of purified recombinant NC protein (not shown). Taken together, these observations provide strong evidence that ALLINIs act by sabotaging capsid assembly and vRNP encapsidation. They also show that proteins differ in their bubbling thresholds and that nucleic acid is not needed to induce bubbling. Microsc. Microanal. 21 (Suppl 3), 2015 546 References: [1] KA Jurado et al., Proc Natl Acad Sci U S A 110 (2013), 8690-5. [2] W Wu et al., Science 335 (2012), 182. [3] N Cheng et al., J Struct Biol 185 (2014), 250-6. [4] This work was funded in part by the Intramural Research Program of NIAMS (A.C.S.), the Intramural AIDS Targeted Antiviral Program (A.C.S), and National Institutes of Health grants AI070042 and GM103368 (A.E.). Figure 1. Tomographic central slices of HIV-1 virions classified according to core morphology (conical or non-conical) and the presence (white arrows) or absence of an eccentric condensate. Bar, 50 nm. Figure 2. Bubblegram imaging. Black arrowhead, filled core; white arrowhead, core with less content; black arrow, eccentric aggregate; white arrows, first bubbles. Bar, 50 nm. Figure 3. Tomo-bubblegrams. Each pair of panels shows a central section from the initial tomogram (top) and the corresponding section from the tomo-bubblegram (bottom). Bar, 50 nm.");sQ1[274]=new Array("../7337/0547.pdf","Affinity Cryo-Electron Microscopy Studies of Viral Particles Captured Directly From Cell Culture","","547 doi:10.1017/S1431927615003530 Paper No. 0274 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Affinity Cryo-Electron Microscopy Studies of Viral Particles Captured Directly From Cell Culture Ci Zhang1*, Frank Vago1*, Fei Guo1, Zheng Liu1, Guimei Yu1, Phil Serwer2, and Wen Jiang1 Markey Center for Structural Biology, Department of Biological Sciences, Purdue University, West Lafayette, IN 47907 2 Department of Biochemistry, University of Texas Health Science Center, San Antonio, TX 78229 *Equal contribution Single particle cryo-electron microscopy (cryo-EM) and 3-D reconstruction have emerged in recent years as a powerful tool for solving high-resolution 3-D structures of viruses and macromolecular complexes. However, sample preparation for single-particle cryo-EM typically relies on a dedicated purification step before EM sample grid preparation, for example, using column chromatography or gradient centrifugation, to obtain homogeneous samples at high concentration. This strategy usually uses large volume of cell culture (hundreds of milli-liters to multiple liters) and takes hours to days to complete. While stable major states of the sample can be successfully prepared, many of the short-lived intermediate states won't be able to survive the lengthy duration and those minor low abundance states are prone to being ignored. For example, until our attempt with T3 phage [1], tailed dsDNA phage intermediates with incompletely packaged dsDNA genome eluded cryo-EM investigation due to their instability and the faint band visibility in the gradient centrifugation due to low abundance. Therefore, improved sample preparation methods, for example, the affinity cryo-EM grids [2-5], are needed to allow sampling of all states, including the minor, transient intermediates, to fully cover the complete process of the targets for better understanding of structures and functions. Four different affinity grid techniques have been developed for single-particle cryo-EM sample preparation: the Ni�NTA lipid monolayer method [5], the streptavidin 2D crystal-based method [4], the functionalized carbon film method [3], and single-step antibody-based method [2]. However, no structures of subnanometer or higher resolutions using these affinity-grid approaches have been reported and there are concerns if the affinity grids will limit the reconstructions to low resolutions. Here we employed the Ni-NTA lipid monolayer based affinity grid to capture His-tagged T7 phage particles directly from small volume of bacterial lysate for single-particle cryo-EM imaging and 3-D reconstruction. We used the T7Select� phage display system (Merck Chemicals) to construct a new T7 phage variant with 6x His-tag at the C-terminus of all 415 copies of major capsid protein gp10 in the phage particles. Plaque assays showed that the new T7His phage has normal infectivity as wild type T7 phage. Lipid monolayer based affinity EM grids were prepared using a 10:1 weight ratio of DOPC and Ni-NTA lipid. To maximally capture all possible T7 particle states including low abundance or shortlived intermediate states, Ni-NTA affinity grids were soaked with cell lysate immediately after the cells being lysed by the inoculated T7 phage. Cryo-EM was then used to gather a large set of micrographs of the captured particles using the Titan Krios microscope at low dose conditions. From the micrographs, we could observe a large number of particles of various sizes, shapes, and aggregation states (Figure 1). Most of the particles can be visually classified into the three major states of T7 capsids, capsid I, capsid II, and mature phage, which have been thoroughly characterized in past cryo-EM studies using particles purified with traditional centrifugation method [6]. Many additional unusual states are also observed, for example, tubes, prolate particles, "Pac-man"-like particles, 1 Microsc. Microanal. 21 (Suppl 3), 2015 548 paracrystalline-like fiber clusters, and lipid wrapped particles. The three major states, capsid I, capsid II, and mature phage, of T7 phage particles represent three different steps in the T7 maturation [6]. We could determine the icosahedral structure of all three states to subnanometers resolutions of capsid I at 6.9 � resolution, capsid II at 8.1 � resolution, and mature phage at 7 � resolution, respectively, based on "gold" standard Fourier Shell Correlation criterion (Figure. 2). From the icosahedral shell densities of all three structures, -helices and -sheets can be identified as rod-like and curved flat shape features, respectively, and the HK97-like protein fold [7] can also be readily recognized. Our results suggest that affinity cryo-EM would be a general approach suitable for imaging and high resolution structural studies of the dynamics process of a wide range of biological systems [8]. References: [1] PA Fang, ET Wright, ST Weintraub et al. Visualization of bacteriophage T3 capsids with DNA incompletely packaged in vivo. J Mol Biol 384 (2008), p. 1384-1399 [2] G Yu, F Vago, D Zhang et al. Single-step antibody-based affinity cryo-electron microscopy for imaging and structural analysis of macromolecular assemblies. J Struct Biol 187 (2014), p. 1-9 [3] MC Llaguno, H Xu, L Shi et al. Chemically functionalized carbon films for single molecule imaging. J Struct Biol 185 (2014), p. 405-417 [4] BG Han, RW Walton, A Song et al. Electron microscopy of biotinylated protein complexes bound to streptavidin monolayer crystals. J Struct Biol 180 (2012), p. 249-253 [5] DF Kelly, D Dukovski & T Walz Monolayer purification: a rapid method for isolating protein complexes for single-particle electron microscopy. Proc Natl Acad Sci U S A 105 (2008), p. 47034708 [6] F Guo, Z Liu, P Fang et al. Capsid expansion mechanism of bacteriophage T7 revealed by multistate atomic models derived from cryo-EM reconstructions. Proceedings of the National Academy of Sciences of the United States of America 111 (2014), p. E4606-14 [7] WR Wikoff, L Liljas, RL Duda et al. Topologically linked protein rings in the bacteriophage HK97 capsid. Science 289 (2000), p. 2129-2133 [8] The authors acknowledge funding from the National Institutes of Health and Welch Foundation. The cryo-EM images were taken in the Purdue Cryo-EM Facility, and the Purdue Rosen Center for Advanced Computing provided the computational resource for the 3D reconstructions. a a Capsid II Capsid I c Virion b c Tube 50 nm b d N"arm& Domain&A& e N"arm& Domain&A& d e f g h i Domain&P& E"loop& Domain&P& E"loop& Figure 1. Gallery of T7His phage particles affinity-captured from E. coli lysate. Figure 2. 3-D reconstructions of capsid I, capsid II and virion affinity-captured from lysate.");sQ1[275]=new Array("../7337/0549.pdf","Using Microscopy to Qualitatively and Quantitatively Assess Crystalline Content in Amorphous Active Pharmaceutical Ingredients and Drug Product","","549 doi:10.1017/S1431927615003542 Paper No. 0275 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Using Microscopy to Qualitatively and Quantitatively Assess Crystalline Content in Amorphous Active Pharmaceutical Ingredients and Drug Product Andrew D. Vogt and Joseph P. Neilly AbbVie Inc, Drug Product Development, NCE Analytical R&D, North Chicago, IL USA. Amorphous active pharmaceutical ingredients (APIs) have improved solubility characteristics over their crystalline analogs. As consequence, amorphous APIs can have improved bioavailability. These improved properties often come at the expense of chemical and physical stability for the API [1]. The presence and quantity of the crystalline form in the amorphous material should be assessed to ensure manufacturing reproducibility and high quality API for its intended use. It is also important to monitor crystalline API in an amorphous drug product to ensure there is no form conversion that might affect the drug product (DP) over its shelf life. Amorphous materials have no long-range order compared to their crystalline analogs and as a result can be identified by the absence of any sharp peaks in an x-ray powder diffraction (XRPD) pattern (Figure 1). The presence of sharp peaks in a diffraction pattern indicates that a material has crystalline character. XRPD has been able to detect crystalline content in amorphous API down to about 2% w/w [2], but this is highly dependent on the API. In addition to XRPD, other techniques that have been used to identify and/or quantify crystalline content in amorphous API include the following [2-5]: Differential scanning calorimetry Fourier transform infrared spectroscopy Near infrared spectroscopy Raman spectroscopy Solid State Nuclear Magnetic Resonance Polarized Light Microscopy Each of these techniques has advantages and disadvantages on their ability to differentiate between crystalline and amorphous phases of an API. Important considerations when characterizing DP with an amorphous API for crystalline content is drug load and formulation matrix. Since there are multiple components in a DP (API and excipients), specificity is very important for identifying a crystalline API and differentiating it from all other components. Thus, the spectroscopic techniques have advantage for specificity, but their sensitivities for a specific API will vary. Polarized light microscopy (PLM) is very good at determining the presence of very small quantities of crystalline material in an amorphous API [6]. However, the technique cannot easily discern crystalline material that is API from other crystalline materials that may be present in the formulation. Hot stage microscopy using PLM can be used to help to overcome this limitation if the melting point of the crystalline material and the melting point of materials in a formulation matrix are known. Figure 2 shows that the crystalline material in this solid dispersion melted before the melting point of the API and, therefore, is not crystalline API. Other imaging/mapping-based techniques that have shown promise in discerning crystalline API from amorphous API are elemental mapping by energy dispersive x-ray spectroscopy and chemical imaging by Raman spectroscopy [7]. Relatively newer technologies including micro-CT and second order near field imaging of chiral crystals (SONICC) are being evaluated for identification and quantification of Microsc. Microanal. 21 (Suppl 3), 2015 550 crystalline form in an amorphous API. The use of these imaging based techniques for distinguishing crystalline API from amorphous API will be discussed. References: [1] J. Bauer, J. Validation Tech. (2009), pp. 63-68. [2] B.A. Sarsfield et al. JCPDS-International Centre for Diffraction Data 2006 ISSN 1097-0002. [3] S Hogan, G Buckton Pharm Res. 18(1) (2001), pp. 112-116. [4] Solid-State NMR Application Note: Durham University, Solid-State NMR Research Service, 2007. [5] U Zimper et al., Pharmaceutics 2 (2010), pp. 30-49. [6] J Bauer et al., Pharm Res. 18(6) (2001), pp. 859-866. [7] J. P. Neilly, et al., Microsc. Microanal. 17 (Suppl. 2) (2011) 1130-1131. 1200 1000 3000 800 2000 600 400 1000 200 0 2.0 6.0 10.0 14.0 18.0 22.0 26.0 30.0 34.0 38.0 0 2.0 6.0 10.0 14.0 18.0 22.0 26.0 30.0 34.0 38.0 Figure 1. X-ray patterns of amorphous (left) and crystalline (right) form of the same API. Figure 2. PLM images of an amorphous dispersion at different temperatures with high birefringent features, suggesting crystalline material. Since the melting point of the material is ~125 �C, the crystalline material is not API. XRPD of the same material indicated the material to be amorphous. Disclosures: Authors are employees of AbbVie and may own AbbVie stock. The design, study conduct, and financial support for this research were provided by AbbVie. AbbVie participated in the interpretation of data, review, and approval of the publication.");sQ1[276]=new Array("../7337/0551.pdf","Advantages of Using Low Voltage for Elemental Mapping by Energy Dispersive X-Ray Spectroscopy in Pharmaceutical Systems","","551 doi:10.1017/S1431927615003554 Paper No. 0276 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Advantages of Using Low Voltage for Elemental Mapping by Energy Dispersive X-Ray Spectroscopy in Pharmaceutical Systems Joseph Neilly1 and John Roth1 1. AbbVie Inc. NCE Analytical Chemistry, North Chicago, IL 60054-6202 The advantages of using low energy primary electron beams (low-kV) for imaging by scanning electron microscopy (SEM) have been well established for many years [1]. Until recently [2], elemental analysis by energy dispersive x-ray spectroscopy (EDS) at low-kV was not practical. Use of a low-kV beam results in reduced x-ray signal and the inability to excite higher energy x-ray lines from higher Z elements. The establishment of larger and more efficient silicon drift detectors has addressed the issue of low x-ray signal, and low-kV elemental mapping by EDS is becoming practical and routine. Since most pharmaceutical products are composed of low Z elements, low-kV beams should be a natural choice for elemental analysis and mapping. A comparison of image quality in elemental maps collected using high (20-30 kV) and low (5-10 kV) primary beams was performed. Several solid oral dosage formulations were selected for testing. Samples tested included tablet coatings, dry blend direct compression tablets and melt extrusion formulations. In addition, a video array from a magnification standard was used to measure spatial resolution of the elemental maps under both beam conditions. All imaging was performed using an FEI Quanta 450 SEM and an Edax Octane EDS system with a 60 mm2 detector. Mapping conditions were set to collect approximately equal numbers x-ray counts in low-kV and high-kV maps. Thus, the lowkV maps were collected for longer times due to the lower efficiency of x-ray generation. Low-kV elemental maps generally had higher spatial resolution than did high energy elemental maps. This was most apparent for maps of lighter elements. Figure 1 compares elemental maps for silicon and chromium from a video array collected at 10 and 20 kV. There is little difference in the resolution of the chromium map at both accelerating voltages. However there are significant differences in resolution of the silicon map at these two voltages. At 20 kV most squares of the array were blurred and not well resolved. At 10 kV squares down to 1 um were resolved. Applying the same comparison to direct compression tablets shows the advantages of higher resolution with lower kV beams. Figure 2 shows the distribution of silicon dioxide in a tablet blend by mapping for silicon. At 20 kV the silicon distribution appears as a variety of larger blurred particles. At 5 kV the granular structure within the larger aggregates becomes visible. Other examples of the improved image quality in x-ray maps collected at lower voltages have been observed. In conclusion, the primary advantage of elemental mapping at lower voltages is the increased resolution due to reduced beam spread within the sample. This not only improves the spatial resolution of elemental mapping, but reduces overlap of data across material boundaries. Lower kV beams also reduce the level of beam damage typical of soft organic materials such pharmaceutical systems. While the x-ray count rate from a low-kV beam is substantially reduced, the use of larger SDD detectors off sets the loss of x-ray signal. Microsc. Microanal. 21 (Suppl 3), 2015 552 References: [1] J Goldstein et al in "Scanning Electron Microscopy and X-Ray Microanalysis", 3rd ed, Plenum New York, 2003, pp 207-209. [2] T Nylese, R. Anderhalt, V. Gorcea, Microsc. Microanal. 19 Suppl 2 (2013), pp. 272-273. Cr Cr Si Si Figure 1. Elemental maps for chromium and silicon from a video array from MRS-3 magnification standard composed of chromium squares in glass matrix collected at 5 kV (left) and 20 kV (right). Figure 2. Elemental maps for silicon from a compressed tablet collected at 10 kV (left) and 20 kV (right). Greater detail is visible in the map collected at lower accelerating voltage. Disclosures: All authors are employees of AbbVie and may own AbbVie stock. The design, study conduct, and financial support for this research were provided by AbbVie. AbbVie participated in the interpretation of data, review, and approval of the publication.");sQ1[277]=new Array("../7337/0553.pdf","Streamlining Identification of Compounds in Pharmaceutical Drug Products","","553 doi:10.1017/S1431927615003566 Paper No. 0277 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Streamlining Identification of Compounds in Pharmaceutical Drug Products T. Nylese1, J. Rafaelsen1, M. Bolorizadeh1, D. Edwards2. 1 2 EDAX, A Div of Ametek, Materials Analysis Division 91 McKee Drive, Mahwah, NJ, 07430 JEOL, 11 Dearborn Road, Peabody, MA 01960 Identifying compounds in pharmaceutical tablets is important for a variety of reasons in a regulated industry. Factors such as drug quality, drug purity, identification of degradants and distribution of materials assure the highest quality of finished drug products. The nature of the active pharmaceutical ingredients (APIs) and excipients in tablet form makes x-ray microanalysis challenging due to electron beam conditions and sample sensitivity to the electron beam, potential for sample damage, and beam spread within the matrix of the material. While techniques such as x-ray phase mapping have been successfully used for compound identification [1], there are limitations due to electron beam interaction with surrounding material. Low kV microanalysis has further been used to optimize collection, limiting the area of analysis to smaller spatial resolution which more distinctly identifies compounds [2]. Yet there are still challenges associated with using this method because it is more difficult to accurately quantify the elements present in the material. This work shows how the spectrum library matching software method facilitates compound identification and matching in pharmaceutical drug products. A library is built from x-ray spectra of known or previously identified compounds; this reference library is then used as a basis for comparison for unknown compound spectra. The method for compound matching is full spectrum energy peak intensity fitting, which removes the dependence on an accurate quantitative analysis which can be challenging for the above identified reasons. In the example used here, data is collected from a multivitamin in tablet form. First, several spectra from unique areas in the electron image are collected, the major elements are identified and the spectrum is labelled. Each spectrum is then entered into the spectrum library, as seen in Figure 1 showing Calcium Phosphate, Magnesium Oxide and Potassium Chloride. Phase mapping of a different area of the sample is then performed to highlight the presence of various phases, or multi-element compounds, with a unique color display. The phase spectra, which are considered unknown materials, are then compared to the match library which provides a conclusive match, with a fit percentage, as compared to the previously identified compounds, Figure 2. The benefits of the spectrum matching method are well suited for regulated industry for several reasons. First, the process follows the documented collection of identified materials to build a reference library and then uses this as the basis for comparison of future materials. Next, the software routine displays a match fit percentage, which removes the need for the analyst to make subjective conclusions, and does not depend upon interpretation of the data by an analyst. Finally, it is a simple software procedure that is well suited for novice analysts who can perform the method and provide documented, substantiated results. Microsc. Microanal. 21 (Suppl 3), 2015 554 Figure 1 shows Calcium Phosphate, Magnesium Oxide and Potassium Chloride compounds from the tablet which were used to create the multivitamin spectrum library. Figure 2 shows a spectrum extracted from the phase map (red) as matched to the library. An 85% fit match confirms it matches Magnesium Oxide. [1] Anderhalt et al, Microscopy & Microanalysis Proceedings (2010). [2] Nylese et al, Microscopy & Microanalysis Proceedings (2013).");sQ1[278]=new Array("../7337/0555.pdf","Mitochondrial Fission Arrest Phenotype in Brain Tissue of Patients and Animal Models of Familial Alzheimer's Disease Revealed with 3D EM","","555 doi:10.1017/S1431927615003578 Paper No. 0278 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Mitochondrial Fission Arrest Phenotype in Brain Tissue of Patients and Animal Models of Familial Alzheimer's Disease Revealed with 3D EM Eugenia Trushina1, Trace Christensen2, Liang Zhang1, Sergey Trushin1, and Jeffrey L. Salisbury2 1. 2. Department of Neurology, Mayo Clinic, 200 First St. SW, Rochester, MN 55905 Electron Microscopy Core Facility, Mayo Clinic, 200 First St. SW, Rochester, MN 55905 Alzheimer's Disease (AD) is a devastating neurodegenerative disorder characterized by progressive cognitive decline that affects the aging population. Hallmarks of AD include deposition of extracellular amyloid plaques, the accumulation of intraneuronal neurofibrillary tangles comprised of hyperphosphorylated tau protein, loss of synapses, memory deterioration and neuronal cell death. Recent data generated in animal models of AD and in AD patients suggest that altered mitochondrial dynamics contribute to the onset and progression of the disease. Mitochondria are dynamic organelles that constantly move within neurites in both anterograde and retrograde directions. Mitochondria function in local energy production to meet cellular demands, while the number and quality of the organelles is determined in part by cycles of fission and fusion. Mitochondrial fission is required for their proper distribution, movement, and quality control. Conversely, the merging or fusion of mitochondria allows the exchange of contents facilitating repair and the increase in size or functional volume of the organelle. The fidelity of fission and fusion machinery depends on a number of recently identified proteins, including mitochondrial fission factor (Mff), mitofusin-1 and 2 (Mfn1, Mfn2), optical atrophy 1 (Opa1), mitochondrial fission protein 1 (Fis1) and dynamin-related protein 1 (Drp1). Alterations in mitochondrial fission and fusion have been shown to be critical to the development of neurodegeneration in Charcot-Marie-Tooth disease type 2A, autosomal dominant optical atrophy, Alzheimer's, Huntington's and Parkinson's diseases. The importance of fission in mitochondrial function was highlighted when deletion of Drp1 was shown to lead to mitochondrial elongation and to an increase in oxidative damage, loss of respiratory function, and neurodegeneration. These observations suggest that mitochondrial division serves as a quality control mechanism and plays an essential role in neuronal health. Nonetheless, there is a delicate balance between accumulation and disposal of defective mitochondria. Loss of those organelles that can still contribute to a positive energy balance may ultimately be detrimental to cellular health and survival. Thus, better understanding of mitochondrial dynamics with respect to the progression of neurodegenerative diseases is needed in order to develop therapeutic strategies to protect neurons from damage. Using transmission electron microscopy, we examined mitochondria in hippocampal tissue from five transgenic mouse models carrying familial AD human mutations for presenilin 1 (PS1, M146L); amyloid precursor protein (APP, K670N, M671L); double transgenic mutant APP/PS1; triple transgenic mutant (3xTg) overexpressing tau (P301L), APP (K670N, M671L) and PS1 (M146V); and tau (P301L). Reconstruction of 3-dimensional renditions from serial thin-sections (3D EM) revealed changes in mitochondrial morphology in AD that was otherwise obscured or difficult to appreciate in individual thin sections. Using this technique, we identified a novel mitochondrial fission arrest phenotype that increased with disease progression and was more pronounced in animals with multiple FAD mutations. We established experimental conditions that mimic the mitochondrial fission arrest phenotype in wild- Microsc. Microanal. 21 (Suppl 3), 2015 556 type disease-free mice. These conditions and brain tissue from FAD animals were used, together with Western blot analysis, fax analysis and immunofluorescence, to characterize the abundance and activity of proteins that play a direct role in mitochondrial fission including Mff, Mfn1, Mfn2, Opa1, Fis1 and Drp1. Taken together our observations suggest that fission arrest occurs in AD in response to altered brain energetics. The figure shows an electron micrograph from a series of serial thin-sections of hippocampal brain tissue from a wild-type non-transgenic mouse. Superimposed on the electron micrograph is 3D reconstruction of mitochondria (green) within a neuropil. Here, the mitochondria have a regular tubular appearance (approximately 1/3 micron in diameter and up to several microns in length). Bar = 1 micron.");sQ1[279]=new Array("../7337/0557.pdf","A Tunable Approach to Visualize BRCA1 Assemblies in Hereditary Breast Cancer","","557 doi:10.1017/S143192761500358X Paper No. 0279 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A Tunable Approach to Visualize BRCA1 Assemblies in Hereditary Breast Cancer Carly E. Winton1,2, Brian L. Gilmore1, Andrew C. Demmert1, Zhi Sheng1, and Deborah F. Kelly1,2 1 2 Virginia Tech Carilion School of Medicine and Research Institute, Roanoke, VA 24016, USA. School of Biomedical Engineering and Science, Virginia Tech, Blacksburg, VA 24061, USA. Invasive breast cancer remains a leading killer of women in the U.S. today (www.cancer.org). Germline mutations in the breast cancer susceptibility protein (BRCA1) are highly correlated with hereditary forms of the disease, accounting for ~25% of all diagnosed cases [1]. Currently, patients with BRCA1 mutations have poor clinical outcomes due to the highly aggressive nature of the tumors and high recurrence rates following conventional therapies [2]. These challenges create a major impetus to elucidate the molecular underpinnings of the disease in an effort to improve treatment options. Under normal cellular conditions, BRCA1 acts as a global genomic surveyor through its diverse roles in DNA repair and RNA synthesis. In the nucleus, BRCA1 can interact with its binding partner BARD1 (BRCA1 Associated RING domain protein) as well as with RNA Polymerase II (RNAP II) [3]. To date, there is no information for the precise manner in which BRCA1 associates with larger protein assemblies, as conventional purification techniques have not yielded complexes suitable for structural analysis. To address this issue, we have developed a new tunable microchip approach that allows us to recover BRCA1-associated protein complexes from the nuclear material of patient-derived breast cancer cells for the first time (Figure 1). This alternative strategy to standard biochemical purification techniques preserves native protein-protein interactions for subsequent biochemical and cryo-Electron Microscopy (EM) analysis. Our new strategy utilizes Silicon Nitride (SiN) microchips coated with Nickel-nitrilotriacetic acid (NiNTA) lipid films that can be easily decorated with a variety of adaptor molecules including His-tagged protein A and a myriad of antibodies, thus creating a tunable system. We employed this system to capture and visualize active BRCA1-transcriptional complexes from the nuclear extracts of hereditary breast cancer cells. We collected images of the tethered BRCA1 assemblies under low-dose condition using TEM. We processed the images using the PARTICLE software package, and the RELION package [4] was then used to calculate a composite 3D density map of the complex. Employing a combination of antibody-labeling and molecular modeling techniques, we could uniquely place the RNAP II core structure into the density map along with the BRCA1-BARD1 RING domains and BRCA1 C-terminal (BRCT) domain. We could further position a short strand of DNA and K63-linked ubiquitin moieties within the map in areas with additional minor density (Figure 2). Complementary biochemical experiments were used to confirm our structural findings and to reveal the first glimpse of BRCA1 assemblies in the context of human cancer. References: [1] Miki Y., Swensen J., Shattuck-Eidens D., Futreal P. A., Harshman K., Tavtigian S., et al., Science. 266 (1994), pp.66-71. [2] Charafe-Jauffret E., Ginestier C., Iovino F., Wicinski J., Cervera N., Finetti P., et al., Cancer Res. 69 (2009), pp. 1302-13. [3] Friedman L. S., Ostermeyer E. A., Lynch E. D., Szabo C. I., Anderson L. A., Dowd P., et al., Cancer Res. 54 (1994), pp. 6374-82. [4] S.H. Scheres., J. Mol. Biol. 415 (2012), pp. 406-418. Microsc. Microanal. 21 (Suppl 3), 2015 558 Figure 1. Patient-derived breast cancer cells were lysed and the nuclear material was collected and enriched in RNAP II, BRCA1, and BARD1. The enriched nuclear material was then loaded onto tunable microchips decorated with Protein A and switchable IgG antibodies. The RNAP II core structure was uniquely placed in the resulting EM density map and regions of BRCA1 RING and BRCT domains were identified using antibody-labeling techniques. Figure 2. The RNAP II core (yellow) fit uniquely in the density map while additional major densities were attributed to the BRCA1-BARD1 RING domains (magenta, green) and BRCT domain (gray). A short strand of DNA (blue) and K63-linked ubiquitins (red, orange) were placed within regions of minor unoccupied densities. We are testing this model using patient-derived cells harboring BRCA1 mutations.");sQ1[280]=new Array("../7337/0559.pdf","mApoE-Functionalized Nanoliposomes Delivering Doxorubicin and Ultrasmall Superparamagnetic Iron Oxide to Glioblastoma Cells Characterized by TEM and Confocal Microscopy","","559 doi:10.1017/S1431927615003591 Paper No. 0280 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 mApoE-Functionalized Nanoliposomes Delivering Doxorubicin and Ultrasmall Superparamagnetic Iron Oxide to Glioblastoma Cells Characterized by TEM and Confocal Microscopy Simona Rodighiero1, Maria Gregori2, Maura Francolini3,1, Elena Vezzoli3, Matteo Tamborini3,4, Massimo Masserini2, Michela Matteoli5,3, Lorena Passoni3. 1 2 Molecular and Cellular Imaging Platform, Fondazione Filarete, Milan, Italy Department of Health Sciences, University of Milano-Bicocca, Milan, Italy 3 Department of Medical Biotechnology and Translational Medicine, University of Milan, Milan, Italy 4 Fondazione Vollaro, Milan, Italy 5 Pharmacology and Brain Pathology Lab, Humanitas Clinical and Research Center, Milan, Italy To overcome the limits of current therapy for glioblastoma (GBM) [1] nanotechnology-based approaches offer attractive and innovative possibilities including improved passage of drugs across the blood brain barrier (BBB) and escaping multidrug resistance by efflux mechanisms. Nanoliposomes (NLs) covalently coupled with a modified apolipoprotein E peptide (mApoE) have been successfully used to enhance the BBB penetration in the context of neurodegenerative diseases [2]. Here, these mApoE functionalized NLs have been modified to encapsulate the anticancer drug doxorubicin (DOXO) or DOXO and ultrasmall superparamagnetic iron oxide (USPIO) and they have been used to target DOXO to GBM cells. The NLs preparations were characterized by transmission electron microscopy (TEM, Fig. 1) and the role of ApoE on the NLs internalization together with its mechanism have been analyzed by confocal microscopy (CM). GBM-derived cell lines U87-MG, A172, T98G were incubated for 4 hours with DOXO-mApoEtargeted or DOXO-non targeted NLs. DOXO intracellular uptake was significantly increased in the presence of the ApoE functionalization giving a more pronounced intracellular accumulation of DOXO in cells incubated with DOXO-mApoE-NLs compared to DOXO-NLs (Fig. 2A, B). These results suggested a mApoE-targeted NLs internalization via receptor-mediated endocytosis. The presence in the incubation medium of dynasore, the inhibitor of dynamin, reduced DOXO intracellular accumulation indicating a role of clathrin-dependent endocytosis in DOXO-mApoENLs uptake (Fig. 2). The USPIO DOXO-mApoE NLs internalization and intracellular trafficking has been further analyzed by both CM and TEM (Figure 3). Inhibition of in vitro cell growth was assayed by MTT test at 72 hours. DOXO-mApoE-NLs were found to inhibit GBM cell viability in a dose-dependent manner with IC50 values of DOXO comprised between 0.5 and 1.5 �g/ml. No cellular toxicity was observed upon incubation with mApoE-NLs. Overall the data obtained support the use of mApoE-targeted nanocarriers for the delivery of chemotherapeutics and/or cytotoxic agents to GBM cells. The possibility to load mApoE-NLs with USPIO nanoparticles to exploit the advantages of correlative microscopy to analyze their intracellular trafficking and the potential use of USPIO mApoE-NLs as contrast agents for Magnetic Resonace Imaging will be discussed. [1] Tanaka et al Nat Rev Clin Oncol (2013); 10(1): 14-26. [2] Re et al. Nanomedicine (2011); (5):551-9. Acknowledgements: Work supported by the European Centre for Nanomedicine (CEN Foundation), Milano, in the framework of the project "Enhancement Packages" (Rif.EP017). Microsc. Microanal. 21 (Suppl 3), 2015 560 A B Figure 1. NLs design and characterization. A) Schematic representation of ApoE- functionalized liposomes encapsulating doxorubicin and USPIO nanoparticles; B) TEM image of ApoEfunctionalized liposomes. A DOXO NLs B DOXO-mApo NLs C DOXO-mApo NLs + Dyn 10 �m Figure 2. DOXO-mApoE NLs internalization in U87-MG cells analyzed by CM. Confocal optical sections showing internalized doxorubicin (red) and the cell, segmented thanks to the SytoBlue 45� labelling (grey). Cells were incubated with the indicated NLs in the absence (A, B) or in the presence of Dynasore (C). A B C D Figure 3. USPIO DOXO-mApoE NLs internalization in U87-MG cells analyzed by CM and TEM. A, B) Confocal optical section of a cell showing internalized doxorubicin (red), USPIO (green) and DAPI (blue). C, D) TEM images of a cell showing internalized NLs and USPIO. B) and D) are the magnified views of A) and C), respectively. Scale bars: 10 �m (A), 2 �m (B, C) and 200 nm (D).");sQ1[281]=new Array("../7337/0561.pdf","Type VII Collagen: From Discovery of the Anchoring Fibril Protein to Clinical Trials Ameliorating the Human Blistering Disease Epidermolysis Bullosa","","561 doi:10.1017/S1431927615003608 Paper No. 0281 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Type VII Collagen: From Discovery of the Anchoring Fibril Protein to Clinical Trials Ameliorating the Human Blistering Disease Epidermolysis Bullosa Douglas R. Keene1 and Sara F. Tufa1 1 Shriners Hospital for Children, Micro-Imaging Center, Portland, Oregon USA Type VII collagen (C7) was identified in 1983, initially as a novel band on an electrophoretic gel separating the components of a tissue homogenate [1]. Joining this group in 1986, our laboratory utilized rotary shadowing to characterize type VII collagen as a dimeric molecule, approximately 790 nm in length, composed of two anti-parallel monomers (inset, Figure 1A) overlapping at their NC2terminal ends by 60nm. Using antibodies prepared in mice, we identified C7 as a component of anchoring fibrils. Anchoring fibrils are long banded structures anchored at both ends within the lamina densa of the dermal-epidermal junction. At the middle of the molecule, arches form within the shallow papillary dermis (Figure 1A). The resulting loops entrap dermal collagen fibrils, forming "staples" which mechanically fasten the epithelium to the underlying dermis. Recognizing that a defect in this adhesive molecule might result in a subepidermal split causing blistering, candidate diseases with blistering phenotypes were evaluated. Recessive Epidermolysis Bullosa (RDEB) results in severe blistering, mutilating scarring, and predilection for squamous cell carcinoma. Mutations were found in the COL7A1 gene resulting in diminished or absent C7 and consequential lack of functional anchoring fibrils (Figures 1B, C). We then evaluated therapies to deliver functional C7 to the basement membrane of C7 deficient skin. Our collaborators at Stanford, led by Paul Khavari, Peter Marinkovich and Alfred Lane, regenerated epidermis from retrovirally transduced autologous RDEB keratinocytes, allowing long-term expression of C7 in human EB skin transplanted onto immunodeficient mice [2]. Meanwhile, our collaborators in Minnesota, led by Jakub Tolar, demonstrated that transplantation of normal bone marrow into C7 deficient EB mice was effective in ameliorating the disease phenotype, with the production of functional C7 and anchoring fibrils at the DEJ [3]. Having tested these two therapies in mice, clinical trials to ameliorate the disease in humans began. Jakub Tolar's group focuses on using bone marrow transplant therapy to deliver WT stem cells to the skin. These cells appear to enter the skin via the circulatory system; they then incorporate into the epithelium and produce functional type VII collagen. Patients are biopsied at 0, 30, 60, 90,180 days then yearly post-transplant and evaluated by ILM and ITEM. Some patients respond to therapy with the production of near normal density of anchoring fibrils (Figure 2) but in a significant number of individuals the anchoring fibril architecture is often thin and wispy, forming only occasional fibrils that are non-banded and mostly without loops. Still, blistering is markedly improved and patient quality of life is decidedly enriched, with significant decrease in blister formation [4]. These individuals give credibility to the suggestion that only 30% of normal C7 is required for skin integrity. The clinical trial at Stanford seeks to treat highly involved lesions by transplanting corrected patient keratinocyte skin grafts onto prepared would beds (Figure 3). Patients are evaluated by ILM and ITEM at day 0, then within 30 days and again at 3, 6 and 12 months post grafting. Complete epidermal regeneration has been observed within 30days. Immunocytochemistry reveals strong C7 expression at Microsc. Microanal. 21 (Suppl 3), 2015 562 the DEJ and a robust anchoring fibril network. Six transplants have been done in each of four patients to date, with encouraging outcomes in all. Figure 1. Anchoring fibrils (formed by C7, rotary shadow inset) entrap banded collagen to "staple" epithelium to dermis in normal skin (A). The severe blistering of Recessive Dystrophic Epidermolysis Bullosa (B) results from a lack of functional C7 and subsequent lack of anchoring fibrils (C). Figure 2. Anchoring fibril immuno-density at day 0 is nil (A) but at day 360 post bone marrow transplant this patient has near normal immuno-density and ultrastructurally identifiable AFs (B). Figure 3. A region of severe blistering has been prepared and grafted (day 0, A, C). After 180 days the grafted region lacks blisters and has gained a population of immuno-identifiable anchoring fibrils (B,D) References: [1] H Bentz et al. (1983). Proc Natl Acad Sci U S A. 80:3168-72. [2] S Ortiz-Urda1 et al. (2002). Nature Medicine 8, 1166 � 1170. [3] J Tolar et al. (2009). Blood. 113:1167-74. [4] J E Wagner et al. (2010). N Engl J Med. 363:629-39. [5] Funding for this work was generously provided by the Shriners Hospitals for Children");sQ1[282]=new Array("../7337/0563.pdf","Characterizing Working Catalysts with Correlated Electron and Photon Probes","","563 doi:10.1017/S143192761500361X Paper No. 0282 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterizing Working Catalysts with Correlated Electron and Photon Probes Eric A. Stach,1 Yuanyuan Li,2 Shen Zhao,3 Andrew Gamalski,1 Dmitri Zakharov,1 Ryan Tappero,4 Karen Chen-Weigart,4 Juergen Thieme,4 Ulrich Jung,3 Anika Elsen,3 Qiyuan Wu,5 Alexander Orlov,5 Jingguang Chen,6 Ralph G. Nuzzo3 and Anatoly Frenkel2 1. 2. Center for Functional Nanomaterials, Brookhaven National Laboratory, New York, NY 11973 Department of Physics, Yeshiva University, New York, NY 10016 3. Department of Chemistry, University of Illinois at Urbana-Champaign, Illinois 61820 4. National Synchrotron Light Source II, Brookhaven National Laboratory, New York, NY 11973 5. Materials Science and Engineering Department, Stony Brook University, Stony Brook 11794 6. Department of Chemical Engineering, Columbia University, New York, NY 10027 Heterogeneous catalysts often undergo dramatic changes in their structure as the mediate a chemical reaction. Multiple experimental approaches have been developed to understand these changes, but each has its particular limitations. Electron microscopy can provide analytical characterization with exquisite spatial resolution, but generally requires that the sample be imaged both ex situ and ex post facto. Photon probes have superior depth penetration and thus can be used to characterize samples in operando (i.e. when they are actively working). But they generally lack spatial resolution and thus give only ensemble average information. We have taken advantage of the recent developments in closed-cell microscopy methods1 to develop an approach that allows us to successfully combine electron, x-ray and optical probes to characterize supported nanoparticle catalysts in operando. By measuring the reaction products at each stage of the reaction, we can directly correlate the information that can be obtained from each approach, and thus gain a deep insight into the structural dynamics of the system. We will describe how we can use this approach to correlate x-ray absorption spectroscopy (both nearedge and extended fine structure), scanning transmission electron microscopy and infrared microspectroscopy to understand how Pt and Pd nanoparticles supported on silica undergo structural changes during the room temperature hydrogenation of the ethylene, and how we can direct measure and describe the reaction products on the surfaces of the nanoparticles as the reaction proceeds. By combining these approaches, we can track the interplay between nanoparticle reduction, coarsening, and the specific surface species at different stages of the reaction.2 (Figure 1) We will also show how this approach can be used to understand the partitioning that occurs in bimetallic nanoparticles during oxidation and reduction at elevated temperatures, with a focus on the NiPt system. The presentation will focus on the development and application of experimental methods, including the high temperature atmospheric pressure electron microscopy, the direct measurement of reaction products using gas chromatography�mass spectrometry and the ability of a newly developed electron microscope for operando microscopy (based on the FEI Talos platform) to characterize bimetallic nanoparticles through energy dispersive x-ray spectroscopy [1-3]. References: [1] de Jonge, N. and Ross, F.M., Nat. Nano., 6 (2011), p. 695. Microsc. Microanal. 21 (Suppl 3), 2015 564 [2] Li, Y. et al, The Complex Structural Dynamics of Catalysts as Revealed by Correlated Imaging and Spectroscopy Probes, submitted. [3] This research project was supported, in part, by DOE BES under Contract No. DE-FG0203ER15476. Development of the micro-cell approach was supported, in part, by an LDRD grant by Brookhaven National Laboratory. Research was carried out at the Center for Functional Nanomaterials and the National Synchrotron Light Sources I and II at the Brookhaven National Laboratory, which is supported by the U.S. Department of Energy, Office of Basic Energy Sciences, under Contract No. DEAC02-98CH10886 and DE-SC0012704 . Figure 1. Experimental results from operando XAFS and STEM. (A) X-ray absorption near edge spectroscopy, (B) Fourier transform magnitudes of Pt L3 edge EXAFS spectra, and (C) STEM images, all measured in operando, during different reaction regimes. EXAFS and STEM results are displayed in (D) for all regimes. The green line indicates the mean particle sizes, as obtained by the STEM image analysis. The red line shows the change of Pt-Pt coordination number obtained from XAFS analysis.");sQ1[283]=new Array("../7337/0565.pdf","In situ Analytical TEM of Ilmenite Reduction in Hydrogen","","565 doi:10.1017/S1431927615003621 Paper No. 0283 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In situ Analytical TEM of Ilmenite Reduction in Hydrogen Arne Janssen1, Eric Prestat1, Matthew Smith1, Sarah J. Haigh and M.G. Burke1 1. Materials Performance Centre, School of Materials, The University of Manchester, Manchester, UK. Ilmenite (FeTiO3) is an important mineral, formed in igneous and metamorphic rocks, and is the main feedstock for the titanium industry. Ilmenite is also an attractive and inexpensive oxygen carrier for chemical-looping combustion (CLC), one of the most promising technologies to accomplish CO2 capture in power plants [1]. The technique uses solid metal oxides as an oxygen carrier, which transfers oxygen from the combustion air to the fuel. The metal oxide reacts with a hydrocarbon fuel in the fuel reactor, to produce CO2 and H2O while reducing the metal oxide. However, detailed information about structural and chemical changes of ilmenite during the reduction process are still not available. The understanding of structural and chemical relationships at the nanometer scale during the oxygen transfer is essential because it can be help to understand the exact reduction mechanism of the oxygen carrier, and hence optimise the combustion process. In situ analytical TEM has huge potential of providing nano-scale information of chemical reactions and structural transformations occurring in atmospheric environments. Due to new developments in in situ holders there is now the possibility of combining in situ TEM studies with the best microanalytical aspects of a modern AEM. In situ reduction experiments have been performed using the new Protochips Atmosphere In Situ Heating and Gas Reaction Cell (Protochips Inc.). The specimens for this investigation were obtained from Norwegian hard rock ilmenite, which was ground with ethanol. Droplets of this suspension were deposited on the surface of the heater E-chip. The reduction of ilmenite was carried out at 470 �C in 950 mbar H2. Analytical transmission electron microscopy investigation has been carried out using a spherical aberration corrected FEI Titan G2 80-200 with ChemiSTEM technology, operated at 200 kV and equipped with a GIF Quantum ER. Energy Electron Spectroscopy (EELS) and X-ray Energy Dispersive Spectroscopy (XEDS) in STEM have been used to map regions of interest to observe chemical changes during the reduction process. Our first results give some insight on the reduction mechanism of ilmenite. STEM-EELS and XEDS spectrum images yielded that the decomposition of ilmenite to titanium and iron take place at relative low temperature, resulting in restructuring of the overall morphology of the initial grains (Fig1&2). Iron particles are formed with a non-uniform distribution around the surface. The density of the particles increases with time and temperature. The formation of the particles is the result of volume diffusion of iron towards the surface of the grains, while Ti is not mobile at temperature below 900 �C [2]. A volume reduction of approximately 40% causes the formation of porosity in the residual TiO2. EEL spectrum imaging of the oxygen K-pre-edge indicate that most of the iron is still covalent bonded with oxygen and not reduced to metallic iron (Fig2e) [3]. The oxidation state of the iron sub-oxides is currently unknown; further EELS studies will be used to clarify the oxidation state of the iron particles. Microsc. Microanal. 21 (Suppl 3), 2015 566 References: [1] H. Leion et al, Chemical Engineering Research and Design, 9 (2008), p. 1017-1026. [2] D.B. Rao, M Rigaud, Oxidation of Metals, 9 (1975), p. 99-116. [3] P.A. van Aken et al, Phys Chem Minerals, 25 (1998), p. 494-498. [4] We acknowledge BP for financial funding. The authors thank Matthew A. Kulzick and Nestor J. Zaluzec for helpful discussions. Figure 1. STEM-HAADF images before (a) and after (b) hydrogen reduction, XEDS elemental maps (c-f) of the decomposed ilmenite. Figure 2. STEM-HAADF images before (a) and after (b) hydrogen reduction, (c) EELS elemental maps showing the Fe, Ti and O distribution in the decomposed ilmenite, (e) O K-pre-edge map at 530 eV indicating covalent bonding of oxygen with iron or titanium.");sQ1[284]=new Array("../7337/0567.pdf","Resolving Surface Structures of Catalytic Ru Nanoparticles during Catalysis","","567 doi:10.1017/S1431927615003633 Paper No. 0284 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Resolving Surface Structures of Catalytic Ru Nanoparticles during Catalysis B.K. Miller and P.A. Crozier School for Engineering of Matter, Transport and Energy, Arizona State University, Tempe, AZ 85287-6106 In situ studies of catalysts in the environmental transmission electron microscope (ETEM) are of great value to elucidate the structure of a catalyst under reaction conditions. This is important, since the structure of many catalysts change at elevated temperature in the presence of reactive gases. One way to more directly relate structure to activity is to measure the activity of the catalyst while it is being observed in the TEM. This is called operando TEM. The relative catalytic activity can be measured by monitoring the amount of product gas produced by the catalytic reaction using electron energy-loss spectroscopy (EELS) and mass spectrometry [1]. This measurement is made possible through the use of a unique sample preparation, shown schematically in Figure 1. Using an aberration-corrected ETEM, it is possible to image the surfaces of nanoparticles while they are catalysing a reaction. The basic nanoparticle structure can be modelled using a Wulff construction; for more detail, see the abstract by Walker et al. [2]. We have used an imagecorrected FEI Titan ETEM to observe Ru particles under various reaction conditions relevant to CO oxidation. While this reaction is simple and well-studied, the most active form of the catalyst is still debated in the literature [2]. It is agreed that the bulk of the active catalyst particles is Ru metal, but the surface structures which yield the highest catalytic activity are still uncertain. Furthermore, the application of this catalyst to fuel cells will depend on the preferential oxidation of small quantities of CO in H2 gas, also known as PROX, which has received less study. Figure 2 shows the surface of a Ru nanoparticle at elevated temperature with several Torr of H2 gas inside the environmental cell. It is clear that such images can reveal the structures present on surface facets of the Ru particles. Figure 3 shows an example of the detailed information which can be obtained, in this case for a PROX experimental condition. In 2 Torr of H2 with only 0.01 Torr of O2, RuO2 lattice spacings are seen in a layer about 5 thick on both the (100) and (101) surface facets, but are absent in the rest of the particle, where Ru metal spacings are clearly resolved. As shown in Figure 3a, these layers are absent prior to introducing the O2 indicating that they have formed due to exposure to a small amount of oxygen in a highly reducing H2 gas. We continue to make atomic resolution observations of nanoparticle surfaces in different gas environments, coupling these observations to the measured composition of the gas inside the ETEM cell, which should reveal structure-activity relationships for this Ru catalyst. References: [1] B.K. Miller, P.A. Crozier Microscopy and Microanalysis 20, (2014), p.815�824. [2] N.P. Walker, B.K. Miller, & P.A. Crozier, ibid. (these proceedings) [3] D. W. Goodman, C. H. F. Peden, & M. S. Chen, Surface Science 601, (2007), p.5663�5665. [4] The support from National Science Foundation CBET-1134464 and the use of ETEM at John M. Cowley Center for HR Microscopy at Arizona State University is gratefully acknowledged. Microsc. Microanal. 21 (Suppl 3), 2015 568 Figure 1. Schematic showing the structure of the TEM sample. a) Silica spheres of about 200 nm diameter are used as the catalyst support. b) The silica spheres are in turn dispersed over glass wool fibers (also silica), which have been formed into a 3 mm pellet. c) In the TEM, a Ta mesh grid, which also has catalyst particles dispersed over it, is placed above the pellet. Images are acquired from particles dispersed on the Ta mesh, while the entire sample contributes to the gas composition within the TEM. Figure 2 (above). Ru particle supported on amorphous silica. The image is a single 0.5 second acquisition with 2 Torr of H2 in the ETEM cell at 200�C with a TEM voltage of 80kV. The electron dose rate is kept below 200 /2 Figure 3 (right). Two images of the same Ru particle in different gas compositions, both at 200�C and 80kV. a) 2 Torr pure H2 showing no indication of oxidation. b) 2 Torr H2 with 0.01Torr O2, showing a 0.5nm layer of RuO2 on the (100) and (101) Ru surfaces.");sQ1[285]=new Array("../7337/0569.pdf","In situ Environmental TEM and DFT Studies on the Highly Stable AuIr Bimetallic Catalyst","","569 doi:10.1017/S1431927615003645 Paper No. 0285 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In situ Environmental TEM and DFT Studies on the Highly Stable AuIr Bimetallic Catalyst Chang Wan Han1, Paulami Majumdar2, Ernesto E. Marinero1, Antonio Aguilar-Tapia3, Rodolfo Zanella3, Jeffrey Greeley2, and Volkan Ortalan1 . School of Materials Engineering, Purdue University, West Lafayette, IN, USA . School of Chemical Engineering, Purdue University, West Lafayette, IN, USA 3 . Centro de Ciencias Aplicadas y Desarrollo Technologico, Universidad Nacional Autonoma de Mexico, Mexico City, Mexico 2 1 Since the report by Haruta et al. in 1987 of high catalytic activity of supported nanocrystalline Au catalysts in CO oxidation at low temperature [1], Au catalysts have attracted significant amount of interest from researchers. Despite the remarkable activity of Au catalysts in various oxidation processes, a wide range of uses of Au catalysts in industry is limited due to the lack of stability against sintering [2,3]. In this regard, stabilization of supported Au nanoparticles (NPs) is of utmost importance in the field of Au catalysis. With the purpose of increasing the stability of Au catalysts, various methods for increasing the stability have been developed [4,5]. One of the successful methods for increasing the stability is through adding a second metal into Au NPs [6�9]. Specifically, R. Zanella et al. reported that AuIr bimetallic NPs on TiO2 support show enhanced stability and activity compared to Au on TiO2 [7]. Considering several reports of the improved catalytic activity as well as on the improved stability of supported Au-based bimetallic catalysts, it can be expected that Au catalysts for industrial applications will be comprise bimetallic NPs rather than pure Au. Therefore, the mechanism behind the increased stability of Au-based bimetallic catalysts needs to be clearly understood. Despite the importance of Aubased bimetallic catalysts, however, the in situ investigations of their dynamic behaviors during the "real" catalytic process have been limited. In this study, we performed in situ environmental TEM and DFT calculations to investigate the highly stable anatase TiO2 supported AuIr bimetallic NPs, prepared by the deposition-precipitation by urea (DPU) method [6,7]. Particularly, as an effort to understand the activity as well as the increased stability of AuIr catalyst, the dynamic behaviors (i.e., morphological variation and/or surface reconstruction) of AuIr NPs during room temperature CO oxidation process were investigated using in situ E-TEM technique. In all the experiments performed, significant care was taken to avoid electron beam damage by using of low-dose irradiation to avoid structural transformation of the Au NPs [10]. A mixture of CO and He (ratio= 1:1) was fed into the differentially pumped environmental cell of the FEI Titan E-TEM and the pressure was maintained at ~ 0.55 Torr. Figure 1 shows the morphological variation of AuIr NPs in the CO and He ambient at room temperature. Fig. 1(a) displays a facetted AuIr NP at the onset of the experiment. Structural reconstruction occurs during the reaction and after 9 minutes of reaction time the morphology as shown in Fig. 1(b) is that of a rounded NP. Considering the fact that low-coordinated Au atoms [11], locating on the corners and edges of the NPs are attributed to the catalytic activity, the morphological variation can be correlated with the measured temporal variation of catalytic activity of AuIr NPs in CO oxidation. We also performed a series of density functional theory (DFT) calculations for single crystal AuIr alloy slabs on the TiO2 surface [Fig. 2], as a first approximation, to understand the higher stability of TiO2 supported AuIr NPs, as compared to Au NP. At the vacuum/metal interface, Au atoms prefer to segregate to the surface, but at the TiO2 interface, oxygen reverses this trend and draws Microsc. Microanal. 21 (Suppl 3), 2015 570 the more oxophilic Ir atoms to the interface. The segregated Ir atoms near the TiO2 surface leads to the increase in stability of AuIr on TiO2. The results of in situ E-TEM analysis and DFT calculations will be provided and the correlation between the morphological changes of AuIr NPs with the activity in CO oxidation. Moreover, the mechanism of the increased stability will be discussed in the aspect of the theoretically calculated adsorption energies of AuIr slabs on TiO2. References [1] M. Haruta et al, Chem. Lett. 4 (1987) p. 405�408. [2] T. Choudhary and D. Goodman, Top. Catal. 21 (2002) p. 25�34. [3] M. Jos�-Yacaman et al, J. Phys. Chem. B. 109 (2005) p. 9703�9711. [4] M. Kirchhoff et al, Nanotechnology. 16 (2005) S401�S408. [5] S.H. Joo et al, Nat. Mater. 8 (2009) p. 126�131. [6] X. Bokhimi et al, J. Phys. Chem. C. 114 (2010) p. 14101�14109. [7] A. Gomez-Cortes et al, J. Phys. Chem. C. 113 (2009) p. 9710�9720. [8] Y. Guan et al, J. Catal. 305 (2013) p. 135�145. [9] X. Liu et al, Chem. Commun. (2008) p. 3187�3189. [10] Y. Kuwauchi et al, Angew. Chemie. 51 (2012) p. 7729�7733. [11] B.. Hvolb�k et al, Nano Today. 2 (2007) p. 14�18. Figure 1. In situ E-TEM images of AuIr NPs on TiO2 after feeding the CO and He gas mixture. Morphological variation of a AuIr NP from the strongly facetted (Fig. 1a) to the rounded (Fig. 1b) is observed. Figure 2. AuIr slab on anatase TiO2 surface showing segregation of Au to the vacuum interface and Ir to the support interface. Adsorption energy reported below figure.");sQ1[286]=new Array("../7337/0571.pdf","Dynamic structural changes in a single catalyst particle during single walled carbon nanotube growth","","571 doi:10.1017/S1431927615003657 Paper No. 0286 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Dynamic structural changes in a single catalyst particle during single walled carbon nanotube growth Pin Ann Lin1,2, Zahra Hussaini1, Juan C Burgos Beltran3, Jose Leonardo Gomez Ballesteros3, Perla B. Balbuena3, Renu Sharma1 1. Center for Nanoscale Science and Technology, National Institute of Standards and Technology, Gaithersburg, MD, USA 2. NanoCenter of Maryland, University of Maryland, College Park, MD, USA 3. Department of Chemical Engineering, Texas A&M University, College Station, TX, USA The interaction of gases with a solid catalyst nanoparticle during catalysis is a non-equilibrium process. For example, during single-walled carbon nanotube (SWCNT) growth, carbon atoms diffuse in and out of the catalyst particle, causing variations in the chemical potential, and possibly structure, of the particle. We have employed an environmental scanning transmission electron microscope (ESTEM), with an aberration corrector, operated at 300 kV to record real-time, atomic-resolution videos (6 frames s-1) with 1000 frames of SWCNT growth from a Co-Mo/MgO system in a C2H2 gaseous environment at synthesis temperatures. The time-resolved videos, generating large data sets, are used to identify individual reaction steps. We have developed methods to analyze these large data sets accurately and efficiently. The simultaneous existence of two regions having different structures (pure Co and Co-carbide) within an individual particle during SWCNT growth (Fig. 1a) was detected using fast Fourier transform (FFT) analysis of the catalyst particle images. In order to determine the catalyst particle's structure over time, an image processing scheme that utilizes a combination of home-built and publicly-available algorithms for image drift correction, noise reduction, and template matching to accurately locate the position of atomic columns has been created. For each position, the distances between neighboring atom columns were averaged, giving a local spacing value which was used to determine whether a local region was Co or Co-carbide (Fig. 1b). Figure 2a shows the fluctuation in Co volume fraction, which suggests that the catalyst particle has a changing stoichiometry as a function of time. In addition to the particle structure measurements, the growth length of the SWCNT was measured (Fig. 2b). We found that when the Co fraction increases � carbon output is greater than carbon intake � the tube growth rate increases. Similarly, when the Co fraction decreases � carbon intake is greater than carbon output- the tube growth rate decreases. The fluctuation of carbon content inside the catalyst particle can be directly related to the nanotube growth, which can then be used to calculate rates for reaction steps, such as bulk or surface diffusion of carbon atoms through a particle and their incorporation into a nanotube. A detailed description of structural analysis approach and quantitative measurements of catalyst carbon content and nanotube growth will be presented. Microsc. Microanal. 21 (Suppl 3), 2015 572 Figure 1. (a) A high resolution image of SWCNT growth from a catalyst particle that contains two different structures (Co and Co2C), which are identified by FFT images of R1 and R2 regions. (b) Blue and violet areas are marked for Co and Co-carbide, respectively, and an intermediate color in between shows the boundaries. Scale bars are 1 nm. Figure 2. Fluctuations in (a) Co volume fraction and (b) nanotube growth length as a function of time.");sQ1[287]=new Array("../7337/0573.pdf","Detailed Atomic Structure of Defects in 2D Materials: From Graphene to Transition Metal Dichalcogenides.","","573 doi:10.1017/S1431927615003669 Paper No. 0287 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Detailed Atomic Structure of Defects in 2D Materials: From Graphene to Transition Metal Dichalcogenides. Jamie H. Warner1 1. Department of Materials, University of Oxford, OX1 3PH, Oxford, UK Understanding the exact details of the position of atoms within defective regions in materials helps model and predict their properties accurately. Two dimensional materials that are only one atom thick, such as graphene and boron nitride, make identifying the positions of single atoms substantially easier than from 3D material structures. Aberration-corrected transmission electron microscopy at low accelerating voltages of 80 kV and below, provides high atomic contrast in both phase contrast TEM imaging as well as in annular dark field scanning transmission electron microscopy. In recent work we have shown how monochromation of the electron source reduces the energy spread and lowers the detrimental effects of chromatic aberration on spatial resolution. The combination of monochromation and image correction in phase contrast HRTEM mode at an accelerating voltage of 80kV yields images of graphene with contrast spots associated with the position of individual atoms, as shown in figure 1, which when fitted with Gaussian profiles can give FHWM's of 80pm, indicating that sub-Angstrom resolution is approached. This led to the ability to fully resolve the position of individual carbon atoms in graphene around defects and measure the changes in C-C bond lengths. Correlating the atomic structural models deduced from HRTEM with density functional theory provided insights into how changes in charge density around defect clusters causes bond elongation and compression [1,2]. In new work this has been extended to high temperature measurements by using an in-situ heating holder with suspended graphene samples. Totally different defect structures are produced at these high temperatures (>600 oC) and are likely caused by changes to the flexibility of graphene and limits to out-of-plane distortions that help accommodate strain at low temperatures. In more recent work we have extended our ultra-high spatial resolution phase contrast HRTEM imaging to studying defects in the transition metal dichalcogenide 2D crystal, MoS2, shown in figure 2. Monolayered MoS2 is three atoms thick and large defect structures are far more complex than in graphene and more challenging to accurately resolve. Sulfur atoms originally in pristine 2H phase of MoS2 can flip positions, adopting 1T like locations, within defect structures to help stabilize Mo atoms. Detecting the contrast of S atoms in 1T like positions requires spatial resolution approaching Angstrom level. The ability to accurately resolve these features requires spatial resolutions close to the Angstrom level and demonstrates the benefit of the higher spatial resolution achieved by monochromation of the electron source. Time dependent images of the migration of single vacancies into line defects is studied and the flipping of S atoms to new 1T positions detected. References: [1] J. H. Warner, E. R. Margine, M. Mukai, A. W. Robertson, F. Giustino, A. I. Kirkland, Science, (2012) 337, pp. 209-212. [2] J. H. Warner, G-D. Lee, K. He, A. Robertson, E. Yoon, A. I. Kirkland, ACS Nano (2013), 7, pp 9860�9866. [3] The author thanks the funding from the Royal Society. Microsc. Microanal. 21 (Suppl 3), 2015 574 Figure 1. Aberration corrected transmission electron microscopy image of monolayer graphene at an accelerating voltage of 80kV and with monochromation of the electron source. Figure 2. Aberration-corrected transmission electron microscopy image of monolayer MoS2 at an accelerating voltage of 80kV and with monochromation of the electron source. Vacancy defects are detected.");sQ1[288]=new Array("../7337/0575.pdf","Multimodal Characterization of Graphene","","575 doi:10.1017/S1431927615003670 Paper No. 0288 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Multimodal Characterization of Graphene Stefanie Freitag1 1 Carl Zeiss Microscopy GmbH, Market Segment Material Sciences, Munich Site, Germany Graphene is a tightly packed layer of carbon atoms that are bonded together in a hexagonal honeycomb lattice. Only one atom thick (335 pm) but up to several micrometer in lateral extent, graphene shows many extraordinary material properties. As an electric conductor, it performs as good as copper. In the ability to transport heat, it outperforms all other known materials. Optically it is highly transparent but completely impermeable for any other atoms. Harder than steel and also showing a high elasticity, graphene unites several astonishing material properties. This makes graphene one of the most scientific investigated and promising materials for technical applications. The graphene market will be split across many application sectors, like super capacitors, displays, wearable technology and printed electronics, each attracting a different type of graphene, thus various microscopic methods need to be used and combined to characterize the properties starting from the prepared graphene to the engineered graphene. Graphene is often prepared by using an adhesive tape. The top surface layers from Graphite can be removed and transferred to another surface. The most important question is, how thick the transferred flakes actually are and whether the process has left additional substances on the sample surface. Two different procedures for preparing graphene are investigated. First, a basic approach with common adhesive tape and an uncleaned surface substrate and second a special adhesive wafer foil and plasma cleaned Si substrate is used. Two samples of graphene flakes were analyzed with light microscopy (LM), electron microscopy (EM) and atomic force microscopy (AFM) by means of the correlative microscopy technique (CorrMic). In a second step it is investigated, how optical interference contrast microscopy (TIC), cathodoluminescence (CL) and helium ion microscopy (HIM) can serve as complementary techniques in visualization, in profiling, and in measurement of conductivity of graphene. The study demonstrates that it is possible to prepare few layer graphene flakes with both adhesive tape techniques. Bright field microscopy can indicate the number of graphene layers and thickness if the graphene is on a thin optically resonant film [1,2,3]. There is a strong difference in contamination between the two preparation methods. Standard adhesive tapes leave residuals on the surface and the flake itself, which often are only resolved by high resolution microscopic techniques like SEM and especially AFM. Since a lot of investigation on graphene is only done by using optical microscopy, such contamination may cause false and/or misleading results. The findings also show that TIC-Profilometry allows for contact � free thickness measurement and profiling of graphene [4]. Correlating the height measured by AFM with the measured phase shift of a thin film is a method to measure refractive index, extinction, and conductivity. Even the energy dispersive back scattered detector (EsB) allows to distinguish between single and multi-layer graphene [5]. The detection of the finest monolayer thickness differences is possible from the BSE signal in the SEM as shown in this study. The noise is greatly reduced compared to the inelastic plasmon signal from the energy filtered signal in the TEM. The high surface sensitivity of the Helium Ion Microscope enables imaging of monolayers of graphene at high resolution. This is essential for the quality control of graphene Microsc. Microanal. 21 (Suppl 3), 2015 576 patterning processes. By means of patterning and preparation of graphene nano ribbons one can create bandgaps in graphene which are necessary for electrical devices. The investigations showed that the bandgap in a graphene ribbon increases as the ribbon width decreases [6]. Graphene quantum dots on the other hand can be imaged with cathodoluminescence (CL). The CL spectrum has its peak were the quantum dots change their shape from circular-topolygonal-shape and their corresponding edge-state [7]. Graphene can be imaged with various microscopic methods, each revealing a different information. All methods have benefits and disadvantages, therefore using these methods in a combination often leads to deeper understanding of the graphene properties. References: [1] P. Blake et al , Appl. Phys. Lett. 91 (2007) 063124 [2] Z. H. Ni et al, Nano Lett. 7 (2007) 2758 [3] Y. Y. Wang et al, Nanotechnology 23 (2012) 495713 [4] M. Vaupel et al, J. Appl. Phys. 114 (2013) 183107 [5] Iwona Joswik-Biala et al, Microsc. Microanal. 20 Suppl 3 (2014) [6] D.S. Pickard et al, Microscopy and Microanalysis 07/2012; 18(S2):800-801 [7] Soo Seok Kang et al, Current Applied Physics 14 (2014) S111-S114 Figure 1. From LM to AFM; increasing resolution and decreasing speed of analysis Figure 2. TIC microscopy: phase profile along dashed line, converted into height of graphene stack. Figure 3. Nano ribbons in graphene. Created with the Helium Ion Microscope. Dr. Dan Pickard, National University of Singapore");sQ1[289]=new Array("../7337/0577.pdf","Nano-Sculpting of Suspended CVD Graphene with Helium and Neon Ion Beams","","577 doi:10.1017/S1431927615003682 Paper No. 0289 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Nano-Sculpting of Suspended CVD Graphene with Helium and Neon Ion Beams D.A. Cullen,1 J. Swett,2 A. Rondinone,3 P. Bedworth,2 S. Heise,2 and S. Sinton2 1. 2. Materials Science & Technology Division, ORNL, Oak Ridge, TN 37831-6064 USA. Lockheed Martin Space Systems Co., Adv. Technology Center, Palo Alto, CA 94304 USA. 3. Center for Nanophase Materials Sciences, ORNL, Oak Ridge, TN 37831-6493 USA. The ability to precisely modify graphene is critical to a variety of application spaces, ranging from nanoscale electronics to chemical sensors [1,2]. The recent development and production of gas field ion source (GFIS) systems yields new opportunities for fabricating sub-10nm feature sizes within graphene sheets, as demonstrated by the etching of suspended graphene nanodevices with a helium ion beam [3]. However, the precision-controlled nanoscale-sculpting of graphene layers is challenging, as complex and interactive milling mechanisms, such as knock-on displacement, electronic interactions, and thermal processes, compete with self-healing and beam-induced deposition. In the present study, we explore the fundamental principles and mechanisms for controllable nanosculpting (ion-milling using a Zeiss Orion NanoFab, Fig. 1a) of suspended graphene films produced by chemical vapor deposition (CVD). The impact of different aspects of the milling process, e.g., ion choice (helium or neon), beam energy, and ion dose, on pore creation and any associated defect formation, will be systematically explored (Fig. 1b). Ion milling efforts are coupled with low-voltage, aberration-corrected scanning transmission electron microscopy (STEM) in a Nion UltraSTEM 100 (Fig. 1c) to quantify the resulting pore size, lattice defects, and defect density, as shown in Fig. 1d-e. Careful measurements performed at various stages of milling (Fig. 2) will reveal any changes in defect density and carbon coordination with increasing ion dose, and explore aspects of electron-beam induced healing [4]. Changes in the density of single-vacancies, divacancies, multi-vacancies, lattice deformations (Frenkel Pairs and Stone-Wales defects), amorphitization, substitutions, and voids, will be discussed in the context of theoretical calculations and atomistic simulations [5]. All of the aforementioned experiments are being conducted on CVD graphene films, which have been robustly characterized by STEM prior to and after ion milling. Future paths towards meeting the ultimate goal of precise and repeatable sculpting of nano-scaled features in graphene will also be discussed [6]. References: [1] Y Zhang et al, Nature 459 (2009) p. 820. [2] F Schedin et al, Nature Materials 6 (2007) p. 652. [3] MC Leme et al, ACS Nano, 3 (2009) p. 2674. [4] R Zan et al, Nano Letters 12 (2012) p. 3936. [5] O Lehtinen et al, Nanotechnology 22 (2011) p. 175306. [6] Research supported by Independent Research & Development funding at Lockheed Martin and through a user project supported by ORNL's Center for Nanophase Materials Sciences (CNMS), which is a Dept. of Energy, Office of Science User Facility. Microsc. Microanal. 21 (Suppl 3), 2015 578 Figure 1. (a) Zeiss Orion NanoFab helium and neon ion microscope. (b) Pore arrays formed by helium ion milling. (c) Nion UltraSTEM 100 high-performance dedicated STEM. (d-e) Atomic-resolution STEM images of pores in graphene formed using a helium ion beam. (a) (b) (c) Figure 2. Example of the progression of pore formation in suspended CVD graphene produced via NanoFab milling, as imaged in the Nion UltraSTEM 100. (a) Pristine graphene, (b) intermediate He ion dose, and (c) high He ion dose.");sQ1[290]=new Array("../7337/0579.pdf","Atomic Scale, 3-Dimensional Characterization of Radiation Effects in Tungsten for Fusion Applications","","579 doi:10.1017/S1431927615003694 Paper No. 0290 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic Scale, 3-Dimensional Characterization of Radiation Effects in Tungsten for Fusion Applications Philip D Edmondson1, Alan Xu2, Luke R Hanna2, Michal Dagan2, Steve G Roberts2 and Lance L Snead1 1. 2. Materials Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN USA Department of Materials, Oxford University, Oxford, UK The refractory metal tungsten is a promising candidate material for plasma facing components (PFCs) in future fusion reactors due in part to its low sputter yield, good thermal conductivity and low activation under transmutation. However, tungsten suffers from a very high brittle-to-ductile transition temperature (BDTT) of 400-500 �C [1]. This inherent brittleness is only exacerbated under irradiation due to the irradiation-induced defects and the formation of second phase precipitates containing the transmutation products Re and Os. [2,3] Here, we discuss the use of novel field ion microscopy (FIM) and atom probe tomography (APT) techniques to investigate radiation effects in irradiated tungsten. In this study, both ion and neutron irradiated samples were characterized. Materials that were ion irradiated were either pure tungsten, or a tungsten-5at.% tantalum alloys and irradiated with 2 MeV W ions at between room temperature and 500 C to damage levels of up to 33 displacements per atom (dpa) as estimated using SRIM using the Kinchin-Pease method with an Ed of 68 eV. Neutron irradiation pure tungsten specimens were irradiated in the High Flux Isotope Reactor (HFIR) at Oak Ridge National Laboratory (ORNL). Ion irradiated samples were analysed in either a 3DAP-LAR FIM or a Cameca LEAP 3000X HR at Oxford; neutron irradiated samples using a Cameca LEAP 4000X HR at ORNL. A series of FIM images taken during a typical evaporation sequence from the [222] planes of a W-Ta specimen are shown in Fig. 1. Out-of-sequence evaporation events are detected, observed as vacant sites in the inner positions of the plane. These preferential evaporation events can then be exploited to identify the Ta atoms in the W-Ta alloy. By imaging multiple consecutive planes, it is possible to build up a 3-dimensional image of the distribution of Ta atoms in the matrix, Fig. 2. Atom probe tomography techniques are ideally suited to the investigation of second phase precipitates in tungsten, Fig 3. Here we will discuss the differences in the use of ion irradiation and neutron irradiation on the formation of these phases, and the composition of the precipitates in the identification of the dominant phases formed, sigma or chi. [1] A. Giannattasio et al., Phil. Mag. V90 (2010) P.3947 [2] D. E. J. Armstrong, Appl. Phys. Lett. V102 (2013) P.251901 [3] PD Edmondson et al., J. Nucl. Mater., In press, doi: 10.1016/j.jnucmat.2014.11.067 [4] PDE, AX, LRH, MD and SGR acknowledge support from the UK's Engineering and Physical Sciences Research Council (EPSRC) under grants EP/H018921/1 and EP/K030043/1. A portion of the Microscopy was conducted as part of a user proposal at ORNL's Center for Nanophase Materials Science, which is an Office of Science User Facility. Microsc. Microanal. 21 (Suppl 3), 2015 580 Fig 2: Reconstruction of data collected from sequential FIM images. Green spheres indicate W atoms; pink indicates tantalum. Fig 1: Sequential FIM images of the evaporation of a [220] plane in a W-5at.% Ta alloy. Out-ofsequence evaporation is observed. a) b) Fig 3: Re atom maps of an ion irradiated W-Re alloys, irradiated at temperatures of a) 300 and b) 500 �C.");sQ1[291]=new Array("../7337/0581.pdf","Clustering and Radiation Induced Segregation in Neutron Irradiated Fe-(3-18)Cr Alloys","","581 doi:10.1017/S1431927615003700 Paper No. 0291 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Clustering and Radiation Induced Segregation in Neutron Irradiated Fe-(3-18)Cr Alloys Mukesh Bachhav1, G. Robert Odette2, Emmanuelle A. Marquis1 1 Department of Materials Science and Engineering, University of Michigan, Ann Arbor, MI 48109 2 Materials department, University of California, Santa Barbara, CA 93106 High chromium ferritic-martensitic (F-M) steels are one of the promising structural material classes for future nuclear power plants. These steels are designed to combine corrosion resistance, conferred by chromium, with low swelling, high resistance to irradiation damage as well as to retain adequate toughness and elevated-temperature strength during service [1]. However, the long-term use of these steels in intense neutron irradiation environments requires reliable predictions of the evolution of their microstructures and mechanical properties. Binary Fe-Cr alloys constitute a model system for high Cr ferritic/martensitic steels and have therefore generated lot of interest by allowing the systematic study on irradiation induced microstructural changes. In the present study, microstructural changes in neutron irradiated Fe-Cr binary alloys are investigated using atom probe tomography (APT). A series of six Fe-Cr alloys of nominal compositions 3, 6, 9, 12, 15, and 18 at.%Cr were irradiated at a neutron fluence (E>1 MeV) of 1.1 x 1021 n/cm2 at 563 � 15K and to a damage level of 1.82 displacements per atom (dpa). Solute distributions revealed precipitation for alloys containing more than 9at.%Cr (Figure 1). Both the Cr concentration dependence of precipitation and the measured matrix compositions are in agreement with the recently published Fe-Cr phase diagrams [2]. An irradiation-accelerated precipitation process is strongly suggested for precipitation. Along with homogenously distributed Cr-enriched clusters of the phase, few clusters involving Si, P, Ni, and Cr, are observed in the matrix [3]. For Fe-6, 9, 12 at.%Cr, Si and Cr are found segregated to dislocation loops and information pertaining to number density, size, and habit plane were analyzed for Fe-6at.%Cr alloy[4]. Grain boundary chemistry for Fe-Cr alloys are quantitatively compared between the as-received and the neutron irradiated alloys. Zones depleted of clusters and Si are found at the interfaces of carbide and nitride precipitates and along grain boundaries in the vicinity of these precipitates. To study stability of clusters and observed features in irradiated samples, annealing is carried out at high temperatures. The results are discussed in the context of equilibrium segregation, radiation-enhanced diffusion, and/or radiation induced segregation. References: [1] R.L. Klueh, D.R. Harries, High chromium ferritic and martensitic steels for nuclear applications, ASTM International, 2001. [2] M. Bachhav, G. Robert Odette, E.A. Marquis, Scripta Materialia, 74 (2014) 48-51. [3] M. Bachhav, G.R. Odette, E.A. Marquis, Journal of Nuclear Materials, 454 (2014) 381-386. Microsc. Microanal. 21 (Suppl 3), 2015 582 [4] M. Bachhav, L. Yao, G.R. Odette, E.A. Marquis, Journal of Nuclear Materials, 453 (2014) 334-339. [5] This work was supported by the University of Michigan College of Engineering, DOE Office of Nuclear Energy's Nuclear Energy University Programs at the University of Michigan and University of California-Santa Barbara Colleges of Engineering, and the ATR-NSUF program Figure 1: 20 nm thick slices showing the distribution of Cr atoms for each of the analyzed alloys: (a) Fe�3Cr, (b) Fe�6Cr, (c) Fe�9Cr, (d) Fe�12Cr, (e) Fe�15Cr, (f) Fe�18Cr Figure 2: 3D reconstruction showing (a) dislocation loop decorated with Si and Cr in neutron irradiated Fe-6Cr alloy (b) segregation of impurities P, Si, Ni in the matrix of Fe-15Cr alloy");sQ1[292]=new Array("../7337/0583.pdf","TEM Analysis of Structural Transformation in Al/Ni Nanomaterials under High Energy Ion Irradiation","","583 doi:10.1017/S1431927615003712 Paper No. 0292 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 TEM Analysis of Structural Transformation in Al/Ni Nanomaterials under High Energy Ion Irradiation Khachatur Manukyan1, Sergei Rouvimov2,3, Christopher E. Shuck4 and Alexander S. Mukasyan3,4 1. 2. Department of Physics, University of Notre Dame, Notre Dame, Indiana 46556, USA Department of Electrical Engineering, University of Notre Dame, Notre Dame, Indiana 46556, USA 3. Notre Dame Integrated Imaging Facility, University of Notre Dame, Notre Dame, Indiana 46556, USA 4. Department of Chemical and Bio-molecular Engineering, University of Notre Dame, Notre Dame, Indiana 46556, USA The paper addresses the effect of irradiation by accelerated ion beam on the structural transformation of Al/Ni multilayer nanomaterials studied by Transmission Electron Microscopy (TEM). The Al/Ni multi-layered nanomaterials are promising nanostructured energetic composite materials [1-2] that exhibit tunable ignition properties to a variety of external excitation methods including friction, shock waves, electrical sparks, and local heating. Because the ignition of such materials depends on their atomic structure and composition, irradiation may provide a novel approach for modification of the reactivity of the nanostructured energetic composite materials. Here we study the structural transformation in Al/Ni layers under irradiation. Magnetron sputtering and electron beam evaporation have been used to fabricate free-standing reactive multilayer nanostructured foils [3]. High energy carbon and aluminum ion beams with different charge states and intensities were used to irradiate the samples. The samples were analyzed by TEM using both high resolution TEM (HRTEM) and High Angle Annual Dark Field (HAADF) scanning TEM (STEM) modes at FEI Titan 80-300 electron microscope. The microscope was operated at 300 keV and equipped with an Oxford Inca EDX detector. A TEM cross-sectional sample that included the NiO-Ni interface was prepared from the top surface by Focus Ion Beam (FIB) using FEI Helios SEM/FIB dual beam equipment. It has been demonstrated that a significant enhancement of reactivity of Al/Ni materials after relatively short-term (40 min) high energy (20 MeV) irradiation by 12C4+ ions (below the ignition threshold), is associated with structural transformations that lead to a decrease in the thermal selfignition temperature and ignition delay time. Indeed x-ray diffraction (Fig. 1) indicates that defect formation in the samples under irradiation leads to a decrease of the diffraction peak intensities of Al and Ni in irradiated materials as compared to the original foils. At the same time, the full width at half maximum (FWHM) of the Al and Ni peaks (not shown) exhibits different trends with irradiation time indicating that longer irradiation can facilitate the growth of Al crystallites while decreasing Ni crystallite size. This observation agrees well with TEM/STEM analysis (Figs. 1 and 2) that evidences the intermixing of Al and Ni at layer interfaces. Both high resolution TEM and electron diffraction indicate that formation of amorphous materials at the interfaces with small (2-3 nm) crystals of the Al3Ni intermetallic phase occur in the amorphous regions. It can be seen that the nuclei of Al3Ni crystals are distributed in the Al-rich phase close to every other Al/Ni interface for the 40 min irradiated foil (Fig. 2). It is interesting that these nuclei line-up perpendicular to the direction of the incident beam. Such structures confirm that the beam induces solid-state diffusion of Ni into the Al layer, where nucleation of Al3Ni phase takes place. Thus, the enhancement of multilayer energetic nanomaterial reactivity is shown to be associated with Microsc. Microanal. 21 (Suppl 3), 2015 584 radiation-induced structural transformations including defect formation and intermixing of metals at the interfaces that lead to presence of amorphous layers at the interfaces with Al3Ni intermetallic nuclei. References: [1] Zhou, X.; et al . ACS Appl. Mater. Interfaces 2014, 6, 3058�3074 [2] Dreizin, E. L.. Prog. Energy Combust. Sci. 2009, 35, 141�167. [3] Weihs, T.P., In Metallic Films for Electronic, Magnetic, Optical and Thermal Applications: Structure, Processing and Properties, K. Barmak and K. R. Coffey, Eds.; Woodhead Publishing, Swaston, UK, 2014, Chapter 6, 160-243 Figure 1 XRD data (left) of irradiated Ni/Al foils as compared to the original one and HAADF STEM image (right) of the Ni/Al foil irradiated for 40 minutes. Inserted is Al and Ni distribution along the vertical line showing intermixing of Al and Ni at the interfaces. Figure 2 TEM images of Al-Ni foils irradiated for 40 (a) and 150 (b) minutes showing the Al and Ni intermixing and the formation of Al3Ni nuclei (a) and small Al3Ni grains (b).");sQ1[293]=new Array("../7337/0585.pdf","TEM Examination of Precipitation Behaviour of M23C6 and Sigma Phases and Dislocations in SS 310S under Creep Deformation at 800�C","","585 doi:10.1017/S1431927615003724 Paper No. 0293 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 TEM Examination of Precipitation Behaviour of M23C6 and Sigma Phases and Dislocations in SS 310S under Creep Deformation at 800�C Babak Shalchi Amirkhiz1, Su Xu1 1 CanmetMATERIALS, Natural Resources Canada, Hamilton, Canada Stainless steel 310S is a candidate alloy for fuel cladding of the Canadian Generation IV Supercritical Water-Cooled Reactors (SCWR). In-depth understanding of the creep properties and microstructural evolution of SS 310S during creep is vitally important for the proposed SCWR challenging design conditions, and both are strongly linked to formation, dissolution and coarsening of different precipitates [1]. To the authors' knowledge, a detailed microstructural investigation on crept specimens of SS 310S at 800�C is not available. In this study, precipitation and deformation behaviours of a commercial SS 310S plate (Fe, 24-26wt.% Cr, 19-22% Ni, 2% Mn, 1.5% Si, 0.045% P, 0.03% S, 0.08% C) in the asreceived (solution treated), and creep conditions at 800�C were examined using TEM. The creep specimens examined were tested under 30, 50, 70 and 80MPa and the rupture times were 3971, 530, 111, and 59 hours, respectively. All samples were examined using the Tecnai Osiris TEM operating at 200kV. The as received microstructure of 310S mainly consists of gamma grains with nanometric precipitates of Ti(CN) type. Figure 1 shows general microstructure of ruptured samples under tested conditions. In addition to extensive discrete grain boundary M23C6 carbides, intragranular M23C6 precipitated heterogeneously on Ti(CN) particles (Fig. 2A and C). These carbides are coherent or semi-coherent Cr23C6 as evidenced by electron diffraction analysis (Fig. 2A). The Cr23C6 carbides located in the vicinity of grain boundaries can grow faster as Cr can diffuse through the grain boundaries at a higher rate and transform into . The intragranular carbides grow along certain crystallographic orientations (i.e. <002>) and finally transformed to . These grains are elongated towards <002> (Fig. 2B). Transformation of carbides to phase occurred in all creep cases and the areas around large grains are free of small precipitates as seen in Fig. 1. When carbides were absent or grew to lose coherency with the gamma matrix or transformed into , they did not effectively hinder the movement of dislocations. Consequently, the network of subgrains was disappeared around large grains as seen in Fig. 1A, B and C. Dislocation density () in crept samples were measured by adjusting a multibeam case in a low index zone and imaging using high angle annular dark-field (HAADF) [2]. A representation of the dislocations in the crept samples is given in Fig. 3. The measured values for samples crept at 80, 70, 50 and 30 MPa were 2.25 1014, 2 1014, 2.67 1013, and 1.36 1014 m-2, respectively. While values are close to 1014 m-2 and generally decreases with decreasing load, the value for the 50MPa sample was one order of magnitude lower as also seen in Fig. 3C. This could be attributed to the ineffectiveness of the carbides in pinning dislocations after growing to a certain size or transforming into . In the 30MPa sample a network of dislocations is formed within all gamma grains as seen in Fig. 1D and Fig. 4D. The lower creep resistance of SS 310 compared to common commercial austenitic stainless steels [3] may be due to extensive precipitations observed under creep [4]. References: [1] T Sourmail, Materials Science and Technology, Vol 17, 1 (2001) Microsc. Microanal. 21 (Suppl 3), 2015 586 [2] D Rojasa, J Garciab, O Prata, L Agudoc, C Carrascod, G Sauthoffa, AR Kaysser-Pyzallab, Materials Science and Engineering A, 528 (2011), p. 1372. [3] S. Xu, S.-M. Jin and P. Le Dreff-Kerwin, Proc. PVP 2013, PVP2013-97290. [4] The authors acknowledge financial support from NRCan program on Energy R&D (PERD) A Cr23C6 Cr23C6 B C Cr23C6 D 80 Cr23C6 Cr23C6 Figure 1. TEM micrographs showing the microstructure of 316S under creep at 800�C in STEMHAADF images superimposed on EDX Cr elemental map highlighting phase and Cr23C6 carbides in samples crept at (A) 80 MPa, (B) 70 MPa. (C) 50 MPa, and (D) 30 MPa. A Cr23C6 Ti(CN) B Cr23C6 Cr23C6 C Ti(CN) Cr23C6 Figure 2. (A) Early growth stage of a Cr23C6 carbide in the 80 MPa sample, growing towards [002], the inset is the corresponding SAD at <001> zone axis; (B) a precipitate in the 70 MPa sample, the inset is SAD of phase (C) HAADF image overlaid with Fe, Cr and Ti EDS elemental maps showing Cr23C6 carbides and Ti(CN) particles within them in 70 MPa sample, inset is SAD at from <121> zone axis of a carbide particle. A B C D Figure 3. HAADF images representing dislocation density in 310S samples crept for 80, 70, 50 and 30MPa at 800�C, respectively. Insets show the orientation of the beam regarding the sample, the zone axes are <122>, <011>, <001> and <011> respectively from image A to D.");sQ1[294]=new Array("../7337/0587.pdf","Correlative TEM and APT of Helium Bubbles in Ion-Irradiated RAFM Steel","","587 doi:10.1017/S1431927615003736 Paper No. 0294 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Correlative TEM and APT of Helium Bubbles in Ion-Irradiated RAFM Steel B. Mazumder1, C. M. Parish2 and M. K. Miller1 1. 2. Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, TN, USA. Materials Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN, USA. Neutron irradiation of structural materials introduces He through (n, ) transmutation, which degrades the mechanical properties through swelling, blistering, and He embrittlement [1]. He bubbles are trapped on grain boundaries and dislocations, degrading properties; recent studies show that trapping on the surfaces of nanoclusters and larger precipitates [2] may mitigate materials degradation. Hence, it is important to investigate He bubble distributions in materials used in current reactors, such as reduced activated ferritic-martensitic (RAFM) steels. (S)TEM and APT are powerful tools to study microstructural features and provide complementary information. Correlative APT-TEM studies [3] have illustrated the direct comparison of the same features visible via each technique, which can provide more reliable spatial and chemical information. Here, preliminary results correlating TEM and APT of helium bubbles are shown. The material used in this study was F82H steel, He-irradiated to ~8000 ppm He, ~0.5 dpa at 500 �C [4]. (S)TEM images of F82H before and after He ion irradiation are presented in Fig. 1a and 1b, respectively. Bubbles or voids, with a size range of 4-16 nm, are visible (circular features). APT at the same depth shows the presence of bubbles, some with diameters < 2 nm [4]. Since the specimens were prepared from different locations, direct correlation is impossible. For direct APT-(S)TEM correlation, a needle-shaped specimen at a depth of 500 nm from the surface was prepared, and both TEM and APT were performed (TEM, Fig. 2a). Within this volume, bubbles with diameters ranging from 2-8 nm (3.8 nm � 0.8 nm average) were measured via TEM; those with diameters < 2 nm were not detected. Subsequently, APT analysis was performed on the same specimen, and Fig. 2b shows the He bubble distribution from the analyzed volume shown in Fig 2a. Bubble sizes in the APT-analyzed volume ranged from 1 � 0.5 � 7 � 1.2 nm, with an effective average bubble diameter of 3.5�0.4 nm. APT data revealed a bubble number density of 7.2�1022 m-3, which is roughly twice that measured by TEM, ~ 3�1022 m-3. This discrepancy is likely because imprecision in determining the TEM-analyzed volume. The size variation of the bubbles measuredby both the techniques are compared in Fig 2c and 2d, and exhibit similar size distributions for large (> 2 nm) bubbles, but APT detected comparatively more bubbles in the smaller 1-2 nm size range. This correlative study provides more precise information on He bubble locations and distributions within the matrix and other microstructures (precipitate, grain boundary, etc.) if present within the analysis volume. Future microstructural engineering for He mitigation will require that the locations of the He bubbles be determined and correlated using both techniques. He bubble features analyzed in (S)TEM (e.g., dislocations or loops, grain boundaries, chemically similar but crystallographically distinct precipitates) and features analyzed in APT (e.g., boundaries with different Gibbsian segregation, extremely small precipitates) can be successfully correlated, opening up new paths for strategy developments in designing of He-tolerant microstructures. This example illustrates a first step toward that goal. Future TEM or STEM tomography experiments will provide more quantitative comparisons. Microsc. Microanal. 21 (Suppl 3), 2015 588 References: [1] GR Odette et al, Annual Review of Material Research 38 (2008) p. 471. [2] Q Li et al, Journal of Nuclear Materials 445 (2014) p. 165, and references therein. [3] I Arslan et al, Ultramicroscopy 108 (2008) p. 1579. [4] B Mazumder et al, Nuclear Materials and Energy 1 (2015) p. 8. [5] Research supported by the U.S. Department of Energy, Office of Basic Energy Sciences, Materials Sciences and Engineering Division. APT and TEM conducted as part of a user proposal at ORNL's Center for Nanophase Materials Sciences (CNMS), which is a DOE Office of Science User Facility. Figure 1. (a) BF-STEM image of F82H steel before He ion irradiation, b) under-focused TEM image of F82H after He ion irradiation revealing the distribution of He bubbles in the matrix and carbides. Figure 2. (a) TEM image in the under-focused condition showing He bubbles in the APT needle-shaped specimen prior to APT analysis. (b) APT reconstruction of the same needle. Density isosurfaces (purple) at 65-70 atoms/nm reveal He bubbles. In TEM, the entire specimen can be imaged, whereas in APT a partial volume is analyzed due to the local electrode and the detector size. Size distribution of the detected He bubbles within the volume analyzed by (c) TEM and (d) APT.");sQ1[295]=new Array("../7337/0589.pdf","Metallographic Characterization of Thermal Damage in Boiler Tubes","","589 doi:10.1017/S1431927615003748 Paper No. 0295 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Metallographic Characterization of Thermal Damage in Boiler Tubes Michael P. Carroll1 1. Lisin Metallurgical Services, Portland, OR USA Failure of boiler tubes within systems such as fossil fueled power plants have long posed economic and operational challenges due to unscheduled down time. A proper failure analysis is critical to diagnose the root cause, recommend follow-on inspections for the extent of further damage, assess life expectancy, and monitor health throughout the remaining life cycle. Due to the hot and corrosive conditions, the most common damage mechanisms are primarily thermal and/or corrosion related. The focus of this talk will be the use of metallography in the assessment of thermally influenced failures, highlighted by case histories detailing each mechanism. Specific failure causes that will be discussed include long term overheating, short term overheating, and thermal fatigue, which are briefly described below. Finally, an overview of remaining life assessment will be related to a combination of metallographic interpretation, mechanical properties, dimensional inspection and temperature monitoring. Long Term Overheating is implied in more boiler tube failures than any other mechanism, which occurs when local temperatures exceed the design limit for time scales ranging from days to years [1]. Long Term Overheating often manifests as creep and stress rupture, which can be diagnosed with metallographic examination. Stages of creep and stress rupture follow a predictable course including nucleation of creep voids at triple points, continued nucleation along grain boundaries, coalescence of voids, and eventual rupture. Thermal oxidation may also be involved, which lowers the effective wall thickness and therefore increases local stresses. Exposure to elevated temperatures for a prolonged period may also produce microstructural changes that aid in determining the peak temperature ranges. For example, spheroidization of pearlite lamellae implies long term exposure to elevated temperatures below the Ac1 temperature, whereas the observation of fresh pearlite and/or bainite is produced by heating above Ac1 and slow cooling. Short Term Overheating (e.g. rapid overheating) occurs at temperatures in excess of approximately 850 degrees F and sometimes much higher, where the yield strength of the alloy is lowered sufficiently to permit relatively fast deformation and fracture. As opposed to long term overheating, cases with short term overheating are normally confined to a single or short series of process excursions [2]. When rupture occurs below the crystallization temperature, the microstructure near the fracture can exhibit severe plastic deformation. At higher temperatures, between Ac1 and Ac3, a microstructure of pearlite and/or bainite is formed due to the cooling conditions following failure [3]. Examination of adjacent tubes can aid in determining if the observed microstructures were localized or widespread. Thermal Fatigue may refer to high or low cycle fatigue, where thermal cycling imposes mechanical stress on the tubes due to expansion and compression. Crack morphology is consistent with mechanical fatigue failure, and appears as straight unbranched cracks that propagate perpendicular to the maximum tensile stress. Depending on conditions within the tube, the cracks may exhibit oxidation as well as other symptoms of long term overheating damage. Microsc. Microanal. 21 (Suppl 3), 2015 590 References: [1] The NALCO Guide to Boiler Failure Analysis, Robert D. Port, Harvey M. Herro, McGraw Hill, New York. [2] Damage Mechanisms and Life Assessment of High-Temperature Components, R. Viswanathan, ASM International, Metals Park, OH. [3] ASM Handbook, Failure Analysis and Prevention, 8th Edition. Page 529. [4] The author acknowledges Mark Lisin, P.E. for providing case histories within this talk as well as continued mentorship in the field of Metallurgical Failure Analysis. Near Long Term Overheating Bulge Away from Long Term Overheating Bulge Figure 1: Optical photomicrographs showing metallographic cross sections taken at (left) and away from (right) long term overheating damage. Both microstructures are ferritic, but creep void formation along grain boundaries is apparent in the overheated area. Fractography of the main fracture surface revealed a predominately intergranular pathway. 2% nital etch. As-Polished 2% Nital Etch Figure 2: Optical photomicrographs showing longitudinal sections through adjacent cracks. The straight, unbranched, wedge shaped cracks are characteristic of thermal fatigue. The microstructure of the tube consisted of lamellar pearlite in a ferrite matrix. Evidence of overheating damage was not revealed by the metallographic examination. As-polished (left) and 2% nital etch (right).");sQ1[296]=new Array("../7337/0591.pdf","Failure Analysis in Support of Deformation Modeling","","591 doi:10.1017/S143192761500375X Paper No. 0296 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Failure Analysis in Support of Deformation Modeling Lisa Deibler1, Edmundo Corona1, Shelley Williams1 1. Sandia National Laboratories, Albuquerque, NM While many aspects of the mechanical behavior of structural metals can be accurately modeled, the prediction of fracture and failure has proved particularly difficult. The annual Sandia Fracture Challenge [1] was launched in 2012 to test the abilities of models to capture ductile deformation and fracture. Every year a unique geometry sample made from a common engineering alloy is specified, and teams are invited to model the deformation and fracture of the sample based on uniaxial tensile characteristics and geometry. The models are then compared to experimental results. The sample for the 2013 Sandia Fracture Challenge was made from A286 steel in the geometry pictured in Figure 1. The sample was compressed during testing, causing shear along the two ligaments marked A and B in Figure 1. A group at Sandia modeling the part found that their prediction of the forcedisplacement curve began to deviate from experimental results before the sample reached its maximum load. Fractography and metallography were employed to better understand if there were microstructural processes that the model was not capturing which could lead to the deviation from experimental data. Examination of the fracture surfaces of the two ligaments, A and B, showed that they were different as seen in Figures 2 and 3, indicating some asymmetry in the sample at least during final failure. The general morphologies of the fracture surface areas were similar, but the micro-scale fracture features were different. This presentation will focus on the results from the right sides of the samples. The region marked A showed a very rough surface with equiaxed dimples, indicating a tensile failure mode. Region B contained vertical smears, as though the surfaces had rubbed together after initial separation. Region C was flat with small, elongated dimples, indicative of shear failure. The fracture surface evidence indicates that failure likely began at areas B and C in shear, and then areas marked A failed last. This theory was confirmed by deforming specimens to various compressive displacements before failure, then taking several sections through the thickness of the sample to determine where cracking began. Figure 4 contains the polished cross-sections of three samples. Sample 1 was unloaded prior to the maximum load, sample 2 was unloaded at maximum load, and sample 3 was unloaded just prior to failure. In all three samples, the top image is of the front surface, the middle image is from a quarter of the way through the sample, and the bottom image is from halfway through the sample. The images of samples 1 and 2 indicate that damage which would not have been captured by the model does not begin to accumulate until the maximum load. The images of sample 3, unloaded just before failure, indicate that extensive cracking does indeed occur at the top and bottom surfaces and at a quarter of the way through the sample (areas B and C in Figure 2). The fracture surface and metallographic analysis indicated that damage did not begin to accumulate in the samples prior to maximum load, and therefore was not the cause of the model's deviation from experiment. Further mechanical testing did show that the A286 plate from which the samples were Microsc. Microanal. 21 (Suppl 3), 2015 592 machined exhibited anisotropic force-displacement curves. It is expected that the inclusion of these anisotropic characteristics of the force-displacement curve in the model will improve its accuracy. Reference: [1] B. L. Boyce et al. Int J Fract 186 (2014) p.5-68 [2] Amy Allen provided SEM images. [3] Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000. 1 mm 1 mm Figure 1: Sample geometry Figure 2: Right side fractured ligament Figure 3: Left side fractured ligament Sample 1 Sample 2 Sample 3 Surface Quarter Center 1 mm Figure 4: Cross sections of right side ligaments prior to fracture.");sQ1[297]=new Array("../7337/0593.pdf","Cross-sectioning Sub-millimetre Sized Defects in Ornamental Chrome-plated Component","","593 doi:10.1017/S1431927615003761 Paper No. 0297 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Cross-sectioning Sub-millimetre Sized Defects in Ornamental Chrome-plated Component A.J. Lockley1 1. Canadian Nuclear Laboratories, Chalk River Labs, Chalk River Ontario Canada Ornamental chrome components are desirable due to their lustrous and smooth appearance; however blemishes can occur and undermine their appearance. Finding the root-cause of such blemishes is of great importance to manufacturers so that process changes or improvements can be made. In one such case an ornamental chrome component had developed rust-coloured stains after field testing. The component comprised a flash chrome plating on a nickel-plate base supported by a steel substrate. A unique combination of metallographic- and microscopymethods were used to isolate the source of the staining. A digital microscope (DM) in conjunction with a Scanning Electron Microscope (SEM) was a synergistic tool set effective in locating and imaging small surface defects. The DM is a recent evolution of the optical microscope that was particularly useful for its ability to scan wide fields of view at lower magnification with the ability to zoom to higher magnification without having to change objective lenses while maintaining superior depth of field compared with standard optical methods. This instrument proved invaluable for the detection of small circular defects frequently associated with surface stains as seen in Figure 1. Circular defects were examined using the SEM to obtain higher resolution images as seen in Figure 2. The circular defects ranged in size from about 100�m to 200�m in diameter. Cross-sectioning a feature between 100�m and 200�m is challenging using standard metallographic techniques as the removal rates of material of such a thickness can occur in seconds during grinding operations. This being the case serial sectioning by grinding can be a risky operation when features of evidence are rare. To increase the probability of successfully cross-sectioning one of these sub-millimetre defects, sections for examination were mounted in a transparent mounting material. Although the mounting material was transparent it was still difficult to detect small features inside the mount. To facilitate visual contact with the small features an optically clear "window" was made on the side of the mount by grinding and polishing a facet on the side of the mount parallel to the surface containing the defect. Serial sectioning by grinding was used but progression was monitored by observation into the mount with the aid of a stereo microscope. Visually aided serial-sectioning was successful in producing metallographic cross-sections through the centre of the circular defects. This revealed that circular defects tended to be isolated to the outer surface of the nickel plating and if they occurred in a region where the nickel plate was sufficiently thick the defects had limited depth as seen in Figure 3. However, in regions where the nickel plate was relatively thin, a pit in the steel substrate was commonly observed as seen in Figure 4. The circular nature of the defects suggest they result from the presence of gas bubbles that prevented a uniform coverage of the surface. Corrosion to the steel substrate would occur in areas where nickel plating thickness was inadequate to provide corrosion protection. As corrosion progressed the corrosion by-products seeped out of the pit and stained the chrome surface. Microsc. Microanal. 21 (Suppl 3), 2015 594 Figure 1 Digital Microscope image showing a circular defect with surrounding rust-coloured staining on a chrome plated surface. Figure 2 An SEM image showing detail of one of the circular defects. Figure 3 Cross-section of the profile of a circular defect in the nickel plate on a steel substrate. Figure 4 Cross-section through a circular defect with a corrosion pit beneath.");sQ1[298]=new Array("../7337/0595.pdf","Preparing and Etching Titanium to Document Surface Affects","","595 doi:10.1017/S1431927615003773 Paper No. 0298 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Preparing and Etching Titanium to Document Surface Affects Luther M. Gammon1, Rodney R. Boyer2 1. 2. Materials Microscopy Consultant, Seattle, WA, USA Titanium Consultant, Seattle, WA, USA Because of its properties, titanium is one of the more difficult materials to prepare for metallographic examination. [1] [2] This presentation is intended to further illustrate this point. The challenge is to understand and minimize preparation artifacts. Artifacts can be introduced if the wrong preparation steps are taken. For example if one used the same metallographic process used on a body centered cubic martensitic steel, then the etched microstructure would not be a true representation of titanium. This presentation will cover: 1) etchant selection and exposure times, 2) effects of mechanical deformation and cold work, 3) heat affected zones and 4) alpha case, a surface enrichment due to oxygen contamination during thermal exposure at temperatures >~540�C. Micrographs and some of the terminology and processes discussed in this presentation can be found in published literature. [1] [2] All materials have unique properties which should be taken into account when preparing and etching cross-sections to bring out its microstructure. While similar to some materials, titanium is one of the most difficult to bring out its microstructure. It is very difficult to not introduce cold work and heat distortion during the sample preparation. There are also a variety of challenges with using etchants with titanium including application and exposure time of the etchant. For an example, one can etch the base microstructure and hide the hydrides A specialized etch is required for hydrides. There is no tolerance for not having a properly prepared sample free of preparation-induced artifacts such as a deformation layer or one caused by heating during preparation which will mask the true microstructure. Selection of the etchant, etchant times, and etchant application method will all impact the final results. Most etchants should be applied by immersing the reagent on to the sample. A chem-milling etchant like Kroll's will work either by swabbing or an immersion (dip). A tint etchant is one that depends on the etching products to remain on the sample, which will align with the grain direction. After application, the tint etchant is very fragile and easily disturbed. If the etchant is swabbed with a brush for example, it will alter and possibly destroy the appearance. After applying the tint etch, just rinse the specimen off with tap water. Do not brush, swab, or wipe off the surface, just rinse and blow dry. With a tint etchant and polarized light the grain aliment will add to or depolarize the light, giving a highly contrasting microstructure. If the specimen doesn't etch it is likely due to sample preparation. The chem-milling etchant feature has an optimum removal depth not to exceed the depth of field of the objective lens, 400 to 1200 microns. Tint etchant products also have an optimum depth build up. If the etch is applied too long, grain contrast will start to fade. The micrographs in Figure 1 demonstrate the clarity differences of the alpha case due to oxygen contamination observed on Ti surfaces, etched with a tint or chem milling etch (Krolls). Microsc. Microanal. 21 (Suppl 3), 2015 596 Figure 1 a,b,c & d: Ti-6Al-4V etched as indicated, with the number-s being the etch time in seconds. These micrographs demonstrate that the ABF and Oxalic etches/tints will normally show an alpha case much more clearly than the Kroll's etch. Kroll's will normally show it, but the difference between the alpha-case and matrix is often much more subtle as shown in the figure. Cracks are seen for more extreme cases. REFERENCES [1] L.M. Gammon, et al, Metallography and Microstructures, ASM Handbook Vol. 9, 2004, ed. G.F. Vander Voort, (ASMI, Materials Park) p. 899-917. 2. L.M. Gammon and R.R. Boyer, Practical Metallography, 4 (2013) p.263-280");sQ1[299]=new Array("../7337/0597.pdf","Microstructure Enhancement Using Ion Beam Milling","","597 doi:10.1017/S1431927615003785 Paper No. 0299 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microstructure Enhancement Using Ion Beam Milling Larry D. Hanke and Dieter Scholz Materials Evaluation and Engineering, Inc. Plymouth, MN, USA Microstructure analysis, typically a key component of any failure analysis, relies on optimum sample preparation to produce reliable data. However, sample preparation has become more challenging in recent years due to high-technology materials, complex assemblies, and smaller components. Although mechanical cross sectioning, polishing, and chemical etching are sufficient for many applications, broad-beam argon ion milling (AIM), using high-energy ion bombardment to remove material or modify the surface of a specimen, provides an additional level of quality and clarity for critical and difficult-to-prepare samples for SEM and light microscopy inspection. With the ion gun directing energetic argon ions toward the specimen at a low angle of ion incidence with respect to the sample surface, material is gradually removed at the atomic level - cleaning and polishing the sample. This technique is especially useful for multiphase materials with low hardness or soft constituents that present a challenge to obtain undeformed, scratch-free, flat specimens by mechanical polishing. Polishing by ion milling at low-angles produces flat surfaces with no deformation or other disturbance of the material microstructure. Figure 1 shows tin plating on a copper substrate with an intermetallic layer and very thin copper oxide layer under the tin. Surface modification to produce contrast or microstructure can be realized with higher milling angles up to 90�. This provides a great advantage for medical devices and other components consisting of noble metals, like gold and platinum, or corrosion-resistant alloys bonded to less noble metals, which are frequently difficult or impossible to prepare with chemical etching. Figure 2 compares traditional preparation of a tantalum-stainless steel weld joint where the fusion zone microstructure is obliterated during etching to a microstructure clearly resolved after ion milling. Microstructure contrast with ion milling is also useful or extremely fine structures that can be obscured by chemical etching. For failure analysis, the ion mill produces specimen surfaces with no chance of smearing over cracks or other fine flaws. Figure 3 shows very fine cracks at the fusion boundary of a laser weld joining dissimilar metal alloys. The ion milling also eliminates potential contamination at flaws that could be introduced during mechanical polishing. The substrate surface flaws in Figure 1 contain silicon oxide, which could be mistaken for polishing media residue if the sample had been mechanically polished. Additionally, the ion mill can be used to directly prepare cross sections by cutting through a sample with the argon-ion beam. This is effective for samples that are difficult to section mechanically, such as semiconductors, multi-layer structures, or material combinations with large hardness differences. Multilayer structures are revealed without the distortion that can occur with mechanical polishing, and sections are located with high accuracy on microscopic components. Finally, the ion mill can be used f to expose internal component layers for failure analysis. Since material removal parameters can be closely controlled with ion milling, subsurface conditions can be revealed with significant mechanical damage or chemical attack. Typical uses of this technique are exposure of surface contamination on packaged integrated circuits and flaws under plating layers on printed circuits. Microsc. Microanal. 21 (Suppl 3), 2015 598 596 Figure 1. Microstructure of delaminating tin plating on copper substrate. Figure 2. Comparison of traditional preparation (a) with ion mill (b) tantalum-stainless steel weld. Figure 3. Ion-milled specimen showing microstructure and cracks at weld joining platinum to stainless steel.");sQ1[300]=new Array("../7337/0599.pdf","Characterization of Hot-Compressed Magnesium Alloys in a Scanning Electron Microscope","","599 doi:10.1017/S1431927615003797 Paper No. 0300 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of Hot-Compressed Magnesium Alloys in a Scanning Electron Microscope Shirin Kaboli, Hendrix Demers, and Raynald Gauvin Department of Mining and Materials Engineering, McGill University, Montr�al, Canada. Under specific operation conditions in a scanning electron microscope (SEM), the secondary electron (SE) and backscattered electron (BSE) contrast include a contribution from the changes in crystallographic orientation of the bulk specimen. The crystallographic contrast is about 5% of the total signal and superimposed on the normal topographical and compositional contrast. In a non-deformed polycrystalline microstructure, the crystallographic contrast is commonly observed between grains of different crystal orientation, known as grain contrast or simply orientation contrast. In a deformed polycrystalline microstructure, local crystallographic contrast is observed inside each grain, indicating the variation of crystal orientation due to deformation. The variation of crystal orientation can be random or regular. If a deformed grain contains a mosaic structure of sub-grains with random crystal orientation relative to the nominal crystal orientation of the parent grain, a random crystallographic contrast is observed inside the grain. This contrast is known as sub-grain contrast since it is attributed to the presence of misoriented sub-grains inside the high angle grain boundaries of a deformed grain. If deformation occurs in a progressive manner inside a grain, the variation of crystal orientation is not random. As a result, a regular crystallographic contrast is observed inside the grain. This contrast is known as bend contour contrast due to the crystallographic origin and similar appearance to the bend contours observed in micrographs of bent thin foils in a transmission electron microscope [1]. In this study, the effects of SEM operation conditions were studied on the appearance of bend contour contrast in a hot-compressed Magnesium (Mg) alloy specimen. The Mg-0.2Al-0.3Ca (wt%) alloy was selected for the uniaxial hot-compression testing at a temperature of 400 �C, a strain rate of 0.01s-1, and a strain of 0.6 using a 100 kN servo-hydraulic materials testing system. The BSE imaging was carried out at 1-3� specimen tilt range, a 10 and 30 keV electron beam energy and a 7 mm working distance using a Hitachi SU-8000 cold-field emission SEM. The electron backscatter diffraction (EBSD) crystal orientation mapping was carried out at 70� specimen tilt, a 20 keV electron beam energy and a 25 mm working distance. Fig. 1 shows the microstructure of the deformed specimen after the uniaxial hot-compression test. Fig. 1a shows the BSE micrograph of the area inside an elongated deformed Mg grain. A large and nonuniform bend contour contrast in the form parallel contours was observed inside this grain. In addition, two horizontal scratches from the mechanical grinding and polishing steps were indicated with red arrows. Using the different SEM operation conditions such as stage tilt and electron beam energy, a number of examinations were carried out to study the origin of the parallel contours of contrast observed in Fig. 1a and differentiate them from the internal deformation substructures such as low angle boundaries. Fig. 1a-c show the BSE micrographs obtained at 1�, 2�, and 3� stage tilt and a beam energy of 30 keV, respectively. Fig. 1d-f show the BSE micrographs obtained at 1�, 2�, and 3� stage tilt and a beam energy of 10 keV, respectively. For stage tilt examinations, the offset of the electron beam axis versus stage tilt center was corrected for a eucentric tilt. The position, width, and BSE intensity of contours varied when the stage was tilted. Also, the position of contours remained constant with a change of the electron beam energy. However, the width and BSE intensity of the contours varied with a Microsc. Microanal. 21 (Suppl 3), 2015 600 change of the electron beam energy. The width of contours decreased with an increase in electron beam energy. These examinations clearly indicated the dependence of the contours to the local crystal orientation relative to the electron beam. Furthermore, the inverse pole figure (IPF) coloring map of the same area is displayed in Fig. 1g. Color variation associated with crystal misorientation and low angle grain boundaries were observed in this grain. These results show that bend contour contrast in the BSE micrograph is related to the local crystal misorientation (i.e. rotation of the Mg crystal across the grain) during deformation [2]. References: [1] D C Joy in "Quantitative scanning electron microscopy", ed. D. B. Holt, (Academic Press, London), pp. 131-181. [2] S Kaboli, H Demers, N Brodusch and R Gauvin, J. Appl. Crystallogr. (2014) Submitted. [3] The authors thank Dr. S Zaefferer at Max-Planck Institute for Metals Research (MPIE) for his technical assistance and contributions to this work. Figure 1. The microstructure of Mg-0.2Al-0.3Ca (wt%) alloy after the uniaxial hot-compression test. Backscattered electron micrographs obtained (a) 1� (b) 2� (c) 3� at a beam energy of 30 keV and (d) 1� (e) 2� (f) 3� at a beam energy of 10 keV. (g) an inverse pole figure coloring map of the same area. The position, width and BSE intensity of contours varied when the stage was tilted. The width and BSE intensity of the contours varied with a change of the electron beam energy. The appearance of the contours of contrast was related to the rotation of the Mg crystal across the grain during deformation.");sQ1[301]=new Array("../7337/0601.pdf","Characterization of a Sub-Grain Boundary Using Accurate Electron Channeling Contrast Imaging","","601 doi:10.1017/S1431927615003803 Paper No. 0301 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of a Sub-Grain Boundary Using Accurate Electron Channeling Contrast Imaging H. Mansour1, M.A. Crimp2, N. Gey1,3,N. Maloufi1,3 1 Laboratoire d'�tude des Microstructures et de M�canique des Mat�riaux (LEM3), Universit� de Lorraine, 57045 Metz, France 2 Michigan State University, Department of Chemical Engineering & Materials Science, 428 S. Shaw Lane, East Lansing, MI 48824, United States 3 Laboratory of Excellence on Design of Alloy Metals for low-mAssStructures (DAMAS), Universit� de Lorraine, France Electron Channeling Contrast Imaging (ECCI) is a powerful technique in scanning electron microscopy (SEM) for observing and characterizing crystallographic defect such as dislocations, stacking faults, and grain boundaries. In order to detect defects, ECCI uses the fact that backscattered electrons are very sensitive to the angle between the incident beam and the crystal lattice. To characterize defects, it is necessary to carry out ECCI under controlled two beam channeling conditions. In the past, these imaging conditions were established using electron channeling patterns (ECPs), selected area channeling patterns (SACPs), or electron backscatter diffraction patterns (EBSD), but all of these approaches have either spatial or angular resolution limitations. We have recently developed a novel approach for collecting high angular (0.1o) and spatial resolution (~500nm) selected area channeling patterns (HR-SACPs) that allows high accuracy control of ECCI channeling conditions [1,2].This approach, termed Accurate ECCI (A-ECCI), has been applied to unambiguously characterize screw dislocations in IF-steel using the gb=0 invisibility criteria [1]. In this contribution, A-ECCI performed in a Zeiss AURIGA 40 FIB SEM is used to analyze a sub-grain boundary in a polycrystalline bcc IF-steel slightly deformed in tension. HR-SACPs and ECC images were collected using a large four-quadrant Si-diode backscattered electron detector. Prior to the HR-SACP collection, the sample was tilted to 70� to determine the orientation of the region of interest by EBSD. Since A-ECCI is carried out at low tilt, the EBSD data was used to determine the approximate orientation of the crystal at 0� tilt. Specific channeling conditions were then established by HR-SACP in conjunction with tilt and rotation of the sample. Figures 1a and b show ECC images of the sub-grain boundary composed of individual dislocations, which appear as white line segments. HR-SACPs collected from either side of the boundary (indicated on Figure 1a) reveal a disorientation of 0.2�, with the rotation occurring along the red dotted line. The spacing between dislocations in the boundary increases from the lower left side to the upper right side of the sub-boundary until the boundary vanishes, indicating that the disorientation decreases along the boundary. Using the dislocations projected length (450nm) and a depth visibility of 60-70nm, the inclination angle was calculated to be about 6�-9�. The dislocation lines show fading dotted contrast, indicating their sense of inclination, which is plotted in red on the stereographic projection in figure 1f. The trace and inclination correspond to the [10-1] direction. The dislocations along the sub-grain boundary were characterized using the gb=0 and g(bxu)=0 invisibility criterion, as shown in figure 1.b-e. The dislocations go out of contrast for g=(10-1), consistent with dislocations having a Burgers vector b=[1-11] and being edge in character. This is also consistent with the Microsc. Microanal. 21 (Suppl 3), 2015 602 misorientation axis of the boundary, and consistent with the boundary being tilt in character. Finally, the average spacing of the dislocations in the boundary is in close agreement with the calculated ~ for tilt boundaries, and b=�[1-11]. This work demonstrates that A-ECCI assisted by HR-SACPs is a powerful technique and experimentally robust method for studying the misorientations of sub-boundaries and dislocations that make them up. The misorientation of low angle boundaries with misorientations as small as 0.2�, below the capabilities of standard EBSD systems, can be characterized along with the nature of the dislocations that make up the boundaries. References: [1] H.Mansour, J.Guyon, M.A Crimp, N.Gey, B.Beausir, N.Maloufi, ScriptaMaterialia 84-85 (2014) p 11-14 [2] J.Guyon, H.Mansour, M.A Crimp, N.Gey, S.Chalal ,N.Maloufi, Ultramicroscopy.149 (2015) 34-44. Figure1. Characterization of a sub-boundary using A-ECCI. (a) ECC Image of the subboundary. Two HR-SACP were acquired from each side of the boundary.(b-e)Burgers vector analysis on the dislocations applying gb=0 and g(bxu)=0 invisibility criterion using different g vectors (b) g=(01-1); (c) g=(1-10); (d): g=(1-21); (e): g=(10-1). (f) Stereographic projection showing {111} and {110} poles.");sQ1[302]=new Array("../7337/0603.pdf","3D Crystallographic Imaging Using Laboratory-Based Diffraction Contrast Tomography (DCT)","","603 doi:10.1017/S1431927615003815 Paper No. 0302 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 3D Crystallographic Imaging Using Laboratory-Based Diffraction Contrast Tomography (DCT) A. Merkle1, C. Holzner1, M. Feser1, S. McDonald2, P. Withers2, W. Harris1, E. Lauridsen3, P. Reischig3, H. Poulsen3, L. Lavery1 1 2 Carl Zeiss X-ray Microscopy, Inc., Pleasanton, CA USA Manchester X-ray Imaging Facility, School of Materials, University of Manchester, Manchester M13 9PL, UK 3 Xnovo Technology ApS, 4600, K�ge, Denmark Traditional X-ray tomography has, for some time, operated under a single absorption-based contrast mechanism. However, in recent years X-ray imaging has experienced a dramatic increase in the range of accessible imaging modalities � extending the classical absorption contrast with e.g. phase contrast, dark-field contrast, fluorescence, diffraction contrast, etc. Common for almost all such new imaging modalities are that they were developed at synchrotron facilities, and then � for some � have since been implemented on laboratory X-ray systems. [1,2] Crystallographic imaging is primarily known from electron microscopy, and particularly the introduction of the electron back-scattering diffraction (EBSD) technique in the early 1990's, has made it a routine tool for research and/or development related to metallurgy, functional ceramics, semi-conductors, geology etc. The ability to image the grain structure in such materials is instrumental for understanding and optimization of material properties and processing. [3] However, the destructive nature of 3D EBSD prevents the technique from directly evaluating the microstructure (and grain-orientation) evolution when subject to either mechanical, thermal or other environmental conditions. Conversely, nondestructive X-ray imaging methods allow for such `4D' time dependent studies, but to date have been primarily the domain of a limited number of synchrotron facilities. [4] Here we present the development and application of a new method, termed diffraction contrast tomography (DCT), to extend the capabilities of lab-based X-ray tomography systems to nondestructively analyze 3D crystallographic information of polycrystalline samples. The technique is implemented on a commercial laboratory X-ray microscope that utilizes a polychromatic divergent beam and a synchrotron-style detection system. During data collection, rotation and imaging of the sample yields a series of diffraction patterns generated by the sample crystallites when the Bragg condition is satisfied. The patterns are then reconstructed to yield grain orientation, center of mass, and size for a large number of grains. This work will present a selection of results of laboratory DCT, including a beta titanium alloy sample and a microstructure of sintered copper spheres. Discussion of the implementation will include the boundary conditions and capabilities of the method, including coupling to in situ environments within the microscope or subjecting samples to extended time evolution experiments (across days, weeks, months). In addition, ways in which the DCT method can be correlatively coupled to related characterization techniques will also be examined, such as by employing focused ion beam serial sectioning for destructive, but complementary, characterization of the same volume following a DCT evolution experiment. Microsc. Microanal. 21 (Suppl 3), 2015 604 [1] A. P. Merkle et al., Ascent of 3D X-ray Microscopy in the Laboratory, Microscopy Today, 21 (2013), p. 10 [2] E. Maire and P. Withers, Quantitative X-ray tomography, International Materials Review, 59 (2014) p. 1 [3] "Electron Backscatter Diffraction in Materials Science," ed. A. Schwartz et al., Springer US (2009) [4] W. Ludwig et al., X-ray diffraction contrast tomography: a novel technique for threedimensional grain mapping of polycrystals. 1. Direct beam case, J. Appl. Crystallogr., 41 (2008), p. 302-309 L L Figure 1. Experimental arrangement for lab DCT. A polychromatic source is located on the right, with X-rays proceeding through an aperture to illuminate the sample located in the center. A series of exposures are collected while the sample is rotated, creating diffraction patterns on the detector on the left, where the primary beam is blocked with a beam stop. Figure 2. A cube plot representation of results from applying the DCT method to a beta titanium alloy sample. Each cube represents the location and relative size of an individual crystallite, with orientation denoted by color as shown in the inverse pole figure.");sQ1[303]=new Array("../7337/0605.pdf","Characterization of Elastic Strain Field and Geometrically Necessary Dislocation Distribution in Stress Corrosion Cracking of 316 Stainless Steels by Transmission Kikuchi Diffraction","","605 doi:10.1017/S1431927615003827 Paper No. 0303 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of Elastic Strain Field and Geometrically Necessary Dislocation Distribution in Stress Corrosion Cracking of 316 Stainless Steels by Transmission Kikuchi Diffraction Arantxa Vilalta-Clemente1, Martina Meisnar1, Sergio Lozano-Perez1, and Angus J. Wilkinson1. 1. Department of Materials, University of Oxford, Parks Road, Oxford OX1 3PH, United Kingdom. Stainless steel alloys such as SUS 316 are widely used for the application in nuclear power plants because of their excellent performance in high-temperature and corrosive environments. In this work, a stress corrosion crack from a sample tested under simulated primary water from a pressurized water reactor has been characterized. Transmission electron backscattered diffraction (t-EBSD) or transmission Kikuchi diffraction (TKD) in the scanning electron microscope (SEM) is emerging as a very promising approach to characterize materials at the nano-scale. TKD involves the analysis of electron transparent samples, similar to those prepared for transmission electron microscopy (TEM). Due to smaller interaction volume of the incident beam with the specimen, the spatial resolution provided by TKD can be below 10 nm, which is a significant improvement over conventional EBSD [1-3]. The TKD geometry is derived from the basic EBSD setup with the main difference being the specimen position and its orientation with respect to the SEM column. The thin foil sample is kept almost horizontal (10 tilt) with respect to the incident electron beam, as opposed to the 70 tilt required for conventional EBSD. A Zeiss Merlin Field-Emission SEM equipped with an EBSD Bruker e-Flash HR detector with Forescatter Diodes (FSDs) below the phosphor screen was used for the data acquisition. The ability to alter the camera tilt provides additional flexibility in setting up the geometry. In this work, we have analyzed TKD patterns using the cross-correlation algorithms presented by Wilkinson et al [4] initially for high angular resolution analysis of EBSD patterns. As in EBSD, changes in elastic strain and lattice rotations cause small shifts in the positions of zones axes and other features in the TKD patterns. Cross-correlation is used to measure the variations between each test pattern and a reference pattern in at least 35 sub-regions distributed across the pattern. The dispersion of shifts across the pattern is used to determine the change in lattice strain and rotation relative to the reference point. Figure1a was obtained using colour coding of the signals from the FSDs and provides information on the orientation contrast within the stainless steel sample, and reveals the location of the crack, the misorientation of the grains on either sides of the crack, the existence of twinning deformation and slip. TKD maps were obtained using 11 nm step size, with the detector binned to record patterns at 160 by 120 pixels. We have used the cross correlation method to study the deformation around the crack tip. Example maps for a selected strain component (12 shear strain) and two rotation components (12 about the sample normal, and 13 about the vertical axis) are shown in Figure 2. The in-plane shear strain 12 distribution shows a maximum in the top grain, at the intersection of the grain boundary with the deformation bands. The 12 lattice rotations are shown in Figure 2b, where larger values are observed in the region below the crack in the bottom grain and between the deformation bands in the top grain. We have used the resulting rotation tensor and the Nye geometrically necessary dislocations (GNDs) to Microsc. Microanal. 21 (Suppl 3), 2015 606 assess the dislocation content in the 316 Stainless Steel sample. Accumulation of GNDs on the deformation bands, twining and slip bands as well near the open crack between the two grains and the grain boundary itself is observed (Figure 1b). The deformation history of this type of samples indicates that the band intersecting the grain boundary is deformed bands with high dislocation density. Furthermore, TKD has been used to measure the grain boundary orientation and establish a gauge for a quantifying plastic deformation at the crack tip and other regions in the surrounding matrix [5]. References: [1] R R Keller and R H Geiss, J. Microsc 245 (2012), p. 245. [2] P W Trimby, Ultramicroscopy 120 (2012), p. 16. [3] P W Trimby et al, Acta Materialia 62 (2014), p. 69. [4] A J Wilkinson et al, Ultramicroscopy 106 (2006), p. 307. [5] M M et al, Journal of Nuclear Materials (submitted). [6] The authors acknowledge funding from EPSRC grant No: EP/J015792/1 & EP/J016098/1 and Areva (UK). Samples were provided by INSS (Japan). (a) (b) crack Grain boundary 1 m Figure 1. (a) FSE image showing the crack and the grain boundary (b) GND density map generated using 11 nm step size and the colour scale is dislocations per m2 on a log10 scale. (a) (b) (c) Figure 2. Example strain and rotations.");sQ1[304]=new Array("../7337/0607.pdf","Transformation and Deformation Characterization of NiTiHf and NiTiAu High Temperature Shape Memory Alloys.","","607 doi:10.1017/S1431927615003839 Paper No. 0304 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Transformation and Deformation Characterization of NiTiHf and NiTiAu High Temperature Shape Memory Alloys. L. Casalena1, D. R. Coughlin1,2, F. Yang1, X. Chen1, H. Paranjape1, Y. Gao1, R. D. Noebe3, G. S. Bigelow3, D. J. Gaydosh3, S. A. Padula3, Y. Wang1, P. M. Anderson1, M. J. Mills1. 1. 2. Department of Materials Science & Engineering, The Ohio State University, Columbus, OH, USA. Los Alamos National Laboratory, Los Alamos, NM, USA. 3. NASA Glenn Research Center, Cleveland, OH, USA. NiTiHf and NiTiAu are exciting candidates amongst an emerging class of high temperature shape memory alloys (HTSMAs), exhibiting properties conducive to actuator applications in demanding automotive and aerospace environments. NiTiHf can be tailored to achieve a highly favorable balance of properties, including high strength, stability, and work output at temperatures up to 300�C, yet at a reduced cost compared to Pt, Pd, and Au containing counterparts [1]. NiTiAu shows potential for work output at much higher temperatures � where the benefits may offset cost � but additional research is needed. These investigations focus on developing a fundamental understanding of the inherent microstructure-property relationships for these NiTi-based HTSMAs. The attractive characteristics seen in many of these systems are strongly influenced by the formation of nano-scale precipitates. Advanced electron characterization techniques are used to explore the interaction of these precipitates with martensite at low temperature, and with dislocations at higher temperature. These insights are further incorporated into microstructural modeling frameworks with the aim of developing accurate simulations of polycrystalline functional response. Two compositions of NiTiHf (Ni51Ti29Hf20 and Ni50.3Ti29.7Hf20) and four compositions of NiTiAu (Ni11Ti49Au40, Ni10.3Ti49.7Au40, Ni10Ti50Au40, Ni9Ti51Au40) were prepared at NASA Glenn Research Center. Small cylindrical bars were cut for mechanical testing and heat treatments. TEM samples were trenched and thinned to electron transparency using a focused ion beam (FIB) on a FEI Nova 600. Dislocation analysis was conducted in scanning transmission electron microscopy (STEM) mode using a FEI Tecnai F20 at 200kV. Atomic resolution high angle annular dark field (HAADF) STEM micrographs were obtained using a probe-corrected FEI Titan 80-300 at 300kV. Investigations have highlighted the powerful effects of temperature, composition, and aging on critical stress and transformation temperature. Figure 1 shows the results of isothermal compression tests on NiTiHf, where the increasing slope of each curve represents the critical stress to form stress-induced martensite, while the decreasing slope represents the critical stress for slip. These alloys have exhibited up to 4% fully recoverable pseudoelastic strain, and yield strengths as high as 2 GPa [2,3]. Aging at intermediate temperature results in the formation of nano-scale H-phase precipitates, as shown in Fig. 2a,b, which have complex but beneficial effects on alloy behavior. Structural analysis of the precipitate was completed in previous publications [4], but recent efforts have been aimed at understanding how the precipitates interact with the surrounding matrix in such a way that the martensitic transformation is near-perfectly accommodated. In addition to attempts to understand precipitate-martensite interactions at low temperatures, dislocation activity at higher temperatures is also of great interest. The creation or movement of dislocations during the martensitic transformation leads to unrecovered strain, which may accumulate with subsequent cycles, leading to functional fatigue through degradation of the shape memory response. Initial dislocation analysis has shown that bands of dislocations with <001>-type Microsc. Microanal. 21 (Suppl 3), 2015 608 Burgers vectors are most commonly observed [3]. In Fig. 2c,d, dislocations labeled 1, 2, and 3 refer to dislocations with Burgers vectors a[001], a[010], and a[100] respectively. Similar investigations into aged samples are currently underway. NiTiAu has been the subject of far less research interest than NiTiHf, possibly due to the high cost of Au, however recent mechanical testing data has revealed work output capability at significantly higher temperatures than what has been observed in similar NiTi-based HTSMAs. In addition, these alloys have shown minimal variation in transformation temperature and mechanical properties as a function of composition spanning stoichiometry. This could be a result of secondary phase formation giving rise to a quasi-equiatomic state, or alternatively weak atomic site preference compared to other NiTi-based SMAs. Further investigations are in progress. [1] H E Karaca et al, Mater Sci Tech 30 (2014), p. 1530. [2] D R Coughlin, PhD Dissertation, The Ohio State University (2013). [3] D R Coughlin et al, Mater Sci Eng A (2015), submitted for publication. [4] F Yang et al, Acta Mater 61 (2013), p. 3335. [5] The authors acknowledge funding from the US Department of Energy, Office of Basic Energy Sciences, under Grant No. DE-SC0001258. Figure 1. 0.2% offset critical stress of aged NiTiHf alloys during isothermal compression testing [2,3]. Figure 2. (a),(b) HAADF STEM images of H-phase precipitates in Ni50.3Ti29.7Hf20 aged 3hrs at 650oC, centered on the [010]H([110]B2) zone, with orientation relationship shown. (c),(d) Dark field STEM diffraction contrast images of as-extruded Ni51Ti29Hf20 post compression testing at 150oC, showing a<001> type dislocations imaged under different diffraction conditions [4]. See text for details.");sQ1[305]=new Array("../7337/0609.pdf","Crystallographic Orientation and Deformation Mechanisms of Laser-processed Directionally Solidified WC-W2C Eutectoids","","609 doi:10.1017/S1431927615003840 Paper No. 0305 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Crystallographic Orientation and Deformation Mechanisms of Laser-processed Directionally Solidified WC-W2C Eutectoids Wei-Ting Chen1 and Elizabeth C. Dickey1 1 Department of Materials Science and Engineering, North Carolina State University, Raleigh, NC, USA Ceramic composites with excellent mechanical performance have been used in applications such as cutting tools and ceramic drilling bits for many decades. Carbides are the most prevalent materials in these applications due to their high hardness, melting temperature, and wear resistance. However, most monolithic carbides suffer from low sinterability and low fracture toughness. WC-Co composites have been widely used for improving densification of WC while also improving fracture toughness, although the addition of the metal content sacrifices hardness, oxidation resistance, and thermal stability [1]. Directionally solidified WC-W2C eutectoids (WC-W2C DSE) at the eutectoid composition are being explored as an alternative composite structure that may provide a path for highly dense materials with high hardness without sacrificing oxidation resistance and thermal stability. WC-W2C DSEs at 40 at % carbon were produced at laser processing rates between 0.12-3.24 mm/s and a laser power of 2000 W. The relationship between laser processing rate (V) and interlamellar spacing ( ) follows the equation: , typical of a eutectoid transformation. The indentation hardness was found to increase with decreasing interlamellar spacing via a power-law relationship with an exponent of -2.4, with a maximum hardness of 28.2�0.33 GPa measured at an interlamellar spacing of 331�36 nm. This value is 28% higher than the highest hardness (22 GPa) measured in WC-Co composites (12 wt % Co) [2]. The present work aims to study microstructure, crystallography and deformation mechanisms in laserprocessed WC-W2C DSE, to understand the role of the microstructure on the hardness scaling. Microstructure and crystallographic orientation relationships of WC-W2C composites are investigated by electron backscattered diffraction (EBSD) in a scanning electron microscope. As shown in Fig. 1(a), a lamellar-typed microstructure is observed in all samples. As evident in Fig. 1(b), the majority of the material has the orientations of WC[ 2 ]//W2C[ 2 ] or WC[01 0]//W2C [01 0] parallel to the growth direction. In Fig. 1(c), the preferred in-plane orientation was found to be WC(0001)//W2C(0001), while higher misorientations may be due to the precipitation of the primary WC prior to the eutectoid reaction [3]. STEM images (Fig. 2) indicate that the interface plane is faceted parallel to WC(0001)//W2C(0001), and that it is semi-coherent, with the Burger's circuit indicating the presences of one misfit dislocation every 39 d(01 ) WC to accommodate the 2.6% lattice mismatch. Deformation mechanisms induced by indentation are studied by using bright-field TEM imaging under several two-beam conditions. As shown in Fig. 3(a)(b), although dislocations are observed in the asprocessed sample, shear banding and grain refinement are the dominant deformation mechanisms in W2C-rich regions. As shown in Fig. 4(a), dislocation networks with Burger's vector parallel to <11 0> bounding stacking faults with displacement vectors parallel to < 11 > are commonly observed in the primary WC phase. The regions containing eutectoid (fine lamellae) microstructure exhibit markedly different deformation mechanisms. Shear banding is suppressed in W2C lamellae. Moreover, as shown in Fig. 4(b), partial dislocations with the Burger's vector along < > are observed within the WC lamellae, and these dislocations and stacking faults are interrupted at the interfaces, which implies that interface density and corresponding microstructural length scale may have significant impacts on hardness of WC-W2C DSE [4]. Microsc. Microanal. 21 (Suppl 3), 2015 610 References: [1] Lee, H.C. et al, J. Mater. Sci. and Eng., 33 (1978), p. 125-133. [2] Michalski, A. et al, Int. J. Refract. Met. & Hard Mater., 25 (2007) , p. 153�158. [3] Kurlov, A.S. et al, Inorg. Mater., 42 (2006), p. 121�127. [4] The authors acknowledge funding from National Science Foundation (Grant # CMMI-1139792). (a) (b) (c) Figure 1 � EBSD (a) phase map (b) orientation map (// growth direction) (c) WC(0001)//W2C(0001) interface orientation map of DSE WC-W2C produced at 0.12 mm/s processing rate. (a) (b) Figure 2 � STEM image of the WC(0001)//W2C(0001) interface. (a) Figure 3 � TEM images of (a) undeformed (b) deformed W2Crich region. Samples were produced at 0.12 mm/s processing rate (b) W2C W2C b=[ ] WC W2C W2C WC Figure 4 � TEM bright field images of the indentation-induced deformed samples by 4.9 N load produced at 0.12 mm/s processing rate in (a) primary WC region and (b) WC lamellae region. The indexing of the dislocations is showed at the right side of each image.");sQ1[306]=new Array("../7337/0611.pdf","Staining Block Copolymers using Sequential Infiltration Synthesis for High Contrast Imaging and STEM tomography","","611 doi:10.1017/S1431927615003852 Paper No. 0306 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Staining Block Copolymers using Sequential Infiltration Synthesis for High Contrast Imaging and STEM tomography T. Segal-Peretz,1,2 J. Winterstein,3 M. Biswas,4 J.A. Liddle,3 Jeffrey W. Elam,4 N. J. Zaluzec,5 P.F. Nealey1,2 1 2 Institute for Molecular Engineering, University of Chicago, Chicago, USA Materials Science Division, Argonne National Laboratory, Argonne, USA 3 Center for Nanoscale Science and Technology, National Institute of Standards and Technology, Gaithersburg, USA 4 Energy Systems Division, Argonne National Laboratory, Argonne, USA 5 Electron Microscopy Center, NST Division , Argonne National Laboratory, Argonne, USA Block copolymers (BCP) are considered promising materials for advanced lithography in the sub-20 nm regime, as well as versatile building blocks for a variety of applications ranging from solar cells to membranes. Since the performance and functionally of BCP films is related to their three dimensional (3D) structure, it is desired to characterize the 3D structure using Transmission Electron Microscopy (TEM) and TEM tomography. A key challenge for TEM imaging of most BCPs is their poor contrast. Therefore, BCPs are routinely stained with conventional staining agents such as OsO4 and I2 to increase their scattering contrast. This staining process can lead to undesired swelling and artifacts.1 Figure 1 shows a comparison of annular dark field (ADF) scanning transmission electron microscopy (STEM) imaging of cylinder-forming and lamellae-forming poly(styrene-block-methyl methacrylate) (PS-b-PMMA) BCP films prior to any staining, and after staining with Al2O3 using sequential infiltration synthesis (SIS). SIS is an emerging technique for growing inorganic materials in polymer and block copolymer films using an atomic layer deposition tool.2 The inorganic material is grown in a block-selective manner within the polar domains of the BCP. During the SIS process, one of the precursors (trimethyl-aluminum) selectively interact with the polar moieties (carbonyl groups), resulting in Al2O3 growth that occurs predominantly in the PMMA domains, as can be seen in the x-ray energy dispersive spectropy (XEDS) STEM mapping of lamellae-forming PS-PMMA film (Figure 2). The introduction of Al2O3 significantly enhances the high-angle scattering from the polar domains, and enables high contrast imaging in ADF-STEM mode (Figures 1 and 2). The enhanced contrast of SIS-treated films under ADF-STEM conditions enables 3D characterization of PS-b-PMMA BCP films using STEM tomography. By tuning the SIS conditions, the Al2O3 can serve as a staining agent with high fidelity to the BCP domains, enabling detailed investigation of the 3D structure. For example, ADF-STEM tomography of cylinder-forming PS-b-PMMA film with perpendicular orientation (Figure 3), revealed the morphology of individual domains across the film's depth, defects in the self-assembled film, and grain boundaries. Typical dislocation defects are observed at the grain boundaries of the film. While most of the dislocations persist the entire film's depth, splitting dislocation, were also observed (Figures 3d and 3e), where a single cylinder splits into two cylinders towards the bottom of the film. Combining SIS and ADF-STEM tomography opens a window for studying the 3D structure of BCPs with high precision, leading to better understanding of their behavior. References: [1] Staniewicz L. et al., J. Phys.: Conference Series (2010), 012077 [2] Peng Q., Tsemg Y. C. and Elam J. W., Adv. Mater. 22 (2010), 5129. Microsc. Microanal. 21 (Suppl 3), 2015 612 [3] This work was supported by U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357. Figure 1: STEM images of as prepared film (a),(b), and after three cycles of Al2O3 SIS (c),(d), of cylinder-forming PS-PMMA (a),(c), and lamellae-forming PS-PMMA (b)(d). Film thicknesses are 50 nm and 80 nm, respectively. Figure 2: XEDS-STEM mapping of lamellaeforming PS-PMMA. (a) ADF-STEM image of lamellae-forming PS-PMMA after three cycles of Al2O3 SIS; (b),(c) are the corresponding elemental mapping of aluminum, and oxygen K x-ray lines, respectively. Scale bars are 50 nm. Figure 3: ADF-STEM tomography of cylinderforming PS-PMMA films treated with three cycles of Al2O3 SIS: (a) xy slice (parallel to the substrate) of the reconstructed volume taken from the middle of the film. (b) Visualization of the 3D reconstructed volume; film thickness is 50 nm. For clarity PMMA domains are colored in blue while PS domains are transparent. (c) 1.1 nm thick yz digitally sliced cross section of the reconstructed volume taken at the dashed yellow line in (a). (d),(e) yz cross section, and visualization of a splitting dislocation defect formed at grain boundaries of the film shown in (c).");sQ1[307]=new Array("../7337/0613.pdf","Understanding the Structure-Process-Property Balance in PC/PEI blends by Morphology (using Scanning Transmission Electron Microscopy in Field Emission - Scanning Electron Microscopy) and Correlative Deformation Mechanics","","613 doi:10.1017/S1431927615003864 Paper No. 0307 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Understanding the Structure-Process-Property Balance in PC/PEI blends by Morphology (using Scanning Transmission Electron Microscopy in Field Emission Scanning Electron Microscopy) and Correlative Deformation Mechanics P. Bajaj1, D. Bajaj2, Carl Strom1, H. Zhou2, and Kelly Leung2 SABIC Innovative Plastics, 1 Noryl Avenue, Selkirk, New York 12158 USA, 2SABIC Innovative Plastics, 1 Lexan Lane, Mt. Vernon, Indiana 47629 USA Nanoparticle dispersion has been reported to have a critical impact on the properties of polymer composite materials [1, 2]. Here in, we present the application of STEM in FESEM to study the phase morphology and correlative deformation phenomenon in Polycarbonate (PC)/ Polyetherimide (PEI) blends to bridge the understanding between functional and failing material performance at different processing conditions. PEI is an amorphous, transparent engineering thermoplastic with a unique chemical structure. The aromatic backbone offers superior heat resistance with high Tg ~216�C, low smoke, hydrolytic and improved solvent resistance and modulus. However, high Tg poses a challenge for processability, brittleness and limit its broader use. The colorability of PEI is also limited due to its natural amber color with yellowness index (YI) > 50. Blending PEI with PC, offers gainful advantage for PEI processability, imparts toughness and excellent colorability to the blend. PC/PEI exhibits a distinctly phase separated morphology with PEI phase forming a discrete phase residing in the continuous matrix of PC with pigment dispersion (Figure 1a). For an immiscible blend, the properties are mainly dominated by the continuous phase, but other variables such as the composition, viscosity of the blend, and processing conditions can influence the nanoparticle pigment dispersion and distribution, discrete phase size and shape, and interfacial free volume (compatibilization) yielding different tensile stress vs strain behaviors (Figure 1b,c). STEM in FESEM was investigated for PC/PEI blends as functional parts (control) compared to parts showing brittle failure, for bulk and edge phase morphology (Figure 2a, c). The specimen was cryomicrotomed and thin sections ~ 100-150 nm were collected on copper grids and vapor stained with RuO4 for 3 minutes. Inset morphology highlights the contrasting differences in nanoparticle pigment dispersion, where the control exhibits a uniform homogenous dispersion and distribution for the pigment nanoparticles, and the failed (brittle) part presents pigment agglomeration and preferentially localization at the interface of PC/PEI (Figure 2b,d). Furthermore, the edge morphology illustrates interesting elongation and stretching of PEI domains indicating surface enrichment with PEI for the failed part morphology compared to the control part (figure 2e, f). These morphological differences resulted in correlative toughness of these blends as depicted in Figure 3a-c. Toughness is a property which describes the absorption and dissipation of energy during deformation prior to fracture. In this study, toughness was measured by the area under the tensile stress versus strain curves. For instance, the control morphology presenting a homogenous dispersion of pigment nanoparticles would promote plastic yielding and deformation processes on a nano- and microscale across the volume of the PEI/PC blend and exhibit a ductile fracture due to energy absorption. On the contrary, agglomeration of pigment nanoparticles at the PC/PEI interface and depletion of PC on the surface plays detrimental to impact toughness. If the adhesion between the particles and the matrix is poor, such as in this study for the failed part due to pigment deposition at the PEI-PC interface, the stress concentration at the particle�matrix interface can interfere with the interfacial compatibilization between PEI and PC, and may act as stress precursors resulting in cracks and brittle fractures. In summary, combining morphological, micromechanical, and understanding of the fracture performance of a polymer resin (by identifying its preferred deformation mechanism, e.g., 1 Microsc. Microanal. 21 (Suppl 3), 2015 614 crazing or shear yielding), optimizations in process, molding and formulation conditions can be carried out to tune the design space geared towards robust materials development. References: [1] E. T. Thostenson, et.al., "Nanocomposites in Context," Composites Science and Technology, 65 (2005) 3. [2] H. Khare and D. Burris D, Polymer, 51 (2010): 719. (a) Stress in N/mm2 (b) (c) Stress in N/mm2 PEI y f f PC Pigment Strain in % f Strain in % f 2 �m Figure 1. (a) STEM in SEM morphology of an immiscible PC/PEI blend with pigment nanoparticles (black dots) dispersed homogenously. Typical tensile stress vs. strain response for (b) a ductile and (c) brittle fracture. Control Failed (a) (c) Bulk (b) (d) 2 �m 1 �m 2 �m 1 �m (e) (f) Pigment agglomeration at the PC/PEI interface Edge 1 �m 1 �m Figure 2. STEM in SEM morphology comparing bulk morphology for (a, b) control (c, d) failed PC/PEI blend. Inset (b) and (d) highlights the differences in pigment nanoparticle dispersion. Edge or surface morphology for control (e) and failed (f) PC/PEI blend. (a) Surface enrichment with PEI (b) (c) PEI PC Pigment Figure 3. Root cause failure analysis mechanism linked to pigment nanoparticle dispersion and blend compatibilization.");sQ1[308]=new Array("../7337/0615.pdf","A Green Method for Graphite Exfoliation Using High-Energy Ball Milling.","","615 doi:10.1017/S1431927615003876 Paper No. 0308 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A Green Method for Graphite Exfoliation Using High-Energy Ball Milling. I. Estrada-Guel1,2, F.C. Robles-Hernandez2, J.M. Mendoza-Duarte1, R. P�rez-Bustamante1 and R. Mart�nez-S�nchez1. 1. Centro de Investigaci�n en Materiales Avanzados (CIMAV). Laboratorio Nacional de Nanotecnolog�a Miguel de Cervantes No. 120, 31109, Chihuahua, Chih., Mexico. 2. Department of Mechanical Engineering Technology, University of Houston, Houston, TX 772044020, USA. Graphite, an allotropic and stable form of carbon, is a useful material constituted by multiple layers joined with covalent bonds linked together by a weak Van Der Walls interaction. The single units named graphenes have attracted considerable attention, because of their excellent mechanical, chemical, thermal, electrical properties and low thermal expansion coefficient [1]. The strong covalent (sp2) bonds in this unique honeycomb structure of graphene and their atomic scale thickness impart them with these unusual properties [2]. Taking advantage of the weakness of these interactions, it is possible to insert ions, atoms or molecules between the layers in order to obtain graphenes from graphite. In the exfoliation process, elimination of the intercalated species leads to a significant expansion up to hundreds of times along the c-axis, forming a highly porous material. Exfoliated graphite (EG) has been synthesized by galvanic, chemical and thermal treatments of the natural graphite. However, the chemical method is widely used because its simplicity and versatility. Usually EG is produced intercalating acid guest species between the stacked graphene layers via liquid-phase by reaction between graphite and 18M H2SO4 in the presence of strong chemical oxidants such as KMnO4, HNO3 or H2O2 [3]. However, reaction leftovers are highly toxic and corrosive materials that required careful manipulation and a special confinement. In the present work, some EG were prepared using a low cost and environmentally friendly mechanochemical route, which avoid the use of highly corrosive and dangerous reagents. Raw materials were: natural graphite and sodium carbonate (a white, odorless powder with an alkaline taste, which is domestically well known for its everyday use as a water softener) and citric acid (a weak organic acid used as a natural preservative/conservative and to add an acidic taste to foods and drinks). Both substances were used in this work as defoliation agents. An equiatomic mixture of sodium carbonate and natural graphite was milled in a SPEX high-energy mill for 1h. The milled and unmilled mixtures were leached with an aqueous citric acid solution (8% wt./wt.), hot refluxed 2h, washed with distilled water and dried at 80�C overnight. Morphological and chemical studies were performed with a scanning electron microscope (JSM5800-LV) and a high-resolution microscope (JSM-7201F). The Fig.1 shows some SEM micrographs of graphite/Na2CO3 mixtures with 0 and 1h of milling where bright zones correspond to high Na2CO3 concentration regions (black arrows). After milling Na2CO3 has been dispersed, obtaining a homogeneous mixture. Fig. 2 displays the composition of 1h-milled samples after leaching process. In Fig. 2 can be observed the reduction of Na2CO3 content after the chemical treatment with citric acid. Note, that aqueous washing solution is composed by sodium citrate, an innocuous substance commonly used as food antioxidant. The Fig. 3 exhibits some SEM micrographs of isolated graphite particles after the leaching process. It is evident the reduction of particle size and an increased level of defoliation with this green process [4]. Microsc. Microanal. 21 (Suppl 3), 2015 616 References: [1] V. Singh, D. Joung, L. Zhai, Progress in Materials Science 56 (2011) p. 1178-1271. [2] M. Yang, Y. Houd, N.A. Kotov, Nano Today 7 (2012) p. 430-447. [3] M. Inagaki, R. Tashiro, Y. Washino, J. Physics and Chemistry of Solids 65 (2004) p. 133-137. [4] This research was supported by CONACYT (Project No. 169262) and the Redes Tem�ticas de Nanociencias y Nanotecnolog�a (124886). a) b) Figure 1. SEM micrographs and corresponding EDS maps: a) Un-milled mixture (0h) and b) sample after 1h of milling. a) b) Figure 2. SEM micrographs and EDS elemental analyses: a) 1h milled mixture and b) 1h leached sample. 0h 1h Figure 3. SEM micrographs of graphite particles after the leaching process.");sQ1[309]=new Array("../7337/0617.pdf","Atomic Resolution STEM-EELS Study of Transition Electronic Localization State Induced by Strain.","","617 doi:10.1017/S1431927615003888 Paper No. 0309 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic Resolution STEM-EELS Study of Transition Electronic Localization State Induced by Strain. Yuze Gao,1 Manuel A. Roldan 2,3, Jincang Zhang,1 Liang Qiao,4 Miaofang Chi,2 David Mandrus,2,5 Xuechu Shen,6 Matthew F. Chisholm2, David J. Singh2, Guixin Cao4,5. Department of Physics, Shanghai University, Shanghai 200444, PR China. Materials Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831-6056, USA. 3. Dept. F�sica Aplicada III & Instituto Pluridisciplinar, Universidad Complutense de Madrid, Ciudad Universitaria, 28040 Madrid, Spain. 4. Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA. 5. Department of Materials Science and Engineering, University of Tennessee, Knoxville, TN 37996 6. National Laboratory for Infrared physics, Shanghai Institute of Technical Physics, Chinese Academy of Sciences, Shanghai 200083, People's Republic of China. 2. 1. Electronic localization in disordered systems is among the most interesting phenomena in condensed matter physics. Depending on the amount, kind of disorder and dimensionality [1], there are two types of localization, classical Anderson localization (AL) and weak localization (WL). The origin of the AL is multiple scattering interference of the waves due to the randomness in the potential, thus altering the nature of the wave functions [2], while WL typically occurs in disordered electronic systems at very low temperatures. Although the WL is generally regarded as a precursor of the AL transition, the underlying physics of this phenomena is not fully understand yet. To study the transition between weak and strong localization, we synthesize high-quality ZrO2 nanopillars with controlled dimensionality embedded in epitaxial La2/3Sr1/3MnO3 (LSMO)/LaAlO3 [110] using state-of-art pulsed laser epitaxy. This peculiar nanostructured system offers a great flexibility to manipulate the superficial structural disorder along the interface by controlling the size of ZrO2 nanopillar, thus strongly affects the electronic transport of the LSMO matrix. To characterize this superstructure and demonstrate the precise control of its electronic properties, we performed a combined study of Electron Energy Loss Spectroscopy (EELS) with the Z-contrast high angle annular dark field (HAADF) imaging technique in the scanning transmission electron microscope (STEM) Nion UltraSTEM 200, a unique tool that allow us to obtain simultaneously composition, chemistry and structure of materials with atomic resolution and sensitivity. These techniques, together with geometric phase analysis allow us to determine the spatial distribution of the epitaxial strain across the samples and clarify its impact on the various electronic localization mechanisms. Microsc. Microanal. 21 (Suppl 3), 2015 618 References: [1] T. Schwartz et al. Nature 446 (2007), p. 52-55 [2] P. Lee and T.V. Ramakrishnan, Rev Mod Phys 57 (1985), p. 287-337 [3] The authors thank S. Dong at Eastsouth University and Z. Gai in ORNL for useful discussions. Work at ORNL was supported by the Department of Energy, Basic Energy Sciences, Materials Sciences and Engineering Division (MAR, LQ, MC, DM, MFC, DJS, GC) and by the European Research Council Starting Investigator Award "STEMOX 239739" (MAR). Figure 1. High resolution high angle annular dark field planar view image showing the lattice inside and outside of the ZrO2 circular pillars embedded in LSMO.");sQ1[310]=new Array("../7337/0619.pdf","Biofabrication of Dynamic, 3-Dimensional, In vitro Models of Disease.","","619 doi:10.1017/S143192761500389X Paper No. 0310 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Biofabrication of Dynamic, 3-Dimensional, In vitro Models of Disease. Richard L. Goodwin1, 2, Pin-Hsuan Chang2,3, Rebecca S. Jones2, Katrina A. Harmon4, Tzlil Perahia5, Brooks A. Lane2, John F. Eberth2,4, Daping Fan4, Robert L. Price4,6, Jay D. Potts4,6, Harold I. Friedman7 and Michael J. Yost8 University of South Carolina, School of Medicine, Biomedical Sciences Dept., Greenville, SC USA University of South Carolina, Biomedical Engineering Program, Columbia, SC USA 3. University of Michigan, Civil and Environmental Engineering Dept., Ann Arbor, MI USA 4. University of South Carolina, School of Medicine, Dept. of Cell Biology and Anatomy. Columbia, SC USA. 5. University of South Carolina, School of Medicine, Columbia, SC USA 6. University of South Carolina, School of Medicine. Instrument Resource Facility, Columbia, SC USA. 7. University of South Carolina, School of Medicine, Dept. of Surgery. Columbia, SC USA. 8. Medical University of South Carolina, Dept. of Surgery, Charleston, SC USA 2. 1. Maintenance of homeostasis is dependent on numerous cell-cell and cell-ECM interactions that take place in a variety of microenvironments that are subject to a wide range of mechanical forces. A mechanistic understanding of these processes is necessary in order to use this knowledge for new and improved therapies. In general, two approaches, in vivo and in vitro, have been used to gain insight into the mechanisms that regulate pathological processes. In vivo approaches are ideal in most cases; however, experiments in whole organisms are often complex and it is difficult to control for all variables. On the other hand, traditional 2D tissue culture models do not have sufficient complexity to adequately model pathology. Thus, there is an enormous gap between simple 2D cell culture and in vivo models. With this as our motivation, we have created models of myocardial, valve, and coronary artery development using a variety of biofabrication techniques, [1-4]. These models have allowed for the elucidation of specific molecular mechanisms that drive the morphogenesis of these tissues [5,6]. We have also used these fabrication techniques in conjunction with a custom made fluid flow bioreactor to investigate the role of mechanical forces in cardiovascular fibrous development, which is extremely difficult to study in vivo [7,8]. These studies and others involving stem cell integration and differentiation have served to fill the gap between in vitro and in vivo approaches [9]. In the present study, we have developed a new fabrication technology that allows for the copolymerization of collagen gels with mammalian primary cells. By adding this technology to our other fabrication protocols we have generated a dynamic, 3D, in vitro model of atherosclerosis, a common vascular disease in which endothelial, smooth muscle, and inflammatory cells form plaques. Our flexible manufacturing technologies allow for the generation of tubular vascular constructs that contain fibroblasts (cells of the adventitia), smooth muscle (cells of the media), and endothelium (cells of the intima). Using a computational fluid dynamics modeling approach, we designed a tube geometry with a central nozzle that is predicted to produce disturbed flow patterns that are associated with atherosclerotic plaque development. This tube geometry was used to investigate the role that hemodynamics plays in atherogenesis. This specific 3D geometry was then manufactured and characterized for growth and remodeling under static and flow conditions. Figure 1 shows the different cell morphologies, the dynamic cell distribution, and the cell specific remodeling that was observed in the tube constructs over the course of these experiments. Microsc. Microanal. 21 (Suppl 3), 2015 620 In another use of our new biofabrication technology, we have generated a 3D model of the foreign body response, which is a process that limits the form and function of implanted biomaterials [10]. In this model, a 3D printed mold is used in conjunction with our collagen/cell casting process to create a silicone implant that is coated with a collagen/cell polymer. Initial characterization of this model indicates that cells take on different cell morphologies, have a dynamic cell distribution, and deposit new type I collagen over a 14-day culture period (Figure 2). This model will be used to investigate mechanisms of the foreign body response and to screen potential new therapeutic interventions. A) B) C) Figure 1. Cytotube Model of Atherosclerosis. Panel A is a confocal image of cells within the wall of the construct at day 7 (Green=F-actin, Blue=DAPI, Red=smooth muscle alpha actin). Panel B shows the distribution of cells over time. Panel C shows tube contraction over time is dependent on cell type. A) B) C) Figure 2. 3D Model of the Foreign Body Response. Panel A is a confocal image of cells within the wall of the construct at day 7 (Green=F-actin, Blue=DAPI, Red=Type I collagen). Panel B shows the distribution of cells over time. Panel C shows the distribution of Type I collagen within the construct. References: [1] Evans, Heather J et al, AJP-Heart and Circulatory Physiology 285 2 (2003), p. H570. [2] Yost, Michael J et al, Tissue Engineering10 2-Jan (2004), p. 273. [3] Goodwin, Richard L et al, Developmental Dynamics 233 1 (2005), p. 122. [4] Nesbitt, Tresa L et al, In Vitro Cellular & Developmental Biology-Animal 43 (2007), p. 1. [5] Nesbitt, Tresa L et al, Developmental Dynamics 238 2 (2009), p. 423. [6] Norris, Russell A et al, Developmental Dynamics 238 5 (2009), p. 1052. [7] Tan, Hong; Biechler, Stefanie V et al, Developmental Biology 374 (2) (2013), p. 345. [8] Biechler, Stefanie V et al, Front Physiol. ;5:225. doi: 2014 10.3389/fphys.2014.00225. eCollection [9] Valarmathi, Mani T et al, Biomaterials 32 11 (2011), p. 2834. [10] Paul DiEgidio et al, Ann Plast Surg. 73(4) (2014), p. 451.");sQ1[311]=new Array("../7337/0621.pdf","Deposition of Cell-Laden Hydrogels in a Complex Geometry Using a 3D BioPrinter","","621 doi:10.1017/S1431927615003906 Paper No. 0311 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Deposition of Cell-Laden Hydrogels in a Complex Geometry Using a 3D BioPrinter Sarah Grace Dennis1 and Michael J. Yost1 1. Medical University of South Carolina, Department of Surgery, Charleston, USA Solid freeform fabrication (SFF) corresponds to a variety of technologies capable of additive layer-bylayer manufacture of three-dimensional constructs through computer-aided design (CAD) and computeraided manufacturing (CAM) [1]. Combining SFF and tissue engineering techniques resulted in cell bioprinting. In bioprinting techniques; cells and other biological materials are incorporated into carrier materials during the fabrication process and concurrently extruded using a dispensing device [2-3]. An advantage of this technology is the ability to incorporate three-dimensional medical technologies, such as magnetic resonance imaging (MRI) and computer tomography (CT), into the designs resulting in patient-specific scaffolds [2]. The Palmetto Printer was designed to deposit biomimetic tissues for tissue engineering. This study demonstrates the bioprinter is capable of producing cell-laden hydrogel constructs with a complex geometry while maintaining high cell viability. Normal human dermal fibroblasts (NHDF's) and human adipose microvascular endothelial cells (HAMEC's) were cultured in FGM-2 and EGM-2, respectively, until confluent. Alginate, the hydrogel, was dissolved in sterile, distilled water to produce a solution with an alginate concentration of 3%. This solution was then autoclaved for sterilization. NHDF and HAMEC cells were suspended in the alginate at a ratio of 4:1 and a concentration of 1.0 million cells per milliliter alginate [4]. The printing substrate was a 100mM CaCl2 gelatin solution comprising of calcium chloride dehydrate, sodium chloride (0.9 wt%), porcine gelatin (3 wt%), and titanium dioxide (1 wt%) in distilled water. The cell-laden alginate was deposited in a bifurcated ellipse pattern to demonstrate the bioprinter's ability to generate viable structures. The constructs gelled for 10 minutes post-printing. They were then stained, following the kit protocol, with a mammalian viability/cytotoxicity assay by Invitrogen Life Technologies. The stained constructs were imaged one hour post-printing with a Leica TCS SP5 AOBS Confocal Microscope System using Z-stack parameters of X over Y depth. Cells that appear green or yellow are counted as live. Red cells are counted as dead. Cell viability was calculated using the # of live cells following equation: Viability = # total cells � 100%. Cell viability for the bioprinted ellipses was quantified by imaging five different sections, counting the live and dead cells. Cell viability was 98.1%. Additive manufacturing is the logical next step in tissue engineering and design. For tissue engineering, the unique features of the ability to precisely place nanoliter size aliquots of cell containing hydrogel as well as the ability to recapitulate a more natural process will move the field forward substantially. The Palmetto Printer is a step along the way of creating workable solid free form fabrication of living tissues as has been demonstrated here. 1. Ferris, C.J., Gilmore, K.G., Wallace, G.G., & Panhuis, M. Biofabrication: An Overview of the Approaches Used for Printing of Living Cells. Appl. Microbiol. Biotechnol. 97 (10), 4243-4258, doi: 10.1007/s00253-013-4853-6 (2013). 2. Khalil, S., Nam, J., and Sun, W. Multi-Nozzle Deposition for Construction of 3D Biopolymer Tissue Scaffolds. Rapid Prototyping Journal. 11 (1), 9-17, doi: 10.1108/13552540510573347 (2005). Microsc. Microanal. 21 (Suppl 3), 2015 622 3. Blaeser, A., Duarte Campes, D.F., Weber, M., Neuss, S., Theek, B., Fischer, H., and JahnenDechent, W. Biofabrication Under Fluorocarbon: A Novel Freeform Fabrication Technique to Generate High Aspect Ratio Tissue-Engineered Constructs. BioResearch. 2 (5), 374-384, doi: 10.1089/biores.2013.0031 (2013). 4. Czjaka, C.A., Mehesz, A.N., Trusk, T.C., Yost, M.J., and Drake, C.J. Scaffold-Free Tissue Engineering: Organization of the Tissue Cytoskeleton and Its Effects on Tissue Shape. Annals of Biomedical Engineering. 42 (5), 1049-1061, doi: 10.1007/s10439-014-0986-8 (2014). 5. The authors acknowledge funding from the National Science foundation EPS-0903795 and the National Institutes of Health NIDCR R01- DE019355 (MJY PI). Figure 1: Palmetto Printer. A Programmable Logic Controller (PLC) coordinates the actions of all printer functions (A). The bioprinter is enclosed in an airtight containment chamber (B) with filtered intake and exhaust (C) that maintains a regulated internal positive pressure to reduce risk of contamination. There are dual UV lights (D) that are mounted at the top of the chamber and are controlled by the PLC. The Janome 2300N XYZ Robot (E) is programmed and controlled by an integrated computer (J). There is a distance laser (I) mounted on the robotic arm's Z-axis dispenser holder. There are three independent dispenser controllers (F) that are programmed manually to regulate the output of the dispenser guns available to be loaded onto the Z axis arm under computer control. Temperature of the substrate (printing surface) is maintained by a water bath controller (K) between 4 and 40. Dual digital cameras (L) are available to monitor printer activity and sample formation. Figure 2: Bioprinted Construct. Image of bioprinted constructs stained with mammalian live/dead fluorescence. Live cells appear green and dead cells appear red taken with Leica TCS SP5 AOBS Confocal Microscope. Inset, photomicrograph of printed ellipse.");sQ1[312]=new Array("../7337/0623.pdf","3D Printable In-Situ Fluorescent Microscope","","623 doi:10.1017/S1431927615003918 Paper No. 0312 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 3D Printable In-Situ Fluorescent Microscope Hany Osman1 1. Indiana University, Department of Pathology and Laboratory Medicine, Indianapolis, Indiana, USA In-vivo and in-situ fluorescent microscopy are commonly used techniques in research laboratories for direct micro-visualization of excised tissue or tissues within live animals [1]. The technology is slowly transitioning to clinical applications, however it remains at a prohibitively high cost [2]. The clinical utility includes in-vivo applications in the gastrointestinal tract and other organ systems, however the technology has greater potential for use in fields such as pathology, dermatology and others. In the current paper we propose the application of 3D design and printing as a cheap method for providing the framework of a fiber optic fluorescent microscope. 3D printing allows the development of highly customizable and unique structural and mechanical parts at extremely low costs. It also allows for the reproduction and modification of the microscope for various experiments and applications using off-the-shelf components. We used the XYZ, da Vinci 1.0 3D printer for printing and Blender 3D modelling software for designing the parts. We also used a Cannon T2i digital single lens reflection (DSLR) camera for the image acquisition, a 20x objective, dichroic mirror, excitation and emission filters and tube lens as shown in Figure 2. The filters and light emitting diode (LED) were selected for utility with acridine orange fluorescent dye, which was applied to the tissue to be viewed. An imaging fiber optic probe was attached to the objective using 3D printed parts for in-vivo or in-situ visualization. The other end was placed on the specimen as shown in Figure 2. Examples of the resulting images captured by the system are shown in Figure 1. The 3D printed components include a filter cube, which holds the dichroic mirror, objective and fiber optic holder that attaches to the filter cube, a camera tube and the light source housing. The 3D printed light source collimator is designed to accommodate the LED and heat sink and also holds an excitation filter and a condenser lens. A condenser lens is placed at a distance according to its focal length for partial collimation of the light from the LED light source to the base of the objective lens. The camera tube is designed with a base adaptor that inserts into the DSLR camera and has slots that hold the emission filter and tube lens. The camera tube and filter cube are printed in black plastic to minimize external light noise. References: [1] Inoue H et al, Endoscopy 32 (6) Endoscopy. 2000 Jun;32(6):439-43 [2] Shin D et al, PLoS One. 2010 Jun 23;5(6):e11218. doi: 10.1371/journal.pone.0011218. Microsc. Microanal. 21 (Suppl 3), 2015 624 Figure 1. Colonic crypts seen using the 3D printed fluorescent microscope after application of acridine orange (left), corresponding hematoxylin and eosin processed slide (right) Fiber optic to specimen Figure 2. Schematic illustrating light path for fluorescent microscope (left), 3D rendered microscope showing printed parts (right)");sQ1[313]=new Array("../7337/0625.pdf","Nanoscale Materials Characterization in Industry: Aerosols Produced from Abraded Composite Restoratives","","625 doi:10.1017/S143192761500392X Paper No. 0313 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Nanoscale Materials Characterization in Industry: Aerosols Produced from Abraded Composite Restoratives Mary I. Buckett1, Axel Bogdan2 and Daniel A. Japuntich3 1 2 3M Company, Corporate Research Analytical Laboratory, St. Paul Minnesota 55144 3M Company, Industrial Adhesives and Tape Division, St. Paul Minnesota 55144 3 3M Company, Safety, Security, and Protective Services Laboratory, St. Paul Minnesota 55144 Nanotechnology has impacted a number of core technology platforms in the manufacturing arena, and has been leveraged to produce game-changing new product opportunities. Commercial manufacturers have introduced engineered nanoparticles (ENPs) into their filler systems to enhance a variety of product properties. For example, 3M has leveraged the high value nanocomposites bring to the dental industry. Nanocomposites exhibit excellent wear characteristics combined with the high esthetics people expect, thus providing long-lasting, good looking restorations. True nanotechnology in dental composites, i.e. the bottom-up design of nano-size filler particles in the size range 1-100 nm, is being used quite successfully in dental restorative products.(1) Commensurate with nanotechnology implementation, it is important � but difficult - to study whether free engineered nanoparticles (ENPs) can be released from a composite product, for example in the form of airborne dust during mechanical manipulations like drilling or grinding. We've addressed this problem with a methodical approach for generating, collecting, and analyzing nano-size aerosol particles from abraded dental composite materials.(2) Standard aerosol sampling instruments were combined with a custom-made sampling chamber to create conditions of a dental drilling environment. In this experiment, we sampled fresh, steady-state aerosol produced from drilling a dental composite and collected data on aerosol size distribution before significant Brownian coagulation could occur. The aerosol particles were counted and sorted by electrostatic classification into discrete size ranges. Aerosol samples in the 7 nm, 20 nm and 75 nm range were routed directly onto TEM grids for morphology, size, and compositional analysis by Transmission Electron Microscopy (TEM). Our test results show that the nano-size aerosol particles produced during the abrasion of dental composite fall into three main categories (in order of prevalence): 1) oil; 2) graphitic carbon or carbon ash; and 3) composite material (ENP + binder). Results from the 3 size categories are summarized in Table 1. During abrasion, aerosol generation seemed independent of the percent filler load of the restorative material and the operator who generated the test aerosol. TEM analysis confirms that `chunks' of filler and resin are generated in the nano-size range (see Figure 1); however free ENPs are not observed in any of the size ranges we investigated. References: [1]. Mitra, S.B., Wu, D., Holmes, B.N.: An Application of Nanotechnology in Advanced Dental Materials, J Am Dent Assoc 134:1382-1390 (2003) [2] Bogdan, A., Buckett, M., Japuntich, D.: Nano-Sized Aerosol Classification, Collection and Analysis - Method Development Using Dental Composite Materials, J Occup and Environ Hygiene DOI: 10.1080/15459624.2013.875183 (2014) Microsc. Microanal. 21 (Suppl 3), 2015 626 TABLE I. TEM Analysis Summary of aerosol particulates observed when grinding dental composite materials in a controlled environment. Size Range 7 nm 20 nm 75 nm Amorphous Carbon Ash No Yes Yes Graphitic Carbon No Yes Yes Dental Composite (Filler attached to Resin) No No Yes Free Dental Filler Particles No No No Oil Yes Yes Yes Figure 1. STEM image at right and x-ray elemental map data showing the composition of the bulk agglomerated dental filler composite used in a commercial dental restorative product.");sQ1[314]=new Array("../7337/0627.pdf","Nanoparticle Safety in the Workplace: Developing a Methodology to Monitor and Remediate Spills.","","627 doi:10.1017/S1431927615003931 Paper No. 0314 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Nanoparticle Safety in the Workplace: Developing a Methodology to Monitor and Remediate Spills. Stefano Rubino1, Brad Lute1, Iris Guo1, Kelly Cadieux1, Byron D. Gates1. 1. Department of Chemistry and 4D LABS, Simon Fraser University, Burnaby, BC, Canada. Recent years have seen a rapidly increasing presence of nanoparticles not only in research applications, but also in industries such as medical and manufacturing. Since nanoparticles often exhibit properties unlike those of their bulk counterparts, they should be treated as unknown substances with potentially dangerous collateral effects until long-term assessments are completed. The field of nanotoxicity has recently received much attention from governments and public agencies, with the establishment of national and international infrastructures targeted at studies on safe use of nanomaterials (for example, the EU funded QualityNano research infrastructure). We propose here a methodology to monitor and remediate spills of nanoparticles in the workplace, with the aim to develop a policy on safe usage and handling of nanomaterials in general. Ideally, the methods developed should be easy to apply, inexpensive and non-destructive to the workplace. Nanoparticle spills are often invisible to the naked eye, but several techniques can be used to detect them on a work surface. In this study, we considered a common laboratory countertop contaminated by spills of solutions containing nanoparticles in different solvents. A complete characterization of the spill would require tools with elemental resolution in the nanometer scale. However, we assume that the contaminant nanoparticle characteristics are already well known to a given workplace, and instead focus on features that allow us to easily distinguish them from the countertop itself or common macroscale contaminants. For example, such features of contaminant nanoparticles could be element(s) not present in the workplace, specific optical properties or a unique shape. We have chosen as representative examples Au nanorods, Se nanowires and FePt nanoparticles. We selected X-Ray Fluorescence (XRF) as an optimal tool to detect spills based on previous results obtained in our research group [1,2]. The detection limits for XRF are established by direct comparison with other investigative analytical tools, such as Inductively Coupled Plasma Mass Spectrometry (ICP-MS), light absorption spectroscopy (Fig. 1), and Scanning and Transmission Electron Microscopy (SEM/TEM) (Fig. 2) coupled with Energy Dispersive X-ray Spectroscopy (EDXS). Once a spill is detected, various remediation strategies can be applied. It was shown that wiping of the spill is ineffective, but also spreads the nanoparticles over a larger area [1,2]. More effective strategies involve the casting of a polymer or rubber on the spill to encapsulate the nanoparticles. Depending on the particles, remediating agent(s), and surface interactions, a varying number of applications of the remediating agent are necessary to remove a significant number of nanoparticles from the contaminated surfaces. The efficacy of the remediation is monitored using the same techniques mentioned above. The method is tested for a wide number of parameters to ensure general applicability in conditions commonly encountered in the workplace. Such parameters include: i) nanoparticle composition, shape and their surface chemistry; ii) solvent composition and polarity; iii) remediation agents; and iv) use of surfactants or other additives prior to applying the remediation techniques. Spills containing more than Microsc. Microanal. 21 (Suppl 3), 2015 628 one type of nanoparticle are investigated as well, to determine the effect of particle interaction on detection and remediation. References: [1] J Zhou, M.Sc. Thesis, Simon Fraser University (2015). [2] BD Gates et al, "Detecting, Handling and Controlling Nanoparticle Contamination in the Workplace, Final Report for WorkSafeBC and Workers' Compensation Board of Nova Scotia", Project number: RS2010-IG40, (2013). [3] We thank Brandy K. Pilapil and Michael Wang for providing FePt and Se nanomaterials used in this work. This research was supported in part by funds from WorkSafeBC through the Innovation at Work program, the Natural Sciences and Engineering Research Council (NSERC) of Canada, the Canada Research Chairs Program (B.D. Gates), and CMC Microsystems through the MNT Financial Assistance program that facilitated access to materials characterization services. This work made use of 4D LABS shared facilities supported by the Canada Foundation for Innovation (CFI), British Columbia Knowledge Development Fund (BCKDF), Western Economic Diversification Canada, and Simon Fraser University. Figure 1. Left) Light absorption spectra of the Au nanoparticle solutions prepared by serial dilution, showing the plasmonic peaks characteristic of Au nanorods. Right) Plot of the integrated intensity of the absorption spectra in the region [680-860] nm around the peak centered at 780 nm as function of the relative nanoparticle concentration. Figure 2. TEM micrographs of Au nanorods (left) and Se nanowires (middle) dropcast on a TEM grid. SEM image of a simulated spill containing both types of nanomaterials (right). The shape and aspect ratio of these nanoparticles makes them easily distinguishable from other sources of contamination.");sQ1[315]=new Array("../7337/0629.pdf","In-Situ Liquid Cell TEM Observations of Silicate Mineral Dissolution","","629 doi:10.1017/S1431927615003943 Paper No. 0315 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In-Situ Liquid Cell TEM Observations of Silicate Mineral Dissolution Donovan N. Leonard1, Raymond R. Unocic2, and Roland Hellmann3,4 1. 2. Materials Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, TN 3. Universit� Grenoble Alpes, ISTerre, F-38041 Grenoble cedex 9, France. 4. CNRS, ISTerre, F-38041 Grenoble, France. The chemical weathering of silicate minerals controls in large part the chemical composition of natural waters, such as lakes, rivers, and oceans. Weathering reactions are an important sink for CO2, and play a critical role in the global carbon cycle, which in turn has implications for climate change. Chemical weathering is also crucial to current environmental issues such as acid mine drainage, and geological CO2 storage and sequestration schemes. Understanding the mechanisms of mineral dissolution at the nanoscale is therefore a prerogative for predicting and modeling chemical weathering at multiple scales, ranging from the microscale up to the global scale. Even though silicate mineral weathering has been intensely studied both experimentally and in natural settings for many decades now, the exact mechanisms that control dissolution are still being debated. During dissolution in aqueous solutions, multi-cation silicate minerals develop a secondary altered surface phase, which can be either crystalline or amorphous. Understanding how this surface layer develops is a key to understanding the mechanism of dissolution [1-2]. Until recently, post-altered mineral (and glass) surfaces have been examined using conventional TEM and other high-resolution techniques, such as atom probe tomography. However, to better understand the operative mechanism(s), the mineral dissolution process should ideally be examined in situ in their native fluid environment. To address this issue, we are currently employing nanoscale measurements via in situ liquid cell TEM to examine the in situ dissolution of a common chain silicate mineral, wollastonite (CaSiO3). Preliminary results show that surfaces retreat on the nmscale and this dissolution can be measured in real time. Focused ion beam (FIB) milling was used to prepare an electron transparent wollastonite lamella measuring 25�m x 4 �m x 100nm thick for the fluid cell experiments. A Protochips Poseidon holder (for use in a Hitachi HF3300 TEM/STEM) with a liquid layer thickness of 5 �m was used. To reduce Ga implantation and amorphous material on the surface of the FIB lamella, a Fischione Nanomill was operated at 900eV to Ar+ mill each side of the sample at 10� for 10min. The FIB sample was then micromanipulated such that the region of interest (ROI) was placed over the 40nm thick SiN membrane of the liquid cell platform and attached to the edge of the Si chip using e-beam deposited C. TEM observations were performed at 300kV and electron dose was minimized by keeping the gun valve closed except when capturing TEM images. The deionized water (DI) flow-rate through the fluid cell was set at rate of 10�L/min. Figures 1a-dshow the FIB fabrication process for the silicate lamella, which was first thinned to electron transparency and then attached to the fluid cell window. Figure 2a shows the bright-field (BF) TEM image of the wollastonite lamella before DI water was introduced into the fluid cell. Note that a crack in the lamella (arrow) occurred during the placement of the top window on the fluid cell. The TEM images shown in Figures 2b-c show evidence for dissolution of the wollastonite localized at the crack after DI water was introduced into the fluid cell. The TEM images and corresponding image intensity profiles across the crack (see inset rectangles in Figure 2) were used to quantify the progression of the Microsc. Microanal. 21 (Suppl 3), 2015 630 dissolution front. The most reactive area of the sample appeared to be the crack faces, and the corresponding change in BF image contrast indicated that the thickness of the lamella changed in these regions after 76 minutes. Future work will concentrate on supplementing the structural retreat rates with nanoscale chemical data from EFTEM and EELS analyses [3]. [1] R Hellmann et al, Phys. Chem. Miner. 30 (2003) p. 192-197. [2] R Hellmann et al, Nature Materials DOI:10.1038/NMAT4172 (2015). [3] Research supported by Oak Ridge National Laboratory's Center for Nanophase Materials Sciences (CNMS), which is a U.S. Department of Energy, Office of Science User Facility. Figure 1. (a) FIB lamella of wollastonite. (b) Thinned window stuck to sample, which was used as fiducial. (c) Lamella after Ar ion milling attached to micromanipulator needle. (d) Wollastonite lamella attached to edge of Si substrate of liquid cell platform with thinned region of interest over SiN window. Figure 2. BF TEM images showing (a) silicate lamella with no fluid in cell, with arrow indicating crack. Crack width measures 61nm. (b) Liquid cell filled with DI water and dissolution observed at crack faces. Crack width measures 95 nm. (c) Imaged 76 minutes after liquid introduced into cell with the crack faces still undergoing dissolution. Crack width measures 136nm. 5�m 5�m 5�m 5�m 2�m 1�m 1�m");sQ1[316]=new Array("../7337/0631.pdf","3D FIB SEM imaging of oil filled chalk: What are the challenges?","","631 doi:10.1017/S1431927615003955 Paper No. 0316 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 3D FIB SEM imaging of oil filled chalk: What are the challenges? K. N. Dalby1, R. P. Harti1, H. O. S�rensen1 and S. L. S. Stipp1 1. NanoGeoScience Group, Department of Chemistry, Nano-Science Center, University of Copenhagen, 2100, Denmark In Europe, fresh groundwater and oil resources are hosted in chalk, which is a sedimentary rock, dominated by calcite (CaCO3) precipitated by marine organisms. The calcite particles have remained small during millions of years of compaction and weathering, creating a complex microstructure with low permeability in spite of often high porosity. Knowledge about pore architecture is required to simulate fluid flow, which can be used to predict how contaminants would behave in groundwater aquifers, or how easily oil could be extracted. Much of the characterisation of chalk porosity was done before the current explosion in high-resolution techniques, resulting in fluid flow models based on information gathered at the centimeter to micrometer scale. However, it is now possible to examine complex porous systems at nanometer resolution [1] and to use the data to improve predictions of petrophysical parameters [2]. X-ray tomography and ptychographic X-ray tomography are pushing resolution boundaries but the focused ion beam scanning electron microscope (FIB SEM) provides some of the highest resolution for imaging natural materials in 3D [3]. A disadvantage of FIB SEM however, is that one cannot image liquid phases because of the need for high vacuum in the analysis chamber. One option is to use a cryogenic stage and freeze the liquids [4]; another is to solidify the oil in place using osmium tetroxide [5], which is the method we explored. An advantage of oil solidification is that freezing is not required and the atomic number contrast is enhanced because Os diffuses into the liquid phase during the stabilisation process (Figure 1). To investigate the feasibility of using the Os fixation of the oil in chalk investigations, we took three fragments from the same �lborg chalk. We imaged one dry, another partially filled with oil and another that was fully filled. We used a dual beam FIB SEM and collected information as we milled through the sample, to produce 3D images. Figure 1 illustrates that there are several factors to consider. The first is the heterogeneous nature of the chalk. The completely dry and partially filled images show similar areas of chalk where calcite dominates but the completely oil filled area is dominated by pores. This is the common "representative volume" issue, where input from lower resolution methods is necessary. The second factor is the milling and imaging environment. The effect of beam damage on pore architecture is an important aspect. The choice of detector can help reduce complications in the subsequent image processing. The third factor is that the osmium could modify the pore surface composition, thus affecting the wetting behaviour and oil distribution [6] and the interpretation of pore geometry. By taking these three factors into account, we can begin to visualize pore scale fluid displacement processes, down to nanometer scale. [7] Microsc. Microanal. 21 (Suppl 3), 2015 632 [1] K N Dalby et al, Microsc. Microanal. (2014) p. 316. [2] D M�ter et al, Applied Physics Letters 105 (2014), p. 043108. [3] T L Burnett et al., Scientific Reports 4 (2014), p. 4711 [4] G Desbois and J L Urai., EMC Vol 2: Materials Science (2008), p807. [5] R P. Harti, University of Copenhagen MSc thesis (2015) [6] J Gosh and G R Tick, Journal of Contaminant Hydrology 155, p. 20 [7] The authors acknowledge funding from P3 Project, funded by Maersk Oil and Gas A/S and the Danish National Advanced Technology Foundation and by the Danish Council for Independent Research (via DANSCATT). B C A 5 �m Figure 1. Heterogeneous nature of chalk on the micrometer scale. Back scattered electron images of dry chalk (A), pores partially filled with oil (B) and completely filled (C). Scale is 5 micrometers and images are taken at the same beam settings (10 kV and 8 nA). In all three images, air filled pores are black and calcite is dark grey. EDXS mapping of these sections shows that in the dry sample (A), the lighter contrast material is redeposited calcite rich in Ga. In the partially filled sample (B), the lighter contrast comes from the Os used to stabilise the oil. In the completely filled sample (C), the lighter contrast is provided by a thick Os film that smears with electron beam and ion beam interference.");sQ1[317]=new Array("../7337/0633.pdf","Cryo-Transmission Electron Microscopy of Sea Spray Aerosols","","633 doi:10.1017/S1431927615003967 Paper No. 0317 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Cryo-Transmission Electron Microscopy of Sea Spray Aerosols Joseph P. Patterson,1 Douglas B. Collins,1 Jennifer M. Michaud,1 Jessica L. Axson,1 Michael D. Burkart1 Kimberly A. Prather,1,2 and Nathan C. Gianneschi,1* 1. 2. Chemistry and Biochemistry, University of California, San Diego, La Jolla, California 92093-0356 Scripps Institution of Oceanography, University of California, San Diego, La Jolla, California 92093 We report the development of the first gas phase cryo-TEM sampling experiments (aerosol-cryo-TEM) allow the detection of hydrated SSA particles by electron microscopy. This unique approach shows the first observation of whole bacteria, phytoplankton, virus particles and marine vesicles within hydrated droplets. Combined with graphene oxide (GOx) supported cryo-TEM of sea surface microlayer (SSML) and bulk seawater, this study reveals the exquisite structural complexity of biogenic marine nanostructures in the ocean and marine atmophere. Bacterial marine vesicles are known to be abundant in coastal seawater and the open ocean. The high local concentrations of carbon in the form of DNA, RNA etc. make them a significant source of carbon and information. This study shows that whole intact bacteria and membrane vesicles exist within sea spray aerosol water droplets. While bacteria remain intact, structural differences within membrane vesicles in the SSML and aerosols indicates that the vesicles are formed during the bubble bursting/aerosol production process and therefore exist as a unique structure in the marine environment. Therefore, bubble bursting leading to the ejection of marine vesicles represents a significant mechanism for the transfer of biological nanostructures from the ocean, into the atmosphere and around the planet. Specifically, the observation of the vesicles in the 50-200 nm size range provides direct evidence for a new mechanism for organic carbon (OC) enrichment in SSA particles, something which has been widely debated in recent literature and previously only ascribed to marine nanogels. This fundamentally changes our knowledge SSA production pathways and their role in cloud formation. While marine vesicles have never been detected in SSA before, lipid-like molecules and vesicles in aerosols have been theorized to play an important role in the evolution of pre-biotic life. This study shows that it is indeed possible for vesicles to be formed and exist in discrete aerosol particles supporing this theory. Thus, this could represent an alternative method for transferring biological species from the ocean and ultimately transported around the planet. We propose this as an alternative to the theorized inverse micelle model for lipids in aerosols. This has implications for how nanostructures comprised of prebiotic species could have resulted in the first single cells. The ability to observe such nanostructures, as well as their formation pathway at the interface of the ocean, represents a major advance in our understanding of factors influencing the transfer of carbon from the ocean to the atmosphere. We anticipate this study will open a new avenue of analysis for aerosol particles, not only for ocean-derived aerosols, but for those produced over land, in the laboratory, or from other aqueous sources where there is interest in the transfer of organic species. References [1] F Azam, F Malfatti, Nature Reviews Microbiology, 5(10), 2007, p. 782. [2] SJ Biller et. al, Science, 343, 2014, p. 183. Microsc. Microanal. 21 (Suppl 3), 2015 634 [3] JP Patterson et. al. Soft Matter, 8, 2012, p. 3322. [3] Stokes MD, et al. Atmospheric Measurement Technology, 6, 2013, p. 1085. [4] Prather KA, et al. Proceedings of the National Academy of Sciences, 110, 2013, p. 7550. Figure 1. Schematic for the preparation of Aerosol-Cryo-TEM grids. Graphene oxide TEM grids are loaded into a humid environment (loading phase). Wet SSA dropets can then passively adhere to the grids (adherence phase). The droplets are then vitrified by rapidly plunging into liquid ethane (vitrification phase). Figure 2. Cryo-TEM images of the marine environment, bulk ocean (blue), sea surface microlayer (green) and sea spray aerosols (orange).");sQ1[318]=new Array("../7337/0635.pdf","Towards Atomic Level Understanding of Transition Alumina Phases and Their Phase Transformations","","635 doi:10.1017/S1431927615003979 Paper No. 0318 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Towards Atomic Level Understanding of Transition Alumina Phases and Their Phase Transformations L. Kovarik1, M. Bowden1, D. Shi2, A. Anderson1, Jianzhi Hu2, J. Szanyi2, J. H. Kwak3, C.H.F. Peden2 Environmental Molecular Sciences Laboratory, Pacific Northwest National Laboratory, P.O. Box 999, Richland, Washington USA 2 Institute for Integrated Catalysis, Pacific Northwest National Laboratory, P.O. Box 999, Richland, Washington USA 3 School of Nanobiotechnology & Chemical Engineering, UNIST, Ulsan 689-798, Korea Transition Al2O3 derived from dehydration of Al hydroxides are highly complex materials with a significant degree of inherent structural disorder. The way in which the disorder is manifested for various heat treatment conditions and for various hydroxide precursors is a highly relevant topic in catalysis, with important implications for rationalization of unique surface chemistry and catalytic behavior of these materials [1]. When heat-treated at relatively high temperatures (>900�C), the complexity of the microstructure is generally associated with the polymorphs of -Al2O3 and -Al2O3, which form in closely inter-growing structures. This has been a main issue in their reliable characterization, and there are currently none or only poorly fitting crystallographic models available. Similarly, the stability of these polymorphs remains poorly understood and actively studied [2]. In this work we address the structural nature of -Al2O3 and -Al2O3, and the mode of their phase transformations using combination of in-situ and ex-situ imaging and spectroscopy techniques. The current work mainly relies on the use of HAADF Scanning Transmission Electron Microscopy imaging, XRD, high-resolution NMR and DFT calculations. The HAADF STEM observations were performed with a probe corrected FEI Titan 80-300, and the thermal treatment was performed under in-situ heating conditions inside the TEM with Aduro Protochips heating holder at 900-1100 �C. Figure 1 shows structural and morphological changes that occur during heat-treatment of transition Al2O3 particles. The in-situ observations enabled us to characterize the evolution of microstructure across several length scales, ranging from macro-scale porosity down to crystallographic level. At the crystallographic level, it is found that the defective nature of the microstructure is due to intergrowth from number structural variants that evolve concurrently during the thermal treatment. The structure of -Al2O3 itself is fund to be a highly complex, lacking a long-range periodicity and requiring a rationalization in terms of several closely related crystallographic variants. The two main variants are identified as 1-Al2O3 and 2-Al2O3. Full crystallographic description was obtained on the basis of quantitative analysis of HAADF STEM images from a number of low indexed zones, as shown in Fig.1(f,g) and XRD analysis. Analogous characterization effort has been performed for -Al2O3. The validity of derived structures was evaluated with NMR measurements and DFT based modeling. It will be shown how energetic degeneracy among the various transition alumina leads to complex intergrowths. In Fig.1(h) we compare the stability of the newly derived 1-Al2O3 and 2-Al2O3 with -Al2O3, -Al2O3, and the thermodynamically stable -Al2O3. In the final part of this talk, we describe the evolution of transition Al2O3 during thermal treatment, and discuss the mode of their phase transformations [4]. 1 Microsc. Microanal. 21 (Suppl 3), 2015 636 References [1] G. Busca, Catalysis Today, (2013), p.1�12. [2] I. Levin & D. Brandon, Journal of the American Ceramic Society, 81(8), (1998), p.1995� 2012. [3] Kovarik, L., Bowden, M., Genc, A., Szanyi, J., Peden, C. H. F., & Kwak, J. H. (2014). The Journal of Physical Chemistry C, 118(31), (2014), p.18051�18058. [4] This research is part of the Chemical Imaging Initiative at Pacific Northwest National Laboratory. The work was conducted in the William R. Wiley Environmental Molecular Sciences Laboratory (EMSL), a national scientific user facility sponsored by DOE's Office of Biological and Environmental Research and located at PNNL. Figure 1. (a,b,c) HAADF observation of microstructural evolution of transition Al2O3 during insitu heat treatment. (d) HAADF image depicting the domains of -Al2O3 and -Al2O3 in a common intergrowth (2hr@900�C), (e,f) Atomic level depiction of 1-Al2O3 along [001] and [100] crystallographic directions. (d) Schematic representation of 1-Al2O3. (e) Comparison of enthalpies of formation for the newly derived -Al2O3 and other relevant Al2O3 [3].");sQ1[319]=new Array("../7337/0637.pdf","In situ Studies of the Reaction-Driven Restructuring of Ni-Co Core-Shell Nanoparticles","","637 doi:10.1017/S1431927615003980 Paper No. 0319 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In situ Studies of the Reaction-Driven Restructuring of Ni-Co Core-Shell Nanoparticles Cecile S. Bonifacio1, Huolin L. Xin2, Sophie Carenco3, Miquel Salmeron3, Eric A. Stach2, and Judith C. Yang1 1. 2. Department of Chemical and Petroleum Engineering, University of Pittsburgh, Pittsburgh, PA 15260 Center for Functional Nanomaterials, Brookhaven National Laboratory, Upton, NY 11973 3. Materials Science Division, Lawrence Berkeley National Laboratory, Berkeley, CA 94720 Core-shell bimetallic nanoparticles (NPs) have novel catalytic, optical and electronic properties compared to their monometallic equivalents [1]. These catalytic properties can be controlled by finetuning chemical composition, surface oxidation, structure and dimension [2,3]. In fact, segregation in the core-shell structure has been demonstrated as a potential route of tuning the NPs catalytic properties through in situ gas reaction studies [4]. However, experimental evidence of elemental distributions during reactions is necessary to confirm this hypothesis. Here we have used in situ imaging and spectroscopy techniques provide direct evidence of the structural and elemental changes which occur in these NPs during reactions. Oxidation and reduction at 0.3 Torr in O2 and H2 gas at 220�C and 270�C, respectively, of the core-shell Ni-Co nanoparticles (NPs) were performed in two cycles in using both an environmental transmission electron microscope (ETEM) and an ambient pressure x-ray photoelectron spectrometer (AP-XPS). These NPs (diameter of 25 nm) were dispersed in hexane and drop-casted on SiN and Si grids for the in situ studies. An aberration-corrected FEI Titan ETEM microscope with a Gatan parallel electron energy loss spectrometer (EELS) operated at 300KV was used for simultaneous imaging, selected area electron diffraction (SAED) pattern and EELS acquisition. SAED patterns and EELS maps provided the structural and elemental distributions of Ni and Co within the particle for quantitative analysis. Figure 1 shows the change in the surface structure of core-shell nanoparticles during the two cycles of oxidation (I-O2 and III-O2) and reduction (II-H2 and IV-H2) in the ETEM [5]. Small clusters (marked with arrows in Figure 1) formed after the 1st reduction cycle and remained on the surface until the end of the experiment. These clusters were identified as cobalt oxide based on the acquired EELS maps (Figure 2a-b). Such results correlate with the increase in Co peak area observed during the 1st cycle of reduction using the surface-sensitive technique of XPS[6] (Figure 2c). Further oxidation and reduction of the particles, labeled III-O2 and IV-H2, respectively, led to detection of Ni species at the surface of the particles (via EELS, Figure 3a and b) and which is correlated to an increase in Ni XPS peak area (Figure 3c). These results clearly show a reaction-driven restructuring of the core-shell NPs by Ni segregation during the 2nd reduction cycle. Quantitative analysis of the EELS results is underway to identify the valence states during the oxidationreduction reactions. Correlation of the core-shell reconstruction with the electronic structure changes from ETEM and AP-XPS will be provided to obtain the optimum reaction conditions, i.e., catalytic properties, of the Ni-Co core-shell NPs in challenging reactions such as selective CO2 reduction [7]. References: [1] S. Alayoglu et al, Nature Materials 7 (2008), p. 333 [2] R. Tiruvalam et al, Faraday Discussions 152 (2011), p. 63 [3] S. Mandal, KM Krishnan, Journal of Materials Chemistry 17 (2007), p. 372 Microsc. Microanal. 21 (Suppl 3), 2015 638 [4] F. Tao et al, Science 322 (2008) p. 932-934 [5] CS Bonifacio et al, Manuscript in preparation [6] S. Carenco et al, Small (2015), In press. [7] CB and JY acknowledge financial support by DOE Basic Energy Sciences (DOE-BES) and the Center for Functional Nanomaterials (CFN) at Brookhaven National Lab supported by the Office of Basic Energy Sciences of the US Department of Energy Contract No. DE-SC0012704. SC was supported by the Director, Office of Science, Office of Basic Energy Sciences, Chemical Sciences, Geosciences, and Biosciences Division, under the Department of Energy Contract No. DE-AC0205CH11231. Figure 1. High angle annular dark field (HAADF) images of the core-shell NPs during the oxidation (IO2 and III-O2) and reduction (II-H2 andIV-H2) cycles. Formation of small clusters (labelled with arrows) on the surface was observed after II-H2 reduction cycle. Figure 2. HAADF (a) shows small clusters outside the shell during I-H2 which were identified as Co based on the analysis of the EELS map (b). Clusters of red pixels on the NP surface in (b) verified the Co cluster formation. The increase in Co signal in the XPS 1st reduction cycle (Red1) (c) correlated with the HAADF and EELS results. Figure 3. EELS map (a) during II-H2 showing green pixels, Ni species, on the surface of the NP. EELS Co and Ni L2,3 edges (b) were extracted from the marked area in (a) showing significant Ni signals. Such results correlate to the increase in XPS relative area from Ni (c) indicating Ni segregation to the surface.");sQ1[320]=new Array("../7337/0639.pdf","Nano-level Structure-Reactivity Relationships of Ni-NiO Core-shell Co-catalysts on Ta2O5 for Solar Hydrogen Production","","639 doi:10.1017/S1431927615003992 Paper No. 0320 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Nano-level Structure-Reactivity Relationships of Ni-NiO Core-shell Co-catalysts on Ta2O5 for Solar Hydrogen Production Qianlang Liu1, Liuxian Zhang1 and Peter Crozier1 1. School for the Engineering of Matter, Transport and Energy, Arizona State University, Tempe Arizona 85287-6106 Tantalum oxide and many tantalate-based systems have been reported to show extraordinarily high activities and quantum yields when decomposing water under ultraviolet (UV) illumination [1]. Although pure tantalum oxide shows some photocatalytic activity, loading with a nickel-based cocatalyst improves the initial H2 production rate by 3 orders of magnitude and results in stoichiometric decomposition of pure water into H2 and O2 [2]. Interestingly, this co-catalyst needs to have a particular microstructure, Ni core-NiO shell, to show high activity. However a detailed atomic-level understanding of the relationship between the catalyst microstructures and the photocatalytic reactivities has not yet been fully explored. Aberration-corrected TEM provides an efficient way to observe fine structures at this level and here we investigate the structure-reactivity relations in different co-catalysts to study both reaction and deactivation mechanisms [3]. A series of controlled Ni-NiO core-shell co-catalyst structures were prepared on a Ta2O5 substrate by tuning the heat treatment conditions. The photocatalytic activity of each structure was tested in pH 7 DI water in a photoreactor system interfaced to a gas chromatography (GC) for H2 detection. An aberrationcorrected FEI Titan (300 kV) was employed to study the fine difference in the initial co-catalyst structures, as well as the evolution of the materials after exposure to UV light and water. Each initial co-catalyst sample was examined and HRTEM images were obtained. The samples varied in particle size and NiO shell thickness. Typical core-shell structures are shown in Fig 1a. Changes in the core-shell morphologies resulted in large changes in H2 production rates and deactivation time. As is shown in Fig 1b, other than the inactive pure NiO sample, H2 production rate first reaches a maximum value then gradually decreases over time for each active core-shell structure. The most active co-catalyst structure has a complicated morphology (Fig. 2a) with an average oxide shell thickness/particle size percentage of ~34%. It gives a maximum H2 production rate of 189 mol/h/g and takes about 6 h to deactivate to 50% max activity. Increased H2 production was found to be related to an increase in the thickness of NiO shell due to suppression of the reverse reaction. HRTEM images reveal that the core-shell co-catalyst structures deactivated primarily due to a loss of metallic Ni from the core structure. Fig. 2 shows that initially both Ni and NiO phases were present in the most active co-catalyst structure. However during deactivation, the catalyst transformed either to structures consisting of NiO nanoblocks (Fig. 2b) or hollow NiO shells (Fig. 2c). The phase transformations occurring during deactivation were associated with Ni diffusion processes that are driven by light illumination. The information we have gained by correlating microstructures with reactivities is now being employed to design improved co-catalysts structures. More analytical data to explore the chemical information will be presented. Microsc. Microanal. 21 (Suppl 3), 2015 640 References: [1] Kudo, A. and Miseki, Y., Chem. Soc. Rev. 38, (2009) 253. [2] Kato, H. and Kudo, A., Chem. Phys. Lett. 295, (1998) 487. [3] Liu, Q., Zhang L., Crozier P.A., Appl. Catal. B: Environ.172,(2015)58. [4] The support from US Department of Energy (DE-SC0004954) and the use of AC-TEM at John M. Cowley Center for HR Microscopy at Arizona State University is gratefully acknowledged. Figure 1. (a) Aberration-corrected TEM image showing a typical Ni core - NiO shell co-catalyst structure. (b) H2 production rates vs. reaction time of different co-catalysts with a series of NiO shell thickness. Figure 2. (a) Initial co-catalyst structure which gives the highest H2 production. FFT indicates both Ni and NiO phases were present initially. After exposure to 16 h UV light and water, two types of deactivated structures emerge: (b) NiO nanoblocks; (c) hollow NiO shell.");sQ1[321]=new Array("../7337/0641.pdf","Atomic Level Element Specific Investigation of Bimetallic Structures by Z-Contrast Imaging","","641 doi:10.1017/S1431927615004006 Paper No. 0321 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic Level Element Specific Investigation of Bimetallic Structures by Z-Contrast Imaging M. Cem Akatay, Sergio I. Sanchez, Steven A. Bradley UOP LLC, A Honeywell Company The sensitivity of HAADF-STEM to atomic number has been used to obtain quantitative information about the structure and chemistry of nanostructures. Electron tomography based on Z-contrast has been used to obtain atomic resolution 3D reconstruction of crystalline nanoparticles [1]. This approach needs two or more projections of the structure that are later processed computationally by discrete tomography to determine the position of each atom in space. This method utilizes crystallographic constraints of the structure and would be of limited use for structures lacking long-range order. Browning et al. utilized quantitative Z-contrast imaging to resolve the structure of sub-nanometer clusters on crystalline metaloxide support [2]. The group has utilized dynamic multislice STEM imaging simulations which greatly benefit from the well-ordered crystalline support of uniform thickness. This method would be of less use on support materials of varied thickness with complex crystal structure. Another setback of this method constitutes itself where quantitative imaging requires calibration of the intensity levels to compensate varying experimental parameters. Different groups have developed theories and best practices to quantify the intensity as accurately as possible, requiring tedious and time-inefficient steps [3, 4]. Our work focuses on the use of Z-contrast imaging to obtain structural and element specific information on sub-nanometer clusters supported on rough surfaces without the need of absolute intensity calibration. The goal is to specify the elemental nature of an atom by differentiating its intensity level from other possible elements. This capability is demonstrated on a Pt-Pd catalyst supported on Al2O3 using an aberration-corrected, sub-�ngstrom resolution scanning transmission electron microscope. This system was chosen as Pt atoms (Z=78) are 1.7 times heavier than Pd atoms (Z=46) resulting in significant contrast difference to be utilized by HAADF-STEM imaging. Further, the studied material has sub-nanometer average cluster size, which simplifies deciphering structure, as clusters composed of fewer atoms have less configuration possibilities. Single atom elemental identification is performed by referencing the lowest atom intensity as Pd and categorizing other atoms based on their intensity relative to the reference atom for each micrograph. Cluster morphology is determined by quantifying the intensity profiles of the clusters with respect to the reference atom. Chemistry of the clusters is estimated based on possible configurations that would satisfy the observed morphology for the existing elements. Pt on Al2O3 is used as a monometallic control sample. The micrograph shown in Figure 1A illustrates Pt atoms along Pt clusters. The cluster is determined to possess a bilayer structure (Figure 1B). Al2O3 supported Pt:Pd (1:1 by weight) catalysts are imaged by aberration corrected STEM (Figure 2A). An example of the results is shown in Figure 2B comparing the intensity profiles of a single atom along with two clusters, one of which is approximately twice as large as the other. The fraction of single Pd atoms is determined by the previously described method for a series of micrographs. The Pt:Pd sample exhibited a higher fraction of single Pd atoms compared to single Pt atoms. The chemical nature of the clusters is also estimated with the aforementioned method and cross-checked by STEM-EDX measurements, both of which found the clusters to be slightly Pt rich. By its bulk loading, the bimetallic sample has 80% more Pd than Pt atoms. Based on this, Pt rich clusters would imply that higher fraction of the single atoms will be Pd, confirming the aforementioned finding. Microsc. Microanal. 21 (Suppl 3), 2015 642 [1] Van Aert et al, Nature (2011), 470 374�377 [2] Browning et al., ChemCatChem (2013), 5 2673 � 2683 [3] LeBeau et al, Ultramicroscopy (2008), 108 1653�1658 [4] Yang et al, Materials Characterization (2003), 51 101� 107 Figure 1. (A) HAADF-STEM micrographs of the Pt/Al2O3 catalyst (B) HAADF-STEM intensity profiles of atoms compared to the line profiles obtained from the cluster. The dashed lines indicate to the unit thickness and its multiples. Figure 2. (A) HAADF-STEM micrographs of the Pt:Pd/Al2O3 catalyst (B) HAADF-STEM intensity profiles of various structures");sQ1[322]=new Array("../7337/0643.pdf","In-situ Study of Coarsening Mechanisms of Supported Metal Particles in Reducing Gas","","643 doi:10.1017/S1431927615004018 Paper No. 0322 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In-situ Study of Coarsening Mechanisms of Supported Metal Particles in Reducing Gas S.Y. Zhang1, Matteo Cargnello2, Christopher B. Murray3, George W. Graham1, and Xiaoqing Pan4 1 2 Dept of Materials Science and Engineering, University of Michigan, Ann Arbor, MI 48109, USA Dept of Chemical Engineering, Stanford University, Stanford, CA 94305, USA 3 Dept of Chemistry and Department of Materials Science and Engineering, University of Pennsylvania, Philadelphia, PA 19104, USA 4 Dept of Chemical Engineering and Materials Science and Department of Physics and Astronomy, University of California - Irvine, Irvine, CA 92697, USA The ability to monitor dynamic material transformation processes in-situ is crucial for understanding structure-property relationships in nano-engineered materials. In the past decade, most of the in-situ gas reaction results were acquired with a dedicated environmental transmission electron microscope (ETEM) where the pressure in the sample area is typically around 10 Torr [1]. To make the in-situ condition more realistic, especially in terms of gas pressure, a MEMS-based closed cell technique has been developed recently [2]. Here we report results utilizing this breakthrough technique to study the coarsening mechanisms of supported metal particles within a conventional spherical aberration-corrected TEM at elevated temperature (above 500 �C) and atmospheric pressure in real time, which provide important knowledge for designing and preparing more active and stable heterogeneous catalysts for specific reactions. The gradual coarsening of metal nanoparticles in supported metal catalysts with time-in-use is one of the most widely recognized modes of catalyst deactivation. In general, particle coarsening occurs via a combination of two thermally activated processes, Ostwald ripening (OR), where larger particles grow at the expense of smaller ones through single-atom release and capture due to the size-dependent difference in surface energy, and particle migration/coalescence (PMC), which involves random motion of entire particles, leading to particle collisions and combination when they are in close enough proximity. Mono-dispersed Pt nanoparticles, either 2.2�0.2 nm or 4.8�0.2 nm in diameter, were prepared as described previously [3]. Catalyst samples containing each, as well as a mixture of both, were subsequently prepared by wet impregnation of Sasol TH100-150 alumina. In-situ observation was carried out with a JEOL 3100 in combination with the Protochips AtmosphereTM system, which consists of a MEMS-based closed cell (two SiN windows, each 30-50 nm in thickness, with a 5 m gap in between), a heating holder, and a gas delivery manifold. The sample was heated under 760 Torr of 5%H2/N2 at 600 �C, 700 �C and 800 �C, each for 5 h, in succession. The use of size-selected samples allows us to cleanly separate the two processes and study the corresponding mechanisms individually. In the sample containing both 2.2 nm and 4.8 nm particles, the dominant mechanism is OR, indicated by the gradual disappearance of the smallest particles in Figure 1. However, OR was successfully Microsc. Microanal. 21 (Suppl 3), 2015 644 suppressed in both of the single-sized samples. Due to the higher mobility of the smaller particles, PMC results in a relatively small change in the particle size distribution of the sample containing the 2.2 nm particles only, as shown in Figure 2, while in the sample containing only 4.8 nm particles, essentially no change was observed after the total 15 h sintering experiment. Ex-situ observations revealed that beam-induced effects, such as surface disorder, particle rotation, and coalescence are pronounced for metal particles smaller than 5-6 nm with long exposure and high electron dose. Examples are shown in sequential images in Figure 3 and Figure 4. To avoid such artifacts, the beam valve needs to be closed between image acquisitions. The images used for particle size distribution measurements here were all taken under relatively low magnification, as a further precaution. References [1] DeLaRiva, A.T. et. al., J. Catal. 308 (2013) 291. [2] Allard, L.F. et al., Microsc. Microanal. 18 (2012) 656. [3] Cargnello, M., et al., Science 341 (2013) 771. [4] The authors gratefully acknowledge funding from Ford Motor Company under a University Research Proposal grant and the National Science Foundation under grants DMR-0723032 and CBET-115940. Figure 1. Sequential images showing the gradual disappearance of the smallest particles in the mixed sized sample at 800 �C. Elapsed time, in minutes, is indicated on the bottom center of each image. Figure 2. Comparison of low (a) and high (b) density areas before and after heating at 800 �C in the sample containing only the 2.2 nm Pt particles. Figure 3. Sequential images showing beam-induced Figure 4. Sequential images showing surface disorder. Elapsed time, in minutes, is indicated on the beam-induced coalescence. Elapsed time, in top right corner of each image. minutes, is indicated on the top right corner of each image.");sQ1[323]=new Array("../7337/0645.pdf","FEI and National Geographic STEM Outreach � Mysteries of the Unseen World","","645 doi:10.1017/S143192761500402X Paper No. 0323 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 FEI and National Geographic STEM Outreach � Mysteries of the Unseen World John Williams1 1. FEI, 5350 NE Dawson Creek Drive, Hillsboro, OR 97124 FEI has partnered with National Geographic on the 3D giant-screen film, Mysteries of the Unseen World. National Geographic is one of the world's largest scientific and educational non-profits with an audience of over 400 million consumers. The goal of Mysteries is inspire curiosity about all things around us that are not visible to the human eye, and in doing so create interest in Science, Technology, Engineering and Math (STEM). Mysteries explores all things that are too fast, too slow, or too small be seen with the human eye. FEI is using Mysteries as a mechanism for STEM outreach to introduce the concepts of the nanoscale, potential applications of nanotechnology, and the uses of electron microscopy. Since its debut in November of 2013, Mysteries has been viewed by over 800,000 consumers at leading museums and science centers around the world. In addition, National Geographic has an extensive educational outreach network to deploy scientific concepts of the film into the classroom. To date, National Geographic has conducted 11 educator workshops, reaching over 750 educators and more than 15,000 students. An additional 13,000 educator DVD's have been distributed by theaters and at events. FEI uses Mysteries as the foundation of its STEM outreach efforts. We introduce students to the science and beauty of life and materials at the nanoscale and the tremendous capability of electron microscopy in revealing and in some cases creating this beauty. FEI is continuing to partner with National Geographic and with leading museums to integrate electron microscopy into their STEM education outreach programs. FEI has promoted the film and the associated microscopy themes through hosted events, interactive kiosks at museums, engagements at major STEM exhibitions such as the USASEF, and through development and deployment of an electron-microscopy-based iPad app. In 2014, FEI reached over one million consumers through Mysteries-related outreach. Microsc. Microanal. 21 (Suppl 3), 2015 646 Figure 1. Mysteries of the Unseen World promotional poster Figure 2. Students at USASEF learning about electron microscopy via the interactive kiosk");sQ1[324]=new Array("../7337/0647.pdf","Using Microscopy for Authentic Science Teaching: A Learning Sciences Perspective","","647 doi:10.1017/S1431927615004031 Paper No. 0324 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Using Microscopy for Authentic Science Teaching: A Learning Sciences Perspective Martin Storksdieck1 1 Oregon State University, Center for Research on Lifelong STEM Learning, Corvallis, OR, USA We are currently experiencing a new national effort to improve science education in K-12, one that is building on a somewhat failed effort from the mid-1990s to introduce "inquiry" into the science classroom. This new movement is build around the adoption and implementation of the Next Generation Science Standards [1], which themselves represent a faithful translation of a National Research Council report that provided the basic guidebook for these new standards, entitled A Framework for K-12 Science Education [2]. In contrast to previous science education standards from the 1990, the new ones are formulated as so-called "performance expectations"; performance expectations specify what a student should be able to do or demonstrate in terms of scientific or engineering competency that can be observed or tested. The NGSS are therefore not curriculum guidelines, but the endpoint of a learning or teaching effort that is oriented towards a student outcome. How this is achieved is left to curriculum developers or teachers. There is much controversy over the adoption and implementation of standards in general, so-called common core standards in particular, and the NGSS as well [3]. Since the NGSS are performance standards, some of the potential controversy is tight to how one would know whether students perform to expectations [4]. However much these questions might currently dominate some of the public discourse around science education reform, the fact remains that a new vision for science education and science learning has been formulated by the Framework, and been channeled into the political process via the NGSS. These standards now provide an exciting opportunity for authentic engagement with science in K-12 [5]. This talk will elaborate ways in which microscopy as an easily accessible scientific instrument/method can be used to further the vision for science education reform. The new opportunities for enriching K-12 science education through microscopy arises through the eight science and engineering practices that now form the central avenue through which students are to engage with a limited set of core disciplinary ideas and so-called cross-cutting concepts in science � those that are common in all disciplines. The practices are: 1. Asking questions (for science) and defining problems (for engineering) 2. Developing and using models 3. Planning and carrying out investigations 4. Analyzing and interpreting data 5. Using mathematics and computational thinking 6. Constructing explanations (for science) and designing solutions (for engineering) 7. Engaging in argument from evidence 8. Obtaining, evaluating, and communicating information. Microsc. Microanal. 21 (Suppl 3), 2015 648 Microscopy allows us to explore the unseen, to use technology to extend our ability to explore natural phenomena. Moreover, it invites the learner to pay attention to details; to distinguish between that which can be seen and which is imputed; to consider the limitation of a scientific instrument; and just as much, explore the intersection between instrumentation and scientific insight. Using microscopy, therefore, not only touches on many of the 8 practices in some form or another, it also creates an opportunity to link engineering, the resulting technology, mathematics and science, and therefore creates an integrated approach towards STEM [6]. Microscopy is not limited to school-based investigations. Informal or out-of-school learning is sometimes more powerful than schooling, and increasingly a new perspective on student-centered, lifelong and 24/7 learning across settings and time emerges as a new paradigm in our knowledge society [7] [8]. Microscopy is not only being used widely in museum settings, it also opens the door to many hobby scientists, and many children explore at home the world with microscopes in their spare time. This is not trivial statement: using scientific instruments as part of spare time activities is a powerful means for creating an identity as a science learner, and maybe even as a future scientists. Research on learning suggests that active engagement and practice are essential in developing scientific understanding [9]. Microscopy, when used as an authentic tool for exploration into the unknown and unseen, can and should therefore be an essential tool in formal and out-of-school science learning. References: [1] National Research Council. Next Generation Science Standards: For States, By States. Washington, DC: The National Academies Press, 2013. [2] National Research Council. A Framework for K-12 Science Education: Practices, Crosscutting Concepts, and Core Ideas. Washington, DC: The National Academies Press, 2012. [3] National Research Council. Guide to Implementing the Next Generation Science Standards. Washington, DC: The National Academies Press, 2015. [4] National Research Council. Developing Assessments for the Next Generation Science Standards. Washington, DC: The National Academies Press, 2014a. [5] National Research Council. America's Lab Report: Investigations in High School Science. Washington, DC: The National Academies Press, 2005. [6] National Academy of Engineering and National Research Council. STEM Integration in K-12 Education: Status, Prospects, and an Agenda for Research. Washington, DC: The National Academies Press, 2014b. [7] National Research Council. Learning Science in Informal Environments: People, Places, and Pursuits. Washington, DC: The National Academies Press, 2009. [8] National Research Council. STEM Learning Is Everywhere: Summary of a Convocation on Building Learning Systems. Washington, DC: The National Academies Press, 2014c. [9] National Research Council. How People Learn: Brain, Mind, Experience, and School. Washington, DC: The National Academies Press, 1999.");sQ1[325]=new Array("../7337/0649.pdf","Magnetic Dynamics Studied by Time-Resolved Electron Microscopy","","649 doi:10.1017/S1431927615004043 Paper No. 0325 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Magnetic Dynamics Studied by Time-Resolved Electron Microscopy F. Carbone1, R. Jayaraman1, Y. Murooka1, G.F. Mancini1, E. Baldini1, P. Huang2, M. Cantoni3, A. Magrez4, D. McGrouther5, T. Giamarchi6, H. M. R�nnow2 1. Laboratory for Ultrafast Microscopy and Electron Scattering, Ecole Polytechnique F�d�rale de Lausanne, CH-1015 Lausanne, Switzerland 2. Laboratory for Quantum Magnetism, Ecole Polytechnique F�d�rale de Lausanne, Switzerland 3. Centre Interdisciplinaire de Microscopie Electronique, CIME, Ecole Polytechnique F�d�rale de Lausanne, CH-1015 Lausanne, Switzerland 4. Competence in Research of Electronically Advanced Materials, Ecole Polytechnique F�d�rale de Lausanne, CH-1015 Lausanne, Switzerland 5. School of Physics and Astronomy, Kelvin Building, University of Glasgow G12 8QQ Scotland 6. Departement de Physique de la Mati�re Condens�e, University of Geneva, Switzerland email: fabrizio.carbone@epfl.ch Keywords: magnetic dynamics, time-resolved electron microscopy, cryo-microscopy For future technology, it is necessary to increase the data density of storage media, and to advance functionality of the various devices. Magnetism at the nanoscale is particularly promising for developing magnetic data storage, magnetic switching, spintronics, and THz technology. While to date magnetic properties have mainly been observed and characterized statically, it has now become essential to monitor and capture their fast dynamics. The dynamics are based on fundamental and characteristic magnetic processes such as domain wall motion, skyrmion lattice formation, magnetic damping, vortex motions, spin-orbit coupling. These occur at their characteristic time scales spreading from ms to fs. On the other hand, nanostructures can play essential roles for advancing functionality such as nanodot, hetero-structures, surface, defects, grain boundaries. Therefore it is often essential to directly monitor the localized nanoprocesses while responding to external stimulations such as electric/magnetic field, electron radiation, and temperature. For this purpose, we have developed time-resolved transmission electron microscopy (TEM), which was recently extended to the fs domain [1]. In contrast to other techniques such as Magnetic Force Microscopy, Holography, spin-polarized SEM, our approach can access time-resolved dynamics of bulk behaviors that happens between ms and fs. A pulsed laser was introduced externally both at the cathode and the sample. Photoelectrons are emitted from the cathode as a pulse train, and accelerated toward the sample to 200kV. The dynamics can be induced by the laser pulse, and can be monitored at the fs time scale in a pump (laser)-probe (electron) geometry. ms dynamics can be recorder with a TV-rate camera or CCD camera, combined with an electron energy filter. We also extended our other capabilities for controlling the sample's temperature down to liquid helium temperature. With this cryo-TEM, for example, vortexes in superconductors MgB2 were revealed [2]. Furthermore, high quality crystals with unique properties are synthesized and were precisely characterized. Here, we will describe the dynamical role of disorder in a large and flat thin film of Cu2OSeO3, exhibiting a skyrmion phase in an insulating material [3]. A 150 nm thin film was prepared by Focused Ion Beam; particular care was devoted into obtaining a large (7000 nm) and flat sample. In Fig. 1, a Lorentz microscopy image of our sample is shown. This micrograph was recorded in our JEOL JEM2100 modified for ultrafast operation. The sample was cooled down to 7 K and Lorentz microscopy was performed in Fresnel mode by controlling the field produced by the microscope objective lens. A field of 400 G was applied to the material and the skyrmion phase was induced. The image shows the regular distribution of over 50000 skyrmions. The effect of thermal and beam-irradiation induced fluctuations will be described showing the impact that disorder has on the skyrmion Microsc. Microanal. 21 (Suppl 3), 2015 650 lattice orientation, and providing a unique point of view on the topology of this magnetic ground state. Fig. 1 Cryo-Lorentz TEM image of the skyrmion lattice forming within the helical domains in Cu2OSeO3. The applied field is 400 G and the base temperature was 7 K. The thin film was thinned down to 150 nm. References: [1] L Piazza, et al., Chemical Physics 423 (2013), p. 79. [2] M J G Cottet, et al., Phys. Rev. B 88 (2013), p. 014505. [3] A A Omrani, et al., Phys. Rev. B 89 (2014), p. 064406. [4] The authors gratefully acknowledge funding from an ERC starting grant.");sQ1[326]=new Array("../7337/0651.pdf","Toward Ultrafast Electron Microscope with Femtosecond Temporal Resolution, Atomic-Level Spatial Resolution, and Single-Shot Imaging Capability","","651 doi:10.1017/S1431927615004055 Paper No. 0326 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Toward Ultrafast Electron Microscope with Femtosecond Temporal Resolution, Atomic-Level Spatial Resolution, and Single-Shot Imaging Capability Katsumi Tanimura The institute of Scientific and Industrial Research, Osaka University, Osaka, Japan Innovative progress in materials science in the last century made it possible to determine the structures of matters at the atomic-level (1 �) resolution, and to detect ultrafast structural dynamics in the temporal domains as short as 10-13 s. However, one important limitation was left unsolved; that the ultimate spatial and temporal resolutions are incommensurate. At the beginning of this century, the new innovation has started to realize direct structural determination of materials in ultrafast processes. In particular, ultrafast time-resolved electron diffraction methods have developed dramatically to capture structural dynamics in short temporal domains, as reviewed in Refs. 1 and 2. The next milestone of ultrafast structural dynamic studies may be to establish the femtosecond (fs) time- resolved electron microscope (U-TEM) with atomic-level spatial resolution and with a single-shot imaging capability [3]. Here I describe the present status of our study to construct such a U-TEM, based on our single-shot fs electron diffractometer using relativistic ultrafast electron pulses. The relativistic electron pulses give two crucial advantages over conventional DC-accelerated electron beams in electron diffraction and electron microscope. First, MeV electrons greatly enhance extinction distance for elastic scattering, and provide structural information almost free from any multiple diffraction effects; the kinematic theory assuming single scattering events can be applied. This feature allows us to obtain ultrafast atomistic-scale structural evolution by combining time-resolved diffraction results with state-of-art theoretical modeling, which is capable of quantitatively reproducing the time evolution of diffraction intensities. Detailed atomistic evolutions have been obtained in laser-generated warm-dense gold [4]. This is a version of electron diffractive imaging. Second, for relativistic electron beams minimal longitudinal pulse broadening will result, even with more than 109 e/cm2 per pulse. This enables us to perform single-shot transmission diffraction at fstemporal resolution, crucial for irreversible processes such as laser-induced solid/liquid transitions. More importantly, this feature opens a possibility to obtain single-shot imaging of atomic structures, by combining appropriate lens systems and high-sensitive detectors. For any atomic motions in fs-temporal domains, single-shot imaging is crucial as we study structural changes beyond ergodic picture. The renovation from the electron diffractometer to microscope needs several challenging developments. These include ultimate stability and monochromaticity of energies of MeV electrons generated by rf photocathode, precise beam control for collimation, focusing and expansion, and high-sensitive detection for imaging. The present status of these developments is discussed in details. References: [1] A. H. Zewail, Annu. Rev. Phys. Chem. 57, 65 (2006). [2] G. Sciaini1and R. J. Dwayne Miller, Rep. Prog. Phys. 74, 096101 (2011). [3] D. J. Flannigan and A. H. Zewail, Accounts Chem. Res. 45, 1828 (2012). [4] S. L. Daraszewicz, et al., Phys. Rev. B 88, 184101 (2013). Microsc. Microanal. 21 (Suppl 3), 2015 652");sQ1[327]=new Array("../7337/0653.pdf","RF Photoinjector Based Time-Resolved MeV Electron Microscopy","","653 doi:10.1017/S1431927615004067 Paper No. 0327 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 RF Photoinjector Based Time-Resolved MeV Electron Microscopy P. Musumeci University of California at Los Angeles Department of Physics and Astronomy, Los Angeles, CA, USA Transmission electron microscopy (TEM) is one of the primary tools for materials characterization and finds many important applications in various area of scientific research. Over the past two decades, significant advances have been made in the spatial resolution achievable in TEM (now well below 1 �) through the implementation of aberration correction [1]. One of the recent trends to further expand the capabilities of TEMs is to dramatically improve the temporal resolution to ultrafast time scales in order to study the dynamics of microscopic processes in real time [2-4]. So far, two approaches for providing temporal resolution to TEMs have been explored. One is the stroboscopic ultrafast electron microscopy (UEM) developed at Caltech [5]. The number of electrons per pulse is reduced to one on average to minimize the collective Coulomb interactions between electrons. This approach has achieved atomic scale spatial resolution and a few hundred femtosecond temporal resolution, but is limited to study reversible processes since the pump-probe event has to be identically reproduced by millions of times for a single image. The other approach is the single-shot method pioneered by Bostanjoglo and co-workers [6] in which each image is taken by an electron pulse containing millions of particles. The most successful example of this approach, the Dynamic Transmission Electron Microscope (DTEM) developed at LLNL [7] has achieved 10 nm - 15 ns spatialtemporal resolution. A recent DOE workshop [8] identified one of the critical needs for a major new instrumental development in the area of time-resolved electron microscopy to achieve single-shot realspace imaging with a spatial/temporal resolution of 10 nm/10 ps. In this paper we discuss the development of a single shot picosecond time resolved transmission electron microscope (SPTEM) with 10 ps temporal resolution and 10 nm spatial resolution based on the use of MeV beams from an RF photoinjector that would improve the current state-of-the-art in temporal resolution in single shot electron microscopy by three orders of magnitude. A throughout design of the instrument has been outlined in a recent Phys. Rev. Applied paper [9]. A schematic of the proposed system is shown in Fig. 1. The final target of the proposed work (which is currently funded under a Phase I STTR with Radiabeam Technologies) is to enable us to take real-space snapshots of irreversible structural changes with few ps shutter speed and tens of nanometer resolution. The key innovations to enable this goal are: (i) the use of high energy (4 MeV) electrons from a high gradient photocathode radiofrequency (RF) electron gun to employ of the highest possible beam brightness in the TEM, and take advantage of the relativistic suppression of space charge interactions; (ii) an X-band RF cavity to dramatically reduce the energy spread of the beam from the RF gun to below 10-5 thus minimizing the effects of chromatic aberrations; (iii) a compact strong electron optical column based on permanent magnet quadrupoles (PMQs) to avoid the issues of large relativistic electron lenses. Current state of the art RF photoinjectors use 100 MV/m peak field gradients to generate very high brightness relativistic electron beams. These sources have played a prominent role in the development of the XFEL and have recently been exploited also in the field of ultrafast electron diffraction [10]. RF photoinjectors have already demonstrated [11] the capability to generate sub-ps beams with ~ pC charge and <30 nm-rad normalized emittance, satisfying the requirements for single shot imaging in a TEM. By Microsc. Microanal. 21 (Suppl 3), 2015 654 lengthening the pulse to 10 ps and further increasing the extraction field on the cathode adopting a novel 1.4 cell gun design, it will be possible to further reduce the emittance to 10 nm, thus reducing the divergence at the sample to < 1 mrad and enabling both incoherent shadow and diffraction contrast imaging mechanisms in the proposed microscope. Figure 1: Schematic of the proposed SPTEM system, including an S-band photocathode RF gun, an X-band regulation cavity, the main solenoid condenser stage, and the PMQ triplet objective lens. One particular aspect where further work is needed is certainly the control of the beam energy spread. In order to achieve the required a few tens of nm spatial resolution, the chromatic and spatial aberrations of the focusing column must be minimized and the beam energy spread should be kept below 1�10-4. This can be achieved introducing an additional RF cavity, operating at X-band, to eliminate the correlated energy spread after the gun. Another issue is the development of compact electron lenses to replace the large and bulky solenoidal lenses required to focus relativistic beams and thus minimize the propagation distance along the column giving less time to the space charge effects to degrade the image quality. In [9] we devised a solution to provide x12 magnification stage with limited aberrations using a permanent magnet quadrupole triplet. Three magnification stages will be included in the final column for a total magnification of x1000 enabling 10 nm spatial resolution with standard electron detectors. References: [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] P. E. Batson, N. Dellby, and O. L. Krivanek, Nature 418, 617 (2002). Future Science Needs And Opportunities For Electron Scattering: Next-Generation Instrumentation And Beyond, DOE workshop March 1-2, 2007. Dynamic Processes in Biology, Chemistry, and Materials Science: Opportunities for Ultrafast Transmission Electron Microscopy, PNNL workshop June 14-15, 2011. W. E. King et al., J. Appl. Phys. 97, 111101 (2005). A. H. Zewail, Science 328, 187 (2010). O. Bostanjoglo, Advances in Imaging and Electron Physics 121, 1 (2002). N. D. Browning et al., in Handbook of Nanoscopy, Chapter 9 (2012) Future of Electron Scattering and Diffraction workshop. http://www.orau.gov/electron2014/ R. K. Li and P. Musumeci. Phys. Rev. Applied, 2, 024003 (2014) P. Musumeci and R. K. Li, ICFA Beam Dynamics Newsletter #59 (2012). R. K. Li et al., Phys. Rev. ST Accel. Beams 15, 090702 (2012).");sQ1[328]=new Array("../7337/0655.pdf","Ultrafast Lattice Dynamics of Granular L10 Phase FePt Measured by MeV Electron Diffraction.","","655 doi:10.1017/S1431927615004079 Paper No. 0328 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Ultrafast Lattice Dynamics of Granular L10 Phase FePt Measured by MeV Electron Diffraction. A. H. Reid1, X. Shen2, R. K. Li2, S. P. Weathersby2, G. Brown2, T. Chase1, R. Coffee2, J. Corbett2, J. C. Frisch2, N. Hartmann2, J. Li3, C. Hast2, R. K. Jobe2, E. N. Jongewaard2, J. R. Lewandowski2, J. E. May2, D. McCormick2, T. Vecchione2, J. Cao4, E. E. Fullerton5, Y.K. Takahashi6, J. Wu2, X. J. Wang2, & H. A. D�rr1 Stanford Institute for Materials and Energy Science, SLAC National Accelerator Laboratory, 2575 Sand Hill Road, Menlo Park, CA 94025, USA. 2. SLAC National Accelerator Laboratory, 2575 Sand Hill Road, Menlo Park, CA 94025, USA. 3. National Synchrotron Light Source, Brookhaven National Laboratory, Upton, New York 11973, USA 4. Physics Department and National High Magnetic Field Laboratory, Florida State University, Tallahassee, Florida 32310, USA 5. Center for Magnetic Recording Research, UC San Diego, 9500 Gilman Drive, La Jolla, CA, 92093, USA. 6. Magnetic Materials Unit, National Institute for Materials Science, Tsukuba 305-0047, Japan. The L10 phase of FePt exhibits extremely high magnetic anisotropy, making it a material of choice for next generation magnetic storage applications. However, these applications rely on new methods for writing magnetic information, such as Heat Assisted Magnetic Recording [1] or All-Optical Magnetic Recording [2]. Both Heat Assisted Magnetic Recording and All-Optical Magnetic Switching use laser induced thermal dynamics to assist in changing the magnetic state. In this paper we study the Ultrafast response of a granular FePt thin film to rapid thermal heating using a 50 fs optical laser pulse. The response of the FePt grains gives direct insights into the nanoscale heat transport within the grains; the thermal properties of the FePt L10 phase and the vibration modes of the grains. FePt grains in a carbon matrix are grown in the L10 phase by co-sputtering with carbon onto an MgO single crystal substrate. The film segregates into FePt grains separated by carbon, see Fig. 1 (a). The grains formed are approximately cylindrical, with a height of 10 nm and a log-normal distribution of diameters with a modal value of 10 nm. The FePt is in the L10 structure, with the crystallographic c axis normal to the substrate. The substrates are removed by chemical etching and the remaining freestanding FePt film is floated onto a copper microscopy grid. The experiment is conducted in a collinear pump-probe geometry. The FePt film is pumped with a 50 fs 800 nm laser pulse focused to a spot of 1 mm diameter with an incident fluence of 5 mJ/cm2. The lattice response is probed by a relativistic electron bunch with a kinetic energy of 2.2 MeV, subpicosecond duration, and 100 fC charge, produced by an rf photocathode gun at the SLAC ASTA Ultrafast Electron Diffraction facility. The electron diffraction pattern formed on a phosphor screen is imaged by an Andor EMCCD camera for various laser�electron time delays (Fig. 1b). The FePt film shows well defined Bragg reflections, implying that individual grains grew epitaxially from the MgO substrate and are aligned crystallographically with one another. The diffraction confirms the grains' crystallographic phase and orientation. The intensity and position of various Bragg reflections are monitored as a function of pump probe delay. By tilting the sample surface to 45 degrees 1. Microsc. Microanal. 21 (Suppl 3), 2015 656 from normal incidence Bragg reflections with components along the crystallographic c-axis are measured--in particular the [111] and [202] reflections. Together with measurements of the Bragg reflections at normal incidence, these allow the full three dimensional expansion dynamics of the grains to be measured in real time. The diffraction measurements reveal that the thermal expansion of granular FePt is anisotropic. A linear thermal expansion is observed for the in-plane a and b directions. However, the thermal expansion along the out-of-plane c axis in near zero. This behavior persists to fluences up to 17 mJ/cm2. The dynamic response of the FePt grains also exhibits a large degree of anisotropy. The laser pulse causes a rapid volume expansion of the grains, driven by the expansion of the grain in the in-plane (crystallographic a & b) directions. A new equilibrium volume is reached after 3 ps. The rapid in-plane expansion is observed to drive a contraction of the lattice in the out-of-plane direction. This sets up a volume conserving bulk oscillation with a period of 7 ps (Fig 1c, d). These experiments offer new insights into the thermal behavior of nanoscale FePt. [1] D. Weller et al. "L10 FePtX-Y media for heat assisted magnetic recording", Phys. Status Solidi A 210, 1237 (2013). [2] C-H. Lambert et al. "All-optical control of ferromagnetic thin films and nanostructures", Science 345, 1337 (2014). [3] This work is supported by the Department of Energy, Office of Science, Scientific User Facilities Division and SLAC Director Program Development Fund. A.H.R. is supported by the Department of Energy, Laboratory Directed Research and Development funding, under contract DE-AC02-76SF00515. Work at UCSD supported by the ONR MURI program. Figure 1. (a) TEM image of the FePt grains. (b) Schematic of the electron diffraction experiment with the FePt sample surface tilted with respect to the incoming laser and electron pulses. (c) Illustration of two dynamical motions observed within the FePt grains.");sQ1[329]=new Array("../7337/0657.pdf","High Spatial/Energy Resolution Band Gap Measurements: Delocalization and Other Effects in a Monochromated Cold FEG Nion Dedicated STEM.","","657 doi:10.1017/S1431927615004080 Paper No. 0329 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 High Spatial/Energy Resolution Band Gap Measurements: Delocalization and Other Effects in a Monochromated Cold FEG Nion Dedicated STEM. R. W. Carpenter1,3, H. Xie2,4, T. Aoki3, and F. A. Ponce4,3. 1 2 Dept. of Chemistry and Biochemistry, Arizona State University, Tempe, USA. School for Engr. of Matter, Transport, and Energy, Arizona State University, Tempe, USA. 3 LeRoy Eyring Center for Solid State Science, Arizona State University, Tempe, USA. 4 Dept. of Physics, Arizona State University, Tempe, USA. Measuring band gaps and delocalization with a STEM microscope requires a small probe, for spatial resolution, and high enough energy resolution and stability to both resolve the gap and minimize the energy width of the ZLP so that the gap structure is not obscured. The Nion microscope used here is located in a specially prepared very stable site and has demonstrated 20 meV energy resolution with a 2 � probe, and small enough ZLP base width to observe small band gaps and lattice phonon losses [1,2,3]. The ratio of Nion to typical Schottky FEG TEM/STEM ZLP width at 1% of maximum intensity is 0.035. This ratio will vary a bit depending on how current is adjusted in the probe, but in general it is small, due the monochromator and cold FEG. Here we show results for band gap measurements relative to edge-on interface position in two cross section specimens. In one specimen the materials on opposite sides of the interface have quite different gaps (Al2O3, Eg ~ 9 eV and GaN, Eg ~ 3.4 eV) and in the other the difference is small ( AlGaN multiquantum well superlattice, Eg ~ 5 eV and AlN, Eg ~ 6.2 eV). The main results for both materials systems were the same: when the probe was on the interface the band gaps for both materials in the couple were visible in the spectrum, and when the probe was moved away from the interface the band gap signal of the material containing the probe remained strong, while the gap signal from the material on the opposite side of the interface deceased until it disappeared when the probe was sufficiently far from the interface. Some of the results for the Al2O3/GaN interface are shown in Fig. 1.The distance from the interface where the GaN gap signals disappears is ~62 �, and the distance for Al2O3 gap signal disappearance is ~ 39 �. Similar results are shown for the MQW superlattice/GaN specimen in Fig. 2. Here, the MQW gap signal disappears about 67� from the interface on the AlN side, but on the MQW side the gap is still between 5 and 6 eV at 17 � from the interface, i.e. 17� is not far enough from the interface for the effect of the larger AlN gap to disappear. In this latter case the gap difference is only about 1 eV, and it was difficult to resolve the two gaps simultaneously because their onsets are not sharp. The magnitude of the delocalization lengths we observed are in remarkably good agreement with the approximate values given by Egerton for losses smaller than 50 eV : ~ 35 � for 10 eV loss and ~ 60 � for 5 eV loss[4]. The shapes of the loss curves of Figs. 1 and 2 introduce some uncertainties into band gap measurements, especially the higher energy band gap of each pair, because the loss curve for that constituent lies on top of the lower gap loss curve, making background correction difficult. This effect can increase the apparent gap by a small amount. Nevertheless our values for AlN and GaN measured away from interfaces are in reasonably good agreement with reported values [5]. A wide range of values have been reported for Al2O3, with the highest values near about 9 eV, and lower values reported for specimens containing impurities or exhibiting Urbach tails [6]. During optical band gap measurements for Al2O3 Tomiki et. al. observed a significant peak at ~ 6 eV loss that they attributed to impurities [6], which is similar to the broad peak also centered at ~ 6 eV that we observed for the Al2O3 loss curve (upper curve of Fig. 1). We also attribute this midgap feature in Al2O3 spectra to impurities. Additional data for other systems will also be presented. Microsc. Microanal. 21 (Suppl 3), 2015 658 References [1] RW Carpenter et al, Micros. Microanal. 18(Supp 2) (2012), p.408. [2] O L Krivanek et al, Microscopy 62(1) (2013), p.3. [3] O L Krivanek et al, Nature 514 (2014), p.209. [4] R F Egerton, Ultramic. 107 (2007) p.575. [5] S Bloom, J Phys Chem Solids 32 (1971) p.2027. [6] T Tomiki et al, Jour Phys Soc (Japan) 62(No. 3) (1993) p.573. 62� into Al2O3 GaN nearly gone At Interface Al2O3 gap GaN gap Figure 1. Three low loss spectra from an Al2O3/GaN interface specimen. Top curve, 62 � into Al2O3 from interface. Middle curve, probe on interface, Bottom curve, probe 39 � into GaN. These spectra are from a line scan of 100 steps, of 2.4 � length. Recorded at 60 kV, 0.005 eV/ch dispersion. Probe size ~ 2 �. 39� into GaN, and Al2O3 gone 67 � into AlN At Interface 17 � into MQW Figure 2. Three LL curves from AlGaN MQW superlattice/AlN interface specimen. Green curves are back ground stripped using a window before the ~ 5 eV MQW gap. Top curve: in the AlN Middle curve: on the interface Bottom curve: in the MQW. These curves are from a line scan with steps recorded 16.7 � apart, at 60 kV, probe size ~ 2 �.");sQ1[330]=new Array("../7337/0659.pdf","Avoidance of Radiation Damage in Vibrational-Mode Energy-Loss Spectroscopy.","","659 doi:10.1017/S1431927615004092 Paper No. 0330 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Avoidance of Radiation Damage in Vibrational-Mode Energy-Loss Spectroscopy. R.F. Egerton1, P. Rez2 and P.A. Crozier3,4 1. 2. Physics Department, University of Alberta, Edmonton, Canada T6G 2E1. Department of Physics, Arizona State University, Tempe, AZ 85287, USA. 3. Center for Solid State Science, Arizona State University, Tempe, AZ 85287, USA. 4 School for Engineering of Matter, Transport and Energy, ASU, Tempe, AZ 85287, USA. New-generation monochromator systems incorporated into transmission electron microscopes allow electron energy-loss spectroscopy (EELS) to be performed on small regions of a thin specimen with an energy resolution in the range 10 � 50 meV. This enhanced performance will be highly useful in characterizing electronic properties of materials (e.g. bandgap studies in semiconductors) and it also makes possible the study of vibrational (phonon) modes of energy loss (vibEELS), which can provide information about the nature of the chemical bonds present in the specimen. The vibrational-mode peaks occur mostly in the energy range 0.1 � 0.2 eV. Although some high-angle component is expected [2 � 4], the recorded signal arises largely from dipole interactions that involve scattering angles below 1 mrad [5], implying that the interaction is delocalized over distances of many nm. Although this delocalization reduces the spatial resolution of vibEELS analysis, it offers the possibility of aloof-mode measurement [6] with the electron probe tens of nm beyond the edge of the specimen, as demonstrated experimentally [1]. In this mode, electronic transitions within the specimen are only weakly excited, so radiation damage to the specimen might be expected to be minimal. Avoidance of radiation damage is important because many of the interesting vibEELS specimens are beam-sensitive, which further limits the spatial resolution of the analysis [5]. We have simulated the signal and damage in aloof mode by taking the energy-loss probability as: dP/dE = t e2(220h2v2)-1 Im{-2/[(E) + 1]} K0[4 r E/(vh)] (1) where t is the specimen thickness, v is the incident-electron speed, (E) is the relative permittivity of the specimen at an energy loss E, and r is distance of the probe from the edge of the specimen. The modified Bessel function K0 represents delocalization of the signal, extending over many nm as shown in Fig. 1. Equation (1) applies equally to the electronic excitations in the specimen that give rise to radiolytic damage but because of the higher values of E, the delocalization is much less; see Fig.1. In fact, we might expect that radiolysis requires an energy exchange of several eV, as evidenced by the fact that polymers can be degraded by ultraviolet radiation but not by visible or infrared light. If this energy threshold is 5 eV, for example, Fig.1 suggests that a probe placed at least 20 nm from the edge of the sample will excite few of the electronic excitations that lead to damage but still generate a substantial vibrational-loss signal. Evaluation of the energy-loss function in Eq.(1) allows us to simulate the effect of changing the threshold energy, for comparison with measurements of dP/dE as a function of r. The response dP/dE can be thought of as the integral of a point-spread function dP/dA over nearby areas A of the specimen. If dP/dA 1/r2 over most of its range [7], dP/dE (dP/dA) dA (1/r2) r dr = Microsc. Microanal. 21 (Suppl 3), 2015 660 log(r) and this behavior is confirmed by plotting K0 against log(r), as in Fig. 1b. The lower limit rmin of this approximation is determined by the probe diameter and cutoff of the Lorentzian angular distribution of scattering. The upper limit rmax is set by dynamical screening (Bohr adiabatic limit) and gives rise to the curvature seen in Fig. 1b, which suggests rmax ~ 1000 nm for a 0.15eV peak and 60keV electrons. So we might expect some vibrational-loss signal to arise from material lying as far as 1�m from the edge of the specimen, with half the signal generated within about 30 nm of the edge, assuming log(r) behavior. In conclusion, it appears that radiolysis damage can be almost completely avoided by positioning an electron probe at least 20 nm from the edge of a specimen, assuming that the probe has no aberration tails, that phonon excitation is non-damaging and that radiolysis requires a minimum energy transfer of a few eV. The spatial resolution of the vibrational-loss signal will then be some tens of nm, somewhat better than the dose-limited resolution (DLR) for the most radiation-sensitive organic materials [5,8]. References: [1] OL Krivanek et al, Nature 514 (2014), p. 209. [2] P Rez, Microsc . Microanal. 20 (2014), p. 671. [3] C Dwyer, Phys. Rev. B 89 (2014), p. 054103. [4] P Cueva and DA Muller, Microsc . Microanal. 20, suppl. 3 (2014), p. 590. [5] RF Egerton, Microsc. Microanal. 20 (2014), p. 658. [6] A Howie, Ultramicroscopy 11 (1983), p. 141. [7] RF Egerton, "EELS in the Electron Microscope", 3rd edition, (Springer, New York) p.226. [8] RFE acknowledges funding from the Natural Sciences and Engineering Research Council of Canada and wishes to thank Michael Bergen for his advice on Matlab programming. (a) (b) Figure 1. (a) K0 function of Eq.(1), showing spectral intensity as a function of distance r of the probe from the edge of the specimen, for a vibrational peak at 0.1 eV and an electronic excitation at 5eV loss, assuming a STEM operating at 60 kV. (b) K0 function plotted against log10(r), for a 0.15eV vibrational peak and for electronic energy losses of 5 eV and 10 eV, assuming an incident energy of 60 keV.");sQ1[331]=new Array("../7337/0661.pdf","Hydrogen Analysis by Ultra-High Energy Resolution EELS","","661 doi:10.1017/S1431927615004109 Paper No. 0331 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Hydrogen Analysis by Ultra-High Energy Resolution EELS H. Cohen1, P. Rez2, T. Aoki3, P.A. Crozier3,4, N. Dellby5, Z. Dellby5, D. Gur6, T.C. Lovejoy5, K. March7, M.C. Sarahan5, S.G. Wolf 1 and O.L. Krivanek2,5 1. 2. 3. 4. 5. 6. 7. Department of Chemical Research Support, Weizmann Institute of Science, Rehovot 76100, Israel Department of Physics, ASU, Tempe, AZ 85287, USA Center for Solid State Science, Arizona State University, Tempe, AZ 85287, USA School for Engineering of Matter, Transport and Energy, ASU, Tempe, AZ 85287, USA Nion Co., 11511 NE 118th St., Kirkland, WA 98034, USA Structural Biology, Weizmann Institute of Science, Rehovot 76100, Israel Physique des Solides, University of Paris-Sud, Orsay 91405, France Hydrogen is a difficult element to detect by electron microscopy (EM). Being very light, its elastic scattering cross-section is much smaller than for other elements. Even worse, it does not give rise to any core loss edges that can be used for detecting it by electron energy loss spectroscopy (EELS), nor to the emission of characteristic X-rays or Auger electrons. In metals, hydrogen typically becomes a loosely bound proton, plus an electron that joins the valence band. In organic materials it is bound more tightly, but it is easily ejected by knock-on or ionization damage. Its detection has been reported by annular bright field (ABF) STEM [1] and other methods such as measuring plasmon energy shifts, but not by a general method applicable to all materials. A reliable EM spectroscopic method for mapping hydrogen would be very useful, especially if it were also able to provide information on its bonding. Ultra-high energy resolution (UHR) EELS is now able to provide such a method. A new type of an electron monochromator [2] has improved the EELS energy resolution about 3-fold, to ~10 meV at 60 keV primary energy, and this has made vibrational spectroscopy possible in the electron microscope [3]. With its single proton nucleus, hydrogen gives the highest energy vibrational peaks and is therefore the easiest element to detect by the vibrational approach, in principle. Fig. 1 shows vibrational spectra of titanium hydride (TiH2) and epoxy resin [3]. The vibrational peak due to the loosely bound hydrogen in TiH2 is at 147 meV and thus lies on a large background due to the zero loss peak (ZLP), but it can be readily extracted using a power-law fit. The main vibrational peak in the epoxy resin is due to more strongly bound hydrogen and occurs at 360 meV. The cross-sections are of the order of 1 pm2 per atom, i.e. similar to the carbon K shell ionization cross section [4]. However, because the vibrational signal is concentrated into narrow peaks rather than an edge tens of eV wide, it is more readily detectable. Due to the low energies, a large part of the vibrational signal is "delocalized" and can be excited even with an "aloof" electron beam positioned tens of nm in the vacuum next to the sample [3]. The TiH2 spectrum was in fact acquired with an aloof beam of ~2 nm diameter stationed ~30 nm in the vacuum outside a TiH2 particle. Provided that the beam is substantially tail-free, radiation damage is then greatly reduced, as the only transitions the aloof beam can excite in the sample occur at energies < 1 eV. This means that if the sample can tolerate infrared light, it will not be greatly damaged by the aloof beam. For the epoxy spectrum, the beam was defocused and covered a sample patch of ~400 nm diameter [3], to minimize radiation damage per unit area. A disadvantage of this approach is that the wide beam is sensitive to sample charging which can act as an imperfect "lens" that adds aberrations worsening the spectrum quality. A more robust approach for minimizing radiation damage is to scan a focused beam rapidly over the sample and de-scan it after the sample, and we are implementing this mode. Microsc. Microanal. 21 (Suppl 3), 2015 662 The best signal-to-noise ratio in vibrational spectra of beam-sensitive materials is usually obtained in the aloof mode, albeit at a reduced spatial resolution. Fig. 2 shows a UHR EEL spectrum of guanine acquired at 60 keV, with the beam parked 60 nm outside the sample. The agreement with the infrared spectrum is excellent, although the EELS energy resolution (~16 meV) is considerably worse. As is typical of vibrational spectroscopies, the different peaks can be assigned to different types of bonds and vibration modes (see the inset in Fig. 2). This opens up many interesting possibilities, including being able to monitor how different types of hydrogen bonds stand up to electron exposure. A further improvement of energy resolution by about 2x should be straightforward with our approach, provided that both the monochromator and the spectrometer are brought up to the rigorous stability standards developed for aberration-corrected microscopes. The spatial resolution of the analysis can most likely also be improved substantially, by focusing on high-angle EELS scattering events [5]. To summarize this exciting new development, UHR EELS promises to make the analysis of hydrogen as routine and potentially more informative than analytical electron microscopy of any other element. References: [1] R. Ishikawa et al., Nature Materials 10 (2011) 278-281. [2] O.L. Krivanek et al., Microscopy 62 (2013) 3-21. [3] O.L. Krivanek et al., Nature 514 (2014) 209-212. [4] P. Rez, Microsc. Micoanal. 20 (2014) 671-677. [5] T.C. Lovejoy et al., Microsc. Micoanal. 20 suppl. 3 (2014)558-559. [6] We gratefully acknowledge the use of facilities within the LeRoy Eyring Center for Solid State Science at ASU, and grants NSF MRI-R2 #959905 and DE-SC0004954 and DE-SC0007694. Figure 1. UHR EEL spectra of TiH2 and epoxy resin. Nion UltraSTEM100MC, 60 kV. Figure 2. Top: UHR EEL spectrum of guanine (Nion UltraSTEM100MC, 60 kV), aloof mode. Bottom: infrared spectrum of guanine.");sQ1[332]=new Array("../7337/0663.pdf","Applications of Plasmon Energy Expansion Thermometry","","663 doi:10.1017/S1431927615004110 Paper No. 0332 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Applications of Plasmon Energy Expansion Thermometry Matthew Mecklenburg1, William A. Hubbard2, E. R. White 2, Rohan Dhall3, Stephen B. Cronin3, Shaul Aloni4, and B. C. Regan2 1. Center for Electron Microscopy and Microanalysis, University of Southern California, Los Angeles, CA, USA. 2. Department of Physics and Astronomy & California NanoSystems Institute, University of California, Los Angeles, CA, USA. 3. Department of Electrical Engineering, University of Southern California, Los Angeles, CA, USA 4. Molecular Foundry, Lawrence Berkeley National Laboratory, Berkeley, CA, USA. Plasmon energy expansion thermometry (PEET) has finer spatial resolution than conventional thermometers such as thermocouples, optical pyrometers, and resistance thermometers [1]. In an electron microscope, a beam electron interacting with a sample sometimes loses energy by creating a plasmon, or charge oscillation. A plasmon's energy is proportional to the square root of the sample material's valence electron density. Thermal expansion changes this density and thereby shifts the plasmon energy [2]. PEET relates the plasmon energy, which can be mapped with few-nanometer spatial resolution using electron energy loss spectroscopy (EELS), to temperature. Such high-spatial resolution thermometry has numerous applications in electron microscopy [3], including heating holder calibration and temperature gradient mapping. In a heating sample holder, whether it is furnace-style or has a microfabricated heater membrane, the temperatures are not always uniform and well-characterized. Both in situ and external temperature calibrations have associated uncertainties. In the former case the thermometer cannot be co-located with the sample (it is often quite remote), and in the latter the calibration may change. PEET can calibrate sample temperature as a function of heater power directly (Fig. 1). Examples of plasmon energy maps acquired for PEET are shown in Figs. 1E-F, along with the simultaneously acquired annular dark field (ADF) images (Figs. 1C-D). Both Fig. 1E and Fig. 1F show a network of grain boundaries, which give 10-20 meV plasmon energy shifts due to the local density decrease. (The detection of these grain boundaries demonstrates that PEET is sensitive to density changes that occur on atomic length scales.) Figure 1E is the reference (room temperature) map. During the acquisition of the Fig. 1F map, the heater power was stepped (Fig. 1A), which produced the plasmon shifts appearing as dark bands. The small heater induces little thermal drift. Combining the data of Figs. 1E-F [1] gives the temperature change (Fig. 1G). The heater power vs temperature plot (Fig. 1H) represents a local, in situ temperature calibration. Here a continuous aluminum film was used, but nanoparticles of aluminum or another material with a well-defined plasmon could be dispersed on a sample to provide point thermometry within the field of view. PEET can also map the temperature gradients in an operating microelectronic device such as the aluminium heater shown in Fig. 2. With increasing power dissipation the temperature gradients in the aluminium wire become steeper. The wire is hottest in the center, much like an incandescent bulb's filament. PEET adds a new temperature mapping capability, on par with structure determination and elemental analysis, to the already extensive capabilities of TEM for characterizing microelectronic devices [4]. Microsc. Microanal. 21 (Suppl 3), 2015 664 References: [1] M Mecklenburg et al, Science 347 (2015), p. 629-632. [2] G Meyer, Zeitschrift fur Physik 148 (1957), p. 61-71. [3] M P Seah et al, Journal of Materials Science 21 (1986), p. 1305-1309. [4] This work was supported by NSF DMR-1206849, and in part by FAME, one of six centers of STARnet, a Semiconductor Research Corporation program sponsored by MARCO and DARPA. Data presented were acquired at the Center for Electron Microscopy and Microanalysis at the University of Southern California. Figure 1. (A) A plot of power vs time during the EELS scan. (B) An intensity scale for the plasmon maps (E) and (F) and temperature map (G). Images (C) and (D) show the ADF images simultaneously acquired with the plasmon maps in (E) and (F). The temperature map resulting from (E) and (F) is shown in (G). Each map is 46x271 pixels with a 5 nm pixel pitch. Figure 2. Four maps at different powers are shown. Each map is 313x256 pixels in size with a 17 nm pixel pitch. The histogram of temperatures from each image is shown below the corresponding image.");sQ1[333]=new Array("../7337/0665.pdf","Aberration corrected STEM-EELS study of the hole distribution in cuprate superconductors","","665 doi:10.1017/S1431927615004122 Paper No. 0333 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Aberration corrected STEM-EELS study of the hole distribution in cuprate superconductors M. Bugnet1, G. Radtke2, S. L�ffler1,3, P. Schattschneider3, D. Hawthorn4, G. A. Sawatzky5, and G. A. Botton1 1. Department of Materials Science and Engineering & CCEM, McMaster University, Hamilton, Canada. UPMC � Paris 6, IMPMC-CNRS UMR 7590 Campus Jussieu, Paris, France. 3. Institute of Solid State Physics and USCTEM, Vienna University of Technology, Vienna, Austria. 4. Department of Physics and Astronomy, University of Waterloo, Waterloo, ON, Canada. 5. Department of Physics and Astronomy, University of British Columbia, Vancouver, BC, Canada. 2. Over the last decade, aberration correctors, monochromators, brighter electron sources and the improved stability of electron optics in the transmission electron microscope have drastically improved the capabilities of analytical electron microscopy. Atomic-scale structural and chemical analyses of a wide range of materials are now routinely available. In particular, electron energy loss spectrometry (EELS) in the scanning transmission electron microscope (STEM) allows to combine both spatial and chemical information, via elemental mapping in crystals at the atomic level [1], by rastering a high current sub�ngstr�m probe over the region of interest of the specimen. Using STEM-EELS, elemental mapping can provide a direct and clear evidence of the atomic-scale structure of a material. In addition, in a first approximation, core-loss EELS can directly probe the unoccupied density of states allowed by transition rules. The energy loss near edge structures (ELNES) contain relevant information on the electronic structure, chemical bonding, and thereby related properties of the material under investigation. It is of particular interest that spatially-resolved information on the electronic structure and properties can be obtained by selecting specific spectral features of the ELNES. Indeed, these fine structures arise from transitions to unoccupied states of a particular energy, thereby allowing the identification of localized states at the atomic scale [2]. The relevance of identifying spatially-resolved fine structures is shown in the case of cuprate superconductors. Indeed, the localization of holes/electrons in these compounds is of primary interest to further understand their physical properties. Although other unoccupied states spectroscopy such as xray absorption spectroscopy (XAS) provide valuable spectral information, performing EELS measurements within the TEM has the unarguable advantage of higher spatial resolution [3,4]. As an example, the chain-ladder superconductor CaxSr1-xCu24O41, made of CuO2 chains and Cu2O3 ladders, is investigated by STEM-EELS at the atomic level (see Figure 1a,b). The STEM-EELS experiments were carried out in two aberration-corrected FEI Titan microscopes equipped with monochromators and high-resolution Gatan EELS spectrometers. The atomic resolution achieved by aberration-corrected STEM-EELS allows the study of the hole distribution atomic column by atomic column, in contrast to XAS that integrates over all Cu-rich planes in the structure. Fine structures at the O-K and Cu-L2,3 edges are in good agreement with existing XAS results. Analysis of the hole distribution is performed from the O-K pre-edge ELNES, which is composed of the O-2p hole band, and the upper Hubbard band (see Figure 1c) [5,6]. Using the spectrum imaging technique available in STEM-EELS, we demonstrate that the holes lie preferentially within the CuO2 chains of the structure. The quantitative analysis aspects of the hole concentration determination are discussed, in comparison Microsc. Microanal. 21 (Suppl 3), 2015 666 with available results from existing data in the literature. Furthermore, the effect of channeling on the quantification is discussed based on simulations. This work highlights the combination of ELNES analyses with atomic resolution in the aberration corrected STEM, to improve the understanding of the electronic properties of cuprate superconductors [7]. References: [1] SJ Pennycook and C Colliex, MRS Bulletin 37 (2012), p.13. [2] S L�ffler et al, submitted. [3] N Gauquelin et al, Nature Communications 4 (2014), p.1. [4] J Fink et al, Physical Review B 42 (1990), p. 4823. [5] A Rusydi et al, Physical Review B 75 (2007), p. 104510. [6] M Bugnet et al, in preparation. [7] The experimental work has been performed at the Canadian Center for Electron Microscopy, a national facility supported by NSERC, the Canada Foundation for Innovation and McMaster University. 60 1.000 231.0 461.0 691.0 921.0 1151 1381 1611 1841 2071 2301 2531 2761 2991 3221 3451 3681 3911 4141 4371 4601 4831 5061 5291 5521 5751 5981 6211 6441 6671 a b Ladders Chains Ladders Chains Ladders Horizontal projection along [001] 40 20 1 nm Zone axis [100] 1.000 381.0 761.0 1141 1521 1901 2281 2661 3041 3421 3801 4181 4561 4941 5321 5701 6081 6461 6841 7221 7601 7981 8361 8741 9121 9501 9881 1.026E+04 1.064E+04 1.102E+04 1.140E+04 1 Channel 526 528 530 532 534 0 Energy loss (eV) c O-K edge (EELS) O-2p Holes Upper Hubbard Band Higher energy transitions Model Beam along [100] Intensity 522 524 526 528 530 532 Energy Loss (eV) 534 536 Figure 1. Spectrum image data set of CaxSr1-xCu24O41 illustrating simultaneously acquired STEMHAADF image (a) and O-K pre-edge EELS spectra projected along [001] (b). Variations in the intensity of the pre-edge fine structures are observed from chains to ladders. (c) Monochromated O-K edge of CaxSr1-xCu24O41 recorded in [100] zone axis. The pre-edge fine structures are fitted with Gaussian functions. nergy loss (eV) 28 530 532");sQ1[334]=new Array("../7337/0667.pdf","Holograms for the Generation of Vortex States with L=500 Fabricated by Electron Beam Lithography","","667 doi:10.1017/S1431927615004134 Paper No. 0334 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Holograms for the Generation of Vortex States with L=500 Fabricated by Electron Beam Lithography Erfan Mafakheri1 Vincenzo Grillo2,3 Roberto Balboni4,Gian Carlo Gazzadi3, Claudia Menozzi1,2 Stefano Frabboni,1,2 Ebrahim Karimi5 and Robert W Boyd5,6 1. Dipartimento FIM, Universit� di Modena e Reggio Emilia, Via G. Campi 213/a, I-41125 Modena, Italy 2. CNR-Istituto Nanoscienze, Centro S3, Via G. Campi 213/a, I-41125 Modena, Italy 3. CNR-IMEM Parco Area delle Scienze 37/A, I-43124 Parma, Italy 4. CNR-IMM, Via P. Gobetti 101, I-40129 Bologna, Italy 5. Department of Physics, University of Ottawa, 25 Templeton, Ottawa, Ontario, K1N 6N5 Canada 6. Institute of Optics, University of Rochester, Rochester, New York 14627, USA Electron Vortex beam with high quanta of OAM are interesting for their magnetic properties and to explore the transition between quantum and classical definition of Orbital Angular Momentum[1]. This justifies the technological effort to produce increasingly complicated hologram encoding for such complicated case. While so far the technology of hologram fabrication was based on FIB [2][3] patterning on SiN with a small exception [4] we are exploring the possibility to produce the highest quanta of OAM ever built by using Electron Beam Lithography (EBL) We demonstrate here the case of a vortex beam with nominal OAM L=500. One extremely interesting problem is that the generation and characterization of such beams is indubitably challenging since the number of pixel necessary for the sampling of the vortex structure is typically very high. For example even if the phase were directly observed on the CCD screen (e.g. by interference) most CCD mounted on modern TEM would produce a poor representation of the vortex due to the limited number of pixel. At the same time providing a large enough carrier frequency will require a very large memory representation while calculating the hologram and poses some limits in term of the FIB possibility. This is where the EBL can produce a big advantage since in principle a larger number of pixel can be controlled and potentially with a finer and more precise pitch. For the actual realization of the hologram we created by FIB the reference circular aperture in the Au layer used to cover the standard SiN support film. The other side was used for EBL patterning. The membrane was spin coated by PMMA while different dose factors were tested. The final processing of the hologram was carried on by exposure in the chamber of Reactive Ion beam Etching. In fig 1a it is possible to see a TEM overview of the hologram while fig 1b is a detail from which we can see that the average pitch size was about 100 nm. While we were not able to capture a full detailed thickness map of the overall hologram, we used the nominal hologram shape to predict its diffraction. We were able to obtain a "reasonable agreement" between the simulation and the experimental intensity of the vortex and this can be considered as a proof that a beam close to L=500 has been generated. In fig 1c we superimposed a simulated intensity to an experimental image at focus. While one of two vortexes was reasonably close to simulations, important deviation can be observed in the opposite diffraction order. They could be probably ascribed to the bending of the SiN membrane. In spite of some experimental difficulties we generated one of the largest quantum angular momentum state not only in electron microscopy and we have opened the road for massive EBL application in holographic masks [5]. Microsc. Microanal. 21 (Suppl 3), 2015 668 References: [1] V. Grillo, G. C. Gazzadi, E. Mafakheri, S. Frabboni, E. Karimi and R W. Boyd Phys. Rev.Lett 114, (2015) 034801 [2] V. Grillo, G C Gazzadi, E. Karimi, E Mafakheri,R W. Boyd, and S Frabboni Applied Physics Letters 104, 043109 (2014) [3] T.R Harvey, J S Pierce, A K Agrawal, P Ercius, M Linck and B J McMorran New Journal of Physics 16 (2014) 093039 [4] Tyler R. Harvey, Gii Brougher, Kurt Langworthy, and Benjamin J. McMorran. Poster P09-05 presented at EIPBN 2013, May 2013 [5] The authors acknowledge the help of Franco Carillo for some of the EBL procedures. a) b) c) Figure 1. a) Overall image of the EBL fabricated Hologram. b) TEM image of a detail of the holographic map 3) Experimental diffraction from the hologram: In the inset we superimposed the simulation on the intensity based on the projected hologram.");sQ1[335]=new Array("../7337/0669.pdf","Synthesization of vortex beams by combining fork-shaped gratings for transmission electron microscopy","","669 doi:10.1017/S1431927615004146 Paper No. 0335 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Synthesization of vortex beams by combining fork-shaped gratings for transmission electron microscopy Ken Harada1, Teruo Kohashi1 and Tomohiro Iwane1 1 Central Research Laboratory, Hitachi Ltd., Hatoyama, Saitama 350-0395, JAPAN Electron vortex beams are considered to be probes for next-generation electron beam devices because vortex beams carry intrinsic orbital angular momentum. Their use as probes should thus improve measurement capability [1, 2, 3]. We fabricated fork-shaped gratings with a 400-nm lattice constant made from a Si3N4 membrane with a 200-nm thickness using a focused ion beam machine (FB-2100, Hitachi High-Technologies Corp.). Electron diffractions from the gratings were observed using a prototype 300-kV field emission electron microscope [4] with an optical system constructed for small angle diffraction (camera length ~150 m or larger) [5]. Ring-shaped spots in the observed diffraction patterns of the vortex beams were a typical feature. Furthermore, the shape and size of the grating openings were superimposed on the shape of the diffraction spots of the vortex beam [6]. A ring-shaped spot is directly usable as an incident beam probe for "scanning" type electron microscopes. For conventional transmission electron microscopes (TEMs), however, the feature of a spot is unusable because of its non-uniform irradiation density. To extend the use of vortex beam to TEMs, we are aiming at developing a plane-wave-like illumination probe with a uniform density. In the present work, we tried to superimpose several ring-shaped spots of different sizes to create uniform illumination in the reciprocal space. Three methods were investigated: (1) superimpose vortex beams with different topological numbers, (2) superimpose vortex beams with the same topological number but different diameters, and (3) superimpose vortex beams with the same topological number but different diameters from a single phase grating. Figure 1 shows the results for the second method. Nine fork-shaped gratings of third order dislocation with different diameters (10, 6.7, 3.3, and 1.7 m) were combined as one grating-device (left panels), and diffraction patterns with superimposed ring-shaped diffraction spots (right panels). The rectangular lattice within the diffraction pattern in the right panels was caused by the four-fold symmetric arrangement of the gratings. This rectangular lattice, dot-like irradiation is undesirable for TEMs. We determined that random positioning of the gratings solved this problem. Figure 2 shows the results for the third method. Single fork-shaped gratings of third order dislocation with two or three superimposed openings with different diameters are shown in the left panels, and their electron diffraction patterns are shown in the right panels. The phase of each grating lattice was a key factor in the harmonization of the superposition of the ring-shaped diffraction spots. With the third method, the grating was single phase, so the superposition was perfectly harmonized. However, the beam intensity for the smallest opening was too weak for it to be visible with the contrast in Fig. 2. Microsc. Microanal. 21 (Suppl 3), 2015 670 The diffraction spots superimposed on the specimen were observed using small-angle electron diffraction when the specimen was positioned at the diffraction plane. These results show that the illumination area can be extended by using the second and third methods. The next step is to irradiate a specimen with a superimposed vortex beam and observe unprecedent objects, hopefully. References: [1] M. Uchida and A. Tonomura, Nature, 464, (2010) p. 737. [2] J. Verbeeck et al., Nature, 469, (2010) p. 301. [3] B. J. McMorran et al., Science, 331, (2011) p. 192. [4] T. Kawasaki et al., Jpn. J. Appl. Phys., 29, (1990) p. L508. [5] K. Harada, Appl. Phys. Lett., 100, (2012) p. 061901. [6] K. Harada et al., Microsc. Microanal., 21, (2014) p. 274. Figure 1. Figure 2. Figure 1. Combined gratings as grating-devices (left) and electron diffraction patterns with superimposed ring-shaped diffraction spots (right): (a) single grating-device, (b) one grating and two grating-devices, (c) one grating and four grating-devices. Several rings were superimposed on diffraction pattern. Figure 2. Fabricated single fork-shaped gratings of third order dislocation with two or three superimposed openings with different diameters (left) and electron diffraction patterns (right): (a) singlediameter grating, (b) double-diameter grating, (c) triple-diameter grating.");sQ1[336]=new Array("../7337/0671.pdf","Propagation properties of Electron Vortex Beams","","671 doi:10.1017/S1431927615004158 Paper No. 0336 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Propagation properties of Electron Vortex Beams K. Saitoh1, H. Nambu2, N. Tanaka1 and M. Uchida3 1 2 EcoTopia Science Institute, Nagoya University, Nagoya 464-8603, Japan Department of Crystalline Materials Science, Nagoya University, Nagoya 464-8603, Japan 3 Advanced Science Research Laboratory, Saitama Institute of Technology, Fukaya 369-0293, Japan Electron vortex beam, an electron beam carrying an orbital angular momentum (OAM), has been attracting a great attention by its unique physical property and application to a new microscopy in materials science [2,3]. In the present paper, we show our latest studies on the generation and propagation dynamics of electron vortex beams [4,5,6]. Electron vortex pairs embedded in a single beam are generated using a nanofabricated holographic grating and observed their propagation dynamics. Figure 1(a) shows successive series of cross sections of a propagating electron beam which has two vortices with the same topological charge of +1. The pair of the phase singularities, which are indicated by arrows, rotates in the same direction as the beam propagates. When the topological charges of the two vortices are opposite to each other, they rotate opposite direction. The experimental results are well explained by the Gouy phase shift and confirmed by numerical simulations of the propagation. The results of this study may help develop an intuitive understanding of electron vortex motion and give a new viewpoint for analyzing electron microscope images and phase change in a crystal. We seek to explore the measurement of electron OAM by using a nano-fabricated, forked grating. We experimentally examine how an electron vortex beam with orbital angular momentum (OAM) undergoes diffraction through a forked grating. The nth-order diffracted electron vortex beam after passing through a forked grating with a Burgers vector of 1 shows an OAM transfer of n. Hence, the diffraction patterns become mirror asymmetric owing to the size difference between the electron beams. Such a forked grating, when used in combination with a pinhole located at the diffraction plane, could act as an analyzer to measure the OAM of input electrons [6]. This may explain the results of recent experimental studies by Verbeeck et al., in which the observation of dichroism is reported in EELS of ferromagnetic Fe thin films using electron vortex beams [2]. We also study the utility of forked gratings for phase retrieval. Forked gratings produce diffracted peaks of electron vortex beams. These peaks have rich information of phase distribution on a plane of the forked grating. We apply the Fourier iterative phase retrieval technique for diffraction patterns obtained using forked gratings. Figures 2(a) and 2(b) show two examples of phase retrievals. Fig. 2(a) shows a phase retrieval of a forked grating without any specimen. The phase retrieval shows a good convergence and robustness compared to that done by a simple circular aperture, indicating that the diffracted vortex peaks carry additional information for the phase retrieval. A simulation study shows a correlation between the specimen shape and distortion manner of the diffracted electron vortex beams, which might be used to map a phase distribution in real space. Microsc. Microanal. 21 (Suppl 3), 2015 672 (e) Fig. 1. Forked gratings generating Electron vortex pairs embedded in a single beam. Binarized computer holograms calculated with topological charges of (+1, +1) (a) and (+1, -1) (b). Electron beam masks of (+1, +1) (c) and (+1, -1) (d), which are nano-fabricated by a focused ion beam instrument. (e) A successive series of cross sections of a propagating vortex pair embedded in a single beam. (g) (i) (k) (h) (j) (l) Fig. 2 (a)-(f): Phase retrieval of a forked grating without specimen. Experimental TEM image of forked grating (a) and diffraction pattern of (b) are used as real space and reciprocal space constraints. The amplitudes of the image (c) and diffraction pattern (d) are well reproduced by the present analysis. A constant phase in the image (e) and phase rotation of each diffracted peaks in the diffraction pattern are visualized. (g)-(l): Phase retrieval of a forked grating with a specimen of a Fe thin film. The introduction of the specimen disturbs circular symmetric vortex peaks. The amplitudes of the experimental image (g) and diffraction pattern (h) are well reproduced by the present analysis as shown in (i) and (j), respectively. References [1] M. Uchida and A. Tonomura, Nature 464 (2010) 737. [2] J. Verbeeck, H. Tian, and P. Schattschneider, Nature 467 (2010) 301. [3] B. J. McMorran et al. Science 331 (2011) 192. [4] K. Saitoh et al. J. Electron Microsc. 61 (2012) 171. [5] Y. Hasegawa et al. J. Phys. Soc. Jpn. 82 (2013) 033002. [6] K. Saitoh et al, Phys. Rev. Lett. (2013) .");sQ1[337]=new Array("../7337/0673.pdf","Electron Singularities, Matter Wave Catastrophes, and Vortex Lattice Singularimetry","","673 doi:10.1017/S143192761500416X Paper No. 0337 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electron Singularities, Matter Wave Catastrophes, and Vortex Lattice Singularimetry T. C. Petersen1*, M. Weyland2,3, D. M. Paganin1, T. P. Simula1, S. A. Eastwood1, A. I. Bishop1 and M. J. Morgan1 1. 2. School of Physics and Astronomy, Monash University, Clayton, Australia Department of Materials Engineering, Monash University, Clayton, Australia 3. Monash Centre for Electron Microscopy, Monash University, Clayton, Australia Singularities in wave fields, such as phase vortices, are of substantial interest for light optics, matter waves and x-ray optics. New theoretical insights [1] and cutting edge nanofabrication techniques have recently created a wide variety of electron vortex beams [2, 3] for the transmission electron microscope (TEM), which can provide spectroscopic information pertaining to the magnetic properties of nanoscale specimens [2]. The rich diffraction physics of singular optics is vast and many concepts can be directly translated to electron microscopy. Interesting physics has been discovered by highlighting differences between electron and optical vortex beams, such as a remarkable interplay between Zeeman coupling, Landau levels and Gouy phases in the TEM [4] or the Faraday effect for electron vortex beams propagating in a magnetic field, which interestingly occurs in vacuum [5]. Our singular electron optics experiments began by considering ways in which to spontaneously nucleate vortices in the TEM using aberrations. Using insights from Berry et al. [6], we have created electron diffraction catastrophes [7] to produce distorted lattices of vortices by imposing probe forming aberrations in a Titan3 80-300 TEM (FEI), which can correct both the illumination and imaging lenses. More recently, we used electron aberrations to produce the elliptic umbilic catastrophe and have measured the non-trivial three dimensional (3D) architecture of the hyperbolic umbilic. The diffraction catastrophe experiments naturally led to measurements of the Gouy phase anomaly [8, 9] for astigmatic electron waves [10] using in-line holography. The Gouy anomaly describes quantized phase changes incurred by rays on the optic axis, which pass through a caustic. By examining the 3D form of the astigmatic caustic in the TEM, we sought to derive an intuitive explanation of the Gouy phase anomaly for focused electron waves, based upon the probability density in the vicinity of line foci. By considering quantum fluctuations near the optic axis, we united several disparate theoretical interpretations within the framework of focussed paraxial electron waves [11]. Specifically, we related the anomaly to Heisenberg momentum fluctuations, statistical confinement, Berry's geometric phase, and Keller's quantized semi-classical phase changes, known as Maslov indices. We predicted direct encoding of Gouy phases in Gabor holograms, which could be integrated using quantum `weak measurements' of the local momentum. Preliminary data demonstrates this approach, as in shown Figure 2, which is more accurate than our original in-line holography results. Our studies of electron caustics informed new approaches for creating vortices in Bose Einstein Condensates (BEC) using quantum gas analogous of aberrated focusing, based upon external trapping potentials, which were confirmed using simulations of the Gross-Pitaevskii equation, including nonlinear interactions [12]. The BEC study created new insights for interpreting recent Bose nova experiments and the detection of dipole-dipole interactions. Microsc. Microanal. 21 (Suppl 3), 2015 674 We are currently motivated by practical applications of vortex lattices for interferometry [13], which we have been developing using light optics. We have derived a quantitative form of `singularimetry', in which a 3-beam lattice of optical vortices provides an algebraic and localized method for measuring phase shifts imparted by an absorbing and refracting specimen in one arm of a 3-beam interferometer. For certain microscopy techniques, it may not be technologically feasible to use multiple beam splitters in this manner. To circumvent this issue, we have developed an alternative differential form of singularimetry, which utilizes vortices and gradient singularities as topological fiducial markers in a structured illumination context. This new approach provides a concise analytic measurement of phase gradients imparted by refracting specimens, yielding quantitative information that is both local and deterministic. We have accurately quantified our phase gradient experiments to demonstrate that lattices of wave field singularities can be used to detect subtle specimen variations with high precision. [1] KY Bliokh, YP Bliokh, S Savel'ev and F Nori, Phys. Rev. Lett. 99 (2007), p. 190404. [2] J Verbeeck, H Tian and P Schattschneider, Nature 467 (2010), p. 301. [3] BJ McMorran et al., Science 331 (2011) p. 192. [4] KY Bliokh, P Schattschneider, J Verbeeck and F Nori, Phys. Rev. X 2 (2012), p. 041011. [5] C Greenshields, RL Stamps and S Franke-Arnold, New J. Phys. 14 (2012), p. 103040. [6] MV Berry, JF Nye and FJ Wright, Phil. Trans. A 291 (1979), p.1382. [7] TC Petersen et al., Phys. Rev. Lett. 110 (2013) p. 033901 [8] TD Visser and E Wolf, Opt. Commun. 283 (2010), p. 3371. [9] G Guzzinati, P Schattschneider, KY Bliokh, F Nori and J Verbeeck, Phys. Rev. Lett. 110 (2013), p. 093601. [10] TC Petersen et al. Phys. Rev. A 88 (2013) p. 043803. [11] TC Petersen et al. Phys. Rev. A 89 (2014) p. 063801. [12] TP Simula, TC Petersen and DM Paganin, Phys. Rev. A 88 (2013), p. 043626. [13] SA Eastwood, AI Bishop, TC Petersen, DM Paganin and MJ Morgan, Opt. Express 20 (2012) p. 13947. Figure 1. Gouy phase jumps measured directly from an astigmatic electron caustic. The experiment has been vertically scaled to match theory, heuristically accounting for partial coherence and image blur. Figure 2. a) Raw intensity of a three beam vortex lattice, b) experimental phase gradients measured at vortex locations c) theoretical vector field at vortex locations; colors indicate gradient magnitude.");sQ1[338]=new Array("../7337/0675.pdf","Holographic Generation of Highly Twisted Electron Beams","","675 doi:10.1017/S1431927615004171 Paper No. 0338 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Holographic Generation of Highly Twisted Electron Beams Vincenzo Grillo,1, 2 Gian Carlo Gazzadi,1 Erfan Mafakheri,1, 3 Stefano Frabboni,1, 3 Ebrahim Karimi,4, and Robert W Boyd4,5 1. 2. CNR-Istituto Nanoscienze, Centro S3, Via G. Campi 213/a, I-41125 Modena, Italy CNR-IMEM Parco Area delle Scienze 37/A, I-43124 Parma, Italy 3. Dipartimento FIM, Universit� di Modena e Reggio Emilia, Via G. Campi 213/a, I-41125 Modena, Italy 4. Department of Physics, University of Ottawa, 25 Templeton, Ottawa, Ontario, K1N 6N5 Canada 5. Institute of Optics, University of Rochester, Rochester, New York 14627, USA Free electrons can possess an intrinsic orbital angular momentum upon free-space propagation [1][2]. Beams with a high number of twists are of particular interest because they carry a high magnetic moment about the propagation axis. Using electron holographic plates obtained by FIB patterning of SiN membranes we generated a beam with an orbital angular momentum up to 200 in the first order of diffraction: the biggest ever demonstrated so far [3]. Using EFTEM mapping in a JEOL 2200 FEG TEM microscope operated at 200keV we carried on a detailed experimental analysis of the thickness pattern in the holographic plate in order to evaluate the introduced phase effect and therefore the wavefunction right at the exit of the hologram. By simple propagation we were able to know the wavefunction at different planes: The intensity pattern was found to be in agreement with the experiments at different focal planes. Fig 1a shows indeed the comparison of the intensity pattern as calculated by propagation and the experiment at Fraunhofer, it is evident a very good agreement. The knowledge of the wavefunction permits to decompose it in different values of Orbital angular momentum (OAM). The OAM spectrum is evaluated by transforming the calculated electron wave function into polar coordinates and performing the Fourier transform in the azimuth coordinates. It is easy to verify that when the OAM is written with respect to the geometric center of one of the vortexes the correct average OAM number of about L=200 is recovered. The broadening of the distribution is due to the loss of intensity at very high frequencies induced by hologram imperfections. In fact an eigenvalue of OAM should have no azimuthal localisation. We were also able to locate the positions of the each singularity: in fig 1c we highlighted in red the singularity centres and superimposed them to the intensity. Singularities are mainly located on the periphery of the dark region and spatially decomposed probably as an effect of the interference of the 0th order beam that cannot be completely spatially separated. An in depth analysis of the radial profile also reveals a large number of ripples outside the main ring, suggesting that the generated beam has a large spectrum decomposition in terms of Laguerre-Gauss Eigenfunctions, especially with respect to the radial number p. These characteristic can be tight down to the hard aperture used to limit the hologram. In fig 1d we fitted for sake of example the wavefunction at a given plane using a "minimal" set of Laguerre Gauss states about L=200 and p=0-50 but it is worth recalling that the actual radial decomposition depends sensibly on the actual electron optics, on the hologram details and on hologram illumination conditions. We expect to apply these beams to interact with magnetic field and to couple with Landau states in the microscope but it is important to take into account the above limitations to plan future experiments. Microsc. Microanal. 21 (Suppl 3), 2015 676 References: [1] J. Verbeeck, H. Tian, and P. Schattschneider, Nature 467, (2010) p 301. [2] B. J. McMorran, A. Agrawal, I. M. Anderson, A. A. Herzing, H. J. Lezec, J. J. McClelland, and J. Unguris, Science 331, 192 (2011). [3] V. Grillo, G. C. Gazzadi, E. Mafakheri, S. Frabboni, E. Karimi and R W. Boyd Phys. Rev.Lett 114, (2015) 034801 Experiment Calculated 0.12 0.10 0.08 |c|2 0.06 0.04 0.02 0.00 -300 -200 -100 0 100 200 300 a) c) b) OAM d) Figure 1. a) Comparison of the experimental and calculated intensity of the hologram diffraction: the +/-1 and 0th order diffraction are visible. b) spectrum of the OAM of one of the generated vortexes. c) phase singularities (red) superimposed to the diffracted intensity: almost all singularities are located to the perifery of the dark region. d) example of possible fitting of the actual wavefunction to a limited set of Laguerre Gauss state. The fitting (above) and the experiment (below) are compared. Even under simplified assumptions on the hologram illumination a minimum set states of a few L (see b) and radial number p=0-50 are necessary to provide a qualitatively reasonable decomposition of the vortex state.");sQ1[339]=new Array("../7337/0677.pdf","Characterization of Advanced Nanomaterials for Lithium Ion Batteries Cathodes","","677 doi:10.1017/S1431927615004183 Paper No. 0339 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of Advanced Nanomaterials for Lithium Ion Batteries Cathodes Nicolas Brodusch1, Pierre Hovington2, Hendrix Demers1, Raynald Gauvin1, and Karim Zaghib2 1 2 Department of Mining and Materials Engineering, McGill University, Montr�al, Qu�bec, Canada. Institut de Recherche d'Hydro-Qu�bec (IREQ), Varennes, Qu�bec, Canada. Nowadays, the high and fluctuating oil prizes combined with the urgent need for clean transport technologies have pushed the leaders in the energy storage industry to develop high capacity batteries to sustain a driving autonomy that could competition oil-based vehicles. For this purpose, safe and efficient materials regarding the charge/discharge cycling as well as the storage capacity need to be developed for each electrode of the battery. At this time, the more promising materials for the positive (cathode) electrode of lithium ion batteries (LIB) in terms of electrochemical properties and safety has been the lithium iron phosphate, LiFePO4 (LPF), powders. However, the bulk electronic conductivity of lithium iron phosphate is quite low, and carbon is generally added in the LPF matrix or at the LPF particles surface to enhance their electrical conductivity [1]. Recently, orthosilicate materials were investigated for cathode electrodes due to the SiO4polyanion potential in reducing the M2+/M3+ redox potential in Li2MSiO4 nanomaterials (M = Mn, Fe, Co or Ni), thus improving the charge/discharge behavior of the cell [2]. Since the last two decades, scanning electron microscopy has faced a revolution in the development of field-emission microscopes with the most powerful in terms of versatility and spatial resolution being the cold-field emission scanning electron microscope (CFE-SEM). This type of SEM provides low voltage surface-sensitive imaging as well as low voltage STEM imaging with transmitted electrons detectors combined with high efficiency x-ray detectors and transmitted diffraction capabilities that permitted to obtain high quality results on a wide range of materials science topics [3, 4]. Figure 1 is a comparison of a LFP powder in pristine condition (Fig.1a, b) and with a carbon layer at the particles surface (Fig.1c, d) where the visibility of the carbon layer is obvious when considering the loss of contrast from the facets of the particles in Figure 1c, d. Micrographs and an x-ray map obtained at low voltage in deceleration mode as well as in STEM mode from a Li2CoSiO4 orthosilicate are shown in Figure 2. The high spatial resolution achieved in these micrographs is striking and demonstrate the high capability of this type of new CFESEMs to characterize and assist the development of new potential nanomaterials for LIB. References: [1] K. Zaghib et al, Positive electrode: lithium iron phosphate, In Encyclopedia of electrochemical power sources (2009), p. 264-296: Elsevier. [2] K. Zaghib et al, Journal of Power Sources, 160 (2006), pp. 1381-1386. [3] M. Guinel et al, Journal of Microscopy, 252 (2013), pp. 49-57. [4] N. Brodusch, H. Demers, and R. Gauvin, Journal of Microscopy, 250 (2013), pp. 1-14. [5] N. Brodusch et al, Surface and Interface Analysis, 46 (2014), pp. 1286-1290. Microsc. Microanal. 21 (Suppl 3), 2015 678 Figure 1. Secondary electron micrographs of a pristine LPF powder (a, b) and the same LPF powder with a surface carbon layer (c, d) recorded at E0 = 2 kV with a stage bias of 10 V (landing energy of 1.9 kV). (a, c) Upper in-lens detector, (b, d) top in-lens detector without energy filtration. Figure 2. (a, b) Secondary electron micrographs with landing voltage of 1 kV in deceleration mode with the upper detector, (c, d) bright-field STEM micrographs at E0 = 30 kV and (e, f) bright-field STEM micrograph and its corresponding EDS-x-ray map showing the Li2CoSiO4 nanoparticles in pink and the CoO by-product particles in yellow. All micrographs were from a Li2CoSiO4 nanopowder used as a cathode material for LIB.");sQ1[340]=new Array("../7337/0679.pdf","Low Accelerating Voltage X-ray Microanalysis � Strategies and challenges.","","679 doi:10.1017/S1431927615004195 Paper No. 0340 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Low Accelerating Voltage X-ray Microanalysis � Strategies and challenges. Peter McSwiggen1 1 McSwiggen & Associates / JEOL USA, Inc., St. Anthony, MN, U.S.A. Low accelerating voltage analysis has become an important strategy in microanalysis because it allows us to push the traditional limitations when attempting to obtain the smallest analytical volumes. Reducing the accelerating voltage reduces the penetrating distance of the beam electrons and thereby reduces the volume being analyzed. Two general strategies have been tried for obtaining the smallest analytical volumes [1]. One approach is to simply use the lowest, reasonable, accelerating voltage that will produce the smallest electron interaction volume. The X-rays will come only from within this volume, and therefore a small analytical volume will be achieved. The optimum accelerating voltage that will produce the smallest interaction volume and still produce a reasonable X-ray count rate, tends to fall in the 5-8 kV range. A lower kV will reduce the penetrating depth of the electrons further. However with lower kV, the diameter of the electron beam itself becomes the critical factor in determining the analytical area, and with a lower kV, the beam diameter becomes larger. Much of this increase in beam diameter comes from the higher required beam current used due to the lower X-ray production rate at the lower kV. Using a lower accelerating voltage also typically requires that a different set of X-ray lines be used in the analysis. Some of the more commonly used X-rays lines are not generated at the low accelerating voltages. A second strategy involves first determining the preferred X-ray lines to use, and then select a accelerating voltage that is only 1-3 kV greater than the critical ionization energy of the highest energy X-ray line being used. This is called the low-overvoltage method [2]. The beam electrons will typically have a higher energy than the previous method, but the fact that they have such a low overvoltage means that once they have entered the sample they quickly drop below the critical ionization energy. Therefore they can only produce X-rays very close to the surface (Fig. 1), even though the beam electrons continue much deeper into the sample. Both methods have their advantages and disadvantages, when comparing (a) the analytical volume, (b) minimum detection limits, (c) secondary fluorescence, and (d) the effect of surface coatings and contamination. With care, it is possible to analyze, quantitatively, features that are in the 200 nm range [3]. However, the critical question becomes, how good are the quality of the analyses using the two different methods. One of the big problems is the transition metals, particularly, when switching from using the K line to the L line for elements like Fe, Ni, Co, and Mn. The quality of the analyses can drop precipitously when using the L lines over the K lines. Using Fe as an example; the problem with using the L line in the analysis is that self-absorption of the Fe L lines is very high, and there is an Fe absorption edge that falls between the Fe L and L lines (Fig.2). Therefore the ratio of Fe L / Fe L varies with the Fe abundance. This becomes a significant problem when these two lines cannot be separated due the low energy resolution of the spectrometer. This is the case with an energy dispersive spectrometer (EDS), and even with a wavelength dispersive spectrometer when using a layered synthetic microstructure (LDE) for the analyzing crystal. Applying the correct mass absorption coefficient becomes very problematic. An alternative is to use a wavelength dispersive spectrometer (WDS) with a TAP crystal. The two X-ray lines are separate and can be measured independently. However the count rates are much lower than on the layered synthetic microstructure (LDE) analyzing crystal. Microsc. Microanal. 21 (Suppl 3), 2015 680 There are various strategies that can be used that will help improve the quality of transition metal analyses, even using the L lines when the L and L cannot be resolved. One option is to use standards that are more similar in composition to the unknown. This reduces the reliance on the matrix corrections routines and thereby reduces the impact of any error in the absorption calculations. If better standards are not available, then a second option might be to use internal standards. Internal standards are phases in the same sample that are large enough to be analyzed under the normal accelerating voltages. Once the larger grains have been analyzed, then can then be used as standards using the low-kV conditions. These internal standards have the potential advantage of being similar in composition to the unknown, and have the added advantage of being identical in their conductive coating, something that can have an important impact when doing low-kV analyses. In order to quantitatively analyze sub-micron grains, a lower than normal accelerating voltage must be used, but working at a lower accelerating voltage can offers many challenges. However, good analyses can be obtained, if one understands where the pitfalls exist and work to minimize their effects. References: [1] P McSwiggen, IOP Conf. Series: Materials Science and Engineering 55 (2014). [2] JT Armstrong, AGU Fall Meeting, San Francisco, Calif., (2011) abstract V31C-2538. [3] JT Armstrong et al, Microscopy and Analysis 27(7) (2013) p. 20-24. Figure 1. Monte Carlo simulation of the analytical volume for a Ni sample using the low-kV method at 7kV using Ni L (a), and the low-overvoltage method at 10kV using Ni K (b). Figure 2. Differences in the amount of absorption of various Fe X-ray lines in an iron absorber.");sQ1[341]=new Array("../7337/0681.pdf","Advantages of Beam Deceleration for Low kV EDS Analysis.","","681 doi:10.1017/S1431927615004201 Paper No. 0341 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Advantages of Beam Deceleration for Low kV EDS Analysis. D. Edwards1, N. Erdman1, D. Guarrera1, M. Shibata1, V. Robertson1, F. Timischl2, Y. Nemoto2 1. 2. JEOL, USA Inc., 11 Dearborn Rd, Peabody, MA 01960 JEOL Technics Ltd, Akishima, Japan Low voltage imaging and microanalysis have been gaining prominence in the last few years due to their distinct advantages for analysis of beam sensitive and charging materials. Recent advances in the SEM electron optics have allowed the SEM to maintain a small probe size at both imaging (small beam current) and analysis (large probe current) conditions at low accelerating voltages [1-3]. In particular in JEOL SEM microscopes one of the methods we have been employing to achieve this goal has been the use of beam deceleration (referred to as GB mode). Beam deceleration not only helps with charge balance but also effectively reduces lens aberrations thus improving overall image resolution. Gentle Beam works by retarding the primary beam voltage using a negatively charged stage bias to a lower landing energy. The landing voltage (Elanding=Egun-Ebias) can be varied with a combination of electron source voltage and specimen bias to achieve the necessary charge balance as well as high resolution performance at ultra-low voltages. The typical values for specimen bias are 0-2kV, though better performance can be obtained with bias values up to 5kV. Beam deceleration also serves as a form of aberration correction; the aberration coefficients (both spherical and chromatic) are reduced when the ratio Elanding/Egun is reduced for a fixed Egun, meaning larger specimen bias enhances image resolution at ultra-low kVs. The use of Gentle Beam function thus preserves all the advantages of high kV imaging (gun brightness, small probe size) with added advantages of reduced charging, reduced specimen contamination and improved surface detail. In this paper the benefits of beam deceleration at the specimen, as it applies to EDS will be demonstrated on the next generation Field Emission and Tungsten filament SEMs. Our extensive experience with imaging using Gentle Beam suggested that it would also be beneficial for elemental analysis of beam sensitive specimens that can't survive extensive counting times due to potential heat damage or specimens that require low voltage microanalysis because of charging. Figure 1 illustrates the improved elemental mapping that can be obtained by using a specimen bias. We have analyzed a sample of slag at 3kV and 90nA showing EDS mapping without application of GB (a,c,e) and with a -5kV bias (b,d,f). The maps demonstrate both enhancements in spatial resolution as well as signal to noise ratio when employing specimen bias. We will further demonstrate application of this technique for composite samples, graphene, dopants in battery materials and nanoparticles on ITO. Our data will show that utilizing this technique the user could minimize acquisition time during EDS mapping while still maintaining good signal to noise ratio and high spatial resolution, making analysis of non-conductive and/or beam sensitive specimens more accessible. References: [1] [2] D.C. Bell and N. Erdman, in Low Voltage Electron Microscopy: Principles and Applications (2012): 1-30. [2] N. Rowlands et al. Microscopy and Microanalysis 15.S2 (2009): 548-549. [3] D. Edwards et al. Microscopy and Microanalysis 20.S3 (2014): 646-647. Microsc. Microanal. 21 (Suppl 3), 2015 682 a b c d e f Figure 1. EDS map of Slag at 3kV and 90nA: (a,c,e) No Gentle Beam; (b,d,f) -5kV Gentle Beam.");sQ1[342]=new Array("../7337/0683.pdf","Nanostructures Viewed through Low Voltage Electron Microscopy","","683 doi:10.1017/S1431927615004213 Paper No. 0342 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Nanostructures Viewed through Low Voltage Electron Microscopy Daniela R. Radu* and Cheng-Yu Lai, Delaware State University, Dover, DE, USA (dradur@desu.edu) Materials that present nanostructural features, either internally--as in nanoporous structures, or externally--as in nanoparticles with various shapes and aspect ratios, pose a challenge in visualizing their nanoscale features via low-voltage microscopy. The LV EM 5, the only benchtop TEM in the market, offers a variety of tools that overcome the typical low voltage drawbacks. These enhancements will be presented in the context of two projects. The first part of the presentation will discuss the use of LV EM 5 in SEM-mode in a project targeting the application of mesoporous silica nanomaterials in carbon capture, with emphasis on the particles analysis. In the second part, the TEM and SEM modes will be presented in the context of a solar research project, where the solar device is fabricated solely through solution processing and both nanoparticle precursors and films are subjected to microscopy investigation. Mesoporous Silica. The ability of amine groups to capture CO2 through a simple chemical reaction, prompted the idea of combining the CO2 chemical absorption capability of aminederivatized surfaces with the adsorption capability of a unique folded nanosheet structure and ultra-large surface area of a novel porous silica platform. From economic perspective, the synthetic path of this novel platform involves a low-cost, scalable process, involving inexpensive starting materials and room temperature synthesis. The synthesis of mesoporous silica nanostructures (MSN) involves a simple condensation reaction of a silica source around a template (surfactant, block copolymer etc.). Various surfactants and pore enhancing molecules have been explored for changing pore size and pore accessibility, while maintaining spherical morphology of the particles. Upon synthesis, the template is removed by calcination or extraction in an acidic solution for cationic surfactants or alcoholic solution for non-ionic surfactants. A survey of various synthetic methodologies has lead to obtaining a unique nanostructure, nanosheet-like silica nanosphere (NSN). Nanoprecursor-Based Solar Cells. Chalcogenide semiconductors, such as copper indium gallium diselenide � Cu(In,Ga)Se2 (CIGS), kesterites � Cu2ZnSn(S,Se)4, (CZTS), and recently proposed iron chalcogenide, Fe2SiS4 and Fe2GeS4, offer bandgaps close to ideal for the absorber material in thin-film photovoltaic (PV) technologies. Progress is continuously made in this arena both in terms of small cell efficiencies and in terms of larger area module efficiencies. Besides the cost aspect on Earth applications, chalcogenide solar cells are also interesting for space applications because of their excellent stability against particle irradiation and development of high efficiency flexible cells. Roll-to-roll printed nanoparticle technologies for flexible solar cells could offer the combination of fast, atmospheric pressure deposition, a lightweight substrate, and a thin, inexpensive absorber layer, each of which decreases the cost and the weight of the final solar cell or module, toward making solar energy affordable. Nanoparticles and nanoparticles-originated polycrystalline layers of Cu2ZnSn(S,Se)4, (CZTS), and Fe2SiS4 and Fe2GeS4 from nanoparticles synthesis to the solar cell performance will be reviewed.1-3 NSN Materials. The synthetic routes reported for porous silica nanospheres obtained with the co-solvent method have employed energy intensive reaction conditions and used semi-scalable Microsc. Microanal. 21 (Suppl 3), 2015 684 methods such as microwave-assisted synthesis.4 The materials synthesized with our conditions require room temperature, in air. Characterization of materials via nitrogen porosity indicates a very large surface area and pore volume (Table 1). Table 1. BET Surface area of synthesized materials Material NSN-1 NSN-2 NSN-3 2 1070 1240 1398 BET Surface area (m /g) Ethyl ether Ethyl ether� Ethyl ether Co-solvent(s) ethanol A comparison of images obtained via SEM and LV SEM show the capability of LV EM 5 to visualize nano-features close in nm range. Nanoprecursor-Based Solar Cells. Nanoparticles structure and size as well as surface properties along with nanoparticle coatings and film thermal treatment, all play an immense role in making a continuous, crack-free, crystalline absorber layer. The instrument facile operation enables analysis of a large amount of samples on daily-basis, which is beneficial for synthetic optimization and fast advancement of research. Furthermore, identification of nanoparticle composition when particle size is below 100 nm becomes cumbersome with XRD due to line broadening in XRD. Investigation of NPs via TEM imaging and electron diffraction patterns highlight the potential of LV microscopy in such research. The LV EM 5 instrument in SEM mode enables early identification of both physical (impurity) and mechanical (cracks) defects in sample surface and cross section to identify showstoppers before transferring the film to the next step in the device fabrication flow. Both projects benefit from the easy access of the LV EM 5. In contrast to high-voltage SEM and TEM counterparts, the LV enables high-throughput analysis. While not being able to substitute for high-voltage microscopy, the bench top LV EM 5 is a versatile, inexpensive and easy to operate system. The instrument represents an excellent educational tool that could be solely operated by students/researchers without the need of a specialized operator. Acknowledgements. The authors thank DOD (Award W911NF-14-1-0071) for funding support in acquiring the LV EM 5. References 1. Cao, Y.; Denny, M. S.; Caspar, J. V.; Farneth, W. E.; Guo, Q.; Ionkin, A. S.; Johnson, L. K.; Lu, M.; Malajovich, I.; Radu, D.; Rosenfeld, H. D.; Choudhury, K. R.; Wu, W., HighEfficiency Solution-Processed Cu2ZnSn(S,Se)4 Thin-Film Solar Cells Prepared from Binary and Ternary Nanoparticles. Journal of the American Chemical Society 2012, 134, 15644-15647. 2. Radu, D. R.; Caspar, J. V.; Johnson, L. K.; Rosenfeld, H. D.; Malajovich, I.; Lu, M. Copper zinc tin chalcogenide nanoparticles. WO2010135622A1, 2010. 3. Radu, D. R.; Caspar, J. V.; Lu, M.; Johnson, L. K.; Cao, Y.; Ionkin, A. S.; Malajovich, I.; Rosenfeld, H. D.; Sun, F.; Tassi, N. G., CZTS absorber layer in thin film p-n junction solar devices from quaternary nanoparticle precursors. Prepr. Symp. - Am. Chem. Soc., Div. Fuel Chem. 2011, 56, 179.");sQ1[343]=new Array("../7337/0685.pdf","Nanoscale Characterization of Li-ion Battery Cathode Nanoparticles by Atom Probe Tomography Correlated with Transmission Electron Microscopy and Scanning Transmission X-Ray Microscopy","","685 doi:10.1017/S1431927615004225 Paper No. 0343 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Nanoscale Characterization of Li-ion Battery Cathode Nanoparticles by Atom Probe Tomography Correlated with Transmission Electron Microscopy and Scanning Transmission X-Ray Microscopy A. Devaraj1, C. Szymanski1, P. Yan1, C.M. Wang1, V. Murgesan2, J. M. Zheng2, J. Zhang2, T. Tyliszczak3, S. Thevuthasan4 [1] Environmental and Molecular Sciences Laboratory, Pacific Northwest National Laboratory, Richland WA, 99354 [2] Energy and Environmental Directorate, Pacific Northwest National Laboratory, Richland WA 99354 [3] Advanced Light Source, Lawrence Berkeley National Laboratory, Berkeley CA, 94720 [4] Qatar Energy and Environmental Research Institute, Qatar, UAE Li-ion batteries are currently powering innumerable consumable electronic devices and are also considered as a candidate for energy storage solution for long range electric vehicles [1]. There is an ongoing effort to engineer Li-ion battery electrodes that can perform with even higher energy storage capacity along with longer term energy storage performance. Specifically next generation cathode material development is of significant focus, because cathodes amount to significant volume of the battery and share a major portion of the overall cost of the battery [1, 2]. In order to develop cathodes that can repeatedly store charge at high capacities for many cycles, a thorough understanding of their starting structure, composition and chemical state as well as the changes of the same as a function of electrochemical cycling is needed. This necessitate a correlative microscopy approach of integrating three powerful microscopy techniques capable of probing composition, structure and chemical state at nanoscale, namely atom probe tomography (APT), transmission electron microscopy (TEM) and scanning transmission x-ray microscopy (STXM) respectively. Nanoparticle morphology of the Li-ion battery cathode materials necessitate specific modifications for the sample preparation and analysis procedures for APT as discussed below. Using this correlative microscopy approach a variety of Li-ion battery cathodes based on LiMO2 (M=Mn, Ni, Co) as well as layered and spinel oxides with multiple metallic elements were investigated both before they are cycled as well as after electrochemical cycling to obtain atomic scale insight towards mechanism of capacity degradation of the Li-ion battery cathode materials as a function of electrochemical cycling. The details of specimen preparation, APT data analysis and APT results from LiMnO2 cathode nanoparticles are discussed here. For conducting APT analysis of the cathode materials, the nanoparticles of LiMnO2 dispersed on a Silicon wafer was transferred on to the Silicon microtip array using the omniprobe nanomanipulator lift-out procedure, by just using electron beam imaging and electron beam assisted Pt deposition (Figure 1(a)). Subsequently needle shaped specimens were fabricated by annular milling in a Helios Nanolab FIB system as shown in figure 1(b). The needle specimens were analyzed in a CAMECA LEAP4000XHR laser assisted APT system using 355nm wavelength UV laser at 100 KHz pulse repetition rate and 40K specimen temperature. The mass-to-charge spectra consisted of both elemental peaks of Li, Mn and O ions along with many molecular ions (figure 1c). The APT reconstruction revealed nanoscale distribution of Li, Mn and O (figure 1d). These APT results were correlated with aberration corrected STEM imaging and diffraction of the same cathode nanoparticles aiding in understanding atomic scale structure. STXM imaging and NEXAFS for metal L edges (Ni, Mn, & Co L edges) and O K edge were performed at Advanced Light Source beamlines 5.3.2.2 and 11.0.2 to understand the chemical state of elements. The NEXAFS spectra obtained on the Microsc. Microanal. 21 (Suppl 3), 2015 686 References [1] Goodenough, J. B.; Kim, Y., Challenges for Rechargeable Li Batteries. Chem Mater 2010, 22 (3), 587-603. [2] Ellis, B. L.; Lee, K. T.; Nazar, L. F., Positive Electrode Materials for Li-Ion and Li-Batteries. Chem Mater 2010, 22 (3), 691-714. [3] A portion of the research was performed using EMSL, a national scientific user facility sponsored by the DOE-BER and located at PNNL. The characterization work was funded by laboratory directed research and development funding as a part of chemical imaging initiative in PNNL. [4] The Advanced Light Source is supported by the Director, Office of Science, Office of Basic Energy Sciences, of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231. cathode materials were compared with NEXAFS spectra from standard oxide specimens to understand the exact oxidation state of individual elements. By correlating the APT reconstructions with STEM images, STXM images and NEXAFS spectra a comprehensive understanding of structure, composition and chemical state of these Li-ion battery cathodes are obtained. Figure 1: (a) Series of scanning electron microscopy images showing nanomanipulation of LiMnO2 nanoparticles on to a Si microtip post followed by Pt capping using electron beam assisted Pt deposition (b) final needle specimen after annular milling in FIB. (c) Mass to charge spectra of LiMnO2 obtained by laser assisted atom probe tomography indicating the different elemental and molecular ions observed (d) APT Reconstructions of LiMnO2 showing distribution of Li, Mn and O.");sQ1[344]=new Array("../7337/0687.pdf","Integrated APT/t-EBSD for Grain Boundary Analysis of Thermally Grown Oxide on a Ni-Based Superalloy","","687 doi:10.1017/S1431927615004237 Paper No. 0344 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Integrated APT/t-EBSD for Grain Boundary Analysis of Thermally Grown Oxide on a Ni-Based Superalloy Y. Chen1, K.P. Rice1, T.J. Prosa1, E.A. Marquis2 and R.C. Reed3 1. 2. CAMECA Instruments, Inc., Madison, WI, USA Dept. of Materials Science and Engineering, University of Michigan, Ann Arbor, MI, USA 3. Dept. of Engineering Science, University of Oxford, Parks Road, Oxford, OX1 3PG, UK Thermal barrier coatings (TBCs) allow turbine engines to be operated at temperatures greater than the melting temperatures of engine components to pursue better propulsive power performance and fuel efficiency [1]. TBCs generally consist of three layers. As illustrated in Figure 1, on the top is a coat made of yttrium-stabilized ZrO2 (or YSZ) which has excellent thermal resistivity. Beneath the YSZ layer is a thermally grown oxide (TGO) scale that consists of -alumina grains. At the coating/substrate interface is a bond coat layer that improves adhesion of the ceramic layers on the superalloy substrate. The TGO layer typically forms as a thin layer during deposition of the YSZ layer and grows during service. The growth of the TGO introduces additional interfacial stresses to the coating that can eventually lead to delamination of the TBC structure. A longer TBC life-time is achieved by slowing down the growth of the TGO layer. In the case of -alumina, the inward diffusion of oxygen along grain boundaries is dominant mechanism for the oxide growth. Alloying with reactive elements (REs), such as Y, Hf, La, has been shown to significantly reduce the growth rate of TGO, and while these elements have been found to segregate to grain boundaries, the exact phenomenon by which oxygen transport is affected is not clear. Atom probe tomography (APT) was previously used to quantify grain boundary chemistry in TGO layers. Significant chemical variations between different grain boundaries were reported, suggesting a strong effect of grain boundary character [2]. Inspired by the ability of mapping needle shaped specimens using electron backscatter diffraction (EBSD) in transmission mode, this work aims to further investigate the relationships between grain boundary chemistry and grain boundary misorientation using the integrated APT/t-EBSD technique [3]. Linking orientation to chemistry will inform on the possible pathways to controlling alumina growth. Mo grids were cut in half and electropolished for mounting specimens. Needle-shaped APT specimens were prepared by a standard lift-out method and annular milling using an FEI Nova FIB/SEM. The integrated EXAX EBSD system allows for in-situ mapping feedback between milling processes. Figure 2 is a typical t-EBSD pattern from an -alumina grain in TGO collected at a tilt angle of 155 degrees using 30 kV electrons. Figure 3a is a t-EBSD map showing multiple grains in an APT specimen for targeting grain boundaries. Two separate APT analyses were performed in region A and B in Figure 3a. In region A, grain boundaries segregated with Zr are visible and the grain misorientation of the two alumina grains on the bottom were subsequently measured using t-EBSD (Figure 3c) by terminating the APT analysis before tip fracture. The same tip was further FIB milled into region B and mapped with tEBSD using a step size of 10 nm (Figure 3d). Two grain boundaries were successfully captured by APT (Figure 3e) and their structures and chemistries can be studies accordingly. In this study we have demonstrated that high-resolution t-EBSD maps can be acquired on needle-shaped APT specimens that consist of alumina grains of size ranging from few hundred nanometers to few micrometers. Transmission EBSD mapping offers the ability to target site-specific grains for APT Microsc. Microanal. 21 (Suppl 3), 2015 688 analysis, and correlate grain boundary chemistries with grain misorientations. In addition, we succeeded in collecting multiple APT datasets and t-EBSD maps from a single APT specimen, thus the evolution of structure across TGO scales can be investigated. References: [1] A. Evans et al., Prog Mater Sci 46 (2001), p. 505. [2] Y Chen et al., Oxidation of Metals 82 (2014), p. 457. [3] K. Babinsky et al., Ultramicroscopy 144 (2014), p. 9. Figure 1. SEM image of TBC with a 3 m thick TGO scale on a Ni-based superalloy Figure 2. A typical EBSD pattern of -alumina collected with 30 kV electrons Figure 3. t-EBSD maps of alumina grains and correlative APT results from region A and B in the TGO scale (see text for full description)");sQ1[345]=new Array("../7337/0689.pdf","A Comparative Analysis of a Si/SiGe Heterojunction-Bipolar Transistors: APT, STEM-EDX and ToF-SIMS","","689 doi:10.1017/S1431927615004249 Paper No. 0345 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A Comparative Analysis of a Si/SiGe Heterojunction-Bipolar Transistors: APT, STEM-EDX and ToF-SIMS Robert Estivill1,2,3, Pascal Chevalier1, Frederic Lorut1, Marc Juhel1, Laurent Cl�ment1, Germain Servanton1, Gregory Avenier1, Adeline Grenier2, Didier Blavette3 1. 2. STMicroelectronics, 850 rue Jean Monnet, 38926 Crolles, France Univ. Grenoble Alpes, F-38000 Grenoble, France CEA, LETI, MINATEC Campus, F-38054 Grenoble, France 3. Groupe de Physique des Mat�riaux � GPM UMR CNRS 6634, Universit� de Rouen, France Due to the complexity of characterising compound semiconductors, including dopant distribution, multiple characterisation techniques are needed. Traditionally time-of-flight secondary ion mass spectroscopy (SIMS) has been the tool of choice for chemical profiling of semiconductor systems. Although it affords a lower limit of detection, it is constrained by a low lateral resolution, making large test zones necessary (several hundred microns). More recently, energy dispersive X-ray - scanning transmission electron microscopy (STEM-EDX) allows local specimen preparation and can generate 2D concentration maps. But due to low sensitivity it cannot quantify light elements (i.e. boron). Because of size effects, large test zones are not always representative of the local chemistry in the device and a complete picture is therefore unavailable. Atom probe tomography (APT) is an analytical 3D microscopy technique which maps the position of atoms in a material allowing composition measurements of a small selected volume. With a sub-nanometre spatial resolution, analysis of localised structures is possible and all elements are detected with the same probability. Initially dedicated to metals, semiconductor applications have escalated in recent years [1]. In this work a Si/SiGe heterojunction bipolar transistor for 55nm BiCMOS (bipolar � complementary metal oxide semiconductor) technology development has been studied. The basic structure is made up of a silicon emitter, silicon-germanium base and silicon collector [Fig. 1(a)]. The base follows a germanium concentration gradient. Both emitter and collector are arsenic doped, the base containing boron. The stack was submitted to a spike anneal and preparation halted at the silicide step. The wafer was then analysed using the three techniques cited above. APT and EDX measurements specimens were prepared by focused ion beam (FIB). For the APT, backside preparation was privileged, due to a low analysis yield when using standard preparation geometry. In the case of APT and STEM-EDX, measurements were made on both the test zone [Fig. 1(b)] and the final device [Fig.1(c)], whereas ToF-SIMS measurements have been made only in the test zones. By using APT is has been possible to validate the information from the other techniques. Moreover it gives the 3D distribution of boron atoms in the device. Due to the feeble concentration (5�10E19 at/cm3) in the silicon-germanium base, although the boron signals are visible, the dopant distribution is not clear due to back-ground noise. To increase the signal-to-noise ratio, multiple event analysis of the APT data has been used [2]. Previously used for carbon detection, here we show it is also possible to apply this method to boron [Fig.2]. Once the data has been processed the dopant position in the structure is made clear. A slight shift in the position of the boron peak is noted relative to the test structure. Comparison of the three techniques shows a similar shape. Comparison of STEM-EDX and APT shows a good agreement in the shape of the germanium profile, with a higher Microsc. Microanal. 21 (Suppl 3), 2015 690 concentration and sharper profile in the device noted in both cases [Fig.3(a-b)]. Also, a small segregation of arsenic at the interface between the emitter and the base only in the device is made apparent. This is attributed to an incomplete cleaning of the interface before deposition of the emitter, resulting in a small oxide layer and accumulation of dopant [3] that must be avoided. In conclusion, APT has allowed a direct comparison between test zones and real device. It is the only technique which allows direct detection of boron distribution in the later. Due to the low dopant concentration multiple-event analysis was used to increase the signal-to-noise ratio. Several substantial differences were noted between the test zones and device. References [1] D. Blavette, S. Duguay, P. Pareige, Int J Mater Res 102, 2011, 1074. [2] L. Yao et al., Phil Mag Lett 93, 2013, 299. [3] G. Servanton, R. Pantel, 41, 2010, 118 [4] This study has been performed at the nanocharacterisation platform (PFNC) of the Minatec Campus and ST Microelectonics, Crolles. The author would like to acknowledge a CIFRE (ANRT) scholarship a) b) c) Figure1. Schematic representation of the gate stack analysed (a) and TEM images of both test zone (b) and device (c). Number of impacts a) b) Figure2. Normalised mass-to-charge spectrum with all impacts (a) and only multiples (b). a) b) c) Figure3. Comparison of germanium (dashed lines) and boron (solid) in device (black) and test zones (red) by APT (a), STEM-EDX (b) and ToF-SIMS (c).");sQ1[346]=new Array("../7337/0691.pdf","CdSe1-xTex Phase Segregation in CdSe/CdTe Based Solar Cells","","691 doi:10.1017/S1431927615004250 Paper No. 0346 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 CdSe1-xTex Phase Segregation in CdSe/CdTe Based Solar Cells Jonathan D. Poplawsky,1 Naba R. Paudel,2 Amy Ng,3 Karren More,1 and Yanfa Yan2 1. 2. Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, TN, USA. Department of Physics and Astronomy, The University of Toledo, Toledo, OH, USA. 3. Department of Chemistry, Vanderbilt University, Nashville, TN, USA. Thin film polycrystalline CdTe-based solar cells are promising devices for solar applications due to their low production costs and high theoretical efficiency (~30%). However, the highest thin film CdTe laboratory cell efficiency is 21.5%, and the best CdTe module efficiency is 17.5%, which are well below the theoretical efficiency [1]. CdTe solar cell devices typically utilize a p-type CdTe absorber layer junctioned with a thin (~50-100 nm) n-type CdS window layer (~2.4 eV band gap) [2]. A photoactive CdTexS1-x alloy layer forms at the interface region, which could enhance the photocurrent in the long wavelength regime due to bowing effects [3]. However, S diffusion into CdTe is limited due to the large lattice mismatch between CdS and CdTe, and there is not enough S diffusion to be effective in increasing the photocurrent [4]. On the other hand, Se can easily diffuse into the CdTe layer due to the relatively close lattice constants of CdTe and CdSe. Therefore, CdTe solar cell devices have been fabricated with a CdSe (~1.7 eV band gap) window layer to enhance the short-circuit current (JSC) by increasing the photocurrent for long wavelength photons. To optimize the device, CdTe has been grown on 50, 100, 200, and 400 nm CdSe layers. External quantum efficiency (EQE) data for samples grown with CdSe layers between 50 and 200 nm show improved photocurrent in both the short and long wavelength regimes compared to CdS/CdTe devices, which drastically improves the JSC [4]. However, devices with a CdSe layer greater than 400 nm have a significantly reduced photocurrent for short wavelength photons, reducing the JSC. To better understand the role of the CdSe layer and Se diffusion on device performance, a series of atom probe tomography (APT), SEM-based electron beam induced current (EBIC), and STEM EDS measurements were performed on these devices. The EBIC measurements, shown in Fig. 1, were performed using a 3 kV, ~20 pA electron beam in a Hitachi S4800 CFEG SEM equipped with a Gatan EBIC system, and include EBIC images for devices with a 100 and 400 nm thick CdSe layer. There is a clear, strong EBIC signal throughout the entire CdTe layer for both samples. However, on closer inspection of the thin conducting oxide (TCO)/CdSe/CdTe interface, there are clear differences between the EBIC and SEM images for these samples. Each layer can be distinguished in the secondary electron (SE) image for the 400 nm thick CdSe sample, while there is no evidence for the presence of a CdSe layer in the SE image for the 100 nm thick CdSe sample. Despite not having a clear CdSe junction layer, the 100 nm sample shows a peak EBIC current at the TCO interface, while the 400 nm sample shows little to no EBIC signal at the TCO interface. To better understand the CdSe/CdTe interface, APT needles were prepared at the TCO/CdSe/CdTe interface region. The reconstructed APT data for devices with 400 and 100 nm thick CdSe layers are shown in Fig. 2a and 2b, respectively. These APT data show that Te and Se atoms easily diffuse between the CdSe and CdTe layers to form a CdSe1-xTex alloy interface layer with x changing with respect to distance away from the interface. There are obvious Te- and Se-rich phases within the originally deposited CdSe layer, with an overall composition of 25% Te and 25% Se for the 100 nm CdSe layer. However, the 400 nm thick CdSe layer device shows a distinct interface between the CdTe and CdSe layers, with a 35% Serich phase. A comparison of the APT data with the EBIC images suggests that the large-grained Microsc. Microanal. 21 (Suppl 3), 2015 692 CdSe1-xTex crystals with x < ~0.3 are non-photoactive when coupled to the CdTe. STEM EDS data of these materials will also be presented. This work has shown that engineering devices with a CdSe1-xTex can produce a photoactive junction partner with p-type CdTe that can improve the JSC. [1] [2] [3] [4] [5] MA Green, et al, Progress in Photovoltaics: Research and Applications 23 (2014) p. 1�9. HH Abu-Safe, et al, Journal of Electronic Materials 33 (2004) p. 128�134. C Li, et al, IEEE Journal of Photovoltaics 4 (2014) p. 1636�1643. NR Paudel and Y Yan, Applied Physics Letters 105 (2014) 183510. Research supported by ORNL's Center for Nanophase Materials Sciences (CNMS), which is a U.S. Department of Energy, Office of Science User Facility. a) b) c) d) e) f) 166 124 83 102 77 52 26 1 0.5 �m 42 1 Figure 1. EBIC and SE images for CdTe devices grown with a 100 nm (a, c, and d) and 400 nm (b, e, and f) thick CdSe layer. The scale bar for (a) and (b) is 5 �m. Figures c-f are the same magnification. Contrast in EBIC images defined by the EBIC-current/probe-current. Figure 2. APT reconstructions of CdTe devices with (a) 400 nm and (b) 100 nm CdSe layers. The corresponding z-axis, 1-dimensional line profiles are shown to the right of the reconstructed needles. The black line in (b) shows the approximate CdSe/CdTe interface before Se diffusion.");sQ1[347]=new Array("../7337/0693.pdf","Dopant and Interfacial Analysis of Epitaxial CdTe Using Atom Probe Tomography","","693 doi:10.1017/S1431927615004262 Paper No. 0347 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Dopant and Interfacial Analysis of Epitaxial CdTe Using Atom Probe Tomography George L. Burton1, David R. Diercks1 and Brian P. Gorman1 1. Metallurgical and Materials Engineering, Colorado School of Mines, Golden, CO. Cadmium Telluride (CdTe) is a II-VI semiconductor with a direct band gap of 1.45 eV at room temperature. It has a high absorption coefficient and can be produced easily from raw materials, making it an ideal material for photovoltaic (PV) devices. Thin film polycrystalline CdTe PV devices have shown exceptional promise, reaching efficiencies of 20% [1], however much research must be accomplished to approach the theoretical efficiency of ~30%. The study of polycrystalline CdTe is hindered primarily by grain boundaries (GBs), which give rise to high recombination rates, current leakage, device shunting, and enhanced migration of dopants [2]. To overcome the complexity and problems associated with GBs, CdTe can be grown in a single crystal form with meticulous control over chemical purity and dopant concentration through molecular beam epitaxy (MBE). Identifying the concentration, uniformity, and location of impurity species, whether added intentionally (i.e. dopants) or unintentionally (i.e. unwanted contaminants), can help to optimize the MBE growth parameters for high efficiency devices. With its unparalleled three-dimensional atomic spatial resolution, atom probe tomography (APT) is particularly suited for determining these impurity properties. Laser-pulsed APT in conjunction with correlative transmission electron microscopy (TEM) was conducted on a number of epitaxial CdTe thin films. TEM images of specimens taken prior to and following APT were used to produce an accurate analysis volume for APT reconstructions [3]. A variety of dopant species were analyzed, including arsenic as shown in Figure 1. The APT reconstruction shows clustering of As, with local concentrations sometimes exceeding 40 at. %, throughout the sample. Additionally, the concentration profile suggests that As is unexpectedly substituting for Cd rather than Te, which implies n-type doping rather than p-type. In Figure 2, a cadmium magnesium telluride (CdMgTe) heterojunction layer was analyzed. In addition to determining the uniformity and concentration of magnesium within the layer, APT analysis revealed a small amount of oxygen contamination. Interfacial impurities such as these can have a particularly detrimental effect on defect densities and carrier lifetimes. The relationship of the growth conditions to the structural and compositional TEM and APT findings and their impacts on material performance will be discussed. References: [1] Martin A. Green et al, Prog. Photovolt.: Res. Appl., 22 (2014), p. 701. [2] S. Girish Kumar and K.S.R. Koteswara Rao, Energy Environ. Sci. 7 (2014), p. 45. [3] B. P. Gorman et al., Microscopy Today, 16 (2008), p. 42. [4] The authors acknowledge funding from the Department of Energy's SunShot Foundational Program to Advance Cell Efficiency (F-PACE II). Microsc. Microanal. 21 (Suppl 3), 2015 694 (a) Concentration (at. %) 50 40 30 20 10 4.0 8.0 Te Cd As 12.0 (b) Distance (nm) Figure 1. (a) APT reconstruction of As-doped CdTe, using 15 at. % As isoconcentration surfaces (b) concentration profile through one As cluster whose region is defined by the grey box in the reconstruction in part (a). 16.0 (a) (b) Concentration (at. %) Te 1e1 1e0 1e-1 1e-2 40 80 120 (c) Mg O Cd Distance (nm) Figure 2. (a) Transmission electron micrograph of epitaxially grown CdTe with a ~21 nm CdMgTe heterojunction. (b) APT reconstruction of same sample using Mg isoconcentration surface. The green points are Te (c) Concentration of Cd, Te, Mg, and O across the CdMgTe heterojunction layer.");sQ1[348]=new Array("../7337/0695.pdf","In-Situ Deuterium Charging for Direct Detection of Hydrogen in Vanadium by Atom Probe Tomography","","695 doi:10.1017/S1431927615004274 Paper No. 0348 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In-Situ Deuterium Charging for Direct Detection of Hydrogen in Vanadium by Atom Probe Tomography S. Dumpala1, D. Haley2, S.R. Broderick1, P.A.J. Bagot2, M. P. Moody2, and K. Rajan1 1. Department of Materials Science and Engineering and Institute for Combinatorial Discovery, Iowa State University, 2220 Hoover Hall, Iowa State University, Ames, USA. 2. Department of Materials, University of Oxford, Parks Road, Oxford OX13PH, UK The study of hydrogen embrittlement is of great interest for several decades, owing to the large reduction in maximum elongation of hydrogen exposed materials [1]. Towards this end, we will be studying the hydrogen interaction with material microstructure. Even though there exist several competing theoretical models, providing a mechanism by which dislocations and cracks are expedited through microstructures [2], there is a lack of experimental evidence to support differing theoretical models, owing to two main limitations. Firstly, hydrogen is weakly interacting with many radiations (e.g. x-ray, electron), and thus is difficult to image. Secondly, the interactions are truly atomistic, and thus require a scale-matched imaging methodology. Thus there exists a need for real-space quantitative analysis for identification of H, at these nano scales. Such information would allow for greater understanding of the dislocation H interaction. With the high chemical sensitivity of Atom Probe Tomography (APT) along with its 3D visualization capabilities, APT is able to overcome these limitations inherent in other methods. Existing challenges in this, owing to hydrogen's high diffusivity, and low solubility in iron-based materials are to maximize signal when charging with D2 when undertaking controlled in-situ D2 charging experiments [3]. In the present work, the existing environmental chamber, which is integrated with the LEAP [4] with a maximum operating pressure range of 50 milli Torr was modified (Fig 1) to allow for atmospheric pressure (~1 Atm) charging and to be able to use D 2 as a feed gas as higher pressure assists in greater deuterium concentration in the material due to theoretical solubility partitioning. Preliminary in-situ deuterium charging experiments performed on the pre-cleaned vanadium samples clearly indicated the significant detection of D2 (Fig. 2). Thus the existing reaction cell design offers several advantages over ex-situ techniques. Firstly the time requirements for transfer can be reduced, owing to the naturally dry (and thus clean) environment, and the presence of a 3-stage vacuum system. Secondly, the existing heating stage allows for hydrogen-1 (protium) to be expelled from the material prior to deuterium charging. References: [1] H. Sugimoto et al, Acta Materiala, 67 (2014) 418. [2] J. Song et al, Acta Materiala 59 (2011) 1557. [3] D. Haley et al, International Journal of Hydrogen Energy 39 (2014) 12221 [4] S. Dumpala et al, Ultramicroscopy 141 (2014) 16 Microsc. Microanal. 21 (Suppl 3), 2015 696 [5] The authors acknowledge the support from Air Force Office of Scientific Research grants: FA955010-1-0256, FA9550-11-1-0158 and FA9550-12-1-0456; and NSF grants: ARI Program CMMI-09389018 and PHY CDI-09-41576. Figure 1. Environmental chamber modified with additional pressure components. Inset shows the close up of the pressure gauge. Figure 2. Atom probe tomography mass spectra of the vanadium sample after the in-situ D2 charging, clearly indicating the direct detection of deuterium with presence of peaks corresponding to various deuterium compounds.");sQ1[349]=new Array("../7337/0697.pdf","High-Throughput SEM via Multi-Beam SEM: Applications in Materials Science","","697 doi:10.1017/S1431927615004286 Paper No. 0349 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 High-Throughput SEM via Multi-Beam SEM: Applications in Materials Science Joseph R. Michael1, Craig Y. Nakakura1, Tomasz Garbowski2, Anna Lena Eberle2, Thomas Kemen2 and Dirk Zeidler2 1. 2. Sandia National Laboratories, Albuquerque, NM 87185 Carl Zeiss Microscopy GmbH, Oberkochen, Germany The maximum image acquisition rate for conventional SEM is limited by many factors related to both the instrumentation used and the physics of secondary electron image formation. The instrumentation limitations are related to maximum scan rates, detection band width and collection efficiency. The signal-to-noise for a given pixel dwell time is dependent mainly upon the beam current of the primary beam. For a given beam current, shorter pixel dwell time results in a reduction in the signal-to-noise ratio for the image. Thus, the only way to have higher throughput is to continually increase the beam current or improve the collection efficiency of the detectors employed. However, it is not possible to increase the beam current continuously due to the eventual loss of resolution associated with electronelectron interactions contributing to beam broadening and to electron optical considerations that require larger convergence angles and further increase in the beam size. Even if the desired resolution could be achieved with higher beam currents, there are some classes of samples that will be damaged or modified by high current imaging. This paper will introduce the multi-beam SEM and then discuss recent advances in its applications for high throughput imaging of materials science samples. Improvements in imaging rate in the SEM can be achieved by use of multiple-beam SEM where an array of electron beams is scanned across a sample to achieve large area imaging while maintaining the required resolution.[1-3] A schematic of the multiple-beam SEM optics is shown in Figure 1. The multiple-beam SEM produces an array of 61 electron beams from a micro-optic element that is illuminated by a single Schottky electron source. The 61 beams are focused on to the sample surface. Secondary electrons generated by the individual beams are collected with very high efficiency and focused into individual beams of electrons that are normal to the sample surface. The secondary electrons generated by each beam are projected through a second electron column onto a multi-detector array and the signal from all the beams is collected simultaneously. Careful design and alignment of the projector column has reduced cross talk between the imaging detectors to levels that are acceptable. A magnetic beam splitter is utilized to separate the primary and secondary electron beams. The total data throughput rate is then the product of the single detector collection rate with the number of electron beams used. In the present case, an array of 61 beams scanned over the sample surface produces images at up to 1.2 gigapixels per second. Practical imaging is achieved by simultaneously scanning all 61 beams such that each beam scans a subset of the total image area while including a small overlap between the images. Thus, large areas are scanned in the time it takes to scan one sub-image leading to the high throughput. Imaging of larger areas is achieved by stepping the sample stage across the area to be imaged, acquiring an image at each step, then stitching the images together. In this manner, large areas (mm2) can be rapidly imaged at a pixel size as small as 3 nm. Figure 2 is an image of a Au on C resolution sample obtained at 3 kV with a single electron beam and demonstrates a resolution of 2 nm. Figure 3 is a multi-field of view (mFOV) of a static random access memory (SRAM) test structure.[4] This image is about 100 m in width and contains 61 separate images all obtained in parallel in 1 second with a pixel size of 4 nm. Figure 4 Microsc. Microanal. 21 (Suppl 3), 2015 698 demonstrates the detail visible in the individual images of the test structure. Potential important applications of high throughput imaging are in the areas of quality control, failure analysis and the potential to study evolving structures over large areas with good temporal resolution. References [1] A. L. Keller et al, Proc. of SPIE, 9236 (2014), 92360B-1. [2] A. L. Eberle et al, Microscopy (Tokyo), 63 (2104),1. [3] A. L. Eberle et al, J. Microsc. Accepted for publication, 2015. [4] http://www.brukerafmprobes.com/a-3553-scmsample.aspx. [5] Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United States Department of Energy (DOE) under contract DE-AC0494AL85000. Figure 1. Schematic of the mSEM electron optics. Figure 2. Au on C image acquired at 3kV. column. Figure 3. mFOV of an SRAM test structure acquired at 1.5 kV with 4 nm/pixel. Figure 4 Imaged formed by the central beam from the mFOV in Figure 3. The width of the image is 12.6 m.");sQ1[350]=new Array("../7337/0699.pdf","Measurement of the Electron Beam Point Spread Function (PSF) in a Scanning Electron Microscope (SEM)","","699 doi:10.1017/S1431927615004298 Paper No. 0350 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Measurement of the Electron Beam Point Spread Function (PSF) in a Scanning Electron Microscope (SEM) Yudhishthir P. Kandel1, Matthew D. Zotta1, Andrew N. Caferra1, Richard Moore2, and Eric Lifshin1 1 2 College of Nanoscale Science and Engineering, SUNY Polytechnic Institute, Albany, NY, USA Nanojehm Inc. Albany, NY, USA A knowledge of the spatial distribution of the electron beam current density, often referred to as the point spread function (PSF), is valuable for understanding the behavior of scanning electron microscopes (SEM) and various other instruments. Previously, a number of attempts at PSF determination have been made based on experimental measurements or electron optical calculations [1-7]. Some of the experimental methods employed knife edge or other scans[1, 8]. Liddle et al. [5] used a TEM image of a reference sample to determine the PSF for an electron beam lithography tool. They assumed an elliptical Gaussian shape for the electron beam and determined its standard deviations in two orthogonal directions using an iterative method to match the reference and blurred images. Babbin et al. developed a test sample that can be used to estimate the PSF using a Fourier transform method[6]. All these approaches are limited in the sense that they do not provide the fine, often irregular, details in the electron beam shape that may not be symmetric or monotonic. A more accurate determination of a PSF is critical, however, if the goal is to improve SEM resolution by deconvolution using the method described by Lifshin et. al [9]. PSF determination described here is based on the availability of a well characterized near planar reference sample with a very small secondary electron mean free path. The intensity map of the reference sample is designated by the matrix X. It is measured under conditions such that the probe size is comparable to the pixel size. Furthermore, a high enough probe current and data collection time is selected to ensure an adequate signal to noise ratio. The reference sample is then imaged with either the same or a different microscope using different operating conditions to get image b, where the probe size may be considerably larger than the pixel size. These imaging conditions typically correspond to practical operating conditions such as the use of large probe currents needed to obtain low noise images in a short time as is the case for a thermionic source SEM. If the PSF, K, is assumed to be position independent, which we have verified experimentally for a range of conditions, then the problem of finding electron beam shape can be posed with the use of, , the regularization functional. As a prerequisite for implementation of this method, procedures have been developed for proper sub-pixel image alignment and the elimination of artifacts present from digitization, brightness, contrast, and gamma settings as well as saturation effects. As an example, Figure 1 shows a three dimensional PSF near the electron beam focus measured with a TESCAN VEGA� LaB6 source SEM operated at 20 keV and 7 pA of probe current. SEM images of an Au-C Pella� sample corresponding to various focus positions and the reference image are also shown. Since the roughness of the sample was less than one micrometer and the PSF did not change significantly over the first one micrometer step shown, the value around the zero displacement setting was found to Microsc. Microanal. 21 (Suppl 3), 2015 700 successfully lead to image restorations with improved spatial resolution. Research is underway to use the full PSF in dealing with structures with a higher degree of roughness. The authors acknowledge the support of Mr. Jeffrey Moskin, President of Nanojehm for providing the resources that made this study possible as well as TESCAN for providing instrumental support. References: 1. 2. 3. 4. 5. 6. 7. 8. 9. E. Kratschmer, S. A. Rishton, D. P. Kern and T. H. P. Chang, JVST B 6 (6), 2074-2079 (1988). M. Sato, JVST B: Microelectronics and Nanometer Structures 9 (5), 2602 (1991). J. E. Barth and P. Kruit, optik 109 (3), 101-109 (1996). A. Goldenshtein, Y. I. Gold and H. Chayet, SPIE 3332, 132-137 (1998). J. A. Liddle, P. Naulleau and G. Schmid, JVST B 22 (6), 2897 (2004). S. Babin, M. Gaevski, D. Joy, M. Machin and A. Martynov, JVST B 24 (6), 2956-2959 (2006). K. Nakamae, M. Chikahisa and H. Fujioka, Image and Vision Computing 25 (7), 1117-1123 (2007). T. A. P. J.W. Elmer, A.T. Teruya, presented at the 2nd International Conference on Electron Beam Welding, Aachen, Germany, 2012 (unpublished). E. Lifshin, Y. P. Kandel and R. L. Moore, Microscopy and Microanalysis 20 (01), 78-89 (2014). Figure 1: (a) shows the general electron beam envelope near focus, result from measurement slices taken at various locations as measured in TESCAN VEGA is shown in (b) as a three dimensional color plot; (c) is the reference image taken with TESCAN MIRA� (c1), (c2), (c3), and (c4) shows the images taken at various focus level.");sQ1[351]=new Array("../7337/0701.pdf","Imaging Contrast with Multiple Ion Beams","","701 doi:10.1017/S1431927615004304 Paper No. 0351 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Imaging Contrast with Multiple Ion Beams Huimeng Wu, Sybren Sijbrandij, Shawn McVey and John Notte Ion Microscopy Innovation Center, Carl Zeiss Microscopy LLC, One Corporation Way, Peabody MA 01960, USA A commercial Ga-FIB/SEM system can directly image samples using an ion beam or e-beam before, after or during milling/depositing process. This capability provides important feedback for process control. Because of the limited spatial resolution and Ga contamination of the Ga+ ion beam, the e-beam is often considered as the primary imaging tool. But ion beam imaging also provides important information about the samples. Orion Nanofab integrates He+, Ne+ and Ga+ focused ion beams on one single platform. He+ and Ne+ ion beams are based on the gas field ion source (GFIS) technology. The images generated by He+/Ne+ ion beams are sub-nm in resolution capturing intricate details of the samples[1]. Nanofab also provides an optional state of art Ga-FIB. With this unique configuration, Orion Nanofab provides a great platform to study ion beam imaging with a variety of ion species. In this study, we investigate three imaging modes: secondary electron (SE), secondary ion (SI) and backscattered ion (BSI), using three ion beams: He+, Ne+ and Ga+. The most common imaging mode of the e-beam and ion beam is via SE. Compared with the SE image of e-beam, ion beam imaging is more sensitive to surface topography, easier charge neutralization with a flood gun, better for passive voltage contrast, and stronger grain orientation contrast due to ion-channeling effects. Other imaging modes via SI and BSI can provide additional information to better characterize the sample and eliminate the need for further analysis. SI images provide material contrast because of the different secondary ion yield of different materials, particularly sensitive to the presence of oxides and carbides. This property enables the studies of corrosion or grain boundary segregation in metallic systems[2]. Also the SI mode allows the electron flood gun to operate in continuous emission mode without multiplex imaging. BSI images provide less surface specification, but the atomic number "Z" contrast. Because BSI has the similar energy as the incident ion beam energy, the signal can pass through several layers. The schematic illustration in Fig. 1a shows three ion beams on a Orion Nanofab and the ion beam/sample interactions. When an ion beam impinges on a sample surface, it emits secondary particles, including low energy secondary electrons, secondary ions, backscattered ions and others. An experimental measurement was carried out to compare the BSI signal and SI signal for He+ and Ne+ ion beams on ten chosen target elements (Fig. 1b). The results show that the Ne+ induced SI signal is more than one order of magnitude larger than He+, as is expected from SRIM-calculations of sputter yield[3]. For a He+ ion beam, the BSI signal is about an order of magnitude larger than the SI signal. For a Ne+ ion beam, the BSI signal intensity is approximately equal to the SI signal. From this measurement, a He+ ion beam is suitable for SE and BSI images, but the SI signal is relatively low. A Ne+ ion beam is useful for acquiring all imaging with SE, SI and BSI signal. Two samples were tested to show that the combination of SE, BSI and SI images can provide complementary information about the samples. Fig. 2a and 2b are tilted He+ ion beam images of carbon thin film on coper grid. The SE image (Fig. 2a) shows the surface topography of the thin carbon film and the SE signal from the materials beneath the carbon film can't penetrate through the film. The BSI image (Fig. 2b) shows better contrast of the underlying material than the carbon film because of the Microsc. Microanal. 21 (Suppl 3), 2015 702 carbon low Z. Fig. 2c and 2d are Ne+ ion beam images of the oxide patterns with Al pads. The SE image (Fig. 2c) shows the very strong voltage contrast. Two grounded Al pads on the bottom corner of the image are very bright. The two floating pads in the center of the image appear almost black. All oxide dielectric patterns are very dark because of charging. While in the SI image (Fig. 2d), all oxide dielectric patterns show good contrast and all four metal Al pads are dark because of the low SI yield. No flood gun was applied during the imaging of either samples. References: [1] BW Ward, JA Notte and NP Economou, Vac. Sci. B. 24 (2006), p. 2871. [2] A Laquerre and MW Phaneuf, Microscopy & Microanalysis, 14 (2008), p. 620. [3] http://www.srim.org/ a b Figure 1. a) Schematic illustration of three ion beams on a) Orion Nanofab and the ion beam/sample interaction. b) Comparison of BSI and SI signal for 25 keV He+ and Ne+ ion beam. Figure 2. Tilted He+ ion beam a) SE and b) BSI images of carbon thin film on coper grid. Ne+ ion beam c) SE and d) SI images of the oxide patterns with Al pads.");sQ1[352]=new Array("../7337/0703.pdf","Low Energy BSE Imaging with a New Scintillation Detector","","703 doi:10.1017/S1431927615004316 Paper No. 0352 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Low Energy BSE Imaging with a New Scintillation Detector J. Kolosov�1, J. Jiruse1, J. Fiala1 and J. Ber�nek1 1 TESCAN Brno, s.r.o., Libusina trida 1, Brno, Czech Republic. More and more applications of the scanning electron microscope (SEM) rely on a low electron energy because it decreases the depth of specimen radiation damage. It also enables a clear visualization of nonconductive samples, surface structures are better resolved, and new types of contrast can be observed [1, 2]. In the case of backscattered electron (BSE) imaging, working at low primary beam energies (and currents) puts high demands on the actual detector design because of the weak signal intensities. The sensitivity of the most commonly used scintillation detectors drops rapidly in the region of energies under 3 keV. We have recently developed a new scintillation type BSE detector with enhanced sensitivity in the low energy region [3]. Thanks to a special surface treatment the dead layers on the detection surface were reduced and the detection limit of the detector was pushed down to 200 eV, see Figure 1. The good collection efficiency together with the high light yield and short decay time of the scintillation material guarantee high signal to noise ratio even at high scanning rates and low probe currents. This is crucial for many applications, like for example 3D analysis of radiation sensitive or non-conductive samples. The benefits of this BSE detector can also be exploited in the semiconductor industry, especially in the field of failure analysis. The fast 3D reconstruction using the new BSE detector at low primary energies gives contrast which offers a better depth resolution. The signal contains information only from the top layers and is not affected by the signal from the bulk, see Figures 2A and 2B. In the surface-enhanced Raman microscopy (SERS) the molecules to be analyzed are adsorbed on rough surfaces, such as partially metal-coated plasma etched polystyrene spheres. This substrate is nonconductive, thus correlative SEM and Raman imaging [4] requires low primary energies and very low probe currents, see Figure 2C. Another application is the 3D reconstruction of resin-embedded biological samples where low primary beam energies and currents are necessary to minimize possible radiation damage and sample charging. Because of large quantities of analyzed material, high imaging speeds are required, i.e. FIB-milling plus milled surface imaging must fit within the limit of 2 minutes, and pixel resolution of the image should be higher than 10 nm per pixel, see Figure 3. References: [1] DC Bell, N Erdman in "Low Voltage Electron microscopy: Principles and Applications", ed. DC Bell and N Erdman, (John Wiley & Sons, Chichester) p. 1. [2] L Frank et al, Materials 5 (2012), p. 2731. [3] J Kolosova et al, Proceedings of the 18th International Microscopy Congress (2014). [4] J Jiruse et al, Microscopy and Microanalysis 20 (2014), p. 990. Microsc. Microanal. 21 (Suppl 3), 2015 704 Figure 1. Detection limits of the new BSE detector. A): Tin on carbon sample imaged at 200 eV primary beam energy and 65 pA probe current. B): Ni grid imaged at various pixel dwell times (from 320 ns to 32 �s), 500 eV primary energy and 500 pA beam current. Imaging at high scanning rates in the low energy region is possible. A B C Figure 2. IC cross section imaged at A) 1 keV and B) 5 keV primary energies showing enhanced surface sensitivity at lower primary beam energies. C): Au-coated plasma-etched polystyrene spheres (SERS substrate). Image was taken at 1 keV primary energy and 45 pA probe current. Figure 3. Stained liver section imaged at 2 keV primary energy and 100 pA probe current. Acquisition times of the images (2048 x 2048 pxl) were 49 seconds. Imaging was done with the low energy BSE detector placed inside the column. Sample courtesy Gatan.");sQ1[353]=new Array("../7337/0705.pdf","Origins and Contrast of the Electron Signals at Low Accelerating Voltage and with Energy-Filtering in the FE-SEM for High Resolution Imaging","","705 doi:10.1017/S1431927615004328 Paper No. 0353 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Origins and Contrast of the Electron Signals at Low Accelerating Voltage and with Energy-Filtering in the FE-SEM for High Resolution Imaging Hendrix Demers1, Nicolas Brodusch1, Patrick Woo2, and Raynald Gauvin1 1. 2. Department of Mining and Materials Engineering, McGill University, Montreal, Quebec, Canada. Hitachi High-Technologies Canada Inc., Toronto, Canada. For developing new technologies, it is important to characterize materials at the nanoscale. To achieve high resolution, field emission scanning electron microscopes (FE-SEM) were developed. These microscopes allow working at low accelerating voltage, below 5 kV, to take advantage of the reduction of the interaction volume with accelerating voltage. Furthermore, their higher gun brightness compared to conventional thermo-electronic emitters [1], allow a probe size at the nanoscale. However, technical problems arise when SEM operates at low kV (source brightness decreases and chromatic aberration increases). Using the deceleration mode minimizes these problems. Further improvement is achieved by using a cold-field emitter, which has a smaller energy spread, a higher brightness, and a smaller probe size than a FE-SEM. At low accelerating voltage, the emission volume of the backscatter (BSE) and the secondary (SEII) electrons signals is close than the one of the SEI signals. Furthermore, the use of a magnetic field above the sample and probe deceleration improve the spatial resolution by collecting mostly high-resolution signals. In addition, the energy-filtering of the electron signals allows the selection of contrast detected: topographical, compositional, structural, or crystallographic. Figure 1 and 2 are illustrations of high resolution and high contrast imaging at two different magnifications of a Li2FeSiO4 powder with two different in-lens electron detectors (upper and top) with a Hitachi SU8230 cold field-emission SEM. The powder was dispersed in a mixture of ethanol and ionic liquid, which provided charge compensation during imaging and allowed using the most suitable accelerating voltage (2.2 kV) to obtain high contrast micrographs [2], to reduce beam spreading, and to increase the compositional contrast. In addition, energy-filtering was applied to the upper and top detectors. Both collected signals were composed of a combination SEs and BSEs, the upper detector collects low angle electrons whereas the top detector collects high angle electrons. In these micrographs, the upper detector signal was filtered to repulse the low energy SEs from the collector which improve the topographical contrast (Fig. 1A and 2A). In these micrographs, the signal is majority composed of high energy SEs (> 9 eV) and low angle BSEs and the contrast is obviously topographic with a compositional component. On the other hand, the top detector was filtered at 80% leading to a signal composed mostly of high energy BSEs (1.76 to 2.20 keV). A high compositional contrast between iron oxide (bright) and ortho-silicate (dark grey) particles was observed with similar high spatial resolution. The high angle BSEs, collected by the top detector, are assumed to come from a small volume around the probe impact point. In addition, high energy BSEs have smaller emission volume, which further improve the spatial resolution. However, the energy-filtering and the collection solid angle of the top detector decrease the signal-to-noise ratio (SNR) of the micrographs (Fig. 1B and 2B). Modern FE-SEMs, like the HITACHI SU-8230, provide low accelerating voltage, deceleration mode, and energy-filtering of the electron signals to allow the characterization of materials at the nanoscale with various types of contrasts. The understanding of the origins of the collected signals and effect of energyfiltering on electron micrograph contrast will extend the imaging capabilities of the FE-SEM towards new nanoscale applications. Microsc. Microanal. 21 (Suppl 3), 2015 706 References: [1] J.I. Goldstein et al, "Scanning Electron Microscopy and X-Ray Microanalysis", (Springer, 3rd edition, 2003). [2] N. Brodusch et al, Microscopy and Microanalysis, 20 (2014), pp.38-39. [3] D.C. Joy, Journal of Microscopy, 208 (2002), pp. 24-34. [4] J. Kim et al, Journal of Vacuum Science Technology B, 25 (2007), pp. 1771-1775. Figure 1. Electron micrographs of a Li2FeSiO4 lithium ortho-silicate powder with energy-filtering (V0 = 2.2 kV) and ionic liquid preparation. A: Upper detector (filter bias = 9 V): a resolution of 5.1 nm and a SNR of 65 were measured with SMART-J [3,4]. B: Top detector with filtering (80%): a resolution of 7.1 nm and a SNR of 2.9 were measured with SMART-J [3,4]. Figure 2. Higher magnification electron micrographs of a Li2FeSiO4 lithium ortho-silicate powder with energy-filtering (V0 = 2.2 kV) and ionic liquid preparation. A: Upper detector (filter bias = 9 V): a resolution of 4.7 nm and a SNR of 2.9 were measured with SMART-J [3,4]. B: Top detector with filtering (80%): a resolution of 5.0 nm and a SNR of 0.5 were measured with SMART-J [3,4].");sQ1[354]=new Array("../7337/0707.pdf","The use of Microcontact Printing of Thiols to Evaluate Attachment of Xylella fastidiosa Under Distinct Conditions of Calcium Availability.","","707 doi:10.1017/S143192761500433X Paper No. 0354 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The use of Microcontact Printing of Thiols to Evaluate Attachment of Xylella fastidiosa Under Distinct Conditions of Calcium Availability. Breno Leite1, Rafal Dziedzic2, L. F. Cruz3, A. L. Gillian-Daniel2, Charlie Nielsen1 and L. De La Fuente3 1. 2. JEOL USA, Inc., Peabody, MA, USA University of Wisconsin, MRSEC Education Group, Madison, WI, USA 3. Auburn University, Department of Entomology & Plant Pathology, Auburn, AL, USA (bleite@jeol.com) Plant Pathogenic bacteria attach to surfaces to form biofilms and effectively colonize the host plant. Xylem vessels internal surface attachment is the first step towards colonization during the infection of grapevines by Xylella fastidiosa [1], causative agent of Pierce's disease, an economically important factor for the California wine industry. Auburn University [2,3] is devoted to identify and test factors that are favorable to attachment and those that are not. The study of this xylem-limited pathogen is critical to establish how pathogenic cells reach massive growth inside vessels and eventually alter plant growth and/or cause death. Massive growth also means enormous production of virulence factors, such as toxins, and extended vessels blockage of normal flow of nutrients and water. The goal of this investigation is to test a methodology that can reduce the time of initial screening of chemical influencing factors and speed up experimental protocols. We accomplished these goals by: 1) monitoring biofilm formation under controlled conditions on glass surfaces coated with gold vs. gold surfaces coated with thiol (SH rich moieties), and 2) comparing obtained results with published results. Layered assemblies were manufactured at the University of Wisconsin, which simulates surfaces that typically interact with pathogens. We used artificial gold surfaces coated with an octadecamethiol monolayer, that causes alterations in their hydrophobicity. The hydrophobicity changes because -SH groups are attracted to gold and leave behind molecule tails that are averse to water. The change to a negative due to SH groups ionization H (S- remains on the surface). Negative sulfur is attracted to positively charged gold. The technique is called Microcontact Printing of Thiols (http://education.mrsec.wisc.edu/294.htm). We tested a penny print (Figure 1A, B) and a designed pattern (not shown). Biofilm growth was tested by growing X. fastidiosa in PD2 growth media liquid cultures: control and supplemented media. Prepared slides were sterilized with ethanol inside the laminar flow hood and transferred to Falcon tubes containing PD2 [4] media and incubated for 5 days at 28�C. After incubation slides were dried and fixed with osmium tetroxide. Treatments were: 1) Control PD2 medium + X. fastidiosa (inoculated media used as positive control); 2) PD2 + 2mM CaCl2 + X. fastidiosa (Complex media PD2 supplemented with calcium) and 3) PD2 + 1.5mM EGTA + X. fastidiosa (Complex media PD2 supplemented with the calcium chelator EGTA). Scanning Electron Microscopy (JSM-6610 JEOL) was used to image surface growth. Images aimed to highlight differences between areas outside and inside the thiol printed areas. Biofilm and/or cell colonization was assessed by processing the data collected with the National Institutes of Health "ImageJ" software (http://imagej.nih.gov/ij/). The colonized area % was correspondent to the area with bacterial cell forms/Total area analyzed. Data has confirmed that: a) X. fastidiosa cells attach preferentially to surfaces with higher degrees of hydrophobicity (Figure 1 B); b) calcium (CaCl2) significantly enhances cell attachment and aggregation (Figure 1 C) and c) EGTA reduces attachment. Results corroborate previously obtained data [2,3,4]; however this new method is less time consuming. Microsc. Microanal. 21 (Suppl 3), 2015 708 Repeatability and standardization will likely be needed to introduce this method as a routine screening step in many labs devoted to test the influence of chemicals and other factors on colonization of many pathogens. Preliminary tests will be then refined for more complete and conclusive work. Data also allowed a better rationalization of how layers are assembled in vivo (Figure 1 D); We foresee the increase of data generation allowing: a) establishment of common basis for attachment of plant pathogens, b) help determine the foundation of the physics and chemistry involved in interactions of cells and their environment. References: [1] A.J. Simpson et al (2000). Nature 13 (406): 151-9. [2] L.F. Cruz et. al. (2012). Appl. Environ. Microbiol. 78 (5): 1321-31. [3] L.F. Cruz et. al.(2014). Appl Environ Microbiol, 12. pii: AEM.02153-14. [Epub ahead of print]. [4] B. Leite et al. (2004). F. of European Microbiology Society 230: 283-290. [5] P.A. Cobine et. al. 8 1 (2013) PLoS ONE: e54936. doi:10.1371/journal.pone.0054936. Authors acknowledge Dr. Maria L Ishida (FDACS) for critically reviewing the manuscript: Figure 1. Region of a penny reproduced by thiol print on the right image (A); Vigorous X. fastidiosa on the thiol print penny areas (B); Area % occupied by cells and biofilm (Image J software ) either on top of a gold surface or thiol printed areas. PD2 (control); CaCl2 supplied and in the presence of EGTA, a calcium chelator (C). Rationalization of why calcium mediation increases surface colonization (D).");sQ1[355]=new Array("../7337/0709.pdf","Rhizomes and Roots of Rare Arctic-Alpine Snowfield Plants on the Edges of Retreating Snowfields at Glacier National Park, Montana","","709 doi:10.1017/S1431927615004341 Paper No. 0355 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Rhizomes and Roots of Rare Arctic-Alpine Snowfield Plants on the Edges of Retreating Snowfields at Glacier National Park, Montana Apple, Martha E.1, M. K. Ricketts1, and L.G. Carlson1 1 Department of Biological Sciences, Montana Tech of the University of Montana, Butte, Montana, USA, 59701. The glaciers and snowfields at Glacier National Park, Montana are disappearing due to climate change [1]. Snowfield plants rely on melting snow from the edges of snowfields and glaciers during the brief summer growing season in this otherwise harsh environment. Since the edges of snowfields and glaciers move as these icy bodies recede, snowfield plants live on changing edges and may lose this habitat altogether with the disappearance of snowfields and glaciers. Glacier National Park is home to rare arctic-alpine plants that inhabit the snowfield's edges and that live downstream and obtain their water, at least in part, from summer snowmelt. Thus, the habitats of these rare plants are endangered. Plants are intrinsically linked to their environments, and plant functional traits are those characteristics of plants that influence their interactions with the environment [2]. Plant functional traits exist on the macroscopic, microscopic, physiological, biochemical, and temporal levels. For example, the presence of rhizomes is a plant functional trait. Rhizomes are macroscopic, underground horizontal stems that occur frequently in snowfield plants. Rhizomes provide a ready reservoir of carbohydrates that if mobilized and metabolized, can be used by snowfield plants for rapid growth; for withstanding adverse conditions before snowmelt; for production of adventitious roots; for clonal reproduction; and for production of buds. These buds can quickly sprout during snowmelt and grow into aerial shoots that may in turn produce flowers or spores. Rhizomes are also a key means of advance for pioneer species, which include plants colonizing a recently revealed snowfield's edge. Phenology is the science of the timing of biological events, and one consequence of climate change is that phenology can be offset. Therefore, rhizome physiology can be seen as a functional trait that is a coordinating mechanism for many phenological events in snowfield plants. In 2012 and 2014, we established geospatially referenced transects perpendicular to the lateral and leading edges of snowfields at Mt. Clements, Siyeh Pass and Piegan Pass at Glacier National Park [3]. We collected leaves for morphometric measurements from 1m2 quadrats placed at 5m intervals along the 50m transects. Functional traits of plants differed significantly with distance from the snowfields. Morphometric traits of leaves collected along an environment gradient extending from the edge of the vast Mt. Clements snowfield to the top of the steep (35�) Mt. Clements moraine were measured with Image-J and dry weights were obtained [4]. Specific leaf area, (mm-2/mg dry weight,) is an indirect measure of leaf density. Leaves with higher SLA generally have less densely packed cells. At the Mt. Clements moraine, community-weighted trait means of SLA decreased significantly with distance from the snow. Therefore, the leaves of the collective snowfield species were thinner, or less dense, near the snow where water availability was greater, and thicker, or with greater density, farther from the snow, where water availability decreased. Thinner leaves are less drought-tolerant than thicker, dense leaves with lower SLA. Leaf circularity was greater closer to the snow and leaf shape became more complex with increasing distance from the snow. Snowfield plants are likely to have herbaceous growth forms, have underground storage organs (including rhizomes, bulbs, corms, and taproots); and to have simple leaf shapes. Cushion plants and shrubs became more common with distance from the snow. Soil Microsc. Microanal. 21 (Suppl 3), 2015 710 qualities as well as snow and water availability may influence the plants, as soils had a significantly greater proportion of fine particles at the edge of the Mt. Clements Moraine snowfield. Plant functional traits also varied on the periglacial patterned ground at Siyeh Pass and Piegan Pass, which takes the form of a brown and green striped landscape of terraces with brown stony treads and green inclined risers. Nitrogen is relatively limited in alpine soils, and nitrogen-fixing members of the legume family (Fabaceae) grew on both the treads and the risers. On the risers, Salix arctica, the arctic willow, and Dryas octopetala, of the rose family (Rosaceae), are co-dominant low-growing ectomycorrhizal shrubs that form extensive mats [5, 6]. Experimental warming of the rhizosphere fungi of S. arctica suggested that fungal community biomass increased with increased belowground allocation of carbon [7]. Alpine ectomycorrhizal symbioses and belowground carbon pools may therefore be influenced by the warmer temperatures associated with climate change. Seedlings and saplings of Pinus albicaulus (Whitebark Pine) and Abies lasiocarpa (Subalpine fir) associated with the risers but not with the treads. Plant communities differed markedly between the risers and the treads, which had a greater percentage of rare arctic-alpine plants. Plants on the stony treads had a greater incidence of aboveground, xeromorphic (drought tolerant) features. Alpine fell-field plants on calcareous soils of the northern Rocky Mountains of Montana and Wyoming had a high incidence of vesicular-arbuscular mycorrhizal colonization [8]. The specific mycorrhizal status of plants on the rocky treads of the periglacial patterned ground at Siyeh Pass and Piegan Pass remains to be investigated and would lend itself well to microscopy. References [1] Brown, J., J. Harper, and N. Humphrey. 2010. Cirque glacier sensitivity to 21st century warming: Sperry Glacier, Rocky Mountains, USA, Glob. Planet. Change. doi:10.1016/j.gloplacha.2010.09.001 [2] Venn, S., Pickering, C., and Green, K. 2014. Spatial and temporal functional changes in alpine summit vegetation are driven by increases in shrubs and graminoids. AoB PLANTS 6:plu008. Doi:10.1093/aobpla/plu008 [3] Apple, M. E. 2012. Measuring Impacts to Rare Peripheral Arctic-Alpine Plants at the Edges of Permanent Snowfields/Glaciers that are Receding due to Climate Change in Glacier National Park. RMCESU Report P12AC10557, MT-02. [4] Schneider, C. A.; Rasband, W. S. & Eliceiri, K. W. (2012), "NIH Image to ImageJ: 25 years of image analysis", Nature methods 9(7): 671-675, PMID 22930834 [5] Treu, R., Laursen, G.A., Stephenson, S.L., Landolt, J.C. and Densmore, R. 1995. Mycorrhizae from Denali National Park and Preserve, Alaska. Mycorrhiza 6(1):21:29. [6] Bjorbaekmo, M.F.M., Carlsen, T., Brysting, A., Vralstad, T., Hoiland, K., Ugland, K.I., Gem, J., Schumacher, T., and Kauserud, H. 2010. High diversity of root associated fungi in both alpine and arctic Dryas octopetala. BMC Plant Biology, 10:244 doi:10.1186/1471-2229-10-224. [7] Fujimura K.E., Egger K.N., Henry G.H , 2008 The effect of experimental warming on the rootassociated fungal community of Salix arctica. Isme Journal.2(1):105-114p. [8] Lesica, P. and Antibus, R. K. 1986. Mycorrhizae of alpine fell-field communities on soils derived from crystalline and calcareous parent materials. Canadian Journal of Botany, 1986, 64(8): 1691-1697, 10.1139/b86-226 [9] The authors acknowledge funding from RM-CESU and acknowledge The Crown of the Continent Research Learning Center at Glacier National Park.");sQ1[356]=new Array("../7337/0711.pdf","Impact of Phenazine-1-carboxylic acid upon Biofilm Development in the Rhizosphere of Dryland and Irrigated Wheat.","","711 doi:10.1017/S1431927615004353 Paper No. 0356 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Impact of Phenazine-1-carboxylic acid upon Biofilm Development in the Rhizosphere of Dryland and Irrigated Wheat. Melissa K. LeTourneau1, Matthew M. Marshall2, Linda S. Thomashow3, and James B. Harsh1 1. 2. Department of Crop and Soil Sciences, Washington State University, Pullman, WA, USA. Pacific Northwest National Laboratory, United States Department of Energy, Richland, WA, USA. 3. Agricultural Research Service, United States Department of Agriculture, Pullman, WA, USA. Rhizobacterial biofilms are important sinks for plant-derived carbon and sources of organic matter in soil. Phenazine-1-carboxylic acid (PCA), a bacterial metabolite that has been shown to promote biofilm development via reduction of iron in culture [1], has been observed in concentrations up to 1 g/g root in the rhizospheres of dryland, but not irrigated, cereal monocultures throughout the low-precipitation zone of the Columbia Plateau [2]. Because soils in this area are particularly susceptible to loss of organic matter and erosion due to widespread, long-term conventional tillage practices, we are investigating the impact of PCA upon storage of plant-derived carbon in rhizobacterial biofilms under both dryland and irrigated conditions. We hypothesize that PCA promotes accumulation of extracellular polymeric substances (EPS) in the rhizosphere, especially under dryland conditions. Our approach involves microscopic comparisons of biofilms formed by a PCA-producing fluorescent pseudomonad strain (PCA+) and an isogenic mutant impaired in PCA synthesis (PCA-) grown on wheat roots. In order to simulate dryland and irrigated soil moisture conditions, the wheat is planted in soilfilled columns within a growth chamber imposing temperature, humidity, and soil-moisture controls. We are comparing root-colonization patterns using confocal microscopy, extent and morphology of biofilm matrices using focused-ion-beam SEM, and nano-structural characteristics of biofilm matrices using helium-ion microscopy. Non-inoculated controls have been established in these experiments to account for endophytic bacteria introduced from the seed and bulk soil bacteria that have survived soil autoclaving. The soil-grown root systems in our experiments have robust rhizo-sheaths (Figure 1), such that representative colonies are easier to identify on roots grown on filter paper (Figure 2). However, we have observed some consistent colony morphologies on soil-grown root surfaces, (Figure 3). Additional sample comparisons using the variety of imaging techniques described above are expected to confirm that the PCA+ strain consistently produces more EPS than the PCA- strain under dryland and irrigated conditions [3]. References: [1] Y Wang, JC Wilks, T Danhorn, et al., J. Bacteriol. 193 (2011), p. 3606. [2] DV Mavrodi, O Mavrodi, JA Parejko, et al., Appl. Environ. Microbiol. 78 (2012), p. 804. [3] The authors acknowledge funding from the USDA Agriculture and Food Research Initiative grant 2011-67019-30212, the DOE Office of Science Graduate Student Research Program, and the Otto and Doris Amen Dryland Research Endowment. We also appreciate the consultation, training, and analyses provided by microscopy and spectroscopy scientists at the DOE Environmental Molecular Sciences Laboratory, and staff at the Franceschi Microscopy and Imaging Center at Washington State University. Microsc. Microanal. 21 (Suppl 3), 2015 712 200 m Figure 1. Soil-grown roots from the growth chamber experiments have robust rhizo-sheaths, which complicate identification of representative colonies at root surfaces. 5 m 5 m 5 m 5 m Figure 2. Representative colony morphologies on roots grown on filter paper. Upper images are PCA+ colonies, while lower images are PCA- colonies. More sample comparisons are required to determine whether these distinct colony morphologies are consistent. 10 m 10 m Figure 3. Colonies on soil-grown roots. The left image is a PCA+ colony, while the right image is a PCA- colony. Larger bacteria in the PCA- colony are most likely endophytes introduced from the seed, or bulk soil bacteria that survived soil autoclaving.");sQ1[357]=new Array("../7337/0713.pdf","Imaging Ventral Cell Plate Formation in Guard Cells","","713 doi:10.1017/S1431927615004365 Paper No. 0357 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Imaging Ventral Cell Plate Formation in Guard Cells Xingyun Qi1, Keiko U. Torii1,2 1. 2. Department of Biology, University of Washington, Seattle, WA, 98195, USA. Howard Hughes Medical Institute, University of Washington, Seattle, WA, 98195, USA. Plant cells, unlike animal cells, are encapsulated in rigid cell walls, which provide the mechanical support but also limit the movement of plant cells. Among the few plant cell types that could move, stomatal guard cells are of great importance. The opening and closure of stomatal guard cells control the trade-off between efficient gas exchange for photosynthesis and water loss via transpiration [Figure 1A]. Therefore, an interesting question is to understand on molecular basis the regulation of the ventral cell wall movement of the guard cells. By analyzing the transcriptome in meristemoid, a stomatal precursor with stem-cell-like properties [Figure 1B-1D], we identified a novel gene with unknown function [1]. The GFP fusion of this gene product driven by its endogenous promoter is specifically expressed in the stomata lineage cells at the late developmental stage from late meristemoids to mature guard cells, with peak expression at the guard mother cells. Interestingly, besides some intracellular highly mobile punctae, the GFP signal is extremely enriched in the new cell plate between two future guard cells. Furthermore, we noted that this protein exhibits a polar distribution at the cell periphery, where the future cell plate will eventually fuse, well in advance of the cytokinesis in guard mother cells. The closest homologue of this novel gene based on DNA alignment is found in the plasmodesmal proteome. Plasmodesmata, the physical channel between two adjacent cells, is formed during cytokinesis [2]. It is known that the plasmodesmata flux could be regulated through dilation/constriction of the neck aperture by the modification of the gatekeeper of plasmadesmata, callose [3,4]. Although continuous plasmodesmata were observed in immature guard cells, no continuous plasmodesmata were found connecting guard cells to sister guard cells or to adjacent epidermal cells [5-7], indicating they were sealed during the maturation of guard cells. We are currently testing our hypothesis that this protein may be involved in ventral cell plate formation as well as plasmodesmata regulation during stomata differentiation. References: [1] LJ Pillitteri et al, Plant Cell. 23(2011), p. 3260 [2] AJ Maule, Curr. Opin. Plant Biol. 11(2008), p. 680 [3] JE Radford, M Vesk and RL Overall, Protoplasma 201(1998), p. 30. [4] A Levy, D Guenoune-Gelbart and BEpel, Plant Signal. Behav. 2(2007), p. 404. [5] JE Pallas Jr and HH Mollenhauer, Science 175(1972),p. 1275. [6] AC Wille andWJ Lucas,Planta160(1984), p, 129. [7] BA Palevitz andPK Hepler,Planta164(1985), p. 473. [8] The authors acknowledge funding from a grant from Gordon and Betty Moore Foundation (GBMF3035) to KUT. KUT is an HHMI-GBMF Investigator. Microsc. Microanal. 21 (Suppl 3), 2015 714 Figure 1. Stomatal development (A) A stoma is a passage for carbon dioxide, oxygen, and water vapor. Shown is a mature Arabidopsis stoma. (B) A protodermal cell can differentiate into a pavement cell or undergo a transition to be a meristemoid mother cell (MMC). An MMC will create a stomatal precursor stem cell called meristemoid (M) by the initial asymmetric cell division. The M reiterates one to three rounds of asymmetric amplifying divisions in an inward-spiral manner before differentiating into a guard mother cell (GMC). The GMC devides symmetrically to produce the paired guard cells surrounding a pore, which form a mature stoma. (C) Wild-type leaf epidermis of Arabidopsis. m, meristemoid; arrowhead, guard mother cell; asterisks, mature stoma. (D) A meristemoid (expressing GFP, green) reiterates asymmetric division and renews itself.");sQ1[358]=new Array("../7337/0715.pdf","Scanless lattice light sheet microscopy","","715 doi:10.1017/S1431927615004377 Paper No. 0358 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Scanless lattice light sheet microscopy Bi-Chang Chen1 1. Research Center for Applied Science, Academia Sinica, Taipei, Taiwan Over the last decade, powerful new microscopes have dramatically sharpened biologists' focus on the molecules that animate and propel life [1]. The techniques have improved biologists' ability to visually track the movements of cells' tiniest structures. Noticeably, a new imaging platform, "lattice light sheet microscopy" demonstrated by Chen et al. has shown the ability to collect high-resolution images rapidly and minimizes damage to cells, meaning it can image the three-dimensional activity of molecules, cells, and embryos in fine detail over longer periods [2]. However, in order to construct such a lattice light sheet microscope, a lot of optics components needed to be used since there are at least three 4-f (f is focal length) system required, which makes optical alignment complicated. Here, we would like to present a simplified lattice light sheet microscopy, which removes two galvanometers in the lattice light sheet microscopy system, meaning that two 4-f systems are exempted to have a lattice light sheet microscopy free from scanning mechanically. Compared to the lattice light sheet microscopy, in order to have volume scanning via selective plane illumination, the galvanometers serve to translate the light sheet through specimen in x and z [3]. Thus, a pairs of galvanometers with 4-f system is required in between the physical mask and excitation objective. Actually, instead of having galvanometers installed in the system, we could simply shift the projected pattern on the SLM since SLM plane is demagnified onto the sample plane. By this way, the entire microscope becomes very compact and easy for alignment. In the scanless lattice light sheet microscope system as shown in Figure 1, an expanded laser of 488 nm laser illuminates a reflective liquid crystal on silicon (LCoS) spatial light modulator (SLM) displaying the calculated phase pattern conjugated to the sample plane and placed before the annular mask as in the lattice light sheet microscopy system. Firstly, a lens (L1) is used in a 2-f configuration to create a diffraction pattern at its focal plane that is the Fourier transform of the incident laser diffracted by the SLM. A custom opaque mask with transmissive annuli is placed at this plane to filter out the unwanted diffraction orders. This desired diffraction pattern then is focused by the excitation objective with the help of a 4-f image setup (L2 and L3) to produce a light sheet and corresponding point spread functions (PSF) at the sample plane. The fluorescence generated within the specimen is collected by an orthogonal detection objective. Both objectives are dipped in a shallow media-filled and temperature controlled bath. A tube lens (L4) images the fluorescence signal filtered by an optical filter onto a camera. Figure 2 shows the experimental measurements of the lattice light sheet intensity patterns at the sample plane of the excitation objective, which is performed under the scanless lattice light sheet microscope for each Bessel functions at varying distance. These xz excitation intensity profiles are measured by shifting the calculated SLM patterns. The results are in good agreement with theoretically predictions, which means this simplified lattice light sheet microscope is feasible for the bio-imaging applications as lattice light sheet microscopy performs. Since lattice light sheet microscopy could collect high-resolution images rapidly and minimize damage Microsc. Microanal. 21 (Suppl 3), 2015 716 to cells, meaning it can image the three-dimensional activity of molecules, cells, and embryos in fine detail over longer periods than was previously possible, it has drawn a lot of attention in the live image field. To improve the performance of the microscope is required in order to make it more accesses to the public. Here, we demonstrated a simplified lattice light sheet microscope without mechanically scanning, which relaxes the critical optical alignments while performing the construction of lattice light sheet microscope [4]. References: [1] D. J. Stephens and V. J. Allan, Science 300 (2003), p. 82-86. [2] B.-C. Chen, et al. Science 346 (2014), p. 1257998/1-1257998/12. [3] L. Gao, et al. Nature protocols 9(2014), p.1083-1101. [4]The control software of the experimental system was licensed from Howard Hughes Medical Institute, Janelia Farm Research Campus Figure 1. Schematic setup for scanless lattice light sheet microscope : a 488 nm CW laser source is used for excitation and expanded to cover the desired phase pattern on SLM (HoloEye PLUTO VIS). The diffracting patterns is focused by L1= 500 mm and filtered by mask. The filtered intensity pattern is imaged onto the rear pupil plane of the excitation objective (Nikon, CFI Apo 40XW NIR) through a 4-f system composed of L2= 125 mm and L3=250 mm. The emitted fluorescence is collated by the detection objective (Nikon, CFI Apo 40XW NIR) placed orthogonally and imaged onto a sCMOS (Hamamatsu, Orca Flash 4.0) with tube lens L4 = 220 mm. Figure 2. The measured xz cross-sectional intensity profile of the lattice light sheet at the sample: (a) single Bessel beam; (b)-(h), the beam distances between each Bessel beam are 4.80, 2.40, 2.00, 1.64, 1.60, 1.48, and 1.20 �m, separately.");sQ1[359]=new Array("../7337/0717.pdf","Rapid High-resolution Brain Mapping with CLARITY Optimized Light Sheet Microscopy (COLM)","","717 doi:10.1017/S1431927615004389 Paper No. 0359 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Rapid High-resolution Brain Mapping with CLARITY Optimized Light Sheet Microscopy (COLM) Raju Tomer1,4, Karl Deisseroth1,2,3,4 Department of Bioengineering, 2Department of Psychiatry and Behavioral Sciences, 3Howard Hughes Medical Institute, 4CNC Program Stanford University, Stanford, CA A major goal in neuroscience is to map the architecture of brain with both high wiring-level details and a brain-wide perspective. This challenge has drawn the attention of generations of scientists, while excellent progress has been made, many challenges and opportunities remain. Electron Microscopy (EM) is the benchmark method for brain mapping studies [1-2]. EM provides a few nanometer imaging resolutions that allows reconstruction of the finest details of neural circuits. However, EM imaging speed is still generally slow (hence restricted to small blocks of tissues) and automated reconstruction has proven to be challenging. On the other end, MRI based technologies such as Diffusion Tensor Imaging (DTI) provides a quick view of the major neuronal fiber pathways in the brain. However, DTI generally lacks in resolution (about a millimeter) and it is still unclear what these tracts actually represent. There is an obvious disconnect in the high wiring level details of EM imaging and the brain-wide perspective of DTI. Light microscopy approaches, with sub-micron resolution and relatively fast imaging speed, can potentially bridge this gap. In addition, light microscopy combines well with the genetic labelling techniques allowing visualization of the sub-cellular details. One major limitation of light microscopy has been the opaqueness of brain and other biological tissue in the visible light spectrum. This has received increased attention over the last decade, resulting in many new chemical tissue clearing methods ([3-4], among others). Our lab recently developed a new method called CLARITY [5-6] that provides excellent tissue clearing while preserving the molecular and structural content - a feature generally lacking in other methods. The basic idea of CLARITY is to build a highly cross-linked network of hydrogel inside the tissue. This is achieved first by infusion of a cocktail of formaldehyde and hydrogel monomer acrylamide into the tissue, followed by thermal initiation of the polymerization reaction inside the tissue resulting in a highly stable network. In the next step, all the cell membrane lipids are removed from the tissue, either by using electric field [5] or by a simple passive clearing protocol [6], resulting in a highly transparent stable tissue-hydrogel hybrid that is amenable to multiple rounds of specific histochemical labelling and imaging. The next major challenge is to develop optimized high-speed and high-resolution microscopy methods to image large intact transparent tissue in entirety. To address this, we developed CLARITY Optimized Light-sheet Microscopy (COLM [6], Figure 1), building upon the 100 years old idea of light-sheet microscopy [7]. The basic idea of light-sheet microscopy is to illuminate the sample with a thin sheet of light and collecting the emitted fluorescence with an orthogonally arranged wide field detection arm. This optical decoupling of illumination and detection provides two key advantages over the point-scanning Confocal and 2-photon microscopy: (a) high imaging speed, owing to the use of fast sCMOS or CCD cameras, and (b) low photo-bleaching. The light-sheet microscopy idea has been revived in the past decade, 1 Microsc. Microanal. 21 (Suppl 3), 2015 718 largely in the context of developmental biology studies [8]. Additional optimizations were needed to achieve high-resolution deep imaging of large intact samples. COLM encompass three major innovations. Firstly, an optically homogeneous sample mounting approach was developed to minimize any optical aberrations. Secondly, synchronized illumination-detection strategy was employed to reduce the background signal. Finally, an adaptive parameter correction procedure was developed to achieve optimal image quality in the entire sample. COLM enables several novel experimentations such as imaging of intact mouse brains, spinal cord, adult Zebrafish and large pieces of primate brains. Combined with genetic labelling methods, COLM has been used for cell types, structural and functional mapping of mouse and zebrafish brains. [9] References: [1] DD Bock et al. Nature 471 (2011), 177-182. [2] KL Briggman, M Helmstaedter, W Denk, Nature 471 (2011), 183-188. [3] HU Dodt et al. Nat. Methods 4 (2007), 331�336 [4] H Hama et al. Nat. Neuroscience 14 (2011), 1481�1488 [5] K Chung et al. Nature 497 (2013), 332-337. [6] R Tomer et al. Nature protocols 9 (2014),1682-1697. [7] H Siedentopf, R Zsigmondy, Annalen der Physik 10 (1903), 1-39. [8] J Huisken, DY Stainier, Development 136 (2009), 1963-1975. [9] We thank the entire Deisseroth lab for helpful discussions. All the tools and methods described are distributed and supported freely (clarityresourcecenter.org/COLM.html) Figure 1. Comparison of confocal, two-photon and COLM. Confocal achieves optical sectioning by employing a pinhole. Two-photon utilizes the fact that only simultaneous absorption of two photons results in fluorescence signal. COLM, built upon light-sheet microscopy principles, achieves optical sectioning by confining the illumination to the plane of interest, and is several hundred times faster with minimal photo-bleaching.");sQ1[360]=new Array("../7337/0719.pdf","Three-Color Two-Photon Three-Axis Digital Scanned Light-Sheet Microscopy (3c2p3a-DSLM)","","719 doi:10.1017/S1431927615004390 Paper No. 0360 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Three-Color Two-Photon Three-Axis Digital Scanned Light-Sheet Microscopy (3c2p3a-DSLM) Weijian Zong1, 2, Zuosheng Liu3, Fuzeng Niu3, Shaowa Li1, Jia Zhao1, Aimin Wang3* and Liangyi Chen1* The State Key Laboratory of Biomembrane and Membrane Biotechnology, Institute of Molecular Medicine, Peking University, Beijing 100871 2. China Department of Cognitive Sciences, Institute of Basic Medical Sciences, Beijing 100850 3. State Key Laboratory of Advanced Optical Communication System and Networks, School of Electronics Engineering and Computer Science, Peking University, Beijing 100871, China * Corresponding author: lychen@pku.edu.cn, wangaimin@pku.edu.cn Light-sheet fluorescent microscopy (LSFM) or selective plane illumination microscopy (SPIM) is a very powerful tool for biologists to do in vivo imaging[1-4]. With low photo-bleaching rate, fast acquisition speed and good axial resolution, LSFM is the optimal choice for long-term observation of model animal development, such as Drosophila embryos, Zebrafish embryos and C. elegans. By combining two-photon fluorescence excitation with LSFM, Truong et al. have created two-photon digital scanned light-sheet microscopy (2P-DSLM), allowing for deep-tissue imaging of highly scattering Drosophila embryos and of fast beating hearts of Zebrafish [2]. To achieve high axial resolution (thin light sheet) and large field of view simultaneously, we have developed a novel two-photon three-axis digital scanned light-sheet microscope (2P3A-DSLM) based on ultrafast axial scanning of illumination focal spot with a tunable acoustic gradient (TAG) index device[5]. Instead of using wave-mixing to create three excitation wavelengths[6], we used two home-made femtosecond fiber lasers with wavelength centers at 780 nm and 1050 nm, respectively. The 780 nm femtosecond pulses were obtained by frequency doubling of a high power 1560 nm mode-locked Er fiber laser. The final spectrum width of about 10 nm, a repetition rate of 100 MHz and a maximum output power of 880 mW (Fig. 1 (b)). A mode-locked Yb-fiber laser with a fiber amplifier was used to generate the 1050 nm pulses with 10 nm spectrum width, 100 MHz repetition rate, and 1.5 W output power (Fig. 1 (c)). The third laser is a commercial Ti-sapphire laser (Coherent), and its wavelength can be tuned between 680 nm and 1080 nm. Therefore, we can simultaneous excite three fluorophores with different colors. Moreover, with the 2P3A configuration [5], we can tailor our lightsheet to any shape between 5�5 m2 and more than 200�200 m2 with constant thickness limited by diffraction and fast imaging rates limited by the detector. The tailorable illumination area allows multi-scale field of view (FOV), and is capable of imaging cells, tissue and live model animals. Overall, our system is flexible in wavelength combination, excitation intensity tuning and variable sizes of field-of-view selection. The thickness of the lightsheet was maintained at ~1 m in spite of the change in illumination area between 5�5 m2 and more than 200�200 m2, which is optimal to resolve subcellular structure in live organisms. We firstly tested its performance by doing three-color three-dimension imaging of C. elegans. DAPI-labeled chromosome, GFP-labeled cytoplasm and dsRED-labeled motor neuron could be excited and acquired simultaneously(Fig.2). Single nuclear and dendrite could easily be resolved (Fig.2). Next we imaged the pancreas islet and blood vessel of a three-day-old zebrafish to evaluate the capability of our system to do fast deep-tissue multiple-color imaging (Fig. 3). -cells are labled with EGFP in the Tg (kdrl:EGFP) transgenic fish line, and vascular epithelial cells are labeled with RFP in Tg (kdrl:RFP) transgenic fish line. At this early stage of zebrafish embryo, pancreas islet is small and consists only a few cells. They are buried about 250 m inside the body of a 3-day-old zebrafish, which is difficult to visualize using one-photon light-sheet microscopy. Two-photon point-scanning microscopy could work but is too slow to catch fast dynamics such as blood flow and calcium waves. Moreover, its high photonbleaching rates renders long-term imaging impossible. Using our 3C2P3A-DLSM with 920 nm and 1050 nm duel-color excitation, the structure of -cells and the blood vessels and their migration and vascularization can be well resolved. 1. Microsc. Microanal. 21 (Suppl 3), 2015 720 An up-to-date light sheet microscopy arrangement (3C2P3A-DSLM) is demonstrated. The 3C2P3A-DSLM combines multiple-color excitation, tailorable light-sheet area and fast three-dimension imaging in one setup. This will facilitate highly flexible two-photon lightsheet imaging within cells, tissue and live model animals of different labels. This research was supported in part by the National Natural Science Foundation of China grants (31327901, 81222020, 31221002, 61475008) and the Beijing Natural Science Foundation (7121008). References: P. A. Santi, "Light Sheet Fluorescence Microscopy: A Review," J Histochem Cytochem 59, 129 (2011). 2. T. Truong, W. Supatto, D. Koos, J. Choi, S. Fraser, "Deep and fast live imaging with two-photon scanned light-sheet microscopy," Nat Methods 8, 757-760 (2011). 3. P.J. Keller and E. H. K. Stelzer, "Digital Scanned Laser Light Sheet Fluorescence Microscopy," Cold Spring Harb Protoc.5 (2010). 4. T.A.Planchon, L.Gao, D.E.Milkie , M.W.Davidson, J.A.Galbraith, C.G.Galbraith, and E Betzig, "Rapid three-dimensional isotropic imaging of living cells using Bessel beam plane illumination," Nat. Methods 8,417�423 (2011). 5. Zong W, Zhao J, Chen X, Lin Y, Ren H, Zhang Y, Fan M, Zhou Z, Cheng H, Sun Y, Chen L. Large-field high-resolution two-photon digital scanned light-sheet microscopy. Cell Res. 2014. doi: 10.1038/cr.2014.124. [Epub ahead of print]. 6. P. Mahou, J. Vermot, E. Beaurepaire, W. Supatto, "Multicolor two-photon light-sheet microscopy," Nat Methods 11, 600-601 (2014). 1. Fig. 1. (a) Schematic illustration of the 3C2P3A-DSLM setup. (b) The spectrum, pulse width and pulse sequence diagram of 780 nm fiber laser. (c) The spectrum, pulse width and pulse sequence diagram of 1050 nm fiber laser. Fig. 2. 3C2P3A-DSLM three-color imaging of C. elegant. (a)-(c) Single-channel images of DAPI labeled chromosome, GFPtransgenic cytoplasm and dsRED-transgenic motor neuron, respectively. (d) The 3D reconstruction and three-channel merged image of (a)-(c). Scale bar: 20 m. Fig. 3. Two-color 3D fast deep-tissue imaging of pancreas islet and blood vessel in zebrafish. Vascular epithelial cell is transgenic with transgenic with Tg (kdrl:RFP) and showed in Red color. -cell is transgenic with Tg (kdrl:EGFP) and showed in Green color. Scan bar: 20 m");sQ1[361]=new Array("../7337/0721.pdf","Easier and Safer Biological Staining: High Contrast UranyLess Staining of TEM Grids using mPrep/g Capsules","","721 doi:10.1017/S1431927615004407 Paper No. 0361 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Easier and Safer Biological Staining: High Contrast UranyLess Staining of TEM Grids using mPrep/g Capsules Benmeradi N1,2, Payre B2 and Goodman SL3 Delta Microscopies, 22, B route de saint Ybars, La c�te blanche, 31190, Mauressac, France Universit� Toulouse, CMEAB Facult� Medecine, 118 route Narbonne, 31062, Toulouse, France 3. Microscopy Innovations LLC, 213 Air Park Rd, Suite 101, Marshfield, WI, 54449, USA 2. 1. Uranyl acetate (UA) has been used for decades in life science electron microscopy as a positive and negative stain [1, 2]. But due to recent regulations (especially in Europe and Japan) there has been considerable effort to find less toxic and non-radioactive replacements, including stains based on Oolong tea extracts [3], platinum blue [4], and gadolinium [5]. This report demonstrates a safe and nonradioactive stain that provide the contrast, broad utility, rapid staining, and ease of use of UA. UA is commonly followed by lead citrate [1,6] in a multistep protocol that also includes multiple rinses. This protocol requires extensive forceps handling to transfer grids into and out of grid boxes, onto and between stains and rinses, and back into storage. Due to their fragility, it is difficult to not damage some grids during all this handling. Moreover when several grids are prepared they are easily mixed-up and obtaining identical stain timing is difficult, thereby reducing experimental reproducibility. We demonstrate a new low-toxicity "Uranyless" stain developed by Delta Microscopy with Chromalys (France) that is composed of lanthanides (La, Dy, Gd) and other compounds that provide a high affinity to stain biological structures at a near neutral 6.4-6.8 pH to not disrupt sensitive structures. We combine UranyLess with mPrep/g capsule processing to make staining easier, more efficient and reproducible. Each mPrep/g capsule holds 1 or 2 grids and connects to lab pipettors to precisely deliver reagent to the grids held within. Capsules connect individually or can be stacked together onto single or multi-channel pipettors to enable staining any desired number of grids (Fig. 1). Since grids are stored and stained in the same capsule, this eliminates the potential for loss, damage or mix-up. Reproducibility is enabled since any number of grids can be prepared simultaneously with the same protocol, or even with multiple protocols by simultaneously drawing reagents from different microtiter wells in one 96-well plate. Tissues were prepared with a standard protocol: 2% glutaraldehyde in 0.2M pH 7.4 PO4 buffer for 2 hrs, buffer rinses, 1 hr buffered 1% OsO4 postfix, serial ethanols, and embedding in Epon/Araldite. 80 nm sections on 200 mesh Cu grids were inserted into mPrep/g capsules. The grids were stained by pipetting into mPrep/g capsules: UranyLess 2 minutes, 3 water rinses, lead citrate 2 minutes, 3 water rinses, and air drying in the capsule. Negative stain specimens were prepared by inserting carbon-coated Formvarfilmed 300 mesh grids into mPrep/g capsules, pipetting in bacteriophage T6 suspension and holding for 1 minute, contrasting with UranyLess for 1 minute, and then removing the mPrep/g capsules and blotting grids in the capsule to remove excess water prior to imaging with an Hitachi HT7700 at 80 keV. UranyLess and mPrep/g capsules provide rapid high contrast positive staining of animal and plant tissues (Figs 2-3), and high contrast negative staining (Fig 4) comparable to UA without radioactivity and with less toxicity. The mPrep/g capsules also provide a sealed staining environment that eliminates the need for hydroxide pellets to reduce lead precipitation [1]. In summary, mPrep/g capsule grid staining with Uranyless and lead citrate staining provides ease, quality and reproducible results. What color is UR-LESS? If its not yellow I will need to reshoot these images. Microsc. Microanal. 21 (Suppl 3), 2015 M&M Abstract 2015 Uranyless and mPrep 722 1a 1b 2a g g 1 �m 2b 2c 2d 5 �m 2e 3 ??? nm 500 nm 4 2 �m 1 �m 1 �m 500 nm Fig 1: a) Two mPrep/g capsules (g) stacked onto a single channel pipettor for staining 2-4 grids. b) Eight labeled mPrep/g capsules on 8-channel pipettor for staining 8-16 grids, from reagents in microtiter plate. Fig 2: Animal tissues: a) kidney, b) intestine, c) ovarian follicle, d) myelinated neuron, e) hepatocyte. Fig 3: Spinach leaf thylakoid plant tissue. Fig 4: Negative stained T6 bacteriophage. ---------[1] JJ Bozzola, LD Russell, "Electron Microscopy" 2nd ed, (Jones and Bartlett, Boston) p.124-9. [2] MA Epstein, SJ Holt. J Cell Biol 19 (1963), p. 335-6. [3] S Sato, et al. J Microsc 229(1) (2008), p. 17-20. [4] S Inaga, et al. Arch Histol Cytol. 70(1) (2007), p. 43-9. [5] M Nakakoshi, et al. J Electron Microsc (Tokyo). 60(6) (2011), p. 401-7. [6] Reynolds ES. J Cell Biol 17(1) (1963): p. 208�12.");sQ1[362]=new Array("../7337/0723.pdf","HSV-1 Scaffolding Protein Bubbles Readily in the Absence or Presence of DNA, Allowing its Localization in Immature and Mature Nucleocapsids","","723 doi:10.1017/S1431927615004419 Paper No. 0362 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 HSV-1 Scaffolding Protein Bubbles Readily in the Absence or Presence of DNA, Allowing its Localization in Immature and Mature Nucleocapsids Dennis C. Winkler, William W. Newcomb, Anastasia Aksyuk, Weimin Wu, Naiqian Cheng, and Alasdair C. Steven Laboratory of Structural Biology Research, National Institute of Arthritis and Musculoskeletal and Skin Diseases, National Institutes of Health, Bethesda, MD 10892 USA. The herpes simplex virus 1 (HSV-1) capsid is an immense macromolecular assembly, ~ 1250 � in diameter and composed of thousands of protein subunits of six different kinds. Their cumulative mass amounts to ~ 200 MDa and the capsid houses a 100 MDa genome. The assembly pathway has much in common with those of tailed bacteriophages, with which HSV shares a (distant) evolutionary relationship [1]. It starts with formation of a procapsid which matures via a major structural transformation, preceded by proteolysis and coupled to DNA packaging. A-capsids are mature but empty; B-capsids are mature but retain a shrunken scaffolding shell; C-capsids are filled with DNA and have ostensibly expelled the scaffolding shell � Fig. 1 [1]. In this study we are applying "bubblegram imaging" [2, 3] to the various kinds of HSV capsids. Among other goals, we aim to determine whether there are any proteins present in the DNA inside the C-capsid, as there are in many bacteriophages. Capsids were produced and isolated as previously described [4]. Cryo-EM was performed on a CM200FEG electron microscope operated at 120 kV, recording "dose series" of micrographs with 1-second exposures recorded 5 seconds apart. Each exposure imparted 15 electrons/�2. A field of purified procapsids is shown in Fig. 2 for the first exposure (left) and the ninth exposure (right). In the latter panel, the particles show extensive bubbling in their inner (scaffolding) shell, whereas the outer shell (capsid) shows only blurring, not bubbling. These observations indicate that the scaffolding protein preUL26.5 is particularly bubbling-prone. They also demonstrate that DNA is not needed to elicit bubbling. A corresponding image pair is shown in Fig. 3 for a mixed population of Acapsids, B-capsids, and C-capsids. The A-capsids blur but show no bubbling. The B-capsids bubble internally, confirming that their internal density consists mainly if not exclusively of the scaffolding protein UL26.5 processed by removal of a C-terminal peptide [1]. In contrast, C-capsids show small bubbles distributed throughout their interior. A plausible (but not the only) candidate is residual scaffolding protein. Originally there are ~ 1800 copies per procapsid and, although the large majority are expelled during DNA packaging, there may be enough left to generate the observed bubbles. In this scenario, these proteins are no longer organized in a spherical shell but are re-distributed throughout the C-capsid interior. References [1] G. Cardone et al in "Viral Molecular Machines", eds. M.G. Rossmann and V.B. Rao (Springer, New York) p. 423. [2] W. Wu et al, Science 335 (2012) p. 182. [3] N. Cheng et al, J Struct. Biol. 185 (2014) p. 250. [4] F. Booy et al, Cell 8 (1991) p. 1007. [5] This research was supported by the Intramural Research Program of the National Institute of Arthritis and Musculoskeletal and Skin Diseases of the National Institutes of Health. Microsc. Microanal. 21 (Suppl 3), 2015 724 Figure 1. Central section of an averaged tomographic reconstruction of the procapsid, showing the full thickness of the inner shell and (schematically) the elongated conformation of preUL26.5. Also, schematic cross-sections of the four capsids [1]. Scale bar = 20 nm. Figure 2. Cryo-micrographs of a field of procapsids, before bubbling (A, first exposure) and after initiating bubbling (B, ninth exposure). Scale bar = 100 nm. Figure 3. Cryo-micrographs of a mixed field of A- , B- and C-capsids, before bubbling (A, first exposure) and after initiating bubbling (B, thirteenth exposure). Scale bar = 100 nm.");sQ1[363]=new Array("../7337/0725.pdf","Improving the Reliability, Ease, and Efficiency of Section Staining in a Diagnostic Laboratory with the mPrep/g System for TEM Grid Processing","","725 doi:10.1017/S1431927615004420 Paper No. 0363 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Improving the Reliability, Ease, and Efficiency of Section Staining in a Diagnostic Laboratory with the mPrep/g System for TEM Grid Processing Craig Radi 1 1 Wisconsin Veterinary Diagnostic Laboratory, Madison, WI. In a clinical diagnostic lab it is important to obtain results efficiently. With Transmission Electron Microscopy (TEM), a common problem is inconsistent staining of tissue sections that can lead to inefficiency. Poor staining can substantially delay diagnoses due to the time to prepare more sections, stain more grids, and then repeat TEM imaging. Most labs have a favorite method, with many using a variation of droplet staining [1]. We recently learned of an entirely new method based on mPrep/g (microscopy preparation for grids) capsules that can simultaneous stain 16 or more grids. In the present study we compare the mPrep/g method to our droplet method. This study's droplet method used a pipet tip box covered with Parafilm that was slightly dimpled to provide a depression to hold stain droplets. Grids were placed section-side down on droplets of 2% aqueous Uranyl Acetate (UA), for 10 minutes, picked up with forceps, and rinsed with a water stream from a transfer pipette. They were then placed on droplets of Reynolds Lead Citrate (LC) for 10 minutes and covered. The grids were then rinsed with .002 Normal NaOH, followed by a water stream rinse, blotted with filter paper, and stored until TEM examination. Two problems that come up using this method at times is damage to the sections as you pick up grid between staining and rinsing and having to add more rinses to the sections. With the mPrep/g method, 1 or 2 grids are placed directly into mPrep/g capsules at the microtome, into tapered grid slots that securely hold them even if a capsule is dropped. For staining, the capsules attach like pipette tips to lab pipettors. This study used a 200 l 8-channel Gilson Pipetman to enable staining of 16 grids in 8 mPrep/g capsules. Grids were stained from reagents held in a 96-well plate (Fig 1). The first row of the 96-well plate held 40 l of UA, the next 3 rows held 400 l of water, then 40 l of LC, then 3 more water rows with 400 l of water. 8 mPrep/g capsules holding 2 grids each were attached to the 8-channel Pipetman, and dipped into the bottom of the UA reagent row. 35 l of UA was then drawn into the capsules and held for 10 minutes. The UA was then expelled back into this row, and the capsules/pipette were moved to the first water row. In this row, 40 l of water was rapidly drawn up and down 30 times (taking about 60 seconds) to rinse and agitate, and then repeated in the next two water rows. Then LC stain was drawn in and held for 10 minutes, followed by expelling and another 2x15 water rinses. The capsules were then rested on filter paper to drain off water, and removed and stored in the capsule storage box (which they came in) until TEM imaging. Figures 2a and 2b show images from two grids prepared by our standard droplet method. For diagnostic analyses these images are difficult to see fine detail due to some precipitates. Figures 3a and 3b (from the same tissue blocks) were prepared using mPrep/g capsule staining. Stain intensity is similar but there is a noticeable reduction of precipitate. Thus, the mPrep/g staining provided more reliable images. While it is possible to re-rinse the poor droplet prepared grids to Microsc. Microanal. 21 (Suppl 3), 2015 726 reduce the precipitate, this would substantially decrease efficiency and may damage the grid. In addition to providing more consistent results, mPrep/g processing was also easier and more efficient. With mPrep/g processing, 16 grids were only handled when inserted into the capsules (and when put in the TEM). They were never damaged during staining, and since the capsules were labeled they were much easier to document. Further, since 16 grids were prepared simultaneously, the overall time and effort required was reduced by an order of magnitude since the time to prepare, store, document and stain each grid took only seconds compared to minutes using the droplet method [2]. 1. 2a. 2b. 3a. 3b. Figure 1. mPrep/g capsules on 8-channel Pipetman, and a 96 well plate used to hold reagents. Figure 2. a) Brain and b) Placenta tissues stained using the droplet method. Figure 3. a) Brain and b) Placenta tissues stained using the mPrep/g system method. [1] J Bozzola, L Russell, "Electron Microscopy: Principles and Techniques for Biologists", Jones and Bartlett Pub. MA, 1998 p. 120. [2] Thanks to Steven Goodman and Tom Strader of Microscopy Innovations for suggestions and assistance with protocol development.");sQ1[364]=new Array("../7337/0727.pdf","Capabilities of a New Sample Preparation Method for Probing Polymer Penetration and Localization into Wood Without Embedding","","727 doi:10.1017/S1431927615004432 Paper No. 0364 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Capabilities of a New Sample Preparation Method for Probing Polymer Penetration and Localization into Wood Without Embedding Brian Dorvel1, Praveenkumar Boopalachandran1, Andrew Bowling2, Xue Chen3, and Steve King3 1 2 Dow Chemical Company, Core R&D- Division of Analytical Sciences, Freeport, TX, U.S.A 77541 Dow Agrosciences, Advanced Technology Development, Indianapolis, IN, U.S.A 46268 3 Dow Chemical Company, Industrial Solutions R&D, Freeport, TX, U.S.A 77541 The changes in the wood structure after modification have been an interest in the wood industry for wood treaters, biologists, and botanists. However, due to the complex wood structure, it is always a challenge to obtain a high quality image of the modified wood with high resolution, due to the wood cutting, polishing, and staining methods. A description of the results using this new preparation technique to observe and measure polymer penetration and morphology in the wood will be presented. Typically, the wood needs to be embedded since it is porous and tough to cross section without fixing the polymer in place, but several polymers are miscible in typical embedding media and may disrupt the localization of the polymer modified wood. Razor blades or normal diamond microtomy knives for sample preparation can also cause sample damage and compression, affecting the resolution. To minimize compression in biological samples, several methods have been employed but mainly using vitreous ice or cryomicrotomy [1][2]. The new sample preparation technique provides excellent resolution without the necessity for cryo temperatures. Figure 1 shows the comparison of the new method (Figure 1A) to cutting with a normal diamond microtomy knife (Figure 1B) and finally a razor blade (Figure 1C) on a piece of Southern Pine without polymer impregnation or embedding. The cross sections of the wood in both normal diamond knife microtomy and the razor blade cannot resolve the internal structure of the wood inside the cell wall. However, the new method provides a smooth sectioning of the wood which shows the middle lamella and the intracellular contents. Throughput is also increased since typical embedding methods may take from 12 hours to over 3 days to infiltrate samples. Several analytical methods may now be utilized including electron microscopy, optical microscopy, and spectroscopy to probe polymer localization in the wood. Figure 2 demonstrates the utilization of some of these methods for probing the morphology and localization of polymer in the wood. By staining the wood with ruthenium tetroxide (to provide contrast) and using the new sample prep method, we can distinguish the untreated wood (Figure 2A, inset) from the polymer rich domain areas inside the wood (Figure 2B, inset), which have reacted with the ruthenium. A second example revolves around fluorescence microscopy. Since wood contains many phenolic compounds, it has an intrinsic autofluorescence in the green and blue wavelengths in the spectrum, and changes in these wavelength intensities may indicate areas with polymer penetration [3]. Figure 2A shows an untreated piece of wood with a polymer treated piece of wood in Figure 2B, where the differences in brightness at the same exposure are clearly evident. In summary, the new preparation method has proven to be effective for studying polymer penetration into the components of wood without embedding. Analytical techniques which may have been limited due to embedding can be fully utilized to probe the internal polymer chemical interactions and localization. The application of this technique has contributed to the increased throughput of polymer Microsc. Microanal. 21 (Suppl 3), 2015 728 formulations screening by decreasing the sample prep time, all which has translated to bringing promising formulations to customers at an accelerated rate. References: [1] C.E. Hsieh, M. Marko, J. Frank and C.A. Mannella: Electron tomographic analysis of frozenhydrated tissue sections. Journal of Structural Biology 138, pp. 63-73, 2002. [2] J. R. McIntosh, "Electron Microscopy of Cells: A new beginning of a new century," The Journal of Cell Biology, vol. 153, pp. 25�32, 2001. [3] N. J. Chaffey, Wood Formation in Trees: Cell and Molecular Biology Techniques. 2003, pp. 86�97. Figure 1. Scanning electron microscopy (SEM) image comparison of untreated wood block sections cut with the new method (A), a normal diamond knife (B), and a stainless steel razor blade (C). All images were taken at 1000x magnification. A B polymer 50 m 50 m Untreated Polymer Treated Figure 2. A comparison of untreated wood (A) and polymer treated wood (B) sectioned using the new method, perpendicular to the grain of the wood. The SEM images (inset) show morphology differences from ruthenium staining between untreated and polymer treated wood. Corresponding fluorescence images are taken at the same exposure level of 100ms.");sQ1[365]=new Array("../7337/0729.pdf","Toward Quantitative In Situ TEM of Materials and Devices in Gases and Liquids at the Atomic Scale","","729 doi:10.1017/S1431927615004444 Paper No. 0365 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Toward Quantitative In Situ TEM of Materials and Devices in Gases and Liquids at the Atomic Scale Seiji Takeda1, Hideto Yoshida1 and Kentaro Soma1 1. Nanoscience and Nanotechnology Center, Institute of Scientific and Industrial Research (ISIR), Osaka University, 8-1 Mihogaoka, Ibaraki, Osaka 567-0047, Japan Among the various observation methodologies, in situ transmission electron microscopy (TEM) has recently advanced with technological developments such as aberration correction of lenses, fast digital cameras, and miniaturized specimen containers with various functions. These advances have enabled observation of a variety of phenomena in materials and devices at higher spatial and temporal resolution, especially in gases and liquids. The first crucial era of in situ TEM started in the 1970s using a large specimen room, which was only available in high-voltage TEM apparatus at that time, for examining plastic deformation and the interaction of point defects in metals and alloys. Electron irradiation was used to intentionally generate point defects in a crystalline thin foil, and then the kinetics of point defects was analyzed using quantitative high-voltage TEM [1], even though point defects have not been identified by direct TEM imaging even now. Environmental TEM (ETEM) was also introduced in highvoltage TEM apparatus by Swann and Tighe to reveal chemical reactions and others properties in situ [2]. Given the recent advances in TEM technology, it is now possible to investigate the essential static and dynamic characteristics of materials and devices by quantitative in situ TEM at the atomic scale. First, we briefly summarize our recent in situ ETEM studies of catalyst materials (Fig. 1) [3]. The conditions necessary for atomic-resolution ETEM using a Cs corrector of the objective lens have been described elsewhere [3]. We studied the gold nanoparticulate catalysts Au/CeO2 and Au/TiO2. It is wellknown that gold, the most stable metallic element, shows remarkable catalytic activity for CO oxidation even below room temperature. Haruta gold catalysts were prepared using the deposition precipitation method, and exhibited high catalytic activity at room temperature. Gold atoms on a crystalline gold surface are frequently displaced by electrons with energy higher than 400 keV. Therefore, ETEM observation was performed with electrons of 80, 200, and 300 keV to suppress the electron-irradiation effect. Systematic acquisition along with both numerical and statistical analyses of the ETEM imaging data led to the intrinsic catalyst structure in the reaction environments. The quantitative analyses further indicated that the activation sites of oxygen molecules at room temperature are most likely to be at the perimeter interface between gold nanoparticles and metal oxide supports. More importantly, the perimeter interface is not structurally rigid. Glimpse of gas molecules that interact with the surface of a gold nanoparticle is now possible with recently developed ETEM. In electron-irradiation-sensitive materials and devices, intrinsic phenomena may be masked by electron irradiation. It is helpful to systematically acquire in situ TEM data as a function of electron current density , electron dose D, and electron energy E. In several applications, the intrinsic phenomena and/or structures can be deduced at the atomic scale by extrapolating the systematically acquired in situ TEM data to both zero and D [3]. It is now realized that another crucial era of in situ TEM has started [4]. For quantitative in situ TEM of time-dependent phenomena, for instance dynamic atomic motions in a chemical reaction, quantitative evaluation and removal of the electron-irradiation-induced phenomena that may appear in the background in in situ TEM data is required. The statistically random displacement of atoms by electron irradiation probably differs from the strongly correlated displacement Microsc. Microanal. 21 (Suppl 3), 2015 730 of atoms on the surface in various reaction environments. Technological development of a system for automated data acquisition and analysis of in situ TEM imaging combined with electron energy loss spectroscopy data, as is conceptually proposed in Fig. 2, is needed to further stimulate the ongoing in situ TEM of materials and devices. References: [1] M. Kiritani and H. Takata, J. Nucl. Mat., 69�70 (1978), 277. [2] H. M. Flower, N. J. Tighe and P. R. Swann in "High Voltage Electron Microscopy: Proceedings of the Third International Conference", eds. P.W. Swann, C. J. Humphreys and M. J. Goringe, (Academic Press, London, 1974). [3] S. Takeda et al, Ultramicroscopy (2015) doi:10.1016/j.ultramic.2014.11.017 and references therein. [4] For instance, P. A. Crozier and T. W. Hansen, MRS Bulletin 40 (2015), 38. [5] The authors acknowledge funding from Japan Society for the Promotion of Science Grant-in-Aid for Scientific Research (A) (No. 25246003) and Specially Promoted Research (No. 19001005). Figure 1. (a) Gas molecules interacting with the catalytic gold surface. (b) Instantaneous structure of the gold catalyst under a reaction condition. Reproduced from [3] with permission from Elsevier. Figure 2. Possible procedure for observing the intrinsic processes of materials and devices in real environments by in situ TEM. The procedure will be most useful in microscopy and diffraction methodologies using probes other than electrons.");sQ1[366]=new Array("../7337/0731.pdf","Recent Progress with AC E(S)TEM and Application to Single Atom Catalysis","","731 doi:10.1017/S1431927615004456 Paper No. 0366 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Recent Progress with AC E(S)TEM and Application to Single Atom Catalysis Pratibha L Gai1,2,4, Leonardo Lari1,2, Michael R Ward1,2, Thomas Martin1,2, Robert W Mitchell1,2,4 David Lloyd1,2, Alec LaGrow1,2, Ian Wright1,2, and Edward D Boyes1,2,3. York JEOL Nanocentre1 and Departments of Physics2, Electronics3 and Chemistry4, University of York, York, YO10 5DD, UK. The development of atomic resolution ETEM [1,2] and now of the E(S)TEM [3,4,5] with 0.1nm image resolution in both modes has required managing multiple scientific and engineering issues. The projects have built on the foundations laid by earlier contributors to the field, most notably Hashimoto and coworkers [6] and Swann [7], with lower resolution TEM imaging developments which nevertheless supported outstanding science. Notably this included studies by Swann of the reduction of haematite to iron [8] and the untangling by Gai [9,10] of the critical contributions of surface defects in the oxidation catalysis of hydrocarbons. The latter work led to a redefinition of the basic criteria for the design of the catalyst, understanding of the commercial process mechanisms and economic operating parameters [11]. With the recent in-house development at York of the novel E(S)TEM system, we have been able for the first time to combine controlled hot stage temperatures and a gas environment around the sample with full (AC) STEM capabilities. These include single atom sensitivity in HAADF imaging, so far down to Z=29, wide angle electron diffraction and full EDX analyses in a double (TEM + STEM) aberration corrected version of the JEOL 2200 FS (S)TEM. The new functionality has been added without reducing the core capability of the original instrument and in important ways, including the vacuum and pumping systems, augmenting it. These developments achieve the goal of continuous dynamic in-situ studies of gas-solid chemical reactions with atomic resolution. There is access to key intermediate phases which may be metastable with respect to reaction conditions of gas and temperature, as well as being susceptible to side reactions if exposed to air during sample transfer for ex-situ studies. Surface science studies are usually either at low gas pressures (>0.1Pa is described as `high pressure' in this literature [12]) and/or uses discontinuous dosing. The ESTEM operates with continuous gas flows and pressures at the sample of a few Pa to a few mbar with even 1Pa corresponding to a gas supply of 104 monolayers per second with the actual supply also depending on the sticking coefficient. The York system is also compatible with enclosed specimen holder based thin window cells for higher pressures or hydrated atmospheres. The `open' architecture of the ESTEM with multiple added stages of powerful differential pumping across a series of beam-line apertures supports the >109 pressure differential between the specimen and the FEG. It also separates the specimen holder design from the gas supply system and supports use of the full existing range of specimen holders which are available. These now include double-tilt hot stages for greater flexibility for proper crystallographic analyses for imaging, diffraction and other forms of analysis down to the single column level. It has been possible to modify the original Gatan 628 furnace hot stage design, with the considerable advantage of retaining the 3mm disc specimen capability [13], and replace the electronics for 0.1nm resolution and single atom imaging but this can be achieved only after a soak time of several minutes and high magnification area retention requires dexterity in control inputs [Fig.1]. On the other hand, MEMs technology provides high lateral stability, both short and long term, and is ideal for particulate catalyst samples placed directly onto the filmed windows, ones custom covered with the correct chemistry or using the regular catalyst support particles in a similar way. The geometry of both types of hot stage have been adjusted to support EDX analysis; with the furnace holders at up to ~400�C and the MEMS ones demonstrated to >650�C [Figs.2,3]. Microsc. Microanal. 21 (Suppl 3), 2015 732 Initial applications of the new system include investigations of the incidence of single atoms on the support between nanoparticles or pre-particle rafts [Fig. 1] and their relevance to catalyst performance [4,5]. Single atoms may be expected to have different properties to those embedded on sites making up crystalline nanoparticle surfaces, or associated with rafts, and it is prudent for the potentially highly reactive single atoms to be treated and analyzed in-situ under controlled reaction conditions, or at least preservation conditions, rather than after ex-situ reaction and transfer through air into a conventional high vacuum electron microscope. Limits are set on the method by the potentially invasive e-beam; to varying degrees in TEM and STEM modes, as a function of the gas atmosphere and it seems support. Au nanoparticle on C-film MEMS chip at 650�C in ESTEM Pt/Pd NP 2nm Pd Pt EDX in ESTEM Fig.1: Single atom and raft imaging of Pt under reaction conditions (Pt/C/H2); atoms imaged at 0.11�0.01nm diam in HAADF ESTEM Fig.2: Au nanoparticle EDX at 650�C Fig.3: EDX data [14] recorded in the ESTEM from a core-shell nanoparticle References [1] P L Gai et al, Science (1995) 267, p661 [2] E D Boyes and P L Gai, Ultramicroscopy (1997) 67, p219 [3] P L Gai and E D Boyes, Microscopy Research and Technique (2009) 72, p153 [4] E D Boyes, M R Ward, L Lari and P L Gai, Annalen der Physik, (2013) 525, p432 [5] P L Gai, L Lari, M R Ward and E D Boyes, Chem. Phys. Letts. (2014) 592, p335 [6] H Hashimoto, T Naikui, T Etoh and K Fujiwara, Jap. J. Appl. Phys, (1968) 7, p946 [7] P R Swann and N Tighe, Jernkont. Ann, (1971) 155, p497 [8] P R Swann and N Tighe, Metall. Trans. B Process Metallurgy, (1977) 8, p479 [9] P L Gai, B C Smith and G Owen, Nature, (1990) 348, p430 [10] P L Gai, J Sol State Chem, (1983) 49, p25 [11] P L Gai and K Kourtakis, Science, (1995) 267, p661 [12] G Somorjai, R L York, D Butcher and J Y Park, Phys. Chem. Chem. Phys, (2007) 9, p3501 [13] J Sagar, L Fleet, M Walsh, L Lari, E D Boyes, O Whear, T Huminiuc and A Hirohata, Appl. Phys. Letts, (2014) 105, p032401 [14] M R Ward, T Hyde, E D Boyes and P L Gai, Chem. Cat. Chem, (2012) 4, p1622 Support from the University of York and the UK Engineering and Physical Sciences Research Council (EPSRC) through Critical Mass Grant EP/J018058/1 and other funding is gratefully acknowledged.");sQ1[367]=new Array("../7337/0733.pdf","XEDS Performance of Atmospheric Membrane Holders in the AEM","","733 doi:10.1017/S1431927615004468 Paper No. 0367 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 XEDS Performance of Atmospheric Membrane Holders in the AEM Nestor J. Zaluzec Electron Microscopy Center, NST Div, Argonne National Laboratory, Argonne, Il, USA The use of SiNx windows has been well established as a technology [1] which can be readily employed to observe microstructural evolution during in situ studies in the TEM/STEM as an alternative to experiments using a differentially pumped environmental TEMs. In previous studies [2-3] using SiNx membrane windows in liquid cells it was shown that the penumbra of the holder was the principle limitation in the use of X-ray Energy Dispersive Spectroscopy (XEDS). This was principally due to the cover/lid design, which can partially or completely block x-ray signal detection. The modifications which were established for that work have recently been extended to gaseous holder configurations. In this work FEI Tecnai F20, and CM200F TEMs equipped with windowless SDD systems from EDAX (Apollo) were used to evaluate the performance of an implementation of this technology on a Protochips Atmospheric Holder both for simple spectroscopy as well as mapping at nominally 1 ATM. Although the penumbra of the holder has been minimized as in the liquid cell design, next in importance is the penumbra created by Si frame, which illustrated in Figure 1. Shown schematically is the 300 �m tall Si frame which overshadows the SiNx window area. The penumbra angle can be simply calculated by noting the beam position D with respect to the Si side wall of height H i.e. = arctan (H/D) . To measure the relative effect of the Si side wall penumbra for a given location, we investigated a 50 nm thick SiNx window which was covered by a uniformly thick Au film the combination of which was supported on 300 �m Si frame sandwich (Figure 2a) with a 5 �m air gap. By tilting the holder to the optimum angle unshadowed angle dictated by the holder penumbra [4], we then measured the AuL Intensity as a function of position across the width (Y) and length (X) of the SiNx film. In Figures 2b and 2c shows the results while traversing the parallel (Y) and perpendicular (X) directions relative to the tilt axis as indicated. In 2b, one can see the relatively uniform intensity profile across the Y axis of the chip, indicating minimal shadowing, while traversing the X axis one, not surprisingly, sees the top half of the window has little to no shadowing while the bottom half gets progressively worse as one approached the Si sidewall. Of course with dual XEDS systems located on either side of the tilt axis the penumbra effect on the each side can be minimized by simple tilting of the holder toward the oppositely oriented detector. In figure 3a and 3b we show the effects of multiple scattering in the 5 �m thick air gap obtained by traversing across an edge into fortuitous holes in the Au metal film with the film on first the top then the bottom window. The gas path introduces a significant scattering profile and decreases the spatial resolution, due to an electron skirt effect. On the top window, the Au signal drops to near zero, but the skirt at the bottom window introduces a significant background level. Indicating that for optimum XEDS and/or STEM the material of interest should be on the top window, while for EELS and/or TEM the materials should be on the bottom window. Finally in figure 4, we confirm that high resolution (< 10 nm) hyperspectral data sets can be obtained at 1 ATM with a 5 �m air gap when the region of interest is located on the appropriate top window. References: [1] de Jonge N., Ross F.M., Nat.Nano 6 (2011), p. 695. [2] Zaluzec N.J. et al, Microsc. Microanal. 20 (2014), p. 323. Microsc. Microanal. 21 (Suppl 3), 2015 734 [3] Lewis E.A. et al, Chem Commun. 50 (2014), p. 9983. [4] Zaluzec N.J. et al, Microsc. Microanal. 21, Sup 2 (2014) these proceedings [5] Research supported by the U.S. DoE, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357 at the Electron Microscopy Center , NST Division of ANL.");sQ1[368]=new Array("../7337/0735.pdf","Opportunities and Challenges for In-Situ Characterization of Photocatalysts in Environmental TEM.","","735 doi:10.1017/S143192761500447X Paper No. 0368 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Opportunities and Challenges for In-Situ Characterization of Photocatalysts in Environmental TEM. Liuxian Zhang1, Qianlang Liu1, Benjamin K. Miller1 and Peter A. Crozier1 School for Engineering of Matter, Transport and Energy, Arizona State University, Tempe, AZ, USA Photocatalytic water splitting has been considered a promising technology for generating sustainable clean energy. Essentially, photocatalytic materials enable the process of converting and storing solar energy in the form of H2 molecules. It is now recognized that atomic level in-situ observations of catalytic materials are critical for understanding structure-reactivity relationships and deactivation processes such as photocorrosion. For photocatalysts, this requires that the system be observed not only in presence of reactant and product species but also during in-situ light illumination. Here opportunities and challenges associated with building a "photo-reactor" inside an environmental TEM (ETEM) are discussed. An optical fiber-based in-situ illumination system was developed and installed in an FEI Tecnai F20 ETEM with light intensity close to 10 suns [1]. The ETEM has a differential pumping system which allows up to 10 Torr gas pressure around the TEM sample. TiO2 based and Ta2O5 based UV absorption photocatalysts were chosen as model systems for in-situ and ex-situ characterizations. TiO2 is relatively abundant with simple crystal structures while Ta2O5 is very efficient with more complicated structures. The oxides are functionalized with Ni-NiO core-shell nanoparticles, which have been reported to be one of the most efficient surface co-catalysts when loaded onto TiO2 or Ta2O5. The combination of heat, light, and reactant gas stimuli allows the structural evolution of a photocatalyst to be followed through activation, reaction, and deactivation. Figure 1 shows C the activation of a Ni metal co-catalyst on TiO2 involving in-situ reduction at 400� in 1 Torr for 4 h. Interestingly, the Ni metal becomes coated with a thin amorphous TiOx layer during the reduction step. During the subsequent Ni oxidation step to form the Ni-NiO core-shell structure, this TiOx mixes with NiO layer. Additional experiments have confirmed that the Ni/TiO2 catalyst reduced ex-situ is also covered with TiOx layer. We have shown that subtle changes may take place under photoreaction conditions. For example, in-situ observation of TiO2 nanocrystals under water and light exposure shows the formation of amorphous surface layers less than 0.5 nm in thickness [2]. We are currently installing an optical fiber into an FEI Titan ETEM equipped with an image corrector to enable sub-angstrom in-situ imaging of photocatalysts under reaction conditions (using the objective aperture port (Fig. 2)). Using EELS and residual gas analysis to detect gas molecules inside the microscope, we are also pursuing operando TEM to directly detect hydrogen during in-situ illumination of the photocatalysts in a water vapor atmosphere. For water splitting it is important to understand the difference in deactivation mechanisms under liquid and vapor phase reaction conditions. As shown in Fig. 3, photo corrosion was observed ex-situ as a transformation of the Ni/NiO core-shell structure to a void-shell 1. Microsc. Microanal. 21 (Suppl 3), 2015 736 structure due to dissolution of Ni under liquid water and UV exposure from a xenon lamp [3-4]. This photocorrosion mechanism does not occur during in-situ experiments with 18 Torr of water vapor and up to 16 h UV light exposure. This highlights the difference between liquid and gas phase reaction conditions. For the low pressure conditions present in the differentially pumped ETEM, operando measurements are critical to connect structure and reactivity. For catalytic activity measurements from ex-situ liquid cell photoreactors, in-situ measurement should ideally be performed in a liquid cell environment to correlate structure and reactivity. Practical issues including electron beam effects, light source alignment, sample stability, etc. must also be understood and controlled. This presentation will discuss many of the technical and scientific opportunities and challenges for in-situ characterization of photocatalysts with emphasis on solar hydrogen production. References: [1] B.K. Miller et al, Microscopy and Microanalysis 2013, 19, 461-469 [2] L. Zhang et al, Nano Lett. 2013 13 (2), 679�684 [3] Q. Liu et al, Appl. Catal. B 2015 172-173, 58�64 [4] L. Zhang et al, J. Phys. Chem. (under review) [5] The support from US Department of Energy (DE-SC0004954) and the National Science Foundation (CBET-1134464) and the use of TEM at John M. Cowley Center for HR Microscopy at Arizona State University is gratefully acknowledged. Figure 1 Ni/TiO2 reduced in 1 Torr flowing H2 for 4 h in-situ Figure 2 Schematic drawing of the development of light illumination system on objective aperture rod for FEI TITAN Figure 3: a) Ni-NiO core-shell structure maintained after 16 h water vapor and light exposure in-situ. b) Ni-NiO core-shell structure transformed to void-shell after ex-situ exposure of 6 h light exposure in liquid water.");sQ1[369]=new Array("../7337/0737.pdf","Functionalization of Graphene","","737 doi:10.1017/S1431927615004481 Paper No. 0369 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Functionalization of Graphene M. F. Chisholm1, Jaekwang Lee1, Junjie Guo1,2, Zhiqing Yang1,3, S. J. Pennycook1, S. T. Pantelides4, and Wu Zhou1 1. 2. ORNL, Materials Science and Technology Division, Oak Ridge, USA. Taiyuan University of Technology, Taiyuan, China. 3. Chinese Academy of Sciences, Institute of Metal Research, Shenyang, China. 4. Vanderbilt University, Nashville, USA. Experimental results and calculations based on imaging of functionalized graphene with stable nanopores, the smallest rotor and `super' crown ethers will be presented and discussed. It is argued that the stabilization of nanopores in graphene is a critical issue that must be resolved to enable functionalization of graphene. Small holes in graphene are known to be subject to reconstruction and partial or total filling by diffusing carbon or other adatoms (known as the self-healing process) (1-3). Using medium-angle annular dark-field (ADF) imaging in an aberration-corrected scanning transmission electron microscope (STEM), we found that silicon atoms stabilize graphene nanopores by bridging the dangling bonds around the perimeter of the hole (4). STEM imaging was performed using a fifth-order aberration-corrected STEM (Nion-UltraSTEM100) with a cold field emission electron source. The microscope was operated at 60 kV to prevent knock-on damage to graphene. The angular range of the collected electrons in medium-angle ADF mode was 58�200 mrad semiangle. Theoretical calculations reveal the underlying mechanism for this stabilization effect is that Si atoms bond strongly to the graphene edge and their tetrahedral coordination preference (5, 6) forces adatoms to form dendrites sticking out of the graphene plane, instead of filling into the nanopore. All carbon dangling bonds in the nanopore seen in Fig. 1 are passivated by silicon atoms. Although a number of bonding configurations are observed, all the Si atoms are calculated to have binding energies larger than 5 eV. Smaller holes stabilized by silicon were also observed. Fig. 2 shows three silicon atoms replacing a single hexagonal ring of carbon atoms in a graphene layer. Under the electron beam, this silicon trimer rotates as a single unit in stepwise jumps while the surrounding carbon atoms remain fixed (7). Theoretical calculations indicate that the energy barrier to rotate the trimer is about 2 eV, corresponding to a tangential force of only 4.3 nN parallel to the graphene layer. This small force, provided by an electron beam to any of the three silicon atoms, generates a torque on the trimer and results in a rotation of 60� in about 140 femtoseconds and demonstrates that the ultimate miniaturization of a mechanical device (switch, oscillator, stirrer) down to a triangular arrangement of three atoms is possible. Oxygen atoms were imaged in graphene (Fig. 3) only at the edges of small holes in highly stable crown ether configurations within the 2D graphene layer (8). Oxygen and carbon in the ether configuration at graphene or graphene oxide zig-zag sheet edges have been proposed to exist but, until this investigation, have not been imaged by electron microscopy. Density functional calculations indicate that the crown ether configurations in graphene should exhibit selectivity for different cations depending on the crown ether ring size, a key property of individual crown ether molecules, but in graphene have the added property of being rigid and planar, prized features sought in preorganized receptors for selective ion binding. Calculations indicate that these "super crown ethers" provide unprecedented binding strength and selectivity. Thus, new supramolecular materials in which metal ions are trapped into arrays within the graphene plane are possible. These new materials should provide electrical, magnetic and optical functionality to graphene. Microsc. Microanal. 21 (Suppl 3), 2015 738 References: 1. A. Barreiro, F. B�rrnert, S. M. Avdoshenko, B. Rellinghaus, G. Cuniberti, M. H. R�mmeli, L. M. Vandersypen, Scientific Reports 3, 1115 (2013). 2. R. Zan, Q. M. Ramasse, U. Bangert, K. S. Novoselov, Nano Letters 12, 3936-3940 (2012). 3. L. Tsetseris, S. T. Pantelides, Carbon 47, 901-908 (2009). 4. J. Lee, Z. Yang, W. Zhou, S. J. Pennycook, S. T. Pantelides, M. F. Chisholm, Proceedings of the National Academy of Sciences 111, 7522-7526 (2014). 5. M. Yin, M. L. Cohen, Physical Review B 29, 6996 (1984). 6. P. M�linon, B. Masenelli, F. Tournus, A. Perez, Nature Materials 6, 479-490 (2007). 7. Z. Yang, L. Yin, J. Lee, W. Ren, H.-M. Cheng, H. Ye, S. T. Pantelides, S. J. Pennycook, M. F. Chisholm, Angewandte Chemie International Edition 53, 8908-8912 (2014). 8. J. Guo, J. Lee, C. I. Contescu, N. C. Gallego, S. T. Pantelides, S. J. Pennycook, B. A. Moyer, M. F. Chisholm, Nature Communications 5, 5389 (2014). 9. Work supported by the U.S. Department of Energy Office of Science, Basic Energy Sciences, Materials Sciences and Engineering Division and through a user project supported by ORNL's Center for Nanophase Materials Sciences, which is sponsored by the Scientific User Facilities Division, Office of Science, Basic Energy Sciences, U.S. Department of Energy. b c a Binding Energy (eV) b 1.5 1.0 0.5 0.0 14-crown-4 Magnetization (u B) Figure6 2. STEM ADF image of a silicon trimer in Figure 1. ADF image of a graphene nanopore graphene nanopore (a) before and (b) after one 4 stepwise rotation. [7] passivated with 17 Si atoms. [4] 2 0 Cu Ni Co Atom Fe Mn Cr e f Binidng Energy (eV) 1.5 1.0 3.0 2.5 B inidng Energy (eV ) c c a h d 2.0 i b 12-crown-3 c 2.4 2.0 1.6 1.2 0.8 0.4 18-crown-6 in graphene 14-crown-4 2.0 1.5 18-crown-6 molecule Li Na K Atom Rb Cs f Li Na K Atom Rb Cs Figure 3. (a) Experimental STEM-ADF image of crown ether in graphene. (b) Calculated structure overlaid on experimental image. (c) Calculated binding energy of metal ions in 18-crown-6 in graphene compared with of flexible 18-crown-6. [8] i");sQ1[370]=new Array("../7337/0739.pdf","Dynamics of Triangular Hole Growth in Monolayer Hexagonal Boron Nitride under Electron Irradiation","","739 doi:10.1017/S1431927615004493 Paper No. 0370 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Dynamics of Triangular Hole Growth in Monolayer Hexagonal Boron Nitride under Electron Irradiation Gyeong Hee Ryu1, Hyo Ju Park1 and Zonghoon Lee1,2 School of Materials Science and Engineering, Ulsan National Institute of Science and Technology (UNIST) and 2Center for Multidimensional Carbon Materials, Institute for Basic Science (IBS), Ulsan 689-798, S. Korea Defects are known to influence the intrinsic properties of materials. In graphene, topological defects have been shown to alter its chemical and physical properties and these have been extensively investigated. [1], [2] Such defects in hexagonal boron nitride (hBN) have been reported, but the formation mechanism of defects is not fully understood. Indeed, most reports have focused on defect formation and characterization [3], [4], [5] in localized regions of exfoliated hBN. At present, it is known that the edges of holes in hBN layers usually adopt zigzag and armchair-type configurations, with the former being more common. [3] Moreover, the zigzag configuration contains two different types of terminated edges due to the intrinsic heterogeneity of hBN, factors that have been shown to affect the material's intrinsic electrical properties.[7] However, the growth mechanisms of hBN defects are also not well established. The dynamics of triangular defects induced by electron beam irradiation were analyzed using aberration corrected TEM with a monochromator to elucidate the mechanism of defect growth in hBN monolayers. A large area monolayer of hBN grown by chemical vapor deposition at atomic resolution is shown in Figure 1. Figure 1a shows initial formation of triangular holes. After prolonged electron beam irradiation, grown triangular holes can be conformed as shown in Figure 1b. The hBN defect growth process was subsequently investigated. Efforts were then directed towards assessing whether the defects maintained their triangular shape after prolonged periods of electron beam irradiation. As summarized in Figure 2, the growth of a triangular hole appeared to be initiated by the removal of B and N atoms near the centers of the defect edges. This experimental observation indicated that triangular defects featuring edges with missing B and/or N atoms are more stable than those with atoms missing near a vertex. Regardless, under prolonged electron beam irradiation, the B and N atoms next to the vacancies were subsequently ejected in a manner that ultimately restored the overall triangular shape of the defect. Moreover, the triangular shape of the defects was maintained even after two holes merged together. In this study, the growth of triangular holes in hBN monolayers was observed using sequential TEM imaging. When a monolayer of hBN was subjected to electron beam irradiation, a vacancy formed initially and grew while maintaining a triangular shape. Such shapes were observed even when such holes merged. This mechanism for the growth of these defects appeared to involve the ejection of B and N atoms near the centers of the defect edges and also the ejection of bundles of atoms. 1 Microsc. Microanal. 21 (Suppl 3), 2015 740 References [1] Banhart, f., et al., ACS Nano 5 (2010), 26-41 [2] Girit, �. �., et al., science 323 (2009), 1705-1708 [3] Alem, N., et al., Phys. Rev. B 80 (2009), 155425. [4] Jin, C., et al., Phys. Rev. Lett. 102 (2009), 195505. [5] Meyer, J. C., et al., Nano Lett. 9 (2009), 2683-2689. [6] Yin, L.-C., et al., Phys. Rev. B, 81 (2010), 153407. [7] This work was supported by Nano Material Technology Development Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT and Future Planning (2012M3A7B4049807). This work was also supported by IBS-R019-D1. Figure 1. Formation and growth of large triangular holes in monolayer hBN by electron beam irradiation. The TEM images show (a) the initial defects and (b) an enlarged area of the same region. The scale bar is 2nm. Figure 2. Sequential atomic resolution images of monolayer hBN showing how the shape and orientation of the defects are maintained upon further growth. The scale bar is 2 nm.");sQ1[371]=new Array("../7337/0741.pdf","Cu Atoms Reknit the Graphene Structures","","741 doi:10.1017/S143192761500450X Paper No. 0371 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Cu Atoms Reknit the Graphene Structures Emi Kano1, 2, Ayako Hashimoto1, 2, 3, 4 and Masaki Takeguchi1, 3, 4 1. 2. Graduate School of Pure and Applied Sciences, University of Tsukuba, Tsukuba, Japan. Global Research Center for Environment and Energy based on Nanomaterials Science, National Institute for Materials Science, Tsukuba, Japan. 3. Surface Physics and Structure Unit, National Institute for Materials Science, Tsukuba, Japan. 4. Transmission Electron Microscopy Station, National Institute for Materials Science, Tsukuba, Japan. The interactions between metals and graphene have recently been studied in order to control the properties of graphene. Our knowledge about these metal-graphene systems are mainly based on theoretical calculations, except for Fe dimers in graphene vacancies [1] or some metal-mediated etching [2] observed by transmission electron microscopy (TEM). Theoretical studies have predicted that transition metal atoms in graphene vacancies have unique properties depending on its electronic state [3]. In contrast to most transition metals, Cu and Au atoms have filled d shells. These electronic states affect the catalytic properties. Bulk Au doesn't show any catalytic properties, but downsized Au nano-particles attract enormous attention as catalyst of CO oxidation. Bulk Cu is known to be the best catalyst for the synthesis of graphene by chemical vapor deposition, and Cu nano-particles are also able to synthesize carbon nanotubes, though these catalytic growth mechanisms are unclear. Atomic scale observations of the interaction between Cu and graphene could prove the predicted properties and gain understanding how Cu act as catalyst to create nano-carbon structures. Here we report the structures and dynamics of Cu atoms in a graphene sheet by aberration-corrected TEM (JEM-ARM200F, JEOL). The samples were obtained by transferring monolayer graphene onto in situ heating chips (E-chips for Aduro, Protochips). It was cleaned up by heating, and the clean surface area was about 200 � 200 nm. Cu was then deposited by ion beam etching system (PECS, Gatan). The accelerating voltage was 80 kV, and a temperature was kept at 150 �C with an in situ heating holder. When observed at room temperature just after the deposition, Cu didn't form crystals; they were dispersed with a lot of oxygen and hydrocarbons. The sample was heated at 150 �C to remove these contaminants. When we started the heating, some of the contaminants got away and Cu atoms aggregated with each other to form clusters or nanoparticles, as shown in Fig. 1a. Some Cu atoms remained on graphene with some contaminants, and then they substituted carbon in a graphene lattice when a high density of electron beam was irradiated. Atoms appeared white in Fig. 1b since it was taken at over-focus. More than 10 Cu atoms were embedded closely in graphene, where additional Cu and C had existed (indicated by red arrow in Fig. 1a). TEM movies were acquired with a speed of 1-2 frames/s to observe the dynamics of embedded Cu atoms in graphene vacancies. Current density was about 250 A/cm2 at Fig. 1b, which was enough to cause in-plane diffusions of Cu atoms in graphene, and was too large to observe atoms simply adsorbed on graphene because the out-plane diffusion barrier was more than ten times smaller than the in-plane one. FFT spots of the electron beam irradiated area gradually broadened as shown in Fig. 2b, compared with Fig 2a. Many small grains were formed which were surrounded by 5-7 defects (Fig. 2c). Defects created near Cu atoms moved to form grain boundaries. The area without irradiation didn't change the structure. Microsc. Microanal. 21 (Suppl 3), 2015 742 These results suggest that the diffusion of Cu atoms was promoted not by heat but by the electron beam irradiations. Figure 3a-b show the consecutive snapshots of the movie. One atom of Cu dimer, shown in Fig. 3a, replaced neighboring C atoms and changed the structure from Fig. 3a to Fig. 3b. The corresponding models are Fig. 3e and 3f, respectively. Since single Cu atom in single vacancy (Fig. 3g) seems most stable, many Cu atoms embedded in closely gradually diffused to be isolated. It is reported that many kinds of metals etch graphene [2]. Cu atoms, however, switch positions with neighboring C atoms (Fig. 3a-b) and reknit the graphene structures rather than etching. Even when small hole-defects were created by irradiation, Cu atoms immediately mended them (Fig. 3i-j). References: [1] He, Z. et al. Nano Lett. 14, 3766�3772 (2014). [2] Ramasse, Q. et al. ACS Nano 6, 4063�4071 (2012). [3] Krasheninnikov, a., Lehtinen, P., Foster, a., Pyykk�, P. & Nieminen, R. Phys. Rev. Lett. 102, 126807 (2009). [4] A part of this work was supported by "Nanotechnology Platform Project" of the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan. Figure 1. TEM images of Cu/graphene. (b) Magnified image of (a). White spots are Cu atoms. Figure 2. FFT of (a) pristine graphene and (b) reknitted graphene. (c) TEM image of b. Figure 3. (a-b, i-j) Consecutive snapshots of TEM movie. (c-d, k-l) smoothed images of a-b and i-j, respectively. (e-h) atomic models of Cu atoms in graphene vacancies (SV: single vacancy, TV: trivacancy, QV: quad-vacancy) shown in (a-d). Scale bar: 0.5 nm.");sQ1[372]=new Array("../7337/0743.pdf","VEELS Study of Boron and Nitrogen Doped Single Layer Graphene","","743 doi:10.1017/S1431927615004511 Paper No. 0372 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 VEELS Study of Boron and Nitrogen Doped Single Layer Graphene F.S. Hage1, D.M. Kepaptsoglou1, T.P. Hardcastle2, C.R. Seabourne2, A.J. Scott2, R. Brydson2, R. Zan3, J. Amani4, H. Hofs�ss4, U. Bangert5 and Q.M. Ramasse1 1. SuperSTEM Laboratory, SciTech Daresbury Campus, Daresbury, WA4 4AD, U.K. Institute for Materials Research, SCAPE, University of Leeds, Leeds, LS2 9JT, UK 3. Department of Physics, Faculty of Arts and Sciences, Nide University, Nide 51000, Turkey 4. II Physikalisches Institut, Georg-August-Universitat Gottingen, Friedrich-Hund-Platz 1, 37077 Gottingen, Germany 5. Department of Physics and Energy, University of Limerick, Limerick, Ireland 2. The recent years have seen a significant increase in research effort directed at all aspects of graphene science and technology. Despite its excellent properties, a lack of a controllable method for electronic structure modification has severely impeded any successful application of graphene in novel electronic devices. With the aim to overcome this limitation, functionalization through substitutional doping with boron and nitrogen atoms has been of particular interest (predicted to result in an effect akin to p and n doping in semiconductors, respectively) [1]. Successful doping of graphene by single boron and nitrogen atoms was recently demonstrated by means of low energy ion implantation [2, 3]. Low energy ion implantation does not only allow for controllable graphene functionalization, but is also compatible with modern semiconductor technology. The present work is concerned with measuring the effect of single atom substitutions on the valence electrons of graphene. This is done by means of simultaneous Valence Electron Energy Loss Spectroscopy (VEELS) and Annular Dark Field (ADF) imagining in the Scanning Transmission Electron Microscope (STEM). Two Medium Angle ADF (MAADF) images of single layer graphene patches are shown in Figs. 1a and b. In following with Bangert et al. [2], the increase (Fig. 1a) or decrease (Fig. 1b) in image contrast unambiguously identifies a single atom nitrogen (Fig. 1b) or boron (Fig. 1c) substitution in the respective graphene lattices. Corresponding loss spectra in Fig. 1c shows a clear increase (N) or decrease (B) in the graphene plasmon peak intensity, as compared to a spectrum from a pure graphene sheet. To a first approximation, the results indicate that the introduction of a single atom dopant results in a significant enhancement (N) or dampening (B) of the graphene electron plasmonic response. This interpretation is supported by a relative reduction (N) or increase (B) in peak width in the Zero Loss Peak (ZLP) subtracted spectra in the inset in Fig. 1c. As the substitutional boron atom is expected to induce a net valence electron deficit, these results seem in perfect agreement with an intuitive picture of the electronic structure rearrangement. Thus, dampening might be rationalized in terms of boron effectively disrupting the delocalized character of graphene valence states. In a broader context, the demonstrated modification of the graphene plasmon resonance hints at the possibility of tailoring future graphene based plasmonic devices down to the atomic scale. Further implications of the presented results will be discussed on the basis of theoretical modelling. The images in Figs. 1a and b were processed with a Double Gaussian image filter as described by Krivanek et al. [4, 5]. All measurements were carried out on a Nion UltraSTEM100 microscope operated at 60 kV. [6] Microsc. Microanal. 21 (Suppl 3), 2015 744 References: [1] LS Panchakarla, KS Subrahmanyam, SK Saha, A Govindaraj, HR Krishnamurthy, UV Waghmare and CNR Rao, Advanced Materials 21 (2009), p. 4726. [2] U Bangert, W Pierce, DM Kepaptsoglou, Q Ramasse, R Zan, MH Gass, JA Van den Berg, CB Boothroyd, J Amani and H Hofs�ss, Nano Letters 13 (2013) p. 4902. [3] Y Xu, K Zhang, C Br�sewitz, X Wu and HC Hofs�ss, AIP Advances 3 (2013) 072120 [4] OL Krivanek, MF Chisholm, V Nicolosi, TJ Pennycook, GJ Corbin, N Dellby, MF Murfitt, CS Own, ZS Szilagyi, MP Oxley, ST Pantelides and SJ Pennycook, Nature 464 (2010), p. 571. [5] OL Krivanek, MF Chisholm, MF Murfitt and N Dellby, Ultramicroscopy 123 (2012), p. 90. [6] SuperSTEM is the UK National Facility for aberration-corrected STEM and is funded by the UK Engineering and Physical Sciences Research Council (EPSRC). Figure 1. (a) Double Gaussian filtered [4, 5] MAADF images of substitutional single nitrogen (a) and boron (b) atoms in graphene. (c) VEEL spectra of nitrogen and boron implanted single layer graphene, compared to that of pure graphene. The inset shows the corresponding ZLP subtracted spectra normalized to the plasmon peak height.");sQ1[373]=new Array("../7337/0745.pdf","Electron ExitWave Reconstruction From a Single Defocused Image Using a Gaussian Basis","","745 doi:10.1017/S1431927615004523 Paper No. 0373 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electron Exit Wave Reconstruction From a Single Defocused Image Using a Gaussian Basis Konstantin B. Borisenko, Christopher S. Allen, Jamie H. Warner and Angus I. Kirkland Department of Materials, University of Oxford, Parks Road, Oxford, OX1 3PH, UK Aberration-corrected high resolution transmission electron microscopy (HRTEM) is a well established tool for studying the atomic structures of materials. Using a series of images collected with different foci it is possible to reconstruct the electron exit wavefunction [1,2]. The accuracy of this type of reconstruction, however, is limited by the ability of the sample to withstand radiation damage during the long acquisition time required for data collection. The usefulness of this method is hence limited for radiation sensitive materials. The signal to noise ratio in the reconstruction also depends on the number of images recorded in the focal series. Usually, 30 or more images are collected, which provides the necessary redundancy in the experimental data for reliable reconstruction of the exit wave phase and amplitude. Special steps are therefore required to reduce radiation damage and increase the accuracy of the reconstruction if such a method is to be applied to radiation sensitive materials. One approach is to use a lower radiation dose to record the focal series [3]. This, however, can lead to problems with aberration measurement and alignment of noisy image data. Another possible approach is to reduce the number of collected experimental data, in principle, down to a single image. However, one of the barriers to successful exit wave reconstruction is that in this case the number of variables in the restored complex exit wave, if each pixel is treated as a variable, is twice as large as the number of the recorded experimental data (number of pixels) in the image. The presence of objective lens aberrations and detector noise further complicates the analysis and can lead to multiple and unstable reconstructions. In the present work we present a new algorithm for exit wave reconstruction from a single defocused image of a weak-phase object and also a more general object. By representing the electron exit wave in a Gaussian basis it is possible to greatly reduce the number of variables needed to find a solution, in a way that is related to compressed sensing [4]. To illustrate this we compare the reconstructed experimental exit wave phase from graphene and silicon nitride using a Gaussian basis with the corresponding exit wave reconstructed using a conventional linear Wiener filter restoration from an extended focal series. In a weak-phase object approximation exit wave has a complex amplitude: W ( x, y ) Ae i ( x , y ) , where ( x, y ) is phase shift, and A is an amplitude, A = 1. We represent the exit wave phase as a sum of n symmetric Gaussian functions G as ( x, y ) G n (a n , bxn , b yn , c n , x, y ) , where the Gaussian functions n ( x bx ) 2 ( y b y ) 2 in two dimensions have general form G (a, b x , b y , c, x, y ) ae independent parameters, defining their position, height and width. 2c 2 , and are described by four Microsc. Microanal. 21 (Suppl 3), 2015 746 In case of coherent or partially coherent illumination the image intensity in the real space is obtained by computing the squared modulus of a convolution of the point spread function of the electron microscope 2 P with the electron exit wave, I ( x, y ) W ( x, y ) P ( x, y ) . The image intensity is therefore represented as a function of parameters of the Gaussians in the exit wave and the parameters of the point spread function of the microscope, or objective lens aberration function in reciprocal space. With known parameters for the lens aberration function, the parameters of the exit wave can be found by minimising the sum of squared differences between the simulated and experimental real space images, or alternatively, by solving an overdetermined system of non-linear equations, I ( x, y, an , bxn , byn , cn ) sim I ( x, y ) exp 0} The reconstructed exit wave phase from a single defocused image of a monolayer graphene sample using the representation in a Gaussian basis is presented in Figure 1. As the exit wave phase is sensitive to the three-dimensional structure of the layer, brighter regions discernible in the exit wave phase may represent out-of-plane ripples around the location of a dislocation dipole, indicated in the image. References [1] WMJ Coene, A Thust, M deBeeck and D VanDyck, Ultramicroscopy 64 (1996), p. 109. [2] WK Hsieh, FR Chen, JJ Kai and AI Kirkland, Ultramicroscopy 98 (2004), p. 99. [3] B Barton, B Jiang, C Song, P Specht, H Calderon and C Kisielowski, Microsc. Microanal. 18 (2012), p. 982. [4] EJ Cand�s and MB Wakin, IEEE Signal Proc. Mag. 25 (2008), p. 21. [5] Financial support from the European Union under the Seventh Framework Program under a contract for an Integrated Infrastructure Initiative (Ref. 312483-ESTEEM2) is gratefully acknowledged. Figure 1. Experimental image of monolayer graphene a) and the reconstructed exit wave phase using representation in a Gaussian basis b). A dislocation dipole is indicated by arrows in the image.");sQ1[374]=new Array("../7337/0747.pdf","Advanced electron microscopy study of fission product distribution in the failed SiC layer of a neutron irradiated TRISO coated particle","","747 doi:10.1017/S1431927615004535 Paper No. 0374 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Advanced electron microscopy study of fission product distribution in the failed SiC layer of a neutron irradiated TRISO coated particle Haiming Wen1, Isabella J. van Rooyen1, John D. Hunn2, Tyler J. Gerczak2, Charles A. Baldwin2, Fred C. Montgomery2 1. 2. Fuel Performance and Design Department, Idaho National Laboratory, Idaho Falls, Idaho, USA Fusion and Materials for Nuclear Systems Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee, USA Tristructural isotropic (TRISO) coated particle fuel has been designed for application in hightemperature gas-cooled reactors (HTGR). TRISO particles for the HTGR fuel development effort underway at Idaho National Laboratory (INL) and Oak Ridge National Laboratory (ORNL) consist of a two-phase uranium oxide-uranium carbide (UCO) fuel kernel, a carbon buffer layer, an inner pyrolytic carbon (IPyC) layer, a SiC layer, and an outer PyC (OPyC) layer [1]. The first in a series of irradiation experiments (AGR-1) clearly shows release of certain metallic fission products, e.g., Ag and Pd, through intact TRISO coatings, with Cs generally well retained [1]. No significant chemical interaction was observed between Pd and SiC for UCO TRISO coated particles, which retained Cs [2]. During post-irradiation examination of TRISO particles from AGR-1 Compact 5-2-3, gamma-activity measurements determined that Cs had been released by two specific particles. The fractional Cs and Ag releases from one of the two particles, Particle 523-SP01, were determined to be ~30% and ~85% respectively. Non-destructive three-dimensional X-ray tomography performed on this particle revealed that the SiC failure was localized to one specific region. Figure 1 (a)-(c) shows x-ray, optical, and scanning electron microscopy (SEM) images of the SiC failure in this particle. The failure mechanism was found to be SiC degradation through chemical reaction with Pd and U initiating at a SiC inner surface exposed by IPyC fracture [3]. To further examine the failure mechanism, fission product distribution and transport, scanning transmission electron microscopy (STEM) specimens were prepared using the focused ion beam (FIB) technique. Figure 1(d) shows the locations where STEM lamellae were sectioned by FIB. STEM and energy-dispersive X-ray spectroscopy (EDS) were conducted. In lamella 7, all precipitates close to the crack are rich in Pd and Si, ~63% of which contain a small atomic concentration of U. Approximately 63% of the Pd-Si and Pd-Si-U precipitates contain Cs, with concentrations ~0.20�1.43 at%. No Ag is evident in any of the precipitates. Ce and Cs have very close peaks in the EDS spectra; because the TRISO particle had no detectable Ce release and significant Cs release, all of the overlapping Cs and Ce peaks are considered Cs peaks. Further examination using EDS with better energy resolution is underway to obtain more accurate Cs concentrations. In lamella 8, most precipitates are composed of Pd and Si, with ~50% also containing a small atomic concentration of U. Approximately 36% of the Pd-Si and Pd-Si-U precipitates contain Cs, with concentrations ~0.20�1.99 at%. Only one precipitate (out of 36) rich in Pd and Ag was found, with 5.34 at% Pd and 2.02 at% Ag. Figure 2 shows representative STEM images of lamella 10. In Figure 2 (a)�(d), precipitates numbered and arrowed were subjected to EDS investigation. Precipitates in the SiC layer are Pd-Si-U, Pd-Si, Pd-U or Pd. Approximately 30% of these precipitates contain Cs, with concentrations usually in the range ~0.40�0.80 at%, highest concentration being 2.20 at%. No precipitates containing only Cs were Microsc. Microanal. 21 (Suppl 3), 2015 748 observed. Ag and/or Cd containing precipitates were identified by EDS point scans 1�8 and 15�16 at grain boundaries or triple junctions. The Ag concentration in these precipitates is 0.22�7.95 at%, and the Cd concentration is 0.20�1.31 at.%. Ag and/or Cd may exist by themselves or coexist with Pd. The number of Cs containing precipitates is significantly higher than that of Ag or Cd containing precipitates; however, Cs concentration in the precipitates is usually significantly lower than Ag or Cd concentration. Dark areas in the SiC layer (Figure 2(d)) were identified by EDS to be pure carbon. Bright precipitates in the pure carbon areas (17, 18 and 19 in (d)) were identified to be Pd2Si or PdSi. These results confirm chemical reactions between Pd and SiC, forming Pd2Si or PdSi and carbon. Note that Cs containing precipitates were seldom found and no Ag or Cd containing precipitates were found in other lamellae sectioned from other locations in the SiC layer of this TRISO particle. In summary, a TRISO coated particle with ~30% Cs and ~85% Ag release had a major crack across the buffer layer and the IPyC layer. Areas in the SiC layer close to the crack tip were corroded by Pd, forming pure carbon areas and Pd2Si or PdSi; such corroded areas may provide pathways for Cs migration. Ag and/or Cd containing precipitates were identified in the SiC layer only in locations close to the crack. References: [1] P.A. Demkowicz et al, Paper HTR2014-31182, Proc. of HTR 2014, Weihai, China, 27-31 Oct. 2014. [2] I.J. van Rooyen et al, Nuclear Engineering and Design 271 (2014), p. 114. [3] J.D. Hunn et al, Paper HTR2014-31254, Proc. of HTR2014, Weihai, China, 27-31 Oct. 2014. [4] This work was sponsored by the U.S. Department of Energy, Office of Nuclear Energy, under DOE Idaho Operations Office Contract DE-AC07-05ID14517. James Madden is acknowledged for the FIB sample preparation. Paul Demkowicz is thanked for his review of this paper. Figure 1. X-ray tomograph (a), optical micrograph (b), and backscattered electron composite image (c) showing localized degradation from Pd and U attack of the SiC layer where it was exposed by an IPyC crack; (d) SEM image showing locations where STEM lamellae were sectioned. Figure 2. STEM images of STEM lamella 10 shown in Figure 1(d). Precipitates numbered and arrowed in (a)-(d) were subjected to EDS analysis. Dark areas in (d) are carbon, and letters A-D indicate locations for EDS analysis. Ag and/or Cd are detected in EDS point scans 1-8 and 15-16.");sQ1[375]=new Array("../7337/0749.pdf","Post-Irradiation Examinations of Irradiation Creep Tested Zircaloy-2","","749 doi:10.1017/S1431927615004547 Paper No. 0375 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Post-Irradiation Examinations of Irradiation Creep Tested Zircaloy-2 Wade Karlsen1, Mykola Ivanchenko1, Ulla Ehrnst�n1 and Ken R. Anderson2 1 2 VTT Technical Research Centre, Espoo, Finland. Bechtel Marine Propulsion Corp., Bettis Laboratory, West Mifflin, PA, USA This work stems from a test program studying the in-pile creep behaviour of Zircaloy-2 materials. The work involves the post-irradiation examinations (PIE) of the Zircaloy-2 specimens creep tested under neutron irradiation at the HALDEN Test Reactor. The multiple specimens provided to VTT for PIE were creep tested to various plastic strain levels with some failing during testing. In-pile creep testing in the HALDEN Test Reactor involved final fluence levels of ~3�1020 n/cm2 to 5�1020 n/cm2 (>1 MeV) and temperatures of ~550 to ~650�F (~288 to ~343�C). The PIE of the various specimens included mainly fractography of the failed specimens using scanning electron microscopy (SEM) and detailed transmission electron microscopy (TEM) characterization. Scanning electron microscopy fractography was carried out using a Philips/FEITM XL30 ESEM with a LaB6 electron source. The fractured specimens showed asymmetrical necking, but the fractures were ductile and similar in appearance (Fig. 1). One of the specimens showed ductile dimple features, but with dimple walls exhibiting split fracture edges, rather than single, sharp edges (e.g., the higher magnification image in Fig. 1b). Dimensional measurements showed that in all cases there was significant localization of deformation, manifested by marked necking of the specimen, while the gauge diameter even 5 mm away from the fracture surface was still quite close to the specified original diameter. Transmission electron microscopy was carried out with a Philips/FEITM CM200 STEM equipped with a field emission gun (FEG) source operated at 200 keV. Images were recorded with a GatanTM 794 retractable multi-scan camera controlled by Gatan's Digital MicrographTM software (v. 1.82.80 Beta). Magnification calibration was verified. Foils for TEM analysis were prepared from two regions: the grip sections for irradiation defect analysis (assumed to be undeformed); and the gauge sections for characterization of deformation-related defects. Material from the grip was sliced perpendicular to the bar axis, while material in the gauge region was sliced at an angle of about 30 degrees off of perpendicular in order to produce samples suitable for 3 mm discs. TEM foils were electrolytically thinned to electron transparency using a twin jet-polisher. At least two adequate foils were produced from each specific location of each creep specimen, with the intent of imaging at least two areas in each of the two foils for each of the specimens. For a broad sampling of the irradiationinduced defects for analyses � and to ensure adequate statistical power to detect differences, at least 30 images were recorded for each material condition. Bright-field TEM images of the general microstructure of irradiation-creep tested Zircaloy-2 specimens are shown in Fig. 2. Figure 3 shows higher-magnification bright-field TEM images of the primary irradiation-induced defects, <a>-type dislocation loops. The measured irradiation produced <a>-type dislocation loop populations were similar in all of the materials, with mean diameters ranging from 6 to 9 nm, and number density from 1.8 to 2.9 x1022/m3 (dislocation density 4.5 to 7.3 x1014/m2). Figure 3 also shows evidence of dislocation channeling. More details of the analysis will be given in the presentation. The results of this recent work presented here augment the work reported by Cockeram, et al. [1] and are directly related to and support the work reported by Kozar, et al. [2]. References: [1] B. V. Cockeram, et al., J. of Nuclear Materials 418 (2011) p. 1. [2] R. W. Kozar, et al., J. of Nuclear Materials 444 (2014) p. 14. Microsc. Microanal. 21 (Suppl 3), 2015 750 (a) (b) Figure 1. SEM fractography of two irradiation creep tested Zircaloy-2 specimens; (b) shows dimple walls exhibiting split fracture edges. Figure 2. Low-magnification bright-field TEM images of irradiation-creep tested Zircaloy-2 � showing overall microstructures, including second-phase precipitates (mainly Zr-Fe-Cr Laves). (a) (b) (c) Figure 3. Higher-magnification bright-field TEM images of the primary irradiation-induced defects observed in the examined specimens: <a>-type dislocation loops formed on prism planes of the hexagonal crystal structure, (a) and (b); and an example of a clear band indicating evidence of dislocation channeling in regions close to fracture surface of the deformed gauge region (c).");sQ1[376]=new Array("../7337/0751.pdf","Inert Gas Measurement of Single Bubble in CeO2","","751 doi:10.1017/S1431927615004559 Paper No. 0376 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Inert Gas Measurement of Single Bubble in CeO2 Lingfeng He1, Janne Pakarinen2, Xianming Bai1, Jian Gan1, Yongqiang Wang3, Anter El-Azab4 and Todd R. Allen1 1. 2. Idaho National Laboratory, Idaho Falls, ID, USA University of Wisconsin, Madison, WI, USA 3. Los Alamos National Laboratory, Los Alamos, NM, USA 4. Purdue University, West Lafayette, IN, USA Uranium dioxide (UO2), an oxide with a fluorite crystal structure, is the main fuel used in commercial light water reactors. Inert fission gases such as Xe and Kr significantly impact the performance of UO2 during reactor operation and in storage. These gases have a large yield of approximately 25% and have a low solubility in UO2, resulting in the formation of large density of fission gas bubbles [1]. Such bubbles cause the fuel to swell, which promotes clad outward creep that shortens the cladding lifetime. Characterization of the inert fission gas content in bubbles can help us understand fuel swelling and fuel pin pressurization. This is performed for Xe bubbles in cerium dioxide, CeO2, which is considered a surrogate for UO2 due to similar crystal structure and properties. Xe ion irradiation of CeO2 was performed at room temperature using a 200 keV Danfysik ion implanter at the Los Alamos National Laboratory. The Xe ion energy was 400 keV and the total fluence was 1�1016 ion/cm2. The Xe implanted CeO2 was annealed in air at high temperature. A 200 kV FEI Titan scanning transmission electron microscope (STEM) with CEOS probe aberration corrector equipped with EDX and EELS was used to characterize Xe bubble microstructure and composition. Figure 1 shows the STEM image and EELS elemental mapping of a Xe bubble. The bright signal of Xe M-edge and dark signal of Ce M-edge at the bubble region indicates that the bubble is filled with Xe. The atomic ratio of Xe to Ce in the yellow dashed square region is 3.75 and 9.58 at.% determined by EDX and EELS, respectively. The local foil thickness was measured through inelastic mean free path (IMFP) using EELS technique. The IMFP, depends on the specimen density , electron energy E0 and collection semiangle (excitation semiangle and collection semiangle ) according to following equations [2]: , , , Based on local TEM foil thickness, bubble size, and specimen density, the number of Ce atoms in the dashed square region was calculated to be around 1.0�106. The number of Xe atoms in the bubble is about 3.8�104 and 9.6�104 in terms of EELS and EDX results, respectively. The replacement of Ce by Xe in the bubble was calculated to be 0.12 and 0.31 according to EELS and EDX results, respectively. The red solid square regions R1 and R2 were selected for EELS spectrum analysis (Figure 2). Weak Xe M4,5-edge at around 700 eV was only detected in R1 and strong Ce M4,5-edge at about 900 eV was found in both R1 and R2. A method introduced by Microsc. Microanal. 21 (Suppl 3), 2015 752 Fortner and Buck [3] was applied for the EELS data analysis. The average branching ratio of Ce, M4/M5 at R1 and R2 is about 1.09 and 1.17, respectively, which correspond to an average valence state of +3.9 and +4, respectively. It indicates O vacancies were formed at bubble region. In short, chemical information and inert gas content in single bubble can be determined by STEM/EDS/EELS techniques. This research was supported as part of the Center for Materials Science of Nuclear Fuels, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science as well as Laboratory Directed Research and Development (LDRD) program at Idaho National Laboratory. The ion implantation was partially supported by the Center for Integrated Nanotechnologies (CINT), a DOE nanoscience user facility jointly operated by Los Alamos and Sandia National Laboratories. References: [1] D. Olander, Fundamental Aspects of Nuclear Reactor Fuel Elements, US Department of Energy, 1976. [2] K. Iakoubovskii et al., Microsc. Res. Tech. 71 (2008) p. 626. [3] J. Fortner and E. Buck, Appl. Phys. Lett. 68 (1996) p. 3817. Figure 1. (a) STEM image of a single Xe bubble in CeO2 after post-irradiation annealing at 1200oC for 1 hour. (b) EELS Xe M-edge mapping. (c) EELS Ce M-edge mapping. The yellow dashed square was selected for composition analysis using EDX and EELS, and the red solid squares were selected for EELS spectrum analysis. Figure 2. EELS spectra of R1 and R2 regions in Figure 1.");sQ1[377]=new Array("../7337/0753.pdf","Exploring Helium Mitigation in Ferritic Alloys by Advanced Microscopy","","753 doi:10.1017/S1431927615004560 Paper No. 0377 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Exploring Helium Mitigation in Ferritic Alloys by Advanced Microscopy Chad M. Parish1, Philip D. Edmondson1, Fred W. Meyer2, Mark E. Bannister2, Baishakhi Mazumder3, and Michael K. Miller3 Materials Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN USA Physics Division, Oak Ridge National Laboratory, Oak Ridge, TN USA 3. Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, TN USA 2. 1. Future nuclear energy systems, fission or fusion, will show significantly higher-energy ("harder") neutron spectra than the thermal-spectrum reactors of today. Hard neutron spectra make the heliumgenerating (n,) reaction more pronounced. Coupled with high operating temperatures, helium embrittlement may become the lifetime-limiting failure mode [1]. Materials design paradigms to provide helium mitigation need to be explored experimentally by high-resolution microscopy. Here, we discuss STEM and APT-based methods to relate microstructural features such as nanoclusters (NCs) and carbides to nanometer-scale helium bubbles. Samples were F82H or 14YWT ferritic alloys. F82H contains multiple carbide populations and 14YWT multiple oxide populations, including 2 nm-scale NCs [2,3]. Samples were irradiated with a stackedenergy He beam to ~8000 ppm and 1 dpa [4] or monochromatic He to ~30,000 appm at ~1 dpa [3] at multiple temperatures. From APT isodensity surfaces, we previously demonstrated imaging of He bubbles, and employed it to measure if they were associated with NCs [3]. We have also recently found accumulation of He bubbles to M23C6-type carbides in F82H [4], Fig. 1. TEM is traditionally used to image voids of bubbles via over/under focused conditions. However, resolution is limited to ~2 nm [5]. HAADF in STEM provides higher resolution, but it is difficult to differentiate bubbles from NCs in these alloys in HAADF mode. Combining BF-STEM in over- or under-focus [6] with in-focus HAADF makes clear differentiation of bubbles / voids from NCs possible, Fig. 2. Further, recent advances in EDS X-ray mapping capability [7] make it possible to map chemically the extremely small NCs or carbides, Fig. 3, allowing the highest level of differentiation between the different features in the materials' helium-damaged microstructure. In summary, determining the size, density, and distribution of He bubbles is necessary for successful deployment of future nuclear power systems. Combining advanced STEM and APT methods allows comprehensive characterization of the He bubble distributions in irradiated materials. [8] [1] SJ Zinkle & NM Ghoniem, Fus. Eng. Des., V51 (2000) P.55. [2] RL Klueh et al., Curr. Op. Sol. State Mater. Sci., V8 (2004) P.239. [3] PD Edmondson et al., J. Nucl. Mater., V434 (2013) P. 210. [4] B Mazumder et al., Nucl. Mater. and Energy, 1 (2015) in press. [5] B. Yao et al., J. Electron Micros., V61 (2012) P. 393. [6] CM Parish & MK Miller, Microsc. Microan., V20 (2014) P. 568. [7] P Schlossmacher et al., Micros. Today, July (2010) P. 14. [8] This work was supported by the U.S. Department of Energy, Office of Science, Basic Energy Sciences, Materials Sciences and Engineering Division. A portion of the Microscopy was conducted as part of a user proposal at ORNL's Center for Nanophase Materials Sciences, which is an Office of Science User Facility. We acknowledge the use of the Analytical Instrumentation Facility (AIF) at North Microsc. Microanal. 21 (Suppl 3), 2015 754 Carolina State University, which is supported by the State of North Carolina and the National Science Foundation. We thank Dr. D. T. Hoelzer for 14YWT sample material. Fig. 1: Atom probe atom maps of a Cr23C6-type carbide. Inset shows isodensity surfaces (presumably He-bubbles) at the carbide. F82H, irradiated at ~50-60�C. Cr atoms in purple. Fig. 2: Aberration-corrected STEM of He-bubbles. Left: underfocus BF with small collection angle (BF); right, HAADF image. Bubbles can be matched between the two modes. ORNL Titan microscope. Fig. 3: High-resolution aberration-corrected HAADF (left) and X-ray map (right) of He-irradiated 14YWT. (This region is between the ion-incidence surface and He stopping band, so no bubbles are seen.) X-ray map colored by multivariate statistical analysis (MVSA): Red: Fe-Cr-W matrix; Green: Cr-W grain boundary segregation; Blue: Y-Ti-O rich nanoclusters. North Carolina State University Titan Microscope.");sQ1[378]=new Array("../7337/0755.pdf","Evolution of Oxide Structures in Friction Stir Welded Alloy MA956","","755 doi:10.1017/S1431927615004572 Paper No. 0378 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Evolution of Oxide Structures in Friction Stir Welded Alloy MA956 L.N. Brewer1,2, B.W. Baker1,3, M.J. Bird1, and S. Osswald,1,4 1. 2. 3. 4. Mechanical and Aerospace Engineering, Naval Postgraduate School, Monterey, CA, USA Metallurgical and Materials Engineering, University of Alabama, Tuscaloosa, AL, USA Mechanical Engineering Department, United States Naval Academy, Annapolis, MD, USA Purdue University, West Lafayette, IN, USA Oxide dispersion strengthened alloys are of great potential importance for advanced fission and fusion reactors because of their high temperature strength, creep resistance, and radiation damage resistance.[1] Unfortunately, these materials are not readily joined by fusion welding as the dispersed oxides agglomerate and coarsen in the weld pool, thus drastically reducing the strength of the welded material. Friction stir welding (FSW) can successfully join these materials through a solid-state bonding process, but there is still some reduction in strength due to evolution of the oxide structure.[2] This paper examines the changes in oxide size and phase as a function of FSW parameters in the alloy MA956. Two plates of MA956 (Fe-19.9Cr-4.8Al-0.39Ti-0.51Y2O3-0.023C) were friction stir welded over a series of tool rotation and translation speeds, from 200-500 RPM and from 25-175 millimeters per minute (MMPM). Cross-sectional samples for electron microscopy, Raman spectroscopy, and x-ray microanalysis were prepared by cutting the plate perpendicular to the weld and using metallographic techniques down to a finish of 0.05um colloidal silica. SEM and EDX analyses were completed using a Zeiss Neon 40 scanning electron microscope at 10-12 keV beam voltage, 60 �m objective aperture, and a probe current of 1 nA. Both EDX maps and point spectra were collected using the EDAX Genesis system and a silicon drift detector. Raman spectroscopy measurements were performed with Renishaw's inVia Raman microspectrometer using 514 and 785 nm laser excitation wavelengths. SEM imaging and x-ray mapping showed a clear increase in oxide particle size in the stir zone of welds with large heat input, i.e. low tool translation speed and high rotation speeds. The yttrium-based oxides in the base material are approximately 10 nm is size as determined by SAXS and STEM measurements. After FSW, yttrium-aluminum oxide particles as large as 350 nm in size were observed. Interestingly, the spacing between coarsened oxides was approximately 3�m for all of the high heat input conditions, suggesting that the oxide coarsening is not just from Ostwald ripening alone. The Raman data shows that the phase of the oxides is also changing as the oxides coarsen. The highest heat input conditions (e.g. 500RPM/25 MMPM) have a majority of yttrium aluminum garnet (YAG) particles with some yttrium aluminum perovskite (YAP) particles present in the stir zone. Lower heat input conditions, e.g. 400 RPM/100 MMPM, do not show signal from the YAG phase but instead primarily from the YAP phase. These results are in agreement with the work of Chen et al., [3] but the present results are the first to show the direct connection between the FSW conditions and the size and phase evolution of the oxides. These results are important because they underscore the need to control oxide phase stability during FSW in order to maintain the strength and radiation damage resistance of the welded material. References: [1] P. Andresen, F.A. Garner, et al., EPRI, Palo Alto, CA, 2012, pp. 1-320. [2] B.W. Baker, T.R. McNelley, et al., Materials Science and Engineering-A, 589 (2014) 217-227. [3] C.L. Chen, P. Wang, et al., Materials at High Temperatures, 26 (2009) 299-303. Microsc. Microanal. 21 (Suppl 3), 2015 756 Figure 1 Figure 2 Figure 1. Secondary electron image (top) of oxides from friction stir weld at high heat input condition (500 RPM/25 MMPM). The white arrow indicates an individual, coarsened oxide (Y-Al-O). (bottom) EDX spectra showing signals from the matrix and the oxide particle. Figure 2. Raman spectral map (top) of particles with dominant features from the YAG phase. The individual numbers correspond to spectra from individual particles. (bottom) The individual Raman spectra that correspond to the numbered oxide particles. Figure 3 Figure 3. Systematic change in Raman spectra as a function of increasing FSW tool rotation speed.");sQ1[379]=new Array("../7337/0757.pdf","Structural Effects in Oxide Dispersion Strengthened (ODS) Steel Neutron Irradiated to 3 dpa at 500�C","","757 doi:10.1017/S1431927615004584 Paper No. 0379 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Structural Effects in Oxide Dispersion Strengthened (ODS) Steel Neutron Irradiated to 3 dpa at 500�C Alexander Mairov1, Jianchao He2, Kumar Sridharan1 1. 2. Department of Engineering Physics, University of Wisconsin-Madison, Madison, USA School of Materials Science and Engineering, University of Science and Technology Beijing, Beijing, China The exposure of a material to a radiation field results in the formation of point defects which, depending on temperature and dose, can then lead to the evolution of extended defects such as dislocation loops, stacking fault, voids, and bubbles [1]. These extended defects can then cause hardening and embrittlement, phase instabilities, irradiation creep, swelling. Advanced reactors are expected to operate at much higher temperatures and higher doses as compared to present LWR. Materials used for cladding and core internals should possess a combination of high creep strength and radiation damage tolerance at high temperatures [2]. Ferritic or ferritic-martensitic (FM) oxide dispersion strengthen (ODS) steels containing Y-Ti-O nanoparticles (~10nm, 0.2 to 0.3 wt.%, also referred to as nanoclusters) provide this combination of properties. These nanoparticles are effective in impeding dislocation movement in the steel matrix at high temperatures, and it is hypothesized that the interfaces between the nanoparticles and the matrix act as sinks for radiation-induced defects and enhance radiation damage tolerance [3]. In this paper, we have investigated structural effects in a neutron irradiated 9%Cr-ODS steel. Neutron irradiations were performed at a nominal temperature of 500 �C in the Advanced Test Reactor (ATR), located at the Idaho National Laboratory (INL, Idaho Falls), up to a dose of 3 dpa. Structural examinations on nanometer length scales were performed using Transmission Electron Microscopy (TEM) and Scanning-TEM coupled with Energy Dispersive Spectroscopy (EDS). TEM samples were prepared using Focused Ion Beam (FIB) lift out techniques with a Quanta 3D FEG FEI FIB: lift-out techniques were used to reduce the volume of the samples, in order to minimize electron aberration caused by magnetic effects stemming from the ferritic/martensitic structure of these steels. TEM and STEM investigations were conducted at the Microscopy and Characterization Suit (MACS) located in Idaho Falls, using a Tecnai TF30-FEG STwin microscope operated at 300 KV. The nominal composition of the ODS steel used in this study is shown in Table 1. The microstructure of the as received steel is complex, containing a preexisting network of dislocations, a fine dispersion of nano-scale oxide particles and coarse carbides. Radiation induced defects were analyzed with 2-beam imaging conditions, and, following the procedure outlined in Yao et al. [4], the samples were imaged close to the [100], [110] and [111] zone axis. Defects such as dislocation loops (Figure 1) and voids caused by irradiation were imaged and quantified. High-angle annular dark field (HAADF) STEM images, which provide strong Z-contrast, were employed to identify the small oxide clusters present in the matrix and quantify them before and after radiation. EDS was used to confirm the presence of Y and Ti in the oxide nanoparticles. Figure 2 shows an HAADF STEM image and EDS point scans performed on precipitates also observed in the irradiated sample [5]. Microsc. Microanal. 21 (Suppl 3), 2015 758 C Si Mn P S Ni Cr W 0.14 0.048 0.05 <0.005 0.004 0.06 8.67 1.96 Table 1. 9%Cr-ODS chemical composition (wt%, Balance Fe). Ti 0.23 Y 0.27 O 0.14 N 0.017 Ar 0.004 Figure 1. Bright field TEM micrograph showing dislocations loops (indicated by white arrows) in the neutron irradiated 9%Cr-ODS steel. These loops were imaged using g002 close to the [110] zone axis. Inset shows the electron diffraction pattern close to [110] zone axis. Figure 2. STEM-EDS spectra (spot analysis) collected from nanoclusters showing mainly Fe, Cr, O, Ti and Y after neutron irradiation to 3 dpa at 500�C. Cu signal are from the grid on which the TEM sample is seated. References [1] S.J. Zinkle, Comprehensive Nuclear Materials 1 (2012), p. 65. [2] T. Allen et al., Materials Today 13 (2010), p. 14. [3] L. L. Hsiung et al., Phys. Rev. B. 82 (2010), p. 184103. [4] B. Yao et al., Journal of Nuclear Materials 434 (2013), p. 402�410. [5] A portion of this work was performed at the Center for Advanced Energy Studies (CAES) in Idaho Falls, Idaho through the Advanced Test Reactor National Scientific User Facility (ATR NSUF) award. Part of this work was performed using the NSF-supported shared facilities at the University of Wisconsin and funding from the DOE Office of Nuclear Energy's Nuclear Energy University Programs.");sQ1[380]=new Array("../7337/0759.pdf","Titanium (Ti) Alloy Electron Beam (EB) Weld Fracture Surface Interpretation","","759 doi:10.1017/S1431927615004596 Paper No. 0380 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Titanium (Ti) Alloy Electron Beam (EB) Weld Fracture Surface Interpretation Steven J. Gentz1 1. NASA, Langley Research Center, Hampton, VA Failure analysis can be defined as collecting, interpreting, and documenting information to determine the failure cause. Data collection and interpretation should be a balance of: visual and microscopy; macro-, micro-, and nano-chemistry; and research and discussion. The following example centers on restrain in concluding a failure cause without evaluating all of the available data. A Ti alloy pressure vessel designed to contain an inert gas was undergoing qualification testing. This vessel consisted of a cylindrical center section and hemispherical domes joined with circumferential EB welds. Pressurization/depressurization was through a fitting EB welded on one dome. The qualification test regime consisted of proof testing, nondestructive examination (NDE), several hundred pressure cycles to the maximum design pressure (MDP), and then pneumatic burst testing. The post proof NDE did not report any detectable defects, the cyclic testing occurred without incident, and the burst pressure was greater than two times the MDP. Following burst testing, the vessel was subjected to metallurgical analysis to examine the fracture surface and remnant hardware for anomalous features. The vessel separated at a dome/cylinder EB weld. The cylindrical section showed outer diameter (OD) axial strain lines and diametric growth (i.e., barreling) plastic deformation. The fractured EB weld showed good tracking and uniformity. Housing wall thickness measurements in the cylindrical section adjacent to the weld fusion zone did not detect deviations from the drawing dimension. The fracture surface presented three distinct zones, which were labeled the initial fracture (Figure 1), fast fracture, and final overload. The initial fracture was approximately 6.5 cm long and centered at the EB cover pass overlap. When viewed with a stereomicroscope, this zone was distinctly different from the adjacent fast fracture region, had an OD shear lip, and numerous tear ridges converging at multiple inner diameter (ID) locations. In addition, a demarcation line was seen at approximately mid span through the thickness. Depending on the light source and viewing angle, this boundary line appeared to turn towards the ID surface. As the vessel was subjected to repeated pressure cycles, this region was suspected to be a fatigue crack approximately 6.5 cm long by 0.3 cm deep. The project manager and structural analyst received this preliminary observation with alarm and skepticism as a defect of this magnitude should have failed well before the recorded burst pressure. Subsequently the fracture surface was examined with electron microscopy and sectioned for metallography. Microscopy determined the preponderance of crack extension in this zone was a combination of ductile dimple and cleavage fracture with no evidence of organized cyclic fatigue. Metallography (Figure 2) determined the reported OD shear lip was associated with the EB cover pass fusion zone, and the demarcation line was a microstructural thermal effect from the cover pass. Figure 3 shows a composite of the metallography section compared with the corresponding location on the matting fracture surface. Although the fracture exhibited unique characteristics, providing premature observations without sufficient data can be counterproductive as they may generate unnecessary confusion. Microsc. Microanal. 21 (Suppl 3), 2015 760 756 Figure 1. Initial Fracture Zone in Circumferential EB Weld Showing OD Shear Lip, Tear Ridges, and Mid Span Demarcation Line (3x magnification) Figure 2. EB Weld Fracture Cross Section (10x magnification, etchant � Kroll's reagent) Figure 3. Composite of Optical, Electron Microscopy, and Metallographic Images Showing Relationship of Demarcation Line to Microstructural Feature (line indicates sectioning location)");sQ1[381]=new Array("../7337/0761.pdf","Space Shuttle Stiffener Ring Foam Failure Analysis, a Non-conventional Approach","","761 doi:10.1017/S1431927615004602 Paper No. 0381 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Space Shuttle Stiffener Ring Foam Failure Analysis, a Non-conventional Approach P.M. Howard NASA Kennedy Space Center Materials Science Division, Chemistry Branch NE-L3 Kennedy Space Center, FL The Space Shuttle Program was all too familiar with the dangers of solid rocket booster (SRB) foam debonding and potentially striking the Shuttle during ascent into space. The Space Shuttle Program used lightweight, rigid polyurethane foam for cryogenic tank insulation and for structural protection of the SRB stiffener rings. Although the foam application on the stiffener ring was not in the flight path of the Shuttle; during ascent, a piece of foam could have become caught in the flight turbulent flow and redirected back into the Shuttle Aft Engine Compartment or the External Tank. During the years of the Space Shuttle Program, tens of thousands of hours were spent optimizing the process parameters and studying the chemistry of the foam in order to produce a successful application. The foam application process called for an initial bond coat of 1/8 inches to be applied and then successive layers until the ten inch thick application was completed. Extensive "trial and error test matrix" studies were conducted to control the application parameters such as the urethane two part mix ratios, operator application techniques, and humidity and temperature control of the solid rocket motor structure as well as the facility. The chemistry of the foam also underwent two formulation changes of the catalyst and the blowing agent in response to environmental regulations; which added additional variables and an overall confusion to the true mode of failures. For example, during the preparation of the SRBs for the STS-117 mission; four foam application attempts failed. The foam failure was submitted to the NASA, Kennedy Space Center Chemistry laboratory for failure analysis. When previous foam applications failed; the classical methods of failure analysis: evaluating the process, performing chemical residue extractions for adhesive failure, bulk mechanical testing and fracture analysis, did not provide the root cause of the failure of the foam. A new approach had to be found. Realizing that foam is the ideal media to document and preserve its own mode of failure; thin-sectioning was immediately identified as a logical approach for foam failure analysis. To observe the foam cell morphology, the foam samples were cross sectioned at 90 degrees relative to the application to observe the three dimensional morphology of the cells. The morphology was examined by polarized light microscopy (PLM) and scanning electron microscopy (SEM). SEM analyses of the foam cells discovered that the bond coat cells had been distorted 40 to 100 times larger than the normal size. This was due to the expansion of the second layer application while the bond coat was still uncured. The failure occurred within a narrow range of approximately 500 micrometers of the bond coat. Using the special cross-section technique and then examining the cell foam morphology, provided a much greater understanding of the foam failure modes than previously achieved. The Shuttle is no longer with us, but this failure analysis approach will be applied to America's next generation of spacecraft using the Space Shuttle legacy flight hardware. Microsc. Microanal. 21 (Suppl 3), 2015 762 Figure 1. Application of the foam to the SRB stiffener rings and the build-up by layers. Figure 2. A two foot section of foam application that failed at less than 10 pounds per square inch (psig) pressure which for a normal application should have been 50 psig. Figure 3. Cross-section examination of the foam revealed that the bond coat had not been cured before the second coating distorted and destroyed structure of the bond coating.");sQ1[382]=new Array("../7337/0763.pdf","Research Projects and Failure Investigations Conducted Through a Commercial Materials Testing Laboratory","","763 doi:10.1017/S1431927615004614 Paper No. 0382 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Research Projects and Failure Investigations Conducted Through a Commercial Materials Testing Laboratory Jaret J. Frafjord1 1. IMR Test Labs, Portland, OR USA This paper will describe work conducted with research staff and failure analysts. There are advantages to using an accredited, commercial materials testing laboratory for research and failure investigations, even if there is little or no onsite expertise in the area of study. The expertise can come in the form of knowledge of equipment and how to best accommodate or design special test fixtures. These labs are fully capable of all mechanical, chemical and metallographic analysis to provide one-stop shopping for all testing needs. The personnel are fully trained and they use the equipment on a regular basis. The equipment is calibrated regularly to NIST standards. The labs are audited regularly and they perform in round robin proficiency tests to ensure accuracy. One project involved conducting an extensive heat treat study to show how the properties changed through a variety of cooling rates. Two thermocouples were attached to the ends of the bar stock and heated in the same furnace. Each set of bar stock was removed and cooled at a different rate and the cooling rate monitored and plotted (Figure 1). Each bar was tested for hardness, microstructure, and tensile strength. Although the ultimate strength and elongation were not greatly affected, the yield strength dropped as the cooling rate increased. Differences in the microstructure is shown in Figure 2. Another project involved testing a pin for a dam system in the Columbia River. The engineering team needed to test the strength of the metal pin at 38F to simulate the river temperature. A bath system was developed to allow the test specimen to soak in a bath of water and dry ice in the testing machine. The specimen soaked for 5 minutes prior to testing, and the tensile test was conducted while the test specimen was submerged in the water. These and other projects will be discussed in detail, showing how the knowledge of equipment and test methods helped ensure proper and accurate results that could be traced to ASTM procedures and NIST standards. Microsc. Microanal. 21 (Suppl 3), 2015 764 Figure 1. Cooling curves of a steel cooled by water, oil, fan, air, and furnace. Figure 2. Micrographs of the steel in the water quenched condition (left) and the air cooled condition (right).");sQ1[383]=new Array("../7337/0765.pdf","Using Visual Communication Elements in Technical Reports","","765 doi:10.1017/S1431927615004626 Paper No. 0383 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Using Visual Communication Elements in Technical Reports Coralee McNee1, Aaron Kueck2 and Dave Christie2 1. 2. Metallurgist, Brooktondale, New York, United States. Curtiss-Wright Corporation, Ithaca Materials Research, Lansing, New York, United States. Visual examination, fractography and photodocumentation are arguably the most important pieces of analysis when running a metallurgical failure investigation. Each of these analyses requires communicating complex systems visually. Therefore, visual communication concepts can prove to be a valuable tool. As failure analysts often work with clients only briefly, their ability to communicate complex ideas effectively to a diverse audience can increase the likelihood of their success. This paper highlights the following key requirements for effective visual communications. Be observant. In failure analysis, it is important to not only observe the fracture surface but the entire part, and it is important to understand how the part interacts with other components or surrounding environments. Approach problems without a preconceived idea of what the solution may be; avoid overlooking details that don't fit the "standard" narrative. Composition counts. Composition includes balance, proportion, repetition, unity, focal point, rhythm and variety. An easy way to create balance in an image is to use the rule of thirds for placing your focal point in an image. Composition applies to sets of images as well. When documenting a single feature across several different imaging techniques, it is often useful to use the same or similar magnification and orientation, implementing unity. Using several different imaging techniques to capture a single feature provides repetition. For example capturing backscatter electron images, secondary electron images, incident light images, and diffuse light images provides variety. Combining unity and repetition in this way provides variety in the documentation of that feature. For example, images in sequence allow the viewer distinguish between relevant differences without explicit explanation as seen in Figure 1. Also, consider a report as a single composition. The information, images, data and presentation should all connect to one another in a way that keeps the interest of the reader. For example, the report should move from macroscale photos to micrographs and from components to systems. Good composition often goes unnoticed and when done properly you are not aware of its function. Repetition and unity should be used to communicate to standard information i.e. scale bars or font sizes in presentations. An added benefit is that careful use of repetition and unity make subtle variety a powerful tool in drawing the viewer's attention. Adding an annotated image to a presentation or using an inset in a diagram can focus your viewer without using intrusive presentation artifices. Good report composition brings the reader along with the failure analyst as the problem is described and documented, analyzed, and finally dispositioned. Microsc. Microanal. 21 (Suppl 3), 2015 766 Figure captions should be descriptive but succinct and guide the reader to the relevant features in the image. Avoid long-winded explanations when a suitable image conveys the idea visually. Annotations on images should provide cues guiding the reader to important features, but without crowding the image or obscuring surrounding features. Complicated annotations may require a pair of images, with and without additions, to provide the complete picture. Engineers and scientists communicate increasingly complex systems and specialized science to increasingly diverse audiences. Visual communication can transcend language, cultural, and technical backgrounds when done well. It is no surprise that when someone really wants to understand how something works they say `Show me.' References: [1] A Ruedinger, "Visual Communication" Art Institute Portland, OR Summer 2007. Lecture. [2] D A Lauer and S Pentak, "Design Basics"6th ed. (Thomson Wadsworth, United States) p. 23. Figure 1. Using three images in sequence allows the viewer to locate the higher magnification image, while providing additional information about the subject.");sQ1[384]=new Array("../7337/0767.pdf","Failure Analysis to Unambiguously Clarify Hydrogen Fracturing in Steels","","767 doi:10.1017/S1431927615004638 Paper No. 0384 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Failure Analysis to Unambiguously Clarify Hydrogen Fracturing in Steels Pohl, Michael1 Ruhr-University Bochum, Faculty of Mechanical Engineering, Institute for Materials, Chair for Material Testing, Bochum, Germany. Hydrogen can get into metallic components and damage them in very different ways (fig. 1). In liquid phase metallurgical processes such as casting and welding, as a result of the dramatic reduction in the solubility of gases during solidification, gas bubbles can form causing porosity which particulary in conjunction with H induced cracking, form "fish-eyes" acting as dangerous internal stress raisers. This type of cracking, without the presence of porosity, which is also known as "flaking", can occur in thin steel wires only a few m in diameter as well as in large 100 tonn steel forgings, in both cases significantly detrimentally affecting both, the manufacturing and service behavior of the component [1]. Apart from metallurgical sources, semi-finished and finished components can also absorb H from galvanic processes or as a result of corrosion. In these cases the H diffuses into regions of higher stress and strain causing cracking, often unexpectedly and after significant periods of time [2]. In failure analysis the clarification of hydrogen induced component damages is based on fractographic examinations in combination with H-analysis. The delayed fracture, which is typical for H-induced damages, is a strong evidence for a damage caused by hydrogen when analyzing damage. If a sample taken from the component breaks during a clamping test, this can only be caused by H which has always been present in the component. If too much of the hydrogen has escaped meanwhile, the fractographic comparison between a sample broken in a clamping test (fig.2) and the primary fracture of the component, as well as the final failure fracture (fig. 3 a + b) is necessary. As these three fractures arise from the identical material, the comparison should lead to an unambiguous result. Finally there is still the possibility to selectively detect the diffusible (damaging) hydrogen using the HCA-method [3]. The consecutive series of the investigation steps is shown schematically in figure 4. References: [1] M. Pohl, Hydrogen in Metals: A Systematic Overview, Pract. Metallogr. 51 (2014) p.291-305. [2] M. Pohl, S. K�hn, Stress Corrosion Cracking of High Strength Steels, Materials Testing 52 (2010) p. 52-55. [3] S. K�hn, F. Unterumsberger, T. Suter, M. Pohl, New Methods to Analyse the Diffusible Hydrogen in High Strength Steels, Materials Testing 55 (2013) p. 648-652. 1. Microsc. Microanal. 21 (Suppl 3), 2015 768 Figure 1. Possible forms of H-Absorption. Figure 2. Generating a fracture for comparison. Figure 3. Generating a H2fracture for comparison. Figure 4. Unambiguously clarify H-induced cracking.");sQ1[385]=new Array("../7337/0769.pdf","Developing Mechanistic Understanding of the Effect of Temperature and Environment on the Hold Time Fatigue Behavior of Haynes 282 Alloy","","769 doi:10.1017/S143192761500464X Paper No. 0385 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Developing Mechanistic Understanding of the Effect of Temperature and Environment on the Hold Time Fatigue Behavior of Haynes 282 Alloy D. Ellis1, T. Hanlon1, C. Shen1, and E. Hall1 1. GE Global Research, Niskayuna, NY, USA Advanced ultra-supercritical coal-powered steam plants achieve significant efficiency gains via higher operating temperatures and pressures than existing plants. A significant challenge is developing materials for these more aggressive operating conditions and environment. This study was undertaken to develop an understanding of the hold time fatigue mechanism of the Ni-based superalloy Haynes 282 and to separate out the environmental and creep effects of the fatigue behavior. Crack growth rate experiments on samples of Haynes 282 were conducted at temperatures of 1200, 1400, and 1600F in air, steam, and vacuum environments (compact tension sample, 25 ksi in1/2, R = 0.1). Foils from tested samples were then prepared from the bulk free surface oxide and from the fracture surface adjacent to and far from the crack tip using focused ion beam scanning electron microscopy (FIB-SEM). Transmission electron microscopy (TEM) was used to identify various phases present in the resulting oxides, with an attempt to correlate structure to crack growth rate. Figure 1 shows crack growth rate data from samples tested at 1200, 1400, and 1600F in air, steam, and vacuum conditions. At 1200�F and 1400�F, crack growth rates are accelerated in steam, and delayed in vacuum, with respect to the air tests. However at 1600F, crack growth rates among the three environments tend to converge. The data suggest a pure environmental fatigue interaction at 1200F, a predominantly environmental-fatigue interaction at 1400F, and a pure creep-fatigue interaction at 1600F. The oxide structure at the free surfaces of all the samples was similar showing internal Al2O3 oxidation, gamma prime phase depletion, and hexagonal Cr 2O3and Ti-rich oxides at the surface of the oxide. Figure 2 shows oxide structure at the crack tip for samples tested at 1400F in air and steam. A key observation is the delayed formation of Al2O3 and Cr2O3 at the crack tip of specimens tested in air versus those tested in steam. Despite having a higher free energy of formation, mixed Ni oxides form first at the crack tip in air, leaving in their wake a large degree of local porosity. In contrast, specimens tested in steam, under a lower oxygen partial pressure, formed thick continuous layers of Cr2O3 directly at the crack tip, with limited porosity. The degree of porosity formed at the crack tip in air is believed to promote a more compliant crack tip, relative to the thick Cr2O3 layer, and is thought to retard crack propagation through crack path deflection and strain accommodation. In summary, environmental-fatigue interactions dominated at temperatures of 1400F and below. At 1600F, some environmental interaction was detected even as creep effects were becoming dominant. Near-field crack tip oxidation characteristics, and not bulk oxidation resistance, drive the variation in the hold time fatigue behavior over the range of temperatures and waveforms evaluated. The transition from crack tip porous NiO and Ni-rich spinel to dense Al2 O3/Cr2O3 may accelerate hold time fatigue in steam environments. Microsc. Microanal. 21 (Suppl 3), 2015 770 References: [1] Acknowledgment: This work was supported by the Department of Energy National Energy Technology Laboratory under Award No. DE-FE0005859. [2] Disclaimer: This report was prepared as an account of work sponsored by an agency of the U.S. Government. Neither the U.S. Government nor any agency thereof, nor any of their employed, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacture, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the U.S. Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the U.S. Government or any agency thereof. Figure 1. Crack growth rate for Haynes 282 at 1200, 1400, 1600F in air, steam, and vacuum. Figure 2. O, Al, Ti, Ni (red, green, purple, blue respectively) TEM EDS maps at crack tip of samples tested at 1400F in air and steam.");sQ1[386]=new Array("../7337/0771.pdf","Metallography � A Powerful Instrument for Material Characterisation, Material Development and Failure Analysis","","771 doi:10.1017/S1431927615004651 Paper No. 0386 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Metallography � A Powerful Instrument for Material Characterisation, Material Development and Failure Analysis Michael Panzenb�ck1, Francisca Mendez-Martin1, Boryana Rashkova1, Patric Sch�tz1 1. Department of Physical Metallurgy and Materials Testing, Montanuniversit�t Leoben, Austria A. Widmanst�tten (1754-1849) was one of the first scientists who developed techniques for studying the microstructure of meteorites by grinding and etching them with nitric acid. Also H.C. Sorby (18261908) used such methods for microstructural investigations of minerals and rocks to identify their origin, as well as to examine steels and meteorites. Other famous and nowadays well known researchers such as R. Hadfield (1858--1940), A. Martens (1850-1914), E.C. Bains (1891-1971), K.H. Ledebur (18371906), and H. Brearley (1871-1948) further developed these basic methods to get more information about the microstructure, especially in case of steels. Many microstructural parts or phases of the FeFe3C phase diagram are called in honour of these scientists, e.g., "Widmanst�tten ferrite", "Sorbite", "Bainite", "Martensite", or in case of cast iron "Ledeburite". Without doubt, a material's microstructure is decisive, because it is the arrangement, size and distribution of different phases which is responsible for the mechanical properties such as strength, ductility and toughness. In order to develop high-performance materials for existing and new applications it is necessary to establish a basic understanding of the material behaviour or its response under service loads. In addition to the standard methods such as light microscopy (LM), scanning electron microscopy (SEM), focused ion beam (FIB), also high-resolution methods (transmission electron microscopy (TEM), atom probe tomography (APT)) have been developed over the last three decades. Moreover, advanced test equipment is used, to identify material behaviour in various conditions and environments. We want to present the role of metallography in the fields of material development [1] and failure analyses [2], as described here for one example. Figure 1 shows an overview of broken supports. Macroscopic investigations reveal dark spots within the crack surface. This indicates that such defects were responsible for the failure of many supports. Further investigations revealed [2] that the area of crack initiation was covered with Zn, which originates from hot dip galvanization. A cut through such broken and unbroken parts in longitudinal direction revealed cracks covered with Zn (Fig. 2). The light-optical micrograph shows the Zn-layer followed by the microstructure of the steel, which consists of ferrite and a small amount of pearlite, as can be seen far away from the Zn-layer. However, it seems that regions near the Zn-layer exhibit a nearly pure ferrite structure. This can be attributed to an effect of decarburisation. The "ferritic" microstructure is caused by the cathodic protection due to zinc during etching with Nital-solution. In this case the cracks can be attributed to a liquid metal embrittlement (LME). Between the years 2000 and 2010 many failures of steel constructions occurred due to LME. In many cases, polluted Zn, used for the hot dip galvanization process, was identified [3] as a trigger for LME. Elements such as Pb, Sn, and Sb were typically identified with high concentration of more 1 wt%. In this case, a content of less than 0.1 wt% was found. In contrast to the Zn layer, high concentrations of Cu (about 0.45 wt %), Sn (0.12 wt %) and Sb (about 0.1 wt %) were found in the steel used for the supports. It is not possible to detect local enrichments of these elements with standard metallographic methods. Therefore, we used APT to prove the enrichment of Cu and Sn on or near grain boundaries. The results Microsc. Microanal. 21 (Suppl 3), 2015 772 are shown in Fig. 3. The preparation of the tip (Figs. 3a, b) was made by using etching methods and subsequent milling by FIB. The process was controlled by electron back-scatter channelling diffraction (EBSD) to ensure that a grain boundary is in the investigated region. The grain boundaries are covered with precipitates of Cu and Sn (Figs. 3c, d). Additionally, ternary cementite is visible. References [1] S. Mayer, G. Hawranek, F. Mendec-Martin, M. Panzenb�ck, S. P�lzl, S. Primig, B. Rashkova, H. Clemens, to be published in Practical Metallography (2015) [2] M. Panzenb�ck, P. Sch�tz, Microscopy and Microanalyses, 20 (2014), p. 1868 [3] A. Durst, A. Luithle, W. Bleck, M. Pohl, Practical Metallography, 48 (2011), p. 6. Figure 1. Macroscopic overview of the fracture surface. Arrows mark the region of crack initiation (dark spots). Figure 2. Light-optical micrograph of a crack after etching with Nital. The crack surface is covered with zinc. Figure 3. TEM images (a) of an atom probe tomography (APT) specimen (prepared by FIB) of a broken support. The results reveal fine Cu- and Sn-rich precipitates at the grain boundaries between a cementite lamella and the iron -matrix. Fe atoms are shown in pink (b), Cu isosurfaces in yellow and Sn isosurfaces in green. The grain boundary between the cementite lamella and the iron -matrix is indicated in red color in (c) and (d).");sQ1[387]=new Array("../7337/0773.pdf","Classifying the Severity of Grain Boundary Corrosion in CoCrMo Biomedical Implants","","773 doi:10.1017/S1431927615004663 Paper No. 0387 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Classifying the Severity of Grain Boundary Corrosion in CoCrMo Biomedical Implants Alex Lin1, Emily Hoffman1 and Laurence D. Marks1 1 Department of Materials Science and Engineering, Northwestern University, Evanston, IL Cobalt-chromium-molybdenum (CoCrMo) alloys with the addition of carbon are popular for biomedical devices, specifically total hip replacements, due to their superior wear resistance, longer service duration and reduced inflammation resulting from such devices. In the mid-2000s, approximately 35% of the hip replacements in the US were metal-on-metal hip replacements based on CoCrMo alloys.[1] Despite its superior mechanical properties and wear resistance compared to metal-on-plastic or ceramics systems, CoCrMo is susceptible to corrosion due to tribological events such as joint movements and its constant exposure to corrosive body fluids. Grain boundaries have been commonly associated with the mechanical behaviour of alloys and their structure is especially significant. The coincidental site lattice (CSL) model is a powerful mathematical tool to characterize grain boundaries and identify `special' boundaries that display superior mechanical behaviour compared to regular high angle grain boundaries.[2] Previous work on stainless steel, copper, nickel, and aluminum alloys has demonstrated that, in accordance with CSL theory fitted with the Brandon criterion ( 15� ), grain boundaries with a low reciprocal coincident lattice density () have a relatively low energy compared to `general' high angle grain boundaries and high boundaries.[3] Solution-annealed high carbon CoCrMo hip implants are imaged using scanning electron microscopy (SEM) to investigate the grain boundaries in the retrieved sample. As shown in Figure 1 (A) and (B), the SEM images containing the grain boundaries at varying degrees of corrosion are classified via electron backscatter diffraction (EBSD) in order to index the crystallographic misorientations of the grains. The grain boundaries identified as special CSL boundaries (3-49) are then analyzed using a 3D optical microscope for their depth profiles. Preliminary EBSD scans show that there is some relation between the corrosion depth of grain boundaries and their geometry given by the parameter. We report that 50 out of the 85 nontwin CSL grain boundaries examined exhibited some form of corrosion resistance and 24 of these grain boundaries are above 11, which has been previously determined as a threshold for corrosion resistance.[4] By SEM imaging and visual analysis, we are able to differentiate between immune, corroded and semi-corroded boundaries which contain shallow, irregular divot-like features as shown in Figure 2. However, upon further examination of the grain boundary statistics, 17 and 29 boundaries seem to display stronger corrosion resistance behaviour than what its geometry predicts. From this observation, it is clear that there is a quantifiable correlation between the severity of corrosion at a grain boundary and its corresponding interfacial energy. Using the depth profiles of these CSL grain boundaries, a Microsc. Microanal. 21 (Suppl 3), 2015 774 relationship between the depth of the divots, semi-corroded and severely corroded grain boundaries and the geometry of these respective grain boundaries can be established. This relationship is crucial as it provides a more refined understanding of the exact geometries that show corrosion resistant behaviour and how to reduce corrosion in CoCrMo hip implants by using grain boundary engineering. References: [1] S. Pramanik, A. K. Agarwal, and K. N. Rai, Trends in Biomaterials and Artificial Organs,19 (2005), p. 15-26. [2] M. D. Sangid, et al, Materials Science and Engineering: A, 527 (2010), p. 7115-7125. [3] G. Palumbo, K. T. Aust. Acta Metall Mater, 38 (1990), p. 2343-2352. [4] P. Panigrahi, et al, J Biomed Mater Res Part B, 104 (2013), p. 850-859. [5] The authors acknowledge funding from the NSF on grant number CMMI-1030703. Figure 1: (A) A SEM image showing preferentially corroded grain boundaries and different grain boundary geometry. (B) The corresponding EBSD map showing the crystallographic orientations of the grains in order to determine grain boundary geometry. Figure 2: A SEM image showing the semicorroded boundaries and asymmetrical divot-like features.");sQ1[388]=new Array("../7337/0775.pdf","Wetting and dewetting of ultra-thin Ni films on Si and SiO2 substrates","","775 doi:10.1017/S1431927615004675 Paper No. 0388 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Wetting and dewetting of ultra-thin Ni films on Si and SiO2 substrates Klaus van Benthem Department of Chemical Engineering and Materials Science, University of California, Davis, CA 95616 Wetting of thin films on substrates and their solid state dewetting behavior at elevated temperatures is governed by the balance between free surface energies of film and substrate, and the respective interface energy. For many materials system, including Ni on silicon, solid-state reactions at the film substrate, defect segregation, and interfacial stress can cause interdiffusion across the interface plane and alter the wetting behavior [1]. For polycrystalline films dewetting is often initiated by grain boundary grooving at locations where grain boundaries intersect the free surface of the thin film. Aberration-corrected scanning transmission electron microscopy (STEM) combined with electron energy-loss spectroscopy is an ideal tool to characterize the cross-sectional interface structure and local bonding configurations, respectively. In situ transmission electron microscopy (TEM) further allows monitoring of changes in film morphology and interface structure as a function of temperature and time. However, the self-diffusion length of Ni at 864 K and 1296 K is 0.04 nm and 25 nm within 1s, respectively. Hence, dynamic transmission electron microscopy (DTEM) is required to evaluate dewetting behavior for homologous temperatures above 0.5. In this presentation recent results obtained for thin polycrystalline Ni films on both Si and SiO2 substrates will be reported. Thin films of Ni with nominal thicknesses ranging between 5 and 30 nm were deposited at room temperature by DC magnetron sputtering. When Ni is co-deposited with Pt on clean Si substrate surfaces, the formation of a presilicide layer below Ni1-xPtxSi films was observed with structure and composition distinctly different from previously observed diffusion layers [2]. It was found that during two-step rapid thermal annealing Ni interstitial diffusion can kinetically dominate over the formation of Ni silicide, which results in a metastable pre-silicide layer (Figure 1). The pre-silicide layer was found to limit diffusion of Ni into the Si substrate and, therefore, allows for the low-temperature growth of Ni2Si and NiSi [2]. Alternatively, when Ni is deposited on thermally grown amorphous SiO2 no interdiffusion between film and substrate takes place. In situ heating experiments revealed formation of holes within the metal film that nucleate at the Ni/SiO2 interface rather than at the free surface of the film. Ni islands were observed to retract, in attempt to reach equilibrium on the SiO2 layer. The formation of graphene layers between the metal film and the SiO2 substrate was observed during film agglomeration, which indicates lowering of the Ni/SiO2 interface energy (Figure 2). Cr, which is considered an impurity in this study, forms surface oxide layers on the free surface of SiO2 and the Ni islands. Cr does not prevent dewetting of Ni at elevated temperatures, but will likely have altered the equilibrium shape of the Ni islands [3]. The kinetics for the dewetting of Ni films from SiO2/Si substrates can be evaluated through DTEM. Initial experiments with laser energies leading to temperatures close to the melting point of Ni reveal metal drop formation on the substrate surface with morphologies typical for spinodal decomposition. In thin parts of the TEM sample metal droplets are ejected form the substrate surface, while interfacial stresses due to the misfit Microsc. Microanal. 21 (Suppl 3), 2015 776 in thermal expansion coefficients between film and substrate cause fracturing of the silicon substrate [4]. References [1] Thron AM, Greene P, Liu K, van Benthem K. Structural changes during the reaction of Ni thin films with (100) silicon substrates. Acta Mater 2012;60:2668�78. [2] Thron AM, Pennycook TJ, Chan J, Luo W, Jain A, Riley D, et al. Formation of presilicide layers below Ni1-xPtxSi/Si interfaces. Acta Mater 2013;61:2481�8. [3] Thron AM, Greene P, Liu K, van Benthem K. In-situ observation of equilibrium transitions in Ni films; agglomeration and impurity effects. Ultramicroscopy 2014;137:55�65. [4] Hihath S, van Benthem K et al., in preparation (2015) [5] This work was supported by a CAREER award of the US National Science Foundation (DMR-0955638), and a UC Laboratory Fee grant (#12-LR-238313). Fig. 1: Annular dark field STEM image of the Ni(Pt)Si layer deposited on the Si substrate. Underneath the silicide film the pre-silicide layers is observed that reveals Ni in different interstitial configurations. Reproduced with permission [2]. Fig. 2: High-angle annular dark field STEM image (left) of a Ni island supported by a roughly 10nm thick thermal SiO2 film. The right image shows the corresponding spectrum image with blue color representing Ni, red representing Cr, and green representing O. Reproduced with permission [3].");sQ1[389]=new Array("../7337/0777.pdf","Domain Structure of Bulk and Thin-Film Ferroelectrics By Transmission Kikuchi Diffraction","","777 doi:10.1017/S1431927615004687 Paper No. 0389 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Domain Structure of Bulk and Thin-Film Ferroelectrics By Transmission Kikuchi Diffraction Matthew J. Burch1, David T. Harris1, Chris M. Fancher1, Jon-Paul Maria1, and Elizabeth C. Dickey1. 1 North Carolina State University, Materials Science and Engineering, Raleigh, NC USA. Non-linear dielectrics are of great interest due to their integration into modern electronics, including tunable and memory devices [1]. Recently, much work has been invested into obtaining bulk-like properties of dielectric thin films at lower processing temperatures than are generally required for maximum performance. In particular, Harris et. al. has shown impressive dielectric properties of thin films at processing temperatures as low as 900�C for barium titanate (BaTiO3), contrasted with bulk, where similar properties would need 1250�C or higher processing temperatures [2]. While dielectric properties (permittivity, tunability, etc.) of ferroelectric oxides are generally considered to be the result of the extrinsic effect of domain wall motion, ferroelectric domains are rarely observed in small grained (<0.4�m) samples [3]. Thus, if these domains are present, their formation and structure is not well defined or understood. This study utilizes modern day diffraction and imaging techniques to understand nano-domains and their impact on the dielectric properties of small-grained bulk and thin film dielectrics. Transmission Kikuchi diffraction (TKD) is a relatively new technique utilizing an electron backscatter diffraction (EBSD) detector in a scanning electron microscope (SEM), where electrons are transmitted through a thin sample and captured similarly to traditional EBSD, as shown in figure 1. This geometry allows for significantly enhanced resolution (<10nm) compared to traditional EBSD (~80nm) and allows for examination of domain structure in samples with grain sizes too small for traditional EBSD. In this study, TKD was used to identify and characterize the domain structure of small-grained samples composed of thin-film BaTiO3 or lead zirconate titanate (Pb(Zr,Ti)O3). Furthermore, we wish to use these studies to find the fundamental limit to the spatial and strain resolution of TKD. Figure 2 shows an example TKD band-contrast image and inverse pole figure (IPF) maps of bulk Pb(Zr,Ti)O3, with the domain structure clearly present in both images. A band contrast image is generated when the intensity of the Kikuchi bands is compared to the overall strength of the TKD pattern. Areas where the Kikuchi bands are less intense, such as grain boundaries, defective areas, or domain boundaries, will appear darker than areas of strongly diffracting Kikuchi bands, such as in the middle of a grain or domain. The striped domains shown in figure 2 correspond to 90� domains, as observed by both the IPF Z (grain orientations relative to out-of-plane direction) and IPF Y (grain orientation relative to the vertical in-plane direction) images. This investigation shows these striped domains are on the order of 60-80 nm in width. Figure 3 shows the TKD patterns within a single grain of a large grained BaTiO3 thin film sample. The orientation change from the first and second TKD patterns is consistent with a 90� domain. These TKD patterns demonstrate the extreme sensitivity of TKD as a characterization technique for domain structures in ferroelectrics as the c/a ratio of BaTiO3 is only ~1.005 [4]. Microsc. Microanal. 21 (Suppl 3), 2015 778 References: [1] K. Tagantsev, et. al., Journal of Electroceramics 11 (2003), p. 5 [2] D. T. Harris, et. al., Applied Physics Letters 103 (2013), p. 012904 [3] G. Arlt and D. Hennings, Journal of Applied Physics 58 (1985), p. 1619 [4] The authors acknowledge funding from the Center for Dielectrics and Piezoelectrics at the North Carolina State University. This material is based upon work supported by the National Science Foundation Graduate Research Fellowship for M. J. Burch under Grant No. DGE-0946818. The authors acknowledge the use of the Analytical Instrumentation Facility (AIF) at North Carolina State University, which is supported by the State of North Carolina and the National Science Foundation. Figure 1: Schematic diagram of transmission Kikuchi diffraction (TKD). The thin (<150 nm) sample is oriented so the incoming electron beam is transmitted through the thin sample, which is then collected by the EBSD detector. $" !" #" Figure 2: (A) Band contrast image of bulk PZT samples with the domains clearly evident. The following images are IPF images, with the 90� domains clearly evident for both the Z-(B) and Y-(C) relative orientation IPF images. !" #" Figure 3: EBSD patterns of a within a single grain of a barium titanate thin film, showing a change in orientation that is consistent with a 90� domain.");sQ1[390]=new Array("../7337/0779.pdf","Origin of Ferroelectricity in Thin Film HfO2 Probed by Revolving STEM and PACBED","","779 doi:10.1017/S1431927615004699 Paper No. 0390 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Origin of Ferroelectricity in Thin Film HfO2 Probed by Revolving STEM and PACBED Xiahan Sang1, Everett D. Grimley1, Tony Schenk2, Uwe Schroeder2 and James M. LeBeau1 1. 2. Department of Materials Science and Engineering, North Carolina State University, Raleigh, NC NaMLab gGmbH/TU Dresden, Noethnitzer Str. 64, D-01187 Dresden, Germany Ferroelectric HfO2 thin films have recently become compelling candidate materials to replace leadcontaining ferroelectrics [1-3]. Though many efforts have sought to control HfO2 ferroelectricity through doping and capping layer confinement, the multiphase nature of the films has limited our understanding of the governing mechanisms. Current hypotheses suggest that ferroelectricity arises in these films through a stabilized non-centrosymmetric Pca21 [2,4] or Pmn21 [4] orthorhombic phase, yet x-ray diffraction studies have been unable to unambiguously refine the space groups of the phases present due to the complex nature of the thin film structures [2]. In this talk, we will present direct experimental evidence of the presence of the ferroelectric orthorhombic Pca21 phase in Gd-doped HfO2 thin films. We utilize a combination of revolving scanning transmission electron microscopy (RevSTEM) [5] and position averaged convergent beam electron diffraction (PACBED) to characterize the structure of the films directly at the nanoscale. In particular, we will demonstrate that the high spatial resolution of STEM is required to uniquely identify the phases present within the film. With the combined accuracy and precision of RevSTEM, we will show that lattice parameters measured directly from real space confirm the presence of both monoclinic and orthorhombic phases. For example, different projections of the orthorhombic phase are shown in Fig. 1 (a) � (d). Moreover, the measured lattice constants (Fig 1. table) were inconsistent with the ferroelectric Pmn21 phase [4], but did not sufficiently delineate between the other candidate phases. To verify the observed orthorhombic phase is non-centrosymmetric, we will discuss results from PACBED, which as has been demonstrated to readily reveal spontaneous polarization [6]. For example, Figs. 2 (a � c) compare [110] oriented HfO2 PACBED patterns for (a) simulated centrosymmetric (here Pbca), (b) non-centrosymmetric Pca21, and experiment (c). The experimental pattern in Fig. 2 (c) shows breaking of mirror symmetry along the vertical plane in good agreement with the simulated pattern for Pca21 (Fig. 2 (b)). Broken symmetry in PACBED is indicative of non-centrosymmetry suggesting that Pca21 is the most likely candidate structure for the orthorhombic phase in HfO2. Finally, we will discuss the possible sources of stabilization for the orthorhombic structure and possible routes to further control behavior [7]. Microsc. Microanal. 21 (Suppl 3), 2015 780 References: [1] T. S. B�scke et al., Applied Physics Letters 99 (2011), p. 102903 [2] J. M�ller et al., Nano Letters, 12 (2012), p. 4318 [3] D. Martin et al., Advanced Materials, 26 (2014), p. 8198 [4] T. D. Huan et al., Physical Review B, 90 (2014), p. 064111 [5] X. Sang and J. M. LeBeau, Ultramicroscopy 138 (2014), p. 28 [6] J. M. LeBeau et al., Applied Physics Letters 98 (2011), p. 052904 [7] Funding for this research was provided by the National Science Foundation (Award Number DMR1350273) and the German Research Foundation (Deutsche Forschungsgemeinschaft) in the frame of the "Inferox" project. This material is based upon work supported by the National Science Foundation Graduate Research Fellowship for EDG under Grant No. (DGE-1252376). XS, EDG, and JML acknowledge use of the Analytical Instrumentation Facility (AIF) at North Carolina State University, which is supported by the State of North Carolina and the National Science Foundation. Authors acknowledge Christoph Adelmann for HfO2 film deposition, and Tina Sturm and Almut P�hl from Leibniz IFW Dresden, Germany for FIB preparation of the lamellas. Figure 1. Experimental RevSTEM images for Gd:HfO2 (scale bar: 1 nm) with zone axes labels assuming a Pca21 structure. The graphic depicts Pca21 unit cell with gold and red circles representing Hf and O atoms respectively. The table shows comparison of DFT simulated lattice parameters with parameters measured from RevSTEM images. Figure 2. Simulated (SIM) Pbca (a), Pca21 (b), and experimental (EXP) (c) PACBED patterns for HfO2 oriented along the [110] zone axis. Patterns in both (b) and (c) have broken mirror symmetry along the vertical plane indicated by the arrows. Black levels in (c) adjusted to account for noise.");sQ1[391]=new Array("../7337/0781.pdf","Role of Interface Structure and Chemistry in Resistive Switching of NiO Nanocrystals on SrTiO3","","781 doi:10.1017/S1431927615004705 Paper No. 0391 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Role of Interface Structure and Chemistry in Resistive Switching of NiO Nanocrystals on SrTiO3 Xuan Cheng1, Jivika Sullaphen1, Matthew Weyland2,3, Hongwei Liu4, and Nagarajan Valanoor1 1. 2. School of Materials Science and Engineering, UNSW Australia, Sydney, Australia Department of Materials Engineering, Monash University, Melbourne, Australia 3. Monash Centre for Electron Microscopy, Monash University, Melbourne, Australia 4. Australian Centre for Microscopy and Microanalysis, University of Sydney, Sydney, Australia Nickel oxide (NiO) is a binary metal oxide that is found to have resistive switching (RS) properties, which makes it a viable candidate for next generation resistive random access memories (RRAMs). The demand for high-density memories has concentrated on RS materials with scalable dimensions. In these nanoscaled devices, the presence of point and line defects, composition inhomogeneity, and atomic interdiffusion interfaces have been shown to have a significant impact on properties. The objective of this study is to investigate the role of interface morphology and chemistry in resistive switching behavior of NiO epitaxial nanocrystals fabricated on (001) SrTiO3 (STO) substrates [1]. NiO nanostructures are found to have {111} and {113} orientated facets. The epitaxial relationship between the nanostructure and the substrate can be confirmed. The crystallographic orientations of the nanostructures are [110]NiO//[110]STO inplane and [100]NiO//[100]STO out of plane, confirming that the nanostructure and substrate are aligned in a cubic-on-cubic fashion. Aberration corrected high angle annular dark field (HAADF) scanning transmission electron microscopy (STEM) reveals the interface between the NiO nanocrystals and the underlying STO substrate to be rough, irregular, and have a lower average atomic number than the substrate or the nanocrystal [1]. Energy dispersive x-ray spectroscopy and electron energy loss spectroscopy (shown in Figure 1) confirm both chemical disorder and a shift of the energy of the Ti L2,3 peaks. Analysis of the O K-edge profiles in conjunction with this shift, implies the presence of oxygen vacancies at the interface. This sheds light into the origin of the previously postulated minority carriers' model to explain resistive switching in NiO [2]. HRTEM and HAADF-STEM reveal that the substrate is deformed (see Figure 2) so as to back-fill around the base of the nanostructure in order to promote wetting. Remarkably, the substrate under the nanocrystal is physically pulled towards the nanostructure, thus forming a rim around the nanocrystal [4]. References: [1] X. Cheng et al, Appl. Phys. Lett. Mater. 2 (2014), p. 032109 [2] J. Sullaphen et al, Appl. Phys. Lett.100 (2012), p. 203115 [3] The research at UNSW was supported by an ARC Discovery and LIEF Grant. The research at Monash Centre for Electron Microscopy (MCEM) used equipment funded by Australian Research Council Grant No. LE0454166 (FEI Titan 80�300 FEGTEM). We thank MCEM, Microscopy Units (EMU) at UNSW and Australian Centre for Microscopy and Microanalysis (ACMM) at USYD for the provision of equipment and technical support. The author would also like to acknowledge Dr. Ye Zhu Microsc. Microanal. 21 (Suppl 3), 2015 782 for his contribution in discussing the results. Figure 1. (a) Ti L edges of the selected points from EELS line scan. (b) ADF intensity plot and EEL spectrum image simultaneously acquired with EELS. (c) Simultaneously acquired O K edges from point a, d, and f. Figure 2. Schematic of top view image (top) and cross-sectional transmission electron microscope (TEM) image (bottom) shows different appearance of substrate in TEM image due to the cut position of NiO in TEM sample preparation. In the top-view image, NiO nanocrystals, the substrate, and the overlapped area due to the substrate deformation are shown as yellow, blue and green colour respectively. The deformed substrate around NiO is indicated with blue dotted lines");sQ1[392]=new Array("../7337/0783.pdf","Local Domain Wall Structure in Polycrystalline BiFeO3","","783 doi:10.1017/S1431927615004717 Paper No. 0392 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Local Domain Wall Structure in Polycrystalline BiFeO3 A. Bencan1, G. Drazic2, H. Ursic1, N. Sakamoto3, B. Jancar4, N. Wakiya5, H. Suzuki3, B. Malic1, D. Damjanovic6 and T. Rojac1 1. 2. Electronic Ceramics Department, Jozef Stefan Institute, Ljubljana, Slovenia. Laboratory for Materials Chemistry, National Institute of Chemistry, Ljubljana, Slovenia. 3. Research Institute of Electronics, Shizuoka University, Hamamatsu, Japan. 4. Advanced Materials Department, Jozef Stefan Institute, Ljubljana, Slovenia. 5. Graduate school of Science and Technology, Shizuoka University, Hamamatsu, Japan. 6. Ceramics Laboratory, Swiss Federal Institute of Technology � EPFL, Switzerland. In multiferroic BiFeO3 (BFO), the domain walls have properties that are fundamentally different from those of the bulk, e.g., it was shown that the domain walls in BFO thin films possess higher electrical conductivity than the surrounding [1]. While detailed structural studies of domain walls were performed on BFO thin films, these issues are only at their early research stage for the bulk BFO ceramics. This is because the preparation of BFO ceramics is non-trivial; Bi2O3 volatility, variation in Fe valence state, possible reactions of BFO with impurity elements, lead to formation of undesired secondary phases [2]. Recently, we demonstrated that the local conductivity at domain walls influences the macroscopic properties of bulk BFO [3]. In order to make further steps in the understanding of the domain wall conductivity and electromechanical response of BFO ceramics, we investigate the local structure of domain walls using different microscopy methods. As opposed to epitaxial thin films, the determination of the type of the domain walls in polycrystalline ceramics with randomly oriented grains is practically impossible by using just AFM/PFM. By identical location approach, we were able to determine the type of individual domain walls in BFO ceramics on a micro level with the combination of SEM/EBSD and AFM/PFM analysis. BiFeO3 ceramics were prepared by reactive sintering of a mechanochemically activated Bi2O3�Fe2O3 powder mixture at 760�C or 820oC in a packing powder. The ceramics had a relative density of 93%. The structure of domain walls in the BFO ceramics was analyzed by different scanning electron microscopy methods on a micro level (SEM, EBSD), down to atomic scale using atomically resolved Cs-probe corrected JEM-ARM200CF microscope. In rhombohedral BFO ceramics, three different domain walls, i.e., 180o on {111} planes, 71o on {110} planes and 109o on {100} planes were observed. Experimentally, a combination of SEM/EBSD and cAFM studies shows that all types of domain walls, i.e., 71o, 109o and 180o, are conductive in BFO ceramics [Figure 1]. We precisely analyzed the local structure of domain walls at the atomic level by measuring the displacement of Fe cations within the periodic structure [Figure 2]. In this way, we were able to determine the local polarization direction, local variations in lattice parameters and potential local strains on the level of unit cells. The complexity of domain wall structure will be discussed and related to the macroscopic, functional response of the polycrystalline BFO. Microsc. Microanal. 21 (Suppl 3), 2015 784 References: [1] J Seidel et al, Nat. Mater. 8 (2009), p.229 [2] T. Rojac et al, J. Am. Ceram. Soc. 94 (2014), p.1993 [3] T. Rojac et al, Adv. Funct. Mater. 2015, doi:10.1002/adfm.201402963 [4] This work was supported by the Slovenian Research Agency (P2-0105 and J2-5483). Figure 1. SEM (a) and corresponding electric current image (b) of 109�, 71� domain walls in BiFeO3 ceramics. The types of domain walls were identified by EBSD analysis. Electric current images at +11 V showed higher electric current signal at domain walls as compared to that of the surrounding domains. Figure 2. HAADF-STEM image of 109o domain wall. Inserts show Fe displacement from the central position.");sQ1[393]=new Array("../7337/0785.pdf","X-Ray Microanalysis of Art Glass Surfaces","","785 doi:10.1017/S1431927615004729 Paper No. 0393 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 X-Ray Microanalysis of Art Glass Surfaces R.B. Simmons Georgia State University, Biology Dept., Atlanta, GA 30303 Briarwillow, LLC; Atlanta, GA 30345 The Studio Glass Movement began in 1962 when Harvey Littleton began exploring the idea of creating art from molten glass in a garage on the grounds of the Toledo Museum of Art in Toledo, OH. Since that time the world of art glass has expanded to include a range of items from large scale sculpture to hand-crafted jewelry. Glass formulations have evolved to bring a host of new colors into artists' palettes and come from both large and small scale manufacturers. Torch technology has also advanced to the point of bringing glass art out of the large scale `hot shop' and into the studios and homes of hobbyists. This expansion of glass art production has introduced both the excitement of the creative process and the hazards associated with such work to many people who often are not familiar with many aspects of the latter. In general, the colorants in glass are metallic compounds (oxides, chlorides, etc.), some of which may be hazardous to both the glass worker and the consumer. This project began as a study of metallic surfaces on glass beads used in a program for providing emotional and psychological support for children in treatment for cancer and other lifethreatening illnesses (Beads of Courage, Inc.). The commonly used `reduction surface' technique in art glass creates a metallic `luster' surface on the glass. Reduction surfaces can be quite colorful and are popular in art glass jewelry. It is this type of surface which can be of concern to both the artist and end user of the product. As a rule these effects are produced when a glass with a high metal content is exposed to a flame with an excess fuel to oxygen ratio (aka reduction flame). This flame type tends to remove oxygen from metal compounds and leave pure metal on the surface. Problems come about in two areas: (1) release of vaporized metal into the workspace of the artist and (2) the chemical nature of the metallic surface which may come into contact with the wearer's skin. A number of different heavy metals are used for this type of art glass used in jewelry, some of which are both toxic and regulated by government agencies. Class color names may be misleading when it comes to disclosure of the types of metals involved. Microanalysis of these popular glasses and dissemination of the results can be of critical importance for both worker safety and consumer protection. While immediate toxicity to the consumer is practically unknown, at least one metal commonly found in reduction surfaces (Pb) is controlled by the Consumer Products Safety Commission and may present complex and expensive legal issues to distributors and consumers of finished art glass, especially if the objects are to be handled by children (1). Gold appearing surfaces are often assumed to be metallic gold, silver surfaces silver, etc. Analysis of these surfaces shows that visual color and color name may, in some instances, be misleading. Samples of `reduction' glass from different manufacturers were purchased on the commercial market. Each was then flameworked to produce the metallic reduction surface, and analyzed to determine the elements present. The samples were examined in a LEO 1450vp scanning electron microscope operating at kV20 and were further analyzed by Energy Dispersive X-ray Microanalysis (EDS) using a Rontec X-Flash X-ray microanalysis system. Results showed that while many surfaces were very similar in appearance the chemical composition on the surface could be quite different. Examples of these surfaces are seen in Figure 1. One heavy Microsc. Microanal. 21 (Suppl 3), 2015 786 metal in particular, Pb, is not considered safe for human contact and has been banned from use in any materials which may be contacted by children under 12. Beads of Courage, Inc. routinely uses artist made glass beads in their support programs for children in treatment for cancer and other serious illnesses and expressed concerns related to the safety of metallic surface beads and the ability of their staff to discern difference glass types. Based on these results Beads of Courage, Inc., made the decision to remove all metallic surface glass beads from circulation in their programs. References 1. Consumer Product Safety Improvement Act of 2008, http://www.cpsc.gov/cpsia.Pdf Figure 1. A. Spectrum from `Psyche' indicating the presence of Ag on the reduction surface. B. Psyche in neutral and reduced form. C. Spectrum of `Silver Plum' showing the presence of Mn. D. Neutral vs. reduced Silver Plum showing the reflective silver-appearing Mn surface. E. Spectrum from Gold Ruby showing the presence of Pb and trace amounts of Se. F. Neutral Ruby Gold glass compared to the gold-to-silver reduction surface.");sQ1[394]=new Array("../7337/0787.pdf","Homogeneity and Sample Preparation from Grams to Microns using NAA, XRF, and SEM-EDS","","787 doi:10.1017/S1431927615004730 Paper No. 0394 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Homogeneity and Sample Preparation from Grams to Microns using NAA, �XRF, and SEM-EDS Abigail P. Lindstrom1, Jeffrey M. Davis2, Rolf Zeisler1, Nicholas WM Ritchie1, and Richard M. Lindstrom1 1. National Institute of Standards and Technology, Materials Measurement Laboratory, 100 Bureau Drive, Gaithersburg, MD 20899 USA 2. PNDetector GmbH, Otto-Hahn-Ring 6, 81739 M�nchen, Germany When a sample is described as homogeneous, we mean that there is a lack of variation in a measured property over the volume of the sample. The notion of homogeneity is somewhat ill defined because all samples are inhomogeneous at sufficiently small length scales. So to be precise, one must say that a sample is homogeneous to a certain level of variation over a certain length scale. Whether a sample can be considered homogeneous depends upon the application and the length scale, mm or �m. During Instrumental Neutron Activation Analysis for the certification of NIST SRM 2711a, a soil reference material, excess variability was found in the Al concentration above that expected from counting statistics. Analyses were done by �XRF as an additional test of homogeneity. To investigate further at shorter length scales, we made additional measurements using �XRF and SEM-EDS x-ray spectrum imaging. For two �XRF preparations, one of which was used for spectrum mapping, a small amount of the sample was embedded in epoxy. For one preparation, the bottom of the mount was examined without polishing. For the second mount, the epoxy was polished. The samples were imaged in an Eagle III �XRF system, which has a nominal spot size of approximately 50 �m. The amount of material probed by the �XRF was significantly smaller than that done by INAA and larger than that done by the SEM (see below). The polished mount was also investigated by x-ray spectrum imaging on a TESCAN MIRA3 Schottky field emission SEM with 4 Pulsetor silicon drift detectors at 15 keV, 1 nA and 51.2 ms/displayed pixel. The point spectra were quantified against standards using NIST DTSA-II [xx] (Figure 2). We also found that there were some differences between the two �XRF preparations. The mount that was examined from the bottom of the epoxy was much higher in fines than the other mount where the mount that had been polished, and thus was sampled from a more representative location was less homogeneous. The xray spectrum mapping showed a wide variety of compositions and almost no homogeneity at all. The take-away message is that homogeneity like beauty is in the eye of the beholder. A sample that is homogeneous on one length scale may be well suited for its intended purpose but totally unsuited to other purposes at other length scales. Microsc. Microanal. 21 (Suppl 3), 2015 788 srm2710a.1a srm2710a.139a srm2710a.483a srm2710a.756a srm2710a.936a srm2710a.1118a srm2710a.1255a srm2710a.1459a srm2710a.1973a srm2710a.1973c srm2710a.1716a srm2710a.1871a srm2710a.2061a Wtd mean Chisq/df ppm Al 59258 60141 59906 58928 59701 59252 58832 59725 60322 60136 59179 59917 59680 59621 1.06 � 492 396 649 532 524 472 479 496 447 502 401 476 512 srm2711a.1a srm2711a.368a srm2711a.484a srm2711a.733a srm2711a.1094a srm2711a.1342a srm2711a.1563a srm2711a.1688a srm2711a.1212a srm2711a.1950a srm2711a.2290a srm2711a.last.a ppm Al 67710 67698 67981 67943 65774 66388 67129 67209 65675 66569 66507 66458 � 408 421 429 416 413 447 408 409 411 401 399 404 66912 3.77 Figure 1. The Al concentration after INAA of SRM 2711a and a comparable material SRM 2710a. Figure 2. Elemental maps of a region on the polished mount collected on the TESCAN MIRA3 at 15 keV and 51.2 ms per displayed pixel (FOV = 512 �m).");sQ1[395]=new Array("../7337/0789.pdf","Quantitative Analysis of Sulfur in Polymer Materials.","","789 doi:10.1017/S1431927615004742 Paper No. 0395 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantitative Analysis of Sulfur in Polymer Materials. Stuart McKernan1, Mary I. Buckett1 and William G. Stratton1 1. 3M Corporate Research Analytical Laboratory, 3M Center, Building 201-BE-16, St. Paul, MN 55144 Energy Dispersive Analysis (EDS) in the TEM relies on being able to detect X-rays produced by the different atomic species within a sample as it is irradiated by the incident electron beam. A number of factors affect how the number of X-rays of each species that are detected relate to the actual chemical composition of the sample; in particular the proportions of X-rays that are absorbed in the specimen and surrounding environment between being generated and being detected and analyzed will have a significant effect on the final composition reported. Typically for light elements in an irregular sample the proportion of detected versus generated is an unknown but very influential quantity, leading to quantification values with large error bars. In this paper we report our efforts to determine optimal criteria to obtain reasonable values for the Sulfur:Carbon ratio and the minimum detectable concentration for Sulfur in these polymer materials. The most accurate composition determinations are obtained from analyses using known standards under the same analytical conditions as the unknown material. This is more accurate than calculating the compositions using the as-provided k-factors and standard correction factors in a "standardless analysis" [1]. To that end standard spectra from a samples of known composition (using as close to ideal acquisition parameters as possible) were created. Test samples were prepared in the form of cured epoxy with various loadings of Jeffamine Sulfonate (JAS). This produced a set of homogeneous S-containing standards. The nominal concentration of JAS in the epoxy for each sample was designed as 5%, 10%, and 20%. The actual concentrations of S in the test samples were determined by halogen analysis by combustion ion chromatography as: 0.22, 0.50, and 0.98 wt% respectively. TEM samples were prepared from the JAS/epoxy by ultramicrotomy to produce thin, parallel-sided sections 100nm thick that were supported on an ultrathin carbon film. The TEM samples were then analyzed in a JEOL 2100F operating at 200kV using a Thermo Scientific Noran EDS System. Sample homogeneity and beam-stability were examined by comparing multiple spectra from different areas of the samples, or the same area during multiple scans. The carbon content of the TEM support film was subtracted from each active spectra by acquiring spectra from an adjacent region of the support film alone for the same time as the actual film spectrum and then subtracting the two spectra. In figure 1, four adjacent areas on a section of the 20% JAS film are outlined and were analyzed separately and together. They show some distinct variations in composition, together with a bright damaged spot. In figure 2, results of multiple analyses from the same area indicate that after the first scan, where the oxygen signal appears to decrease slightly, the signals (and particularly the S content) appear to remain stable during the multiple EDS scans. For the standard spectrum, using the highest S-content (20 wt% JAS) TEM sample, multiple spectra were acquired for 1000s from different areas. Those summed spectrum were then used to define the experimental C and S k-ratios in the NSS 3.0 software. Additional spectra from the 20% JAS content and the 10% and 5% content were then quantified using these ratios. Comparisons between the EDS results and the Halogen Analysis are shown in figure 3. Using this standards-based analysis reasonably good agreement between the EDS composition and that obtained by halogen analysis was obtained for the 20% JAS sample (corresponding to a sulfur content of Microsc. Microanal. 21 (Suppl 3), 2015 790 ~1 wt %). The samples appeared to be sufficiently robust under the electron beam to produce consistent composition analysis even with 1000 s dwell times. Both the 20% JAS and 10% JAS samples produced samples with detectable S content at the 1 wt% S and 0.5 wt% S level. The 5% JAS produced a sample with 0.22 wt% S that is below the detection limit by EDS under these conditions. [2] References: [1] D B Williams and C B Carter "Transmission Electron Microscopy", (Plenum, New York) 1996. [2] The authors acknowledge Jay Lomeda of the Corporate Research Materials Lab at 3M for the preparation of the samples used in this work. Scan area point1 point2 point5 C-K 95.150 97.305 98.046 97.205 96.572 O-K 3.981 2.464 1.555 2.408 2.409 S-K 0.868 0.231 0.398 0.387 1.019 + point3 point4 Figure 1. Four adjacent areas outlined in blue analyzed independently, and together outlined in orange Figure 2. Plot of the concentration of O and S as a function of scan number from the same area. Sulfur (wt%) Halogen Analysis 5 % JAS 0.22 � 0.001 10 % JAS 0.50 � 0.002 20 % JAS 0.98 � 0.02 Sample Sulfur (wt%) (S)TEM EDS 0.06 � 0.04 0.75 � 0.47 0.86 � 0.25 Detectable by (S)TEM Standardless EDS value EDS Analysis (for comparison) No 0.015 � ~5.0 Yes 0.164 � ~5.0 Yes 0.159 � ~5.0 Figure 3. Comparison of cured epoxy blocks with different JAS content.");sQ1[396]=new Array("../7337/0791.pdf","Quantification of the Boron Speciation and Cu Oxidation States in Alkali Borosilicate Glasses by Electron Energy Loss Spectroscopy","doi:10.1017/S1431927615004754","791 doi:10.1017/S1431927615004754 Paper No. 0396 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantification of the Boron Speciation and Cu Oxidation States in Alkali Borosilicate Glasses by Electron Energy Loss Spectroscopy Guang Yang1, Shaodong Cheng1, Chao Li1, Mingying Peng2, Yuanzheng Yue3 Electronic Materials Research Laboratory, Key Laboratory of The Ministry of Education& International Center for Dielectric Research, Xi'an Jiaotong University, Xi'an, China 2. State Key Laboratory of Luminescent Materials and Devices, South China University of Technology, Guangzhou, China 3. Section of Chemistry, Aalborg University, DK-9000 Aalborg, Denmark Alkali borosilicate glasses have been widely used for a long time due to their good chemical/physical properties. In borosilicate glasses, boron has two structural configurations: trigonal BO3 and tetrahedral BO4 units1. The BO4 tetrahedra participate in the three-dimensional network structures of borosilicate glasses, therefore they are preferred for some special applications, such as nuclear waste immobilization. Several techniques have been developed to quantify the fraction of BO4 in borosilicate glasses, including X-ray absorption near edge structure (XANES), Raman spectroscopy and nuclear magnetic resonance (NMR)2. Among these tools, 11B magic-angle spinning (MAS) NMR is one of the most reliable methods to quantify the BO4 fraction, N4= [BO4]/([BO4] + [BO3]), in boron glasses. However, solid-state NMR instrumentation is not commonly found in all research laboratories. Electron energy loss spectroscopy (EELS) can also be applied to quantify N4 since the BO3 and BO4 units exhibit characteristic features in the corresponding spectra (Fig. 1a). The quantification of N4 by EELS has been tried but the obtained N4 data was lower than the actual N4 values measured by NMR, due to the electron beam irradiation damage which causes the transformation of BO4 into BO3 units during the signal acquisition3. In this work, we have developed a method based on EELS data acquisition and analysis, which enables determination of the boron speciation in a series of ternary alkali borosilicate glasses with constant molar ratios. A script in DigitalMicrograph for fast acquisition (~0.05s) of EELS has been designed, from which the boron K-edge spectra can be obtained with minimum electron irradiation damage (Fig. 1b). The fraction of BO4 tetrahedra can be obtained by fitting the experimental data with linear combinations of reference spectra with special criteria. The measured BO4 fractions (N4) obtained by EELS are consistent with those from 11B MAS NMR data, suggesting that EELS be an alternative and convenient way to determine the N4 fraction in glasses. Three optically transparent colorful (red, green and blue) glasses were synthesized by sol-gel method. All glasses have the same composition but the annealing atmosphere is different. The exhibiting color is believed to be due to the different oxidation states of Cu in these glasses. XRD and XPS have been applied to these glasses but no crystalline phase or Cu signal was detected. SEM and TEM analysis shows that nano-sized precipitates are homogeneously distributed in the glass matrix. The precipitates were analyzed by STEM-EELS. From the STEM images it is observed that the nano-precipitates are brighter than the glass matrix, with the combination of EELS spectra imaging analysis they are found to be Cu rich precipitates. The oxidation states of Cu are quantified by fitting the Cu L2,3 edge spectra from the precipitates with 1. Microsc. Microanal. 21 (Suppl 3), 2015 792 the reference spectra from single valenced Cu reference compounds (Fig. 2). The oxidation states of Cu in the precipitates are found to be 0, +0.7 and +1.94 for the red, green and blue glass, respectively. Density functional theory (DFT) was used to find the qualitative correlation between the oxidation states of Cu and the apparent color of the glass. The calculated optical absorption and reflection spectra could verify the Cu oxidation states obtained by EELS. References: [1] H. Sauer et al., Ultramicroscopy, 49 (1993)198 [2] M. Moesgaard et al., Chem. Mater., 22 (2010)4471 [3] G. Yang et al., Phys. Chem. Glasses-B, 47 (2006) 507 [4] The authors acknowledge the funding from National Natural Science Foundation of China (51202180), the Fundamental Research Funds for the Central Universities in China and the 111 Project of China (B14040). Figure 1. (a) Plots of boron K-edge reference fingerprint spectra of the BO3 and BO4 units found in the minerals vonsenite and rhodizite1, respectively. (b) Experimental boron K-edge spectra of the LBS (red), NBS (blue) and KBS (green) glasses. Figure 2. Cu K-edge EELS spectra of (a) Cu reference materials and (b) NPs in different glasses and the corresponding curve fitting. The solid lines are the summed spectra, and the dotted lines are fitted curves");sQ1[397]=new Array("../7337/0793.pdf","Elemental Analysis of Rare-earth Magnet Utilizing Cathodoluminescence","","793 doi:10.1017/S1431927615004766 Paper No. 0397 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Elemental Analysis of Rare-earth Magnet Utilizing Cathodoluminescence Susumu Imashuku1,2 and Jun Kawai1 1. 2. Department of Materials Science and Engineering, Kyoto University, Kyoto, Japan. Institute for Materials Research, Tohoku University, Sendai, Japan. Insulating materials and semiconductors emits light in the ultraviolet to infrared region by the bombardment of electrons. This phenomenon is called Cathodoluminescence (CL). CL phenomenon has been applied to analysis in many areas: an evaluation of defect distribution in semiconductor, an investigation of geological history of minerals, and so on. Especially, materials containing rare earth elements have high luminescence intensity. The detection limit of CL analysis is as low as ppb order. In the present study, we tried to apply CL analysis for evaluating rare earth elements in magnets. We also tried to sort magnets containing rare earth elements from their luminescent colors with a portable CL spectrometer we realized [1]. Analysis of rare earth elements in magnets is quite important for recycling due to high cost of rare earth elements. Samples analyzed in the present study were a samarium-cobalt magnet and a neodymium magnet. We measured chloridized magnets because magnets themselves do not show CL phenomenon. The magnets were dissolved in 35% hydrochloric acid (HCl), and the solutions were dried in vacuum. Then, the vacuum-dried residues were measured with a SEM-CL system we customized. The SEM-CL system consisted of conventional SEM (JEOL, JSM-5610LVS), optical fiber, and optical spectrometer (Ocean optics, QE65Pro). The accelerating voltage and electron beam current of the SEM-CL system were set to 15 kV and 2 A, respectively. Measurement duration was set to 100 seconds. Pressure in a chamber of the SEM-CL system was about 10-3 Pa during the measurement. A schematic view of the portable CL is shown in Fig. 1. +z plane of a single crystal of LiTaO3 with 6 mm � 6 mm in x-y plane and 5 mm in z-axis was attached on a Peltier device with silver paste. A gold wire was set in a needle holder. The needle holder was attached to the LiTaO3 crystal with silver paste. We captured CL images of the vacuum-dried residues through a glass viewport using a digital singlelens reflex camera equipped with a close-up lens with a focusing distance of 100 mm. Exposure time of the camera was set to 100 sec. Measurements were carried out during cooling the LiTaO3 crystal after heating at 125 �C. Pressure of the sample chamber was set to 1 Pa. Figure 2 (a) shows CL spectrum of a vacuum-dried residue of the samarium-cobalt magnet alone with CL spectrum of samarium chloride (SmCl3). Peak positions of the vacuum-dried residue of the samarium-cobalt magnet were the same as those of SmCl3. Figure 2 (b) shows CL spectrum of a vacuum-dried residue of the neodymium magnet. We confirmed that the neodymium magnet contained neodymium (Nd) (19.3%), praseodymium (Pr) (4.8%), dysprosium (Dy) (1.1%), terbium (Tb) (2.3%) by inductively-coupled plasma atomic emission spectrometry. By comparing a CL spectra of NdCl3, PrCl3, DyCl3, and TbCl3 as shown in figure 2 (b), peaks of the vacuum-dried residue of the neodymium magnet were indexed as shown in figure 2 (b). More than 1% of rare earth elements in the neodymium magnet were detected by SEM-CL analysis. Thus, we succeeded in detecting above 1% of rare earth elements in magnets by SEM-CL analysis by vacuum-drying solutions dissolved the magnets into HCl. We next tried to sort the magnets by CL images captured with the portable CL spectrometer. Figure 2 Microsc. Microanal. 21 (Suppl 3), 2015 794 (c) show CL image of the vacuum-dried residue of the samarium-cobalt magnet. It produced orange luminescence which corresponds to a color of the strongest peaks of the vacuum-dried residue of the samarium-cobalt magnet as shown in figure 2 (b). As for the vacuum-dried residue of the neodymium magnet, intensity was too low to detect its luminescence. We extracted iron ions from the solution the neodymium magnet dissolved using methyl isobutyl ketone [2] because the neodymium magnet contained 66 % of iron. After iron ions were removed, the solution was vacuum-dried. The CL image of the vacuum-dried residue of the iron removed solution of the neodymium magnet was captured through a filter cut light with wavelengths less than 700 nm. The vacuum-dried residue of the neodymium magnet produced red-purple color as shown in figure 2 (d). The red-purple color corresponds to illumination of NdCl3 (approximately 890 nm) because the LED light produced a red-purple color when we captured an image of LED light with wavelength of 850 or 940 nm through the filter. Thus, the portable CL spectrometer can sort a samarium-cobalt magnet and a neodymium magnet by their luminescent colors. References: [1] S Imashuku, N Fuyuno, K Hanasaki and J Kawai, Rev. Sci. Instrum. 84 (2013), p. 126105. [2] Y Kakita, Bunseki Kagaku, 16 (1967), p. 624. Figure 1. Schematic views of a portable CL spectrometer. Figure 2. CL spectra of (a) a vacuum-dried residue of the samarium-cobalt magnet alone with SmCl3, and (b) a vacuum-dried residue of the neodymium magnet alone with NdCl3, PrCl3, DyCl3, and TbCl3 by SEM-CL analysis. CL images of (c) a vacuum-dried residue of the samarium-cobalt magnet and (d) a vacuum-dried residue of the iron removed solution of the neodymium magnet using a portable CL spectrometer. (d) was captured through a filter cut light with wavelengths less than 700 nm.");sQ1[398]=new Array("../7337/0795.pdf","Toward 5D Imaging in an In-Situ Environmental TEM","","795 doi:10.1017/S1431927615004778 Paper No. 0398 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Toward 5D Imaging in an In-Situ Environmental TEM Huolin L. Xin1, Lili Han1, and Ruoqian Lin1 1. Center for Functional Nanomaterials, Brookhaven National Laboratory, Upton, NY 11973 In situ environmental TEM (ETEM) is a rapidly evolving area and has experienced impressive developments in imaging naomaterials' transformation in response to the change of environments over the past years due to the availability of dedicated environmental TEMs [1] and holder based systems [2]. For example, with help of MEMS-based local heating holders, it is now possible to monitor the subatomic scale changes of nanocatalysts in gases at elevated temperatures [3]. With the tremendous progress made on the sample environment side, the throughput of analytical STEM imaging and spatially resolved EELS spectroscopic imaging becomes a major bottleneck for various applications where real-time 2D and 3D compositional and bonding information is highly needed. Here, in this talk, we present the development of in situ 4D STEM-EELS tomography and its application in the first series of experiments that attempt to unravel the time-dependent 3-D chemical restructuring process of bimetallic nanocatalysts upon oxidation. Fig. 1 shows the in situ STEM-EELS imaging of a single Fe-Co nanocatalyst upon heating at 500 degree C in 0.2 mbar of oxygen performed in an environmental Titan. The elemental maps visualize in projection the segregation and oxidation of Fe and Co. Both the maps and the analysis of the Fe/Co L2,3 near edge fine structures suggest that Fe segregates out and gets oxidized prior to that of Co. This is in agreement with the fact that Co is more noble than Fe (the standard reduction potential of Co is lower than that of Fe). In spite of that, the reaction pathway regarding how cobalt penetrated the Fe oxide shell remain elusive due to the limited 2D projection information. To reconstruct the 3D chemical pathway, we performed in situ STEM-EELS tomography at a series of important reaction stop points using a custom-made MEMS-based high-tilt tomography heating holder (DenSolution inc., Netherlands). This constitutes a discrete version of 5D imaging. In the following, we will describe our development of quantitative STEM-EELS tomography. Compared with STEM-EDX, the advantage of STEM-EELS is its capacity to extract spatially resolved bonding information. Figure 2 demonstrates this capability using an example of a partially oxidized FeCo catalyst. By applying statistical classification to the Fe L2,3 near edge fine structures, the metallic and oxidized iron component maps can be readily extracted (Fig. 2). In addition, one aspect that EELS traditionally falls short of is that EELS has difficulty working with thick samples due to multiple scattering. We used a quantitative dual-EELS method developed in refs. [4] to correct for both multiple inelastic and elastic scattering artifacts in the tilt series. The fulfilment of the projection requirement not only improves the reconstruction resolution but also allows tomograms to be rendered quantitatively. Fig. 3 gives a reconstruction example of the 3D mosaic structure of the Fe-Co bimetallic nanocatalysts at the end point of its chemical restructuring. [5] References: [1] R. Sharma and K. Weiss, Microscopy Research and Technique 42 (4), 270 (1998). [2] HL Xin et al, Microscopy and Microanalysis 19 (06), 1558 (2013). [3] HL Xin et al, Nano Letters 14 (6), 3203 (2014). [4] HL Xin et al, Utramicroscopy, 19, 38 (2014) and HL Xin et al, Phys. Rev. B., 90, 214305 (2014) Microsc. Microanal. 21 (Suppl 3), 2015 796 [5] Research carried out at the CFN/BNL, which is supported by the U.S. Department of Energy, Office of Basic Energy Sciences, under Contract No. DE-SC0012704. Figure 1. Time-lapsed in situ STEM-EELS imaging of a single Fe-Co nanocatalyts using MEMS-based tomography heating holder. Figure 2. Extraction of Fe metallic and oxidized component maps using linear decomposition of the Fe L2,3 edge. Results acquired at -30.9 degree tilt for a partially oxidized Fe-Co catalysts (not the same catalyst in Figure 1). Figure 3. The STEM-EELS extracted elemental reconstruction of a fully oxidized Fe-Co nanocatalysts.");sQ1[399]=new Array("../7337/0797.pdf","In-situ Transmission Electron Microscopy to Probe the Electrochemical Deposition of Nanostructured Materials","","797 doi:10.1017/S143192761500478X Paper No. 0399 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In-situ Transmission Electron Microscopy to Probe the Electrochemical Deposition of Nanostructured Materials Jie Yang1, Carmen Andrei2, Gianluigi A. Botton2, 3 and Leyla Soleymani1,4 School of Biomedical Engineering, McMaster University, Hamilton, Canada Canadian Centre for Electron Microscopy, McMaster University, Hamilton, Canada 3. Department of Materials Science and Engineering, McMaster University, Hamilton, Canada 4. Department of Engineering Physics, 1280 Main Street West, McMaster University, Hamilton, Canada, L8S 4L8 2. 1. Nanomaterials and nanostructured materials have been used extensively in biosensing platforms, and their morphology strongly influences the detection limit and dynamic range of these sensing systems [1]. This makes it very important to precisely tune the biosensing substrate at the biomolecular scale or the nanoscale. To develop nanostructures with desired morphology, a detailed understanding of the growth mechanism is very essential. Considerable effort has been devoted to investigate the mechanisms involved in electrochemical deposition, but controversy regarding the growth modes still remains due to the lack of direct experimental proof of the nucleation and growth kinetics [2]. Real-time observation of the growth process through high resolution in-situ microscopy allows direct and precise study of the structural evolution; however, some difficulties in real-time imaging in liquid environments remain [3]. Here we have used an electrochemically-biased liquid cell inside a transmission electron microscope (TEM) to investigate the growth of Au nanoparticles on carbon microelectrodes. The liquid cell has built-in electrical and fluidic circuitry, which enables us to simultaneously collect electrical signals and electron micrographs during the whole process [4]. We evaluated this system by applying cyclic voltammetry to carbon electrodes modified with platinum nanoparticles in sulfuric acid solutions. In addition, we studied the effect of electron beam dose on the thickness of the liquid layer using Electron Energy Loss Spectroscopy (EELS) and TEM imaging. With this approach, we captured the electrochemical growth process of Au nanoparticles in chloroauric acid solutions under a fixed potential. The cyclic voltammagrams obtained using the liquid cell are comparable to those obtained using standard electrochemistry cells, indicating the system's capability to study in-situ electrochemical processes. Bubbles can be generated in two ways: for large area and using a high electron beam current density ( in the order of 0.63 pA/nm2) or applying a negative enough bias voltage. Bubbles grow under a beam current density of 0.026 pA/nm2 or higher and shrink under 0.0061 pA/nm2 or lower (Figure1). The electron beam also induces homogenous nucleation and growth of Au nanoparticles in solutions of chloroauric acid, while dissolves the grown particles if the beam current density is tuned high enough. Therefore suitable conditions are identified to conduct the in-situ electrochemical studies, under which high resolution imaging is possible inside bubbles having a thin electrolyte layer. This enables the initial stages of electrochemical deposition to be captured (Figure 2). Quantitative analysis of the electrochemical current versus time curves (Figure 3a) and images (Figure 3b, c, d, e) was done to study the nucleation/growth mechanism. The growth of Au exhibits a progressive nucleation and growth. In conclusion, we have applied an effective in-situ microscopy system to investigate electrochemical process under different solution, electron beam, and potential conditions. We also evaluated the electron beam effects and demonstrated how to control electrolyte thickness for high resolution imaging [5]. Microsc. Microanal. 21 (Suppl 3), 2015 798 References: [1] L Soleymani et al, Nature Nanotechnology 4 (2009), p. 844. [2] A Radisic et al, Nano Letters, 6 (2006), p. 238. [3] T J Woehl et al, Ultramicroscopy, 127 (2013), p. 53. [4] R R Unocic et al, Microscopy and Microanalysis, 20 (2014), p. 452. [5] The authors acknowledge funding from NSERC under the Discovery Program. Dr. G. Silveira and A. Duft is thanked for their helpful contributions to this work. b d c a e Figure 1. Evolution of a bubble in water under beam current density of 0.026 pA/nm2; Images a, b, c, d, e were acquired at 15 s intervals under TEM mode. a c b Figure 2. Thickness distribution of liquid-bubble interface measured by EELS: a. TEM image showing the bubble generated in water; b. Zoomed-in HAADF image of the interface indicated by the red circle in a, and EELS was conducted through a line shown by the red arrow; c. Measured relative thickness (t/ ) of the liquid along the line in b and inserted plot shows the calculated absolute liquid thickness. d e c b a Figure 3. Current transient curve and corresponding images showing the in-situ deposition of Au on carbon from 180 nm-thick chloroauric acid under a fixed potential of -0.6V for 200s. Bright field STEM images b, c, d, e are related to the time points indicated by the arrows in the current vs. time curve in a). t/ m");sQ1[400]=new Array("../7337/0799.pdf","Quantitative Structural Analysis of Fuel Cell Catalysts and Carbon Supports by TEM and Cryo-STEM Tomography","","799 doi:10.1017/S1431927615004791 Paper No. 0400 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantitative Structural Analysis of Fuel Cell Catalysts and Carbon Supports by TEM and Cryo-STEM Tomography Elliot Padgett1, Nina Andrejevic1, Zhongyi Liu2, Koji Moriyama3, Ratandeep Kukreja4, Yi Jiang5, Veit Elser5, David A. Muller1. 1. 2. School of Applied and Engineering Physics, Cornell University, Ithaca, NY. Chemical and Materials Systems Laboratory, General Motors, Warren, MI. 3. Honda R&D Co. Ltd. Automobile R&D Center, Haga-machi, Tochigi, Japan. 4. GM Global Powertrain Engineering, Pontiac, MI. 5. Department of Physics, Cornell University, Ithaca, NY. The 3D microstructure of electrode materials plays a key role in electrochemical energy conversion and storage systems. In fuel cells, reactant transport to the catalyst through the carbon support and ionomer governs the limiting reactions but remains poorly understood. Two carbon materials commonly used as fuel cell catalyst supports are Vulcan, which has catalyst on the carbon surface, and Ketjen Black (HSC), which is porous and has catalyst in both the carbon interior and surface. Catalyst activity and degradation mechanisms depend on the distribution of catalyst particles on the support, which cannot be unambiguously determined by bulk measurements. We use TEM and STEM tomography to measure the size and spatial distribution statistics of platinum catalysts on carbon supports. Comparing measurements of total surface area by tomography with electrochemical measurements of the chemically active surface area, we find that catalyst particles embedded in the carbon interior are still chemically accessible. While TEM allowed rapid data acquisition, STEM allowed more reliable segmentation, reducing the systematic error in measured surface area from 37% uncertainty in TEM to 6% in STEM. Quantitative measurements derived from tomography, especially surface areas and volumes, are sensitive to the segmentation of materials in reconstructed tomograms. Image contrast mechanisms limit the quality of segmentation that can be achieved. While bright field TEM provides high contrast for lowZ materials such as carbon, phase contrast creates speckle noise in high-resolution images of amorphous materials that broadens the distribution of intensities, (Fig 1a,b) blurring the reconstruction and compromising segmentation. Low angle annular dark field (LAADF) STEM provides an intensity scale that is monotonic with thickness, a low-noise background, and easier segmentation (Fig 1c,d). We produced tomograms of different loadings of platinum particles on Vulcan and HSC carbon supports (Fig 2). Maintaining the sample at cryogenic temperatures using a LN2-cooled tomography holder enables investigation of beam-sensitive materials such as ionomers [1] and supresses carbon contamination, which is problematic (especially for STEM) when carbon and low-Z elements are of interest. We identified platinum using threshold segmentation to determine the size and surface area distributions of hundreds of catalyst particles per sample (Fig 3). We identified the carbon surface using thresholding and morphological filtering to distinguish surface vs. interior platinum particles and measure the distance from catalyst to the carbon surface in HSC. TEM and STEM measure qualitatively similar surface area distributions (Fig 3), but the large threshold error in TEM limits its quantitative interpretation. In STEM it is clear that interior particles contribute roughly twice as much total surface area, but for particles larger than 2-3nm, most platinum surface is on the carbon exterior. We combine tomographic and electrochemical surface area measurements to better understand catalyst activity. We measure smaller surface areas from electrochemistry than tomography. Fig 3c indicates the fraction of platinum surface that is chemically inaccessible. Vulcan, with platinum only on the carbon exterior, and HSC, with two thirds of the platinum surface embedded in the carbon, show similar electrochemically active surface areas, implying that much of the embedded catalyst is chemically accessible. [2] Microsc. Microanal. 21 (Suppl 3), 2015 800 References: [1] DA Cullen et al, Journal of The Electrochemical Society 161 (2014), pp. F1111-F1117. [2] Funded by GM & Honda. EM Facility support from the NSF MRSEC program (DMR 1120296). Figure 1. (a) TEM and (c) LAADF STEM images of platinum nanoparticles on HSC carbon taken with a 200kV FEI Tecnai F20. Corresponding histograms (b,d) from larger images show intensity distribution for carbon support film (blue Gaussian) and HSC carbon (red Gaussian). Figure 2. Isosurface renderings of (a) TEM (FEI T12, 120kV) and (b) cryo-STEM (FEI F20, 200 kV) tomograms of platinum particles (orange) on HSC carbon support (grey). Figure 3. Cumulative surface area of all platinum particles above a minimum diameter in (a) TEM and (b) STEM reconstructions, with particles on HSC surface and interior separated. Error bars represent statistical uncertainty; shaded areas represent uncertainty due to platinum segmentation. (c) Specific surface areas from TEM tomography and hydrogen adsorption/desorption for Pt on HSC and Vulcan.");sQ1[401]=new Array("../7337/0801.pdf","A Quasi In Situ HRTEM Study of the Air Stability of (Ni/Co)MoS2 Hydrodesulfurization","","801 doi:10.1017/S1431927615004808 Paper No. 0401 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A Quasi In Situ HRTEM Study of the Air Stability of (Ni/Co)MoS2 Hydrodesulfurization Catalysts. G.M. Bremmer1, L. van Haandel2, E.J.M. Hensen2, J.W.M. Frenken1, P.J. Kooyman3 1. 2. Huygens-Kamerlingh Onnes Laboratory, Leiden University, Leiden, The Netherlands. Schuit Institute of Catalysis, Eindhoven University of Technology, Eindhoven, The Netherlands. 3. ChemE, Delft University of Technology, Delft, The Netherlands. (Ni/Co)MoS2 catalysts are widely used for the hydrodesulfurization (HDS) of fossil fuels, in the form of nanometer-sized slabs on a high-surface-area support (e.g. -Al2O3). As environmental legislation pushes oil refineries towards producing transportation fuels with progressively lower sulfur levels, research on HDS catalysts is an ongoing effort both in industry and in academia [1]. Traditionally, the preparation and characterization of catalyst samples is performed ex situ, subjecting the material to ambient air. Earlier work by Kooyman and Van Veen revealed the destructive effect of air on these HDS catalysts. Their HRTEM study showed a decrease in size of the MoS2-slabs after careful exposure to ambient air [2]. Promoting the MoS2-slabs with Ni/Co atoms, which are incorporated in the slab structure, has the effect of enhancing catalytic activity and selectivity [3,4]. The stability towards ambient air of Ni/Co-promoted MoS2-catalysts has not yet been studied. As the Ni/Co promoter atoms have such a significant effect on the catalytic behavior, these atoms might also influence the stability of the slabs as a whole. To investigate this issue, we analyzed the slab lengths of a series of -Al2O3-supported (Ni/Co)MoS2 catalysts using HRTEM without any exposure to air, using a protective atmosphere sample transfer holder [5]. After 24 h of contact with ambient air and after 1 month in air, the samples were measured again (Fig. 1). Using the same approach of controlled air exposure, XPS measurements were conducted on an identical set of samples, to obtain information on changes in the elemental composition. The XPS measurements reveal that during the air exposure, the MoS2-slabs get oxidized. As shown in Fig. 2, the atomic structure of MoS2-slabs consists of one layer of Mo atoms sandwiched between two layers of S atoms. Using TEM measurements, the length of these slabs can be determined when the slabs are in plane with the electron beam, as shown in Fig. 2b. In HDS catalysis, the Mo atoms located on the very edge of the slabs are the catalytically active sites, which can get oxidized in air to form MoO3. Locally, the planar structure is then lost, causing the slabs to appear shorter on TEM micrographs. Quantitative analysis of the resulting images shows that Ni/Co-promoted MoS2-catalysts do decay due to oxidation (Fig. 3), indicating the necessity of shielding these samples from ambient air. References: [1] S Brunet et al., Applied Catalysis A: General 278 (2005), p. 143. [2] PJ Kooyman and JAR van Veen, Catalysis Today 130 (2008), p. 135. [3] EJM Hensen et al., Catalysis Letters 84 (2002), p. 59. [4] AK Tuxen et al., Journal of Catalysis 295 (2012), p. 146. [5] HW Zandbergen et al., Electron Microscopy 1998, in: Proc. ICEM 14, Cancun, Mexico, 31 Aug.�4 Sept. 1998, Symposium W, Volume II, (1998), p. 491. Microsc. Microanal. 21 (Suppl 3), 2015 802 [6] This work was supported by the Netherlands Organisation for Scientific Research (NWO/OCW) as part of the Frontiers of NanoScience (NanoFront) program. 1 1 1 2 2 2 Figure 1. HRTEM images of CoMoS2 catalyst slabs on -Al2O3 a) without any contact to air, b) the same sample region after 24 h of contact to air, c) the same sample region after 1 month in air. The sample region labeled `1' shows a stack of slabs that decreases in size after 24 h, while the sample region labeled `2' shows a slab that has completely disappeared after 24 h. a) = Mo =S c) b) Figure 2. a) Atomic structure of a MoS2-slab, showing the single Mo layer, sandwiched between two layers of S atoms. b) Side view of the slab, as it is visible in TEM measurements. c) A stack of slabs. Figure 3. Average slab length as a function of exposure time to ambient air, showing a decrease in average length for all samples. Sulfidation pressures during sample preparation are indicated in the legend.");sQ1[402]=new Array("../7337/0803.pdf","Tracking Ionic Transport and Electrochemical Dynamics in Battery Electrodes Using in situ TEM-EELS","","803 doi:10.1017/S143192761500481X Paper No. 0402 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Tracking Ionic Transport and Electrochemical Dynamics in Battery Electrodes Using in situ TEM-EELS Feng Wang,1,* Wei Zhang,1 Peng Gao,1 Khim Karki1 1 Sustainable Energy Technologies Department, Brookhaven National Laboratory, Upton, NY 11973 (*fwang@bnl.gov) The design and development of safe, high energy electrodes for next-generation batteries, requires better understanding how electrode function by real time tracking of electrochemical dynamics in active constituent, i.e. individual particles, and the collective electrochemical behavior of particles assembled in an electrode. Due to the inherent heterogeneity of the electrodes, this type of characterization requires developing new tools and techniques for measurements at a range of relevant length scales. Most of the available in-situ techniques, such as those based on hard X-ray scattering, are suited for bulk measurement at electrode level, but very often have no adequate spatial resolution to probe local structural changes in single particles or interfaces. High-resolution transmission electron microscopy (TEM) imaging and energy-loss spectroscopy (EELS), capable of exceptional spatial resolution [1], has been unsuitable for in-situ studies until very recently, when in-situ electrochemical cells were developed for operation inside the transmission electron microscope. In the last several years, in-situ TEM has been widely applied for studying electrochemical reactions in different electrode systems and at scales spanning from a single particle to composite electrodes [2-4]. Figure 1a shows one type of electrochemical cell specialized for studying lithium reactions in nanoparticles being loaded onto carbon-coated TEM grids [4]. The carbon film, being used as the support of nanoparticles, also provides the pathway for electronic and ionic transport. The counter electrode, lithium metal coated with natively formed surface lithium oxynitride (acting as electrolyte), is attached to a piezo-driven biasing probe that is built into the TEM�STM sample stage. The cell design is simple while widely applicable for studying electrochemical reactions in various anode and cathode materials. The special design of the electrodes, only consisting of thin carbon film and single layer of nanoparticles, allows the use of the low energy-lying Li K-edge EELS spectrum as a lithium probe, to identify Li species through fine-structure fingerprinting technique (Fig. 1b)[1], and to track the evolution of chemical states of Li during electrochemical reactions [4]. The developed electrochemical cell was recently utilized for studying lithium transport and electrochemical dynamics in single FeF2 particles and thin films (Figure 2). The results from the real time observation provided the 1st experimental evidences showing that lithium moves fast across the surface of the FeF2 nanoparticles, but penetrates into the bulk of FeF2 particles at a much slower rate, taking a few minutes for a 10-nm particle, because of the insulating nature of FeF2. Diffusion of Li into the bulk is prohibited, and so the reaction front propagates, layer-by-layer, into the bulk, during which iron percolating network is formed to provide electronic transport pathway [4, 5]. With these in-situ TEM studies, we were able to identify the rate-limiting factors of electrochemical reactions in conversion-type electrodes, and use knowledge to design new mixed -cation cathodes, in which a 2nd cation was introduced to improve the local electronic and ionic transport properties and thereby reduce the cycling hysteresis [6]. Very recently we developed a liquid-type electrochemical cell from a flow liquid platform (designed by Hummingbird Scientific) [7], and utilized it to investigate the electrodeelectrolyte interactions in the liquid electrolyte through correlative in-situ TEM-EELS and synchrotron X-ray measurements. The versatility of in-situ TEM technique for studying ionic transport and electrochemical dynamics in battery electrodes will be discussed along with our recent results on several conversion and intercalation cathode materials [8]. Microsc. Microanal. 21 (Suppl 3), 2015 804 [1] F. Wang, et al., "Chemical Distribution and Bonding State of Lithium in Intercalated Graphite with Optimized Electron Energy-Loss Spectroscopy", ACS Nano, 5, 1190 (2011). [2] C.M. Wang, et al. "In situ transmission electron microscopy and spectroscopy studies of interfaces in Li ion batteries: challenges and opportunities." J. Mater. Res. 25, 1541(2010). [3] J.Y. Huang, et al. "In situ observation of the electrochemical lithiation of a single SnO2 nanowire electrode. Science 330, 1515 (2010). [4] F. Wang, et al., "Tracking of Li Transport and electrochemical reaction in nanoparticles", Nat. Commun. 3, 1201 (2012). [5] F. Wang et al., "Ternary Metal Fluorides as High-Energy Cathodes with Low Cycling Hysteresis", Nat. Commun. (accepted). [6] F. Wang et al., "Conversion reaction mechanisms in lithium ion batteries: study of the binary metal fluoride electrodes", J. Am. Chem. Soc. 133, 18828 (2012). [7] K. Karki, P. Gao, W. Zhang, Daan Hein Alsem, Norman Salmon, and Feng Wang, "Liquid Electrochemical Cell for in-situ TEM Studies of Batteries" Microsc. Microanal. proceeding (2014). [8] This work is financially supported by the Laboratory Directed Research and Development (LDRD) program at Brookhaven National Laboratory. Figure 1. (a) Schematic illustration of in-situ electrochemical cells for operation in the transmission electron microscope, (b) experimentally measured Li K-edge EELS spectra for a series of lithium compounds, which may be used as fingerprints for differentiating different lithium species. Figure 2. Tracking lithium transport and reaction in FeF2 by in-situ STEM: (a) selected annual dark field scanning TEM (ADF-STEM) images of one single nanoparticle in comparison to the phase-field simulation (on the right), (b) ADF-STEM image of FeF2 film showing the movement of the reaction front across the area, as indicated by the white dashed lines, (c) moving distance of the reaction front versus time, indicating ~100x higher speed of lithium diffusion along the surface than that in the bulk.");sQ1[403]=new Array("../7337/0805.pdf","Effects of Quantized, Transient Chromatic Aberrations in Ultrafast Electron Microscopy","","805 doi:10.1017/S1431927615004821 Paper No. 0403 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Effects of Quantized, Transient Chromatic Aberrations in Ultrafast Electron Microscopy Dayne A. Plemmons, Alyssa J. McKenna, and David J. Flannigan Department of Chemical Engineering and Materials Science, University of Minnesota, Minneapolis, MN, 55455, USA High-resolution in situ electron microscopy is currently an extremely active area of research owing, in part, to the development and advancement of new instrumentation and methods for studying specimens at the atomic scale in liquids, at elevated temperatures and pressures, and under electrical biasing and during mechanical deformation. Notably, the increased experimental flexibility associated with employment of spherical aberration-correction systems � that is, relaxed requirements on accelerating voltages and pole-piece gaps � has facilitated studies of atomic-scale dynamic processes occurring under operando conditions with millisecond temporal resolution [1,2]. However, wide ranges of atomic-scale structural dynamics occur on timescales much shorter than one millisecond, the motions of which therefore cannot be resolved with current digital detector technologies. Extension of the capabilities of electron microscopy to the ultrafast temporal regime is made possible by application of the pump-probe methodology in which a pulsed-laser is used to generate discrete photoelectron packets from the emission source to probe the specimen at a precisely controlled delay after optical excitation (pump). This approach, known as ultrafast electron microscopy (UEM), has been used to investigate dynamic phenomena via sub-picosecond temporally-resolved imaging, diffraction, and spectroscopy [3]. As in conventional TEM, the spatial resolution of UEM is dependent upon the coherence and current of the beam emanating from the electron source. One strategy for optimizing the coherence of photo-generated beams relies on populating each packet with, on average, a single electron. This method circumvents Coulombic space-charge broadening at the electron source (i.e., the Boersch effect), and the resulting beam coherency can be comparable to that observed during conventional thermionic emission [4]. This has been demonstrated via pulsed imaging (i.e., employing discrete photoelectron packets without pumping the specimen) of lattice fringes in various specimens and illustrates the potential for resolving real-space, angstrom-scale ultrafast dynamics with UEM [5,6]. While the single-photoelectron approach circumvents deleterious space-charge effects, the narrow energy spread can be compromised when the photon pulses and the electron packets are spatiotemporally overlapped at the specimen; light absorption and emission can occur in the specimen near-field, thus generating large populations of photoelectrons (relative to the elastically-scattered electrons) having discrete energies differing by integer multiples of photon quanta [7]. This results in a radially symmetric projection of the discrete electron energy distribution in the image plane due to the velocity dependence of the Lorentz force and, consequently, produces annular chromatic aberrations in the real-space intensity distributions. Here, we describe how the induced electron-energy envelope resulting from absorption and emission of photons in the optically-pumped specimen near field becomes the dominant effect limiting spatial resolution of UEM images during initial excitation. Moreover, select spatial frequencies become exaggerated due to the quantized nature of the near-field interaction. Application of the effect to representative lattice-fringe images of model nanostructures indeed limits resolving power and suggests that this transient, quantized chromatic aberration will need to be addressed in order to achieve angstrom-scale, real-space ultrafast imaging with UEM in the initial Microsc. Microanal. 21 (Suppl 3), 2015 806 moments of specimen excitation. References: [1] H. Zheng, Y. S. Meng and Y. Zhu, MRS Bull. 40 (2015), 12. [2] H.-G. Liao, D. Zherebetskyy, H. Xin, C. Czarnik, P. Ercius, H. Elmlund, M. Pan, L.-W. Wang and H. Zheng, Science 345 (2014), 916. [3] D. J. Flannigan and A. H. Zewail, Acc. Chem. Res. 45 (2012), 1828. [4] M. Aidelsburger, F. O. Kirchner, F. Krausz and P. Baum, Proc. Natl. Acad. Sci., U.S.A. 107 (2010), 19714. [5] H. S. Park, J. S. Baskin, O.-H. Kwon and A. H. Zewail, Nano Lett. 7 (2007), 2545. [6] B. Barwick, H. S. Park, O.-H. Kwon, J. S. Baskin and A. H. Zewail, Science 322 (2008), 1227. [7] B. Barwick, D. J. Flannigan and A. H. Zewail, Nature 462 (2009), 902. [8] This work was supported primarily by the National Science Foundation through the University of Minnesota MRSEC under Award Number DMR-1420013 and is based upon work supported by the National Science Foundation Graduate Research Fellowship Program under Grant No. DGE-1348264. Additional support was provided by a 3M Nontenured Faculty Award under Award Number 13673369, and acknowledgment is made to the Donors of the American Chemical Society Petroleum Research Fund for partial support of this research under Award Number 53116-DNI7. Figure 1. (a) Conical ray diagram depicting focusing of electrons of quantized energies (�) arising from absorption and emission of photons in the specimen near field. Rays corresponding to elasticallyscattered electrons (ZLP) are shown for reference. (b) Calculated low-loss EEL spectrum occurring at t = 0 (i.e., precise spatiotemporal overlap of the laser pulse and electron packet at the specimen). Each sideband occurs at integer multiples of the incident photon energy. (c) Resulting annular chromatic point-spread function (PSF) at t = 0. (d) Overall PSF shown in (c) and the effects of spherical aberration. The observed shoulder arises from the annular chromatic PSF. (e) Result of applying the annular chromatic PSF to a TEM bright-field image wherein lattice fringes are observed. The image (with FFT) is representative of an expected UEM image at t 0, while the t = 0 image is blurred due to the quantized, chromatic aberration (scale bars: 5 nm, 2.5 �-1).");sQ1[404]=new Array("../7337/0807.pdf","Effects of Electron-Gun and Laser Parameters on Collection Efficiency and Packet Duration in Ultrafast Electron Microscopy","","807 doi:10.1017/S1431927615004833 Paper No. 0404 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Effects of Electron-Gun and Laser Parameters on Collection Efficiency and Packet Duration in Ultrafast Electron Microscopy Erik Kieft,1 Karl B. Schliep,2 Pranav K. Suri,2 and David J. Flannigan2 1 2 FEI Achtseweg Noord 5, 5651 GG Eindhoven, The Netherlands. Department of Chemical Engineering and Materials Science, University of Minnesota, 421 Washington Avenue SE, Minneapolis, Minnesota 55455, USA. Real-space imaging of full-morphological, angstrom-scale structural dynamics occurring on the femtosecond (fs) timescale is possible with ultrafast electron microscopy (UEM) via the stroboscopic pump-probe methodology enabled by interfacing an otherwise standard TEM with a short-pulsed laser system [1-3]. The technology has matured to the point that commercial systems, such as the FEI Tecnai Femto, are now available [4]. The base platform for the Tecnai Femto is FEI's Tecnai G2 20 200 kV instrument with a thermionic electron gun. In UEM mode, the LaB6 is operated cold such that electron emission occurs only during photo-illumination. Despite this, nanosecond single-shot imaging and diffraction � wherein >107 electrons per packet reach the specimen � are possible with a Tecnai-based UEM equipped with a conventional thermionic electron gun [5,6]. While this illustrates that sufficient photoelectrons for such experiments are generated in an otherwise unmodified Tecnai TEM, the parameter space for application-specific optimal electron-gun and photoelectron-generating laser properties remains ill-defined. Quantitatively and systematically mapping such parameter space is challenging due to the large number of variables affecting photoelectron packet properties [7-9]. Here, we describe the effects of thermionic gun configuration (e.g., Wehnelt aperture diameter and relative position of the LaB6 emission source) and photoelectron-generating laser-pulse duration on electron collection efficiency (CE) and temporal duration (t50), respectively. Specifically, we use General Particle Tracer (GPT) code [10] to simulate the evolution of fs photoelectron packets generated from a cold LaB6 source in an otherwise standard Wehnelt assembly, as is the case for UEM mode in the Tecnai Femto instrument. Considered in the simulations were the effects of electron path lengths, energy spread, and Coulombic interactions, and we tracked the temporal and energy spreads, brightness, and CE as a function of various parameters. Properties of the photoelectron-generating laser pulse (e.g., spot size on the LaB6, duration, and energy) and geometry of the gun (e.g., relative axial LaB6 position and Wehnelt aperture diameter) were systematically varied in order to investigate the resulting electron packet properties. Three discrete UEM modes were considered: single electron (1 electron/packet), burst (103 electrons/packet), and single shot (107 electrons/packet). Results of the simulations suggest that there exists an optimal LaB6 position relative to the Wehnelt aperture wherein a CE of 100% can be achieved for the single-electron mode and, additionally, that CE increases approximately linearly up to and beyond 90% for both burst and single-shot modes with increasing Wehnelt aperture diameter. Moreover, we found that only marginal increases in the overall temporal resolution can be achieved by decreasing the photoelectron-generating laser-pulse duration from 280 to 10 fs owing to evolution of the packet properties in the gun region and the acceleration tube. The goal of these simulations is to provide a starting point for experimentally mapping parameter space accessible with UEM instruments based on otherwise standard self-biasing thermionic electron guns while also Microsc. Microanal. 21 (Suppl 3), 2015 808 providing insight into challenges associated with achieving high temporal and spatial resolutions [11]. References: [1] A. H. Zewail, Science 328 (2010), 187. [2] D. J. Flannigan and A. H. Zewail, Acc. Chem. Res. 45 (2012), 1828. [3] L. Piazza, D. J. Masiel, T. LaGrange, B. W. Reed, B. Barwick and F. Carbone, Chem. Phys. 423 (2013), 79. [4] D. J. Flannigan and O. Lourie, Microsc. Anal. 27 (2013), S5. [5] S. T. Park, D. J. Flannigan and A. H. Zewail, J. Am. Chem. Soc. 133 (2011), 1730. [6] U. J. Lorenz and A. H. Zewail, Science 344 (2014), 1496. [7] A. Gahlmann, S. T. Park and A. H. Zewail, Phys. Chem. Chem. Phys. 10 (2008), 2894. [8] M. Aidelsburger, F. O. Kirchner, F. Krausz and P. Baum, Proc. Natl. Acad. Sci., U.S.A. 107 (2010), 19714. [9] B. W. Reed, J. Appl. Phys. 100 (2006), 034916. [10] S. B. van der Geer, O. J. Luiten, M. J. de Loos, G. Poeplau and U. van Rienen, Inst. Phys. Conf. Ser. 175 (2005), 101. [11] This work was partially supported by the DOE Advanced Research Projects Agency-Energy (ARPA-E) under Contract Number 0472-1595 and in part by a 3M Nontenured Faculty Award under Award Number 13673369. Acknowledgment is made to the Donors of the American Chemical Society Petroleum Research Fund for partial support of this research under Award Number 53116-DNI7. a) b) Wehnelt Small tip, 16 �m flat cylinder Anode Accelerator ... ... Ztip DWeh Large tip, 150 �m flat C1 aperture Collection Efficiency (CE) c) 100 Single-electron Burst d) 100 e) 500 CE (%) CE (%) t50(fs) 75 50 25 0 100 200 75 50 25 0 Single-electron Burst Single-shot 0 1 10 fs 80 fs 280 fs 400 300 200 200 300 400 Ztip (m) 300 400 DWeh(mm) 2 3 4 Ztip (m) Figure 1. (a) Schematic of the thermionic gun showing the Wehnelt cylinder and LaB6 with aperture diameter (DWeh) and gun-tip height (Ztip) defined. (b) Schematic of the thermionic gun geometry in the Tecnai Femto with pertinent components labeled and collection efficiency (CE) illustrated. (c) Effect of Ztip position on CE for the single-electron and burst modes for DWeh = 0.7 mm. (d) Effect of DWeh at optimal Ztip position on CE for each mode studied. (e) Effect of photoelectron-generating laser-pulse duration (10, 80, and 280 fs) on temporal resolution (t50) as a function of Ztip position.");sQ1[405]=new Array("../7337/0809.pdf","Ultrafast Point Projection Electron Microscopy","","809 doi:10.1017/S1431927615004845 Paper No. 0405 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Ultrafast Point Projection Electron Microscopy Pratistha Shakya and Brett Barwick Department of Physics, Trinity College, Hartford, Connecticut, USA By combining the techniques of point projection electron microscopy with an ultrafast field emission tip source a simple compact ultrafast electron microscope can be constructed. The microscope uses a tungsten field emission tip as an electron source and by focusing a femtosecond laser pulse onto the tip1,2 electron pulses with 100 fs are created. A schematic of the microscope setup is shown in Fig. 1. The microscopes operation can be switched to a standard CW mode by increasing the DC field applied to the tip. The magnification, M, of the microscope3 is directly related to the ratio of the tip-to-specimen distance d and the distance from the tip to the detector D by, M= . (1) In our apparatus the distance D is fixed at 0.1 m, however, d, can be varied from 10-2 m to 10-5 m giving magnifications of 10� to 10,000�. To test the capabilities of the microscope, a Holey Carbon Quantifoil TEM grid was imaged. It has holes of different shapes and sizes with the smallest having a diameter of 1 micron. The images seen on the electron detector screen are shown in Fig. 2. For Fig. 2a, the tip voltage is -84 V, and there is a large signal entirely due to CW electron emission. When the tip voltage is reduced to -63 V, there is almost zero CW emission and then by irradiating the tip with femtosecond laser pulses an image can be made using femtosecond electron pulses which is shown in Fig. 2b. We investigated the temporal duration of the electron emission from the tip using an autocorrelation experiment2 and found that the electron pulses are ~100 fs in duration. We also show that at low tip voltages the relatively low kinetic energy of the electrons results in electrostatic lensing near the specimen, which can cause the electrons to be deflected, as shown in Fig. 2d. While the deflections in Fig. 2d are due to the charging of the TEM grid, the images prove the sensitivity of the microscope to weak electric fields. While normally electrostatic distortions are considered detrimental to image formation in a conventional high energy TEM, we plan to use this sensitivity to follow the dynamics of ultrafast laser induced fields. Combining the techniques of a point projection microscope with an ultrafast field emission source gives the microscope some advantages over traditional electron microscopes. Because of the very small distances between the electron source and sample the electron pulse has no time to broaden or disperse. This may allow temporal resolutions on the 10 femtoseconds to be attained, while also having spatial resolution better than ~10 nm at low tip voltages3. This high temporal and spatial resolution will allow us to follow dynamics in a number of nanoscale systems. Microsc. Microanal. 21 (Suppl 3), 2015 810 References: [1] B. Barwick, J. Handali, and E. Quinonez, "Femtosecond photoelectron point projection microscope", Review of Scientific Instruments 84, 103710 (2013). [2] B. Barwick, C. Corder, J. Strohaber, N. Chandler-Smith, C. Uiterwaal, and H. Batelaan, "Laser-induced ultrafast electron emission from a field emission tip", New Journal of Physics 9, 142 (2007). [3] V. T. Binh, V. Semet, and N. Garcia, "Low-energy-electron diffraction by nano-objects in projection microscopy without magnetic shielding", Applied Physics Letters 65(19), 24932495 (1994). [4] The authors acknowledge funding from Connecticut Space Grant and Trinity College FRC Grant. Fig. 1. Setup for the femtosecond point projection microscope. The probe excitation beam (red wave) is used to excite the field emission tip, producing pulsed electron packets. The pump beam (blue wave) is used to excite the sample. The pump and the probe laser beams enter the microscope from opposite sides. The pulsed electron packets are accelerated to the sample and the specimen image is then projected on the screen. -84 V -63 V 50 eV (a) (b) (c) (d) Fig. 2. CCD camera images of different TEM grids when operating the microscope with DC emission (a) and with femtosecond pulsed laser (b, d). Images (a) and (b) were taken in 2000� magnification. The CW mode required a tip voltage of -84 V and the pulsed laser mode required -63 V on the tips. Figure 2(c) is an optical micrograph of a 120�m�120�m TEM grid at 63� magnification. Figure 2(d) was taken at the same magnification to show distortion on a TEM grid due to electrostatic lensing. This effect can be reduced by coating the specimen with materials of higher conductivity.");sQ1[406]=new Array("../7337/0811.pdf","Quantitative Determination of Thermal Fields and Transformation Rates in Rapidly Solidifying Aluminum by Numerical Modeling and In-situ TEM","","811 doi:10.1017/S1431927615004857 Paper No. 0406 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantitative Determination of Thermal Fields and Transformation Rates in Rapidly Solidifying Aluminum by Numerical Modeling and In-situ TEM Can Liu1, Kai Zweiacker1, Joseph T. McKeown2, Thomas LaGrange3, Bryan W. Reed3, Geoffrey H. Campbell2 and J�rg M.K. Wiezorek1 1. 2. 3. Department of Mechanical Engineering and Materials Science, University of Pittsburgh, Pittsburgh, PA 15261 Materials Science Division, Lawrence Livermore National Laboratory, Livermore, CA 94551 Integrated Dynamic Electron Solutions, Inc, Pleasanton, CA 94566 The dynamic transmission microscope (DTEM) offers incomparable nano-scale spatiotemporal resolution that is advantageous for characterizing irreversible transient processes in various material systems [1]. Previously, we utilized bright field imaging and diffraction in DTEM with nano-second temporal resolution to characterize the dynamics of rapid solidification (RS) in pure Al and Al-Cu alloy thin films in single-shot, single-image acquisition mode [2-4]. Recent developments in DTEM instrument enable a single-shot, multiple-image acquisition mode (Movie-mode) [5], which significantly reduces potential uncertainty and experimental error when measuring parameters such as interface velocity during rapid solidification process. However, it is inherently challenging to measure the temperature evolution during the RS transformation cycle while performing in-situ DTEM observations. Knowledge of the thermal field evolution in correlation to the solidification front velocities would greatly benefit quantitative understanding of RS process in material systems. Here we report results of enthalpy-based numerical modeling performed in a COMSOL� multiphysics environment that have been validated by experimental observations to quantitatively determine the thermal field evolution during RS. Figure 1 represents a schematic illustration of a possible timing sequence for Movie-mode DTEM experimentation and an example of an image series obtained for pulsed-laser induced melting and subsequent re-solidification in 80-nm-thick pure Al thin film. The image series includes nine images spanning from 0 �s to 20.4 �s after the melting laser pulse, covering onset of melting to completion of solidification using a 50-ns electron pulse duration and a 2.5-�s interframe time spacing. Several image sequences were used to calculate the evolution of the melt-pool size (radius, R) and interface velocity during RS. Figure 2 shows the thermal field distribution predicted by simulations during RS and the comparison between the experimentally determined time evolution of the melt pool area and solidification front velocity from the time-resolved image sequences and those computed by simulations. The details of the thermal field and associated velocity evolution behavior will be discussed. References [1] King, W.E., et al., Journal of Applied Physics (2005) 97: 111101. [2] Kulovits, A.K., et al., Philosophical Magazine Letters (2011) 91: 287. [3] McKeown, J.T., et al., Microscopy and Microanalysis (2012) Vol. 18 (Suppl. 2): 602. [4] McKeown, J.T., et al., Acta Materialia (2014) 65: 56. [5] LaGrange T., et al., Micron (2012) 43: 1108. [6] Work was performed under the auspices of the U.S. Department of Energy, Office of Basic Energy Sciences, Division of Materials Sciences and Engineering for FWP SCW0974 by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344.The research activities at the Microsc. Microanal. 21 (Suppl 3), 2015 812 University of Pittsburgh received support from the National Science Foundation, Division of Materials Research, Metals & Metallic Nanostructures program through Grant No. DMR 1105757. i) ii) Figure 1. i) Schematic illustration of Movie-mode DTEM with selected pre-delay time and interframe time spacing and ii) an example of image sequence showing laser-induced melting and RS in Al thin films recorded by in-situ Movie-mode DTEM using a 50-ns electron pulse duration and a 2.5-�s interframe time spacing i) ii) iii) Figure 2. i) Temperature distribution in the pure Al film simulated by COMSOL� multi-physics model, ii) comparison between the experimentally determined converted radius evolution of the melt pool and simulated radius evolution and iii) simulated solidification velocity compared with velocity calculated from experimental observations");sQ1[407]=new Array("../7337/0813.pdf","Dynamical Effects on Atomic-Resolution Imaging and Diffraction of the Tetragonal and Orthorhombic Phases of LaFeAsO","","813 doi:10.1017/S1431927615004869 Paper No. 0407 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Dynamical Effects on Atomic-Resolution Imaging and Diffraction of the Tetragonal and Orthorhombic Phases of LaFeAsO Pranav K. Suri,1 Jiaqiang Yan,2 David Mandrus,2 and David J. Flannigan1 1 Department of Chemical Engineering and Materials Science, University of Minnesota, Minneapolis, MN 55455 2 Department of Materials Science and Engineering, University of Tennessee, Knoxville, TN 37996 The discovery of high-Tc superconductivity in the lightly-doped iron pnictides has led to an explosion in theoretical, synthetic, and characterization work [1,2]. Doping in these materials suppresses the temperatures at which the crystallographic phase transition and antiferromagnetic (AF) ordering occur relative to the undoped parent compounds. At sufficient doping levels, emergence of the superconducting dome is observed with the suppressed AF and so-called nematic states bordering this transition [3]. Of particular interest are the 1111-type compounds, which have been found to exhibit Tc > 50 K. Consequently, efforts have focused on elucidating the origin of such emergent properties in these and other strongly-electron correlated materials. Results suggest that there is a complex interplay between structural transformations and spin/charge/orbital ordering, the precise nature of which remains unknown with respect to the superconducting state [4]. With the advent of Cs-corrected microscopy along with advances in in situ TEM capabilities and spectroscopic techniques, it is now possible to directly observe structural phase transitions and probe materials properties on the atomic scale [5,6]. Here, we describe our studies of the 1111type compound LaFeAsO using a battery of high-resolution and in situ TEM techniques in real (imaging), reciprocal (diffraction), and energy (spectroscopy) space. This material, which when electron-doped exhibits high-Tc superconductivity, undergoes a crystallographic phase transition at 160 K from tetragonal to orthorhombic and also shows AF stripe ordering below 140 K [1,2]. To study the room-temperature tetragonal phase, we used an aberration-corrected FEI Titan G2 60-300 (S)TEM and an FEI Tecnai G2 F30 TEM to obtain atomic-resolution images, selectedarea electron diffraction (SAED) patterns, and X-ray energy-dispersive spectroscopic (XEDS) profiles. For the orthorhombic and stripe-ordered AF phases, we used a Gatan 626 cryo-holder. In support of the experiments, we performed image simulations using a multislice method [7]. The combined imaging and diffraction experiments show that, regardless of the structural phase, dynamical effects on different regions of the same specimen suppress Z-contrast and produce qualitatively-similar scattering. For example, Figure 1 (left panel) shows a high-angle annular dark-field STEM (HAADF-STEM) image of (La,As) and (Fe,O) atomic columns as viewed along the [001] zone axis. It can be seen that the Z-contrast predicted to be present via image simulation is absent. Further, the right panel displays an SAED pattern along the [001] direction at 296 K showing a (100) reflection, whereas the same reflection is absent in the FFT of an atomic-resolution bright-field conventional TEM (CTEM) image also obtained at 296 K and, thus, the tetragonal phase. The implications for these and other results on directly observing the crystallographic phase transition in LaFeAsO via in situ TEM will be discussed [8]. Microsc. Microanal. 21 (Suppl 3), 2015 814 References: [1] Y. Kamihara, T. Watanabe, M. Hirano and H. Hosono, J. Am. Chem. Soc. 130 (2008), 3296. [2] J. Paglione and R. L. Greene, Nat. Phys. 6 (2010), 645. [3] R. M. Fernandes and J. Schmalian, Supercond. Sci. Technol. 25 (2012), 084005. [4] R. M. Fernandes, A. V. Chubukov and J. Schmalian, Nat. Phys. 10 (2014), 97. [5] P. E. Batson, N. Dellby and O. L. Krivanek, Nature 418 (2002), 617. [6] D. A. Muller et al., Science 319 (2008), 1073. [7] E. J. Kirkland, Advanced Computing in Electron Microscopy, (Springer, New York, 2010). [8] This work is supported by the DOE Advanced Research Projects Agency-Energy (ARPA-E) under Contract Number 0472-1595, and by a 3M Nontenured Faculty Award under Award Number 13673369. Acknowledgment is made to the Donors of the American Chemical Society Petroleum Research Fund for partial support of this research under Award Number 53116-DNI7. Part of this work was carried out in the College of Science and Engineering Characterization Facility, University of Minnesota, which has received capital equipment funding from the NSF through the UMN MRSEC program under Award Numbers DMR-0819885 and DMR-1420013. Part of this work was carried out in the College of Science and Engineering Minnesota Nano Center, University of Minnesota, which receives partial support from NSF through the NNIN program. Figure 1. (Left Panel) Room-temperature HAADF-STEM image of LaFeAsO along the [001] zone axis. The LaFeAsO tetragonal crystal structure (top) and a simulated HAADF-STEM image (bottom) with Z-contrast are overlaid on the image with the atomic columns labeled. (Right Panel) SAED pattern showing the [001] orientation and presence of the (100) Bragg spot. Figure 2. Atomic-resolution bright-field CTEM images and corresponding FFTs at 296 K (tetragonal) and 93 K (orthorhombic) from an LaFeAsO crystal viewed along the [001] zone axis. The d-spacings that give rise to spots in the observed FFTs are labeled. Note the absence of the (100) spot in the FFTs, regardless of temperature.");sQ1[408]=new Array("../7337/0815.pdf","Phase Determination of Black TiO2 Nanoparticles","","815 doi:10.1017/S1431927615004870 Paper No. 0408 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Phase Determination of Black TiO2 Nanoparticles Meng-kun.Tian,1 Masoud. Mahjouri-Samani, 2 Gyula. Eres,2 Ritesh.Sachan,3 Matthew F. Chisholm3, Kai Wang, 2 Alexander. A. Puretzky,2 Christopher. M. Rouleau, 2 Mina. Yoon, 2David. B. Geohegan,2 and Gerd Duscher1,3 1.Dept. of Materials Science and Engineering, University of Tennessee, Knoxville, TN, USA 2.Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, TN, USA 3.Material Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN, USA The increased photocatalytic activity of black TiO2 nanoparticle [1,2] is attributed to the bandgap narrowing in the surface layer produced by crystallization of amorphous material in a reducing atmosphere. These nanoparticles were found to have unique crystalline core-amorphous shell structures of TiO2 stoichiometry. Here we report a form of black TiO2 nanoparticles with different core�shell structure characterized by highly spatially resolved characterization methods. These particles were transformed from amorphous materials by post annealing. Spectra of monochromated electron energy-loss spectroscopy (EELS) show significant deviations of reference rutile Ti-L2,3 and O-K energy-loss near edge structure (ELNES) in the near surface area (2.5nm-1nm away from vacuum) of a rutile nanoparticle, while no clear deviations are observed in the interior (7nm-4nm away from vacuum). Through a linear combination and linear least-square fits of these spectra, we find that deviation of ELNES originates from high concentration of Ti2O3. This result indicates a stoichiometric rutile core-Ti2O3 shell structure. Fig 1a shows the HRTEM image of black TiO2 (rutile with diameter>15nm) nanoparticle. EELS spectra were taken from 7nm to 1nm away from surface along the direction highlighted by the arrow. As shown in Fig 1b, spectrum taken from 7nm to 4nm looks similar with each other, showing rutile characteristic. However, a significant changes of the EELS spectra taken at 2.5 nm and 1nm is evident: the two L3-eg peaks have roughly the same maximum intensity (2.5nm away from vacuum), but ELNES from spectra taken at 1nm away from vacuum further changed into smeared out anatase signature. These smeared out feature have been attribute to the existence of Ti3+ component [3, 4]. In order to explore these structure changes quantitatively, we use the linear combinations of pure rutile and pure Ti2O3 reference EELS spectra to fit the experimental data, as shown in Fig 1c. We use the linear least-square fitting to find the best fit between experimental and these reconstructed spectra yielding the TiO2 concentrations as coefficients. The fit errors are very close to the noise level (max<13%). From the fit, Ti2O3 component in this nanoparticles are responsible for the deviations of rutile ELNES: 33% Ti2O3 at 2.5nm, 60% Ti2O3 at 1nm. Even though deviations of Ti-L2,3 ELNES are not easily observed in the range of 7nm-4nm, the least-square fits show a low Ti2O3 concentration in the particle inner (11%-18%). This is consistent with a core-shell structure. The ratio of the maximum intensity of the first two peaks (a and b for Ti2O3, and a' and b' for TiO2) uniquely identifies Ti2O3 and TiO2. Specifically, for Ti2O3 a/b<1, while for TiO2 a/b>1, as Microsc. Microanal. 21 (Suppl 3), 2015 816 shown in Fig 2a. No obvious changes were observed in O-K ELNES at each location except near surface area where the high Ti2O3/TiO2 ratio gives rise to a slight change in the ratio of the maximum intensities of the first two peaks in the O-K edge at 2.5nm and a significant change 1nm away from the vacuum, as shown in Fig 2b and f. Ti2O3 coefficients from fitting the O-K ELNES are consistent with those of Ti-L2,3 ELNES within a 2% error at each location except 1nm where signal-to-noise ratio of O-K edges signals is high. Thus, the fit of ELNES is highly reliable, showing a Ti2O3 shell and stoichiometric rutile core of a form of black TiO2 nanoparticles not reported previously. References [1]. X. B Chen et al. Science 331 (2011), p. 746-750. [2] A. Naldoni et al. J Am Chem Soc 134 (2012), p. 7600-7603. [3] A. S.Sefat, et al. J Solid State Chem 178 (2005), p. 1008-1016. [4] A.Ohtomo, A. et al. Nature 419 (2002), p. 378-380. Figure 1(a) HRTEM image of rutile particle viewed in [001] zone axis. EELS spectra (inset) taken from particle inner to surface along the direction highlighted by the arrow. (b) Linear combination of Ti-L2,3 ELNES from pure rutile and Ti2O3 at different ratios. (c)-(e) Linear least-square fit of EELS spectra taken from 7nm (represent for particle inner), 2.5nm and 1nm. The difference between the experimental and reconstructed EELS spectra are given at the bottom of each image. Figure 2 (a) O-K ELNES from pure rutile and Ti2O3. (c)-(e) Linear least-square fit of EELS spectra taken from 7nm, 2.5nm and 1nm. Difference between the experimental and reconstructed EELS spectra are given at the bottom of each image.");sQ1[409]=new Array("../7337/0817.pdf","Scanning Transmission Electron Microscopy Investigation of the Structure of Multilayered Perpendicular Magnetic Tunnel Junctions","","817 doi:10.1017/S1431927615004882 Paper No. 0409 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Scanning Transmission Electron Microscopy Investigation of the Structure of Multilayered Perpendicular Magnetic Tunnel Junctions Danielle Reifsnyder Hickey1 and K. Andre Mkhoyan1 1. Department of Chemical Engineering and Materials Science, University of Minnesota, Minneapolis, MN 55455, United States The composition and morphology of multilayered magnetic materials has important consequences for magnetic memory applications. CoFeB/MgO interfaces, specifically, exhibit perpendicular magnetic anisotropy, which can be utilized in perpendicular magnetic tunnel junctions (p-MTJs) [1]. Such tunnel junctions are used for read heads, magnetic random access memory, and logic elements [1-3]. The ability to control magnetism using an electric field alone has increased their potential [4], and significant effort has been devoted to fabricating new compositions, characterizing the composition and orientation of the nanostructures that form, and decreasing the layer thickness to enable an electric field to penetrate into the structure despite the screening of electrons [1-4]. Devices based on p-MTJs have great promise, and the realization of predicted performance levels depends on the characteristics of the interfaces, which can vary according to the preparation conditions. Of particular interest is the quality of the interfaces between the various layers. Additionally, MTJ structures often must be compatible with high-temperature processing conditions to improve their performance or be successfully integrated into functional devices [3]. The relative crystallinity and orientation of the layers, the segregation of phases within a layer during annealing, and the migration or diffusion of metals across layers can all affect the performance of devices [3,5]. In certain cases, structural changes during high-temperature treatment can completely destroy the perpendicular magnetic anisotropy [5]. Here, we present scanning transmission electron microscopy (STEM) imaging and elemental analysis characterizing p-MTJ interfaces. The samples are thin films deposited onto oxidized silicon wafers by ultrahigh-vacuum magnetron sputtering. For sample preparation, crosssectional lamellae are prepared by focused-ion-beam milling. Thinning of the lamellae is critical to achieve electron-transparent specimens. Because p-MTJs are often sandwiched between transition metal elements with higher atomic numbers (higher Z values), the electron scattering in STEM due to the metal layers can be significant. The CoFeB and MgO layers contain relatively low-Z elements and are only 0.5-2 nm thick, compared to the high-Z metal layers that are 10-20 nm thick. Analytical STEM analysis, including electron energy loss spectroscopy (EELS) and energy dispersive X-ray spectroscopy (EDS), is uniquely suited to the compositional and morphological analysis of such fine features. Figure 1 shows high-angle annular dark-field (HAADF) STEM images of a p-MTJ. In Figure 1A, the thin-film sample has been patterned and cross sectioned into a lamella for STEM analysis (from top to bottom: protective platinum and carbon layers, the p-MTJ structure, and Si regions of increasing thickness). Figure 1B presents a higher-magnification STEM image of the p-MTJ structure (top: two metal layers; middle: the CoFeB/MgO/CoFeB p-MTJ; and bottom: three metal layers) [6]. Microsc. Microanal. 21 (Suppl 3), 2015 818 Referenc ces [1] S Ikeda et al., Na Mater. 9 (2010) 721. at. ( W . 009) 242501 . [2] WG Wang et al., Appl. Phys. Lett. 95 (20 [3] T Liu et al., Sci. Rep. 4 (2014 5895. u R 4) [4] W-G Wang, et al Nat. Mater 11 (2012) 64. l., r. [5] HD Gan et al., Ap Phys. Le 99 (2011 252507. G ppl. ett. 1) [6] The authors grate a efully acknowledge the preparation of p-MTJ sa p amples by th research g he group of Prof. Weigang Wang in the Department of Physics at the Univer W D o rsity of Ariz zona, and fun nding provided by the Center for Spintronics, a STAR d C RNET prog gram admin nistered by the y Semiconductor Resea arch Corporation. S es J (A) of -sectional lam mella Figure 1. HAADF STEM image of p-MTJ samples: ( image o the crossand (B) image of the p-MTJ sand i dwiched betw ween two mu ulticompone metal lay ent yers.");sQ1[410]=new Array("../7337/0819.pdf","Dopant Quantification by Atomic-scale Energy Dispersive X-ray Analysis","","819 doi:10.1017/S1431927615004894 Paper No. 0410 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Dopant Quantification by Atomic-scale Energy Dispersive X-ray Analysis Zhen Chen1, Scott D. Findlay1, Matthew Weyland2,3, Adrian J. D'Alfonso4 and Leslie J. Allen4 1. 2. School of Physics and Astronomy, Monash University, Victoria 3800, Australia. Monash Centre for Electron Microscopy, Monash University, Victoria 3800, Australia. 3. Department of Materials Engineering, Monash University, Victoria 3800, Australia. 4. School of Physics, University of Melbourne, Victoria 3010, Australia. Atomic resolution energy dispersive X-ray (EDX) mapping in scanning transmission electron microscopy (STEM) has been realized recently [1, 2]. Strong dynamical scattering of electrons along the well-aligned atomic columns leads to a nonlinear relationship between X-ray counts and elemental concentration, necessitating detailed image simulation when quantification is sought [3, 4]. The introduction of dopants, a circumstance of great interest for practical applications in material science, offers further complications. For quantitative work in high angle annular dark field (HAADF) imaging, the simple fractional occupancy procedure, i.e. `virtual crystal approximation' as depicted in figure 1(a), is not always sufficient [5]. In particular, for atomic resolution analysis, it is possible that columns with the same concentration but different depth distributions of the substitutional dopants will produce quantitatively different signals, as depicted in figure 1(b). We will explore the limits this imposes on precise concentration determination via atomic-resolution STEM EDX. Our case study is a GaAs/Al0.5Ga0.5As heterostructured nanowire, with preliminary experimental data on morphology shown in figure 1(c) and (d). To explore the uncertainties in the quantification of the Al due to different Al depth distributions, we constructed random Al/Ga configurations after the fashion of the schematic in figure 1(b) and simulated the resultant EDX signals. Figure 2(a) shows a histogram of the simulated EDX signals for 40 such configurations for concentrations of 45%, 50% and 55% aluminum by atomic number in a column 40 atoms long, using 302 keV electrons and a probe-forming aperture angle of 15 mrad. The average Al-K signal over a unit cell from different Al/Ga configurations shows a Gaussian-like distribution, and the mean and standard deviation are given in table 1. It is seen that the 3 statistical confidence limits, i.e. uncertainty bounds within a 95% confidence interval, overlap for the Al-K X-rays intensity from samples with 5% concentration change. In other words, it is difficult to distinguish variations in intensity based on different configurations for the same concentration from genuine differences in concentration at this level. Note too that the fractional occupancy simulation is more than 3% larger for Al-K signal. The channeling effect can in principle be reduced by using a large probe forming aperture, though whether this is possible in practice will depend on the capability of the aberration-corrector available. Figure 2(b) and the right hand side of table 1 show the results for a 50 mrad probe-forming aperture semiangle. The Al-K X-rays intensity from 45%, 50%, and 55% Al concentration is then clearly separated beyond 3 and, furthermore, the intensity from fractional occupancy is identical to the average intensity of different Al/Ga depth distributions. The fractional occupancy simulation procedure is much more reliable for the larger convergence angle, a boon for composition characterization by EDX as it is much more efficient in terms of computation time [6]. Microsc. Microanal. 21 (Suppl 3), 2015 820 References: [1] AJ D'Alfonso et al, Physical Review B 81 (2010), 100101. [2] PG Kotula et al, Microscopy and Microanalysis 18 (2012), p. 691-698. [3] G Kothleitner et al, Physical Review Letters 112 (2014), 085501. [4] Z Chen et al, Submitted for publication (2014). [5] E Carlino and V Grillo, Physical Review B 71 (2005), 235303. [6] The authors wish to thank Jenny Wong-Leung and Jenny Nian-Jiang from the Australian National University for the nanowire specimens and acknowledge support from the Australian Research Council's Discovery Projects funding scheme (Projects DP110102228 and DP140102538) and DECRA funding scheme (Project DE130100739). Figure 1. Structure model and morphology of GaAs/Al0.5Ga0.5As nanowire. (a). Structure model used for simulation, [110] zone-axis. (b) Schematic of possible Al/Ga depth distributions. (c) STEM-HAADF image. (d) Enlarged image of the white rectangular region in (c) showing the heterostructure. (a) 12 10 Al 45% Al 50% Al 55% (b) 10 8 6 4 2 0 Al 45% Al 50% Al 55% Frequency 8 6 4 2 0 1.05 1.10 1.15 1.20 1.25 1.30 1.35 1.40 Frequency 5 7.0 7.2 7.4 7.6 7.8 8.0 8.2 8.4 8.6 8.8 Frational intensity (x10 ) Fractional intensity (x106) Figure 2. Histogram of Al-K X-rays intensity for different Al depth distributions. For a convergence angle of (a) 15 mrad and (b) 50 mrad. The legends show the Al concentration. Convergence angle Al content Mean Standard deviation Fractional occupancy 45% 1.07�10�5 3.79�10�7 1.11�10�5 15 mrad 50% 1.19�10�5 4.44�10�7 1.24�10�5 55% 1.32�10�5 3.37�10�7 1.36�10�5 45% 0.71�10�5 3.87�10�8 0.71�10�5 50 mrad 50% 0.79�10�5 2.68�10�8 0.79�10�5 55% 0.87�10�5 2.43�10�8 0.87�10�5 Table 1. Average intensity of Al-K X-rays across different Al/Ga depth distributions.");sQ1[411]=new Array("../7337/0821.pdf","Quantitative symmetry determination and symmetry mapping using convergent beam electron diffraction technique","","821 doi:10.1017/S1431927615004900 Paper No. 0411 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantitative symmetry determination and symmetry mapping using convergent beam electron diffraction technique Kyou-Hyun Kim1, Jian-Min Zuo2,3 Advanced Process and Materials R&BD Group, Korea Institute of Industrial Technology, Incheon, 406-840, Korea 2 Department of Materials Science and Engineering, University of Illinois at Urbana-Champaign, Urbana, Illinois 61801, USA 3 Frederick Seitz Materials Research Laboratory, University of Illinois at Urbana-Champaign, Urbana, Illinois 61801, USA The symmetry recorded in convergent beam electron diffraction (CBED) patterns is in general determined by direct visual inspection [1], which does not provide uniform measurement. Furthermore, experimental CBED patterns are often noisy and deviate from the ideal symmetry because of the sample geometry and defects. Thus, the imperfection in experimental CBED patterns can lead to uncertainty in the symmetry determination [2]. Here, we propose a symmetry quantification method for CBED patterns using the profile R-factor (Rp) [3, 4] and the normalized cross-correlation coefficient () [5]. We have also developed computer algorithms to automate these procedures. We demonstrate that the method proposed here is highly effective and provides a more precise way to determine the symmetry in CBED patterns. The symmetry quantification method can be also combined with a scanning electron diffraction technique for symmetry mapping [6]. Figure 1 shows the image processing procedures for mirror symmetry quantification. First, the symmetry related two diffraction discs (A, A') are selected about the mirror plane (yellow line) as shown in Fig. 1(a). The template A is used as the reference motif so that the symmetry element is calculated by comparing with template A'. The template A and A' are first aligned (Fig. 1(c)), and A' is flipped horizontally to obtain a mirror image (A'm, Fig. 1(h)). For the rotational operation, the template A' is simply rotated by 360/no with respect to n-fold rotation. The circular mask (Figs. 1(d) and (i)) is finally used to remove areas affected by CBED disk edge and to obtain the final templates (Figs. 1(e) and (j)). Then, we applied the profile R-factor (Rp) and the normalized cross-correlation coefficient () to quantify the similarity between A (=IA(x, y)) and A'm (=IB(x, y)) as given in Eqs. (1) and (2). (1) R p 1 I x, y I x, y I ( x, y ) B A 2 A 2 , (2) x, y x, y I A x, y I A I B x, y I B 2 I A ( x, y ) I A x, y I B ( x, y ) I B 2 The symmetry quantification algorithm was then combined with the scanning electron diffraction technique for symmetry mapping. Figure 2(a) shows a typical Bragg diffraction contrast for a stacking fault in a strained Si crystal. The mirror selected for quantification is along the yellow line as indicated in Fig. 2(b). The symmetry variation was mapped using the A/A', B/B' and C/C' disk pairs, and the symmetry distribution was mapped for 152 x 72 nm2 from Fig. 2(a). Figures 2(c) and (e) are the calculated symmetry maps for Rp and , respectively. The grid in the symmetry maps becomes bright as the symmetry of the investigated grid matches the selected symmetry (i.e., mirror). In both maps, the dark contrast indicates symmetry breaking from the selected mirror symmetry. For example, the profile of Rp and values were selected along the line indicated in Fig. 2(a) and plotted in Figs. 2(d) and (f), respectively. In the area of stacking fault, the Rp rapidly increases from 0.19 to 0.79, and the drops Microsc. Microanal. 21 (Suppl 3), 2015 822 significantly from 0.98 to 0.18. Thus, the symmetry breaking is detected across the stacking fault and near the stacking fault. The details of symmetry quantification and mapping can be found in the reference [7]. We have proposed a symmetry quantification method by using Rp and . The result for the Si single crystal shows that Rp with ~ 0.1 and with ~ 0.98 can be used to determine a symmetrical pattern. In addition, the Si single crystal has a constant symmetry over the scanning area while the strained Si sample shows large symmetry variation over a stacking fault. We believe that this study provides a powerful tool for symmetry study in real materials. References: [1] T. Hahn, "International tables for crystallography", Volume A, ed. 5 (Springer). [2] K.-H. Kim, David A. Payne and J.-M. Zuo, Phys. Rev. B 86 (2012), 184113. [3] E. Jansen et al, J. Appl. Cryst. 27 (1994), 492-496. [4] B.H. Toby, Powder Diffraction 21 (2006), 67-70. [5] J.P. Lewis, Vision Interface 95 (1995), 120-123. [6] J. Tao et al, Phys. Rev. Lett 103 (2009). [7] K.-H. Kim and J. -M. Zuo, Ultramicroscopy 124 (2013), 71. [8] This work was supported by DOE (DEFG02-01ER45923). We thank Jerome Pacaud for providing the strained silicon sample. Figure 1. Image processing procedures used for mirror symmetry quantification. Two diffraction discs related by mirror are selected as indicated by the dotted circles A and A' in the (a). Each disc is then processed to give two templates A and A'm as shown above. Figure 2. (a) A bright field image of strained silicon showing a stacking fault, and (b) the selected CBED pattern. The symmetry maps for Rp and are shown in Figs. 7(c) and (e), respectively. The (d) and (f) show the Rp and profile across the stacking fault along the line indicated in the (a).");sQ1[412]=new Array("../7337/0823.pdf","Concentration at Detection Limit of Dopant for Semiconductor Samples Using Dual SDD Analysis System.","","823 doi:10.1017/S1431927615004912 Paper No. 0412 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Concentration at Detection Limit of Dopant for Semiconductor Samples Using Dual SDD Analysis System. K. Fukunaga1, N. Endo1, M. Suzuki2 and Y. Kondo1 1 2 JEOL Ltd., 3-1-2 Musashino, Akishima, Tokyo 196-8551, Japan Thermo Fisher Scientific Japan, 3-9 Moriya-cho, Kanagawa-ku, Yokohama, 221-0022, Japan An energy dispersive spectrometer (EDS) analysis system using a Si drift detector (SDD) can enhance sensitivity of compositional analysis by using a large solid angle detector [1,2]. The use of multiple EDS detector system dramatically enlarge the effective area of the detector, resulting in the short measurement period. Up to now, transmission electron microscopy (TEM)-EDS has not been greatly used in the analysis of elements for semiconductor doping. This is because the detection limit by Si (Li) - EDS is of the order of 1500-2000 ppm, which is insufficient for semiconductor industry. There has consequently been a strong desire to see "limit of detection (LOD)", that is, how small an amount of elements we can be detected, using a multiple SDD analysis system. In this study, we explore the concentration at LOD for As dopant in Si device using FETEM with a dual SDD analysis system. We used an As-doped Si wafer as a sample. The depth profile of As concentration in the Si wafer has been verified by Secondary Ion Mass Spectrometry (SIMS). We made a cross sectional lamella sample for TEM observation using an ion milling (Ion Slicer EM-09100IS, JEOL). The thickness of the TEM sample is estimated to be 140 nm judging from the count rate of X-ray and probe current. We measured several data sets of EDS spectrum imaging using a 200kV FETEM (JEM-2800 JEOL) equipped with two SDD detectors (sensor area = 100 mm2). The concentration profiles were extracted and converted from the data set with a quantitative analysis system (NSS 3, Thermo Fisher Scientific). Figure 1 shows cross sectional STEM-HAADF image of the sample. The yellow, brown and blue rectangles show the regions of analyzed, As rich and pure Si. The locations for As rich and pure Si were determined from As depth profile of SIMS. Figure 2 shows EDS spectra from the brown rectangle region. We can see the peaks for As-K and As-L. Figure 3 shows concentration profiles of As in Si using As-K and As-L lines. Theoretical estimation of LOD is determined with competition between the net counts for dopant element and the fluctuation of background tail of Si peak at the dopant peak energy. If we take the net counts of dopant element is three times larger than the fluctuation of the background, which confirms the presence of the dopant with 99.7% certainty, we define an approximate dopant concentration (Cdopant) at LOD as 3k C dopant (1) I net P / B ,where the Inet is net count of Si-K, the P/B is ratio of net count of Si (P) and background count at energy of As lines (B), and k is K factor for lines of dopant, which is unity in this calculation. The theoretical Cdopant are calculated to be 30 ppm for As-K and 80 ppm for As-L from equation (1). These are value of blue lines (1nA, 80 min) shown in Fig. 3(a) and 3(b). The values for the calculation are listed as Inet = 6,800,000, P/B for As-K = 1,480, P/B for As-L = 220. The Cdopant for As-K is smaller than that for As-L, although net count of As-L = 9,800 is larger than that of As-K = 5,100. This is due to that the P/B of As-K is better than that of As-L. Therefore, in order to increase the sensitivity, it is important not only to increase the number of counts, but it is effective to choose a line of X-ray having a high P/B. Microsc. Microanal. 21 (Suppl 3), 2015 824 References: [1] S. Kawai, I. Ohnishi, T. Ishikawa, K. Yagi, T. Iwama, K. Miyatake, Y. Iwasawa, M. Matsushita, T. Kaneyama and Y. Kondo, Microscopy & Microanalysis 20 (S3) (2014), p. 1150-1151. [2] I. Ohnishi, S. Kawai, T. Ishikawa, K. Yagi, T. Iwama, K. Miyatake, Y. Iwasawa, M. Matsushita, T. kaneyama and Y. Kondo, Proceedings of IMC18 (Prague 8-12 Sept.) (2014). [3] This equation is equivalent listed in "Introduction to Analytical Electron Microscopy" (ed. J J Hren, J I Goldstein and D C Joy, Prenum Press) Chap.3 (1979) p.113 with 3 times standard deviation. Figure 1. STEM-HAADF image of cross sectional sample made from As-doped Si wafer. Figure 2. EDS spectra from the brown rectangular region in Fig. 1. (a) As-L, (b) As-K. Figure 3. Profiles of As concentration in Si using (a) As-K and (b) As-L lines. The yellow arrows are theoretical estimation of Cdopant at LOD (blue line). Red lines indicate the concentration profile measured with SIMS.");sQ1[413]=new Array("../7337/0825.pdf","Application of Bremsstrahlung Background Calculation and Automated Element Identification to TEM EDS Spectra","","825 doi:10.1017/S1431927615004924 Paper No. 0413 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Application of Bremsstrahlung Background Calculation and Automated Element Identification to TEM EDS Spectra F. Eggert1, P.P. Camus1, A. Sandborg1, N.J.Zaluzec2 1 EDAX Inc., Ametek Materials Analysis Division, 91 McKee Drive, Mahwah, NJ, US-07430 2 Electron Microscopy Center, NST Div, Argonne National Laboratory, Argonne IL, USA Qualitative analysis is a major benefit of Energy Dispersive Spectrometers (EDS) based on collection of whole spectral X-ray distribution in one step. An automated element identification approach, EXpertID, was developed for evaluation of SEM spectra with utilization of decision loops based on internal quantitative calculation assessments and repeated fit of spectra reconstructions and simulations [1]. The required calculation of bremsstrahlung distribution for background approximation follows [2] and was modified with the absorption terms E and AE [3]: N Ebr = E E 0 - E (E0 - E )2 AE a + b +c E E E E0 Photon energy E and primary electron energy E0 NE br Number of bremsstrahlung counts in channel with photon energy E E Detector efficiency with photon energy E AE Self absorption in specimen with photon energy E a, b, c Coefficients for spectra fit If let a always zero, for each spectrum segment the coefficients b and c are possible to determine at the respective flanking two preselected fit points (regions) containing no X-ray peaks. There is no limit to the number of fit regions that are possible to use. The coefficients change with each section. The c is additionally zero between spectrum begin and first fit point and also until the end. A different X-ray excitation and absorption physics must be considered with application for TEM specimen spectra. The primary electron energy with typically 200 kV is much higher than with SEM. Because the TEM specimens are very thin, the self absorption AE is much reduced and depends on specimen thickness. The model was extended by introduction of a TEM specimen layer thickness and density. The absorption calculation for thin TEM specimen follows the simple X-ray absorption law and is even possible to neglect in many cases. Therefore the detector efficiency E becomes the crucial part of the background approximation. Modern SDD detectors for TEM have no vacuum window (windowless) [4]. The knowledge of the detector absorption layers is then the most important part of absorption effects. The windowless SDD spectra have usually many more counts in the low energy region and the measured background is very close to the primarily emitted Bremsstrahlung quanta distribution. Some investigations are presented to show an improved TEM-spectra background approximation for windowless SDD. The detector chip front contact structure has the major influence to the detector efficiency and its knowledge is a basic requirement. Actually it seems that using a windowless SDD in the TEM is a setup to investigate the SDD front contact structure. Finally the background calculation is a core part of automated element identification, which is based mainly on repeated spectra reconstructions. A good approximation also in low energy parts of the Microsc. Microanal. 21 (Suppl 3), 2015 826 spectrum is required to get proper element decisions with corresponding line assessments (K->L and L->M coupling of lines from same element). Some examples are presented with application of the modified and improved background calculation and automated element identification EXpertID for TEM EDS spectra evaluation. References: [1] F. Eggert, IOP Conf. Ser.: Mater. Sci. Eng. 7 (2010) 012007. [2] L.Lifshin, Ottawa, Proc.9.Ann.Conf.Microbeam Analysis Soc. (1974) 53. [3] F.Eggert, Experim.Techn.d.Physik 33 (1985) 441. [4] EDAX Technical Note, 2014 June [5] Research supported in part by the U.S. DoE, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357 at the Electron Microscopy Center, NST Division of Argonne National Laboratory");sQ1[414]=new Array("../7337/0827.pdf","Materials Selection for Ultra-Thin Diamond-Like Carbon Film Metrology and Structural Characterization by TEM.","","827 doi:10.1017/S1431927615004936 Paper No. 0414 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Materials Selection for Ultra-Thin Diamond-Like Carbon Film Metrology and Structural Characterization by TEM. Guilherme P. Souza1, Kurt C. Ruthe1 , Lifan Chen2, Liang Hong2 and Haifeng Wang2 Western Digital Corporation, Magnetic Heads Operations, Metrology & Materials Characterization, Ayutthaya, Thailand, 13160 2. Western Digital Corporation, Magnetic Heads Operations, Materials Characterization, Fremont, CA, USA, 94539 Critical dimension metrology by transmission electron microscopy (TEM) plays a paramount role in ultra-thin 10-30� diamond-like carbon (DLC) films used in hard-disk drive manufacturing, where performance is traded against reliability as the nominal thickness continues aggressive scaling downward [1]. TEM sample preparation requires that a protective coating is deposited prior to sitespecific focused ion beam (FIB) cross-sectioning. This coating is also critical in providing a high contrast delineating marker of the DLC top surface for imaging contrast-based metrology. The criteria that such protective coating must be inert to the DLC, as well as free of coarse structure, narrows the materials selection significantly. Cr is one of the most common coating materials used due to its legacy from SEM coaters for high resolution applications as it forms a continuous, quasi-amorphous film. As the literature on protective layer materials selection for ultra-thin DLC film is non-existent, we disclose original research that has led to a breakthrough in the industry and expanded the current understanding. DLC film was deposited on NiFe substrate using the conventional filtered cathodic arc (FCA) process [2]. TEM imaging and EELS were performed using a Schottky field emission gun TEM operating at 200 kV, and a post-column spectrometer. Samples, having the DLC film as outer surface layer, were initially coated prior to FIB with either Cr or Cr2O 3 by ion beam sputtering. C and Cr quantified elemental profiles across a Cr-coated DLC film are shown in Fig.1(a). C is clearly skewed towards the Cr layer, indicating interaction. C-K edge extracted from the interfacial region reveals that carbon exists as (Cr-) carbide, with a characteristically intense * event [3] (Fig.1b). On the other hand, profile skewness is not present when the same DLC is coated with Cr2O3 (Fig.1a), and the C-K edge does not possess the intense * indicative of carbide (Fig.1b), retaining a more DLC-like shape instead. Thus the interaction between C and the protective coating material is central to the interpretation of spectroscopic data. TEM-based film thickness metrology of the DLC layer is also affected by its interaction with Cr, as shown in Fig.2(a). Using identical film thickness metrology definitions, the Cr2O3 -coated DLC film is ~ 10� thicker than the Cr-coated counterpart. The excess film thickness comes from the fact that C formed carbide with Cr. To evaluate the degree of interaction between DLC and Cr2O3 (and Cr), evaporated Au film was deposited on the DLC. Evaporated Au is widely known to be the most gentle and thus the least interacting film deposition technique available. Results show that the Au-coated DLC film thickness is comparable to that of the Cr2O 3 -coated film, confirming that Cr2O3 deposited by the ion beam sputter deposition technique is likewise gentle and non-interacting to the DLC film. However, Au formed large grains (Fig.2b) which disturbed the continuity of its interface with the DLC. Fig.2(c) shows a cross-section of the same DLC sample without any protective coating applied, which was achieved by a very elaborate, manual sample preparation methodology on a Ta coupon substrate that cannot be reproduced for industrial-scale metrology. The uncoated DLC thickness is equivalent to that of evaporated Au-coated and Cr2O3 -coated DLC samples, thus further validating the non-interacting nature 1. Microsc. Microanal. 21 (Suppl 3), 2015 828 of Cr2O3 protective layer for ultra-thin DLC films. Other coating materials tested in this study, such as Ti, Ir, W, and Ta, all show some degree of interaction with C varying between Cr and Cr2O 3, and therefore are not as ideal as Cr2O3 for use as protective marker coating materials for DLC. In conclusion, ultra-thin DLC characterization by TEM is strongly dependent on the choice of protective material applied prior to cross-sectioning. In particular, the interplay between oxidation and carbidization leads to metrology inaccuracy and misinterpretation of structural information, as well as potential metrology control excursions given that Cr film stability is dependent on vacuum conditions and target material cleanliness. Cr2O3 makes a more reliable, stable and robust protective coating system in that regard, allowing for TEM/EELS characterization of the true, artifact-free DLC film at � scale. References: [1] AC Ferrari, Surf. Coat. Tech. 180-181 (2004), p. 190. [2] J Robertson, Mater. Sci. Eng. R37 (2002), p. 129. [3] X Fan et al, Appl. Phys. Lett. 75-18 (1999), p. 2740. (b) (a) Figure 1. (a) EELS profiles for C, Cr, and Cr2O3 across the DLC. Cr and C form a carbide, as shown by the pronounced C-K edge * event (b) compared to the more DLC-like character retained in the case of the Cr2 O3 -coated sample. C-K edges were taken from the interface between DLC and protective layers. (a) (b) (c) Figure 2. (a) Comparison between Cr- and Cr2O3 -coated DLC showing that the Cr-coated DLC is ~ 10� thinner due to C/Cr interaction; not seen for the Cr2 O3 -coated DLC. Evaporated Au-coated (b) and uncoated (c) DLC films have the same thickness as compared to Cr 2O3 -coated DLC, confirming that Cr2O3 protective layer does not interact or change ultra-thin DLC films for TEM characterization.");sQ1[415]=new Array("../7337/0829.pdf","STEM characterization of Gold-Copper anisotropic nanocrystals","","829 doi:10.1017/S1431927615004948 Paper No. 0415 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 STEM characterization of Gold-Copper anisotropic nanocrystals Lourdes Baz�n-D�az 1,2, Rub�n Mendoza-Cruz1, 2, J.Jes�s Vel�zquez-Salazar1, Ulises Santiago1, Daniel Bahena Uribe3, Miguel Jos�-Yacam�n1. 1. Department of Physics & Astronomy, The University of Texas at San Antonio, One UTSA Circle, San Antonio, TX 78249, USA. 2. Instituto de F�sica, Universidad Nacional Aut�noma de M�xico, A. P. 20-364 Distrito Federal C.P. 01000 M�xico 3. Cinvestav. Av. Instituto Polit�cnico Nacional 2508, Col. San Pedro Zacatenco, Delegaci�n Gustavo A. Madero, M�xico D.F. C�digo Postal 07360 Alloy nanoparticles are an important group of nanomaterials exhibiting size, shape, structure and composition dependent properties. In bulk, gold-copper alloys exhibits ordered phases Au3Cu (L12), AuCu (L10), AuCu3 (L12) [1]. Theses phases could be modified in temperature and composition at the nanoscale. For instance, ligands play an important role in the synthesis of bi-metallic nanoparticles because they influence the final shape and size of the nanoparticle. The most common method used to synthesize alloyed nanoparticles is wet-chemistry that make use of phosphoric acids, polymeric chains or thiol groups as surfactants in organic solvents such as toluene that control the spatial and shape distribution of the nanoparticles. However, the knowledge of the internal structure of the crystals provides insights to understand the crystal growth of the nanosystem. In recent years it has been reported the relationship between shape and internal features such as stacking faults and twin boundaries on nanowires and decahedral particles [2]. This features (twins, stacking faults and other defects) can modified the final shape of nanoparticles. Some other factors that affect in the same way are the concentration of the reactants and the temperature ranges. Therefore, in this work we present a systematic study on gold-copper bimetallic system to analyze the internal structure of ultrathin nanowires, decahedral nanostars and nanocubes. Modifying few conditions during the synthesis, such as metal concentrations and surfactant used, namely hexadecylamine (HDA), octadecylamine (ODA), oleylamine (OLA) or 1-dodecanethiol (DDT), we obtained Au-Cu nanocrystals with the different morphologies aforementioned. Through high resolution transmission electron microscopy (HRTEM), High Angle Annular Dark Field (HAADF) imaging and Energy Dispersive X-Ray Spectroscopy (EDS) we have obtained atomic resolution that allow us to describe the internal structure of the synthesized particles. Fig. 1 shows a coiled ultrathin Au-Cu nanowire with diameter below 3 nm synthesized at low temperature. There have been few reports about similar nanostructures synthesized at high temperature [3-5]. However, to our knowledge, no previous reports have been done about this system so it represent by itself a novel and exceptionally interesting structure, since multiple twins and stacking faults are present along the whole wire and determine its growth. Images in Fig. 2 correspond to pentagonal nanostars synthesized by using HDA (Fig. 2a) or OA (Fig.2b) and gold-copper nanocubes (Fig. 2c). From the high resolution images it was inferred that, in the case of the nanostars, the center of the particle consist of 5-fold-symmetry decahedral seeds and each branch grows along the twinning plane as is shown. Furthermore, the edge of each branch is reconstructed by stepped planes, which give rise to {433} high index facets. On the other hand, when OLA is used, we obtain novel pentagonal nanoparticles similar to those made with HDA but the nanocrystals are smaller and the branches are rounded (Fig.2b). They also exhibit a core with 5-fold symmetry, but unlike the first Microsc. Microanal. 21 (Suppl 3), 2015 830 ones, the branches grow along the [001] direction instead of the twinning plane of the 5-fold core. The branches show no twinning and its edges are formed by {111} and {110} stepped planes. In turn, goldcopper nanocubes were synthesized by using 1-dodecanethiol. The mean diameter of these cubes is around 8 nm. They present flat {200} flat faces and rest on the [001] axis zone as is shown in Fig. 2c. In conclusion, we are able to obtain novel gold-copper nanocrystals with different morphologies by modifying the synthesis conditions and the surfactant employed. Undoubtedly, the knowledge of the structural arrangement at nanoscale level as well as the distribution of the elemental composition will bring a new way to understand the atomistic structures when bimetallic alloys are grown with twin boundaries and stacking faults. References: [1] Taherkhani F et al., Alloy Compd 617 (2014), p. 746. [2] Ying C et al., Cryst. Growth Des 11 (2011),p.5457. [3] Teng X, et al., Angew Chem Int Ed 47 (2008), p. 2055 [4] Velazquez-Salazar, J. J et al., ACS nano 5 (2011), p. 6272. [5] Hong, X et al., Chem Commun 47 (2011), p. 5160. [6] This project was supported by a grant from the National Institute on Minority Health and Health Disparities (G12MD007591) from the National Institutes of Health. Fig. 1. Ultrathin gold-copper nanowire. Twinning and stacking faults rule its growth and final structure. Fig. 2. Nanostars synthesized by using HDA (a) and OLA (b). It is noted the 5-fold symmetry of the core and the different direction of growth. (c) Nanocubes synthesized by using 1-dodecanethiol.");sQ1[416]=new Array("../7337/0831.pdf","EFTEM Contrast Tuning and EELS Fine Structure Analysis for Characterization of Carbon Containing Ultra-Low-k Dielectric Materials","","831 doi:10.1017/S143192761500495X Paper No. 0416 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 EFTEM Contrast Tuning and EELS Fine Structure Analysis for Characterization of Carbon Containing Ultra-Low-k Dielectric Materials Brenda Prenitzer1, Stephen Schwarz1, Brian Kempshall1, Imen Rezadad1,2 1. NanoSpective, Inc., Orlando, Florida, USA. 2. Physics Department, University of Central Florida, Orlando, Florida, USA. Miniaturization of transistors is but a single leg of the race to improve speed in integrated circuits (IC). Interconnect is equally as crucial to the performance of advanced technology node devices. As production devices are emerging at the 14nm node, there is a corresponding and necessary reduction in interconnect pitch. Blazing fast processing speeds and densely packed routing intensifies the imperative to use advanced ultra-low-k (ULK) dielectric materials to reduce the RC time delay, minimize crosstalk and lower power consumption. Technological evolution has ushered in the integration of progressively lower dielectric constant materials starting with fluorinated dense materials through carbon-laden polymers, organically modified SiOx and into the era of porous silicate materials with incorporated organic species. While these porous insulators have proven to perform admirably, there are concurrent collateral complexities for process integration and as well as for characterization. These materials are less thermally and mechanically robust than their traditional SiOx predecessors. Nevertheless, the reliability of the IC is dependent on the ability of the dielectric to maintain its integrity through the severe processing steps and beyond.1 A multitude of characterization opportunities are generated by the use of ULK materials in the interlevel dielectric (ILD) and the required complex support network of films for diffusion barrier, adhesion, etch stop to name a few. Some properties of interest, e.g., overall morphology, composition, etc. are handily conquered by conventional analytical microscopy; however, some of the most important properties of ultra-low-k dielectrics are dependent on the pore structure, permittivity, density and chemistry. Methods that are well suited for thin film analysis on a blanket wafer find their limits on a fully processed wafer or die. In general, most characterization methods either lack sufficient sensitivity or lateral spatial resolution to adequately characterize ULK dielectrics. Ever increasing sophistication in tools and innovation in techniques is required to keep pace. Electron energy loss spectroscopy (EELS) and energy filtered transmission electron microscopy (EFTEM) are applied to the characterization of the ILD and etch stop layers (ESL) of ICs at a technology node of 32nm or smaller. EELS is used to probe valence and conduction bands to determine critical information about the local bonding environment of Si, O, N, C and H in ULK and ESL layers. The spectra can be used as a chemical fingerprint. Examples of contrast enhancement using EFTEM contrast tuning in carbon containing materials will be shown. Direct imaging of nanometer size pores through the thickness of the TEM specimen is confounded by the fact that the image is a 2D projection of a 3D volume. With pore sizes on the order of 1-2nm, the projected volume will contain the superposition of several pores. Filtered imaging enhances the ability to observe the pore structure in ULK dielectrics by forming images with the relatively few electrons that have suffered inelastic interactions corresponding to the selected energy of position of the spectrometer slit. References [1] Sheng-Wen Chen et al., Materials 5 (2012), 377-384 [2] James M. Howe, et al., Journal of Electron Microscopy 53(4) (2004), 339-351 Microsc. Microanal. 21 (Suppl 3), 2015 832 Figure 1 EFTEM energy filtered series. The enhanced contrast allows the direct observation of the multilayered dielectric structures as well as the porosity in the ILD. Figure 2. Low and core electron energy loss spectra. The fine structure for each material is distinctive and is indicative of how the density of unoccupied states changes with atomic bonding.");sQ1[417]=new Array("../7337/0833.pdf","The OTO Specimen Preparation Method for Optimal Scanning Electron Microscopy Imaging of Pseudomonas aeruginosa","","833 doi:10.1017/S1431927615004961 Paper No. 0417 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The OTO Specimen Preparation Method for Optimal Scanning Electron Microscopy Imaging of Pseudomonas aeruginosa Judy Tsz Ying Lee1, Nanthawan Avishai2, and Kwai Ping Connie Tam1,3 1. 2. Departments of Ophthalmic Research, Cleveland Clinic Cole Eye Institute, Cleveland, OH, USA. Swagelok Center for Surface Analysis of Materials, Case Western Reserve University, Cleveland, OH, USA. 3. Departments of Ophthalmic Research, Cleveland Clinic Lerner College of Medicine, Cleveland, OH, USA. Pseudomonas aeruginosa, a ubiquitous gram-negative rod-shaped bacterium, has been intensively studied as an opportunistic human pathogen. It is one of the most common pathogens for nosocomial infection in immunocompromised individuals, e.g. cystic fibrosis patients [1]. Scanning electron microscopy (SEM) is a useful tool for obtaining detailed surface topography of microorganisms. For instance, it has been used for studying the ultrastructural basis of the resistance of P. aeruginosa to antiseptics, disinfectants and antibiotics [2]. Several methods for SEM sample preparation have been developed in order to enhance contrast, reduce structural damage and preserve cell structure in the native state. These techniques include glutaraldehyde fixation, negative staining, cryo-techniques, critical point drying, coating specimens with gold or osmium, and OTO staining. OTO staining has mostly been used in the preparation of biological tissues to provide bulk conductivity, which enables enhanced contrast [3]. Here, we report a specimen preparation protocol for optimal SEM imaging of P. aeruginosa using the OTO staining method. In our optimized protocols, P. aeruginosa in the mid-exponential growth phase was washed and diluted with 0.9% sodium chloride to a cell density of 108 cfu/ml. It was then fixed with 2.5% glutaraldehyde in PBS and deposited onto a 0.4 m-pore-size polycarbonate membrane. After fixation for 1 hour at room temperature and overnight at 4�C, it was post-fixed with 1% OsO4 for 1 hour, followed by 1% thiocarbohydrazide (TCH) for 5 min, and 1% OsO4 again for 5 min, with thorough washing between each step. The sample was dehydrated with graded ethanol series, and then impregnated with 50% hexamethyldisilazane (HMDS) in ethanol, and finally 100% HMDS. The air-dried sample was sputtered with gold before imaging. As a comparison, we have also processed the sample without the steps of 1% TCH followed by 1% OsO4. Compared to the specimen prepared with the conventional method (Fig.1), the OTO method revealed significantly higher resolution ((Fig.2), which allows the observation of bacterial surface details. Our protocol also helps to preserve the cells during sample processing, as cell shrinkage was only observed in the conventional method. The improved image quality allows detection of subtle effects of any treatments to the bacterial surface morphology. Microsc. Microanal. 21 (Suppl 3), 2015 834 References: [1] J.M. Plotnikova, L.G. Rahme and F. M. Ausubel, Plant Physiol. 124 (2000), 1767-1774. [2] U. Tattawasart, et al. J. of Antimicrobial Chemotherapy. 45 (2000), 145-152. [3] S. V. Buravkov, V. P. Chernikov, and L. B. Buravkova. Bulletin of Experimental Biology and Medicine. 151 (2011), 378-382. A B 1 �m 500 nm Figure 1. SEM images of P. aeruginosa using the conventional method for sample preparation at magnifications of 25,000x (A) and 50,000x (B). The SEM images were taken with FEI Helios Nanolab 650 using 2 KV, 50 pA with TLD mode 2. Loss of surface components and cell shrinkage were observed. A B 1 �m 500 nm Figure 2. SEM images of P. aeruginosa using the OTO method for sample preparation at magnifications of 25,000x (A) and 50,000x (B). The SEM images were taken with FEI Helios Nanolab 650 using 2 KV, 50 pA with TLD mode 2. Bacterial surface components were well preserved and cell shrinkage was not observed.");sQ1[418]=new Array("../7337/0835.pdf","Thin Film and Multilayer Scintillators for Low Voltage Backscattered Electron Imaging in the Scanning Electron Microscope","","835 doi:10.1017/S1431927615004973 Paper No. 0418 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Thin Film and Multilayer Scintillators for Low Voltage Backscattered Electron Imaging in the Scanning Electron Microscope NC Barbi1, MB Tzolov2, OE Healy1, RB Mott1, WM McDonald2 1 2 PulseTor LLC, Pennington, NJ USA Lock Haven University, Department of Geology and Physics, Lock Haven, PA USA Scintillators produce a number of photons proportional to the average energy of the incident electrons. Low energy electron detection using scintillators therefore starts at a disadvantage compared to imaging at higher energies. The predominant scintillator used in electron microscopy is crystalline Ce-doped Yttrium Aluminum Garnet (YAG), a poor electrical conductor that must be coated with a conductive layer on the impact surface to drain the excess charge induced by the very electrons which it must detect. Fortunately, this conductive layer (Al or ITO) can be very thin, on the order of 10-20 nm, and therefore has little effect on electrons with energy greater than 10kV. At low energies, however, absorption in the conductive film becomes a factor. The combined result of lower yield at low energies and the absorption by the conductive layer is that BSE imaging is seldom used below 5kV. New detector designs, using scintillators coupled to Silicon PhotoMultipliers (SiPMs), allow different sensor segments to be placed around a central hole in the detector, meaning that different scintillator types can be used in the same device [1]. In such "Scintillator on Multiplier" (SoM) detectors, electrically conductive ZnO, for example could be used without coating as a scintillator for low energy electrons in one or more sensor locations, and YAG (Ce) in the remainder. Figure 1 repeats the illustration of a single detector using different scintillator materials. Because there is an SiPM coupled to each scintillator segment, the imaging signal can be arbitrarily derived from a single sensor, one type of sensor, or all the sensors combined. We have extended this idea further by developing thin films of zinc tungstate (ZT) showing photoluminescence quantum efficiency in the range of 70%, and which exhibit sufficient conductivity to drain the charge typically encountered in SEM imaging. The nanostructure of these films leads to improved scintillation properties compared to the bulk material, and indeed, compared to ZnO. The absence of an additional top coating allows electrons to directly interact with the surface of the film and produce scintillation. Figure 2 shows improved imaging realized with ZT at 2kV compared to YAG, both images being collected simultaneously. The SNR as a function of incident beam energy (Figure 3) shows the advantage of thin ZT for low energy, while confirming the superiority of bulk YAG for higher energies. Figure 4 shows an image taken at 1 kV with ZT. For comparison, the emission spectrum of zinc tungstate is compared with that of YAG and YAP (Yttrium Aluminum Perovskite) in Figure 5. The improvement provided by thin film ZT for low energies, the superiority of YAG (Ce) for higher energies, and the flexibility of the deposition process together suggest that integrating ZT with YAG into a multilayer structure would be advantageous. We have constructed the first multilayer ZT/YAG scintillator. This structure will combine the efficient interaction of low energy electrons with a surface ZT film and the efficient collection of high energy electrons by the significantly thicker YAG scintillator underneath. This compact scintillator structure can extend the application of BSE imaging Microsc. Microanal. 21 (Suppl 3), 2015 836 to low voltage, while maximizing efficiency through optimum use of the available sensor space in the detector. [1] N.C. Barbi, et al Microsc. and Microanal., 2011 4X4 mm SSPM YAG:Ce 4X4 mm SSPM ZnO:Ga 4X4 mm SSPM ZnO:Ga 4X4 mm SSPM YAG:Ce Figure 1: Example of SoM detector configuration having a segmented configuration enabling use of different scintillator materials Cu Al Al C Cu C Figure 2: Cu grid - Al stub - C tape 2kV 2.5nA 80 second integration time Figure 3: SNR as a function of kV for ZT and YAG images Figure 4: 1kV BSE image of sintered stainless using ZT scintillator steel 2o Figure 5: Emission spectra for YAG, ZT, and YAP ZT, YAP and YAG");sQ1[419]=new Array("../7337/0837.pdf","Investigation of Image Contrast of Energy-Filtered BSE Image at Ultra Low Voltage","","837 doi:10.1017/S1431927615004985 Paper No. 0419 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Investigation of Image Contrast of Energy-Filtered BSE Image at Ultra Low Voltage Y. Hashimoto1, A. Muto1, T. Walters2, E. Woods2 and D. C. Joy3,4 1. Nano-Technology Systems Div., Hitachi High Technologies America, Clarksburg, MD 20871, USA Institute for Electronics and Nanotechnology, Georgia Institute of Technology, Atlanta, GA 30332, USA 3. Material Science and Engineering, University of Tennessee, Knoxville, TN 37831, USA 4. Center for Nanophase Materials Science, Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA 2. To meet the increasing demands to clarify the compositional differences in advanced composite materials, we reported a new imaging method using energy filtered BSE signals at Ultra Low Voltages (ULV) [1][2]. This method was applied to carbon nanotube (CNT) and polytetrafluoroethylene (PTFE) composite films to confirm its ability to differentiate distribution of CNT and PTFE components at 0.3 kV as shown in Figure 1. We confirmed that the difference of surface potential affects the contrast between CNT and PTFE more strongly than the differences of average atomic number and surface morphology for this specimen. To apply this method to any other materials, it will be necessary to investigate the effect of the difference in average atomic number and surface morphology on the contrast because the signal behavior under ULV condition does not follow the conventional theory used over 1 kV [3]. In this study, we investigated the effect from the difference of average atomic number by fundamental experiments and simulation techniques. Figure 2 shows a general view of the Hitachi SU8230 FE-SEM and a schematic of signal detection system using the deceleration mode. To control the signal detected by the top detector, an energy filter called "top filter" is used. This works as a high pass filter, so it allows only electrons with higher energy than the filtering voltage to be detected. High angle BSE (HA-BSE) signal can be detected by the top detector with an appropriate combination of landing voltage, decelerating voltage, and filtering voltage. A fundamental experiment of the contrast caused by the difference of average atomic number on BSE imaging at ULV was carried out using a basic test specimen. This consists of four pieces of Si wafer, and 50 nm thick films of three types of materials with different atomic weights (C, Cr, Pt) are deposited on three pieces. Figure 3(a) shows a BSE image of the test specimen at 0.3 kV. In this image, the contrast between Si, Cr, and Pt is observed. Figure 3(b) shows comparison of image brightness between each material against landing voltage. The contrast does not follow the conventional theory used over 1 kV. For example, Pt has the highest Z number and it is brightest at 1 kV, but it is also the darkest at 0.3 kV. This tendency partially conforms to a previous study [3]. We will disclose more details including simulation results in the session. Acknowledgement The CNT / PTFE composite film specimen was kindly supplied by Prof. Yoshiyuki SHOW of the Department of Electrical and Electronic Engineering, School of Engineering, Tokai University. Microsc. Microanal. 21 (Suppl 3), 2015 838 References : [1] Y. Hashimoto et al., Microsc. Microanal. 19 (Suppl 2), 1176-1177 (2013). [2] Y. Hashimoto et al., Microsc. Microanal. 20 (Suppl 3), 34-35 (2014) [3] Ilona M�llerov�, Scanning, 23, 379 (2001) (a) (b) CNT PTFE CNT PTFE Fig.1 BSE images of CNT / PTFE composite film at 0.3 kV detected by (a)Upper detector, (b)Top detector Fig.2 General view of Hitachi SU8230 FE-SEM and a schematic of signal detection system using the deceleration mode (a) (b) Fig. 3 (a)BSE image of test specimen at 0.3 kV, (b)Comparison of image brightness between each material against landing voltage");sQ1[420]=new Array("../7337/0839.pdf","Low-Energy Electron Diffractive Imaging Based on a Single-Atom Electron Source","","839 doi:10.1017/S1431927615004997 Paper No. 0420 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Low-Energy Electron Diffractive Imaging Based on a Single-Atom Electron Source I.-S. Hwang1, W.-T. Chang1, C.-Y. Lin1,2, and W.-H. Hsu1,3 1. 2. Institute of Physics, Academia Sinica, Nankang, Taipei, Taiwan. Department of Physics, National Taiwan University, Taipei, Taiwan. 3. Department of Materials Science and Engineering, National Tsing Hua University, Hsinchu, Taiwan. Imaging of the atomic structures of two-dimensional materials and organic materials is a challenge for current electron microscopes because of low imaging contrast and high radiation damage for highenergy electrons. It has been a general trend to develop electron microscopy of lower energies. Thanks to the progress in aberration-correction techniques, transmission electron microscopes with voltages down to 15-40 kV have recently been demonstrated. However, it becomes very difficult to achieve atomic resolution when the electron energy is reduced below 10 keV. An alternative approach is phase retrieval imaging, which requires a sufficiently coherent source, detection of high-angle diffracted patterns with a sufficient resolution, and a sufficiently small detection area on the sample. There is no need to fabricate high-quality lenses with a large numerical aperture. In this work, we propose several experimental schemes of low-energy electron coherent diffractive imaging (CDI). A great advantage of low energy electrons over high-energy electrons and x-ray is that the cross-sections of interaction with the atomic potentials are very large, so that the diffraction pattern has good signal-to-noise ratios even at high scattering angles and thus high-resolution images can be reconstructed. We plan to use noble-metal covered W(111) single-atom tips (SATs) as the electron sources for lowenergy electron CDI. High brightness and fully spatial coherence of electron beams emitted from this type of SATs have been demonstrated [1-3]. This type of SATs can be reliably prepared and regenerated in vacuum [2,3]. Their pyramidal structures are thermally and chemically stable, ensuring their long operation lifetime. To achieve a sufficiently small detection area on the sample, two types of schemes are proposed here. The first one is based on a lens-less imaging technique. Fig. 1 shows a schematic of an electron point projection microscope (PPM) with a MCP-screen mounted on a rail. The detector screen can be moved in a certain range behind the sample to record the transmission patterns at different positions. Highmagnification images can be acquired when the detector screen is moved to the far end. High-angle diffraction patterns can be obtained when the detector is moved close to the sample. One example is shown in Fig. 2(a). The central pattern is the projection image or hologram of the sample, and six surrounding patterns are observed, which are related to diffraction of a single domain of monolayer graphene. The diffraction patterns are shown individually with enhanced contrast in Fig. 2(b). The central and diffracted patterns resemble the bright-field images and multiple dark-field images, respectively, in selected-area diffraction of TEM when the electron beam is defocused on the sample. The lens-less scheme is simple and can achieve a high imaging contrast due to the use of low-energy electrons (20-500 eV). However, it can be applied to samples of thickness less than 1 nm. Thus, there is a need to develop a low-keV electron diffraction microscope with energies between 1 and 10 keV. Here we propose several imaging modes for a transmission-type low-keV EM (Fig. 3). A CDI scheme with a parallel beam to illuminate a sample is illustrated in Fig. 3(a). Fig. 3(b) illustrates a scheme with a slightly convergent beam to illuminate the sample. This scheme can reach a spot size fulfilling the Microsc. Microanal. 21 (Suppl 3), 2015 840 oversampling requirement more easily than that in Fig. 3(a). Fig. 3(c) illustrates a scheme of in-line electron holography, same as the one originally proposed by Gabor [4]. In this case, an object is illuminated with a focused electron beam of a large converging angle in front of or behind the object. [1] TY Fu, LC Cheng, CH Nien and TT Tsong, Phys. Rev. B 64 (2001) 113401. [2] HS Kuo, IS Hwang, TY Fu, JY Wu, CC Chang and TT Tsong, Nano Lett. 4 (2004) 2379. [3] CC Chang, HS Kuo, IS Hwang and TT Tsong, Nanotechnology 20 (2009) 115401. [4] D Gabor, Nature 161 (1948) 777. [5] The authors acknowledge funding from Academia Sinica of R. O. C. (AS-99-TP-A02). Figure 1. Schematic of a lens-less electron diffraction microscope with a retractable MCP. Figure 2. Transmission patterns recorded on a monolayer graphene with a lens-less electron diffraction microscope. (a) Diffraction patterns of a monolayer graphene sample. The electron energy is 270 eV. (b) Six dark-field patterns with enhanced contrast are shown. Figure 3. Proposed schemes of a low-keV electron microscope. (a) CDI with a parallel beam to illuminate the sample. (b) CDI with a slightly convergent beam to illuminate the sample. It also allows operation of scanning electron microscopy if the secondary electrons are collected by an electron detector. (c) In-line holography.");sQ1[421]=new Array("../7337/0841.pdf","Level Set Method for Tip Shape Evolution Simulation for Atom Probe Tomography","","841 doi:10.1017/S1431927615005000 Paper No. 0421 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Level Set Method for Tip Shape Evolution Simulation for Atom Probe Tomography Jie Bao1, Zhijie Xu2, Robert Colby3, Suntharampillai Thevuthasan4, Arun Devaraj3 1 2 Nuclear Science Division, Pacific Northwest National Laboratory, Richland, WA, USA; Computational Mathematics Division, Pacific Northwest National Laboratory, Richland, WA, USA; 3 Environmental Molecular sciences Laboratory, Pacific Northwest National Laboratory, Richland, WA, USA; 4 Qatar Energy and Environmental Research Institute, Qatar, UAE 1. M�ller, E. W.; Panitz, J. A.; McLane, S. B., The atom-probe field ion microscope. Preview of Scientific Instruments 1968, 39 (1), 83-86. Invented in 19671, atom probe tomography (APT) was designed to visualize chemical heterogeneity present in a small volume of a specimen with a sub-nanometer spatial resolution 2-5. Numerical simulation plays an important role in studying the sample geometry evolution and corresponding electrical field changes during field evaporation of needle shaped specimens in APT. The level set method can provide sub-grid accuracy on tracking the solid-vacuum interface and much higher computational efficiency than other models that require an atomic level grid. The modeling details are introduced in our previous work 6. The proposed approach is applied to predict the tip evaporation of the specimen, which is (100) chromium (Cr) single crystalline thin film grown on a (100) single crystalline magnesium oxide (MgO) substrate. Through comparison with experimental observation, the level set method shows the capability of providing atomic level accuracy while using a relatively coarse simulation grid that is about 5 times the atomic cell. This leads to a huge advantage in computational efficiency. The scanning transmission electron microscopy (STEM) image of the tip geometry before and after evaporation is shown in Fig. 1(a), and the comparison between STEM image and simulation results is shown in Fig.1(b). Additionally level set simulations were used to investigate the dynamic tip shape evolution of oxide multilayer materials. Due to their interesting interfacial properties oxide multilayered structures have acquired a lot of attention in several scientific and industrial fields, such as magnetic storage media and microelectronics. The dynamic tip shape evolution of a composite material consisting of alternating layers of CeO2 and ZrO2 in two orientations (topdown and side-ways) was studied by the numerical simulations. In order to study how the sample geometry and electrical field changes affect the species density distribution on the detector, a trajectory simulation model by Runge-Kutta method is integrated into the level set solid-vacuum interface tracking model. Because the grid size of the level set method is around 5 times the atomic volume, a local grid refinement method is used for initialization of the starting positions of evaporated atoms for the trajectory calculation. Fig. 2 shows the simulation results for the CeO2 and ZrO2 horizontal multilayer sample evaporation. Fig. 2(a) and (b) show the tip geometry and electric field evolution during evaporation after 0 and 5000 atoms are evaporated respectively. Fig. 2(c) shows the end position of atoms flying over the counter electrode, and Fig. 2(d) shows the atom density distribution on the selected region as shown on Fig. 2(c). Because the evaporation strength of ZrO2 is about double that of CeO2, CeO2 is evaporated much faster and easier than ZrO2, which causes concaves for CeO2 on the tip surface. The concaves on the tip surface lead to higher density of electrical field lines, and make the trajectories of Ce closer to each other. This is consistent with the narrower band of Ce and wider band of Zr in the density distribution as shown in Fig. 2(d), which is fairly well matched to the experiment observations and measurements. Similarly, Fig. 3 shows the simulation results for the CeO2 and ZrO2 vertical multilayer sample evaporation. Reference Microsc. Microanal. 21 (Suppl 3), 2015 842 2. Vurpillot, F.; Bostel, A.; Blavette, D., Trajectory overlaps and local magnification in threedimensional atom probe. Applied Physics Letters 2000, 76 (21), 3127-3129. 3. Cerezo, A.; Godfrey, T. J.; Smith, G. D. W., Application of a positionsensitive detector to atom probe microanalysis. Review of Scientific Instruments 1988, 59 (6), 862-866. 4. Blavette, D.; Bostel, A.; Sarrau, J. M.; Deconihout, B.; Menand, A., An atom probe for threedimensional tomography. Nature 1993, 363, 432-435. 5. Blavette, D.; D�conihout, B.; Chambreland, S.; Bostel, A., Three-dimensional imaging of chemical order with the tomographic atom-probe. Ultramicroscopy 1998, 70 (3), 115-124. 6. Xu, Z.; Li, D.; Xu, W.; Devaraj, A.; Colby, R.; Thevuthasan, S., Simulation of Heterogeneous atom probe tip shapes evoltion during field evaporation using a level set method and different evaporation models. Computer Physics Communications 2015, In press. 7. This work was funded as a part of laboratory directed research and development as a part of chemical imaging initiative in pacific northwest national laboratory. RC acknowledge funding from EMSL distingusihed Wiley Postdoctoral Fellowship. (a) (b) Fig. 1: (a) scanning transmission electron microscopy (STEM) side view of the tested specimen before and after evaporation, (b) Comparison between the simulation results and the observation of the experiment. Fig. 2: Simulation results for the CeO2 and ZrO2 horizontal multilayer sample evaporation; (a) and (b): the tip geometry and electric field evolution during evaporation after 0 and 5000 atoms are evaporated respectively; (c): end position of atoms flying over counter electrode; (d): atoms density distribution on the selected region as shown on (c). Fig. 3: Simulation results for the CeO2 and ZrO2 vertical multilayer sample evaporation; (a) and (b): the tip geometry and electric field evolution during evaporation after 0 and 5000 atoms are evaporated respectively; (c): end position of atoms flying over counter electrode; (d): atoms density distribution on the selected region as shown on (c).");sQ1[422]=new Array("../7337/0843.pdf","Atom Probe Tomography Analysis of Bulk Chemistry in Mineral Standards","","843 doi:10.1017/S1431927615005012 Paper No. 0422 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atom Probe Tomography Analysis of Bulk Chemistry in Mineral Standards F. A. Laiginhas1, A. Perez-Huerta1, R. L. Martens2, T. J. Prosa3, D. Reinhard3 1. 2. Dept. of Geological Sciences, University of Alabama, Tuscaloosa, AL 35487 USA Central Analytical Facility, University of Alabama, Tuscaloosa, AL 35487 USA 3. CAMECA Instruments, Inc., Madison, WI, 53711 USA Atom Probe Tomography (APT) analysis of materials is an established technique for atomic level compositional analysis. Extensive research has been performed on many alloys, compounds, multilayered thin films, integrated circuits, and even polymers [1]. While some geological materials, mainly minerals and biominerals, have been previously studied [2-4], the use of APT for the resolution of bulk chemistry of minerals and its relationship to stoichiometry are insufficiently constrained. Often, the mass spectra from natural geological samples exhibit many complex ionic species, and their interpretation can lead to differing determinations of composition. Utilizing the lift-out method for APT specimen preparation [5] (Fig. 1), APT analysis was performed on mineral standards of magnetite and pyrite with known elemental compositions [Magnetite (Fe3O4): Fe = 71.88%, O = 27.53%; Mn = 0.18% by mass; Pyrite (FeS2): S = 53.45%, Fe = 46.55% by mass], to correlate to APT. Results indicate that the APT bulk atomic composition for magnetite underestimates the content of oxygen relative to iron based on the chemical formula, while overestimating the oxygen content based on the elemental composition of the standard (Fig. 2). For pyrite, APT results match the composition of sulfur and iron based on the molecular formula, but underestimate the composition of iron related to the measured elemental content in the standard (Fig. 2). Based on these findings, two preliminary conclusions are presented: 1) For sulfides, and minerals with no oxygen, the atom probe is giving atomic proportions according to the molecular formula; and 2) For oxides, the atom probe underestimates the proportion of oxygen, likely due to complex interactions with cations, relative to the molecular formula. Overall, our preliminary results indicate the need of further exploring the use of APT to obtain bulk chemistry for mineral samples, while providing basic operational procedures for sample preparation and standard development to be applied to more complex geologic samples in the future. References: [1] T.F. Kelly and M.K. Miller, Rev. Sci. Instrum. 78 (2007) p. 031101. [2] K.H. Kuhlman et al., Ultramicroscopy 89 (2001) p. 169. [3] J.W. Valley et al., Nature Geoscience 7 (2014) p. 219. [4] L. Gordon and D. Joester, Nature 469 (2011) p. 194. [5] D. Lawrence et al., Microscopy and Microanalysis 12 (2006) p. 1742. Acknowledgments: This research was funded by the United States National Science Foundation (EAR-1402912) granted to APH. Microsc. Microanal. 21 (Suppl 3), 2015 844 Figure 1. Example of focus ion beam (FIB) mineral sample tips, magnetite (left) and pyrite (right) for atom probe (APT) analysis. Figure 2. Example of APT mass spectra of magnetite (top) and pyrite (bottom) including the results for the bulk atomic composition.");sQ1[423]=new Array("../7337/0845.pdf","Atom Probe Tomography Characterization of Engineered Oxide Multilayered Structures","","845 doi:10.1017/S1431927615005024 Paper No. 0423 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atom Probe Tomography Characterization of Engineered Oxide Multilayered Structures M. I. Nandasiri1, N. Madaan1, A. Devaraj1, J. Bao2, Z. Xu2, T. Varga1, V. Shutthanandan1, and S. Thevuthasan3 [1] Environmental and Molecular Sciences Laboratory, Pacific Northwest National Laboratory, Richland WA, 99354 [2] Computational Material Sciences, Pacific Northwest National Laboratory, Richland WA, 99354 [3] Qatar Environment and Energy Research Institute, Qatar Foundation, Doha, Qatar The high temperature operation of solid oxide fuel cells (SOFC) is one of the main challenges we have to overcome, especially for commercializing SOFC for portable power generating applications [1]. Solid state electrolytes with enhanced oxygen ionic conductivity at low and intermediate temperatures are needed to lower the operating temperature of SOFC [2]. Thus, there is an ongoing need to develop new electrolytes or modify existing electrolytes to enhance the ionic conductivity. Recently, oxide multilayer hetero-structures with enhanced ionic conductivity stimulated a great interest as SOFC electrolytes [3]. Here, we investigate the influence of engineered nano-scale interfaces on the ionic conductivity of doped ceria and zirconia multilayer thin film electrolytes by utilizing state-of-the-art characterization techniques including atom probe tomography (APT). The multilayer thin films with alternative layers of samaria doped ceria (SDC) and scandia stabilized zirconia (ScSZ) were grown using oxygen plasma-assisted molecular beam epitaxy (OPA-MBE) to understand the effect of nano-scale interfaces on oxygen ionic conductivity through these films. The number of layers in the SDC/ScSZ multilayer thin films was varied from 2 to 20 by keeping the total film thickness constant at 140 nm. Oxygen ionic conductivity measurements were carried out as a function of temperature on well characterized samples using four probe surface impedance spectroscopy. Although these measurements demonstrate significantly higher ionic conductivity in multilayer thin films in comparison to a single layer thin film or bulk polycrystalline materials, the mechanisms associated with the enhanced ionic conductivity through nano-scale interfaces is not well-understood. Elemental inter-diffusion and dopant segregation across multiple interfaces could play important roles in the oxygen ionic conductivity through these hetero-structures. As such, we carefully characterized the structural and chemical properties of the materials utilizing various bulk and surface sensitive capabilities including x-ray diffraction (XRD) and x-ray photoelectron spectroscopy (XPS). In particular, the interfaces in multilayer thin films were carefully characterized using APT to study the elemental distributions along with the elemental inter-diffusion and dopant segregation at the interfaces. Laser assisted APT can provide quantitative three-dimensional chemical analysis of dielectric materials with lateral and depth resolutions in the order of 0.2-0.3 nm and chemical sensitivity up to parts-per-million levels with field-of-view on the order of 100 � 100 � 100 nm3 [4, 5]. Although conventionally APT has been extensively used to characterize metals and alloys, it is comparatively in its infancy in characterizing oxides and insulators especially composites consisting of heterogeneous structure [6]. Oxide multilayer structures add additional Microsc. Microanal. 21 (Suppl 3), 2015 846 complications to the characterization of the doped ceria/zirconia multilayers. We have used APT to map the elemental distribution in these films and interfaces and Fig.1 shows 3-D reconstruction of APT experiments carried out parallel to the interfaces and perpendicular to the interfaces along with the XPS depth profiling of these multilayer thin films. Fig. 1: Three dimensional reconstructions of APT data (Red ions correspond to CeO2 layer and blue ions correspond to ZrO2 layer) along with the XPS depth profiles of SDC/ScSZ multilayer thin films. The APT reconstructions demonstrate asymmetric tip shape evolution and artifacts associated with trajectory aberrations and we carried out level set model calculations to explain these artifacts [7]. The asymmetric evolving shape predicted by the level set model will be qualitatively compared with the experimental data collected by aligning the hetero-structured interfaces perpendicular and parallel to the tip axis. [1] Singhal, S.C., Solid oxide fuel cells for stationary, mobile, and military applications. Solid State Ionics, 2002. 152-153: 405-410. [2] Wachsman, E.D. and K.T. Lee, Lowering the Temperature of Solid Oxide Fuel Cells. Science, 2011. 334(6058): 935-939. [3] Emiliana, F., P. Daniele, and T. Enrico, Ionic conductivity in oxide heterostructures: the role of interfaces. Science and Technology of Advanced Materials, 2010. 11(5): 054503. [4] Kelly, T. F., Larson, D. J., The second revolution in atom probe tomography. MRS Bulletin, 2012. 37 (2): 150-158. [5] A. Devaraj, R. Colby, W. P. Hess, D. E. Perea, S. Thevuthasan, The Role of Photoexcitation and Field Ionization in the Measurement of Accurate Oxide Stoichiometry by Laser Assisted Atom Probe Tomography. Journal of Physical Chemistry Letters, 2013. 4(6): 993-998. [6] A. Devaraj, R. Colby, F. Vurpillot, S. Thevuthasan, Understanding Atom Probe Tomography of Oxide-Supported Metal Nanoparticles by Correlation with Atomic-Resolution Electron Microscopy and Field Evaporation Simulation. Journal of Physical Chemistry Letters, 2014. 5: 1361-1367. [7] Z. Xu, D. Li, W. Xu, A. Devaraj, R. Colby, S. Thevuthasan, B. P. Geiser, D. J. Larson, Simulation of heterogeneous atom probe tip shape evolution during field evaporation using a level set method and different evaporation models. Computer Physics Communications, 2015. 189: 106-113.");sQ1[424]=new Array("../7337/0847.pdf","Creating Isoconcentration Surfaces in Low-Chemical-Partitioning, High Solute-Containing Alloys","","847 doi:10.1017/S1431927615005036 Paper No. 0424 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Creating Isoconcentration Surfaces in Low-Chemical-Partitioning, High Solute-Containing Alloys B.C. Hornbuckle1, M. Kapoor1, and G.B. Thompson1 1 Department of Metallurgical & Materials Engineering, The University of Alabama, Tuscaloosa, AL 35487 USA The proximity histogram or proxigram has become a prevalent means of quantifying chemical partitioning in atom probe data sets. The proxigram is generated by the creation of an isoconcentration or isodensity surface around a cluster or precipitate from which the composition normal to the interface (both into and out of the surface) is measured [1]. The advantage of the proxigram is the ability to obtain compositional data normal to this surface even if the surface is anisotropic or has a varying radius of curvature. These surfaces also provide clear visual representation of precipitate features that can allow direct comparison to other microscopy techniques, such as transmission electron microscopy. In high chemical partitioning systems, the atom map is usually sufficient for the user to identify the element that should be used to create the isosurface. In cases where chemical partitioning is weak in the atom map, selection of which species and its corresponding isoconcentration value can be difficult to determine, see Figure 1. For low solute systems (<~10 at.%), various clustering algorithms can be used to identify the partitioned species from which the proxigram can be created. Unfortunately, for high-solute alloys, the clustering algorithms are not as robust because the probability that the first, second and so forth nearest neighbor is the solute atom itself is quite high making statistical distinction for cluster identification difficult. This poster presents a methodology to identify the correct species and isoconcentration value when low chemical partitioning and a high solute content exists in an alloy. We have used the 50.3Ni-29.7Ti-20Hf (at.%) alloy as our case study. Upon solutionizing, water quenching and aging, the alloy precipitates the H-phase from a B2 matrix [2]. These precipitates are visible in the (S)TEM-HAADF image, Figure 1(a), but the atom map, Figure 1(b), does not delineate the precipitate and matrix based on any of the chemical species' partitioning behavior. To identify the H-phase precipitates, the following procedure was developed and is general in its application. (1) Select the bulk composition as the starting values for each species' isoconcentration value. This will likely be incorrect as it does not account for partitioning between the matrix and precipitate but provides for a common reference. (2) Based on the definition of the proxigram, the user-defined isoconcentration value must coincide with the composition at the interface (0 on the x-axis) of the proxigram. Through user-defined iterations of isoconcentration values, these isoconcentration values will eventually converges to the outputted isoconcentration value at the interface, Figure 2. From those proxigrams, the specific composition for each respective species within each phase can now be extracted, i.e. the Ni composition is found from the created Ni isoconcentration proxigram, the Ti composition from the Ti isoconcentration created proxigram, and so forth. (3) Using those values for the phase compositions, a new isoconcentration surface is generated, Figure 3, that delineates the phases. (4) By placing a 1D profile through each phase, one can determine which species enrich each phase. In our case study, one must select either Ni, Ti, or Hf as the final rendered proxigram. The proxigram that shows agreement with the 1D profile identifies this correct species, which in this example was Ti. Microsc. Microanal. 21 (Suppl 3), 2015 848 References: [1] O.C. Hellman, et al. Microscopy and Microanalysis 6 (2000) 437. [2] F. Yang, et al. Acta Materialia 61 (2013) 3335. [3] The authors gratefully acknowledge funding from NASA grant NNX09AO61A Figure 1: (a) STEM-HAADF of Hphase precipitates in a B2 matrix. (b) Atom map of an equivalent alloy visually showing no obvious partitioning. Figure 2: (a) Ni isoconcentration surface and proxigram at 48.39 at.%. (b) Ti isoconcentration surface and proxigram at 28.24 at.%. (c) Hf isoconcentration surface and proxigram at 18.71 at.%. The circle denotes the isoconcentration value at the interface for that species and is in agreement with the userselected isoconcentration value used to create the corresponding surface. Unfortunately, these isosurfaces do not clearly represent the visual delineation of the precipitate and matrix noted in Figure 1(a). Figure 3 describes the creation of the correct isoconcentration species and value to generate the correct rendering. Figure 3: The isoconcentration surfaces whose value matches the precipitate compositions determined in the proximgrams of Figure 2 for each species. (a) Ni isoconcentration surface and proxigram at 46.27 at% (b) Ti isoconcentration surface and proxigram at 23.44 at.%. (c) Hf isoconcentration surface and proxigram at 22.69 at.%. (d) 1D profile which shows agreement with the Ti isosurface and proximity which indicated that Ti was the correct species and isoconcentration value to use for reporting final phase compositions.");sQ1[425]=new Array("../7337/0849.pdf","Experimental Evaluation of Conditions Affecting Specimen Survivability in Atom Probe Tomography","","849 doi:10.1017/S1431927615005048 Paper No. 0425 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Experimental Evaluation of Conditions Affecting Specimen Survivability in Atom Probe Tomography T.J. Prosa1, D. Lawrence1, D. Olson1, S. Strennen1, I. Martin1, D.J. Larson1, R.L. Martens2, J. Goodwin2, A. Portavoce3 and D. Mangelinck3 1. 2. CAMECA Instruments, Inc., Madison, WI, USA. Metallurgical and Materials Engineering, University of Alabama, Tuscaloosa, AL, USA. 3. Aix-Marseille Universit�, IM2NP, F-13397 Marseille Cedex, France. It is a miraculous quirk of nature that any material can survive application of the extreme electric fields necessary to extract atoms from a specimen surface one-at-a-time without bulk rupture of the material itself. The field of atom probe tomography (APT) relies upon this natural quirk and continues to find an ever widening variety of materials can be successfully investigated [1], yet the issue of premature specimen failure remains critical for continued field growth and adoption [2]. Since the primary goal of an ordinary APT analysis is to gather compositional information related to a materials science problem, improving the success rate of the investigation is a secondary luxury when a single, high quality dataset is sufficient for the analysis need. In the current study, analysis yield improvement is the primary purpose of the investigation. We report on our results to date relating how analysis conditions affect yield for our standard specimen. For this study we chose a material that has many material characteristics common to the microelectronics industry but also with a history of low (but non-zero) yield in our laboratory. As illustrated in Figure 1, the material consists of a 12 nm oxide grown on a Si <100> substrate with an additional 100 nm of phosphorous doped poly-silicon was grown and implanted with boron [3,4]. Historically, the analysis yield of the doped and implanted poly-silicon region was relative high while that of the oxide was extremely low. Our goal was to statistically analyze the analysis yield for this specimen type and then consider how various variables, both specimen preparation and analysis conditions, effect yield. The initial considerations included specimen size, as measured by the tip diameter at the oxide, cap size, detection rate (DR), and laser pulse energy. For the initial investigation, some 66 specimens were manufactured with a similar geometry and a variety of oxide dimensions to look for correlation between analysis yield and specimen size. In Figure 1, the protective nickel cap is visible at the specimen apex with the bright 12 nm oxide observable 100 nm below. A narrow shank angle geometry was chosen due to its reproducibility. This allowed for predictable analysis evolution near and through the primary region of interest (the 12 nm SiO2). Based on these 66 analysis attempts, a number of statistically significant conclusions were drawn for this sample type: first, yield through the poly-silicon is very high. In fact, additional data not reported here puts the overall yield for this region near 95%. Specimens rarely fail here regardless of analysis conditions. Second, DR does affect yield through the oxide. Lowering the DR from 0.3% to 0.1% increases yield from 24% to 82% which is statistically significant at better than a 95% confidence level. Third, specimens fail much more often through the low-to-high field Si/SiO2 interface than the high-to-low field SiO2/Si interface. In fact, every specimen that survived through the top high-to-low interface continued analysis into the substrate. Fourth, the physical size of the oxide interface does not affect yield. This is illustrated in Figure 2. Microsc. Microanal. 21 (Suppl 3), 2015 850 References: [1] E.A. Marquis et al., Current Opinion in Solid State and Materials Science 17 (2103), p. 217. [2] D.G. Brandon, Field Ion Microscopy, Eds. J. Hren, S. Ranganathan (Plenum, 1968) p. 64. [3] A. Portavoce et al., Defect and Diffusion Forum 289-292 (2009), p. 329. [4] A. Portavoce et al., Diffusion and Defect Data 264 (2007), p. 33. Table 1. Analysis Yield through Various Regions of the Standard Structure Region 1) Nickel Cap/Poly Si Interface 2) Poly Si 3) Si/SiOx Interface DR = 0.1% Si/SiOx Interface DR = 0.3% 4) SiOx/Si Interface aTotal Successes 43 42 14 6 20 Attemptsa,b 61 48 17 25 20 Region Yield 70% 88% 82% 24% 100% Number of Specimens = 66 from Ni Cap Evaluation (too few ions) = 5 bExcluded Ni Cap N i Implanted Si 100 nm SiO2 12 nm N i SiO S i N i S i B Si 100 nm S i 5 nm SiO2 20 nm Figure 1. A description of the standard structure (far left), SEM image of a typical APT specimen before analysis, TEM of the low yielding SiO2 region of the structure, and atom map revealing the phosphorus doping and boron implant (far right). 220 0.1% Success 0.1% Fail 0.3% Success 0.3% Fail SiO2 Diameter (nm) 200 180 160 140 120 100 80 60 Figure 2. Ordered plot of successful analysis yield through the SiO2 as a function of specimen diameter. Acquisition was attempted with ion detection rates of either 0.3% (circles) or 0.1% (squares).");sQ1[426]=new Array("../7337/0851.pdf","Atom Probe Tomography of Zircon and Baddeleyite Geochronology Standards","","851 doi:10.1017/S143192761500505X Paper No. 0426 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atom Probe Tomography of Zircon and Baddeleyite Geochronology Standards D.A. Reinhard1, D.E. Moser2, I.R. Barker2, D. Olson1, I. Martin1, Katherine P. Rice1, Y. Chen1, D. Lawrence1, T.J. Prosa1 and D.J. Larson1 1. 2. CAMECA Instruments Inc., Madison, WI USA University of Western Ontario, London, Ontario, CAN N6A 5B7 Atom probe tomography (APT) makes it possible to study the compositional structure of geological materials at the nanoscale [1]. It is an analytical technique whereby single atoms (or small groups of atoms) are ionized under the presence of a large electric field and removed from the surface one-at-atime. Each removal (field evaporation event) is synchronized to a timing pulse and projected to a 2D position-sensitive detector providing ion identification through time-of-flight mass spectroscopy and, ultimately, a composition map of the surface [2]. Performed iteratively over many millions of ions, the evolving surface information is reconstructed as a 3D map of individual ions from which information can be extracted for a for a variety of real-space analyses. In general, optimal APT acquisition conditions are material dependent, thus experimentation is often necessary to achieve acceptable levels of data quality and analysis yield for an unfamiliar sample. As APT is a destructive technique, it may be advantageous to perform optimization on non-precious, standard materials prior to sacrificing the true material of interest. In the present work, we explore the effect of various laser-pulse energies and annealing conditions on zircon (BR266) [3] and baddeleyite (Phalaborwa) [4] standards and report progress to date. Multiple APT specimens of both standards were made using standard lift-out techniques [2]. For many of the successful analyses, post-acquisition high-resolution SEM imaging was used to describe final specimen shape and constrain the reconstructed 3D ion maps (see Figure 1) [2]. Measurements of the final specimen apex radius, as a function of voltage and laser pulse energy, provides an estimate of the average electric field at the time of evaporation. Over the range of attempted pulse energies (50-800 pJ), the evaporation field of zircon was observed to change by ~12%. Using a default k-factor of 3.3 [2], estimated evaporation fields ranged from ~25-28 V/nm. Baddeleyite was found to behave similarly. Because limited analysis yield is an issue [5], we explored post-specimen preparation treatments in hopes of improving the structural integrity. Annealing conditions ranging from 100-500 �C for 1-4 hours are believed gentle enough to leave the primary structures unchanged while relaxing some of the internal stresses which might lead to premature specimen failure. Figure 2 shows atom maps from analyzed annealed and as-received specimens. No difference is observed in the size and distribution of detected Pb-clusters between the annealed and un-annealed specimens. References [1] J.W. Valley et al., Nature Geoscience 7 (2014), p. 219. [2] D.J. Larson et al., "Local Electrode Atom Probe Tomography" (Springer, New York 2014). [3] R.A. Stern, Report 14, Geological Survey of Canada, Current Research, 2001-F. [4] L.M. Heaman, Chemical Geology 261 (2009), p. 43. [5] D.J. Larson et al., Microscopy and Microanalysis 20(S2) (2014), p. 2088. Microsc. Microanal. 21 (Suppl 3), 2015 852 Figure 1: High-resolution SEM imaging of a zircon specimen after acquisition. These images were used to estimate the final tip radius, the shank angle and the sphere-to-cone ratio to constrain the reconstruction to allow for an estimate of the evaporation field for future acquisitions. Note: charging of the insulator specimen can cause aberrations in the image and limit the accuracy of the radius measurement. Figure 2: Y and Pb ion maps from annealed and un-annealed specimens showing similar spatial features.");sQ1[427]=new Array("../7337/0853.pdf","Atom Probe Tomography of Feldspars and Aluminosilicate Glasses","","853 doi:10.1017/S1431927615005061 Paper No. 0427 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atom Probe Tomography of Feldspars and Aluminosilicate Glasses Lyle M. Gordon Environmental Molecular Spectroscopy Laboratory, Pacific Northwest National Laboratory, 3335 Innovation Blvd. Richland, WA 99354 Feldspars are a group of alumniosilicate minerals that make up as much as 60% of the Earth's crust. The most common feldspars can be expressed in terms of three endmembers (Figure 1), specifically, potassium feldspar (Orthoclase), sodium feldspar (Albite) and Calcium feldspar (Anorthite). At high temperatures, solid solutions exist between the distinct endmembers; however, during cooling miscibility gaps result in phase separation (exsolution) of the solid solution into multiple stable phases. The phase separation and the formation of the resulting microstructure could occur either by nucleation and growth of a second phase or by spinodal decomposition, both followed by coarsening. Identifying the phase transformation and microstructure formation mechanisms and determining the geological history of these materials requires detailed analysis of the microstructure and quantitative compositional measurements of the exsolved phases. To characterize the fine length scale of the microstructure and limited degree of chemical variation simultaneously, an analytical tool with high spatial resolution and chemical sensitivity is required. Atom-probe tomography (APT) is uniquely capable of characterizing the nanoscale chemistry of the exsolution microstructure. APT is an established technique in metallurgical and semiconductor research, however, with the recent advent of ultra-violet laser pulsing analysis of a range of minerals has become possible, however, only a few have been analyzed to date including iron oxides [1], apatites [2], and olivines [3]. Initial results from atom probe tomographic investigation into the structure and chemistry of natural and synthetic alkali and plagioclase feldspars of varied compositions and microstructures will be presented. Both stable endmembers and unstable compositions within the miscibility gap have been analyzed (Figure 1 & 2). APT mass-to-charge ratio spectra (Figure 2) demonstrated high spectral resolution, m/m, of greater than 1000 at full width-half max of the Si2+ peak. Runs with greater than 107 ions were routinely collected with reliable detection of trace ions below 100 ppm. Analytical results under a range of experimental conditions to determine the ideal operating parameters for APT of feldspars will be presented. Further, comparison of data from crystalline feldspars minerals to compositionally similar glassy materials will be conducted to identify the influence of crystallinity on APT. [1] Gordon L. M. & Joester D. Nanoscale chemical tomography of buried organic-inorganic interfaces in the chiton tooth. Nature 469 (2011) p .194-197. [2] Gordon L.M., Tran L., & Joester D. Atom probe tomography of apatites and bone-type mineralized tissues. ACS nano 6 (2012) p. 10667-10675. [3] Arey, B. et al. Atom Probe and TEM Investigation of Natural Olivines. Microscopy and Microanalysis. 18 S2 (2012) p. 658-659. [4] The research was performed using EMSL, a DOE Office of Science User Facility sponsored by the Office of Biological and Environmental Research and located at Pacific Northwest National Laboratory. Microsc. Microanal. 21 (Suppl 3), 2015 854 Figure 1. Ternary phase diagram of feldspar. Endmember and solids-solution (not necessarily stable) are labeled with mineralogical names. Composition and locality of analyzed samples (Albite, yellow circle and Bytownite, red circle) are indicated on the diagram. Figure 2. Labeled atom probe mass-to-charge state ratio spectra for representative samples of Amelia Albite and Stillwater Bytownite, corresponding to the points indicated in Figure 1.");sQ1[428]=new Array("../7337/0855.pdf","Catalysis and Atom Probe Tomography: Recent Progresses and Future Developments towards the Analysis of Nanoporous Samples","","855 doi:10.1017/S1431927615005073 Paper No. 0428 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Catalysis and Atom Probe Tomography: Recent Progresses and Future Developments towards the Analysis of Nanoporous Samples C�dric Barroo1, Andrew P. Magyar2 and David C. Bell1,2 1. 2. School of Engineering and Applied Sciences, Harvard University, Cambridge MA USA Center for Nanoscale Systems, Harvard University, Cambridge MA USA Since the infancy of microscopy and surface science, many advanced techniques have been utilized to better understand catalytic systems. Previous studies focus either on the catalytic reaction (reactivity and selectivity) or on the catalyst itself, focusing on structural/morphological reconstructions as well as changes in the chemical composition of the catalyst. Atom probe tomography (APT) is a powerful tool for the characterization of both the three-dimensional structure and composition of catalysts at the atomic-scale. Alloys catalysts are of particular interest in applied catalysis because the synergistic effects between the different metals can drastically influence their activity as compared to their constituting pure metals [1]. At the microscopic scale, the surface composition of the very first atomic layers can differ from that of the bulk. The surface composition can be modified via physicochemical treatments, yielding different catalytic behaviours, which are highly dependent on the composition of the alloy, the temperature and duration of treatments, the crystallographic orientation, and the presence of surface or bulk defects. Among others, surface segregation/depletion, core-shell structure and formation of nano-islands can be studied by APT and can inform the nanoscale engineering of catalyst surface compositions (see [2] and references therein). For such studies, a needled shape catalyst is used, which can then be directly measured via APT. More recently, developments have been made regarding the sample preparation allowing the analysis of more realistic catalysts by atom probe tomography. Conventional tip samples used in APT can be used as a support for nanoparticles of catalysts using an electrophoresis method [3] and this method has been successfully used for the analysis of Pt, Pt-Co, Ir@Pt and Ag@Pd nanoparticles [1,3]. Eventually, embedding methods and FIB tip preparation can be used to study well-dispersed nanoparticles [4]. Current research focuses on the use of nanostructured materials, such as powders and nanoporous catalysts, for energy applications. As an example, nanoporous gold catalysts (npAu) exhibit a high activity and selectivity for alcohol oxidation reactions, which has been attributed to the presence of traces of silver, originating from the sample preparation. The systematic study of such a catalyst requires a control of its microstructure. However, the very morphology of this class of catalysts and the presence of pores make the APT analysis and the reconstruction impossible via traditional techniques. The sample preparation thus requires new developments. A glass micro-encapsulation technique has also been developed to study nanoparticles and powders by APT [5] and provides a rapid, robust, and inexpensive way to analyze nano- or micro-particles using LEAP, however the amorphous nature of the glass can also present challenges for this approach. The present study reports the analysis of nanoporous gold with traces of silver. This has been made possible by the combination of SEM, TEM and APT analysis. To study those catalysts, different methodologies are used to fill the pores and then allow the analysis by atom probe: atomic layer deposition (ALD) (an example of APT analysis of Al-doped ZnO (AZO) film deposited by ALD is presented Fig1a), e-beam evaporation, and in situ deposition in the FIB, prior to the regular lift-out method. The npAu samples are studied at different stages of the catalytic reaction to trace the evolution Microsc. Microanal. 21 (Suppl 3), 2015 856 of the catalyst's structure from its initial state through its final form. The stages corresponds to 1) the initial catalyst prior to any treatment, 2) the catalyst after ozone pre-treatment, 3) the sample right after the first moment of the reaction, and 4) the active catalyst, namely the catalyst after establishment of a steady state regime of conversion. The combined approach can allow correlation of the nano-structure of a nanoporous gold matrix, derived from TEM with the 3D nano-structure of the same system obtained from APT both before and after exposure to reaction conditions. These results directly indicate a coarsening of the nanogold matrix due to the reaction event (Fig1b). The behavior of the silver atoms within the gold backbone is also studied in details, and the formation of Au-Ag alloy is supposed to be responsible of the activity of the catalyst. [6] References: [1] K. Tedsree, et al., Nature Nanotech. 6 (2011) 302 [2] C. Barroo, P. A. J. Bagot, G. D. W. Smith, T. Visart de Bocarm�, "Investigating Nano-structured Catalysts at the Atomic scale by Field Ion Microscopy and Atom Probe Tomography", in AtomicallyPrecise Methods for Synthesis of Solid Catalysts; RSC Catalysis Series, 2015, pp. 248-295 [3] T. Li, et al., ACS Catal. 4 (2014) 695 [4] P. Felfer, et al., Angew. Chem. 126 (2014) 11372 [5] D. C. Bell, et al., Microscopy and Microanalysis, 19 [S2] (2013) 954 [6] This work was supported as part of the Integrated Mesoscale Architectures for Sustainable Catalysis - IMASC, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences under Award # DE-SC0012573. C.B. acknowledges postdoctoral fellowships through the Belgian American Educational Foundation (BAEF) as well as WallonieBruxelles International (Excellence grant WBI.WORLD) foundations. Figure 1. a) 3D reconstruction of Al-doped ZnO (AZO) deposited by atomic layer deposition on a presharped APT (red balls corresponds to Al atoms, and grey to Zn atoms) � b) SEM evidence of coarsening of npAu after reaction (scale bar: 400 nm).");sQ1[429]=new Array("../7337/0857.pdf","Background Recovery through the Quantification of Delayed Evaporation Multi-Ion Events in Atom-Probe Data","","857 doi:10.1017/S1431927615005085 Paper No. 0429 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Background Recovery through the Quantification of Delayed Evaporation MultiIon Events in Atom-Probe Data Karen Kruska1, Daniel K Schreiber2 1 Fundamental & Computational Sciences Directorate, Pacific Northwest National Laboratory, Richland, WA, USA 2 Energy & Environment Directorate, Pacific Northwest National Laboratory, Richland, WA, USA The reliable quantitative analysis of oxides by atom probe tomography (APT) is an on-going issue. Analysis conditions have been shown to dramatically affect the measured composition, while rarely achieving a completely satisfactory measurement. In most cases, the optimal analysis conditions have been identified as a combination of low laser energy and high electric field. However, these analysis conditions also result in a relatively high probability for multi-ion evaporation within a single pulse, which can degrade the overall analysis and possibly introduce an unintentional compositional bias. Saxey proposed graphical methods for interpreting multi-ion evaporation events (multihits) through correlation histograms [1,2]. An example of these so-called "Saxey plots" is shown in Figure 1 for NiO (LEAP 4000XHR, =355 nm, T=40K, 2-50 pJ/pulse laser energy range) in addition to conventional mass spectra for these same data. Color coding identifies ordinarily ranged (red) and unranged (green) portions of the data. Note that the data have been filtered to events with exactly two detected ions/pulse to maximize the S/N ratio in this representation. When the mass spectra is filtered to only unranged ions in Figure 1c, what remains is unidentifiable background that is typically discarded. However, when plotted on the correlation histogram these same data are readily identifiable as correlated, delayed evaporation events by the diagonal green lines extending from the lower-left to upper right of Figure 1a. Identification of these background ions is enabled by leveraging the shared delay in evaporation (t) of the correlated ion pair, as described by Saxey[1]. This is accomplished by plotting the mass-to-charge state ratio (m/n) of one ion versus the difference in the square roots of the two m/n ratios and creating a histogram, as shown in Figure 2a for the same NiO data. Identification and quantification of the various peaks follows, enabling reliable recovery of >80 % of the "background" present in the original unranged portion of the multihit mass spectrum. Similar data are also displayed for Fe3O4 in Figure 2b. The results of these quantifications are summarized in Table 1. Given the small percentage of unranged ions that are part of an event consisting of exactly two detected ions, it is unsurprising that the overall impact on the measured composition is minimal. However, careful implementation of these new analysis methods could provide novel insights into the origins of non-stoichiometric oxide measurements by APT and more broadly explore the role of background and multihit events in systematic composition errors for other materials that exhibit a propensity for correlated evaporation. References: [1] Saxey, D W, Ultramicroscopy 111, (2011) p. 473-479 [2] Santhanagopalan, D et al., Ultramicroscopy 148, (2015) p. 57-66. Microsc. Microanal. 21 (Suppl 3), 2015 858 Figure 1. (a) Correlation histogram and (b)-(c) corresponding mass spectra for NiO wherein the data is filtered to only display data from exactly 2 detected ions in a given pulse. Color coding indicates ions that were identified by conventional ranging (red) and ion pairs that were not identifiable (green). Figure 2. Recovered mass spectra from the unranged portions of NiO and Fe3O4 multihit data. O Concentration As-Ranged 47.5 � 0.1 at.% 51.7�0.1 at.% O Concentration Unranged Multihits 57.4 � 0.1 at.% 49.6�0.1 at.% Corrected O Concentration 47.8 � 0.1 at.% 51.7 � 0.1 at.% Recovered Background 82% 67% NiO Fe3O4 Table 1. O concentration using standard ranging and by adding unranged, identifiable multihit ions.");sQ1[430]=new Array("../7337/0859.pdf","Atom Probe Tomography Analysis of Bulk Chemistry in Mineral Standards","","859 doi:10.1017/S1431927615005097 Paper No. 0430 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atom Probe Tomography Analysis of Bulk Chemistry in Mineral Standards F. A. Laiginhas1, A. Perez-Huerta1, R. L. Martens2, T. J. Prosa3, D. Reinhard3 1. 2. Dept. of Geological Sciences, University of Alabama, Tuscaloosa, AL 35487 USA Central Analytical Facility, University of Alabama, Tuscaloosa, AL 35487 USA 3. CAMECA Instruments, Inc., Madison, WI, 53711 USA Atom Probe Tomography (APT) analysis of materials is an established technique for atomic level compositional analysis. Extensive research has been performed on many alloys, compounds, multilayered thin films, integrated circuits, and even polymers [1]. While some geological materials, mainly minerals and biominerals, have been previously studied [2-4], the use of APT for the resolution of bulk chemistry of minerals and its relationship to stoichiometry are insufficiently constrained. Often, the mass spectra from natural geological samples exhibit many complex ionic species, and their interpretation can lead to differing determinations of composition. Utilizing the lift-out method for APT specimen preparation [5] (Fig. 1), APT analysis was performed on mineral standards of magnetite and pyrite with known elemental compositions [Magnetite (Fe3O4): Fe = 71.88%, O = 27.53%; Mn = 0.18% by mass; Pyrite (FeS2): S = 53.45%, Fe = 46.55% by mass], to correlate to APT. Results indicate that the APT bulk atomic composition for magnetite underestimates the content of oxygen relative to iron based on the chemical formula, while overestimating the oxygen content based on the elemental composition of the standard (Fig. 2). For pyrite, APT results match the composition of sulfur and iron based on the molecular formula, but underestimate the composition of iron related to the measured elemental content in the standard (Fig. 2). Based on these findings, two preliminary conclusions are presented: 1) For sulfides, and minerals with no oxygen, the atom probe is giving atomic proportions according to the molecular formula; and 2) For oxides, the atom probe underestimates the proportion of oxygen, likely due to complex interactions with cations, relative to the molecular formula. Overall, our preliminary results indicate the need of further exploring the use of APT to obtain bulk chemistry for mineral samples, while providing basic operational procedures for sample preparation and standard development to be applied to more complex geologic samples in the future. References: [1] T.F. Kelly and M.K. Miller, Rev. Sci. Instrum. 78 (2007) p. 031101. [2] K.H. Kuhlman et al., Ultramicroscopy 89 (2001) p. 169. [3] J.W. Valley et al., Nature Geoscience 7 (2014) p. 219. [4] L. Gordon and D. Joester, Nature 469 (2011) p. 194. [5] D. Lawrence et al., Microscopy and Microanalysis 12 (2006) p. 1742. Acknowledgments: This research was funded by the United States National Science Foundation (EAR-1402912) granted to APH. Microsc. Microanal. 21 (Suppl 3), 2015 860 Figure 1. Example of focus ion beam (FIB) mineral sample tips, magnetite (left) and pyrite (right) for atom probe (APT) analysis. Figure 2. Example of APT mass spectra of magnetite (top) and pyrite (bottom) including the results for the bulk atomic composition.");sQ1[431]=new Array("../7337/0861.pdf","The Importance of Location, Location, Location: As True for Rhodococcus fascians as for Real Estate","","861 doi:10.1017/S1431927615005103 Paper No. 0431 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Importance of Location, Location, Location: As True for Rhodococcus fascians as for Real Estate T. Sawyer1 and M.L. Putnam2 1. 2. Electron Microscopy Facility, Oregon State University, Corvallis, OR, USA Botany and Plant Pathology, Oregon State University, Corvallis, OR, USA Rhodococcus fascians is a Gram-positive pleomorphic bacterial pathogen of plants which causes growth deformities known as leafy galls (Figure 1). The majority of plants affected are those produced for amenity use, hence grossly malformed plants, for which there is no cure, must be destroyed at the nursery, leading to losses up to and exceeding hundreds of thousands of dollars per nursery on an annual basis An epiphytic phase of the bacteria was first described in 1936 [1] by observation of inoculated pea seedlings thin-sectioned, stained, and examined under a compound light microscope. Much later, scanning electron microscopy [2] was used to determine where on the plant surface the bacteria were located, but the focus was on aerial structures. In neither study were bacteria observed in cell layers deep within the tissues or in the vascular system, nor is it known whether it is possible for R. fascians to systemically infect plants. Whether the bacteria can colonize roots is also unknown, although previous studies in our laboratory have shown that leafy galls form after inoculation of soil of pot-grown plants. It is important to understand whether R. fascians colonizes the vascular tissue or roots: if so, this would provide new insights into the infection process, epidemiology and management of disease caused by the bacteria. We are now examining the various organs and tissues of peas inoculated as seeds to determine the physical location of R. fascians cells within and on the plants. Phaseolus vulgaris seeds were semi-immersed in a suspension of 108 colony forming units (CFU) per ml of bacterial cells for one hour, then placed individually on a solid growth medium in glass test tubes and incubated under continuous light for two weeks, after which they were examined. Peas were inoculated with a pathogenic isolate, a non-pathogenic isolate, or sterile buffer. After the infection period a 5 mm sections roots and stems were removed from the seed for scanning electron microscopy processing. Individual parts were chemically fixed in a modified Karnovsky formulation over night at 4C. Specimens were dehydrated in a serial solution of acetone from 10% 100% in preparation for critical point drying. Samples were mounted on a stub, sputter coated and imaged with a Quanta 600 FEG scanning electron microscope, (FEI, Hillsboro, OR, USA). Cells of R. fascians were clearly visible within the cortical cells of a pea root that had emerged after inoculation (Figure 2). [1] M. Lacey, Ann. Appl. Biol. 23 (1936) pp. 743-751. [2] K. Cornelis et al, MPMI 14 (2001) pp. 599-608. Microsc. Microanal. 21 (Suppl 3), 2015 862 Figure 1. A leafy gall on a young Osteospermum plant. Bar = 1 cm. Figure 2. A cluster of R. fascians growing within the cortical cells of a pea root. A shows the location of the cells enlarged in B.");sQ1[432]=new Array("../7337/0863.pdf","Investigation into the Stabilization of Soil Organic Matter by Microbes","","863 doi:10.1017/S1431927615005115 Paper No. 0432 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Investigation into the Stabilization of Soil Organic Matter by Microbes Alice C. Dohnalkova1, Rosalie K. Chu1, Malak Tfaily1, Alex R. Crump2, William B. Chrisler2, Tamas Varga1, and Bruce W. Arey1 1 2 1,2 Environmental Molecular Sciences Laboratory, Fundamental and Computational Sciences Division Pacific Northwest National Laboratory, Richland, WA, USA A better understanding of below ground carbon (C) flux is of fundamental importance to predict how changing climate will influence the C balance of forest (and other) ecosystems [1]. The root system of higher plants is associated not only with soil environment composed of inorganic and organic matter, but also with a vast community of metabolically active microorganisms. Rhizosphere is the zone of soil immediately surrounding the plant roots, with the microbial population considerably higher than that of root free soil environment. Soil organic carbon pools are often defined either as labile or as recalcitrant, referring to its stability against decomposition of soil organic matter (SOM). We studied the microbial role in production and stabilization of SOM in laboratory setup of column-grown Pinus resinosa mesocosm systems [2], by (a) imaging by light and electron microscopy, with (b) high resolution chemical analysis by Fourier transform ion cyclotron resonance-mass spectroscopy (FTICR-MS), and (c) crystallographic X-ray analyses of the microbially-induced mineral weathering, to determine SOM resistance to decomposing activities. Our main area of interest was to characterize the stabilization of microbially-produced extracellular polymeric substances (EPS), a remarkable dynamic entity that plays critical functional role in a wide variety of geomicrobial processes in soil. EPS is primarily associated with physical adhesiveness, therefore with biofilm formation, cell adhesion to solid surfaces, with creation of protective microhabitats against adverse environmental conditions, and facilitating mineral aggregation. Additionally, due to its immense absorptive capacity, EPS is capable of binding, accumulating, and sequestering dissolved organic matter and metals from the environment and it is consequently able to influence a wide range of biogeochemical processes. These include the dissolution and precipitation of minerals, as well as redox and/or complexation reactions. Generally composed of microbially-secreted heterogeneous combinations of high-molecular-weight polysaccharides, lipids, phosphate, proteins, and nucleic acids, EPS mass can consist of up to 95% of bound and unbound water. Due to this extreme hydration, EPS is among the most difficult biological structures to preserve and characterize in its native state and presents a major challenge for obtaining accurate high-resolution images via electron microscopy [3]. As a field follow-up study with a goal to relate our laboratory tree columns system to a more natural setup, we implanted mesh bags filled with biotite as used in the designed soil in the previous lab mesocosms experiment [2, 4]. The bags were placed in the rhizospheric zone of Pinus ponderosa young seedlings in Wenatchee national forest, following the experimental setup [5]. The material was allowed to incubate with the natural inoculum of microbial community present in the soil over period of 6 months. After that, the bags were removed, and the material was imaged and analyzed for microbial presence and specific associations with minerals, characterized for organic carbon compounds, for microbially-induced mineral weathering, and for microbial community genetic signatures. These results were then related for recurring features in both the laboratory column experiments and the field study. Microsc. Microanal. 21 (Suppl 3), 2015 864 The above scheme of imaging and analytical techniques will serve to characterize the rhizosphere interactions resulting in the environmental processes such as soil organic matter persistence in ecosystems, relevant to carbon sequestration, and its stability to conditions applicable to responses to global change. References: [1] Litton, C.M., Giardina, C.P., 2008. "Below-ground carbon flux and partitioning: global patterns and response to temperature". Functional Ecology 22, 941-954. [2] Dohnalkova A, et al., 2014. "Correlative Imaging and Analyses of Soil Organic Matter in the Rhizosphere". Microscopy and Microanalysis, (20), S3, 1192�1193 [3] Dohnalkova A, et al., 2011. "Imaging Hydrated Microbial Extracellular Polymers: Comparative Analysis by Electron Microscopy." Applied and Environmental Microbiology 77(4):1254-1262. [4] Balogh-Brunstad, Z., C.K. Keller, R.A. Gill, B.T. Bormann, and C.Y. Li, 2008b. "The effect of bacteria and fungi on chemical weathering and chemical denudation fluxes in pine growth experiments." Biogoechem. doi: 10.1007/s/10533-008-9202. [5] Wallander H. et al., 2013. "Evaluation of methods to estimate production, biomass and turnover of ectomycorrhizal mycelium in forests soils e A review" Soil Biology & Biochemistry 57, 1034e1047 [6] This research was performed at the Environmental Molecular Sciences Laboratory (EMSL), a national scientific sponsored by the Department of Energy's Office of Biological and Environmental Research, located at PNNL. Figure 1. A - Mesh bag field experimental setup for microbial inoculation of soil minerals in ponderosa pine ecosystem. B � detail of the 200 um nylon mesh bag containing biotite minerals. C � SEM image of resulting microbial EPS deposited on biotite. The intricate network of stabilized residual carbohydrates-based polymers maintain their integrity even after solvent extraction and dehydration.");sQ1[433]=new Array("../7337/0865.pdf","The Development and Structure of Cornalean Flowers and Fruits.","","865 doi:10.1017/S1431927615005127 Paper No. 0433 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Development and Structure of Cornalean Flowers and Fruits. Brian A. Atkinson1, Teresa Sawyer2, Gar W. Rothwell1,3, and Ruth A. Stockey1 1. 2. Department of Botany and Plant Pathology, Oregon State University, Corvallis, Oregon, USA. Electron Microscope Facility, Oregon State University, Corvallis, Oregon, USA. 3. Department of Environmental and Plant Biology, Ohio University, Athens, Ohio, USA. The order Cornales (dogwood order) is the earliest diverging lineage within the most diverse group of flowering plants the asterids, which number over 80,000 living species [1]. This order contains an impressive diversity of flower and fruit morphologies, which makes Cornales an important resource for understanding the early evolutionary diversification of asterids. Fortunately, the Cornales is exceptional in having an excellent fossil record dating back to the Late Cretaceous (100-66 Ma), thus providing a rich source of crucial data for a wide range of evolutionary studies [2, 3]. Recent investigations have generated several hypotheses of cornalean evolution and systematics that remain untested by comparative morphological and anatomical studies, and fossils provide critical data both for testing those hypotheses and for reconstructing the early evolutionary history of the group [3]. Due to the systematic importance of the Cornales, the paleobotany lab at Oregon State University has begun an in-depth initiative to study the early evolution of this order. Paleobotancial studies of Cornales have already begun in which Late Cretaceous cornalean flowers and fruits are being characterized. However, in order to develop a broader understanding of cornalean reproductive biology, it is essential to study flowers and fruits of living representatives. Preliminary results from fossil analyses suggest that a number of Late Cretaceous species may be closely related to Nyssa (Nyssaceae) and Davidia (Davidiaceae). Therefore, it is necessary to analyze developing flowers and fruits of Nyssa sylvatica and Davidia involucrata. The aim of this study is to characterize the morphology and anatomy of young and mature flowers and fruits, and to relate them to comparable features of the fossil species. The overarching goal is to reveal developmental processes and patterns that lead to significant evolutionary changes in fruit structure. Samples are taken from Nyssa and Davidia trees that are cultivated on the Oregon State University campus. During November 2014, mature fruits Nyssa and Davidia were collected and sectioned transversely and longitudinally before chemical fixation. Mature fruits were fixed in Formalin - Acetic Acid - Alcohol (FAA) for several days. Fruits were then dehydrated in an ethanol series (30% for two days, 50% for three days, 70% for four days, and 95% for four days). After dehydration the specimens infiltrated and embedded with ethanol-glycol methacrylate (Technovit 7100 embedding kit, Electron Microscopy Sciences, Hatfield, Pennsylvania, USA) plastic, at sequential ratios: 2:1, 1:1, and 2:1. Then specimens were fully embedded in the plastic. Fruits were sectioned using an AO 820 rotary microtome at a thickness of 5 microns. Microtome sections were stained with toluidine blue (pH 4.2) and mounted on microscope slides with mounting medium. During the spring mature flowers will be collected and prepared using the same steps as above. Prepared specimens will be photographed using a BetterLight digital scanning camera (Better Light, Placerville, California, USA). Samples collected for scanning electron microscopy were chemically fixed with a diluted Karnovsky solution for 60 hours at 40C. Flowers were dehydrated in an ethanol serial solution for critical point Microsc. Microanal. 21 (Suppl 3), 2015 866 drying. Images were acquired using a Quanta 600 FEG, FEI, Hillsboro, OR, USA. The results from this study will provide important data regarding the morphology, anatomy, and development of cornalean flowers and fruits. We hypothesize that there are different developmental processes and patterns of developmental change that lead to mature structural differences between Nyssa and Davidia. These studies will provide a test of this hypothesis. Furthermore, the data yielded from this study will provide a framework for working with fossil flowers and fruits that represent different stages of development. Using a multidisciplinary approach consisting of developmental and paleobotanical research we have an opportunity to shed light on the evolutionary processes that have shaped the evolution and diversity of reproductive organs within Cornales [4]. References: [1] APG, Botanical Journal of the Linnean Society 161 (2009), p.105. [2] C Fan and QY, Xiang American Journal of Botany 90 (2003), p. 1357. [3] QY Xiang et al, Molecular Phylogenetics and Evolution 59 (2011), p. 123. [4] The authors acknowledge funding from the National Science Foundation, grant number DGE1314109.");sQ1[434]=new Array("../7337/0867.pdf","Morphology, Anatomy, and Development of Cunninghamia lanceolata (Cupressaceae) Pollen Cones.","","867 doi:10.1017/S1431927615005139 Paper No. 0434 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Morphology, Anatomy, and Development of Cunninghamia lanceolata (Cupressaceae) Pollen Cones. Brian A. Atkinson1, Teresa Sawyer2, Gar W. Rothwell1,3, and Ruth A, Stockey1 1. 2. Department of Botany and Plant Pathology, Oregon State University, Corvallis, Oregon, USA. Electron Microscope Facility, Oregon State University, Corvallis, Oregon, USA. 3. Department of Environmental and Plant Biology, Ohio University, Athens, Ohio, USA. Cupressaceae (Cypress family) is the most widespread and one of the most economically important conifer lineages [1]. Some well-known representatives of Cupressaceae are Sequoia (redwoods), Juniperus (junipers), and False Cedars (Thuja, Calocedrus, Chamaecyparis). This family has a long and rich fossil history that dates back to the Early Jurassic [2]. Due to its impressively long and complex evolutionary history, Cupressaceae has been a subject of numerous morphological, anatomical, paleobotanical, and evolutionary studies. The majority of past studies have focused on seed cones within Cupressaceae [3]. In contrast, pollen cone studies are largely lacking, thus our understanding of pollen cone evolution within Cupressaceae is hindered. There is a growing number of paleobotanical studies from several geological time intervals [2, 4] that are beginning to shed light on the evolution of pollen cones. Recently, the paleobotany lab at Oregon State University has recovered cupressaceous pollen cone fossils from the Eocene (ca 55 Ma) of Western Canada. Coupled with detailed structural and developmental studies of living cones these fossils will provide a greater understanding of the evolution of Cupressaceae and more fully characterize changes in conifer pollen cones through time. Preliminary results reveal that the fossil pollen cones are assignable to the extant genus, Cunninghamia. This genus represents the earliest diverging lineage within Cupressaceae with an impressive fossil record of seed cones, wood, and leaves. To further our understanding of the evolution of this genus and of the Cupressaceae as a whole, studies of living pollen cones also have been initiated. The immediate goals of this study are as follows: 1) to characterize the morphology and anatomy of young pollen cones in bud stage, 2) to characterize the morphology and anatomy of mature pollen cones, and 3) to reveal any significant changes in cone structure through the entire range of development. The over-arching goal of this research is to provide a structural framework for future studies regarding the morphology, anatomy, and evolution of cupressaceous pollen cones. Two trees of Cunninghamia lanceolata that are cultivated in Corvallis, Oregon and at Ohio University, Athens, Ohio provided material for study. Two major stages of development are of interest: the late-bud stage and mature pollen cone stage. During January 2015, buds of pollen cone clusters were collected and sectioned transversely and longitudinally before chemical fixation. Sectioned buds were fixed in Formalin - Acetic Acid - Alcohol (FAA) for twelve days. Afterwards, the sectioned pollen cones were dehydrated in an ethanol series (30% for two days, 50% for three days, 70% for four days, and 95% for four days). After dehydration the specimens were infiltrated and embedded with ethanol-glycol methacrylate (Technovit 7100 embedding kit, Electron Microscopy Sciences, Hatfield, Pennsylvania, USA) resin, at sequential ratios: 2:1, 1:1, and 2:1. Then specimens were fully embedded in the plastic. The embedded specimens were sectioned using an AO 820 rotary microtome at a thickness of 5 microns. Microtome sections were stained with toluidine blue (pH 4.2) and mounted microscope slides Microsc. Microanal. 21 (Suppl 3), 2015 868 with mounting medium. Prepared specimens will be photographed using a BetterLight digital scanning camera (Better Light, Placerville, California, USA). Pollen cone clusters in bud stage will also be dissected, sectioned, and imaged using light microscopy and Scanning Electron Microscopy. Later during the growing season in spring, mature cones will be collected and dissected and studied using SEM, (Quanta 600 FEG (FEI, Hillsboro, OR, USA). Samples for SEM will be chemically fixed in a modified Karnovsky solution for 60 hours at 40C. Pollen cones will be dehydrated as above in preparation for critical point drying. Samples will mounted to a stub with carbon tape, sputter coated and imaged. The results from this study will provide important data regarding the morphology, anatomy, and development of pollen cones of Cunninghamia. Due to the phylogenetic position and outstanding fossil record of Cunninghamia, we predict that differences between the young and mature stages of pollen cones will meaningfully improve our understanding of conifer reproductive evolution. Furthermore, the data yielded from this study will provide a framework for working with fossil pollen cones that represent different stages of development. Using a multidisciplinary approach consisting of structural and paleobotanical research we have an opportunity to shed light on the evolution of cupressaceous pollen cones [5]. Images will include early pollen cone development, cross sections of pollen cones, pollen with light and electron microscopes. References: [1] A Farjon and D Filer in "An Atlas of the World's Conifers: An Analysis of Their Distribution, Biogeography, Diversity and Conservation Status" Brill (2013) p. 17. [2] I Escapa et al, Review of Palabotany and Palynology 151 (2008), p. 110. [3] BA Atkinson et al, American Journal of Botany 101 (2014), p. 2136. [4] GR Hernandez-Castillo et al, International Journal of Plant Sciences 166 (2005), p. 339. [5] The authors acknowledge funding from the National Science Foundation, grant number DGE1314109.");sQ1[435]=new Array("../7337/0869.pdf","Why Phragmites australis Canes Grown in Udono Reed Bed are the Best Materials for the Reeds of Japanese Wind Instrument `Hichiriki'. A Structural and Biomechanical Study","","869 doi:10.1017/S1431927615005140 Paper No. 0435 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Why Phragmites australis Canes Grown in Udono Reed Bed are the Best Materials for the Reeds of Japanese Wind Instrument `Hichiriki'. A Structural and Biomechanical Study Masahiro Kawasaki1, Tadashi Nobuchi2, Yuta Nakafushi3, Masateru Nose4, and Makoto Shiojiri5 1. 2. JEOL USA Inc., 11 Dearborn Road, Peabody, MA 01960, USA Kyoto University, Kyoto 606-8501, Japan * Present address: 1-68 Kunobe, Shiga, 520-2353, Japan 3. School of Science and Engineering, University of Toyama, Toyama, 930-8555, Japan 4. Faculty of Art and Design, University of Toyama, Takaoka, Toyama 933-8588, Japan 5. Kyoto Institute of Technology, Kyoto 606-8585, Japan Present address: 1-297 Wakiyama, Kyoto 618-0091, Japan Hichiriki (Figure 1) is a double reed woodwind instrument in Japanese ancient imperial court music `gagaku' since the 7th century and one of UNESCO Intangible Cultural Heritages. For more than 1,200 years, the best reeds for the hichiriki have been made only out of canes of Phragmites australis (P. australis) (common reed) harvested from "Udono" which is a limited reed bed of riverbanks near Kyoto along the Yodo River. This resembles that the best reed for clarinet, oboe, or bassoon is manufactured from a cane of Arundo donax (A. donax) (giant reed) which grows only in a few areas of the Var in France due to its very mild climate. Now, misgivings that environmental disruption in Udono causes ecocide of the P. australis for hichiriki reeds and may bring a catastrophic crisis to gagaku are expressed [1]. In this study, plant anatomy was examined for choice canes of P. australis grown in different reed beds in Japan as well as morphology, and the local indentation hardness and Young's modulus of tissues on the cross-sections of some representatives of hichiriki reeds were measured, searching why the canes from Udono's P. australis are the best materials for hichiriki reeds. Specimens for light microscopy were prepared by cutting the sample cane pieces to transverse sections about 30 m thick with a sliding microtome, followed by double staining with safranin and fast green FCF [2]. The hardness and Young's modulus of different parts on the cross-sectioned surfaces of four hichiriki reeds were measured with a Vickers indentation load of 5 mN using a nanoindentation system (Fischerscope, H100C-XYp). Figure 2 shows light micrographs of the sections of the canes from different reed beds. T is the wall thickness measured from the image. The structure of the cane is indicated in Fig. 2(e). P. australis almost resembles A. donax in plant anatomical structure but not in outer shape and size of cells. This is the reason why A. donax cannot be used for hichiriki reeds. Figure 3 illustrates the hardness measurements of four hichiriki reeds. We conclude that the good canes for hichiriki reeds have an outer diameter of about 11 mm, a wall thickness of about 1 mm and comparatively homogeneous structure where harder materials, such as epidermis, hypodermis, sclerenchymatous cells, and vascular bundle sheaths with hard walls, are orderly deployed with softer materials such as parenchyma cells and vascular bundles. This structure has smaller differences of hardness and Young's modulus between the hard and soft materials in the reed, providing the best music performance. Hichiriki players say that it has become harder and harder to get good materials for reed, similar to the players of clarinet and oboe using A. donax reeds. However, they can be still found in Udono but be hardly found in other reed beds. We thank Ms. Hitomi Nakamura, the Reigakusha Gagaku Ensemble, for providing the samples, the photograph of her hichiriki and valuable comments from a view of musical performance. Microsc. Microanal. 21 (Suppl 3), 2015 870 [1] M. Schuring et al, 2013. Pipers 381 (2013)10. [2] M. Kawasaki et al, Micros Res. Tech. (2015) in print. Fig. 1. Hichiriki. Its length is about 180 mm. Fig. 2. Light micrographs of P. australis stems from different reed beds; Kitakami Riv. near Sendai, Watarase Riv. near Nikko, Turumi Riv. in Tokyo, Uji Rev. in Kyoto, and Udono. Ep: epidermis. Hp: hypodermis. Vb: vascular bundle. Bs: vascular bundle sheath. Cor: cortex. Pa: parenchyma cell. Ct: cells with thin cell wall. Sc: sclerenchymatous cells. Pc: pith cavity. Fig. 3. Indentation hardness (a) and Young's modulus (b) of different parts (shown in the inset in b) in different hichiriki reeds (indicated in the inset in a). A photograph of reed (Rozetsu) R 5-1 is shown in a.");sQ1[436]=new Array("../7337/0871.pdf","Analysis of Plant Responses to Titanium Dioxide (TiO2) Nanoparticles","","871 doi:10.1017/S1431927615005152 Paper No. 0436 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Analysis of Plant Responses to Titanium Dioxide (TiO2) Nanoparticles Kaydee Smith1,2, Kajal Ghoshroy3, and Soumitra Ghoshroy1,2 1 2 Electron Microscopy Center, University of South Carolina, Columbia, SC 29208 USA Department of Biological Sciences, University of South Carolina, Columbia, SC 29208 USA 3 Department of Biological Sciences, University of South Carolina, Sumter, SC 29150 USA Recently, the amount of nanoparticles in consumer products has dramatically increased. Because of their broad scale use, these nanomaterials can be expected to be present in the environment, raising concerns about their effects on the lives of plants and animals. If these nanomaterials reach the environment, plants could potentially interact with them and take-up the particles into their system. Titanium dioxide (TiO2), a nanomaterial, is used in the production of cosmetics, and has been shown to have UV light shielding capabilities [2]. The effect of TiO2 nanoparticles on metabolic processes has been studied and was found to affect processes such as hormone metabolism [1]. However, the structural effects of TiO2 nanoparticles are still unclear. This study focuses on observing the effects of 5-15 nm TiO2 nanoparticles on the physiology and structure of cells and organelles of two plant species [jalapeno (Capsicum annuum) and corn (Zea mays)] using light and electron microscopy techniques. Plant responses were a result of nanoparticle exposure through the root systems. The plants were grown from seeds in a growth chamber with controlled light, humidity and temperature. Six plants of the same height from each species were collected, their roots washed thoroughly, and transferred to bowls containing a soil-free liquid culture medium (Hoagland solution) (Fig. 10). On the seventh day, the solution in three of the four bowls was replaced with three concentrations of TiO2 (515nm) nanoparticles (300 mg/L, 600 mg/L, and 1000 mg/L) dissolved in water. The final dish was filled with water and served as the control. The plants were maintained for at least seven days for the uptake of TiO2. Root and leaf samples from each group were collected and washed for study using light and electron microscopy. Root samples collected for SEM were fixed in glutaraldehyde, post-fixed in osmium tetroxide, dehydrated in series of ethanol, critically dried in a critical point dryer, mounted on aluminum stubs, gold sputter coated, and observed under the Tescan Vega3 SEM (includes a ThermoNoran EDS detector) for the presence of titanium. Samples for TEM and light microscope (LM) analysis were fixed in glutaraldehyde, post-fixed in osmium tetroxide, dehydrated in series of ethanol, incubated in acetone, and embedded in EMBed 812. Sectioned blocks were used for LM and TEM analyses. The root mass of corn and jalapeno plants exposed to high levels of TiO2 was lower than that of the control indicating inhibition of root growth (Fig.9). Results from SEM analyses exhibited structural damage of root epidermal cells in corn and jalapeno plants exposed to 1000 mg/L TiO2 (Fig.2). EDS analyses indicated presence of titanium particles on the surface of both corn and jalapeno plant roots exposed to 1000 mg/L TiO2 (data not shown). The control groups did not show any signs of structural damage (Fig. 1). Light microscopy revealed possible vacuolation in the corn root cells exposed to large amounts of TiO2 (Fig. 4). This was most evident when comparing the corn control to the 1000 mg/L TiO2 sample (Fig.3). TEM analysis uncovered changes in the roots of both jalapeno and corn 1000 mg/L TiO2 groups. In comparison to the control, the 1000 mg/L TiO2 corn root cells became highly vacuolated with no apparent presence of TiO2 particles inside the cells (Fig.5 & 6). Therefore corn, a monocot plant, may not take-up titanium nanoparticles. This phenomenon is yet to be understood. TEM analysis of the jalapeno plants treated with 1000 mg/L TiO2 revealed clusters of particles (possibly TiO2) within the root cells, but no vacuolization (Fig.7 & 8). While the root surface structural damage seen in SEM images appears to be the same in both species, TEM analysis demonstrated that TiO2 might have varying effects on the cells of corn and jalapeno plants. Further experiments are underway to evaluate the chlorophyll levels in the leaves of each species for each experimental group. Microsc. Microanal. 21 (Suppl 3), 2015 872 References: 1. Laxminath, Tumburu, et al. "Phenotypic and Genomic Responses to Titanium Dioxide And Cerium Oxide Nanoparticles In Arabidopsis germinants."Environmental Toxicology and Chemistry 34.1 (2015): 70-83. 2. Li, Xiaozho, et al. "Layer-by-layer assembled TiO2 films with high ultraviolet light-shielding property." Thin Solid Films 571.1 (2014): 127-133. Fig.1: Control jalapeno root tip with no visible structural damage (SEM) Fig.2: 1000 mg/L TiO2 jalapeno root tip with visible structural damage (SEM) Fig.3: Corn control root cross section without structural damage (LM) Fig.4: 1000 mg/L TiO2 corn root tip cross-section with possible vacuolization (LM) Fig.5: Corn control root cells showing no vacuolization (TEM) Fig.6: 1000 mg/L TiO2 corn root cells with visible vacuolization (TEM) Fig.7: Jalapeno control root cells Fig.8: 1000 mg/L TiO2 jalapeno root with no visible particle clusters (TEM) cells with visible particle clusters within the cells (TEM) Fig.10: Method for exposure of plants to TiO2 Fig.9: Dry root mass in grams of corn and jalapeno control and experimental groups");sQ1[437]=new Array("../7337/0873.pdf","Correlative Light and Electron Microscopy Techniques: Challenges and Successes.","","873 doi:10.1017/S1431927615005164 Paper No. 0437 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Correlative Light and Electron Microscopy Techniques: Challenges and Successes. Erin S. Stempinski1, Xufeng Wu2, and Christine A. Brantner1,3 1. Electron Microscopy Core Facility, National Heart Lung and Blood Institute, National Institutes of Health, Bethesda, Maryland, USA. 2. Light Microscopy Core Facility, National Heart Lung and Blood Institute, National Institutes of Health, Bethesda, Maryland, USA. 3. Center for Microscopy and Image Analysis, The George Washington University, Washington, D.C., USA. Correlative light and electron microscopy (CLEM) combines the strengths of light microscopy (LM) and electron microscopy (EM) to form a more complete picture of a cellular process. LM allows for a multitude of fluorescent tags and a wide field of view to select rare events, and EM allows for increased resolution and visualization of cellular ultrastructure [1]. We used three methods to examine external and internal morphologies of cells: cell cultures grown on coverslips, resin sections on coverslips, and resin sections on coated slot grids. Coverslip methods used indium-tin-oxide (ITO) coverslips with fiducial markers. The coverslips were coated with a 0.1% (w/v) Poly-L-lysine solution for 30 min, rinsed in water, and dried on filter paper overnight. Cells were grown and labeled on the coverslips depending on the requirements of the projects and imaged in a laser scanning confocal microscope (LSCM). All samples were imaged on a ZEISS Sigma HD VP Scanning Electron Microscope (SEM). Areas of interest on coverslip samples noted in LM were found in the SEM using ZEISS Shuttle and Find software. Slot grids were placed in a STEM holder and imaged using a backscatter detector. Correlation and overlay of images was performed using multiple software packages. Cell cultures on ITO coverslips were fixed, dehydrated, and either critical-point dried or treated with hexamethyldisilazane (HMDS). Fine cellular processes of N2A cells were broken or lost in the critical point dryer (CPD), but retained when the cells were en bloc stained with uranyl acetate and dried with HMDS. Challenges of working with the ITO coverslips included fragility, expense, and the requirement of a special holder to be made for the CPD. Some samples were embedded in LR White resin and serial sections were placed on ITO coverslips with fiducial markers using uncoated slot grids. The coverslip was immunolabeled for LM, imaged with a LSCM, post-stained using 1% uranyl acetate and lead citrate, then imaged in the SEM. Sections that were 200 nm in thickness had more fluorescent staining in the LSCM than 100 nm sections. However, the 200 nm sections did not provide the necessary resolution in the SEM to image certain ultrastructural features (Figure 1). For both cell culture and resin sections on coverslips, the ZEISS Shuttle and Find software allowed for simple and fast localization of the cell or region of interest between LM and EM. Samples that used the resin sections on slot grids method were cells cultured on MatTek dishes with gridded coverslips. Cells were fixed, en bloc stained, dehydrated, and infiltrated with Epon resin in the dishes. For some samples, the coverslip was removed from the dish and placed over a round disc mold filled with resin. For other samples, BEEM capsules were placed in the dish. After polymerization, serial sections were cut and placed on coated slot grids. There were challenges releasing the MatTek coverslips from the polymerized resin puck. They did not release when immersed in liquid nitrogen, when heated on a hot plate, nor when separated using a razor blade. Coating the dishes with carbon Microsc. Microanal. 21 (Suppl 3), 2015 874 prior to growing the cells did not enhance the separation of the coverslip from the resin. Some glass did release from the resin puck when the coverslip was scored with a diamond scribe. These problems were prevented when BEEM capsules were placed over the region of interest. However, the grid pattern on the coverslip could not be seen once the coverslip was in resin. Therefore, the region of interest needed to be marked on the underside of the coverslip with a diamond scribe prior to processing. DIC images of the region of interest prior to EM processing facilitated both close trimming of the block face near the region of interest and the ability to find the particular cells of interest. The SEM provided a wide field of view, facilitating the ability to find the cells of interest within the section. Imaging parameters at the LM level determined the success of the LM-EM overlays. Successful overlays had LM images at magnifications similar to those taken at the EM level and LM image resolutions (i.e. pixel numbers) similar to EM image resolutions (Figure 2). Correlative microscopy is a powerful tool that can be used in the understanding of biological processes. It necessitates robust techniques and attention to detail to ensure success. These techniques can be flexibly adapted for many types of samples based on the information desired [2]. References: [1] AA Mironov and GV Beznoussenko, Journal of Microscopy 235 (2009), p.308-321. [2] Special thanks to the Drs. Panpan Yu, Xueting (Tina) Jin, Lois Greene, and Chad Donaldson for providing project ideas and cell cultures. This research was supported by the Intramural Research Program of the NIH, NHLBI. Figure 1. Correlation of 200 nm sections on ITO coverslips with (A) SEM, (B) LSCM, and (C) overlay of SEM and LSCM. Fine ultrastructural detail in the brightly staining region in (B) is not evident in the SEM. Bar = 1 �m. Figure 2. Overlay of LM and EM of N2A cells. Cells were imaged live, fixed, cytoskeleton extracted, and processed for EM. Some shifting occurred due to the cells moving prior to fixation and during processing. Bar = 50 �m.");sQ1[438]=new Array("../7337/0875.pdf","Characterizing Microtubule Organization in the Arabidopsis Thaliana Root Apical Meristem via Correlative Microscopy","","875 doi:10.1017/S1431927615005176 Paper No. 0438 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterizing Microtubule Organization in the Arabidopsis Thaliana Root Apical Meristem via Correlative Microscopy Katherine Celler1, Chris Ambrose2, Yuan Ruan1, Bradford Ross3, Geoffrey Wasteneys1 1 2 Department of Botany, University of British Columbia, Vancouver, Canada Department of Biology, University of Saskatchewan, Saskatoon, Canada 3 BioImaging Facility, University of British Columbia, Vancouver, Canada Growth in plants occurs at shoot and root apices, with new tissues arising from stem cell centres known as meristems, areas analogous to animal stem cell niches (Figure 1A). Meristem size depends on a balance between the rate of uncommitted stem cell proliferation and cell differentiation, determined by different hormone levels and their cross-talk. Recent discoveries indicate that the cytoskeleton plays an important part in this process, with microtubules (MTs) regulating the hormone auxin and its transport via complex feedback mechanisms [1]. In addition, MT organization in dividing cells has also been implicated in meristem maintenance, with the Arabidopsis thaliana MT-associated CLASP protein playing a critical role [2, 3]. CLASP accumulates at specific cell edges, enabling MT growth around these edges and promoting the formation of MT bundles that span adjacent cell faces [4]. Unlike the transverse conformation MTs typically adopt in elongating cells (Figure 1B), CLASP promotes transfacial MT bundle formation (Figure 1C). These transfacial bundles are strongly associated with maintaining the capacity for division, an important stem cell feature. To date, imaging of plant MTs has mainly employed immunofluorescence on fixed samples [5] or the use of fluorescent fusions to study MT dynamics using confocal laser scanning microscopy [6]. Though informative, these methods do not allow for resolving of individual MTs within bundles and the precise progression of MT configurations (or bundle association with CLASP) during stem cell differentiation is as yet not understood. To characterize the developmental switches that lead to changes in MT organization a correlative approach is necessary. High pressure freezing, freeze substitution and imaging with fluorescence light microscopy (fLM) followed by transmission electron microscopy (TEM) or tomography will enable the structural analysis of MTs at Arabidopsis thaliana cell edges at high resolution and in three dimensions. Preliminary analysis has been done on an Arabidopsis thaliana sample containing the CLASP protein tagged with a histidine tag and GFP. An fLM image of a high pressure frozen, freeze substituted and thick-sectioned root apical meristem indicates fluorescent signal at cell edges (Figure 2). Further immunolocalization studies on this same sample (using Ni-NTA-Nanogold labelling) will confirm whether the GFP signal corresponds to the CLASP protein. Tomographic analyses can provide information about the 3-dimensional organization of microtubules at these locations. Study of multiple cells in different stages of development will provide a picture of MT progression from transfacial to transverse orientation, and the role of CLASP in these transitions. Microsc. Microanal. 21 (Suppl 3), 2015 876 Figures Figure 1. (A) Schematic of the plant root apical meristem and associated cell types. The quiescent (organizing) centre sends signals to the stem cells to prevent them from differentiating. Stem cells divide asymmetrically to produce daughter cells that divide a number of times in the meristematic zone before they stop dividing and terminally differentiate. (B-C) Spatial microtubule organization in cells: in elongating cells, MTs adopt a transverse organization (B), whereas in dividing cells (stem cells and dividing daughter cells), MT-associated protein CLASP enables microtubule growth around sharp cell edges (C). Figure 2. Fluorescent light micrograph of Arabidopsis thaliana containing CLASP-His-GFP. The image shows a high pressure frozen, freeze substituted and thick-sectioned root apical meristem. Green fluorescent signal is present at cell edges. References [1] Ambrose, C., et al., Dev Cell, 2013. 24(6): p. 649-59. [2] Ambrose, J.C., et al., Plant Cell, 2007. 19(9): p. 2763-2775. [3] Ruan, Y. and G.O. Wasteneys. Curr Opin Plant Biol, 2014. 22: p. 149-58. [4] Ambrose, C., et al., Nat Commun, 2011. 2: p. 430. [5] Wick, S.M., et al., J Cell Biol, 1981. 89(3): p. 685-90. [6] Marc, J., et al., Plant Cell, 1998. 10(11): p. 1927-1939.");sQ1[439]=new Array("../7337/0877.pdf","Cryogenic Sample Preparation Preserves Elemental Composition for Correlative Light and X-ray Fluorescence Microscopy","","877 doi:10.1017/S1431927615005188 Paper No. 0439 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Cryogenic Sample Preparation Preserves Elemental Composition for Correlative Light and X-ray Fluorescence Microscopy Qiaoling Jin1, *, Barry Lai2, Si Chen2, Sophie Charlotte Gleber2, Lydia Finney2, David Vine2, Tatjana Paunesku3, Gayle Woloschak3, Stefan Vogt2, Chris Jacobsen 1,2 1 Department of Physics &Astronomy, Northwestern University, 2145 Sheridan Road, Evanston, Illinois, United States of America 2 X-ray Science Division, Advanced Photon Source, Argonne National Laboratory, 9700 South Cass Avenue, Argonne, Illinois, United States of America 3 Department of Radiology, Northwestern University, 303 E. Chicago Ave., Illinois, United States of America * Presenter: Qiaoling.Jin@northwestern.edu Synchrotron based x-ray fluorescence microscopy (XFM) is well suited to determine the detailed spatial distribution of most biologically important elements in biological cells and tissues with submicrometer resolution and attomgram sensitivity [1, 2, 3, 4]. Cryogenic sample preparation coupled with cryogenic scanning XFM systems such as the Bionanoprobe at Argonne National Laboratory (ANL) presents one of the most reliable approaches for studies of cellular elemental homeostasis [5]. However, due to the very limited availability of XFM microprobes capable of cryoscannig as well as the necessity to perform other correlative studies, different sample preparation protocols have been developed [6,7]. Aldehyde based chemical fixation and rapid freezing based cryofixation, followed by different dehydration protocols, are commonly used. Both methods have been originally developed and extensively studied in the field of transmission electron microscopy for the preservation of ultrastructure and protein antigenicity [8]. When they are adapted to prepare cultured adherent mammalian cells for XFM studies, parallel comparison among different sample preparation approaches is currently underexplored on the preservation of cellular elemental content and distribution [7]. To illustrate possible artifacts associated with each specific approach, we decided to compare how elemental content and distribution is preserved in cryofixed versus chemically fixed cells, with both of them subsequently dehydrated and scanned by XFM under ambient temperature. Mouse embryonic fibroblast NIH/3T3 cells grown on silicon nitride windows were plunge frozen in liquid nitrogen cooled liquid ethane using a FEI Vitrobot Mark IV plunge freezer. They were then cryogenically transferred and imaged using a cryo light microscope, and the cryojet XFM microprobe installed at 2-ID-D at Advanced Photon Source (APS) of ANL. Visible cryo light microscope images (Fig. 1A) indicated wellmaintained cell morphology. The 2-dimensional elemental maps revealed homogeneous distribution of most freely diffusible ions K and Cl, with comparable content to that found in living cells [9]. There were more Ca ions in the cytoplasm than in the nucleus, an indication of undamaged cellular membrane before and during plunge freezing [9]. Fe has characteristic perinuclear distribution [7]. Zn concentration was higher in the nucleus than in the cytoplasm, presumably due to the presence of a large quantity of Zn binding proteins in the nucleus. We then subjected one of the plunge frozen samples to be freeze dried (PFFD) and compared to cells which were chemically fixed by 4% paraformaldehyde (PFA) and dried in the air. Both of them were scanned by XFM microprobe at APS beamline 2-ID-E. The total elemental contents in each cell were obtained through region-of-interest analysis. They were then averaged from all cells in the same sample and plotted as fractions to the content in PFFD sample. As shown in Fig. 1B, 2-dimensional maps of PFFD sample showed very similar distribution pattern of most Microsc. Microanal. 21 (Suppl 3), 2015 878 874 elements including P, S, Cl, K, Fe and Zn to the frozen hydrated cells, with some possible redistribution of Ca. Compared to PFFD sample, 4% PFA fixed cells had the severe loss of K (>99% loss) and Cl (70% loss), and higher content of Ca (200% increase). In addition, 4% PFA fixed cells showed less preserved amount of some tightly bound ions such as P, S and Fe, with about 50% S and Fe and 60% of P detected compared to PFFD sample. Although cell-to-cell variation and statistical error might factor in to this apparent reduction, the comparable level of Zn prompted us to believe that a certain degree of loss might exist for these elements. We conclude that plunge freezing followed by freeze drying preserved the contents and distributions of most elements at a level comparable to frozen hydrated cells. It should be the method of choice whenever possible. If conventional chemical fixation has to be chosen, the results on diffusible ions and certain other elements must be carefully interpreted [10]. Figure 1. Comparison of different sample preparation approaches on the preservation of elemental contents and distributions. Panel A: visible cryo light microscope image (VLM) and 2-dimentional elemental maps of NIH/3T3 cells which were plunge frozen and imaged frozen hydrated by the cryojet XFM microprobe at 2-IDD at APS. Panel B: 2-dimentioanl elemental maps of plunge frozen and freeze dried NIH/3T3 cells (PFFD, 1st row) and 4% PFA fixed NIH/3T3 cells (4% PFA, 2nd row). Panel C: fractions of average elemental content in 4% PFA fixed cells to the average content in PFFD sample. Scale bars: 20 �m. Color scale in false colors spans from black (no signal) to red (maximum signal). References: [1] Paunesku T et al, Int. J. Radiat. Biol. 85 (2009), p. 710. [2] Vogt S and Ralle M, Anal. Bioanal. Chem. 405 (2012), p. 1809. [3] Hummer AA and Rompel A, Metallomics 6 (2013), p. 597. [4] McRae R et al, Chem.Rev. 109(2009), p. 4780. [5] Chen S et al, J. Synchrotron Rad. 21(2014), p. 66. [6] McRae R et al, Journal of Structural Biology. 155 (2006), p. 22. [7] Matsuyama S et al, X-ray spectrometry 39(2010), p. 260. [8] Vanhecke D et al, Methods in Cell Biology 88 (2008), p. 151. [9] Chandra S et al, PNAS, 86 (1989), p. 1870. [10] The authors acknowledge funding from NIH grant R01 GM104530.");sQ1[440]=new Array("../7337/0879.pdf","Methodology Development at NYULMC Microscopy Core - Correlative Light and Electron Microscopy Applications","","879 doi:10.1017/S143192761500519X Paper No. 0440 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Methodology Development at NYULMC Microscopy Core - Correlative Light and Electron Microscopy Applications Alice F. Liang1, 2,*, Chris Petzold1, Kristen Dancel-Manning1, Yan Deng1 and Michael Cammer1 1. 2. OCS Microscopy Core, New York University Langone Medical Center, New York, USA Department of Cell Biology, New York University Langone Medical Center, New York, USA * Corresponding author The Microscopy Core at New York University Langone Medical Center (NYULMC) established in 2005 is based on electron microscopy services. Light microscopy services were implemented in 2009 and the Core became a fully functional Microscopy Core. Similar to most Microscopy Cores, our light microscopy (LM) services focus on training for the correct use of microscopes, and keeping all the microscopes in a nice working state. Image analysis is another major LM core service, which helps users to better understand their imaging data. Our electron microscopy (EM) services are very broad, ranging from all different kinds of sample preparation to image acquisition. Given the fact of rapid development of light and electron microscopes, catching up with new technology is one of the key points for a modern Microscopy Core. The advantage of combining light and electron microscopy services in a single core provides a unique position for the Microscopy Core to carry projects over from light microscopy to electron microscopy. So, correlative light and electron microscopy (CLEM) becomes a hot selling point for our Microscopy Core services. Fluorescence microscopy has opened the way to study protein localization, interaction, and function in live cells. A huge effort has been put in this field during the last 20 years to improve the resolution of light microscopes, and super-resolution microscopes are good examples to localize protein at single molecule levels. However, precise localization of protein at high resolution and at the ultrastructural level can only be studied by electron microscopy (EM). By combining light and electron microscopy imaging techniques, the protein in live cells or tissue can be studied at higher resolution at the ultrastructural level, making correlative light and electron microscopy a very powerful tool [1, 2]. Project variety is the nature of the core facility. Experimental design of complex procedures, such as CLEM for each individual project by utilizing current available instruments and reagents, is a challenge. We have successfully carried on multiple projects by using different CLEM strategies [1], such as: neighboring sections, hybrid labeling, correlative overlay and Tokuyasu cryosections. To study cortical inhibitory neuron maturation and synaptic development, we use hybrid labeling strategy to double immunolabel GFP and Somatostatin (SST, a growth hormone-inhibiting hormone) of brain vibrotome sections of the Lhx6-eGFP transgenic mouse, to detect: (1) GFP positive cells under fluorescent microscope following correlated ultrastructural immunolabeling using Nanogold as marker; (2) SST positive cells using HRP/DAB as marker (Fig. 1). Double positive cells (Fig. 1) and single positive cells (data not shown) were clearly distinguished by silver enhanced gold particles and DAB darkened cells at ultrastructure level. To study melanoma invasion, we used the frozen section correlative overlay strategy to localize the melanocyte migration in the skin of Braf and P10 knock out mouse (Fig. 2). The gridded bottom dish was adopted for several projects to study centrosome mutation [3] and HIV infection (not published) using correlative overlay strategy; and Tokuyasu cryosections combined with super resolution microscopy were used to study the protein localization in intercalated disc of mouse heart [4]. Microsc. Microanal. 21 (Suppl 3), 2015 880 References: [1] K. Cortese et al, The Journal of Histochemistry and Cytochemistry 57, (2009), 1103-12. [2] B. N. Giepmans, Histochemstry and Cell Biology 130, (2008), 211-217. [3] J. Li et al, Nature 495, (2013), 255-259. [4] E. Agullo-Pascual et al, Cardiovascular Research 104 (2), (2014), 371-381. Figure 1. Correlative light and electron microscopy study using brain vibrotome section to double immunolocalize GFP (Nanogold) and SST (HRP/DAB). Double positive cells were shown clearly at ultrastructural level. Figure 2. Melanoma cells invasion in mouse skin was identified by using frozen section correlative overlay strategy.");sQ1[441]=new Array("../7337/0881.pdf","Investigating Kinetochore Structural Dynamics during Mitosis by Employing Combinations of Several Microscopic Techniques","","881 doi:10.1017/S1431927615005206 Paper No. 0441 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Investigating Kinetochore Structural Dynamics during Mitosis by Employing Combinations of Several Microscopic Techniques Valentin Magidson1,4, Jie He1, Jeffrey G. Ault1, Christopher B. O'Connell1,5, Nachen Yang1, Haixin Sui1,2, Bruce F. McEwen1, and Alexey Khodjakov1,3 1 2 Wadsworth Center, New York State Department of Health, Albany, NY 12201, USA School of Public Health, State University of New York, Albany, NY 12201, USA 3 Rensselaer Polytechnic Institute, Troy, NY 12180, USA 4 Current address: National Cancer Institute, Frederick, MD 21702, USA 5 Current address: Nikon Instruments, Melville, NY 11747, USA The kinetochore attaches a chromosome to the mitotic spindle and harnesses forces that move the chromosome [1]. Recent studies indicate that kinetochores possess compliant linkages that contribute to mitotic checkpoint signaling [2]. Deformation of these elements are proposed to account for the differences in the distance between populations of labeled inner and outer plate proteins (intrakinetochore stretch) observed by light microscopy (LM). However, the underlying structural basis for intrakinetochore stretch remains unknown. To investigate this, we employ combinations of several microscopic techniques to investigate the change of the kinetochore from an expanded form to a contracted one during microtubule (MT) interaction, and how this change is reversed by drugs affecting MT polymerization. Fluorescent markers for outer and inner kinetochore proteins and for the spindle poles determine the orientation of sister kinetochores with respect to poles and to the centromere and allow precise determination of intrakinetochore stretch for individual kinetochores. By superimposing LM fluorescence on the distribution of gold particles in the corresponding EM images, we compare the shapes of kinetochores with high verses low values of intrakinetochore stretching (Figures 1 and 2). The human cell line RPE1 is used which is chromosomally stable, maintaining a near-diploidy number of 46 chromosomes. The microscopic techniques employed in different combinations are immuno-LM (GFP-tagged and FluoroNanogold-tagged), immuno-EM (FluoroNanogold-tagged), DIC, serial section EM, and EM tomography. An important point of this study is that we can locate individual kinetochores and their sisters, among the 92 kinetochores present in a single cell, in both the LM images and EM images, and superimpose the LM image over the EM one. We found that kinetochores are expanded early in spindle formation and are contracted during metaphase when the chromatin between sister kinetochores is the most stretched, creating the most tension. Drugs that effect MT polymerization reverse this process by expanding contracted kinetochores that have lost their MT attachments. The changes in size occur at the outer kinetochore region, while the inner region remains unchanged. Rather than a plate, the outer kinetochore region radically expands and contracts, depending on its interactions with MTs. References: [1] I. M. Cheeseman, Cold Spring Harb. Perspect. Bol. 6(7) (2014), a0 15826. [2] X. Wan et al, Cell 137(4) (2009), 672. Microsc. Microanal. 21 (Suppl 3), 2015 882 Figure 1. (A-C) Examples of metaphase kinetochores with various level of intra- and inter-kinetochore stretching. Positions of the centroids and exact values of CenpA-Hec1 distance (Delta) as well as interkinetochore distances (Hec1-Hec1) are shown for each kinetochore/centromere. Notice that kinetochores can remain compact even on extremely stretched centromeres (A). Also, deformation of the centromere is often asymmetric (B). Finally, separation of the inner and outer- kinetochore proteins consistently occurs along the axis of the attached microtubule bundle (denoted with yellow lines). Figure 2. (A) Histogram of CenpA-Hec1 distances (Delta) in metaphase cells. (B) Direct LM/EM comparison of Hec1 distribution for metaphase kinetochores with high (k1, CenpA-Hec1=173 nm) vs. average (k2, CenpA-Hec1=137 nm) intrakinetochore stretching. As evident from the distribution of gold particles, the outer plate is more compact (gold particles form a tighter cluster) in the moderatelystretched k2. In highly stretched k1, Hec1 extends along the attached MT bundle. Notice that while GFP is not directly visualized in EM, its position is revealed when LM and EM images are superimposed. Color crosses mark centroids of Hec1 (red) and CenpA (green) distributions. Bar = 500 nm.");sQ1[442]=new Array("../7337/0883.pdf","Cryo-Ultramicrotomy of Complex Molecular Fluids","","883 doi:10.1017/S1431927615005218 Paper No. 0442 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Cryo-Ultramicrotomy of Complex Molecular Fluids Min Gao Liquid Crystal Institute, Kent State University, Kent, Ohio, USA. Complex molecular fluids (CMFs, e.g., liquid crystals, crude oil, and detergents) have tremendous impacts on the modern world, but the detailed understanding of CMF behaviors at the molecular level is often surprisingly limited. This can be attributed to their complicated structure and the lack of effective nanoscale structural probes for CMFs in general. In this paper, we summarize our recent effort to apply cryo-ultramicrotomy to cryo-TEM studies of a representative and challenging CMF group - liquid crystals (LCs). Our results show that cryo-ultramicrotomy can be used as a reliable specimen preparation technique allowing subnanometer resolution direct imaging of CMFs in general. We also discuss on the advantages and disadvantages of the technique based on comparative studies using different techniques including thin film plunge freezing and replica techniques. Liquid crystals (LCs) are mesophases with orientational order only (nematic mesophases) or both orientational and 1D (or 2D) positional orders (smectic mesophases) [1]. LC structures are very sensitive to surface and interface, making it difficult to preserve the native structure in electron beam transparent thin films. In addition, due to the very weak intermolecular interaction, LC materials normally show severe radiation damage under electron beam. As a result, a replica TEM technique, freeze fracture TEM (FFTEM), has been the dominating electron microscopy technique for LCs despite its relatively low resolution (a few nanometer). We use cryo-ultramicrotomy in both thermotropic and lyotropic LCs. A thermotropic LC typically consists of single- or multiple-component complex molecules with phase transitions driven mainly by temperature; while a lyotropic LC is a solution of complex molecules in certain solvent (most often, water) and the amount of order is predominantly controlled by the solution concentration. We applied the so-called cryo-electron microscopy of vitreous section (CEMOVIS) technique to lyotropic LCs, i.e., high pressure freezing is employed to obtain vitrified "bulk" lyotropic LCs, followed by cryoultramicrotomy and cryo-TEM. While for thermotropic LCs, plunge freezing is used to quench the desired structure stabilized at a higher temperature. Two examples are shown here: lyotropic disodium cromoglycate (DSCG), known also as an anti-asthmatic drug (Fig. 1); and Au nanoparticle doped 5CB, a commonly used thermotropic LC (Fig. 2). We compare the "bulk" cryo-ultramicrotomy results with those obtain using "thin film" plunge-freezing [2]. Figure 1 shows comparative study of the DSCG solutions. The uniform contrast of the suspended 6.2% DSCG and the stripes with bright/dark contrast in 15% DSCG match the isotropic and uniaxial nematic structures, respectively. The dark stripes can be understood as the elongated chromonic aggregates formed by face-to-face packing of the DSCG molecules in water. In general, the cryo-ultramicrotomy "bulk" method yield similar results to the "thin film" plunge-freezing. But the bulk approach minimize the surface anchoring effect played by the LC/air and LC/carbon interfaces. We also present our results on the influence of cryoprotectant, which is often needed for low concentration solutions. Figure 2 shows comparative study of Au nanoparticle doped 5CB. The nanoparticles of 1-3 nm in size seem to be distributed randomly in 5CB without agglomerations. However, Fig. 2a clearly shows that Microsc. Microanal. 21 (Suppl 3), 2015 884 the Au nanoparticles are strongly attracted by the carbon supporting film used in "thin film" plunge freezing and can decrease the concentration of Au nanoparticles in the suspended 5CB. In addition, we show that STEM Z-contrast imaging improves the visibility of the nanoparticles, which is especially useful for cryo-sectioned TEM specimens due to the damages from the sectioning [3]. References: [1] PG De Gennes and J Prost, "The Physics of Liquid Crystals", (Oxford University Press, New York). [2] M Gao et al, Microsc Res Tech 77 (2014), 754. [3] The TEM-related experiments were carried out at the cryo-TEM lab of the Liquid Crystal Institute, Kent State University. The author thanks Dr. Oleg D. Lavrentovich, Dr. Chenming Xue, and Dr. Quan Li for providing the samples. Figure 1. Comparative cryo-TEM study of DSCG lyotropic chromonic LCs. (a) � (e) Cryo-TEM results of DSCG solutions prepared by the "thin film" approach (plunge frozen specimens). (a) A typical image of 6.2% DSCG, showing uniform contrast corresponding to the isotropic phase. (b) A typical image of 15% DSCG. (c) Corresponding FFT pattern of the nematic structure shown in Fig. 1b. (d) A magnified image of the marked local area in Fig. 1a. (e) An image of the aggregates perpendicular to the thin film surface observed in 15% DSCG. (f) and (g) CEMOVIS images of nematic regions in 15% DSCG with 10% dextran. The hollow arrow in (g) points out a domain of aggregates perpendicular to the specimen surface. Figure 2. Comparative cryo-(S)TEM studies of Au nanoparticle doped 5CB prepared by "thin film" plunge-freezing (a) and cryo-ultramicrotomy (b and c).");sQ1[443]=new Array("../7337/0885.pdf","Optimized and Improved Immunogold Labeling on Ultrathin Sections of Nervous Tissue Following High Pressure Freezing, Freeze Substitution and Low Temperature Embedding","","885 doi:10.1017/S143192761500522X Paper No. 0443 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Optimized and Improved Immunogold Labeling on Ultrathin Sections of Nervous Tissue Following High Pressure Freezing, Freeze Substitution and Low Temperature Embedding Wei-Ping Li1, Sarada Viswanathan2, Yalin Wang1, Loren L. Looger2 & Patricia K. Rivlin1 1 2 Electron Microscopy Shared Resource, Howard Hughes Medical Institute/Janelia Research Campus, Looger Laboratory, Howard Hughes Medical Institute/Janelia Research Campus, Ashburn, VA 20147 Immuno-electron microscopy is a powerful technique for localizing proteins in cells and tissues. To enable labeling of ultrathin sections of nervous tissue, we developed an immunogold-silver labeling protocol that optimizes antigen detection and preserves neuronal ultrastructure. Specimens from different species including mouse, rat and dragonfly were processed with high pressure freezing, freezesubstitution, and embedded in Lowicryl HM20 resin at -50 (HPF/FS/HM20). Indirect immuno-gold labeling was applied on ultrathin sections for detection of target antigens. To visualize primary antibodies, secondary antibodies were conjugated to 5nm or 10 nm gold particles. Gold labeling was followed by strictly controlled silver-enhancement (i.e. duration and temperature). This approach is reliable and sensitive; a positive signal is readily observed at low magnification on the electron microscope (Figure 1). 5 nm and 10 nm gold particles are enlarged proportionally during silverenhancement while keeping their uniform shape (Figure 2). This makes this approach suitable for double labeling studies, especially when both antigens are located in the same cellular compartment. In our lab, this protocol was used successfully in co-localization analysis, quantification and 3D reconstruction with double labeling on serial ultrathin sections (Figure 3). We provide examples from our studies and compare the sensitivity of immunolabeling with different size gold particles at the light and electron microscopic level. Reference: [1] Sarada Viswanathan et al. High-performance probes for light and electron microscopy. Nature Methods (2015), in press. Microsc. Microanal. 21 (Suppl 3), 2015 886 882 Figure 1: Low magnification image of immunogold-silver labeling on mouse brain following HPS/FS/HM20. Bar = 2 �m. A B Figure 2: Double immunogold-silver labeling on mouse brain following HPS/FS/HM20. A) Localization of two target antigens in two neuronal dendrites. B) Co-localization of two antigens in one neuronal dendrite (upper). Bar = 200 nm. Figure 3. Double immunogold-silver labeling on mouse brain and serial 3D reconstruction of labeled dendrites. Bar = 1 �m.");sQ1[444]=new Array("../7337/0887.pdf","Agar Assisted Embedding: Visualizing Plant Pathogen Interactions at The Root - Soil Interface","","887 doi:10.1017/S1431927615005231 Paper No. 0444 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Agar Assisted Embedding: Visualizing Plant Pathogen Interactions at The Root � Soil Interface J. Kilcrease, J. Hammond United States Department of Agriculture, Agricultural Research Service, U.S. National Arboretum Floral and Nursery Plants Research Unit, Beltsville, MD 20705 Soil pathogens such as viruses are wide spread and have detrimental effects on many economically important plant species. A suite of plant viruses exist that are transmitted to plant root systems through soil or by soil-borne vectors including parasitic nematodes, fungi, and protists [1]. These types of viruses and the interactions between the virus, vector (if present), and host plant are not well known and prove to be difficult to study largely in part due to problems in concurrent analysis of plant and soil [1]. Visualizing interactions of virus-soil complexes that are unstable or easily damaged presents a challenge. The key to successfully imaging the delicate root-soil interface in its native state including host, virus, and vector is maintaining sample integrity throughout the fixation � dehydration � embedment processes. Therefore the objective of this research is to; 1) utilize a liquid agar fixative medium initially to aide in the stabilization of soil / biological sample complexes in their native conformation and 2) determine how Plantago asiatica mosaic virus (PlAMV) virus particles move from infected roots into surrounding soil. PlAMV is a member of the genus Potexvirus that affects a number of plant species, including lilies, causing mosaic or ringspot symptoms [2]. The Potexvirus group is named after one of its members potato virus X (PVX) and is for the most part homogenous in nature with filamentous virus particles that are 470-580 nm in length and approximately 12 nm wide with a single-stranded, positive sense RNA genome [2]. While PlAMV was first discovered in the Russian Far East, it was reported in 2010 in the Netherlands to reduce lily cut-flower production by almost 80% [3]. More recently the virus has been found on Asiatic and Oriental lilies imported to the United States and if not controlled could significantly impact domestic cut flower lily production [4]. A vector has not been found, but it has been reported that the virus can spread via mechanical transmission by root-to-root contact and through soil at the root-soil interface [5]. Transmission electron microscopy (TEM) methods were used to investigate the effects of PlAMV in root-soil samples of Nicotiana benthamiana and Asiatic / Oriental lilies (Lilium sp.). These findings demonstrate the architecture / interactions of the virus in the root and at the root-soil interface utilizing an agar assisted fixation protocol to stabilize the natural conformation of soil in direct contact with roots. Lilies with known PlAMV infection along with manually inoculated N. benthamiana plants were used and root balls with soil intact extracted from pots. A liquid solution of 3% agar in 0.05M phosphate buffer was prepared with glutaraldehyde added (after heating) to a final concentration of 3%. The solution was then dripped onto small feeder roots surrounded by an intact soil matrix. The agar fixative solution was allowed to solidify and the agar embedded section carefully removed from the root ball. This extracted area was placed into a vessel containing the same liquid agar fixative solution and held at 4�C overnight. Agar embedded root-soil samples (Fig. 2) were further trimmed, stained with osmium tetraoxide, and embedded into LX-112 resin. Ultrathin sections were obtained utilizing a DiATOME Ultra 45 knife and imaged with a JEOL JEM-100CX transmission electron microscope at 80kV. Phosphotungustic acid (PTA) was utilized to negatively stain leaf dips and root grinds from infected plants. Microsc. Microanal. 21 (Suppl 3), 2015 888 Successful confirmation of PlAMV infection in Lilium sp. and N. bentamiama was determined via leaf dips (Fig. 3), negatively stained soil preparations, and PCR (data not shown). The root-soil embedded tissue using the agar assisted procedure resulted in composite samples containing both root and soil entities with undisturbed architecture (Fig. 1 A-C). To date we have not detected virus extrusion or free PlAMV particles outside of the root wall, however this method of utilizing a liquid agar fixative solution has been demonstrated to be very useful in the TEM visualization of the rootsoil interface. It could also be used in instances of soil-borne vector and/or other pathogen interaction with roots as well as a stabilizer in pre-embedment processing for bacterial colony matrices, fungal networks, and other unstable targets. Further studies of viral interactions and architecture in soil-borne scenarios are likely to reveal more information about host-pathogen-vector etiology [6]. References: [1] A. G. Roberts, eLS - John Wiley & Sons Ltd, Chichester, Plant Viruses: Soil-borne (2014). [2] K. Komatsu, et. al. Arch. Virol. 153 (2008) 193-198 [3] Anonymous. https://www.vwa.nl/txmpub/files/?p_file_id=2001424, accessed Feb. 10, 2015. [4] J. Hammond, Plant Disease 99 (2015) 292. [5] C.G.M. Conijn, Acta Hort 1027 (2014) 213-229. [6] Acknowledgements: Funding by Oak Ridge Institute of Science and Engineering & USDA A 1 B 2 C 2 3 3 Fig. 1. TEM images of root and the root � soil interface in plants infected with Plantago asiatica mosaic virus. A: N. benthamiana root with PlAMV virus accumulation (1) near cell wall. B, C: Root � soil interface demonstrating undisturbed soil particulate (2) outside of root epidermal wall (3) of N. benthiama and Lilium sp. respectively. Scale bars = 500 nm. Fig. 2. (Left) Root-soil matrix of lily in native conformation embedded in agar - fixative solution prior to dehydration. Fig. 3. (Right) PlAMV virus particles from infected N. benthamiama leaf negatively stained with PTA. Scale bar = 500 nm.");sQ1[445]=new Array("../7337/0889.pdf","The Atmospheric Scanning Electron Microscope (ASEM) observes the Cultured Fluorescent Neuron","","889 doi:10.1017/S1431927615005243 Paper No. 0445 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Atmospheric Scanning Electron Microscope (ASEM) observes the Cultured Fluorescent Neuron Tatsuhiko Ebihara1, Hidetoshi Nishiyama2, Mitsuo Suga2 and Chikara Sato1, 1. 2. JEOL National Institute of Industrial Science and Technology (AIST), Tsukuba, 305-8566, Japan Ltd., 1-2 Musashino 3-chome, Akishima, Tokyo 196-8558, Japan Correspondence should be addressed to Chikara Sato (ti-sato@aist.go.jp) Neural network by neurite and synapse, is the bases of brain functions and development. The morphological changes of neurons and synapses are a lead to solution of the brain development and memory formation. Optical microscopic observation of cultured neurons, is important method for understanding brain. As synapses are around 1 m, optical microscopic observation faces difficulties for subsynaptic structures. One solution is super-resolution optical microscopy, the other, we subject the possibility of electron microscopic observation. The new Atmospheric Scanning Electron Microscope (ASEM) is Correlative Light-Electron Microscope (CLEM), in which the optical microscope (OM) and SEM quasi-simultaneously observe samples in water solution under open atmosphere [1]. In this system, an inverted SEM observes the underwater sample from beneath an open dish while an optical microscope (OM) observes it from above (Fig. 1). The disposable dish with an electron transparent SiN film window can hold a few milliliters of culture medium, and allows various types of cells to be cultured in a stable environment, e.g., a CO2 incubator. The film with poly-L-lysine coating allows primary neural culture of mouse hippocampus (Fig. 2). Since ASEM observes sample in water, samples need only brief pretreatment, but no treatment against vacuum is required. Immuno-staining is also as easy as those for normal OM [2]. ASEM would realize the easy EM observation and prevent the effects caused by the hydrophobic pretreatments for traditional EM. We developed the heavy metal staining conditions to visualize cell shape and intracellular structures. Double staining by uranyl acetate and tannic acid, stained plasma membrane and cell nuclei. Tungstic acid stained mitochondria and other fibers. We imaged primary neural cultures whose postsynaptic spines are transgenetically labeled by green fluorescence. In combination with OM and heavy metal staining (EM), fine structures of neurites could be distinguished in pre- and post- synaptic structures (Fig. 2). We also immunostained cultured neurons, and studied the distribution of structural proteins in growth cone. Microsc. Microanal. 21 (Suppl 3), 2015 890 References [1] H. Nishiyama et al., J Struct Biol 169, (2010) p. 438-449. [2] Y. Maruyama et al., J Struct Biol 180, (2012) p. 259-270. Figure 1. Configuration of the ASEM and cultured, fixed and stained cells on the ASEM dish. The SEM has a totally inverted structure, with electron gun at the bottom. Div.14 Cultured Neuron D B C 10m A 5m B 2m C 1m D Figure 2. Correlative imaging of cultured neuron (Day in vitro 14). (A) Phase contrast (B) Fluorescent image of rectangle area in (A). (C) EM image. (D) Merged image of OM and EM, the same area in Fig B's rectangle.");sQ1[446]=new Array("../7337/0891.pdf","Radiation Damage of Biological Specimen in Environmental Electron Microscopy","","891 doi:10.1017/S1431927615005255 Paper No. 0446 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Radiation Damage of Biological Specimen in Environmental Electron Microscopy Justus Hermannsd�rfer1 and Niels de Jonge1, 2 1. 2. INM � Leibniz Institute for New Materials, Saarbr�cken, Germany. Department of Physics, University of Saarland, Campus A5 1, 66123 Saarbr�cken, Germany Proteins labeled with nanoparticles can be studied in their native context of the plasma membrane in intact cells in liquid using scanning transmission electron microscopy (STEM) [1,2]. One approach is the usage of a microfluidic chamber enclosing the cell entirely in liquid and imaging at 200 keV beam energy [1]. An alternative approach is to study the cell maintained under a thin liquid layer using environmental scanning electron microscopy (ESEM) at 30 keV [2]. An important question is to which extent the effect of electron beam induced radiation damage changes the structure of cells in liquid. In order to examine what type of radiation damage is of influence, one should first look at the specific question that the microscopic study aims to address. We are interested in the dimer formation of the epidermal growth factor (EGF) receptor (EGFR), a transmembrane receptor playing a critical role in the pathogenesis and progression of many different types of cancer. In Comparison to super-resolution fluorescence microscopy, the enhanced resolution of STEM (with a spatial resolution of a few nanometres) enables us to distinguish between monomers, dimers and clusters of labelled proteins while the cell remains intact and in its natural liquid environment. However, due to its invasive nature, electron microscopy entails several complications like interactions of the electron beam and/or radiolysis reactions with the solvent. Hence, the main limiting factors for the imaging of biological samples are changes in the structural integrity of a cell caused by radiation damage. What matters in our experiment is that the locations of proteins are determined with sufficient resolution to examine the compositions of protein complexes and as close as possible to the native situation. Possibly, some radiation damage to the biological structure can be tolerated as long as the protein positions remain the same. In our case, we specifically label certain protein species with nanoparticles. Tolerance for radiation damage can thus be evaluated based on the shift of nanoparticles. According to literature the limit of radiation damage for biological molecules in a water vapor environment [3] equals ~1 e-/Ų, which we consider as the lower limit for unfixed cells in saline buffer solution. In a previous study, we demonstrated the ability to detect gold labeled EGFR proteins with a resolution of 4 nm in fixed cells for 200 keV Liquid STEM. The electron dose applied was in the range of 7�10� e-/Ų for which no radiation damage was observed [1, 2]. However, the effect of electron beam irradiation on cells in liquid has remained mostly unexplored. We examined the electron dose tolerance of COS7 cells in liquid. Gold nanoparticles of 10 nm diameter were adhered to EGFRs via a streptavidin-biotin bond. The cells were fixed with glutaraldehyde and kept in pure water during electron microscopy. Two experimental approaches were used. 1) Cells fully enclosed in liquid in a microfluidic device at 200 keV. 2) Cells maintained in an open wet environment using ESEM at 30 keV (Fig. 1). Several regions on different cells were repeatedly imaged in order to assess the effect of increasing electron dose. The electron dose limit for radiation damage was defined from the onset of shifts of the labels > 2 nm. It was found in our ESEM experiments that the label positions remained mostly unchanged between two images recorded with a total electron dose of 430 e/Ų (Fig. 2A). Yet, most of the nanoparticles at the edges of the image had shifted while recording a series of images with a total dose of 8815 e-/Ų (Fig. 2B). In conclusion, fixed COS7 cells can be studied in liquid with electron microscopy, while the positions of the labeled EGFRs do not significantly change for an electron dose sufficient to detect the labels. Microsc. Microanal. 21 (Suppl 3), 2015 892 References: [1] de Jonge et al, Proc. Natl. Acad. Sci. 106, (2009), p. 2159. [2] D.B. Peckys et al, Sci. Rep. 3 (2013), p. 2626. [3] V.R. Matricardi, R.C. Moretz, D.F.Parsons, Science 177, (1972), p. 268. [4] We thank E. Arzt for his support through INM. Research in part supported by the Leibniz Competition 2014. Figure 1. Experiment to test radiation damage of ESEM of a eukaryotic cell in hydrated state. Schematic of the setup. Live eukaryotic cells are grown on a supporting silicon nitride membrane and incubated with specific protein labels consisting of gold nanoparticles (AuNPs). Imaging is done by scanning a focused electron beam over the cell. Transmitted electrons are recorded with the STEM detector located beneath the sample. The cell is maintained in a saturated water vapor atmosphere (740 Pa, 3 �C), while a thin layer of water covers the cell. Figure 2. Micrographs of the same region on a COS7 cell with gold nanoparticles conjugated to epidermal growth factor receptors in the plasma membrane. (A) ESEM-STEM image showing the Au NP label as bright spots on the background shapes in grey of the biological material. The locations of Au NPs detected in a second image at the same position are shown in blue color. The two images were recorded with a total electron dose of 2x215 e-/Ų. Most Au NP positions overlapped between the two images, only a few streaks are visible at the top of the image indicating shifts. (B) Image showing the shifts of Au NPs after the recording of a series of images with a total dose of 8815 e-/Ų. The largest shifts were observed at the edges of the image. The mean grey value of the imaged area decreases with increasing dose due to the destruction of the eukaryotic cell. Scale bar: 500 nm.");sQ1[447]=new Array("../7337/0893.pdf","The 3D-Structure Analysis of Spermatids in the Seminiferous Epithelia by the Serial Block-Face SEM Method","","893 doi:10.1017/S1431927615005267 Paper No. 0447 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The 3D-Structure Analysis of Spermatids in the Seminiferous Epithelia by the Serial Block-Face SEM Method Yuji Hasebe1, Tomohiro Haruta1, Natasha Erdman2, Mistuo Suga1, Hideo Nishioka1, Toshiaki Suzuki1 1. 2. JEOL Ltd, 3-1-2 Musashino, Akisima, Tokyo 196-8558 Japan JEOL USA inc, 11 Dearborn Road Peabody, MA 01960 USA In a seminiferous epithelium, spermatogenesis is observed from a spermatogonium to sperm. In the spermatogenesis, cells undergo meiosis and mitosis while the spermatogonium differentiates into the sperm. These cell divisions are not complete and the cells are still connected by intercellular bridges even after cell divisions until sperm formations. In order to understand the network of the cells, it is necessary to analyze more than several tens of the cells, which are connected by the intercellular bridges. Therefore, it is difficult to analyze 3-dimensional structure of the network of the cells by TEM-tomography and confocal laser scanning microscopy (CLSM), because TEM does not have enough analysis area, and CLSM does not have enough spatial resolution. To solve this problem, we applied a serial block-face SEM (SBF-SEM) method1) to analyze the connections of the cells. In the SBF-SEM method, an ultramicrotome is installed in a specimen chamber of a SEM. The surface of a resin embedded specimen is cut at predetermined thickness, and the SEM image of the exposed specimen surface is captured. By repeating this process and stacking the captured SEM images, it is possible to reconstruct the 3-dimensional cell structure in high resolution at nm level for large volume of sub millimeter cubic. In this study, we used field-emission SEM, JSM-7100F (JEOL Ltd., Japan), equipped with the Gatan 3View 2XP(Gatan Inc., USA) (Figure 1). We took about 1,000 SEM images of the sliced seminiferous epithelium, and reconstructed its 3-dimensional structure. We also extracted the cells connected by the intercellular bridges from the reconstructed structure. The analyzed area of this data is 80 m (X direction), 90 m (Y direction), and 70 m (Z direction) (Figure 2). We recognized about 50 connected spermatids, and found the spermatid with 6 intercellular bridges, meaning number of cell division is 6 for this cell. We also inferred order of cell divisions for these 50 cells from number of the intercellular bridges. We report the details of the analysis in the presentation. References [1] Denk W. and Horstmann H., Serial block-face scanning electron microscopy to reconstruct three-dimensional tissue nanostructure. Plos Biol (2004), 2, P. 1900-1909. Microsc. Microanal. 21 (Suppl 3), 2015 894 Figure1. This figure shows JSM-7100F, equipped with the Gatan 3View 2XP. 20 m Figure2. This figure shows orthoslices of a seminiferous epithelium.");sQ1[448]=new Array("../7337/0895.pdf","Crustacean Cuticle: Synthesis and Remodeling of a Dynamic Extracellular Matrix During molt Cycle","","895 doi:10.1017/S1431927615005279 Paper No. 0448 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Crustacean Cuticle: Synthesis and Remodeling of a Dynamic Extracellular Matrix During molt Cycle Jasna Strus1, Nada Znidarsic1, Andrej Blejec1,2, Magda Tusek Znidaric2 1 University of Ljubljana, Department of Biology, Ljubljana, Slovenia 2 National Institute of Biology, Ljubljana, Slovenia Cuticle is mostly considered as a nonliving part of crustacean body functioning as a mechanical and permeability barrier in different environments. In this contribution we want to expose the idea that cuticle is a living part of the animal body perforated by numerous pore channels with cytoplasmic extensions delivering various components to the newly synthesized extracellular matrix which is constantly remodeled during molt cycle. Terrestrial isopod crustaceans developed an elaborate cuticular matrix with different degree of calcification as an adaptation to various land habitats, ranging from sea shores to deserts. Samples of surface cuticle were examined with transmission electron microscope in different molting stages of sea slaters Ligia palassii from Friday Harbor, WA, Ligia exotica from Wilmington, NC and Ligia italica from Piran Bay in Slovenia. Cytochemical localization of CaATPase [1] and EDX analysis was performed in cuticles at different molt stages. Ultrastructural study of cuticles reveals a highly dynamic structure of epidermal cells which secrete cuticle during molt cycle [2]. In premolt animals epidermal cells change shape and cell polarity is established by apical extensions of long cytoplasmic processes and short microvilli (Fig.1). Chitin microfibrils are secreted at the tips of microvilli and exocytotic vesicles are present in between. Exocuticle is secreted as loosely arranged layers of branching chitinous fibers forming a fan�like pattern. Proteinaceous component of the cuticle is secreted from small vesicles which bud off the Golgi complexes in the apical part of epidermal cells (Fig.2). Calcification of cuticle starts already in the premolt and continues after exuviation. The localization of Ca ATPase activity in spherules and epidermis demonstrates intense calcium fluxes in tergal integument of intramolt animals. In intermolt animals lead deposits were present in the exocuticular pore canals, along the basolateral membranes in mitochondria and vesicles of epithelial cells [3]. The results of ultracytochemical reactions were confirmed by X-ray microanalysis of tissues in intramolt animals. The analysis of smaller spherules at the base of the old cuticle revealed the presence of calcium and phosphorous while no calcium was detected in larger spherules close to the new cuticle. Lead precipitates were detected in the spherules but not in the matrix of the ecdysal space. Extensive nanotubular structures of 25 nm diameter were observed in the ecdysal space of intramolt animals which interconnect the spherules and newly formed epicuticle (Fig.3). They contain electron dense material at the sites of attachment to the epicuticle. Immunocytochemical analysis of nanotubules did not confirm the presence of tubulin in the ecdysal space. Thicker fibers extend from hypodermis, cross the newly formed cuticle and are connected to spherules in the ecdysal space (Fig.4). They were described as massive arrays of fibers running from tendon cells through the new cuticle and ecdysal space up to the distal layers of the detached cuticle in adult premolt L. italica and in premolt intramarsupial specimens of P. scaber. [4]. Our results show that crustacean cuticle is a very dynamic extracellular matrix which is frequently renewed and constantly remodeled in terrestrial isopod crustaceans. Epidermal cells with their Microsc. Microanal. 21 (Suppl 3), 2015 896 cytoplasmic processes extending through cuticular pore channels provide synthesized materials for the construction of the new cuticle. At the same time the materials resorbed from the old cuticle accumulate in the spherules of the ecdysal space with extensive nanotubular structures which are presumably involved in the formation of additional epicuticular layers [5]. References: [1] Ando T, et al, Acta Histochem cytochem 14 (1981), p. 705. [2] Strus J and A Blejec In "Crustacean Issues 13", Eds. B.Kensley and R.C.Brusca, (A.A.Balkema Publishers, Rotterdam) p. 343 [3] Strus J and P Compere (1996). Pflugers Arch - Eur J Physiol 431, p. 251 [4] Znidarsic N et al, Zookeys 176 (2012), p.39. [5] The authors acknowledge funding from the SRA and Short term visiting Scholarship in Friday Harbor Laboratories within UW-Ljubljana University Exchange Project, Grant ER176713. Fig. 1 Ligia italica in premolt stage: epithelial cell (EC) with exocytotic vesicles (EV) covered by a three-layered new cuticle (En-endocutiocle, Ex-exocuticle, Ep-epicuticle) and the old cuticle (OC) above cuticular scale (CS) is detached. Fig.2. Ligia italica in premolt stage: Epithelial cells (EC) with microvilli (M), cytoplasmic extensions (CE) and exocytotic vesicles (EV) secrete components of the new cuticle (NC) Fig.3 Ligia palassii in intramolt stage: nanotubular structures (NT) with electron dense tips at the surface of epicuticle (Ep) Fig.4. Ligia palassii in intramolt stage: tendinous fibers (TF) extending through the exocuticle (Ex) and epicuticle (Ep) connecting to the spherules (S) in the ecdysal space (ES)");sQ1[449]=new Array("../7337/0897.pdf","Structure and Assembly of the Bacillus anthracis Exosporium","","897 doi:10.1017/S1431927615005280 Paper No. 0449 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Structure and Assembly of the Bacillus anthracis Exosporium Cynthia M Rodenburg, Sylvia A McPherson, Charles L. Turnbough Jr., Terje Dokland Department of Microbiology, University of Alabama at Birmingham, Birmingham, AL, USA Bacillus anthracis, the causative agent of anthrax, is a highly pathogenic Gram-positive soil bacterium that is also considered a serious threat as a potential bioterrorism agent. Like other bacilli, B. anthracis forms protective spores upon starvation. The spores are the main pathogenic agents, easily spread and highly resistant to environmental stress. When the spores enter a mammalian host, they germinate and undergo vegetative growth, followed by production of anthrax toxin, which leads to systemic bacteremia and toxemia that is often fatal. The spore is a complex, multilayered structure, consisting of a nucleoid surrounded by a peptidoglycan cortex and a proteinaceous coat [1]. Spores of B. anthracis and other pathogenic bacilli are surrounded by an exosporium, a balloon-like layer that acts as the outer permeability barrier of the spore and is thought to contribute to spore survival in the environment, uptake by potential hosts, and as a virulence determinant. The exosporium consists of a hair-like nap and a paracrystalline basal layer [1]. The filaments of the nap are comprised of trimers of the collagen-like glycoprotein BclA, while the basal layer contains at least 20 different proteins. One of these proteins, BxpB, forms tight complexes with BclA and is required for attachment of essentially all BclA filaments to the basal layer [2]. Another basal layer protein, ExsB, is required for the stable attachment of the exosporium to the spore [3]. ExsY is required to complete exosporium assembly during sporulation [4]. The exosporium also harbors several enzymes, including alanine racemase (Alr), which functions as an anti-germinant, presumably functioning to prevent premature germination. We have used a combination of cryo-electron microscopy (cryo-EM), cryo-sectioning (Tokuyasu method) and crystallographic analysis of negatively stained exosporium fragments to compare wildtype spores and spores with deletions of specific exosporium genes, including bclA, bxpB, exsB and exsY, to study the structure and assembly of the B. anthracis exosporium [5]. Cryo-EM reveals the intact spore in its native form and shows clearly the bilayered basal layer and the nap filaments with their distal knobs (Fig. 1A, B). The filaments appear to connect to protrusions in the basal layer, more clearly seen in negatively stained cryo-sections (Fig. 1C). bclA spores lack the nap, while bxpB spores lack both the nap and the protrusions, but retain the bilayered basal layer (Fig. 1D, E). The exosporium of exsB spores is more fragile and is easily sloughed upon grown in liquid media, but looks otherwise completely normal and contains a normal complement of BxpB protrusions and BclA filaments (Fig. 1F). Taken together with crystallographic analysis of isolated exosporium fragments, these data suggest that the protrusions are made of trimers of BxpB (Fig. 1G, H). The protrusions interact with a crystalline layer of hexagonal subunits formed by other, still undetermined basal layer proteins (Fig. 1G). Although bxpB spores retain the hexagonal subunits, the basal layer is not organized with crystalline order and lacks basal layer protrusions and most BclA filaments, indicating a central role for BxpB in exosporium organization. Microsc. Microanal. 21 (Suppl 3), 2015 898 References: [1] Henriques, A.O. and Moran Jr., C.P., (2007). Annu. Rev. Microbiol. 61, 555�588. [2] Steichen, C. T., Kearney, J. F., and Turnbough, C. L., Jr. (2005). J. Bacteriol. 187, 2868-5876 [3] McPherson, S., Li, M., Kearney, J. and Turnbough Jr., C.L. (2010). Mol. Microbiol. 76, 1527�1538. [4] Boydston, J.A., Yue, L., Kearney, J.F., and Turnbough, Jr., C.L. (2006) J. Bacteriol. 188, 7440�7448 [5] Rodenburg, C.M., McPherson, S.A., Turnbough Jr., C.L. and Dokland, T. (2014). J. Struct. Biol. 186, 181-187. Figure 1. (A) Cryo-electron micrograph of the wildtype B. anthracis spore. (B) Cryo-EM (high-pass filtered) and (C) negatively stained cryo-section of the wildtype exosporium, showing the nap (n) and basal layer (bl). (D�F) Cryo-sections of the bclA (D), bxpB (E) and exsB (F) exosporium. (G) 2D crystal average of the negatively stained, wildtype exosporium, showing the hexagonal subunits. BxpB trimers are indicated in purple. (H) Model for the arrangement of BxpB (purple) and BclA (green).");sQ1[450]=new Array("../7337/0899.pdf","Tomographic Analysis of EmaA Adhesin Glycosylation in Aggregatibacter actinomycetemcomitans","","899 doi:10.1017/S1431927615005292 Paper No. 0450 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Tomographic Analysis of EmaA Adhesin Glycosylation in Aggregatibacter actinomycetemcomitans Alison Watson1, Hannah Naughton1, Michael Radermacher1, Keith P. Mintz2, and Teresa Ruiz1 1 2 Departments of Molecular Physiology & Biophysics, University of Vermont, Burlington, VT Microbiology & Molecular Genetics, University of Vermont, Burlington, VT The numbers of antibiotic-resistant bacteria has rapidly increased in recent years, thus current research is directed at developing new therapies targeting colonization. Host adherence is frequently considered the - first step in the infectious process and is crucial for colonization. A. actinomycetemcomitans is a Gram bacterium involved in periodontal diseases and other systemic infections (e.g. pneumonia, and endocarditis). These infections are associated with numerous virulence determinants, including extracellular matrix adhesins. EmaA (extracellular matrix protein adhesin A), a trimeric autotransporter adhesin, mediates the binding of this bacterium to collagen. EmaA forms antennae-like structures that protrude from the bacterial surface. The 30 nm distal region of the structure, which is further divided into three subdomains (SI, SII, SIII), contains the N-terminal sequences of the three monomers and encompasses the collagen binding activity [1]. Deletion of this region or any of the subdomains abrogates collagen binding. Furthermore, EmaA is glycosylated, and this post-translational modification is essential for the function of adhesin. Interestingly, this glycoprotein uses a similar synthesis pathway as the O-polysaccharide (O-PS) synthesis of lipopolysaccharides (LPS). Disabling sugar synthesis, transport, and glycan ligation in this pathway changes EmaA function [2]. Yet, the link between the modifications of the protein by glycosylation remains unclear. The aim of this study is to analyze the 3D structure of the functional domain of the EmaA adhesin from a mutant strain with a disrupted glycosylation mechanism. The waaL mutant strain lacks the O-antigen ligase, WaaL, which is an essential component of the O-PS glycosylation pathway. Structural comparison of the glycosylated and non-glycosylated adhesins will help to determine the structural role of this modification in collagen binding. Bacteria, grown as previously described [3], were adsorbed onto hydrophilic carbon-coated grids that had been pretreated with a colloidal gold solution (fiducial markers) and were negatively stained with Nano-W. Tomographic single-axis tilt series (~ 50) were acquired over a �64� angular range in 2� intervals using a Tecnai12 electron microscope at 42000x nominal magnification, which corresponds to a calibrated 3.08 �/pixel at the specimen scale (Fig. 1A). Tilt series were processed using IMOD [4] to generate tomograms (Fig. 1B). Individual EmaA adhesins (>30) were selected for further processing. Reconstructions of EmaA adhesin subvolumes (Fig. 2) were calculated using EMIRA and visualized in Chimera [5,6]. Subvolumes were low-pass filtered and aligned to a reference volume of the wild type EmaA adhesin (Fig. 2, center). Alignment parameters were transferred and applied to each of the unfiltered subvolumes and an average was calculated. This process was repeated multiple times to refine the translational and rotational shift parameters. Aligned subvolumes were averaged together and compared to the wild type EmaA volume (Fig. 2, right). We observed that the waaL mutant cells produced fewer EmaA adhesins than the wild type cells. This is consistent with what has been previously described [2], and indicates that glycosylation of EmaA may be significant for the stability, as well as function of the adhesin. Glycosilation, however, does not seem to affect the characteristic bending pattern of the antennae-like structures. Based on the present analysis, the structure of the EmaA adhesins present in the waaL mutant A. actinomycetemcomitans strain does Microsc. Microanal. 21 (Suppl 3), 2015 900 not appear to be significantly different from wild type. The functional region maintains the complex architecture composed of 3 subdomains (SI-SIII). SI has a globular shape ( =5 nm), while SII and SIII have a cylindrical shape and are separated by a linker region of 3 nm in length. However, there appears to be a subtle bend between the SI-SII regions (Fig. 2, right). This bend is reminiscent of structural changes observed in a different the G162S substitution mutant strain that exhibits greatly reduced collagen binding activity [3]. Currently, we are increasing the number of subvolumes in the data set to resolve the structure at higher resolution to better evaluate the significance of this potential structural difference. References: [1] Yu, C., Mintz, K.P., Ruiz, T., J. Bacteriol. 191 (2009), p.6253. [2] Tang, G., Ruiz, T., Mintz, K.P., Infect. Immun. 80 (2012), p.2868. [3] Azari, F., Radermacher, M., Mintz, K.P., Ruiz, T., J. Struct. Biol. 177 (2011), p.439. [4] Kremer, J.R., Mastronarde, D.N., McIntosh, J.R., J. Struct. Biol. 116 (1996), p.71. [5] M. Radermacher, Micoscopy & Microanalysis, 19 (2013), Supplement S2 p.762. [6] Petterson, E.F., Goddard, T.D., Huang, C.C., et al., J. Comput. Chem. 25 (2004), p.1605. [7] This work was supported by NIH grant DE024554, (T.R. & K.P.M) and benefitted from developments from NIH grant GM078202 (M.R.). Figure 1. Electron tomography of the waaL mutant bacteria: A) 0� projection of the tilt-series; B) projection of the central slices of the reconstructed tomogram. Arrows point to EmaA. Bar = 150 nm. Figure 2. Alignment and averaging of waaL mutant bacteria EmaA subvolumes extracted from tomograms: left) 0� projection and surface representation of two adhesin subvolumes; center) depiction of alignment procedure to the wild type reference subvolume (purple, [1]); right) average of selected subvolumes. Reference volume is shown as a purple mesh overlay. Bar = 10 nm.");sQ1[451]=new Array("../7337/0901.pdf","3D Reconstruction of Mitochondrial Complex I Analyzed After Biogenesis in the","","901 doi:10.1017/S1431927615005309 Paper No. 0451 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 3D Reconstruction of Mitochondrial Complex I Analyzed After Biogenesis in the Absence of Assembly Factor N7BML (NDUFAF2) Christine T. Nolan1, Katarzyna Kmita2, Volker Zickermann2, Teresa Ruiz1, Michael Radermacher1 1. 2. Department of Molecular Physiology and Biophysics, University of Vermont, Burlington VT, USA Institute of Biochemistry II, Medical School, Goethe University, Frankfurt/M, Germany Complex 1(NADH:ubiquinone oxidoreductase) is the first and largest electron transport chain enzyme in bacteria and in the inner membrane of mitochondria. Complex 1 consists of two arms forming an "L" shape. One arm lies in the inner membrane of the mitochondria, while the other protrudes out into the mitochondrial matrix. After the matrix arm binds NADH, two electrons are transferred via a series of iron-sulfur clusters to ubiquinone with concomitant translocation of four protons across the membrane. The reduced ubiquinone is then transferred to complex III, cytochrome C is reduced and further processed in complex IV. In the process of oxidizing NADH, a total of 10 electrons are translocated across the membrane, creating a membrane potential that powers ATP production by complex V. The minimal bacterial complex I contains 14 subunits that are preserved through all species. Eukaryotic complex 1 has more than 21 additional (accessory) subunits and possesses a molecular weight of approximately 1 MDa. The function of a large number of accessory subunits is still unknown. Complex I assembly is a multi-step process supported by assembly factors that transiently bind to subassemblies of the enzyme [1]. At least five assembly factors are involved in the biogenesis of eukaryotic complex I. NDUFAF2 is involved in the late stages of complex I assembly, specifically in the final assembly of the matrix arm, although this factor is not absolutely essential for the process. The absence of NDUFAF2 seems not to affect the ubiquinone reductase activity of the assembled complex I, however, in N. crassa mitochondria an increase of the production of reactive oxygen species has been reported [2]. N7BML is a yeast ortholog of NDUFAF2 [3]. We have analyzed purified complex I from the yeast Yarrowia lipolytica, lacking assembly N7BML (nb7ml). Mitochondrial extracts showed significantly decreased complex I content, but the purified complex showed a ubiquinone reductase activity comparable to wild type [3]. All subunits were present in the assembled complex as determined by dSDS PAGE and proteomic analysis [3]. Structural studies were performed by 3D electron microscopy of single particles, using the random conical tilt reconstruction technique. The sample was applied to carbon coated grids at a concentration of 0.0015 mg/ml and stain embedded with 2% PTA. Pairs of images were recorded with a nominal magnification of 52kX and tilt angles of approximately 55� and 0� using a 100 kV accelerating voltage. Images were scanned on a SCAI flatbed scanner with 7 �m pixel size, then binned by a factor of 3, which resulted in a 4.02 � final pixel size at the specimen scale. Image processing was carried out using SPIDER version 5.0 with extensions. A total of 9815 pairs of particles were selected. The 0� images were processed first by a single-reference alignment followed by correspondence analysis. This process was continued with three iterations of multi-reference alignment, each followed by correspondence analysis and classification. Nine final classes were obtained, and nine 3D reconstructions were calculated from the corresponding tilt images. The 3D reconstructions were refined using 3D reference based projection alignments based on Radon transforms [4]. Microsc. Microanal. 21 (Suppl 3), 2015 902 Five of the reconstructions (1-4,6) closely resemble the wild type enzyme, with small conformational variations and differences in orientation. Classes 7-9 have low occupancy and are consequently not well defined. Class 5 shows a mixture of particles that need further analysis. At the current resolution no significant structural differences can be observed between the mutant and wild type complex I [5], indicating that the lack of assembly factor 2 does not introduce major structural changes. Our findings are consistent with the observation that nb7ml complex I was completely assembled and showed close to normal ubiquinone reductase activity [3]. Higher resolution might still reveal small structural differences that could affect the stability or regulation of complex I without affecting its basic function. References: [1] McKenzie M, Ryan MT, IUBMB Life 62 (2010), p497 [2] Pereira B., Videira A., Duarte M., Mol. Cell. Biol. 33 (2013), p2623 [3] Kmita et al., submitted [4] Radermacher M., Scanning Microscopy 11 (1997), p171 [5] Radermacher M. et al. J Struct Biol. 154(3) (2006), p269. [6] The authors acknowledge funding from NIH, grant 2RO1 GM068650 to M.R, and from DFG grants ZI 552/3-1 to V.Z. and EXC 115 to V.Z. a b c Figure: a) Averages of the nine classes of the 0�images of nb7ml complex I. Numbers in the upper right corner are the number of images in the classes. b) Surface representation of the volumes calculated from the tilt images of the nine classes. c) Comparison of the holo enzyme (left) and the volume of class 6 (right), both low-pass filtered to the same resolution (35�). At this resolution, the two volumes show no significant differences, confirming that complex I is fully assembled. Differences visible are within the range of normal flexibility observed in samples of the complex. Scale bars 10nm.");sQ1[452]=new Array("../7337/0903.pdf","Three-Dimensional Visualization of Herpesvirus Envelopment at High Resolution Using STEM Tomography and Serial Sectioning on High Pressure Frozen/Freeze-Substituted Cells","","903 doi:10.1017/S1431927615005310 Paper No. 0452 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Three-Dimensional Visualization of Herpesvirus Envelopment at High Resolution Using STEM Tomography and Serial Sectioning on High Pressure Frozen/FreezeSubstituted Cells Martin Schauflinger1, Clarissa Villinger2,3, Jens von Einem2 and Paul Walther3. 1. 2. Electron Microscopy Core Facility, University of Missouri, Columbia, Missouri, USA. Institute of Virology, University Medical Center Ulm, Ulm, Germany. 3. Central Facility for Electron Microscopy, Ulm University, Ulm, Germany. Human cytomegalovirus (HCMV) is a clinically highly relevant Herpesvirus. Lacking a suitable animal model, HCMV's strict species specificity is a major obstacle for the characterization of biological and pathogenic viral phenotypes. Electron microscopy is a suitable tool to characterize viral phenotypes, unraveling viral morphogenesis. We used advanced EM technologies to visualize the HCMV final envelopment processes in the viral assembly complex. Three-dimensional visualization using STEM tomography and serial sectioning was used to study HCMV final envelopment at high resolution. Infected cells were rapidly cryo-immobilized by high pressure freezing in order to overcome the disadvantages of traditional chemical fixation. High pressure freezing in combination with freeze substitution grants superior preservation of subcellular structures and is favorable for the examination of dynamic processes like virus envelopment. STEM tomography was applied as described [1] to visualize virus-vesicle interactions with high resolution (Fig. 1). A limiting disadvantage of STEM tomography is that only a relatively small volume (the Z height is limited to ~1 micron) of the infected cell can be imaged. The more traditional method of serial sectioning was used for three dimensional visualization of the whole HCMV assembly complex [2] (Fig. 2). The viral assembly complex is the specific area in the cytoplasm of HCMV infected cells in which final envelopment of viral particles occurs. The assembly complex exhibits distinct local membrane compositions: early endosomes in its central region, and vesicles characterized by Golgi markers in its periphery. This led us to speculate that budding events might occur preferentially in either the central or the peripheral area of the assembly complex. Using the 3D high resolution afforded by both STEM tomography and serial sectioning on high pressure frozen HCMV-infected cells, it was possible to determine the distribution of enveloped and non-enveloped virus particles throughout the area of the assembly complex. Contrary to our hypothesis, quantitative analysis revealed that the events of final envelopment are equally distributed within the assembly complex irrespective of the local membrane composition [2, 3]. References: [1] Villinger et al. in "Electron Microscopy", ed. J Kuo (Humana Press, Totowa), pp. 617�638. [2] Schauflinger M, Villinger C, Mertens T, Walther P, and von Einem J. Cell Microbiol 15 (2013), 305�314. [3] This work was supported by the DFG priority research program SPP1175. Microsc. Microanal. 21 (Suppl 3), 2015 904 Figure 1. Three-dimensional reconstruction of membranes and viral capsids from STEM tomography of HCMV-infected fibroblasts at 5 days post infection. Areas with high density, such as membranes and capsids, occur bright in the dark-field STEM tomogram. HCMV capsids before (lower right), during (black arrow), and after (white arrow) final envelopment were reconstructed and superimposed on a single slice of the tomogram. The leaflets of viral membranes are depicted in blue and grey respectively; viral capsids are indicated in red. Red lines indicate the cut-away plane in which the segmented structures have been revealed to better visualize the membranes around the capsids. Bar 200 nm. Figure 2. Micrographs from the HCMV assembly complex were taken with a nominal magnification of 12,000x from a total of 28 sections and aligned into an image stack. Shown here is a single micrograph in X�Y direction, while the image stack is indicated by a slice through the Z-axis. Cellular and viral structures were segmented on all sections and superimposed. Enveloped virions (green), non-enveloped particles (red), multivesicular bodies (transparent blue). Bar 1 um.");sQ1[453]=new Array("../7337/0905.pdf","Dictionary Based Reconstruction of the 3D Morphology of Ebola Virus","","905 doi:10.1017/S1431927615005322 Paper No. 0453 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Dictionary Based Reconstruction of the 3D Morphology of Ebola Virus Alaa AlAfeef 1,2, W. Paul Cockshott1, and Ian Maclaren2 1 2 School of Computing Science, University of Glasgow, Glasgow, G12 8QQ, UK SUPA School of Physics and Astronomy, University of Glasgow, Glasgow, G12 8QQ, UK The filoviruses Ebolavirus (EBOV) and Marburgvirus cause fatal hemorrhagic fevers in humans and other mammals, known as Ebola virus disease (EVD), with mortality rates approaching 90% [1] which makes it listed as world health organization risk group 4 pathogen. Ebola virus caused the 2013�2015 epidemic in West Africa, which has resulted in at least 9,268 confirmed deaths and 23,034 suspected cases as of 18 Feb 2015 [2]. Consequently, considerable effort is focused on developing therapeutics and vaccines to prevent infections. Cryo electron tomography (cryo-ET) is now an increasingly important technique for obtaining structural details of complex viral component organization at subnanometer resolution. Specifically, with reference to EBOV, Cryo-ET can be used for studying the arrangement of the internal viral components of EBOV such as the glycoprotein spikes (GP) on the surface of EBOV which are the target of multiple neutralizing antibodies [3]. This is vital for improved understanding of the principles of infection, strategies to combat the viral infection. Electron Tomography (ET) typically involves the acquisition of a set of two-dimensional projection images by rotating the specimen over a tilt range (typically restricted to +/- 70�) using bright field Transmission Electron Microscopy (BF-TEM), followed by alignment and reconstruction using established algorithms to reconstruct a 3D volume that represents the physical morphology of the specimen under investigation. However, reconstruction still suffers from distortion (especially the missing wedge artifacts) due to the limited range of angles sampled. Also, specifically for tomography on sensitive biological materials, low signal-to-noise ratio is a problem due to the need to limit the electron dose over the entire tilt series, resulting in noisy 3D representations. Compressed sensing (CS) has recently been applied to ET, and such approaches are both more noise tolerant and reduce out of plane distortions from the missing wedge [3]. In this work, we reconstruct a cryo-tilt series acquired at approximately 2� intervals from �60 of an Ebola entry-competent virus-like particles (VLPs), using datasets published previously in reference [6], using a new compressed sensing based algorithm, DLET [5]. These results are compared to a 3D reconstruction of the same datasets obtained using the well-known SIRT algorithm. DLET utilizes a dictionary-based learning approach and is very noise-tolerant, allowing reliable reconstruction from far fewer projections than are normally required by traditional ET methods. Figure 1 shows a comparison between SIRT and DLET for reconstruction using tilt series 1. Please note that this and the subsequent figure are produced without any segmentation to show the raw results of the reconstruction and to emphasize the clear difference between the results of the two methods. In Figure 1 it is clear, and it can be noted that DLET gives a noticeable much clearer visibility of the details of the virus and its morphology, including the clearly visible GP spikes surrounding particles it. Figure 2 shows a volume rendering of another reconstruction for a second other tilt series. Here the DLET shows a more much clearer reconstruction with a significant decrease of reconstruction noise. Consequently, Microsc. Microanal. 21 (Suppl 3), 2015 906 we show that the use of this dictionary-based reconstruction algorithm gives major benefits for the reconstruction the 3D volume of viruses from ET datasets of limited tilt range, relatively low signal to noise, and low contrast levels. It is anticipated that the application of this technique will present major advantages in discerning the nanoscale details of virus structure and infection behavior as compared to established tomographic reconstruction algorithms. References: [1] Feldmann, Heinz, and Thomas W. Geisbert,The Lancet 377.9768 (2011), p.849. [2] World Health Organization, "Ebola Situation Report - 18 February 2015" (2015). [3] Qiu, Xiangguo, et al., Sci. Transl. Med., 4.138 (2012), p.138ra81. [4] Saghi, Zineb, et al., Nano letters 11(2011), p.4666. [5] A. AlAfeef et al., J. Phys.: Conf. Ser 522 (2014), P.012021. [6] Tran, Erin EH, et al., Journal of virology, 88.18 (2014),p.10958. [7] AAA acknowledges funding for a PhD studentship from the Lord Kelvin/Adam Smith Scholarship of the University of Glasgow. This work was made possible by generous help from Erin E. H. Tran and Sriram Subramaniama from National Institutes of Health, Bethesda, Maryland, USA. a b c Figure 1. a) 0 Projection of Ebola-VLP b)XY-Orthoslice obtained through the constructed volume using SIRT and c)DLET. The GP spikes are clearly visible surrounding particles as indicated by arrows a in (c). b c Figure 2. a) 0 Projection of Ebola-VLP and gold fiducial markers (b) Volume rendering of reconstructed volume using SIRT and (c) DLET. Smearing due to markers is reduced in (c).");sQ1[454]=new Array("../7337/0907.pdf","Whole-Cell Tomography Using a Conventional Scanning Electron Microscope","","907 doi:10.1017/S1431927615005334 Paper No. 0454 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Whole-Cell Tomography Using a Conventional Scanning Electron Microscope Taiga Okumura1, Minami Shoji1, Akiko Hisada1, Yusuke Ominami2, Tatsuo Ushiki3, Masato Nakajima3 and Takashi Ohshima1 1. 2. Hitachi, Ltd., Central Research Laboratory, Tokyo, Japan Hitachi High-Technologies Corporation, Ibaraki, Japan 3. Graduate School of Medical and Dental Sciences, Niigata University, Niigata, Japan Traditionally, ultrastructure of biological samples has been elucidated predominantly using transmission electron microscopy (TEM). For TEM observation, the samples are embedded in resin and cut into ultrathin sections using an ultramicrotome. These sample preparation procedures, however, require skills to make adequate electron-transparent thin sections on a small grid, which is obstacle to TEM examination. To overcome the difficulty, we developed plate-transmission electron microscopy (PlateTEM), by which internal structure of biological samples is observed using a transmitted electron (TE) mode of scanning electron microscopy (SEM) with a scintillator plate [1,2]. Furthermore, whole-cell tomography was conducted using this technique in order to image the three-dimensional (3D) structure of the samples. In this technique, the samples on a transparent scintillator plate are irradiated with an incident electron beam, and the plate emits scintillation light (photons) produced by electrons that pass through the samples (Fig. 1). Some of the electrons are backscattered and absorbed by dense materials inside the samples, which means that the quantity of penetrating electrons and emitted photons varies with the density of the samples. Thus Plate-TEM achieves TE imaging without the sample-sectioning process. In addition, the sample surface can be observed simultaneously by detecting secondary and/or backscattered electrons (SEs/BSEs). Plate-TEM cannot directly reveal the 3D structure because the compositions of biological samples overlap with each other in TE images. Therefore we made a new sample stage to tilt the samples for Plate-TEM tomography. The new tilt-stage was fabricated by installing a small motor on a sample stage and attaching a scintillator plate to the motor shaft. Photons are emitted from the underside of the plate and detected by a photodetector. A tilt-series (�60� with an increment angle of approximately 2�) was acquired using a conventional SEM. Image alignment and 3D reconstruction were carried out with IMOD [3], and image visualization was performed with Avizo (Mercury Computer Systems). Indian muntjac thymus cells (Mm2T) were examined using the Plate-TEM tomography. The cells were cultivated on the scintillator plates, fixed in glutaraldehyde, postfixed in osmium tetroxide, dehydrated in an ethanol series, critical point-dried, and coated with carbon to avoid specimen charging. Figure 2A and 2B shows the SE and Plate-TEM images of the same cell, respectively. The cell surface was observed in the SE image, whereas several organelles such as mitochondria were recognized in the Plate-TEM image. Furthermore, the tomography with Plate-TEM images generated the volume-rendered 3D model (Fig. 3A) and the slice image (Fig. 3B). Thus this technique enables us to inspect 3D distribution and forms of organelles from various viewing angles. In conclusion, Plate-TEM tomography is a valuable technique for exploring 3D internal structure of biological samples using a conventional SEM without destructive sample-sectioning processes. Microsc. Microanal. 21 (Suppl 3), 2015 908 References: [1] Y Ominami et al, Scanning Microscopies (2014) 923608. [2] Y Ominami et al, Microscopy and Microanalysis (2014) 32�33. [3] JR Kremer, DN Mastronarde and JR McIntosh, J. Struct. Biol. 116 (1996) 71�76. Electron detector Incident electron Transmitted electron SE/BSE Cell Scintillator plate Photodetector Tilt for tomography Photon Figure 1. Schematic of Plate-TEM. A B 5 m 5 m Figure 2. SE (A) and Plate-TEM (B) images of the Mm2T cell. A B Figure 3. Volume-rendered 3D model (A) and Slice image (B) of the Mm2T cell.");sQ1[455]=new Array("../7337/0909.pdf","3D Imaging of the Early Embryonic Chicken Heart After Altered Blood Flow with Focused Ion Beam Scanning Electron Microscopy","","909 doi:10.1017/S1431927615005346 Paper No. 0455 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 3D Imaging of the Early Embryonic Chicken Heart After Altered Blood Flow with Focused Ion Beam Scanning Electron Microscopy Madeline Midgett1, Claudia S. L�pez2, Sandra Rugonyi1 1. 2. Biomedical Engineering, Oregon Health & Science University, Portland Oregon, USA Multiscale Microscopy Core, OHSU Center for Spatial Systems Biomedicine, Oregon Health & Science University, Portland Oregon, USA Introduction: Embryonic heart formation is a finely orchestrated interplay between genetic and environmental factors, with rapid transformations that occur at the tissue, cell, and subcellular levels. Myofibrils play an essential role in cardiac tissue remodeling as hemodynamic load increases throughout development. Myofibrils mature by aligning in the direction of stretch and increasing the amount of contractile units [1]. Concurrently in early development, a portion of endocardial cells in the cushion bulges of extracellular matrix transform into mesenchymal cells that invade the cardiac jelly through the process of endothelial mesenchymal transition (EMT). As EMT progresses, the original plump and tightly organized endocardium disassembles and rearranges to form an intact flattened layer [2]. Endocardial and myocardial cell changes during development ultimately shape the heart. Altered blood flow is known to lead to congenital heart defects. However, the early remodeling that precedes defects in the mature heart has not been fully characterized. This study used a well-established hemodynamic intervention called outflow tract (OFT) banding in the chicken embryo to constrict the diameter of the early embryonic OFT, the distal portion of the developing heart. Banding increases the hemodynamic load (pressure and shear stresses) on cardiac tissues. 3D shape and organization of endocardial cell and myofibril components of the OFT were then characterized to determine the ways in which normal tissue remodeling is detrimentally modified by increased hemodynamic load. Methods: Fertilized White Leghorn chicken eggs were incubated until stage HH18. 10-0 nylon suture was passed under and secured in a knot around the mid-section of the OFT in banded embryos. Embryos were collected at HH24 (~24 hrs later) and processed for transmission electron microscopy (TEM) and focused ion beam scanning electron microscopy (FIB-SEM) techniques. TEM imaging was performed on a FEI Tecnai 12 Spirit system interfaced to an EagleTM 2K CCD multiscan camera. FIB-SEM imaging was done on a FEI Helios 650 NanoLabTM DualBeamTM. Each embryo was removed from the egg and immediately immersed in fixative to maximally preserve and contract the tissue, and then processed as previously described [3]. TEM imaging was used to identify areas of interest, and FIBSEM was used to acquire serial (3D), high-resolution images [3]. Image stacks were then aligned using Amira software to create isotropic image volumes. Endocardial cells and myofibril material were manually segmented from the images in order to reconstruct and quantify the 3D shape and orientation. Results and Discussion: TEM and FIB-SEM images demonstrated a disrupted OFT endocardium in banded tissue compared to the control (Figure 1). Control tissue showed a tightly arranged endocardium with plump and round cells, while the altered arrangement in banded tissue displays hallmarks of a more advanced EMT configuration. 3D endocardial cellular reconstructions display dramatic cell projections and gaps between cells in the banded embryo compared to the control (Figure 1 B, D). This suggests that the altered hemodynamic load after banding modifies the progression of EMT in the OFT. The degree of cell elongation was quantified using Amira software, where the elongation factor of 3D cellular surfaces Microsc. Microanal. 21 (Suppl 3), 2015 910 (based on a ratio of eigenvector values) of banded endocardium (n=1 embryo) was 60% larger than control endocardium (n=1 embryo). Segmented myofibril material from FIB-SEM image stacks through the OFT myocardium demonstrated an altered myofibril alignment in banded tissue compared to the control (Figure 2). Control tissue had disordered myofibril orientations, while the myofibrils in banded tissue appeared more aligned and organized circumferentially around the OFT. This architectural myofibril reorganization indicates that the altered hemodynamic load after banding modifies the normal alignment processes of myofibrils in the OFT. These changes in normal remodeling are likely the initial steps of a detrimental cardiovascular development path that leads to heart defects in the mature heart [4]. References: [1] M Hagopian and D Spiro. J Cell Biol 44(1970), p.683-7. [2] A Person D, S Klewer, and R Runyan. Int Rev Cytol 243(2005), p. 287-335. [3] M Rennie, C Gahan, C S L�pez, et al. Microsc Microanal. 20(2014), p. 1111-1119. [4] The authors awknowledge funding from an internal OHSU Center for Spatial Systems Biomedicine award and Charles Patrick Memorial Scholarship. Electron microscopy was performed at the Multi-scale Microscopy Core (MMC) with technical support from the OHSU-FEI Living Lab and the OHSU Center for Spatial Systems Biomedicine. Figure 1. Stitched TEM images of the endocardium (A&C), and corresponding 3D cell segmentations displaying each endocardial cell in a different color (B&D) from a control (A&B) and a banded embryo (C&D). CJ, cardiac jelly; E, endocardium; *, cell projections and gaps; scale bar, 5 m. Figure 2. FIB-SEM images of the myocardium layer (A&C), and corresponding myofibril segmentations displaying myofibrils in different colors (B&D) from a control embryo (A&B) and a banded embryo (C&D). Myofibrils in FIB-SEM images, arrow; scale bar, 5 m.");sQ1[456]=new Array("../7337/0911.pdf","Boosting Contrast of Cryo-EM Images Without a Phase Plate","","911 doi:10.1017/S1431927615005358 Paper No. 0456 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Boosting Contrast of Cryo-EM Images Without a Phase Plate Michael S. Spilman1, Hua Guo1, Benjamin E. Bammes1, Liang Jin1, and Robert B. Bilhorn1 1 Direct Electron LP, 13240 Evening Creek Dr., Ste. 311, San Diego, CA 92128 USA Transmission electron microscopy (TEM) imaging of frozen-hydrated biological requires careful selection of exposure in order to balance the goal of maximizing image contrast with the necessity of minimizing radiation damage [1]. In general, increasing the total exposure of each image improves lowfrequency contrast at the expense of high-resolution information due to radiation damage. One of the paradigm-shifting features of several new CMOS-based direct detection cameras is the ability to collect movies--a continuous stream with negligible dead time between frames [2,3]. Several studies have already shown how movies can be exploited to correct stage drift and beam-induced specimen motion. However, movies also provide intrinsic dose fractionation, allowing microscopists to choose their image exposure ex post facto by using subsets of frames from each movie. We have implemented a new image processing method for using these direct detection movies to generate images with both high contrast and high resolution. This method does not require additional instrumentation such as a phase plate. The method applies a low-pass filter to each movie frame based on the expected radiation damage at the corresponding cumulative specimen dose. Briefly, a movie is acquired of a specimen at 2-3� the normal total electron exposure. To correct specimen drift (which is consistent across the entire image), the frames from the movie are iteratively aligned, and to correct beam-induced specimen motion and charging (which are local effects that vary across the image), subregions for each frame are iteratively aligned [4]. To ameliorate the effects of radiation damage, lowpass filters are applied to each frame based on expected damage rate of the specimen. For biological specimens, this damage rate can be estimated from previous radiation damage studies [5,6]. We have demonstrated the benefits of this method by using images of frozen-hydrated Brome mosaic virus (BMV). Images generated based on our method show improved isotropic high-frequency SNR along with significantly improved low-frequency contrast compared to conventional imaging (Fig. 1). We processed a data set using both the conventional method and our new "damage compensation" method to generate de novo three-dimensional reconstructions of BMV. Resolution was determined by the gold-standard Fourier Shell Correlation (FSC). Our new method improved the resolution significantly from 4.4 � to 4.1 � resolution, thus demonstrating the power of damage compensation with a direct detection camera for high-resolution structural studies. We have applied this method to both large macromolecular complexes (such as viruses) and much smaller complexes (including GroEL and small RNA particles). Compared to conventional imaging techniques, the visual contrast of the particles is noticeably higher using our damage compensation method. Additionally, in cases where the exposure rate is low enough, individual electron events can be isolated and counted to further improve image quality. Since electron counting occurs in software as an image processing operation, this technique can be applied to images from any camera with sufficient single electron signal-to-noise ratio. We have implemented our own counting algorithm, which further boosts the contrast of images primarily by eliminating Landau noise. Microsc. Microanal. 21 (Suppl 3), 2015 912 References: [1] RM Glaeser, and RJ Hall, Biophys J 100 (2011), pp. 2331-2337. [2] L Jin and R Bilhorn, Microsc Microanal 16 (2010), pp. 854-855. [3] BE Bammes et al, Microscopy & Microanalysis Conference Proceedings (2012), LB-36. [4] AF Brilot et al, J Struct Biol 177 (2012), pp. 630-637. [5] BE Bammes et al, J Struct Biol 169 (2010), pp. 331-341. [6] LA Baker et al, J Struct Biol 169 (2010), pp. 431-437. [7] The authors acknowledge funding from the National Institutes of Health, grant number 8R44GM103417-03. Figure 1. Damage compensation applied to Brome mosaic virus (BMV). Data was collected with a DE12 Camera System on a JEOL 3200FSC with in-column energy filter, at 61,000� magnification. Images were collected over 1.5 seconds at 25 fps and a total exposure of 53 e-/�2. Left-Top: An example cropped region and SNR plot (from the particle stack) of typical image at 0.68 �m defocus, processed conventionally (frames 2-12, ~17 e-/�2 exposure) and with damage compensation (frames 2-36, ~53 e/�2 exposure). In each case, the first two frames were discarded. Left-Bottom: Gold-standard FSC curves for reconstructions using damage compensated data, conventional data, and conventional images using alignments from damage compensated data. Right: Comparison of beta sheets for the final density map in each case. Data, 3D reconstructions, and atomic model courtesy of Wah Chiu, Zhao Wang, and Corey Hyrc (Baylor College of Medicine).");sQ1[457]=new Array("../7337/0913.pdf","The best of both world, how a "Christmas tree" TEM can please biologist and material scientists.","","913 913 doi:10.1017/S143192761500536X doi:10.1017/S143192761500536X Paper No. 0457 Microsc. Microanal. 21 (Suppl 3), 2015 Microsc. Microanal. 21 (Suppl 3), 2015 �� Microscopy Society of America 2015 Microscopy Society of America 2015 The best of both world, how a "Christmas tree" TEM can please biologist and material scientists. Eric Hanssen1, Mat P Watts2, Edward R Smith3, Prajakta Gosavi4, Paul A Gleeson4 and John W Moreau2 Advance Microscopy Facility, Bio21 Institute, The University of Melbourne, Melbourne, Australia School of Earth Science, The University of Melbourne, Melbourne, Australia 3. Department of Nephrology, The Royal Melbourne Hospital, Melbourne, Australia 4. Department of Biochemistry and Molecular Biology, The University of Melbourne, Melbourne, Australia 2. 1. Historically the "soft and squishy" (eg. Life science) and the "hard and dry" (eg. Material science) scientific worlds have not mixed really well in the electron microscopy/microanalysis field with poor overlap of capabilities and knowledge within service centers usually dedicated to one or the other. But more and more often the need to save space, save funds and staff sees the two worlds collide. Usually this result in a lot of compromise on the equipment side but can also benefit both parties. We will take the example of a FEI Tecnai F30 that was originally set up for cellular biology as a cryoTEM, twin pole pieces, cryo box and tomography capability. Recently it has been upgraded with STEM and STEM-EELS. The cryo capability is mainly used by the biologist and the STEM and/or EELS are essentially used in material science. The former will provide a priceless advantage in some analysis of harder but beam sensitive materials while the later will enable tomography of thicker specimen (~1micron vs 200-300nm) as well as the possibility of identification of particulate if used in conjunction with EELS. Case 1: STEM tomography of Golgi apparatus. This organelle is described as a stack of lamellae forming a structure of about 1-2 micron in diameter and involved in protein trafficking. The standard, even single section tomography gives a poor representation of the overall architecture and connectivity of this structure. Increase in expression level a Golgi associated protein, protein X, change the morphology of the Golgi at optical microscopy level. STEM tomography reveals some insight in the new connectivity and the extent of the defects created by the increase of protein X expression (Figure 1) and underline the importance of this protein in overall protein maturation and trafficking Case 2: Identification of fetuin-A containing calciprotein particle (CPP) in plastic sections. Accumulation of CPP is believed to drive inflammation and extra osseous mineral deposition. The activation of the inflammosome is triggered by internalization of the CPP, this study aim to understand the degradation pathway if the CCP. STEM-EELS was performed to identify the Ca and P components of the CPP and follow it during endocytosis. (Figure 2) Case 3: Study of valence change in highly susceptible -MnO2: -MnO2 mineral, which is analogous to minerals produced by manganese oxidizing bacteria; is strongly oxidizing (containing Mn(IV)) and has been implicated in the redox transformation of a variety of organic and inorganic compounds in the environment. Oxidizing Mn(III/IV) minerals are also thought to be the primary environmental species involved in the release of toxic Cr(VI) by the oxidation of the less toxic and more insoluble Cr(III). We used a Gatan cryo Holder to maintain the sample temperature below -170 degres C in order to assess if cryo conditions were susceptible to stop the valence change due to beam damage. (Figure 3) Microsc. Microanal. 21 (Suppl 3), 2015 Microsc. Microanal. 21 (Suppl 3), 2015 914 910 A B Figure 1. A) STEM tomography model from a 1 micron thick section showing a normal Golgi apparatus. Each color represent an independent cisternae. The ER is represented in red and is not directly linked to the Golgi. B) STEM tomography of protein X knock down. Each color represent an independent system. The extent of the green model shows a high level of interconnectivity between some cisternae of the Golgi apparatus and the ER. Scale bar: 500nm A B C Figure 2 A) thin section of thru a monocyte in TEM bright field. The square represents the region acquire in STEM (B) and STEM-EELS (B inset) showing the Calcium signal. C) STEM-EELS spectra of an individual crystal. A B Figure 3. A) Manganese IV standard EELS spectrum (aqua) compared EELS spectra of -MnO2 maintained at room temperature or in cryo conditions. No valence change is observed in cryo conditions. B) Cryo STEM-EELS image of a sample containing both manganese II (red) and manganese IV (green)");sQ1[458]=new Array("../7337/0915.pdf","Adhesion Pili from Enterotoxigenic Escherichia coli Share Similar Biophysical Properties Despite Their Different Assembly Pathways","","915 doi:10.1017/S1431927615005371 Paper No. 0458 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Adhesion Pili from Enterotoxigenic Escherichia coli Share Similar Biophysical Properties Despite Their Different Assembly Pathways Narges Mortezaei,1 Chelsea R. Epler,2 Paul P. Shao,2 Mariam Shirdel,1 Bhupender Singh,1,3 Annette McVeigh,4 Bernt Eric Uhlin,3 Stephen J. Savarino,4,5 Magnus Andersson1 Esther Bullitt2 Department of Physics, Ume� University, SE-901 87 Ume�, Sweden. Department of Physiology & Biophysics, Boston Univ. School of Medicine, Boston, MA 02118, USA. 3The Laboratory for Molecular Infection Medicine Sweden (MIMS) and Department of Molecular Biology, Ume� University, SE-901 87 Ume�, Sweden. 4 Enteric Diseases Department, Infectious Diseases Directorate, Naval Medical Research Center, Silver Spring, MD 20910, USA. 5 Department of Pediatrics, Uniformed Services University of the Health Sciences, Bethesda, MD 20814, USA. 2 1 Enterotoxigenic Escherichia coli (ETEC) cause diarrhea, enabled by the expression of different adhesion pili. These long, thin, helical filaments, also called fimbriae, share structural similarities, but exhibit varied force requirements for unwinding the filaments into thin fibrillae. The ability of adhesion pili to unwind reduces the shear forces acting on bacteria during intestinal transit, thereby facilitating the sustained adhesion that is necessary for initiation of disease. There is limited genetic similarity between pilins from distinct adhesion pili,.Those in the gastrointestinal (GI) tract can be equally similar to pilins on bacteria that colonize the GI or urinary tracts. Additional diversity arises from pilus assembly via one of two pathways, either the chaperoneusher pathway (CUP), or the alternate chaperone pathway (ACP), independent of the environmental niche of the bacteria. For example, different ETEC express and assemble CFA/I pili via ACP, and CS20 pili via CUP. Our work [1,2] shows that the structures of these two ETEC pili, CFA/I and CS20, share similar structural properties, with diameters of 7-8 nm, lengths of 1-3 �m, 3.2 subunits per turn of the helix, and 0.8 or 0.9 nm rise per subunit, respectively. A three-dimensional reconstruction of CS20 pili from electron cryomicroscopy data has been computed and compared to both CFA/I pili and to P-pili expressed on E. coli that cause urinary tract infections (UTIs). Results show that the inclination of subunits in pili from the GI pathogens is similar, but CUP assembled pili are more similar with respect to the number of subunit-subunit contacts between layers of the helix (Figure 1). Physical properties of CS20 pili were measured using force measuring optical tweezers. The unwinding force of CS20 was assessed to 15 +/- 1 pN. (Figure 2) These results are compared to forces measured for other pili expressed on bacteria in the GI and urinary tracts. First, when the subunit orientation is steeper and the number of contact points is the same, the unwinding force required is higher. Second, when the subunit orientation is the same and the number of contact points is reduced, the unwinding force is lower. In addition, dynamic force measurements are used to examine the velocity at which filament unwinding requires increased force, defining its `corner velocity'. We see a broad distribution of corner velocities Microsc. Microanal. 21 (Suppl 3), 2015 916 that appear to be inversely related to the force of unwinding. Thus, higher corner velocities are measured on pili with lower unwinding forces. In summary, new data on CS20 pili are providing structural and biophysical details that allow comparative analysis between adhesion pili that are essential virulence factors on pathogenic bacteria that cause diarrheal disease or urinary tract infections. References: [1] Mortezaei et al., 2015. Molecular Microbiology 95(1), 116�126. [2] Li et al., 2009. Proc. Natl. Acad. Sci 106(26), 10793-10798. CS20 CFA/I P-pili Figure 1. Three-dimensional helical reconstructions of adhesion pili from pathogenic bacteria: cryoEM reconstruction of CS20 pili; negative stain reconstruction of CFA/I pili, and cryoEM reconstruction of P-pili. CS20 and CFA/I are expressed on bacteria that cause diarrheal disease, and P-pili are expressed on bacteria that cause urinary tract infections. Magnification bar, 5 nm. Figure 2. Force spectroscopy data showing that steady-state CS20 pilus unwinding occurs at 15 pN force (black curve). The pilus can be re-wound (blue curve) after a dip in force corresponding to nucleation required for re-winding.");sQ1[459]=new Array("../7337/0917.pdf","X-ray Tomographic Microscopy of Drosophila Brain Network and Skeletonized Model Building in the Three-Dimensional Image","","917 doi:10.1017/S1431927615005383 Paper No. 0459 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 X-ray Tomographic Microscopy of Drosophila Brain Network and Skeletonized Model Building in the Three-Dimensional Image Ryuta Mizutani1, Rino Saiga1, Kentaro Uesugi2, Akihisa Takeuchi2, Yasuko Terada2 and Yoshio Suzuki2 1. 2. Department of Applied Biochemistry, Tokai University, Hiratsuka, Kanagawa 259-1292, Japan. Japan Synchrotron Radiation Research Institute (JASRI/SPring-8), Sayo, Hyogo 679-5198, Japan. The brain consists of a large number of neurons that make up a three-dimensional network. The first step to understanding brain functions is to analyze the structure of this network. Although three-dimensional structures of brain tissues have been reported, their structures are difficult to comprehend. This is because of a lack of quantitative descriptions of the three-dimensional network, which should be represented with three-dimensional Cartesian coordinates, rather than a three-dimensional distribution of intensities. Here, we report on x-ray tomographic microscopy of the brain network of the fruit fly Drosophila melanogaster and its analysis by skeletonized-model building [1]. Cephalic ganglion dissected from a wild-type Canton-S adult fly was stained with aurate by reducedsilver impregnation, as described previously [2,3]. The brain was then sequentially immersed in ethanol, n-butylglycidyl ether, and epoxy resin (Burnham Petrographics). The obtained sample was mounted using a nylon loop (Hampton Research) and incubated at 90�C for 16 h to cure the resin. X-ray microtomography equipped with Fresnel zone plate (FZP) optics was performed at the BL37XU beamline of the SPring-8 synchrotron radiation facility. An FZP with an outermost zone width of 100 nm and diameter of 310 m was used as an x-ray objective lens. Transmission images produced by 8keV x-rays were recorded using a CMOS-based imaging detector (Hamamatsu Photonics). The tomographic slices were reconstructed from the x-ray images to determine the three-dimensional distribution of the linear attenuation coefficient. The spatial resolution was estimated to be 160 nm by using three-dimensional square-wave patterns. An example of the obtained structure is shown in Fig. 1. Neuronal processes were clearly visualized as a network structure. In order to analyze the structure, its three-dimensional network should be further delineated in terms of Cartesian coordinates by building a skeletonized model. The model can be built by using a method like those used in crystallographic studies of macromolecular structures [1]. Since such an analysis should start by constructing an overall model that can facilitate structural analysis at a higher resolution, an initial model was built from a three-dimensional image obtained with other equipment of microtomography having a wider viewing field at BL47XU of SPring-8 [1]. Automatic tracing was applied to sparsely distributed structures such as those of peripheral nerves, while manual tracing was performed to build the neuropil model. The resulting model (Fig. 2) consists of neuronal processes with a total length of 378 mm in a volume of 0.220 � 0.328 � 0.314 mm3. The neuronal process model was classified into groups on the basis of the three-dimensional structures. Anatomical segments can be extracted from the model by specifying neuronal processes in a group-bygroup manner. Figure 2 shows some of the structures of the optic lobe, which is responsible for visual information processing. Neuronal processes of the medulla and second optic chiasma, which are proximal to the compound eye, exhibit periodical structures corresponding to repeated units of photoreceptors. On the opposite side of the optic lobe, neuronal processes are assembled into several Microsc. Microanal. 21 (Suppl 3), 2015 918 tracts pointed toward central brain regions. These structures represent information paths from the inputs on the eye to the integrative process in the central brain. Refinements of these models by using the high resolution data should reveal finer three-dimensional aspects of the Drosophila brain. [1] R Mizutani et al, J. Struct. Biol. 184 (2013), 271�279. [2] R Mizutani et al, Tissue Eng. C 14 (2008), 359�363. [3] R Mizutani et al, J. Synchrotron Radiat. 15 (2008), 374�377. Figure 1. Stereo rendering of a frontal view of the Drosophila brain. Networks including the midline structure called the central complex are illustrated. Scale bar: 10 m. (b2) (b1) (b) (a) (b3) (b4) Figure 2. Skeletonized model of the left hemisphere of the Drosophila brain. The entire model is shown in (a). The optic lobe (b) is composed of the medulla (b1), second optic chiasma (b2), lobula plate (b3), and lobula (b4). Neuronal process groups are color-coded.");sQ1[460]=new Array("../7337/0919.pdf","Three-Dimensional Neuronal Structure of Human Cerebral Cortex Determined by Synchrotron-Radiation Microtomography","","919 doi:10.1017/S1431927615005395 Paper No. 0460 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Three-Dimensional Neuronal Structure of Human Cerebral Cortex Determined by Synchrotron-Radiation Microtomography Rino Saiga1, Susumu Takekoshi2, Chie Inomoto2, Naoya Nakamura2, Akio Tsuboi2, Motoki Osawa2, Makoto Arai3, Kenichi Oshima3, Masanari Itokawa3, Kentaro Uesugi4, Akihisa Takeuchi4, Yasuko Terada4, Yoshio Suzuki4 and Ryuta Mizutani1 1. 2. Department of Applied Biochemistry, Tokai University, Hiratsuka, Kanagawa 259-1292, Japan. Tokai University School of Medicine, Isehara, Kanagawa 259-1193, Japan. 3. Tokyo Metropolitan Institute of Medical Science, Setagaya, Tokyo 156-8506, Japan. 4. Japan Synchrotron Radiation Research Institute (JASRI/SPring-8), Sayo, Hyogo 679-5198, Japan. Neuronal circuits are responsible for brain functions including cognition, reasoning, and decision making. Neurons constitute neuronal circuits by forming three-dimensional networks in the cerebral cortex tissue. Therefore, the functional mechanism of human brain can be revealed by visualizing and analyzing the three-dimensional structure of the cerebral cortex. Here, we report on the threedimensional structure of human cerebral-cortex tissue, as determined by synchrotron-radiation microtomography at resolutions up to 100 nm. Post-mortem cerebral tissues were collected with informed consent from the legal next of kin using protocols approved by ethical committees of the related organizations. The cerebral tissues were subjected to Golgi impregnation, as described previously [1]. The stained tissues were sequentially immersed in ethanol, n-butylglycidyl ether, and epoxy resin. The samples were transferred to a borosilicate glass capillary filled with resin and kept at 90�C for 40 h to cure the resin. X-ray microtomography equipped with Fresnel zone plate (FZP) optics was performed at the BL37XU beamline of the SPring-8 synchrotron radiation facility. An FZP with an outermost zone width of 50 nm and diameter of 250 m [2] was used as an objective lens, and an x-ray guide tube [3] as a beam condenser. A total of 900 transmission images were recorded with a CMOS-based detector using monochromatic radiation of 8 keV. Tomographic sections were reconstructed with the convolution back-projection method of the RecView program (available from http://www.el.u-tokai.ac.jp/ryuta/). Spatial resolution was estimated to be 100 nm by using test objects [4]. The obtained three-dimensional structure of human cerebral tissue is shown in Figure 1. Neurons and neuronal processes were clearly visualized as a three-dimensional distribution of x-ray attenuation coefficients. Dendritic spines were observed as small protruding structures from the dendrites. These spines have claviform heads with longitudinal lengths of 300-950 nm and cross-sectional diameters of 150-350 nm. The heads are connected to the dendrites through thin cords with typical widths of 150 nm and lengths of 300-1500 nm. These spines are responsible for neurotransmission in neural circuits. Simple-projection x-ray microtomography with a wider field of view (1 mm) was performed at the BL20XU beamline of SPring-8. A total of 1800 transmission images per dataset were recorded using 12keV radiation. The spatial resolution was estimated to be 1.2 m by using test objects. The obtained structure (Fig. 2a) illustrated complicated neuronal networks that cannot be comprehended at a glance. As a result, the image had to be further analyzed to reveal the neuronal circuits embedded in the cerebral tissue. Such neuronal network models can be built by placing and connecting nodes in the three- Microsc. Microanal. 21 (Suppl 3), 2015 920 dimensional map of x-ray attenuation coefficients that represent electron densities. Series of nodes corresponding to neuronal processes and somas were traced to construct skeletonized models of neurons (Fig. 2b). Since the resultant models are represented in three-dimensional Cartesian coordinates, the distances between neuronal processes or somas can be readily calculated from the coordinates. This will allow us to determine neuronal circuits in the human brain tissue by analyzing the positional relationships of the neurons [1,5]. [1] R Mizutani et al, Cerebral Cortex 20 (2010), 1739�1748. [2] Y Suzuki et al, Jpn. J. Appl. Phys. 44 (2005), 1994�1998. [3] Y Suzuki et al, J. Phys. Conf. Ser. 463 (2013), 012028. [4] R Mizutani et al, Nucl. Instrum. Meth. A 621 (2010), 615�619. [5] This work was supported in part by Grants-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (nos. 21611009, 25282250, and 25610126). The synchrotron radiation experiments were performed at SPring-8 with the approval of the Japan Synchrotron Radiation Research Institute (JASRI) (proposal nos. 2013A1384, 2014A1057, and 2014B1083). Figure 1. Stereo rendering of the three-dimensional structure of a pyramidal neuron and neuronal processes in the human cerebral cortex. Scale bar: 5.0 m. (a) (b) Figure 2. (a) Three-dimensional structure of human cerebral tissue. Scale bar: 50 m. (b) Skeletonized model of the tissue structure. Neurons are color-coded. Soma positions are indicated with closed circles.");sQ1[461]=new Array("../7337/0921.pdf","A Reverse Engineering Approach for Imaging Spinal Cord Architecture � Large Area High-Resolution SEM Imaging.","","921 doi:10.1017/S1431927615005401 Paper No. 0461 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A Reverse Engineering Approach for Imaging Spinal Cord Architecture � Large Area High-Resolution SEM Imaging. C. A. Brantner1, M. Rasche4, K. E. Burcham3, J. Klingfus3, J. E. Sanabia3, C. E. Korman1, 2, and A. Popratiloff1 1. GW Center for Nanofabrication and Imaging, The George Washington University, Washington, D.C., USA. 2. Department of Electrical and Computer Engineering, The George Washington University, Washington, D.C., USA. 3. International Applications Center, Raith America, Inc., Troy, NY, USA. 4. Raith GmbH, Dortmund, Germany. The CHIPSCANNER, an e-beam lithography instrument, is a scanning electron microscope (SEM) instrument with a laser interferometer stage and electron detectors that has been used for the imaging of integrated circuit boards in the microchip industry. High-resolution, large-area images are created by capturing sequential SEM images and stitching them together for further analyses. The electronics industry uses this technology: to protect against counterfeit products, to verify designs and to recover obsolete designs. Here we show a biological application for the CHIPSCANNER (Raith, Germany). It can be used for imaging tissue samples by generating large area 2D images from many images. We used back-scattered electrons and a cross-section of a rat spinal cord in Epon to show that a biological sample could be imaged in the electron beam lithography instrument. The architecture of a spinal cord is complex and an examination of cellular structures can only be understood by knowledge of the tissue organization as a whole. Thus low magnification images as well as high magnification images are required for understanding. Here, we propose acquisition of high-resolution 2D registration of adjacent images that includes the entire spinal cord structure. Image acquisition has challenges to be overcome. The software on many SEM instruments has the capability to collect images in an array and tile them together. There are difficulties in such operations: pixel-to-pixel spacing, limited field of view (FOV), large numbers of images needed to cover large areas of interest, long collection times, stage movement errors, beam position and current drift (causing positional inaccuracy). The CHIPSCANNER overcomes these difficulties with its laser interferometer stage providing accurate sample location allowing for precise stitching together of adjacent images. Responsible for beam deflection and signal recording, the pattern generator efficiently collects and stores images at resolutions up to 50,000 pixels by 50,000 pixels. Figure 1 shows the smooth junction between four images of a microchip that the CHIPSCANNER has imaged and stitched together [1]. Large area images with smooth junctions are important for biology and histology could benefit by visualizing the whole cross section of a tissue. Spinal cord from a rat was dissected, fixed, and embedded in EPON. Cross-sections were cut and further embedded in resin between Aclar sheets. A resin embedded section was mounted on a stub and imaged in the CHIPSCANNER. Several sample preparation protocols including traditional transmission electron microscope (TEM) and serial blockface protocols were examined to find the set of conditions that allowed for the best preservation of the ultrastructural details as well as the best imaging conditions in the instrument [2]. Figure 2 shows an image from the CHIPSCANNER of the spinal cord. Microsc. Microanal. 21 (Suppl 3), 2015 922 We have shown that the CHIPSCANNER can be used to generate high-resolution 2D images from spinal cord samples and is a viable method for obtaining detailed information from large biological samples with an electron beam instrument. TEM is the "gold" standard for resolution and contrast of samples, yet the FOV and area that can be imaged is relatively small compared to the size of the tissue that the cells are organized into. Focused Ion Beam SEM (FIB SEM) is another method to image cells and tissue at high resolution, however it is not capable of large area imaging as can be accomplished in the CHIPSCANNER. There are methods available to image tissues in the field of Histology that scan stained, vibratomed sections on slides using a light microscope and a camera. This scanning and imaging of a slide with a light microscope is similar to the operation of the e-beam CHIPSCANNER. References. [1] J Klingfus, KE Burcham, M Rasche, T Borchert and N Damnik, ISTFA Conference Proceedings (2011), p. 373-376. [2] T Deerinck, E Bushong, A Thor, M Ellisman, (2010) NCMIR methods for 3D EM: A new protocol for preparation of biological specimens for serial block face scanning electron microscopy. Microscopy: 6-8. Figure 1. Diagram showing images of a microchip taken with the CHIPSCANNER that have been stitched together. The pink scale bar represents 9 �m. This system relies on the precision locations of the stage and the e-beam, instead of algorithms in the software for stitching images together to display a large 2D area of the sample. Figure 2. A high resolution image of rat spinal cord taken in the CHIPSCANNER. Left is a 200 �m x 100 �m area where the image contains 40,000 pixels x 20,000 pixels, corresponding to 1 pixel being 5 nm x 5 nm. The scale bar is 20 �m. Right is a close-up section from the same image, where the scale bar is 2 �m.");sQ1[462]=new Array("../7337/0923.pdf","Simplified Biological Tissue Analysis with Combination of Atmospheric Scanning Electron Microscope and Light Microscope","","923 doi:10.1017/S1431927615005413 Paper No. 0462 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Simplified Biological Tissue Analysis with Combination of Atmospheric Scanning Electron Microscope and Light Microscope Akiko Hisada1 and Yusuke Ominami2 1 2 Hitachi Central Research Laboratory, Saitama, Japan Hitachi High-Technologies Corporation, Ibaraki, Japan Correlative light and electron microscopy has recently been considered to be a powerful tool for analyzing cell biology. For any observation that uses different types of microscopes in succession, specimens should be processed appropriately. Viable cells or fixed cells maintained in liquid can be observed with a light microscope. When those cells are successively observed with a conventional electron microscope, they should be processed so that they remain intact when placed in a high vacuum and irradiated with an electron beam. However, cells easily change shape if processed using drying. To keep the focus on cell samples that are processed differently, various procedures to create a trace of the visual field under a light and electron microscope have been proposed. Another effective approach is to use an isolated atmospheric specimen chamber with an electron microscope. This device can prevent the deformation of cells caused by drying, making it easier to maintain the same field of view. Several types of specimen chambers have been designed for scanning electron microscopes (SEMs) [1-4]. To this end, we have developed a novel atmospheric scanning electron microscope (ASEM) that enables bulky biological specimens to be observed without preparations such as dehydration, drying, and electroconductive processing [5, 6]. In this study, we evaluated the ASEM images of individual sections of rat organs to establish an observation procedure combining ASEM and light microscopy. An inner specimen chamber combined with a thin membrane enabling electron beam penetration is inserted in the main chamber of a desktop type SEM (figure 1, a and b). A specimen inside the inner chamber is kept at ambient atmospheric pressure (figure 1, b) or negative pressure (figure 1, c). Electron beam dispersion with the gas can be reduced by adjusting the negative pressure [6]. Rat organs were excised and fixed with neutral buffered formalin. After washing with water, sections were examined under an optical stereo microscope, and then identical sections were observed with ASEM. To evaluate the ASEM images, sections were processed with conventional H-E staining. As a result of ASEM observation, it was possible to visualize the cells of hydrated tissues without staining (figure 2). For example, we observed goblet cells of the intestinal mucosa surface, enterocytes of a mucosa section, and cartilage cavities and chondrocytes of a hyaline cartilage of the trachea. We have established an observation procedure to directly examine a specimen under ASEM after making an optical stereo microscope observation without additional preparation. In conclusion, observing a section of the hydrated bulky tissue at a cellular level was simplified. Furthermore, we propose a new workflow to determine points of interest by making ASEM observations prior to timeconsuming histological analysis (figure 2). Microsc. Microanal. 21 (Suppl 3), 2015 924 References: [1] Danilatos GD, Scanning 7 (1985), p. 26. [2] Thiberge S et al., Proc Natl Acad Sci USA. 101 (2004), p. 3346. [3] Ogura T, Biochem Biophys Res Commun. 377 (2008), p. 79. [4] Suga M et al., Ultramicroscopy 111 (2011), p. 1650. [5] Ominami Y et al., Proc SPIE, 9236 (2014), p. 923604. [6] Ominami Y et al., Microscopy, Epub 2014 Dec 23. (a) (b) Vacuum pumps SEM BSE detector (c) Primary electrons BSE detector Vacuum Specimen Specimen holder Stage Atmosphere Inner chamber Negative pressure Vacuum pump SiN membrane (20 nm) Gas Atmosphere (50 � 100 �m) Specimen Vacuum Main chamber Figure 1. Schematic view of desktop type ASEM. (a) Primary electrons penetrating membrane. (b) Imaging mode at atmospheric pressure. Inner chamber is kept at 10 5 Pa (blue). (c) Negative pressure from 103 Pa to 105 Pa (yellow). Excision Fixation Sampling Paraffin sectioning Staining Examination 2 - 3 days Visual observation/ Optical stereoscopic microscope Atmospheric scanning electron microscope (ASEM) Optical transmission microscope 1 mm Rat colon 100�m 100�m Organ - Tissue Tissue - Cell Histochemistry Figure 2. Proposed workflow for processing a piece of hydrated tissue for histological analysis.");sQ1[463]=new Array("../7337/0925.pdf","3D X-ray Microscopy: A New High Resolution Tomographic Technology for Biological Specimens","","925 doi:10.1017/S1431927615005425 Paper No. 0463 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 3D X-ray Microscopy: A New High Resolution Tomographic Technology for Biological Specimens Leah Lavery1, Arno Merkle1 and Jeff Gelb1 1. Carl Zeiss X-ray Microscopy, Pleasanton, CA, United States of America. A new field of 3D X-ray Microscopy (XRM) has emerged bringing dramatic resolution and contrast improvements to X-ray tomographic imaging of biological specimens for correlative studies and hierarchical structure investigations of hard and soft tissue. An X-ray microscope uses an X-ray source rather than a visible light source to view the internal structure of opaque specimens. Analogous to computed tomography (CT) a specimen can be imaged without physical sectioning and a complete 3D view of the object is generated. Yet X-ray microscopes provide superior spatial resolution down to the nanoscale and tunable phase contrast to image nature's vast diversity from cells to entire organisms ex vivo up to tens of centimeters in size. Light microscopy and immunostaining techniques have become essential practice for biological and biomedical research providing functional information on the distribution of gene products such as proteins and ribonucleic acids from primarily 2D images. More recent developments, such as Light Sheet Fluorescence Microscopy (LSFM), have provided an attractive option for developmental biology and other research when observing millimeter-sized live specimens in 3D due to an optical architecture that enables fast acquisition with low phototoxicity [1]. Despite the revolutionary advancements of light microscopy, visible light has physical penetration limits due to specimen thickness or opacity. Microcomputed tomography (microCT), an alternative technique capable of producing full 3D images, uses multiple X-ray projections to reconstruct a 3D representation of an object. MicroCT has been highly useful for large objects such as the human body in vivo and applications requiring resolution no greater than ~5-10 �m. However, demand is growing for applications that require 3D imaging with high resolution (single micron and below) and high contrast for hard and soft tissue as well as cell-level information. Laboratory XRM, which emerged in the past decade from the foundations of synchrotron-based X-ray imaging technology, produces direct 3D tomographic information from opaque specimens with resolutions well into the sub-micron range, even achieving 10's of nm resolution for certain architectures and applications [2]. This presentation will cover 3D XRM technique, followed by a survey of recent application areas for XRM within the life sciences where it acts as a complementary and correlative bridge between the contrast and resolution standards set by light and electron microscopy. By integrating an imaging detector with sufficiently small and tunable pixel size coupled to optimized scintillation materials, the standard trade-off of sample size versus resolution found in conventional laboratory microCT are improved with XRM. High total system spatial resolution may be maintained while reducing the dependence on geometric penumbra. This, in turn, reduces the dependence of resolution on geometric magnification, relaxing restrictions on sample placement relative to the source and source spot size [3]. By employing such a geometry, samples of considerable size (tens of centimeters for low density objects) may be imaged in 3D with high resolution. In addition, as a result of small effective detector pixel dimensions and tunable detector and source positions, propagation Microsc. Microanal. 21 (Suppl 3), 2015 926 phase contrast may be employed to effectively highlight interfaces in samples that exhibit low absorption contrast, such as unstained tissue and other organic specimens. Research topics currently being explored and facilitated by the use of XRM in the life sciences are diverse, including developmental biology, soft tissue, agriculture, hard-tissue, bio-engineered materials, including 3D scaffold materials, and many others. Bone research has been represented in a number of recent studies, including fatigue loading of human bone and blood-bone interaction in hematopoietic stem cells. Through the use of phase-contrast imaging, additional applications have emerged, in particular the ability to quantify the 3D structure of cartilage, which necessitates the ability to visualize low-density soft tissue (cartilage) alongside higher-density tissue (bone). Also, bone represents an example of a hierarchical material, where it is useful to quantify 3D microstructure across several length scales, as depicted in Figure 1. Bio-mechanical studies utilizing XRM have been explored on bone as well as in the dental field, in particular to explore the interaction of the tooth-bone-periodontal ligament system. Researchers have shed light on this interesting bio-mechanical system and have used XRM as a way to explore the 3D structure under real (and varying) loading environments through the use of various mechanical loading rigs. XRM has enabled a number of applications in biological research by providing micron and nanoscale information across a wide array of sample sizes. However, no single microscopy modality is capable of imaging structures in 3D across an entire range of length scales and contrast mechanisms. In response to this, efficient correlative microscopy methods have been proposed that use several imaging solutions to analyze a single sample. Non-destructive XRM will contribute greatly to this correlative imaging landscape by providing a bridge between light and electron microscopy in length scale, contrast mechanisms and sample preparation requirements. One active area where XRM is being used in a correlative workflow is in the inspection and identification of known sub-volumes of stained embedded samples (for EM) prior to serial sectioning for 2D or 3D imaging in the SEM, FIB/SEM or TEM. This significantly increases efficiency by navigating to a sub-volume of interest, investing time imaging identified regions of interest with the electron microscope instead of searching for a needle in the haystack. New applications such as this point to a bright future for XRM as a common technique in the research laboratory. Figure 1 Hierarchical structure in bone with 3D renderings of datasets. Images were collected using ZEISS XRM technology with the exception of ZEISS FIB-SEM dataset (far right). References: [1] Selchow, O., Photonik Int., 11 (2013), pp. 44�47. [2] A. Merkle and J. Gelb. Microscopy Today 21, (2013) pp. 10-15. [3] A. Tkachuk, et al., Z. Kristallogr. 222, (2007) pp. 650-655.");sQ1[464]=new Array("../7337/0927.pdf","Cryo Transmission Electron Microscopy of Three Research-Grade Liposomal Formulations of Doxorubicin","","927 doi:10.1017/S1431927615005437 Paper No. 0464 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Cryo Transmission Electron Microscopy of Three Research-Grade Liposomal Formulations of Doxorubicin William Monroe1, Angel Paredes1, Patrick Sisco2, Kristen Ahlschwede1 and Sean Linder2 U.S. Food and Drug Administration, National Center for Toxicological Research, Jefferson, Arkansas, USA 2. U.S. Food and Drug Administration, Office of Regulatory Affairs, Jefferson, Arkansas, USA Nanotechnology has sparked a rapidly growing interest in medicine with the potential to solve a number of issues associated with conventional therapeutics, including poor water solubility, lack of targeting capability, nonspecific distribution, systemic toxicity, and low therapeutic index. Over the past several decades, remarkable progress has been made in the development and application of engineered nanoparticles to treat diseases more effectively. One of these key nanomaterials is the liposome, which is currently being used to encapsulate various chemotherapeutic agents. This encapsulation is engineered to increase the efficacy of the chemotherapeutic agent by protecting it against degradation, prolonging circulation time, improving cellular uptake, and reducing side effects that limit dosage amounts. As the commercialization of liposomes has increased, regulatory-based methodologies for their characterization have not been well documented. As such, to ensure the safety and quality of commercialized liposomal products, the development of methodologies to characterize these materials is imperative [1]. In this work, we address this concern by investigating a method to characterize the particle size and size distribution associated with liposome based drugs. More specifically, this study uses three researchgrade liposomal formulations of doxorubicin with sizes of 100, 200, and 300 nm. Particle size and size distribution were analyzed using electron cryomicroscopy (cryoEM) which is an established technique that preserves small biomolecular structures in their most native state. The 100 nm liposomes were confirmed to be monodisperse and close to their expected size. The cryoEM image in Figure 1 shows that the 200 nm liposomal formulation is polydispersed and larger than anticipated. These results were similar to those obtained by dynamic light scattering (DLS). These methods and results will guide the scientists in manufacturing and regulatory agencies to verify the formulation characteristics and ensure batch to batch consistency. These three formulations will be used as test material for future pharmacokinetic studies [2]. 1. The information in these materials is not a formal dissemination of information by the FDA and does not represent agency position or policy. Microsc. Microanal. 21 (Suppl 3), 2015 928 References: [1] Report by the Arkansas Regional Laboratory, U.S. Food and Drug Administration, July, 2014 [2] Gabizon, A., H. Shmeeda, and Y. Barenholz, Pharmacokinetics of pegylated liposomal Doxorubicin: review of animal and human studies. Clin Pharmacokinet, 2003. 42(5): p.419-36 Figure 1. Left. Representative CryoEM image of 200 nm liposomal doxorubicin formulation. Right. Size distribution of the same sample. Scale bar = 200 nm.");sQ1[465]=new Array("../7337/0929.pdf","Using a White Light Confocal Profiler for Ancient Diet Reconstruction","","929 doi:10.1017/S1431927615005449 Paper No. 0465 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Using a White Light Confocal Profiler for Ancient Diet Reconstruction Ashley Remy�, Christopher Schmidt�, Ruggero D'Anastasio� 1 2 University of Indianapolis, Indianapolis IN USA Museo Universitario Universit� "G. d'Annunzio", Chieti e Pescara, Italy Dental-based dietary reconstruction is a process of examining occlusal surfaces of ancient teeth by means of Dental Microwear Texture Analysis (DMTA). Diets indicate subsistence strategies such as lifeways including either hunting and gathering, fishing, agriculture, or pastoralism. DMTA uses a White Light Confocal Profiler (WLCP) to interpret surface microtopography. In general, foragers have rougher surfaces while farmers, who processed their food more, have smoother surfaces and their microfeatures aligned in a common direction. Herculaneum, an ancient Roman town, gives a unique opportunity to employ these methods of diet reconstruction. In AD 79, Herculaneum, along with Pompeii, was destroyed by the eruption of Mt. Vesuvius. Archaeological excavations recovered skeletal remains of approximately 300 individuals along the coast of Herculaneum [1]. The nature of these individuals' deaths allows us the rare opportunity of having a snapshot in time. The recovery of numerous subsistence artifacts and actual food items (including copious seeds and nuts) allowed us to predict that these people should have a diet that is somewhat rougher than might be expected for what is generally considered a fishing village diet. Data were collected with the WLCP and studied via scalesensitive fractal geometry software. The variables include Complexity (surface roughness), Anisotropy (similarity of feature orientation), and Textural Fill Volume (amount of surface removed) on high resolution casts of molar Phase II occlusal wear facets. Phase II facets were selected because they are the area on the tooth where food is crushed during mastication. Magnification was 100X. Data were studied with Sfrax� and Toothfrax� software, and the statistical procedures employed ANOVA. Our results indicated that our prediction was correct, Herculaneum did consume a somewhat harder diet, with values of 1.46 for complexity and .0033 for anisotropy, although some of this hardness is biased by the children who had a harder diet than the adults. It is unclear at this point why children have greater roughness, but this has been seen in Medieval children as well. Overall, the Herculaneum diet appears to be diverse and ranges well beyond fish consumption (which usually leads to smooth tooth surfaces). Ongoing study involves interpreting the data in the context of this group of people who died simultaneously so we can better understand how diets vary within the population. Microsc. Microanal. 21 (Suppl 3), 2015 930 Figure 1 and 2. Examples of DMTA from the individuals of Herculaneum as seen with fractal software. References: [1] Capasso, L. (2001). I fuggiaschi di Ercolano: paleobiologia delle vittime dell'eruzione vesuviana del 79 dC.");sQ1[466]=new Array("../7337/0931.pdf","Testate Amoebae Diversity, from the Atlantic Forest Aquatic and Edaphic Environments, Collected Within the Rio de Janeiro State, Brazil.","","931 doi:10.1017/S1431927615005450 Paper No. 0466 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Testate Amoebae Diversity, from the Atlantic Forest Aquatic and Edaphic Environments, Collected Within the Rio de Janeiro State, Brazil. In�cio D. da Silva-Neto1, Anderson G. S. Silva1, Pedro H. C. Nunes1, Gisele R. Santos1, Marcelo H. de O. Sales1, Thiago da S. Paiva1, Luiggia G. B. R. Araujo1 and Breno Leite2 Dep. de Zoologia, Inst. de Biologia, Universidade Federal do Rio de Janeiro, Brasil JEOL USA, Inc., Peabody, MA, USA (idsnet@biologia.ufrj.br) 2. 1. Atlantic Forest is considered one of world's most species-diverse ecosystems, with high endemism rate and often referred as a diversity hot-spot. [1, 2]. Due to anthropic activity since the 1500's, it has been constantly reduced until only ca. 7.9% of its original area remains intact [2]. In the state of Rio de Janeiro, the Atlantic Forest originally covered 98% of its territory, while currently its extension is less than 17%. Considering such scenario and the eventual possible implications of huge diversity losses, it is important to study species diversity within an alpha-taxonomy (i.e. descriptive taxonomy) context. The present study is part of a Brazilian program called BIOTA-FAPERJ (E26/110.022/2011), which aims to investigate the diversity of microeukaryotic organisms present in various environments of the Atlantic Forest in Rio de Janeiro, using morphological and molecular approaches. The program also has the objective to investigate the biotechnological potential of these organisms in improving environment quality. Testate amoebae are free-living, heterotrophic, microeukaryotic organisms which are polyphyletic, representing various lineages among the eukaryote tree, most associated to the Arcellinida (Amoebozoa), but slightly related to Labyrinthulomycetes (Stramenopiles) Rhizaspididae (Rhizaria) [3]. Such microeukaryotes have their bodies protected by a shell, of which morphology is largely used for species identification. While carrying out this study, testate amoebae were isolated from: a) sediments from water samples collected from streams, ponds, and bromeliad tanks; and b) from soils obtained from various locations in the Rio de Janeiro Atlantic Forest. Samples were split into Petri dishes and raw mass cultures were kept in the laboratory. The organisms were morphologically identified under stereomicroscope. The testate amoebae were photographed under differential interference contrast (DIC) and then observed a under scanning electron microscopy (SEM). We documented more than 80 species (Figure 1), mostly assigned to Arcellinida and Euglyphida, many are new occurrences, of which the most frequent were Arcella hemisphaerica hemisphaerica, A. hemisphaerica undulata, A. braziliensis, A. vulgaris, Centropyxis aculeata, C. aculeata oblonga, C. spinosa, C. platystoma, Difflugia corona, D. acuminata, D. cylindrus, D. elegans, D. achlora, D. penardi, Ciphoderia ampulla, Euglypha filifera, E. ciliata, Trinema lineare, Lesquereusia globulosa, L. modesta minima e Protocucurbitella coroniformis var. ecornis. The comparison with other geographic regions of the world, by tabulating the number of testate amoebae species, indicated high diversity within these areas. This study is the one of the first steps to fill the current gap in the knowledge of microeukaryotic diversity in Brazil. References: [1] Myers N (1988) The Environmentalist 8: 1�20. [2] Myers N (1990) The Environmentalist 10: 243�256. [3] Adl SM et al., (2012) Journal of Eukaryotic Microbiology 59: 429�493. Authors acknowledge Dr. Maria Ishida (FDACS) and Dr. Natasha Erdman (JEOL USA Inc) for critically reviewing the manuscript. Microsc. Microanal. 21 (Suppl 3), 2015 932 Figure 1. Arcella vulgaris stretching its pseudopodia (from life), scale bar = 25 �m (A); Arcella gibbosa with a three-lobed pseudostome (arrow), scale = 25 �m (B); Cyclopyxis impressa with pseudopodia and pseudostome shown in the same image (from life), scale = 100 �m (C); Centropyxis spinosa with various shell apertures (arrows) (D); Euglypha sp. with regular, but complex scales (E); Lesquereusia extranea, with a very peculiar pattern of "scales" (F); Difflugia bacillifera showing pieces taken from the environment build its shell, a diatom frustule is shown (arrow) (G); Quadrulella symmetrica var. tubulata with protruding pseudopod (large arrow). Soft body can be seen inside shell (small arrow) amoeba can be seen inside (from life), scale bar = 25 �m (H); Quadrulella symmetrica var. tubulata exhibiting the pattern of shell plates (I).");sQ1[467]=new Array("../7337/0933.pdf","Inactivation of Bacterial Endospores in an Artificial Tissue for Electron Microscopy Analysis","","933 doi:10.1017/S1431927615005462 Paper No. 0467 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Inactivation of Bacterial Endospores in an Artificial Tissue for Electron Microscopy Analysis Ryan M. Hannah1, Brian M. Leroux1 and Robert K. Pope1 1 Electron Microscopy Laboratory, National Biodefense Analysis and Countermeasures Center, Battelle National Biodefense Institute, 8300 Research Plaza, Fort Detrick, MD 21702. Removal of tissue samples that contain, or potentially contain, pathogenic organisms from containment laboratories for electron microscopy analysis poses unique challenges. While the extraction of nucleic acids or proteins for molecular biology or immunology analysis is fairly straightforward, these methods of extraction destroy ultrastructure, and are not suitable for samples intended for microscopy. Fixation of tissue samples containing pathogenic organisms traditionally requires several days to inactivate the organisms. Furthermore, sterility testing for pathogenic organisms can take from 2-21 days, depending on the organism. This lengthy fixation with sterility testing is not feasible for the rapid turnaround required for forensic samples. Previously, data demonstrated that treatment of bacterial endospores with 4% paraformaldehyde/1% glutaraldehyde for 240 minutes inactivates most bacterial endospores. [1] The objective of this study was to deduce the time required to inactivate bacterial endospores in an artificial tissue matrix. Bacillus subtilis endospores were used for this experiment due to their known resistance to inactivation [1]. Endospores were embedded in an artificial tissue (4% agarose/20% beef liver) prepared as either 1 cm3 or 0.5 cm3 cubes. These sizes were used to mimic the maximum size of tissue pieces typically collected for light and electron microscopy. The cubes were treated with universal EM fixative (4% paraformaldehyde/1% glutaraldehyde) for the following times 1, 2, 4, 8, 24, 48, and 72 hours, and 4, 5, 6, and 7 days (data from days 4-7 is not shown on the graph). Following fixation, the cubes were washed with buffer three times, ground, and plated to enumerate viable spores. Inactivation of the endospores varied for the different sized cubes. The spores in the 0.25 cm3 pieces were completely inactivated in four hours, while the spores in the 1 cm3 pieces were completely inactivated in eight hours. The results of these studies will reduce the total time required to fix and inactivate tissue samples to enable their removal from biocontainment. Future expansion of this work will include the timed inactivation of several species of endospores, including Bacillus anthracis. References [1] CA Brantner, et al, Inactivation and ultrastructure analysis of Bacillus spp. and Clostridium perfringens spores. Microscopy and Microanalysis. 20 (2014), 238-244. [2] "This work was funded under Agreement No. HSHQDC-07-C-00020 awarded by the Department of Homeland Security Science and Technology Directorate (DHS/S&T) for the management and operation of the National Biodefense Analysis and Countermeasures Center (NBACC), a Federally Funded Research and Development Center. The views and conclusions contained in this document are those of the authors and should not be interpreted as necessarily representing the official policies, either expressed or implied, of the U.S. Department of Homeland Security. In no event shall the DHS, NBACC, or Battelle National Biodefense Institute (BNBI) have any responsibility or liability for any use, misuse, inability to use, or reliance upon the information contained herein. The Department of Homeland Security does not endorse any products or commercial services mentioned in this publication." Microsc. Microanal. 21 (Suppl 3), 2015 934 Figure 1. Procedure for the production of artificial tissue (4% agarose/20% beef liver) containing Bacillus subtilis endospores. The artificial tissue was placed into fixative for various times, washed with buffer, ground and the spores enumerated for viability. Figure 2. Decrease in spore viability after timed treatment in universal EM fixative solution. Spores in artificial tissue were fixed for times ranging from one hour up to seven days (days 4-7 not shown on graph). The time for inactivation varied for the different sized pieces of artificial tissue with the spores in 0.25 cm3 cubes being inactivated in four hours and the spores in 1 cm3 cubes inactivated in eight hours.");sQ1[468]=new Array("../7337/0935.pdf","Aspergillus fumigatus Biofilms: a Comparison of Processing Techniques for Scanning Electron Microscopy of Fungal Mycelium and Extracellular Matrix.","","935 doi:10.1017/S1431927615005474 Paper No. 0468 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Aspergillus fumigatus Biofilms: a Comparison of Processing Techniques for Scanning Electron Microscopy of Fungal Mycelium and Extracellular Matrix. Lydia-Marie Joubert1, Jose AG Ferreira2,3, David A Stevens3 and Lynette Cegelski4 1. Cell Sciences Imaging Facility, Stanford University School of Medicine, Stanford, CA 94305, USA. School of Medicine, Faculdade de Sa�de e Ecologia Humana-FASEH, Vespasiano, Brazil. 3. Division of Infectious Diseases and Geographic Medicine, Stanford University, Stanford, CA, USA. 4. Department of Chemistry, Stanford University, Stanford CA 94305, USA. 2. Introduction: Biofilms are matrix-enclosed microbial populations adherent to each other and/or to surfaces or interfaces [1,2]. It has recently been shown that Aspergillus produces in vitro an extracellular matrix with typical biofilm characteristics under static and shaken, submerged conditions [3]. Aspergillus fumigatus is frequently isolated from cystic fibrosis (CF) patients, and Aspergillus biofilms may be one of the most important virulence factors in CF and invasive pulmonary aspergillosis [4, 5]. In-depth analysis of Aspergillus biofilms is therefore necessary to improve antifungal targets for treating complex A. fumigatus biofilm-associated diseases [6]. SEM analysis of the 3D architecture of hydrated biofilms is commonly affected by standard fixation and drying techniques [7], and stabilization of proteins through aldehyde cross-linking, with post-fixation of lipids with osmium-tetroxide (OsO4), help maintain overall biofilm structure. Retention of fine features is generally accomplished through critical point drying (CPD) or hexamethyldisilazane (HMDS) [8]. Environmental SEM, using Ruthenium Red as contrasting agent, or Variable Pressure (VP)-SEM using ionic liquids, have been reported to improve imaging of hydrated biofilms and their natural in situ 3D architecture [9, 10, 11]. In this study we investigated the effect of processing techniques and reagents on SEM analysis of the cellular mycelium and extracellular matrix (ECM) of two modes of biofilm growth of A. fumigatus. Processing parameters evaluated were (1) time in primary aldehyde fixatives, (2) including OsO4 as secondary fixative, (3) final drying through CPD or HMDS and (4) hydrated structure with VP-SEM. Methods: A. fumigatus biofilms were grown in RPMI 1640 culture medium on 12mm circular plastic rotating bioreactor disks, or as a floating biofilm mat close to the water-air interface. After 2 days of growth, disks and biofilm mats were removed from culture medium, rinsed in phosphate-buffered saline and fixed in 4% paraformaldehyde (PFA) with 2% glutaraldehyde (GA) in 0.1M sodium cacodylate buffer. Table 1 summarizes the processing parameters evaluated. Hydrated samples were observed with a Hitachi 3400-N SEM operated at 15kV, 60Pa, using Backscattered Electron (BSE) detection. Dried samples were sputter-coated (50�, Au/Pd) before imaging with a Hitachi 3400N SEM operated at 10kV under high vacuum, and a Zeiss Sigma FESEM using InLens Secondary Electron (SE) detection at 2kV. Results: Post-fixation with OsO4 generally improved ultrastructure, while also enhancing SE and BSE detection for SEM analysis. Also, shorter periods (less than 1hr) in both aldehyde and OsO4 fixatives resulted in improved separation of fine structural features (Figure 1), while longer fixation times caused a collapse of fungal mycelium and fibers in the extracellular matrix. CPD resulted in improved preservation of especially ECM, whereas biofilms dried with HMDS showed more collapsed hyphae connected by sheets of ECM lacking fibrous ultrastructure. Inherent to VP-SEM is the poor signal to noise ratio due to gas and moisture in the specimen chamber, as well as the occurrence of water coating hyphae as an electron dense sheet, which obscures fine cellular features while retaining 3D structure. Microsc. Microanal. 21 (Suppl 3), 2015 936 Conclusions: Revisiting standard processing protocols for EM analysis of microbial biofilms emphasizes the complexities involved in visualizing the attached lifestyle of microbial communities. Fixation and dehydration-induced artifacts should be thoroughly analyzed to determine the appropriate combination of techniques that will best reveal specific structural aspects of biofilms. Using VP-SEM instead of highvacuum SEM may better reveal hydrated 3D architecture, but limit ultrastructural analysis of individual cells and extracellular matrix. Our results suggest that shortened times of aldehyde fixation and OsO4 post-fixation, followed by CPD, is optimal for high-resolution ultrastructural SEM analysis of cellular features and extracellular matrix of Aspergillus biofilms. References: [1] JW Costerton, et al, Annual Rev Microbiol. 49 (1995), p.711. [2] H Lappin-Scott, S Burton and P Stoodley, Nature Reviews Microbiology 12 (2014), p.781. [3] FM Muller, M Seidler and A Beauvais, Med Mycol. 49 (2011), Suppl 1, S96. [4] JJ Speirs JJ, CK van der Ent and JM Beekman, Curr Opin Pulm Med. 18 (2012), p. 632. [5] AM de Vrankenrijker, et al. Clin Mircobiol Infect. 17 (2011), p.1381. [6] S Kaur and S Singh, Med Mycol. 52 (2014), p.2. [7] M Alhede, et al, FEMS Immunol Med Microbiol. 65 (2012), p.335 [8] DF Bray, J Bagu and P Koegler, Microscopy Research and Technique 26 (1993), p.489. [9] JH Priester, et al, J Microbiol Methods 68 (2007), p.577. [10] K Weber, et al, FEMS Microbiol Lett. 350 (2014), p.175. [11] Y Asahi, et al. AMB Express 5 (2015), p.6. [12] The authors acknowledge Beckman Center (LMJ), The Child Health Research Institute, Stanford Interdisciplinary Initiatives Program (DAS), a gift from Mr John Flatley (DAS) and NIH Director's New Innovator Award (LC, grant DP2OD007488) for financial support. BIOFILM CULTURE (x2): FLASK versus DISK 2%GA + 4%PFA + 1%OsO4 24hrs fix CPD HMDS 2%GA + 4% PFA (no OsO4) 24hrs fix CPD HMDS 45min fix CPD HMDS 45min fix CPD HMDS Hydrated: VP-SEM (60Pa) Hydrated: VP-SEM (60Pa) Table 1: Summary of fixation and drying parameters used to process A. fumigatus biofilms for SEM. Figure 1: SEM images illustrating typical biofilm characteristics after optimal fixation periods in PFA, GA and OsO4, followed by CPD. In the dense mycelium (A) hyphae are connected by fibrous ECM (arrows) (B, C), which can be seen closely adherent to the cell surface at high magnification (D).");sQ1[469]=new Array("../7337/0937.pdf","Sure, We Can Image That! An Attempt to Image the Edge of a Growing Bacterial Colony","","937 doi:10.1017/S1431927615005486 Paper No. 0469 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Sure, We Can Image That! An Attempt to Image the Edge of a Growing Bacterial Colony Garnet Martens1 and Kevin Hodgson1 1 Bioimaging Facility, University of British Columbia, Vancouver, BC, CANADA A while back we received a request to image a growing bacteria colony. Now given that we have years of live-cell imaging experience, we imagined the conditions required to collect data from a bacterial colony would be easier to control than that of mammalian cells. Bacteria don't move too much and they don't need CO2. Our initial efforts using LSCM (Fig. 1) seemed to indicate we were on the right track but then things turned completely upside down. Almost all our efforts were in vain. Applying a coverslip to obtain high enough resolution to see individual cells (Fig. 1) created an anaerobic environment prohibiting cell growth. We tried overnight imaging with a spinning disk confocal (Fig. 2) several times and we retried imaging with a LSCM (weird drying artifacts in the agar). We had some initial success imaging with a GoPro camera mounted inside the incubator (Fig. 3). Some of the best data came from mounting a DSLR/macro lens inside a custom chamber (Fig. 4) and imaging the plate inside the incubator (Fig. 5). What we really wanted to see were the individual cells at the edge of the growing colony. So, what do cunning microscopists do when the equipment in your facility cannot do the job? We ask for an equipment demo, that's what we do. After a few trials we were finally able to image some growing colonies using a relatively new instrument called the Incucyte. We are still working out the bugs (pun intended) of this system but the combination of phase and fluorescence is finally giving us some decent data (Fig. 6a). The first conclusion we have found is that the edge of the colony exists as a single cell layer at least 10 cells deep (white arrow in Fig 6a and b). Moving toward the center of the colony the cells begin to pile up in a second step, then a third step until finally they are so deep that the colony looks homogenous. At the end of the day we have learned some valuable lessons in patience and we will keep this project in mind the next time the thought "sure, we can image that" comes to mind. Microsc. Microanal. 21 (Suppl 3), 2015 938 Fig. 1 LSCM edge of GFP expressing E. coli colony. Imaged with 60X oil. Fig.2 Individual E. coli that simply refused to grow into colony forming units. Imaged with spinning disk confocal Fig. 3 Lawn of E. coli after exposure to three different antibiotics. Imaged with a GoPro Hero 3+. Fig. 4 Our custom chamber to allow macro photography within our tissue incubator. Fig. 5 Lawn of E. coli after 3- hour exposure to antibiotics. Imaged with a Nikon D7000 at 35mm Fig. 6 Growing GFP expressing E. coli colonies. Imaged with the Incucyte (20X) (a) and LSCM (60X oil lens with 2X digital zoom) (b)");sQ1[470]=new Array("../7337/0939.pdf","Atomic-Level Fabrication of Crystalline Oxides in STEM","","939 doi:10.1017/S1431927615005498 Paper No. 0470 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic-Level Fabrication of Crystalline Oxides in STEM Q. He,1,2 S. Jesse,1,3A.R. Lupini,1,2 D.N. Leonard,2 M.P. Oxley,4 O. Ovchinnikov,1,4 R. Unocic,1,3 A. Tselev,1,3 M. Fuentes-Cabrera3,6, B. G. Sumpter,1,3,6 S.J. Pennycook,5 S. V. Kalinin,1,3 and A.Y. Borisevich1, 2 1 2 Institute for Functional Imaging of Materials, Oak Ridge National Laboratory, Oak Ridge, TN 37831 Materials Sciences and Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831 3 The Center for Nanophase Materials Science, Oak Ridge National Laboratory, Oak Ridge, TN 37831 4 Dept of Physics and Astronomy, Vanderbilt University, Nashville, TN 5 Dept of Materials Science and Engineering, The University of Tennessee Knoxville, Knoxville, TN 6 Computer Science and Mathematics Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831 Manipulation and control of the matter at the atomic level is one of the ultimate goals in nanoscience. The demonstration of atomic manipulation of xenon atoms by STM [1] was one of the seminal achievements in this direction. Beyond the obvious impact of demonstrating the smallest artificially fabricated structure, this opened the possibilities for the fabrication of novel atomic structures including quantum corrals,[2] standing electronic waves,[3] atomic switches,[4] molecular cascades for atom based computing,[5] and quantum holographic devices.[6] While the STM and non-contact AFM approaches are limited to material surfaces, an alternative paradigm for nano-patterning in bulk material is offered by electron beams. E-beam lithography in scanning electron microscope (SEM) geometry[7] has been demonstrated to fabricate three dimensional structures at the nanometer scale. However atomic-level fabrication is only expected with highly energetic e-beams in (scanning) transmission electron microscope ((S)TEM). Recently in-situ fabrication of metallic nanowires of 2D MoS2 and MoSe2 materials have been demonstrated. However, there are very few reported studies of fabricating bulk material in (S)TEM. For several materials it has been shown that an amorphous area several tens of nm in size can be converted into a polycrystal[8] or single crystal.[9] Recently, atomic rearrangements in amorphous materials have been reported.[10] These observations suggest that (S)TEM beam can in principle be used to achieve sub-nanometer level bulk nanofabrication. In this work, we demonstrate atomic-level sculpting of 3d crystalline oxide nanostructures from metastable amorphous precursor in a scanning transmission electron microscope (STEM). SrTiO3 nanowires can be fabricated epitaxially from the crystalline substrate following the beam path. This method can be used for producing crystalline structures as small as 1-2 nm and the process can be observed in situ with atomic resolution. Two interesting observations are made, as shown in Figure 1: Firstly, there is a certain threshold electron dose required, beyond which the crystallization is not observed. Secondly, the nucleation of the crystal preferably happens at the interface between the amorphous film and the crystalline substrate, and the new crystal grows epitaxially. Assuming an unphysically low thermal conductivity of the amorphous SrTiO3 material, Joule heat induced by electron beam was estimated to be at most 50K, which is still too low to be responsible for the phase transformation. The energy transfer is thus likely to be knock-on in nature. Atomistic molecular dynamics (MD) simulations verified the feasibility of beam-induced crystallization if the high-energy excitation is applied in the vicinity of the amorphous-crystalline interface. We further demonstrate fabrication of arbitrary shape structures via control of the position and scan speed of the electron beam. Combined with broad availability of the atomic resolved electron microscopy platforms, these observations suggest the feasibility of large scale implementation of bulk Microsc. Microanal. 21 (Suppl 3), 2015 940 atomic-level fabrication as a new enabling tool of nanoscience and technology, providing a bottom-up, atom-by-atom, complement to 3D printing. [11] [1] [2] [3] [4] [5] [6] [7] [8] [9] D. M. Eigler, E. K. Schweizer, Nature 1990, 344, 524-526. M. F. Crommie, C. P. Lutz, D. M. Eigler, Science 1993, 262, 218-220. M. F. Crommie, C. P. Lutz, D. M. Eigler, Nature 1993, 363, 524-527. D. M. Eigler, C. P. Lutz, W. E. Rudge, Nature 1991, 352, 600-603. A. J. Heinrich, C. P. Lutz, J. A. Gupta, D. M. Eigler, Science 2002, 298, 1381-1387. C. R. Moon, C. P. Lutz, H. C. Manoharan, Nature Physics 2008, 4, 454-458. C. Vieu, F. Carcenac, A. Pepin, Y. Chen, M. Mejias, A. Lebib, L. Manin-Ferlazzo, L. Couraud, H. Launois, Applied Surface Science 2000, 164, 111-117. aI. Jencic, M. W. Bench, I. M. Robertson, M. A. Kirk, Journal of Applied Physics 1995, 78, 974-982; bA. Meldrum, L. A. Boatner, R. C. Ewing, Journal of Materials Research 1997, 12, 1816-1827. aY. Zhang, J. Lian, C. M. Wang, W. Jiang, R. C. Ewing, W. J. Weber, Physical Review B 2005, 72, 094112; bG. Zhu, G. A. Botton, in The 15th European Microscopy Congress, Manchester Central, United Kingdom, 2012. K. Zheng, C. Wang, Y.-Q. Cheng, Y. Yue, X. Han, Z. Zhang, Z. Shan, S. X. Mao, M. Ye, Y. Yin, E. Ma, Nat Commun 2010, 1, 24. The research is sponsored by the Division of Materials Sciences and Engineering, Office of Basic Energy Sciences, U.S. Department of Energy. BGS and MF-C were supported by the Center for Nanophase Materials Sciences which is sponsored at Oak Ridge National Laboratory by the Office of Science, Basic Energy Sciences, U.S. Department of Energy. Calculations made use of resources at the Oak Ridge Leadership Computing Facility at the Oak Ridge National Laboratory, which is supported by the Office of Science of the U.S. Department of Energy under Contract DE-AC05-00OR22725. (b) [10] [11] (a) Figure 1. HAADF-STEM images of the e-beam fabricated epitaxial SrTiO3 rods. (a) A lower magnification HAADF image of a series of SrTiO3, demonstrating the effect of electron dose and the importance of the substrate. The formation of the SrTiO3 rods is more complete with higher electron dose. However for the left most rod, which was grown using the lowest electron dose, there is still growth near the interface. (b) A higher magnification HAADF image of the region highlighted in (a), showing the epitaxial character of the growth.");sQ1[471]=new Array("../7337/0941.pdf","TEM In situ Plastic Deformation of Silver Nanowires","","941 doi:10.1017/S1431927615005504 Paper No. 0471 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 TEM In situ Plastic Deformation of Silver Nanowires J. Eduardo Ortega1, Diego Alducin1 and Arturo Ponce1. 1. Department of Physics and Astronomy, University of Texas at San Antonio, San Antonio, Texas 78249, United States. Silver nanowires present a series of positive characteristics that give them versatility and a wide range of applications. In this investigation we report the plastic deformation and failure of silver nanowires, produced using the polyol method [1], using a specialized AFM holder (Nanofactory Inst.) to perform a mechanical three point bending test in situ TEM (JEOL 2010F). This holder has a silicon tip and a reference cantilever to measure the applied load. With this arrangement it is possible to observe the AFM tip, the contact radius with the region of interest, changes on the shape of the nanostructure as well as measure the applied force in the range of nanonewtons. The mechanical test of the nanowires was recorded on video by a CCD camera coupled to the microscope at 15 frames per second (fps). The samples rested on the gold bondwire, (Figure 1) and were finely approached towards the AFM tip at a 0.5 nm step by means of a piezoelectric material. From the video it was possible to analyze the mechanical response of the nanowire, in both regimens elastic and plastic, to an applied load making it possible to estimate variables such as the nanowire's suspension length, deflection angle, rupture point and the displacement of the AFM tip, needed to estimate the Young's Modulus (E). AFM imaging and indentation experiments were performed using a Multimode NanoScope V using silicon nitride triangular cantilevers. The normal force was calibrated by recording the deflection of the cantilever as a function of the scanner displacement while in contact with a sapphire substrate. AFM nanoindentation measurements are based on the force plots acquired (on seven nanowires) at five different points of the nanostructure; the tensile modulus in each plot was obtained using a modified Hertz model that correlate the data of the applied force and the indentation depth of the tip. [2] Figure 2 shows the height profile of a silver nanowire and the elasticity distribution values obtained for the experimental indentations. As measured from the recorded experiment and the acquired TEM images, the beam length between supported ends was 475 nm. The average diameter of the nanowire was observed to be 60 nm with an average side of 29 nm. The load of the tip on the nanostructure can be calculated using Hooke's law with the value 155 N/m for the spring constant (k) and the deflection (x) of the tip. To obtain the Young modulus (E) of the Ag NWs during in situ three point bend testing the equation for the maximum deflection of the midpoint of the nanowire =(FL3)/(48EI) where `I' is the moment of inertia of a pentagonal beam as observed from HRTEM images acquired. The calculated Young's modulus of the beam during one of this experiments was calculated to be 109.5 GPa. As the applied load increase the nanowire deflects until it reaches fracture and completely separates. During the deformation event it was possible to detect contrast lines related to surface stresses moving briskly along the beam, reminiscent of a thin membrane on the surface of the structure. An observation of the area of the nanowire at the vicinity with the load, reveled a region without contrast lines which can be associated to the compressive stress, and underneath that region, another with the associated to tensile stress. Nevertheless, failure of the pentagonal nanostructure started on the side where the load was applied. Finally, as the yield stress is reached the beam diameter decreases, leading to a plastic regime which continues until a maximum deflection of 65 nm where and ends with a fracture by Microsc. Microanal. 21 (Suppl 3), 2015 942 necking. Experiments realized with the Multimode's silicon nitride tip attained a Young's modulus of 97.2 � 10.9 GPa which is in agreement with reported data of 94 GPa for this diameter scale [3]. This increase on resistance, with respect of the bulk material, is attributed to a surface effect and the unique five-fold twin microstructure of the silver nanowire [4]. References: [1] Wu B, Heidelberg A, Boland JJ. Nano Lett 6, (2006), p. 468. [2] H. Butt, B. Cappella and M. Kappa, Surf. Sci. Rep. 59, (2005), p. 1. [3] Yong Zhu et al, PHYSICAL REVIEW B 85, (2012), 045443. [4] This project was supported by NSF PREM DMR #0934218 and the Department of Defense #64756RT-REP and the NIH RCMI Nanotechnology and Human Health Core (G12MD007591). Figure 1. Three point bending configuration of the Ag NW with a length of 576nm. The AFM tip has an approximated length of contact on the Ag NW of 60nm. Figure 2. Figure 4 � a) Contact-Friction image of a silver nanowire used for the nanoindentation experiments. b) Height profile of the nanowire highlighted in a). c) Histogram of the effective elastic modulus computed for the approach force plots obtained in contact mode.");sQ1[472]=new Array("../7337/0943.pdf","In Situ Visualization of Metallurgical Reactions in Nanoscale Cu/Sn Diffusion Couples","","943 doi:10.1017/S1431927615005516 Paper No. 0472 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In Situ Visualization of Metallurgical Reactions in Nanoscale Cu/Sn Diffusion Couples Qiyue Yin 1, Fan Gao 2, Zhiyong Gu 2, Eric A. Stach 3, Guangwen Zhou 1 1. Department of Mechanical Engineering & Multidisciplinary Program in Materials Science and Engineering, State University of New York at Binghamton, USA 2. Department of Chemical Engineering, University of Massachusetts Lowell, Lowell, USA 3. Center for functional Nanomaterials, Brookhaven National Laboratory, Upton, USA Pb-free nano solders offer exceptional opportunities for making nanoscale contacts as needed with the miniaturization of devices as well as the production of nanosized circuits. Pb-free solders have been extensively produced in bulk, powder and thin films. However, the knowledge of many properties of Pbfree nanosolders, including diffusion, intermetallic reaction mechanisms and phase evolution is still very limited. Using a two-segment copper-tin (Cu-Sn) nanowire as a model system, in which Sn acts as the solder element and Cu serves as a functional element, we perform an in-situ TEM study to elucidate the phase/structural transformation of various intermetallic compounds (IMCs) and its dependence on the relative length of the Sn and Cu segments as well as the formation and growth of Kirkendall voids by the reactive diffusion between Cu and Sn. The Cu-Sn two-segment nanowires are fabricated by room-temperature sequential electrodeposition assisted with polycarbonate nanoporous membrane templates [1]. The as-prepared Cu-Sn nanowires are kept as a solution in ethanol, followed with ultrasonic dispersion and drop casting onto a lacey carbon film supported by TEM Mo grid, which is then mounted onto a Gatan heating holder controlled by a Gatan temperature ramping stage. In-situ TEM observations of the soldering reaction in the Cu-Sn nanowires are carried out on a JEOL JEM2100F TEM operated at 200 kV. STEM energy dispersive Xray analysis and electron diffraction are performed to confirm the chemical composition and structure evolution of the IMCs during the soldering reaction. A single nanowire is selected for in-situ heating TEM observations. Fig. 1(a) shows the morphological evolution of a nanowire while it is being heated from room temperature (RT) to 495.5 � with the C holding time of ~ 1 min for each temperature interval unless specified. The dashed line denotes the original interface region of the two segments with the Sn segment on the left and the long Cu segment on the right. The Sn/Cu length ratio of the two segments is about 1:3. The nanowire has no obvious changes in morphology until it is being heated to ~ 200 � at which a bulge starts to become visible in C, the Cu segment near the Sn/Cu interface. Concurrent with the bulge formation, the diameter for the Sn segment near its end on the left is shrinking (see the corresponding TEM images at 200 � and 205 � C C). This trend becomes more obvious as the temperature is raised to 207 � the TEM image obtained at this C; temperature shows that a void is formed at the left end of the Sn segment, where the outer layer of the void is a native amorphous Sn oxide layer formed from the sample preparation. With further increase in temperature, the bulge grows larger with more Sn depleted from the end of the Sn segment, as seen from the void growth. Meanwhile, new voids are formed in the Cu segment near the bulge (see the image corresponding to the temperature of 215.5 � The voids in the Cu segment migrate away from the C). bulge and merge, forming a larger void with the continued annealing at 217.5 � for 5 min. Further C increase in the annealing temperature to 364 � and then to 495.5 � results in the void growth that C C Microsc. Microanal. 21 (Suppl 3), 2015 944 eventually leads to the breakage of the Cu segment. The above in-situ TEM observation reveals that the Cu-Sn metallurgical reaction occurs around 200 � and the bulge formation around this temperature C range can be attributed to the formation of the IMC -phase Cu6Sn5, as identified by electron diffraction shown in Fig. 1(b). The -Cu6Sn5 bulge is formed initially in the Cu segment, suggesting that Sn diffuses into the Cu segment to form the IMC. However, once the IMC is formed, the reaction is mainly limited to the Sn segment by Cu diffusion into Sn through the formed IMC. This is evidenced by the void formation in the Cu segment near the IMC/Cu interface due to the Kirkendall effect, i.e., the diffusion of Cu in the IMC is faster than that of Sn in the IMC. With the continued IMC growth, Sn is gradually consumed by reacting with incoming Cu atoms supplied by solid-state diffusion of Cu through the IMC layer, which results in the void growth at the end of the Sn segment. One can see that the void in the Sn segment has no noticeable change after the temperature is raised above 217.5 � and thereafter, C indicating that Sn has been completely consumed to form the IMC at this reaction stage. The continued growth of the Kirkendall void in the Cu segment at the temperature above 217.5 � suggests that more C Cu diffuses to the reacted region to form Cu-rich IMCs while the nanowire is being heated to the higher temperatures. The morphology and length of the Cu segment on the right of the Kirkendall void remains unchanged during the entire annealing process, demonstrating that there is little diffusion of Sn to the Cu segment due to the slower diffusion rate of Sn in the IMC than that of Cu in the IMC [2]. A full description and discussion of how the phase transformation pathway varies with the Sn/Cu length ratio will be presented and compared. References: [1] F. Gao et al, The Journal of Physical Chemistry C 113.22 (2009): 9546-9552. [2] Q. Yin et al. Nanoscale (2015) in press. [3] This work was supported by the National Science Foundation under NSF Collaborative Research Award Grant CMMI-1233806. This research was performed in part at the Center for Functional Nanomaterials, Brookhaven National Laboratory, which is supported by the U.S. Department of Energy, Office of Basic Energy Sciences, under Contract No. DE-AC02-98CH10886. Figure 1. (a) In-situ TEM observation of the morphological evolution of a Cu-Sn two segmented nanowire upon heating from room temperaeture (RT) to 495.5 � The red dashed line delineate the C. interface area between the Cu and Sn segment. The red arrow on the TEM image corresponding to the annealing temperature of 200 � denotes the formation of a bulge in the Cu segment near the Sn/Cu C interface. (b) TEM image with SAED pattern showing the -phase Cu6Sn5.");sQ1[473]=new Array("../7337/0945.pdf","Visualization of Gold Nanoparticle Self-assembly Kinetics","","945 doi:10.1017/S1431927615005528 Paper No. 0473 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Visualization of Gold Nanoparticle Self-assembly Kinetics Taylor J. Woehl1 and Tanya Prozorov1 1. Emergent Atomic and Magnetic Structures, Division of Materials Sciences and Engineering, Ames Laboratory, Ames, IA 50011, USA. Various types of colloidal nanoparticles are known to self-assemble into hierarchical mesostructures via anisotropic interparticle interactions, such as dipolar [1] and electrostatic interactions [2]. Self-assembly strategies are typically designed in an empirical manner or based on equilibrium models of these interparticle interactions, and kinetic considerations are often overlooked. Due to a lack of high spatial and temporal resolution in situ observations, little is known about the kinetics of nanoparticle selfassembly. Previous work has suggested that the kinetics of nanoparticle self-assembly process can be likened to multistep chemical [3] and polymerization [4] reactions, where interparticle interactions determine the rate constants for assembly of differently sized and shaped mesostructures [5, 6]. However, it is not clear how the interplay of interparticle interactions and assembly kinetics affect the hierarchical self-assembly process and resulting mesostructure morphology. Here we use real-time nanoscale observations to measure the self-assembly kinetics of colloidal gold nanoparticles into one dimensional chains. Gold nanoparticles suspended in acetate buffer were observed with in situ liquid cell scanning transmission electron microscopy (STEM) to self-assemble into chains of up to 20 nanoparticles over times of several minutes. Real-time kinetic data and in situ nanometer resolution images revealed the formation pathway and rate of gold nanoparticle self-assembly to be dependent on the imaging electron beam current (Figure 1). The rate of self-assembly increased proportionally with the beam current, and the self-assembly pathway changed from sequential attachment of nanoparticles to formed chains at low beam currents, to chain-chain attachments at high beam currents. Experimental measurements of the nanoparticle diffusion coefficient revealed that the self-assembly process was diffusion driven, where the nanoparticle mobility dictated the self-assembly rate and pathway. Importantly, through these systematic beam current experiments we revealed that the nanoparticle mobility was the underlying control factor for the self-assembly kinetics. We expect these conclusions will shed light on the role of other mobility-controlling factors on nanoparticle selfassembly, such as nanoparticle size, shape, and suspending solvent [7]. References: [1] Z.Y. Tang, N.A. Kotov, M. Giersig, Science, 297 (2002), p. 237-240. [2] H. Zhang, D.Y. Wang, Angewandte Chemie-International Edition, 47 (2008), p. 3984. [3] Q. Zhou et al, Crystengcomm, 15 (2013), p. 5114. [4] K. Liu et al, Science, 329 (2010), p. 197. [5] J. Zhang et al, Journal of the American Chemical Society, 128 (2006), p. 12981. [6] R.L. Penn, Journal of Physical Chemistry B, 108 (2004), p. 12707. [7] We thank Cari Dutch and William Ristenpart for assistance in developing the image analysis algorithms for single particle tracking and diffusivity measurements. T.P. acknowledges support from the Department of Energy Office of Science Early Career Research Award, Biomolecular Materials Program. This work was supported by the U.S. Department of Energy, Office of Basic Energy Science, Division of Materials Sciences and Engineering. The research was performed at the Ames Laboratory, Microsc. Microanal. 21 (Suppl 3), 2015 946 which is operated for the U.S. Department of Energy by Iowa State University under Contract No. DEAC02-07CH11358. Figure 1. Electron beam dependent gold nanoparticle self-assembly kinetics. Time lapsed annular dark field (ADF) STEM images showing self-assembly of gold nanoparticle chains over 150 s at beam currents of (a) and (b) . The scale bar in the final panel of (b) is . (c)-(d) The number of nanoparticle chains with sizes ranging from 1 � 5 nanoparticles, as a function of time for beam currents of (c) and (d)");sQ1[474]=new Array("../7337/0947.pdf","Thermally Driven Cation Exchange at Solid State between Cu2Se and CdSe Nanocrystals: an In-Situ TEM Study.","","947 doi:10.1017/S143192761500553X Paper No. 0474 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Thermally Driven Cation Exchange at Solid State between Cu2Se and CdSe Nanocrystals: an In-Situ TEM Study. Andrea Falqui1a, Alberto Casu2, Alessandro Genovese,1a,2 Liberato Manna2, Paolo Longo3, Gianluigi A. Botton4, Sorin Lazar5, Mousumi UpadhyayKahaly1b, Udo Schwingenschloegl1b 1. a King Abdullah University of Science and Technology, 23955-6900 Thuwal, Kingdom of Saudi Arabia: Biological and Environmental Sciences and Engineering Division; b Physical Sciences and Engineering Division; 2. Istituto Italiano di Tecnologia, via Morego 30, 16163 Genova, Italy 3. Gatan, Inc., 5794 W Las Positas BLVD, Pleasanton, California, United States of America 4. Department of Materials Science and Engineering, McMaster University, Hamilton, Ontario, Canada 5. FEI Electron Optics, Achtseweg Noord 5, Eindhoven, 5600 KA, Netherlands. Cation exchange (CE) reactions in crystals concern the partial or the complete substitution of a cation species, while the anion one keeps its place in the crystalline structure and is essentially preserved in the reaction [1]. CE is commonly based on the very rapid direct reaction occurring in liquid solution between inorganic colloidal nanocrystals (NCs) and cationic species. Due to the highly dynamic nature of the phenomenon as well as to the liquid environment in which it takes place, its direct imaging looks very hard to perform. Copper chalcogenides have been found to be inclined to CE reactions, since in these materials the formation of a great number of copper vacancies offers effective pathways for cation interdiffusion and exchange [2]. Between copper chalcogenides, Cu2Se (NCs) nanocrystals are colloidal nanomaterials well known for their semiconducting, optoelectronic and thermoelectric properties. In particular, antifluorite Cu2Se phase becomes superionic when heated above a threshold temperature, so that Cu cations are able to diffuse randomly with liquid-like mobility around the rigid Se sublattice. Using an in-situ TEM approach, our work first shows what happens to only Cu2Se NCs when the temperature is increased at 400�C: Cu2Se NCs expel free Cu species, forming then Cu-vacancies in cation sublattice, with consequent change of stoichiometry into Cu2-xSe, and with both a concomitant small decrease of the cubic lattice constant and a modification of the electronic structure. Besides, this change in the electronic structure of Cu-depleted Cu2-xSe can be correlated to the appearance of a plasmon absorption band in the electron energy loss (EEL) spectrum, showing a peak at about 1.1 eV, detected only by high-resolution EEL Spectrometry. Ab-initio simulations of the EEL spectrum have also confirmed the appearance of this low-loss energy feature as a consequence of the Cu vacancies formation in the Cu2Se cation sublattice. Furthermore, we report that the free Cu species expelled from Cu2-xSe NCs at 400�C can be exploited to perform solid state CE reactions at local scale by simple thermal activation. In particular, we present the results of thermally driven CE between cubic Cu2Se NCs and wurtzite CdSe nanorods (NRs) or nanowires (NWs), when deposited on the same amorphous substrate (constituted either by thin carbon film or Si3N4 membrane). After this in-situ heating at 400�C, Cu2-xSe NCs show again the Cu-depletion already observed when they were put as only species on the TEM grid and heated at the same temperature. Moreover, CdSe NRs or NWs undergo a pervasive chemical transformation, revealing a total loss of Cd species with concomitant substitution with Cu. Total exchanged-CdSe NRs and NWs experience also a complete structural transformation from wurtzite to antifluorite crystal structure, preserving the close-packing direction of Se atoms in the structures, namely [0001]hcp and [111]fcc. Microsc. Microanal. 21 (Suppl 3), 2015 948 These studies evidence that: a) the shape features of both the nanosized species are basically preserved during the CE reaction, but with CdSe NRs/NWs length decrease; b) in absence of Cu2Se NCs, CdSe NRs or NWs do not undergo any transformation in the same thermal range; c) the CE reaction occurs also in regions not irradiated by the electron beam, indicating to be a simple thermally driven process; d) the CE reaction likely happens due to a migration of the Cu species expelled from the Cu2Se NCs on the amorphous substrate irrespective of its chemical nature, until reaching the CdSe NRs or NWs. Finally, we show how by using the in-situ heating approach, it is feasible following a CE process that, when happening in a liquid environment, is so fast that only the ultimate and stable reaction's product can be studied and directly imaged. References [1] D.H. Son, S.M. Hughes, Y.D. Yin, A.P. Alivisatos, Science 306 (2004), p.1009. [2] A.O. Yalcin et al., Nano Lett. 14 (2014), p.3661. [3] The authors acknowledge funding from the ERC TRANS-NANO project with contract number 614897. Figure 1. Mixture of the Cu2Se NCs and CdSe NRs: a) at RT, before heating; b) at RT, after heating at 400�C; Electron Diffraction patterns of a representative region featuring Cu2Se NCs and CdSe NRs at RT pre- (c) and post- (d) thermal treatment at 400�C with superposition of 1D profile of electron diffraction signal, as obtained by integration over the full round angle in the reciprocal space; e) comparison between these integrated linear profiles collected at RT pre- (magenta) and post- (orange) treatment at 400�C, revealing the phase transformation occurred to the CdSe NRs component of the original mixture. The RT pre-heating electron diffraction profile is indexed as a combination of a cubic Cu2Se and wurtzite CdSe, while after 400�C heating only cubic Cu2-xSe phase is found.");sQ1[475]=new Array("../7337/0949.pdf","Improved Environmental Control and Experimental Repeatability with New In-Situ Devices","","949 doi:10.1017/S1431927615005541 Paper No. 0475 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Improved Environmental Control and Experimental Repeatability with New In-Situ Devices Trevor Moser1,2, Tolou Shokuhfar2,3, and James Evans1 1 2 Pacific Northwest National Laboratory, Env. Mol. Sci. Lab., Richland, WA 99354 Michigan Technological University, Dept. of Mechanical Engineering, Multiscale Technologies Institute, Biochemistry and Molecular Biology Program, Houghton, MI 49931 3 Universities of Illinois at Chicago, Dept. of Physics and College of Dentistry, IL, 60607 Abstract: As liquid cell experiments in electron microscopy have increased in popularity, a number of challenges have emerged surrounding current microfluidic platform limitations. Commercially available liquid flow cell holders trap a small volume of liquid between two thin electron transparent membranes, most commonly silicon nitride supported by a thicker silicon substrate [1]. Often a spacer material, usually either an inert metal or polymer thin film, is used in an attempt to define the fluid path length. This assembly is then loaded within a chamber at the tip of an instrument compatible holder and sealed hermetically with a series of O-rings. Liquid or gas is delivered to the viewing area via microfluidic tubing that travels the length of the holder and empties into wells surrounding the silicon devices without exposure to the vacuum of the instrument. Sample inflow/outflow wells are arranged such that the silicon devices containing the electron transparent membranes are located between wells, and any sample introduced into the hermetic chamber must flow either between or around the silicon devices before exiting through the outflow well. In this way, liquid samples can be imaged while protected from the high vacuum environment of an electron microscope, and exchange/introduction of sample material can be presented to the imaging area with flow capabilities. While these designs have proven effective, a number of limitations have been identified in recent years that continue to limit experimental reproducibility of liquid cell experiments. The dimensions and design of current electron transparent membranes and support geometries confine the effective viewing area to a maximum of 0.01mm2, with many experiments restricted to 0.0025mm2 or less. Additionally, the pressure differential between the vacuum of the instrument and the environmental sample results in significant outward bulging of the electron transparent membranes, increasing the thickness of the liquid cell and further limiting the overall effective imaging areas [2]. For example, many dose related experiments studying water hydrolysis and reaction kinetics have been shown to be volume sensitive, and membrane bulging complicates attempts to quantify beam related damage and reproduce growth, or other dynamic experiments [3]. Finally, the thickness of the liquid layer is also difficult to reproduce even when using patterned spacers of known thickness. Environmental contaminants or the sample itself can function as unintended spacers which dictate the experimental thickness if they become trapped between the surface of one device and the spacer of another, or are themselves larger than the intended nominal spacing. Thickness increases due to sample size or entrapment can be mitigated somewhat by assembling the device and flowing sample particles between the windows once within the instrument. However, holder geometries allow for significant bypass of flow around the chips, which in many cases can be 2-3 orders of magnitude greater than the intended fluid thickness between the membranes, hindering the amount of sample that can be detected in the already very limited viewing area. We will present a new platform for overcoming the limitations listed above, in an attempt to increase the usability of in-situ holders and simplify interpretation and repeatability of liquid cell results. We will Microsc. Microanal. 21 (Suppl 3), 2015 950 highlight results from using this platform to study both soft/biological samples and materials/chemistry systems to demonstrate the versatility of the approach. References: [1] N. de Jonge and F. Ross Nature Nanotechnology 6 (2011) p.695 [2] K. Jungjohann, et al Microscopy and Microanalysis 18 (2012) p. 621 [3] T. Woehl, et al ACS Nano 6 (2012) p. 8599 [4] A portion of the research was performed using EMSL, a national scientific user facility sponsored by the Department of Energy's Office of Biological and Environmental Research and located at Pacific Northwest National Laboratory (PNNL). PNNL is operated by Battelle for the U.S. Department of Energy under Contract DE-AC05-76RL01830. Support was provided by the Department of Energy's Office of Biological and Environmental Research Mesoscale Bioimaging Pilot Program under grant number FWP 66382. The project was partially supported by the National Science Foundation, Award No. 1350734. Figure 1: a) A standard silicon device for use in commercial in-situ holders. A single silicon nitride window can be seen centered on the device. b) Low-magnification bright field STEM image of an assembled liquid cell. Significant bulging of the windows is apparent by the contrast gradient seen across the window. Imaging is often limited to corners where electron transmission is maximized. c) Bright field STEM image of the corner of a silicon nitride window similar to that seen in (b). The contrast gradient indicates thickness changes as a result of window bulging, and highlights the lack of equivalent imaging areas within a typical liquid cell.");sQ1[476]=new Array("../7337/0951.pdf","In situ STEM Investigation of Shape-Controlled Synthesis of Au-Pd Core-Shell Nanocubes","","951 doi:10.1017/S1431927615005553 Paper No. 0476 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In situ STEM Investigation of Shape-Controlled Synthesis of Au-Pd CoreShell Nanocubes Nabraj Bhattarai and Tanya Prozorov Emergent Atomic and Magnetic Structures, Division of Materials Sciences and Engineering, Ames Laboratory, Ames, IA 50011, USA. Properties and applications of nanoparticles (NPs) strongly depend on their shape, size and crystalline structure. The fabrication of NPs with well-controlled shape, size and crystallinity with sufficient yield represents one of the main challenges of the nanoscience and nanotechnology. Despite the vast research, the nucleation and shape-selective growth mechanism of NPs from solution are not fully understood. The in situ scanning transmission electron microscopy (STEM) analysis has become one of the most important technique allowing dynamic investigation of NP growth.[1-3] We studied shape- and sizecontrolled in situ formation of Au-Pd core-shell nanocube from gold NPs. The in situ STEM study was carried out by using a Continuous Flow Fluid Cell holder platform with the reagents sealed between the two 50 nm-thick electron-transparent silicon nitride (SiN) membranes. Visualization of controlled shape-selective synthesis of Au-Pd core-shell nanocubes was performed with the FEI Tecnai G2 F20 STEM operated at 200 kV, using the high angle annular dark field (HAADF) STEM imaging mode. The ex situ STEM micrograph of core-shell nanocube (Figure 1), shows ~ 20 nm cubes with uniform shapes and sizes. These nanocubes are formed from Au octahedral NPs with fast growth of Pd along <111> directions than along <100> direction, when Pd is added onto it and reduced by ascorbic acid (AA).[4] The liquid cell reaction visualized in the static mode is presented in Figure 2a, b. In this case, Au NPs were mixed with Pd precursor and AA and deposited in the SiN window and hermetically sealed. The liquid cell holder platform was inserted into the microscope and STEM analysis was carried out. The interaction of electron beam with the sample and the effect of electron dose are currently investigated. As seen in Figure 2a, b, while some core-shell NPs are formed with Au as core and Pd as the shell, the Pd layer is not uniform. Initially there were very few particles but after an hour several NPs were seen. More than 60% of NPs consisted predominantly of Pd (represented by black circles) indicating excess of Pd in the system. On the other hand, in the areas not directly exposed to the beam, well defined shape and sized NPs are present. In another static in situ experiment, the Pd concentration was reduced, yielding the core-shell structure shown in an inset of Figure 2a. The shape and size of these nanocubes are in good agreement with those obtained ex situ. In order to elucidate the reaction mechanism of core-shell nanocube formation from Au NPs and visualize the reaction in real time, further experiments will involve controlled delivery of the reactants in situ. References: [1] K. L. Jungjohann et al, Nano Letters, 2013, 13, 2964-70. [2] H.-G. Liao et al, Science, 2012, 336, 1011. [3] T.J. Woehl et al, Scientific Report, 2014, 4, 6854. [4] N. Bhattarai et al, Journal of Nanoparticle Research, 2013, 15, 1660. [5] T.P. acknowledges support from the Department of Energy Office of Science Early Career Research Award, Biomolecular Materials Program. This work was supported by the U.S. Department of Energy, Office of Basic Energy Science, Division of Materials Sciences and Engineering. The research was performed at the Ames Laboratory, which is operated for the U.S. Department of Energy by Iowa State University under Contract No. DE-AC02-07CH11358. Microsc. Microanal. 21 (Suppl 3), 2015 952 Figure 1: Ex situ HAADF STEM micrograph of Au-Pd core-shell nanocube as seen in S/TEM grid. The Z-contrast imaging showed that the nanocube is ~20 nm in size with the core made from Au (higher atomic number) and the shell made from Pd (lower atomic number). Figure 2: Low magnification in-situ STEM imaging of the formation of Au-Pd core-shell nanostructures from octahedral Au NPs. (a) Low magnification image shows that core-shell nanostructures are obtained with Au inside (core) and Pd outside (shell). A random area in Figure (a) is zoomed in and is presented in Figure (b) where the core-shell structure is clearly seen (represented by red circled regions). The coreshell ex-situ nanocubes are relatively smaller in size ~ 20 nm, while the in-situ (red circled) are larger in size ~60 nm. Some of the Pd nanocrystals were seen (represented by black black circles). Size controlled AuPd core-shell nanocube obtained from another static in situ experiment by changing the concentration of Pd precursor is inset in Fig. 2a, which shows the nanocube is made from Au core and some finite layer of Pd shell, indicating that the in situ experiment is close agreement with the ex situ experiments.");sQ1[477]=new Array("../7337/0953.pdf","In-situ STEM Observation of Strain Field Movement in a LiMn2O4 Nanowire Battery","","953 doi:10.1017/S1431927615005565 Paper No. 0477 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In-situ STEM Observation of Strain Field Movement in a LiMn2O4 Nanowire Battery Soyeon Lee1,2, Yoshifumi Oshima2,3, Eiji Hosono4, Haoshen Zhou4, Ryoji Kanno5 and Kunio Takayanagi2,6 1. Quantum Nanoelectronics Research Center, Tokyo Institute of Technology, 2-12-1-H-51 Ohokayama, Meguro-ku, Tokyo 152-8551, Japan. 2. JST�CREST, 7-gobancho, Chiyoda-ku, Tokyo 102-0075, Japan. 3. School of Materials Science, Japan Advanced Institute of Science and Technology, 1-1 Asahidai, Nomi, 923-1292, Japan. 4. Energy Technology Research Institute, National Institute of Advanced Industrial Science and Technology, Umezono, 1-1-1, Tsukuba, 305-8568, Japan. 5. Department of Electronic Chemistry, Tokyo Institute of Technology, G1-1 4259 Nagatsuta, Midori-ku, Yokohama 226-8502, Japan. 6. Department of Physics, Tokyo Institute of Technology, 2-12-1-H-51 Oh-okayama, Meguro-ku, Tokyo 152-8551, Japan. To understand lithium ion transport behavior is an issue for improving the performance of lithium ion batteries. Lithium insertion or extraction reaction in most lithium ion battery (LIB) electrode materials causes lattice parameter change. It leads local strain field change. Thus, measuring strain field during charge-discharge cycles allows us to know how lithium ions move inside a LIB material. There are several methods to measure local strain field [1-3]. Among them, scanning moir� fringe (SMF) method is reported to use interference between scan raster of scanning transmission electron microscope (STEM) probe and a lattice plane of a crystal [4]. SMF method has some advantages: SMF image can be taken easily by controlling scan raster distance and direction, and strain field movement can be detected by controlling scan speed. In this study, we have demonstrated in-situ observation of strain field movement inside LiMn2O4 cathode during a charge-discharge cycle by using SMF method. The in-situ observation was performed by using our developed nanowire battery which consists of nanowire LiMn2O4 cathode, ionic liquid electrolyte and Li4Ti5O12 anode [5-6]. The nanowire-battery was loaded in our homemade electrical biasing double-tilt TEM holder. The LiMn2O4 nanowire in the nanowire-battery was observed by annular bright field (ABF) imaging [7-8] of STEM using an aberration corrected TEM, R005 at 300 kV. The convergent semi-angle was 30 mrad. The inner-outer semi-angle of ABF detector was 15-30 mrad. Simultaneously electrochemical properties were measured by cyclic voltammetry. The voltage was scanned from 2.20 to 4.50 V vs Li/Li+ at scan rate of 0.55 mV/s. The measurement was performed by source-measurement unit, Keithley 2635A. The SMF image was taken by tilting the scan direction slightly from the (111) lattice plane of the LiMn2O4 nanowire in order to enhance the sensitivity of the strain field (Fig. 1): the strain change of about 0.1 % was detected. As shown in Fig. 1, the moir� fringe is rotated when the spacing of the lattice fringe is expanded or shrunk. The moir� fringe was straight without charging or discharging the nanowire battery. However, we found Microsc. Microanal. 21 (Suppl 3), 2015 954 that the moir� fringe was bent into the shape of mirrored "S" when lithium extraction or insertion reaction occurred. The "S" moir� indicates that the lattice spacing is gradually expanded and shrunk. Such an "S" moir� was observed discretely in the series of STEM images, indicating that the strain field was not caused by the nanowire own structure. Since upper-half and lower-half of "S" moir� was observed in sequentially obtained two STEM images, it indicates that the strain field moved along the nanowire during a charge-discharge cycle. From the series of STEM images, the estimated velocity of the strain field moving was same order with the velocity to lithium movement estimated from simultaneously obtained cyclic voltammetry curve. Therefore, we conclude that the strain field movement corresponded to the movement of lithium ions due to charging or discharging. From the analysis of SMF images, the detail of lithium transport behaviour will be discussed [9]. References: [1] M. J. Hytch, et al, Ultramicroscopy, 74 (1998), p. 131. [2] K. Tsuda, et al, J. Electron Microsc., 56 (2007), p. 57. [3] M. J. Hytch, et al, Nature, 453 (2008), p. 1086. [4] S. Kim, et al, Appl. Phys. Lett. 103 (2013), p. 033523. [5] S. Lee, et al, J. Phys. Chem. C, 117 (2013), p. 24236. [6] S. Lee, et al, ACS nano, 9 (2015), p. 626. [7] S. Lee, et al, J. Appl. Phys. 109 (2011), p. 113530. [8] S. Lee, et al, Jpn. J. Appl. Phys. 51 (2012), p. 020202. [9] This work was supported by the Japan Science and Technology Agency (JST) under the CREST project. Figure 1. Schematic illustration of scanning moir� fringe method used in this study (rotation moir�). Scan raster (blue) is slightly tilted from lattice plane (green). When lattice spacing is changed (case 1 and case 2), the angle and spacing of moir� fringes (as shown by red arrows) are changed largely.");sQ1[478]=new Array("../7337/0955.pdf","On the Weak Forces on Nanoparticles","","955 doi:10.1017/S1431927615005577 Paper No. 0478 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 On the Weak Forces on Nanoparticles Diego Alducin, Miguel Jos�-Yacam�n, and Arturo Ponce Department of Physics and Astronomy, University of Texas at San Antonio, San Antonio, Texas 78249, USA "As we go down in size, there are a number interesting problems that arise", Feynman famously said in the American Physical Society of 1959 at the California Institute of Technology. What he meant by that, was then little understood, but today as we continually research the field of nanotechnology we can quantitatively observe these "interesting problems". Weak forces, such as the Van der Waals forces, clearly affect the domain in which nanoparticles are observed. Molecular attractions, electrostatics, forces of adhesion and friction are some of the forces involved in the kinematics of nanoparticles. In this work, we manipulated a series of capped gold nanoparticles and quantitatively measured the forces that allowed different types of motions. Combining a scanning probe tool within a transmission electron microscope we manipulated a gold nanoparticle and recorded the sliding and rolling motions of the nanoparticle across a gold surface. Determination of the contact mechanics and kinematical motion of nanoparticles is essential to improve the manipulation and understanding of materials at the nanoscale. Using in situ TEM techniques for probing the nanoparticles to measure and observe the forces involved during manipulation offered significant contributions into the way future problems can be approached. During this experiment we introduced a Nanofactory nano-indenter holder with a silicon AFM cantilever into a transmission electron microscope. Using this holder, several nanoparticles placed on a PZT controlled gold wire was approached to the cantilever. The interaction between the cantilever, the nanoparticles and the gold wire produced different types of motions. The forces within these motions were quantified due to the deflection of the cantilever with a known spring constant (k = 5.45 N/m) and due to the live visual feedback given by the recording of the motion in the TEM. In order to study the mechanism of manipulation using the AFM tip and to measure the forces involved, initially static equilibrium conditions must be established, the free body diagram of these forces is sketched in Figure 1(b). The Hertzian model of contact mechanics with an adhesion force present between elastic bodies was first introduced by Bradley[1]. The rolling and sliding of the particles are associated to particle-substrate friction force that needs to be overcome to produce motion[2]; equivalently adhesion force plays an important role in the detachment of the contact object with the particle, either tip or substrate surface. In this way, the forces considered, based on the diagram of the Figure 1(b). The rolling motion of the particle over the surface was recorded from which a series of frames were extracted for their force analysis in Figure 2. Our results open new insights about weak forces at individual nanoparticles in real time. The data obtained through the direct imaging of the kinematics of a nanoparticle rolling provides a more detailed and quantifiable information about these forces. The interactions between the capped nanoparticle, the tool and the surface showed how fundamental the force of adhesion is at this scale, more specifically how they change as more complex motions are introduced. This Microsc. Microanal. 21 (Suppl 3), 2015 956 work will help provide more insight into how this "weak" forces are much more important at this scale. Figure 1. (a) TEM micrograph of the AFM tip approaching to a gold nanoparticle deposited in a gold wire substrate. (b) Schematic diagram of the experiment and the forces included in the manipulation of the nanoparticle. Figure 2. Rolling motion of the capped gold nanoparticle on gold surface by an AFM tip. (a) initial position, zero degrees of rolling, (b) 17.4� rotation (c) 48.6� rotation (d) 90.3� rotation (e) 127.7� rotation (f) 140.2� rotation. All rotation angles are measured with respect to the initial position (a). The total displacement of the rolling is 204 nm. [1] Bradley RS, Phil Mag 13 (1932), p.853 [2] Saito S, et al, J Appl Phys 92 (2002), p.5140 [3] This project was supported by grants from the National Center for Research Resources (5 G12RR013646-12) and the National Institute on Minority Health and Health Disparities (G12MD007591) from the National Institutes of Health. We also thank support from NSF grants DMR-1103730 and NSF PREM Grant # DMR 0934218.");sQ1[479]=new Array("../7337/0957.pdf","In situ Observation of Annealing Effects in Ga(NAsP) Multi Quantum Well Structures","","957 doi:10.1017/S1431927615005589 Paper No. 0479 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In situ Observation of Annealing Effects in Ga(NAsP) Multi Quantum Well Structures Rainer Straubinger1, Andreas Beyer1, Lennart Duschek1, Tatjana Wegele1, Wolfgang Stolz1 and Kerstin Volz1 Philipps-Universit�t Marburg, Faculty of Physics and Materials Science Center, 35032 Marburg, Germany The epitaxial growth of multi component semiconductor layers such as Ga(NAsP) enables the improvement of laser and transistor devices due to the individually tunable band gap and lattice constant. Furthermore, the material system shows great potential for realizing optical light sources on Si substrate. Due to the low temperature growth conditions used for this metastable material system, point defects are incorporated in the material upon growth. In particular the thermal annealing procedure after growth can reduce the number of crystal defects and thus increase the efficiency of active regions of the devices [1]. One drawback of the thermal treatment is the gradual blur of the interfaces between the quantum wells and the barriers. To improve the understanding of the annealing process and in turn to improve the quality of the material system Ga(NAsP), (scanning) transmission electron microscopy ((S)TEM) investigations are indispensable. Especially the in situ observation of annealing effects during (S)TEM investigations can help to understand the physics behind these processes as many aspects of the formation of such complex materials are still unknown. Thereby the growth and annealing conditions can be optimized to improve the optical properties. From a more fundamental point of view, also the quantitative determination of the chemical composition of the quaternary alloy on an atomic scale as well as an understanding of the local effects like strain or atomic disorder on STEM contrast presents a true challenge, which is addressed in this study. The investigated samples for these investigations were grown by metal organic vapor phase epitaxy (MOVPE) in a commercially available horizontal reactor system (AIX 200). Growth was performed with hydrogen as carrier gas at a temperature of 575�C and a reduced reactor pressure. GaP/Si templates serves as pseudo substrate consisting of about 100 nm thick GaP layer nucleated on an exactly oriented Si(001) substrate. A more detailed description of the defect-free nucleation is given in Reference [2, 3]. For the in situ observation we use the atmosphere gas environmental cell from Protochips Inc. The environmental chip has nine electron transparent windows out of SiN with a diameter of around 10 micron. This allows us, in combination with our double Cs-corrected JEOL JEM 2200 FS operating at 200 kV, to get high resolution images during the annealing process up to 1000�C under nitrogen environment. The images are acquired in scanning transmission electron microscopy (STEM) mode by using a high angle annular dark field detector (HAADF). Using the HAADF detector in combination with a highly focused beam which scans the sample column by column, the intensity of the final image is correlated to the atomic number of the element in the sample. This enables an intuitive way to distinguish between the different regions in the structure of the area of interest (compare figure 1 B, C). To realize this project the first step is to bring a electron transparent sample into the cell. Therefore we cut a regular 20 micron long lamella out of the sample using the focused ion beam (FIB) JIB 4601 F (JEOL) and prepare the area of interest with the H-bar method. This method has the advantage that the electron transparent part of the lamella is protected in between thicker regions. After the thinning process the lamella is lifted with the electron transparent part exactly over one of the SiN windows (compare figure 1 A) and the lamella is fixed by using two tungsten depositions. So far the ex situ observation using TEM has shown that there is a drastic change in the structure of 1. Microsc. Microanal. 21 (Suppl 3), 2015 958 the quantum well, after the annealing up to 975�C. Figure 1. B shows the cross section STEM image of the as grown Ga(NAsP) sample. The brighter line is the Ga(NAsP) quantum well embedded in (BGa)(AsP). One can clearly see that there is no intensity fluctuation as well as no significant change in the width along the quantum well. Figure 1. C shows in comparison the same sample after the thermal treatment up to 975�C. The destroyed area in the quantum well is marked with a black square. The quantum well shows a widening as well as a darker spot in the middle of the widening. At the moment it is not clear what the driving force for this change at the specific position in the quantum well is. It may be that the inhomogeneity of the composition or a thickness fluctuation triggers the change of the quantum well during thermal treatment. As already mentioned the special combination of the atmosphere gas environmental cell and the in situ STEM observation allows us to investigate the thermal treatment at high resolution conditions in real time. These results as well as the quantification of the composition of the quaternary alloy will be presented and discussed [4]. References: [1] S. Gies, et al, Journal of Crystal Growth, 402 (2014), p. 169. [2] K. Volz, et al, Journal of Crystal Growth, 315 (2011), p. 37. [3] B. Kunert, et al, Thin Solid Films, 517 (2008), p. 140. [4] Support of the German Science Foundation (DFG) in the framework of the GRK 1782 ("Functionalisation of semiconductors") is gratefully acknowledged. Figure 1. (A) Shows an SEM image of the electron transparent FIB lamella on its final position over the SiN window. (B) Cross section STEM image of the as grown and (C) annealed Ga(NAsP) quantum well structure.");sQ1[480]=new Array("../7337/0959.pdf","When is Si3N4 not Si3N4? When it is a Low Stress SiNx Membrane Window","","959 doi:10.1017/S1431927615005590 Paper No. 0480 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 When is Si3N4 not Si3N4? When it is a Low Stress SiNx Membrane Window Nestor J. Zaluzec Electron Microscopy Center, NST Div, Argonne National Laboratory, Argonne, Il, USA The use of SiNx windows has been well established as a support media which can be readily employed to observe microstructural evolution during a wide range of studies in the TEM/STEM and SEM's. It is a stable film which is usually grown on top of thin (100-300 �m thick) Silicon frames. SiNx can be chemically treated, heated, and is relatively inert. Generally these films are grown using a chemical process and can be made routinely and reproducibly into films ranging from 10 nm to 1000 nm thick. There are a range of chemical synthesis methods which are used to grow these films which include: The various chemical routes above all have been used to grow nitride films, however depending upon the route employed one can introduce stress in the films which for thin window application often results in rupture, particularly for the very thin self supporting windows which are used in electron microscopy. As a result one of the most common processes to create "low stress" nitride films involves a Low-Pressure Chemical Vapor Deposition (LPCVD) nitride process that uses dichlorosilane and ammonia. Chemically the process is suppose to create nominally pure Si3N4. As a practical issue, one finds that these films are not pure Si3N4, but rather also incorporate both Cl and O. In figure 1, is shown representative spectra taken from 2 different thickness of SiNx films (50 & 200 nm), the chlorine is always present even after plasma cleaning and always to the same magnitude ( ~ 0.5%) relative to Silicon. Measured over multiple films from an individual supplier (thus fabrication process) the Si/Cl ratio is independent of thickness up to ~ 200 nm. This implies that the Cl (and O) are both incorporated into the SiNx film during growth. In figure 2 is a comparison of the O and Cl signal from low stress SiNx films procured from 4 different suppliers. All films tested contained both O and Cl but to varying amounts, which can be seen by the variable high of the Cl K peak), in addition, some of the films contain moderate amounts of Carbon. While the amounts of Cl and O are small, their peaks can become problematic when the SiNx film are used as nanoparticle supports during experiments which involve elemental microanalysis using either X-ray or Electron Energy Loss Spectroscopy. These media "system peaks" can be readily detected / observed and result in a spectral overlap of characteristic signal from small nanoparticles. For example, various L and M shell lines partially overlap the Cl K peaks ( ~ 2.6 keV). Fortunately the Cl/Si ratio appears to be constant for a given supplier, thus affording a method by which the single might be compensated for by scaling to the Si K signal. Microsc. Microanal. 21 (Suppl 3), 2015 960 It is therefore, important , during any microanalysis experiment to characterize the elemental composition of these film, prior to quantification. This also holds true for various "carbon" films which are purchased in bulk from suppliers, which also show contamination/incorporation of various atomic species such as O, P, Ca, K, depending upon the vendor and fabrication method. References [1] Research supported by the U.S. DoE, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357 at the Electron Microscopy Center , NST Division of ANL Figure 1.) SiNx film showing SiL, NK, OK, SiK, and ClK x-ray lines, for two different film thicknesses normalized to the SiK line intensity. Note the decrease in the NK signal for the 200 nm film in this figure is due to x-ray absorption effects. Figure 2.) SiNx films from 4 different suppliers showing SiL, CK, NK, OK, SiK, and ClK x-ray lines. Left spectra normalized at NK. Right spectra normalized at SiK");sQ1[481]=new Array("../7337/0961.pdf","Modifications to Membrane Heater Holders to Enable XEDS in the AEM","","961 doi:10.1017/S1431927615005607 Paper No. 0481 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Modifications to Membrane Heater Holders to Enable XEDS in the AEM N. J. Zaluzec1, A. Janssen2, M. G. Burke2, D. Gardiner3, S. Walden3 1. 2. Electron Microscopy Center, NST Div, Argonne National Laboratory, Argonne, Il, USA Materials Performance Centre, University of Manchester, Manchester M13 9PL, UK 3. Protochips Inc, Raleigh NC. USA Membrane and furnace technology is frequently used today for in situ holder construction to enable elevated temperature TEM/STEM studies. Today, it is becoming increasingly important to not only visualize temperature dependent microstructural evolution but also quantify elemental changes. While both technologies provide high and stable temperatures, neither in their standard configurations readily facilitate X-ray Energy Dispersive Spectroscopy (XEDS) principally due to their sidewall penumbras [1], which can partially or completely block x-ray signal detection similar to that in liquid cell holders [2]. This penumbra can be most easily appreciated by reference to Figure 1. In this work we modified a membrane holder as illustrated in Figure1c and tested it in an FEI Tecnai F20, Tecnai F30 and CM200F TEMs equipped with windowless SDD systems from either EDAX (Apollo) or Oxford (Xmax). To measure the relative performance of a given design, we use a 100 nm thick SiNx window on 200-300 �m Si frame, which is inserted and centered in the holder. The penumbra effect can be assessed by measuring the dependence of the SiK line intensity as a function of holder tilt. In Figure 2a shows the results from a unmodified membrane holder while 2b in a modified configuration. In 2a, one can see the lack of counts at lowest tilts and the clear break in the slope of the Intensity vs Tilt curve where the penumbra no longer limits the detector solid angle at ~ 25�. The initial nearly linear variation is due to the shadowing of the detector by the penumbra. Once the full detector solid angle is no longer limited then the variation for a uniformly thick film will vary as the thickness along the incident beam direction. For this geometry the intensity variation will be an inverse cosine function of tilt (t*=t/cos()). This can be seen in Figure 2b and is not detectable in 2a, due to the extremely limited operational regime. Due to their dimensions and limited space in TEM pole piece gaps, there is little that can be done with furnace heaters until their size is reduced, and results from testing that type of holder are not shown here. Whereas adapting a membrane holder to minimize the penumbra has been successful the application of windowless XEDS to high temperature studies also requires one to establish the limitations due to infrared (IR) radiation emitted by the thermal heating. Figure 3 shows that up to 400�C there is no detectable difference; however, above 450�C a gradual distortion (Figure 4) due to detection of the IR signal because of the absence of an absorbing window is clear. Above 525�C unless the detector is protected from IR by a suitable window, the spectroscopy for a windowless SDD is not realizable in our systems. Further work is currently in progress to investigate low background (Be) shields as a means to mitigate system peaks arising from the holder body. References: [1] Zaluzec N.J. Microsc. Microanal. 20 (2014), p. 1318. [2] Zaluzec N.J et al, Microsc. Microanal. 20(2) (2014), p. 323. [3] Research supported by the U.S. DoE, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357 at the Electron Microscopy Center , Nanoscience and Technology Microsc. Microanal. 21 (Suppl 3), 2015 962 Division of Argonne National Laboratory, and EPSRC Grants #EP/G035954/1 and EP/J021172/1, and DTR Agency Grant HDTRA1-12-1-003. 1A 1B 1C Figure 1. A, Furnace Heater Geometry; B, Membrane Heater Geometry; C, Modified Membrane Geometry. 2A 2B 3 Figure 2. A, Standard Geometry; B, Modified Geometry. Figure 3. Comparison 25 vs 400 Co. 4 Figure 4. Effects to the low energy XEDS data due to IR signal at elevated temperatures.");sQ1[482]=new Array("../7337/0963.pdf","Annular Bright-Field Electron Microscopy Tracking Solid-State Chemical Reaction","","963 doi:10.1017/S1431927615005619 Paper No. 0482 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Annular Bright-Field Electron Microscopy Tracking Solid-State Chemical Reaction Peng Gao1, Ryo Ishikawa1, Eita Tochigi1, Akihito Kumamoto1, Naoya Shibata1, Yuichi Ikuhara1 1 Institute of Engineering Innovation, School of Engineering, The University of Tokyo, Tokyo 113-8656, Japan To track the structural evolution during solid-state chemical reactions and phase transformations, in situ X-ray scattering, nuclear magnetic resonance and electron microscopy are the most commonly used techniques. Although the bulk-based measurement methods enable valuable insights into the evolution of long-range structure, they have inadequate spatial resolution to probe phase boundaries propagation or localized atomic displacements. The atomic number (Z)-contrast high angle annular dark-field (HAADF) in aberration corrected scanning transmission electron microscopy (STEM) makes it possible to capture phase boundaries motion at atomic resolution [1] and track individual atoms migration in realtime [2], but it is almost blind to the light atoms (e.g., lithium) in chemical compounds. In contrast, the recent advancements in the annular bright field (ABF) imaging [3], allow us to quantitatively determine the composition and occupancy of both heavy and light elements columns, providing a unique and powerful tool/technique to study the structures of functional materials and dynamics processes therein. Here we demonstrate an approach based on ABF electron microscopy to visualize a reduction reaction driven by electron beam radiolysis in manganese oxides. Spinel Mn3O4 has three different Mn columns (Fig.1a) and two O columns (Fig.1b). Mn1 columns are much brighter in Z-contrast image Fig. 1a because they contain twice as many cations as Mn2 and Mn4 columns. The O1 columns should be slightly misaligned based on the spinel structure and lean to Mn1 sites as shown in Fig. 1b. As a transition-metal oxide, it's well known that the incident electron beam in electron microscope can knock oxygen out of manganese oxide (via Knotek-Feibelman mechanism), leading to reduction. After exposure to electron beam for a while, spinel structure frame with high concentration of oxygen vacancies becomes unstable. Meanwhile, gained energy from the incident electron beam, the cations become very active and hence the structure tends to convert into rock salt with less nonstoichiometry. Partial cations at Mn4 sites gradually fill the vacant Mn3 columns indicated by the arrows in the Fig. 1c. The other Mn4 atoms fill the Mn2 columns as in a rock salt structure all the cation columns have the same intensity. During cations migration, the oxygen anions shift accordingly. In the rock salt structure, the anions prefer to stay at symmetric positions that are the middle points of two neighboring cations. Therefore, upon reduction O1 columns shift away from Mn1 and the distance of two O1 columns (indicated in the Fig.1b) shortens. Fig.1d shows the distance is decreased by ~1.2% every 10 seconds, equating to an averaged velocity ~0.35 pm/s for O1 columns shift. The reduction firstly occurs at the surface where the oxygen ions can easily be released and then gradually propagates into interior with further exposure. The reaction front planes (i.e., phase boundaries between reduced surfaces and pristine core) that are perpendicular to the viewing direction are also identified from the ABF image. This methodology demonstrated herein enables real-time tracking full atoms motions during a reduction of manganese oxides. Thanks to the high sensitivity to both light and heavy atom columns, the ABF imaging can achieve unprecedented details of dynamic processes. With further improved stability of environmental sample holders in future, ABF electron microscopy will ultimately allow various in situ studies of solid state transformation and electrochemical/chemical reactions at atomic resolution and lead to beneficial technological solutions to industry [4]. Microsc. Microanal. 21 (Suppl 3), 2015 964 References: [1] P. Gao et al, Nature Communications 4 (2013) 2791. [2] R. Ishikawa et al, Physical Review Letters, 113 (2014) 155501. [3] S. Findlay et al, Applied Physics Express, 3 (2010) 116603. [4] The authors gratefully acknowledge the financial support through a Grant-in-Aid for Scientific Research on Innovative Areas "Nano Informatics" (Grant No. 25106003) from Japan Society for the Promotion of Science (JSPS), and "Nanotechnology Platform" (Project No. 12024046) from the Ministry of Education, Culture, Sports, Science and Technology in Japan (MEXT). Figure 1. High resolution HAADF (a) and simultaneously recorded ABF (b) images of Mn3O4 particle from a cold FEG GRAND ARM at 300 kV with spatial resolution of 45 pm. Four types of cations columns and two types of anions columns are labeled respectively. The distance between two O1 columns is calculated to measure the shift of O during reduction. (c) The atomistic model illustrating spinel Mn3O4 structure along the viewing direction of [111]. Purple: Mn. Green: O. (d) Sequential series of ABF images showing the spinel Mn3O4 converts into rock salt MnO and O2. The arrows indicating the Mn3 sites are gradually filled. (e) The distance between two O1 columns [indicated in (b)] is reduced ~3.5% within 40 s, corresponding to ~28 pm.");sQ1[483]=new Array("../7337/0965.pdf","Ex Situ and In Situ (S)TEM of Iron Oxide Nanoparticles Synthesized by Decomposition of an Organometallic Precursor","","965 doi:10.1017/S1431927615005620 Paper No. 0483 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Ex Situ and In Situ (S)TEM of Iron Oxide Nanoparticles Synthesized by Decomposition of an Organometallic Precursor Ryan Hufschmid1,2, Hamed Arami,1 Kannan M. Krishnan,1 and Nigel D. Browning2 Department of materials Science and Engineering, University of Washington, Box 352120, Seattle, WA 98195-2129, USA 2. Fundamental and Computational Sciences Directorate, Pacific Northwest National Laboratory, PO Box 999, Richland, WA 99352, USA Developments in in situ Transmission Electron Microscopy (TEM) have enabled the observation of many chemical and materials processes in relevant environmental conditions and at the necessary length and timescales. The understanding of nanoparticle nucleation and growth phenomena is one area that has benefited tremendously from this technology. By encapsulating a liquid sample between impermeable, electron transparent membranes, i.e. Si3N4[1] or Graphene[2], nanoparticles can be synthesized inside the TEM and imaged throughout the reaction. In this way platinum[3], gold [4],[5], and lead sulfide[6], for example, have been synthesized in situ. In these systems the electron beam facilitates reduction of an aqueous metallic salt precursor, so calibration and control of electron dose is crucial[7]. There has been some investigation of surfactant mediated nanoparticle growth[8], but for the most part in situ TEM techniques have not been applied to the wide variety of organic-phase nanoparticle syntheses. For many nanomaterials, for example the synthesis iron oxide nanoparticles, organic phase reactions provide the necessary synthetic control[9]. Superparamagnetic iron oxide nanoparticles have desirable magnetic properties, combined with general biocompatibility and abundance in nature, making them attractive for a variety of biomedical applications[10]. Because properties are dependent on, for example, particle size, size distribution, morphology, crystallinity, and immediate environment[11], it is important to understand how synthetic conditions affect the growth and nucleation of nanoparticles. The most direct method for characterizing these effects is in situ TEM. Comparing the results of in situ syntheses, driven by electron beam reduction, to ex situ thermal decomposition can provide insight into the relationship between electron dose and temperature. Superparamagnetic iron oxide nanoparticles can also be expected to display unique behavior in solution; in addition to the usual solvation forces and electron beam effects, magnetic forces are also present in this system, increasing the complexity of nanoparticle interactions. Superparamagnetic iron oxide nanoparticles were synthesized by the thermal decomposition of iron(III) oleate, in 1-octadecene, with excess oleic acid acting as a surfactant. Thermal characteristics of the precursor were evaluated using thermal gravimetric analysis (TGA), differential scanning calorimetry (DSC) (both TA Instruments). Nanoparticle phase has been characterized ex situ using - 2 powder Xray diffraction (Bruker F8 Focus), and Raman spectroscopy (Renishaw inVia). Size and size distribution have been determined by fitting Vibrating Sample Magnetometry (VSM) results, and with TEM. Scanning TEM (STEM), bight field TEM, and selected area diffraction were performed on a 300kV Cscorrected FEI Titan with HAADF detector and GATAN CCD. For ex situ syntheses, nanoparticles in organic solvents are deposited on carbon films. In situ experiments were performed using a Hummingbird Scientific (Lacey, WA) liquid stage, with 50nm Si3N4 chips. 1. Microsc. Microanal. 21 (Suppl 3), 2015 966 We have characterized size, size distribution, crystallinity, and phase for iron oxide nanoparticles synthesized under a variety of synthetic conditions. Monodisperse particles can be synthesized with size controlled between approximately 2 and 30 nm by varying the excess surfactant (i.e. oleic acid) ratio. Nanoparticles are generally single crystalline, as confirmed by atomic resolution TEM and STEM (Figure 1 A, B), and monodisperse. Monodisperse nanoparticles self-assemble into close packed arrays. We discuss the considerations for imaging nanoparticles in organic solvents in situ (Figure 1D), and present some of the first results of iron oxide nanoparticles interacting in their synthetic environment [12]. Figure 1. High resolution TEM (A, 5 nm scale bar) and STEM (B, 2 nm scale) images of iron oxide nanoparticles. STEM images of iron oxide nanoparticles self-assembled on a carbon film (C, 50 nm scale), and in solution (iron oleate, 1-octadecene, and oleic acid) in the liquid stage (D, 50 nm scale). References: [1] M. J. Williamson et al, Nat. Mater. 2 (2003) p. 532�536. [2] J. M. Yuk et al, Science 336 (2012) p. 61�64. [3] H. Zheng et al, Science 324 (2009) p. 1309�1312. [4] N. de Jonge et al, Ultramicroscopy 110 (2010) p. 1114�1119. [5] J. M. Grogan, L. Rotkina, and H. H. Bau, Phys. Rev. E 83 (2011). [6] J. E. Evans et al, Nano Lett. 11 (2011) p. 2809�2813. [7] P. Abellan et al, Chem. Commun. 50 (2014) p. 4873. [8] L. R. Parent, et al. ACS Nano 6 (2012) p. 3589�3596. [9] J. Park, et al. Nat. Mater. 3 (2004) p. 891�895. [10] K. M. Krishnan, IEEE Trans. Magn. 46 (2010) p. 2523�2558. [11] H. Arami et al, Med. Phys. 40 (2013) p. 071904. [12] This work was supported by NIH grants 1RO1EB013689-01/NIBIB, 1R41EB013520-01 and 1R42EB013520-01. Part of this work was conducted at the University of Washington NanoTech User Facility, a member of the NSF National Nanotechnology Infrastructure Network (NNIN). Research performed at Pacific Northwest National Laboratory (PNNL) was supported by the Chemical Imaging Initiative under Contract DE-AC05-76RL01830 operated for the Department of Energy by Battelle. A portion of this work was performed at EMSL, a national scientific user facility sponsored by the Department of Energy's Office of Biological and Environmental Research and located at PNNL.");sQ1[484]=new Array("../7337/0967.pdf","Diffraction-Ring Contraction as a Method of In Situ Thermometry in TEM","","967 doi:10.1017/S1431927615005632 Paper No. 0484 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Diffraction-Ring Contraction as a Method of In Situ Thermometry in TEM Daniel R. Cremons1 and David J. Flannigan1 1 Department of Chemical Engineering and Materials Science, University of Minnesota, Minneapolis, MN, 55455, USA In situ and modus operando transmission electron microscopy (TEM) enable visualization of atomicscale phenomena occurring under conditions that mimic device operation and chemical reaction environments. These techniques have extended the capabilities of static TEM to new systems and dynamic processes previously experimentally inaccessible due to technical limitations. In situ heating experiments, made possible by specially-designed specimen holders, are being driven by interest in understanding heat transport, phase transitions, and thermal properties on the nanoscale [1-5]. For most heating holders, the specimen temperature is indirectly measured via incorporation of a thermocouple into the device. As such, the true steady-state temperature of the specimen in relation to the thermocouple or thermistor reading can be a potential source of error. Typically, in situ specimen thermometry methods involve calibrating the current applied to a resistively-heated holder to the melting point of a well-characterized material. As it is a binary measurement at the melting point, however, application over the entire temperature range accessible with the holder can be challenging. Thus, ideal methods for in situ lattice thermometry would enable continuous use over a wide temperature range and would directly reflect the true specimen temperature. Here, we present a method for directly and continuously measuring specimen temperature in TEM via coefficient of thermal expansion (CTE) and sub-pixel analysis of parallel beam electron diffraction (PBED) patterns. To accurately measure lattice expansion, and thus temperature, from PBED patterns, one must determine the center position of the direct beam. This can be done using a circular Hough transform (CHT) allowing for sub-pixel determination of diffraction ring radii [6]. Here, we used polycrystalline aluminum films as a test of the method due to its large CTE and thus significant diffraction ring contraction. By comparing measured ring contraction at specific heating holder settings to that expected from the CTE of aluminum, an accurate calibration of the thermocouple reading with respect to lattice temperature from 20 to 340 �C was obtained. This indicates that, for the specific heating holder used, the thermocouple reading is an accurate indicator of true specimen temperature. In addition, the {200} and {111} planes follow the same trend with respect to the CTE, indicating the aluminum film is expanding isotropically over the temperature range measured. The Gatan Model 652 Mark II Double Tilt Heating Holder used in the experiments described here has a resistively-heated crucible for changing specimen temperature. Due to the Debye-Waller effect, one would expect the intensity of Bragg spots or diffraction rings to decrease with increasing heating holder temperature. No such effect was observed over the experimental temperature range because measured intensity changes of the Bragg spots are a convolution of the Debye-Waller effect and a change in excitation error due to uncontrolled specimen holder tilting. Imprecision in the mechanical control of the -tilt axis leads to uncontrolled tilting during crucible heating. By following the center-of-mass of a single-crystal silicon diffraction pattern over the course of heating, the real-space tilting of the specimen at each temperature can be estimated [7]. Microsc. Microanal. 21 (Suppl 3), 2015 968 References: [1] J. G. Brons and G. B. Thompson, Thin Solid Films 558 (2014) 170-175. [2] M. Gandman, et al., Phys. Rev. Lett. 110 (2013) 086106. [3] Z. M. Peng, et al., J. Catal. 286 (2012) 22-29. [4] T. Brintlinger, et al., Nano Lett. 8 (2008) 582-585. [5] K. H. Baloch, et al., Nat. Nanotechnol. 7 (2012) 315-318. [6] D. R. G. Mitchell, Ultramicroscopy 108 (2008) 367-374. [7] The authors acknowledge support from 3M in the form of a Nontenured Faculty Award (Grant #13673369), Seagate, and from the Donors of the American Chemical Society Petroleum Research Fund in the form of a Doctoral New Investigator Grant PRF#53116-DNI7. Thin film deposition was carried out in the Minnesota Nano Center, University of Minnesota. Parts of this work were carried out in the Characterization Facility, University of Minnesota, which receives partial support from the NSF through the MRSEC program. Figure 1. (a) Parallel-beam diffraction pattern of as-deposited aluminum film at 1.5 m camera length. (b) Hough transform of the (111) diffraction ring at a Hough radius, rHough = rring. (c) Full radial integration line scans of (111) peaks for ten temperature points of one trial showing the reciprocal lattice-vector contraction due to heating. (d) Comparison of fitted experimental diffraction ring shifts to theoretical shifts based on the CTE for aluminum. (e) Estimated uncontrolled tilting at each thermocouple reading for a single-crystal silicon specimen via center-of-mass determination.");sQ1[485]=new Array("../7337/0969.pdf","Studying Perovskite-based Solar Cells with Correlative In-Situ Microscopy","","969 doi:10.1017/S1431927615005644 Paper No. 0485 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Studying Perovskite-based Solar Cells with Correlative In-Situ Microscopy J. A. Aguiar1, S. Wozny2, W. Zhou2, H. Guthrey1, H. Moutinho1, A. G. Norman1, C. S. Jiang1, J. Berry1, K. Zhu1, T. Holesinger3, and M. M. Al-Jassim1 1. 2. National Renewable Energy Laboratory, Golden, CO, 80401 University of New Orleans, New Orleans, LA, 70148 3. Los Alamos National Laboratory, Los Alamos, NM, 80465 Hybrid organic-inorganic perovskite based solar technologies are generating a great deal of interest in the materials community. In this work, we plan to discuss our ongoing work to characterize new synthesis routines and processes to generate sustainable and reliable perovskite-based materials whose properties are significantly better than the current state of the art. In particular, we are utilizing the latest advances in high-resolution analytical and in-situ microscopy to characterize these emerging photovoltaic materials and their interfaces, defects, and discrete paths to crystallization. There are longstanding interests in the relation between growth, microstructure, defects, electronic structure, and electro-optical activity for two reasons. The first reason is studies suggest there is a great deal of variability in the growth of these materials. The fundamental origins however remain nascent due to the current novelty of these materials and the complexities associated with studying beam-sensitive materials. Furthermore, beyond static conditions as shown in Figure 1, observing transient behavior associated with the growth of these materials is of increasing importance to set future research directions for generating next generation perovskite-based solar cells. The second reason is a more complete understanding of the microstructure, growth defects, and doping behavior and how they affect the efficiency of devices will be crucial in developing both sustainable and efficient perovskite-based solar technology. This talk will present the correlations between our current ongoing observations and measurement of the optical properties, microstructure, defects, and crystallization associated with perovskite-based solar cells. The guided use of the latest high-resolution state-of-the-art cathodoluminescence and in-situ (scanning) transmission electron microscopy (S/TEM) techniques to examine growth, crystallization, and material stability of this exciting class of materials will also be discussed at length [1]. Reference: [1] This work was supported by the National Renewable Energy Laboratory as a part of the NonProprietary Partnering Program under Contract No. DE-AC36-08-GO28308 with the U.S. Department of Energy. Microsc. Microanal. 21 (Suppl 3), 2015 970 Figure 1. (a) Above is an overview diffraction contrast TEM image of an as grown hybrid organicinorganic perovskite film. (b) Based on the image contrast and subsequent chemical profiling with lower voltage STEM-based imaging we have identified that the perovskite films grown to date are not phase pure, but contain regions of PbI2 (dark contrast) and CH3NH3PbI3 (light contrast). (c) The crystallinity of the film is further revealed with selected area electron diffraction, where both PbI2 and CH3NH3PbI3 diffraction spots are revealed. (d) These results highlight the potential use of in-situ based microscopy where the effects of humidity, temperature, and solvent chemistry could be varied to lock-in on possible combinations to grow phase pure and potentially more stable perovskite based solar cells.");sQ1[486]=new Array("../7337/0971.pdf","Quantitative Analysis of Nucleation and Growth Behavior from in situ Liquid Cell Studies","","971 doi:10.1017/S1431927615005656 Paper No. 0486 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantitative Analysis of Nucleation and Growth Behavior from in situ Liquid Cell Studies Anton V. Ievlev1,2, Stephen Jesse1,2 , Sergei V. Kalinin1,2, Thomas J. Cochell3, and Raymond R. Unocic1,2 1 2 Institute for Functional Imaging of Materials, Oak Ridge National Laboratory, Oak Ridge, TN 37831 Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, TN 37831 3 Chemical and Materials Engineering, University of Kentucky, Lexington, KY 40506 Developing a fundamental understanding of nanocrystal nucleation and growth have important implications for a wide range of materials used in catalysis, energy storage, photonic, and electronic applications where the ability to tune crystal size, morphology, and surface chemistry will dictate nanoscale properties and functionality. Here we develop a universal approach for directly probing the kinetic laws, anisotropy, and particle-particle interactions from operando experiments of crystal growth and advanced data analytics. Direct observations of nanocrystal growth from liquid phase precursors can be visualized at high spatial resolution using in situ liquid cell scanning transmission electron microscopy (S/TEM) [1]. In this approach, a solution is encapsulated between electron transparent silicon nitride membranes then irradiated with an electron beam to induce a chemical reduction reaction by radiolytically generated aqueous electrons (eaq-). It has been shown that the electron dose can directly influence the growth behavior from reaction-controlled growth to diffusion-limited growth as interpreted by growth kinetic measurements [2-3]. In the present study a 1 mM K2PtCl6 + 0.5 M H2SO4 solution was used to study the nucleation and growth mechanisms of Pt nanostructures by performing controlled irradiation studies with an aberration corrected FEI Titan S/TEM operating at 300kV. The focused STEM probe, generally used for imaging, is used here to both stimulate and directly image the chemical reduction processes. Figure 1 shows a sequence of annular dark field (ADF) STEM images acquired during electron beam irradiation. There are multiple nucleation sites that appear at different time-intervals, which are located either on the top membrane (in focus) or bottom membrane (out-of-focus) within the liquid cell. From the in situ dataset, we utilized quantitative image analysis methods to analyze the nucleation and growth behavior of the Pt nanostructures by first performing a binary threshold to each frame followed by automated edge detection, center of mass, and particle tracking algorithms. This approach was used to track two particles indicated in Figure 1c and Figure 2 shows the image analysis results for both particles. The time-resolved edge outlines reveal how the shapes of the particles evolve as a function of time (and cumulative electron dose). The time and angular dependence (in polar coordinates) are shown for the growing particles in terms of absolute and normalized radii. The color scale in Figure 2 correlates to the value of the particle radius for each angle and at each time step. Maps of the absolute radius show a monotonic increase, which suppresses data analysis. Normalization over certain time steps allow for characteristic features of the growth caused by the interaction between particles to be revealed. Cooperative dynamics of particle growth can be presented as a 3D graph (Figure 3). The particles exhibit a diffusion-limited growth behavior. There is a decrease in the growth velocity up to the point of nanoparticle contact, which is suggestive of an influence of overlap of the diffusion fields. The interface of the particles Microsc. Microanal. 21 (Suppl 3), 2015 972 away from the point of contact and not impeded by an overlap in diffusion fields and continues to grow at a constant velocity. The combination of in situ liquid cell microscopy and quantitative image analysis provides new insight into the basics mechanisms that control the nanocrystal growth behavior and reaction kinetics from liquid phase precursor solutions [4]. References: [1] HG Liao, K Niu, and H Zheng. Chemical Communications 49 (2013) p. 11720-11727. [2] TJ Woehl et al. ACS Nano 6 (2012) p. 8599-8610. [3] TJ Woehl et al. Nano Letters 14 (2014) p. 373-378. [4] Research supported by Oak Ridge National Laboratory's Center for Nanophase Materials Sciences (CNMS), which is a U.S. Department of Energy, Office of Science User Facility. Quantitative image analysis software available at CNMS. Figure 1. a-f) Time-lapsed ADF STEM images showing the electron beam induced nucleation and growth of Pt from a 1 mM K2PtCl6 + 0.5 M H2SO4 solution. The electron dose rate is 196 e-/nm2s and the scale bar is 200 nm. Figure 2. Quantitative image analysis showing the growth evolution of Particle 1 and Particle 2 (Figure 1c) as a function of time. Edge detection used to illustrate shape-evolution during particle growth and extract absolute and normalized values of particle radius as a function of time (shown in polar coordinates) Figure 3. Particles nucleation and growth graph. Each point on the graph corresponds to the particle at the certain time step. Color correlates with particle area.");sQ1[487]=new Array("../7337/0973.pdf","Liquid Cell TEM of Al Thin Film Corrosion under Potentiostatic Polarization","","973 doi:10.1017/S1431927615005668 Paper No. 0487 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Liquid Cell TEM of Al Thin Film Corrosion under Potentiostatic Polarization See Wee Chee1,2, Jeung-Hun Park3, Ainsley Pinkowitz1, Brent Engler1, Frances M. Ross4, David Duquette1 and Robert Hull1 Department of Materials Science and Engineering, Rensselaer Polytechnic Institute, Troy, NY 12180, USA. 2. Center of BioImaging Sciences, Department of Biological Sciences, National University of Singapore, Singapore 117557. 3. Department of Materials Science and Engineering, University of California Los Angeles, Los Angeles, CA 90095, USA. 4 IBM TJ Watson Research Center, Yorktown Heights, NY 10598, USA Controlling the detrimental effects of metal corrosion is a key concern in many aspects of our modern society. In research looking to elucidate the mechanisms that govern corrosion behavior of a metal, electrochemical techniques play an essential role [1]. While it has been shown that liquid cell transmission electron microscopy can visualize in situ the corrosion structures that form in metal thin films immersed in aqueous media [2], integration with electrochemical measurements is needed if crucial information is to be obtained about the underlying processes. Electrochemical cells for in situ TEM have seen significant development due to the needs of battery research [3], but there are technical challenges in translating the method to corrosion research. In particular, the working electrode in most corrosion experiments is the metal of interest itself. However, in the electrochemical cells for liquid cell TEM, the sample is attached to pre-patterned thin film electrodes, commonly made of Au or Pt. The exposed contact between an active metal and a noble metal can lead to rapid galvanic corrosion of the active metal (see Figure 1). Here, we will describe experiments in which a blanket Al thin film (100 nm thick) is deposited over one of the two blank chips (without pre-patterned electrodes, see Figure 2(a)) that make up the microfluidic cell in a Hummingbird Scientific liquid flow holder. The film makes contact with one of the three metal leads as the working electrode and the other two leads are used as reference and counter electrodes respectively. The electrolyte is a mixture of 0.1M Na2SO4 and 0.001M NaCl dissolved in de-ionized water. The solutions are de-aerated by bubbling nitrogen gas overnight and fluid flow is maintained at 5 �l per min using a syringe pump. Potentiostatic polarization is carried out with a Gamry Reference 600 potentiostat with concurrent observations recorded at TV rate (30 frames per sec) in a FEI CM30 TEM operated at 300kV. Figure 2 (c)-(e) shows successive images in which the dissolution of individual grains (seen with dark diffraction contrast) in the film is observed as the potential is raised from -200 mV in 0.5 mV/s steps. As expected, the dissolution takes place in the anodic portion of the curve. These experiments demonstrate direct observation of localized corrosion under potentiostatic control, but the design suffers from two issues. First, corrosion is most rapid at the contact between the film and metal lead, which can lead to loss of electrical contact during an experiment. Second, corrosion takes non-uniformly over the entire film, making it challenging to quantitatively correlate the electrochemical information with the observed phenomena at the silicon nitride window. Strategies to mitigate these issues will be discussed. 1. Microsc. Microanal. 21 (Suppl 3), 2015 974 References: [1] G.S. Frankel and M. Rohwerder in "Encyclopedia of Electrochemistry, Volume 4, Corrosion and Oxide Films" ed. M. Stratmann and G.S. Frankel (Wiley-VCH, Weinheim), p. 687. [2] S.W. Chee et al. Microsc. Microanal. 20 (2014) p. 462, Chem. Comm. 51 (2015), p. 168 [3] C.M. Wang, J. Mater. Res. 30 (2015), p.326. [4] The authors acknowledge funding from the National Science Foundation, grant number DMR1309509. Figure 1. Light microscope image of Al film deposited through a shadow mask over Au electrodes of an electrochemical chip after 12 mins in 0.1 M NaCl solution. Notice that areas in contact with bare Au experienced significant Al dissolution. (e) (a) (b) (d) (c) Figure 2. (a) Light microscope image of the Al working electrode (b) Polarization curve of an Al thin film in 0.1 M Na2SO4 and 0.001 M NaCl solution. (c-e) Snapshots of film microstructure at different points of the polarization (from -200 mV to 500 mV) showing the dissolution of single grains (indicated with white arrows). Note that this is the second time that the sample has been polarized in the solution. Scale bar is 50 nm.");sQ1[488]=new Array("../7337/0975.pdf","Structural Analysis of Carbon Nanotubes of Various Diameters Grown by Spray Pyrolysis using Raman Spectroscopy","","975 doi:10.1017/S143192761500567X Paper No. 0488 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Structural Analysis of Carbon Nanotubes of Various Diameters Grown by Spray Pyrolysis using Raman Spectroscopy E.G. Ordo�ez-Casanova1, 2, M. Rom�n-Aguirre, 2, A. Aguilar-Elguezabal, 2, F. Espinosa-Maga�a, 2 Universidad Aut�noma de Ciudad Ju�rez. AV. del Charro 450 N., C.P. 32310, Ju�rez, Chih, M�xico. Centro de Investigaci�n en Materiales Avanzados, S.C., Miguel de Cervantes 120, Chihuaua, Chih., M�xico 31109 2 1 Carbon nanotubes of various diameters and length were grown through the alternative spray pyrolysis method [1-2], using propanol, buthanol and cyclohexanol as the carbon source, at temperatures higher than 700�C. In this investigation,we present a comparative analysis of the structures and morphology of the three samples by Raman spectroscopy in order to verify whether it is possible, by using this method, to obtain carbon nanotubes of small diameters at a low cost. The Raman spectra were acquired by the LabRam Horiba HR system using a He- Ne laser at 632.8 nm and 14.2 mW, equipped with a CCD detector column at 75 � C. The resolution obtained was of approximately 1 cm -1. According to the literature, this study presents the responses found at different frequency regions: (1) The peaks occurring at a low frequency < 200 cm-1 are characteristic only of SWNT assigned to RBM (Radial Breathing Mode), and their frequency depends essentially on the diameter of the nanotubes [3]. (2) The peak at 1340 cm- 1 is assigned to the poorly-organized residual graphite. In this region the peak is related to the so-called D band (Band Disorder) [4-5].(3) The frequencies between 1500 cm-1 and 1600 cm - 1 are related to the G band and are highly characteristic of single nanotubes and multiple wall, These frequencies correspond to a splitting of the graphite extension mode (stretching mode) [3-5]. (4) A second order mode vibration, between 2450 cm-1 and 2650 cm-1 , is assigned to the first on-tone mode and to the D mode, sometimes called G� [3-5]. Figure 1(a) shows the Raman spectrum of the propanol sample at high frequencies, from 500 to 3000 cm . The characteristic D band and G band peaks are observed. Band D is at 1321 cm -1 , and it is more intense than the G band at 1582 cm-1 given. This indicates the presence of poorly organized graphite and the presence of possible defects in carbon nanotubes. The presence of the band G` 2635 cm -1, corresponds to second order dispersal processes [6]. The fact that both bands, D and G`, have high intensity peaks indicates structural imperfections in the sample. Figure 1(b) shows a Raman spectrum of low frequency region. Three peaks around 161 cm -1 , 223 cm -1 and 267cm -1 are featured. The first two peaks indicate the presence of single-walled nanotubes with different chirality, and the third peak nearby 267 cm -1, corresponds to the response of the quartz substrate used for the growth of nanotubes. -1 Figure 2(a) displays the Raman spectra of the butanol sample in the high frequency region. We can observe that Band D is at 1340 cm -1 , and Band G is given at 1587 cm -1 and 2663 cm-1. Bands G` and D , as propanol , are stronger than Band G, thus indicating a high density of structural imperfections, possibly poorly-organized graphite. In Figure 2(b), the low frequency region, we can observe peaks at 154 cm -1 and 229 cm -1 and 294cm � 1. By considerating only the first two peaks, the possible presence of single-walled nanotubes of various types of chirality can be confirmed. The peak around 294 cm-1 is related to the substrate.The cyclohexanol sample, where carbon nanotubes were grown at temperatures of 850 C, can be observed in Figure 3(a). Here, Band D is located at 1332 cm- 1, Band G at 1597 cm-1 Microsc. Microanal. 21 (Suppl 3), 2015 976 972 and Band G` at 2657 cm -1. In this sample, the intensity of Band D is lower, indicating that the sample has a low density of structural imperfections as well as poorly-organized graphite.Figure 3(b). Shows low energies where two predominant curves around 147 cm-1 and 186 cm-1 are observed, indicating two possible types of single-walled nanotubes with different diameters. The other peaks present are related to the quartz substrate. In conclusion; carbon nanotubes were grown at a temperature of 700�C for propanol and buthanol. In the case Cyclohexanol, they were grown at 850�C. Raman spectroscopy reveals that the three samples containing the presence of possible carbon nanotubes of small diameters answer to regions of very low frequency. However, all samples with structural defects by analysis of their responses occurred at high energies, which challenges us to continue testing to get better quality nanotubes and small diameters at a low cost. References: [1] Ordo�ez-Casanova, E. G., Rom�n-Aguirre, M., Aguilar-Elguezabal, A.,Espinosa-Maga�a, F. (2013). Synthesis of Carbon Nanotubes of Few Walls Using Aliphatic Alcohols. Materials, 6 (6), 2534-2542. [2] Ordonez-Casanova, E., et al. "Characterization of Few-Walled Carbon Nanotubes Using Alcohols Aliphatic as Carbon Source." Microscopy and Microanalysis 19.S2 (2013): 1608-1609. [3] Dresselhaus, M. S., Dresselhaus, G., Saito, R., & Jorio, A. (2008). Raman spectroscopy of carbon nanotubes. Contemporary Concepts of Condensed Matter Science, 3, 83-108. [4] Bachilo, S. M., Strano, M. S., Kittrell, C., Hauge, R. H., Smalley, R. E., & Weisman, R. B. (2002). Structure-assigned optical spectra of single-walled carbon nanotubes. Science, 298(5602), 2361-2366. [5] Bergstrom Jr, R., & Knoesel, E. (2007). Raman Spectroscopy of Carbon Nanotubes. Bulletin of the American Physical Society, 52. [6] Dresselhaus, M. S., Dresselhaus, G., & Jorio, A. (2007). Raman spectroscopy of carbon nanotubes. The Journal of Physical Chemistry C, 111(48), 17887-17893 . 151cm-1 1340 cm-1 D D IG/ID= .34 1321 cm-1 287cm-1 223cm-1 Buthanol Intensity (a.u.) RBM/Propanol 1587cm-1 G Intensity (a.u.) Intensity (a.u.) PROPANOL 2663 cm-1 G` 1582 cm-1 G G` -1 2635 cm 500 100 200 300 400 500 1000 1500 2000 2500 3000 Raman shift (cm-1) RAMAN shift (cm-1) 500 1000 1500 2000 2500 3000 RAMAN shift (cm-1) Figures 1a, Raman spectrum of the propanol sample at high energies.. 1b Raman spectrum of the propanol sample at low energies.2a, Raman spectrum of the buthanol sample at high energies. -1 1597 cm -1 154 cm RBM/Buthanol Cyclohexanol 850� C Intensidad (u.a) -1 1332 cm Intensidad (u.a) -1 229 cm -1 294 cm -1 2657 cm Intensidad (u.a) RBM/Cyclohexanol 850 �C 147 cm-1 186 cm-1 500 1000 1500 2000 2500 3000 200 400 300 Raman Shift (cm-1) RAMAN shift(cm-1) RAMAN shift(cm-1) Figures 2 b Raman spectrum of the buthanol sample at low energies.3a Raman spectrum of the cyclohexanol sample at high energies.3b. Raman spectrum of the cyclohexanol sample at low energies.");sQ1[489]=new Array("../7337/0977.pdf","Cu/MWCNT Interfaces in Room Temperature Aged Nanocomposites Revealed by HRTEM and EELS","","977 doi:10.1017/S1431927615005681 Paper No. 0489 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Cu/MWCNT Interfaces in Room Temperature Aged Nanocomposites Revealed by HRTEM and EELS M.E. Mendoza1, I.G. Sol�rzano2, T. Aoki3, D.J. Smith3 1 2 Materials Division, Inmetro, Duque de Caxias, Rio de Janeiro, 25250-020, Brazil. Department of Materials Engineering, PUC-Rio, Rio de Janeiro, PO Box 238097, Brazil 3 Department of Physics, Arizona State University, Tempe, AZ 85287-1504, USA The success of a metal-matrix bulk nanocomposite depends upon the nature, volume fraction, distribution of reinforcement, and the interface with the matrix. Multi-walled carbon nanotubes (MWCNT) have significantly improved the mechanical properties of composites when reinforced. However, transport properties (namely electric) have been observed to decrease [1,2]. Little information about the Cu-CNT interface structure and chemistry (responsible for the mechanical and transport properties) has been reported, mainly due to the lower solubility of C in Cu. In this study, Cu-MWCNT bulk nanocomposites were produced by chemical synthesis followed by thermo mechanical processing using spark plasma sintering (SPS). This present investigation reports the structural and spectroscopic characterization of a Cu-0.5 wt (%) and Cu-5 wt (%) MWCNT nanocomposites using TEM and EELS. Nanocomposite powders were synthesized according to the procedure described elsewhere [3]. Bulk pellets were produced by sintering the powder material using a Dr. Sinter lab SPS 1050 at 600�C by 5 min, heating rate of 100 o C/min, 70 MPa compression under 10-3 Torr. A JEOL 4000EX operating at 400 kV and a JEOL ARM200F operating at 200KV were used for characterization. Sample preparation was done using a Nanonova focused-ion-beam instrument. After consolidation into pellets and sintering, the Cu matrix presented a good consolidation state, exhibiting equiaxed grains with characteristic annealing twins and also some heterogeneous grain growth, with grain size in the 50nm�3m range, as shown in Fig.1a. An important finding is the transformation of most MWCNTS into graphite and also amorphous carbon, mainly at copper grain boundaries. It is believed that high current density, as well as high temperature and pressure during processing, played a major role to catalyze this transformation. However, some MWCNTs with diameters above 50 nm were found (inset Fig. 1a). Fourier analysis has shown localized regions (5-20 nm length) at the Cu-MWCNT interface where inter-diffusion between Cu and C appears to have taken place, as observed in the HRTEM image of Fig. 1b. This finding agrees with simulations reported by Dorfman et al [4]. Note that although the sample was analyzed at 400 kV, it did not suffer detectable structural changes and offered good stability during TEM operations. EELS line scan across the interface confirmed the presence of inter-diffusion region containing C, Cu and O, as shown in Fig 2. A 10-nm layer of Cu2O between the MWCNT and Cu matrix suggests an oxidation process suffered by the thin foil at room temperature over a period of about 30 months. In addition, such long aging at RT has allowed Cu2O precipitation in the copper matrix (Fig. 3). References [1] K. T. Kim, et al. Materials Science and Engineering A 430 ( 2006) 27-33. [2] K.Tae, et al. Materials Science and Engineering A 449-451 (2007) 46-50. [3] M.E. Mendoza, I.G. Solorzano, E.A. Brocchi. Materials Science and Engineering A 544(2012)21-26. [4] S. Dorfman et al. Materials Science and Engineering C, (2001) 191-193. [5] The authors acknowledge NSF (USA), AFOSR (USA) and CNPq (Brazil) for financial support. Microsc. Microanal. 21 (Suppl 3), 2015 978 a) b) Copper Cu + C C Fig. 1a) BF-STEM image of Cu-0.5 wt% MWCNT showing significant C segregation at Cu grain boundaries. Fig 1 b) HRTEM image at the Cu/ transformed MWCNT interface. a) b) Fig. 2a. Cu-5 wt% MWCNT: relative composition of C, Cu and O at the interface region measured by EELS. b) ADF STEM of the measured region. Fig. 3a. Bright-field TEM image of Cu matrix exhibiting extensive agglomeration and Cu2O precipitates. b) HRTEM of Cu2O precipitate.");sQ1[490]=new Array("../7337/0979.pdf","Graphene Related Nanostructures Synthesized by High-Energy Ball Milling","","979 doi:10.1017/S1431927615005693 Paper No. 0490 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Graphene Related Nanostructures Synthesized by High-Energy Ball Milling I. Estrada-Guel1,2, O. Anderson-Okonkwo1, F.C. Robles-Hernandez1 1. Department of Mechanical Engineering Technology, University of Houston, Houston, TX 772044020, USA 2. Centro de Investigaci�n en Materiales Avanzados (CIMAV). Laboratorio Nacional de Nanotecnolog�a Miguel de Cervantes No. 120, 31109, Chihuahua, Chih., M�xico The size and structure of the nanoparticulate materials imparts unique characteristics [1] perceptible at macro scale, leading ours to a new universe of possible areas of application. A good example is carbon allotropes as fullerenes, graphenes, etc. These structures have been the focus of research looking for new synthesis routes and potential applications. One important problem with its synthesis is the generation of soot in form of amorphous carbon. Soot is considered a byproduct that in some cases requires tedious, expensive and wasteful procedures for removing. On the other hand, soot can be visualized as the precursor of fullerenes or technically defined as the black solid product of incomplete combustion or pyrolysis of fossil fuels and other organic materials. This material plays important roles as industrial filler and pigment (carbon black) and a traffic-related air pollutant (diesel soot) [2]. It is primarily composed of carbon (>80%) in form of agglomerated primary particles with diameters on the order of 10�30 nm and crystalline and amorphous domains. In spite of the wide interest in fullerenes, very little is known about structures found in soots, the relation between them or their stability. It is known that some structural changes of soot particles may also be induced by laser irradiation [3]. On the other hand, carbon nanostructures (e.g. graphenes, nanotubes, etc.) are synthesized by evaporation methods. Here in this research we present alternative methods to synthesize this nanostructures using solid state methods such as high-energy ball milling that are cost effective and environmentally safe. For carbon, the crystallinity changes with ball milling time the general conclusion is that milling induces nanocrystalline-amorphous phase transitions [4]. However, there exist few reports upon the effect of milling over amorphous and carbon graphitic structures [5]. In this work, some evidence of microstructural changes of soots after high-energy ball milling is presented. The raw material is commercially available fullerene soot in form of a byproduct from fullerene synthesis. Soots were milled using a SPEX 8000M for intervals from 0 to 10h. After milling, samples were characterized by Raman spectrometry in a Horiba-XploRA device and morphological studies were performed via SEM in a JSM-7201F and TEM with a JEM2200-FS microscope. The Fig.1 shows two SEM-TEM micrographs of original soot (0h) and as-milled samples. Raw material (Fig. 1a) presents wooly particles characteristic of evaporation residues. These particles form agglomerates and clusters. The TEM observations show that the soot particles observed by SEM are agglomerates with limited crystallinity. The particles observed in the soot are in the range of 20-50 nm. After milling (Fig. 1b), the morphology of the particles changes abruptly in form of laminar aggregates with domains lower than 20 nm. Raman studies show that as the sample is milled, the D and G band begins to develop indicating the fact that mechanical milling is capable of inducing crystallization of carbon soot into complex structures such as graphitic carbon and graphene. These in-situ changes are induced by the high energetic energy released by the impacts of milling media. The high intensity in the 2D band in milled samples indicates that graphitic carbon was fractured leading to the development of graphene layers. The TEM micrograph Microsc. Microanal. 21 (Suppl 3), 2015 980 shown in Fig. 3 corroborates the presence of stacked graphene sheets, showing an interplanar spacing of 0.335 nm. References: [1] B. Munkhbayar et al. P. Tech 234 (2013) p. 132-140. [2] A. Sadezky et al. Carbon 43 (2005) p. 1731-1742. [3] L. Hu et al. Carbon 44 (2006) p. 1725-1729. [4] C.P. Marshall, M.A. Wilson. Carbon 42 (2004) p. 2179-2186. [5] A.D. Lueking et al. Carbon 45 (2007) p. 2297-2306. [6] I. Estrada-Guel thanks the CONACYT support under project 169262 and the Redes Tem�ticas de Nanociencias y Nanotecnolog�a (124886). a) b) Figure 1. SEM-TEM micrographs of a) Raw soot (0h) and b) sample after 4h of milling. 0.335 nm Figure 2. Raman spectra of soot samples as a function of milling time and HRTEM image of 4h milled sample.");sQ1[491]=new Array("../7337/0981.pdf","Electron Microscopy and Photocatalytic Studies of 1D TiO2 Nanostructures Synthesized by Two Different Routes","","981 doi:10.1017/S143192761500570X Paper No. 0491 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electron Microscopy and Photocatalytic Studies of 1D TiO2 Nanostructures Synthesized by Two Different Routes Dwight Acosta1,Julieta Cabrera2, Hugo Alarc�n2, Alcides L�pez2, Juan Rodr�guez 2 Roberto Candal3 1. 2. Instituto de F�sica, UNAM , M�xico Universidad Nacional de Ingenier�a, Lima, Per� 3. Universidad de Buenos Aires, Argentina Nanotubes/nanorods (1D) TiO2 nanostructures of around 8 nm in diameter were synthesized by alkaline hydrothermal treatment of sol-gel made TiO2 or P-25 TiO2 . Anatase like 1D TiO2 nanostructures were obtained in both cases. The 1D nanostructures made using seeds from Sol Gel TiO2 nanopowders turn on rodlike nanostructures and presents lower surface area than the nanostructures made from commercial TiO2 P-25 (97 y 279 m2/g, respectively). In both cases, the 1D structures shown lower photocatalytic activity than P25 nanopartcles. However, the rodlike nanostructures obtained from TiO2 Sol Gel seeds displayed slightly higher efficiency than the original seeds. Despite the higher surface area shown by the nanostructures, the photocatalytic efficiency did not improve with respect to their precursor seeds. This phenomenon can be associated with the presence of other of structures like particles, nanoribbons or a kind of sheets with lower crystallinity and even amorphous phases. TiO2 nanomaterials are well-studied and commonly used materials for liquid and gas-phase photocatalytic applications due to its high performance photocatalysis for water splitting and for degradation of organics [1]. In the last years one-dimensional nanostructures such as nanotubes, nanorods, nanowires, nanobelts, etc. of inorganic materials have attracted great attention because they could offer larger surface area in comparison to nanoparticles [2]. In this work, we report de synthesis of nanotube and rodlike shape TiO2 nanostructures by hydrothermal synthesis using seeds of: TiO2 nanopowders synthesized by So Gel method in our laboratory and commercial TiO2 P-25. We have studying the effect of hydrothermal treatment time and the influence of the starting materials in the morphology, thermal stability and their photocatalytic activity. The obtained nanotubes and nanorods were compared with their corresponding starting materials to evaluate their photocatalitic performance under the degradation of an organic pollutant as Rhodamine B (RhB). Photocatalytic efficiency for degradation of RhB was carried out under the radiation of Ultravitalux 220 W OSRAM ultravitalux lamp, with a measured mean radiation Intensity of 60 W/m2 in the UV-A range. An aqueous solution with initial volume of 150 mL was prepared with an amount of 0.050 g catalyst and RhB 10 ppm, the solution was stirred first in the dark for 30 min to ensure that the RhB was adsorbed to satura-tion on the catalysts. The decrease of the RhB concentration was determinate as a function of the irradiation time from the change in absorbance at 564 nm. Microsc. Microanal. 21 (Suppl 3), 2015 982 978 FE-SEM images (left) and TEM and HRTEM images (right) of 1D nanostructures obtained from P25 TiO2 hydrothermally treated for 18, 24 and 40 h with further acid treatment and annealed at 400�C (a, b and c) TEM images of 1 d TiO2 nanoestructures obtained from Sol Gel NP hydrothermally treated for 18 h (a), after acid treatment (b) and after annealing at 400 �C (c). A comparison of the degradation of RhB solutions of P25 nanoparticles and the corresponding 1D nanostructures (left) and Sol Gel nanoparticles and the corresponding 1D nanostructures (right) can be done from these graphics. References [1] Jih-Jen Wu, Chi-Chung Yu J. Phys. Chem. B, 108 11 (2004) 3377 [2] Yas Yamin, Ni. Keller, Val�erie Keller J. of Photoch. and Photobiology A: Chemistry (2010)");sQ1[492]=new Array("../7337/0983.pdf","Strain Relaxation in InAs Quantum Dots and its Suppression by Indium Flushing","","983 doi:10.1017/S1431927615005711 Paper No. 0492 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Strain Relaxation in InAs Quantum Dots and its Suppression by Indium Flushing H. Xie1,2, F. A. Ponce2, R. Jakomin3,4, M. P. Pires3,5, R. Prioli3,6 and P. L. Souza3,6. 1. 2. School for Engineering of Matter, Transport, and Energy, Arizona State University, Tempe, USA. Department of Physics, Arizona State University, Tempe, USA 3. Instituto Nacional de Ci�ncia e Tecnologia de Nanodispositivos Semicondutores, PUC-Rio, Brazil 4. Campus de Xerem, UFRJ, Duque de Caxias, Brazil 5. Instituto de F�sica, UFRJ, Rio de Janeiro, Brazil 6. Pontificia Universidade Cat�lica do Rio de Janeiro, Rio de Janeiro, Brazil InAs quantum dot (QD) system has been explored for intermediate band solar cells (IBSCs) to overcome the Shockley-Queisser limit of one-bandgap solar cells. The intermediate bandgaps introduced by the QDs provide additional transition paths for light absorption which ideally increases the current in IBSCs while preserving the output voltage [1]. AlGaAs is used as matrices to approach the optimum bandgaps for high photovoltaic efficiency [2]. Lattice mismatch, which is necessary for QDs formation by Stranski-Krastanov growth mode [3], can result in plastic relaxation and degrade the photocurrent by the presence of dislocations. The plastic relaxation can be prevented by increasing the temperature for a brief period after the InAs QDs are partly covered by a capping layer. The effect of this so called indium flushing [4] technique on removal of plastic strain relaxation is studied here. The InAs/AlGaAs QD structures were grown by metalorganic chemical vapor deposition (MOCVD). InAs dots formed as a result of Stranski-Krastanov growth. We capped the QDs by a 10 nm GaAs capping layer in sample A and a 5 nm GaAs capping layer in sample B. The temperature was raised after deposition of the capping layer. The structure consisted of ten layers of InAs dots separated by 70-nmthick Al0.3Ga0.7As barriers. Two-beam diffraction contrast images and high resolution transmission electron microscopy (HRTEM) images of the QDs in each sample were recorded in a Philips CM200FEG electron microscope. The morphology of the InAs dots was studied from high-angle annular dark field (HAADF) images which were taken in a JEOL 2010 F electron microscope. Two-beam diffraction contrast images of sample A and sample B under g = 220 are shown in Figure 1a and 1b. Moir� fringes in the QDs of sample A (Fig. 1a) show that the lattice parameters of the dots and the matrix are different suggesting strain relaxation. In sample B (Fig. 1b), bend-contour contrast is observed suggesting the dot is strained. HREM of a QD in sample A (Fig. 1c) shows a 60 degree dislocation (left) and a Lomer dislocation (right) at the hetero-interface confirming strain relaxation in sample A. The HAADF images (Fig. 2) show that the QDs in sample A can be as thick as the 10 nm GaAs capping layer while the QDs in sample B are limited to 5 nm in height. The critical height of QDs to form dislocation loops is calculated based on the balance of the lattice misfit force and the dislocation line tension. Strain relaxation is suppressed by limiting the height of QDs below the critical height using thin capping layers in indium flushing [5]. References: [1] A. Luque and A. Mart�, Phys. Rev. Lett. 78 (1997) p. 5014. [2] R. Jakomin et al, J. Appl. Phys. 116 (2014) p. 093511. [3] D. Leonard et al, Phys. Rev. B 50 (1994) p. 11687. [4] Z. Wasilewski et al, Cryst. Growth 202 (1999) p. 1131. Microsc. Microanal. 21 (Suppl 3), 2015 984 [5] We gratefully acknowledge the use of facilities within the LeRoy Eyring Center for Solid State Science at Arizona State University a) c) b) Figure 1. Two-beam diffraction contrast images of QDs in sample A (a) and sample B (b) and HREM images of a QD in sample A (c). Figure 2. HAADF images of QDs in sample A (left) and sample B (right). a) b) Figure 3. (a) The simplified model of a QDs with a dislocation loop expending across the QD. (b) The critical height of QDs to form dislocation loops vs the misfit strain between dots and matrices. Experimental data are also shown in the graph.");sQ1[493]=new Array("../7337/0985.pdf","Peculiar Plasmon Peak Position in Electron Energy Loss Spectrum of Hexagonal Boron Nitride/Graphene Double Layer","","985 doi:10.1017/S1431927615005723 Paper No. 0493 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Peculiar Plasmon Peak Position in Electron Energy Loss Spectrum of Hexagonal Boron Nitride/Graphene Double Layer Nicholas Cross1, Lei Liu2, Ali Mohsin2 Gong Gu2, Gerd Duscher1,3 1. Department of Materials Science and Engineering, the University of Tennessee, Knoxville, TN 37996, United States 2. Department of Electrical Engineering and Computer Science, the University of Tennessee, Knoxville, TN 37996, United States 3. Materials Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831, United States Van der Waals (vdW) heterostructures of two-dimensional materials have attracted considerable research interest [1]. We report the observation of a large blue shift in the + plasmon peak of the electron energy loss spectrum of a hexagonal boron nitride (h-BN)/graphene vdW heterostructure with regard to those of single-layer graphene, single-layer h-BN, and bilayer h-BN. Samples for this study have been grown by an atmospheric-pressure chemical vapor deposition process modified from previous work [2]. First, graphene is grown on a copper foil substrate, followed by a hydrogen etch step that result in fresh zigzag-oriented edges of graphene islands and holes in the islands. Ammonia borane (NH3BH3) is then used as the precursor to grow h-BN. In contrast to previous work, where the precursor charge was controlled to ensure the synthesis of strictly in-plane heterojunctions of graphene and h-BN, the precursor charge is significantly increased to result in the formation of bilayer hBN and h-BN/graphene double layer regions, along with areas of single-layer graphene and single-layer h-BN on the same sample. Atomic-resolution imaging and electron energy loss spectroscopy (EELS) of the samples is accomplished using a state-of-the-art fifth order aberration-corrected scanning transmission electron microscope (STEM) Nion UltraSTEM100, with the capability to obtain high-angle annular dark field (HAADF) images that can distinguish intensity differences between low-Z elements (e.g. N, B, and C). Images are acquired with an accelerating voltage of 60 kV, and EELS performed with an energy dispersion of 0.3 eV/channel and an energy resolution of 0.6 eV. Figures 1A, B, C, and D show typical HAADF images of single-layer graphene, single-layer h-BN, bilayer h-BN, and an h-BN/graphene double layer, along with their respective electron energy loss spectra. The spectra are fit with both Lorenztian and Gaussian distributions to obtain a noise-free representation of the spectra and give confident peak positions and widths. Table 1 shows that the plasmon peaks of the hBN/graphene double layer are within 0.1 eV from those of graphene, single-layer h-BN, and bilayer hBN. While the + plasmon peaks of graphene, single- and bilayer h-BN are consistent with those reported in the literature [3,4], the + plasmon peak of the h-BN/graphene double layer, a vdW heterostructure, is appreciably blue shifted with regard to each of graphene, single- and bilayer h-BN. This suggests that the in-plane bonds of the two two-dimensional sheets interact, resulting in alteration of the electronic structure. Microsc. Microanal. 21 (Suppl 3), 2015 986 Figure 1. HAADF images (2 nm 2 nm) and EELS of (A) graphene, (B) single-layer h-BN, (C) doublelayer h-BN (C), and (D) h-BN/graphene double-layer. EELS plots graph the experimental data (red lines/fluctuating spectrum), model data (blue lines/ smooth spectrum) (summation of all of the Lorenztian and Gaussian peaks used to fit experimental data) and the noise (orange line/fluctuates around the x-axis) (difference between experimental and model data). The dashed line is an indicator to aid in recognizing peak shift. Table 1. Peak positions of and + surface plasmons for all four types of regions in Fig. 1. References: [1] A. Geim, and I. Grigorieva, Nature 499 (2013), p. 419. [2] L. Liu, J. Park and D. Siegel, Science 343 (2014), p. 163. [3] T. Eberlein, U. Bangert, and R. Nair, Physical Review B 77 (2008), p. 233406. [4] R. Arenal, O. St�phan and M. Kociak, Physcial Review Letters 95 (2005), p. 127601. [5] We acknowledge financial support by NSF (DMR-1410940). The atomic resolution characterization of this research was conducted at the Center for Nanophase Materials Sciences (CNMS2014-339), which is a DOE Office of Science User Facility.");sQ1[494]=new Array("../7337/0987.pdf","Nanostructured Ruthenium Disulfide Catalyst High Active in the HDS of DBT","","987 doi:10.1017/S1431927615005735 Paper No. 0494 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Nanostructured Ruthenium Disulfide Catalyst High Active in the HDS of DBT C. Ornelas1*, C. Leyva-Porras2, D. Carrillo3, A. Aguilar-Elgu�zabal1 and L. Alvarez-Contreras1 1 Centro de Investigaci�n en Materiales Avanzados (CIMAV), Laboratorio Nacional de Nanotecnolog�a, Miguel de Cervantes No.120, C.P. 31109, Chihuahua, Chih., M�xico 2 Centro de Investigaci�n en Materiales Avanzados, S.C. (CIMAV); Alianza Norte No. 202, Parque de Investigaci�n e Innovaci�n Tecnol�gica (PIIT), C. Aeropuerto km. 10 Apodaca, N.L. M�xico. C.P. 66600 3 Institute of Engineering and Technology, Autonomous University of Juarez, UACJ, Ave. del Charro #610 norte, C.P. 32320, Cd. Ju�rez, Chihuahua, M�xico Recently our research group achieves to develop a way to synthesized very high catalytic activity RuS2 catalysts, for the hidrodesulfuration (HDS) of dibenzothiofene (DBT) process [1, 2], in agreement to predicted in the past years by different groups [3, 4, 5, 6]. Low sulfurations levels was related with the low catalytic activity in the ruthenium sulfide catalyst due to remaining of metallic ruthenium in the catalyst, this poor sulfuration was associated with hydrogen used during the activation process which is a high reducing agent [7, 8, 9], nevertheless our group achieved excellent catalytic activity with very high reduction atmosphere of 98% H2/H2S [10] this means that different characteristics are responsible of high activity in the catalyst for HDS reaction. Nanoestructured RuS2 catalyst was synthesized according to previous work [9, 10] and characterized by electron microscopy, this nanoestructured catalyst has especial interest because it�s high activity in the HDS of DBT process many times better than industrial catalysts. The catalyst was synthesized following the procedure used in previous work [9, 10] and was activated by 2 hr in a tubular furnace at 673K whit reduction atmosphere of H2S. The resulted catalysts were tested in the HDS of DBT reaction. The HDS studies were carried out in a Parr model 4560 high-pressure batch reactor. 0.25 gram of catalyst was placed in the reactor with a solution of 5 vol. % of DBT in decaline. The reactor was pressurized to 3.1 MPa with hydrogen and heated up until 623 K. After the working temperature was reached, sampling for chromatographic analysis was performed during the course of each run to determine conversion versus time dependence. Electron Microscopy HRTEM JEOL JEM2200FS+Cs and SEM JEOL JSM-7001F were used to characterize the catalyst, micrographs obtained by SEM (fig 1a) ) show a porous structure of ruthenium sulfide with some excess of sulfur (S/Ru ratio = 3.1), that means that the sulfidation process was good enough to make complete sulfidation of all the ruthenium atoms, diffraction patterns obtained by HRTEM show ruthenium sulfide as pyrite type crystalline structure (RuS2), the morphology of the obtained catalyst is small crystals agglomerated and sinterized type sheets made of nanocrystals between 2 and 10 nanometers principally less than 5 nanometers with very high porosity between the particles and other sheets, the figure 1 b) to d) show clearly this morphology. The use of pure H2S gas flow in the sulfidation process allows to synthesized nanostructured RuS2 catalysts with a lot high active sites for the HDS of DBT reaction responsible of the good catalytic activity. Microsc. Microanal. 21 (Suppl 3), 2015 988 References: [1] Ornelas Guti�rrez Carlos El�as, et al. Mexican patent MX/a/2011/013529 [2] Ornelas Guti�rrez Carlos El�as, et al. US Patent 13/444,411. [3] Pecoraro T. A. and Chianelli R. R., Journal of Catalysis, 67 Issue 2 (1981), 430-445. [4] Raje A. P, et al, Applied Catalysis A: General 150 (1997), 297-318. [5] Liaw S-J., et al, Applied Catalysis A: General 151 (1997), 423-435. [6] Chianelli R. R., Berhault G., Torres B., Catalysis Today, 147, (2009) 275�286. [7] Navarro R., et al, 1999. Fuel Processing Technology, 61, pp. 73�88. [8] Castillo-Villalon P., et al, 2008. Journal of Catalysis, 260, pp. 65�74. [9] Blanchard J., et al, 2009. Catalysis Today, 147, pp. 255�259 [10] Ornelas C., et al, 2014. Microsc. Microanal. 20 (Suppl 3), 2014 pp. 1980-1981 Presenting author's email: e-mail address * carlos.ornelas@cimav.edu.mx b) a) c) d) Figure 1. RuS2 Catalysts, (a))SE SEM Image, (b)-d))Z contrast micrographs by HRTEM of catalyst.");sQ1[495]=new Array("../7337/0989.pdf","Template Synthesis of Hollow Carbon Nanofibers","","989 doi:10.1017/S1431927615005747 Paper No. 0495 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Template Synthesis of Hollow Carbon Nanofibers Yian Song1, Jiake Wei1,2 and Jingyue Liu1 1. 2. Department of Physics, Arizona State University, Tempe, AZ, 85287, USA The Institute of Physics, Chinese Academy of Sciences, Beijing, 100190, China Hollow nanostructures, especially hollow carbon nanostructures with adjustable inner space and large surface area, have attracted increasing interest because of their potential applications in catalysis, energy storage, and drug delivery [1]. To better control the morphology of hollow nanostructures, template based synthesis approach is of interest [1]. ZnO nanostructures can serve as appropriate templates since various morphologies can be synthesized [2]. Moreover, the ZnO template can be removed by using reductive gas treatment at moderate temperatures. Here we report the synthesis of hollow carbon nanofibers (HCNFs) from ZnO nanowires (NWs) by using ethanol as the carbon source. The key synthesis steps are schematically illustrated in Figure 1. First, the ZnO NWs were produced by a modified physical vapor deposition process. Then the as-synthesized ZnO NWs reacted with an ethanol vapor at a temperature between 500 �C to 700 �C to form a layer of carbonaceous material conformably covering the ZnO NWs. The thickness of the deposited layer can be adjusted by controlling the deposition time and temperature. The coated ZnO NWs were then reduced by 5% H2 (with N2 or Ar as the carrier gas) at 850 �C for 3 hours to completely remove the ZnO. The final high temperature treatment further graphitized the HCNFs and increased their total surface area. Figure 2 shows SEM images of the as-prepared ZnO NWs (a) and the final HCNFs after the high temperature treatment (b). Figure 2c shows an annular dark-field image of the HCFs confirming that the produced carbon fibers are hollow. By analyzing many similar images of the different regions of the sample it was estimated that the average wall thickness of the HCFs is approximately 4-5 nm. Figure 2d, a high resolution bright-field STEM image, clearly reveals short segments of wavy lattice fringes, similar to those of graphite. The total surface area of the synthesized HCFs was measured to be approximately 1970 m2/g, far exceeding the original total surface area (~ 12-18 m2/g) of the ZnO NWs. Electron energy loss spectroscopy (EELS) technique was used to probe the electronic structure of the synthesized HCFs. Figure 2e shows the EELS spectra of the carbon K-edge acquired from a HCNF after high temperature treatment, amorphous carbon support of the TEM grid, and graphite (taken from reference [3]). The first edge of the carbon K-edge spectra corresponds to the 1s to * transition and the 2nd edge corresponds to 1s to * transition. The relative peak intensities of these two peaks can be used to quantitatively measure the sp3/sp2 ratio of the synthesized HCNFs. Figure 2e suggests that after the high temperature treatment the HCNFs were highly graphitized with an electronic structure close to that of crystalline graphite. Spatial resolved EELS analysis of the HCNFs after different reduction or chemical treatments can provide useful information on the evolution of the electronic properties of the HCNFs. The approach reported here can be used to produce various types of carbon hollow nanostructures for practical applications, for example, lithium based battery or carbon based nanocatalysts. References: [1] X Lou, et al, Advanced Materials 20 (2008), p. 3987. [2] Z Wang, Journal of Physics: Condensed Matter 16 (2004), p. R829. Microsc. Microanal. 21 (Suppl 3), 2015 990 [3] K. Andre Mkhoyan, et al, Nano Letters, 9 (2009), p.1058 [4] The authors acknowledge funding by the College of the Liberal Arts and Science of Arizona State University and the use of facilities in the John M. Cowley Center for High Resolution Electron Microscopy at Arizona State University. The help from Mr. Daniel Mieritz and Professor Don Soe for suface area measurement is appreciated. Figure 1. Illustration of the synthesis process for producing carbon hollow nanofiber. Figure 2. SEM images of (a) ZnO NWs and (b) HCNFs after high temperature treatment; (c) ADF image of the HCFs showing the thickness of the HCNF walls; (d) bright-field STEM image of the wall of the HCF showing the segmented graphite-like fringes; (e) carbon K-edge EELS spectrum of a HCF together with amorphous carbon and graphite. The spectra in (e) were normalized by the intensity of the carbon * peak, and both the HCF and the reference graphite peaks were shifted upward for clarity.");sQ1[496]=new Array("../7337/0991.pdf","Atomic Resolution Study of W-Doped VO2 Nanowires","","991 doi:10.1017/S1431927615005759 Paper No. 0496 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic Resolution Study of W-Doped VO2 Nanowires Hasti Asayesh-Ardakani1, 2, Anmin Nie1, 2, Peter M. Marley3, Yihan Zhu4, Patrick J. Phillips2, Sujay Singh5, Farzad Mashayek6, Ganapathy Sambandamurthy5, Ke-bin Low7, Robert F. Klie2, Sarbajit Banerjee3, Gregory M. Odegard1 and Reza Shahbazian-Yassar1, 2,6 1 Department of Mechanical Engineering-Engineering Mechanics, Michigan Technological University, Houghton, MI 49933-1295, USA 2 Department of Physics, University of Illinois at Chicago, Chicago, IL60607-7059, USA 3 Department of Chemistry, University at Buffalo, State University of New York, Buffalo, New York 14260-3000, USA 4 Advanced Membranes and Porous Materials Center, King Abdullah University of Science & Technology, Thuwal, 23955-6900, Kingdom of Saudi Arabia 5 Department of Physics, University at Buffalo, State University of New York, Buffalo, New York 14260-3000, USA 6 Department of Mechanical and Industrial Engineering, University of Illinois at Chicago, Chicago, IL60607-7059, USA 7 Research Resource Center, University of Illinois at Chicago, IL60607-7059, USA Metal-Insulator Transition (MIT) in VO2 has attracted attention of many theorists and experimentalists for more than fifty years since the discovery of the phenomena by Morin [1]. The distinctive aspects of this phenomena are structural phase transition, sharp resistivity and optical transparency changes by several order of magnitudes at ~ 340 K [2]. These distinctive properties have inspired many applications such as thermo/electrochromics, Mott transistors, memristors, thermal actuators, gas sensors, strain sensors and temperature sensors. Recent efforts focus on controlling of phase transition and domain structures in finite size VO2, which results in different material properties and play a critical role in device applications. In this work, we have focused on the effect of tungsten (W) dopant in MIT of individual singlecrystalline VO2 nanowires by use of aberration corrected scanning transition electron microscopy. The high-resolution Z-contrast imaging of individual single-crystalline WxV1-xO2 nanowires indicates W dopant atoms in the structure as shown in Figure 1a. The strain map analysis of high-resolution images reveals the effect of dopants in MIT of VO2 (Figure 1b-c). The dopants create anisotropic stress and structural distortions into the VO2 structure. This stress facilitates the phase transition form monoclinic structure to tetragonal structure. We also verified the experimental observation by Density Functional Theory (DFT) calculations. Microsc. Microanal. 21 (Suppl 3), 2015 992 References: [1] FJ Morin, Phys Rev Lett 3 (1959), p. 34. [2] V Eyert, Ann Phys (Berlin) 11 (2002), p. 650. Figure 1. (a) Atomic resolution HAADF image of WxV1-xO2 nanowires. Insets correspond to the FFT of (a) which indicates that (a) has been acquired along the [301] zone axis of the monoclinic structure. The spots with higher intesity implie the existance of W in each column as compared to other spots. (bc) strain maps relate along to and perpendicular to (113) lattice planes. --");sQ1[497]=new Array("../7337/0993.pdf","Influence of Nonlinear Intensity Attenuation in Bright-Field TEM Images on Tomographic Reconstructions of Micron-Scaled Materials","","993 doi:10.1017/S1431927615005760 Paper No. 0497 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Influence of Nonlinear Intensity Attenuation in Bright-Field TEM Images on Tomographic Reconstructions of Micron-Scaled Materials Jun Yamasaki1, Michihiro Mutoh2, Shigemasa Ohta3, Shuichi Yuasa3, Shigeo Arai2, Katsuhiro Sasaki4 and Nobuo Tanaka2 1. Research Center for Ultra-High Voltage Electron Microscopy, Osaka University, Ibaraki, Osaka, Japan. 2. EcoTopia Science Institute, Nagoya University, Nagoya, Japan. 3. JEOL Ltd., Akishima, Tokyo, Japan. 4. Department of Quantum Engineering, Nagoya University, Nagoya, Japan Nowadays three-dimensional (3D) analyses of nanometer-sized and sub-micron-sized objects have been widely achieved by tomography in transmission electron microscopes (TEM). One of the next methodological targets should be quantitative 3D reconstructions in which not only the shape but also the internal density is correctly reproduced. This is, however, generally hindered by the nonlinearity between projection thickness and image intensity. In the case of mass-thickness contrast in bright-field TEM (BF-TEM) images, the ideal exponential attenuation with increasing thickness is disturbed by multiple scatterings. The nonlinearity in the tilt series should induce an inaccurate density distribution in the 3D reconstruction. In the present study, the nonlinear attenuation effect has been analyzed using amorphous carbon microcoils (CMCs) [1] shown in Fig. 1(a). Their well-defined shapes and compositional homogeneity are quite useful to estimate the mass-thickness [2]. The intensity attenuation has been measured along the line in Fig. 1(b) taken by the high-voltage electron microscope (HVEM) [3]. The results measured at the acceleration voltages of 400, 600, 800 and 1000 kV are converted to the plots of the electron transmittance T with increasing thickness in Fig. 2. At a glance, all the data seem to obey the linear attenuation regardless of the differences in the electron energy. However, the least square line for the data at 400 kV has a considerable value of the negative intercept at zero thickness. It is considered that such nonlinear attenuation should induce failures in conversion from intensity to thickness and thus inhibits correct 3D reconstructions. The influence of the nonlinearity on tomographic reconstructions has been examined using a specially-developed 360�-tilt sample holder for elimination of the missing-wedge effect [2]. Figure 3 shows the reconstruction results from the tilt series taken at 400 kV and 1000 kV. Although the 3D shape of the CMC has been reconstructed well in both cases, the internal density is not uniform but has gradient from the center at 400 kV. Moreover, there is slight increase of the vacuum level at the inside region of the coil. The inaccurate density reconstruction should result from the nonlinearity pointed in Fig. 2. Judging from the plot for 600 kV electrons in Fig. 2, the linearity is valid at least down to lnT of -0.4, which corresponds to about 2/3 of the electron transmittance. This information should be beneficial in actual tomography experiments of carbon-based materials because one can foresee quality of the reconstruction from the minimum transmittance in a single BF-TEM image prior to the tilt series acquisition. References: [1] S. Motojima et al., Appl. Phys. Lett. 56 (1990), p. 321. Microsc. Microanal. 21 (Suppl 3), 2015 994 [2] J. Yamasaki et al., Microscopy, 63 (2014), p. 345. [3] N. Tanaka et al., Microscopy 62 (2013), p. 205. Figure 1. Carbon microcoils. (a) SEM image (taken by Microphase Co., Ltd.) and (b) BF-TEM image taken by HVEM. Figure 2. Attenuation of electron transmittance T in BF-TEM images with increasing thickness. Figure 3. 3D reconstructions of the CMC from the tilt series of BF-TEM images taken at (a) 1000 kV and (b) 400 kV.");sQ1[498]=new Array("../7337/0995.pdf","Atomic-Scale Characterization of the Reduction of -Fe2O3 Nanowires","","995 doi:10.1017/S1431927615005772 Paper No. 0498 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic-Scale Characterization of the Reduction of -Fe2O3 Nanowires Wenhui Zhu1, Jonathan P Winterstein2, Renu Sharma2 and Guangwen Zhou1 1. Department of Mechanical Engineering & Multidisciplinary Program in Materials Science and Engineering, State University of New York, Binghamton, NY 13902, USA 2. Center for Nanoscale Science and Technology, National Institute of Standards and Technology, Gaithersburg, MD 20899, USA Reduction of metal oxides is a reaction of removing lattice oxygen and plays an role in producing active materials for a variety of applications ranging from catalysis to electronic devices [1]. The reduction of iron oxides has been investigated intensively because of its vital role in heterogeneous catalysis. The variable valence states of Fe cause a rather complicated Fe-O phase diagram, leading to the possibility of multiple phases being formed during redox reactions. It has, therefore, been a longstanding challenge in understanding the reduction mechanism of Fe2O3. Here, we present observations of an iron oxide superlattice structure resulting from oxygen vacancy ordering and stacking fault formation as well as the phase transformation pathway during the reduction of -Fe2O3 nanowires The -Fe2O3 nanowires were prepared by the thermal oxidation of pure iron [2]. The reduction was conducted in a vacuum chamber at T = 500 with the pressure of H2 gas (99.999% purity) at 270 Pa. The morphologies of the nanowires both before and after the H2 reduction were examined using a scanning electron microscope (SEM). The crystal structure and valence state of the nanowires were characterized by TEM imaging, diffraction, and electron energy-loss spectroscopy using an aberrationcorrected TEM operated at 300 kV. Fig. 1 shows SEM images of the nanowires before and after the reduction reaction. It can be seen that the as-prepared nanowires have a smooth surface morphology while the reduced nanowires show the formation of bulges on the surface. Fig. 2 (a) illustrates a bright-field TEM image of a single reduced nanowire with the presence of bulges on the surface. Fig. 2 (b) presents a HRTEM image at relative low magnification of the parent nanowire as marked with the red square in Fig. 2 (a).The upper-right inset in Fig. 2 (b) is an optical diffractogram of the HRTEM image, which shows the presence of superlattice diffraction spots in addition of the fundamental reflections associated with the -Fe2O3 structure of the parent nanowire. The lower-right inset in Fig. 2 (b) is simulated diffraction pattern using the -Fe2O3 structure with the removal of oxygen atoms (i.e., the ordering of oxygen vacancies) of every 10th {3 3 00} plane. Fig. 2 (c) illustrates the formation of another type of defects, i.e., stacking faults, with the shifting of atomic planes, forming the banded structure in the parent region of the reduced nanowire. There have been several reports suggesting that the formation of the modulated structure in -Fe2O3 is caused either by oxygen-vacancy ordering [3] or by stacking fault formation [4]. Here we find that both oxygen-vacancy ordering and stacking faults appear during the reduction process. In addition to the formation of the modulated structure by ordering of oxygen vacancies or stacking faults, phase transitions are also observed in the reduced nanowires. Fig. 3 (a) shows a TEM image of a reduced nanowire. The diffraction pattern (inset in Fig. 3(a)) shows that both the parent and the bulge area have a cubic structure, which can be indexed either as -Fe2O3 (space group: P4132 ; nearly cubic with a small tetragonal distortion) or Fe3O4 (space group: Fd 3 m ) due to their close lattice constants. Electron energy loss spectroscopy (EELS) was used to further distinguish the two structures. Fig. 3 (b) Microsc. Microanal. 21 (Suppl 3), 2015 996 shows the Fe L edge. The L3/L2 ratio of the parent nanowire is ~ 5.0, higher than that of the bulge area (~ 4.5), indicating their different oxidation states, i.e., the parent nanowire is -Fe2O3 while the bulge area is Fe3O4. Due to the similar lattice constants of the two phases, their interface is coherent and the Fe3O4 bulge forms epitaxially with the -Fe2O3 nanowire (Fig. 3 (c)), where the -Fe2O3 phase is formed by the conversion reaction of the initial -Fe2O3 nanowire. Fig. 3(d) is a simulated HRTEM image of the -Fe2O3/Fe3O4 interface, which matches well with the experimental one shown in Fig. 3(c). In conclusion, TEM analysis indicates the reduction process involves the formation of a superlattice caused by both oxygen-vacancy ordering and stacking faults and a topotactic reaction from -Fe2O3 to Fe3O4. References: [1] Luo, W., et al., Journal of Physics D: Applied Physics, 2007. 40(4): p. 1091. [2] Yuan, L., et al., Nanoscale, 2013. 5(16): p. 7581-7588. [3] Chen, Z., et al., Chemistry of Materials, 2008. 20(9): p. 3224-3228. [4] Cai, R., Chinese Physics B, 2013. 22(10): p. 107401. Figure 1. (a) SEM image of the as-prepared -Fe2O3 nanowires, with the inset showing the smooth surface and tapered tip of the nanowires [2], (b) SEM image of the reduced nanowires with the inset showing bulge formation on the parent nanowires. Figure 2. (a) a single -Fe2O3 nanowire after the H2 reduction, (b) HRTEM image showing the presence of superlattice due to oxygen-vacancy ordering , the insets are the experimental and simulated diffractograms, and (c) HRTEM showing the presence of stacking faults (image filtered to reduce noise). Figure 3. (a) Iron oxide Nanowires with bump along the parent nanowire, the inset shows the [110] zone-axis diffraction pattern from both the parent nanowire and bulge area , (b) EELS of the two areas with different L3/L2 ratios of Fe L edge, indicating that the parent nanowire is -Fe2O3 and the bump area is Fe3O4, (c) HRTEM image showing the -Fe2O3/Fe3O4 interface area, (d) The simulated HRTEM of -Fe2O3 and Fe3O4, with the inset of the structure model, showing a coherent interface.");sQ1[499]=new Array("../7337/0997.pdf","Atomic Structure of Amorphous 2D Carbon Structures as Revealed by Scanning Transmission Electron Microscopy","","997 doi:10.1017/S1431927615005784 Paper No. 0499 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic Structure of Amorphous 2D Carbon Structures as Revealed by Scanning Transmission Electron Microscopy Jani Kotakoski, Franz Eder, Giacomo Argentero, Stefan Hummel, David Lindner, and Jannik Meyer Faculty of Physics, University of Vienna, Boltzmanngasse 5, 1090 Wien, Austria The atomic structure of disordered materials is one of the remaining challenges in materials science due to the difficulties related to imaging non-repeating arrangements of atomic positions. In fact, since over 80 years, the popular concept of the atomic structure of amorphous materials has been for a large part based on the drawings of a random network by Zachariasen [1]. Only very recently, the first possibility to image the complete atomic structure of a disordered material appeared in the form of two-dimensional materials. So far, two different materials have been described: a truly 2D silica glass [2,3] and graphene amorphized utilizing the imaging electrons in a transmission electron microscopy (TEM) experiment at a voltage of 100kV [4,5]. While the silica structure offers us a glimpse to the atomic arrangements in a naturally forming amorphous structure, introducing the disorder step-by-step during atomic resolution imaging allows the study of all of the intermediate states between a perfect crystal and a completely amorphized material. We show [5] that the change from a crystal to a glass happens suddenly, and at a surprisingly early stage. Right after the transition, the disorder manifests as a vitreous network separating individual crystallites, similar to the modern version of the crystallite theory. However, upon increasing disorder, the vitreous areas grow on the expense of the crystallites and the structure turns into a random network (See Fig. 1 a-c). Thereby, our results show that both of these two models for amorphous structures can be correct, and can even describe the same material at different degrees of disorder. In addition to the already-reported materials, in this contribution, we show two additional disordered 2D carbon materials: graphene amorphized with ion irradiation (in contrast to electron irradiation) and evaporated thin carbon films with mainly sp2-hybridized bonding (Fig. 1 d). Further, we explore the structural differences in graphene membranes amorphized at different acceleration voltages during a TEM experiment. While the study of amorphous materials is itself interesting from the fundamental science perspective, the controlled introduction of disorder into graphene is also of technological relevance. For example, patterning of amorphous patches into graphene can, at least in principle, be utilized to create allgraphene devices for nanoelectronics and spintronics, and their use can be explored for applications in thermoelectrics. Additionally, because disorder enhances the chemical reactivity of graphene, selectively amorphized graphene structures may have use in the field of sensor applications [6]. References: [1] Zachariasen, W., Journ. Am. Chem. Soc 54 (1932), p. 3841. [2] Lichtenstein, L. et al, Angew. Chem. 51 (2012), p. 404. [3] Huang, P. Y. et al, Nano Lett. 12 (2012), p. 1081. [4] Kotakoski, J. et al, Phys. Rev. Lett. 106 (2011), p. 105505. Microsc. Microanal. 21 (Suppl 3), 2015 998 [5] Eder, F. et al, Sci. Rep. 4 (2014), p. 4060. [6] The authors acknowledge funding from the European Research Council (ERC) Project No. 336453PICOMAT, Austrian Science Fund (FWF) through Grant No. P25721-N20, M1481-N20, and I1283N20. Figure 1: a-c. Transition from a crystalline graphene (a) through a crystallite structure (b) into a random network (c) due to electron irradiation. d. Atomic structure of a sputtered 2D carbon membrane.");sQ1[500]=new Array("../7337/0999.pdf","Sample preparation artifacts in nuclear materials and mitigation strategies","","999 doi:10.1017/S1431927615005796 Paper No. 0500 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Sample preparation artifacts in nuclear materials and mitigation strategies A. Aitkaliyeva1, J. W. Madden1, B. D. Miller1, J. I. Cole1, J. Gan1 1 Idaho National Laboratory, Idaho Falls, Idaho, 83415 Diverse microstructures form in nuclear materials upon exposure to radiation. The defects produced during irradiation of materials can alter their mechanical properties and lead to embrittlement of reactor structural materials during service life. Therefore, it is imperative to know various radiation effects in reactor materials since it can aid in understanding in-reactor degradation behavior, accounting for irradiation effects in design, and producing new generation radiation-tolerant materials. Characterization of radiation-induced changes in reactor materials at the nano and atomic scales is typically conducted in transmission electron microscopes (TEM). Three most commonly used sample preparation techniques include electro-polishing, broadbeam ion milling, and focused ion beam (FIB) approach. However, preparation of samples using conventional sample preparation techniques, such as electro-polishing and ion milling, requires close-in, hands-on manipulation of the sample for extended periods of time. This is not feasible with highly radioactive nuclear materials. Electro-polishing involves the usage of electrolytic solution and doesn't introduce any defects that can be confused with irradiation-induced defects, and therefore it has traditionally been preferred sample preparation method for irradiated materials. Over the past several years, the popularity of electro-polishing began to diminish and FIB-based lift-out technique became preferable preparation techniques for irradiated materials. Figure 1 compares microstructure of unirradiated MA957 specimen prepared for TEM analysis using a) electro-polishing and b) FIBbased lift-out technique. Both techniques yield good quality specimens without any notable sample preparation artifacts in case of MA957. FIB has a major disadvantage when it comes to nuclear materials characterization � beam damage. Material removal in FIB is achieved via Ga ion milling, which can lead to formation of various defects in the target material such as vacancies, interstitials, defect clusters, intermetallic phases, and Ga precipitates. The extent of sample preparation-induced damage can interfere with specimen analysis and in case of nuclear materials preclude accurate defect microstructure characterization. Damage accumulated during FIB-based sample preparation can look very similar to the damage produced in nuclear materials irradiated to low displacements per atom (dpa) levels and/or lower temperatures. Minimization of FIB damage is typically achieved by using low-energy (5 keV and 2 keV) cleaning. However, in some materials even extensive 2 keV cleaning step is not sufficient in elimination of Ga damage. A number of unirradiated materials were examined and the results Microsc. Microanal. 21 (Suppl 3), 2015 1000 indicate that FIB damage is more extensive in some materials than the others. These formed damaged layers, consisting of both amorphous material and implanted Ga ions, should be eliminated in order to perform an in-depth examination of material microstructure after irradiation. The obtained results show that the energy of impinging particles should be reduced to below 2 keV for complete elimination of sample preparation artifacts. However, only the most advanced FIB systems are capable of reducing ion energy below 2 keV and these systems are not optimal for preparation of highly activated materials due to shorter working distances of 2 mm (as opposed to 10 mm in FEI Quanta microscopes). Therefore, post-processing of the FIB lamella has to be considered to minimize or eliminate this damage and ensure high quality analysis results. Post-FIB processing can be achieved using a variety of advanced ion milling systems, such as PIPS-2 and NanoMill, in which energy of the Ar ion beam can be reduced to <500 eV. Figure 2 illustrates advantages of adding post-processing to the FIB preparation procedure by providing comparison of Fe-12Cr specimen microstructure (a) after jet-polishing, (b) FIB-based lift-out, and (c) a combination of FIB-based lift-out technique with post-processing in PIPS-2. Figure 1. Bright-field TEM micrographs of unirradiated MA957 after: a) electro-polishing and b) preparation in FIB tool. Figure 2. Bright field TEM micrographs of unirradiated Fe-12Cr sample prepared using: a) FIB and b) FIB with post-processing in PIPS-2.");sQ1[501]=new Array("../7337/1001.pdf","Microstructural Characterization of Hydrogen Irradiated Austenitic Stainless Steel","","1001 doi:10.1017/S1431927615005802 Paper No. 0501 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microstructural Characterization of Hydrogen Irradiated Austenitic Stainless Steel Hyung-Ha Jin1, Eunsol Ko1, Junhyun Kwon1 1. Nuclear Material Safety Research Division, Korea Atomic Energy Research Institute, 989-111 Daedeok-daero, Yuseong-gu, Daejeon, 305-353, Republic of Korea Irradiation assisted stress corrosion cracking by neutron irradiation has been recognized as a significant concern for extended operation of commercial nuclear reactors [1, 2]. Recently, the retention of helium and hydrogen has been raised for a promising factor affecting the IASCC susceptibility of nuclear internals [3, 4]. In this work, we are attempting microstructural characterization of hydrogen irradiated austenitic stainless steels to get more information on effect of hydrogen on microstructural changes in austenitic stainless steel. A commercial austenitic stainless steel (SS316 type) was used for this work. An ion irradiation experiment was carried out with a multi-purpose ion implanter in the Korea Institute of Geoscience & Mineral Resources (KIGAM). In the ion irradiation, hydrogen ions (H2+) were used with various energy ranges from 50keV to 490keV for the development of uniform radiation damage and implanted ion concentration in the experimental sample. Since the ion-irradiated layer was calculated to be about 1 �m in depth by "Stopping Range of Ions and Matter (SRIM)" [5]. TEM lamellae were prepared by FIB milling and low energy argon ion milling. A transmission electron microscope (TEM) equipped with an energy dispersive spectrometer (EDS) system was used for the analyzing of radiation induced defects and radiation induced segregation (RIS) behavior at grain boundaries. According to a low magnification TEM image in Figure 1, radiation damage layer was formed up to a depth of around 1 �m. In the ion irradiated austenitic stainless steel, dislocation loops were observed in the matrix as shown in Figure 1(b). Most dislocation loops were identified to be Flank faulted loops with a burgers vector of 1/3<111>. Cavities by implanted hydrogen were also observed in the matrix as shown in Figure 1(c). The formation of unknown phase at the grain boundary was observed in the hydrogen irradiated sample. Figure 2(a) is a montage of TEM images taken near grain boundary region in the hydrogen irradiated region. HRTEM analysis in Figure 2(b) clearly indicates that the unknown phase has a body centered crystal (BCC) structure. The unknown phase is identified to be ferrite () or martensite () with the BCC structure. Since implanted hydrogen was expected to be built up to 100000 appm (~ 0.1-0.2 wt%) in the hydrogen irradiation sample, the high density of hydrogen seems to play a role in the formation of ferrite or martensite at grain boundary. The EDS result in Figure 2(c) shows that the BCC phase has low Cr and high Ni concentration compared with the matrix. It was found that strong RIS was developed at the interfaces between the transformed ferrite or martensite and austenitic () matrix as well as grain boundaries through the hydrogen irradiation. The strong RIS seems to lead to high Ni and low Cr concentration in the transformed ferrite or martensite. In conclusion, the implanted hydrogen during the irradiation is expected to enhance or transformation and to develop strong RIS near grain boundary. References: [1] E.P. Simonen, S.M. Bruemmer, JOM 50 (1998), p. 52. [2] E.A. Kenik, J.T. Busby, Mater.Sci.Eng. R. 73 (2012), p. 67. Microsc. Microanal. 21 (Suppl 3), 2015 1002 [3] K. Fujimoto et al, Proceedings of the 12th international Conference on Environmental Degradation of Materials in Nuclear Power system (2005), p.299. [4] D.J. Edwards et al, J. Nucl.Mater. 384 (2009), p. 249. [5] J.F. Ziegler, J.P. Biersack, U. Littmark, The Stopping and Range of Ions in Solids, Pergamon, New York, (1985). [6] This research was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (No. 2012M2A8A4025886). Figure 1. A Low magnified TEM image (a), TEM images showing Frank faulted loops (b) and cavities (c) after hydrogen ion irradiation. Figure 2. A Low magnified TEM image near grain boundary (a), HRTEM image and corresponding FFT images indicating the formation of BCC ( or ) in vicinity of grain boundary (b) and EDS result showing Ni enrichment of Cr depletion in the transformed BCC ( or ) phase.");sQ1[502]=new Array("../7337/1003.pdf","LAMDA: Irradiated-Materials Microscopy at Oak Ridge National Laboratory","","1003 doi:10.1017/S1431927615005814 Paper No. 0502 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 LAMDA: Irradiated-Materials Microscopy at Oak Ridge National Laboratory Chad M. Parish1, N. A. P. Kiran Kumar1, Lance L. Snead, Philip D. Edmondson1, Kevin G. Field1, Chinthaka Silva1, A. Marie Williams1, Kory Linton1 and Keith J. Leonard 1 1. Materials Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN USA Oak Ridge National Laboratory's Low Activation Materials Development and Analysis (LAMDA) laboratory is a dedicated facility containing specialized instruments for the study of irradiation-induced effects on materials properties. Located in the Materials Science and Technology Division, LAMDA consists of several interconnected contamination-zone and clean area suites. Originally created as a facility for plutonium studies and then thermophysical properties of graphite for high temperature gas cooled reactors in the 1960's, the role of LAMDA has changed with the emphasis placed on "low activation" materials. Currently, LAMDA is involved in both fundamental and applied research on radiation-induced changes in structural materials, reactor internals, diagnostic materials and sensor components for both current and advanced reactor designs for both fission and fusion systems. LAMDA allows for the examination of low activity radiological samples (< 100 mR/hr at 30 cm) without the need for remote manipulation. LAMDA typically utilizes small, compact samples to allow researchers to leverage cutting-edge characterization and test equipment to study materials phenomenon not possible at a hot cell facility. The LAMDA facility is maintained as a low alpha contamination facility, but certain equipment is available for fuel-related studies. LAMDA contains a strong electron microscopy subprogram, which we describe here. A full complement of equipment and instruments are available for metallographic and microstructural analysis of materials. This includes X-ray tomography and three irradiated-materials-dedicated FEI DualBeam FIB-SEM instruments for 3D materials studies and the preparation of specimens for TEM. LAMDA contains a new 200kV Schottky JEOL JEM-2100F S/TEM with EDS and Gatan Quantum Image Filter, and leverages ORNL's other TEM tools as well as SEMs and atom probe. The LAMDA lab works in conjunction with the Irradiated Materials Examination and Testing (IMET) and Irradiated Fuels Examination Laboratory (IFEL) hot-cell facilities at the laboratory to study materials irradiated at either the High Flux Isotope Reactor (HFIR) or other experimental and commercial reactors. The LAMDA, IMET, IFEL and HFIR facilities are all part of the National Science User Facility. LAMDA performs work supported by various Department of Energy (DOE) programs including the Fusion Materials, Light Water Reactor Sustainability, DOE-Naval prime contractors, as well as collaborations with other international research facilities, universities and companies. Figure 1 shows the IMET facility, where irradiated capsules are opened and the specimens sorted and subjected to first-stage preparation (e.g., inspection, cutting, grinding). Figure 2 shows one of the FEI DualBeam instruments, which is located in a lead-shielded room and equipped for remote operation, and can be used to prepare TEM specimens of irradiated fuels. Figure 3 shows the newly installed JEOL 2100F TEM/STEM instrument. Figure 4 shows Re-rich precipitates in HFIR-irradiated tungsten that was prepared in LAMDA through FIB processing, showing that materials with relatively high induced levels of activation can be handled with appropriate safety controls for the researchers. [1] [1] Supported by Office of Fusion Energy Sciences, US Department of Energy under contracts DEAC05-00OR22725 with UT-Battelle, LLC. CM200 microscopy research (Figure 4) conducted as part of a user proposal at ORNL's Center for Nanophase Materials Sciences, which is an Office of Science User Microsc. Microanal. 21 (Suppl 3), 2015 1004 Facility. Figure 1: Irradiated Materials Examination and Testing (IMET) hot cell facility. Figure 2: FEI Quanta DualBeam FIB-SEM system. A second operator control station, separated from the instrument by a lead wall, protects the operator from strongly gamma-emitting specimens. Figure 3: JEOL 2100F TEM/STEM instrument. Equipped with EDS and EELS+EFTEM, this tool is now available for irradiated materials and fuels work in LAMDA. Figure 4: Tungsten irradiated at ~750�C to 7�1025 n/m2, 2 dpa, ~6at%(Re+Os) transmutation. Radiation-induced precipitation (perhaps or phases) and voids are visible.");sQ1[503]=new Array("../7337/1005.pdf","Analysis of Radiation Damage in Magnesium Aluminate Spinel by Means of Cathodoluminescence.","","1005 doi:10.1017/S1431927615005826 Paper No. 0503 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Analysis of Radiation Damage in Magnesium Aluminate Spinel by Means of Cathodoluminescence. Iwona Jozwik1, Jacek Jagielski1,2, Grzegorz Gawlik1, Przemyslaw Jozwik1,2, Renata Ratajczak2, Gerard Panczer3, Nathalie Moncoffre4, Anna Wajler1 and Agata Sidorowicz1. Institute of Electronic Materials Technology, Wolczynska 133, 01-919 Warsaw, Poland. National Centre for Nuclear Research, Soltana 7, 05-400 Otwock/Swierk, Poland. 3. Institut Lumi�re Mati�re ILM, UMR5306 Universit� Lyon 1-CNRS, Universit� de Lyon, Villeurbanne, France 4. Institut de Physique Nucl�aire de Lyon IPNL, Universit� de Lyon, Universit� Lyon 1, CNRS/IN2P3, UMR 5822, Villeurbanne, France. 2. 1. Magnesium aluminate spinel (MgAl2O4) is one of the oxides envisaged to be used in manufacturing of inert matrix fuel. Although the material is recognized for its radiation resistance, most of the experiments were performed on single crystals mainly by the use of Rutherford Backscattering/channeling (RBS/C) method and not much is known about the effects of irradiation of this material in polycrystalline form. The very recent concept is to use luminescence techniques as an experimental method able to measure the level of disorder in polycrystals, as it may be applied to both single and polycrystalline solids. Damage accumulation was reproduced by using an ion irradiation method, which is the most convenient method of simulation of radiation defects created during the exposure of material to a radiative environment. MgAl2O4 single crystal samples (commercially available) and polycrystalline samples obtained by the hot-pressing (Astro, Thermal Technology) of magnesium aluminate nanopowder were irradiated with 320 keV Ar+ ions at fluencies ranging from 1x1012 to 2x1016 cm-2 in order to create various levels of radiation damage. Cathodoluminescence (CL) measurements were performed using an EMSystems setup mounted in Auriga CrossBeam Workstation (Carl Zeiss NTS). The electron beam energy was reduced to 10 kV in order to limit the analyzed volume of the material as much as possible to the damaged layer (~200 nm). As the result of measurements carried out, a set of cathodoluminescence spectra has been recorded. The raw signal for each sample with different fluencies of Ar+ ions was analyzed in terms of signal intensity of the selected band. The spectra recorded for polycrystalline samples (Figure 1a) exhibit two broad bands, positioned around 420 nm and 520 nm (consisting in fact of a few emission bands).The presence of the first one is assigned to transitions from the excited states to lower energy states of the Mn2+ ion usually present in the synthetic spinels [1]. The second band positioned around 520 nm is typical for the non-stoichiometric compositions of synthetic spinels, and appears when the vacancy concentration increases [1]. Introduction of the anion vacancies causes a repartitioning of the Mn2+, and the emission at 520 nm is characteristic of the rearrangement of the activator rather than the direct influence of the vacancies. The intensity of both bands rapidly decrease with the ion irradiation fluence. Similar spectra were observed in the case of monocrystalline samples. Additionally, for single crystals, the damage-build up as a function of accumulated ion fluence was established through RBS/C. The results of CL (reduced signal intensities of the selected bands) and RBS/C analysis for single- and polycrystalline samples were then processed using Multi-Step Damage Accumulation (MSDA) model [2-4], and are presented in Figure Microsc. Microanal. 21 (Suppl 3), 2015 1006 1b. That allowed for the determination of damage build-up kinetics, and finally cross-section for radiation damage build-up. The phenomena responsible for the rapid decrease of the luminescence of the material are different than those related to the creation of displaced atoms present in the crystal channels. The rate of changes resulting from the formation of non-luminescent recombination centers is very different than that resulting from the creation of single defect clusters or dislocations. Consequently CL technique appears as a complementary tool bringing new possibilities in the damage accumulation studies in single- and polycrystalline materials [5]. References: [1] RL Mohler and WB White, Mat. Res. Bull. 29 (10) (1994), p. 1109. [2] J Jagielski and L Thom�, Vacuum 81 (2007), p. 1352. [3] J Jagielski and L Thom�, Appl. Phys. A 97 (2009), p. 147. [4] S. Moll et al, J. Appl. Phys. 106 (2009), p. 073509. [5] This work was partially sponsored by the National Science Centre (Poland) under the contract number DEC-2011/03/D/ST8/04490 and by 09-133 French � Polish collaboration program. a) b) Figure 1. The CL spectra recorded on MgAl2O4 polycrystalline samples (a), and radiation damage accumulation kinetics based on CL measurements analyzed with MSDA model for single- and polycrystals of MgAl2O4 (b).");sQ1[504]=new Array("../7337/1007.pdf","Stability of Precipitates in Zirconium Alloys under Self-ion Irradiation","","1007 doi:10.1017/S1431927615005838 Paper No. 0504 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Stability of Precipitates in Zirconium Alloys under Self-ion Irradiation Xu Wang1,2, Mengqiao Zhao2, Feifei Zhang1,2, Lumin Wang2,1 Collge of Energy, Xiamen University, Xiamen, Fujian 361102, China Department of Nuclear Engineering and Radiological Science, University of Michigan, Ann Arbor, Michigan 48109, United States 2. 1. Zirconium alloys are widely used in nuclear power reactors as fuel cladding and core structure materials due to their low neutron capture cross section, good corrosion resistance and mechanical strength. To meet the requirement of high burnup of the nuclear fuel, new generation Nb-containing zirconium alloys, such as M5, E110, E635 and Zirlo, were developed. Many ion and neutron irradiation experiments have been conducted on zirconium alloys to evaluate their radiation resistance [1,2]. The stability of the precipitates in the zirconium alloys is one of the most important concerns because it is believed to cause an accelerated rate of uniform corrosion besides other negative effects. The material used in this study is zirconium based ally containing small amounts of Fe and Nb. The samples were irradiated to 4.55�1015 Zr++/cm2 (20 dpa at the damage peak) using 3 MeV Zr++ with a defocused beam in the Ion Beam Laboratory at Texas A&M University, using a 1.7 MV Tandetron accelerator. The irradiations were conducted at 320oC, 360oC and 400oC, respectively. Cross-section TEM samples were prepared by focused ion beam (FIB) lift-out method using a FEI Helios 650 NanolabDualbeam FIB. Scanning transmission electron microscopy (STEM) was conducted using a JEOL 2100 spherical aberration (Cs)-corrected high resolution STEM operated at 200keV. TEM analysis and characterization were conducted using a JEOL 3100R05 double Cs-Corrected TEM. All the electron microscopy and sample preparation were conducted at University of Michigan Electron Microbeam Analysis Laboratory (EMAL). Fig.1 shows the HAADF, BF STEM images and EDS mapping of as received zirconium alloy. Precipitates are shown as black dots in the BF-STEM image (Fig.1b), and as white dots in HAADF image due to the higher mass of the main components. EDS mapping indicates that Nb and Fe are enriched in precipitates. Hcp Zr (Cr,Fe)2 and fcc ZrFe2 laves phase are two types of precipitates which were commonly reported in zirconium alloys without Nb, such as Zr-4. In Zr-Nb-Fe ternary system, two Laves phases (Zr(Nb,Fe)2 and (Zr,Nb)Fe2), fcc (ZrNb)2Fe intermetallic and two bcc solid solutions (-Zr and -Nb) were reported [3]. HRTEM images and corresponding FFT images of precipitates in three different zone axes are shown in Fig.2. According to these images, the precipitates are identified as bcc -Zr phase with lattice parameter a=0.35nm. Fig. 3a is a HRTEM images of the boundary of precipitate and matrix in the sample irradiated to 15 dpa at 320oC. It is clear that the precipitate is disordered while the matrix remains crystal. The inset is the corresponding selected-area electron diffraction (SAED) pattern showing that the precipitate has become amorphous after ion irradiation. Fig.3b and 3c are HRTEM images of samples irradiated at 360 oC and 400 o C respectively. It is apparent that the crystal structure of precipitates survived the irradiation at 360 oC and 400 oC. To sum up, the amorphization of Nb enriched precipitates in the zirconium alloy were observed after self-ion irradiation to 20 dpa at temperature lower than 360 oC. References: [1] H.H. Shen, S.M. Peng, X.S. Zhou, Chin.Phys. B 23(3) (2014), p. 036102. [2] R.W. Gilbert, M. Griffiths, G.J.C. Carpenter, J. Nucl. Mater. 135 (1985), p. 265. [3] C. Ramos, C. Saragovi, M.S. Granovsky, J. Nucl. Mater. 366 (2007), p. 198. [4] This research is supported by China General Nuclear Power Group. Microsc. Microanal. 21 (Suppl 3), 2015 1008 (a) (c) Nb (b) Zr Fe Fig. 1. (a) HAADF STEM images and (b) BF STEM images of unirradiated zirconium alloy; (c) EDS mapping shows Nb and Fe are enriched in precipitates. (a) [ ] (b) [ ] (c) [ ] Fig. 2. HRTEM images and corresponding FFT images (inset) of precipitates in different zone axes, (a) [111]; (b) [011]; (c) [001] (a) Matrix (b) SPP (c) Matri x SPP Matrix SPP Fig. 3. HRTEM images from the boundary of precipitates and matrix of the samples irradiated up to 20 dpa at (a) 320oC, (b) 360 oC and (c) 400 oC.");sQ1[505]=new Array("../7337/1009.pdf","The Evolution of Precipitates in High Cu and High Ni RPV Welds under Longterm Thermal Ageing","","1009 doi:10.1017/S143192761500584X Paper No. 0505 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Evolution of Precipitates in High Cu and High Ni RPV Welds under Longterm Thermal Ageing J. J. H. Lim1, S. Lozano-Perez2, M. G. Burke1, P. D. Styman3, I. Maclaren4, K. Wilford5, C. R. M. Grovenor2 1. 2. Material Performance Centre, The University of Manchester, Manchester, UK Department of Materials, The University of Oxford, Oxford, UK 3. National Nuclear Laboratory, Didcot, UK 4. School of Physics and Astronomy, The University of Glasgow, Glasgow, UK 5. Rolls Royce, Derby, UK Under neutron irradiation, solute-enriched clusters form in low alloy steels and welds that are used for manufacturing reactor pressure vessels (RPV) in light water reactor (LWR) systems. This can result in a significant increase in the yield strength of the steel and an increase in the ductile-to-brittle transition temperature (DBTT) compared to the non-irradiated steel. To date, the behavior of RPV materials is evaluated through extensive surveillance irradiation tests performed throughout the lifetime of the nuclear power plant. Thus, it is important to be able to predict and also understand the mechanism(s) by which this embrittlement occurs. Over the past 35 years, there have been significant efforts to identify and characterize these Cu-Mn-Ni-Si-enriched clusters, as RPV degradation can seriously limit the operating lifetime of a NPP. The analysis of neutron-irradiated 0.3 wt.% Cu RPV welds using atom probe field-ion microscopy first demonstrated that Cu- Mn-NiSi solute clusters, also known as Cu-enriched clusters (CEC), form, due to the combined effects of irradiation and the supersaturation of Cu in Fe. All RPV steels and welds now have strictly controlled Cu limits to avoid this degradation. However, several studies have shown that solute clusters are formed during neutron irradiation. These clusters were significantly enriched with Mn, Ni and Si. Although the formation of a MnNi-phase in Fe is predicted by thermodynamics calculation [1,2], no evidence has been presented to date to demonstrate that a MnNi precipitate can form in low alloy steels during ageing at a temperature of 365�C. In this work, precipitates that form in a high Cu-Mn-Ni low alloy steel under long-term thermal ageing at a temperature ~60C-deg higher than a pressurized water reactor has been studied using advanced electron microscopy and atom probe tomography (APT) techniques. Cu precipitates were observed in the early stage of ageing, i.e. 1000 hours [3]. These CRPs undergo a martensitic transformation from BCC-9R-3R-FCC structure as reported in [4]. One such thermally-aged Cu precipitate with a 9R structure was imaged and is shown in the high resolution TEM image in Figure 1. After prolonged ageing at 365�C, a Mn-Ni phase nucleated at the interface of the Cu precipitate and bcc Fe matrix as a "shell" surrounding the Cu as the core, as shown in Figure 2. Our studies also show that further ageing leads to the change in the morphology of these "core-shell" precipitates, as shown in Figure 3. EDX and EELS spectrum images have also demonstrated that a thicker MnNirich shell had formed at Cu precipitates associated with dislocations or grain boundaries due to preferential channeling of solutes along dislocations and grain boundaries. Further results obtained from TEM and APT will be compared and discussed in the presentation. References: [1] Liu, CL, et al, Mat. Sci. & Eng. A 238(1) (1997), p. 202. [2] Xiong W, et al, (2014), MRS Comm. [3] Morley, A, Department of Materials. Oxford, University of Oxford. DPhil Thesis (2008). [4] Lim, JJH, Department of Materials. University of Oxford, UK. DPhil Thesis submitted (2014). Microsc. Microanal. 21 (Suppl 3), 2015 1010 Figure 1: High resolution transmission electron image taken along <111>Fe showing a twinned 9R structure Cu precipitate (8 nm) in WV steel weld after thermal ageing for 49,800 hours at 365 �C. The angle between the (009)9R basal and (114)9R twin planes is 62.5 � 1�. The 2-layer fringe spacing at the position of a hexagonal-type stacking faults is indicated. Similarly, the 4-layer fringe spacing at the position of a cubic-type stacking fault is also indicated. Figure 2: EDS spectrum images corresponding to a thermally-aged precipitate nucleated at the matrix of low alloy steel weld aged for 18,620 hours. Figure 3: EELS multivariate statistical analysis reconstructed spectrum images corresponding to a thermally-aged precipitate nucleated at the matrix of low alloy steel weld aged for 49,800 hours.");sQ1[506]=new Array("../7337/1011.pdf","Amorphization of Indium Phosphide Single Crystals by Swift Heavy Ion Bombardment","","1011 doi:10.1017/S1431927615005851 Paper No. 0506 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Amorphization of Indium Phosphide Single Crystals by Swift Heavy Ion Bombardment A.S. Khalil Mining and Metallurgy Department, Tabbin Institute for Metallurgical Studies (TIMS), Helwan, POB 109, Cairo, Egypt Irradiation of single crystalline compound semiconductors with energetic heavy ion beams produces disorder in the irradiated crystal lattices. Indium Phosphide (InP) is an important technological III-V compound semiconductor used in space solar arrays and a variety of electronic devices. In this investigation, electron transparent and bulk samples cut from InP (001) wafer were bombarded by 200 MeV Au+ ions (~ 1MeV/amu) at fluencies ranging from 5x1010 to 1x1014 ion/cm2 in order to study the amorphization process of the InP (complete atomic disorder of the original InP crystal lattice). As shown in figures 1(a) to 1(c) for thin foil samples investigated by TEM, the progression of amorphization is followed with increasing the fluence of irradiating Au+ ions. Complete amorphization occurs at fluence of > 5x1013 as confirmed by the inset showing the selected area electron diffraction pattern were the diffused rings and the absence of any diffraction spots are synonymous with amorphous state. This observation was further supplemented by Rutherford Backscattering Spectrometry (RBS/C technique) for the bulk InP samples which showed that at irradiation fluence of 1x1014 ion/cm2 the sample surface was completely amorphous. Generally, amorphization in swift heavy ion irradiated compound semiconductors like InP can best be described by a combination of both heterogeneous mechanism (direct impact amorphization were each ion creates an amorphous ion track) and homogenous mechanism (defect accumulation were each ion track consists of an agglomeration of point defects). Models have been developed wherein an impinging ion can produce both a combination and coexistence of both amorphous-ion tracks and point defects-ion tracks which, when these ion tracks accumulate and overlap, the crystal converts to amorphous state [1]. By invoking the Hecking model [2] for the accumulation of damage we were able to show that the amorphization process proceeds by the accumulation and overlap of individual ion tracks. An Indicator of the amorphization process are the values of relative disorder ()min defined according to the following formula: ( )min = Yirradiated - YUnirradiat ed , here Y is the yield from the RBS/C measurements. Yrandom - YUnirradiat ed The value ()min gives an estimate of the relative amount of disorder in the crystal and it reaches unity for a complete amorphous state. Plotting ()min versus the Au+ ion fluence as depicted in figure 2 we conclude that the progression of amorphization might favor the homogenous mechanism [3]. This was further confirmed by HRTEM observations on ion track cores which indicate that individual ion track cores are not amorphous as shown in figure 3. Microsc. Microanal. 21 (Suppl 3), 2015 1012 References: [1] W. J. Weber, Instruments and Methods B, 166/167 (2000), p. 98. [2] N. Hecking, K. Heidemann and E. Kaat, Nuclear Instruments and Methods B, 15 (1986), p. 760. [3] J. Nord, K. Nordlund and J. Keinonen, Physical Review B 65 (2002), article number 165329. Figure 1. TEM micrographs for three different irradiating fluencies. In (a) for 5x1010ion/cm2, the ion tracks are well separated and in (b) overlap leads to amorphous pockets at 1x1013 ion/cm2. These will grow with increasing ion fluencies till complete amorphization occurs at fluence of 1x1014 ion/cm2 shown in (c). Figure 2. The plot of ()min versus ion fluencies, the best fit can be described by the composite Hecking model which gives more weight to homogenous amorphization. Figure 3. HRTEM micrograph of an ion track core, the continuation of lattice fringes implies that the ion track is not amorphous.");sQ1[507]=new Array("../7337/1013.pdf","In Situ Observation of Single Ion Damage in Electronic Materials","","1013 doi:10.1017/S1431927615005863 Paper No. 0507 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In Situ Observation of Single Ion Damage in Electronic Materials Daniel Bufford1, Remi Dingreville1 and Khalid Hattar1 1. Sandia National Laboratories, Albuquerque, NM, USA Radiation damages crystalline materials by displacing atoms from their lattice sites, a process that may degrade material properties. In semiconductor components, device performance may deteriorate gradually with increasing fluence due to accumulated damage, however, sub-micrometer length scale device components are also susceptible to sudden problems. These "single event effects", so called because they are caused by a single particle [1], range from transient errors (e.g. flipping of a memory bit), to permanent losses of device functionality, as in gate rupture. With increasingly shorter length design rules, a single cascade may affect an entire element, or even span multiple elements. Even in metals, single ions are capable of creating dislocation structures, or even craters and cavities, all of which reduce electrical conductivity and overall performance. Although some models exist to simulate this radiation-induced defect production in Si, many do not extend past cascade quenching, i.e. time scales on the order of tens of picoseconds. Defect clusters in Si are known to evolve on much longer time scales, stretching to seconds or minutes [2]. In order to better inform computational models currently under development to simulate this behavior on longer time scales, this study observed and characterized defects produced in commonly used electronic materials by single ions as a function of ion atomic mass, energy, and sample temperature. Ion irradiation performed in situ inside of a TEM can provide real time observation of defects produced by irradiation, alongside the powerful suite of microstructural characterization tools the microscope offers. The resulting experimental data approach the time and length scales intended to be addressed by the model. To this end, Si and Au were irradiated during in situ observation, and the resulting defects were characterized and allowed to age for different times at a range of relevant temperatures. In this work, gold samples were prepared by depositing films onto NaCl substrates by pulsed laser deposition, then floating the films off in water and collecting them on Mo mesh TEM grids. These samples were then annealed for 4 hours at 400 �C in a vacuum furnace (~10-7 torr) to coarsen the grains for improved imaging and analysis. Si samples were prepared from single crystal wafers by grinding, polishing, and low-angle ion milling. Additional Si samples were prepared by the small angle cleavage technique [3], which avoids the surface damage caused by ion milling. The in situ ion irradiation experiments were completed at the in situ ion irradiation TEM (I3TEM) facility located at Sandia National Laboratories [4]. This facility includes a JEOL 2100 TEM connected to an EN Tandem Van de Graaff accelerator (0.8-6 MV), among other equipment. To reliably distinguish single ion strikes, the ion beam flux incident on the sample was at most 109 ions cm2 s-1, which corresponds to approximately 1 ion s-1 within the 330 nm � 330 nm viewing field at �100,000 magnification. Video was collected during these experiments to identify the locations of and damage resulting from single ion strikes (see Figure 1), and defects were characterized using bright- and dark-field imaging techniques. Single ions in Si produce locally disordered regions in the vicinity of the cascade. With sufficient ion fluence, these regions overlap and eventually lead to a disordered layer. In contrast, Au remains crystalline during irradiation, with cascades generating an array of point defects, dislocation loops, and stacking fault tetrahedra (Figure 2). The behavior of these various types of radiation-induced defects in the two materials in this study will be compared to previous experimental work, and to various previously developed models. Microsc. Microanal. 21 (Suppl 3), 2015 1014 References [1] JR Schwank, et al, IEEE Transactions on Nuclear Science 60 (2013), p.2074-2100. [2] PD Edmondson, et al, Solid State Phenomena 108-109 (2005), p.145-150. [3] JP Mccaffrey, Ultramicroscopy 38 (1991), p.149-157. [4] K Hattar, et al, Nuclear Instruments & Methods in Physics Research Section B-Beam Interactions with Materials and Atoms 338 (2014), p.56-65. [6] The authors thank X. Zhang (Texas A&M University), D.L. Buller and B.R. Muntifering for their assistance. Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of the Lockheed Martin Corporation, for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-AC0494AL85000. Figure 1. Sequential transmission electron micrograph snapshots from video collected in situ during ion irradiation of Si with 46 keV Au1-. (a,b) Two sequential frames capturing the creation of a defect cluster. (c) Difference image: here features present in (a) but not (b) appear light, and those in (b) but not (a) appear dark. Unchanged features make up the flat, lighter grey background. Hence, the defect created defect (arrow) is more easily seen. Figure 2. Sequential transmission electron micrograph snapshots from video collected in situ during ion irradiation of Au with 2.8 MeV Au4+. (a) Here a several defect clusters are visible from previous single Au ions. (b) Another cluster of defects appears in the center. (c) The difference image shows the extent of the damage caused by the ion in (b).");sQ1[508]=new Array("../7337/1015.pdf","A Combined Effect of Electron Beam and Stress on Plastic Flow of Amorphous Silica Microparticles.","","1015 doi:10.1017/S1431927615005875 Paper No. 0508 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A Combined Effect of Electron Beam and Stress on Plastic Flow of Amorphous Silica Microparticles. Sanjit Bhowmick, Douglas Stauffer, Ryan Major, Oden Warren and S. A. Syed Asif Hysitron, Inc., Minneapolis, Minnesota 55344, USA Radiation induced plastic flow in amorphous silica glass is an important subject in glass science and technology and have been studied for decades by many researchers using high energy ions and particles. However, the deformation behavior of such material irradiated by low energy electrons is not well understood. In comparison to heavier particles and ions, electrons have much higher penetration depths and therefore can generate uniform damage and structural changes throughout the sample. In this study, we investigate plastic flow of silica particles under a combined effect of compressive stress and electron beam inside a scanning electron microscopy. To prepare the particle samples for compression experiments, silica microparticles of diameter ~1 �m were mixed in water, ultrasonicated for 10 minutes, and dispersed on silicon substrates. In situ compression experiments were conducted using a PI 85 SEM PicoIndenter (Hysitron, Inc., Minneapolis, MN) with 5 �m flat punch diamond probe. TriboScan software was used to record and analyze load-displacement data. The load-displacement plots and real-time video of deformation were synchronized and captured during the experiment, which aided post-experimental analysis. Quasistatic compression experiments were conducted at Pmax = 0.05 mN, 1 mN and 4 mN under different beam intensities. Figures 1a-f show deformation behavior of the particles before and after the experiments with beam off and beam on conditions. Considerable variation in the plastic strain has been observed with peak loads and beam condition. In these tests, plastic strain was calculated as /D, where D is the diameter of the particle and is the amount of compression along the indentation axis. At 1 mN load and beam off condition, a particle showed negligible strain (<0.05%), fig. 1a. However, similar diameter particles deformed plastically to 47% and 67% strain when the beam was kept on and probe currents of 30 pA and 480 pA were applied (fig. 1b and c). In another set of experiments at 4 mN peak load and beam off condition, a particle deformed plastically to 36% strain and fractured creating a wedge-shaped missing segment (fig. 1d). When an electron beam with probe current of 30 pA was applied, a similar particle deformed plastically to 58% strain with a surface crack (fig. 1e). However, no crack was observed when a high probe current of 480 pA was applied during a test (fig. 1f). It is important to note that all the particles used in this study were exposed to electron beam before starting a test. So, it can be assumed that irradiation induced damage or defects in all the particles before loading were similar. This leads to a conclusion that both the applied load and beam energy are playing significant role in enhancing the structural changes which causes large plastic flow. Electron beam also shows an instantaneous effect on plastic flow as illustrated in figure 1g. The total energy consumed by a particle during a test can be considered as Ut = Ue (elastic deformation energy) + Up = plastic deformation energy + Ur = beam energy. The variation of strain as a function of load and beam intensity is shown in fig. 1h. Densification, anisotropic deformation, and radiation induced plastic flow will be discussed as possible deformation mechanism in the presentation. An important implication of this study is that electron irradiation under applied stress can induce significant instability and reduction in strength in silica resulting in lower lifetime in many devices where silica is an integral component. Microsc. Microanal. 21 (Suppl 3), 2015 1016 References: [1] P. M. Ajayan and S. Iijima, Phil. Mag. Lett., 65 (1992), p. 43. [2] D. Stauffer and S. Bhowmick, R. Major, O. Warren and S. Asif, Microscopy and Microanalysis, 20 (2014), p. 1544. [3] K. Zheng, Nature Communication, 1 (2010), p 1. [4] E. Snoeks, A. Polman and C. A. Volkert, APL, 65 (1994), p 112646. (a) (b) (c) (d) (e) (f) 4 (g) (h) Figure 1: a-f: Images of silica particles after compression with beam off (a and d) and beam on (b, c, e, f) conditions. Fig. g shows instantaneous effect of beam off, beam on, and variation of probe current. Fig. h shows plastic strain as a function of load and beam intensity.");sQ1[509]=new Array("../7337/1017.pdf","Quantitative Stress Measurements Near Blocked Slip Bands in Austenitic Stainless Steel Using High Resolution Electron Backscatter Diffraction.","","1017 doi:10.1017/S1431927615005887 Paper No. 0509 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantitative Stress Measurements Near Blocked Slip Bands in Austenitic Stainless Steel Using High Resolution Electron Backscatter Diffraction. D.C. Johnson1, M.D. McMurtrey2, G.S. Was1 1. Dept. of Nuclear Engineering and Radiological Science, University of Michigan, Ann Arbor, MI 48109 2. Dept. of Materials Science and Engineering, University of Virginia, Charlottesville, VA, 22904 The objective of this study is to quantitatively measure the stress profile ahead of a blocked slip band in an unirradiated austenitic stainless steel. Austenitic stainless steel is the main structural material for nuclear reactor core components and a potential material for advanced reactor concepts. Defect clusters, loops, and voids produced under irradiation change the deformation mode of these alloys from uniform slip to heterogeneous slip through dislocation channeling. When these dislocation channels interact with a grain boundary, the pile up of dislocations induces a stress at the head of the pile-up. It is believed that this stress increase is strongly correlated with stress corrosion cracking behavior. To date, no realistic quantitative measurements of this stress increase in stainless steels have been made [1]. As a precursor to quantitative measurement of localized stresses in irradiated samples, a method was developed for measuring the stress ahead of a blocked slip band in unirradiated material utilizing high resolution electron backscatter diffraction (HREBSD) patterns and a cross correlation software, CrossCourt3 (CC3), for data analysis. During cross correlation, a low stress reference pattern (taken from the same grain as analysis, but removed from high stress areas) is compared with every other pattern collected during the scan. Differences in the EBSD patterns can be used to generate a distortion tensor which is subsequently used to calculate the stress at each point. This technique was employed for stress measurements performed by Wilkinson et al [2] on a commercial purity titanium alloy. For this study, a Fe-21Cr-32Ni tensile bar was electropolished for EBSD analysis. A constant extension rate tensile (CERT) test was performed in an argon environment at 288�C to 0.7% total plastic strain. EBSD was then performed using a Philips XL30FEG and TSL 6 software. Fig. 1 shows an area of the sample investigated where the slip bands are completely arrested at the grain boundary. This should correspond to the highest stress state since transmission of the slip band would relax some of the stress. The EBSD patterns taken from this area (Fig. 2) show high quality Kikuchi bands which are necessary for accurate stress analysis. Using CC3, the slip bands become visible when observing the geometric mean peak height map, Fig. 3. This plot is generated by correlating the intensity and rotation of each EBSD pattern to the reference with the value for auto correlation being set to 1. Therefore the cross correlation technique is more sensitive to small changes in the pattern quality and can extract more information than image quality or inverse pole figure maps. The stress state as calculated by the CC3 software is shown in Fig 4. Visible here is plume of elevated stress values leading from the head of the slip band into the adjacent grain. The highest stress magnitude observed is at the interaction point, and drops significantly as the distance from the grain boundary increases. Stress values deviate from the matrix stress values over the first 2-3 microns. The magnitude of the stress for these measurements is qualitatively reasonable [3], with peak values in the range of 1-2 GPa for the von Mises stress. Less than 0.1% of the scan points had stress values greater than 10 GPa, but these points also had mean angular errors greater than 5 x 10-3 radians and were removed from the analysis. Given the distribution of stress and the magnitude of measured values, HREBSD has been shown to be a plausible technique for analyzing unirradiated austenitic stainless steel. Similar analysis will be done in the future for irradiated samples to study the effects of irradiation on the stress state ahead of dislocation channels. Microsc. Microanal. 21 (Suppl 3), 2015 1018 References: [1] MD McMurtrey et al, Plasticity 56 (2014), p. 219. [2] TB Britton and AJ Wilkinson, Acta Materialia 60 (2012), p. 5773. [3] TC Lee, IM Robertson, and HK Birnbaum, Philos Mag A 62 (1990), p. 131. Figure 1: Slip band-grain boundary interaction Figure 2: EBSD pattern taken from region in Fig. 1 for analysis in CC3 Figure 3: Geometric Mean Peak Height from CC3 analysis Figure 4: von Mises stress distribution at head of slip band");sQ1[510]=new Array("../7337/1019.pdf","Microstructural Changes in a Highly Irradiated ODS Ferritic MA957 Alloy","","1019 doi:10.1017/S1431927615005899 Paper No. 0510 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microstructural Changes in a Highly Irradiated ODS Ferritic MA957 Alloy Dan Edwards1 and Mychailo Toloczko1 1. Pacific Northwest National Laboratory, Reactor Materials & Mechanical Design Group, Richland WA 99301 To further enable the design of nuclear fuel and duct cladding that can survive neutron doses of >300 displacements per atom (dpa), a variety ferritic/martensitic steels and oxide dispersion strengthened (ODS) ferritic alloys are being evaluated for their response to neutron irradiation over a range of temperatures and neutron doses. Here we report on the microstructural characterization of an ODS ferritic alloy (MA957) pressurized creep tube neutron irradiated in a fast reactor at 943K to 110 dpa. MA957 alloy is an Fe-13.9 wt%Cr with minor additions of Ti and yttrium as Y2O3. This is one of the highest irradiation doses achieved for this alloy that contained enough material to perform both mechanical property and microstructural characterization. The issue of microstructural stability of the ODS ferritic alloys under neutron irradiation remains one of the most active areas of research in the development of advanced nuclear reactors [1-4]. A JEOL 7600F FEG-SEM and a Cs probe corrected JEOL ARM200CF were used to assess the changes in the microstructure due to neutron irradiation. TEM samples were prepared in an FEI Quanta 3D FIB/SEM. Texture of the creep tube was documented via EBSD and precession electron diffraction on the TEM lamella prepared from the unirradiated archive tube and irradiated tube. The second phase particles were characterized in the JEOL ARM200CF using a combination of HAADF, STEM BF, elemental mapping using a JEOL Centurio SDD system, and a Gatan Model 965 GIF with Dual EELS capability. Dislocation imaging was performed using a combination of both conventional TEM and STEM imaging, one of the first known attempts to characterize the dislocation structures of irradiated pressurized creep tubes using aberration corrected STEM. Various combinations of convergence angles and HAADF collection angles were explored to determine the best parameters for imaging dislocations as well as the ODS particles and other second phases in the microstructure. The microstructure of the unirradiated MA957 creep tube consists of a "bamboo" grain structure, that is, high aspect ratio grains with an average length of ~10 �m and a diameter of ~300 nm. The tubes possess a strong <011> texture aligned with the axis of the extruded tube. The dislocation structure is fairly dense, and forms arrays of straight line dislocations that span the narrow dimension of the grains. A moderate density (~1021 per m-3) of Ti oxynitrides (Ti-N-O) were found dispersed throughout the grain structure, with a size ranging from 15-200 nm. These particles are a separate distribution from the 1-2 nm diameter Y-Ti-O nanoxides proven to exist by atom probe tomography measurements; these latter nanoparticles are responsible for the high strengthen and microstructural stability of this alloy. Careful examination in both CTEM and STEM mode revealed that each of the Ti-N-O particles down to sizes less than 10 nm had an attached cavity, presumably due to a cavitation effect from extrusion at higher reductions. These cavities were often aligned at the ends of the particles in the tube axis direction. Irradiation produced substantial changes to the Ti-N-O particles, leading to what appears to be irradiation-induced dissolution of the particles at all sizes. These changes are shown in Figure 1(a) and (b), which compares elemental maps taken from the unirradiated archive tube and the irradiated tube, Microsc. Microanal. 21 (Suppl 3), 2015 1020 respectively. Though not shown clearly at this magnification, the particles in Figure 1(a) all have smooth edges except where cavities are attached to the particle interfaces; the particles in Figure 1(b), taken from the irradiated sample, exhibit jagged and uneven edges when viewed at higher magnifications. The cavities observed in the unirradiated tube were not found in this irradiated tube, possibly due to be removed by radiation enhanced diffusion. Dislocation loops have formed in addition to the pre-existing dislocation structure, which appears to have been altered to produce a more tangled network than seen in the unirradiated material. Despite these changes, the overall microstructure appears to remain intact during irradiation to such a high neutron dose, however, continued irradiation may lead to additional changes if the Ti-N-O particles continue to dissolve. References: [1] M.B. Toloczko, et al., Journal of Nuclear Materials, Vol. 453, 2014, 323-333. [2] G.R. Odette, JOM, 66, No. 12, 2014. [3] S.J. Zinkle and J.T. Busby, Mater Today 12(11) (2009) pp. 12�19. [4] C. Parish and M.K. Miller, Microsc. Microanal, 20(2) (2014) pp. 613-626. [5] The utilized PED system was transferred from Portland State University. A portion of the research was performed using EMSL, a national scientific user facility sponsored by the DOE's Office of Biological and Environmental Research and located at PNNL. A B Figure 1. Elemental maps confirm the presence of a moderate density of Ti-N-O particles located in clusters and as isolated particles. The image (a) is from the unirradiated archive, in (b) from the tube irradiated at 943K to 110 dpa. A low density of smaller Y-Ti-O particles are also present throughout the grains in each condition.");sQ1[511]=new Array("../7337/1021.pdf","Sulfidation of Metal Oxide Crystallites: An Ex-Situ TEM Study","","1021 doi:10.1017/S1431927615005905 Paper No. 0511 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Sulfidation of Metal Oxide Crystallites: An Ex-Situ TEM Study C.E. Kliewer, S.L. Soled, and S. Miseo ExxonMobil Research and Engineering Co., 1545 Route 22 East, Annandale, NJ 08801 The concentration of sulfur in diesel is becoming regulated to increasingly lower levels.1-3 Thus, new catalytic systems are necessary to effectively address this challenge. To effectively develop these new materials, it is important to better understand the complex nanostructures existing in the current hydrodesulfurization (HDS) catalysts.4-11 This study looks into that issue by using exsitu TEM time-temperature-transformation (T-T-T) data to follow the development of the active phase in a Ni-promoted Mo-rich HDS catalyst. Oxide particles were first crushed into fines using an agate mortar and pestle. The fines were dusted onto standard, 200 mesh, holey-carbon-coated TEM grids. The grids were transferred into a Philips CM200F where the oxide particles were imaged in the bright field TEM mode at an accelerating voltage of 200 kV. The locations of the oxide particles on the grid were "mapped" during the TEM examination so that those same particles could be re-located and re-examined with subsequent thermal treatments. A fixed bed reactor designed specifically for ex-situ TEM studies12 was used to treat the grid. Examination of the fresh oxide agglomerates revealed numerous structures (Figure 1a). The metal oxide crystallites are represented by the darker phase, while a binder phase is presented by the light gray, mottled structure. The TEM grid holding the oxide crystallites was then treated at 200 �C for 8 h under a 10% H2S/H2 environment. An examination of the same crystallites revealed a significant difference in the material. While the binder phase did not change, newly formed particles were apparent (Figure 1b). Elemental analysis of these particles indicated that they were enriched in Ni and S. This provided visual evidence of the low temperature Ni phase partitioning that can occur during the sulfidation process and is thus consistent with open literature studies discussing the formation of detrital Ni and Co sulfide particles during the sulfidation of Mo-based catalysts.4,5,7,9,10 The grid was given a final sulfidation treatment at 375 �C for 4 h under a 10% H2S / H2 environment and then returned to the TEM for re-examination. Significant changes were again evident within the agglomerate (Figure 1c). First, the Ni sulfide particles observed in Figure 2b continued to growth with time and temperature. Second, distinct layered structures became evident. The presence of the layered features clearly indicated the Mo-rich sulfide particles had formed within the system. It is well known that the Mo-rich sulfide particles must be decorated with the promoter atoms to be fully active.13 Thus, these data indicate the importance of an overabundance of Ni in the oxide in order to have sufficient Ni remaining to coat the exterior of the Mo-rich sulfide particles and allow catalysis to occur. References: 1. K.G. Knudsen et. al. "Catalyst and Process Technologies for Ultra Low Sulfur Diesel" Applied Catal. A: General (1999) 189 205-215. Microsc. Microanal. 21 (Suppl 3), 2015 1022 2. C. Song "An Overview of New Approaches to Deep Desulfurization for Ultra-Clean Gasoline, Diesel Fuel, and Jet Fuel" Catalysis Today (2003) 86 211-263. 3. M.C. Kerby et. al. "Advanced Catalyst technology and Applications for High Quality Fuels and Lubricants" Catalysis Today (2005) 104 55-63. 4. R. Candia et. al. "Effect of Sulfiding Temperature on Activity and Structures of Co-Mo/Al2O3 Catalysts II" Bull. Soc. Chim. Belg. (1984) 93 763-773. 5. R. Prada Silvy et. al. "Influence of the Activation Temperature on the Degree of Sulfidation, Dispersion, and Catalytic Activity of Co-Mo/Al2O3 Catalysts" Indian Journal of Technology (1987) 25 627-638. 6. E. Prayen et. al. "Morphology Study of MoS2- and WS2-Based Hydrotreating Catalysts by HighResolution Electron Microscopy" J. Catal. (1994) 147 123-132. 7. S. Eijsbouts "On the Flexibility of the Active Phase in Hydrotreating Catalysts" Appl. Catal. 158 (1997) 53-92. 8. L.S. Byskov et. al. "Sulfur Bonding in MoS2 and Co-Mo-S Structures" Catal. Lett. 47 (1997) 177182. 9. S. Eijsbouts et. al. "Changes of MoS2 Morphology and the Degree of Co Segregation during the Sulfidation and Deactivation of Commercial Co-Mo/Al2O3 Hydroprocessing Catalysts" Ind. Eng. Chem. Res. 46 (2007) 3945-3954. 10. S.D. Kelly et. al. "Structural Characterization of Ni-W Hydrocracking Catalysts Using In-Situ EXAFS and HRTEM" J. Catal. (2009) 263 16-33. 11. Y. Okamoto et. al. "Effect of Sulfidation Atmosphere on the Hydrodesulfurization Activity of SiO2supported Co-Mo Sulfide Catalysts: Local Structure and Intrinsic Activity of the Active Sites" J. Catal. (2009) 268 49-59. 12. C.E. Kliewer et. al. "Ex-Situ Transmission Electron Microscopy: A Fixed Bed Reactor Approach" Microscopy and Microanalysis (2006) 12 (2) 135-144. 13. M. Daage and R.R. Chianelli, "Structure-Function Relations in Molybdenum Sulfide Catalysts: The "Rim-Edge" Model" J. Catal. 149 (1994) 414-427. a b c Figure 1: BF TEM image showing: (a) oxide crystallites (dark phase) and binder (lighter gray phase with mottled texture) , (b) original crystallites after a 200 �C, 8 h treatment in a 10% H2S / H2 environment, and (c) original crystallites after a 375 �C, 4 h treatment in a 10% H2S / H2 environment. Arrows indicate the same region in each image. Solid arrow shows region of binder that does not change with sulfidation, while broken arrow presents region of significant phase partitioning with sulfidation.");sQ1[512]=new Array("../7337/1023.pdf","Synthesis and Characterization of LiNbO3 nanofibers","","1023 doi:10.1017/S1431927615005917 Paper No. 0512 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Synthesis and Characterization of LiNbO3 nanofibers M. Cristina Maldonado-Orozco, J. Enrique Sosa-M�rquez, M. Teresita Ochoa-Lara, F. EspinosaMaga�a Centro de Investigaci�n en Materiales Avanzados, S.C., Laboratorio Nacional de Nanotecnolog�a. Miguel de Cervantes 120, Chihuaua, Chih., M�xico 31109 Ferroelectric oxides ABO3 in which A is an alkali or rare-earth element, and B represent a transition metals, such as LiNbO3 with a perovskite structure, are widely used in the fields of nonlinear optics, pyroelectric detectors, electro-optical modulators, thin-film capacitors, and optical memories [1,2]. Because their properties are dependent not only on its chemical composition but also on its structure, shape, and size, it has been found that reduction of the grain size to the nanoscale leads to distinct properties from those of the bulk. Over the last few decades, one dimensional nano materials such as nanotubes and nanofibers, have attracted great attention due to their unique structure and properties, i.e. large specific surface area and chemical/mechanical stabilities. Thus nanofibers can be used as building blocks in nanotechnology [3,4]. Previously, several ceramic nanowires have been synthesized by various processes, e.g. solution method, laser ablation and chemical vapor deposition (CVD). Recently, there has been an intense research effort on electrospinning of ceramics since it is a straightforward way to synthesize nanostructures. The synthesis of LiNbO3 nanofibers was carried out by the electro-spinning method. A detailed description of the procedure can be found in the literature [5]. In this work, the precursor solution was composed of poly(vinylpyrrolidone) (PVP), niobium ethoxide (Nb(OCH2CH3)5 and lithium hydroxide (LiOH), dissolved in ethanol. The solution was heated at 30�C with stirring for 2 hours and then delivered into a metallic needle at a constant flow rate of 0.3 mL/h by a syringe pump. The metallic needle was connected to a high-voltage power supply and a grounded aluminum foil was placed 15 cm from the needle tip. With an applied high-voltage of 13 kV, the precursor solution jet was accelerated toward the aluminum foil, leading to the formation of VOSO4/Nb(OCH2CH3)5/LiOH/PVP fiber composite, together with a rapid evaporation of the ethanol. The composite nanofibers were then annealed at 700 �C for 2 hour, with a slope of 3�C/min to obtain LiNbO3 nanofibers. The presence of pure phase is confirmed by XRD analysis, as shown in Fig. 1, for calcined fibers, showing the formation of crystalline LiNbO3. Diffraction-peak identification is performed on the basis of the PDF2 release 2010 ICDD database. Fig. 2 shows SEM micrograph of as-spun fibers. Cylindrical and randomly oriented fibers were obtained with diameter about 60-150 nm. Fig. 3 shows TEM micrograph from an isolated, calcined LiNbO3 nanofiber, after dispersing the sample in isopropanol, with a few �m long, where the nanoparticles composing the fiber are clearly seen. Fig. 4 shows the same fiber at higher magnification, where well defined planes are observed, indicating a good crystallinity. References [1] Y. Xu, Ferroelectric Materials and Their Applications (North-Holland, Amsterdam, 1991). [2] M. E. Lines and A. M. Glass, Principles and Application of Ferroelectrics and Related Materials (Clarendon, Oxford, 1977). [3] Y. Dzenis, Science 304 (2004) 1917. Microsc. Microanal. 21 (Suppl 3), 2015 1024 [4] D. M. Carrillo-Flores, M. T. Ochoa-Lara, and F. Espinosa-Maga�a. Micron 52 (2013) 39-44. [5] G. Faggio, V. Modafferi, G. Panzera, D. Alfieri, S. Santangelo, J. Raman Spectroscopy 42 (2011) 593-602. Intensity (a.u.) (012) (104) (110) (024) (116) (211) (122) (202) (006) (113) (018) (214) (300) (125) 10 20 30 40 50 60 70 2 (208) Figure 1. XRD pattern of asspun VOSO4/PVP composite. Figure 2. SEM images of as-spun and calcined VOSO4/Nb(OCH2CH3)5/LiOH/PVP composite. Figure 3. TEM image of LiNbO3 nanofibers. Figure 4. TEM image of a nanoparticle composing the nanofiber.");sQ1[513]=new Array("../7337/1025.pdf","ZnO Nanofibers Easily Synthesized by Electrospinning. A New Formula.","","1025 doi:10.1017/S1431927615005929 Paper No. 0513 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 ZnO Nanofibers Easily Synthesized by Electrospinning. A New Formula. J.E. Sosa-Marquez, M. C. Maldonado-Orozco, F. Espinosa-Maga�a, M.T. Ochoa-Lara Centro de Investigaci�n en Materiales Avanzados, S.C., Laboratorio Nacional de Nanotecnolog�a, Av. Miguel de Cervantes #120, Complejo Industrial Chihuahua, C.P. 31109, Chihuahua, Chih., M�xico. ZnO is well known as a wide band gap (3.37 eV) semiconductor and for its special properties like piezoelectricity and photoconductivity, photoluminescence [1], among others, which can result in innovative solutions in high technology, such as lasers, solar cells, gas and chemical sensors, etc. However, the reduction of ZnO size to nanometric scale (nanotubes, nanoparticles, nanofibers, nanowires, etc.), increases the range of technological applications. Particularly ZnO nanofibers, due to its electrical and optical properties, make it a good candidate for to develop solar cells, nano sensors and opto-electronic devices. The aim of this work was to synthesize and to study ZnO nanofibers. The synthesis was realized by Electrospinning Technique (Fig. 1), a simple, continuous and scalable technique to produce polymer composite and metal oxide nanofibers with high aspect-ratio and controlled morphologies [2]. An electrospinning equipment Nabond Unit_Standard was used for producing the nanofibers. Previously, two separate precursor solutions were prepared, one of them containing zinc acetate (Zn (CH3COO)2�2H2O) (0.5M) and absolute ethanol, and the other containing polyvinylpyrrolidone (PVP) and water (30%) and absolute ethanol (70%), used as solvents. The solutions were mixed and the resultant solution was stirred for 24 hours and immediately it was loaded in a syringe of 10 ml, connecting to the electrospinning equipment with a thin hose. A needle was placed at the end in the other extreme of the hose and at 15 cm from collecting surface. A 20 kV voltage was applied between the needle and the collector, resulting in a layer formed by the polymeric fibers. The material produced was heated in air in a furnace at 600 �C for 1 hour to remove the polymer layer for generating the ZnO structure. The ZnO nanofibers characterization was performed by the High Resolution Transmission Electron Microscopy (HRTEM), Energy Dispersive Spectrometry (EDS), Scanning Electron Microscopy (SEM), X Ray Diffraction (XRD) and Thermo-gravimetric Analysis (TGA) techniques. In Fig. 1 (a), it can be observed in the SEM images, the precursor PVP fibers with an average diameter of around 163 nm. TGA study showed the temperature range to eliminate the organic material and to get the reaction for obtaining ZnO nanofibers. After some tests, the best calcination temperature was 600�C. Then the material was examined by the XRD technique and the result is illustrated in Fig. 2(a), showing that the produced material matches with ZnO Wurzite phase. The calcined material was observed in HRTEM (Fig. 2(b)) showing nanofibers with an average diameter of 100 nm. An EDS study complements the results (Fig. 2 (b)). Fig. 3 illustrates a union in a nanofiber. Results indicate that the method employed to produce ZnO nanofibers in this work, allows obtaining them with a good quality. Microsc. Microanal. 21 (Suppl 3), 2015 1026 Acknowledgements: Carlos Ornelas Guti�rrez, Daniel Lardiz�bal Guti�rrez, and Carlos Roberto Santill�n Rodr�guez. Centro de Investigaci�n en Materiales Avanzados, S.C., Laboratorio Nacional de Nanotecnolog�a, Av. Miguel de Cervantes #120, C.P. 31109, Chihuahua, Chih., M�xico. References [1] B. Sen, M. Stroscio and M. Dutta, Journal of Electronic Materials, 40 (2011), p.2015. [2] J. Sundaramurthy, et. Al., Biofuel Research Journal 2, 2 (2014) p. 44 Fig. 1 (a) Electrospinning method diagram and (b) Sem Micrograph of Polymeric ZnO nanofibers 14000 13000 12000 11000 10000 9000 8000 7000 6000 5000 4000 3000 2000 1000 0 20 (101) (100) (002) Intensity (arb. units) (110) (102) (103) 30 40 50 60 (200) (112) (201) 70 80 2 Fig. 2 (a) XRD of the material showing Wurzite Structure and (b) EDS of a nanofiber showing two different zones Fig. 3 HRTEM micrograph of a detail of a union into a nanofiber.");sQ1[514]=new Array("../7337/1027.pdf","Cr(Al)N/Al2O3 nanocomposite coatings fabricated by differential pumping cosputtering","","1027 doi:10.1017/S1431927615005930 Paper No. 0514 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Cr(Al)N/Al2O3 nanocomposite coatings fabricated by differential pumping cosputtering Masahiro Kawasaki1, Masateru Nose2, Ichiro Onishi3, Kenji Matsuda4, and Makoto Shiojiri4,5* JEOL USA Inc., 11 Dearborn Road, Peabody, MA 01960, USA. Faculty of Art and Design, University of Toyama, Takaoka, Toyama 933-8588, Japan. 3. JEOL Ltd. 3-1-2 Musashino, Akishima, Tokyo 196-8558, Japan 4 School of Science and Engineering, University of Toyama, Toyama 930-8555, Japan 5. Kyoto Institute of Technology, Kyoto 606-8585, Japan. * Present address: 1-297 Wakiyama, Kyoto 618-0091, Japan. 2. 1. Nanocomposite films of metal nitrides such as TiN/Si3N4, TiN/BN, and CrN/AlN have attracted substantial attention as new hard coating materials. It is difficult to prepare composite films consisting of nitride and oxide by conventional reactive sputtering methods. Nose et al. developed a differential pumping cosputtering (DPCS) system with two chambers A and B, which can fabricate different nanocomposite films [1]. We elucidated the process and mechanism of film growth in the DPCS system, using Cr(Al)N/SiOx nanocomposite layer deposited on the under buffer layers grown on a Si substrate [2-4]. Here, we report the mechanical property and structure of Cr(Al)N/Al2O3 layers prepared at various conditions in the DPCS system, to demonstrate its usefulness for fabricating superhard coatings. Cr50Al50 and Al2O3 targets were set in chambers A and B respectively, and a (001) Si wafer was used as the substrate. The substrate was heated at 250oC. First, three depositions were successively performed on the Si substrate for making the transition or buffer layers to promote adhesion between the composite layer and substrate. Except for a substrate rotational speed of =12 rpm, the preparation conditions of the gas flow and RF power for these transition layers were the same as that used for the previous Cr(Al)N/SiOx nanocomposite coating [2-4]. Next, the main deposition was carried out for 810 min by operating both the CrAl chamber A and the Al2O3 chamber B, on the transition layers rotated at the same speed . The CrAl sputtering and the Al2O3 sputtering were performed with flows of Ar (10 sccm)+N2 (20 sccm) and Ar (20 sccm) respectively, at different RF powers for preparing composite layers with different compositions: e.g. 200 W and 100 W, respectively so as to obtain a nominal composition of Cr(Al)N/17 vol.Al2O3. The structure was observed by analytical electron microscopy using JEOL JEM-2800 and ARM200F microscopes, and the indentation hardness HIT and Young's modulus E* of the films were measured using a nanoindentation system (Fischerscope, H100C-XYp) at room temperature. We got the following conclusion from the experiments such as shown in Figures 1-4. (1) The transition or buffer layers prepared successively on the Si substrate by sputtering from the CrAl target with flows of (i) Ar (10 sccm), (ii) Ar (10 sccm)+N2 (10 sccm), and (iii) Ar (10 sccm)+N2 (20 sccm) were layers of composed of bcc Cr crystallites and a-Al2O3 particles (layer C), Cr crystallites, NaCl-type CrN crystallites, and a-Al2O3 particles (D), and Cr(Al)N crystallites and a-Al2O3 particles (E), respectively. These layers, where the composition gradually changes from metal (Cr) to nitride (CrN), are appropriate to the adhesion between the metal substrate and the composite nitride layer. The multilayered structure composed of Cr (or CrN) layers and oxide layers, which were found in the layers prepared at =1 rpm, was not observed in the present layers prepared at first rotational speeds. (2) The main layer (F) grew in a columnar structure normal to the substrate surface. Each column comprises Cr(Al)N crystallites and a-Al2O3.particles. The Cr(Al)N crystallites and a-Al2O3 particles were homogeneously dispersed and Microsc. Microanal. 21 (Suppl 3), 2015 1028 any multilayered structure was not formed, unlike the nanocomposite layers prepared at = 1 rpm. (3) HIT and E* increased with increasing within a measured range of 1~12 rpm. In the multilayered structure prepared at low rotational speeds, the nitride crystal lattices in the same layer can easily deform without interrupt by oxides and the oxide particle layers also act on softening the coating. With increasing oxide fraction, HIT and E* increased to reach maximum values and then decreased. This shows that the hardness of nitride coatings is improved by fabricating the nanocomposite layer. The Cr(Al)N/17 vol. %Al2O3 and Cr(Al)N/17 vol %SiOx layers prepared at =12 rpm were superhard coatings with HIT = 44~46 GPa and E*~350 GPa. They had the structure described in (2). The fine a-Al2O3 particles can work as obstacles against the lattice deformation of Cr(Al)N. The increasing amount of amorphous oxide (>17 vol. %) reduced the supremacy of hard nitride and consequently softens the coatings. Thus, we demonstrated that the DPS system allows us to fabricate superhard nanocomposite coatings with a hardness of as high as 45 GPa. Since we have prepared a composite layer with a hardness of 48 GPa, DPCS has potential to fabricate harder coating layers by controlling preparation condition and searching target materials. [1] M. Nose et al, J. Vac. Sci. Technol. A 30 (2012) 011502. [2] M. Kawasaki et al, ACS Appl. Mater. Interfaces 5 (2013) 3833. [3] M. Kawasaki et al, Appl. Phys. Lett. 103 (2013) 201913. [4] M. Kawasaki et al, M&M 2014 Abstract, Hartford CT. 0383-000160. Figure 1(top left). (a) STEM-HAADF image of an area including layers A-F. (b) HAADF intensity profile, (c-g) EELS intensity profiles, and (h-l) EDS intensity profiles along the line indicated in (a). Figure 2(top middle). HR-TEM image of an area including layers E and F. Figure 3(top right). (a) STEM-HAADF image of a cross section, parallel to the substrate surface, of layer F. (b) EDS spectrum from area in (a). (c-f) EDS maps of O-K, Al-K, Cr-K, and N-K signals of the same cross section, respectively. Figure 4 (bottom left). HIT and E* of Cr(Al)N/17vol.%Al2O3 composite layers and Cr(Al)N layers prepared at different substrate rotational speeds. HIT and E* of Cr(Al)N/17 vol.%SiOx, and Cr(Al)N/38 vol.%SiOx layers are added for reference.");sQ1[515]=new Array("../7337/1029.pdf","Synthesis of Hexagonal Bar Shape of ZnO Particles by Using Hydrothermal Treatment","","1029 doi:10.1017/S1431927615005942 Paper No. 0515 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Synthesis of Hexagonal Bar Shape of ZnO Particles by Using Hydrothermal Treatment A. Medina1a, Salom�n E. Borjas1b, P. G. Mart�nez1b, G. Gonz�lez1c, L. B�jar1d, C. Aguilar2, J.L. Bernal3 Instituto de Investigaciones Metal�rgicas, 1b Instituto de F�sica y Matem�ticas, 1c Facultad de Ciencias F�sico y Matem�ticas, 1d Facultad de Ingenier�a Mec�nica, Universidad Michoacana de San Nicol�s de Hidalgo, Morelia, Michoac�n, M�xico, C.P. 58000 2 Depertamento de Ingenier�a Metal�rgica y Materiales. Universidad T�cnica Federico Santa Mar�a. Av. Espa�a 1680, Valpara�so, Chile. 3 Automotive Mechanics Department. Universidad Polit�cnica de Pachuca. Zempoala, Hidalgo. M�xico 1a In the last decade, the ZnO have took bigger importance because it is widely used as additive in several materials and products including plastics, ceramics, glass, cement, rubber lubricants, paint, adhesives, sealant, pigments, fire retardants [1]. Recently, it has also received special attention in applications such as heat-protecting windows, optoelectronic, gas sensing, transparent electrodes and solar cell applications [2]. ZnO is a versatile semiconductor material with a wide direct band gap of 3.3. eV and has been produced with different morphologies like wires, tubes, belts, bows, rods, rings, springs, stars, flowers, sheet networks, disks, columns, needles, and nuts [3]. Different synthesis process of ZnO have been used; however, both the effect of the temperature treatment and Zn source on the ZnO morphology by using hydrothermal method has not been studied. In this work, zinc nitrate hexahydrate (SigmaAldrich, purity 99 %) was used. Firstly, 5.95 g of Zn source was dissolved in 21.62 g of distilled water. Then, a second solution was prepared by dissolving 0.80 g of NaOH in 21.62 g of distilled water. The second solution was added slowly to the first solution and stirred for 30 min. The resulting suspension with the molar composition of Zn source: NaOH:H2O = 1:1:120 was hydrothermally treated in an autoclave at 160 �C for 1 day under static conditions. The synthetized particles were analysed by X-ray Diffraction (XRD), field emission scanning electron microscopy (FE-SEM) and dispersive energy spectroscopy (EDS). The results showed the formation of ZnO hexagonal crystal structure which can appreciate in figure 1. The image 2(a) show the morphology of the ZnO hexagonal particles with a 5 �m of large and 2 �m of thickness and the figure 2(b) shows that the hexagonal particles are formed by Zn and O elements. The molar ratio of Zn:NaOH of 1:1 was determinate in the formation of hexagonal shape of ZnO crystals by using hydrothermal method at high temperature aged obtained a new synthesis procedure for ZnO hexagonal particles by controlling its homogeneity, morphology and bar size of the particles. References [1] Y.Zhang, et al, Nanoscale Res. Lett. Volume 3 (2008) p. 201 [2] Daniel Vanmaekelbergh and Lambert K Van Vugt, Nanoscale Volume 3 (2011) p. 2783 [3] Kushwaha et al, AIP Advances, Volume 3 (2013) p. 042110-1. Microsc. Microanal. 21 (Suppl 3), 2015 1030 Figure 1. XRD pattern of ZnO particles synthesized at 160 �C. Figure 2. a) SEM image of ZnO hexagonal particles, b) EDS spectrum from the hexagonal particles.");sQ1[516]=new Array("../7337/1031.pdf","Fe Doped-ZnO Nanofibers Synthesized by Electrospinning.","","1031 doi:10.1017/S1431927615005954 Paper No. 0516 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Fe Doped-ZnO Nanofibers Synthesized by Electrospinning. J.E. Sosa-Marquez, M. C. Maldonado-Orozco, F. Espinosa-Maga�a, M.T. Ochoa-Lara Centro de Investigaci�n en Materiales Avanzados, S.C., Laboratorio Nacional de Nanotecnolog�a, Av. Miguel de Cervantes #120, Complejo Industrial Chihuahua, C.P. 31109, Chihuahua, Chih., M�xico. . ZnO semiconductor, with a 3.37 eV band gap, has proven to be a useful and promising material for electronic and optical applications, specially, in the nanotechnology area. However, doped ZnO with transition metals offers more scientific and technological possibilities, such as diluted magnetic semiconductors or to modify its optical and electrical properties with metals as Al and Ga [1]. ZnO, in nanoscience, has been studied in many different ways, such as films, nanoparticles, nano-rods, nanotubes and recently nanofibers. ZnO nanofibers have showed an excellent surface/volume relation, flexibility, easy to produce, etc. The synthesis of Fe doped ZnO nanofibers has been accomplished by A. Baranowska-Korczyc et al [2]. They reported a method to synthesize it, using as precursors zinc acetate (C4H6O4Zn_2H2O, CHEMPUR) and iron acetate(CH3CO2)2Fe. In addition, they dissolved poly(vinylalcohol)(PVA) in distilled water. Both solutions were mixed and processed in an electrospinning equipment to perform the nanofiber synthesis. There are many techniques to produce nanofibers, but most authors agree that the electrospinning technique has shown to be easier, more versatile and more efficient than others. In this report, Fe doped ZnO nanofibers synthesized by Electrospinning Technique, using an Electrospinning Equipment Nabond Unit_Standard, is described. The synthesis process required the preparation of a solution formed by the combination of two precursor, one of them containing zinc acetate (Zn (CH3COO)2�2H2O) and iron acetate(CH3CO2)2Fe (0.5M) in absolute ethanol. The other precursor solution was prepared with polyvinylpyrrolidone (PVP), water (30%) and absolute ethanol (70%). The solutions were mixed and stirred for 24 hours. A transparent and yellow substance was obtained, which was loaded in a syringe of 10 ml, connecting to the electrospinning equipment with a thin hose. A needle was placed at the end in the other extreme of the hose, at 15 cm from collecting surface. A 20 kV voltage was applied between the needle and the collector. A layer was formed by the polymeric fibers over the collector. The ZnO nanofibers characterization was performed by High Resolution Transmission Electron Microscopy (HRTEM), Energy Dispersive Spectrometry (EDS), Scanning Electron Microscopy (SEM), X Ray Diffraction (XRD) and Thermo-Gravimetric Analysis (TGA) techniques. In Fig. 1, it can be observed the SEM images, the precursor PVP fibers with an average diameter of around 300 nm. Fibers were heated in air in a furnace at 600 �C for 1 hour to eliminate the PVP layer and to synthesize the ZnO structure, in agreement with the TGA study. Next, the material was examined by the XRD technique and the result is illustrated in Fig. 2, showing that the produced material matches with ZnO Wurzite phase. However, two lines unexpected, belonging to Fe2O3 structure appeared. They are very small, but it is possible that a small excess of the iron acetate amount and the heat process Microsc. Microanal. 21 (Suppl 3), 2015 1032 favored the growing of this phase. On the other hand, Fig. 3 exhibits HRTEM results, showing an example of nanofiber with an average diameter of 50 nm. Acknowledgements: Carlos Ornelas Guti�rrez, Karla Campos Venegas and Daniel Lardiz�bal Guti�rrez. Centro de Investigaci�n en Materiales Avanzados, S.C., Laboratorio Nacional de Nanotecnolog�a, Av. Miguel de Cervantes #120, Complejo Industrial Chihuahua, C.P. 31109, Chihuahua, Chih., M�xico. References [1] Min-Chul Jun, Sang-Uk Park and Jung Hyuk Koh, Nanoscale Research Letters, 7 (2012), p.639. [2] A. Baranowska-Korczyc, K. Fronc, J.B. Pelka, K Sobczak, D. Klinger, P. Dluzewski, and D. Elbaum, Radiation Physics and Chemistry, 93 (2013) p. 21 Fig. 1 SEM micrograph showing polymeric Fe doped ZnO nanofibers Fig. 2 XRD of the material showing Wurzite Structure and some lines of Fe2O3 phase Fig. 3 HRTEM micrograph of Fe doped ZnO Nanofiber");sQ1[517]=new Array("../7337/1033.pdf","Chemical and structural characterization of PbS-SiO2 Core-shell structures synthesized by ultrasonic wave-assisted chemical bath deposition.","","1033 doi:10.1017/S1431927615005966 Paper No. 0517 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Chemical and structural characterization of PbS-SiO2 Core-shell structures synthesized by ultrasonic wave-assisted chemical bath deposition. P. Moreno-Wong1, J. Alvarado-Rivera1,2, Gemma Moreno1*, M. C. Acosta-Enr�quez3, S. J. Castillo3. 1. 2. Departamento de F�sica, Universidad de Sonora, Hermosillo, Sonora, M�xico. Consejo Nacional de Ciencia y Tecnolog�a, M�xico, D.F., M�xico. 3. Departamento de Investigaci�n en F�sica, Universidad de Sonora, Hermosillo, Sonora, M�xico. Heavy metal sulphures such as lead sulfide (PbS) are semiconductors used in electronic devices. By controlling the size, shape and chemical components of this type of materials, it's optical and electronic properties can be manipulated [1]. PbS has a small hole which is almost equal to electron mass, this leads to a large exciton with a Bohr radius of 20nm, approximately. In nanometric size, electrons and holes and, by consequence, the exciton, can be strongly confined [2]. The ability for adjusting the energy emitted by the lead's photon in a quantum system, in the region near IR of the electromagnetic spectrum makes this material desirable for the production of quantum dots, nanotransistors, solar cells, among other optoelectronic devices. The commonly used formula to synthesize PbS contains trietanolamine as complexing agent. In this work we exchange that compound for citric acid (C6H8O7), which is commonly found and easily biodegraded. Monodisperse silica microspheres were synthesized by sol-gel process, using the St�ber method [3]. By varying the concentration of the precursors, the size of the silica particles can be controlled. The following precursors were used: tetraethylorthosilicate (TEOS, Si(OC2H5)4) as precursor of silica, ethanol and deionized water as solvents and ammonia hydroxide (NH4OH) as a catalyst. The synthesis took place in a glass beaker at room temperature with magnetic stirring for 24 hours. Once the reaction is finished, the obtained microspheres were washed using centrifugation and deionized water. For the core-shell PbS-SiO2 structures, the monodisperse silica microspheres were used as a base for PbS nanoparticles to grow on by using ultrasound assisted chemical bath. The following precursors compounds were used: lead acetate (Pb(C2H3O2)2) 0.5M and thiourea (SC(NH2)2) 1M as a PbS precursors, sodium hydroxide (NaOH) to enhance OH- ion release, this way controlling the reaction's pH, citric acid (C6H8O7) as complexing agent and deionized water. We included the colloidal spheres solution and placed the beaker in an ultrasonic bath for sonication. The obtained samples of decorated spheres were washed with deionized water; afterwards they were dried in a conventional oven at 75�C. For comparison, a sample without silica microspheres was synthesized to obtain only PbS nanopowders. Monodisperse silica of ~550 nm was obtained; they were analyzed by Dynamic light scattering using a Zetasizer Nano Range analyzer (Malvern). PbS was synthesized in an ultrasonic bath with an innovating, less harmful and less costly formula than conventional ones. In Figure 1, X-Ray Diffraction patterns of both samples with and without silica are presented. It is observed that for both samples PbS is obtained, and for the PbS-SiO2 sample the presence of hydrocerussite (Pb(CO3)2(OH)2) was identified. It seems that the citric acid in the presence of the microspheres promotes the formation of the hydrated lead carbonate as a secondary reaction product. X-ray photoelectron spectroscopy (XPS) of the PbS powders was carried out in Perkin Elmer PHI 5100 spectrometer and the results are displayed in Figure 2. Analysis of the high resolution spectra of C 1s, O 1s, S 2p and Pb 4f photoelectron lines shows Microsc. Microanal. 21 (Suppl 3), 2015 1034 that PbS powders it does contain hydrocerussite and lead oxide. Pb 4f7/2 is located at 138.3 eV and it corresponds to lead in hydrocerussite [4]. Scanning electron microscopy measurements were performed on the PbS decorated silica spheres, the samples were gold coated for observations. It can be seen in Figure 3 a) the laminar crystals of hydrocerussite, which have sizes larger than 1 m. At higher magnifications it can be observed that the silica spheres are irregularly coated and small crystals are scarcely dispersed over the sphere surface. EDS analysis showed that the surface is rich in lead, which it can indicate PbS starts to growth on this new layer [4]. It can be concluded that citric acid can be used as a complexing agent for PbS synthesis by ultrasonic wave-assisted chemical bath deposition [5]. References: [1] Rogach, A. L. et al, Small 3 (2007), p. 536�557. [2] Zhao, X. S., et al, Langmuir 21 (2005), p. 1086�1090. [3] Bogush, G.H., Tracy, M.A. and Zukoski IV, C.F., Non-Crystalline Solids 104 (1988), p. 95-106. [4] NIST X-ray Photoelectron Spectroscopy Database, NIST Standard Reference Database 20, Version 4.1, http://srdata.nist.gov/xps/Default.aspx. [5] The authors acknowledge the participation of Dr. B�rbara Bermudez Reyes from the Research Center of Aeronautic Engineering and Innovation of the University of Nuevo Le�n (CIIIA-UANL), M�xico, for her valuable assistance with SEM characterization. . Figure 1. XRD pattern of a) PbS powder, b) PbS-SiO2 system. a) Figure 2. XPS high resolution spectra of C 1s, O 1s, S 2p and Pb 4f photolectronic lines of PbS. [4] b) 0.5 m Figure 3. SEM images of the PbS-SiO2 core-shells.");sQ1[518]=new Array("../7337/1035.pdf","Chemical Synthesis and Structural/Analytical Characterization of CuNi-Al2O3 Nanocomposites.","","1035 doi:10.1017/S1431927615005978 Paper No. 0518 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Chemical Synthesis and Structural/Analytical Characterization of CuNi-Al2O3 Nanocomposites. M.I Ramos.1; E. Brocchi. 1; R. Navarro.1; I.G Solorzano.1; N. Suguihiro.2 Department of Chemical and Materials Engineering, PUC- Rio de Janeiro, Brazil 2. Brazilian Center for Physics Research, Rio de Janeiro, Brazil. 1. It is well established that nanostructured materials show advanced mechanical and transport properties [1]. These materials, however, can be further enhanced through the inclusion of nanoparticles in the structure of a particular matrix, thereby forming nanocomposites [2]. Under this perspective, CuNi alloys, which are highly ductile material, can have its hardness considerably improved with the incorporation of Al2O3 nanoparticles [3]. Different methods for producing nanocomposites in laboratory scale have been studied from a variety of physical and chemical methods. The present study is centered on a chemical route process aiming at obtaining metallic nanoparticles of homogeneous composition in which a fully homogeneous dispersion of a nano scale ceramic phase (Al2O3) has been incorporated. The synthesis procedure can be divided into two steps. First, thermal decomposition of a nitrate solution containing, Ni(NO3)2, Cu(NO3)2 and Al(NO3)3 in an adequate proportion of them. The second step is the preferential hydrogen reduction (copper and nickel) of the oxides mixture obtained in the first step. Based on the nitrates mixture composition the Ni, Cu and Al2O3 contents can be pre-established. The obtained nanocomposite powder was analyzed by XRD for phase identification of the product powder. SEM-EDS analysis was employed as a first step for an overall appreciation of the microstructure; extend of particles agglomeration and also for obtaining elemental chemical mappings. TEM techniques have been extensible used under diffraction a phase contrast modes, using a LaB6 Jeol 2010 and a FEG Jeol instruments, both operating under 200 kV accelerating potential. A detailed study of particle morphology and size distribution of the synthesized powder shows it is primarily constituted of spheroidal-like particles with diameters in the 5 to 60 nm range, some exhibiting faceting, Figure 1. TEM/STEM analytical techniques, including EDS elemental mapping , shows that the entire synthesis procedure has been successful since it has produced metallic nanoparticles with homogeneous composition in Ni and Cu with a fine dispersion of smaller Al2O3 nanoparticles, with an average 10nm in size. A typical example is displayed in Figure 2, a bright field /dark field pair of a composite nanoparticle with an inner structure suggesting a core/shell assembly, a unique structural feature which is consisted of an inner core containing Al2O3 and an external shell of CuNi. Figure 3. Shows the EDS spectrum acquired from the edge of the particle shown in Fig 2. Finally, micro hardness measurements carried out on consolidates bulk pellets of CuNi/-Al2O3 nanocomposites confirmed the expectations that mass addition of Al2O3, even about 1%, can result in a much higher hardness as compared with binary CuNi alloy, 250 Hv and 147 Hv respectively [4]. References: [1] E. A. Brocchi, R. C. S. Navarro, M. S. Motta, F. J. Moura and l. G. Sol�rzano; Materials Chemistry and Physics, vol. 140, pp. 273 - 283, 2013. [2] M.S. Motta, E. A. Brocchi and I.G. Solorzano; Mater, Sci, Eng C15, 2001 [3] M.I. Ramos: M.SC disertation, PU-Rio de Janeiro, Brazil, 2013 [4] The authors acknowledges CNPq (Brazil)for financial support and LABNANO (CBPF-Brazil). Microsc. Microanal. 21 (Suppl 3), 2015 1036 Figure 1. a) Bright field TEM of CuNi-Al2O3 nanoparticles agglomerate;b) Histogram of composite nanoparticles size distribution Figure 2. TEM images of nanoparticle boxed in Fig.1: a) Bright field b) Centered Dark Field. Figure 3. EDS spectrum from the edge of shell nanoparticle shown in the Fig2.");sQ1[519]=new Array("../7337/1037.pdf","Recycled Al Reinforced with Oxide Nanoparticles Produced by Stir-Casting Method","","1037 doi:10.1017/S143192761500598X Paper No. 0519 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Recycled Al Reinforced with Oxide Nanoparticles Produced by Stir-Casting Method A. Santos-Beltr�n1, V. Gallegos-Orozco1, M. Santos-Beltr�n2, F.J. Baldenebro-Lopez 1, C.D. G�mez-Esparza2, I. Ronquillo-Ornelas1 and R. Mart�nez-S�nchez2 1 Universidad Tecnol�gica de Chihuahua Sur, Km. 3.5 Carr. Chihuahua a Aldama, C.P. 31313. Chihuahua, Chih. M�xico 2 Centro de Investigaci�n en Materiales Avanzados (CIMAV), Miguel de Cervantes No. 120, C.P.31109, Chihuahua, Chih., M�xico Aluminum alloys reinforced with hard nanoparticles named Metal Matrix Nanocomposites (MMNCs) are very attractive in many applications in the industry, this kind of materials exhibit improved mechanical properties with relatively low contents of reinforcement. Automotive and aerospace industries are demanding these composites for critical applications taking into account their low density and high temperature resistance characteristics. MMNC's are materials reinforced with hard particles (e.g. oxides and nitrides) with size ranging from 10 nm to 100 nm. In the present work, nanocomposites based aluminum with hard nanoparticles of TiO2 and CeO2 were fabricated by combining two techniques such as mechanical milling and the stir-casting method. Compared to other routes, melt stirring process has some important advantages, e.g., the wide selection of materials, better matrix� particle bonding, easier control of matrix structure, simple and inexpensive processing, flexibility and applicability to large quantity production and excellent productivity for nearnet shaped components [1,2]. Nanoparticles and metallic powders, in the weight ratio of Recycled Al/nanoparticles = 3, were separately milled using a Spex ball mill in uncontrolled atmosphere during 2h. The device and milling media used were made from hardened steel. The milling ball to powder weight ratio was set to 5:1. Consolidated samples were added into molten recycled Al using a resistance furnace equipped with a graphite stirring system. Each cylinder was hot extruded in a direct extrusion system at 550 �C. The specimens in both asmilled and as-sintered conditions were studied by scan electron microscopy (SEM) and atomic force microscopy (AFM). The SEM bright-field image (see Fig. 1a) shows the microstructure of the Al-TiO2 nanocomposite, the inset shows a close up image of the TiO2 nanoparticles dispersed into the recycled Al matrix; these particles are in the size range of about 80 to 100 nm. Fig. 1b shows the AFM topography image of the Al-TiO2 composite after the hot extrusion process, the image reveal a homogeneous crystallite size distribution of about 50 to 100 nm. The inset shows the profile of the crystallite. Fig. 2a shows the SEM bright-field image of the microstructure of the Al-CeO2 nanocomposite after the hot extrusion process; the image also shows the presence of some fiber-shaped CeAl intermetallic compound. In the inset is clearly observed the CeO2 nanoparticles dispersed into the recycled Al matrix. The Figure 1b shows the crystallite size distribution where most of these crystallites are below 100 nm in size. The presence of these hard nanoparticles dispersed into the recycled Al prevents by the pinning effect, the excessive increase of the crystallite during thermo-mechanical extrusion process. The combined effect of hard nanoparticles dispersion and the small crystallite size, improved the mechanical properties of the recycled Al matrix. References [1] C. Suryanarayana, Nasser Al-Aqeeli. Progress in Materials Science 58 - Issue 4 (2013), 383-502. [2] H. Abdoli, H. Asgharzadeh, E. Salahi. J. Alloys. Compd 473 (2009), 116�122. Microsc. Microanal. 21 (Suppl 3), 2015 1038 a) b) Fig.1. a) SEM bright-field image that shows the Al-TiO2 microstructure and the inset shows the TiO2 nanoparticles dispersion into the Al matrix. b) AFM image that shows the crystallite size of the Al-TiO2 nanocomposite matrix and inset shows the height profile of the crystallite. a) b) Fig.2. a) Microstructure of the Al-CeO2 sample (SEM bright-field image), the inset shows a homogeneous nanoparticle dispersion into the Al matrix, b) SEM bright-field image that shows the crystallite size distribution.");sQ1[520]=new Array("../7337/1039.pdf","Microstructure and mechanical properties of Al-composites reinforcing with Gr/Ni nanoparticles processed by high-energy ball milling.","","1039 doi:10.1017/S1431927615005991 Paper No. 0520 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microstructure and mechanical properties of Al-composites reinforcing with Gr/Ni nanoparticles processed by high-energy ball milling. J.M. Mendoza-Duarte, R. Mart�nez-S�nchez, C. Carre�o-Gallardo, I. Estrada-Guel. Centro de Investigaci�n en Materiales Avanzados (CIMAV). Laboratorio Nacional de Nanotecnolog�a Miguel de Cervantes No. 120, 31109, Chihuahua, Chih., M�xico. Aluminum (Al) composites have been applied in the automotive and aerospace industry [1] due their low density and high strength. These materials can be produced with oxides, carbides or nitrides dispersed in the metallic matrix by mechanical milling. Recent studies show an increment in mechanical properties with different reinforcing materials as nanotubes [2-4], graphenes [5] and carbon [6]. These additives induce different reinforcement levels, depending how do they interact with material dislocations [7]. Graphite (Gr), an allotropic form of carbon, is used as reinforcing additive because it is cheap, abundant and can be easily comminuted. D. Yadav et al. [8] added nickel (Ni) particles to an Al matrix by friction stir processing, finding that Ni helps to obtain grain refinement and increase the mechanical properties of the prepared composites. This work deals with the synthesis of some Al-based composites reinforced with Ni nanoparticles covered with Gr by mechanical milling. These composites were prepared and their morphology and mechanical properties were evaluated as a function of milling intensity. The first step was cover Ni particles with Gr by mechanical milling, this process was carried out using a SPEX-8000M mill using steel balls with a ratio (milling media to powder) of 5:1 (in wt.), mixtures of 90% Gr and 10% Ni were milled during 0, 1, 2, 3 and 8h periods. After, the Al composites were synthesized following same procedure using mixtures of pure Al powder and previously metallized 1%Gr-Ni particles (in wt.). The milled composites were compacted using a cylindrical die under 900MPa and sintered 4h at 623K in order to prepare samples for mechanical tests. Micrographs in Fig. 1 show an important increment of the particle size of composite particles after milling, forming equiaxal particles due Al ductility and milling media impacts. In Fig. 1a some Gr/Ni particles embedded in the Al matrix are shown an attached EDS analysis confirms their composition. The Fig. 2.1 presents the stress�strain curves of the studied Al-Ni/Gr composites compared with milled and un-milled pure Al sample used as a reference. Pure Al milled sample (Alp 2h) shows a notable increment on its mechanical response (compared with un-doped and un-milled Alp 0h), this is caused by the increment of the dislocations density and grain size reduction achieved by milling. On the other hand, with Gr/Ni addition, a more important increment is obtained, reaching a maximum value with Gr/Ni 4h addition. Hardness tests were also carried out and the results are detailed in the Fig. 2.2. In the graph, it is noticeable a higher hardness value of pure Al milled sample in comparison with the un-milled reference. Also, the highest increment was found with the composite milled by 2h with Gr/Ni 4h addition. Microsc. Microanal. 21 (Suppl 3), 2015 1040 References: [1] S. Hurley, Met. Bull. Mon. (1995) 54-56 [2] Y. Wu, G.-Y. Kim, A.M. Russell, Mater. Sci. Eng. A 538 (2012) 164-172. [3] S. Salimi, H. Izadi, A.P. Gerlich, J. Mater. Sci. 46 (2011) 409-415. [4] K. Hansang, L. Marc, Nanotechnology 23 (2012) 415701. [5] R. P�rez-Bustamante, D. Bola�os-Morales, J. Bonilla-Mart�nez , I. Estrada-Guel, R. Mart�nezS�nchez, J. Alloys Comp. (2014) 578-582. [6] Halil Arik. Mat. Design (2004) 31-40. [7] S. Goussous, W. Xu, X. Wu, K. Xia, Compos. Sci. Tech. 69 (2009) 1997-2001. [8] Devinder Yadav, Ranjit Bauri Mat. Sci. Eng. A 528 (2011) 1326-1333. [9] This research was supported by CONACYT (Project Nr. 169262) and the Redes Tem�ticas de Nanociencias y Nanotecnolog�a (124886). 0h 1h 2h 4h 8h a) Element C K Al K Ni L Weight% 4.47 37.78 57.75 Atomic% 13.49 50.81 35.70 Figure 1. SEM micrographs (SE detector) of Al-Gr/Ni particles of milled composites. (a) Cross section image (RE detector) of Al Gr/Ni4h sample after 2h of milling. (MPa) 500 450 400 350 300 1) 100 2) 80 200 150 100 50 0 0.00 AlCNi 8h M2h AlCNi 4h M2h AlCNi 2h M2h AlCNi 1h M2h AlCNi 0h M2h Alp2h Alpunmilled 0.05 0.10 0.15 0.20 HRF 60 250 40 20 AlCNi0h AlCNi1h AlCNi2h AlCNi4h AlCNi8h Alp 0 1 2 4 6 8 0 Milling Time (h) Figure 2. 1) Compressive test curves of samples milled 2h and 2) Plot of hardness results as a function of milling intensity and Gr/Ni addition.");sQ1[521]=new Array("../7337/1041.pdf","Evaluation of Mechanical Properties of Aluminum Alloy (Al-2024) Reinforced with Carbon-Coated Silver Nanoparticles (AgCNP) Metal Matrix Composites","","1041 doi:10.1017/S1431927615006005 Paper No. 0521 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Evaluation of Mechanical Properties of Aluminum Alloy (Al-2024) Reinforced with Carbon-Coated Silver Nanoparticles (AgCNP) Metal Matrix Composites C. Carre�o-Gallardo1, J.M. Mendoza-Duarte1, C. L�pez-Mel�ndez2, I. Estrada-Guel1 and R. Mart�nez-S�nchez1. Centro de Investigaci�n en Materiales Avanzados (CIMAV). Laboratorio Nacional de Nanotecnolog�a. Miguel de Cervantes No. 120, 31109, Chihuahua, Chih., M�xico. 2. Universidad La Salle Chihuahua, Prol. Lomas de Majalca No. 11201, C.P. 31020 Chihuahua, M�xico. Aluminum matrix composites (AMCs) are emerging as advance engineering materials due to their strength, ductility and toughness. The aluminum matrix is getting strengthened when it is reinforced with the hard ceramic particles like SiC, Al2O3, and B4C etc. Aluminum alloys are still the subjects of intense studies, as their low density gives additional advantages in several applications. These alloys have started to replace cast iron and bronze to manufacture wear resistance parts. MMCs reinforced with particles tend to offer enhancement of properties processed by conventional routes. The alloys primarily utilized today in transport aircraft are 2024-T4 and the alloys having still higher strength (2014-T6, 7075-T6, 7079-T6 and 7178-T6). Aluminum alloy 2024 has good machining characteristics, higher strength and fatigue resistance than both 2014 and 2017. It is widely used in aircraft structures, especially wing and fuselage structures under tension. It is also used in high temperature applications such as in automobile engines and in other rotating and reciprocating parts such as piston, drive shafts, brake- rotors and in other structural parts which require light weight and high strength materials [1] Raw materials were Al2024 alloy as the metal matrix and AgC nanoparticles with a mean size of 15 nm as reinforcement agent. Initial Al2024 burhs were produced by machining a commercial solid bar. These coarse powders were mixed with AgcNP in differents concentrations, chemical composition of Al2024 is shown in Table 1 and the addition of the reinforcement particles are show in Table 2. The mechanical milling of powder mixture was performed in a high-energy horizontal attritor mill (ZOZ CM01 Simoloyer), under an inert atmosphere of pure argon gas. The milling container was made of stainless steel and milling media was made of hardened chrome steel, 2 ml of methanol was used as a process control agent to avoid excessive cold-welding of the powder particles. The Al2024-AgCNP composite powders used to prepare the extruded rods were ball milled for 10 hours. A blank (reference) sample without any reinforcement addition was produced under the same conditions for comparison purposes. The milling ball-to-powder weight ratio was set at 20:1. The behavior of tensile test is presented in Fig. 1, the composites with nanometric particulates exhibited a higher yield and tensile strength compared with Al2024 reference alloy observed tendency is that yield strength and UTS values of samples increases as AgCNP content increases. The experimental values obtained in extrusion condition are higher than those obtained after solution treated at 495�C and water quenched, due extrusion process and because the solution treated does not provide an increase in 1. Microsc. Microanal. 21 (Suppl 3), 2015 1042 resistance to aluminum alloy 2024, a similar behavior was found in the UTS, with the subsequent aging process, the yield strength and UTS value of the composite was higher than that other two conditions due to formation of precipitates (S). Figure 2 shows the morphology, distribution and amount of such precipitates in the composites reinforced after aged processing. These small needles are homogeneously dispersed into the alloy matrix; this could indicate that the nucleation of the (S) precipitates was induced by the aging treatment. References: [1] Feng YC, Geng L, Zheng PQ, Zheng ZZ, Wang GS. Fabrication and characteristic of Al-based hybrid composite reinforced with tungsten oxide particle and aluminum boratewhisker by squeeze casting. Materials & Design 2008; 29: 2023�6. [2] The authors acknowledge to Red Regional de Aeron�utica of program CONACyT grupos interinstitucionales regionales emergentes (GIRE) Regs. 235620 and the Redes Tem�ticas de Nanociencias y Nanotecnolog�a (152992) 520 340 320 500 480 460 440 Yield strength [MPa] 300 280 260 240 220 200 180 160 0.0 0.5 1.0 Al Al Al /Ag NP Solution trated at 495�C C UTS [MPa] 420 400 380 360 2024 340 320 300 Al Al Al 2024 2024 2024 /Ag NPSolution treated at 495�C C /Ag NP Extruded C /Ag NPAged at 190�C C /Ag NP Extruded 2024 C 2024 /Ag NP Aged at 190�C C 1.5 2.0 2.5 3.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 AgCNP [wt %] AgCNP [wt %] Figure 1. Variation of ultimate tensile strength (UTS) and yield strength of the pure Al2024 alloy and prepared composites as a function of AgCNP concentration in the three conditions. Figure 2. TEM micrograph shows the precipitates (S) interacting with AgCNP.");sQ1[522]=new Array("../7337/1043.pdf","Synthesis of Ti-Nb-Ta-Zr Alloys Foams by Powder Metallurgy","","1043 doi:10.1017/S1431927615006017 Paper No. 0522 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Synthesis of Ti-Nb-Ta-Zr Alloys Foams by Powder Metallurgy C. Aguilar1, A. Medina2a, L. B�jar2b, D. Guzm�n3, S. Lascano1, H. Carre�n2a, I. Alfonso4 1 Depertamento de Ingenier�a Metal�rgica y Materiales. Universidad T�cnica Federico Santa Mar�a. Av. Espa�a 1680, Valpara�so, Chile. 2a Instituto de Investigaciones Metal�rgicas, 2b Facultad de Ingenier�a Mec�nica, Universidad Michoacana de San Nicol�s de Hidalgo, Morelia, Michoac�n, M�xico, 58000 3 Departamento de Metalurgia, Universidad de Atacama, Av. Espa�a 485, Copiap�, Chile 4 UNAM, Campus Morelia, Antigua carretera a P�tzcuaro 8701, Morelia, Mich., M�xico. 58190. Human has some degenerative diseases which lead to degradation of mechanical properties of the bone [1]. It is know that biomaterials are solution for bone replacement. Titanium and titanium-based alloys are used due to its superior biocompatibility and corrosion resistance in conjunction with low density and an elastic modulus much more similar to that of human bones [2]. The elastic modulus of titanium and titanium-based alloys are much higher than of human bone (10-30 GPa) [3]. The elastic modulus mismatch would cause bone loss, implant loosening and premature failure of the artificial hip [4]. Therefore, an important requirement of biomaterials is a modulus close to that of natural bones. Torres et al. [5] reported that elastic modulus values decreased with the porosity increased in cp Ti foams. This work, study the application of NaCl to synthetize foams of Ti-2Nb-4Ta-8Zr. Ti based alloys were produced by mechanical alloying using a planetary mill for 4 and 12 h of milled. Foams were obtained using NaCl (60%v/v) as space-holder with a mean particle size of 35 m. Powders and space-holder were mixed and uniaxial pressing was performed at 420 MPa of compaction pressure. Green compacts were introduced in distilled water at 60�C for remove NaCl. Four cycles of 2h each were applied to remove NaCl particles. Finally samples were sintered at 1300�C for 3 h in Ar atmosphere. Scanning electron microscopy and porosity analysis were carried out. Figure 1 shows the morphology of Ti2Nb-4Ta-8Zr alloy powders milled at 4 and 12 h. Figure 1(a) shows the agglomerated particles with sizes between ~ 3 to 15 m with an irregular morphology and the figure 1 (b) shows a size between 5 to 30 m, this size could be due to cold weld. Figure 2 shows the foams where there are two types of pores. The distribution of big and small pores is homogeneous and the shape is irregular. The figure 2(a) shows the pores produced by space holder have sizes between 100 to 200 m during 4 h of milled and the pores produced by 12 h of milled with a sizes smaller than 5m are showed in figure 2b. The morphology of pores may fall in two categories: (1) spherical shape (2) shape based on the space holder materials (Mostly in irregular shape) These results shows that is possible to synthetize Ti2Nb-4Ta-8Zr foams using NaCl. Foams present a homogeneous distribution of big pores and small pores with shape irregular. The pores produced by 4 h of milled have sizes between 100 to 200 m and the pores produced by 12 h of milled process with a size smaller than 20 m and finally it is important to tell that this method of synthesis can be applied to synthesize different foams of biomaterials. Microsc. Microanal. 21 (Suppl 3), 2015 1044 References [1] Torres Y., et al., Metallurgical and Materials Transactions B Volume 42 (2011), p. 891 [2] Nadakuduru V., N. et al., Materials Science and Engineering A Volume 528 (2011), p. 4592. [3] Xiaopeng W, et al., Journal of the Mechanical Behavior of Biomedical Materials Volume 4 (2011), p. 2074. [4] Zhou Y L, et al., Materials Science and Engineering A Volume 371 (2004), p. 283. [5] Torres Y, et al., Journal of Materials Processing Technology Volume 212 (2012), p. 1061. Figure 2.SEM images of Ti-2Nb-4Ta-8Zr powder milled (a) 4h and (b) 12h. Figure 4. SEM images of foams formed (a) 4 h and (b) 12 h.");sQ1[523]=new Array("../7337/1045.pdf","Synthesis of Ti-Nb-Ta-Mn Alloys by Mechanical Alloying","","1045 doi:10.1017/S1431927615006029 Paper No. 0523 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Synthesis of Ti-Nb-Ta-Mn Alloys by Mechanical Alloying C. Aguilar1, A. Medina2a, L. B�jar2b, D. Guzm�n3, S. Lascano1, H. Carre�n2a, I. Alfonso4 1 Depertamento de Ingenier�a Metal�rgica y Materiales. Universidad T�cnica Federico Santa Mar�a. Av. Espa�a 1680, Valpara�so, Chile. 2a Instituto de Investigaciones Metal�rgicas, 2b Facultad de Ingenier�a Mec�nica, Universidad Michoacana de San Nicol�s de Hidalgo, Morelia, Michoac�n, M�xico, 58000 3 Departamento de Metalurgia, Universidad de Atacama, Av. Espa�a 485, Copiap�, Chile 4 UNAM, Campus Morelia, Antigua carretera a P�tzcuaro 8701, Morelia, Mich., M�xico. 58190. Due to the technical advancement, the average age and life expectancy of human beings have been continuously increase, this may produce degenerative diseases, which leads to degradation of the mechanical properties of the bone. Metallic materials are widely used as implants under load-bearing conditions [1,2]. The 316L stainless steels, Co-Cr and Co-Cr-Mo alloys have some disadvantages like high elastic modulus (E) (around 200 GPa, which is very high compared with E of bones, which is around 2 to 30 GPa) [3]. Ti-based alloys are the beneficial materials to re-place the damaged bone due to their excellent mechanical properties, corrosion resistance and biocompatibility. This work studies the synthesis of Ti-30Nb-13Ta-6Mn placed in 250 ml hardened steel vial and milled in argon atmosphere through planetary mill. Also 2 wt. % of stearic acid was added as lubricant in the milling process. The powders were milled for 4 and 50 h. Further the Ti-Nb-Ta-Mn powders synthetized have been characterized by X-ray diffraction, scanning and transmission electron microscopy. Figure 1 shows the morphology of Ti-30Nb-13Ta-6Mn alloy milled powders at 4 and 50 h. Figure 1(a) shows the agglomerated particles with equi-axial morphology with maximum size of ~60 m and figure 1(b) shows an irregular morphology with a size 40 m for 4 and 50 h of milling respectively. Figure 2 shows the XRD pattern of Ti-30Nb-13Ta-6Mn powders milled for 4 and 50h. The XRD pattern reveals that there is no formation of intermetallic compounds between Ti, Nb, Ta and Mn. After 50 hrs of milling, the strongest Nb, Ta and Mn peaks were disappears due at the formation of a solid solution with Ti. It is possible to observe that there are some peaks of Ti and Mn present inside the amorphous hump. The 4 h of milled powders of all zones exhibits the presence of crystalline phases without amorphous phases as showed in figure 3(a). By other hand at 50 h of milling, the TEM micrograph clearly shows the mixing of amorphous phases and crystalline phases as showed in figure 3(b). The results indicate that the Ti-30Nb-13Ta-6Mn alloy formed through mechanical alloying. The energy given by mechanical alloying process was sufficient to increase the solubility between Ti, Nb, Ta and Mn. The increased surface energy (due to the reduction of crystallite size) and elastic strain energy (due to the increase of dislocations density) were responsible for the increase in solid solubility. At 50h of milling the Ti-30Nb13Ta-6Mn alloy is a mixing between crystalline and amorphous phases. References [1] Niinomi, M. Science and Technology of Advanced Materials, Volume 4 (2003), p. 445. [2] Xiong, J., et al. Acta Biomaterialia, Volume 4 (2008), p. 1963. [3] Nomura, N., et al. Materials Science and Engineering: C, Volume 25 (2005), p. 330. Microsc. Microanal. 21 (Suppl 3), 2015 1046 Figure 1. SEM images of Ti-30Nb-13Ta-6Mn powder milled (a) 4h and (b) 50 h. Figure 2. XRD pattern of Ti-30Nb-13Ta-6Mn milled at 4 and 50 h. Figure 3. TEM images of Ti-30Nb-13Ta-6Mn powders milled at (a) 4 and (b) 50 h.");sQ1[524]=new Array("../7337/1047.pdf","Microstructural Investigation of Multi-state Resistive Switching Characteristics in Multi-layered Platinum/Tantalum Oxide using In-situ TEM","","1047 doi:10.1017/S1431927615006030 Paper No. 0524 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microstructural Investigation of Multi-state Resistive Switching Characteristics in Multi-layered Platinum/Tantalum Oxide using In-situ TEM Seong-Il Kim, Seung-Pyo Hong and Young-Woon Kim Research Institute of Advanced Materials, Department of Materials Science and Engineering, Seoul National University, Seoul, Korea Resistive switching (RS) characteristics of transition metal oxides have been attracted great attention for the next generation non-volatile memory, called resistive random access memory (ReRAM) [1]. Tantalum oxide (TaOx), one of the most widely studied RS systems, has been reported to show both unipolar and bipolar RS behaviors with multi-state RS characteristics [2-3]. In general, multi-level resistance states is referred when it shows more than 3 distinctive states, for example, low resistance state (LRS), high resistance state (HRS) and states in between them. The bipolar multi-state RS behavior of TaOx has been reported by Hur et al [3]. The unipolar multi-state RS characteristics, on the other hand, are rarely reported. The unipolar multi-state RS characteristics in multi-layered Pt/TaOx will be reported in this presentation. The multi-state RS can provide a simple way to develop multi-bit devices in nonvolatile memory storage systems by simple process of forming alternate layers of Pt and TaOx. A 40-nm-thick TaOx thin film was deposited onto the e-beam-evaporated 200-nm-thick Pt/Ti/SiN/Si substrates by reactive DC magnetron sputtering at room temperature. Subsequently, a 40-nm-thick Pt layer was sputter-deposited onto the former TaOx thin film with same condition above. One more set of TaOx and Pt layers were repeated on the top of the pre-layered Pt/TaOx to make Pt-top-electrode (TE)/TaOx/Pt-middle-electrode (ME)/TaOx/Pt-bottom-electrode (BE) thin film. All depositions were performed in vacuum without exposing to air and retractable patterned metal mask was applied on the substrate during the depositions. By carefully controlling the level of the sweeping voltage, multi-state RS characteristics were obtained from current-voltage (I-V) measurement in the probe station. Microstructural and elemental changes were investigated by using transmission electron microscopy (TEM) equipped with energy dispersive spectroscopy (EDS) and electron energy loss spectroscopy (EELS). In the typical semi- and double-logarithmic plots of the I-V curve of Pt/TaOx/Pt/TaOx/Pt structure, a pristine state showed LRS in the sample, unlike other ReRAM devices. The electro-forming process, which required initiating the repeatable RS with a compliance current (100), was not necessary in this performance. Reset and set process occurred with same bias polarity at ~ � 0.9 V and ~ � 3.2 V, respectively. With coarse controlling the sweeping voltage, two major resistance states of LRS and HRS were observed, as can be seen in Figure 1, and there showed additional resistance states, which represent resistance ratio of more than 10, were obtained, when the sweeping voltage was manipulated in between voltage. It was believed that RS behavior could be originated from the formation of conducting path through the metal oxide or near the metal/metal oxide interface. Thus, several different conducting paths from each layer or interface in the stacked configuration could explain the multi-state RS. The pristine state of the alternate stacked structure, which was confirmed to be LRS through TaOx, was observed by TEM. According to TEM and EDS investigation in Figure 2, the device was manufactured as planned and amorphous TaOx was formed. Microstructural investigation was made using in-situ probing stage to identify the multi-state RS mechanism. Sequential change of microstructure was recorded and compared, which will be presented in the talk as a movie clip. Microsc. Microanal. 21 (Suppl 3), 2015 1048 References: [1] H. -S. P. Wong et al, Proceeding of the IEEE 100 (2012) p. 1951. [2] F. Kurnia et al, Phys. Status Solidi RRL 5 (2011) p. 253. [3] J. H. Hur et al, Nanotechnology 23 (2012) 225702. [4] This research was supported by the Nano�Material Technology Development Program through the National Research Foundation of Korea funded by the Ministry of Science, ICT & Future Planning (2011-0019984). Figure 1. The semi- and double-logarithmic plots of I-V curve: (a) Unipolar RS behavior of Pt/TaOx/Pt/TaOx/Pt sample. (b) Multiple resistance states of Pt/TaOx/Pt/TaOx/Pt sample. Figure 2. (a-b) TEM images of the pristine Pt/TaOx/Pt/TaOx/Pt sample, current path illustrated during applying the voltage. (c-g) High angle annular dark field (HAADF) image and EDS elemental mapping of (d) titanium, (e) platinum, (f) tantalum, and (g) oxygen.");sQ1[525]=new Array("../7337/1049.pdf","Generating Large Magnetic Field in a High Resolution Electron Beam Lithography","","1049 doi:10.1017/S1431927615006042 Paper No. 0525 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Generating Large Magnetic Field in a High Resolution Electron Beam Lithography C.H. Wan1, P. Guo1, Q. T. Zhang1, A. Rudzinski2, T. Kliem2, R. Jede2, X. Y. Sun3, X.F. Han1,* 1. Institute of Physics, Chinese Academy of Sciences, Beijing National Laboratory of Condensed Matter Physics, Beijing, 100190, Peoples Republic of China 2. Raith GmbH, Dortmund, D-44263, Germany 3. GermanTech Co, Ltd, Beijing, 100190, Peoples Republic of China Since discovery of giant magnetoresistance (GMR) [1, 2], spintronics have achieved great progress not only in laboratory research but also in commercial applications two typical examples of which are GMR reader [3] and magnetic random access memory (MRAM) based on tunneling magnetoresistance (TMR) effect [4]. Due to small sizes in three dimensions of the elementary device, it requires accurate lithographic technique, typically electron beam lithography (EBL) to pattern raw films into desired shape as well as to make electrodes with much larger sizes which occupy large area of a wafer and only function as electrical pads during subsequent measurement process. In this talk we will report about first results and measurements achieved with an electron beam lithographic system (EBL, eLINE Plus CAS) containing a unique in situ setup to not only fabricate but also characterize and measure GMR/TMR based devices. The eLINE Plus CAS is additionally equipped with the capability of generating a large lateral magnetic field, which is indispensible in measurement of spintronic devices or materials, besides of its original SEM and EBL functionalities. This unique EBL instrument, thus, could be not only used as a microscope or a lithography system but also as a analytical instrument for in-situ magnetoelectric transport measurements. Figure 1 shows the structure of the magnet in chamber whose pole tips are placed underneath pole piece of SEM column. The maximum lateral field realized at current stage is about 400 Oe as exciting current is elevated as 2.5 A. This large field has already been applied in in-situ measurement GMR or TMR devices. As shown in Figure 2, utilizing the magnet as well as the other two tungsten probers already installed in the chamber, we have measured magnetotransport properties of a TMR device whose TMR ratio is nearly the same as that measured outside chamber. We will further report about the characterization of the homogeneity of the magnetic field and next steps taken in this project. It is worthy of accentuating that magnetic field should be switched off as the SEM is used and remanence of the magnet is too small to disturb ordinary operation of the SEM and EBL whose microscopic resolution and lithographic resolution could still remain in the order of 3 nm and 10 nm, respectively. Combination of magnetic field with ordinary EBL or SEM not only endows them with capability of in-situ magnetoelectric measurement capability but also shed light on hybridization of SEM with magnetic force microscope or magneto-optical Kerr microscope, advancing the development of Microsc. Microanal. 21 (Suppl 3), 2015 1050 characterization techniques for magnetic materials or spintronic devices. Figure 1. Magnet prober and SEM polepiece arrangement within EBL system chamber Figure 2 Field dependence of a TMR device measured in EBL chamber. References [1] M. N. Baibich et al. Physical Review Letters, 61(1988), 2472. [2] P. Gr�nberg, et al. Physical Review Letters, 57 (1986), 2442. [3] . Fert. Reviews of Modern Physics, 80 (2008), 1517. [4] Y. Huai, AAPPS Bulletin, 18(2008), 33.");sQ1[526]=new Array("../7337/1051.pdf","Densification and Microhardness of Spark Plasma Sintered ZrB2+SiC Nano-Composites*","","1051 doi:10.1017/S1431927615006054 Paper No. 0526 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Densification and Microhardness of Spark Plasma Sintered ZrB2+SiC NanoComposites* Naidu V. Seetala1, Marquavious T. Webb1, Lawrence E. Matson2, HeeDong Lee3, Carmen M. Carney2, Thomas S. Key3 1. Department of Mathematics & Physics, Grambling State University, Grambling, LA 71245, USA Air Force Research Laboratory, Materials and Manufacturing Directorate, Wright-Patterson Air Force Base, Dayton, OH 45433, USA. 3. UES, Inc., Dayton, OH 45432, USA 2. Ultra-High-Temperature Ceramics (UHTCs) such as ZrB2 and HfB2 with incorporation of SiC nanofiller are useful as structural materials for applications in propulsion and thermal protection systems such as turbine-engine hot section components, leading edge of hypersonic vehicles, where extremely high heat fluxes generate very high temperatures and steep temperature gradients [1]. Spark plasma sintering (SPS) technique is used for densifying the UHTCs under the influence of uniaxial pressure and pulsed direct current [2]. Here, we study the densification, grain growth, and microhardness of ZrB2 nanocomposites with 15% and 20% SiC consolidated using SPS. Nano-powders of ZrB2 (43 nm hexagonal 99% pure) and SiC (40 nm cubic 99% pure) were obtained from US Research Nanomaterials, Inc, Houston, Tx. The nano-powders of ZrB2 with 15 or 20 vol% SiC were mixed in isopropanol using Thinky planetary mixer and ultra-sonication followed by drying in vacuum at 50�C. The Spark Plasma Sintering (SPS) is performed at 2,100 �C to 1,800 �C at 32-40 MPa at Wright-Patterson Air Force Base, Dayton, OH or at SPS NanoCeramics, LLC, Morton Grove, IL. We used SEM/EDXS for granular and elemental analysis, Archimedes method for density measurements, and microhardness tester for Vickers hardness. Figure 1 shows the SEM view graph of SPS consolidated ZrB2+15Vol%SiC nano-composite. The EDXS analysis revealed that though the SiC is distributed throughout the sample, the darker grains have higher SiC composition compared to the rest of the sample. The EDXS analysis also revealed that there is some oxygen content found on Zr rich whiter grains. The x-ray diffraction analysis (reported elsewhere) showed ZrO2 phase due to oxidation of some Zr during SPS consolidation, which is minimized by Argon-gas purged prior to SPS consolidation. Table 1 shows the densification, microhardness, grain size, and EDX elemental distribution of SPS consolidated ZrB2+15Vol%SiC and ZrB2+20Vol%SiC nano-composites at different temperatures and pressures. The density of the composites seems to be influenced by the content of low dense SiC and ZrO2 in the composites. Figure 2 shows the densification (compared to the theoretical value) dependence on the wt% elemental composition O/Zr ratio obtained from EDXS analysis. It indicates that the densification decreases with increasing ZrO2 contents. Figure 3 shows that the microhardness is inversely proportional to the pressure and directly proportional to the temperature applied during SPS consolidation. Fig. 4 shows that the average grain size obtained from the SEM viewgraphs of the SPS consolidated ZrB2+SiC nano-composites is proportional to both the pressure and the temperature used during the SPS consolidation. The results suggests that there is a need to optimize the SPS parameters to reduce the SiC segregation in order to obtain uniform distribution of SiC throughout the ZrB2+SiC composite, and eliminate Zr oxidation observed during SPS treatment. _______________ * This work is supported by the Air Force Contract FA8650-13-C-5800. Microsc. Microanal. 21 (Suppl 3), 2015 1052 Table 1: Density, microhardness, grain size, and EDX elemental distribution in SPS consolidated ZrB2+15Vol%SiC and ZrB2+20Vol%SiC nano-composites SiC Pressure Temp. Densification Hardness Grain Size EDX wt% (Vol%) (MPa) (�C) (%) (VHN) (�m) O/Zr 20 40 1950 88.2 1242 2.1 0.033 20 35 1850 86.2 1455 1.6 0.079 20 32 2100 87.2 2304 1.75 0.152 20 32 1800 75.9 1815 0.6 0.330 15 35 1850 89.7 1819 1.78 0.331 15 35 1850 92.2 1845 1.61 0.032 15 32 1800 75 2094 0.544 0.022 Figure 1: SEM of SPS consolidated ZrB2+15Vol%SiC nano-composite. Figure 2: Densification vs O/Zr EDX wt% ratio Figure 3: Hardness vs Pressure/Temperature. References: Figure 4: Average grain size vs P*T product. [1] W. G. Fahrenholtz, et al, J. Am. Ceram. Soc. 90 (2007) 1347-1364. [2] R. K. Enneti, et al, Int. J. Refractory Metals & Hard Matter. 31 (2012) 293-296.");sQ1[527]=new Array("../7337/1053.pdf","Metal/ceramic Interface Structures and Segregation Behavior in Aluminum-based Composites.","","1053 doi:10.1017/S1431927615006066 Paper No. 0527 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Metal/ceramic Interface Structures and Segregation Behavior in Aluminum-based Composites. Xinming Zhang1, Tao Hu1, Jorgen F. Rufner1, Thomas LaGrange2, Geoffrey H. Campbell2, Enrique J. Lavernia1, Julie M. Schoenung1, Klaus van Benthem1 1. Department of Chemical Engineering and Materials Science, University of California, Davis, CA 95616 2. Physical and Life Sciences Directory, Lawrence Livermore National Laboratory, 7000 East Ave, Livermore, CA 94550 Trimodal Al alloy (AA) matrix composites consisting of ultrafine-grained (UFG) and coarse-grained (CG) Al phases and micron-sized B4C ceramic reinforcement particles exhibit combinations of strength and ductility that render them useful for potential applications in the aerospace, defense and automotive industries. A critical requirement for the application of trimodal composites is the creation of a strong interfacial bond between ceramic reinforcement and metal matrix to allow for effective load transfer. Different interfacial structures, including presence of amorphous layers between B4C and the UFG Al phase [1�3], formation of nanocrystalline grains with diameters of 30-50 nm at the Al/B4C interface [1,2] and segregation of Mg to Al/B4C interfaces [3,4], have been reported in separate literatures. However information on the relationship between different interface morphologies (e.g., amorphous layers, nanocrystalline grains, etc.) and their chemical composition profiles remains absent in the literature. The composite powder was fabricated via mixing cryomilling [5] and unmilled powder together and consists of 10 wt.% B4C, 30 wt.% CG 5083 AA, and the balance of UFG 5083 AA. The consolidation of mixed powder was achieved via hot isostatic pressing. Electron-transparent TEM samples were prepared by focused ion beam (FIB) sectioning. TEM imaging was performed with either a JEOL 2500 or a FEI F20 UT Tecnai. Electron Energy Loss Spectroscopy (EELS) was done with either a JEOL 2100 or a F20 UT Tecnai. Elemental distribution maps were acquired using an FEI Super-X windowless EDXS detector (FEI Company, Hillsboro, OR) installed on the Titan G2 80-200 instrument. All microscopes are operated at 200 kV. Figure 1a is a bright field TEM image demonstrating a layered structure between a B4C particle and the CG Al phase. Figure 1b is a HRTEM image of the same interface that reveals a double-layered interface consisting of an amorphous layer and a Mg-rich layer. Figure 2 shows integrated EELS intensity line profiles across the Al/B4C interfaces as a function of relative distance. Chemical composition profiling suggests the existence of Al oxide in conjunction with B4C, while single or multiple layers of Mg-rich oxide are observed in between Al oxide and the alloy matrix. HRTEM and EELS experiments revealed an ultra-thin amorphous aluminum oxide layer separating a poly-crystalline MgO layer from the B4C dispersoids. Precession assisted electron diffraction mapping was performed for the UFG Al/B4C interface and revealed no preferred orientation of the grains in the UFG region with respect to the B4C particle. Hence, grain orientation or texturing did not associate with the segregation of Mg at these interfaces. References: [1] [2] [3] [4] Li Y et al, Mater Sci Eng A 527 (2009), p. 305. Yao B et al, Compos Part Appl Sci Manuf 41 (2010), p. 933. Li Y et al, Acta Mater 59 (2011), p. 7206. Li Y et al, Acta Mater 58 (2010), p. 1732. Microsc. Microanal. 21 (Suppl 3), 2015 1054 [5] Witkin DB, Lavernia EJ. Prog Mater Sci 51 (2006), p. 1. Figure 1. (a) Bright field TEM image of the interface between the B4C particle and the Al matrix showing the existence of Mg segregation layer; (b) HRTEM image showing a double-layered interface consists of amorphous layer and Mg-rich layer. Figure 2. Integrated EELS intensity line profiles across (a) the CG Al/B4C interface and (b) the UFG Al/B4C interface.");sQ1[528]=new Array("../7337/1055.pdf","Preliminary Investigations of Chemical & Morphological Inhomogeneities in La0.6 Sr0.4CoO3- Single-Crystalline Perovskite Thin Films by ACTEM and STEM-EELS","","1055 doi:10.1017/S1431927615006078 Paper No. 0528 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Preliminary Investigations of Chemical & Morphological Inhomogeneities in La0.6 Sr0.4CoO3- Single-Crystalline Perovskite Thin Films by ACTEM and STEM-EELS Burcu �g�t1, Michael L. Machala1, Ai Leen Koh2, Robert Sinclair1, William C. Chueh1 1 2 Department of Materials Science and Engineering, Stanford University, Stanford, CA 94305-4034 USA Stanford Nano Shared Facilities, Stanford University, Stanford, CA 94305-4045, USA Perovskite-structured materials are employed as cathodes for carrying out the oxygen reduction reaction in solid oxide fuel cells (SOFCs) [1]. Among them, La0.6Sr0.4CoO3- (LSC) exhibits high electronic conductivity and oxygen diffusivity. At the same time, however, LSC suffers from chemical stability issues, particularly the exolution of SrO species from the bulk to the surface, and decomposition to ternary and binary oxides [1,2]. Well-ordered, single-crystalline thin films are ideal model systems to investigate the nanoscale variations in chemistry and microstructure in LSC [3,4]. In this work, dense ~250-nm-thick LSC was deposited using pulsed-laser deposition [5] on a (100)oriented yttrium-stabilized zirconia (YSZ) single-crystalline substrate, with a 35-nm-thick samarium doped ceria (SDC) buffer layer using KrF laser in 5 mTorr of oxygen. During deposition, the substrate temperature was 700 oC, and fluency was 1.5 Jcm-2 keeping the repetition rate as 5 Hz. Substance-totarget distance was kept as 80 mm. After the deposition process, the material system has been annealed at 550oC in 1 torr of O2 atmosphere for 6 hours and then quenched. We apply aberration-corrected transmission electron microscopy (ACTEM) and electron energy loss spectroscopy in scanning transmission electron microscopy mode (STEM-EELS) in an FEI Titan 80-300 environmental TEM at an accelerating voltage of 300 kV. TEM samples were prepared by tripod polishing followed by Ar ion milling. ACTEM experiments were performed using negative Cs imaging conditions (-11 �m and overfocus), which allows light elements to be visualized with bright contrast [6]. STEM-EELS line spectra were acquired using convergence and collection angles of 9.3 mrad and 18.7 mrad, respectively. Figure 1a shows a TEM bright field (BF) image of the YSZ/SDC/LSC system. Figure 1b shows a higher magnification image of Figure 1a inset, revealing the single crystalline nature of LSC. Despite the single crystallinity, significant roughness was observed on the surface of LSC (figure 1c). Within these surface features, channel-like domains exist, where the structure deviates from single crystallinity (Figure 1d). Figures 2a to 2c display the annular dark field STEM (ADF-STEM) images of three representative area and the points where the spectra in Figures 2d to 2f were acquired. The EELS spectra are background subtracted by curve fitting. The energy-loss near-edge structures (ELNES) of La-M4,5 and Co-L2,3 ionization edges vary in intensity going from SDC/LSC layer with either Co or La appear to be deficient toward the surface in Fig. 2d and 2e, respectively. In Figure 2f, EELS spectra seem to reveal that La-L and Co-M edge intensities are approximately constant along the area outside of the channel. Further investigation on 17 different channels revealed similar behavior to Figure 2d and 2e. In summary, our investigations of LSC reveal the existence of channel-like structures that underlies surface morphological features. The EEL spectra show how the chemical composition likely varies in the channel-like structures. Additional experiments will be performed on plan-view and cross-sectional samples to understand the thermodynamic phenomenon leading to the occurrence of these channels, and the surface properties of the LSC film. Microsc. Microanal. 21 (Suppl 3), 2015 1056 References: [1] J. Janutschewsky et al, Adv. Func. Mater. 19, (2009), p. 3151-3156. [2] Z. Cai et al, Chem. Mater., 24, (2012), p. 1116-1127. [3] L. Dieterle et. al., Adv. Energy Mater., 1, (2011), p. 249-258. [4] N. Tsvetkov et al, J. Mater. Chem. A, 2, (2014), p. 14690. [5] D.N. Mueller et. al. Nature Comm., 6, (2015), p. 1-8. [6] C. L.Jia et al., Science, 299, (2003), p. 870-873. Figure 1. (a) TEM Bright field image showing yttrium stabilized zirconia (YSZ), samarium doped ceria (SDC) and La0.6Sr0.4CoO3- (LSC) layers. (b) ACTEM image of the Figure 1a inset showing the single crystalline nature of LSC in a [110] projection. (c) BF image of another region containing the channellike regions. (d) ACTEM image acquired from the region marked in figure 1c where the atomic structure becomes disordered. Figure 2. (a to c) Annular dark field STEM (ADF-STEM) images of the YSZ/SDC/LSC material system and the marked points where corresponding STEM-EELS spectra in figure 2d to 2f were acquired. (d to f) STEM-EELS spectra acquired from the points marked from 1 to 16 (figure a to c) showing the intensities of Co-M4,5, and La-L2,3 ionization edges for comparison (d,e) inside and (f) outside the channels.");sQ1[529]=new Array("../7337/1057.pdf","Low Dose Electron Microscopy of Interlayer Expanded Molybdenum Disulfide Nanocomposites.","","1057 doi:10.1017/S143192761500608X Paper No. 0529 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Low Dose Electron Microscopy of Interlayer Expanded Molybdenum Disulfide Nanocomposites. Hector A. Calderon1, Yanliang Liang2, Hyun Deog Yoo2, Yifei Li2, F. C. Hernandez3, C. Kisielowski4, Yan Yao2,5 1. 2. Departamento de F�sica, ESFM-IPN, Ed. 9 UPALM-Zacatenco, M�xico D.F. 07738, Mexico. Department of Electrical and Computer Engineering, University of Houston, Texas 77204, USA. 3. College of Technology, University of Houston, Houston, Texas 77204, USA. 4. Molecular Foundry and JCAP, Lawrence Berkeley National Laboratory, Berkeley, CA 94720, USA 5. Texas Center for Superconductivity,University of Houston, Houston, Texas 77204, USA. Corresponding authors: H. A. C. (email: hcalder@esfm.ipn.mx) and Y. Y. (email: yyao4@uh.edu) Mg rechargeable batteries represent a safe and high-energy battery technology. However there is a lack of suitable cathode materials due to the slow solid-state diffusion of the highly polarizing divalent Mg ion. Recently, a different method has been proposed [1] i.e. interlayer expansion as a general and effective atomic-level lattice engineering approach to transform inactive intercalation hosts into efficient Mg storage materials without introducing adverse side effects. As for this work, the corresponding characterization using electron microscopy is reported. Transmission electron microscopy has been performed in the TEAM 05 microscope (NCEM-LBNL) in conditions of low dose rate in TEM mode together with a routine in MacTempas � in order to apply the exit wave reconstruction procedure with 40 experimental images [2]. The samples are prepared by inserting a controlled amount of PEO (polyethylene oxide) into the lattice of MoS2 in order to increase the interlayer distance. PEO is a rather beam sensitive substance that evaporates as the sample is observed in high dose rate. Four samples are prepared as described elsewhere [1]. Briefly, samples peo1- and peo2- have different MoS2: peo molar ratios i.e., Li0.16MoS2(PEO)0.49 and Li0.13MoS2(PEO)0.98, respectively. Sample com-MoS2 represents the commercially available compound and res-MoS2 is a control sample that was determined to be Li0.21MoS2(H2O)1.0. Figure 1 shows a typical experimental image of the peo1- sample taken with a dose rate of approximately 20 e-/�2s for a 1 s exposure. Some features are still visible and in order to recover all of the sample characteristics, an exit wave reconstruction procedure (EWR, MacTempas �) is performed with 40 experimental images that differ in the focussing condition (-20 nm to + 20 nm). The result for this series of images is given in Fig. 2 that shows the corresponding phase image. As can be seen, all important features are now visible i.e., layers of both the inorganic MoS2 compound and the polymer. Attempting to image this composite structure fails when the dose rate increases beyond 100 e-/A2s, the polymer becomes unstable and the sample changes in a rather short time. Figure 2 shows a summary of observations for the involved samples with the layer spacings that are indicated. The com-MoS2 sample shows the expected value for the layer spacing which is similar to the control sample (res-MoS2) i.e., 0.61 and 0.62 nm respectively. Interestingly, the interlayer distance in the peo1- and peo2- samples gives values rather close to the expected interlayer expansion, it varies from 1.22 nm (peo1-) to 1.40 nm (peo2-). This work is completed with image simulation and modelling of the observed samples. References: [1] Y. Liang, H. D. Yoo, Y. Li, S. Jing, H. A. Calderon, F. C. Robles, L. C. Grabow, Y. Yao. Nano Letters, under revision, February 2015. Microsc. Microanal. 21 (Suppl 3), 2015 1058 [2] Electron Microscopy was performed at the Molecular Foundry, which is supported by the Office of Science, Office of Basic Energy Sciences of the U.S. Department of Energy under Contract No.DEAC02-05CH11231. The research is partially supported by CONACYT (Proyecto FOINST. 75/2012, 148304 and 129207), IPN (COFAA, SIP). a b Figure 1. peo1-MoS2 sample. (a) Low dose experimental image. (b) Phase image after EWR procedure. Figure 2. Representative phase images of the investigated samples. (a) Commercial MoS2, (b) resMoS2, (c) peo1- and (d) peo2-MoS2, respectively.");sQ1[530]=new Array("../7337/1059.pdf","Structure evolution of Mo2C catalysts upon exposure to oxygen","","1059 doi:10.1017/S1431927615006091 Paper No. 0530 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Structure evolution of Mo2C catalysts upon exposure to oxygen Jacob Held1, Mark Sullivan1, Aditya Bhan1, K. Andre Mkhoyan1 1. Department of Chemical Engineering & Materials Science, University of Minnesota, Minneapolis, MN 55455. Molybdenum carbide (Mo2C) formulations have recently received renewed interest as catalysts that can be functionally tuned via surface termination alteration (e.g. Mo, C, or O) [1-4]. In this work, we demonstrate oxygen co-processing as a means of reversibly tuning catalyst acid site density in situ. Oxygen co-feed was observed to reversibly alter the dehydration rate of isopropyl alcohol by a factor of about 30 as shown in Figure 1a. Oxygen adatoms both suppressed metallic functionality and generated Br�nsted acidity. Nitrogen physisorption measurements in conjunction with XRD peak broadening showed that catalyst particles were composed of 2-5 nm crystallites with BET surface areas of 60 - 100 m2 gcat-1. However, catalyst X-ray diffractograms (Figure 1b) were inadequate for accurate catalyst phase determination due to the indistinguishability of the proposed hexagonal close-packed and orthorhombic Mo2C phases. Here we report on the evolution of the crystal structure of Mo2C catalysts using analytical scanning transmission electron microscopy (STEM). Crystal structures of the catalyst particles were analyzed by high-angle annular dark-field (HAADF) imaging using an aberration-corrected monochromatic FEI-Titan G2 60-300 equipped with a Gatan Enfinium EEL spectrometer and a Super-X EDX system. A representative sample of post-reaction Mo2C catalyst particles was mechanically crushed in an agate mortar and suspended in dimethylformamide. The resulting suspension was then sonicated to break up agglomerates, and was drop cast onto a holey/thin carbon film on a copper grid. HAADF-STEM imaging, shown in Figure 2, revealed that the catalyst particles consist of 1-3 nm crystallites with 1-2 nm pores visible throughout, which corroborates the XRD analyses. Analysis of high-resolution HAADF-STEM images of particles tilted to low-order zone axes were used to evaluate the structure and symmetry of individual crystallites, confirming the presence of orthorhombic Mo2C in post-reaction samples with lattice parameters that agreed well with neutron scattering analyses by reported by Parth� et al [5]. References: [1] P. Liu et al., J. Phys. Chem. B 110 (2006) 19418. [2] K.J. Leary et al., J. Catal. 101 (1986) 301. [3] S. K. Bej et al., Appl. Catal. A 264 (2004) 141. [4] H. Ren et al., ChemSusChem 6 (2013) 798. [5] E. Parth� et al., Acta Crystallogr. 16 (1963) 202. [6] J. S. Lee et al., Journal of Catalysis 112 (1988) 44. [7] This research was supported by MRSEC program of the National Science Foundation under Award Number DMR-1420013 and by the U.S. Department of Energy under award number no. DE-SC0008418. Microsc. Microanal. 21 (Suppl 3), 2015 1060 Figure 1. a) Propylene synthesis rate from IPA as a function of co-fed O2 partial pressure, showing a reversible 30 fold control of Br�nsted acid site density. b) X-ray diffractograms of Mo2C catalysts before and after reaction including reference diffractograms for orthorhombic -Mo2C [5] and FCC -Mo2C phases [6]. The positions of the highest intensity MoO2 and MoO3 peaks are also included. Figure 2. a) HAADF-STEM image of a post-reaction catalyst particle exhibiting visible porosity in the range of 1-2 nm. b) High magnification of boxed region of image in a) showing crystallinity, size of grains, and porous features. c) Schematic view down the [010] zone axis of orthorhombic Mo2C [5]. Mo atoms are in pink, while C atoms are in black. d) Representative high resolution HAADF-STEM image of [010] zone axis of the same sample. Two stacking faults on the (100) planes are marked with lines.");sQ1[531]=new Array("../7337/1061.pdf","TEM and EDX Studies on the Structural and Compositional Evolution of PtNi3 Concave Nanocubes","","1061 doi:10.1017/S1431927615006108 Paper No. 0531 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 TEM and EDX Studies on the Structural and Compositional Evolution of PtNi3 Concave Nanocubes Lihua Zhang1, Chengyu Wang2, and Jiye Fang2 1 Center for Functional Nanomaterials, Brookhaven National Laboratory, Upton, NY 11973, USA Department of Chemistry, State University of New York at Binghamton, Binghamton, New York 13902 USA 2 It has been demonstrated that the optical, biological, and chemical properties of metal nanoparticles are highly influenced by size, composition and morphology [1]. These systems are promising candidates for a wide variety of applications in catalysis, sensing, electronics, photonics, and medicine. Pt and its alloys have been receiving a great deal of attention because of their unique catalytic properties and a wide variety of economically driven applications. Recent work on Pt3Co nanocubes [2] and Pt3Ni nanoframes [3] showed enhanced catalytic activity with a strong dependence on the structure and composition of the nanocrystals. In this work we utilize high-resolution transmission electron microscopy (HRTEM) and electron dispersive X-ray (EDX) spectroscopy to study the structural and compositional evolution of PtNi3 concave nanocubes before and after annealing in order to understand the difference in their electrochemical properties. High-resolution TEM images of as prepared PtNi3 concave nanotubes were observed along three representatives zone axes [001], [110], and [111], and showed that the initial PtNi3 concave nanocubes are fcc nanocrystals(Fig.1 a-c), while after annealing most PtNi3 concave nanocubes become hollow nanocrystals(Fig.1 d-f). EDX elemental mapping results showed that in as prepared nanocubes, Pt forms a cubic frame and the Ni EDX map indicates that it forms a sphere located inside and outside of the Pt cubic frame (Fig.2 a-c). After annealing, it was found that Pt-rich frame still exists but Ni EDX map shows much less Ni than initial nanocubes, and the Pt/Ni ratio after annealing is much higher than in initial nanocubes (Fig.2 d-f). We conclude that the PtNi3 concave nanocubes developed from Ni-rich nanocrystals to Pt-rich nanoframes after annealing. The detail of the structure and composition evolution of PtNi3 nanocubes before and after annealing will be discussed. References [1] Y. Xia, Y. Xiong, B. Lim, S. E. Skrabalak, Angew. Chem. Int. Ed.2009, 48, 60�103. [2] Chenyu Wang, Cuikun Lin, Lihua Zhang, Zewei Quan, Kai Sun, Bo Zhao, Feng Wang, Nathan Porter, Yuxuan Wang, and Jiye Fang. Chemistry a European Journal. 2014, 20 (6), 17531759. [3] C. Chen, Y. Kang, Z. Huo, Z. Zhu, W. Huang, H. Xin, J. D. Snyder, D. Li, J. A. Herron, M. Mavrikakis, M. Chi, K. L. More, Y. Li, N. M. Markovic, G. A. Somorjai, P. Yang, V. R. Stamenkovic, Science, 2014, 343,1319. [4] The HRTEM and STEM/HAADF-EDS studies were carried out at the Center for Functional Nanomaterials, Brookhaven National Laboratory, which is supported by the US Department of Energy, Office of Basic Energy Sciences, under contract no. DE-SC0012704. Microsc. Microanal. 21 (Suppl 3), 2015 1062 (a) 001 (b) 110 (c) 111 (d) 001 (e) 110 (f) 111 Figure.1 Bright field images from as prepared concave nanocubes (a)-(c), and from annealed nanocubes (d)-(f) along orientation [001], [110], and [111] respectively. (a) (b) Ni Ni Pt Pt (c) (d) (e) Pt Ni (f) Figure.2 Elemental mapping and linescan of Ni and Pt from as prepared concave nanocube (a)-(c) and from annealed nanocube (d)-(f).");sQ1[532]=new Array("../7337/1063.pdf","Machine-Learning Aided Evolution Studies of Nano-composite Electrodes and Nano-particle Catalysts for Fuel Cell Applications","","1063 doi:10.1017/S143192761500611X Paper No. 0532 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Machine-Learning Aided Evolution Studies of Nano-composite Electrodes and Nano-particle Catalysts for Fuel Cell Applications David Rossouw1, Lidia E. Chinchilla1, Sagar Prabhudev1 Tyler Trefz2, Natalia Kremliakova2, and Gianluigi A. Botton1. 1 2 McMaster University, Department of Materials Science and Engineering, Hamilton, ON, Canada. Automobile Fuel Cell Cooperation, 9000 Glenlyon Parkway, Burnaby, BC V5J 5J8 Canada Automotive vehicles powered by proton exchange membrane fuel cells are approaching commercial viability but further improvements are necessary for a practical technology that can be mass-produced cost effectively. The bottleneck limiting their commercialization is the notorious cathodic cell reaction, called the oxygen reduction reaction (ORR), which is a multi-electron pathway reaction with inherently sluggish kinetics. Platinum (Pt) has been used as a catalyst to accelerate the ORR, but despite its exceptional catalytic activity, it is cost prohibitive for commercialization. The search for more affordable fuel cell cathode materials has focused on controlling the surface structure and composition of novel multi-metallic catalytic nanoparticles on high surface area support membranes. However, such nano-structured heterogeneous systems are notoriously challenging to characterize. In addition to difficulties associated with interpreting spatially and spectrally overlapping analytical signals, samples can be beam-sensitive, so care must be taken to limit the beam dose. Here, we overcome these challenges by using a machine learning technique, called independent component analysis (ICA) implemented in HyperSpy [1], applied to electron energy loss spectroscopy (EELS) signals obtained using a transmission electron microscope (TEM) [2]. We characterize a proposed fuel cell cathode material, promising improved corrosion resistance and durability, containing nano-particulate platinum on a NbOx-carbon hybrid support. We unambiguously separate spatially and spectrally overlapping platinum and niobium EELS signals then map their distribution in the electrode before and after fuel cell loading (Fig. 1). The separation and mapping of the various electrode components provides valuable insight into the role of the various material components in the fuel cell ORR. The ICA technique has also been used to study gold-platinum (AuPt) nano-particles which show great promise as durable catalysts in the ORR at the fuel cell cathode. Unlike Pt alloyed with 3d transition metals [3,4], AuPt nano-particles do not undergo leaching in the harsh chemical environment in the fuel cell. Recent studies have revealed that AuPt particles, less than 10 nm in size, exhibit properties notably different from AuPt bulk alloys. However, progress in understanding the relationship between nanoscale composition and enhanced electrocatalytic properties, has been hindered by the lack of advanced analytical techniques with both sufficiently high spatial and chemical specificity [5]. Here, ICA is used to separate the spectrally and spatially overlaying gold and platinum EELS signals acquired from Pt@Au (Pt clusters seeded on Au) nano-particles. These particles were annealed in-situ using a dedicated TEM sample holder and nanoscale phase transformations were tracked over the course of a thermal treatment. This in-situ study of nano-scale compositional and morphological changes in individual Pt@Au nanoparticles provides valuable insight towards the understanding of the enhanced surface reactivity of these nanoparticles [6]. Microsc. Microanal. 21 (Suppl 3), 2015 1064 References [1] www.hyperspy.org [2] F. de la Pe�a et al, Ultramicroscopy 111 (2011), p. 169. [3] S. Prabhudev et al, ACS Nano., 7 (2013), p. 6103. [4] G-z. Zhu et al, J. Phys. Chem. C., 118 (2014), p. 22111. [5] V. Petkov et al, Nano Lett., 12 (2012), p. 4289. [6] D.R. acknowledges support from the Royal Society's Newton International Fellowship scheme and F. de la Pe�a & P. Burdet for many useful discussions on the ICA technique. S. P acknowledges support from Cory Chiang in the synthesis of Au-Pt nanoparticles. GAB is grateful for funding from NSERC under the CaRPE-FC network and to AFCC for partially supporting this work. Figure 1. (a) TEM-EELS analysis of a fuel cell cathode material containing nano-particulate platinum and niobium oxide on a NbOx-carbon hybrid support. (b) The raw spectral data are noisy due to limited beam dose to avoid sample damage. (c) ICA finds three main components in the spectral mixture, belonging to platinum and niobium. 1065 Microsc. Microanal. 21 (Suppl 3), 2015 Figure 2. (a) TEM-EELS analysis of a AuPt nanoparticle ensemble. (b) Trace evidence of gold and platinum edges are present in the noisy raw EELS spectra. (c) ICA successfully separates the spectrally and spatially overlapping gold and platinum edges in the EELS signals present in the raw data. Microsc. Microanal. 21 (Suppl 3), 2015 1066");sQ1[533]=new Array("../7337/1067.pdf","Atomic-scale Characterization of Restructured PtCu Nanocubes","","1067 doi:10.1017/S1431927615006121 Paper No. 0533 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic-scale Characterization of Restructured PtCu Nanocubes Cecile S. Bonifacio1, Junjun Shan2, Franklin(Feng) Tao2, and Judith C. Yang1 1. 2. Department of Chemical and Petroleum Engineering, University of Pittsburgh, Pittsburgh, PA 15260 Department of Chemical & Petroleum Engineering and Department of Chemistry, University of Kansas, Lawrence, KS 66047 Bimetallic nanoparticles are of increasing interest due to the mutual influence of different neighboring atoms that leads to catalytic behavior different than that of a monometallic cluster[1]. The catalytic properties of these materials are dependent on the catalyst atoms at the edge and/or corner[2], i.e., surface of the nanoparticle. Motivated by previous studies [3,4] resulting to surface compositional change of bimetallic catalyst through metal segregation and reconstruction; post-synthesis reaction in the gas phase with ambient pressure x-ray photoelectron spectroscopy (AP-XPS) was used to tune the composition of two types of PtCu nanocubes(NCs) with different surfaces. Here we have used microscopy and spectroscopy techniques to provide direct evidence of atomic-scale elemental distributions within the NCs post-synthesis gas reactions. PtCu bimetallic NCs (8 nm in size flat shell or regular nanocubes (RNC) and 12 nm in size concave nanocubes (CNC)) supported in Al2O3 were drop-casted on holey carbon TEM gold grids. These grids were exposed to single and sequential gas reactions in H2 and H2-CO at 200 �C, respectively, in situ in an AP-XPS. A FEI Titan scanning transmission electron microscope (S/TEM) with a ChemiSTEMTM energy-dispersive x-ray spectrometer (EDS) system operated at 200 KeV was used to obtain the elemental distributions post-gas reactions of the NCs. The acquired EDS maps were compared with APXPS data and predicted models using periodic density functional theory (DFT) calculations of the resulting structure of the NCs before and after gas-reactions. Figure 1 and 2 shows the EDS line scans and associated DFT models before and after the reactions for the RNCs and CNCs, respectively. After reaction with H2 at 200 �C, the initial Pt skin of the RNC was retained but the structure was transformed with more Pt-Cu alloy distributed on the surface and some Cu below the Pt skin. This structure is similar to the DFT results of Cu at the subsurface layer for the RNCs(Figure 1c). The removal of the Pt skin, change from mesa-type to spherical shape, and redistribution of the Pt and Cu into a homogeneous alloy of the RNCs occurred after exposure to CO (Figure 1d). On the other hand, a significant change occurred to the CNCs after the single reaction with H2 at 200 �C with the formation of a 2 nm Pt shell and Pt-Cu alloy core (Figure 2c). Further reaction with CO resulted to an increased concentration of Cu at the subsurface of the CNC(Figure 2d). The increase in Cu signal at the subsurface concurs with the DFT prediction of Cu clusters/patch formation in the shell of the CNC after the sequential reaction in H2-CO. The atomic elemental distributions obtained through EDS showed that the resulting surface structure and composition of the NCs depend on the gas used in the reactions[5]. The surface segregation of the bimetallic NCs drove the observed surface reconstructions during post gas reactions. Further analysis of the EDS maps, AP-XPS and DFT calculations are underway to determine the process and driving forces of surface reconstruction at the atomic level for the flat and concave nanocubes [6]. References: Microsc. Microanal. 21 (Suppl 3), 2015 1068 [1] R. Ferrando, J. Jellinek, R.L. and Johnston, Chemical Reviews 109 (2008), p. 845 [2] J.A. Rodriguez, Surface Science Reports 24 (1996), p. 225 [3] F. Tao and M. Salmeron, Science 331 (2011), p. 171 [4] D. Hansgen et al, Nature Chemistry 2 (2010), p. 484 [5] J. Shan et al, Nature Communications (2015), Manuscript in preparation [6] CB and JY acknowledge financial support by DOE Basic Energy Sciences (DOE-BES) and Dr. Karen Bustillo (NCEM) for technical support. The microscopy work was done in the National Center for Electron Microscopy (NCEM), Molecular Foundry at Berkeley National Lab supported by the Office of Basic Energy Sciences of the US Department of Energy Contract No. DE-AC02-05CH11231. Figure 1. High angle annular darkfield (HAADF) image(a) of the RNC and EDS line scans across the NC before(b) and after reaction in H2 at 200 �C (c) and two sequential reactions in H2-CO (d). Gray and blue circles in the DFT models correspond to Pt and Cu atoms, respectively. Figure 2. High angle annular darkfield (HAADF) image(a) of the CNC and EDS line scans across the NC before(b) and after reaction in H2 at 200 �C (c) and two sequential reactions in H2-CO (d). Gray and blue circles in the DFT models correspond to Pt and Cu atoms, respectively.");sQ1[534]=new Array("../7337/1069.pdf","Characterization of Core/Shell Bi-Metallic Cube-Shaped Nanoparticles with Scanning Transmission Electron Microscopy","","1069 doi:10.1017/S1431927615006133 Paper No. 0534 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of Core/Shell Bi-Metallic Cube-Shaped Nanoparticles with Scanning Transmission Electron Microscopy Dalaver H. Anjum1, Akshaya Samal2, and Manuel A. Roldan-Gutierrez1 Imaging and Characterization Lab, King Abdullah University of Science & Technology (KAUST), Thuwal, Makkah 23955, Kingdom of Saudi Arabia (KSA) 2 KAUST Catalysis Center, King Abdullah University of Science & Technology (KAUST), Thuwal, Makkah 23955, Kingdom of Saudi Arabia (KSA) Bi-metallic nanoparticles (NPs) demonstrate superior catalytic performance as compared to their single metal counterparts because of "synergistic" effects. One of the most interesting bi-metallic NP systems is gold-palladium (Au-Pd) due to their wide-range of catalytic and energy applications [1]. Depending upon the synthesis conditions, the elemental distribution of these metals leads to different shape alloy NPs namely the core/shell, segregated, uniform mixing, and multi-shell [2]. In this report, we present a scanning transmission electron microscopy (STEM) analysis of core/shell cube-shaped Au-Pd NPs. These NPs are supposed to have the Pd metal in their cores which are then surrounded by Au and Pd metal shells. A TEM-instrument of model Titan G2 80-300CT S/TEM from FEI Company was employed and was used by setting the electron beam energy of 300 keV to accomplish the underlined analysis. At first, conventional high-angle annular dark-field STEM (HAADF-STEM) analysis was performed to identify the distribution of Au and Pd metals in these NPs. Second, electron tomography (ET) analysis, also in HAADF-STEM mode, was applied to determine structures of their cores and shells. Third, the spectrum imaging (SI) line profiles were acquired by combining the STEM with X-ray energy dispersive spectroscopy (EDS) techniques to confirm the presence of outer Pd shell on NPs since its thickness was of merely a couple of nanometers (nm). STEM and SI datasets were acquired in Digital Micrograph Software Package from Gatan, Inc. While the ET datastes were acquired in Xplore3D and were reconstructed in Inspect3D Packages both from FEI Company. The results from STEM and ET analyses of above mentioned Au-Pd NPs are shown in Fig. 1 (A-D). It can be noticed from the HAADF-STEM micrograph (Fig. 1A) that the NPs have cube shape morphology. Such HAADF-STEM micrographs were acquired at a relatively low camera length of 58 mm so that the Z-contrast dominates in acquired micrographs [3]. Hence, owing to this imaging condition, it is safe to state that the dark-color core region of about 50 nm in length is containing the Pd metal and this region is surrounded by bright-color 15 nm thick Au metal shell. A couple of more observations can also be made from the micrograph shown in Fig. 1 A. First, the presence of thin layer (~ 2 nm in length around the NPs "1" and "2") surrounding the Au shell and, second, the morphology of the Au shell seems to be of irregular shape. However, it cannot be exerted with full confidence that the 2 nm layer of Pd-metal is indeed present as it may be just due to the misalignments of NPs with respect to the electron beam. This is why these both questions were attempted with the single-axis ET analysis and the results (slices from the tomograms) are shown in Fig. 1B-D. ET analysis revealed that the Au-shells around Pd-cores are indeed of irregular shapes (Fig. 1D). Furthermore the Au-shells can be of either concave or convex morphology as seen in the middle slices of NP "2" and NPs "1" & "3" in Fig. 1C, respectively. However, the presence of the thin Pd-shells around the Au-shells could not 1 Microsc. Microanal. 21 (Suppl 3), 2015 1070 been verified with the ET analysis. This can be due to the "missing-cone" artifact in the single-axis tomography datasets [3]. On the other hand the SI analysis of these NPs, whose results are shown below in Fig. 2 (enclosed by a "red" box), revealed the presence of Pd-shell around Au-shell. It should be noted that the SI datasets were acquired with 1 nm pixel size and 10 s of dwell time at each pixel to acquire the EDS spectra. Then the EDS-signals from both Au and Pd metal were plotted and superimposed on the same NP. In conclusion, STEM is an excellent technique for characterizing the structure and morphology of core/shell NPs. Single-axis ET analysis of NPs has limitations in verifying the presence of less than a couple of nanometer thick layers around them. Although, it can be improved further by a factor of at least 10% with the acquisition of dual-axis ET datasets [3]. Nevertheless the best way possible for this type of characterization of core/shell NPs can be achieved by combining the ET and SI analyses in the form of 4-dimensional (4D) analysis in an S/TEM instrument. References: [1] R. G. Chaudhuri and S. Paria, Chemical Reviews, 112, 2373�2433 (2012) [2] D. Ferrer, A. Torres-Castro, X. Gao, S. Sepu�lveda-Guzma�n, U. Ortiz-Me�ndez, and M. Jose�Yacama�n, Nano Letters, 7 (6), 1701-1705 (2007) [3] P.A. Midgley and M. Weyland, Ultramicroscopy, 96, 413�431 (2003) Fig. 2 Fig. 1: STEM and ET analyses of core/shell Au-P NPs. (A): HAADF-STEM micrograph showing cube shape core/shell NPs. (B): Top-region of the tomogram, reconstructed with SIRIT routine, of the region in Fig. 1 (A) by tilting the specimen from -74� to +74� with an angle-increment of 1�. (C): Middle-region of the tomogram. (D): Bottom-region of the tomogram. Fig. 2: SI line-profile analysis of an Au-Pd NP acquired with STEM-EDS technique. HAADF-STEM micrograph of a NP is shown along with its corresponding Pd and Au line-profiles. Pd and Au EDSsignals are superimposed on the NP to elucidate the three core/shell structure of these cube shape NPs.");sQ1[535]=new Array("../7337/1071.pdf","Crystal Plane Effect of CeO2 in Metal-CeO2 Nanocatalysts for CO Oxidation","","1071 doi:10.1017/S1431927615006145 Paper No. 0535 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Crystal Plane Effect of CeO2 in Metal-CeO2 Nanocatalysts for CO Oxidation Samantha Mock and Ruigang Wang* Department of Chemistry and Materials Science and Engineering Program, Youngstown State University, Youngstown, OH 44555. Email*: rwang01@ysu.edu In oxide-supported metal heterogeneous catalysts, it has been recognized that metal catalysts supported on different oxides have different catalytic properties, depending on the reducibility of the oxide [1], the surface acidity and basicity of the oxide [2], and the metal-oxide interaction/reaction [3]. Charge transfer or redox reactions are essential steps for the catalytic reaction at elevated temperature, therefore the reducible oxide support presents a potential advantage compared to conventional non-reducible oxide support, e.g. silica and alumina. Reducible metal oxides are solid state compounds exhibiting variable valence or oxidation states of the metal, such as CeO2, TiO2, VOx, FeOx, CoOx, MnOx, HfOx, PrOx, TbOx, SmOx. Many transition and lanthanide metals possess variable oxidation states because of the occupancy of 3d and 4f orbitals, respectively. In this paper, we report a facile hydrothermal synthesis of CeO2 nanorods and nanocubes, and catalytic activity characterization of 10wt% Ni/CeO2 nanorods and nanocubes, aiming to understand the support crystal plane effect of CeO2 in metal-CeO2 nanocatalysts for low temperature CO oxidation. CeO2 nanorods and nanocubes were prepared using a hydrothermal method [4-5]. Typically 0.1M Ce(NO3)36H2O and 6M NaOH mixtures were heated to 90~210 �C and held for 48 hrs in a sealed 200 mL Teflon-lined autoclave (~50 % fill). Then the autoclave was cooled to room temperature before the solid products were recovered by suction filtration. The materials were washed thoroughly with distilled water to remove any co-precipitated salts, then washed with ethanol to avoid hard agglomeration in the nanoparticles, and dried in air at 50 �C for 12 hrs. Transmission electron microscopy (TEM) characterization was performed using a JEOL2100 operated at 200 kV and equipped with an EDAX detector and annular dark-field detector. Hydrogen temperature programmed reduction (H2-TPR) study was examined using hydrogen chemisorption on the Quantachrome iQ and Micrometrics 2920 to explore how much hydrogen adsorbs as a function of temperature. The catalytic oxidation of CO was conducted by using a fixed bed plug flow reactor system. 1vol%CO/20vol%O2/79vol%He with a 70 mL/min flow rate was supplied through mass flow controller and passed through the catalyst bed. The catalyst (~100 mg) was mixed with quartz wool (coarse, 9 m) and filled in the quartz tube reactor. The reaction temperature was programmed between room temperature and 350oC and monitored by thermocouple. The reactant CO and product CO2 were analyzed by using an on-line gas chromatograph (SRI multiple gas analyzer GC, 8610C chassis) system. Figure 1 (a) and (b) shows typical TEM images of CeO2 nanorods and nanocubes prepared by a hydrothermal method, respectively. Figure 1 (c) compares the Raman spectra of CeO2 nanorods and nanocubes, showing a higher concentration of oxygen vacancy in rods sample. Figure 1 (d) compares the shape (crystal plane) effect of CeO2 support on hydrogen consumption of 10wt%NiCeO2 nanorods and nanocubes. All metal-loaded samples show improved low-temperature activity, compared to pure CeO2 nanorods and nanocubes. When comparing the low temperature hydrogen consumption over the temperature range from room temperature to 350oC, Ni-CeO2 nanorods show higher hydrogen consumption compared to Ni-CeO2 nanocubes. This could be attributed to the Microsc. Microanal. 21 (Suppl 3), 2015 1072 interfacial anchoring effect of metals on different crystal planes on CeO2 with different shapes. We will present the atomic level interfacial structure and chemical composition of Ni-CeO2 nanorods and nanocubes using HRTEM, EDX and EELS in details. References [1] Wang, Y.G. et al, J. Am. Chem. Soc. 135 (2013) 10673. [2] Glazneva, T.S. et al, Kinetics and Catalysis 49 (2008) 859. [3] Haruta, M. et al, J. Catal. 115 (1989) 301. [4] Wang, R. et al, RSC Adv. 3 (2013) 19508. [5] Wang, R. et al, RSC Adv. 4 (2014) 3615. [6] This work is supported by American Chemical Society Petroleum Research Fund (#52323) and National Science Foundation (CHE-1362251). The use of TEM facilities at the Center of Excellence in Materials Science and Engineering at Youngstown State University are gratefully acknowledged. 5 0 n m 5 0 n m (a) CeO2 nanorods 0.10 0.08 0.06 0.04 (b) CeO2 nanocubes 215 C o m=99.9 mg H2 consumption Intensity (a.u.) 0.02 0.00 -0.02 0.10 0.08 0.06 0.04 764 C (b) 10% Ni-CeO2 nanocubes o 210 C o m=96.5 mg CeO2 rods CeO2 cubes 200 400 600 800 1000 -1 762 C (a) 10% Ni-CeO2 nanorods o 0.02 0.00 1200 1400 -0.02 0 200 400 600 o 800 1000 Wave Number (cm ) Temperature ( C) (c) Raman spectra (d) H2-TPR Figure 1 TEM images (a)/(b) (scale bar: 50 nm) and Raman spectra (c) of CeO2 nanorods and nanocubes, and H2-TPR profiles (d) of 10wt% Ni/CeO2 nanorods and nanocubes under a 5%H2/95%Ar gas atmosphere.");sQ1[536]=new Array("../7337/1073.pdf","Democratizing the Micro-Scale: A Simplified, Miniaturized SEM for K-12 and Informal Student Scientists","","1073 doi:10.1017/S1431927615006157 Paper No. 0536 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Democratizing the Micro-Scale: A Simplified, Miniaturized SEM for K-12 and Informal Student Scientists Lindsey Own1, Shay� Whitmer1, Lawrence Own2,3, and Christopher S Own3 1. The Evergreen School, Shoreline, WA 2. HiveBio, Seattle, WA 3. Voxa, Shoreline, WA Learners in K-12 schools and informal learning environments have historically had significantly less access to critical technologies than practicing scientists. They can observe pollen grains in basic light microscopes while studying plants, but cannot observe the surface structures that affect the pollens' dispersal. They can learn about structure-function relationships in insect anatomy, but not observe mechanosensory bristles on the eye of a fly they caught in their own classroom or learning space, nor the nanostructures enabling moths to hang upside-down on the ceiling. While today's student scientists have access to a great wealth of micro- and nano-scale images via the internet, they have so far been unable to take those images themselves. For K-12 student scientists and informal learners, direct observation of a great range of scientific phenomena has been impossible. Using a new scanning electron microscope (SEM) called MochiiTM developed at Voxa in Seattle, WA, specifically designed for simplicity of use by non-specialists, students at The Evergreen School and at HiveBio Community Lab have begun directly imaging their own samples to personally observe scientific phenomena previously inaccessible via standard classroom light microscopy. (Fig. 1 & 2) Having a high-resolution microscope in the classroom or informal learning environment permits much deeper exploration on the part of the students, rather than simply being passively presented with information and evidence from other sources. Working with this new technology is both highly engaging and accesses deep learning mechanisms through active learning. As more and more evidence is presented that active learning results in deeper conceptual understanding for learners[1], opportunities such as directly imaging one's own collected and prepared specimens become clearly beneficial to improving learning outcomes. Education researchers have further found that using real-world examples, and finding direct relevance and influence of the concepts they study, leads to greater learning outcomes and confidence for girls in their science education.[2] Personal preparation of students' own samples also leaves room for critical mistake-making and improving from those mistakes, as well as individual student-driven comparisons.[3] Pollen from different flowers, proboscises from different insects, root hairs from plants grown in different conditions demonstrate the great range of structures and strategies in the natural world that nature has optimized over millions of years of earth's history. A classroom SEM offers direct access to all of these structures for deeper science learning. At Microscopy and Microanalysis 2015, we will present examples of learning opportunities developed utilizing the Voxa portable SEM, as well as SEM images taken by students of samples prepared and loaded by those students. Microsc. Microanal. 21 (Suppl 3), 2015 1074 References: [1] Michael, Joel. Where's the evidence that active learning works? Advances in Physiology Education. 1 Dec 2006. 30:4. http://advan.physiology.org/content/30/4/159.short [2] Kulturel-Konak, Sadan et al. Review of Gender Differences in Learning Styles: Suggestions for STEM Education. Contemporary Issues in Education Research. Mar 2011. 4:3. http://www.cluteinstitute.com/ojs/index.php/CIER/article/view/4116/4171 [3] Mosatche, Harriet et al. Effective STEM Programs for Adolescent Girls. Afterschool Matters. Spring 2013. http://files.eric.ed.gov/fulltext/EJ1003839.pdf Figure 1. A 7-year-old student scientist exploring pollen grains in the MochiiTM via iPad interface. Figure 2. A MochiiTM image taken by the student scientist in Fig. 1.");sQ1[537]=new Array("../7337/1075.pdf","What Does Quantitative Mean In Atomic-Resolution EDS STEM?","","1075 doi:10.1017/S1431927615006169 Paper No. 0537 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 What Does Quantitative Mean In Atomic-Resolution EDS STEM? NR Lugg1, G Kothleitner2,3, B Feng1, N Shibata1 and Y Ikuhara1 1. 2. Institute of Engineering Innovation, School of Engineering, The University of Tokyo, Tokyo, JAPAN Institute for Electron Microscopy and Nanoanalysis, Graz University of Technology, Graz, AUSTRIA 3. Centre for Electron Microscopy, Graz, AUSTRIA Energy-dispersive x-ray spectroscopy (EDS) in scanning transmission electron microscopy (STEM) is quickly becoming ubiquitous in atomic-resolution chemical mapping [1-3] due to increased detector sizes and efficiencies. Much work has been done to try and bring lower resolution EDS mapping to a quantitative scale [4], by carefully calibrating measured x-ray counts so that they represent quantitative information about the specimen � the density of atoms in the area probed for instance. However, at atomic resolution, this is only the first "layer" of quantification. The majority of atomic-resolution EDS mapping to date has been qualitative � identifying in columns specific elements appear. This has been partially due to the lower efficiencies of the detectors available and the inability to obtain significant signal. However a more significant problem exists: the complex elastic and thermally scattering an atomically-sized probe undergoes when interacting with a crystal � i.e., electron channeling � means that the signal detected has a highly non-linear response to the density of atoms that are located under the probe [1,5]. Therefore, in general, even if the x-ray counts have been appropriately calibrated � i.e., we have achieved the first layer of quantification � one still cannot extract quantitative information about the specimen from the atomic-resolution maps acquired. We call this the second "layer" of quantification. In Fig. 1 we see simulated EDS STEM maps, calibrated so that the signal is in units of atoms/nm3. From these maps we can see that when the STEM probe is on-column, the measured "density" is much higher than the real density. Furthermore, if one considers the average density from the map, we also get a value much higher than the real value. This is due to channeling and, in this presentation, we discuss how this increase occurs in detail [6]. In high-angle annular dark-field (HAADF) STEM imaging, quantification has been achieved by comparing experimental intensities with simulation [7]. This quantification-by-comparison has also been achieved in atomic-resolution STEM electron energy-loss spectroscopy [8] (and by extension, it is possible in EDS STEM) � an impressive effort given the complexity of such calculations. Advanced deconvolution techniques have also been developed which remove the effects of channeling from chemical maps [9]; they are faster to implement then direct simulation and assume nothing about the ionization interactions involved giving them some advantage over direct comparison. However, the biggest drawback of all these techniques is that they requires one to have a model structure to begin with and the information is not directly extracted from the experimental data. Despite the success of the aforementioned techniques, the best-case scenario would be if one could obtain quantitative information directly from experimental maps. Here we explore how tilting the specimen may help in achieving this [6]. Although the idea of tilting the specimen to reduce channeling is not new, we show here that this rule-of-thumb holds for EDS STEM and that tilting the specimen does indeed reduce the effects of channeling. We show examples of how tilting allows one access to quantitative information using two test cases: single crystal strontium titanate (Fig. 2) and bicrystal yttrium stabilized zirconia with a low-angle grain boundary. The downside of tilting is that one loses atomic resolution in the direction tilted. To ameliorate this deleterious side effect of tilting, we consider Microsc. Microanal. 21 (Suppl 3), 2015 1076 a precession series of tilted images and show that one may retain atomic resolution while also obtaining quantitative information about the specimen. [1] AJ D'Alfonso, B Freitag, DO Klenov and LJ Allen, PRB 81 (2010), p. 100101. [2] PG Kotula, DO Klenov and HS Von Harrach, Microsc Microanal 18 (2012), p. 691. [3] G Kothleitner, MJ Neish, NR Lugg, SD Findlay, W Grogger, F Hofer, LJ Allen, PRL 112 (2014), p. 085501. [4] M Watanabe and D Williams, J Microsc 221 (2006), p. 89. [5] LJ Allen, AJ D'Alfonso, SD Findlay, JM Lebeau, NR Lugg and S Stemmer, J Phys: Conf Ser 241 (2010), p. 012061. [6] NR Lugg, G Kothleitner, N Shibtata and Y Ikuhara, Ultramicroscopy in press (2014), doi:10.1016/j.ultramic.2014.11.029. [7] J LeBeau, SD Findlay, LJ Allen and S Stemmer, PRL 100 (2008), p. 206101. [8] HL Xin, C Dwyer and DA Muller, Ultramicroscopy 139 (2014), p. 38. [9] NR Lugg, M Haruta, MJ Neish, SD Findlay, T Mizoguchi, K Kimoto and LJ Allen, APL 101 (2012), p. 183112. [10] The authors acknowledge funding from MEXT and the JSPS. Figure 1. Simulated EDS STEM maps of the Sr, Ti and O K lines in strontium titanate. Number in parentheses indicates actual density (atoms/nm3) for specific element. Numbers next to upper-left (green), upper-central (red) and central (cyan) pixels indicate "densities" measured for those probe positions. Number in lower-right corner indicates the "density" measure from averaging an entire map. An accelerating voltage of 200 kV, probe convergence angle of 22 mrad and thickness of 600 � have been assumed. Figure 2. Mean values of tilted EDS STEM maps for the Sr, Ti and O K lines in strontium titanate for a series of thickneses and tilt angles (along the <100> direction). Units are multiples of the actual density of the respective element. A sample tilted image (300 �, 3�) is shown next to each plot. An accelerating voltage of 200 kV and probe convergence angle of 22 mrad have been assumed.");sQ1[538]=new Array("../7337/1077.pdf","Channelling and atomic resolution STEM using X-ray emissions with absorption","","1077 doi:10.1017/S1431927615006170 Paper No. 0538 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Channelling and atomic resolution STEM using X-ray emissions with absorption C.J.Rossouw MCEM, Monash University, Victoria 3800, Australia. Although quantitative analysis using soft characteristic X-rays is beset by relatively high absorption effects, the concomitant skewing (depending on detector-specimen geometry) of the detected X-ray signal towards ionizations which have occurred closer to the surface may have advantage in retaining coherent information compared with an incoherent contribution. It is envisaged that strong absorption of soft X-rays may enhance coherent contrast in both lattice images and channelling patterns compared to that derived from more energetic X-rays, particularly with specimen-detector geometry at a grazing angle of detection to enhance this effect. A Bloch wave approach provides a framework in which both ionization localization and absorption of generated X-rays may be accounted for, and enables explicit separation of coherent and incoherent contrast. An initially unexpected outcome is an ability to quantify the degree of cross-talk between projected columns of atoms in lattice resolution mode. It turns out that the incoherent component is responsible for most of the cross-talk. Fig. 1 compares thickness-averaged signals from an aberration-free focused 200 keV coherent beam of 1.2 � resolution placed on an Au column in the <100> projection with a channelling response from a collimated beam. As the interaction becomes more delocalized, the variation in response is diminished, although this is somewhat masked by thermal smearing. Note that the signal is developed within the first 30 �, after which it decays rapidly with thickness. The build-up of an incoherent background is solely a reflection of the absorptive potential and is invariant with crystallographic site and interaction delocalization. Geometry-adjusted effective mean free paths for X-ray absorption are assumed to be 10, 0.5 and 0.1 m respectively for L, M and N excitations. <100> channelling patterns for L and N shell excitations as well as the N channelling ratio with a pair of X-ray detectors placed symmetrically above and below the specimen are shown in Fig. 2. The higher absorption of N shell excitations yields stronger top/bottom ratio contrast. Fig 3 shows calculated STEM lattice images for these excitations. A gedanken experiment allows isolation of a central column of atoms within the superlattice, and crosstalk is observed as the beam is scanned across neighbouring columns. With the coherent wavefunctioninjected onto atomic columns, the central maximum is rapidly attenuated at depths greater than 50 �, and a hole is formed at about 100 � thus eliminating almost entirely coherent ionization events at greater depths. This rapid and selective absorption of means the top few atoms alone yield coherent ionization intensity in Au. The ratio of contrast achieved by the top and bottom detectors is also shown in Fig. 3. Other materials with higher Debye temperatures and lower atomic numbers (such as Si or TiAl alloys) may allow coherent ionization effects to be monitored to greater depths, and this may also be helped by specimen cooling to inhibit incoherent scattering. It is my view that quantitative analysis of X-ray channelling patterns may ultimately prove more robust and indeed more useful than the pursuit of atomic resolution in STEM. For instance, the determination of an interstitial site of 1% Cr in mullite [1] would be well-nigh impossible by STEM, but was readily achieved by channeling effects. STEM is compromised in both real and reciprocal space, whereas channeling is not compromised in reciprocal space and has a resolution determined by itself. [1] C.J. Rossouw and P.R. Miller, American Mineralogist 84 (1999) 965-969. Microsc. Microanal. 21 (Suppl 3), 2015 1078 Fig. 1 Au L, M and N shell responses showing total and incoherent signals. The N shell X-rays are more strongly absorbed (see dashed lines). Left is a coherent beam focused onto an Au column and right is channeling response from a collimated beam (see text). Fig 2 L- and N-shell Au channelling patterns 200 � (top) and 500 � (bottom). Right, incoherent pattern and bottom, ratio of N emissions top/bottom due to variation in X-ray absorption (7 % contrast). Fig 3 M-shell STEM lattice images 200 �. Total, coherent, incoherent, total ratio top/bottom (25 % contrast) and wavefunction at 100 �. Note appearance of central hole in , log scale.");sQ1[539]=new Array("../7337/1079.pdf","Absolute-Scale Quantitative Energy Dispersive X-ray Analysis in Aberration-Corrected Scanning Transmission Electron Microscopy","","1079 doi:10.1017/S1431927615006182 Paper No. 0539 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Absolute-Scale Quantitative Energy Dispersive X-ray Analysis in AberrationCorrected Scanning Transmission Electron Microscopy Zhen Chen1, Adrian J. D'Alfonso2, Matthew Weyland3,4, Daniel J. Taplin1, Scott D. Findlay1 and Leslie J. Allen2 1. 2. School of Physics and Astronomy, Monash University, Victoria 3800, Australia School of Physics, University of Melbourne, Victoria 3010, Australia 3. Monash Centre for Electron Microscopy, Monash University, Victoria 3800, Australia 4. Department of Materials Engineering, Monash University, Victoria 3800, Australia Energy dispersive X-ray spectroscopy (EDX) has long been capable of determining elemental concentration ratios with a sensitivity of a few atomic percent at sub-micron resolution. This is achieved by comparison with reference specimens and taking signal ratios, thereby eliminating factors such as ionization cross section, fluorescence yield and detector geometry [1] that must otherwise be measured or calculated, and by avoiding low order zone axis orientations and the associated strong dynamical electron scattering ("channeling"). Atomic resolution EDX analysis in scanning transmission electron microscopy (STEM) has recently become possible [2-5] through improvements in aberration-corrected electron optics and X-ray detector design. However, channeling must be reckoned with in atomic resolution imaging since on-axis conditions are necessary for direct structure interpretation. Moreover, relative concentration may be less informative than the absolute number of atoms on this length scale. In this talk we will demonstrate absolute-scale, quantitative agreement between the numbers of X-ray counts in experiment and first-principles simulations in STEM EDX using an atomically-fine electron probe [6]. A SrTiO3 [001] test sample was used. The experiment was carried out using an aberrationcorrected FEI Titan3 electron microscope operating at 302 kV. A conventional Si(Li) detector was used, and obtaining good counting statistic necessitated averaging the EDX signal over several unit cells, forfeiting atomic-resolution information but gaining stability against noise and aberrations in the probeforming optics [7]. Simulations were carried out using the freely available STEM software package [8], incorporating calculation of the ionization cross-section and of channeling of the electron probe. For comparison with experiment it is further necessary to account for such factors as beam current, detector efficiency/geometry and fluorescence yields. Results are shown in figure 1 for a range of thicknesses and two different probe-forming aperture angles (15.2 mrad and 21.5 mrad). An independent measurement of sample thickness was essential. Position-averaged convergent beam electron diffraction (PACBED) has been used for precision thickness determination [7]. Figure 2 shows a comparison between experimental and simulated PACBED patterns showing good visual agreement. To make this more objective, an L2-norm measure was used to compare simulated and experimental PACBED patterns. The central plot in figure 2 shows that such plots have a clear global minimum, which proved to be in good agreement with the visual approach. However, to obtain clear features in the PACBED pattern, we used a probe-forming aperture of 9.2 mrad, smaller than that used for the EDX experiments. An alternative approach is to quantify the intensity in the annular dark field (ADF) image [9] recorded simultaneously with the EDX image. The results of this approach are seen to be in excellent agreement with the PACBED method in the right hand graph in figure 2 [10]. Microsc. Microanal. 21 (Suppl 3), 2015 1080 References: [1] JI Goldstein et al in "Principles of Analytical Electron Microscopy", ed. D.C. Joy et al, (Plenum Press, New York, 1986) p. 155. [2] AJ D'Alfonso et al, Phys. Rev. B 81 (2010), 100101(R). [3] M Chu et al, Phys. Rev. Lett. 104 (2010), 196101. [4] PG Kotula et al, Microscopy and Microanalysis 18 (2012), p. 691. [5] G Kothleitner et al, Phys. Rev. Lett. 112 (2014), 085501. [6] Z Chen et al, Submitted for publication (2014). [7] JM LeBeau et al, Ultramicroscopy 110 (2010), p. 118. [8] LJ Allen et al, Ultramicroscopy in press (2015) doi:10.1016/j.ultramic.2014.10.011. [9] JM LeBeau et al, Phys. Rev. Lett. 100 (2008), 206101. [10] The authors thank Dr N.J. Zaluzec, Argonne National Laboratory, and A. Sandborg, EDAX Inc., for assistance with the geometry of our EDX detector. This research was supported by the Australian Research Council under the Discovery Projects funding scheme (Projects DP110102228 and 140102538), its DECRA funding scheme (Project DE130100739) and its LIEF funding scheme (Project LE0454166). Figure 1. Absolute-scale comparison of X-ray counts between experiment and simulation for the Sr Kshell and Ti K-shell peaks for two different probe-forming aperture semiangles . Figure 2. Left: Visual comparison of PACBED pattern between experiment and simulation ( = 9.2 mrad) (area 9). Centre: L2-norm analysis automating PACBED thickness determination. Right: Comparison of thicknesses determined by PACBED and by quantitative ADF STEM.");sQ1[540]=new Array("../7337/1081.pdf","Theoretical Evaluation of Atomic-Resolution X-ray Analysis toward Quantification","","1081 doi:10.1017/S1431927615006194 Paper No. 0540 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Theoretical Evaluation Quantification of Atomic-Resolution X-ray Analysis toward M. Watanabe* * Dept of Materials Science and Engineering, Lehigh University, Bethlehem. PA 18015. The latest aberration-corrected scanning transmission electron microscopes (STEMs) in combination with the large solid-angle silicon-drift X-ray detectors (SDDs) improve limited efficiencies of signal generation and collections. It is no longer dream to acquire atomic resolution X-ray maps and to achieve single atom sensitivity in X-ray analysis by using the latest aberration-corrected instruments. Obvious next challenge is quantification of such atomic-resolution X-ray maps. There are several attempts to perform quantification of the atomic resolution X-ray maps [e.g. 1, 2]. However, quantified results are deviated from expected values from the structures. For example, Ga composition does not reach to 100% at Ga only columns in a [100]-projected GaAs compound [1]. Clearly, X-ray signals are generated from the specific atomic column where the incident electron is focused but also surrounding columns. For appropriate quantification, it is essential to determine how much X-ray signals are generated from the target and neighboring columns. In this study, X-ray signal generation in an oriented crystalline structure is simulated theoretically toward quantitative analysis. In order to simulate X-ray generation, first, it is required to know how the incident electron probe propagates in an oriented crystalline material. The electron propagation, called the electron wave function, can be simulated by multislice calculation. In this study, the xHREM code [3] was used to simulate the electron wave function. Figure 1 shows the simulated electron propagation at a Ga column in the [100]-projected GaAs. An image at the bottom in Fig. 1 indicates the electron distribution at one of sliced planes. The electron distribution is not homogeneous but strongly related to atomic arrangements, which are scattering sources. Once the electron distributions are simulated, X-ray signals can be simulated at each sliced plane. As shown in Fig. 2 (a schematic cross-section view of the GaAs structure), X-ray spectra were simulated at each sliced layer by porting the X-ray generation engine of legacy Desktop Spectrum Analyzer (DTSA) codes [4]. An X-ray spectrum at a certain depth from the top surface of the specimen was obtained by adding individual spectra. Since the DTSA codes were used, X-ray absorption is also taken care of. X-ray maps were constructed by repeating this process at various positions. In Fig. 3, a set of simulated results are summarized for different specimen thicknesses (10, 40, 80 and 160 unit-cell depths, corresponding to 5.6, 22.4, 44.8 and 89.6 nm, respectively): simulated high-angle annular dark-filed (HAADF) images in top row, simulated ratio maps of Ga K to total K intensities (Ga K and As K) in middle row, and normalized total K intensities in bottom row. Because X-ray generation and detection for the Ga K and As K line due to close atomic numbers of Ga (31) and As (33), the maps of Ga K/total intensity ratio approximately indicate the Ga concentration fraction. Although the individual atomic columns are still separated in the intensity ratio maps even at thicker region, more mixing can be seen in generated X-ray signals between atomic columns as the specimen thickness increases. Furthermore, the difference in X-ray intensities between on- and off-atomic columns is reduced as the specimen thickness increases. For Quantification of atomic-resolution X-ray maps, these X-ray generation behaviors must be modeled as a function of the specimen thickness. Microsc. Microanal. 21 (Suppl 3), 2015 1082 References [1] M. Watanabe, Microscopy 62 (2013), 217-241. [2] G. Kothleitner et al., Phys. Rev. Lett. 112 (2014), 085501. [3] K. Ishizuka., J. Electron Microsc. 50 (2012), 291. [4] C.E. Fiori et al. Public domain DTSA software package (1992). [5] The author wishes to acknowledge financial support from the NSF through grants DMR-0804528 and DMR-1040229. Figure 1: Simulated electron propagation at a Ga column position of [001]-projected GaAs (top) and an extracted intensity distribution (bottom). Figure 2: A schematic cross-section view of the [001]-projected GaAs, showing individual slices for X-ray spectrum generation. Figure 3: A set of simulated HAADF images (top), ratio maps of Ga K over total K intensities (middle), and normalized total K intensity maps (bottom) at different thicknesses.");sQ1[541]=new Array("../7337/1083.pdf","Quantitative EDS of Surface Modified Pd Powders for Hydrogen Storage.","","1083 doi:10.1017/S1431927615006200 Paper No. 0541 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantitative EDS of Surface Modified Pd Powders for Hydrogen Storage. Joshua D. Sugar1, Mark Homer1, Paul G. Kotula2, Patrick J. Cappillino3, Markus Ong4, and David B. Robinson1 1. 2. Sandia National Laboratories, Livermore, CA, USA. Sandia National Laboratories, Albuquerque, NM, USA. 3. University of Massachusetts Dartmouth, Chemistry and Biochemistry Department, North Dartmouth, MA, USA. 4 Whitworth University, Physics Department, Spokane, WA, USA. Palladium and its alloys are known to be useful for applications such as catalysis [1], electrocatalysis [2], and hydrogen isotope storage and separation [3, 4]. Atomic-scale surface and subsurface layers of other metals have surface hydride energetics that may enhance the kinetics of hydrogen absorption and desorption [5]. In addition, increasing the surface area/volume ratio by making Pd alloys nanoporous enhances surface-limited reaction rates. When Pd is alloyed with a higher melting temperature metal such as Rh, the temperature range over which the nanoporous structure remains morphologically stable is extended [6]. We use a variety of techniques to fabricate surface modified Pd powders at different length scales, including, the consolidation of dendrimer-encapsulated nanoparticles [7], surfactant templates [8], and atomic layer electroless deposition (ALED) [9]. It is crucial to understand the spatial distribution of alloying elements in these structures because of the large effect it has on thermal stability, hydrogen storage properties, and the kinetics of hydrogen uptake and release. Quantitative EDS was performed on Rh and Pt coatings applied to Pd powders at several length scales. Surface-modified powder particles with 5 nm diameter were fabricated using dendrimer encapsulation [10], 100-nm diameter particles were fabricated with surfactant templates [11], and ALED was used on powders >1 �m in diameter [9]. In all cases, the same analysis routines that utilize a combination of multivariate statistical analysis and multiple-least squares fitting was used to quantify the composition and thickness of the surface modified layer(s). We will discuss the importance of large-area SDD EDS detectors for acquiring data on these surface modified layers that range from tens of nm to sub-nm dimensions. References: 1] S. Tanaka, et al., Chemical Communications 47 (2011) [2] M. Watanabe and S. Motoo, Journal of Electroanalytical Chemistry 60 (1975) [3] R. Lasser and K.H. Klatt, Physical Review B 28 (1983) [4] M. Li and W.F. Yang, International Journal of Hydrogen Energy 34 (2009) [5] J. Greeley and M. Mavrikakis, Journal of Physical Chemistry B 109 (2005) [6] D.B. Robinson, et al., International Journal of Hydrogen Energy 35 (2010) [7] R.W.J. Scott, et al., Journal of Physical Chemistry B 109 (2005) [8] D.B. Robinson, et al., International Journal of Hydrogen Energy 34 (2009) [9] P.J. Cappillino, et al., Langmuir 30 (2014) [10] P.J. Cappillino, et al., Journal of Materials Chemistry 22 (2012) [11] M.D. Ong, et al., Chemistry of Materials 24 (2012) [12] Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000. Microsc. Microanal. 21 (Suppl 3), 2015 1084 Figure 1: (a) Lower magnification HAADF STEM image of 5 nm-diameter Pd particles made with dendrimer encapsulation. The green-boxed region in (b) is selected for EDS analysis. The Rh concentration is mapped in (c), and a line profile for the region marked with a blue arrow is shown in (d). The core-shell compositional structure of the particles is visible in the Rh concentration map. Figure 2: (a) A HAADF STEM image shows several nanoporous, 100 nm-diameter Pd particles made with surfactant templates. The green-boxed region in (b) is selected for EDS analysis. The Rh concentration is mapped in (c), and a line profile for the blue arrow is shown in (d). Again, the core-shell compositional structure of the particles is visible. Figure 3: (a) HAADF STEM image of ~1 �m-diameter Pd particles that have been surface modified with ALED. These samples were prepared using ultramicrotomy. The green-boxed region in (b) is selected for EDS analysis. The Rh concentration is mapped in (c), and a line profile for the blue arrow is shown in (d). The surface enrichment in Rh is visible.");sQ1[542]=new Array("../7337/1085.pdf","Computational Method for Composition Determination of Multilayer Epitaxial Semiconductor Structures Using Standards-Based Energy-Dispersive X-Ray Spectrometry","","1085 doi:10.1017/S1431927615006212 Paper No. 0542 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Computational Method for Composition Determination of Multilayer Epitaxial Semiconductor Structures Using Standards-Based EnergyDispersive X-Ray Spectrometry Monika Rathi1, Nan Zheng2 and Phil Ahrenkiel2 1. University of Houston, Mechanical Engineering, Houston, TX, U.S.A. 2. South Dakota School of Mines & Technology, Nanoscience & Nanoengineering, Rapid City, SD, U.S.A. Accurate measurements of composition provide critical information in understanding and optimizing epitaxial growth of compound semiconductors and alloys particularly used in optoelectronic devices. Energy-dispersive X-ray spectrometry (EDX) is widely used technique in transmission electron microscopy (TEM) to rapidly identify compositions of multilayer structures. It is easy to derive qualitative conclusions instantly with this method, quantitative analysis of the generated data remains a challenge that requires specialized software tools. We have developed Igor Pro procedures for analysis of both EDX data, using reference standards for calibration. For EDX, spectra acquired from known-composition standards allow extraction of the spectral generation rates weighted by the detector response. It is required that the standards data are obtained from the same electron-lens configuration (TEM) and all the reference and unknown samples are of same crystal structure. The standard data set can be obtained once and used multiple times with different unknown samples. Our EDX algorithm uses multivariate statistical analysis combined with a generalization of the -factor method of Watanabe and Williams. For a thin specimen, after correcting for absorption, the detected spectral intensity (number of counts per unit energy-time) at energy E can be written (1) d E k n De T wk xk E The sum is over constituent elements. The coefficients are the products of atomic concentration n , electron dose De , and foil thickness T with the atomic ratios wk , and the xk E are weighted generation rates. Treating energy E as a discrete parameter, and relabeling the coefficients as yk , we obtain the representation d X y . A collection of spectra can be combined into the matrix form D X Y . Singular-value decomposition allows factorization as D R C , where the columns of C contain the eigenvectors of D D . ( D is the transpose of D.) The contribution of each factor to the spectra is represented in a column of R; the relative significance of each towards accurate reproduction of the data matrix is proportional to the eigenvalues. Factors or low significance are then eliminated, along with their corresponding loadings, contained in the rows of C. Using spectra obtained from selected standards with known compositions, we find a matrix Z satisfying Y Z C . Noting that: (2) n De T k Z c k the normalization method in [1] can be applied: Zc w k Z c k (3) The procedure requires knowledge of at least one foil thickness among the spectra included in D (not necessarily with known composition). The components of Z are found by minimizing Microsc. Microanal. 21 (Suppl 3), 2015 1086 w c Z , (4) k ik j ij i, j identifies the standards. The least-squares minimization where the index can be formulated as an eigenvalue problem. Multiple solutions Z m to (4) may be found, which are combined linearly to most closely satisfy (2). The resulting matrix Z can then be applied to related specimens, using the factor loadings c R L d computed from their measured spectra as input, then using y Z c . The output then contains the computed thicknesses of all included spectra. This only has quantitative importance if an absorption correction is applied, in which case the algorithm is iterated until convergence. In addition, the weighted generation matrix can be computed, using X R Z L , allowing the simulation of spectra for arbitrary, related compositions. Flow chart in figure 1 describes the process in detail. 2 k , 2 Figure 1. Flow chart describing the computational logic to solve compositions Various tests were run on III-V multicomponent alloys, using readily available III-V binary endpoint compounds as reference standards. We mainly describe application of the technique to the III-V ternaries GaP1-yAsy, InP1-yAsy, GaxIn1-xAs, and GaxIn1-xP alloy and quaternary GaxIn1-xP1-yAsy alloys. But we believe that this technique can be used for any alloy if the required reference standards are available. References: [1] M Watanabe and D B Williams,, J Microscopy (221) p. 89. [2] M Rathi et al, Microsc Microanal. 19 (2013), p. 66. [3] J C H Spence and J M Zuo, Electron Microdiffraction, (Plenum, New York) p. 68.");sQ1[543]=new Array("../7337/1087.pdf","Practical Considerations in Quantitative Nanoscale Energy-Dispersive X-ray Spectroscopy (EDX) and Its Application in SiGe","","1087 doi:10.1017/S1431927615006224 Paper No. 0543 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Practical Considerations in Quantitative Nanoscale Energy-Dispersive X-ray Spectroscopy (EDX) and Its Application in SiGe Weihao Weng1, Frieder H. Baumann1, Yue Ke1, Rainer Loesing1, Anita Madan1, Zhengmao Zhu1 and Ahmad D. Katnani1 1. IBM Microelectronics Division, Zip 40E, 2070 Route 52, Hopewell Junction, NY 12533, USA Email : wweng@us.ibm.com TEL : +1-845-894-5944 In advanced semiconductor technology, epitaxial SiGe is used in P-channel Field-Effect Transistor (PFET) as source/drain. It serves as either a strain inducer or a "container" for B dopants, either of which plays a vital role for Complementary Metal Oxide Semiconductor (CMOS) device. Understanding the structure and composition of SiGe can provide crucial information for process development and device performance enhancement. At present, the Ge atomic concentration (%Ge) in SiGe is determined by Secondary Ion Mass Spectrometry (SIMS), High-resolution Rutherford Backscattering Spectrometry (HRBS), Low energy Electron induced X-ray Emission Spectrometry (LEXES) and/or X-ray Diffraction (XRD) on large SiGe structures of hundreds of microns in size. For small SiGe structures at the nanometer scale, like those in source/drain in CMOS device, Energy-dispersive X-ray spectroscopy (EDX) in the Transmission Electron Microscope (TEM) is used. EDX offers exceptional advantages in quantification, including high spatial resolution, good precision and accuracy. This quantitative EDX technique is based on the Cliff-Lorimer C I ratio method for a two-element SiGe system: Ge kGeSi Ge and CGe CSi 100% , where C represents C Si I Si the weight percentages, I represents X-ray intensities, and k is termed the Cliff-Lorimer factor. Here k is not a constant, but a sensitivity factor, which varies with electron microscopes, accelerating voltages, samples and so forth. This work presents our efforts in understanding the variables that influence EDX quantification results and in developing a standard methodology to quantify %Ge using EDX in the TEM for semiconductor research and development. A standard Si1-xGex blanket test sample (Figure 1) was characterized by HRBS and XRD, and found to be Si0.664Ge0.336 with a relative error smaller than 1%. It is then used to calibrate k factor. Effects of sample tilting, sample thickness, electron voltage and electron fluence on quantitative EDX were evaluated. As shown in Figure 2, tilting off the zone axis greatly affects the quantification if the K-line is used, but not as much if the L-line is used. After 2 degree tilting around either [001] axis or [110] axis of SiGe, %Ge reaches a plateau, indicating that electron channeling effect is minimized. [1] Table 1 shows the impact of sample thickness on %Ge determination. As expected for a typical TEM sample that is usually thinner than 80nm, sample thickness effects (i.e. X-ray absorption and fluorescence) can be neglected. It is well known that electron radiation damage in TEM can significantly change the physical and/or chemical state of the tested sample. [2] Figure 3(a) demonstrates that after an electron fluence of 2.54E+09 nm-2 the Si-K intensity of the sample seems to decrease due to electron radiation damage. Figure 3(b) shows that reducing electron energy can effectively diminish the electron radiation damage on SiGe sample. The critical electron fluence on Si0.664Ge0.336 at 120kV is found to be 3.46E+10 nm-2 as compared with 2.54E+09 nm-2 at 200kV, allowing for longer acquisition time and statistically more accurate %Ge determination. Using the standard methodology, namely a thin sample tilted by >2� away from zone axis, low electron energy, below critical electron fluence and a calibrated k factor, %Ge Microsc. Microanal. 21 (Suppl 3), 2015 1088 of a typical SiGe source/drain diamond structure [3] can be obtained in Figure 4. In addition, EDX measurement on the zone axis of the SiGe sample can be performed by setting electron beam in precession mode. [4] The established methodology of quantitative nanoscale EDX technique can be applied to other material systems in semiconductor industry, such as TiSi, NiSi, AsSi etc. References [1] J.C.H. Spence et al, Journal of Microscopy 130 (1983) p. 147. [2] R. H�bner et al, Thin Solid Films 519 (2010) p. 203. [3] C.H. Lin et al, IEDM (2014). [4] Y. Liao et al, Ultramicroscopy 126 (2013) p. 19. [5] Acknowledgements: The authors thank John Bruley (IBM), Yun-Yu Wang (IBM), Jinghong Li (IBM), Alexandre Pofelski (STMicroelectronics) and Germain Servanton (STMicroelectronics) for valuable discussions. 33.6% Figure 1. HAADF-STEM micrograph of the standard SiGe blanket test sample. It has [110] in-plane direction and [001] out-of plane direction. (a) (b) Figure 2. %Ge as a function of sample tilting angles. Due to electron channeling effect, %Ge measured close to the zone axis strongly depends on tilt, and can be off by up to 20% if the K-line is used. This effect is much smaller if the L-line is used. 120kV 200kV 33.6% 33.6% Figure 3. %Ge and normalized intensity as a function of electron fluence at (a) 200kV and (b) 120kV, respectively. Critical electron fluence (as arrowed) on Si0.664Ge0.336 is found to be 2.54E+09 nm-2 for 200kV, and 3.46E+10 nm-2 for 120kV. a b Table 1. %Ge as a function of sample thickness. Figure 4. (a) HAADF-STEM micrograph and (b) %Ge map of a SiGe diamond structure. The %Ge varies from 48.7% to 55.6% within this diamond with an average %Ge at 51.0%.");sQ1[544]=new Array("../7337/1089.pdf","Quantification of Phase Compositions and Diffusional Profiles in Simulated Solid-State Welds of Ti-17 via Super-X Energy-Dispersive X-Ray Spectroscopy","","1089 doi:10.1017/S1431927615006236 Paper No. 0544 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantification of Phase Compositions and Diffusional Profiles in Simulated SolidState Welds of Ti-17 via Super-X Energy-Dispersive X-Ray Spectroscopy Jonathan Orsborn1,2, Robert E.A. Williams2, Hamish Fraser1 1. Center for the Accelerated Maturation of Materials, Department of Materials Science and Engineering, The Ohio State University, Columbus, OH, U.S.A. 2. Center for Electron Microscopy and Analysis, The Ohio State University, Columbus, OH, U.S.A. Recent advances in welding technology have enabled the aerospace industry to further reduce the weight of aircraft, by reliably welding titanium alloys. Solid-state welding, in which two metal pieces are welded without either one melting, has recently been proven to produce reliable and consistent welds, and is rapidly becoming the preferred method for joining titanium. Rapid heating rates, short hold-times, plastic deformation, thermal gradients, and fast cooling rates all contribute to the non-equilibrium nature of the process. Thus, a very wide variety of phase transformations, both equilibrium and nonequilibrium, may occur, affecting the microstructure of the welded metal. Since the mechanical properties of titanium alloys are particularly dependent upon microstructure, it is extremely important to understand these phase transformations and the microstructures they produce in order to better inform integrated computational materials engineering(ICME) models. Commercial titanium alloys have two equilibrium phases: the low-temperature hexagonal close-packed (hcp) phase, , and the hightemperature body-centered cubic (bcc) phase, . Additionally, non-equilibrium phases, such as martensitic 'and ", and athermal and isothermal may also develop depending on thermo-mechanical history [1]. The composition of these various phases and the diffusional profiles that develop during thermo-mechanical processing is a critical piece of information for ICME modeling and a metric that may be quantified more accurately using modern silicon drift detectors for improved XEDS characterization. This study centers on quantifying the bulk compositions of the various phases present, as well as the diffusional profiles that exist between phases of the alloy Ti-17, or Ti-5Al-4Cr-4Mo-2Sn-2Zr, after being subjected to conditions mimicking those of a solid-state weld. To simulate sold state weld conditions, samples were processed, using a GleebleTM 1500, to heat round bars at 100�C/s to nearly 1000�C and hold for a short time (0-10s). The samples were compressed, while at temperature, then cooled at ~30�C/s, similar to the conditions of a solid-state. Imaging in the SEM revealed thin, needlelike intragranular precipitates, on the order of 5-25 nm wide, as shown in Figure 1. However, due to the small size of the precipitates, phase identification was not possible in a bulk, SEM sample. In order to determine the phase of the fine needle precipitates, site-specific TEM samples were prepared using a dual-beam FIB. The TEM lamellae were milled using 30 kV Ga+ ions for coarse material removal and 5 kV Ga+ ions for final thinning and clean up. The thin foils were then plasma-cleaned and low-energy Ar+ ions milled using a Fischione NanomillTM, using accelerating voltages of 900 V and 500 V, for 10 min on each side, at each voltage to remove amorphous damage [2]. TEM analysis was performed using a 300 kV FEI TitanTM equipped with Super-X XEDSTM and a monochromator. This combination of technological advances has enabled unprecedented XEDS with high spatial resolution. While the physics of XEDS has not changed, the markedly better signal to noise provided more statistical certainty from quantification. The extreme stability of the stage and the high count-rate capability of Super-X XEDSTM system help to enable spatially accurate XEDS measurements, on this ultra small size scale. Microsc. Microanal. 21 (Suppl 3), 2015 1090 Figure 2(a) displays a HAADF-STEM image of the alpha lath in a beta matrix, while Figures 2(b-f) show the complimentary XEDS compositional maps. Figure 3(a) shows the HAADF-STEM image and a schematic line indicating the line scan direction for Figure 3(b). The XEDS line profile shows the Al concentration to be slightly elevated and Cr and Mo were reduced in the -phase. The data also showed a build-up of -stabilizing elements, in the -phase, at the / interface. This would be consistent with the diffusional nature of -phase precipitating from the matrix. K-factors were also determined experimentally from samples of solutionized Ti-17 to refine the XEDS quantification [3]. To validate and confirm XEDS, the alloy composition was also determined independently via wet-chemical analysis. Experimentally determined k-factors were used to quantify the compositional maps and profiles to reduce the error in quantification. Experimental k-factor determination and quantification refinement will be discussed. [4] References: [1] G. L�tjering G. and J.C. Williams in "Titanium", (Springer, Berlin). [2] R.E.A. Williams, Ph.D. Dissertation (2010). [3] G. Cliff and G. W. Lorimer, Journal of Microscopy 103 (1975), p. 203-207. [4] Support for this work, from Tom Broderick at GE Aviation, is greatly appreciated. Figure 1. SEM backscattered electron micrograph of the intragranular, needle-like precipitates. Figure 2. a) HAADF-STEM image of -lath in -matrix, with b-f) corresponding quantified elemental maps of solutes (brightness levels were increased, for visibility). Figure 3. a) HAADF-STEM image showing the line used for b) the compositional profile traversing the -lath (the phase-neutral alloying elements, Sn and Zr, were not included in the plot for clarity).");sQ1[545]=new Array("../7337/1091.pdf","Effect of Specimen Geometry on Quantitative EDS Analysis with Four-Quadrant Super-X Detectors","","1091 doi:10.1017/S1431927615006248 Paper No. 0545 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Effect of Specimen Geometry on Quantitative EDS Analysis with Four-Quadrant Super-X Detectors W. Xu1, J. H. Dycus1, X. Sang1, A. A. Oni1, J. M. LeBeau1 1. Department of Materials Science and Engineering, North Carolina State University, Raleigh, NC 27695, USA. In recent years, X-ray signal collection efficiency in energy dispersive X-ray spectrometry (EDS) has been significantly improved by incorporating state-of-the-art four-quadrant Super-X detectors in aberration-corrected microscopes [1, 2] and has enabled routine atomic-resolution EDS elemental mapping, as shown in Figure 1a. The breakthrough in detector technology has sparked interest in quantifying the chemical information directly from atomically resolved EDS maps [3-5]. In order to obtain an atomic-resolution EDS map, the specimen however must be tilted to a low index zone axis, resulting in different orientation with respect to each detector. This will inevitably result in different Xray signals arriving at each detector due to their different absorption distances and effective take-off angles. Such a variation could largely affect the ability to accurately quantify the overall EDS spectrum from four Super-X detectors without precautions as evidenced in this work. Therefore, understanding the X-ray signal variation from each detector under tilted specimen condition becomes both experimentally and theoretically important as a starting point for atomic-resolution EDS quantification. In this talk, we will show that the specimen geometry can have a strong effect of on quantitative EDS when using a four-quadrant Super-X detector configuration. Ni3Al is an ideal prototype material to investigate geometry effects as the Ni and Al characteristic X-ray peaks are well separated and strong absorption of Al-K occurs by Ni [6]. As shown in Figure 1b, a wedge-polished Ni3Al sample was studied using a Titan G2 S/TEM with the sample thin region facing to detectors 3 and 4 in an azimuth angle of 45 degrees. Figure 2a-c show the total intensity of Al-K, Ni-K and Ni-L X-ray signals received from each detector in the same thickness region, respectively. Special care was made to minimize the channeling effect from the specimen. As seen, X-ray peaks obtained from each detector not only vary with the tilt angle, but also systematically shift about 5 degrees towards the positive tilt angle. More importantly, a large deviation of intensity ratios of Al-K/Ni-K and Ni-L/Ni-K peaks from different detectors was observed with same specimen tilt as seen in Figure 2d-f. Furthermore, the ratio deviation also varies with the degree of specimen x-tilt, and appears to be non-symmetric, which is correlated with the wedge shape of the specimen. Although the deviation of X-ray signal from four Super-X detectors may be averaged out in the overall spectrum, significant change of the overall intensity ratio occurs at a higher tilt angle even larger than 10-15 degrees. In view of the large quantification uncertainty with the tilt sample, the correction of X-ray signal from overall spectrum for all Super-X detectors is essential and will be discussed. The limitation on tilt angle regarding the absorption, spurious X-ray and fluorescence effect in the Super-X detector configuration will also be presented. The effect of sample shape on the detected X-ray signals from Super-X detectors will be further compared for wedge, FIB and twin-jet polished specimens [7]. References: [1] P. Schlossmacher, et al, Microscopy and Analysis (nanotechnology supplement) 24 (2010), p. 55. [2] L.J. Allen, et al, MRS Bulletin 37 (2012), p. 47. Microsc. Microanal. 21 (Suppl 3), 2015 1092 [3] G. Kothleitner et al, Physical Review Letters, 112 (2014), p. 085501. [4] P. Lu et al, Scientific Reports, 4 (2014), p. 3945. [5] M. Watanabe, Microscopy, 62 (2013), p. 217. [6] D.B. Williams and J.I. Goldstein in "Analytical Electron Microscopy 1981", ed. R.H. Geiss, (San Francisco Press, San Francisco), p. 39. [7] This work is supported by the Air Force Office of Scientific Research (Grant: FA9550-14-1-0182). The authors also acknowledge the Analytical Instrumentation Facility (AIF) at North Carolina State University, which is supported by the State of North Carolina and the National Science Foundation. Figure 1. (a) HAADF image and the corresponding atomic-resolution EDS maps (Ni-K and Al-K) for Ni3Al oriented along a [100] direction. The EDS maps are filtered using a 3 points average. (b) Schematic illustration of the geometric orientation relationship between the wedge-shape specimen and four-quadrant Super-X detectors. Figure 2. Deviation of X-ray signals from Super-X detectors in terms of (a) Al-K, (b) Ni-L and (c) NiK intensities, and their intensity ratio of (d) Al-K/Ni-K, (e) Al-K/Ni-L and (f) Ni-L/Ni-K. Such deviation also varies with the specimen x-tilt, and appears to be non-symmetric. All spectra were obtained from a wedge-polished Ni3Al specimen in the same thickness region (0.5 mean free path from EELS) and under the same microscope condition (probe-corrected STEM with a 0.11nA and 14 mrad probe). The orientation relationship between the specimen and Super-X detectors is shown in Figure 1b.");sQ1[546]=new Array("../7337/1093.pdf","Influence of Convergence Angle and Finite Effective Source Size for Quantitative Atomic Resolution EDXS","","1093 doi:10.1017/S143192761500625X Paper No. 0546 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Influence of Convergence Angle and Finite Effective Source Size for Quantitative Atomic Resolution EDXS J. H. Dycus1, S. D. Findlay2, W. Xu1, X. Sang1, L. J. Allen3, and J. M. LeBeau1 1. Department of Materials Science and Engineering, North Carolina State University, Raleigh NC 27695, USA 2. School of Physics and Astronomy, Monash University, Victoria 3800, Australia 3. School of Physics, University of Melbourne, Parkville, Victoria 3010, Australia Recent progress in energy dispersive x-ray spectroscopy (EDS) has enabled atomic resolution chemical mapping [1] which has become an invaluable tool for chemical analysis, particularly for identifying heavy elements [2]. Although recent development of Super-X technology has enabled qualitative identification of the atoms present in an atom column, quantitative atomic resolution analysis remains an active area of development. Many contributions to the atomic resolution signal have been studied, such as thickness and tilt; however, the effects of many other parameters remain unreported. In this talk, we will first examine the effects of probe convergence angle on atomic resolution EDS signal via simulation and experiment. Using a probe corrected FEI G2 Titan 60-300 kV STEM equipped with a Super-X detector, atomic resolution EDS maps of <100> strontium titanate (STO) were collected using probe-forming convergence angles of 14 and 20 mrad with otherwise identical operating parameters. Each map was then lattice averaged using a template with the size of the unit cell translation vector in a similar approach as in Ref. [3]. The averaged unit cells were then replicated, producing a tiled representation of the averaged unit cell (Figure 1). By simulating the same structure using �STEM [4], we observe agreement between simulated and experimental atomic maps. Unsurprisingly, both experiment and simulation reveals that 20 mrad produces maps with better spatial resolution; however, the signal from the atom columns appears stronger for the case of 14 mrad, especially for the light atom columns of pure O. Importantly, the difference in the O signal from the Ti/O columns is greater than O in pure O columns due to channeling [5]. We will discuss how the relationship between channeling and convergence angle raises important considerations for atomic level quantification. Contrast dependence on the convergence angle is further explored in addition to finite effective source size and thickness. Simulations for a different system, Ni3Al, display a similar behavior after considering the finite effective source size. By using a test example of <100> Ni3Al, the maximum, mean, and minimum counts are plotted as a function of thickness for both 14 and 21 mrad, shown in Figure 2a. Without accounting for the finite effective source size, 21 mrad exhibits higher maximum intensity for the tightly spaced Ni (left). But incorporating the source size dramatically alters the signal from the atom columns (right), now displaying higher Ni maximum signal for 14 mrad rather than 21 mrad while the background remains lower. The influence of convergence angle and thickness will additionally alter the ratios of x-ray lines which directly impact quantitative elemental analysis. The impact on peak ratios and possible mechanisms for convergence angle will be presented [6]. Microsc. Microanal. 21 (Suppl 3), 2015 1094 References: [1] A. J. D'Alfonso et. al., Physical Review B 81 (2010), p. 100101. [2] J. H Dycus et. al., Applied Physics Letters 102 (2013), p. 081601. [3] P. Lu et. al., Microscopy and Microanalysis 20 (2014), p. 1782-1790. [4] L. J. Allen et. al., Ultramicroscopy, in press, doi:10.1016/j.ultramic.2014.10.011. [5] B. D. Forbes et. al., Physical Review B 86 (2012), p. 024108�9. [6] JHD, WX, XS, and JML gratefully acknowledge support from the Air Force Office of Scientific Research (Grant No. FA9550-14-1-0182). SDF and LJA acknowledge support under the Australian Research Councils Discovery Projects funding scheme (DP140102538 and DP110102228). JHD acknowledges the National Science Foundation Graduate Research Fellowship (Grant DGE-1252376). The authors acknowledge the Analytical Instrumentation Facility (AIF) at North Carolina State University, supported by the State of North Carolina and the National Science Foundation. Figure 1. (a) Experimental EDXS lattice averaged maps replicated to a size of 4x4 unit cells. Line profiles display the intensity along atomic columns in the direction by the arrows. (b) Simulated EDXS maps with a 1 � finite effective source size. Maps were filtered using a `fire' colormap and no smoothing has been applied. Figure 2. Simulated signal for Ni3Al<100>. (a) Maximum, mean, and minimum signals plotted without (a) and with (b) incorporating the finite effective source size.");sQ1[547]=new Array("../7337/1095.pdf","TEM Based High Resolution Electron Diffraction Techniques for Threedimensional Nanostructure Determination","","1095 doi:10.1017/S1431927615006261 Paper No. 0547 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 TEM Based High Resolution Electron Diffraction Techniques for Threedimensional Nanostructure Determination Jian-Min Zuo1,2, Yifei Meng1,2, Piyush Vivek Deshpande1,2, Yang Hu1,2, Kyou-Hyun Kim1,2, Hui Xing1,2,3, Peng Zhang4 and Haifeng Wang4 1 2 Dept of Materials Science and Engineering, University of Illinois, Urbana-Champaign, IL 61801 Frederick Seitz Materials Research Laboratory, University of Illinois, Urbana-Champaign, IL 618013 3 School of Materials Science and Engineering, Shanghai Jiaotong University, China 4 Western Digital Technologies, Inc., 44200 Osgood Road, Fremont CA, 94539 Many technologically important materials possess complex nanostructures and structure dependent properties. Transmission electron diffraction (TED) is an appropriate technique for complex nanostructure analysis because it is highly sensitive to local structure and it can be obtained using a small electron beam [1-4]. Compared to the scanning electron microscope (SEM) based electron backscattering diffraction technique, the small interaction volume in TED allows for high spatial resolution. Traditionally, TED is performed either by using parallel beam illumination with the help of a selected area aperture for selected area electron diffraction (SAED) or by using a focused beam for convergent beam electron diffraction (CBED). Electron nanodiffraction can be performed in a modern TEM instrument using a parallel electron beam of nanometers in diameter with help of an additional condenser lens (minilens) [3, 4]. For complex nanostructure analysis, there is a need to record multiple diffraction patterns by scanning from desired sample areas. Two major approaches have been developed thus far, one uses the built-in STEM scan coil and performs electron nanodiffraction in STEM mode, and the other uses external scan and precession driver coupled with an external camera[5]. For example, Alloyeau et al. [6] and Ganesh et al. [7] have reported the use of a small probe controlled by the GatanTM software "STEM Diffraction Imaging" to acquire diffraction patterns on a pixel by pixel basis. For three-dimensional microstructure determination, EBSD coupled with FIB cross-section or nondestructive X-ray diffraction techniques have been developed. More recently, Liu et al. reported a TEM based technique using conical scan [8]. Here, we report on a TEM based scanning electron nanodiffraction (SEND) technique that uses the built-in TEM deflection coils. Diffraction pattern recording and beam scanning are automated using a DigitalMicrograph� script to control the TEM deflection coils and camera readout [9]. In a conventional TEM with a LaB6 source, SEND can be performed in low dose mode using electron beams of 2~5 nm in full-width at half-maximum (FWHM) and 0.1 pA or less in beam current. Electron diffraction pattern indexing is achieved by a combination of correlation analysis of the recorded diffraction patterns, diffraction peak search and peak indexing using both length and angle information. We demonstrate the performance of this approach nanostructure determination in several nanostructure materials (see Figure 1 for an example). By coupling with a tomographic holder, SEND can be performed with large sample tilts and thus the opportunity for the determination of three-dimensional domains at nanometer resolution [10]. References: [1] Cowley, J.M., "Electron Diffraction Techniques." 1993: International Union of Crystallography. Microsc. Microanal. 21 (Suppl 3), 2015 1096 [2] Spence, J.C.H. and J.M. Zuo, "Electron microdiffraction". 1992: Plenum Press. [3] Zuo, J.M. et al, Science, 2003. 300(5624): p. 1419-1421. [4] Zuo, J.M. et al. Microscopy Research and Technique, 2004. 64(5-6): p. 347-355. [5] Rauch, E.F. et al, Zeitschrift Fur Kristallographie, 2010. 225(2-3): p. 103-109. [6] Alloyeau, D., et al, Ultramicroscopy, 2008. 108(7): p. 656-662. [7] Ganesh, K., et al, Microscopy and Microanalysis, 2010. 16(SupplementS2): p. 1728-1729. [8] Liu, H.H., et al, Science, 2011. 332(6031): p. 833-834. [9] Tao, J., et al, Physical Review Letters, 2009. 103(9). [10] The work reported here is supported by DOE BES under contract DEFG02-01ER45923 and a grant from Western Digital. [001] [011] [111] Figure 1. Orientation maps obtained from a high entropy alloy (Al0.5CoCrCuFeNi) at two orientations of 30 degrees apart. The sample was first fabricated into a needle shape using FIB. SEND was performed using 25x25 sampling points over 240x240 nm2 using an electron probe of 8 nm in FWHM The diffraction patterns were indexed using software developed at University of Illinois and colored according to the inverse pole figure at top-right corner. Two individual experimental diffraction patterns are shown on left.");sQ1[548]=new Array("../7337/1097.pdf","Towards Identification of Oxygen Point Defects by Means of Position Averaged CBED","","1097 doi:10.1017/S1431927615006273 Paper No. 0548 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Towards Identification of Oxygen Point Defects by Means of Position Averaged CBED R. dos Reis, C. Ophus, J. Ciston, P. Ercius, U. Dahmen National Center for Electron Microscopy/Molecular Foundry, LBNL, Berkeley, CA 94720, USA Point defects strongly affect the physical properties of materials and have a determinant impact on their performance. Extensive experimental research has been devoted to their characterization in previous decades. However, the understanding and control of point defects is limited and remains one of the challenging issues in the design of new materials. In this work, we explore the possibilities and limitations of detecting oxygen vacancy sites in Zinc Oxide (ZnO) by means of Position Averaged CBED [1]. PACBED patterns are very sensitive to thickness variation and to small tilts and deformations of the unit cell, with the advantage of being independent of aberrations. In particular, ZnO nanostructures contain a large density of intrinsic defects, especially oxygen vacancies (VO), which play an important role in the green light emission. VO sites exhibit different local relaxations depending on their charge state [2]. Zn atoms surrounding the site in a neutral charge state (VO�) relax inward approximately 11% (1.952� vs 1.733�). On the other hand, a strong outward relaxation of about 19% is found for the charged vacancy (VO++). In order to evaluate the effect of vacancies on a PACBED pattern, Figure 1 presents multislice simulations for an ideal ZnO crystal, ZnO with VO++-type vacancy, and with VO�-type vacancy. A layer thickness of 100� (~30 unit cells) was used for all cases, applying the atomic configuration shown in the first row of Fig. 1. A sublayer containing a single point defect replaced an ideal ZnO slab in the middle of the supercell. Figures 1 (a-c) show the results along the 1010 zone axis. Figures 1(d-f) show the simulations along the 1120 zone axis. The local distortions caused by the presence of point defects can be noticed as a subtle intensity variation within the PACBED discs. To illustrate this effect, Figure 2 plots differential maps for the simulated PACBED patterns. It compares layers containing VO� and VO++ to an ideal ZnO layer along 1010 (Fig 2(a-b)) and 1120 (Fig. 2(c-d)), respectively. Differential maps highlight the characteristic features of each ideal case, otherwise unnoticeable. They also show that the range of intensity variation falls within �25% of the total PACBED intensity. The feasibility of the method was tested by acquiring experimental data from ZnO nanowires as 4DSTEM datasets [3]. CBED patterns were recorded at many probe positions with millisecond dwell times by using a Gatan K2-IS direct electron detection camera installed on the TEAM I microscope [3]. Ideal PACBED patterns were constructed by averaging the probe images over each ZnO unit cell along the 1010 zone axis. Figure 3(a) presents a pattern fitted from the central area of a simultaneously acquired HAADF image (Fig. 3(c)). A mean thickness of 220� was determined by comparison with simulated patterns (Fig. 3(a) bottom). Figure 3(b) shows a differential map comparing the PACBED of a single unit cell (center) to that of its eight first neighbors. It is observed that the signal to noise ratio is sufficient to detect the variations within the patterns (�25% of the intensity). The presence of O vacancies may be determined by refining the positions of the PACBED discs and by matching with simulated patterns. Detailed analysis is currently underway and the issues concerning this method will be addressed alongside the code development to improve the precision [4]. Microsc. Microanal. 21 (Suppl 3), 2015 1098 [1] JM LeBeau et al. Ultramicroscopy 110, p.118 (2010). [2] P. Erhart et al. Phys. Rev. B 73, 205203 (2006). [3] C. Ophus et al. Acta Cryst A70, C1455 (2014). [4]Work at the NCEM/Molecular Foundry was supported by the Office of Science, Office of Basic Energy Sciences, of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231. RR acknowledges the support of CAPES/BR foundation Process No. 1204713-9 and thanks Dr. D.L. Baptista and Dr. C.I.L. Sombrio for providing ZnO nanowires for this study. Figure 1: Simulated PACBED patterns using multislice for an ideal ZnO, a sample containing a single VO� and, VO++, along 1010 zone axis (a�c) and along 1120 zone axis. The structural model for each type of defect is shown on the top row. Figure 2: Differential maps for simulated PACBED comparing the ideal ZnO to the layers including vacancies. (a) ZnO - VO� (b) ZnO - VO++ for 1010 zone axis. (c) ZnO - VO� and (d) ZnO - VO++ for 1120 zone axis. Figure 3: (a) Experimental PACBED constructed from the central unit cell of HAADF image (c), alongside the simulated pattern with thickness of 220�. (b) Differential map comparing the central PACBED to eight first neighbors. (c) HAADF acquired simultaneously to the probe images and used to fit the unit cells along 1010 zone axis (scale bar of 3�).");sQ1[549]=new Array("../7337/1099.pdf","Mapping Valence Electron Distribution of Iron-Based Superconductors using Quantitative CBED and Precession Electron Diffraction","","1099 doi:10.1017/S1431927615006285 Paper No. 0549 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Mapping Valence Electron Distribution of Iron-Based Superconductors using Quantitative CBED and Precession Electron Diffraction Lijun Wu1, Chao Ma1,2, Binghui Ge1,2, Weiguo Yin1, and Yimei Zhu1 1. Condensed Matter Physics and Materials Science Department, Brookhaven National Laboratory, Upton, NY 11973, USA. 2. Beijing National Laboratory for Condensed Matter Physics, Institute of Physics, Chinese Academy of Sciences, Beijing 100190, China. The discovery of superconductivity with transition temperature (Tc) up to 56 K in Fe-based superconductors (FeSCs) has provided a grand opportunity to explore high-temperature superconducting materials beyond copper containing oxides. Like in the cuprates, superconductivity in FeSCs emerges when the parent antiferromagnetic phase is suppressed, typically by introduction of dopant atoms, such as Co- or Ni- doped BaFe2As2. Unlike the oxygen anions sitting at the middle of the Cu-Cu bonds in the cuprates, the anions in FeSCs are positioned above and below the Fe planes. This low-symmetry setup facilitates the anion polarization. The electronic oscillation of the anions has been shown to be a critical relevant degree of freedom in a one-orbital model for FeSCs. Whether this holds in real materials and how it interacts with the other active degree of freedom in the Fe planes are urgently needed to be elucidated. We address these important questions by probing the Co concentration dependence of the valence electron density distribution in Ba(Fe1-xCox)2As2 (BFCA), a prototypical FeSC using quantitative CBED (Fig. 1(a-c)). With the combination of measured low-order structure factors (SFs) and density function theory (DFT) calculated high-order SFs, we retrieve the charge density distribution for undoped (x=0, Tc=0) and optimally doped BFCA (x=0.1, Tc=22.5K) through multipole refinement. The resulting three-dimensional (3D) and two-dimensional (2D) difference electron density map (difference between experimentally measured electron density and superimposed isolated spherical atomic electron density) are shown in Fig.1(d-h). The Co concentration dependence is characterized by a significant increase in the out-of-plane component of the electronic dipole moment around the As anions and in the out-of-plane components of the quadruple moment around the Fe cations (Fig.1(i,j)), echoing the significant increase in Tc and indicating a strong dipole-quadruple interaction between As and Fe atoms. The observations show a strong correlation among Tc, the electronic polarization of the anions, interorbital charge transfer on the Fe cations, and Fe-Fe bond polarization in BFCA. These observations provide direct support for the proposal that the large polarizability of the anions is critical to iron-based superconductivity by solvating the repulsive Coulomb interaction between electrons in the Fe plane [1]. Accurate determination of the doping and temperature dependences of the Fe-quadrupole and As-dipole polarizations requires a systematic study of valence electron distributions for different doping concentration and temperature. Usually the low-order SFs were measured by quantitative CBED, highorder SFs by single crystal x-ray diffraction (SCXRD). For FeSC, however, high quality single crystals are currently not available for SCXRD. We therefore use large angle CBED (LACBED, Fig. 2(a-c)) and precession electron diffraction (Fig. 2(e,f)) to measure high-order SFs. References: [1] C. Ma, L. Wu, W. Yin, H. Yang, H. Shi, Z. Wang, J. Li, C. C. Homes, and Y. Zhu, Phys. Rev. Lett. 112 (2014), p. 077001. Microsc. Microanal. 21 (Suppl 3), 2015 1100 [2] The work is supported by the U.S. DOE, under contract No. DE-SC0012704. Figure 1. (a,b) Experimental energy-filtered (a) and calculated (b) CBED patterns of BaFe2As2 showing the hh0 systematical row reflections at room temperature. (c) Line scans of the intensity profile from the experimental pattern (dots) and calculated one (solid line) after the structure factor refinement. (d,e) Valence electron density map of Ba(Fe1-xCox)2As2 in (100) FeAs plane for (d) x=0 and (e) x=0.1 as well as the three-dimensional map for (f) x=0 and (g) x=0.1. The isovalues of difference charge density on the isosurface are 0.1 e/�3 (red) and -0.1 e/�3 (blue), respectively. The color legend indicates the magnitude of the charge density. The red line denotes the shortest As-As distance between the FeAs layers. (h) The structure model. (i) The radius dependence of the out-of-plane dipole moment inside the sphere centered at the As anion and inside the sphere centered at the middle point of the next-nearest FeFe bond illustrated as the white dashed circle in (d) and (e) (red solid and dashed lines). (j) The radius dependence of the quadrupole moments inside the sphere centered at the Fe anion. Figure 2. (a,b) Experimental energy-filtered (a) and calculated (b) LACBED patterns of BaFe2As2, showing the high-order reflections at 90K. (c) Intensity line profile from the experimental pattern (dots) and calculated one (solid line) after high-order structure factors refinement. (d,e) Electron diffraction patterns without precession (d) and with precession (e, precession angle = 3.1�) for BaFe2As2. With precession, not only more high-order reflections (up to 10,0,0) are measurable, but also their intensities are close to the structure factor squared, thus suitable for electron density refinement.");sQ1[550]=new Array("../7337/1101.pdf","Transmission Kikuchi Diffraction (TKD)via a horizontally positioned detector.","","1101 doi:10.1017/S1431927615006297 Paper No. 0550 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Transmission Kikuchi Diffraction (TKD)via a horizontally positioned detector. Fundenberger J.-J.1,2, Bouzy E.1,2, Goran D.3, Guyon J.1,2, Morawiec A.4, Yuan H.1,2 Laboratoire d'Etude des Microstructures et de M�canique des Mat�riaux (LEM3), UMR CNRS 7239,Universit� de Lorraine, 57045 Metz, France 2. Laboratory of Excellence on Design of Alloy Metals for low-MAss Structures (DAMAS), Universit� de Lorraine,57045 Metz, France 3. Bruker Nano GmbH, Am Studio 2D,12489 Berlin, Germany 4. Institute of Metallurgy and Materials Science, Polish Academy of Sciences, Reymonta 25, 30059 Krakow, Poland Orientation mapping techniques are become indispensable tools for materials characterization. Indeed, they allow for quantification of phase amounts, local crystallographic textures, grain boundary characters, grain sizes, densities of geometrical necessary dislocations et caetera. In particular, the EBSD mapping technique has been a tremendous success because it helps in quick characterization of bulk crystalline materials. However, its lateral spatial resolution is limited. Moreover, as the specimen is highly tilted during the EBSD measurement, the resolution is `anisotropic': it is three times worse in the direction perpendicular to the tilt axis than in the direction of the axis. Consequently, this method is not suitable for application to ultrafine or nano-grained materials. One alternative is to use transmission electron microscopy (TEM) to get orientation maps from Kikuchi patterns using large angle convergent nano-beam [1] or from spot patterns using low angle nano-beam with precession [2]; the lateral resolution of these techniques is in the range of 5-10nm. Recent developments in orientation mapping based on SEM in transmission mode (labeled TKD for Transmission Kikuchi Diffraction) pushed the lateral resolution to about 10nm, that is nearly to the level of TEM-based techniques [3]. With TKD, high resolution maps are obtained using the same hardware and software as in standard EBSD systems [4]. In this presentation we will describe a new configuration for the TKD technique, and we will show that it allows us to achieve a lateral resolution even better than that of the conventional configuration. The novel feature of our configuration is the position of phosphor scintillator. In the conventional TKD, it is positioned nearly vertically, whereas we positioned it horizontally. The scintillator was removed from the nose of the EBSD camera housing, and a mirror was placed to transfer the light emitted by the scintillator to the camera (see figure 1).The mirror is positioned so the optics of the camera is focussed on the scintillator. A thin foil is mounted on a C-shaped transmission sample holder which is on a threeaxis mechanical stage. With this device, both the working distance and the sample-scintillator are independently adjustable. The Kikuchi patterns are free of any geometrical distortion. They are processed either on-line using commercial software or off-line using an in-house software originally developed for TEM-based orientation mapping [1]. Various high resolution orientation maps with a step size of 5nm will be shown as they were obtained, i.e., without any clean-up or noise reduction. For example, an ultrafine grained silicon specimen with 10nm-wide twins inside the grains. The twins are clearly distinguishable on the recorded orientation map. Lateral resolution tests have been carried out on a titanium aluminide with a lamellar microstructure. This specimen has been prepared using EBSD and FIB so the interfaces are edge-on. One-nanometer 1. Microsc. Microanal. 21 (Suppl 3), 2015 1102 step size scans have been made through the interfaces using an electron beam current of 300 pA. The tests show that the physical lateral resolution, i.e. the minimum distance for which only one Kikuchi pattern is discernable, is about 5nm. The effective lateral resolution, which takes into account the ability of the indexing software to disentangle superimposed Kikuchi patterns, is even better than that value. The reason for which our system with on-axis detection leads to an improved lateral resolution compared to the conventional system is that the intensity of inelastic electron scattering decreases with increasing scattering angle. Hence, more intense Kikuchi patterns are expected at low scattering angles. Consequently, smaller electron probe size can be used, and solvable patterns are obtained from smaller electron-material interaction volumes. Summarizing, a conceptually simple and flexible configuration with on-axis detector of transmission Kikuchi patterns is presented. It allows us to achieve lateral resolution better than five nanometers. Due to this feature, the new TKD configuration seems ideal for investigating nano-grained materials. References: [1] Fundenberger J.J., Morawiec A., Bouzy E., Lecomte J.S. Ultramicroscopy 96(2003),p. 127. [2] E.F. Rauch, M. V�ron, J. Portillo, D. Bultreys, Y. Maniette, S. Nicolopoulos, Microscopy and Analysis 22 (2008) S5�S8. [3] Keller R.R., Geiss R.H., Journal of Microscopy 245 (2012), p.245. [4] Trimby P.W. Ultramicroscopy 120 (2012), p.16. [5]Funderberger J.J., Bouzy E., Goran D., Guyon J., Yuan H., Morawiec A. submitted to Ultramicroscopy. Figure 1. Schematic of the new TKD configuration with horizontally positioned detector.");sQ1[551]=new Array("../7337/1103.pdf","Utility of Precession Electron Diffraction Patterns with Varying Degrees of Nonparallel Illumination from the Same Nominal Sample Area","","1103 doi:10.1017/S1431927615006303 Paper No. 0551 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Utility of Precession Electron Diffraction Patterns with Varying Degrees of Nonparallel Illumination from the Same Nominal Sample Area Peter Moeck1, Jack C. Straton1, Danny J. Edwards2, and Ines H�usler3,1 Nano-Crystallography Group, Department of Physics, Portland State University, Portland, OR 972070751, USA 2. Reactor Materials and Mechanical Design Group, Pacific Northwest National Laboratory, Richland, WA 99354, USA 3. Department of Physics, Humboldt University Berlin, Newtonstra�e 15, 12489 Berlin, Germany Diffraction of a primary electron beam that is precessing, i.e. Precession Electron Diffraction (PED), in a Transmission Electron Microscope (TEM) [1-8], has over the last twenty-odd years become "an invaluable tool, not only for the determination of unknown crystal structures but also in aiding the analysis of local microstructure" [8]. Seven years after its first demonstration [9], the "scanning version of PED" [4-6] rivals in popularity electron backscattering in a scanning electron microscope. The precession axis of the primary electron beam coincides ideally with the optical axis of the TEM and the (sharp) apex of the resulting hollow electron cone intersects the sample at a singular point. This ideal is, however, for a variety of reasons not readily achievable in practice. When the spherical aberration of the objective lens of the microscope is not corrected for and the alignment is not perfect, more or less blunt apexes of electron cones (that may not be completely hollow and round) result in PED. These cones intersect the sample over some area (rather than in a single point) and define the spatial resolution of the PED technique. The cones have also been observed to "wander" (or "wiggle and dance" [7]) over some sample area [3,4]. Correction of spherical aberration of the (probe forming) objective lens pre-field may reduce this "wandering" to approximately 1 nm [7]. Utilizing the recently derived relation t loc ( g 0 ( )) 2 g 0 ( ) , where tloc is the local 2 crystal sample thickness, is the electron wavelength, g0 is the magnitude of the reciprocal lattice vector that determines the radius of the zero-order Laue circle, is the precession half angle (5 � 60 mrad), and is the convergence angle (0.5 � 5 mrad) [10], one can in principle assess the size of these sample areas. This is because the radii of the observed zero-order Laue circles in PED patterns must for a plane-parallel sample plate fall on an "equal thickness" curve in Figure 1. Observed deviations from this functional relationship are indications that the sample is not a plane parallel plate and that the apex of the electron cone is blunt. Utilizing wedge shaped samples (that often arise naturally by the crushing of inorganic crystals), one can then assess the illuminated sample area (as a function of the precession angle) after one has determined the wedge angle by some appropriate technique such as the spacing of two-beam extinction fringes. (The linear increase with convergence angle and the quadratic increase with precession angle [1,7,8] of the size of a blunt electron cone apex intersection with the sample may need to be considered in this context.) In ref. [10] we proposed to utilize this relation (and its generalizations to higher order Laue zones) for the determination of those reflections that can be utilized straightforwardly for structural fingerprinting from PED patterns in a TEM [2]. That idea is based on Boris Vainshtein's critical thickness (tcri) criterion for quasi-kinematic diffraction: , where is the volume of the crystal's unit cell and F hkl is the t cri Fhkl square root of a reflection's intensity (i.e. a kinematical structure factor amplitude) [11]. Large precession angles lead generally to instantaneous "few-beam" conditions [2] that are better approximations to exact two-beam conditions (for which Vainshtein's inequality has been derived originally). For candidate crystal 1. Microsc. Microanal. 21 (Suppl 3), 2015 1104 structures, and F hkl will be known so that one needs to make sure that conditions of t loc t cri are fulfilled for those reflections that one utilizes for structural fingerprinting from PED patterns in a TEM (with fixed ). In summary, observed Laue circle radii in PED patterns as function of the degree of non-parallel illumination can be utilized on the basis of our ideas for both determination of the local sample thicknesses and assessments of the illuminated sample areas. Quasi-kinematic structural fingerprinting of nanocrystals from selected PED spot intensities is also possible on the basis of our ideas. As the precession angle magnitude has so far been rarely exploited as an experimental degree of freedom and because locally determined sample thicknesses are quite valuable, it is safe to predict that these kinds of ideas will prove their worth in the future in a range of TEM investigations of crystalline materials (including 6D electron tomography [12], which covers direct and reciprocal space simultaneously) [13]. [1] R. Vincent and P. Midgley, Ultramicroscopy 53 (1994) 271. [2] P. Moeck and S. Rouvimov, Zeitschrift f�r Kristallographie 225 (2010) 110, open access: http://www.degruyter.com/view/j/zkri.2010.225.issue-2-3/zkri.2010.1162/zkri.2010.1162.xml [3] Y. Liao and L. D. Marks, Ultramicroscopy 117 (2012) 1. [4] P. Moeck et al, Proc. 11th IEEE International Conference on Nanotechnology, August 15-18, 2011, Portland, Oregon, USA, pp. 754-759, DOI: 10.1109/NANO.2011.6144300 [5] P. Moeck et al, Cryst. Res. Technol. 46 (2011) 589. [6] A. S. Eggeman and P. A. Midgley, Adv. Imaging Electron Phys. 170 (2012) 1. [7] A. S. Eggeman et al, Zeitschrift f�r Kristallographie - Crystalline Materials 228 (2012) 43. [8] P. Midgley and A. Eggeman, IUCrJ 2 (2015) 126; open access: http://journals.iucr.org/m/issues/2015/01/00/ro5004/ro5004.pdf [9] E. F. Rauch et al, Microsc. Anal. 22 (2008) S5. [10] P. Moeck et al, Microsc. Microanal. 18 (Suppl. 2), 2012, p. 562. [11] B. K. Vainshtein, Fundamentals of Crystallography, Symmetry and Methods for Structural Crystallography (Springer, 2nd edition, 1994) p. 345; see also B. K. Vainshtein, Acta Cryst. B 47 (1991) 145. [12] A. Eggeman et al, Acta Cryst. A 70 (2014) C368. [13] The first author acknowledges discussions with Prof. Dr. rer. nat. habil Wolfgang Neumann and members of Wolfgang's former research group at the Department of Physics of Humboldt University, Berlin/Germany, where he worked on sabbaticals in 2010 and the summer of 2012. Figure 1. Predicted zero-order Laue circle radii in PED patterns versus degree of non-parallel illumination for plane parallel crystal plate samples with thicknesses between 4 and 64 nm.");sQ1[552]=new Array("../7337/1105.pdf","Comparison of Secondary, Backscattered and Low Loss Electron Imaging for Dimensional Measurements in the Scanning Electron Microscope [1,2]","","1105 doi:10.1017/S1431927615006315 Paper No. 0552 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Comparison of Secondary, Backscattered and Low Loss Electron Imaging for Dimensional Measurements in the Scanning Electron Microscope [1,2] Michael T. Postek1, John Villarrubia2, Andras E. Vladar3 and Atsushi Muto4 1-3. 4. National Institute of Standards and Technology, Physical Measurement Laboratory, Gaithersburg, MD 20899 Hitachi High Technologies America, Clarksburg, MD 20871 Scanning electron microscopes (SEM) are used for dimensional metrology and process control in many production environments. The accuracy of these SEM measurements has always been important, but is often overshadowed by two other main measurement drivers: throughput and precision. It is slow and often tedious to achieve accuracy and, so it is often ignored in production. Accuracy of a measurement is becoming more of an abiding concern as sub-10 nm semiconductor structures are routinely produced. Hence, the metrology error budget has shrunk, and has become only a couple of atoms, i.e., virtually nonexistent. Clever new measurement and signal collection methods applied to sub-10 nm metrology must be sought for all types of semiconductor nanostructures, nanomaterials and nano-enabled materials to ultimately achieve the needed accurate measurements. Achieving good SEM measurement accuracy depends on the quality of the acquired image influenced by vibration, drifts, sample contamination and charging, etc., and accounting for specimen-electron beam interactions. New acquisition methods and successful mitigation of detrimental effects can alleviate some of the imaging problems. But, another key element is the application of advanced electron beam-solid state interaction modeling, such as the NIST JMONSEL [3] model to interpret and account for the physics of the signal generation, and help to understand and minimize the various contributions to measurement inaccuracy. This work is a fundamental comparison of secondary (SE), backscattered (BSE) and low-loss (LLE) electron signals acquired on a new instrument that has high-angle BSE and energy-filtered LLE detectors. Early work indicated that the LLE signal could be advantageous for metrology [4]. When that work was first done, it was very difficult to obtain the needed information because of poor signal-tonoise ratio (SNR) and other instrument-specific geometric limitations. LLE imaging is difficult because LLE represent a small and hence inherently noisier subset of all BSE that have undergone only minimal inelastic interactions with a sample and therefore carry high-resolution, surface-specific information [58]. The use of the conventional backscattered electron signal was shown to be beneficial at low landing energies using a microchannel-plate electron detector [9]. In that work, collection and comparison of the BSE generated images of line structures measured about 10 % narrower, compared to the width of measured SE images [9]. Due to the enhanced emission of low-energy (typically less than 10 eV) electrons at the sides and corners, there are common circumstances in which the SE intensity increases more abruptly at an edge than the BSE intensity. If width assignments are based on an intensity threshold, SE images would then be interpreted as showing a wider feature than the BSE image. It was anticipated that LLE signal would provide results similar to BSE results. A Hitachi SU 8230 [2] FESEM, equipped with a high-angle and energy-filtered backscattered electron detector, was used to compare the SE, BSE and LLE signals for dimensional measurements of the NIST RM 8820 magnification calibration sample [10]. The design of the new in-lens energy filtered detector improves the LLE signal-to-noise ratio and reduces the geometrical limitations of the early LLE detectors. Work is progressing to apply the NIST JMONSEL model to interpretation of the differences in measurements Microsc. Microanal. 21 (Suppl 3), 2015 1106 between the modes of electron collection and to ascertain whether better measurement algorithms can be applied to such measurements. For the first time, point-by point measurement data were able to be obtained simultaneously on a sample using these electron collection modes. Preliminary results (Figure 1) show about a 3 nm difference between the arbitrary 50 % intensity thresholds of SE and LLE images of the nominally 100 nm wide 100 nm tall poly-Si lines on a Si substrate. Therefore, these data are consistent with the results of the earlier work but, in this case, were acquired simultaneously and not serially. Pixel to pixel correlation is now possible. Modeling, to verify and understand this difference further, is currently in progress. Clearly, this points to serious measurement issues encountered by blindly applying measurement algorithms without considering the underlying physics provided by applying model-based metrology. The potential value of BSE and LLE has not been fully exploited for dimensional metrology, but has not been forgotten. Some of the early results and further experimental and modeling work coupled with modeling are sufficiently promising that prompt continued exploration into the possibilities that LLE affords to metrology in standards development and to determine the necessary information related to design parameters necessary for its implementation. SE LLE Figure 1. (Left) SE image of RM 8820, the nominal pitch of the lines shown is 425 nm. (Right) SE and LLE linescans from Reference Material 8820 images simultaneously recorded at 2 kV. The nominal pitch of the lines measured is 200 nm. References: [1]Contribution of the National Institute of Standards and Technology; not subject to copyright. [2]Certain commercial equipment is identified in this report to adequately describe the experimental procedure. Such identification does not imply recommendation or endorsement by the National Institute of Standards and Technology, nor does it imply that the equipment identified is necessarily the best available for the purpose. [3] J.S. Villarrubia et al., Proc. SPIE 6518 (2007) p. 65180K. [4]M. Postek et al., SCANNING 23(5), (2001), p. 298. [5]O. C. Wells, Appl Phys Lett 16(4), (1970), p.151. [6]O. C. Wells, Appl Phys Lett 19(7), (1971), p. 232. [7]O. C. Wells, Scan Electron Microsc, 1, IITRI Chicago, (1972), p. 43. [8]O. C. Wells, Appl. Phys. Lett. 49(13), (1986) p. 764. [9]M. Postek et al., Rev Sci. Instrum, 61(12), (1990), p. 3750. [10]https://www-s.nist.gov/srmors/view_detail.cfm?srm=8820");sQ1[553]=new Array("../7337/1107.pdf","Variation in Band Gap Contrast in Natural Molybdenum Disulphide (MoS2) with BSE Collection Angle and Stage Bias using a Segmented Annular BSED.","","1107 doi:10.1017/S1431927615006327 Paper No. 0553 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Variation in Band Gap Contrast in Natural Molybdenum Disulphide (MoS2) with BSE Collection Angle and Stage Bias using a Segmented Annular BSED. B.J. Griffin1,2, D.C. Joy3 and J.R. Michael4 1 Centre for Forensic Science, The University of Western Australia, Crawley, WA Australia 6009 2 Centre for Microscopy, Characterization and Analysis, The University of Western Australia, Crawley, WA Australia 6009 3 Center for NanoPhase Materials Science, Oak Ridge National Laboratory, Oak Ridge, TN 37831 4 Sandia National Laboratories, PO Box 5800, Albuquerque, NM 87185-0886 Molybdenum disulphide (MoS2 or the naturally occurring `molybdenite') is a wide and variable band gap material. In the crystalline form it exhibits a near-perfect basal cleavage. Backscattered electron imaging of freshly cleaved surfaces exhibit a contrast that details lattice strain effects that give rise to variation in the band gap. This effect is particularly well illustrated in Rutherford backscattered ion images (figure 1) relative to conventional BSEI (Figure 2) but the image contrast is consistent. MoS2 has been widely studied as a potential material in modern photonics and well understood imaging is important. Imaging of a freshly cleaved MoS2 sample using a segmented annular BSED (DBS) with and without stage biasing has provided images with markedly different contrasts as a function of working distance, signal collection angle and stage bias. One set of images, from the different detector rings, are provided in figure 3, collected with a high stage bias (4kV). The first image from the innermost annulus shows strong but diffuse regions in the sample. Imaging with the second annulus, this `diffuse' contrast is reversed. With the third annulus, the image "contrast' has less dynamic range but sharply defines complex band gap variation. Finally, in the image from the outermost annulus a rupture in the sample has contrast that is reversed relative to the images but otherwise the image retains the well-defined band gap contrast. Monte Carlo modelling constrains the depth of image data to ~ 0.2 micron (CASINO v4.2) at a beam landing energy of 5 keV. Modelling of the DBS collection angles using proprietry FEI software (1) indicates collection angles of A: 7 � 12, B: 12 � 19, C: 19 � 27, and D: 27 � 34 degrees for the BSED rings imaged in figure 3, respectively. The sample has been also imaged with a landing energy of 1keV and 10keV, a range of WD and stage bias of 0-4 kV in 1kV increments, and using the ET-SED, upper column `mirror' detector and the pole-piece TLD. A simple polished rock sample has been imaged under the same conditions and Z contrast is maximized in the image from the outermost BSE ring with the stage biased at 4kV (figure 3). With a 4kV stage bias the emitted electrons clearly have significantly modified energy characteristics, changing the nature of the collected signals, as is illustrated. It seems that the lower energy SE are directed up the column towards the `mirror' detector above the TLD. The optimum mass and band gap contrast is found to lie in the outer rings under these conditions, suggesting that the band gap contrast relates to the higher energy SE that are usually coincident onto the ET-SED. Under biased stage conditions, the annular BSED also collects components of the emitted SE signal. The inversion of contrasts in the inner BSED rings will be discussed. A further uncertainty is the reversal of contrast of the hole in the MoS2. References (1) FEI is acknowledged for access to their DBS modelling software. (2) Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United States Department of Energy (DOE) under contract DEAC0494AL85000. Microsc. Microanal. 21 (Suppl 3), 2015 1108 Figure 1: Rutherford backscattered ion image of band gap variation in a cleaved MoS2 sample using a 37 kV He ion. (a) 0V stage bias � MoS2 (b) 4kV stage bias � MoS2 Figure 2: BSEI of band gap variation in the same cleaved MoS2 sample using a conventional FESEM at 20 kV. (c) signal profile from (b) (d) 4kV stage bias - rock Figure 3: From top: ET-SE, DBS-A, DBS-B, DBS-C, DBS-D and TLD images of biased (4kV) and unbiased cleaved MoS2 and a polished rock sample (with vertical carbon ink line down the left-hand side) together with the signal profile along the line shown in the top ET-SEI from the biased condition.");sQ1[554]=new Array("../7337/1109.pdf","Monte Carlo Simulations of Signal Electrons Collection Efficiency and Development of New Detectors for ESEM.","","1109 doi:10.1017/S1431927615006339 Paper No. 0554 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Monte Carlo Simulations of Signal Electrons Collection Efficiency and Development of New Detectors for ESEM. V. Nedla 1, I. Konvalina 1, M. Oral 1, J. Hudec 1 Environmental Electron Microscopy Group, Institute of Scientific Instruments of ASCR, Kr�lovopolsk� 147, 61200 Brno, Czech Republic This paper presents a study of the collection efficiency under various conditions, detector configurations and types of detected electrons in our environmental scanning electron microscope (ESEM) AQUASEM II. It is based on a Monte Carlo simulation of signal electron trajectories in vacuum and gaseous environment in a specially designed ionization detector of secondary electrons equipped with two circular electrodes, a drawing of which is in Fig. 1. Four configurations of detection system were used for simulations. First, electrode A was connected to +350 V and used to collect signal, while electrode B was grounded. Second, electrode A was grounded, while electrode B was connected to +350 V and used to collect signal. Third, +350 V was applied on electrodes A and B while signal was detected only from electrode A. Fourth, the same configuration as the third one, but signal was detected only from electrode B. The simulation consisted of two steps. First, the spectrum of the signal electrons was computed by a Monte-Carlo method in the Geant4 [1]. Second, a dependence of the amplification coefficient on the energy and the pressure was obtained in the EOD (Electron Optical Design). The emission of the signal electrons from the gold surface was simulated using the Geant4 for 1 000 000 particles, the beam energy was 20 keV. The trajectories of signal electrons emitted from the sample to water vapor between the sample and the ionization detector and the signal amplification of the detected electrons were calculated using the EOD [2] software with a Monte Carlo plug-in [3]. 10 000 signal electrons with the energies of 2.7, 4.7, 7.6, 50, 100, 500, 1000, 5000, 10 000, 15 000 and 20 000 eV were traced through a 2 mm thick water vapor region (the sample-to-detection electrode distance) with the pressure of 50, 100, 200, ..., 1000 Pa. The signals detected by the individual segments of the detector provide information about the final energy of the signal electrons. The distribution of the electrostatic fields between the detector and the sample, the spatial distribution of the signal electron emission from the sample as well as the gas type and the pressure all have a significant influence on the detected type of electrons. Simulations of the electron collection efficiency for selected primary beam energies were run for a gold sample, taking into account the energy distribution of the signal electrons. The computed dependencies are in Fig. 2A. The results show, that electrode A collects both lowand high-energy electrons at the pressure of up 200 Pa, while for the range of 200 Pa to 600 Pa the proportion of the high-energy electrons decreases significantly, almost vanishing above 600 Pa. In the case of the second and fourth configuration, with the results in Figs. 2B and 2D, the low-energy electrons are significantly detected in the whole range of pressures; an increasing pressure causes the proportion of the high-energy electrons to increase, especially above 20 keV. In the third configuration, there is such an influence of the electric field of 1 Microsc. Microanal. 21 (Suppl 3), 2015 1110 electrode A on the low-energy signal electrons, that their collection by electrode B is limited significantly. Predominantly high-energy electrons with energies of around 20 keV are detected, as it is shown in Fig. 2C [4]. Figure 1. A) Detection electrodes of the ionization detector for ESEM, B) Experimental setup in ESEM AQUASEM II. Figure 2. The dependence of the normalized number of electrons detected by electrodes A and B, see Fig.1, on the water vapor pressure and the energy of the signal electrons for four selected configurations (First-A, Second-B, Third-C, Fourth- D). References: [1] Agostinelli S et al., Nucl. Instr. and Meth. A 506 (2003), p. 250 [2] Lencov� B, et al. Physics Procedia 1 (2008), p.315. [3] Nedla V, et al., Nucl. Instrum. Methods in Physics A, 645 (1) (2011), p. 79. [4] This work was supported by the grant No. GA 14-22777S and LO1212 together with the European Commission (ALISI No. CZ.1.05/2.1.00/01.0017).");sQ1[555]=new Array("../7337/1111.pdf","Electron Microscopy in Air: Transparent Atomic Membranes and Imaging Modes","","1111 doi:10.1017/S1431927615006340 Paper No. 0555 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electron Microscopy in Air: Transparent Atomic Membranes and Imaging Modes Yimo Han1, Kayla Nguyen2, Yui Ogawa2, Hao Shi3, Jiwoong Park2, David A Muller1,4 1. 2. School of Applied and Engineering Physics, Cornell University, Ithaca, USA Chemistry and Chemical Biology Department, Cornell University, Ithaca, USA 3. Physics Department, Cornell University, Ithaca, USA 4. Kavli Institute at Cornell for Nanoscale Science, Ithaca, USA Environmental Scanning Electron Microscopy (ESEM) where differential pumping and a pressurelimiting aperture [1] has enabled electron imaging in partial atmosphere environments. The typical imaging gas path length (GPL) is 2~5 electron mean free paths (mfp), resulting in lower contrast than a pure vacuum SEM. More recently, the use of thick (100nm) SiNx membranes to separate the electron optics in vacuum and the specimens in atmosphere has led to Atmospheric Scanning Electron Microscopy (ASEM) [2] where imaging in liquid or air without a specimen chamber is made possible by placing specimens in contact with the membrane. However, the resolution and contrast in both ESEM and ASEM are compromised by the multiple electron scattering. By keeping the sample away from the electron transparent membrane, thinner membranes can be used (and reused). This has led to the airSEM [3,4], where a thin SiNx window is used to separate electron optic and air from an optically-aligned sample (Fig 1(a)). The airSEM is able to image specimens in air with high throughput � a few minutes per sample at low magnification [3]. The key to high quality imaging is minimizing the multiple scattering. The mfp of low energy electrons (lower than 10 keV) in SiNx is a few nanometers, much shorter than that in air (tens of microns), indicating the electron scattering within the SiNx window is the main limitation on image contrast for the GPL at which the airSEM operates. Graphene is an ideal window material due to its high mechanical strength, single atomic layer thickness and long electron mfp, which keeps the combined path length to within one mfp, a single-scattering regime comparable in resolution and contrast to vacuum SEM. Monte Carlo simulations (Fig 1 (b)) show the image contrast of gold nano particles imaged through a 2layer graphene window is 2.1 times greater than the contrast through the thinnest commercially available (5 nm) SiNx windows. We have fabricated and tested bilayer graphene windows. The experimental airSEM images at 7 keV of gold nanoparticles in Fig 1 (c) and (d), the signal-to-noise ratio (SNR) for our graphene window are 2.85 times greater rather than for a 10 nm SiNx window. In addition to preserving a smaller probe in airSEM, different imaging modes have been developed to improve the resolution and contrast. A secondary electron detector will not work in air, but by collecting the ion current in air generated mainly by secondary electrons, a surface detector can show better resolution and contrast than a standard backscatter electron (BSE) detector. With the surface detector, we observed single-layer graphene on a copper foil in Fig 2 (b), where the contrast follows that of the secondary electron coefficients of Cu and carbon. We have taken in-air images of untreated biological samples, such as e.coli bacteria in Fig 2 (d), which were prepared simply by drop casting living e.coli bacteria on a conductive optical microscope slide. The surface detector is sensitive to surface features, and shows typical secondary electron edge contrast. We built high-resolution bright field (BF) [4] and dark field (DF) airSTEM detectors for imaging the internal structure of very thin samples. Images of e.coli bacteria in air using BF and DF airSTEM detectors, show their internal structure clearly (Fig 3 (a) and (b)). In addition, spectroscopic detectors, such as EDS and cathodo luminescence (CL), were combined with airSEM without the need for special sample preparation. Fig 3 (c) is an X-ray map taken from an adobe plaster from the Jordan Valley, showing the elemental composition of the grains. Microsc. Microanal. 21 (Suppl 3), 2015 1112 References: [1] D. Stokes, Principles and Practice of Variable Pressure/Environmental Scanning Electron Microscopy (VPESEM). Wiley and Sons, West Sussex, (2008). [2] M. Suga, C. Sato, et al., Ultramicroscopy 111 1650-1658 (2011). [3] b-Nano: http://www.b-nano.com/ [4] K, Nguyen, M. Holtz, D. A. Muller, Microscopy and Microanalysis, 19 (S2) 428-429 (2013). [5]Work supported by the Cornell Center for Materials Research, and NSF MRSEC (DMR-1120296). a) X-ray detector Vacuum BSE Window b) 1 Relative Contrast I/I0 0.8 0.6 0.4 0.2 2-layer graphene window c) SiNx d) Graphene e) window 220 200 180 160 e) window 140 SiNx window 120 100 Surface detector 80 Air 60 Sample STEM detector 0 0 10 20 Window Thickness (nm) 1 �m 40 20 0 0 400 d (nm) 0.1 0.2 0.3 0.4 0.5 Figure 1. (a) Schematic of airSEM; (b) Monte Carlo simulation for 7 keV electrons of the relative contrast for a 2-layer graphene window (Green bar) and SiNx windows with different thickness (Red dots). The contrast has been normalized to that of the gas path without a window, I0. Inset is the TEM image of a 2-layer graphene window, scale bar: 1 �m. AirSEM BSE images of gold nanoparticles at 7 keV electron beam energy with (c) 10 nm SiNx window and (d) 2-layer graphene window; (e) Line profiles of gold particle edges, which are indicated by red line in (c) and green line in (d). a) Scanned electron beam b) Amp c) + bias ne phe Gra - bias d) Cop per 10 �m 10 �m 1 �m Bias sample Figure 2. a) Schematic of the surface detector b) Surface detector image of single-layer graphene grown on Cu foil; c) Single-layer graphene images with positive and negative specimen bias voltage d) Unstained drop-cast living e. coli bacteria image by surface detector. a) b) c) 30 �m 1 �m 500 nm Figure 3. High-resolution a) BF airSTEM image and b) DF airSTEM image of uranium-stained e.coli bacteria; c) X-ray compositional map of an adobe plaster from Jordan Valley.");sQ1[556]=new Array("../7337/1113.pdf","Advances In Variable Pressure Imaging and Detection","","1113 doi:10.1017/S1431927615006352 Paper No. 0556 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Advances In Variable Pressure Imaging and Detection Luyang Han1, Christina Berger1, Markus Boese1, Alexander Thesen1, Fang Zhou1, Stefan Meyer1 and Erik Essers1 1. Carl-Zeiss Microscopy, FE-SEM Department, Oberkochen, Germany The variable pressure scanning electron microscopy (VPSEM) has gained considerable interest since its commercial introduction more than 30 years ago [1]. Unlike a conventional SEM, which functions only in high vacuum, the VPSEM can be operated under a gas pressure in the specimen chamber. This permits observation of non-conductive samples without coating, as well as outgassing specimens, which are normally not compatible with high vacuum. However, due to the skirt effect [2] of the primary electron beam, resolution of a VPSEM is usually significantly reduced compared to a conventional SEM. Furthermore the VPSEM is also particularly troublesome for EDS applications, because additional X-ray signal is produced by the electron skirt. In this contribution we will introduce the newly developed VP implementation of next-generation Zeiss SEM, which features not only a very restricted skirt effect under high gas pressures, but also enables the detection of pure secondary electron (SE) signal using all Zeiss GeminiTM technology in-lens detectors. The essential component of our VP implementation is illustrated in Figure 1. It is composed of a coneshaped pressure limiting aperture called "beam sleeve" and a BSE detector, which are mounted on a pneumatically movable arm. This retractable VP module can be inserted pneumatically with high reproducibility. When inserted, the pressure-limiting aperture permits up to 500 Pa gas pressure in the specimen chamber, while the pressure inside the column and the objective lens is maintained at a much better vacuum. This enables the usage of the 8 kV beam booster during VP operation, and thereby VP in-lens detection and high resolution at low beam energy [3]. The Zeiss InLens and EsB detectors can now detect SE and BSE signals in the VP mode thanks to the beam booster. This improves the SNR of the SE signal significantly, and it makes additionally available the low angle BSE signal. By a voltage of up to 400 V, which can be applied to the pressure limiting aperture, secondary electrons are attracted by the aperture and directed to the InLens detector. For the detection of the high angle BSE signal, a 5segment diode detector is also integrated in the retractable VP module. Three factors enable us to achieve high resolution in our VP implementation. First of all, the electron optic system is not influenced by the VP operation. This results in the same lens aberrations and the same beam spot size compared to the high vacuum mode. Secondly, the cone-shaped pressure limiting aperture helps to keep beam gas path length (BGPL) short. Even for EDS applications the BGPL is kept below 3 mm , which significantly reduces the background X-ray signal. Lastly, the InLens detector detects the true SE signal, which carries only the surface information. Figure 2 demonstrates some VP application images, which are taken with the retractable VP module. The left image shows natural fibers coated with Ag nanoparticles imaged at 80 Pa and 10 kV at relative low magnification. The InLens detector and the EsB detector are used to obtain surface topography and material contrast, respectively, and these two kinds of contrast are clearly separated. The middle image shows a medium magnification of filter paper, imaged at 80 Pa and 3 kV. The charging effect is completely eliminated and an excellent SNR is obtained thanks to the InLens detector. The right image shows at a very high magnification an Au/Pd thin film deposited on glass. Although the Au/Pd thin film Microsc. Microanal. 21 (Suppl 3), 2015 1114 is conductive, the film is not grounded but floating on an insulator. Imaging this sample in high vacuum will cause charge buildup and constant shifting of focus. In VP mode the fine grain structure smaller than 10 nm can be clearly resolved. Overall, the new VP implementation permits operation at higher pressure, with high resolution, pure SE contrast and improved signal detection efficiency. References: [1] G. D. Danilatos, Scanning 3 (1980) p. 215. [2] G. D. Danilatos, "Foundations of Environmental Scanning Electron Microscopy", (Academic Press) p. 109. [3] E. Essers, United States Patent US 6,590,210 B1 Figure 1. Pneumatically retractable VP module with a 5-segment BSE detector and the cone-shaped pressure limiting aperture. Left: Sketch Right: Photo of the inserted beam sleeve under the objective lens. InLens EsB InLens InLens 40 �m Figure 2. VP application examples imaged with the retractable VP module. Left: Ag nanoparticle coated natural fibers imaged with InLens and EsB detectors with low magnification at 80 Pa and 10 kV, sample courtesy of SBUK and Dr. Frank Simon, Leibniz-Institute for Polymer Research Dresden. Middle: Filter paper imaged with medium magnification at 80 Pa and 3 kV. Right: Au/Pd thin film deposit on glass imaged with high magnification at 40 Pa and 5 kV.");sQ1[557]=new Array("../7337/1115.pdf","Quick Freeze Substitution Processing of Biological Samples for Serial Block-face Scanning Electron Microscopy.","","1115 doi:10.1017/S1431927615006364 Paper No. 0557 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quick Freeze Substitution Processing of Biological Samples for Serial Block-face Scanning Electron Microscopy. Richard Webb1 and Robyn Webb1 1. Centre for Microscopy and Microanalysis, University of Queensland, Brisbane, Queensland, Australia Until recently the only way to obtain three-dimensional electron microscopy data over a large area of a biological sample was via serial sectioning, a tedious and technically challenging technique. Serial block-face scanning electron microscopy (SBF-SEM) has recently been used as a technique to obtain similar results to serial sectioning transmission electron microscopy but in a simpler and semi-automated way. It involves working with a resin embedded sample similar to that used for transmission electron microscopy. However, the imaging is performed on the block-face rather than the section cut from it. Image collection is done using a backscattered electron detector in a scanning electron microscope. An image is taken of the block-face and then a thin section is cut from the block using a diamond knife. The newly exposed block-face is then imaged, another section cut from it, it is imaged again etc. This process of imaging and sectioning is performed over and over building up a three-dimensional data set. Two commercial instruments are available that will perform SBF-SEM: the Gatan 3View and the FEI Teneo Volumescope. SBF-SEM suffers from a couple of major problems. As the sample is embedded in resin, a nonconductive material, they charge badly and this can have a significant impact on the imaging quality. As a result a lot of this work has been carried out operating the SEM in variable pressure mode. However, this also effects the image quality. Also standard processing protocols used for preparing samples for biological electron microscopy do not leave enough metal in the sample to produce a strong backscattered electron signal to give good imaging. As a result new protocols have been developed that deposit substantial amounts of metal into the sample in an attempt to overcome both these issues. Nguyen et al [1] and Tapia et al [2] have published protocols that use a combination of metals, including osmium-thiocarbohrdrazide-osmium, en bloc uranyl acetate and en bloc lead aspartate. Most groups are now following the "NCMIR Methods for 3D EM" [3] with small variations on it. The problem with this protocol is that the samples are fixed chemically at room temperature. It has long been known that rapid freezing followed by freeze substitution preserves the morphology of biological samples in a form that is close the native state, avoiding most of the artefacts induced by processing chemically at room temperature. In 2011 McDonald and Webb [4] published a method for performing quick freeze substitution in which they could process samples in less than 3 hours, a process that up until that time was usually performed over several days. The metals introduced into the sample by freeze substitution, usually osmium tetroxide and uranyl acetate, do not impart enough conductivity to make samples produced in this way useful for SBF-SEM. Unfortunately the metals used in the NCMIR methods paper [3] were all in aqueous form so are not useable in a freeze substitution processing protocol. New protocols that utilize the quick freeze substitution in combination with metals in an organic solvent such as acetone or methanol have been shown to produce samples that work well for SBF-SEM (Figs. 1 and 2). In fact, these samples can be readily viewed with the SEM operated in high vacuum mode. Chemicals such as lanthanum chloride or imidazole can be included into the freeze substitution media with the osmium tetroxide and these impart Microsc. Microanal. 21 (Suppl 3), 2015 1116 a high electron density to the sample. A double osmium method can be performed by using tannic acid [5]. Once warmed to room temperature other metal containing solutions such as uranyl acetate, phosphotungstic acid [6] and lead acetate [7] are used to give the sample more electron density. By utilizing the quick freeze substitution and rapid embedding methods [8] the entire process can be completed in a day. . References: [1] J Nguyen et al, PNAS 108 (2011), p. 1176. [2] J Tapia et al, Nature Protocols 7 (2012), p. 193. [3] T Deerinck et al, (2010) available online at: http://ncmir.ucsd.edu/sbfsem-protocol.pdf. [4] K McDonald and R Webb, Journal of Microscopy 243 (2011) p. 227. [5] N Jimenez et al, Journal of Structural Biology 166 (2009) p. 103. [6] H Kushida, Journal of Electron Microscopy 16 (1967) p. 287. [7] H Kushida, Journal of Electron Microscopy 15 (1966) p. 90. [8] K.McDonald, Microscopy and Microanalysis 20 (2014) p. 152. Figure 1. Single image from a 3D data set of a sponge larva taken using SBF-SEM. The sample was processed by quick freeze substitution in imidazole and osmium tetroxide in acetone, followed by tannic acid in acetone, another osmium tetroxide, en bloc uranyl acetate and en bloc lead acetate. Figure 2. 3D data set of cultured insect cells infected with baculovirus, processed in a manner similar to the sample shown in Figure 1.");sQ1[558]=new Array("../7337/1117.pdf","The Acquisition of Large Datasets with Serial-Block Face Scanning Electron Microscopy following Light Microscopy Image Acquisition: Workflows from Sample Preparation to Image Acquisition","","1117 doi:10.1017/S1431927615006376 Paper No. 0558 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Acquisition of Large Datasets with Serial-Block Face Scanning Electron Microscopy following Light Microscopy Image Acquisition: Workflows from Sample Preparation to Image Acquisition Christel Genoud, Adrian Wanner, Rainer Friedrich Friedrich Miescher Institute for Biomedical Research, Basel, Switzerland 3D imaging of tissue ultrastructure by serial block face scanning electron microscopy (SBEM) is a promising approach to reconstruct large neuronal circuits in order to analyze their computational functions as well as to localize molecules in cells in culture. However, various steps in SBEM can still be improved to exploit the full potential of the method. In order to correlate light microscopy signals with ultrastructure to reconstruct for example synaptic connectivity among neurons in the zebrafish olfactory system or to identify the cell compartment in which a protein of interest is located, we have developed tools and refined various steps in the imaging pipeline for SBEM with an ultramicrotome in the vacuum chamber. These modifications are necessary to acquire a SBEM dataset corresponding to the field of view that has been imaged by fluorescence imaging in a reasonable time frame. First, we optimized protocols to acquire stacks of EM images with high signal-to-noise ratio. The protocols include modifications of the protocol by Deerinck et al for SBEM [1] and the tiling of large fields of view. Second, we developed software workflows for stitching of tiles, image registration, and other basic tasks based on TrakEM2 [2]. Third, software tools have been used to identify structures in stacks of EM images that were visualized by fluorescence microscopy in the same samples before fixation. Using Matlab and TrakEM2, a semiautomated approach has been used to strictly register in three dimensions 2 modalities of imaging, in our case fluorescent and SBEM stacks. Forth, a company has been initiated that offers tracing and synapse annotation in stacks of EM images by highly trained tracers. Applications of these methods include the reconstruction of all neurons in the olfactory bulb of a zebrafish larva. In summary, these methods have substantially increased the speed and quality of our SBEM approaches and are likely to be useful for a wide spectrum of applications in which acquisition of large datasets are necessary and the correlation with fluorescent data acquired in vivo are required. Microsc. Microanal. 21 (Suppl 3), 2015 1118 Figure 1. Once the fluorescent stack is acquired (upper left) and the SBEM stack is acquired and stitched (upper right), the 2 modalities are registered (lower panel) in order to match the activity recorded in vivo and the ultrastructural morphology of the neurons References: [1] TJ Deerinck, E Bushong, A Thor, MH Ellisman, Microscopy (2010), p. 6-8 [2] A Cardona, S Saalfeld, J Schindelin, I Arganda-Carreras, S Preibisch, M Longair, P Tomancak, V Hartenstein, RJ Douglas, Plos ONE 7 (2012)");sQ1[559]=new Array("../7337/1119.pdf","In Situ Tomography of Membrane Proteins Enabled by Advanced Cryo-FIB Sample Preparation and Phase Plate Imaging","","1119 doi:10.1017/S1431927615006388 Paper No. 0559 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In Situ Tomography of Membrane Proteins Enabled by Advanced Cryo-FIB Sample Preparation and Phase Plate Imaging Miroslava Schaffer1, Benjamin D. Engel1, Marc Wehmer1, Tilak Gupta1, Eri Sakata1, Tim Laugks1, Julia Mahamid1, J�rgen M. Plitzko1 and Wolfgang Baumeister1 Max Planck Institute of Biochemistry, Department of Molecular Structural Biology, Am Klopferspitz 18, 82152 Martinsried, Germany The development of cryo-focused ion beam (cryo-FIB) microscopy into a highly reliable sample preparation technique for frozen-hydrated specimens has recently enabled cryo-electron tomography (CET) of eukaryotic cells with unprecedented resolution and image quality [1,2,3]. The ability to prepare distortion-free lamellas of homogenous, user-defined thickness has allowed in situ studies of cellular structures in their native state, revealing new insights into biological processes [4]. Additionally, cryo-fluorescence microscopy has been combined with cryo-FIB for the in situ targeting of fluorescently-labelled cellular structures [5]. While cryo-FIB milling has increased the accessibility of cellular volumes, thereby widening the biological applications of CET, independent developments in low-dose TEM imaging have pushed the boundaries of CET to new detection limits. The latest generation of direct detection cameras greatly reduces noise and thereby improves sensitivity. Additionally, a new type of TEM phase plate, the Volta phase plate (VPP), can strongly increase in-focus phase contrast of thin biological specimens with relatively little additional experimental effort [6]. However, sample charging in lamellas prepared by cryo-FIB has proven to be highly detrimental to VPP imaging. In this work, we show that this obstacle can be overcome by adding a step to the cryo-FIB workflow. Coating the surfaces of a freshly prepared lamella with a thin, sputtered layer of platinum (~1 nm) sufficiently reduces charging during CET without strongly impeding image quality. However, in cases where the small contrast reduction due to this Pt coating of the lamella must be avoided, an alternative method can be employed. Prior to FIB milling, a sandwich-type coating layer is deposited by applying a conventional sputtered Pt layer, followed by a thick organometallic protective layer deposited by the FIB gas injection system, and then a final sputtered Pt layer (~20nm). Charging on the subsequentlymilled lamella in close proximity to this thick protective layer is sufficiently reduced to allow CET in combination with VPP imaging. Here, we describe the modified cryo-FIB sample preparation workflow. We further show that the combined application of this workflow with CET, using both direct electron detection and VPP imaging, has enabled the first in situ visualization of integral membrane proteins in eukaryotic cells. [1] M Marko et al, Nat Methods 4(3) (2007) p.215. [2] A Rigort et al, PNAS 109(12) (2012) p.4449. [3] E Villa et al, COSTBI 23(5) (2013) p.771. [4] B Engel et al, eLife (2015) 4:e04889. [5] Y Fukuda et al, Ultramicroscopy 143 (2014), p.15. [6] R Danev et al, PNAS 111(44) (2014) p.15635. 1. Microsc. Microanal. 21 (Suppl 3), 2015 1120 Figure 1. Cryo-FIB sample preparation of Chlamydomonas reinhardtii cells. SEM image (a) and TEM image (b) of a finished lamella. Figure 2. Slice from a reconstructed tomogram revealing the chloroplast of a Chlamydomonas reinhardtii cell. Overview (a) and detail (b) images of thylakoid membranes, clearly showing ATP synthases bound to the exterior thylakoid surfaces and photosystem II complexes projecting into the thylakoid interiors.");sQ1[560]=new Array("../7337/1121.pdf","Imaging of Vitrified Biological Specimens by Confocal Cryo-Fluorescence Microscopy and Cryo-FIB/SEM Tomography","","1121 doi:10.1017/S143192761500639X Paper No. 0560 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Imaging of Vitrified Biological Specimens by Confocal Cryo-Fluorescence Microscopy and Cryo-FIB/SEM Tomography Alexander Rigort1, Robert Kirmse1, Volker D�ring1, Michael Schwertner2 and Ralf Wolleschensky1 1. 2. Carl Zeiss Microscopy, Jena, Germany, Email: alexander.rigort@zeiss.com Linkam Scientific Instruments, Tadworth, UK The investigation of vitrified biological specimens (i.e. samples that are plunge or high-pressure-frozen) enables the visualization of cellular ultrastructure in a near-native fully hydrated state, unadulterated by harmful preparation methods. Here, we focus on two recent cryo-imaging modalities and discuss their impact on cryo-correlative workflows. First, we present confocal cryo-fluorescence microscopy, utilizing a novel confocal detector scheme with improved signal-to-noise ratio (SNR) and resolution. Second, we show volume imaging of multicellular specimens by cryo-focused ion beam scanning electron microscopy (FIB/SEM) . Confocal laser scanning microscopes (LSM) are renowned for their optical sectioning capability, a feature enabled by utilizing a pinhole that rejects out-of-focus light. Closing the pinhole improves lateral resolution, but also causes less light to reach the detector leading to reduced signal-to-noise ratios. In cryo-fluorescence microscopy the situation is aggravated by the fact that currently no immersion optics are available and consequently only numerical apertures below NA 1 are possible. Here we combined Airyscan, a novel detector module (available for ZEISS LSM 780, 800 and 880) together with a cryocorrelative stage (Linkam CMS196) for cryo-fluorescence imaging of vitrified Dictyostelium discoideum cells prepared on electron microscopy grids. The Airyscan detection module allows the spatiallyresolved detection of fluorescence light otherwise rejected by the pinhole in a standard confocal system. We show that under cryo-conditions even without immersion optics a significant increase in resolution and SNR can be obtained with Airyscan compared to standard confocal images (Figure 1). In FIB/SEM tomography three-dimensional volumetric data from biological specimens is obtained by sequentially removing material with the ion beam and imaging the milled block faces by scanning with the electron beam. Only recently, it has been shown that this imaging method can be applied also to frozen-hydrated specimens [1]. Cryo-FIB/SEM tomography allows the mapping of large multicellular specimens in the near-native state and is particularly suited to analyze samples that require remaining hydrated. We demonstrate this approach by imaging the cryo-immobilized nematode Pristionchus pacificus (Figure 2). Both methods by themselves promise significant advantages for biomedical research by being able to investigate biological specimens in the near-native fully hydrated state. Yet correlating both imaging modalities, LSM and FIB/SEM of vitrified samples, has the potential to provide even deeper insights into biological context. In contrast to correlative workflows using resin-embedded samples, cryoimaging workflows do not have to find the optimal balance between preserving cellular ultrastructure and maintaining the functional integrity of fluorophores. Moreover, major mechanisms leading to irreversible bleaching of fluorescent molecules are suppressed at cryo temperatures [2]. The correlation between cryo light and electron microscopy data will greatly benefit from an increase in resolution in fluorescence imaging. Cryo-Airyscan is a first step in that direction and can deliver three-dimensional Microsc. Microanal. 21 (Suppl 3), 2015 1122 optical sectioning data that can be used to reliably target cellular structures in a FIB/SEM microscope, before the structural context is explored by cryo-FIB/SEM tomography. References: [1] Schertel A et al., "Cryo FIB-SEM: volume imaging of cellular ultrastructure in native frozen specimens." J Struct Biol. 2013 Nov;184(2):355-60. doi: 10.1016/j.jsb.2013.09.024. [2] Kaufmann R, Hagen C, Grunewald K. Fluorescence cryo-microscopy: current challenges and prospects. Current opinion in chemical biology. 2014;20:86-91. DOI: 10.1016/j.cbpa.2014.05.007 Figure 1. Confocal cryo-fluorescence beampath (a). Comparison between confocal cryo-fluorescence (b) and Airyscan (c) image of vitrified D. discoideum cells expressing mRFP-Tubulin. Scale bar: 10 �m. Figure 2. Cryo-FIB/SEM tomography of high-pressure frozen P. pacificus (a). In-lens SEM micrographs showing the milled cryo-block faces exhibiting a cross-sectional view of the worms' intestine (b,c). Scale bars: (a,b) 10 �m, (c) 1 �m.");sQ1[561]=new Array("../7337/1123.pdf","Growth of Transition Metal Oxides in Solution under Liquid Cell Electron Microscopy and Electron Beam Effects","","1123 doi:10.1017/S1431927615006406 Paper No. 0561 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Growth of Transition Metal Oxides in Solution under Liquid Cell Electron Microscopy and Electron Beam Effects Wen-I Liang1,2, Haimei Zheng1,3 1. 2. Materials Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, CA 94720 Department of Materials Science and Engineering, National Chiao Tung University, Hsin-Chu, Taiwan 300 3. Department of Materials Science and Engineering, University of California, Berkeley, CA 94720 Liquid cell electron microscopy provides an unique platform for the study of nanomaterials growth and transformations in their working environment at the nanometer or atomic scale. Many unseen mechanisms of colloidal nanocrystal growth and self-assembly have been discovered; the electrochemical reactions including lithiation-delithiation of cathode materials and metal dendrite deposition are also visualized. Although numerous successful implements of liquid cell transmission electron microscopy (TEM) have been demonstrated, there is limited understanding of the complex chemistry between precursor, solvent and electron beam so far. Especially, the role of electron beam in the growth of nanocrystals in an organic solvent under TEM is still unclear. The study of nanocrystal growth in solution under the electron beam is not only important for the understanding of electron beam induced nanoparticle formation, but also crucial for managing the electron beam effects in the study of electrochemical processes and other chemical reactions using TEM. We used the growth solution of 0.3 mole precursors (nickel(II) acetylacetonate and iron(III) acetylacetonate) dissolved in 1 ml of surfactant/solvent (oleylamine, oleic acid, and benzyl ether) then loaded several microliter solution in a closed liquid cell with SiNx membranes (15nm on each side of window). The liquid cell was then sealed and loaded in JEOL in-situ 3010 LaB6 TEM operating at 300 kV. During growth, the beam intensities were recorded at the order of 105 A/m2. The movies were captured 10 fame per second using Gatan Orius 833 camera and analyzed by ImageJ software. Under electron beam induction, directional dendritic growth is found (Figure 1). By tracing each branches, we summarized the growth kinetics in Figure 1b. In Figure 1c, the as-grown sample under high-angle annual dark field (HAADF) image shows a light contrast layer encapsulating the bright dendrite, where the contrast is attributed largely to thickness (Z number for Ni and Fe are similar). Energy dispersive X-ray spectroscopy (EDS) mapping provides us abundant information on elemental distribution. As shown in Figure 1d, the dendrite is shown as blue contour and surrounding area are green from the X-ray signal integration, suggesting more Fe can be found in surrounding. Crystal structure is identified under high-resolution TEM. It is worth noting that low dose imaging condition is employed in order to minimize the perturbation of electron beam also prevent solution drying during image. Figure 1e shows several lattice image captured during growth and the crystal structure of Ni1+xFe2-xO4 has been identified. The reaction mechanisms of precursor and solvent with the electron beam will be discussed in great details [1]. References: [1] TEM facilities at the Molecular Foundry of Lawrence Berkeley National Laboratory is supported by the U.S. Department of Energy (DOE) under Contract No. DE-AC02-05CH11231. HZ thanks the Microsc. Microanal. 21 (Suppl 3), 2015 1124 1118 funding support from U.S. DOE Office of Science Early Career Research Program. WIL acknowledges the funding support from Ministry of Science and Technology (MOST) in Taiwan (NSC 102-2119-I009-502). Figure 1. a: Time sequencial images of Ni-Fe-O dendritic growth in a liquid cell under TEM. Scale bars equal to 20nm. b. kinetic growth of each dendrite. c. EDS mapping, showing the elemental distribution. d. EDS spectrums acquired from dendrite (blue) and surrounding region (green). e. High-resolution TEM imaging captured during growth.");sQ1[562]=new Array("../7337/1125.pdf","The Stability of Gold Nanoparticles in Liquid Scanning Transmission Electron Microscopy Experiments Studied under Varied Conditions.","","1125 doi:10.1017/S1431927615006418 Paper No. 0562 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Stability of Gold Nanoparticles in Liquid Scanning Transmission Electron Microscopy Experiments Studied under Varied Conditions. Andreas Verch1, Justus Hermannsd�rfer1 and Niels de Jonge1,2. 1. 2. INM � Leibniz Institute for New Materials, Campus D2 2, 66123 Saarbr�cken, Germany Department of Physics, University of Saarland, Campus A5 1, 66123 Saarbr�cken, Germany Studying dynamical processes such as the self-assembly of nano-particles in liquid with nanometer resolution is possible using a recently established technology for transmission electron microscopy (TEM) in liquid [1]. The sample containing liquid and solid parts are placed in a microfluidic chamber between two electron transparent membranes. The principal setup is schematically shown in Fig. 1. However, interactions between the liquids and the electron beam dramatically increase the complexity of the system and complicate the analysis of the observed processes. High energy electrons excite electronic states in the specimen and the ample liquid resulting in the generation of highly reactive, transient species such as solvated electrons, hydrogen-, and hydroxide radicals. Their appearance initiates a cascade of reactions in which various other powerful oxidizing and reducing agents are formed. These species are often able to attack the solid specimen chemically, which might change its surface properties or actually dissolve it. In many cases this behavior is not desired and it might even render the experiment useless. In order to design experiments resembling the unperturbed specimen as closely as possible, it is thus essential to improve our understanding of the processes occurring in liquid electron microscopy experiments, and to learn how to reduce the impact of the electron beam on the specimen. In a liquid TEM experiment, the steady state concentrations of the different reactive species and even the pH are significantly influenced by parameters such as the initial pH of the solution and the imaging conditions, e.g. the electron dose, as simulations based on equilibrium and rate constants of the aforementioned transient and other involved species have suggested.[2]. Hence, a pH-sensitive experiment, for example, might have an unexpected outcome, since the pH of the solution may be changed as a result of the electron beam interactions. In this work, we studied the stability of gold nanoparticles in a liquid scanning TEM experiment as function of several varied solution parameters, and imaging conditions. We varied the pH, the sodium chloride concentration, the ionic strength of the aqueous phase, and the electron dose introduced into the solution. Our experiments were conducted using a liquid TEM holder (Poseidon, Protochips, NC, USA) in a CS-corrected STEM/TEM (ARM200f, JEOL. Japan) at 200 kV acceleration voltage. The microscope was operated in scanning mode in order to obtain a better handle on the doses introduced into the area observed. Our experiments showed that the addition of sodium chloride strongly influenced the stability of the gold nanoparticles. At neutral pH conditions the addition of chloride ions decreased the redox potential for gold oxidation, leading to an accelerated dissolution of the nanoparticles. The influence of the pH on the underlying reactions became obvious when the pH is changed. While a decrease of the pH merely the gold dissolution rate rates, high pH values significantly altered the behavior of the gold nanoparticles in presence of chloride ions. Instead of dissolving, the nanoparticles drew nearer and eventually coalesced within a few second. This observations confirmed the complex interplay of oxidizing and reducing species with the original nanoparticles, the dissolved-, and the precipitated gold. Video stills of experiments at different pH conditions illustrating this behavior are shown in Fig. 2. References: Microsc. Microanal. 21 (Suppl 3), 2015 1126 [1] N. de Jonge and F. M. Ross, Nature Nanotechnology 6 (2011), pp. 695. [2] N. M. Schneider et al., The Journal of Physical Chemistry C 118 (2014), pp. 22373. We thank Protochips Inc, NC, USA for providing the microchips and the Liquid STEM specimen holder. We thank E. Arzt for his support through INM. Research in part supported by the Leibniz Competition 2014. Figure 1. Principle setup of a liquid STEM experiment. The liquid is enclosed between two electron transparent membranes. The interaction of the electron beam with the liquid results in the formation of various reactive species. These transient species start a cascade of subsequent reaction, which might eventually also alter the specimen. Figure 2. Stills of videos recorded with Liquid STEM showing the behavior of gold nanoparticles at different solution pH in the presence of 2 mol L-1 NaCl. (A) At pH 12 the gold nanoparticles approach each other and eventually grow together. The electron dose rate per image was 12 e- �-1 s-1. (B) The gold nanoparticles rapidly dissolve at pH 7. Electron dose: 3 e- �-1 s-1. Both scale bars represent 200 nm.");sQ1[563]=new Array("../7337/1127.pdf","Automated and Shaped-Controlled Liquid STEM Nanolithography","","1127 doi:10.1017/S143192761500642X Paper No. 0563 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Automated and Shaped-Controlled Liquid STEM Nanolithography Raymond R. Unocic1,2, Andrew R. Lupini1,3, Albina Y. Borisevich1,3, Sergei V. Kalinin1,2, and Stephen Jesse1,2 1. 2. Institute for Functional Imaging of Materials, Oak Ridge National Laboratory, Oak Ridge, TN 37831 Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, TN 37831 3. Materials Science & Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831 There is a range of scanning probe and electron beam nanolithography platforms currently available for producing nano-scaled structures with complex shapes and chemistries from solid or gaseous phase precursors [1-2]. Recent studies have shown that the same concept can be applied to liquid phase precursors using liquid cells and electron beams with manual manipulation [3-4]. In the present work, an automated electron beam control system has been developed to precisely control the position and residence time of the STEM probe from a Cs aberration-corrected FEI Titan STEM operating at 300kV. The position of the STEM probe is controlled by applying bias waveforms to the X- and Y- STEM scan coils such that the focused STEM probe can be positioned in the X-Y plane either in a fixed position with a pre-defined dwell time or rastered in the X-Y plane with a variable translational velocity. This system has been used for nanolithographic patterning of nm-scale 3D structures from liquid phase precursor solutions. An in situ liquid cell TEM holder was used to encapsulate a 10 �M H2PdCl4 aqueous solution between two 50 nm thick silicon nitride membranes. Radiolysis occurs when the electron beam interacts with the solution, forming complex radical species [5]. As a result, aqueous electrons (eaq-) are generated in the liquid, which effectively act as a reducing agent to reduce metallic Pd from the H2PdCl4 complex [6]. Thus, the principals of radiation chemistry within closed liquid cells combined with automated electron beam control can be exploited to nanolithographically-pattern structures of metallic Pd. The Institute of Functional Imaging of Materials (IFIM) acronym was used to demonstrate the concept of liquid STEM nanolithography. The focused STEM probe was rastered through the liquid cell containing the 10 �M H2PdCl4 growth solution and Pd metal was reduced forming Pd nanocrystals directly onto the silicon nitride membrane and within the liquid cell through a nucleation and growth mechanism. An annular dark field (ADF) STEM image of the patterned "IFIM" text is shown in Figure 1a. As a result of chemically sensitive ADF STEM imaging, whereby intensity is proportional to the atomic number and thickness, the nanolithographically patterned Pd has a higher intensity when compared against the lower atomic number background of the silicon nitride membranes and the H2PdCl4 growth solution and therefore the electron beam reduced Pd structures can readily be characterized following patterning. It is clear by the size-scale and legibility of the IFIM letters shown in Figure 1a, that the nanolithography technique can be used to precisely control the position of the STEM probe and resultant Pd crystal deposition. Corresponding higher magnification images shown in Figures 1b-d, reveal that the pattern is composed of Pd nanocrystals having individual crystal sizes on the order of tens of nanometers with the exception of a few larger crystals (Figure 1b). It is interesting to note that the Pd nanocrystals are primarily confined within the local irradiated region of the IFIM pattern, e.g., there is limited nucleation and growth of Pd outside of this area. For liquid STEM nanolithography, precision-controlled nanosculpting is Microsc. Microanal. 21 (Suppl 3), 2015 1128 governed by the interplay between electron dose, electron beam exposure time, diffusivity of eaq-, and nucleation and growth kinetics of the electron beam deposited crystals [7]. References: [1] R Garcia et al, Nature Nanotechnology 9 (2014) p. 577-587. [2] I Utke et al, Journal of Vacuum Science and Technology B 26 (2008) p. 1197-1276. [3] JM Grogan et al, Nano Letters 15 (2014) p. 359-364. [4] MWP van de Put et al, Small 11 (2015) p. 585-590. [5] NM Schneider et al, Journal of Physical Chemistry C 118 (2014) p. 22373-22382. [6] KL Jungjohann et al, Nano Letters 13 (2013) p. 2967-2970. [7] Research supported by Oak Ridge National Laboratory's Center for Nanophase Materials Sciences (CNMS), which is a U.S. Department of Energy, Office of Science User Facility. (RRU, SVK, and SJ) Additional support provided by the Division of Materials Sciences and Engineering Division, Office of Basic Energy Sciences, U.S. DOE. (ARL and AYB) Figure 1. a) ADF STEM image showing the "IFIM" text nanolithographically patterned within a STEM liquid cell using a 10 �M H2PdCl4 precursor growth solution. Scale bar is 500 nm. b-d) Higher magnification ADF STEM images from several regions of the pattern showing the detailed nanocrystalline nature of the deposited Pd.");sQ1[564]=new Array("../7337/1129.pdf","Manipulating Size and Shape of Silica Nanoparticles with Liquid-Phase Transmission Electron Microscopy","","1129 doi:10.1017/S1431927615006431 Paper No. 0564 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Manipulating Size and Shape of Silica Nanoparticles with Liquid-Phase Transmission Electron Microscopy Jovana Zecevi1 , Krijn P. de Jong1 and Niels de Jonge2, 3 1. 2. Utrecht University, Inorganic Chemistry and Catalysis, Utrecht, The Netherlands. INM - Leibniz Institute for New Materials, Saarbr�cken, Germany. 3. Department of Physics, University of Saarland, Saarbr�cken, Germany Recently developed liquid-phase transmission electron microscopy (LP-TEM) holds great potential for dynamic imaging of relevant processes taking place in liquids at nanoscale, such as the evolution of nanoparticles during synthesis or structural changes of nanomaterials under working conditions [1]. LPTEM studies performed in the past years mainly focused on the growth of metallic nanoparticles from solutions, and pointed to important effects the electron beam can have on systems under observation [2]. Amongst various electron beam effects observed in LP-TEM, radiolysis of water and aqueous solutions leads to the formation of species such as eaq -, H �, OH �, H2 , H2O2 , H +, OH -. These species can locally change the chemistry in the LP-TEM cell, triggering reactions such as reduction of metal ions. In fact, electron beam induced nucleation of metal nanoparticles has been one of the main focuses of the LPTEM studies so far [2]. The goal of our work was to study the influence of the electron beam on silica nanoparticles in water. We imaged about 40 nm-diameter silica spheres in a water environment. To prepare a sample, a droplet of sonicated silica-ethanol suspension (~0.5 l) was placed on top of a Si microchip with SiN window (Protochips Inc., NC, USA), allowing the ethanol to evaporate within a minute, so that the silica spheres firmly attached to the SiN window. A 0.3 l droplet of water was then placed on the microchip, and the LP-TEM cell was closed by placing a second microchip on top, with its SiN window facing downward (Fig. 1). Having the silica spheres fixed to the SiN window allowed us to study electron beam effect on individual silica nanoparticles, as opposed to having freely moving silica particles in water, which were reported to create several hundreds of nanometers deposits on the SiN window upon scanning with a focused electron beam [3]. Imaging was performed in scanning TEM (STEM) mode at 200 keV beam energy at different magnifications (using an ARM200F, JEOL, Japan). Fig. 2a shows a STEM image (20 s dwell time) of a silica agglomerate fixed to the top SiN window. Upon longer exposure to focused scanning electron beam (50 frames, total of ~5 min), elongation occurred of the spherical individual silica particles, and of the silica agglomerate, in the direction of the line scanning (Fig. 2b). This observation points towards the presence of a two-fold effect of the focused electron beam on the silica particles and their environment. Firstly, the formation of OH - species upon electron beam induced water radiolysis might have led to a local increase of the pH inducing a local dissolution of the silica nanoparticles. Secondly, a fraction of the dissolved silica seems to have been redeposited on the sides of the silica agglomerate in horizontal direction, i.e. in the scanning direction. Deposition of silica upon electron beam irradiation in a liquid environment has been observed in the earlier study [3] but not with a directional component as observed here. To further test this effect, we changed the direction of scanning by a 90 degrees rotation, and exposed the particles from Fig. 2b to another ~ 5 min of scanning, now with a vertical line scan direction. As can be seen in Fig. 2c, the elongated silica particles from Fig. 2b returned to a nearly spherical shape, while the elongation of the whole agglomerate was reduced. The size of the agglomerate also seems to have been reduced. Possibly, Microsc. Microanal. 21 (Suppl 3), 2015 1130 dissolution and redistribution of silica species has taken place within the interparticle pores of the agglomerate, leading to the fusion of individual silica particles, a loss of pore space, and a reduction of the total volume of the agglomerate. In conclusion, these results show that a focused electron beam can break silica bonds, and can induce the re-deposition of silica onto silica nanoparticles. The electron beam can thus reshape and resize silica nanoparticles in aqueous environment. Observed phenomena can potentially be utilized for nanoscale manipulations, and are also relevant for future LP-TEM investigations of silica based samples. References: [1] N. de Jonge, F. M. Ross, Nat. Nanotech. 6 (2011) p. 695. [2] J. E. Evans et al. Nano Lett. 11 (2011) p. 2809. [3] M. W. P. van de Put et al., Small (2014) DOI: 10.1002/smll.201400913 [4] J.Z. and K.P.dJ. acknowledge funding from the NRSCC and an ERC Advanced Grant. NdJ thanks Protochips Inc., NC, USA, for providing the LP-TEM system. Figure 1. Schematic representation of experimental setup, showing liquid-phase TEM holder containing liquid cell composed of two Si microchips with 50 nm thin SiN window and water layer in between. Silica sample was deposited on the top microchip to enhance the imaging resolution. Figure 2. Electron beam induced morphological changes of agglomerate of silica particles in water, imaged with STEM. (a) STEM image of an agglomerate recorded at 600,000x magnification. (b) STEM image of silica agglomerate after 5 min scanning in horizontal direction. (c) STEM image of silica agglomerate from (b) after 5 min scanning in vertical direction.");sQ1[565]=new Array("../7337/1131.pdf","Materials Science Applications of Aberration Corrected TEM and/or STEM.","","1131 doi:10.1017/S1431927615006443 Paper No. 0565 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Materials Science Applications of Aberration Corrected TEM and/or STEM. Gustaaf Van Tendeloo1, Artem Abakumov1, Sara Bals1, Sandra Van Aert1, Jo Verbeeck1 1. EMAT Research Group, Department of Physics, University of Antwerp, B-2020, Belgium The introduction of aberration corrected (AC) lenses has evidently introduced large changes in the electron microscopy world. Even more important though are the changes in the world of materials science, solid state physics and solid state chemistry. The new possibilities of AC-EM have boosted the development of new materials and new products or provided extra information about the functioning of these products. Major improvements originate from the combination of an increased signal to noise ratio, the possibility to work at variable voltages, the development of better detectors, the improvement of algoritms for 3D electron tomography, the incorporation of monochromators for EELS analysis and the improved recording of EDX spectra. A recent example is the search for the origin of voltage decay in high capacity layered oxide electrodes in Li-based batteries. Lithium rich layered oxides (Li1+xNiyCozMn1-x-yO2) are attractive electrode materials with very high energy densities (16% over todays commercial cells). However they are known to suffer from voltage decay upon cycling. Using a double corrected TEM-STEM instrument we could study the migration of cations between the metal layers and the Li-layers and prove unambiguously that the trapping of the metal ions in the interstitial tetrahedral sites is at the origin of the voltage decay (see figure 1). Expanding the study to Li2Ru1-ySnyO3 and Li2RuO3 we could figure out that the slowest decay occurs for cations with the largest ionic radius [1]. Also more fundamental aspects in nanoscience can now be tackled. The combination of electron tomography and EELS has allowed us to map the valency of the Ce ions in CeO2-x nanocrystals in three dimensions. There is a clear facet-dependent reduction shell at the surface of the ceria nanoparticles; {111} surface facets show a low surface reduction, whereas at {001} surface facets, the cerium ions are more likely to be reduced over a larger surface shell (see figure 2). Our generic tomographic technique allows a full 3D data cube to be reconstructed, containing an EELS spectrum in each voxel. This possibility enables a three-dimensional investigation of a plethora of structural and chemical parameters such as valency, chemical composition, oxygen coordination, or bond lengths, triggering the synthesis of nanomaterials with improved properties [2]. Even in identifying or restoring famous paintings advanced EM can now play a crucial role. We investigated the deterioration of historical chrome yellow paint in the paintings of Van Gogh, in order to allow an optimal reconstruction of his works. This is particularly done by a quantitative analysis and mapping of the EELS data of chromium. The original pigments in late 19th Century consisted of PbCrO4, PbCr1-xSxO4, and PbSO4 particles. During aging, the particles have gradually evolved to core�shell PbCrO4�Cr2O3, or core�shell PbCrO4�PbSO4 particles together with some remaining PbSO4. An artificial aging process speeds up the evolution and shows that they will evolve towards Cr2O3, PbCrO4�PbSO4� Cr2O3, or PbSO4� Cr2O3 core�shell structures. Greenish Cr2O3 deteriorates the originally clean yellow colour. All those intermediate or final states have been observed and analysed by AC STEM [3]. The authors acknowledge financial support from the European Research Grants (ERC Grants) for GVT, SB and JV. [1] M. Sathiya et al. Nature Materials (2014) DOI: 10.1038/NMAT4137 [2] B. Goris et al. ACS Nano 8 (2014) p. 10878 [3] H. Tan et al. Angew. Chem. Int. Ed. 52 (2013) p. 11360 Microsc. Microanal. 21 (Suppl 3), 2015 1132 Figure 1. [1 1 0] HAADF-STEM image of pristine Li2Ru0.75Ti0.25O3 (left) and after 50 chargedischarge cycles. Extra cations at the tetrahedral interstices locally appear after cycling (marked with arrowheads). Figure 2. (a and d) HAADF-STEM reconstructions of near-perfect and truncated octahedral ceria nanoparticle. (b and e) The corresponding 3D visualizations and slices through the 3D reconstructions showing the valency results for Ce3+ and Ce4+, indicating a thicker Ce3+ layer with more oxygen vacancies at the {001} truncation. (c and f) Slices through the Ce3+ and Ce4+ volumes.");sQ1[566]=new Array("../7337/1133.pdf","50pm Aberration Corrected In-situ Electron Microscopy - How Ion behaves in Lithium Ion Nanowire Battery","","1133 doi:10.1017/S1431927615006455 Paper No. 0566 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 50pm Aberration Corrected In-situ Electron Microscopy - How Ion behaves in Lithium Ion Nanowire Battery K. Takayanagi1,4, S.Lee2,4, Y.Oshima3,4, Y.Tanishiro2,4 1. 2. 3. 4. Physics Department, Tokyo Institute of Technology, Tokyo, Japan. Quantum Nanoelectronics Research Center, Tokyo Institute of Technology, Tokyo, Japan. School of Materials Science, JAIST, Ishikawa, Japan. CREST-JST, Tokyo, Japan. Lithium ion batteries (LIBs) attract much interest, and fundamental understanding of mechanisms enabling longer life for rapid charge-discharge and large capacity in use, particularly, of automobiles. Charge-discharge cycle causes structural phase transition [1] of electrode materials, and rapid cycle results generally in an irreversible cycle. Any irreversible change of the electrode structure can result in capacity fading and/or fracture of the electrode [2]. In order to study dynamical process of charge-discharge cycle, in-situ electron microscopy with electrochemical information has been devised to give rich results [3, 4]. An aberration corrected (AbC) electron microscopy (Roo5 microscope) [5] has been applied to solve lithium ion diffusion and phase transition process in LIB, as shown in Fig.1. The electron microscope allowed us to detect the number of lithium ions within each atomic columns of our specimen as demonstrated in Fig.2 [6]. The LIB we developed for in-situ study consists of a free standing LiMn2O4 (LMO) nanowire, ionic liquid electrolyte (ILE) and Li4Ti5O12 of the negative electrode. The LMO nanowires had lengths of 0.1~0.2mm and diameters of 100nm~200nm. The finest electrode has a single LMO nanowire (Fig.3). The nanowire-LIB was found to work without capacity fading at high charge/discharge rate (20 minutes for charge of 10nC), and structure of LMO transformed reversibly while the cyclic voltammetry between 3.5 � 5.5 V (Li/Li+), as shown in Fig.4. The nanowire contained three areas having different ion density; lithium-rich, phase boundary (PB), and lithium-poor areas. These areas were moved in a cyclic manner along the nanowire to help the transport of lithium ions (Fig.5). Because of this phase boundary, the nanowire electrode had neither fracture, nor capacity fading. Therefore, the LMO nanowire electrode offers opportunity of long life even at high charge rate. The electron microscope (Roo5) has achieved a sub-50pm resolution, and has capability of resolving lithium ions which are diffusing within the spinel crystals [7]. Future issue and challenge of in-situ AbC microscopy is to image every step of the dynamic procedure of ionic Microsc. Microanal. 21 (Suppl 3), 2015 1134 charge and potential transport, which cannot be detectable by other method. References: [1] M. M. Thackeray et al, Mater. Res. Bull. 18 (1983) p.461. [2] J. Vetter et al, J. Power Sources 147(2005) p.269. [3] S. Lee et al, J. Phys. Chem. C 117 (2013) p.24236. [4] S. Lee et al, DOI: 10.1021/nn505952k (2014) [5] H. Sawada et al, J. Electron. Microsc. 58 (2009) p.357. [6] S. Lee et al, J. Appl. Phys., 109 (2011) p.113530. [7] S. Lee et al, Jpn. J. Appl. Phys. 51 (2012) p.020202. Fig.1 In-situ system of nanowire LIB. developed for Roo5 electron microscope Fig.2 Individual lithium ion imaging by Annular Bright-Field detector Fig.3 Single nanowire LIB developed for in-situ electron microscopy Fig.4 Cyclic voltammograph of a single nanowire LIB in Fig.3 Fig.5 Phase transition of a LMO nanowire during charge-discharge cycle");sQ1[567]=new Array("../7337/1135.pdf","Depth-Resolution Imaging of Crystalline Nano Clusters on/in Amorphous Films Using Aberration-Corrected TEM","","1135 doi:10.1017/S1431927615006467 Paper No. 0567 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Depth-Resolution Imaging of Crystalline Nano Clusters on/in Amorphous Films Using Aberration-Corrected TEM Jun Yamasaki1, Akihiko Hirata2, Yoshihiko Hirotsu3, Kaori Hirahara4 and Nobuo Tanaka5 1. Research Center for Ultra-High Voltage Electron Microscopy, Osaka University, Ibaraki, Osaka, Japan. 2. Advanced Institute for Materials Research, Tohoku University, Sendai, Japan. 3. Institute of Scientific and Industrial Research, Osaka University, Ibaraki, Osaka, Japan. 4. Department of Mechanical Engineering, Osaka University, Suita, Osaka, Japan. 5. EcoTopia Science Institute, Nagoya University, Nagoya, Japan. The improved image resolution in the depth direction (the electron incident direction) should be another advantage using aberration-corrected TEM. We have demonstrated the availability for obtaining 3D information of nano materials in the previous papers [1,2], in which the detection of tilt of a carbon nanotube (CNT) and the selective imaging of each side-wall lattice of a single-wall CNT were successfully achieved. In the present study, the features and capabilities of the depth-resolution imaging were investigated using image simulations and experiments for two types of samples [3]. The first sample was gold clusters attached on an amorphous carbon film. The experimental through-focal series indicated that the focal plane for the cluster was shifted 3 nm from that for the supporting film (Fig. 1(a) and (b)). This difference is due to the depth-resolution imaging of the cluster and film, the mid-planes of which are separated by 3 nm along the depth direction. On the basis of this information, the three-dimensional configuration of the sample, such as the film thickness of 2 nm, was successfully illustrated in Fig. 1(c). The second sample was a Zr66.7Ni33.3 metallic glass including a medium-range-order (MRO) structure, which was approximately considered to be a crystalline cluster with a diameter of 1.6 nm. In the experimental through-focal series, the lattice fringe of the MRO cluster was visible at limited focal conditions (Fig. 2(a) and (b)). Image simulations reproduced well the focal conditions (Fig. 2(d) and (e)) and also indicated a structural condition for the visualization that the embedded cluster must be apart from the mid-plane of the matrix film. Similar to the case of the first sample, this result can be explained by the idea that the "effective focal planes" for the film and cluster are at different heights. This type of depth-resolution phase contrast imaging is possible only in aberration-corrected TEM and when the sample has a simple structure and is sufficiently thin for the kinematical scattering approximation. References: [1] N. Tanaka, J. Yamasaki, et al., Nanotechnology, 15 (2004), 1779-1784. [2] K. Hirahara, et al., Nano Lett., 6 (2006), 1778-1783. [3] J. Yamasaki, et al., Ultramicroscopy (2014), doi:10.1016/j.ultramic.2014.11.005. [4] This study was partly supported by Grant-in-Aids for Scientific Research on Priority Areas "Materials Science of Bulk Metallic Glasses" (No.18029011) by MEXT and on Innovative Areas "3D active site science" (No.26105001) by JSPS. Microsc. Microanal. 21 (Suppl 3), 2015 1136 Figure 1. (a),(b) Aberration-corrected TEM images of a Au cluster on an amorphous carbon film taken at different focal positions, the distance of which is 3 nm. (c) Schematic of the sample configuration derived based on the focal distance and the diameter of the Au cluster. Figure 2. (a),(b) Aberration-corrected TEM images of a Zr66.7Ni33.3 glass taken at the different focal positions indicated below the images. (c) The approximate trace of the lattice fringe in the cluster indicated by the arrow in (b). (d),(e) Image simulations reproducing (a) and (b), respectively. They were calculated using a structure model of amorphous Zr66.7Ni33.3 including inside the crystalline cluster shown in (f).");sQ1[568]=new Array("../7337/1137.pdf","Strain Control at Two-Dimensional Oxide Interfaces Probed by Aberration-Corrected STEM-EELS","","1137 doi:10.1017/S1431927615006479 Paper No. 0568 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Strain Control at Two-Dimensional Oxide Interfaces Probed by AberrationCorrected STEM-EELS David J. Baek1, Di Lu2, Yasuyuki Hikita3, Harold Y. Hwang2,3 and Lena F. Kourkoutis4,5 1. 2. School of Electrical and Computer Engineering, Cornell University, Ithaca, NY, USA. Department of Applied Physics, Stanford University, Stanford, CA, USA. 3. Stanford Institute for Materials and Energy Sciences, SLAC National Accelerator Laboratory, Menlo Park, CA, USA. 4. School of Applied and Engineering Physics, Cornell University, Ithaca, NY, USA. 5. Kavli Institute at Cornell for Nanoscale Science, Cornell University, Ithaca, NY, USA. Perovskite oxides display a wide spectrum of functional properties such as superconductivity, colossal magnetoresistance, and ferromagnetism that can potentially be utilized for electronic device applications. Advances in the atomic-scale synthesis of these materials have enabled the design of novel oxide heterostructures exhibiting properties that are not found in their bulk constituents. Strain, for example, imposed on the overlaying epitaxial oxide film by a carefully chosen substrate, can alter the material's properties. Here, we use an alternative approach to controlling the strain in Nd0.5Sr0.5MnO3 (NSMO) films by inserting a flexible perovskite-like buffer layer (TSDA: Sr3Al2O6) between the film and the SrTiO3 (100) (STO) substrate [1]. Atomic-resolution mapping of the structure and chemistry at the NSMO/TSDA/STO interface reveals regions with uniform and defect-free buffer layers separated by regions where the buffer layer is lacking and where strain in the film is partially relieved by the formation of dislocations. The 1-nm wide dislocation cores are Mn rich and show a reduced Mn valence. Previous work showed that when NSMO films are grown directly on STO substrates, misfit dislocations appear at the interface due to the 2.84% lattice mismatch (aSTO=3.905�, aNSMO=3.797�). The average spacing between the dislocation cores was found to be ~30nm [2]. However, in this work, the flexible nature of the TSDA buffer layer allows strain relief and the average spacing between dislocations is greatly enhanced to ~100nm. Aberration-corrected ADF STEM imaging and spectroscopic mapping reveal that the dislocation cores are only present in areas that lack the uniform growth of the TSDA layer and which therefore lack an alternative strain release mechanism. The elemental maps in Fig. 1 correspond to an area with uniform growth of TSDA and interestingly, we observed selective Nd diffusion into the TSDA layer which suggests that the open crystal structure of TSDA allows preferential diffusion into the cation sites. However, in Fig. 2, the ADF STEM image and the elemental maps illustrate the missing TSDA layer around the dislocation core. From the Mn and Ti maps, the dislocation core is observed as being occupied by excess Mn and the corresponding site being Tideficient. Previously, similar diffusion of cations from the film into the open structure of the dislocation core was reported [3]. When the Mn-L2,3 edge is integrated parallel to the interface, at the dislocation core, the peaks shift to lower energies suggesting a reduced Mn valence state. When compared with MnL2,3 reference spectra, the Mn in the dislocation core corresponds to Mn3+ while that in NSMO is Mn3.5+. The Mn valence change was further confirmed by performing multivariate curve resolution (MCR) to extract the two distinct Mn components as presented in Fig. 2. A map of the spatial extents of these two components is obtained by a non-least squares fit to the Mn-L2,3 edge spectroscopic map using the reference spectra extracted by MCR. [1] D Lu et al, MRS Spring (2014), K12.11. Microsc. Microanal. 21 (Suppl 3), 2015 1138 [2] C. -P Chang et al, Nat. Commun. 5 (2014), p. 1. [3] M Arredondo et al, Adv. Mater. 22 (2010), p. 2430. [4] This work was supported by the Cornell Center for Materials Research with funding from the NSF MRSEC program (DMR-1120296) and the Department of Energy, Office of Basic Energy Sciences, Division of Materials Sciences and Engineering, under contract DE-AC02-76SF00515. Figure 1. Upper figure: low magnification HAADF-STEM image showing the overall structure of the material under study. Lower figures from left to right: simultaneously recorded ADF image, elemental maps of Ti, Mn and Ti, Nd (white circles: selective diffusion of Nd), concentration profile extracted from the elemental maps with red arrows indicating the location of Nd diffusion (scale bars: 2nm). Figure 2. Upper figures: Closer look into the TSDA layer and the dislocation core, followed by elemental maps of Ti and Mn around the defect. Lower figures: Background-subtracted edge spectra with horizontal averaging (leftmost ADF indicates the area used for averaging). Rightmost mapping shows two distinct Mn components that are extracted by multivariate curve resolution. For the defect component, the major peaks of the Mn-L2,3 edge shift to lower energies as shown in the left spectra (scale bars: 2nm).");sQ1[569]=new Array("../7337/1139.pdf","The Use of Auger Spectroscopy for the Elemental Analysis of Presolar Silicate Grains.","","1139 doi:10.1017/S1431927615006480 Paper No. 0569 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Use of Auger Spectroscopy for the Elemental Analysis of Presolar Silicate Grains. Christine Floss1 1 Laboratory for Space Sciences and Dept. of Physics, Washington University, St. Louis, MO 63130, USA. Presolar silicate grains, formed in the outflows of evolved stars and in the ejecta of stellar explosions (supernovae), survived the formation of the solar system and are found in low abundances in primitive extraterrestrial materials. They are recognized by their isotopic signatures, which differ substantially from those observed in other solar system materials. These grains can be studied in the laboratory to gain a better understanding of stellar evolution and nucleosynthesis of the elements. They also provide information about conditions in the stellar sources in which they formed and the environments they have traversed subsequent to formation, including the interstellar medium (ISM), the early solar nebula and the parent bodies of the meteorites in which they are found. The isotopic composition of a presolar grain provides information about its stellar origin [1], but additional elemental and/or mineralogical information can constrain the conditions under which a grain condensed, and is necessary for comparison with astronomical data [2]. Auger spectroscopy is a well-established surface analytical technique in the materials sciences, which takes advantage of the emission of electrons with characteristic energies (Auger electrons) from a sample irradiated by an electron beam. Like X-ray energies measured in an electron microprobe or scanning electron microscope, the kinetic energies of Auger electrons provide information about sample composition. Because Auger electrons originate only from the top few nanometers directly under the electron beam, the spatial resolution for Auger spectroscopy is primarily dependent on the primary electron beam diameter and is on the order of tens of nanometers, ideally suited for the elemental characterization of presolar silicates, which typically have diameters of 200�300 nm. The development of Auger spectroscopy for the analysis of presolar grains is discussed by [3]. Specific analytical protocols were established for the analysis of presolar silicates to reduce the possibility of electron beam damage, which can occasionally produce artifacts in the Auger spectra. These include use of a low beam current, beam rastering over the grain of interest and acquisition of multiple specral scans that are added together to obtain a single Auger spectrum. Auger spectra are typically subjected to a 7point Savitsky-Golay smoothing and differentiation routine prior to peak identification (Fig. 1). Sensitivity factors for the major rock-forming elements have been determined from olivine and pyroxene standards in order to quantify data from the spectral measurements. In addition, highresolution (10�20 nm) elemental distribution maps can provide detailed qualitative information about the distribution of elements within and around the grains of interest (Fig. 2). The elemental compositions of over 400 O-rich presolar grains have been measured by Auger spectroscopy. The vast majority of the grains analyzed are ferromagnesian silicates. While some grains have stoichiometries consistent with either olivine or pyroxene, a significant fraction have nonstoichiometric compositions. Irradiation, shock and sputtering processes in the interstellar medium can induce changes in the structures, chemical compositions and porosities of the grains [4] and may play a role in transforming stoichiometric grains into grains with non-stoichiometric compositions. Microsc. Microanal. 21 (Suppl 3), 2015 1140 Alternatively, non-stoichiometric compositions may be due to condensation under kinetic conditions. Presolar silicates also have higher than expected Fe concentrations. Secondary alteration has played a role in some meteorites [e.g., 5], but most of the meteorites with high presolar silicate abundances show little evidence of alteration, suggesting that much of the Fe is primary, the result of non-equilibrium condensation of the grains in the stellar envelopes where they formed. These inferences are consistent with modeling work on the formation of dust in O-rich stars, which suggests that most silicate dust forms during short high mass-loss episodes that occur after thermal pulses in thermally pulsing AGB stars [6]. The stellar environment during these episodes will be highly variable, with strong stellar winds and rapid temperature drops, leading to grain formation under kinetic conditions. References [1] Nittler L. R. et al., Astrophys. J. 682 (2008), p. 1450. [2] Henning T., Ann. Rev. Astron. Astrophys. 48 (2010), p. 21. [3] Stadermann F. J. et al., Meteorit. Planet. Sci. 44 (2009), p. 1033. [4] Carrez P. et al., Meteorit. Planet. Sci. 37 (2002), p. 1599. [5] Floss C. and Stadermann F. J., Meteorit. Planet. Sci. 47 (2012), p. 1869. [6] Gail H.-P. et al., Astrophys. J. 698 (2009), p. 1136. [7] Floss C. and Stadermann F. J., Geochim. Cosmochim. Acta 73 (2009), p. 2415. Figure 1. Differentiated Auger spectrum of a presolar ferromagnesian silicate from the CR3 chondrite QUE 99177 [7]. Figure 2. Secondary electron image and Auger elemental maps of a presolar composite grain from the Adelaide ungrouped carbonaceous chondrite [5].");sQ1[570]=new Array("../7337/1141.pdf","Practical Aspects of the Electron Probe Analysis of Boron in Silicates using a LSM Device with Large 2d","","1141 doi:10.1017/S1431927615006492 Paper No. 0570 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Practical Aspects of the Electron Probe Analysis of Boron in Silicates using a LSM Device with Large 2d George B Morgan VI Electron Microprobe Laboratory, University of Oklahoma, Norman, USA The development of Layered Synthetic Microstructures (LSM; also called Layered Dispersion Elements-LDE and, inaccurately, PseudoCrystals-PC) through the 1980's provided electron probe analysts using wavelength-dispersive spectrometry (WDS) access to the K-alpha emissions of the ultralight elements down at least to boron (Z=5) in the periodic table[1]. Although quite a bit has been reported subsequently about analytical methods for boron[2], little discussion has been directed at optimizing analysis on the basis of the type of diffraction element used or the effect of matrix type on the intensity and shape of the background (Bremsstrahlung) surrounding B K in silicate or borate materials. Regardless of the type of diffraction device, the B K peak is very broad and measuring its intensity from most silicate and borate matrices is complicated by sharply increasing background intensity on the high-energy side of the peak (HEB: High-Energy Background) (Figure 1). Using an LSM with 2d in the range of ~147-200 �, diffraction of B K occurs at 2<<45o (~27.4-19.8o) and so removes most of the polarization-induced changes in peak shape as a function of orientation from crystalline materials with symmetries between those of the cubic and triclinic systems. The original LSMs with 2d in the 145-200 range are Mo-B4C interlayer devices. The first principal task of analysis with these devices is choosing a peak position and background offsets (or single offset and slope) that minimize or eliminate intensity from internal fluorescence of the B4C interlayers (Figure 1). Pulse height analysis methods need to utilize the Differential Mode in order to eliminate high-order diffractions of higher-energy x-ray emissions that, although reputedly are eliminated by LSMs, occur both under the B K peak and in the HEB. Differences in absorption and background intensity, slope, and shape make boron metal and simple boride compounds poor choices as standards for analyzing silicate or borate materials. WDS scanning of a variety of silicate materials demonstrates that the intensity and slope of the background around the B K peak changes with increasing average atomic number (ZAvg) and density () of the sample, making the match between standard and sample important. Among silicates, the background slope decreases with increasing ZAvg and such that significant intensity from internal fluorescence is counted in Ferich, high-Z materials; this yields significant fictive boron unless the offset on the HEB is widened (or the background slope is modeled appropriately) relative to methods appropriate for materials with lower ZAvg. The analysis of crystals is made more straightforward by using crystalline primary standards that match the chemistry and coordination environments for boron in the unknowns as closely as possible. Average Z and density effects also lead to an increase in background intensity and slope for glasses relative to crystalline materials such that the use of a glassy primary standard for analyzing minerals may lead to fictive positive boron contents. Conversely, the use of crystalline standards for glass analysis can lead to inaccurately low results for boron that can include negative net peak intensity at zero concentration. The best approach for glass analysis is to use a high-boron (10 wt% B2O3) silicate glass as the primary standard, and then two or more secondary standards - one with nil and another with intermediate boron content - to evaluate accuracy over a range of concentration down to the detection limit. Moreover, using an anhydrous glass standard for boron, the analysis of a pair of initially identical anhydrous and hydrated B-free glasses produced 0.4-0.6 wt% fictive B2O3 in the hydrated glass that is Microsc. Microanal. 21 (Suppl 3), 2015 1142 not seen in the anhydrous glass, perhaps due to a decrease in background intensity accompanying lowered Z and , and/or increased adsorption by higher oxygen content. With the above concerns taken into account, routine analysis of boron in silicates easily obtain minimum detection limits in the 0.2-0.3 wt% B2O3 range with 30-60 second count times using 15 kV acceleration (and lower concentration at lower accelerating voltage). Such limits, however, depend strongly on the thickness of the entrance window to the proportional counter. Polymer support grids or inadequately stretched polypropylene can reduce intensity of the B K peak by as much as 70%, thereby increasing the minimum detectable concentration. [1] G.F. Bastin and H.J.M Heijligers, in "Electron Probe Quantitation", ed. K.F.J Heinrich and D.E. Newbury, (Plenum, New York), p. 145-175. [2] J.J. McGee and L.M. Anovitz, in "Boron Mineralogy, Petrology and Geochemistry", ed. E.S. Grew and L.M. Anovitz, (Mineralogical Society of America, Washington D.C.), p. 771-788. B-Nitride ~181.8 eV Sin = 0.34200 12345- (ZAvg ~11.04, ~2.4 g/cc) (ZAvg~10.72, ~2.6 g/cc) (ZAvg ~11.46, ~2.7 g/cc) (ZAvg ~15.36, ~3.3 g/cc) (ZAvg ~18.69, ~4.7 g/cc) 4 3 2 5 1 207 eV 155 eV 207 eV 155 eV Figure 1. Wavelength-dispersive spectrometer scans (15 kV, 100 nA) across B K region using LSM with 2d ~200 on quartz (SiO2), tourmaline [(Na0.6Ca0.1)(Fe1.6Mg1.2Al0.2)Al6(BO3)3Si6O18(OH)4], hambergite (Be2BO3(OH), danburite (CaB2Si2O8), and boron nitride. Intensity observed on quartz, which is nominally B-free, is due to internal fluorescence within the Mo-B4C interlayers of the LSM. Figure 2. WDS scans showing decrease in the background slope around B K with increasing ZAvg and density of the sample. Note that at the chosen peak position (dashed line), the background slope for the default peak offsets used for B-bearing silicates with ZAvg ~11.6 (heavy line) does not eliminate internal fluorescence from the B4C interlayers of the LSM for high-density, Fe-rich silicates (circled).");sQ1[571]=new Array("../7337/1143.pdf","CHILI, a Nanobeam Secondary Neutral Mass Spectrometer with Extraordinary Spatial Resolution, Sensitivity, and Selectivity: First Results.","","1143 doi:10.1017/S1431927615006509 Paper No. 0571 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 CHILI, a Nanobeam Secondary Neutral Mass Spectrometer with Extraordinary Spatial Resolution, Sensitivity, and Selectivity: First Results. A. M. Davis1,2,4, T. Stephan1,2,3, R. Trappitsch1,2, M. J. Pellin1,2,3,4, D. Rost1,2,3, M. R. Savina2,3, and N. Dauphas1,2,4. 1 2 Department of the Geophysical Sciences, The University of Chicago, Chicago, IL 60637, USA. Chicago Center for Cosmochemistry. 3 Materials Science Division, Argonne National Laboratory, Argonne, IL 60439, USA. 4 The Enrico Fermi Institute, The University of Chicago, Chicago, IL 60637, USA. CHILI, the Chicago Instrument for Laser Ionization, is a nanobeam secondary neutral mass spectrometer using laser resonance ionization that is designed for a lateral resolution as small as 10 nm, a useful yield (atoms detected per atom desorbed) of 30�50%, and nearly complete suppression of isobaric interferences from both monatomic and molecular ions [1 and references therein]. It is equipped with an Orsay Physics Cobra liquid metal ion gun with a lateral resolution of 2.5 nm for desorbing from small spots and rastered areas (less than a few �m) and a 351 nm laser for desorbing from spots or rastered areas from the �m scale upwards. An optical microscope is built into CHILI and is used to image the sample and introduce the ablation laser. An Orsay Physics e�CLIPSE-plus field emission electron gun with a resolution of 4 nm is also used for imaging samples. CHILI is equipped with six tunable Ti:sapphire lasers that we designed and built, pumped by three Photonics DM-40 40 W Nd:YLF lasers (527 nm) for simultaneous isotopic measurement of two or three elements. The Ti:sapphire lasers are tunable from ~700�1000 nm, but can be frequency doubled (350�500 nm), tripled (233�333 nm) or quadrupled (205�250 nm) with nonlinear optical crystals. We anticipate a lateral resolution of ~10 nm for resonance ionization using the Ga+ primary beam, as this is the approximate analytical volume for sputtering; for laser ablation, the minimum spot size will be ~1 �m. CHILI is housed within a dedicated laboratory with controlled temperature and humidity. The analytical sequence is the following: release atoms from the surface by sputtering; eject secondary ions by applying high accelerating voltage to the cloud; apply the normal accelerating voltage; fire the photoionization lasers; and mass analyze the photoions. CHILI's removable sample holder is capable of mounting a variety of common types of samples, including 1" diameter polished sections, �" and 1 cm diameter SEM stubs, and TEM samples. As an initial test of CHILI, we analyzed some presolar SiC grains from mount KJG#2. The grains were extracted from the Murchison CM2 meteorite more than 20 years ago [2] and represent samples of stars that lived and died before the solar system formed. In contrast to recent work on KJG grains [3�5], the samples in this study were not additionally treated with concentrated acids to remove parent-body or terrestrial contamination. The grains were mounted on a high purity gold foil by depositing them from a suspension and pressing them into the gold with a sapphire window. Prior to analysis, energy dispersive X-ray images of the mount were acquired in a scanning electron microscope to locate SiC grains on the gold foil; 22 grains were selected for this study. Using previously developed ionization schemes, the Ti:sapphire lasers were tuned for resonance ionization of strontium, zirconium, and barium. Isotopic standards used in this study are NIST SRM 855a with 180 ppm Sr, Zr metal, NIST SRM 1264a with 690 ppm Zr, and terrestrial BaTiO3, all of which were assumed to be of normal terrestrial isotopic composition. Data were corrected for dead time effects; the detector dead time was found to be 600 ps, which was determined by analyzing Zr metal by ion sputtering with varying primary ion beam current. Instrumental isotopic fractionation was determined to be smaller than the statistical error of typically a Microsc. Microanal. 21 (Suppl 3), 2015 1144 few tens of except for 91Zr, which showed enhanced laser resonance ionization by ~210 due to an odd-even effect as had been previously observed [6] and which is corrected for by analyzing standards. Ten grains had sufficient strontium and/or barium content for isotopic analysis with CHILI. However, none of the grains had detectable zirconium. Strontium and barium isotopic ratios, normalized to 86Sr and 136Ba, respectively, are displayed in Fig. 1 as values (deviation from standards in ). For comparison, Fig. 1 shows data obtained with an earlier generation instrument (CHARISMA) on SiC grains that had undergone acid cleaning [3�5]. Most of the SiC grains analyzed in the present study show a similar range of isotope ratios as in the previous studies, and are consistent with formation in asymptotic giant branch stars [3�5]. There seems to be a tendency to plot closer to normal than the previously analyzed grains; residual parent-body or terrestrial contamination could be responsible for such a trend. One of the grains has substantial excesses in 88Sr and 138Ba. This grain is likely from a core-collapse supernova, and the pattern is due to a neutron burst shortly after the supernova explosion. Two grains show significant depletions in 87Sr and 88Sr, patterns that have not yet been explained. A wide variety of cosmochemical problems will be explored with CHILI, including isotopic and chemical studies of presolar grains, refractory inclusions from meteorites, samples of the Sun returned to Earth by the Genesis spacecraft, and cometary and interstellar dust returned by the Stardust spacecraft. References [1] T Stephan et al. Lunar Planet. Sci. 45 (2014), #2242. [2] S Amari et al. Geochim. Cosmochim. Acta 58 (1994), 459. [3] N Liu et al. Astrophys. J. 786 (2014), #66. [4] N Liu et al. Astrophys. J. 788 (2014), #163. [5] N Liu Dissertation, The University of Chicago (2014), 181 pp. [6] GK Nicolussi et al. Science 277 (1997), 1281. 1500 Mainstream grains [12] Mainstream grains CHILI Sr-depleted grains X-grain 87 1500 1000 1000 ( Sr/ Sr) 500 136 13X 84 86 87 88 86 8X 0 ( Ba/ 0 -500 -1000 130 -500 -1000 Ba) 500 132 134 135 136 137 138 Mass number Mass number Figure 1. Sr and Ba isotopic patterns of presolar SiC grains. Error bars are �2.");sQ1[572]=new Array("../7337/1145.pdf","Electron Microprobe Analysis of Complex Y-REE-Ta-Nb-Ti Minerals from the Petaca Pegmatite District, New Mexico","","1145 doi:10.1017/S1431927615006510 Paper No. 0572 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electron Microprobe Analysis of Complex Y-REE-Ta-Nb-Ti Minerals from the Petaca Pegmatite District, New Mexico Michael N. Spilde1, William P. Moats2, Steve Dubyk3 and Brain Salem 4 1. 2. Institute of Meteoritics, University of New Mexico, Albuquerque, New Mexico, 87131, USA 8409 Fairmont Drive NW, Albuquerque, New Mexico, 87120, USA 3. 1828 Quiet Lane SW, Albuquerque, New Mexico, 87105, USA 4. PO Box 27, Tijeras, New Mexico, 87059, USA High-tech electronic applications, such as high-energy density batteries and super strong magnets, have brought about renewed interest in locating strategic deposits of rare earth element (REE) minerals. All types of REE deposits are being investigated around the world. Granitic pegmatites represent an uncommon source of REE minerals, in addition to providing other industrial minerals such as feldspar or beryl. The Petaca Pegmatite District, in north-central New Mexico, contains at least 88 individual rareelement pegmatite bodies in an area about 7 km wide by 24 km long, as described by Jahns [1]. Many of the pegmatite bodies were mined for sheet and scrap mica from about 1870 through the end of World War II, with some mining continuing into the late 1950's and perhaps as late as 1965. REE-bearing Nb-Ta oxides are among the more interesting minerals that occur in the Petaca district. However, the identities of many of the minerals are sometimes tentative, and for many years, published information simply classified all black, metamict minerals as "samarskite." Due to the uncertainty of which particular REE-bearing Nb-Ta oxides actually occur in the Petaca District, we are in the process of conducting a district-wide survey of the Petaca REE minerals. To date, Y-REE-Nb-Ta-Ti oxides and other minerals have been sampled from 30 individual pegmatites for analysis by electron microprobe. Based on the results of our microprobe study, several new minerals were identified or confirmed for the district, including euxenite-(Y), samarskite-(Y), polycrase-(Y), xenotime-(Y), betafite, microlite, pyrochlore, and uranmicrolite. See Table 1 for formulas. The Y-REE-Ta-Nb-Ti oxide minerals examined in this study are highly complex. Figure 1 shows examples of the complexity of the minerals, including primary inclusions (Fig. 1A), exsolution (Fig. 1B), overgrowth (Fig. 1C), and alteration (Fig. 1D). Zoning within minerals, such as Ta exchanging for Nb across a columbite crystal, reflects changes in the composition of the pegmatite fluid during primary growth of the crystal. Furthermore, late enrichment of titanium and tantalum in the late stages of pegmatite crystallization resulted in Ta-rich minerals (struverite, and microlite) forming around the margins of pre-existing minerals (Fig. 1C), extending from the existing AB2O6 structure. Secondary alteration by late pegmatitic fluids further complicates the picture. Alteration by these late fluids is apparent in Fig. 1D where xenotime replaces samarskite along fractures, and infiltration of fluids rich in U, Ta, and Ca has formed microlite that overprints all the other phases. Differentiating between Y-REE-Ta-Nb-Ti oxide minerals is hampered not only by the complex chemistry of the minerals, but also by the fact that they are often metamict and display weak or no X-ray diffraction patterns. Ewing [2] used a statistical approach to classify the Y-REE-Ta-Nb-Ti oxides, and Ercit [3] further refined the technique using principle component analysis (PCA). We take the same approach, using a three-group statistical model. The PCA plot in Figure 1E indicates that most of the differentiation is found in light (LREE) vs. heavy REE (HREE) associated with primary crystallization. Microsc. Microanal. 21 (Suppl 3), 2015 1146 For example polycrase has higher levels of HREE but euxenite has lower levels of HREE, while samarskite exhibits higher levels of LREE. Ta, Ca, and Pb, which are associated with microlite and pyrochlore resulting from late stage alteration, are strongly anti-correlated with all other elements. Table 1. Y-REE-Ta-Nb-Ti Minerals identified at the Petaca pegmatites. Mineral Name Aeschynite-(Y) Betafite Columbite-(Mn) Euxenite-(Y) Microlite Monazite-(Ce) Polycrase-(Y) Pyrochlore Samarskite-(Y) Uranmicrolite Xenotime-(Y) Ideal Formula (Y,Ca,Fe)(Ti,Nb)2(O,OH)6 (Ca,U)2(Ti,Nb,Ta)2O6(OH) (Mn,Fe)(Nb,Ta)2O6 (Y,Ca,Ce)(Nb,Ta,Ti)2O6 (Na,Ca)2Ta2O6(O,OH,F) (Ce,La,Nd,Th)PO4 (Y,Ca,Ce,U,Th)(Ti,Nb,Ta)2O6 (Na,Ca)2Nb2O6(OH,F) (Y,Fe3+,U)(Nb,Ta)5O4 (U,Ca)2(Ta,Nb)2O6(OH,F) (Y,REE)PO4 References: [1] Jahns, R. H., New Mexico Bureau Mines and Mineral Resources Bulletin 25 (1946), 294 p. [2] Ewing, R.C., Canadian Mineralogist 14 (1976), p. 111. [3] Ercit, T. S., Canadian Mineralogist 43 (2005), p. 1291 Figure 1. Backscattered electron (BSE) micrographs of Petaca minerals. Scale bars are 100 �m. (A) Thorite inclusions (white) in monazite (medium gray). Compare cleavages in the crystalline monazite at the left with the shattered appearance in the metamict, inclusion-rich region on the right. (B) Exsolution of polycrase (light) from samarskite (medium gray) and alteration by pyrochlore (gray). (C) Overgrowth of Ta-bearing rutile (struverite) on columbite (light). (D) Dark gray zircon associated with samarskite (light), altered by xenotime along fractures (medium gray) with microlite veinlets (white). (E) Plot of the scores of canonical variables 1 and 2 for the Y-REE-Ta-Nb-Ti minerals.");sQ1[573]=new Array("../7337/1147.pdf","Exploring Low-dimensional Carbon Materials by High-resolution Electron and Scanned Probe Microscopy.","","1147 doi:10.1017/S1431927615006522 Paper No. 0573 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Exploring Low-dimensional Carbon Materials by High-resolution Electron and Scanned Probe Microscopy. Jannik C. Meyer, Jani Kotakoski, Giacomo Argentero, Clemens Mangler, Bernhard Bayer, Christian Kramberger-Kaplan, Franz Eder, Stefan Hummel, Kenan Elibol, Andreas Mittelberger University of Vienna, Physics department, Boltzmanngasse 5, 1090 Vienna, Austria The microscopic characterization of low-dimensional carbon materials such as graphene or carbon nanotubes by high-resolution electron microscopy is a particular challenge owing to their intrinsically low contrast and high susceptibility to radiation damage. However, the recent developments in aberration-corrected electron optics opened a route to atomically-resolved studies of these materials at reduced electron energies, below the knock-on threshold of ca. 80kV for carbon atoms in graphene [1]. Current results are presented for 60kV studies of graphene and under ultra-high vacuum conditions using the Nion UltraSTEM100. In these conditions, graphene and related samples remain stable up to extremely high doses on the order of 10 e /nm . The use of very high doses, enabled by high sample stability, opens the door to new types of measurements such as an analysis of charge redistribution at single point defects [2], which provides only a very weak signal, or the analysis of electrons scattered to high angles [3] from single atoms, which has a very small cross section. Besides the use of low voltages, we have recently developed a new approach to use low doses and automated image acquisition on large areas of the sample. The theoretical description is given in [4], using simulated data. For the experimental realization we have developed automated low-dose acquisition where images are recorded at atomic or near-atomic resolution from a fresh area of the sample, and in particular, without depositing any dose for focusing into the region of interest. The focus is set for reference only at the corners of the selected region. Fig. 1 shows an example of a large area map of graphene where all sample regions, except for those at the corners, have not been exposed to the primary beam at all prior to data acquisition. At the current status we reliably obtain 2 Angstrom resolution by this approach, shown by the first set of reflections in graphene, whereas 1 Angstrom would be possible with perfectly tuned conditions. The second part of the presentation describes scanned probe studies of free-standing graphene membranes using a novel two-probe scanning tunneling microscope (STM). In this experiment, two STM probes are brought into contact with the graphene membrane from opposing sides, and at the closest point, the two probes are separated only by the thickness of the membrane. In this way, we have studied the tip-induced deformations in the graphene membrane and have revealed different regimes of stability of few-layer graphene. Fig. 2 shows results demonstrating the working principle of the setup and further results are shown in Ref. [5]. References: [1] J. C. Meyer et al, Phys. Rev. Lett. 108 (2012), p. 196102. [2] J. C. Meyer et al, Nature materials 10 (2011), p. 209. [3] G. Argentero, et al, Ultramicroscopy, in press (2014) (DOI:10.1016/j.ultramic.2014.11.031) [4] J. C. Meyer, J. Kotakoski, C. Mangler, Ultramicroscopy 145 (2014), p. 13. [5] F. Eder et al, Nano Letters 13 (2013), p. 1934. 10 2 Microsc. Microanal. 21 (Suppl 3), 2015 1148 [6] The authors acknowledge funding from the European Research Council (ERC) Project No. 336453PICOMAT, Austrian Science Fund (FWF) through Grant No. P25721-N20, M1481-N20, and I1283-N20 Figure 1. (a) Montage of 88 images obtained obtained automatically on a ca. 200nm x 150nm area of the sample. A focus reference was set only at the corners. (b) Example of one image showing the graphene lattice. Scale bar is 1nm. Figure 2. Initial results from the 2-tip test setup. (a) Suspended few-layer graphene, with only one tip (second tip is retracted). (b) STM image observed from one side, with the other tip placed into contact with the graphene membrane (curved lines in a-b are likely due to grain boundaries, and show that in fact the same area is imaged in both cases). (c) Observation of graphene membrane deformations, where the second tip is strongly pushing into the membrane. Scale bars are 200nm (a+b), 500nm (c). Height color scale is 20nm (a+b), 100nm (c).");sQ1[574]=new Array("../7337/1149.pdf","Towards the Electron Spectroscopy Graphene Fingerprint","","1149 doi:10.1017/S1431927615006534 Paper No. 0574 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Towards the Electron Spectroscopy Graphene Fingerprint Lester Lampert, Andrew Barnum, and Jun Jiao Department of Mechanical & Materials Engineering and Department of Physics, Portland State University, Portland, OR, USA. Graphene research continues to demonstrate that this single atom-thick layer of carbon atom material has unprecedented properties for various applications within electronic and optical devices. Observed phenomena from graphene-based devices are largely dependent upon coupling of graphene with the underlying substrate. Especially, the number of layers of graphene, the defect density, and dopants may affect the performance of the devices. To date, Raman spectroscopy has been a dominant tool for graphene property characterizations by establishing fingerprint spectra to determine relative doping levels, number of layers, stacking symmetry, temperature, etc. [1]. Recently, Auger electron spectroscopy (AES) has been slowly gaining popularity for graphene characterization due to its intrinsic surface sensitivity and its ability to measure electron interactions with the surrounding environment. Although there is a wealth of AES data available for graphite and highly ordered pyrolytic graphite (HOPG) [2], including some work with carbon nanotubes [3], as well as some recent work to determine the number of graphene layers by utilizing a calculated electron inelastic mean free path [4], there has yet to be a satisfying description of some AES spectra phenomena observed with graphene. Here, we report AES, X-ray photoelectron spectroscopy (XPS) and Raman spectroscopy of CVD graphene, amorphous carbon, and CVD diamond towards the development of an AES fingerprint spectrum of graphene on various substrates. Amorphous carbon (a-C) films were deposited with a carbon rod coater (target thickness = 4 nm) onto Si/SiO2 wafers. CVD diamond samples (diamond on oxide � DOI) were purchased from MTI Corporation (2 m thick). CVD graphene was grown with a vertical cold wall furnace held under vacuum at 750�C during the annealing/growth phase with a mixture of H2, CH4, and Ar gases. AES and XPS data was acquired with a PHI Versaprobe II AES/XPS/UPS microprobe. Raman spectra were collected with a Horiba HR800 Raman spectrometer microprobe (ex. = 532 nm). AES survey and C KVV spectra as depicted in Fig. 1 were acquired for different carbon allotropes: amorphous carbon (a-C), CVD graphene (sp2) supported by SiO2, and CVD diamond (sp2/sp3). From Fig. 1b, it can be seen that the D parameter for each spectra is 22.8, 23.1, and 21.8 eV (bottom to top). While the D parameter is similar for a-C and CVD graphene on SiO2, the shoulder at 265 eV is suppressed for a-C. This is in good agreement with previously published results to denote the difference between amorphous carbon and a graphitic-like AES spectra [3]. Also, at ~255 eV and ~260 eV there exists a prominent peak for the CVD diamond spectra, which may indicate the contributions from sp3 carbon. Fig. 2 demonstrates the Raman and XPS spectra of the samples investigated in Fig. 1. From the Raman spectra, all allotropes share a peak ~1590 cm-1 (G band), which is indicative of each sample having some sp2 carbon and is reflected in the presence of peaks ~252 eV in Fig. 1b. AES of graphene on different substrates is depicted in Fig. 3. In Fig. 3a, it can be seen the CVD graphene on SiO2 has a noticeable peak at 240 eV attributed to a two-hole plasmon replica effect [5]. D parameters for Fig. 3b are 23.1, 21.8, and 22.8 (bottom to top). The D parameter of graphene/Cu is more diamond-like and that of graphene/Ni is more a-C-like, most likely due to graphite-like structure. Coupled with established techniques such as XPS and Raman, new properties of graphene can be elucidated from an understanding of the AES spectra of graphene within different environments. Microsc. Microanal. 21 (Suppl 3), 2015 1150 References: [1] A. C. Ferrari et al, Nature Nanotechnology 8 (2013), p. 235-246. [2] H. Hashimoto et al. Surf. Interface Anal. 35 (2003), p. 19-23. [3] K. B. K. Teo et al, J. Vac. Sci. Technol. B 20 (2002), p. 116-121. [4] M. Xu et al, ACS Nano 4 (2010), p. 2937-2945. [5] E. Perfetto et al, Physical Review B 76 (2007), p. 233408 (1-4). [6] This study is supported in part by Oregon Metals Initiative and NSF awards No. 1229663 and No. 1240121. Figure 1. (a) Differentiated AES survey spectra of a-C film (bottom, red line), CVD graphene on SiO2 (middle, blue line), and CVD diamond (top, green line); (b) C KVV differentiated AES spectra of a-C film (bottom, red line), CVD graphene on SiO 2 (middle, green line), and CVD diamond (top, cyan line). Figure 2. (a) Raman spectra and (b) XPS survey spectra of a-C film (bottom, red line), CVD graphene on SiO2 (middle, blue line), and CVD diamond (top, green line). Figure 3. (a) Non-differentiated and (b) differentiated AES C KVV spectra for CVD graphene on SiO2 (bottom, red line), Cu (middle, blue line), and Ni (top, green line).");sQ1[575]=new Array("../7337/1151.pdf","STEM and EELS study of the Graphene/Bi2Se3 Interface","","1151 doi:10.1017/S1431927615006546 Paper No. 0575 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 STEM and EELS study of the Graphene/Bi2Se3 Interface D.M. Kepaptsoglou1, D. Gilks2, L. Lari2,3, Q.M. Ramasse1, P. Galindo4, M. Weinert5, L. Li5, G. Nicotra6 and V.K. Lazarov2 1 2 SuperSTEM Laboratory, STFC Daresbury Campus, Warrington, WA4 4AD, UK Department of Physics, University of York, Heslington, York, YO10 5DD, UK, 3 York JEOL Nanocentre, University of York, Heslington, York, YO10 5BR, UK 4 Department of Computer Science and Engineering, Universidad de C�diz, 11510 Puerto Real, Spain 5 University of Wisconsin Milwaukee, Milwaukee, 53211,WI, USA 6 Institute for microelectronics and microsystems, CNR Catania, 95121 Catania CT, Sicilia, Italy Bi2Se3 is a 3D topological insulator (TI) that has attracted a lot of research interest due to its exotic properties [1], associated with topologically-protected helical two-dimensional surface states and onedimensional bulk states associated with line defects such as dislocations. Recent theoretical studies [2] have shown that when graphene is placed near Bi2Se3 the strong spin orbit interaction due to proximity effects will open the band gap in graphene for 0.2 eV. Therefore TI/graphene heterostructures are promising platform for developing electronic and spintronic graphene based devices. It has been recently shown that when Bi2Se3 is grown either on epitaxial graphene or free-standing graphene flakes [1,2], a rich grain structure develops due to the spiral nature of the film growth. The grain boundaries and growth-induced screw dislocations in this system provide grounds for interesting new physics that can be accessed by a combination of scanning tunneling microscopy and transmission electron microscopy techniques such as STEM-HAADF and EELS. In this work we investigate the nature of the graphene/Bi2Se3 interface in order to understand the complex epitaxy between the film and substrate as this ultimately determines the structure and functional properties of the graphene/Bi2Se3 interface topological states. The Bi2Se3 films were grown on epitaxial graphene/SiC(0001) and freestanding graphene (Figure 1a) by chemical vapor deposition. The structure and electronic structure of the Bi2Se3/epitaxial graphene/SiC(0001) interface were studied in cross-section by state-of-the-art aberration-corrected STEM and atomically-resolved EELS. HAADF imaging (Figure 1) confirmed the presence of an epitaxial carbon layer at the Bi2Se3/epitaxial graphene/SiC(0001) interface. Combined imaging and atomically-resolved EELS maps confirm that both the Bi2Se3/carbon and carbon/SiC(0001) interfaces are atomically sharp with a Se termination of the Bi2Se3 layer. Analysis of the C K fine structure across the interface stack confirms that the observed interface carbon layer is in fact epitaxial graphene, indicating that the bonding between the Se atomic plane and the epitaxial graphene has a Van der Waals nature. This weak bonding is further corroborated by strain analysis of the HAADF images, showing the absence of strain at the Bi2Se3/epitaxial graphene/SiC(0001) interface. Such weak bonding would be the key factor for the multiple epitaxial relations which leads to both low- and high-angle boundaries observed in Bi2Se3 thin films when grown on graphene substrate [3,4]. References: [1] Y Ando, J. Phys. Soc. Jpn 82 (2003), p.102001 [2] L. Kou et al., Nano Lett. 13 (2012), p. 6251 Microsc. Microanal. 21 (Suppl 3), 2015 1152 [3] Y Liu et al., Phys. Rev. Lett. 110 (2013), p.186804 [4] Y Liu et al., Nat Phys, 10 (2014), p. 294 [5] The authors acknowledge funding from EPSRC via research grants EP/K013114/1 and EP/K032852/1. SuperSTEM is the U.K. National Facility for Aberration-Corrected STEM funded by the EPSRC. Part of this work was performed at BeyondNano CNR-IMM, supported by MIUR under the project Beyond-Nano (PON a3_00363) Bi Se 2 3 graphene Figure 1. a) planar view (HAADF STEM) image of a Bi2Se3 film grown on epitaxial graphene showing the spiral growth of the Bi2Se3 film and b) cross-sectional view (HAADF STEM image) of the Bi2Se3/epitaxial graphene/SiC(0001) interface showing an epitaxial graphene layer. a) b) Figure 2. a) cross-sectional view (HAADF STEM image) and EELS maps confirming the presence of an epitaxial graphene layer at the Bi2Se3/epitaxial graphene/SiC(0001) interface and C K spectra across the interface showing the sp2 bonding character of the epitaxial graphene layer.");sQ1[576]=new Array("../7337/1153.pdf","Core Structure of Dissociated Dislocations in Bi2Te3 Nanowires","","1153 doi:10.1017/S1431927615006558 Paper No. 0576 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 D.L. Medlin1, K.J. Erickson1*, S.J. Limmer2**, W.G. Yelton2, and M.P. Siegal2 1 2 Core Structure of Dissociated Dislocations in Bi2Te3 Nanowires Sandia National Laboratories, Livermore, CA 94551, USA Sandia National Laboratories, Albuquerque, NM 87185, USA * Present affiliation: Hewlett-Packard Laboratories, Palo Alto, CA 94304, USA ** Present affiliation: Energizer Household Products, Westlake, OH 44145, USA The weak interlayer bonding and possibilities for large Burgers vectors in Bi2Te3 and related layered chalcogenides raise interesting questions regarding the structure of dislocations in these materials. The most extensively studied type of dislocation in Bi2Te3 possesses a Burgers vector of type ! < 2110 >. ! This defect lies in the basal plane and can move easily by glide [1,2]. However, as illustrated in Figure 1, the structure can also incorporate dislocations with a large Burgers vector component lying normal to the basal plane. In this presentation, we discuss HAADF-STEM observations we have made of such defects in electrochemically deposited nanowires of Bi2Te3 [3]. The analysis presented here was made in the course of a broader study exploring how growth and annealing conditions control the resulting crystallinity and internal microstructure in Bi2Te3-based nanowires being developed for thermoelectric applications [4,5]. We focus specifically on understanding the core structures of ! < 0111 > ! dislocations, which possess a remarkably large Burgers vector of 1.048 nm. Figure 2 shows an array of such defects forming a low angle tilt boundary near the center of a nanowire. This array efficiently accommodates the misorientation of the Bi2Te3 basal planes, which here is about 5.5�, by removing a "half plane" consisting of a Bi2Te3 quintuple unit for each dislocation. Higher magnification imaging (Figure 3) shows that the cores of the ! < 0111 > dislocations are ! dissociated into pairs of partial dislocations that bound an intermediate faulted region. Circuit analysis ! ! shows that the partial dislocations are of type !"[0552] and !"[0001]. Analysis of the magnitudes of the perfect and partial dislocations suggests that the dissociation is driven in part by the ability of the partial dislocations to reduce the strain energy of the initial ![0111] dislocation. The vertical alignment of the ! quintuple layers on either side of the faulted region can vary, as is seen by comparing Figures 3a and 3b. ! This change in the alignment can arise through glide of the !"[0552] partial dislocation on an inclined 0115 type plane. The intermediate faulted region forms a seven plane thick, septuple unit, consistent with a local patch of Bi3Te4, rather than the normal Bi2Te3 quintuple layer structure. As we discuss, these observations suggest a mechanism for accommodating tellurium deficiency, which can arise, for instance, through post-growth thermal treatments. [1] S. Amelincx and P. Delavignette, Nature 185 (1960) 603-604. [2] N. Peranio, O. Eibl, Phys. Stat. Solidi A 206 (2009) 42-49. [3] D.L. Medlin et al., Journal of Materials Science 49 (11) (2014) 3970-3979. [4] S.J. Limmer et al., Journal of the Electrochemical Society 159(4) D235-D239. [5] M.P. Siegal et al., J. Materials Research 29(2) (2014) 182-189. [6] Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000. Microsc. Microanal. 21 (Suppl 3), 2015 1154 Figure 1. Bi2Te3 crystal structure. Burgers vectors of the ! < 2110 > and ! ! < 0111 > dislocations are indicated, ! respectively, by the dashed and solid red arrows. Figure 2. HAADF-STEM image of a Bi2Te3 nanowire. A low angle grain boundary, vicinal to the (0001) planes, runs horizontally across the middle of the wire. Tilt misorientation between the grains is accommodated by an array of dissociated ! < 0111 > dislocations, ! which are indicated by arrows. Figure 3. Higher magnification view of the dislocation cores marked as (a) and (b) in Figure 2. The specimen is ! ! imaged along a < 1010 > type orientation. In both cases partial dislocations, of type !"[0552] and !"[0001], bound a 7-layer Bi3Te4 fault. The two cores differ in the offset of the Bi2Te3 quintuple units on either side of the fault. In (a) the lower quintuple unit remains continuous through the fault region; in (b) the quintuple units are offset by two planes.");sQ1[577]=new Array("../7337/1155.pdf","Fabrication and Simultaneous Electrical Measurement of Graphene Nanoribbon Devices Inside a S/TEM.","","1155 doi:10.1017/S143192761500656X Paper No. 0577 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Fabrication and Simultaneous Electrical Measurement of Graphene Nanoribbon Devices Inside a S/TEM. Julio A. Rodr�guez-Manzo1, Zhengqing John Qi1, Matthew Puster1,2, Adrian Balan1, A. T. Charlie Johnson1 and Marija Drndi1. 1. 2. Department of Physics and Astronomy, University of Pennsylvania, Philadelphia, USA. Department of Materials Science and Engineering, University of Pennsylvania, Philadelphia, USA. The high-energy fine electron probe attainable in a transmission electron microscope (operated in TEM or STEM mode) could be thought of as an exquisite subtractive fabrication tool in terms of size (< 1 nm) and spatial resolution (< 1 nm). For devices that require drilling, cutting, stamping or molding at this level, and whose functionality truly stems from their nanoscale dimensions, the in-situ/operando S/TEM approach offers a top-notch platform where the physical properties (e.g., conductivity) of a material can be correlated with its structure and chemistry--with atomic resolution with the best conditions. The range of electron energies (~ 80-300 KeV) and the probe current density (~ 109 A m-2) attainable in a standard S/TEM equipped with a field emission gun are especially well suited to modify 2D materials. In this context, we discuss two examples in which graphene nanoribbon (GNR)-based devices were fabricated and modified with the electron probe while simultaneously electrically biased for electrical characterization. Experiments were carried out in a FEI Titan (aberration-corrected objective lens) or a JEOL 2010F. GNR-based devices were prepared with standard lithographic techniques [1,2] and mounted in sample holders containing electrical feedthroughs. For the latter we used either a home-made sample holder or commercially available sample holders from Protochips Inc. or Hummingbird Scientific. In the first example, we describe how to fabricate GNR-nanopore devices, which are promising candidates for next-generation DNA sequencing [1]. Such devices normally comprise a nanopore with a diameter of 2-10 nm formed with the electron probe at the edge or in the center of a 100-nm-wide GNR on a 50-nm-thick silicon nitride membrane. We discuss the changes on GNR conductance when such devices are irradiated with 200 keV electrons and the differences between irradiating with a homogenous (TEM mode) versus a scanned converged beam (STEM mode). By minimizing the average electron dose in STEM mode delivered to GNR to ~ 102 C m-2, we were able to minimize electron irradiation-induced damage and make nanopores in highly conducting GNR. The resulting devices, with unchanged resistances after nanopore formation, were tested outside the TEM column. They sustain micro ampere currents at low voltages ( 50 mV) in buffered electrolyte solution and exhibit high sensitivity, with a large relative change of resistance upon changes of gate voltage, similar to pristine GNR without nanopores (Figure 1). In the second example, we describe how to use the electron probe to sputter carbon atoms from predefined areas in electrically-connected free-standing graphene sheets to obtain GNR with widths < 10 nm [2]. We show cuts in graphene extending hundreds of nanometers with widths of a few nanometers, as well as different cut shapes (e.g., lines, sharp corners and circles). This approach allows us to correlate the lattice and edge structure of sub-10-nm wide GNR with their electrical properties (Figure 2). Microsc. Microanal. 21 (Suppl 3), 2015 1156 We believe these graphene-based examples illustrate the possibility to use the electron probe of a S/TEM to fabricate and modify 2D-materials-based devices with nanometer functional features and the advantages of performing in-situ S/TEM electrical measurements. Finally, we discuss some of the challenges (e.g., limiting leakage currents) involved in these types of experiments where chips fabricated with standard lithographic techniques are coupled to sample holders with electrical contacts and discuss different chip configurations suitable for the in-situ S/TEM electrical analysis of 2D materials. References: [1] M. Puster et al, ACS Nano 7 (2013), 11283. [2] Z. J. Qi et al, Nano Letters 14 (2014), 4238. [3] This work was supported by NIH grants R21HG004767 and R01HG006879, and NBIC through NSF NSEC DMR08-32802. JZQ and CJ acknowledge SRC contract #2011-IN-2229, NSF AIR Program ENG-1312202. Part of this work was done at the CFN in BNL, supported by the U.S. DOE, Contract No. DE-AC02-98CH10886 (FEI-Titan ACTEM through proposal 31972). Figure 1. From left to right: scheme of GNR-nanopore device for DNA sequencing, optical image of TEM-compatible chip with 4 GNR-nanopore devices, SEM image of a GNR without a nanopore, and STEM image of a GNR-nanopore device. Figure 2. From left to right: SEM image of a GNR with electrical contacts, HAADF STEM image showing a straight cut made in STEM mode, HRTEM of a GNR, and GNR resistance as a function of width from data extracted from in-situ S/TEM experiments.");sQ1[578]=new Array("../7337/1157.pdf","Kink Band Formation in High-strength Bulk Metallic Nanolaminates","","1157 doi:10.1017/S1431927615006571 Paper No. 0578 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Kink Band Formation in High-strength Bulk Metallic Nanolaminates T. Nizolek1, N.A. Mara2, R.J. McCabe3, J.T. Avallone1, I.J. Beyerlein4, T.M. Pollock1 1. 2. Materials Department, University of California Santa Barbara, Santa Barbara, CA 93106, USA Institute for Materials Science and Center for Integrated Nanotechnologies, Los Alamos National Laboratory, Los Alamos, NM 87545, USA 3. Material Science and Technology Division, Los Alamos National Laboratory, Los Alamos, NM 87545, USA 4. Theoretical Division, Los Alamos National Laboratory, Los Alamos, NM 87545, USA Advances in severe plastic deformation processing have enabled the production of bulk, high strength (>1 GPa), copper-niobium nanolaminates in volumes suitable for small structural applications. Recent investigations into the deformation behavior of these materials have shown that nanolaminates with sub100 nm layer thicknesses deform primarily by kink band formation during layer parallel compression. These kink bands (see Figure 1) consist of regions of material that have been sheared and rotated to form distinct bands of misoriented layers. While a single kink band is shown in Figure 1, these bands can form complex networks of localized deformation (see Figure 2). Despite this pronounced strain localization, no crack formation is detected and further deformation continues at a constant flow stress (as high as 1.2 GPa for 30 nm material) [1]. In addition to forming during uniaxial compression, kink banding has been observed to occur during a wide variety of mechanical tests and forming operations (such as bending) that involve layer parallel compression. While kink bands have only recently been observed in metallic nanolaminates, kinking is a well-known phenomenon in plastically anisotropic materials such as fiber composites [2], oriented polymers [3], and even biological materials such as abalone shell and wood [4,5]. The common characteristic of these materials, and a requirement for kink band formation, is mechanical anisotropy. While Cu-Nb nanolaminates are composite materials and may be expected to possess anisotropy due to their lamellar microstructure, only nanolaminates with sub-100 nm layers form pronounced kink bands. This observed length scale dependence of kink band formation is unexpected and is not predicted by models developed for kinking in other materials such as fiber composites. Raising further questions regarding the applicability of existing models for kink band formation is the lack of a pronounced load drop during kinking of metallic nanolaminates [1]. This is in stark contrast to the dramatic loss of load carrying capacity that characterizes kinking in fiber composites. Thus while kinking is appropriately classified as a failure mechanism in fiber composites, kinking in metallic nanolaminates can be considered to be a non-detrimental deformation mechanism. In order to investigate the origin of the length scale dependence of kinking and to compare kink band formation in metallic nanolaminates to kinking in other systems, in-situ mechanical testing as well as pre-test and post-test microscopy has been conducted. Microstructural analysis of the undeformed material reveals that a pronounced difference in grain morphology exists between the sub-100 nm nanolaminates and the larger layer thickness nanolaminates which do not form kink bands. Compression tests conducted at 45� to the layered structure confirm that an increase in grain aspect ratio corresponds to a significant increase in plastic anisotropy. For example, while nanolaminates with a layer thickness of 30 nm and a highly elongated grain morphology exhibit flow stresses in excess of 1200 MPa when Microsc. Microanal. 21 (Suppl 3), 2015 1158 compressed parallel to the layers, the stress required for layer parallel shear is only 230 MPa. From these results it is concluded that the observed length scale dependence of kink band formation is a result of microstructure-driven changes in the level of plastic anisotropy. The sequence of events leading to kink band formation is investigated using in situ scanning electron microscopy (SEM) compression testing. These tests reveal that kink bands nucleate at stress concentrations and propagate across the specimen, eventually intersecting a free surface. After propagation, continued deformation of the compression specimen is accommodated via broadening of the kink bands. From these tests it is concluded that kinking in metallic nanolaminates occurs in two distinct stages: propagation and broadening. Transmission electron microscopy demonstrates that void formation and layer debonding do not occur during these stages. This indicates that kink band formation in these materials must be treated as a constant volume plastic deformation process, a result which is not consistent with the assumed kinematics in fiber composite model for kinking [2]. Using these insights, we propose modifications to the existing description of kink band formation [6]. References: [1] T. Nizolek, et al, Adv. Eng. Mater. (2014) DOI: 10.1002/adem.201400324 [2] B. Budiansky, N.A. Fleck, J.C. Amazigo, J. Mech. Phys. Solids 46 (1998), p. 1637 [3] G.E. Attenburrow, D.C. Bassett, J. Mater. Sci. 14 (1979), p. 2679 [4] R. Menig, et al, Acta Mater. 48 (2000), p. 2383 [5] L. Benabou, Mech. Mater. 42 (2010), p. 335 [6] The authors wish to acknowledge support by the UC Lab Fees Research Program, Award #238091. Figure 1. Backscattered SEM image of a kink band in a uniaxial compression specimen of Cu-Nb nanolaminate material with 30 nm nominal layer thickness. The compression direction and the original layer direction are vertical. Figure 2. Polarized light micrograph of a complex network of kink bands in a uniaxial compression specimen of Cu-Nb nanolaminate material with 15 nm nominal layer thickness. The compression direction and the original layer direction are vertical.");sQ1[579]=new Array("../7337/1159.pdf","Nanostructures Formation in Al-B4C Neutron Absorbing Materials after Accelerated Irradiation and Corrosion Tests","","1159 doi:10.1017/S1431927615006583 Paper No. 0579 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Nanostructures Formation in Al-B4C Neutron Absorbing Materials after Accelerated Irradiation and Corrosion Tests Feifei Zhang1, 2, Jonathan Brett Wierschke2, Xu Wang1, 2, Lumin Wang2, 1 College of Energy, Xiamen University, Xiamen, Fujian 361102, China Department of Nuclear Engineering and Radiological Science, University of Michigan, Ann Arbor, Michigan 48109, United States 2. 1. Al/B4C metal matrix composite (MMC) is an important neutron absorbing material used in both wet storage pools and dry storage casks of spent nuclear fuel for preventing criticality. Because of the high neutron absorption cross-section of 10B, the material can effectively absorb fast and thermal neutrons, but at the same time suffers neutron irradiation damage from the nuclear spent fuel. Moreover, interactions of fast and thermal neutrons with boron lead to the production of several transmutation species such as helium, lithium and others, according to the transmutations reactions induced by neutron irradiation, including the 10B (n, 4He) 7Li reaction and others. The most abundant product of these reactions is energetic helium (~1.47MeV), which may induce further radiation damage and be precipitated out in the material as helium bubbles. The formation and coalescence of the helium bubbles in boron carbide and on boundaries, especially phase boundaries, would affect the corrosion performance of the material in its working environment. The degradation of the material may lead to boron loss which is the critical element to control the criticality of spent fuel. To evaluate the radiation and corrosion effects for interim storage of two hundred years accelerated He+ irradiation and corrosion tests were carried out. The bulk MMC sample was irradiated with 400keV He+ to 1.5�1017 ions/cm2 and 50keV He+ to 3.5�1016 ions/cm2 at room temperature. Autoclave corrosion tests of the ion-irradiated samples were carried out at 400 for 112 days, and humidification of the argon was accomplished by passing the argon through water. Microstructures of irradiated and corroded material were characterized by SEM, TEM and STEM. TEM samples were prepared by Focus Ion Beam (FIB) lift-out method. TEM analysis and characterization were conducted with a JEOL HRTEM 3011 and a JEOL 2100F STEM and JEOL 3100R05 TEM/STEM at the Electron Microbeam Analysis Laboratory (EMAL), University of Michigan. SEM images in Figure 1 show the evidence of corrosion. It is apparent that the boron carbide particles on the surface were totally corroded and disappeared after the corrosion test. Fiber-like nanostructures formed at the bottom of the hole, where was the interface between Al matrix and B4C particles. TEM investigation was taken on the cross-section sample in order to study the newly formed nanostructures. Figure 2a indicates that the nanostructures formed at both surface and the interior. The interior nanostructures are along the Al/B4C phase boundaries. After He+ irradiation, helium bubbles formed and coalesced at the phase boundaries as noted in Figure 2b [1]. These helium bubbles may lead to formation of open channels causing the reaction of interior material with water. Analysis of EDS point spectra shows that there is a newly formed phase at the interface [2] after the sample was placed at 400 for 112 days. The SAD pattern and high resolution BF STEM images revealed the crystalline nature of the nanofibers. Reference [1] F. Zhang, et al, Microsc. Microanal. 19 (Suppl 2), 2013. [2] Z. Zhang, et al, Metallurgical and Materials Transactions A, 43A (2012) 281. [3] This work has been supported by the US DOE NEUP Program under the contract No. 00102642. Microsc. Microanal. 21 (Suppl 3), 2015 1160 a b c Figure 1. (a) SEM image of the sample surface after ion irradiation and corrosion test; (b) SEM image of a hole where a boron carbide particle was dissolved; (c) SEM image at high magnification showing the nanostructures at bottom of the hole. a #2 Surface Interior #1 c #2 d #1 b Figure 2. (a) HAADF image and (b) BF STEM image of the near surface area of a corroded sample, the nanostructures were marked by arrow, (c) TEM BF image of "#2" area at high magnification and corresponding SAD pattern, (d) high resolution BF STEM image of the nanocrystal. The insets are EDS result of point #1 (area near nanostructures) and #2 (on the nanostructure).");sQ1[580]=new Array("../7337/1161.pdf","A Study of Glass Composition and Crizzling in Two Claude Laurent Glass Flutes from the Library of Congress","","1161 doi:10.1017/S1431927615006595 Paper No. 0580 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A Study of Glass Composition and Crizzling in Two Claude Laurent Glass Flutes from the Library of Congress Andrew Buechele1, Lynn Brostoff2, Isabelle Muller1, Carol Lynn Ward-Bamford2 and Xiaogang Xie1 1. 2. The Catholic University of America, Vitreous State Laboratory, Washington, DC, USA. Library of Congress, Washington, DC, USA. Claude Laurent, a French clockmaker, started making musical flutes out of glass in the early 19th century. He was granted a patent in 1806 for "une flute en cristal" and won a silver medal that year at the Paris Industrial Exposition for his invention. Laurent believed that glass had many desirable qualities as a material for fabrication of flutes. Certainly they were impressive to look at, and many were presented to European heads of state during Laurent's lifetime, but, according to Dayton C. Miller (1866-1941), an Ohio physicist and amateur flutist, they were not exceptional in their tonal quality. Miller was also a collector who amassed over 1700 flutes and wind instruments which he donated to the Library of Congress in 1941. Within that collection are 18 of the approximately 140 Laurent glass flutes known to exist worldwide [1]. The flutes have become the subject of a research project directed toward a deeper understanding and appreciation of these unique instruments, and toward providing valuable data for their preservation. Although Laurent referred to his flutes as "crystal," XRF analysis carried out at the Library of Congress has shown that only two of the 18 Miller flutes are made of leaded crystal glass. Most of the rest are made of potash glass, an alkali glass with potassium as the principal component. Some of the flutes in the collection exhibit evidence of glass deterioration, which results when the humidity in the air diffuses into the glass and hydrates and leaches out alkali ions from the glass structure. This endangers the integrity of glass matrix and can be visible as crizzling, a fine network of microscopic cracks in the alkali-depleted surfaces. In advanced stages, flaking is evident in this layer. Both conditions scatter light and cause the glass to appear frosty to the naked eye. The Vitreous State Laboratory (VSL) at The Catholic University of America (CUA) was approached by Dr. Lynn Brostoff of the Library of Congress and asked to help to characterize the potash glass more completely and better understand the degradation process. Initial SEM-EDS examination of a very small unpolished chip taken from a previously damaged extra foot joint of a flute identified as 1235 confirmed that the major components of the glass are K2O, SiO2, and CaO, plus a small amount of Na2O. Subsequently, additional material was collected from the damaged joint of 1235 to provide sufficient material (~60 mg) to conduct Direct Coupled Plasma (DCP) and X-ray fluorescence (XRF) spectroscopic analysis of the glass as noted in Table 1. Quantitative results confirm the major components present, as well as a number of minor and trace elements. At the same time that material was gathered from 1235, a second flute, 717, which also exhibits crizzling, was made available for microsampling at the site of previous damage. A perforation was noted at the base of a decorative Vgroove in the surface of flute 717, which may have been facilitated by crizzling. Cautious pressure in the perforated area at the base of the groove resulted in the extraction of two small chips that contain portions of both the original inner and outer surfaces of the flute body. These chips were mounted in epoxy and polished to expose a cross-section perpendicular to the longitudinal axis of the flute, so that both inner and outer surface could be observed in the SEM. Figure 1 displays a backscattered electron image (BEI) micrograph of the mount, showing modified layers about 30�m thick on both the inner and Microsc. Microanal. 21 (Suppl 3), 2015 1162 outer surfaces of either chip. Compositional profiles were acquired by energy dispersive spectroscopy (EDS) throughout the modified layers of both chips of flute 717. Figure 2 is representative of the behavior seen in all profiles. In brief, in the modified layer, potassium decreases to about 1/3 its value in the original glass, while sodium drops below the detections limit for analytical conditions used; at the same time, oxygen seems to increase by 4 to 5 wt%, perhaps indicating some water retention. The aforementioned transitions are relatively abrupt, while calcium and silicon remain relatively constant across the reaction interface. Comparison of compositions of the glasses used in the fabrication of flutes 1235 and 717 using SEM-EDS on mounted, polished chips also shows that K2O is about 4 wt% higher in 717, which is consistent with a greater degree of crizzling seen in this flute. The layers observed in the above microsamples represent the result of about 200 years of exposure to ambient atmospheric moisture, as well as the breath of flutists. To further study and understand the degradation process in the glass that are taking place on both interior and exterior surfaces of flute joints replications of the two representative Laurent compositions determined above are being melted to engender model samples. Methods of accelerated testing used at VSL to assess the durability of nuclear waste-glass candidates will be applied to induce attack of these replications. In addition to providing a window on the kinetics of the aging reactions in the flute glasses, a correlation of the short term results of an accelerated test with the actual effects of aging under ambient atmospheric conditions will be possible. References: [1] "Anatomy of the Flute", Library of Congress Magazine, 3, No. 5, (2014), p. 4. Flute ID 1235 717 Analyses (wt%) DCP XRF SEM SEM Al2O3 0.36 0.66 0.34 0.00 CaO 3.38 3.48 3.69 3.49 K 2O 15.54 16.49 17.18 21.02 Na2O 0.77 0.68 0.63 0.68 SiO2 79.04 77.88 78.03 74.45 Others 0.92 0.80 0.13 0.36 Table 1. Summary Analyses of the Flute Samples (others include Cl, SO3, B2O3, BaO, Fe2O3, Li2O, MgO, MnO, NiO, P2O5, Rb2O, TiO2, ZnO, ZrO2, in the range of 0.002 to 0.17). Figure 1. BEI micrograph of chips from flute 717 wall showing a modified layer. Figure 2. EDS compositional profiles throughout the modified layer in flute 717.");sQ1[581]=new Array("../7337/1163.pdf","Characterization of Void-Dominated Ductile Failure in Pure Ta","","1163 doi:10.1017/S1431927615006601 Paper No. 0581 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of Void-Dominated Ductile Failure in Pure Ta Blythe G. Clark1, Joseph R. Michael2, Bonnie B. McKenzie2, Jay Carroll2, Hojun Lim2, Brad L. Boyce2 1. 2. Sandia National Laboratories, Physical, Chemical, and Nano Sciences Center, Albuquerque, NM. Sandia National Laboratories, Materials Science and Engineering Center, Albuquerque, NM Prediction of materials reliability requires a physical understanding of the failure process. In the case of void-dominated ductile failure, this includes an understanding of the processes and dominant microstructural variables that lead to void nucleation, growth, and coalescence. Mechanistic descriptions of void-dominated ductile failure are largely based on studies from several decades ago that relied on optical microscopy techniques [1] [2]. This talk will discuss how modern techniques, such as scanning electron microscopy (SEM), electron backscattered diffraction (EBSD), site-specific focused ion beam (FIB) sample preparation, and transmission electron microscopy (TEM), can enable new insight into the ductile failure process, particularly at the nano to microscale. Because this work is part of a collaborative effort to develop predictive computational models for failure of BCC metals, experiments are focused on characterizing the void-dominated ductile failure process in pure Ta. In metal alloys the initiation of voids is primarily driven by decohesion at second-phase particles or inclusions [3][4][5]. However, in pure ductile metals where these particles and inclusions are not present, such as in the 99.9% Ta used in this study, the process is inherently different. Initial tensile deformation experiments and post-mortem fractography as noted in Figure 1 show that Ta exhibits significant ductility, but with a valley and ridge surface [6] instead of the classic hemispherical dimpling commonly observed in ductile metal alloys [7]. Through SEM analysis of the fracture surfaces, it was determined that the mating fracture surfaces were mirrored: valleys mated to valleys, and ridges mated to ridges. Thus the deformation was void dominated, but the mechanism of void initiation was unclear. To further study the void initiation process in Ta and the progression of damage towards failure, interrupted tensile specimens as a function of percent remaining strength were prepared. By polishing the samples to their mid-plane as seen in Figure 2a, where triaxial stresses are highest and thus voids likely to be present, we studied the progression of void structures as a function of strain. Using EBSD analysis to characterize local misorientations in voided regions, we determined that voids in Ta are formed within grains, as opposed to at grain boundaries, along bands of high misorientation with respect to the tensile axis. Figure 2b shows elongated, inclined [001] subgrains alternating with regions of [122] associated with voids in a Ta tensile bar elongated to 60% remaining strength. We hypothesize that the dislocation processes giving rise to localized regions of high misorientation within grains are promoting the collection of vacancies, thus inducing void nucleation. This hypothesis is being further explored through site-specific FIB sample preparation and TEM analysis of dislocation structures in deformed Ta samples. Microsc. Microanal. 21 (Suppl 3), 2015 1164 References: [1] K.E. Puttick, Philosophical Magazine 4 (1959), p. 954�969. [2] H.C. Rogers, Transactions of the Metallurgical Society of AIME 218 (1960), p. 498�506. [3] A. Needleman, Journal of Applied Mechanics 54 (1987), p. 525�31. [4] D. Broek, Engineering Fracture Mechanics 5 (1973), p. 55�66. [5] A.S. Argon, J. Im, and R. Safoglu, Metallurgical Transactions A 6A (1975), p. 825�37. [6] B. L. Boyce, B. G. Clark, P. Lu, J. D. Carroll, and C. Weinberger,aMetallurgical and Materials tungsten, which is also BCC refractory metal and tantalum's neighbor on the periodic table, is brittle at Transactions A 44A (2013), p. 4567-4580. room temperature failing by a low-energy cleavage [7] R. H. Van Stone, T. B. Cox, J. R. Low, Jr. and J. process. Under dynamic loading, tantalum fails by a A. Psioda, International Materials Reviews 30 spallation process which progresses by the nucleation of (1985), p. 157-180. distributed nanoscale voids, followed by void growth [19,20] and link-up Khalid Hattar [8] The authors would like to thank Drs. Corbett Battailefurtherofmotivate thevoids. for useful discussions. and clusters of current study, a preliminary To observation of the fracture surface tantalum was [9] Sandia National Laboratories is a multi-program laboratory managedof and operated by Sandia compared with classic ductile, dimpled rupture morpholCorporation, a wholly owned subsidiary of Lockheed (Figures 1(b)Corporation, for of the fracture Department of ogy Martin and 2). This examination the U.S. surface contract pure tantalum loaded in quasistatic tension Energy's National Nuclear Security Administration underofquestion the DE-AC04-94AL85000. calls into validity of traditional void coales(a) (b) cence models for ductile fracture for this material. As shown in Figure 2(a), tantalum deformed quasistatically at room temperature is clearly ductile when loaded in tension, exhibiting a 92 pct reduction in area before separation. However, the fracture surface bears little resemblance to typical ductile dimples associated with void coalescence, such as those shown for 304L stainless steel in Figure 1(b). Ductile dimples are typically hemispherical in nature.[2,6] However, in tantalum, a series of elongated ridgelines and valleys form in the fracture surface. The ridgelines typically range from 10 to 50 lm long, much larger than typical ductile dimples. Also, broad, nearly planar facets form between ridges and valleys, reminiscent of brittle cleavage or intergranular decohesion. Finally, a serrated thin film extends along many of the ridgelines, (arrows in Figure 2). From these observations, the question arises as to whether tantalum fracture is associated with a conventional ductile void coalescence mechanism, or some other mechanism. Figure 1. SEM images comparing (a) a ductile dimple fracture surface of 304L stainless steel, with (b) a valley and ridge fracture surface of 99.9% Ta. Both materials were deformed quasistatically at room temperature. II. EXPERIMENTAL PROCEDURE (a) (b) Several techniques were used to observe the failure process in tantalum. These techniques included in situ deformation experiments in the scanning electron microscope (SEM) to observe surface slip, crack nucleation, and crack propagation; ex-situ tests where the tensile test was interrupted in the necking regime; and metallographic cross sections that were examined both optically, using the SEM, and using electron backscatter diffraction (EBSD). In addition, the fracture surface was examined post-mortem in the SEM, and focused ion beam cross sections of the fracture surface features were examined using both SEM and the transmission electron microscope (TEM). A. In Situ SEM Tensile Experiments Annealed 99.9 pct tantalum sheet material was procured from Goodfellow Corporation (Oakdale, PA) with dimensions of 300 9 300 9 1.5 mm (Goodfellow product number 000521). The typical chemical analysis for this product is (in ppm): Al 5, Ca 2, Co 1, Cr 5, Cu 2, Fe 30, Mg 5, Mn 2, Mo 100, Na 10, Nb < 500, Ni 3, Si 10, Sn 2, Ti 20, V 5, W 100, and Zr 10. No inclusions or second phase particles were found in either SEM or TEM inspection. Fig. 1--Classic images of the ductile failure process at different magnifications and with different imaging techniques. (a) optical cross section of cavities formed during deformation of rolled copper 1959 study of Puttick[1], (b) secondary electron SEM image of ductile dimples formed on the fracture surface of 304L stainless steel during room temperature quasistatic loading, (c) Sawtooth-type final rupture observed with in situ TEM of high-purity gold at a loading rate ranging from 1 to 10 lm/s from the 1983 study of Wilsdorf[8]. 100 �m Figure 2. (a) Example of a Ta tensile bar deformed quasistatically to 40% remaining strength and polished to the mid-plane for void analysis. (b) EBSD data collected around a localized region of deformation-induced voids in a Ta tensile bar deformed quasistatically to 60% remaining strength. Orientation4568--VOLUME 44A,is given in the stereographic triangle. Here, orientations TRANSACTIONS A with respect color key OCTOBER 2013 METALLURGICAL AND MATERIALS were plotted to the tensile axis. the groundwork for future improvements in failure modeling of tantalum and other metals that fail by a similar process. Tantalum, a body-centered-cubic (BCC) refractory metal, is often studied for its high-temperature behavior, or its behavior under shock loading conditions.[18�20] The current study seeks to evaluate the failure process of tantalum under low rate loading at room temperature both as a means to understand the material's reliability for applications where these environments are relevant, and as a point of departure to later examine highertemperature behavior. Tantalum possesses good ductility at room temperature and maintains modest ductility even down to cryogenic temperatures.[21,22] In contrast, 10 �m");sQ1[582]=new Array("../7337/1165.pdf","Circuit Editing and Failure Analysis Applications using a Three-Ion-Beam (Ga, He and Ne) System and Gas Injection System (GIS)","","1165 doi:10.1017/S1431927615006613 Paper No. 0582 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Circuit Editing and Failure Analysis Applications using a Three-Ion-Beam (Ga, He and Ne) System and Gas Injection System (GIS) Deying Xia1, Huimeng Wu1, Bernhard Geotze1, David Ferranti1, Lewis Stern1 and John Notte1 1. Carl Zeiss Microscopy, LLC, Ion Microscopy Innovation Center, One Corporation Way, Peabody, MA 01960 The focused ion beam (FIB) is widely used in semiconductor industries for circuit editing (CE), failure analysis (FA) and nanofabrication. Gallium FIB is most developed for CE and FA for feature size >10 nm. In general, after the FIB processing, the structural characterization is performed using SEM. A gas injection system (GIS) is integrated into a FIB or SEM microscope system to mill or deposit the materials in electron or ion induced processing in CE and FA. Ga FIB induced processing cannot fabricate small enough metal or insulator nanostructures to meet the shrinkage of feature size in current semiconductor processing. In the Zeiss NanoFab system, the Ga ion beam is used to perform the fast and large scale milling, and the helium ion beam is used to obtain the high resolution images. In addition, the neon ion beam is used to mill or assist deposition of smaller functional structures. Appropriate precursors for ion induced etching or deposition are introduced via an integrated GIS. In this paper we evaluate applications in CE and FA using a combination of three different focused ion beams. The first example is the enhanced etching of silicon (Si) for the backside CE using a Ga ion beam in combination with XeF2 gas. A relatively thick layer (~10�m) of Si layer was left in a larger pre-etched pit for backside CE. If the Ga ion beam is used alone, it would take much longer time to etch down the metal layer. We used multiple-step Ga milling to minimize typical sidewall re-deposition. It took a dose of 46 nC/�m2 to etch down to the metal layer as shown in Figure 1A. With the assistance of XeF2 gas, it only took 0.4 nC/�m2 to begin exposure of the metal layer as shown in Figure 1B. The enhancement factor is as high as 100 and it takes much less time, 10 min with XeF2 comparing to 600 min without XeF2. More importantly, we were able to use end-point detection for CE for this case. Figure 1C shows that the dose of 0.4 nC/�m2 initiates exposure of the metal layer. This event is apparent from the signal of the secondary electrons collected on the Everhardt Thornley (ET) detector and represents a reliable method to stop etching at the target layer. Figure 1D gives another example of enhanced etching of SiO2 film on Si substrate with a combination of the neon ion beam and XeF2 gas. From this curves, we can see the enhancement factor is larger than 3 for both thick and shallow SiO2 layer. Figure 1E shows an additional example of neon ion beam milling with high accuracy for front-side CE for milling a box 150nm�150nm and 400nm deep. It is obvious that endpoint information from curve peak and bottom of Figure 1F can be obtained for metal and insulator layers. The neon ion beam can be also used to cut the fine metal line with good control in CE and FA applications. Metal and insulator nanostructures for CE and FA can be deposited with ion beam induced processing. Helium and neon ion beams used in conjunction with precursor chemistries delivered through a GIS deposit fine metal lines such as Pt, W and Co for conductive electrical connections for CE. Figure 2A shows helium ion deposited 10 nm Co lines with 50 nm pitch on metal fingers. Insulator materials can be deposited from precursors such as TEOS and PMCPS. Figure 2B shows an array of PMCPS squares with 200 nm thickness deposited using a helium ion beam. Both metal and insulator structures prepared in helium or neon ion beam induced processing exhibit good corresponding properties: ultralow resistivity as 100 ��cm for metal line and ultrahigh resistivity as 1013 �cm for insulator pad. Microsc. Microanal. 21 (Suppl 3), 2015 1166 A B 5�m 100 4�m 400 300 200 100 C 80 D 60 shallow 500-nm 500-nm XeF2 shallow XeF2 40 20 0 0 0.2 0.4 0.6 Dose, nc/�m2 0 0 Dose, nc/�m2 2 4 6 8 E F 100nm Dose, nc/�m2 Figure 1. (A)-(B) helium ion image of Ga ion beam etching Si from backside of circuit (A) multiplestep milling with Ga ion beam only with total dose 46 nC/�m2; (B) XeF2 assisted etching; (C) end-point detection for XeF2-assisted Si etching for tested chip as in (B); (D) comparison of endpoint detection for with and without XeF2 to etch SiO2 film on Si substrate using neon ion beam; (E) tilted helium image of etching box on front side of tested chip; (F) endpoint curves for Ne milling processing in (E). A 50 nm B 200nm 2�m Figure 2. Helium ion image of deposited metal and insulate: (A) tilted view, four Co lines on four finger structures; (B) top view, insulator pad arrays of PMCPS.");sQ1[583]=new Array("../7337/1167.pdf","Digital Image Correlation of Heterogeneous Deformation in Polycrystalline Material with Electron Backscatter Diffraction","","1167 doi:10.1017/S1431927615006625 Paper No. 0583 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Digital Image Correlation of Heterogeneous Deformation in Polycrystalline Material with Electron Backscatter Diffraction Javier Esquivel1, Michael D. Sangid1 1. Purdue University, School of Aeronautics and Astronautics, West Lafayette, IN, USA. This work establishes the ability to conduct digital image correlation (DIC) investigations at length scales that elucidate slip system activation and quantification as the primary deformation mechanism in polycrystalline materials. DIC allows for a computational method of strain field measurements using multiple images to track random speckle patterns on material surfaces. Self-assembling gold nanoparticles provide sub-micron resolution speckle patterns to study microstructure influences on deformation using scanning electron microscopy (SEM). The complex microstructures in aerospace grade aluminum give rise to varied deformation fields, which can be studied using electron backscatter diffraction (EBSD). Specimen preparation techniques, speckle patterns, and image correlation analysis are discussed. Experimental identification of strains at grain level can help validate computational crystal plasticity finite element models, which in turn provide better predictive computational models. Component failure is a result of deformation accumulating in small regions within a part. In fact, strain localization is a precursor to material failure. In this research, the fundamentals of strain localization are enabled through the set-up of a unique experimental analysis. The microstructure attributes, especially grain boundaries (GBs), are responsible for heterogeneous deformation. As defects within the microstructure localizes deformation and leads to incompatible strain. DIC represents a powerful tool when combined with characterization of the microstructure, to assess the heterogeneous deformation, damage mechanisms, or fatigue behavior [1-3]. In the following research, a specimen is constructed (Fig. 1a), polished, and fiducial marks are placed on the specimen in the form of microhardness indents to establish a spatial reference for microanalysis (Fig 1b). The spatial crystallographic orientations are measured via EBSD and overlaid on the specimen according to the fiducial marks in Fig. 1c. A gold nanoparticle speckle pattern is applied to the material as shown in Fig. 2 according to the novel procedure developed by Kammers and Daly [4]. For this technique, the sample is hydroxylated to increase the oxide and hydroxide groups via a basic solution bath. The specimen is incubated in silane for 1 day and then a colloidal gold solution for a week to attach silane linker molecules and gold nanoparticles to the functionalized silane molecules, respectively. Reference images of the speckle pattern are taken at 5000x in the SEM via a grid of 6x5 images (24.79 x 18.6 �m per image) within the 100 �m by 100 �m fiducial markers, as shown in Fig. 3a. The Al specimen was loaded and unloaded exhibiting a 1.4% residual strain. Images of the deformed speckle pattern were acquired and the deformation was correlated using VIC-2D software. The strain fields of DIC are superimposed on the crystallographic orientation (Fig. 3b), thus elucidating the GBs in the material. Further, through the crystallographic orientations, the slip planes and GBs are indexed [1]. The results allowed for quantification of local plastic deformation relative to the material's microstructure, in the form of slip bands and represented resolution necessary to display slip system activation. The results are important, as localization of deformation at the slip band level is crucial to failure processes, such as fatigue. Further, the results display spacing in between slip bands, the amount of deformation accommodated by each slip band, and the discontinuous nature of slip. Microsc. Microanal. 21 (Suppl 3), 2015 1168 [1] W. Abuziad, M.D. Sangid, J. Carroll, H. Sehitoglu and J. Lambros, Journal of the Mechanics and Physics of Solids 60 (2012), p. 1201-1220. [2] A.D. Kammers and S. Daly, Experimental Mechanics 53 (2013), p. 1743-1761. [3] F. Di Gioacchino and J. Quinta da Fonseca, Experimental Mechanics 53 (2013), p. 743-754. [4] A.D. Kammers and S. Daly, Experimental Mechanics 53 (2013), p. 1333-1341. [5] The authors gratefully acknowledge German A. Parada for initial set-up of the speckle pattern procedure during undergraduate research at Purdue. The authors graciously acknowledge support for this work from Rolls-Royce Corporation and the National Science Foundation, CMMI-1334664. Prof. Daly is thanked for providing stimulating discussions and suggestions for the SEM-DIC methodology. Figure 1. (a) Al specimen. (b) Microhardness indentations displaying fiducial marks to elucidate regions of interest. (c) Superimposed information of crystallographic orientations via EBSD. Figure 2. Gold nanoparticle surface layer preparation (a) Standard oxide and hydroxide groups. (b) Increased oxide and hydroxide groups after basic solution bath. (c) Silane linker molecules after MPMDMS bath. (d) Gold nanoparticles attach to functionalized silane molecules. Figure 3. (a) 100 �m by 100 �m region of interest � 30 images acquired at 5000x. (b) Concurrent measurements of material deformation via DIC in % axial strain and crystallographic orientations via EBSD to measure 1.4% residual strain.");sQ1[584]=new Array("../7337/1169.pdf","Metallographic Preparation of Incoloy 800 Tube Material for EBSD","","1169 doi:10.1017/S1431927615006637 Paper No. 0584 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Metallographic Preparation of Incoloy 800 Tube Material for EBSD A.J. Lockley Canadian Nuclear Laboratories, Chalk River Laboratories, Chalk River, Ontario, Canada, K0J 1J0 Incoloy 800 is a nickel-iron-chrome alloy used where high temperature� and corrosion -resistance is required with typical configuration and application being tube material for heat-exchangers. During manufacturing, the boilers tubes are subject to cold-work introducing residual strain into critical regions of the tube. In characterizing the impact of this cold-work on the microstructure, tube material was prepared using metallography for examination by Electron Backscatter Diffraction (EBSD). One challenge associated with sample preparation for this application can be attributed to the the wall thickness of the tube, being only 1 millimeter (Figure 1). Sectioning, mounting and polishing of the sample may introduce additional strain, and inflate results obtained by subsequent EBSD. Most such artifacts can be circumvented by electro-polishing, however; this approach brings a different set of complications: electronic differences from inclusions leaving large pits, and heavy rounding of sample edges (an area of specific interest). Standard metallographic mounting methods introduce other complications as well. Modifications to sample preparation, specifically mounting and the final polish, were found to mitigate these complications. To help maintain edge retention several parallel specimens were cut and assembled in a high-conductive silver epoxy to bind and surround the specimens. The primary benefit of the silver epoxy is that its conductive properties encasing the specimens acts to minimize charging and drift in the scanning electorn microscope (SEM) and improve overall EBSD quality. The assembly in silver epoxy was supplemented with an epoxy casting resin to complete the mount to a standard 30 mm diameter metallographic mount configuration (Figure 2). A compression mounting material was not used because it nominally requires compressive loading, from a hydraulic cylinder, to be applied to both the mounting material and the specimen as part of the forming process. Previous laboratory experience has shown that delicate material, such as small springs or the thin walls of tube specimens, may be subject to deformation during this type of mounting process. Therefore, this method was avoided to minimize the introduction of strain resulting from any such deformation. The mounted specimens were prepared using standard metallographic methods, however; the final polish employed a chemical-attack polish comprising cooper (II) chloride, hydrochloric acid and colloidal silica [1] that removed surface damage induced by previous polishing steps. Optical examination using an inverted metallograph was used to inspect the effectiveness of the attack-polish. A prepared surface should appear featureless but structure and surface details, including preparation damage such a light scratching, can be observed by defocusing the microscope and viewing using a small aperature (Figure 3). Such observable damage should be eliminated by further polishing prior to EBSD analysis. Microsc. Microanal. 21 (Suppl 3), 2015 1170 This preparation method has been proven successful in providing strain orientation maps with minimal preparation damage and gives the additional benefit of edge retention to enable strain detail at the surfaces (Figure 4). References [1] [2] G. Petzow, Metallographic Etching, 2nd Ed., ASM international, 1999. EBSD images provided by C. Mayhew, Surface Science Lab, CNL, Canada. Figure 1 Schematic of a single section of a tube wall specimen. Figure 2 Photograph of a metallographic mount containing tube cross-sections. Figure 3 Optical micrograph of Incoloy 800 microstructure. Image is slightly de-focused to reveal detail. Figure 4 Strain-orientation map near the outside surface of the tube (top of image).");sQ1[585]=new Array("../7337/1171.pdf","The Atomic Interfacial Structure between 2 and Phases within a TiAl Alloy in Lamellar Form","","1171 doi:10.1017/S1431927615006649 Paper No. 0585 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Atomic Interfacial Structure between 2 and Phases within a TiAl Alloy in Lamellar Form Fang Liu1, Guo-zhen Zhu1 1. State Key Laboratory of Metal Matrix Composites, School of Materials Science and Engineering, Shanghai Jiao Tong University, Shanghai, 200240, P. R. China. Due to the attractive properties such as high strength-to-weight ratio and excellent high temperature creep and oxidation resistance, TiAl-based alloys with fully lamellar microstructure have been widely applied in producing aerospace and car engines [1-3]. The distribution of 2 and phases, strongly affected by their interface, plays an important role in determining their mechanical properties at high temperature [4-6]. Although many researchers have investigated it by transmission electron microscopy [4, 7], the atomic interfacial arrangement and the chemical bonding nature have not been fully clarified. Therefore, we aim on analysing of the detailed interfacial structure in TiAl alloys. We studied TiAl-based alloys synthesized by arc melting (at China Iron and Steel Research Institute Group). The lamellar structure was achieved by the heat treatment including 1340 � for 0.5 h, 1 h and 2 C h followed by air cooling. The microstructure morphology was studied by the optical microscopy and scanning electron microscopy. The samples were mechanically polished and etched by the aqueous reagent of 2% hydrofluoric acid and 10% hydrogen nitrate prior to the optical characterization. The TEM samples were prepared by the twin-jet electro-polishing technique with an electrolyte of 4% perchloric acid. The detailed atomic structure was investigated using a JEOL ARM200F Scanning Transmission Electron Microscopy (STEM) with a probe corrector. The lamellar structure, consisting of 2 (Ti3Al, DO19) and (TiAl, L10) phases, was successfully synthesized through the above approaches. The lamellar spacing increases with increasing annealing time. A well-defined orientation relationship of {111}//(0001)2 & <1-10>//<11-20>2 was additionally confirmed using the selected area diffraction (SAD) technique (see Figure 1). There are always three sets of diffraction patterns co-existing within the same regions. The diffraction labeled by the yellow rectangle in figure 1 is from 2 phase. The other two diffraction patterns, indicated by the red and green rectangles, belong to <011] and <1-10], respectively. The atomic arrangement of phase viewed along <011] and <1-10] is different (see the STEM-high-angle annular dark-field (HAADF) images in figure 2 (b) and (c)) due to the ordered L10 structure with a slightly larger c axis. Although the c axis is not equivalent to the other two axes in structure, the c/a ratio is only in the range of 1.01-1.03 depending on the Al content. As shown in the STEM-HAADF images in figure 2, 2 phase has brighter Z-contrast compared to the phase. Viewed from <11-20>2, the Ti column has slight brighter contrast compared to the Ti-Al column. The interface between 2 and phase is atomic sharp, as shown in figure 2 (d) and (e). In addition, we always detected the twining (figure 2 (e)) and pseudo-twining (figure 2(f)) of phase viewed along <011]. The pseudo-twinning refers as the mirror structure of phase forming at sides of 2 phase with a dimension of ~2 nm. The detailed atomic arrangement of the 2/ interface is currently under investigated by atomic resolution STEM and electron energy-loss spectroscopy techniques (EELS). Different atomic models can be built to study their interface energy by applied first principles computation. Microsc. Microanal. 21 (Suppl 3), 2015 1172 References: [1] E. A. Marquis, J. M. Hyde, Materials Science and Engineering R: Reports 69 (2010) p. 37-62. [2] E. A. Loria, Intermetallics 8 (2000), p. 1339-1345. [3] H. Clemens, H. Kestler, Advanced Engineering Materials 2 (2000), p. 551-570. [4] G. J. Mahon, J. M. Howe, Metallurgical Transactions A 21 (1990), p. 1655-1662. [5] F. Appel, M. Oehring, R. Wagner, Intermetallics 8 (2000), p. 1283-1312. [6] L. M. Hsiung, A.J. Schwartz, T. G. Nieh, Scripta Materialia 36 (1997), p. 1077-1022. [7] L. Zhao, K. Tangri, Philosophical Magazine A 64 (1991), p. 361-386. [8] Thanks for the supplying samples from China Iron and Steel Research Institute Group. We also thank Prof. Ji Zhang for helping and benefit discussions. Figure 1. The lamellar structure. (a) the bright field image (b) the corresponding diffraction pattern. Figure 2. The atomic structure of lamellar microstructure. (a) the low magnification HAADF image (b)-(c) shows the atomic structure of phase when the beam is parallel to <1-10] and <011], respectively (d) the interface of and 2 (e)-(f) the twining and pseudo-twining, respectively.");sQ1[586]=new Array("../7337/1173.pdf","New Technology and Approaches for the Acceleration and Enhancement of Microstructural Characterization using Electron Backscatter Diffraction","","1173 doi:10.1017/S1431927615006650 Paper No. 0586 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 New Technology and Approaches for the Acceleration and Enhancement of Microstructural Characterization using Electron Backscatter Diffraction Matthew M. Nowell1, Stuart I. Wright1, Travis Rampton2, Ryan J. Stromberg3, Sanjit Bhowmich3, Masateru Shibata4 and Natasha Erdman4 1. 2. EDAX Inc., Draper, UT, USA EDAX Inc., Mahwah, NJ, USA 3. Hysitron Inc., Minneapolis, MN, USA 4. JEOL USA Inc., Peabody, MA, USA Electron Backscatter Diffraction (EBSD) has developed into a well-established microstructural characterization technique that provides quantifiable information on the grain size and shape, grain boundary character, preferred orientation, and local strain state of crystalline materials. Recent developments in pattern detection technology have enabled EBSD data acquisitions speeds to increase to rates greater than 1,400 analysed EBSD pattern per second. To achieve these speeds, acquisition hardware must be coupled with efficient software algorithms that can rapidly both detect the band positions within the EBSD patterns and use this band position information to accurately determine the crystallographic orientation. This image processing requirement is made more difficult at faster acquisition speeds due to typical reduction of the effective EBSD detector resolution. EBSD mapping results from a Nickel alloy, with a cubic crystal structure, collected at 1400 indexed patterns per second with a 99% indexing success rate and a Titanium alloy, with a hexagonal crystal structure, collected at 1150 indexed patterns per second with a 97% indexing success rate are shown in Figure 1. Optimization of both camera acquisition and software image processing settings are required to obtain high speed data with acceptable indexing rates. Sample preparation and SEM operating conditions also play key roles in obtaining faster EBSD acquisition speeds with acceptable indexing success rates. For successful EBSD pattern indexing, the signal to noise ratio (SNR) in the pattern must be sufficiently high to allow accurate band detection. Poor sample preparation reduces SNR while quality sample preparation increases SNR. SNR is also dependent on the camera amplification settings used to collect the digital image. As the amplification level (gain) is increased, SNR decreases. This can be tolerated up until band detection is no longer reliable. An alternative approach is to increase SEM beam current, which increases the incident backscattered electron signal and reduces the time necessary to obtain a useable pattern. With this approach, less electron gain is required and higher SNR patterns are collected. Modern FEG SEM instruments are able to deliver high beam currents without sacrificing spatial resolution. While 1400 indexing points per second represents a significant increase in acquisition speed for EBSD mapping data, other approaches are able to provide qualitative and quantitative microstructural information more rapidly. One such approach is the use of an EBSD detector as an electron imaging device [1]. With this approach, multiple regions of interest (ROI) are defined within the EBSD phosphor screen, and each ROI is used as a backscattered electron detector and imaging channel. Multiple ROIs (up to 25 with this implementation) can be imaged simultaneously, and these images can provide orientation, deformation, topographic, and atomic number contrast images of the microstructure. Because the camera can be operated at pixel resolutions less than those required for band detection, Microsc. Microanal. 21 (Suppl 3), 2015 1174 acquisition speeds greater than 2000 frames per second can be achieved. This approach can also be used to determine the position of grain boundaries, which can in turn be used to reduce the number of measurements needed to represent a given microstructure. The microstructural information obtained can also be correlated with other characterization information. Figure 2 shows an example of the plastic strain field that develops around a nanoindentation. An in situ nanoindentation using PI 87 SEM PicoIndenter (Hysitron, Inc., Minneapolis, USA) was made near a grain boundary, and the plastic strain field, as visualized by the Local Orientation Spread (LOS) map [2] does not propagate through this boundary. As the deformation response is strongly influenced by the local orientation and grain boundary character, the correlation of nanoindentation location with measured microstructure helps better understand local material response [3]. This correlation can be also extended to other microanalysis characterization techniques. For example, Energy Dispersive Spectroscopy (EDS) collected at a lower voltage to improve spatial resolution can be correlated with the EBSD data collected at a higher voltage. [1] S. Wright et al, Ultramicroscopy 148 (2015), p. 132. [2] S. Wright, M. Nowell, and D. Field, Microscopy and Microanalysis 17 (2011), p. 316. [3] C. Fizanne-Michel et al, Materials Science and Engineering A 613 (2014), p. 159. Figure 1. EBSD orientation maps of (left) a nickel alloy collected at 1400 indexed patterns per second with an indexing success rate of 99% and (right) a titanium alloy collected at 1150 indexed patterns per second with an indexing success rate of 97%. Figure 2. EBSD orientation map (left) and local orientation spread map (right) showing plastic deformation response to nanoindentation near selected grain boundary.");sQ1[587]=new Array("../7337/1175.pdf","Electron Microscopy Analysis of Grain Boundary Corrosion in Ni-Cr Binary Alloys Exposed to High Temperature Hydrogenated Water","","1175 doi:10.1017/S1431927615006662 Paper No. 0587 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electron Microscopy Analysis of Grain Boundary Corrosion in Ni-Cr Binary Alloys Exposed to High Temperature Hydrogenated Water Matthew J. Olszta1, Daniel K. Schreiber1 and Stephen M. Bruemmer1 Energy and Environment Directorate, Pacific Northwest National Laboratory, PO Box, 999 Richland, WA USA. Intergranular (IG) corrosion and stress corrosion cracking are important issues in current and advanced nuclear systems, and the understanding the controlling mechanisms are of great concern to the nuclear industry. In order to better elucidate fundamental aspects, this study focuses on exposure of coupons of varying Cr content in Ni-Cr binary alloys. Detailed evaluation of Cr concentration effects on protective film formation and localized corrosion provides underpinning processes controlling behavior of complex Ni-base alloys in light-water reactor service. Coupons of Ni-5Cr, 10Cr, 20Cr and 30Cr binary alloys were polished to a colloidal silica finish to minimize surface damage and expose the bulk grain boundaries. The alloys were exposed to simulated pressurized water reactor primary water environments at 320-360�C for 500-3500 h [1]. After exposure, cross-sections of each specimen were characterized using low kV scanning electron microscopy (SEM) in backscatter mode as well as transmission and scanning transmission electron microscopy (TEM/STEM) using both electron energy loss and energy dispersive spectroscopies (EELS and EDS, respectively). Alloys with lower Cr concentrations (Ni-5Cr and Ni-10Cr) exhibited IG corrosion to depths on the order of 10s of micrometers (with the deepest for each reaching ~30 �m), whereas when the Cr concentration reached 20 at% the IG corrosion was observed on the order of a few micrometers. Typical examples of these corrosion structures are shown in Figure 1. At 30 at% Cr, there was no IG corrosion, as a thin protective surface oxide was observed. Representative TEM/STEM analyses at the leading exposure front for both the lowest and highest Cr content, Ni-based binary alloys is shown in Figures 2 and 3 illustrating Cr depletion and diffusion induced grain boundary migration (DIGM) [2]. At the leading IGA in the Ni-5Cr, the Cr depletion is on the order to 10s of nanometers wide with depletion as low as 0.5 at% Cr (Figure 2). The final position of the grain boundary is observed in the annular darkfield (ADF) image adjacent to the oxidation front and the initial boundary position is indicated by the extent of the Cr depletion. ADF and energy filtered images of the Ni-30Cr illustrate that while DIGM also occurs on the order of hundreds of nanometers (with depletion down to 1-2 at% Cr), there is no IG corrosion present (Figure 3). A thin film of chromia passivates the grain boundary and prevents localized corrosion. Analysis of the various morphologies and levels of depletion of Cr along grain boundaries assists in the understanding Ni-Cr alloy performance in nuclear reactor components. [1] D.K. Schreiber et al, Microscopy & Microanalysis (2013), 19(3), 678-687. [2] R.W. Balluffi and J.W. Cahn, Acta Metallurgica (1981), Vol. 29, 493-500. The authors acknowledge funding from the Department of Energy, Office of Basic Energy Sciences under contract DE-AC06-76RLO 1830. 1. Microsc. Microanal. 21 (Suppl 3), 2015 1176 Ni5Cr Ni10Cr Intergranular attack Intergranular attack Ni20Cr Ni30Cr Intergranular attack No intergranular attack Figure 1. SEM backscatter images of Ni-5Cr, Ni-10Cr, Ni-20Cr and Ni-30Cr exposed at 600 h in hydrogenated water at 360�C illustrating the exposure depths (or lack thereof in Ni-30Cr). Cr depletion Leading intergranular attack Figure 2. STEM EDS elemental maps of exposed Ni-5Cr showing the extent of Cr grain boundary depletion ahead (down to 0.5 at%) of the leading intergranular attack front. Ni-L Surface Cr depletion Grain boundary migration Thin chromia layer Cr-L O-K Figure 3. Annular DF and GIF elemental maps of exposed Ni-30Cr showing the extent of Cr grain boundary depletion near the surface as well as protective chromia formation above the grain boundary.");sQ1[588]=new Array("../7337/1177.pdf","The Effects of Electron Beam Melting on the Microstructure and Mechanical Properties of Ti-6Al-4V and Gamma-TiAl","","1177 doi:10.1017/S1431927615006674 Paper No. 0588 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Effects of Electron Beam Melting on the Microstructure and Mechanical Properties of Ti-6Al-4V and Gamma-TiAl Tait McLouth1, Yuan-Wei Chang1, John Wooten2, Jenn-Ming Yang1 1. 2. Department of Materials Science and Engineering, University of California, Los Angeles, U.S.A. CalRam Inc., Simi Valley, CA, U.S.A. Titanium alloys have been used extensively in the aerospace and biomedical industries due to their high strength to weight ratios, elevated temperature mechanical properties, excellent biocompatibility, and good corrosion resistance [1-4]. Alloys, such as Ti-6Al-4V (Ti-6-4) can be used for replacement hip joints, knee joints, and bone plates because of the aforementioned properties. Titanium-aluminum intermetallic alloys such as -TiAl are attractive for high temperature turbine engine components because of their good thermal stability and low density [5]. Recently, both of these alloys have been manufactured with additive manufacturing (AM) because traditional methods such as casting and forging present problems and limitations [5]. AM provides more design flexibility for titanium alloys, a great benefit when considering the complexity of certain parts made for biomedical implants or jet engines. Electron beam melting (EBM) is a powder processing AM technique that produces fully net shaped parts from a bed of powder, and it is the main focus of this study. In this research, the microstructure and mechanical properties of both Ti-6-4 and -TiAl were studied before and after the EBM process. X-Ray diffraction (XRD), nanoindentation, and micropillar compression were performed to gain an understanding of the effects of the EBM manufacturing process. Microstructural evaluation was performed with the use of a scanning electron microscope (SEM). Figure 1 (a)(b) shows the microstructure of Ti-6-4 and -TiAl, respectively. Both alloys form a fine lamellar microstructure of alternating phases; these needle like Widmanst�tten structures serve to strengthen the alloys by reducing crack propagation through the material. Micropillars prepared by focused ion beam (FIB) milling were compressed by a nanoindenter in order to gather the yield strength and Young's modulus. Stress/strain curves for micropillars are shown in Figure 2(a) and (b) for Ti-64 and -TiAl, respectively. Tabulated values for the experimentally calculated compressive yield strengths, hardnesses, and Young's Moduli are shown in Table 1. From Table 1, it is clear that the EBM manufacturing process has a positive effect on the mechanical properties of Ti-6-4 and -TiAl. Compared to a cast sample of Ti-6-4 that underwent identical testing, the EBM sample displayed yield strengths that were 39% higher on average. This is due to the microstructure that is formed upon cooling. Specifically for Ti-6-4, the phase that forms enhances the mechanical properties, as it has a higher strength than the phase and also acts as a strengthening phase [6]. Referring to Figure 1, very fine spacing of the lighter phase can be observed. The mechanical properties found for -TiAl agree well with calculated and experimental values from the literature with a Young's modulus of 179 � 5 GPa. From this research it can be concluded that the manufacturing process plays a significant role in the final mechanical behavior of a material. In the case of Ti alloys, there seems to be a strengthening effect due to the faster cooling rate and favorable microstructure that forms as a result. Future work to be performed will involve a TEM analysis of the deformation mechanisms of both materials as well as an analysis of the base powder from which EBM samples are produced. Microsc. Microanal. 21 (Suppl 3), 2015 1178 References: [1] Y. Okazaki, S. Rao, Y. Ito, T. Tateishi, "Corrosion resistance, mechanical properties, corrosion fatigue strength and cytocompatibility of new Ti alloys without Al and V," Biomaterials 19 (1998) 1197-1215. [2] Y. Okazaki, E. Nishimura, H. Nakada, K. Kobayashi, "Surface analysis of Ti-15Zr-4Nb-4Ta alloy after implantation in rat tibia," Biomaterials 22 (2001) 599-607. [3] E. Eisenbarth, D. Velton, M. M�ller, R. Thull, J. Breme, "Biocompatibility of beta-stabilizing elements of titanium alloys," Biomaterials 25 (2004) 5705-13. [4] M. Niinomi, "Biologically and Mechanically Biocompatible Titanium Alloys," Mater. Trans. 49 (2008) 2170-8. [5] S. F. Franzen, Joakim Karlsson, "-Titanium Aluminide Manufactured by Electron Beam Melting," Sanna Fager Franzen, Joakim Karlsson (2010). [6] William F. Smith, "Structure and Properties of Engineering Alloys", Second Ed. (New York, NY: McGraw-Hill, 1993) 201-245. (a)! (b)! !10m! !5m! Figure 1: SEM images of microstructures for (a) Ti-6-4 showing the V rich phase (lighter) and Al rich phase (darker) and (b) -TiAl showing the 2-Ti3Al phase (lighter) and -TiAl phase (darker) (a)$ 1600 1400 (b)$ 1600 1400 Stress (MPa) 1000 800 600 400 200 0 0 0.01 0.02 0.03 0.04 0.05 Stress (MPa) 1200 1200 1000 800 600 400 200 0 0 0.02 0.04 Pillar 1 Pillar 2 Pillar 3 Pillar 1 Pillar 2 Pillar 3 Pillar 4 Pillar 5 Strain Strain 0.06 0.08 0.1 Figure 2: Micro-compressive stress-strain curves of (a) Ti-6-4 and (b) -TiAl Material Yield Strength EBM Ti-6-4 1135 � 12 MPa Cast Ti-6-4 812 � 26 MPa EBM -TiAl 620 � 21 MPa Table 1: Mechanical properties of tested samples Young's Modulus 114 � 6 GPa 116 � 2 GPa 179 � 5 GPa Hardness 4.5 � 0.3 GPa 4.1 � 0.2 GPa 5.3 � 0.2 GPa");sQ1[589]=new Array("../7337/1179.pdf","Electron Microscopy Investigations on Catalyst Coatings","","1179 doi:10.1017/S1431927615006686 Paper No. 0589 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electron Microscopy Investigations on Catalyst Coatings Yuan Zhao1, Larry Cirjak2, Dogan Ozkaya3 1 Johnson Matthey Process Technologies, Inc. 115 Eli Whitney Blvd, Savannah GA, USA 2 Johnson Matthey Process Technologies, Inc.785 N. Freedom Street, Ravenna, OH, USA 3 Johnson Matthey Technology Centre, Blounts Court, Sonning Common, Reading, RG49NH, UK Johnson Matthey develops and manufactures high technology catalysts and chemicals, many of which provide environmental and quality of life benefits, such as automobile emission control catalysts, oil refinery catalysts and additives, catalysts for fine chemicals and fuel cell catalysts etc. In the development and utilization of catalyst materials, especially the precious metals, small particle size and high dispersion on the support are very important to optimize the catalysis performance as well as to get the right combination of porosity and interconnectivity.[1-5] Most of the catalysts are made into products by the use of washcoat coating technology. The catalysts are usually formed into a washcoat with several components and coated on a substrate. The way coating is applied depends on functionality and it can be more than one pass which will result in a system with several layers with different functionalities. One of the biggest problems for transmission electron microscopy is to be able to see all the layers in one sample so that their properties can be related to performance. Here we investigated a two-layer system which contains Ni on alumina coating as the bottom layer followed by an alumina only layer by using a novel marking method to locate positions of layers. One such washcoat structure is shown in Figure 1 where the SEM backscattered electron images of a washcoat with two layers are shown. The 1st layer is at the surface and does not contain any Ni but the layer below contains Ni. The way the Ni diffuses into the first layer is very important. SEM resolution limits the observation and microanalysis, and higher resolution analysis is required in case Ni diffuses in small particles. S/TEM would normally be used to investigate the features at high magnifications, but scratching washcoat off substrate and microtoming it would mix the layers and the correlative information between different layers would be lost. The method applied here is to deposit a gold layer on the surface using a sputter coater. The layer beneath the gold layer is the 1st layer. Once the gold layer is deposited, then the washcoat is flaked off and embedded in resin and microtomed. The thin slices can be observed in S/TEM by HAADF imaging as shown in Figure 2. The top layer with high brightness is gold. The 1st and 2nd layers could be found corresponding to the layers observed on SEM images. By using this method, the change in chemistry from the top layer to the bottom one could be studied precisely in S/TEM, such as the Ni diffusion on the top layer (Figure 3). Also, the adhesion and structure between the layers could be studied at higher magnifications. It is clear the adhesion is so good that the only way the layers could be distinguished is the nanoparticulate Ni content of the layers. In this presentation, a few case studies of electron microscopy applications in industrial catalysts and materials research and development are presented. The examples of electron microscopy technique development work, such as particle size and distribution analysis, HAADF, bright field and secondary electron imaging optimizations on various samples, will be discussed. References: [1] L. C. Gontard, D. Ozkaya, and R. E. Dunin-Borkowski, Ultramicroscopy 111, 101 (2011) [2] L. C. Gontard, R. E. Dunin-Borkowski, R.K.K. Chong, D. Ozkaya et al, J. Phys. Conf. Ser. 26, 203 (2005). Microsc. Microanal. 21 (Suppl 3), 2015 1180 [3] Y. Zhao, Y. Tang, G. Vaughan, D. Ozkaya, Microscopy and Microanalysis 18 (S2), 1364 (2012) [4] Y. Zhao, T. E. Feltes, J. R. Regalbuto et al, Catalysis Letters 141(5), 641 (2011) [5] Y. Zhao, T. E. Feltes et al, Catalysis Science & Technology 1(8), 1483 (2011) Figure.1. The SEM backscattered electron images of catalyst washcoat. Figure.2. The S/TEM HAADF image of washcoat thin slice marked with gold layer. Figure.3. The interface between layers. The only difference is the lack of Ni nanoparticles at 1st layer.");sQ1[590]=new Array("../7337/1181.pdf","Spatially-Resolved EELS Analysis of Surface Chemistry of Metallic Lithium for the Development of Li-Air Battery.","","1181 doi:10.1017/S1431927615006698 Paper No. 0590 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Spatially-Resolved EELS Analysis of Surface Chemistry of Metallic Lithium for the Development of Li-Air Battery. M.L. Trudeau1, R. Veillette1, P. Bouchard2 and K. Zaghib2 1 Materials Science, Hydro-Qu�bec Research Institute, 1800 Boul. Lionel-Boulet, Varennes, Qu�bec, Canada, J3X 1S1 2 Energy Storage and Conversion, Hydro-Qu�bec Research Institute, 1800 Boul. Lionel-Boulet, Varennes, Qu�bec, Canada, J3X 1S1 Li-air battery is considered one of the most desired technologies for electric vehicles because of its high theoretical specific energy (more than 3500 Wh kg-1, including oxygen), which could enable a driving distance of > 500 km [1]. However, there are many technical challenges that hinder the practical use of lithium-air batteries. To address these technical challenges, significant research has been dedicated in the last 5 years on investigations of electrolytes, electrode materials and structures, catalysts, binders and anodes [1]. Despite progress in understanding the mechanisms associated with the electrode reactions, the current lithium-air technology is far from the requirements for use in electric vehicles, especially with respect to the cycle life and the power capability [1]. The further development of Li-air battery will necessitate a better understanding of the microstructure and chemical nature of Li metallic sheet. Metallic lithium is highly sensitive to hydrogen, oxygen, nitrogen, and carbon dioxide, which are the major components of wet air, and is more likely to form rapidly LiOH, Li2O, Li2O2, Li3N, and Li2CO3 species [2, 3] when exposed to air and/or moisture. Figure-1a shows XPS Li1s spectra acquired by depth-profiling through the different surface layers found on a pure metallic Li sheet while Figure-1b gives a representation of these different layers with typical thicknesses. Because of this very high surface reactivity and the low hardness of Li, it is thus very difficult to prepare thin sample from a specific Li sheet for direct TEM observations and chemical analysis of the material structure and surface chemical nature. In this work we present a first direct SR-TEMEELS study of the surface chemical nature of Li sheets. We have used a Hitachi air-protection cryogenic FIB holder to prepare thin Li samples and study the surface chemistry using SpatiallyResolved EELS (SR-EELS). The thin samples were prepared from different Li sheets using FIB (NB5000 from Hitachi) with the sample temperature at ~ -90 �C. Figure-1c shows a typical FIB sample that was obtained. The samples were transferred under vacuum using the same holder into a Hitachi HF-3300 cold FEG E-TEM for EELS analysis at low temperature without air contact. Spatially-resolved EELS was done using a high resolution GIF system (Gatan Quantum ER). Figuire-2a shows a TEM image selected with the rectangular slit while Figure-2b), c) and d) show the spectrum images for the Li-K-edge, O-K-edge and C-K-edge respectively. Figure-3a) presents again a Li-K-edge spectrum image while Figure-3b) shows four extracted spectrum taken at different positions on the sample. One can see the clear the change in the Li-K-edge as a function of the depth from the surface (#1) to the last one (#4) that is representative of metallic Li. [1] M.H. Cho, J. Trottier, C. Gagnon, P. Hovington, D. Cl�ment, A. Vijh, C.-S. Kim, A. Guerfi, R. Black, L. Nazar and K. Zaghib, J. Power Source 268, 565-574 (2014) [2] N. Brodusch, K. Zaghib and R. Gauvin, Micros. Res. & Tech. 78, 30-39 (2015) Microsc. Microanal. 21 (Suppl 3), 2015 1182 [3] D. Jeppson, J. Ballif, W. Yuan and B. Chou, 1978. Lithium literature review: Lithium's properties and interactions. Richland, Washington: Hanford Engineering Development Lab. [4] The authors would like to thank the assistance of the staff at Hitachi High Technologies for some samples observation. a) a) b) c) 50 nm Binding Energies (eV) Figure 1. a) XPS spectra as a function of depth for a typical metallic Li sheet; b) A representation of the typical chemical species (passivation layers) at the surface of a metallic Li sheet; c) example of a Li sample taken from such a Li metallic sheet through cryo-fibbing. b) Li K-edge c) O k-edge d) C K-edge a) 50 nm 50 eV 50 eV 50 eV Figure 2. SR-TEM-EELS images for a Li FIB samples. a) TEM image selected with the rectangular slit; b) Spectrum image for the Li K-edge; c) Spectrum image for the O K-edge and d) Spectrum image for the C K-edge Figure 3. a) SR-TEM-EELS images for a Li FIB sample with b) 4 EELS spectrum taken at different position/depth (from the near the surface at the top (#1) to the bottom (#4)). Spectra #4 is representative of metallic Li.");sQ1[591]=new Array("../7337/1183.pdf","High Efficiency Chemical State Surface Analysis Imaging with XPS","","1183 doi:10.1017/S1431927615006704 Paper No. 0591 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 High Efficiency Chemical State Surface Analysis Imaging with XPS Sankar Raman1, John F. Moulder1 and John S. Hammond1 1. Physical Electronics USA, Chanhassen, Minnesota, USA Surface analysis techniques are now commonly used for the characterization of surface modified materials such as biocompatible polymers and bio-implants, multilayer ultra-thin structures such as semiconductor dielectric gate structures and advanced data storage media as well as MEMS and NEMS structures. The most commonly used surface analysis techniques are Scanning Auger Microscopy (SAM), Time-of-Flight Secondary Ion Mass Spectrometry (TOF-SIMS) and X-ray Photoelectron Spectroscopy (XPS). The average analysis depth of all three techniques is less than 5 nm while the spectroscopic spatial resolution of the three techniques is 8 nm, 80 nm and 8 �m respectively. Today the XPS technique is unique in its ability to provide quantitative chemical state information but the information from XPS imaging information has generally been limited to semi-quantitative elemental information. XPS instruments have been designed with "microscope" modes that can translate the spatial distributions of photoelectrons to kinetic energy resolved images (with elemental identification) with a few microns spatial resolution. The high spatial resolution in the "microscope" mode is achieved by limiting the solid angle of collection. This mode limits the count rate at a high spatial resolution from acquiring the desired high sensitivity chemical spectroscopic information. A newer approach to XPS imaging is the microprobe XPS mode with a monochromatic scanning x-ray microprobe XPS. This approach uses a raster scanned, microfocused, x-ray source to define a microprobe spatial resolution of < 8 �m while allowing the energy analyzer to operate with a high solid angle for higher collection efficiency. Recently a 128 channel detector mode has been added to this scanning XPS system to allow the acquisition of an energy dispersed "snapshot" spectrum with chemical state information at each pixel. We will present two applications of this new scanning microprobe technology for high efficiency quantitative surface chemical state imaging. The surface chemistry of bond pads for semiconductor devices plays a vital role in the bond strength and device reliability. Figure 1a shows a 10 �m diameter scanning x-ray induced secondary electron image (SXI) of the bond pad structure. Figure 1b shows a rapid 12 minute Si 2p image of the bond pad with 128 channel spectra acquired at each pixel. Figure 1c shows the three basis spectra from three regions of the bond pad that are used for LLS fitting to define the chemical state images of silicon oxynitride (Figure 1d), silicon oxide (Figure 1e) and a mixture of tungsten silicide and silicon oxide (Figure 1f). Further data will be discussed in the presentation illustrating chemical state quantification. Development of fuel cell technologies has included the use of polymeric materials as proton exchange membranes (PEM) coated on either side of the polymer thin film with Pt electrodes. The lifetime of these fuel cell structures is influenced by the chemical modifications as a function of depth of the polymer films. Based on the 170 �m thickness of the PEM, XPS chemical state imaging of crosssections of the film can provide useful information on the chemical state degradation as a function of the use of the fuel cell. Figure 2a shows the optical micrograph of a cross-section of a used PEM structure. Figure 2b shows a 9 �m diameter scanning x-ray induced secondary electron image of the same area as the optical micrograph. Figure 2c shows the results of 8 minute C 1s chemical state imaging of the cross section showing overlay images of the PEM polymer chemistry (blue) and the chemical modified carbon material in the regions of the two electrodes (red). Further data will be discussed in the presentation of Microsc. Microanal. 21 (Suppl 3), 2015 1184 the chemical state modifications from the cross-sections of the polymer material. The presentation therefore illustrates the high efficiency of chemical state XPS imaging with a scanning x-ray microprobe. Figure 1: a) SXI of bond pad, b) Si 2p peak area map, c) reference spectra extracted from the Si map data, chemical state images of d) silicon oxynitride, e) silicon oxide and f) a mixture of tungsten silicide and silicon oxide Figure 2a Figure 2b Figure 2c Figure 2a shows the optical micrograph of a used PEM structure. Figure 2b shows a 10 �m diameter scanning x-ray induced secondary electron image (SXI) of a similar area as the optical micrograph. Figure 2c shows the results of 8 minute C 1s chemical state imaging of the cross section showing overlay images of the PEM polymer chemistry (blue) and the chemical modified carbon material in the regions of the two electrodes (red).");sQ1[592]=new Array("../7337/1185.pdf","An Analysis of the Particle Formation and Growth Process of the Nanocrystalline Diamond (NCD) Particles on Various Interlayers","","1185 doi:10.1017/S1431927615006716 Paper No. 0592 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 An Analysis of the Particle Formation and Growth Process of the Nanocrystalline Diamond (NCD) Particles on Various Interlayers Dong-Bae Park1, Chan-Hyoung Kang2, Dong-Hyuk Cha3 1,3 Precision Analysis & Development Lab, Cooperative Equipment Center, Korea Polytechnic University, Shihung , Korea 2 Departments of Advanced Materials Engineering, Korea Polytechnic University, Shihung , Korea Since diamond has good material properties such as high hardness, excellent wear resistance and high thermal conductivity, it is widely used in cutting tools or wear resistant materials. Two major types of industrial diamond are Microcrystalline Diamond (MCD) and Nanocrystalline Diamond (NCD). MCD is easy to make and cheap, but it has some disadvantages such as high surface roughness and high residual stress. To overcome these difficulties, many researches on NCD have been actively conducted in these days [1, 2]. According to the research, using an interlayer during the NCD deposition process can block the cobalt dissolution and carbon absorption, and can reduce the residual stress [3, 4]. In this paper, we analyze the particle formation and growth process of the NCD particles on two types of interlayer, tungsten (W) and titanium (Ti), which are known as good materials for interlayer. The procedure of the investigation is as follows: for the Ti interlayer, at first Ti was sputtered on silicon (Si) substrates. The sputtering process had been performed on a DC magnetron sputter until depth of the sputtered interlayer became 1. Secondary, the sputtered Ti layer was scratched in an ultrasonic bath contains nano size diamond powder. At third, NCD was deposited on the scratched Ti layer at 600 in 2.45 microwave plasma CVD system. For the W interlayer, the same process was conducted. Fig.1 shows the AFM (Atomic Force Microscopy) images of the surfaces of the W interlayers and Ti interlayer. The surface of W has a barley grain-like shape (Fig.1 (a)), while that of Ti a needle-like shape (Fig.1 (b)). The roughness average measurement of interlayer is W=3.39nm, and Ti =8.9nm. Fig.2 shows the SEM (Scanning Electron Microscope) images of NCD layers deposited on the interlayers. At t=0.5 hour, the NCD/W particles (Fig.2 (a)) are generated slower than the NCD/Ti particles (Fig.2(c)), while at t=2 hours the growth of NCD/W particles (Fig.2 (b)) is faster than that of NCD/Ti (Fig.2 (d)) particles. In the case of the W interlayer, NCD particles were coalesced and evolved to a film within 0.5 hour. It is mainly due to the fact that the diffusion of carbon species on the W interlayer was fast. The slower diffusion of carbon on the Ti interlayer is considered due to be slower film growth than that on the W interlayer. To obtain depth data, the two specimens are cut by a FIB (Focused Ion Beam) and the sections are observed by a STEM (Scanning Transmission Electron Microscope) as shown in Fig.3. The figure reveals that the gap of void on the NCD/W (Fig.3 (a)) is narrower than that of the NCD/Ti (Fig.3 (b)), and the thickness of the NCD/W is higher than that of the NCD/Ti. From the above results it can be concluded that the organizational structure of the NCD/W is better than that of the NCD/Ti. The reasons for this are as follows: The size of particles of the NCD/W in the particle formation process is bigger. In the growth process, therefore, the coalesced speed is faster and the gap of void is also narrower than that of the NCD/Ti. In this paper, the formation and growth process of the NCD on the W and Ti layers are Microsc. Microanal. 21 (Suppl 3), 2015 1186 investigated which is very helpful to analyze many types of NCD deposition processes on various interlayers. References : [1] J. E. Field, Reports on Progress in Physics 75 (2012) 12. [2] D. M. Gruen, Annu. Rev. Mater. Sci. 29 (1999) 211. [3] B.-K. Na, C. H. Kang, J. Kor. Inst. Surf. Eng. 46 (2013) 68. [4] H. Guo, Y. Qi, X. Li, J. Appl. Phys., 107 (2010) 033722. (a) W (b) Ti Fig. 1 AFM images of the W and the Ti interlayer. (a) NCD/W (at 0.5h) (b) NCD/W (at 2h) (c) NCD/Ti (at 0.5h) (d) NCD/Ti (at 2h) Fig. 2 SEM images of the NCD/W and the NCD/Ti. (a) NCD/W (at 2h) (b) NCD/Ti (at 2h) Fig. 3 STEM images of the interface of the NCD/W and the NCD/Ti.");sQ1[593]=new Array("../7337/1187.pdf","Characterization of the Thickness and Distribution of Latex Coatings on Polyvinylidene Chloride Beads by Backscattered Electron Imaging","","1187 doi:10.1017/S1431927615006728 Paper No. 0593 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of the Thickness and Distribution of Latex Coatings on Polyvinylidene Chloride Beads by Backscattered Electron Imaging Clifford S Todd1, Douglas E Beyer2 1. 2. The Dow Chemical Company, Analytical Sciences, Midland, MI USA The Dow Chemical Company, Packaging and Specialty Plastics R&D, Midland, MI USA Polyvinylidene chloride (PVDC) copolymer resins including SARANTM resins produced by The Dow Chemical Company are commonly used to produce barrier films for food packaging applications [1]. PVDC polymers are particularly advantageous in these applications because they have excellent barrier properties to the flow of oxygen and water vapor over a wide range of environmental conditions. PVDC resins are commonly formulated with a variety of solid additives, including stabilizers, lubricants, extrusion processing aides, colorants, nucleating agents and the like [1]. These are commonly added as small particles in a blending operation. While such a dry blending process is convenient, it does have its drawbacks. In a dry blend the components exist as individual particles, commonly with different particle sizes and/or densities. As a consequence, dry blended formulations are susceptible to segregation of the components during transportation and handling of the resin blend. Such maldistribution of additives can in turn have adverse effects on resin extrusion or film performance. In response, Dow developed a novel process to incorporate solid materials in latex form onto PVDC resin bead surfaces using a coagulation process [2]. This process gives a very uniform distribution of the additive on bead surfaces. Since it is locked onto the surface of the bead it also prevents segregation of the blend components. We developed microscopy techniques in response to the need to characterize the thickness and distribution of these latex coatings on PVDC resin beads. Using backscattered electron imaging chlorine-bearing material such as PVDC appears brighter than latex, composed predominantly of carbon and hydrogen. The depth reached by backscattered electrons is a function of the accelerating voltage used. As beam electrons enter a sample inelastic scattering events reduce their energy, eventually bringing them to rest. Higher energy electron beams travel deeper than lower energy beams before coming to rest. The depth reached by BSEs follows a similar trend. At a lower voltage of 5 keV the BSE signal comes mostly from the latex on top of the PVDC particle surface. The underlying PVDC can be seen in only a few patches where latex is absent or very thin (Figure 1A). At a higher voltage of 20 keV, the beam passes through the layer of latex; the underlying bright PVDC is evident (Figure 1C). The image collected at 10 keV is intermediate between these two; thick latex patches still appear dark, but thinner latex-coated areas appear brighter. Images such as these can be directly compared from sample to sample in order to assess relative differences in latex thickness and coverage. In order to quantitatively calibrate the BSE image brightness to latex thickness on the PVDC particle surface Monte Carlo simulations were used to model electron-sample interactions [3]. The backscatter coefficient is defined as the fraction of beam electrons that escape the sample as BSEs. Figure 2 shows the backscatter coefficient calculated from the Monte Carlo simulations as a function of accelerating voltage and latex layer thickness. The backscatter coefficient has been normalized between that in pure latex (zero) and pure PVDC (one) at the given accelerating voltage. At each voltage in Figure 2 the trend lines indicate that when the latex coating is thinner the backscatter coefficient is higher (brighter BSE image). The BSE coefficient becomes lower (darker BSE image) when the latex layer is thicker, as Microsc. Microanal. 21 (Suppl 3), 2015 1188 expected. The latex thickness at which the BSE image becomes dark is a function of accelerating voltage. If a normalized backscatter coefficient of 50% is chosen as the threshold, for imaging done at 5 keV a latex layer 100 nm thick or more would lead to a dark BSE image. Similarly, for 10 keV BSE images, dark areas on beads must have more than 0.35 microns of latex on the surface. For 15 keV BSE images, dark areas have more than 0.75 microns of latex; for 20 keV BSE images dark areas have more than 1.3 microns of latex on the surface. This thickness calibration was validated by cross-sectioning using a focused ion beam SEM. � TMTrademark of The Dow Chemical Company References: [1] BA Howell, and DE Beyer, Encyclopedia of Polymer Science and Technology. DOI: 10.1002/0471440264.pst391.pub2. [2] SM Kling, Patent US6627679 B1. [3] D Drouin et al, Scanning 29 (2007) 92�101. Figure 1. Latex-coted PVDC beads. BSE images of the same field of view at (A) 5, (B) 10 and (C) 20 keV accelerating voltages. At higher accelerating voltages the BSE signal from the underlying PVDC is detected through progressively thicker latex coatings. Figure 2. Modeled BSE coefficient as a function of latex layer thickness and accelerating voltage.");sQ1[594]=new Array("../7337/1189.pdf","In Situ TEM Studies of Li and Na ion Transport and Li/Na-Induced Phase Transitions in Crystalline Materials","","1189 doi:10.1017/S143192761500673X Paper No. 0594 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In Situ TEM Studies of Li and Na ion Transport and Li/Na-Induced Phase Transitions in Crystalline Materials Reza Shahbazian-Yassar1, 2, 3, 4 Department of Mechanical Engineering-Engineering Mechanics, Michigan Technological University, 1400 Townsend Drive, Houghton, Michigan, United States. 2. Department of Materials Science and Engineering, Michigan Technological University, 1400 Townsend Drive, Houghton, Michigan, United States. 3. Department of Mechanical and Industrial Engineering, University of Illinois at Chicago, Chicago, IL 60607, United States. 4. Department of Physics, University of Illinois at Chicago, Chicago, IL 60607, United States. Although batteries are inherently simple in concept, surprisingly their development has progressed much slower than other areas of electronics. This slow progress is due to the lack of suitable electrode materials and electrolytes, together with difficulties in mastering the interfaces between them. Working out the drastic impact of interfaces on ions transport and understanding the interrelationship between the electrode microstructure and battery performances is important for the design of high performance lithium ion batteries. For instance, it is still not well understood that how defect structures such as twin boundaries affect the transport of ions within crystalline materials. Also, triggered by the recent exploration of alternative technologies to Li-ion batteries, sodium has become a viable alternative to Liion batteries. Na-ion batteries hold the potential to become the technology of choice for large-scale electrochemical energy storage thanks to the high sodium abundance and low costs. However, the investigation of suitable electrode materials for sodium ion batteries is still in its infancy and few electrode materials meet the performance requirement for practical applications. In-situ transmission electron microscopy (TEM) has been shown to be a very powerful technique in shedding light to some of the mysteries in electrochemical performance of new materials. Using this technique, here, we report our new findings on Zn-Sb intermetallic alloys, SnO2 nanowires, and MnO2 nanorods. Sb-based intermetallics, such as Zn-Sb system, which have been proved to be promising anode materials for Li-ion batteries, are also capable of storing of sodium ions. We investigated the microstructural changes and phase evolution of the Zn-Sb intermatellic nanowires using in-situ TEM1. Our results indicate that the reaction between Zn-Sb and sodium proceeds through a different pathway during the first compared to the subsequent cycles. After first sodiation, the initial single crystal Zn4Sb3 nanowire change into the crystallized Na3Sb and NaZn13 nanoparticles; upon extraction of Na+ (desodiation), the Na3Sb and NaZn13 phases develop into the crystallized NaZnSb and Zn particles. Atomic resolution imaging shows that NaZnSb has a layered structure, which provides channels for fast Na+ diffusion. We also explored the lithiation behavior of the individual SnO2 nanowires containing twin boundary (TB). Our in-situ TEM results show that the Li transport pathways will totally change when the (10 1) TB exists inside the SnO2 nanowires comparing with the single crystal SnO2, in which the lithium ions preferred to diffusion along the [001] direction.1 Direct atomic-scale imaging of partially lithiated TBSnO2 nanowire shows that the lithium ions prefer to intercalate in the vicinity of the (10 1) TB, which acts as a conduit for lithium ion diffusion inside the nanowires. - - 1. Microsc. Microanal. 21 (Suppl 3), 2015 1190 Furthermore, we utilized aberration-corrected scanning transmission electron microscopy (ACSTEM) to single -MnO2 nanowire to image the tunneled structures. Cross-sectional ACSTEM shows that the nanowire has a squared cross section and 2�2 tunnels align parallel to its growth direction [001]. An open cell design inside TEM for dynamic observation of MnO2's lithiation/delithiation process is also demonstrated. It is found that upon lithiation, the -MnO2 nanowire shows different orientation-sensitive morphologies. That is, -MnO2 unit cell expands asynchronously along [100] and [010] directions, resulting in macroscopic difference under [010] and [100] zone axes observations. Electron Energy Loss Spectroscopy further confirms such an asynchronous expansion property via quantification of Mn valence during lithiation. These findings provide fundamental understanding for how Li and Na can be transported within crystalline materials. We also have shown that the induced phase transformation can be different in the presence of defects. References: [1] A. Nie, R. Shahbazian-Yassar, et al. Nano Lett. 15 (2015), p. 610. [2] This work was supported by the National Science Foundation (Award No. CMMI-1200383 and DMR-1410560;) and the American Chemical Society-Petroleum Research Fund (Award No. 51458ND10).");sQ1[595]=new Array("../7337/1191.pdf","Electrochemical reactions in an all-solid-state Li-ion battery observed by ex situ and in situ spatially-resolved TEM EELS","","1191 doi:10.1017/S1431927615006741 Paper No. 0595 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electrochemical reactions in an all-solid-state Li-ion battery observed by ex situ and in situ spatially-resolved TEM EELS Kazuo Yamamoto1*, Atsushi Shimoyamada1, Takeshi Sato1, Ryuji Yoshida1, Hisanori Kurobe1, Hiroaki Matsumoto2, Tsukasa Hirayama1, Yasutoshi Iriyama3 1 2 Japan Fine Ceramics Center, 2-4-1 Mutsuno, Atsuta-ku, Nagoya, Aichi, 456-8587, Japan. Hitachi High-Technologies Corporation, 11-1 Ishikawacho, Hitachinaka, Ibaraki, 312-0057, Japan. 3 Department of Materials, Physics and Energy Engineering, Nagoya University, Furo-cho, Chikusa-ku, Nagoya, Aichi, 464-8601, Japan. All-solid-state lithium-ion batteries (LIBs) with incombustible solid electrolytes have advantages of safety, reliability, lifetime, cost, and energy density. However, a serious problem is the large interfacial resistance of Li-ion transfer at the electrode/solid-electrolyte interfaces. One effective solution is an in situ formation of electrode active materials from parent solid electrolytes. Because the electrodes grow from the solid electrolytes with Li insertion reaction, both materials become connected to each other at an atomic scale, leading to a low interfacial resistance. Such electrodes were discovered in Li2O-Al2O3TiO2-P2O5-based solid electrolytes (LATP) [1]. However, the electrochemical reactions, such as the structural growth mechanism and electronic structure changes, are still unclear. Here, we used ex situ and in situ spatially resolved TEM EELS (SR-TEM-EELS) to directly observe the nano-scale Li profiles and the influence on other elements, Ti and O [2]. Figure 1(a) shows a schematic illustration of the prepared LIB sample. A Si- and Ge-doped LATP sheet (LASGTP, 90-�m thick) was used as the solid electrolyte. The 800-nm thick LiCoO2 positive electrode was deposited on one side of the sheet by PLD. On the other side, the Pt current-collector was directly deposited. For the ex situ SR-TEM-EELS experiment, cyclic voltammetry (CV) was performed for 50 cycles in a vacuum with a sweep rate of 40 mV min-1 (Fig. 1(b)), and the negative electrode was irreversibly formed near the LASGTP/Pt interface by decomposition with the Li insertion. After the CV, the TEM sample of the negative side region was prepared by FIB. Figure 2(a) shows the TEM image around the negative side. A slightly uniform contrast layer (about 400 nm) was observed near the Pt. Electron diffraction showed this region was amorphous structure. The SR-TEM-EELS images around the Li-K-edge, Ti-L-edge, and O-K-edge were recorded by the GIF CCD camera (Figs. 2(b) - 2(d)). The Li signals increase in the 400-nm-width region. It means that the amorphous negative electrode was formed in this region. In the spectrum image of Ti-L-edge (Fig. 2(c)), we can observe clear chemical shifts of the L2 and L3 edge lines, which shows that the Ti electronic state changed from Ti4+ to Ti3+ due to the Li insertion reaction. The spectrum image of O-K-edge in Fig. 2(d) also shows the spectrum shifts, indicating that the O also contributed to the electron charge compensation due to the Li insertion. We will also report the in situ observation results showing the changes of the Li concentration profiles during charge and discharge processes. Microsc. Microanal. 21 (Suppl 3), 2015 1192 In summary, we successfully observed the electrochemical reactions using ex situ and in situ SRTEM-EELS, and revealed how the negative electrodes were formed by Li insertion reaction and what influenced on the other important elements. References: [1] Y. Iriyama et al., Electrochem. Commun. 8 (2006) 1287. [2] K. Yamamoto et al., J. Power Sources 266 (2014) 414. [3] This work was supported by the RISING project of New Energy and Industrial Technology Development Organization (NEDO), Japan. (a) Pt in-situ formed negative electrode solid electrolyte (LASGTP) positive electrode (LiCoO2) Au (b) 400 300 Current [nA] ca. 90 m 200 100 0 1st 10th 50th charge Li+ -100 -200 -300 0 discharge 0.5 1 1.5 2 Applied voltage between Au and Pt [V] FIG. 1 All-solid-state LIB sample and cyclic voltammogram (CV). (a) Illustration of the LIB sample. (b) Cyclic voltammogram measured in a vacuum with a sweep rate of 40 mV min-1. As the negative electrodes grow with cycling, the charge current peak becomes higher and sharper in the CV. (a) 800 nm Pt negative electrode (b) Li K-edge (c) Ti L-edge (d) O K-edge low Intensity high LASGTP Li+ 500 nm 60 70 450 460 470 530 540 550 560 570 Energy loss [eV] FIG. 2 TEM image and the SR-TEM-EELS images. (a) TEM image around the negative-electrode/LASGTP interfaces. SR-TEM-EELS images from the region of (a), (b) Li-K-edge, (c) Ti-L-edge, and (d) O-K-edge.");sQ1[596]=new Array("../7337/1193.pdf","In Situ Scanning Transmission Electron Microscopy (STEM) of Individual Electrochemical Intercalation Events in Graphite","","1193 doi:10.1017/S1431927615006753 Paper No. 0596 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In Situ Scanning Transmission Electron Microscopy (STEM) of Individual Electrochemical Intercalation Events in Graphite Jared J. Lodico1, E. R. White1, William A. Hubbard1, Erick Garcia1, Bradley Parks1, Brian Zutter1, and B. C. Regan1 1. Department of Physics & Astronomy and California NanoSystems Institute, University of California, Los Angeles, California 90095 USA Graphite intercalation compounds (GICs) are formed when ions or molecules (intercalants) are inserted between the carbon layers of a graphite host. With some electrolytes a reversible charge transfer process occurs during intercalation, making GICs attractive materials for batteries. The demand for improved batteries has highlighted the need for in situ measurements probing electrode-electrolyte interactions [1]. With in situ scanning transmission electron microscopy (STEM) we observe the reversible electrochemical intercalation of multi-layered (~20-100 layers) graphene in 96% sulfuric acid (H2SO4). Pristine, natural graphite is most commonly A-B (Bernal) stacked, where every other graphene layer B is horizontally offset from its neighbors A by the carbon-carbon bond length. Upon intercalation the crystal structure changes, and the graphene layers above and below the intercalant layer can shift. These shifts change the intensity of a coherently scattered electron beam. For example, if the GIC adopts the stacking A|A|A..., where | represents a layer of intercalant, the first order diffraction signal is expected to increase by a factor of ~4 while the second order signal remains unchanged [2]. Dark field STEM imaging can locate areas experiencing this shift as it occurs. Figure 1A illustrates the ideal imaging conditions, where only the first order Bragg reflections hit the annular dark field (ADF) STEM detector. Imaging GICs in this regime highlights changes in the graphite stacking order. We mechanically exfoliated graphite, transferred chosen flakes onto instrumented, electrontransparent membranes (Fig. 1B), added H2SO4, and assembled vacuum-tight fluid cells for in situ STEM [3]. Figure 2 shows three ADF images taken from a 25 minute video of a graphite flake undergoing electrochemical intercalation. The graphite was cycled from -16.4 mV, the open circuit potential (OCP), to 0.9 V relative to a platinum pseudo-reference electrode using a Gamry 600 potentiostat. The platinum counter and pseudo-reference electrodes are outside the field of view in Fig. 2 but visible in Fig. 1B. Figure 2A shows a STEM image of a graphite flake at the OCP. Electrochemical intercalation of the graphite caused the contrast variations seen in Fig. 2B. Because the pixels in STEM images are acquired serially (here with a pixel dwell time of 65 �s), fast intercalation events give spatial discontinuities in the observed contrast. Figures 2B-C were acquired sequentially and provide an example of such an event. The STEM raster scanned left to right and then top to bottom. The top ~25% of Figs. 2B-C are nearly identical, but at the time indicated by the red arrow in Fig. 2C the graphite abruptly shows contrast that is more uniform than in Fig. 2B, but much like that of Fig. 2A. The graphite returned to an approximation of its initial state almost instantaneously. Microsc. Microanal. 21 (Suppl 3), 2015 1194 The Fig. 2C inset shows the cyclic voltammogram section obtained simultaneously with the frames 2B-C. Time advances to the left, and black dotted lines indicate frame boundaries. A distinct deintercalation peak is coincident with the contrast change in Fig. 2C, showing a clear correlation between an electrical transport event and the structural changes in the graphite. References: [1] J. -M. Tarascon and M. Armand, Nature, 414 (2001) p. 359. [2] B. Shevitski et al., Physical Review B 87 (2013) 045417. [3] E. R. White et al., Appl. Phys. Express 4 (2011) 055201. [4] This work was supported by NSF DMR-1206849, and in part by FAME, one of six centers of STARnet, a Semiconductor Research Corporation program sponsored by MARCO and DARPA. The authors acknowledge the use of instruments at the Electron Imaging Center for NanoMachines supported by NIH 1S10RR23057 and the CNSI at UCLA. Figure 1. (A) ADF STEM conditions for capturing only the 1st order Bragg reflections. The camera length is adjusted so that the central beam and the 2nd order Bragg reflections miss the ADF detector. (B) A graphite flake is connected to the bottom platinum electrode on an electron transparent window. The other electrodes are used as counter and pseudo-reference electrodes. Figure 2. (A-C) ADF images of a single-crystal graphite flake intercalating and deintercalating for the second time. (A) Before intercalation. (B) Intercalating the graphite caused new contrast variations to appear. (C) The graphite sheet deintercalated at the point indicated by the red arrow. The inset shows the simultaneously-acquired electrical transport data. A (downward) current peak coincided with the contrast discontinuity in (C). The scan rate was 2 mV/s.");sQ1[597]=new Array("../7337/1195.pdf","Can Na+ Transport Faster Than Li+ inside Zn-Sb Intermetallic Nanomaterials?","","1195 doi:10.1017/S1431927615006765 Paper No. 0597 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Can Na+ Transport Faster Than Li+ inside Zn-Sb Intermetallic Nanomaterials? Anmin Nie1, 3, Yingchun Cheng4, Robert F. Klie3, Sreeram Vaddiraju2 and Reza Shahbazian -Yassar1 Department of Mechanical Engineering-Engineering Mechanics, Michigan Technological University, 1400 Townsend Dive, Houghton, Michigan 49931, United States 2. Artie McFerrin Department of Chemical Engineering, Texas A&M University, 3122 TAMU, College Station, TX 77843, United States 3. Department of Physics, University of Illinois at Chicago, Chicago, Illinois 60607, United States 4. Mechanical and Industrial Engineering Department, University of Illinois at Chicago, Chicago, Illinois 60607, United States Triggered by the recent exploration of alternative technologies to Li-ion batteries, sodium has strongly broken into energy storage research field thanks to the natural abundance and environmental benignity of sodium resources1, 2, 3. These advantages make Na-ion battery an attractive and potential alternative to the well established Li-ion battery. However, the development of Na-ion battery is currently a challenge because of potential disadvantages, including larger size of Na+ and higher redox potential of Na/Na+ compared to Li analogues. Here, an in-depth comparative study between the electrochemical de/lithiation4 and de/sodiation of Zn4Sb3 nanowires has been conducted by using in situ transmission electron microscopy. Surprisingly, we found that sodium ions transport can be 10~100 times faster than lithium ion inside individual Zn4Sb3 nanowires. In addition, the cracks were often observed in the first few cycles during de/lithiation of the Zn4Sb3 nanowire. However, there was no crack formed even after dozens of cycles during their de/sodiation. Our in situ study indicates that the Zn4Sb3 nanowires exhibit much better rate capability and cyclablility in Na-ion battery compared to Li-ion systems. The underlying reason has also been addressed from the thermodynamic and kinetic aspects of ions transport in Zn-Sb intermetallics. 1. Microsc. Microanal. 21 (Suppl 3), 2015 1196 Reference: [1] Slater, M. D. et al, Adv. Funct. Mater. 23, (2013), 947 [2] Pan, H. et al, Energy Environ. Sci. 6, (2013), 2338. [3] Palomares, V. et al, Energy Environ. Sci. 5, (2012), 5884. [4] Nie, A. et al, Nano lett. 14, (2014), 5301. [5] The authors acknowledge the funding from the National Science Foundation (Award No. CMMI-1200383) and the American Chemical Society-Petroleum Research Fund (Award No. 51458-ND10). The acquisition of the UIC JEOL JEM-ARM200CF is supported by an MRI-R2 grant from the National Science Foundation (Award No. DMR-0959470). Support from the UIC Research Resources Center is also acknowledged. Figure 1. (a) Low-magnification TEM image of the in situ electrochemical testing setup and a schematic depiction of the nano-battery. (b) Atomic scale HAADF image of Zn4Sb3 nanowire taken with the [1-10] zone axis. (c) Typical reaction front travel distance vs time curves for sodiation and lithiation, respectively. (d) Comparison of sodiation and lithiation rates of Zn4Sb3 nanowires. Sodiation rates of Zn4Sb3 are about 10~100 times faster than their lithiation rates.");sQ1[598]=new Array("../7337/1197.pdf","Reaction Mechanism and Kinetic of Graphene Supported Co3O4 Nanocubes with Lithium and Magnesium Studied by in situ TEM","","1197 doi:10.1017/S1431927615006777 Paper No. 0598 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Reaction Mechanism and Kinetic of Graphene Supported Co3O4 Nanocubes with Lithium and Magnesium Studied by in situ TEM Jinsong Wu1,2, Langli Luo1,2, Qianqian Li1,2, Vinayak P. Dravid1,2 Qunli Rao1,3 and Junming Xu1,4 1. 2. NUANCE Center, Northwestern University, Evanston, IL USA Department of Materials Science and Engineering, Northwestern University, Evanston, IL USA 3. Department of Materials Science and Engineering, Shanghai Jiao-Tong University, Shanghai, China 4. College of Electronic Information, Hangzhou Dianzi University, Hangzhou, China While most of current research on energy storage is focused on lithium-ion battery, the alternatives like magnesium and aluminum has many obvious advantages. For example, magnesium is a nature abundant element being the 5th most abundant element in the earth's crust. It is environmentally friendly, low price and has many safe characters, i.e. it is stable enough in ambient atmosphere to handle. However, currently there are many serious limitations in magnesium electrochemistry that prevents magnesiumion battery being an efficient system for energy storage. In this work, by using high resolution in-situ transmission electron microscopy (TEM) the diffusion of multivalent ions and the solid-state reactions with Co3O4 nanocubes and graphene have been studied, in order to explore the reaction mechanism for multivalent-ion batteries, in direct comparison to that of lithium-ion battery. Nano-composite materials, especially graphene-based nanostructure1-2 have been developed for high-capacity anode materials showing enhanced high electron and lithium ion conductivity by graphene. Meanwhile, the emerging insitu transmission electron microscopy (TEM) techniques with localized electrical measurement capabilities provide a practical platform to investigate electrochemical reactions in Li-ion battery materials by building a full or half "nano-cell" inside the TEM specimen chamber17. Such real-time observations of dynamic composition and microstructural evolution in the electrochemical reaction have provided many novel clues to understand the lithiation/de-lithiation mechanisms at nano or even atomicscale for several novel anode materials18. Herein, we report a morphological and structural study of graphene sheets supported Co3O4 nanocubes during the electrochemical reaction with lithium, and magnesium. Upon charging with lithium-ions, the Co3O4 nanocubes decompose to small Co metal nanoparticles (2-3 nm) and embedded in as-formed Li2O matrix; reversely, the CoO nanoparticles formed on the site of Co accompanying the decomposition of Li2O in the discharging process. The lithiation process is dominated by surface diffusion of Li+ and graphene sheets enhance the Li+ diffusion leading to a fast charging process. However, upon charge with magnesium, the Mg2+ diffusion is sluggish and there is no sign of conversion reaction between Mg and Co3O4 at room temperature. Instead, a thin film consisting of metal Mg nanoparticles is formed on the surface of graphene due to a process similar to metal plating. The Al3+ diffusion is even more sluggish and there is no electrochemical reaction between Al and Co3O4 can be observed at room temperature. The finding may shed light on the development of batteries with high energy density based on multivalent ions other than lithium. References: [1] H. Wang, et al. Journal of the American Chemical Society 132, (2010), p.13978. [2] X. Zhu, et al. ACS Nano 5, (2011), p.3333. [3] J.Y. Huang, et al. Science 330, (2010), p.1515. [4] L. Luo, et al. Scientific Reports 4, (2014), article number 3863. [5] L. Luo, et al. ACS Nano 8, (2014), p. 11560. Microsc. Microanal. 21 (Suppl 3), 2015 1198 [6] This work was supported as part of the Center for Electrochemical Energy Science, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences under Award # DEAC02-06CH11357. This work was also supported by the NUANCE Center new initiatives, and made use of the EPIC facility (NUANCE CenterNorthwestern University), which has received support from the MRSEC program (NSF DMR-1121262) at the Materials Research Center, The Nanoscale Science and Engineering Center (EEC-0118025/003), both programs of the National Science Foundation; the State of Illinois; and Northwestern University. Figure 1. Time-resolved TEM images of electrochemical lithiation process (a) to (d); and delithiation process (e) to (h) of Co3O4 nanocubes on graphene. The scale bar is 10nm. Figure 2. Microstructural evolutions in the magnesiation process. (a) The MgO/Mg probe and the Co3O4/Graphene nanocomposites loaded on a Au tip, prior to the electrochemical magnesiation. (b) TEM image of the reaction front of magnesiation after the electric bias of -5 V has been applied. The front moves forward about 240 nm in 20 minutes. (c) SAED pattern of Co3O4/Graphene nanocomposites before the magnesiation reaction. (d) SAED pattern of Co3O4/Graphene nanocomposites after the magnesiation reaction.");sQ1[599]=new Array("../7337/1199.pdf","Technologists' Forum: Safety in the Microscopy Laboratory","","1199 doi:10.1017/S1431927615006789 Paper No. 0599 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Technologists' Forum: Safety in the Microscopy Laboratory E. Ann Ellis1 and Beverly E. Maleeff2 1. 2. Consultant in Biological Electron Microscopy, Thomasville, GA GlaxoSmithKline, King of Prussia, PA Adherence to safety guidelines is a critical aspect of working in any laboratory, and particularly in microscopy laboratories where proper equipment use and maintenance, chemical handling, specimen handling and waste disposal are of the utmost importance. Microscopy labs often deal with hazardous materials such as osmium tetroxide and embedding media on an everyday basis. Safety education of the staff in these labs is essential for the health and safety for everyone involved with these labs. Education is an essential part of all laboratory safety! In general, the most important aspect of laboratory safety is to read, understand and implement the regulations and standard operating procedures in place in that lab or institution. In addition, there may be local or federal laws that impact the conduct of laboratory research. It is the responsibility of each employee and his/her direct supervisor to ensure that all governing rules are provided and made clear before any work is undertaken. Microscopy labs are often defined as either biological or materials science-oriented, dependent on the institution and the physical location of the equipment. While there are general laboratory rules that must be followed regardless of the types of specimens examined, each has rules and regulations specific to the discipline. For example, in a biological microscopy laboratory there must be biological hazard containers for sharps, broken glass, and discarded specimens; chemical fume hoods for the use of resin components; biological cabinets for handling unfixed animal tissues and fluids; spill kits; appropriate air handling systems; and appropriate signage on the outside of the laboratory door to notify anyone entering the laboratory about the proper personal protective equipment (PPE) needed to work safely. In a materials science laboratory, similar regulations must be in place, including containment systems specific to the type of specimens being prepared and examined. Chemical fume hoods are essential components of all electron microscopy laboratories since osmium tetroxide and embedding resins are important but hazardous materials, used every day in the operation of the labs. Maintenance of fume hoods at appropriate operating levels is a continuous part of the housekeeping in that these facilities should be monitored daily for operation and not be allowed to become a cultured storage area for hazardous wastes. A KimwipeTM check (tape a KimwipeTM to the sash with the sash opened to the proper operating height) should be performed daily at the hood operating opening to have a visual clue to whether there is a proper directional airflow. In the event that there is no or reduced airflow, work should be stopped and the proper environmental health and safety staff contacted. Hood flow face velocities should be checked on a schedule determined by local regulations. Sash height at a face velocity of 100 � 20 ft3/minute should be marked on the outside of the hood along with the date that it was last checked [1]. This flow rate of 100 ft3/min is adequate for toxic substances such as osmium, sodium cacodylate and lead and if the microscopist spends a major part of the work day using highly toxic chemicals, the flow rate should be increased to 150 ft3/min [2]. Use of cryogens has increased as our specimen preparation procedures have evolved and this presents additional hazards from both freezing burns to potential asphyxiation. Liquid nitrogen evaporates to Microsc. Microanal. 21 (Suppl 3), 2015 1200 gaseous nitrogen which can displace room air and care must be taken to insure adequate ventilation in areas where liquid nitrogen is worked with and stored. Radiation safety is equally important in a laboratory where radiation-generating instruments are used. In the United States, this is usually governed by state or local laws. At a minimum, this should consist of checking for radiation leakage at the time of installation, at yearly intervals thereafter, and at the time of any service including preventative maintenance. Any deviation from an expected reading must be reported immediately to your institutional radiation safety officer and corrected as soon as possible, and before the instrument is certified for use. Waste disposal is a critical part of working in any laboratory, and no more important than in a microscopy lab. Biological and chemical waste must be addressed individually, and are typically handled using institutional or local regulations. Specific care must be taken with carcinogenic or toxic substances, including, but not limited to, resin components, sodium cacodylate, strong acids and bases, and the dust and shards generated from trimming resin blocks or materials samples. Best laboratory practice includes the recording of all baseline equipment settings, equipment deviations, and maintenance and service records in appropriate log books or online databases so that any changes from baseline can be identified and corrected as required. In a lab that is subject to United States regulatory oversight, this is a legal requirement [3]. The most important safety rule is to use common sense! Plan your work, address the potential safety hazards associated with that work, and assemble everything needed to accomplish that work safely. If something does not seem right, then stop immediately and assess the situation. Contact your supervisor or safety officer before proceeding. References: [1] RS Stricoff and WB Walters. 1990. Laboratory health and safety handbook: a guide for the preparation of a chemical hygiene plan. John Wiley & Sons, Inc., New York. [2] VC Barber, 1994. in Electron Microscopy Safety Handbook,. VC Barber and JA Mascorro, eds. San Francisco Press, Inc., San Francisco. p. 3. [3] Code of Federal Regulations, Title 21: Food and Drugs, Part 58: Good laboratory practice for nonclinical laboratory studies. US Food and Drug Administration. [4] This paper does not necessarily reflect the approved practices of GlaxoSmithKline.");sQ1[600]=new Array("../7337/1201.pdf","Large-Scale 3D Heterogeneity Analysis of CryoEM Data Using Likelihood-Based Classification in Frealign","","1201 doi:10.1017/S1431927615006790 Paper No. 0600 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Large-Scale 3D Heterogeneity Analysis of CryoEM Data Using Likelihood-Based Classification in Frealign Dario Oliveira dos Passos1 and Dmitry Lyumkis1 1. The Salk Institute for Biological Studies, Laboratory of Genetics, La Jolla, CA 92037, USA Single particle cryo-electron microscopy (cryoEM) is an important component of a structural biologist's toolkit, as improvements in instrumentation, software, automation, and specimen preparation are making this technique increasingly powerful for the analysis of large (>100 kDa) macromolecules and macromolecular complexes [1]. A typical single-particle dataset consists of many individual noisy images (often thousands or tens of thousands, and referred to as "particles"), each of which is a characteristic view of a unique macromolecular assembly. One of the biggest advantages of the methodology is the ability to analyze heterogeneous macromolecular assemblies, i.e. those that exhibit either conformational mobility within distinct parts or compositional heterogeneity exhibited by loosely and sub-stoichiometrically associated components. By employing classification techniques, it is possible to place each individual particle into one of several, potentially many, groups, according to homogeneity. Although different approaches have been proposed in the past to address specimen heterogeneity through classification (reviewed in [2], also see [3]), this problem nevertheless remains challenging, and is one of the major areas of methods development [1]. Here we describe a maximum likelihood-based method for the analysis of macromolecular heterogeneity in single particle cryoEM, which is implemented within the Frealign package [4]. Particle alignment parameters are determined by maximizing a joint likelihood that can include hierarchical priors, while classification is performed by expectation maximization of a marginal likelihood. We show that this algorithm has multiple advantages over the methods previously described, specifically: 1. The implementation within the Frealign package for refinement ensures that over-fitting is minimized through the application of a weighted target function and a resolution-limited approach to raw particle refinement and classification. 2. The refinement of particle orientations can be readily separated from the refinement of classification parameters. We show that this leads to an improvement in the accuracy of classification and additionally speeds up the classification. 3. Focused classification can be readily performed by applying a mask onto a region of interest in the dataset. This enables one to specifically analyze mobility and/or compositional heterogeneity within a defined region of interest, while ignoring heterogeneity in the rest of the density. We provide a step-by-step protocol for implementing the algorithm with cryoEM data. We also provide experimental results with synthetic data � wherein the correct classification parameters are known � showing that separating the refinement of particle orientations from the refinement of classification parameters directly benefits and improves the accuracy of classification. We then apply the algorithm to the analysis of the Large Ribosomal Subunit-Associated Quality Control Complex (RQC), a macromolecular assembly consisting of the 60S large ribosomal subunit and multiple loosely bound components. This dataset exhibits a profound amount of heterogeneity, both compositional and conformational in nature. First, we show that it is possible to recover <3� information for the ribosomal core (Figure 1A-C). Subsequently, we show that, using a global classification approach, it is possible to Microsc. Microanal. 21 (Suppl 3), 2015 1202 recover several distinct conformational and compositional states of the large mobile components, including a sub-nanometer resolution reconstruction of the non-ribosomal proteins that are loosely associated to the periphery. Finally, we zoom in on the remaining areas of heterogeneity using a focused classification approach and show that it is possible to deconvolute the mobility of a 180 kDa protein bound to the large 60S ribosomal subunit, which was not possible using a global classification approach. Moreover, we show that small regions of mobility, corresponding only to a single RNA helix, can also be deconvoluted such that the RNA pitch becomes clearly evident (Figure 1D-G). Taken together, this shows that it is possible to recover structural information lost to mobility even for small regions. References: [1] X.-C. Bai, G. McMullan, S. H. W. Scheres, How cryo-EM is revolutionizing structural biology. Trends Biochem. Sci. 40, (2015), p. 49�57. [2] C. M. Spahn, P. A. Penczek, Exploring conformational modes of macromolecular assemblies by multiparticle cryo-EM. Curr. Opin. Struct. Biol. 19 (2009), p. 623�631. [3] S. H. Scheres, A Bayesian View on Cryo-EM Structure Determination. J. Mol. Biol. 415, (2012), p. 406�418. [4] D. Lyumkis, A. F. Brilot, D. L. Theobald, N. Grigorieff, Likelihood-based classification of cryo-EM images using FREALIGN. J. Struct. Biol. 183, (2013), p. 377�388. Figure 1. Analysis of Experimental RQC data using Frealign. (A) Reconstruction of the complete RQC dataset, with the 60S ribosomal subunit colored by local resolution. The dotted regions refers to panels D-G. (B,C) Density shots of ribosomal components at 3 A resolution, including (B) an alphahelical region with clearly resolved side-chains, and (C) an RNA helix with clearly resolved base pairs. (D) Reconstruction of the data without classification within a mobile ribosomal RNA segment. The same reconstruction, but at a lower threshold, is shown in red to indicate the extent of mobility within the region. (E) A small spherical mask used for classification is placed around the region corresponding to the mobile segment. (F) Four resulting classes from focused classification, showing different levels of mobility. (G) The best 3-D class with clear helical density for the mobile RNA segment.");sQ1[601]=new Array("../7337/1203.pdf","Ultrafast Transmission Electron Microscopy with nanoscale Photoemitters","","1203 doi:10.1017/S1431927615006807 Paper No. 0601 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Ultrafast Transmission Electron Microscopy with nanoscale Photoemitters Armin Feist, Katharina E. Echternkamp, Jakob Schauss, Sergey V. Yalunin, Sascha Sch�fer, Claus Ropers 4th Physical Institute- Solids and Nanostructures, University of G�ttingen, 37077 G�ttingen, Germany Ultrafast transmission electron microscopy (UTEM) is a rapidly developing technique which aims to elucidate laser-triggered ultrafast processes at the nanometer scale [1]. In UTEM, ultrafast imaging and spectroscopy with sub-picosecond temporal resolution is achieved by utilizing electron pulses in a stroboscopic laser-pump/electron-probe scheme (Fig 1a). The previously employed flat-surface photocathodes put a severe limit on the achievable spatial resolution due to their intrinsically large electron source size, demanding novel pulsed electron source concepts. Here, we report on the development and first applications of a novel UTEM instrument at the University of G�ttingen which incorporates a tip-shaped photocathode for the generation of high-quality electron pulses [2-4]. Specifically, in a modified JEOL JEM-2100F microscope, we obtain localized singlephoton photoelectron emission from the apex of a ZrO-coated tungsten tip using 3.1 eV optical pulses. Using the local depression of the work function due to the Schottky effect, photoelectron emission is largely confined to the front facet of the crystalline tip. At the sample position, the laser-triggered electron pulses can be focused down to 3 nm spot sizes (Fig.1b, inset), which results in a root-meansquare (rms) emittance of less than 4 nm mrad. The temporal structure of the electron pulses is probed by electron-photon cross-correlation, utilizing inelastic electron scattering in laser-driven optical near-fields [5] around a gold nanotip. An electron traversing an optical near-field experiences an energy gain or loss in multiples of the photon energy (Fig. 1b,d), giving rise to multiple photon sidebands in the transmitted electron energy spectrum. Varying the temporal delay between the optical driving pulse and the arrival time of the electron bunch at the near-field, allows to map the temporal bunch shape. For, on average, less than one electron per pulse, electron-photon cross-correlation results in a temporal width of the electron pulses as small as 300 fs (Fig. 1c, upper panel). At larger bunch charges, space-charge induced spectral and temporal broadening sets in, leading to a strongly correlated chirped electron pulse structure (lower panel, Fig. 1c). Low-emittance ultrashort electron pulses, as reported here, are particularly suited for time-resolved imaging applications which require, either, large transverse coherence lengths, as for example in Lorentz microscopy and electron holography, or small probe sizes. In a first application of nanoscale probing, we investigated quantum interference effects in the inelastic electron near-field scattering [6]. In particular, the population of certain photon sidebands is almost completely suppressed for specific driving field strengths. We quantitatively describe these scattering spectra taking into account quantum path interferences between different photon absorption/emission sequences which lead to the same final state. In conclusion, we report on the development of a novel ultrafast transmission electron microscope employing a tip-shaped photo-emitter. Implications of coherent near-field scattering processes for the optical control of free electron states are discussed. Microsc. Microanal. 21 (Suppl 3), 2015 1204 1198 Figure 1. a) Laser-pump/electron-probe scheme. Ultrashort electron pulses are generated by photoemission and probe at well-defined temporal delays the state of a sample after laser excitation. b) Schematics of inelastic electron scattering in the laser-driven near-field of a nanostructure. Inset: achievable electron spot diameter employing laser-triggered Schottky field emission. c) Electron energy spectra after inelastic near-field scattering for single-electron (top panel) and multi-electron pulses (bottom panel). Photon-order sidebands are only observed for temporally overlapping photon and electron pulses at the optical near-field around a gold nanotip, giving rise to the electron-photon crosscorrelation traces shown on the right . (d-e) Raster-scan of the intensity of gain-scattered electrons imaging the mode structure of the laser-driven near-field. The driving laser field is polarized either in the tip direction (d) and perpendicular to the tip direction (e). Broken red line: geometrical contour of the nanotip. References: [1] A.H. Zewail, Science 328 (2010), 187. [2] C. Ropers et al., Phys. Rev. Lett. 98 (2007), 043907. [3] P. Hommelhoff, C. Kealhofer, M. Kasevich, Phys. Rev. Lett. 97 (2006), 247402. [4] M. Gulde, et al., Science 345 (2014), 200. [5] B. Barwick, D. J. Flannigan, A. H. Zewail, Nature 462 (2009), 902. [6] A. Feist et al., submitted. [7] We gratefully acknowledge funding by the "Deutsche Forschungsgemeinsschaft" (DFG) through CRC 1073 project A5.");sQ1[602]=new Array("../7337/1205.pdf","Observing Liquid Flow in Nanotubes","","1205 doi:10.1017/S1431927615006819 Paper No. 0602 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Observing Liquid Flow in Nanotubes Ulrich J. Lorenz1 and Ahmed H. Zewail1 1. Physical Biology Center for Ultrafast Science and Technology, Arthur Amos Noyes Laboratory of Chemical Physics, California Institute of Technology, Pasadena, CA 91125 Advances in microfabrication have led to the emergence of the field of nanofluidics, which studies fluid transport in nanometer-scale structures [1-2]. Nanoscale confinement may considerably modify the properties and dynamics of a liquid. For example, the flow of water through carbon nanotubes has been reported to be enhanced by several orders of magnitude [3-4]. However, the magnitude of the enhancement has remained a point of contention [5]. A particular obstacle to settling this controversy is the great challenge of studying flow in a single nanotube [6]. Here, we demonstrate how this can be achieved by using 4D Electron Microscopy, which combines the spatial resolution of a Transmission Electron Microscope with the (ultra)fast time resolution of modern laser systems [7]. We directly image the flow of liquid lead through individual ZnO nanotubes to capture a range of flow phenomena and to characterize the nanoscale viscous friction involved [8]. Lead filled ZnO nanotubes [9] on a graphite thin film were studied at 363 K, well below the melting point of lead at 600.64 K. Experiments were performed with the UEM-1 instrument [7]. Briefly, a laser pulse was used to melt the lead core in situ, and the ensuing dynamics of the hot, pressurized liquid were imaged with a short electron burst, fired a short instant later. Figure 1A shows a micrograph of a typical nanotube after exposure to several laser pulses. On the left, lead has leaked through an imperfection in the tube wall and formed a spherical particle, while voids are visible in the remaining lead core on the right. It is evident that the laser fluence is sufficiently high to melt the lead core and to initiate dynamics that are driven by the expansion of the liquid. Among the many flow phenomena that we observe is the process displayed in Fig. 1B-F. When the continuously filled tube in (B) is irradiated with a laser pulse, part of the molten lead column is forced through a leak in the tube wall. A single-shot image recorded at 96 ns (C) shows that a spherical extrusion has formed. After the tube has cooled, the extrusion is slowly reabsorbed on the timescale of several minutes [(D) to (F)]. This suggests that the lead atoms on the surface of the extrusion and in the channel connecting it to the inside of the tube are sufficiently mobile to allow this slow diffusion process to occur. For the partially filled nanotube in Fig. 2A (55 nm inner diameter), laser-heating simply triggers the expansion of the lead column. The complete expansion dynamics, obtained from over 100 individual single-shot experiments, are shown Fig. 2B. During the first 30 ns, the meniscus advances rapidly with a speed of about 4 m/s and then recedes more slowly within another 300 ns, which agrees well with the typical cooling timescales that we measure for similar nanotubes. To analyze the expansion dynamics, we developed a simple model, using an approach similar to Washburn's law for the dynamics of capillary filling [10], but assuming that the expansion of the lead column drives the dynamics, not the capillary force. From a fit of the initial expansion with an expression thus obtained (blue line and dots), we determine that the flow is enhanced by at least one order of magnitude. We believe that the approach developed here will find fruitful application in the study of a range of nanoscale flow phenomena and will greatly benefit the emerging field of nanofluidics [11]. Microsc. Microanal. 21 (Suppl 3), 2015 1206 References: [1] JCT Eijkel and A van den Berg, Microfluid. Nanofluid. 1 (2005) p. 249. [2] L Bocquet and E Charlaix, Chem. Soc. Rev. 39 (2010) p. 1073. [3] M Majumder et al, Nature 438 (2005) p. 44. [4] JK Holt et al, Science 312 (2006) p. 1034. [5] M Whitby and N Quirke in "Handbook of Nanophysics: Nanotubes and Nanowires", ed. KD Sattler (CRC Press, Taylor & Francis Group, Boca Raton, 2011) p. 11.1. [6] A Siria et al, Nature 494 (2013) p. 455. [7] DJ Flannigan and AH Zewail, Acc. Chem. Res. 45 (2012) p. 1828. [8] UJ Lorenz and AH Zewail, Science 344 (2014) p. 1496. [9] C-Y Wang, N-W Gong and L-J Chen, Adv. Mater. 20 (2008) p. 4789. [10] P-G de Gennes, F Brochard-Wyart and D Quere, Capillarity and Wetting Phenomena: Drops, Bubbles, Pearls, Waves (Springer, New York, 2004). [11] Supported by NSF grant DMR-0964886 and Air Force Office of Scientific Research grant FA9550-11-1-0055 in the Physical Biology Center for Ultrafast Science and Technology at Caltech, which is supported by the Gordon and Betty Moore Foundation. UJL was partly supported by a postdoctoral fellowship from the Swiss National Science Foundation. A B C D E F Before 96 ns 96 ns 10 s 110 s 210 s Figure 1. A lead-filled ZnO nanotube and the temporal evolution of an extrusion. (A) Micrograph of a lead-filled ZnO nanotube after irradiation with several laser pulses. Scale bar, 100 nm. (B to F) When the filled nanotube in (B) is laser-heated, molten lead forces its way through a leak in the tube wall. At 96 ns, an extrusion is visible (C), which is slowly reabsorbed on a time scale of several minutes after the tube has cooled [(D) to (F)]. Scale bar, 200 nm. 150 B Expansion (nm) A 100 50 0 -50 Expansion Contraction Before 54 ns After 0 100 200 Time (ns) 300 400 Figure 2. Expansion dynamics of liquid lead in a single nanotube. (A) Single-shot images of a partially filled tube with an inner diameter of 55 nm. At a delay of 54 ns after the heating laser pulse, the lead column has expanded by 140 nm, before returning to its original length at long times. Scale bar, 100 nm. (B) The complete dynamics obtained from a series of such experiments. The gray lines have been inserted to guide the eye. From a fit of the initial expansion using a simple model of the dynamics that we developed in analogy to Washburn's law (blue line and dots), we estimate that viscous friction is reduced by at least one order of magnitude.");sQ1[603]=new Array("../7337/1207.pdf","Ultrabright Femtosecond Electron Sources: Ultrafast Structural Dynamics in Labile Organic Crystals","","1207 doi:10.1017/S1431927615006820 Paper No. 0603 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Ultrabright Femtosecond Electron Sources: Ultrafast Structural Dynamics in Labile Organic Crystals Germ�n Sciaini1,2,*,�, Meng Gao1,2, Cheng Lu1, Hubert Jean-Ruel1,2, Lai Chung Liu1,2, Alexander Marx2, Ken Onda3,4, Shin-ya Koshihara5,6, Yoshiaki Nakano7, Xiangfeng Shao7, Takaaki Hiramatsu8, Gunzi Saito8, Hideki Yamochi7, Ryan R. Cooney1,2, Gustavo Moriena1,2, and R. J. Dwayne Miller1,2,� Max Planck Institute for the Structure and Dynamics of Matter and, Hamburg Centre for Ultrafast Imaging, Luruper Chaussee 149, 22761 Hamburg, Germany. 2. Departments of Chemistry and Physics, 80 St. George Street, University of Toronto, Toronto, Ontario, M5S 3H6, Canada. 3. Interactive Research Center of Science, Tokyo Institute of Technology, Nagatsuta, Midori-ku, Yokohama 226-8502, Japan. 4. PRESTO, Japan Science and Technology Agency, Honcho, Kawaguchi 332-0012, Japan. 5. Department of Chemistry and Materials Science, Tokyo Institute of Technology, okayama, Meguroku, Tokyo 152-8551, Japan. 6. CREST, Japan Science and Technology Agency (JST), 5-3, Yonbancho, Chiyoda-ku, Tokyo 1028666, Japan. 7. Research Center for Low Temperature and Materials Sciences, Kyoto University, Sakyo-ku, Kyoto 606-8501, Japan. 8. Faculty of Agriculture, Meijo University, Shiogamaguchi 1-501 Tempaku-ku, Nagoya 468-8502, Japan. *. Present address: Department of Chemistry, 200 University Ave. W, University of Waterloo, Ontario, N2L 3G1, Canada. The progress in the development of fs-structural probes during the last twenty years has been tremendous. Current ultrafast structural techniques provide the temporal and spatial resolutions required for the stroboscopic observation of atoms in motion. In regards to femtosecond electron sources, different compression approaches and ultra-compact designs have made possible the generation of ultrashort and ultrabright electron pulses. With an effective brightness only one hundredfold below that of fs-hard X-ray Free Electron Lasers, ultrabright femtosecond electron sources have revealed unprecedented results in the study of photoinduced ultrafast structural dynamics [1, 2]. A brief overview of field along with a recent femtosecond electron diffraction (FED) study of the photoinduced insulatorto-metal phase transition of organic charge-transfer salt (EDO-TTF)2PF6 [3] will be presented. Here, we implemented a low repetition rate (10 Hz) and ultra-bright femtosecond electron source in order to avoid cumulative heating effects, and obtain a movie of the relevant molecular motions driving this photoinduced insulator-to-metal phase transition. We were able to record time-delayed diffraction patterns that allow us to identify time-dependent changes over hundreds of Bragg peaks (see figure 1). Model structural refinement calculations indicate the formation of a transient intermediate structure (TIS) in the early stage of charge delocalization (during the initial 2 ps). The molecular motions directing the formation of TIS were found to be distinct from those that, assisted by thermal relaxation, convert the system into a metallic-like state on the 100-ps timescale. In terms of design engineering, ultra-compact femtosecond electron diffraction setups still offer the most convenient and economic means to obtain 100-fs electron pulses with an areal electron density similar to that achievable by DC-RF compression schemes ( 5 e-/�m2) [4-6]. The most compact and simplest 1. Microsc. Microanal. 21 (Suppl 3), 2015 1208 1202 FED setup is comprised of a photocathode, an anode aperture and a cylindrical magnetic lens that is placed right after the sample [7]. N-body simulations show that the extracting DC electric field strength (E) plays a major role. There is a great reduction in the electron pulse duration when going from E = 10 MV/m to E = 20 MV/m. Reaching a DC field strength of 20 MV/m is challenging but possible [8], and under this condition such ultra-compact setup would enable sub-100 fs (FWHM) temporal resolution [9]. The presented findings illustrate the potential of ultrabright femtosecond electron sources for capturing, with atomic resolution, dynamical processes of relevance for the understanding of chemical reactions, phase transitions, and protein structure-function correlations. References: [1] G Sciaini and R J D Miller, Rep. Prog. Phys. 74, 096101 (2011). [2] R J D. Miller, Science 343, 1108 (2014). [3] M Gao et al., Nature 496, 343 (2013). [4] T Van Oudheusden et al., Phys. Rev. Lett. 105, 264801 (2010). [5] M Gao et al., Opt. Express 20, 12048 (2012). [6] R P Chatelain et al., Appl. Phys. Lett. 101, 081901 (2012). [7] C Gerbig et al., Res. Opt. Sci. IT3D.3 (Optical Society of America, 2012). [8] A Descoeudres et al., Phys. Rev. Spec. Top. - Accel. Beams 12, 032001 (2009). [9] Funding for this project was provided by the Natural Sciences and Engineering Research Council of Canada and the Canada Foundation for Innovation. This work was supported in part by a Grant-in-Aid for Scientific Research on Innovative Areas (grant number 20110006) and the Global Centre of Excellence (G-COE) programme for Chemistry from The Ministry of Education, Culture, Sports, Science and Technology in Japan and by Creative Scientific Research (grant number 18GS0208) from The Japan Society for the Promotion of Science. a" b" Figure 1. a) Top left panel: HT-LT denotes the difference between the diffraction pattern of hightemperature (HT) phase and that of the low-temperature (LT) phase. The rest of the panels show the difference between the diffraction patterns of the photoinduced and the initial LT phases as a function of the time delay between the optical excitation and electron probe pulses. b) Relative intensity changes for a few selected Bragg reflections. This figure was adapted from reference 3.");sQ1[604]=new Array("../7337/1209.pdf","Development of MeV Ultrafast Electron Scattering Instruments at SLAC National Accelerator Laboratory","","1209 doi:10.1017/S1431927615006832 Paper No. 0604 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Development of MeV Ultrafast Electron Scattering Instruments at SLAC National Accelerator Laboratory R. K. Li1, A. H. Reid1, S. P. Weathersby1, G. Brown1, M. Centurion2, T. Chase1, R. Coffee1, J. Corbett1, J. C. Frisch1, M. Guehr1, N. Hartmann1, C. Hast1, L. V. Ho1, K. R. Jobe1, E. N. Jongewaard1, J. R. Lewandowski1, A. M. Lindenberg1, J. E. May1, D. McCormick1, X. Shen1, K. Sokolowski-Tinten3, T. Vecchione1, J. Wu1, J. Yang2, H. A. D�rr1 and X. J. Wang1 SLAC National Accelerator Laboratory, 2575 Sand Hill Rd, Menlo Park, California 94025 USA University of Nebraska-Lincoln, 855 N 16th Street, Lincoln, Nebraska 68588, USA 3. University of Duisburg-Essen, Lotharstra�e 1, 47048 Duisburg, Germany 2. 1. Recent years have witnessed rapid development of ultrafast electron and x-ray scattering instruments, aimed at enabling the studies of structural dynamics on the fundamental time and length scales [1, 2]. Ultrafast electron diffraction (UED) and microscopy (UEM) using a few mega-electron volts (MeV) electron beams is a promising new R&D area which has the potential to open many opportunities for groundbreaking science [3-7]. Electrons and x-rays are complementary tools to study material structures. While x-rays primarily interact with electrons in matter, electrons are sensitive to both electrons and nuclei. Compared with hard x-ray, a few MeV electrons have 105 times larger scattering cross sections, 103 times shorter wavelength, and 103 times less radiation damage per elastic scattering [8]. The strong interaction with matter makes electrons the ideal choice to capture information from nanoscale or smaller samples, particularly monolayer and gas phase samples. Electrons are also charged particles and can be easily focused by electromagnetic lenses to form high spatial resolution real-space images. MeV UED and UEM can achieve much higher spatio-temporal resolution compared to their keV counterpart at 100-300 keV energies. State-of-the-art MeV electron sources have one order of magnitude higher field gradient, thus the generated electron beams have much higher brightness. Moreover, space charge effects, the essential limit to the machine performance, between the source and sample, are dramatically suppressed by 102-103 times following the 23 scaling, where is the Lorentz factor and is the normalized speed. To exploit the synergy and complementarity with the Linac Coherent Light Sources (LCLS), SLAC National Accelerator Laboratory recently launched a UED/UEM Initiative aiming at developing the world's leading ultrafast electron scattering instruments. A MeV UED setup has recently been built, commissioned, and brought into full operation at SLAC's Accelerator Structure Test Area (ASTA). The platform, as shown in Fig. 1, consists of a complete photocathode RF gun system, including a LCLStype S-band photocathode RF gun, a 5 mJ femtosecond Ti:Sapphire laser, the laser-RF synchronization system, a high power modulator and an S-band klystron. This system tremendously benefits from many common key technologies developed for LCLS. The setup is now serving an active ultrafast science program and supports R&D on future UED/UEM instruments and accelerators. The UED machine has already generated exciting science results: The machine provided very high SNR and stable intensity in the diffraction patterns, which enabled us to study the lattice dynamics and diffuse scattering of Bismuth polycrystalline samples at high precision. It sparked more interest in this Microsc. Microanal. 21 (Suppl 3), 2015 1210 extensively studied, but yet not fully understood system. As another example, we've observed ultrafast lattice dynamics in FePt nano-cluster thin films following laser excitation. The samples consist of 5-10 nm size FePt grains and are currently being developed for the next generation magnetic data storage media. The laser induced ultrafast electron and spin dynamics is currently being actively investigated using soft x-rays at the LCLS. While x-ray holography at LCLS will enable direct imaging of magnetic switching, the current UED and future UEM instruments open up complementary studies of the currently inaccessible lattice dynamics on comparable nanometer length scales. The next milestone of machine develop will be performing gas phase MeV UED experiment. MeV electrons travel at a speed very close to that of the pump laser, thus the velocity mismatch issue, which is a major constrain for keV UED, is essentially eliminated with MeV beams. For example, for 100 keV electrons, to keep the velocity mismatch below 100 fs the gas target needs to be 40 um or thinner, while a few mm thick gas sample can be used for 2.5 MeV electrons. We will also upgrade the machine repetition rate from 120 Hz to 360 Hz to minimize the data accumulation time. There are several other exciting targets in the Roadmap of UED/UEM development: (1) to improve the temporal resolution from 100 fs to <20 fs level to study fast chemical processes; (2) to improve the probe size in diffraction mode from a few hundred �m to sub-�m, which greatly eases the sample fabrication processes; (3) to develop single-shot UEM revealing real-space images with ps-nm spatiotemporal resolution. References: [1] M. Chergui and A. H. Zewail, ChemPhysChem 10, 28 (2009). [2] R. J. D. Miller, Science 343, 1108 (2014). [3] X. J. Wang, Z. Wu, and H. Ihee, in Proceedings of PAC03 (IEEE, Portland, OR, 2003), WOAC003. [4] W. E. King et al., J. Appl. Phys. 97, 111101 (2005). [5] P. Musumeci and R. K. Li, in ICFA BD Newsletter No. 59, ed by J. M. Byrd and W. Chou (2012). [6] http://science.energy.gov/~/media/bes/pdf/reports/files/Future_of_Electron_Scattering.pdf [7] H. D�rr and X. J. Wang eds. SLAC-R-1039 (2014). [8] R. Henderson, Q. Rev. Biophys. 28, 171 (1995). Figure 1. (a) The MeV UED beamline at SLAC's ASTA. Diffraction patterns of two recently studied system: (b) Bismuth and (c) FePt.");sQ1[605]=new Array("../7337/1211.pdf","Inconsistent Normalized Intensities for Quantitative STEM: Detector Scans and Single Electron Counting","","1211 doi:10.1017/S1431927615006844 Paper No. 0605 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Inconsistent Normalized Intensities for Quantitative STEM: Detector Scans and Single Electron Counting Xiahan Sang1, James M. LeBeau1 1. Department of Materials Science & Engineering, North Carolina State University, Raleigh, NC 27606 STEM images placed on an absolute scale relative to the incident beam intensity have helped solve numerous materials science problems. Two normalization methods have been proposed: the detector scan (DS) method [1], which measures the incident beam intensity directly on the STEM detector, and the single electron count (SEC) method that uses the signal for a single electron and separately measured beam current to determine absolute intensities [2]. In this talk, we will demonstrate that a drastic difference in normalization can occur between the two methods as shown in Figure 1a. For a fair comparison, two major error sources need to be taken into consideration. First, the HAADF detector experiences extremely low beam current (< 0.1 e-/s) in the SEC method compared with the DS approach (> 10 e-/s). As a result, detector non-linearity can play an important role. Second, the detector non-uniformity, which is known to cause 10%-20% error, must be taken into account for both methods. We explore the cause of the DS/SEC difference by imaging the HAADF with very low beam current (0.073 e-/s, measured using a calibrated CCD). For short pixel dwell times (Figure 2a, 16 s), the detector response yields discrete events, with only one or two electrons hitting the detector. Because the FEI hardware/software averages the detector signal, the single electron intensity is inversely related to the dwell time. As an example, at 16 �s the single electron intensity corresponds to 755 arb. units as measured by the difference between the first peak and the background. At intermediate dwell times (Figure 2b), three discrete peaks are observed sitting on a broad peak that corresponds to pseudocontinuous illumination during the well probe dwell time. For very large dwell time (Figure 2c, 256 s), only the pseudo-continuous broad peak is observed and represents the incident beam intensity, or 628 arb. units. As the influence of detector non-linearity and non-uniformity are incorporated, the results of the SE detector maps can be directly compared with the DS method. From the SEC approach, the apparent beam current is 0.052 e-/s, which is the ratio of the beam intensity (628) from 256 s detector image and product of the single electron intensity (47.2, scaled from 16 �s) and dwell time 256 �s. Compared with the CCD measured beam current (0.073 e-/s), a correction factor of 1.4 is required to account for the intensity discrepancy. Thus, under these imaging conditions, the apparent single electron signal does not capture the true intensity of a single electron. After accounting for the mismatch, for example, we achieve significantly better agreement between SEC and DS methods (red vs. black in Figure 1b). Finally, we will discuss other sources of error that contribute to the correction factor, including the detector electronics and control software that can further improve the method agreement [3]. Microsc. Microanal. 21 (Suppl 3), 2015 1212 References: [1] J. M. LeBeau and S. Stemmer, Ultramicroscopy 108 (2008), p. 1653. [2] R. Ishikawa et al., Microscopy and Microanalysis 20 (2014), p. 176. [3] The authors acknowledge the use and support of the Analytical Instrumentation Facility at North Carolina State University, which is supported by the State of North Carolina and the National Science Foundation. Figure 1. (a) Normalized intensity of a STEM image using two different methods: detector scanning and single electron count. (b) Line profiles extracted from the translucent white lines from each map. Figure 2. Detector mapping and corresponding histogram with different dwell times: (a) 16 s, (b) 64 s, (c) 256 s. The red arrows correspond to discrete electron counts while the black arrows correspond to the incident beam intensity.");sQ1[606]=new Array("../7337/1213.pdf","Quantitative Annular Dark-Field Imaging of Single-Layer Graphene","","1213 doi:10.1017/S1431927615006856 Paper No. 0606 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantitative Annular Dark-Field Imaging of Single-Layer Graphene Shunsuke Yamashita1,2, Shogo Koshiya1, Kazuo Ishizuka1,3 and Koji Kimoto1,2 1. 2. National Institute for Materials Science, 1-1 Namiki, Tsukuba, Ibaraki, Japan Department of Applied Chemistry, Kyushu University, 1-1 Namiki, Tsukuba, Ibaraki, Japan 3. HREM Research Inc., 14-48 Matsukazedai, Higashimatsuyama, Saitama, Japan Quantitative annular dark-filed (ADF) imaging in scanning transmission electron microscopy (STEM) has received much attention because it enables us to identify even the type and number of atoms of local structures. A quantification method of ADF imaging was proposed by LeBeau and Stemmer, in which the signal at each pixel is placed on an absolute scale by normalizing the current reaching an ADF detector by the incident probe current [1]. The method realized direct comparison between experimental and simulated ADF images without any arbitrary scaling parameters. They also reported nonlinear responses of the signal detection system of the microscope, thus the proposed method has a limitation that it can be used only on conditions that the signal detection system shows linear responses. In this study, we evaluated response properties of an ADF signal detection system and established a quantification procedure applicable to wider experimental conditions. Using the procedure, we acquired quantitative ADF images of single-layer graphene and compared with simulated images to investigate how accurately the scattering intensities match between experiments and simulations [2]. We used a Titan3 microscope (FEI) equipped with spherical aberration correctors (DCOR and CETCOR, CEOS) operating at an acceleration voltage of 80 kV. An ADF detector (Model 3000, Fischione) and an analog-to-digital (A/D) converter (DigiScan II, Gatan) were used. The response properties of the ADF signal detection system were evaluated by irradiating the ADF detector directly with the incident probe and investigating relationship between an input signal (incident probe current I0 [pA]) and an output signal (ADF signal SADF [count]). The incident probe current I0 was measured in each experiment using a charge-coupled device (CCD) camera (UltraScan, Gatan), whose conversion efficiency was measured in advance. The response property depends on the contrast setting of the ADF detector, which corresponds to the voltage of a photo multiplier tube. The evaluation was conducted at various contrast settings between 50�100%. After evaluating the response properties, curve fitting was carried out to obtain a conversion function which converts SADF into current reaching the ADF detector, which is called ADF detector current IADF [pA]. In the present quantification procedure, a quantitative contrast QADF [%], i.e. IADF normalized by I0, was calculated from SADF using both the conversion function and I0. Acquired ADF images were converted into quantitative ADF images using the customized DigitalMicrograph (Gatan) scripts. Experimental conditions such as convergence semiangle of the probe and ADF detection angle range were precisely measured because they were important for quantitative comparison with simulated images. The STEM image simulation was performed using a multislice program (xHREM with STEM Extension, HREM). The simulation conditions were adjusted to the experimental conditions and quantitative contrast of the experimental ADF images were compared with that of simulated images. In STEM experiments, a commercial CVD graphene (TEM2000GL, ALLIANCE Biosystems) was used as a specimen. The response properties at higher contrast settings showed nonlinearities as previously reported [1]. Thus, we obtained the conversion function that adequately reproduces experimental results by using a curve fitting. The conversion function enables us to convert IADF from SADF acquired under any value of Microsc. Microanal. 21 (Suppl 3), 2015 1214 the contrast settings between 50 and 100% because fitting parameters in the conversion function were also curve fitted as a function of the contrast setting. Figure 1 shows (a) a quantitative ADF image of graphene with 1�4 layers, (b) a high-resolution quantitative ADF image of single-layer graphene and (c) line profiles of quantitative contrast, respectively. The mean quantitative contrast, which was measured by averaging the value in areas including more than one unit cell, at a single-layer region was 0.054%. The mean value of a simulated image was 0.053%, thus the mean quantitative contrast exhibited excellent agreement between experimental and simulated images. We have established a quantification procedure, in which the quantitative contrast is given as the ADF detector current normalized by the incident probe current. Since the quantification procedure fully implements the nonlinear responses for the first time, it allows us to acquire quantitative ADF images even under higher contrast settings, i.e. higher sensitivity conditions, which is indispensable to observation of nanomaterials. We applied the quantification procedure for observation of single-layer graphene and compared quantitative contrast between experiments and simulation. Consequently, it was revealed that the mean quantitative contrast of single-layer graphene showed good agreement. The quantitative ADF imaging could allow us to analyze atomic numbers of attached atoms on a graphene. References: [1] J M LeBeau and S Stemmer, Ultramicroscopy 108 (2008), p. 1653. [2] S Yamashita et al, Microscopy (2015) in press (doi: 10.1093/jmicro/dfu115). [3] This study was partly supported by the JST Research Acceleration Program and the Nano Platform Program of MEXT, Japan. The authors thank Dr. Nagai, Mr. Kurashima and Ms. Ohwada for support in the STEM experiments. Figure 1. (a) Quantitative ADF image of graphene with 1�4 layers. (b) High-resolution quantitative ADF image of single-layer graphene. (c) Line profiles of quantitative contrast along A-A' in (a) and BB' in (b), respectively.");sQ1[607]=new Array("../7337/1215.pdf","Improving the SNR of Atomic Resolution STEM EELS & EDX Mapping while Reducing Beam-damage by using Non-rigid Spectrum-image Averaging.","","1215 doi:10.1017/S1431927615006868 Paper No. 0607 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Improving the SNR of Atomic Resolution STEM EELS & EDX Mapping while Reducing Beam-damage by using Non-rigid Spectrum-image Averaging. Lewys Jones1, Richard Beanland2 Sergio Lozano-Perez1, Karim Baba-kishi3 and Peter D. Nellist1 1. 2. Department of Materials, University of Oxford, Oxford, UK Department of Physics, University of Warwick, Coventry, UK 3. Department of Applied Physics, Hong Kong Polytechnic University, Hong Kong The aberration corrected scanning transmission electron microscope (STEM) can now routinely image crystalline specimens at atomic resolution. Of the imaging modes available the relative ease of interpretation of the annular dark-field mode (ADF) makes this a popular choice, offering thickness / atomic-number type contrast. However, for composition studies either energy dispersive x-ray analysis (EDX) or electron energy-loss spectroscopy (EELS) are added to many STEMs. Unfortunately the achievable signal-noise ratio (SNR) of these signals significantly limits quantitative work. There are three common approaches pursued to increase the SNR of elemental maps; increasing the beam-current, increasing the dwell-time or installation of a higher collection efficiency detector or spectrometer. However, the first two approaches have significant drawbacks. Increasing beam current can rapidly cause beam damage [1] or accelerate carbon contamination, while scanning slower additionally introduces scanning distortions, stage-drift or focal-drift into the frame [2]. Further, it is not always possible to increase beam-current (such as in monochromated systems), while increased dwelltimes make recording larger fields of view prohibitively time consuming for the operator. In our new proposed method, instead of increasing either probe-current or dwell-time, signal is accumulated by recording multiple sequential spectral images. Using the simultaneously acquired ADF images, the data-cubes are then realigned and corrected for scanning-distortion using a custom non-rigid registration algorithm (available online as `Smart Align' [3]). To test our proposed multi-frame approach two data-sets were recorded (Figure 1). The sample used was Pb2ScTaO6 prepared for STEM imaging by ion beam milling. This material exhibits incomplete long-range ordering on the Sc:Ta sublattice. Imaging was carried out near an anti-phase boundary. Simultaneous hardware synced spectra were recorded using a Gatan Quantum GIF and Oxford Instruments X-max 80mm2 detector. A single 256x256 probe-position spectrum image was recorded with a dwell-time of 0.01 s/pix, and for comparison five separate 0.002s/pix spectrum images (same total electron dose). The ADF images from the multi-frame acquisition were first realigned (rigid registered), cropped, then non-rigid-registered. Next, identical scan-distortion corrections were applied to the spectral data before being averaged to yield the restored spectra. To produce maps the EELS data were de-noised using principle component analysis before edge extraction. For the EDX maps characteristic x-rays were integrated after subtracting a linear background. The ADF immediately shows the beam-damage and scanning-distortions which arise from using the slower scan speed. The fast-scanned, distortion-corrected, and averaged ADF image however shows improved SNR and straight lattice planes as expected. The location of the anti-phase boundary is now clearly resolved. The SNR of the Sc L-edge EELS maps show an improvement from 2.1 to 5.1 with the Sc positions relating directly with the ADF image. EDX maps are less clear but in the multi-frame average spectrum image the Pb lattice begins to become apparent. As sample damage is reduced so Microsc. Microanal. 21 (Suppl 3), 2015 1216 drastically, more frames could be recorded to improve SNR further if desired. This was not done here to allow a fair comparison at fixed total dose conditions Our multi-frame spectrum image approach has demonstrated both reduced beam-damage and improved SNR for a fixed electron dose. This new approach allows for more beam-sensitive samples to be mapped, SNR to be increased and/or field-of-view and sampling to be improved [4]. References: [1] R. Egerton, P. Li, and M. Malac, Micron 35 (2004), p. 399�409. [2] L. Jones and P. D. Nellist, Microscopy & Microanalysis. 19 (2013), p. 1050�1060. [3] Demo code available free of charge for academic / non-commercial use at www.lewysjones.com . [4] This research was supported by the European Union Grant Agreement 312483 - ESTEEM2. Figure 1. Conventional and proposed approaches to acquiring spectral image data. Example shows an anti-phase boundary region in a Pb2ScTaO6 perovskite. The left column shows a single scan with 256x256 pixels at 0.01s/pix. The right column shows the restored data from a set of 5 frames scanned at 0.002s/pix after non-rigid scandistortion compensation and averaging. Rows show (top) the simultaneous ADF images, (middle) the EELS Sc L-edge map, and (bottom) the EDX Pb L+M map. The slightly smaller field of view (225x214 pix) is due to cropping after drift realignment. Both data were recorded at 30M magnification.");sQ1[608]=new Array("../7337/1217.pdf","Correction of Linear and Nonlinear Raster Distortion from Orthogonal Image Pairs.","","1217 doi:10.1017/S143192761500687X Paper No. 0608 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Correction of Linear and Nonlinear Raster Distortion from Orthogonal Image Pairs. Christopher T. Nelson1, Andrew M. Minor1. 1. Department of Materials Science and Engineering, University of California, Berkeley and National Center for Electron Microscopy, Molecular Foundry, Lawrence Berkeley National Laboratory, Berkeley, CA USA 94720 Experimental measurements using rastered scan techniques like STEM are subject to significant spatial distortion from the accumulation of error in the probe position. These distortions lead to degradation in spatial analysis including any technique requiring multi-image alignment such as time series, tomography, or multi-image averaging. The case where this error is constant over the timeframe of the experiment has received much prior treatment [1]. However, data often contains significant non-constant positional error which requires nonlinear corrections. In this work we present a method for analyzing and compensating for both linear and non-linear error using pairs of rastered datasets rotated orthogonal to one another. It requires no a-priori assumptions of the sample features or symmetry. The efficacy of this technique is demonstrated using HAADF STEM images (Figure 1) although it is applicable to any rastered data. The proposed method of nonlinear error correction is predicated on the comparative correctness of information transfer along the fast-scan direction, each dataset using the "good" axis to correct the "bad" axis of the other. The nature of this error can be seen in the lattice strain analysis of the atomic scale STEM images in Figure 1. The images are taken of the same region of a perovskite film/substrate interface with the scanning axis rotated 90�. The lattice spacing in the direction of the slow scan axis exhibits large modulations with standard deviations ~1.7 times larger than the relatively uniform fast scan axis. This leads to the primary operating principle for this method: applying shifts to each scan line to best match the data between the two orthogonal images. The corrected image, shown at the bottom of Figure 1, appears as a best-case combination of the fast-scan axis from the two input images. This is further highlighted by comparing the minimal data spread in the histogram of the lattice-spacings of the film in the corrected image. Determination of the linear and non-linear corrections is done by subregion alignment of the two images. Difference () vectors are calculated between aligned reference points (Figure 2a) and this difference is fit as two components. The first is a linear correction which applies shear and scale operations to the two datasets bringing them both to a best-match intermediate image. This component appears as a plane in the images in Fig 2. The nonlinear correction is the local x or y offset applied along the entire axis of the image which minimizes , subject to appropriate smoothing according to the reference point density and positional uncertainty. This component appears in Fig 2 as vertical and horizontal stripes. It is readily observable in Figure 2b that these two components clearly describe the observed difference vectors for these images. Results for these images show the linear distortions contribute up to 2.37� and 3.83� of deviation in x and y, respectively. Nonlinear distortions contribute up to 1.35� and 2.00� in x and y. In this case the nonlinear distortions are of the same order as the linear components and the correction of both is necessary for any practical spatial analysis or image alignment. Microsc. Microanal. 21 (Suppl 3), 2015 1218 Figure 1. (a-b) Atomic scale HAADF STEM images from the same region with the scan axis rotated 90�. (c) Atomic scale HAADF STEM image corrected for linear and nonlinear distortion. (d-f) Corresponding lattice spacing maps in the [100] and [001] directions. (g-i) Corresponding histograms of the [100] and [001] lattice spacings. Figure 2. (a) Illustration of two linearly distorted orthogonal images such as from sample drift. A difference vector (x,y) is calculated between reference points. (b) Difference vector () maps in x and y as measured (left), the best fit linear component (middle), and the best fit nonlinear component (right). References: [1] X. Sang, and J.M. LeBeau, Ultramicroscopy 138 (2014), p. 28. [2] The authors gratefully acknowledge the Singapore-Berkeley Research Initiative for Sustainable Energy (SinBeRISE) and the Molecular Foundry, Lawrence Berkeley National Laboratory, which is supported by the U.S. Dept. of Energy under Contract # DE-AC02-05CH11231.");sQ1[609]=new Array("../7337/1219.pdf","Ptychographic Imaging in an Aberration Corrected STEM","","1219 doi:10.1017/S1431927615006881 Paper No. 0609 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Ptychographic Imaging in an Aberration Corrected STEM Andrew R. Lupini1,3, Miaofang Chi2, Sergei V. Kalinin2,3, Albina Y. Borisevich1,3, Juan Carlos Idrobo2, Stephen Jesse2,3. 1 2 Materials Science & Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831 Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, TN 37831 3 Institute for Functional Imaging of Materials, Oak Ridge National Laboratory, Oak Ridge, TN 37831 One of the traditional limitations of electron microscopy is that although the phase and the amplitude of the transmitted electron beam both contain information about the sample, the detectors are only able to record the intensity. Furthermore most Scanning Transmission Electron Microscope (STEM) imaging modes rely on integrating the transmitted beam over some range of scattering angles. For example, the high-angle annular dark field (HAADF) mode provides `Z-contrast' images because the amount of scattering to high angles depends on the atomic number. The STEM is also able to record bright field (BF) images which, through the principle of reciprocity, are equivalent to conventional TEM images. However, in order to obtain a sufficiently coherent phase-contrast image, the collection angle for BFSTEM has to be much smaller than the convergence angle, meaning that BF-STEM is expected to be much less efficient than TEM. One solution to this predicament is to use a pixelated detector to record the scattering to every angle as a function of probe position. The resulting four dimensional (4D) data set consisting of a 2D image or `Ronchigram' for each probe position in a 2D array provides equivalent information to recording the set of TEM images as a function of tilt angle and can be used to reconstruct the complex exit wavefunction [1]. The experimental configuration is illustrated in Figure 1. Other imaging modes are available in STEM, including differential phase contrast (DPC) which has been shown to allow imaging of polarization at atomic resolution [2] and annular bright field (ABF) [3] which is particularly useful for locating light atoms. These images can also be recorded on a pixelated detector, with the advantage that the particular range of scattering to integrate over can be chosen after the experiment in post-processing and several images can be formed simultaneously (Figure 2). Collecting all of the transmitted beam also allows highly efficient phase contrast images to be recorded and to apply filters that would not be possible with a single detector [4]. We have recorded data sets on both conventional CCDs and on a direct electron detector able to detect single electrons [5] attached to an aberration corrected FEI Titan operating at 300 kV. A set of example images generated from such a data set are shown in Figure 2. Images were collected on a DE-12 camera (Direct Electron, LP, San Diego, CA), equipped with a 4096 x 3072 pixels Direct Detection Device (DDD�) sensor. Each exposure was acquired with continuous frame streaming at 768x768 pixels binned x2 at 300 frames per second. We have developed a custom beam control system to synchronize the STEM scanning to the Ronchigram acquisition. This system is able to synchronize to a variety of triggers at the hardware level, to scan unconventional scan patterns, and to scan at a variable speed within a single image [6]. Another common use for the electron Ronchigram is to accurately measure aberrations in order to allow efficient alignment of an aberration corrector. These methods usually need an amorphous region or special tuning sample, even if the sample of interest is crystalline. However, the 4D data set also contains equivalent information and routes to extract aberrations and other imaging data will be discussed. [7] Microsc. Microanal. 21 (Suppl 3), 2015 1220 References: [1] J. M. Rodenburg, B. C. McCallum and P. D. Nellist, Ultramicroscopy, 48 (1993), p. 304. [2] N. Shibata et al, Nature Physics, 8 (2012), p. 611. [3] R. Ishikawa, et al, Nature Materials 10 (2011), p. 278. [4] T.J. Pennycook, et al, Ultramicroscopy (2014), in press. [5] A.-C. Milazzo, et al. Ultramicroscopy 104 (2005), p. 152. [6] S. Jesse, et al, these proceedings. [7] Research supported by Division of Materials Sciences and Engineering Division, Office of Basic Energy Sciences, U.S. DOE (ARL, AYB) and by Oak Ridge National Laboratory's Center for Nanophase Materials Sciences (CNMS), which is sponsored by the Scientific User Facilities Division, Office of Basic Energy Sciences, U.S. Department of Energy (MC, SVK, JCI, SJ). The authors acknowledge technical assistance from Liang Jin and Benjamin Bammes of Direct Electron, and Hans Christen and Christianne Beekman for providing samples. (A) Pixelated Detector (B) HAADF ` Sample Scanned Probe Electrons (C) Aperture Figure 1. (A) Schematic illustration of the data acquisition scheme. The pixelated detector records the transmitted intensity as a 2D image for each probe position on the sample. (B) Typical Ronchigrams from a La0.67Sr0.33MnO3/SrTiO3 sample. Scale bar is 30 mrad. (C) Example detector configurations (same scale). Figure 2. Integrating the Ronchigram data set over a range of angles is used to form multiple images of the sample from Figure 1(B) using the detector configurations in (C): From left to right, a low-angle dark field image, a BF image with a large detector, an annular BF image, and a BF image with a small detector. Scale bar is 2 nm.");sQ1[610]=new Array("../7337/1221.pdf","Atomic resolution ptychographic phase contrast imaging of polar-ordered structures in functional oxides","","1221 doi:10.1017/S1431927615006893 Paper No. 0610 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic resolution ptychographic phase contrast imaging of polar-ordered structures in functional oxides Ian MacLaren1, Hao Yang2, Lewys Jones2, Peter D. Nellist2, Henning Ryll 3, Martin Simson 4, Heike Soltau 4, Yukihito Kondo5, Ryusuke Sagawa5, Hiroyuki Banba5 1. 2. SUPA School of Physics and Astronomy, University of Glasgow, Glasgow G12 8QQ, UK University of Oxford, Department of Materials, Parks Rd, Oxford, UK 3. PNSensor GmbH, Otto-Hahn-Ring 6, 81739 M�nchen, Germany 4. PNDetector GmbH, Sckellstra�e 3, 81667 M�nchen, Germany 5. JEOL Ltd., 3-1-2 Musashino, Akishima, Tokyo 196-8558, Japan Scanning transmission electron microscopy (STEM) imaging has traditionally relied on (mostly) incoherent imaging using electrons scattered to high angles [1,2]. This gives strong contrast for imaging heavier atoms, but is only an incomplete technique for in ferroelectrics or polar-ordered functional oxides, since the oxygens are invisible in the presence of the dominant scattering from the heavy cations. In recent years, however, there has been a resurgence in the use of low angle scattered electrons using modes like annular bright field imaging [3] and standard bright field imaging [4,5]. This allows the direct imaging of oxygen atoms, which can be used for a full quantification of the local polarization [5], as has previously been performed using negative CS imaging using high resolution TEM [6]. In principle, however, it is possible to reconstruct the phase at atomic resolution using ptychographic methods from suitable diffraction patterns with overlapping diffraction discs recorded at each scan point in a STEM image [7]. However, to do so practically requires the rapid recording of diffraction data, both to allow the acquisition of images with reasonable numbers of pixels in a sensible time scale, as well as to minimize the distortion images in real-space due to sample-drift. We show that this has now become possible by exploiting a direct electron detector running at 1000 frames per second (fps) or greater, which is a massive improvement over earlier CCDs running at < 30 fps. This was applied to an unusual antiphase boundary in Nd,Ti codoped BiFeO3, which has been wellcharacterised previously by conventional atomic resolution STEM [8,9] and HRTEM techniques. The atomic structure and chemistry of this boundary is consequently well-understood making this an ideal test object for atomic resolution ptychography. Atomic resolution datasets were recorded using a JEOL ARM200F, equipped with a cold FEG source and fitted with a PNDetector pnCCD (S)TEM camera running at up to 2000 fps. Reconstructions were performed using algorithms published previously [7]. The resulting phase image is shown in Figure 1 (top) and is compared to the HAADF image acquired simultaneously. The phase image shows a contrast similar to that previously seen in negative CS images of the same structure and can be assigned to atoms by comparison with the HAADF image. The strong red peaks in the phase image are coincident with the green peaks in the HAADF image and must be mixed B-site (Fe/Ti) / O columns. The broad yellow/orange peaks in the phase image are coincident with the bright red peaks in the HAADF image and must be A-site (Bi/Nd) columns. Finally, the weaker yellow peaks in the phase image are not seen in the HAADF image, but sit in positions which would fit expectations for pure O columns. It will clearly be possible to determine the atom positions quantitatively from such images and therefore perform atomic resolution polarization calculations, which will be reported at the conference and compared with the results calculated from atomic Microsc. Microanal. 21 (Suppl 3), 2015 1222 resolution BF/HAADF STEM and negative CS TEM imaging. It is nevertheless, clearly demonstrated that atomic resolution ptychography can be added to the roster of techniques for quantitative materials characterization at the atomic scale, provided suitable detectors are used. References: [1] AV Crewe, J Wall and LM Welter: J. Appl. Phys. 39 (1968) p. 5861. [2] P Hartel, H Rose and C Dinges: Ultramicroscopy 63 (1996) pp. 93�114. [3] M Hammel and H Rose: Ultramicroscopy 58 (1995) pp. 403�415. [4] JM LeBeau et al., Phys. Rev. B, 80 (2009) 174106. [5] I MacLaren, et al., APL Mater. 1 (2013) 021102. [6] CL Jia et al., Nature Mater., 7 (2008) pp. 57-61. [7] TJ Pennycook et al., Ultramicroscopy (2015) in press (doi:10.1016/j.ultramic.2014.09.013). [8] I MacLaren et al., APL Materials, 1 (2013) 021102. [9] I MacLaren et al., APL Materials, 2 (2014) 066106 [10] The authors gratefully acknowledge funding from the EPSRC under grant numbers EP/M009963/1 and EP/M010708/1. This work was supported by the EU FP7 Grant Agreement 312483 (ESTEEM2). Figure 1. (top) Reconstructed phase image of the boundary. (bottom) HAADF image recorded simultaneously with the diffraction dataset, but independent from the calculation of the phase image. Atom column positions are overlaid (red � iron, purple � bismuth, blue � titanium, yellow � oxygen).");sQ1[611]=new Array("../7337/1223.pdf","Practical Measurement of X-ray Detection Performance of Large-Angle Silicon Drift Detectors Toward Quantitative Analysis in the Newly Developed 300 kV Aberration-Corrected Grand ARM","","1223 doi:10.1017/S143192761500690X Paper No. 0611 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Practical Measurement of X-ray Detection Performance of Large-Angle Silicon Drift Detectors Toward Quantitative Analysis in the Newly Developed 300 kV Aberration-Corrected Grand ARM M. Watanabe*, T. Sasaki**, Y. Jimbo**, E. Okunishi**, H. Sawada** * Dept of Materials Science and Engineering, Lehigh University, Bethlehem. PA 18015. ** JEOL Ltd., 3-1-2 Musashino, Akishima, Tokyo, 196-8558, Japan. X-ray analysis in (scanning) transmission electron microscopes (S/TEMs) has been rather limited due to the miserable signal availability in comparison with electron energy-loss spectrometry (EELS). However, X-ray analysis is brought back into the attention because of relatively simple nature and robust aspects, especially after the aberration-corrected STEM instruments become available. Additionally, the recent silicon drift detector (SDD) technology offers more flexible detector-instrument configurations, so that the limited availability of X-ray signals can be improved significantly. The latest aberration-corrected 300-kV JEOL Grand ARM instrument has been developed to achieve the best image resolution and analytical spatial resolution. The Grand ARM is equipped with two large solid angle SDDs: SDD1 is at the traditional geometry (perpendicular to the specimen-holder rod axis) with the take-off angle of 25 degree and SDD2 is placed at the new configuration (along with the holder rod axis) with the take-off angle of 29 degree. The pole-piece is designed to maximize X-ray collection by positioning 2 SDDs as close as possible to the optical axis. In this study, X-ray collection performance of 2 SDD systems was systematically measured and atomic-resolution X-ray analysis was also performed. For measurement of X-ray detection performance, the NiOx thin film was used [1]. Figure 1 compares two X-ray spectra from the NiOx film measured by the two SDDs in the Grand ARM at 300 kV. The SDD2 with the new geometrical configuration shows better collection efficiency. By comparing X-ray spectra obtained from the previously calibrated HB 603 STEM, the collection angles of two detectors were determined. The conventional geometry SDD1 exhibits 0.593 sr whereas the new configuration SDD2 shows 0.994 sr. When two SDDs are in the Grand ARM, X-ray signals can collected at the solid angle of 0.5 sr, which is equivalent to 25% of the half sphere above the specimen. Using the NiOx film, the peak-to-background (P/B) ratio in Fiori definition and inverse hole count (IHC) of two SDDs were also measured in the Grand ARM and are compared with those in 300 kV HB603 and 200 kV aberration-corrected STEMs (Fig. 2). Although the P/B ratio of the both SDDs in the Grand ARM is comparable to that in other aberration-corrected instruments, the IHC values are superior to those in other instruments. Figure 3 summarizes the holder tilt-angle dependence of the P/B ratio measured in ARM. It should be noted that the positive x-tilt is toward the SDD1 and the positive y-tilt is towards the SDD2. The P/B ratios plotted against the x-tilt and y-tilt are shown in Fig. 3(a) and (b), respectively. Fig. 3(c) shows the P/B as a function of both x and y tilts, which means that the specimen is tilted between the both SDDs in positive direction. The SDD2 exhibits lower P/B values than the SDD1 but less sensitive to the specimen tilt angle. Although the P/B values in the SDD2 is slightly degraded, the SDD2 shows higher solid angle than the SDD1, which implies that this new geometry can improve X-ray collection efficiency. A set of X-ray maps obtained from SrTiO3 using the Grand ARM is shown in Fig. 4. The maps with 256x256 pixels were acquired only for 2.5 min in total with the probe current of 80 pA. Microsc. Microanal. 21 (Suppl 3), 2015 1224 These atomic-resolution X-ray maps can be acquired with the limited current for very short acquisition time, which can be beneficial for characterization of irradiation sensitive materials. References [1] R.F. Egerton & S.C. Cheng, Ultramicrosc. 55 (1994), 43. [2] The author (MW) wishes to acknowledge financial support from the NSF through grants DMR-0804528 and DMR-1040229. Figure 1: X-ray spectra from NiOx measured by two SDDs in Grand ARM. Figure 2: P/B plotted against IHC measured in different instruments, including Grand ARM. Figure 3: Tilt angle dependence of P/B measured in Grand ARM: (a) x-, (b) y- and (c) x/y-tilts. Figure 4: Atomic-resolution X-ray maps measured from SrTiO3 in Grand ARM.");sQ1[612]=new Array("../7337/1225.pdf","Characterizing Atomic Ordering of High Entropy Alloys Using Super-X EDS Characterization","","1225 doi:10.1017/S1431927615006911 Paper No. 0612 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterizing Atomic Ordering of High Entropy Alloys Using Super-X EDS Characterization Robert E. A. Williams1,2, Brian Welk1,2, Jake Jensen1,2, Bryan D. Esser1, David W. McComb1, Hamish L. Fraser1,2 1. Department of Materials Science and Engineering, The Ohio State University, 477 Watts Hall, 2041 College Road, Columbus, OH 43210, USA 2. Center for Accelerated Maturation of Materials, 1305 Kinnear Road, Columbus, OH 43212, USA High entropy alloys (HEA), more recently referred to as compositionally complex alloys (CCA), are a new group of alloys receiving a great deal of attention because of the potentially remarkable balance of properties they are expected to exhibit. They offer new pathways to lightweighting in structural applications, new alloys for intermediate and elevated temperature components, and new magnetic materials[1,2]. To realize their potential, however, requires considerable alloy development that will rely on application of integrated computational materials (science and) engineering (ICME), which requires accurate computational models predicting their performance in addition to a detailed knowledge of, for example, their deformation mechanisms. Often, these alloys consist of a mixture of ordered and disordered phases, and because of the compositional complexity, it is necessary to know the nature (i.e., degree of order, site occupancy, and presence of anti-site defects) in the ordered phases if effective models of the deformation behavior are to be developed[3]. These various metrics require accurate compositional measurements at the atomic scale. The work to be presented involved determining the degree of order present in a model B2 compound consisting of Al- 25 at. %Ni- 25 at. %Co. In the past, ALCHEMI has been used to indicate the trend in sub-lattice occupancy but has not been able to provide determinations at to the actual degree of order[4]. Such determinations would require a direct characterization of the actual sub-lattice compositions. Recent technological improvements in STEM probe aberration correction and large collection angle, silicon drift detectors (SDD) for XEDS collection have provided unprecedented quality in XEDS spectral images. The incorporation of a DCOR, for probe correction up to the 5th order, permits the use of larger probe forming apertures while reducing the probe diameter and the SDD have significantly improved collection efficiency over previous generation SiLi detectors. The benefits from this design are multifaceted in that the probe diameter is decreased and the probe current is increased, resulting in better spatial resolution and an increased analytical signal for detection by SDDs. As technological advancements have improved the quality of the data collected, the physics of XEDS quantification has not yet evolved. Quantification of apparent atomic resolution XEDS suffers from the same difficulties as traditional XEDS quantification, moreover collection of apparent atomic resolution XEDS is performed on zone, an undesirable condition that maximizes electron channeling. Aside from the normal difficulties associated with XEDS quantification, the apparent atomic resolution XEDS data is complicated by column "cross-talk" from delocalization and contributions from phonon scattering [47]. The results shown in Fig 1 (a, b) were collected using a FEI Titan ThemisTM 60-300 equipped with a DCOR probe corrector and Super-X EDS collection system with ~200 pA beam current and 25 �s dwell time/pixel. In Fig. 1 (a), the XEDS maps of the elemental distributions over the two sub-lattices of the Microsc. Microanal. 21 (Suppl 3), 2015 1226 B2 compound in the sample are shown, overlaying the HAADF image. Clearly, spatially-resolved spectral collection, at the atomic scale, has been achieved. The compositional profiles obtained are shown in Fig. 1 (b), from which it is clear that one sub-lattice is occupied mainly by Al atoms and the other sub-lattice consists mainly of Ni and Co. As mentioned previously, quantification of this data is not trivial and a variety of physics-based and gaussians-based methods have been attempted to accurately extract chemical compositions from the XEDS spectral data. These results and methods will be discussed as well as suggestions for improvement. References: [1] B. Cantor, et al., Materials Science and Engineering: A 375 (2004): p. 213-218. [2] J-W. Yeh, et al., Advanced Engineering Materials 6, no. 5 (2004): p. 299-303. [3] J-W. Yeh, et al., Metall Mater Trans A 35 (2010): p. 2533. [4] D. H. Hou, et al., Philosophical Magazine A, 74(3), (1996): p. 741-760. [5] B. D. Forbes, et al., Physical Review B 86(2), (2012): 024108. [6] N. R. Lugg, et al. Applied Physics Letters 101(18), (2012): 183112. [7] L. J. Allen, and S. D. Findlay. Ultramicroscopy in press (2014). [8] G. Kothleitner, et al., Physical Review Letters 112(8), (2014): 085501. Fig. 1. a) XEDS map (overlaying a HAADF image) of the atomic columns in the model B2 compound with the beam parallel to <100>. b) XEDS compositional profiles plotted across a set of columns showing that one of the sublattices consists mainly of Al, while the other mainly Ni and Co.");sQ1[613]=new Array("../7337/1227.pdf","Overcoming Traditional Challenges in Nano-scale X-ray Characterization Using Independent Component Analysis","","1227 doi:10.1017/S1431927615006923 Paper No. 0613 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Overcoming Traditional Challenges in Nano-scale X-ray Characterization Using Independent Component Analysis David Rossouw1, Pierre Burdet1, Francisco de la Pe�a1, Caterina Ducati1, Benjamin R. Knappett2, Andrew E. H. Wheatley2, Paul A. Midgley1. 1. Department of Materials Science and Metallurgy, University of Cambridge, 27 Charles Babbage Road, Cambridge, CB3 0FS, UK. 2. Department of Chemistry, University of Cambridge, Lensfield Road, Cambridge, CB2 1EW, UK. Nano-heterostructures are inherently challenging to characterize due to the presence of spatially and often spectrally overlapping signals when using energy dispersive X-ray (EDX) spectroscopy or electron energy loss spectroscopy (EELS) techniques. In addition, inherently low signal yields from such small volumes and electron beam damage often limits signal quality. New image and spectral processing routes are needed to address these issues [1]. Here, we use a machine learning technique, called independent component analysis (ICA) [2], to unmix EDX signals originating from core-shell nano-magnetic particles. The ICA method, executed in HyperSpy [3], blindly decomposes the mixed signals into three components, which are found to accurately represent the isolated X-ray signals originating from the bi-metallic core, shell and supporting film (Fig. 1). The composition of the core is verified by and is in excellent agreement with the separate quantification of bare bi-metallic seed nanoparticles from the same synthesis. This validation provides the crucial evidence needed to justify the use of the technique in the analysis of individual phases in heterogeneous nano-scale volumes, without the need for their mechanical separation or separate analysis at each step in a multi-step synthesis procedure. The machine learning approach also efficiently handles noisy data, and can differentiate true spectral and spatial information from underlying noise, minimising the required sample beam dose [4]. Microsc. Microanal. 21 (Suppl 3), 2015 1228 References: [1] G. Lucas, P. Burdet, M. Cantoni, and C. Hebert, Micron, 52 (2013), p. 49. [2] C. Jutten and J. Herault, Signal Processing 24 (1991), p. 1. [3] www.hyperspy.org [4] D.R. acknowledges support from the Royal Society's Newton International Fellowship scheme. B.R.K. thanks the UK EPSRC for financial support (EP/J500380/1). F.d.l.P. and C.D. acknowledge funding from the ERC under grant no.259619 PHOTO EM. P.A.M and P.B. acknowledges financial support from the European Research Council under the European Union's Seventh Framework Programme (FP7/2007-2013) / ERC grant agreement 291522-3DIMAGE. P.A.M. also acknowledges financial support from the European Union's Seventh Framework Programme of the European Commission: ESTEEM2, contract number 312483. Figure 1. (a) An EDX spectrum image obtained from a cluster of core-shell nanoparticles. (b) The nanoparticles are comprised of a bi-metallic Pt/Fe core surrounded by an iron oxide shell on a carbon support. (c) ICA decomposes the mixed EDX signals into components representing the core (IC#0), shell (IC#1) and support (IC#2).");sQ1[614]=new Array("../7337/1229.pdf","Gold and Arsenopyrite Exsolution and Limits of Arsenic Solubility in Pyrite Investigated by SEM, EPMA, and LA-ICPMS","","1229 doi:10.1017/S1431927615006935 Paper No. 0614 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Gold and Arsenopyrite Exsolution and Limits of Arsenic Solubility in Pyrite Investigated by SEM, EPMA, and LA-ICPMS Karsten Goemann1, Aleksandr S. Stepanov2, Sebastien Meffre2, and Ross R. Large2 1 2 Central Science Laboratory, University of Tasmania, Hobart, TAS 7000, Australia CODES, University of Tasmania, Hobart, TAS 7000, Australia The mineral pyrite (Py, FeS2) is an important carrier of gold in ore deposits, both as solid solution in the pyrite lattice ("invisible gold"), as well as in form of gold inclusions in Py ("free gold"). Form of Au and particle size have substantial impact on Au liberation and recovery from such ores. At least for Carlin-type and epithermal deposits, Au solubility in Py depends on the As content [1]: CAu = 0.02 CAs + 4 x 10-5 (mol%). Above this solubility limit, Au forms free nanoparticles as confirmed by high resolution transmission electron microscopy and secondary ion mass spectrometry [1]. A range of studies have been published investigating As solubility in Py under different conditions, see e.g. [2], however considerable disagreement remains about phase boundaries and stability. Pioneering experiments in the dry Fe-As-S system found As solubility in Py to be <0.5 wt% [3], however under hydrothermal conditions Py with up to 20 wt% As has been produced which was interpreted as metastable phase [4]. First principles calculations indicate that Py with up to 6 wt% As could coexist in equilibrium with arsenopyrite [5]. The aims of this study are two-fold: 1. Determine the smallest Au particle size we can detect with routine techniques used to characterize gold deportment in pyrite, namely mineral liberation analysis (MLA), a scanning electron microscopy (SEM) based technique for automated mineralogy, field emission (FE)SEM combined with energy dispersive x-ray spectrometry (EDS), and laser ablation inductively coupled plasma spectrometry (LA-ICPMS). 2. Investigate the formation of Au nanoparticles in relation to As contents in Py from the Konkera deposit in Burkina Faso. MLA was performed on a FEI Quanta 600 fitted with two EDAX Genesis Sapphire SUTW Si(Li) EDS at 25kV accelerating voltage. High resolution backscattered electron (BSE) imaging and x-ray mapping were conducted on a Hitachi SU-70 FESEM fitted with an Oxford AZtec XMax80 EDS system at 5 to 20kV. Major element compositions of Fe, S, and As were acquired using a Cameca SX100 electron probe microanalyzer (EPMA) at 15-20kV. Trace element compositions and distributions were studied using an Agilent 7500A quadrupole ICP-MS and a New Wave UP-213 nm Nd:YAG laser ablating in a He atmosphere in a custom, small-volume ablation cell. Pyrite from Konkera contains between <0.02 and 4 wt% As and hosts fine-grained arsenopyrite (Apy, FeAsS), chalcopyrite (Cpy, CuFeS2) and gold inclusions, see Fig. 1. The tungsten source MLA only found the largest Au grain shown in Fig. 1A, which is 1.3�1.1 �m2 in size. FESEM BSE imaging at 5 kV accelerating voltage revealed >30 Au grains in this area, 20�1100 nm in size. Many grains possess a complex, spongy texture (Fig. 1C). Outlines are blurry due to averaging in depth (Fig. 1D), limiting accuracy of dimensional measurements and estimation of mass and volume of the Au particles. Monte Carlo simulations [6] at 5kV for Au yield an interaction volume size of 60 nm which is why for smaller particles EDS identification using the AuM peak is impeded by the SK interference from the host Py. At 20 kV, subsurface Au particles become visible and the AuL peak is available, but spectra become even more dominated by the host phase due to the interaction volume increase. In LA-ICPMS, the beam diameter is too large for direct analysis of individual Au nanoparticles. However, they appear as spikes in the transient signal when mapping. Assuming the particles consist of 100% Au+Ag their size can be calculated, yielding similar Au grain sizes as measured by FESEM, with a lower size detection limit of around 150 nm. Microsc. Microanal. 21 (Suppl 3), 2015 1230 Au grains are often located on grain boundaries between Apy and low-As Py (Fig. 1C) or between higher-As and low-As Py. Quantitative EPMA mapping of 10 typical areas of these assemblages gives average As concentrations of 3.2�5.2 wt% (Fig. 2), which is similar to the highest As values of 3.7�3.9 wt% found in the host arsenian Py. This could indicate their formation by decomposition of arsenian Py, expelling As and correspondingly also Au above the new solubility limit to form Apy and native Au, respectively. The texture may indicate a dissolution-precipitation mechanism of Py recrystallization and limited role of solid state diffusion. Apy stoichiometry depends on the temperature. Above 600�C, it has the composition FeAs0.9S1.1, with more sulphur rich compositions becoming stable at lower temperatures [3]. The stoichiometry of the small Apy crystals in these assemblages is Fe1.02As0.86-0.91S1.06-1.11, which would correspond to a higher temperature formation and is richer in As compared to the composition of Fe1.02As0.78-0.86S1.11-1.19 found for larger arsenopyrite crystals in other parts of the sample. References [1] M Reich et al, Geochimica et Cosmochimica Acta 69 (2005), p. 2781. [2] AP Deditius et al, Geochimica et Cosmochimica Acta 140 (2014), p. 644. [3] LA Clark, Economic Geology 55 (1960), p. 1345. [4] ME Fleet and AH Mumin, American Mineralogist 82 (1997), p. 182. [5] M Reich and U Becker, Chemical Geology 225 (2006), p. 278. [6] Using the Casino Monte Carlo software, http://www.gel.usherbrooke.ca/casino/index.html Figure 1. FESEM BSE images at various magnifications and accelerating voltages. A: 7 kV, B: 20 kV, C&D: 5 kV. White arrows in B indicate Au nanoparticles. Figure 2. EPMA element maps and BSE image, field of view 29.5 x 22.1 �m. The table shows the average composition for the map determined by individually quantifying all pixels.");sQ1[615]=new Array("../7337/1231.pdf","Characterization of Stannous Fluoride Uptake in Human Dentine by Super-X XEDS and Dual-EELS analysis","","1231 doi:10.1017/S1431927615006947 Paper No. 0615 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of Stannous Fluoride Uptake in Human Dentine by Super-X XEDS and Dual-EELS analysis Isabel N. David1, Robert E.A. Williams1, Daniel Huber1, Jonathan S. Earl2, David W. McComb1 1 2 Center for Electron Microscopy and Analysis, The Ohio State University, Columbus, OH, USA GlaxoSmithKline Consumer Healthcare R&D, Weybridge, Surrey, England, UK Stannous fluoride (SnF2) is a common additive to dental products and has been shown to reduce the dental hyper-sensitivity in patients. In order to elucidate and better understand the permeability and mass transport mechanisms, analytical electron microscopy (AEM) characterization was performed on human dentin exposed to SnF2 [1]. In particular, techniques such as S/TEM-HAADF imaging, Super-X XEDS and dual-EELS have been used to investigate the ultrastructure and chemistry of the inner dentine tubule surface. Results on the characterization of the Sn-reacted product on the inner surface of dentine microtubules, as well as the role of dentine "nano-tubules" that branch off from the primary microtubule will be discussed. In an attempt to simulate in-vivo dentine exposure, samples of coronal dentine were treated with either deionized water (control) or a 1% w/w SnF2 solution (pH 5.5) in a 5-day cycling model incorporating a daily 2 minute exposure to the SnF2 treatment. The dentine was stored in artificial saliva for the remainder of the study. At the end of each day a specimen was removed and allowed to dry in air resulting in a control and 5 samples with exposure from 1�5 days. A dual-beam focused ion beam (FIB) methodology was developed to prepare consistently thin, cross-sectional specimens of dentin tubules for subsequent analysis. S/TEM-HAADF imaging and EELS/XEDS spectral imaging (SI) were used to determine directly the reacted product observed on the tubule surface following exposure to SnF2. Figure 1(a,b) illustrates the FIB trenching and thinning to produce a thin, cross-sectional dentin lamellae. FIB sample preparation techniques will be discussed. Following FIB sample preparation, a monochromated TitanTM 60-300 STEM equipped with a Super-X XEDS collection system was used to investigate the dentine tubules as shown in Figure 2(a). Characterization of multiple tubules revealed that they were unevenly covered by a coating on the inner tubule surface. Super-X XEDS spectral imaging of the various daily exposures revealed the coating was Sn-rich. The 1% w/w SnF2 solution treated samples, shown in Figure 2(b), produced a Sn signal on all inner tubule surfaces, including that of the revealed "nano-tubule". The Super-X XEDS signal was then used to locate Sn rich regions for subsequent dual-EELS analysis to determine Sn oxidation state when compared against SnO, SnO2 and SnF2 powder standards. The combination of FIB, Super-X XEDS SI, and dual-EELS has been applied successfully to characterize the reacted product formed from exposure of stannous fluoride to dentin with nanometer scale spatial resolution. Super-X XEDS spectral imaging showed clearly the location and presence of Sn and was used to identify regions for subsequent dual-EELS characterization. Dual-EELS was used to identify the oxidation state of Sn by comparing to Sn standards. This work has resulted in unparalleled characterization of dentine tubules and demonstrated that FIB, Super-X XEDS and dual-EELS are potent tools for characterizing salient nano-scale features in human dentine. Microsc. Microanal. 21 (Suppl 3), 2015 1232 References: [1] C.R. Parkinson, J.S. Earl, J. Clin. Dent. 20(5) (2009), p. 152-7. Figure 1. Secondary electron micrographs of 1% w/w SnF2 solution treated human dentine samples. (a) FIB preparation site with specific orientation (b) Thinned TEM sample Figure 2. Super EDS on 1% w/w SnF2 Solution treated human dentine after 4 days. (a) HAADF STEM image of dentine tubule, (b) EDS Sn Map. The same region is pictured in (a) and (b), and the red rectangle surrounding a "nano-tubule" with the presence of Sn.");sQ1[616]=new Array("../7337/1233.pdf","Three-Dimensional Imaging of Point Defects in Functional Materials Using Quantitative STEM","","1233 doi:10.1017/S1431927615006959 Paper No. 0616 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Three-Dimensional Imaging of Point Defects in Functional Materials Using Quantitative STEM Jinwoo Hwang Department of Materials Science and Engineering, The Ohio State University, Columbus, OH 43212 Many important properties of nanoscale functional and energy-related materials are determined by individual point defects, such as impurity dopant atoms and vacancies. Therefore, to understand the materials physics and exploit the properties of emerging materials, it is crucial to precisely determine the 3D location, distribution, segregation, pairing, and clustering of point defects. Electron microscopy has been previously used to identify individual point defects. For example, annular dark-field (ADF) imaging in STEM is sensitive to the atomic number of the material, and therefore can provide intuitive information on the lateral positions of the impurity atoms [1]. To complete the 3D information, however, the precise depth location (the direction parallel to the electron beam) of the point defects must be uncovered as well. However, acquiring depth information at the atomic scale is challenging. Even with high probe convergence angles, the depth resolution of STEM remains above ~10 nm [2], which makes it difficult to determine the 3D position of point defects using conventional approaches, such as focal series reconstruction or confocal microscopy. These conventional techniques also require the acquisition of multiple images of the same area of the sample, and therefore may suffer from low precision due to sample drift, or radiation damage due to long beam exposure. The development of a 3D electron microscopy technique that combines resolution, precision, and efficiency is therefore required. We present a new 3D atomic scale imaging technique based on quantitative STEM. Quantitative STEM measures the ADF signals on an absolute scale, which can be directly compared to simulated images and therefore allows for straightforward interpretation of the image in terms of the number of atoms and atomic species [3]. We show that by including the probe channeling information in the analysis of the atomic column intensities in quantitative STEM images, the 3D structural information of point defects can be determined with unparalleled depth precision [4]. Also, our technique only requires one single ADF image per sample area, and therefore it can achieve substantially higher efficiency in data acquisition and reduce radiation damage to the sample. For example, we determined the 3D positions of individual Gd dopant atoms and their nanoscale clustering in SrTiO3. Here, the quantified error (2D Gaussian function in the inset in Fig. 1A) sets practical limit to the dopant visibility and depth precision in the experiment (Fig. 1A). Possible sources contributing to the error include experimental instabilities, such as specimen drift and surface roughness of the sample. Multislice simulations show that the intensities of the atomic columns containing dopants monotonically depend on the number and the depth position of the dopants when the SrTiO3 thickness is < 4 nm (Fig. 1B). The deviation between the experimental and simulated data points in Fig. 1B is due to the same experimental error shown in Fig. 1A. Therefore we can calculate the distance between each experimental and simulated data points in Fig. 1B, and then weight it with the 2D error function to acquire the expectation value and uncertainty of the depth position of the dopants [4]. The depth uncertainty of the measurement was less than one SrTiO3 unit cell (u.c.) (Fig. 1D). We will also discuss recent results and challenges of the comprehensive 3D mapping of individual dopants that control the performance of the novel semiconductor devices. Identifying the 3D Microsc. Microanal. 21 (Suppl 3), 2015 1234 information of lighter dopants that have atomic numbers close to that of the host atom (e.g. Al or In doped GaN) poses additional challenges, as the intensity in ADF imaging primarily depends on the atomic number. Achieving higher depth resolution and precision will be critical for this work. We advance the 3D quantitative STEM technique in an FEI Titan STEM with a probe aberration corrector that ensures sub-Angstrom lateral resolution, and a high-brightness electron source that provides an excellent signal-to-noise ratio, and therefore allows for substantially higher precision and efficiency in 3D imaging. The new quantitative STEM experiments also utilize the multiple ADF detectors in our STEM. The atomic column image intensities containing depth information varies depending on the collection angle of the ADF detector. Therefore the acquisition of multiple ADF images using different collection angles can provide additional depth information, and lead to a maximized precision for pinpointing the depth of impurity atoms [e.g. 5]. We also expect our new 3D imaging technique to help identify the point defects in other important functional materials, including the dopant atoms in novel quantum devices and the vacancies at oxide interfaces that may govern the properties. References [1] P. M. Voyles, et al., Nature 416 (2002) p. 826. [2] P. M. Voyles, D. A. Muller, and E. J. Kirkland, Microsc. Microanal. 10 (2004) p. 291. [3] J. M. LeBeau, S. D. Findlay, L. J. Allen, and S. Stemmer, Phys. Rev. Lett. 100 (2008) p. 206101. [4] J. Hwang, et al., Phys. Rev. Lett. 111 (2013) p. 266101. [5] J. Y. Zhang, J. Hwang, B. J. Isaac, and S. Stemmer (submitted). Figure 1. (A) Experimental Sr column intensity (Isr) vs. mean Ti-O column intensity ( ITi-O ) plot for Gddoped SrTiO3, and fit for undoped SrTiO3 (solid line). The inset shows the 2D error function calculated from the undoped SrTiO3 data. The error cutoff is also shown as dashed lines. The data points above the upper limit of the error cutoff indicate the columns that contain at least one dopant. (B) Multislice simulations (blue circles) of column intensities for the blue rectangular region in Fig. 1A (thickness = 5 u.c.). The labels (numbers) indicate the dopant position in the column as defined in (D). Experimental points, labeled a-g are also shown. The large dashed triangle indicates the dopant clusters shown in (CE). (C) Isr map (top) and ADF image (bottom) of the area containing c-f columns. (D) A schematic and (E) 3D rendering of dopant clustering in c-e columns. The most probable dopant position is shown in red and the expectation values and uncertainties are labeled (unit = u.c.).");sQ1[617]=new Array("../7337/1235.pdf","Removing Elastic Scattering Effects from Chemical Maps Taken under Incoherent Conditions","","1235 doi:10.1017/S1431927615006960 Paper No. 0617 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Removing Elastic Scattering Effects from Chemical Maps Taken under Incoherent Conditions Y Zhu1, H Tan2, HL Xin3, A Soukiassian4, DG Schlom4, DA Muller5 and C Dwyer6 1. 2. Monash Centre for Electron Microscopy, Monash University, Victoria 3800, Australia CEMES-CNRS, Universite de Toulouse, nMat Group, BP94347, 31055, Toulouse Cedex 4, France 3. Center for Functional Nanomaterials, Brookhaven National Laboratory, Upton, New York 11973, USA 4. Department of Materials Science and Engineering, Cornell University, Ithaca, New York 14853, USA 5. School of Applied and Engineering Physics, Cornell University, Ithaca, New York 14853, USA 6. Ernst Ruska-Centre for Microscopy and Spectroscopy with Electrons, Peter Gruenberg Institute, Forschungszentrum Juelich, Juelich 52425, Germany Incoherent imaging conditions in scanning transmission electron microscopy (STEM) are generally highly favorable for image interpretation, since the image contrast tends to a slowly-varying (often monotonically) function of specimen thickness and defocus (over reasonable values). Here we define an incoherent condition as that under which the signal (IBF, ADF, EELS, EDX, etc.) is accurately given by multiplying the probe intensity inside the specimen by some object function. For EELS, these conditions are largely satisfied by using a collection angle significantly larger than the probe convergence angle (e.g., E0 = 100 keV, ~ 30 mrad, ~ 80 mrad). Nonetheless, even under these highly favorable conditions, strong elastic and thermal diffuse scattering can still cause the EELS signal to be "lost" outside the detector, making image interpretation non-trivial. This occurs particularly for energy losses less than a few hundred eV, and particularly for heavier and/or thicker specimens which tend to cause strong elastic and thermal diffuse scattering. Fig. 1 shows such an example for chemical maps of a BTO/STO multilayer [1]. For example, in Fig. 1, the original plasmonloss map exhibits strong minima, and the original Ba-N map shows rather poor maxima, both of which are the result of strong elastic scattering. On the other hand, the largely incoherent conditions lend themselves to a relatively simple procedure to correct, or compensate, for these elastic scattering effects. The "correction map" is derived from the zero- and low-loss intensity, which gives an incoherent bright-field image, the reciprocal of which is taken to obtain the correction map [1]. Despite its simplicity, the procedure has been demonstrated to perform very well under a large range of atomic number and specimen thickness [1,2]. For example, in Fig. 1, the plasmon-loss minima are entirely removed, the chemical contrast of the Ba-N map is enhanced, and the maps at higher energy losses show only small changes, as expected. Acknowledgements: YZ and CD acknowledge financial support from the Australian Research Council (DP110104734). The work at Cornell was supported by the U.S. Department of Energy, Office of Basic Energy Sciences, Division of Materials Sciences and Engineering (DE-SC0002334). Microsc. Microanal. 21 (Suppl 3), 2015 1236 References: [1] Y Zhu, A Soukiassian, DG Schlom, DA Muller, C Dwyer, Appl. Phys. Lett. 103 (2013), 141908. [2] H Tan, Y Zhu, C Dwyer, HL Xin, Phys. Rev. B 90 (2014), 214305. Figure 1. Correction of elastic scattering artifacts in chemical maps of a BTO/STO multilayer. The incoherent bright-field (IBF) and the correction map are shows at left. The color maps at right are the original maps (top row) and corrected maps (bottom row). [Data taken on Nion 100 kV UltraSTEM with a beam convergence semi-angle of 30 mrad and an EELS collection semi-angle of 80 mrad. The DualEELS mode of a Gatan Enfinium spectrometer was used: -50�462 eV in 0.5 sec, 320�832 eV in 10 msec].");sQ1[618]=new Array("../7337/1237.pdf","Identifying Atomic Reconstruction at Complex Oxide Interfaces Using Quantitative STEM","","1237 doi:10.1017/S1431927615006972 Paper No. 0618 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Identifying Atomic Reconstruction at Complex Oxide Interfaces Using Quantitative STEM Jared M. Johnson1, Justin Thompson2, S. S. A. Seo2, and Jinwoo Hwang1 1. 2. Department of Materials Science and Engineering, The Ohio State University, Columbus, OH 43212 Department of Physics and Astronomy, University of Kentucky, Lexington, KY 40506 We present a novel electron microscopy characterization of atomic reconstruction at epitaxial complex oxide interfaces. One of the most important aspects of oxide heterostructures is the discontinuity in polarity at the interface, which can lead to many exciting new properties that cannot be observed in bulk materials, such as metal-insulator transitions and superconductivity (e.g. [1]). The conflict of polarity may also result in atomic and electronic reconstructions at the interface, which may directly affect the transport properties of the interface. However, identifying the exact atomic reconstruction at the interface is a challenging task because the reconstruction is usually confined to an interfacial region thinner than 1 unit cell. Cross-sectional TEM or STEM images can reveal some information about the interfacial reconstruction, but it is difficult to obtain the reconstruction pattern along the depth direction (parallel to the beam) from the projected image. Therefore the identification of interfacial reconstruction requires a technique that can acquire atomic scale 3-dimensional (3D) information. In this work, we show that the exact stoichiometry and atomic reconstruction at the KTaO3/GdScO3 interface can be revealed using 3D quantitative STEM. Quantitative STEM measures the STEM annular dark-field signals on an absolute scale, which can be directly compared to simulated images. It enables direct interpretation of the STEM images in terms of the number of atoms and atomic species, and therefore provides information on the exact number and chemistry of the atoms in each atomic column [2]. Recently, Hwang and co-workers have advanced this technique to reveal atomic scale 3D information [3]. By including the probe channeling information in the analysis of quantitative STEM images, the 3D positions of impurity atoms in the atomic columns of SrTiO3 have been identified with depth uncertainty less than 1 unit cell. The analysis of quantitative STEM images depends on the comparison between the experimental and simulated images, so the use of STEM image simulation that fully captures the dynamic electron scattering is critical. KTaO3 and GdScO3 both have polar surfaces and therefore exhibit conflicting net charges at the interface, which drives the atomic/electronic reconstruction. Among many potential mechanisms of the reconstruction, our preliminary result shows that the reconstruction involves an intermixing region that consists of one KxGd1-xO layer and one TaySc1-yO2 layer [Fig. 1] [4]. The TaScO layer at the bottom shows ordered, alternating Ta- and Sc- rich columns, while the KGdO layer at the top shows an even distribution of atomic column intensities, which implies a more randomized atomic distribution [Fig. 1(A) and (C)]. We will show that by comparing the quantitative STEM images of the interface to multislice STEM simulations, the exact stoichiometry and the atomic ordering (sequence) along the depth direction at each atomic column can be precisely determined [e.g. Fig. 1(B) and (C)]. The atomic reconstruction may also involve the change in the oxygen positions; therefore identifying the oxygen positions at the interface is also crucial. For this, we use position averaged convergent beam electron diffraction (PACBED) that can determine the oxygen atom displacements in a unit cell with a great precision [5]. By comparing the experimental and simulated PACBED patterns, PACBED can also provide precise TEM sample thicknesses [Fig. 1(D)], which is crucial for the analysis of quantitative Microsc. Microanal. 21 (Suppl 3), 2015 1238 STEM images. The quantitative analysis requires very thin (< 10nm) TEM samples with damage-free surfaces, which we achieve by using the chemical-mechanical sample polishing technique [6], which provides far superior results compared to samples prepared by FIB. References [1] A. Ohtomo and H. Y. Hwang, Nature 427 (2004) p. 423. [2] J. M. LeBeau, S. D. Findlay, L. J. Allen, and S. Stemmer, Phys. Rev. Lett. 100 (2008) p. 206101. [3] J. Hwang, et al., Phys. Rev. Lett. 111 (2013) p. 266101. [4] J. Thomson et al., Appl. Phys. Lett. 105 (2014) p. 102901. [5] J. Hwang, J. Y. Zhang, J. Son, and S. Stemmer, Appl. Phys. Lett. 100 (2012) p. 191909. [6] P. M. Voyles, J. L. Grazul, and D. A. Muller, Ultramicroscopy 96 (2003) p. 251. Figure 1. (A) Experimental STEM ADF image of KTaO3/GdScO3 interface showing atomic reconstruction at the interface [4]. (B) (Left) A model of the squared region in (A), and (right) multislice simulation of the model with 3.2 nm sample thickness. (C) Qualitative comparison between the experimental and simulated line profiles for (top) KGdO layer and (bottom) TaScO layer. Fully quantitative comparison will require an experimental image on an absolute scale, and the thickness match between the experimental sample and the model. (D) (Left) Experimental PACBED pattern from the KTaO3 region in (A) that matches (right) the simulated (multislice) PACBED with 32 nm sample thickness.");sQ1[619]=new Array("../7337/1239.pdf","Revealing Unit Cell Level Distortions in Random Oxide Solid Solutions by Scanning Transmission Electron Microscopy and the Projected Pair Distribution Function","","1239 doi:10.1017/S1431927615006984 Paper No. 0619 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Revealing Unit Cell Level Distortions in Random Oxide Solid Solutions by Scanning Transmission Electron Microscopy and the Projected Pair Distribution Function Xiahan Sang1, Everett D. Grimley1, Changning Niu1, Douglas L. Irving1 and James M. LeBeau1 1. Department of Materials Science & Engineering, North Carolina State University, Raleigh, NC 27606 Picometer-scale lattice distortions critically determine a range of properties in oxide solid solutions including ferroelectricity, piezoelectricity, high temperature superconductivity, and thermoelectricity. Through solid solution alloying of constituent elements, these properties can be tailored for specific applications. Direct real-space visualization of local crystal distortions combined with atomic-level chemical information in solid solutions can thus provide key insights towards understanding the interplay between the lattice degrees of freedom. In this talk, we present an investigation of atomic-level displacements in (La0.18Sr0.82)(Al0.59Ta0.41)O3 (LSAT) where the A sub-lattice contains La/Sr, and the B sub-lattice contains Al/Ta. A spatially averaged face centered cubic unit cell has been determined by diffraction methods for this solid solution. Significant local unit cell distortion, however, is expected due to complex local chemical and charge variation on both A sub-lattice (La3+ and Sr2+) and B sub-lattice (Al3+ and Ta5+). The recently developed revolving scanning transmission electron microscopy (RevSTEM) technique removes sample drift distortion that enables investigation of picometer-level distortion within the 2D projection of the three dimensional crystal [1]. Figure 1b shows a typical RevSTEM image with evident intensity variation across atom columns. Weak columns are found to be Al-rich B sub-lattice atom columns as confirmed with atomic resolution energy dispersive X-ray spectroscopy (EDS) (Figure 1a). The average nearest like neighbor (NLN) around each atom column of A sub-lattice (Figure 1c) and B sub-lattice (Figure 1b) was calculated using atom column fitting and indexing technique as described in Ref. [2], and then used to render the rounded rectangles representing the magnitude of NLN. Larger unit cells are rendered red while smaller unit cells are rendered blue. Figure 1c and 1d show a significantly larger average NLN variation for A site than B site. Additionally, dark B sub-lattice columns tend to have small unit cells and bright B sub-lattice columns tend to have large unit cells. Further, we discuss the development and application of the projected pair distribution function (pPDF) to quantify distortion and correlation. This approach enables measurement of a function that is related to the PDF used in diffraction, but obtained directly from real-space atom column locations as schematically shown in Figure 2a. The partial pPDF can be determined for A sub-lattice and B sublattice (Figure 2b). The large variation for B sub-lattice than A sub-lattice is confirmed by significantly broader peaks in pPDF for B sub-lattice (Figure 2c). Through correlation analysis, we also will show that A sub-lattice cations move closer to B sub-lattice atom columns that are rich in Al atoms and away from those rich in Ta. We will demonstrate using density function theory (DFT) that the observed distortion is due to the local bonding configuration for the different sub-lattices. The local charge density of oxygen is shown to depend on the nearest B-site atom type, and is ultimately determined to provide the driving force for the structural distortion observed on the A sub-lattice. Finally, we will discuss how this approach may be applied across a range of complex oxides to understand their properties and phenomena [3-4]. Microsc. Microanal. 21 (Suppl 3), 2015 1240 References: [1] X. Sang and J. M. LeBeau, Ultramicroscopy 138 (2014), p. 28. [2] X. Sang, A. A. Oni and J. M. LeBeau, Microscopy and Microanalysis 20 (2014), p. 176. [3] X. Sang et al., Appl. Phys. Lett. 106 (2015), p. 061913. [4] The authors acknowledge the use and support of the Analytical Instrumentation Facility at North Carolina State University, which is supported by the State of North Carolina and the National Science Foundation. Figure 1. (a) Atomic resolution EDS highlighting Al-rich column with dark intensity in HAADF (b) sub-section of a RevSTEM image along <100> with the labels `A' and `B' denoting the corresponding sub-lattices. Average A-A (c) and B-B (d) NLN distance around each sub-lattice atom column. The indicated scale bars represent 1 nm. Figure 2. (a) Schematic of projected pair distribution function (pPDF). (b) pPDFs for A (red) and B (gray) sub-lattices calculated based on nth like-neighbor atom columns. (c) Comparison of first and second nearest like-neighbor peaks.");sQ1[620]=new Array("../7337/1241.pdf","Decomposing Electron Diffraction Signals in Multi-Component Microstructures","","1241 doi:10.1017/S1431927615006996 Paper No. 0620 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Decomposing Electron Diffraction Signals in Multi-Component Microstructures Alexander S Eggeman1, Duncan Johnstone1, Robert Krakow1, Jing Hu2, Sergio Lozano-Perez2 Chris Grovenor2 and Paul A. Midgley1 1. 2 Department of Materials Science and Metallurgy, University of Cambridge, Cambridge, UK . Department of Materials Science, University of Oxford, Oxford, UK The question of overlapping signals in measurements recorded in the transmission electron microscope is not a new one. The very name, transmission indicates that the recorded image will be a projection through a given thickness of the sample. In this process the ability to distinguish detail along the beam direction is usually lost. One common approach to separating the combined signals in TEM is through tomography, where multiple projections of the sample from different angles (typically through a systematic tilt-series) are combined to create a volume representation of the sample [1]. In this way some unique signal can be localized within the volume. In this work an alternative approach is used, namely the use of statistical decomposition of the recorded data in order to identify the character and extent of the unique signals existing within the TEM data. Such techniques are known as multivariate statistical analyses (MSA), among which principal component analysis (PCA) [2] and non-negative matrix factorization (NMF) [3] are two widely known and commonly employed methods. The technique takes a set of measurements and determines the subset of signals that can be combined in differing proportions (potentially both positive and negative) to accurately describe the whole. One of the earliest examples of this approach in TEM was to study STEM-EELS data [4]. In this situation a sample of titanium dioxide and tin oxide was scanned with the electron beam and an EELS spectrum recorded at each point. Given the close proximity in energy between the major edges for the two metals it was extremely difficult to separate their contributions quantitatively using conventional analytical approaches. However through MSA both the physical localization of each component, and their relative contribution to each EELS| spectrum could be determined and so the microstructural within the scanned area analyzed. While MSA has been shown to be extremely powerful for such spectroscopic analyses, there is no fundamental difference between spectral data and the structural information contained in a diffraction pattern. Using the scanning diffraction analysis developed by Nanomegas and Grenoble University (a technique referred to as ASTAR or ACOM) [5] it is possible to record spatially resolved diffraction data by scanning an electron beam across the sample and recording a whole diffraction pattern at each point. Like the EELS example shown previously, the presence of multiple different crystals within a sample (which can encompass differently oriented grains, bent or otherwise strained regions as well as different phases) all can give rise to complex diffraction patterns containing multiple different signals, and which can be decomposed to determine the individual diffraction signals. An initial proof of principal study for this technique was performed on a nickel superalloy sample. Here a carbide precipitate was found embedded within the nickel matrix. This particular system showed a strong crystallographic registry between the matrix and precipitate, making simple separation of the multiple diffraction patterns impossible. Instead through MSA the unique diffraction signal from all of Microsc. Microanal. 21 (Suppl 3), 2015 1242 1236 the phases in the structure could be identified and localized. In this study the process was taken further by recording a tilt-series of such data, in this way the diffraction signal from each phase allowed a volume reconstruction of each of the major components under scrutiny in the microstructure (shown in Figure 1a). Furthermore the inter-related information about the morphology of the microstructure and the crystalline orientation of the components allowed interface relationships within the structure to be unambiguously determined (Figures 1b and 1c), leading to the identification of a hitherto unreported crystallographic registry between the carbide and matrix phases [6]. Further studies have been undertaken looking into a variety of different materials systems; from interfaces in twinned gold nanoparticles and semiconductor nanowires; through to the distribution of oxide phases in zirconium layers. In this latter case the separation of the principal diffraction signals from different regions confirmed the existence of a proposed sub-oxide phase [7] and by identifying the overlapping grains with different oxygen content further insight into the nature of the EELS signal recorded from such samples is expected. [1] P. A. Midgley and R. E. Dunin-Borkowski, Nature Materials 8 (2009), p. 271-280 [2] H. Abdi and L.J. Williams, Wiley interdisciplinary reviews: computational statistics 2 (2010), p. 433-459 [3] D. D. Lee and H. S. Seung, Nature 401 (1999), p. 788-791. [4] F. de la Pe�a. et al. Ultramicroscopy 111 (2011), p. 169-176 [5] E. F. Rauch et al., Microsc. Anal. 22 (2008), p. S5-S8 [6] A. S. Eggeman, R. Krakow and P. A. Midgley, Nature Communications in press. [7] J Hu et al., Micron 69 (2015), p. 35-42 a b c Figure 1. a) Reconstructed volume of superalloy sample showing carbide (blue) and -phase (green) precipitates within a nickel matrix (orange). The localization of unique diffraction signals was used as input for the reconstruction. The cube side is approximately 200nm. b) and c) pole-figures for the (220) planes in the carbide and the (026) planes in the nickel matrix respectively across the tilt-series (each tilt is indicated by a different colour). The correspondence highlights the crystallographic registry between the two structures.");sQ1[621]=new Array("../7337/1243.pdf","Dealing With Multiple Grains in TEM Lamellae Thickness for Microstructure Analysis Using Scanning Precession Electron Diffraction","","1243 doi:10.1017/S143192761500700X Paper No. 0621 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Dealing With Multiple Grains in TEM Lamellae Thickness for Microstructure Analysis Using Scanning Precession Electron Diffraction A. Valery1,2, E. F. Rauch2, A. Pofelski1, L. Cl�ment1, F. Lorut1 1. 2. STMicroelectronics, 850 rue Jean Monnet, F-38920 Crolles, France. SIMAP/GPM2 laboratory, 101 rue de la Physique, 38402 Saint Martin d'H�res, France. Materials microstructure is a source of variability in devices performance considering today's transistor feature sizes. As the well-known Electron BackScatter Diffraction (EBSD) technique [1] shows limitations to characterize grains orientation smaller than a hundred nanometers, new tools have been developed over the past few years to characterize the texture of crystalline materials at nanometer scale. Among them is the ASTAR tool based on the acquisition and indexation of Precession Electron Diffraction (PED) patterns acquired in scanning mode [2]. Although its use has proved to be efficient to analyze nano-crystalline materials [3], indexation issues appear when crystal grains size is significantly smaller than the lamella thickness. For such cases, the acquired images are composed of a superimposition of several diffraction patterns. Strong expertise and time is required to prepare ultra-thin TEM lamellae adapted for the analysis of current nanometer scale transistor devices so that only the feature of interest remains in the thickness while left crystalline. As an alternative, a dedicated procedure is proposed to overcome crystal overlapping issues and exploit volume information: each diffraction pattern is iteratively re-indexed after subtraction of the reflections related to the main--or previous-- solution. In this work, this technique is used to study the microstructure of a material when several crystalline phases remain in the TEM lamella thickness. The sample observed was a planar section of nickel silicide (NiSi) thin films (~30nm) on Silicon-OnInsulator (SOI) where some monocrystalline silicon (Si) remained in the thickness of a typical 50-70nm lamella. A scheme of the studied sample is shown in figure 1a. The lamella was prepared using a H450 FEI dual beam system. Experiments were carried out using a FEI Tecnai G2 F20 S-Twin FEG (S)TEM operating at 200 keV. A quasi-parallel probe of 4nm was used to scan the sample, with a convergence semi-angle measured to be at 0.4mrad. For the acquisition, the sample was oriented along the silicon [001] direction. Scanning electron diffraction and orientation analysis were performed using the ASTAR/DigiSTAR system. Figures 2a shows an example of a diffraction pattern acquired during the scanning. At least two different diffraction arrays can be observed and retrieved as illustrated in figures 2b and 2c. The Si phase comes as the first indexation result and some reflections are clearly recognized as the signature of [001]oriented silicon (figure 2b). After subtraction of the Si-related reflections, the remaining diffraction pattern free of any Si footprint is available for indexation. Thanks to this re-indexation, a more reliable orientation mapping of NiSi phase [4] can be produced as displayed in figure 1c. In figure 2c, the orientation of a NiSi grain is correctly indexed as a specific simulated template matches the diffraction pattern with a good correlation index. The indexation clearly identifies the morphological aspect of NiSi grains all along the surface. An interesting result is the epitaxial relationship between the two phases as most of the NiSi crystals are oriented along the [001] direction such as Si. To summarize, a specific method to extract and exploit the multi-information present in diffraction patterns was examined. The experiment demonstrated that this technique allows us to perform the Microsc. Microanal. 21 (Suppl 3), 2015 1244 relevant characterization of NiSi grains, even if the TEM lamella was not thin enough to include the only phase to observe. This alternative approach can be an efficient answer to avoid the preparation of TEM lamellae as thin as features size and thereby provide keys for microstructure analysis of nanocrystals. References [1] V. Randle, Materials Characterization 60 (2009), p.913�922. [2] E. F. Rauch, M. V�ron, Materials Characterization 98 (2014), p.1�9. [3] A. Valery et al, International Microscopy Congress Proceedings (2014), p.3205-3206. [4] M. Kh. Rabadanov and M. B. Ataev, Inorganic Materials 38 (2002), p.120-123. Figure 1. Orientation analysis of the observed TEM lamella, including: (a) a scheme of the prepared planar TEM lamella; (b) the orientation map of the Si remaining in the sample; (c) the orientation map of the NiSi characterized through re-indexation; (d) and (e) the color codes for Si (cubic system) and NiSi (orthorhombic system), respectively. Figure 2. Illustration of a diffraction pattern (re)indexation, with: (a) a diffraction pattern corresponding to a point in the mapping illustrated above; (b) the diffraction spots firstly indexed as Si by template matching; (c) the diffraction spots indexed as NiSi after image treatment; (d) and (e) the ASTAR correlation index maps of the [001]-oriented Si and [001]-oriented NiSi after re-indexation, respectively.");sQ1[622]=new Array("../7337/1245.pdf","Principles and Applications of Energy-Filtered Scanning CBED for Ferroelectric Domain Imaging and Symmetry Determination","","1245 doi:10.1017/S1431927615007011 Paper No. 0622 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Principles and Applications of Energy-Filtered Scanning CBED for Ferroelectric Domain Imaging and Symmetry Determination Yu-Tsun Shao1, Kyou-Hyun Kim2 and Jian-Min Zuo1,3 1. 2. Department of Materials Science and Engineering, University of Illinois, Urbana, Illinois 61801, USA Advanced Process and Materials R&BD Group, Korea Institute of Industrial Technology, Incheon, 406-840, Korea 3. Frederick Seitz Materials Research Laboratory, University of Illinois, Urbana, Illinois 61801, USA The presence of nanodomains is considered to play an important role in determining the optical properties, dielectric permittivity and polarization switching in relaxor-based ferroelectric crystals. Thus, understanding the complex nanodomain structure could provide important clues about the mechanisms that produce the extraordinary piezoelectric responses in relaxor-based ferroelectric crystals. However, techniques such as X-ray diffraction (XRD), neutron diffraction and optical microscopy only provide information about the macroscopic or average symmetry by using probe sizes that are much larger than the size of nanodomains. Determining the polarizations of nanodomains are in general performed using piezoresponse force microscopy (PFM). Atomic resolution scanning TEM (STEM) is also used to determine polarization directions in ferroelectric thin films. For the complex nanodomain structure in relaxor-based ferroelectric single crystals, convergent beam electron diffraction (CBED) has been considered as a powerful technique by using a finely focused probe of a few nanometers in diameter or less [1]. Here we use a recently-developed technique called scanning CBED which has the capability of quantifying as well as mapping the local symmetry variations at nanometer scale [2]. This is based on automated recording of the CBED patterns on the CCD camera while scanning over the region of interest with nanometer-sized electron beam and correlation analysis of recorded patterns. The relaxor-based ferroelectric single crystals of (1-x)Pb(Zn1/3Nb2/3)O3-xPbTiO3 (PZN-xPT) are studied here. This system together with PMN-xPT (M for Mg) have attracted much research interest due to their superior piezoelectric properties (d33>2,500 pm/V and k33>0.9) in the morphotropic phase boundary (MPB) region compared to commercial piezoelectric ceramics [3]. The lead-based relaxor-based ferroelectric crystals are known to have three types of macroscopic symmetry of R (rhombohedral), M (monoclinic), and T (tetragonal) at room temperature dependent on composition. The local symmetry in MPB region of relaxor-based ferroelectric crystals, however, is still under debate. In our study, the scanning CBED was performed at JEOL 2010F TEM, the electron probe had a diameter of about 2.3 nm in FWHM (full-width at half-maximum). The scanning CBED was performed using a step size of 2 nm, and we recorded 15 x 15 data points in a local area of the size 28 nm x 28 nm inside the specimen. Each CBED patterns were energy-filtered using the post-column Gatan Imaging Filter (GIF) with 10 eV energy window to reduce inelastic scattering effects. Figure 1(a) shows the bright field image of the PZN-8%PT sample, the yellow square indicates the scanned region. To analyze the patterns, we calculated the correlation coefficients among different CBED patterns in each scanned dataset. To quantify the similarity between two images, we used the normalized cross-correlation coefficient and profile R-factor, and Rp, respectively, by pixel-to-pixel comparison. We then set the threshold of manually to obtain a stack of templates from the original scanned image stack. By correlating the generated templates with the original scanned images we thus have the correlation map. Figure 1(b) is the correlation map of scanning CBED patterns, which maps out various polarization Microsc. Microanal. 21 (Suppl 3), 2015 1246 nanodomains. Figure 2(a) is the representative CBED patterns of the PZN-8%PT single crystal, the numbers correspond to Fig. 1(b). Figure 2(b) is the simulated CBED pattern on (001) zone axis based on the Bloch-wave method. Thus, scanning CBED combined with correlation analysis and Bloch wave simulations provides a new method for imaging nanodomains in ferroelectric crystals at about 2nm resolution. References: [1] K. Kim, D. A. Payne, and J. M. Zuo, Phys. Rev. B 86 (2012), 184113. [2] K. Kim and J. M. Zuo, Ultramicroscopy 124 (2013), 71. [3] S. E. Park and T. R. Shrout, J. Appl. Phys. 82 (1997), 1804. [4] This work is supported by U.S. Department of Energy, Office of Basic Energy Sciences under contract DEFG02-01ER45923. (a) (b) 0 6 Figure 1. (a) Bright-field image of a 55nm PZN-8%PT sample. The yellow square represents the region of scanning (b) correlation map of scanning CBED patterns. (a) 0 A B 6 A (b) B A' B' A' B' Figure 2. (a) Representative energy-filtered CBED patterns of PZN-8%PT single crystal, the numbers correspond to the numbers in Fig. 1(b). To quantify the mirror symmetry, we compared two pairs of diffraction disks within each pattern, A-A' and B-B', by calculating the and Rp of the mirror-applied A (and B) with A' (and B') using methods described in ref. [2]. The highest values of (, Rp) we calculated from A-A' of 0th- and 6th-template are (0.95, 0.08) and (0.65, 0.29), respectively. For B-B', the values are (0.90, 0.22) and (0.87, 0.34). (b) Simulated CBED pattern using the Bloch-wave method.");sQ1[623]=new Array("../7337/1247.pdf","A Statistical Dictionary Approach to Automated Orientation Determination from Precession Electron Diffraction Patterns","","1247 doi:10.1017/S1431927615007023 Paper No. 0623 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A Statistical Dictionary Approach to Automated Orientation Determination from Precession Electron Diffraction Patterns A. Wang1 , A.C. Leff2 , M.L. Taheri2 and M. De Graef1 1. 2. Dept. of Materials Science and Engineering, Carnegie Mellon University, Pittsburgh PA 15213, USA Dept. of Materials Science and Engineering, Drexel University, Philadelphia PA 19104, USA The precession electron diffraction (PED) technique has been widely used for structure determination since Vincent and Midgley proposed it in 1994 [1]. By acquiring electron diffraction patterns using electron beam precession, the usually strong dynamical scattering effects can be averaged out. Moreover, PED patterns can be indexed with a higher angular resolution than conventional diffraction patterns since more reflections can be collected. In order to determine the crystal orientation from a PED pattern, one would need to measure distances and angles between diffraction spots, a process that rapidly becomes tedious for multigrain samples. A faster approach would be to use a physics-based model to produce simulated diffraction patterns which are pre-calculated according to the microscope settings and the crystal parameters; this is known as a dictionary-based approach. The crystal orientation for a given PED pattern is then obtained by finding the best match against the dictionary. Rauch et al. [2] have developed an automated orientation and phase mapping method in TEM by using PED, which has been implemented in NanoMEGAS' commercial software ASTAR. This method makes use of template-matching algorithms, where the templates are pre-calculated based on the crystal symmetry using kinematical scattering within the fundamental zone in Euler angle space. The locations and intensities of the diffraction spots in an experimental PED pattern are then cross-correlated with the template patterns to find the best match. However, the extensive use of image filtering in ASTAR can make the indexing process somewhat less reproducible because the filtering parameters used prior to cross-correlation may yield variable indexing results. Additionally, the method described in [2] does not always reliably identify orientations from patterns taken on or near grain boundaries because diffraction spots from both orientations are present in the patterns. In [3], an alternative and simple indexing framework based on normalized inner products between the experimental and dictionary patterns was proposed. This approach can determine the crystal orientation but can also differentiate the grain interior from the grain boundary without extensive image processing. Moreover, this approach has the advantage of operating on the raw data, without prior application of image processing tools. The first step is the computation of normalized inner products for each experimental PED pattern with all the dictionary patterns. The set of Euler angles of an experimental pattern is then obtained from the dictionary pattern that results in the largest inner product. The second step is to use a decision tree classifier to cluster the diffraction patterns using the statistical distribution of the normalized inner products and then to segment the grains and identify anomalous regions. This technique was successfully demonstrated with EBSD indexing for a Ni-base alloy. In the present study, we apply the method described in [3] to index PED patterns in a Cu sample containing a twin boundary and compare the results with those obtained using the ASTAR approach. Fig. 1.(a) shows a virtual bright field image constructed from PED patterns acquired across an area of 60 � 80 samples, using a 10.4 nm step size. The data was acquired using a JEOL 2100 LaB6 operated at 200 keV, using a 10 nm spot size and a precession angle of 0.60 . An example PED pattern acquired from Microsc. Microanal. 21 (Suppl 3), 2015 1248 the upper grain and its best match from the dictionary are presented in Fig. 1.(b) and (c), respectively. As shown in Fig. 1.(d), the image pixels are clustered well to differentiate the two grains and the twin boundary is clearly visible. Also, the sample edge is clearly defined. The Euler angle results obtained from the dictionary approach are shown in Fig. 1.(e). Based on the two cluster angles, the misorientation between the two grains is a 62.55 rotation about the [0.62, 0.56, 0.55] axis (3.4 from [111]), veryfying the boundary to be a twin boundary. The y-axis orientation map obtained from ASTAR is shown in Fig. 1.(f); the edge of the sample is not visible, presumably due to residual intensity from the screen phosphor for beam positions outside of the sample. Our dictionary approach shows promising results for the indexing of orientations from PED patterns. Further investigations for multigrain thin films are currently ongoing. References: [1] R. Vincent and P.A. Midgley, Ultramicroscopy 53 (1994), p.271. [2] E.F. Rauch and M. Veron, Materwiss. Werksttech. 36 (2005), p.552 [3] Y.H. Chen, S.U. Park, D. Wei et. al., Microscopy and Microanalysis, under review. [4] A. Wang and M. De Graef are grateful for AFOSR support under MURI contract #FA9550-12-1-0458. A.C. Leff and M.L. Taheri are grateful for support from: US DOE Basic Energy Sciences Early Career program (DE-SC0008274); National Science Foundation Faculty Early Career Program (#1150807); and DOE Nuclear Energy University Program (NE0000315). Figure 1. (a) Virtual bright field image of a Cu sample with a twin boundary, image area 60 � 80 pixels, pixel size 10.4 nm; (b) Experimental PED pattern from the upper grain; (c) Best matching pattern from the dictionary; (d) Color map resulting from the cluster classifier (red pixels are grain interior, light green twin boundary, blue vacuum area); (e) Euler angle distribution, showing two clusters; (f) ASTAR color map.");sQ1[624]=new Array("../7337/1249.pdf","Correlation of Electron Diffraction between t-EBSD in the SEM, CBED in the TEM and ACOM using ASTAR in the TEM using GaN Nanowires","","1249 doi:10.1017/S1431927615007035 Paper No. 0624 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Correlation of Electron Diffraction between t-EBSD in the SEM, CBED in the TEM and ACOM using ASTAR in the TEM using GaN Nanowires Roy H. Geiss Department of Chemistry, Colorado State University, Fort Collins, CO 80523-1872, USA Transmission electron backscattered diffraction, t-EBSD, in the scanning electron microscope, SEM, was initially described in 2010 [1]. After slow initial acceptance, the number of publications in which it is used is increasing markedly. Many of the first applications took advantage of the higher resolution of t-EBSD in the study of thin films [2], [3]. More recently it is being used for pre-screening thin films and foils to determine which grains are best oriented to study certain defects in transmission electron microscopy, TEM [4], [5]. A major advantage of using t-EBSD in the SEM to screen thin films is that it is much easier and quicker to map the orientation of large areas of samples than with conventional diffraction techniques in the TEM. Automated crystal orientation mapping, ACOM, using precession electron diffraction, PED, in the TEM [6] offers an alternate approach to mapping the orientation of large sample areas without having to reposition the sample between the SEM and TEM. In this work electron diffraction using t-EBSD in the SEM, convergent beam electron diffraction, CBED, in the TEM and ACOM in the TEM will be compared using data obtained from the same GaN nanowires applying the three techniques. The actual purpose of the study was to determine the growth properties of the nanowires, including the growth axis and fluctuations in orientation along that axis. To facilitate locating the same nanowire in both the SEM and TEM for t-EBSD and CBED, a single tilt specimen holder for a JEM 2100F TEM was modified to be used in both the 2100F TEM and a JSM 6500F SEM with no repositioning of the sample. To ensure that the same nanowires were studied at AppFive, the GaN nanowires were dispersed on annotated carbon coated TEM grids. (ACOM analysis will be done on the same wires at the AppFive, LLC research facility in Tempe, AZ and be available for the meeting). The t-EBSD patterns were collected in the JSM 6500F SEM using a EDAX Hikari Plus camera. The sample was tilted to -30 degrees with the stage tilted +60 degrees, Figure 1. In the JSM 6500F this allows the working distance, WD, to be varied from 5 mm up. The patterns obtained for this experiment were obtained at a 10 mm WD using 25 kV. Images were collected using either the SE detector in the SEM or a forward scattered detector, FSD, attached to the Hikari camera. Figure 2 shows an SE image of the area with the four GaN nanowires used in this study. The t-EBSD patterns were collected using line scans along the length of the wires; 10 nm steps were used resulting in collections of 200 � 400 data points from a 2 � 4 micron line scan. Images of the individual diffraction patterns from each point were collected as well as the Hough data. Individual patterns were indexed and the axial orientation of the wire was determined. It is assumed that the wires are lying flat on the carbon substrate, but this is not always the case. However, most lie within a few degrees of being flat so that the exact orientation of the nanowire axis can be determined to within a few degrees. Individual t-EBSD patterns from nanowires #4 and #7 clearly show that the axial orientations of the two nanowires are more than a few degrees different, Figures 3 and 6. Analysis software can be used ascertain point-to-point and point-to-origin Microsc. Microanal. 21 (Suppl 3), 2015 1250 misorientations of the patterns along a line scan of an individual wire. Figure 4 shows such a plot for wire #4. In general an array of CBED diffraction patterns from an area in a TEM can only be obtained by manually moving the beam and saving the patterns. Also, in contrast to t-EBSD patterns, CBED patterns take considerably more time to analyze. Examples of a CBED pattern and a t-EBSD pattern from the same nanowire, #7, are given in Figures 5 and 6, respectively. They show the same orientation, allowing for the approximately 45 degree CCW rotation between them. The merits of collection and analysis of t-EBSD patterns in the SEM will be compared to the collection and analysis of both CBED patterns and ACOM - ASTAR data in the TEM. References: [1] RH Geiss, et al, Microsc Microanal 16(S2) (2010) p.1742�43. [2] PW Trimby, et al, Ultramicroscopy 120(0) (2012) p.16�24. [3] N Brodusch, et al, J Microscopy 250(1) (2013) p.1�14 [4] S. Suzuki, JOM, 65(9), (2013), p.1254-1263 [5] R. de Kloe, EDAX blog, October 2014 [6] D Viladot, et al, J Microscopy 252(1) (2013) p.23-34 Figure 1. Experimental t-EBSD Figure 2. SE image of GaN wires Figure 3. t-EBSD from wire #4. Figure 4. Misorientation plots Figure 5. CBED from wire #7 Figure 6. t-EBSD from wire #7");sQ1[625]=new Array("../7337/1251.pdf","Quest for an Optical Circuit Probe","","1251 doi:10.1017/S1431927615007047 Paper No. 0625 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quest for an Optical Circuit Probe Jun Xu1, Anshuman Kumar1, Kin Hung Fung2, Dafei Jin1, and Nicholas X. Fang1 1. 2. Dept of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, USA Dept of Applied Physics, The Hong Kong Polytechnic University, Hong Kong What is common in near field optics and the electronic circuit probe techniques? Typical electronic test probes are conveniently used to connect test equipment such as oscilloscopes to an RF integrated circuit. Likewise in near field optics, it is highly desirable if we have similar precision test probes with both high spatial and spectral resolution to study the optical phenomena. Such light-matter interaction in nanostructures involving single and collective emitters can enhance our understanding and design for cavity quantum electrodynamics. Recently, low-energy excitation by swift electrons has been applied to explore the optical response of nanostructured materials with unmatched spatial resolution. For example, the measurement of photon emission rate using cathodo-luminescence techniques can provide a direct quantification of local radiative energy transfer from electron to photons at spot size close to 10 nanometers, which compliment other experimental techniques such as photoluminescence. Such tightly localized excitation source and a high-resolution mapping technique is particularly suited to study the spontaneous emission rate of a plasmonic nanostructure with sub-wavelength mode volume. In this invited talk, we will discuss our effort of developing near field optical and electron probes for characterizing optical nanstructures for light trapping, mixing and extraction. A fundamental research in the physics of metamaterials is the interaction between electron and the cavity plasmon [1], which plays an important role in many applications such as nanolasers [2] and nanoantennas [3]. Dark plasmonic modes, which lead to "forbidden" photon transition, are of particular interest as they promise to controllable stimulated emission at nanoscale. Recently, many novel phenomena associated with dark modes are also observed. For instance, it has been shown that the Fano resonances associated with the coupling between dark and bright modes can lead to extraordinary far-field phenomena [4]. Our preliminary research show that photon emission of bright and dark modes in single metal bowtie nanoantenna can be selectively excited using a focused electron spot as shown in Fig. 1. The measured high resolution spectrum shows unexpected high photon counts associated with signatures of the dark modes, in contrast to common wisdom that such dark modes only couple weakly to the far field. With an eigendecomposition response theory [5], we successfully explained the phenomena as a result of hybridization between localized plasmons. These probes could provide brand new insight of quantum metamaterials with unprecedented fine spectral and spatial resolution. In the second part of this talk, we present a potential application of local electron energy loss spectroscopy as a quantitative optical probe for photon emission rate engineering. For example, optical excitation on hexagonal boron nitride (hBN) nanostructures has sparked remarkable research interests, since such materials could accommodates highly dispersive surface phononpolariton modes. Our theoretical model on graphene-hBN heterostructures suggests that the graphene plasmon couples differently with the phonons of the two Reststrahlen bands in the mid- Microsc. Microanal. 21 (Suppl 3), 2015 1252 IR, owing to their different hyperbolicity. This also leads to distinctively different interaction between an external quantum emitter and the plasmon-phonon modes in the two bands, leading to substantial modification of its spectrum. The coupling to graphene plasmons allows for additional gate tunability in the Purcell factor, and an induced transparency in its emission spectra. References: [1] F. J. G. de Abajo, Rev. Mod. Phys. 82, 209�275 (2010). [2] D. J. Bergman, and M. I. Stockman, Phys. Rev. Lett. 90, 027402 (2003). [3] P. Chaturvedi, et al., ACS Nano 3, 2965�2974 (2009). [4] V. A. Markel, J. Opt. Soc. Am. B 12, 1783�1791 (1995). [5] K. H. Fung, et al, Phys Rev. B, 89, 045408(2014). [6] The authors acknowledge the financial support by the NSF (grant CMMI-1120724) and AFOSR MURI (Award No. FA9550-12-1-0488). K.H.F. acknowledges financial support from Hong Kong RGC grant 509813. Figure 1. Cathodoluminescence spectroscopy of plasmonic nanoantennas. (a), Schematic diagram of the setup and the process of photon generation by electron beam. Electrons are incident from top of the sample. Photons emitted from the nanoantenna are collected collimated by a parabolic mirror before detection. (b) SEM picture of a fabricated bowtie antenna. (c) Measured CL spectra at three locations indicated as colored dots in b. The inset shows a panchromatic spatial image collected for the whole spectrum detected by the photodetector. Bright color corresponds to high photon counts.");sQ1[626]=new Array("../7337/1253.pdf","Quantum and Time-Resolved Nano-Optics using Auto-Correlated Cathodoluminescence in a STEM","","1253 doi:10.1017/S1431927615007059 Paper No. 0626 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantum and Time-Resolved Nano-Optics using Auto-Correlated Cathodoluminescence in a STEM Sophie Meuret1, Luiz Tizei1, Romain Bourrellier1, Thomas Auzelle2, Fran�ois Treussart3, Bruno Daudin2, Alberto Zobelli1 Bruno Gayral2 and Mathieu Kociak1 1. 2. Laboratoire de Physique des Solides, University Paris-Sud, Orsay France CEA-CNRS Group "Nanophysique et Semiconducteurs," F-38054 Grenoble Cedex 9, France 3. Laboratoire Aime Cotton, CNRS, Universite Paris Sud, and ENS Cachan, 91405 Orsay cedex, France The interest for single photon emitters (SPE) has tremendously grown over the last decades, due to their possible applications in quantum information. Famous SPE are for example InAs/GaAs quantum dots or Nitrogen Vacancy (NV) centers in diamond. A SPE emits only one photon at the time, and therefore it is a natural candidate for solid quantum bits implementation. The usual way to characterize them is to perform an intensity interferometry experiment (Hanbury Brown and Twiss (HBT) interferometry). Such an experiment measures the autocorrelation function g(2)() of emitters. The g(2)() function of a SPE presents a dip at very short delay (g(2)(0) < 1), a phenomenon called the anti-bunching. Here, we used a unique home-made cathodoluminescence (CL) set-up in a scanning transmission electron microscope (STEM) coupled to an HBT experiment allowing nanometer resolution. The experiments are detailed in ref. [1,2] and exemplified in figure 1. Other than the spatial resolution, one of the main benefits of CL over photoluminescence (PL) is the facility to study UV emission due to broadband character of the electron excitation. In this presentation, we will first show that CL-g(2)() had allowed to characterize a new UV-SPE in hexagonal Boron nitride, possibly a new emitter for quantum device. Then, in order to go further in the understanding of this new technics, we will see that even if the interaction mechanisms of PL and STEM-CL with materials are close enough to give the same emission spectra [3], they may lead to huge differences in their g(2)() function, called respectively PL-g(2)() and the CL-g(2)(). Indeed the interaction of electrons with mater produces a plasmon, which will decay into multiple electron-hole pairs at the gap energy (e-h), while the PL-photon mater interaction produces only one e-h. Thus, if there is more than one SPE in the sample, one electron can excite simultaneously multiple centers leading to the synchronization of emission and thus to the emission of bunches of photons. Therefore, if the number of excited centers is larger than one, the CL- g(2)() function will present a huge bunching effect (g(2)(0) >> 1) in stark contrast to the expected flat PL-g(2)() function (g(2)(0) = 1) see figure 3. We will see how, in addition to be a new physical phenomenon, the CL bunching effect is a way to measure the lifetime of emitters at a nanometer scale, without needing pulsed electron guns. It allows the correlation in no time of HADF images, emission spectrum and lifetime measurement. [1] Zagonel and al., Nano Letters 11, 568-73 (2011) [2] Tizei and al., PRL 110, 153,604 (2013) [3] Mahfoud and al., J. Phys. Chem. Lett., 4090-94 (2013) [4] K. Hara and al., Phys. Status Solidis C 8 n�7-8, 2509 (2011) Microsc. Microanal. 21 (Suppl 3), 2015 1254 Figure 1. Study of a localized defect in h-BN:a) Sketch of a STEM column fitted with a CL attachment. The CL signal is sent b) to an spectrometer which records the emission spectrum at each pixel or c) to an Hanbury Brown and Twiss interferometer for CL-g(2)() measurement [2]. A filtered energy map at 4.09 eV of a BN flake is shown in b). Very localized emission of a common defect [4] (spectrum displayed in b) and d)). d) Evidence of single photon emitter behavior. Up Filtered (blue square on the spectra) imaging of the localized BN-defect. Middle: Spectrum taken after the CL-g(2)() acquisition. Bottom= autocorrelation CL-g(2)() of the localized BN-defect study. Figure 2. CL-g(2)() of defects in h-BN for different excitation currents when there is more than one defect excited at the same time. A huge bunching effect (g(2)(0) >> 1) is clearly visible.");sQ1[627]=new Array("../7337/1255.pdf","Correlative STEM-Cathodoluminescence and Low-Loss EELS of Semiconducting Oxide Nano-Heterostructures for Resistive Gas-Sensing Applications","","1255 doi:10.1017/S1431927615007060 Paper No. 0627 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Correlative STEM-Cathodoluminescence and Low-Loss EELS of Semiconducting Oxide Nano-Heterostructures for Resistive Gas-Sensing Applications Derek Miller1, Sheikh Akbar1, Pat Morris1, Robert E.A. Williams1,2, David W. McComb1,2 Materials Science and Engineering Dept., The Ohio State University, Columbus Ohio, United States. Center for Electron Microscopy and AnalysiS (CEMAS), The Ohio State University, Columbus Ohio, United States. 2. 1. Recent advances in resistive-type oxide gas sensors have been made primarily by the combination of multiple materials into nano-heterostructures that often show enhanced or unique properties compared to each pure material constituent [1,2]. It has been shown to be especially useful to use highly crystalline one-dimensional nanorods and nanowires decorated with either discrete oxide particles or a continuous coating generating a core-shell structure [3]. The operating principle of these sensors is simply measuring a change in resistance of a film deposited between two or more electrodes, which varies with the number of charge carriers accepted or donated between the surface and the nearby gas molecules. Two different resistance-dominating mechanisms exist in these sensors when the nanowires are deposited as a random network film [1]. In both mechanisms a depletion layer is formed at the surface and a larger depletion layer creates a higher resistance. The first mechanism considers the potential energy barrier at the interface between two nanowires that an electron must overcome to move through the film. The second mechanism considers electrons moving along the axis of a nanowire and the constriction of the cross-sectional area that does not lie in the depletion region. This can be equated to pushing a current through a smaller diameter wire, which increases the resistance. The second axial mechanism can be engineered through decoration or coating of, for example, p-type Cr2O3 onto an n-type SnO2 nanowire. The p-n junctions created can affect this mechanism greatly, enhancing the sensor response and even making the material more selective toward specific gases and reducing cross-interference. There has been little attempt to characterize the electronic interactions of the core and coating materials. Choi et al. have shown that the selection of a coating material with a higher or lower Fermi energy than the core in n-n and p-p junctions can make SnO2 nanowires more selective to oxidizing or reducing gases, respectively [3,4]. Still, no direct evidence exists to show how well the band edges and defect states (shallow and deep levels) near the heterojunction interface agree with bulk measurements taken over a large sample area. Direct evidence of the electronic structure at a high spatial resolution would enhance the currently over-simplified models and allow better bottom-up design and materials selection. In this study single-crystal n-type SnO2 nanowires have been coated by magnetron sputtering and chemical vapor deposition with n-type TiO2, WO3, ZnO, Nb2O5, and MoO3 as well as p-type Cr2O3, Co3O4 and NiO. Coating thickness and continuity is also varied. Each nano-heterostructure material is tested for gassensing performance toward several reducing and oxidizing gases as a random network film as well as single-nanowire sensors fabricated via Pt deposition in a dual-beam FIB. Initial results have shown that the random network film switches from n-type to p-type response when a thick enough p-type coating is applied. However, single nanowire measurements show only n-type behavior no matter the coating thickness. These effects are likely due to choice of conduction paths through the nanostructures. These sensor measurements form the basis for the study but direct characterization of the electronic structure can greatly aid in the explanation of the results. Microsc. Microanal. 21 (Suppl 3), 2015 1256 A monochromated FEI Titan 80-300 S/TEM equipped with a Gatan Vulcan cathodoluminescence (CL) detector and a Gatan Tridiem EELS system has been used for high spatial resolution mapping of defect states and band edges, respectively. Knowledge of the defect states is essential when evaluating a material for gas sensing applications because the surface defects control the reaction with the gases. CL studies performed in the SEM of SnO2 powders and nanowires assigned peaks to a number of different oxygen vacancies and structural defects but many aspects remain uncertain [5]. The SEM-CL spatial resolution results in sampling many particles or nanowires at once. In contrast, the STEM-CL system allows spatial mapping of CL spectra from specific nanowire facets that can be correlated to the surface defects on each crystalline plane. Furthermore, the electronic defect states present in the coating materials in nanocrystalline form can be mapped and compared to bulk materials. This information adds fundamental knowledge to the behavior of these materials as they exist in this widely-used nano-heterostructure form and helps refine the models of gas-surface reactions. As the CL data does not include full band-gap transitions, monochromated EELS was used to map the band edge onsets across the heterojunction interface in order to draw a more realistic heterojunction band diagram. These measurements can be correlated with the CL data to help predict charge carrier movement across the heterojunction interface and therefore the behavior of the depletion region between the two materials. The single-nanowire electrical measurements show non-linear and sometimes asymmetric behavior and can be used to draw information about the built-in voltage at the heterojunction interface. I-V curves taken in different gas atmospheres and varied elevated temperatures also help to draw information about how the depletion region and built-in voltage behave during real sensor tests. The future challenge remains to correlate this high-resolution electronic structure information to the resulting gas sensor behavior [6]. References: [1] D. R. Miller, et al., Sensors Actuators B Chem., vol. 204, pp. 250�272, 2014. [2] H.-J. Kim and J.-H. Lee, Sensors Actuators B Chem., vol. 192, pp. 607�627, Mar. 2014. [3] S.-W. Choi, et al., J. Mater. Chem. C, vol. 00, pp. 1�7, Jan. 2015. [4] S.-W. Choi, et al., ACS Appl. Mater. Interfaces, vol. 7, no. 1, pp. 647�52, Jan. 2015. [5] G. Korotcenkov and B. K. Cho, Sensors Actuators B Chem., vol. 196, pp. 80�98, Jun. 2014. [6] The authors acknowledge funding from NASA Space Technology Research Fellowship Grant No. NNX13AL72H, and from the Ohio State University Institute of Materials Research (IMRFG). Figure 1. SEM images of nano-heterostructures characterized showing Co3O4 nanoislands decorating SnO2 nanowires (left) and Cr2O3 nanoislands and strips decorating SnO2 nanowires (right).");sQ1[628]=new Array("../7337/1257.pdf","Investigating the Origin of Luminescence in Zinc Oxide Nanostructures With STEM-Cathodoluminescence","","1257 doi:10.1017/S1431927615007072 Paper No. 0628 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Investigating the Origin of Luminescence in Zinc Oxide Nanostructures With STEM-Cathodoluminescence Edward R. White1, Ashley Howkins2, Charlotte K. Williams1, and Milo S. P. Shaffer1 1. 2. Department of Chemistry, Imperial College London, South Kensington, London, SW7 2AZ, UK. Experimental Techniques Centre, Brunel University London, Uxbridge, UB8 3PH, UK. ZnO nanostructures display luminescence in the UV and across the visible spectrum, and show promise as future nanoscale electronic, optoelectronic, and sensing devices [1]. The visible luminescence arises from surface or bulk states at energies inside the ZnO bandgap, however, a fundamental understanding of the luminescent sources is still lacking. The assignment of particular defects to different visible emission peaks is a highly controversial and active area of current research [1,2]. Here, we perform the first spatially resolved scanning transmission electron microscopy cathodoluminescence (STEM-CL) measurements on ZnO nanostructures, and show the emergence of CL spectral peaks associated with morphological changes in ZnO nanorods. Further studies using parallel techniques in the TEM sensitive to intrinsic and extrinsic defects (e.g. HRTEM, atomic resolution HAADF-STEM, EELS, and EDS) will likely conclusively reveal the origin of emission in ZnO and other technologically relevant, luminescent nanostructures. Unlike photoluminescence, cathodoluminescence can probe the emission properties of individual nanostructures since the electron beam can be condensed to nanometer scale. SEM-CL has been used to study defect emission in individual ZnO structures previously, with studies showing, for example, green emission associated with Zn vacancies [2] and a blueshift in the near-band-edge emission with decreasing particle size[3]. However, SEM cannot resolve atomic features to correlate with emission changes, as SEM resolution is limited to a few nanometers. Cathodoluminescence measurements done in the TEM have this unique capability. Here, we use a Gatan Vulcan STEM-CL system to measure changes in cathodoluminescence spectra within individual ZnO nanostructures. Figure 1 shows STEM images and a CL spectrum image acquired across an 800 nm long ZnO nanorod. The spectrum image contains 100 individual spectra, each acquired as the electron beam rasters over an area 8 nm long � the width of the nanorod (50 nm). The emission onset at 375 nm is consistent with the near-band-edge in ZnO [1]. Comparison of the CL spectrum image and the STEM images of the same rod illustrates the dramatic changes in visible emission with morphological changes in the nanorod. The most obvious spectral feature in Figure 1d is the orange emission at ~600 nm. Previously, orange emission has been attributed to surface dislocations, oxygen interstitials, and zinc vacancies; a definitive source remains elusive [1]. Comparing Figures 1c-d it is evident the orange emission is coincident with an increase in HAADF signal. Figure 2 illustrates this more clearly; the amplitude of Gaussians fit to the UV near-band-edge and orange deep-level emissions are plotted as a function of distance along the rod. Other CL spectral peaks also correlate with structural features in the nanorod. There is strong blue emission centered at ~450 nm associated with the tip of the nanorod. There is weak, broad emission centered at ~435 nm in the region with the strong emission at 600 nm. Finally there is weak, but sharp, violet emission at 400 nm, associated with strain in lower portion of the rod, visible in the ADF STEM Microsc. Microanal. 21 (Suppl 3), 2015 1258 image (Figure 1b). Correlating these spectral peaks and morphological changes with further defect investigations promises to reveal the origin of visible luminescence in ZnO nanostructures [4]. References: [1] AB Djurisi and YH Leung, Small 2 (2006), p. 944. [2] F Fabbri et al, Scientific Reports 4 (2014), p. 5158. [3] CW Chen et al, Applied Physics Letters 88 (2006), p. 241905. [4] The authors acknowledge funding from EPSRC grant number EP/K035274/1, and Brunel University and the Experimental Techniques Centre for funding and making available the STEM-CL system, respectively. Figure 1. STEM images and CL spectrum image acquired along the length of a ZnO nanorod. Simultaneously acquired (a) panchromatic CL, (b) ADF, and (c) HAADF images. (d) CL spectrum image of the ZnO nanorod pictured in (a)-(c). Figure 2. Increase in deep-level (DL) emission intensity relative to the near-band-edge (NBE) intensity associated with morphological changes in the ZnO nanorod. (top) Plot of emission intensity vs. distance along the nanorod for the NBE and DL peaks. (bottom) HAADF image of the nanorod. The green dotted line is added as an aid to the reader to show that the increase in the DL emission is coincident with an increase in HAADF signal.");sQ1[629]=new Array("../7337/1259.pdf","Combined EELS and Cathodoluminescence analysis in a STEM microscope of GaN / InGaN quantum wells for LED applications","","1259 doi:10.1017/S1431927615007084 Paper No. 0629 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Combined EELS and Cathodoluminescence analysis in a STEM microscope of GaN / InGaN quantum wells for LED applications P Longo1, DJ Stowe1, RD Twesten1 and A Howkins2 1 2 Gatan Inc., 5794 W Las Positas Blvd, Pleasanton, CA, 94588, USA ETC, Brunel University, Uxbridge, London, UK The correlation between a material's luminescence properties and its nanoscale morphology, microstructure and local chemistry offers great benefit in the understanding of many technologically important materials and devices. This has encouraged a growing interest in performing cathodoluminescence (CL) microscopy at high spatial resolution in a STEM microscope [1,2]. Here, we use combined electron energy loss spectroscopy (EELS) and CL analysis of the GaN / InGaN quantum well (MQW) from a light emitting diode (LED) to investigate the role of In clustering in luminescence efficiency; sub-nanometer compositional information is correlated with the luminescence from individual quantum wells with the MQW. III-nitride semiconductors are technologically important materials with GaN / InGaN MQWs being the source of light emission in current generation blue and white LEDs. However, efficient white LEDs based entirely of III-nitride semiconductors remain elusive due to poor efficiency at green emission wavelengths, the so-called `green-droop'. The reason for low efficiencies at high In content is not well understood; auger electrons fluctuations in quantum well and In composition are two of the many proposed mechanisms [3]. Thus, understanding how the structure, composition and luminescence (intensity and emission wavelength) of MQW structures are correlated is ultimately very important for improving device performance. For the results of this paper, compositional information was obtained from sub-nanometer scale EELS analysis using means of MLLS fitting to extract the composition and the local luminescence was measured simultaneously using CL. The CL light was acquired using miniature elliptical mirrors (solid angle of about 7.3 sr) integrated into the tip of a conventional cryogenic TEM holder. Light is coupled out of the holder through two optical fibres to an optical spectrometer fitted with a PMT and CCD detectors. This combined EELS / CL system offers the advantage of the best in spectral resolution (up to 4 meV), spatial resolution analysis and sensitivity to microstructural changes. Simultaneous EELS / CL data was collected with the sample at -171C minimizing the influence of the electron beam on the sample and increasing the spatial resolution of the Cl data as result of the enhanced rate of radiative recombination within the QWs. The analysis was carried out across the GaN / InGaN MQWs and superlattice layers where each layer is just a couple of nanometers wide. Variations of the luminescence quantum efficiency by more than an order of magnitude was observed between regions separated by only a few nanometers and free extended defects; we analyse how the composition measured by EELS affects the luminescence. Figure 1a shows an ADF STEM image taken during the simultaneous acquisition of the CL and EELS data. The image shows the alternating InGaN bright and GaN dark layers. The interface GaN / InGaN as shown in the ADF STEM image in Figure 1a appears to be fairly abrupt. Figure 1b shows a CL band pass image obtained by integrating the signal at 450 nm in the CL spectrum shown in Figure 1c over a 30nm wavelength window across the entire area in the ADF STEM image in Figure 1. The change in intensity observed in Figure 2 could be caused by variations in the alloy concentration. Local In clustering or change in the composition can easily cause such variations in the CL emission. Change in composition at a subnanometer level has been obtained using EELS and further details of the analysis will be given during the oral presentation. Microsc. Microanal. 21 (Suppl 3), 2015 1260 References: [1] Kim S. K., Brewster M., Qian F., Li Y., Lieber C. M. and Gradecak S., Nanoletters 9, 3940, 2009 [2] Zagonel L. F., Mazzucco S., Tence' M., March K., Bernard R., Laslier B., Jacopin G., Tchernycheva M., Rigutti L., Julien F. H., Songmuang R. and Kociak M., Nanoletters 11, 2011, 568 [3] Yang T.-J., Speck J. S. and Wu Y. �R., Proc. SPIE 8986, Gallium Nitride Materials and Devices IX, 898611 (March 8, 2014) Figures 1: a) Analog STEM Image acquired during the simultaneous acquisition of the EELS and CL data. It clearly shows the alternating InGaN and GaN multilayers. The former is bright whereas the latter dark in the image. The InGaN / GaN interfaces across the entire image appear fairly uniform; b) band pass image obtained by integrating the signal at 450nm in the CL spectrum in Figure 1c over the 435 nm � 465 nm wavelength window. This shows the good spatial resolution that can be obtained using CL in STEM microscope. The change in intensity can be probably caused by variations in the local composition and this can be assessed using EELS; c) CL spectrum extracted from the selected region in the band pass image. The emission band at ~ 450 nm is well pronounced.");sQ1[630]=new Array("../7337/1261.pdf","Low kV High Resolution Scanning Electron Microscopy Study of Si Nanowire Surfaces","","1261 doi:10.1017/S1431927615007096 Paper No. 0630 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Low kV High Resolution Scanning Electron Microscopy Study of Si Nanowire Surfaces C. Alfonso1, L. Roussel1, A. Chara�1, C. Dominici2, A. P. C. Campos2, L.Y. Han3, F. Zhou3 1. Aix-Marseille Universit� � CNRS, IM2NP, Facult� des Sciences, Campus de St J�r�me, F-13397 Marseille, France 2. Aix-Marseille Universit�, CP2M, Facult� des Sciences, Campus de St J�r�me, F-13397 Marseille, France 3. Carl Zeiss Microscopy, GmbH, Carl-Zeiss-Stra�e 22, 73447 Oberkochen, Germany The characterization of small objects like nanometric gold nanocrystals on nanowire (NW) surface allowed us to highlight the improvements of new imaging detectors in SEM. We have employed different detectors for both conventional imaging signals (SE and BSE), coupled with analytical techniques. In previous work [3], saw-tooth faceting and non-homogeneous gold repartition have been characterized by coupling Scanning Transmission Electron Microscopy - High Angle Annular Dark Field (STEM-HAADF), electron tomography and Energy-dispersive X-ray spectroscopy (EDS). This work deals with surface faceting and post-growth repartition of the gold used as catalyst during Molecular Beam Epitaxy (MBE) growth of silicon NWs. Indeed, because of the high surface/volume ratio, NWs surface investigation is fundamental to explain NWs properties and to functionalize them in order to develop new applications as, for example, the detection of chemical or biological species [1, 2]. In this paper, we show that new FESEM at low kV can provide very similar information not only about the surface morphology but also about its composition with high resolution. Fig 1 and 2 show the first results obtained with the Zeiss GeminiSEM 500 ultra-high resolution FESEM, just installed in CP2M Aix-Marseille University. For imaging, this system is completely equipped with In-lens SE and the Energy Selective Backscattered detector (EsB), annular-BF-DF-HAADF-STEM, five quadrant multi-mode solid state BSE, variable pressure (VP) SE and VPBSE detectors. For the chemical and crystallographic analyses, EDS, WDS, and EBSD accessories are additionally installed on this microscope achieving the full imaging and analytical capabilities. Fig 1 shows clearly the transition to six to twelve faces from the bottom to the top of the NW. Moreover, the saw-tooth faceting of one over two faces is also visible with a high resolution. This finer faceting is generally difficult to resolve with common SEMs. In Fig 2, because of filtered in-lens BSE imaging at low kV, a strong chemical contrast is achieved. Gold particles appear bright. It is now possible not only to distinguish the catalyst on the NW but also to resolve nanometric gold nanocrystals on the different surfaces and to analyze their repartition. Such a resolution, coupled to chemical sensitivity, was only possible, up to now, with STEM-HAADF. But, SEM analysis has the advantage of not having any previous preparation such as nanowire dispersion and/or thin foil preparation. Moreover, SEM provides a larger field of view and a better statistics in NW characterization. The low voltage acquisition was adequate for our surface studies since it presents less delocalisation. Microsc. Microanal. 21 (Suppl 3), 2015 1262 More studies are under development to compare analytical performance of this new SEM with the TEM one. At this stage, we believe this new equipment as an indispensable tool to investigate objects at the mesoscopic scale. There is a huge application potential to investigate nano-objects at different scales without destructive manipulation. References: [1] J. Wang, Biomolecule-functionalized nanowires: From nanosensors to nanocarriers, Chem. Phys. Chem. 10 (11) (2009) 1748-1755. [2] G-S. Park et al., Full Surface Embedding of Gold Clusters on Silicon Nanowires for Efficient Capture and Photothermal Therapy of Circulating Tumor Cells, Nano Lett. 12, (2012) 1638-1642 [3] T. David et al., Gold coverage and faceting of MBE grown silicon nanowires, J. Cryst. Growth 383 (2013) 151-157 Figure 1. In-lens SE detector image of a Si NW clearly showing different facets and its fine structure on the surface. The surface sensitive imaging at 1.5 kV reveals highly resolved crystalline facets on the Si NW. Figure 2. EsB detector image of a Si NW, unveiling a gold catalytic particle on top of the pillar and small gold nanoparticles on the surface. High resolution compositional imaging is only possible using the filtered in-lens BSE imaging at low kV. Both pictures shown in Fig.1 and 2 are acquired in parallel and at low kV.");sQ1[631]=new Array("../7337/1263.pdf","An Analytical Scattering Model for Low Energy Annular Dark Field Transmission Scanning Electron Microscopy","","1263 doi:10.1017/S1431927615007102 Paper No. 0631 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 An Analytical Scattering Model for Low Energy Annular Dark Field Transmission Scanning Electron Microscopy Taylor Woehl,1 Jason Holm,1 and Robert Keller1 1. Material Measurement Lab, NIST, Boulder, CO 80305 Scanning transmission electron microscopy (STEM) is a powerful tool for chemical and structural analysis of materials on the nano- and atomic-scale. Often the atomic scale resolution afforded by aberration corrected STEM is not needed, and in some cases the high beam currents and electron energies used can damage composite nanomaterials containing organic components. A relatively new approach for STEM imaging when atomic resolution is not required is STEM imaging in a scanning electron microscopy (SEM) equipped with a transmission detector (t-SEM) [1-3]. This approach has the advantage of being a comparatively easy to use and low cost solution for performing STEM imaging with a resolution of several nanometers. While the increased amount of scattering by the lower-energy electrons (typically 30 keV) can increase signals for imaging and spectroscopy, it also leads to artifacts in samples on thick substrates, such as contrast inversion, low signal-to-noise, and convoluted image contrast [1-3]. Monte Carlo simulations have predicted contrast of annular dark field (ADF) t-SEM images of nanoparticles on thick layers of carbon support material [1]; however, the models did not give much insight into the physical mechanisms leading to the contrast. An analytical model to predict image contrast would help to identify imaging conditions that are void of imaging artifacts and optimize contrast and resolution. Here we employ an analytical electron scattering model to show that for decreasing collection angles, the decrease in ADF t-SEM contrast of gold nanoparticles on carbon films is consistent with increased thickness contrast from the carbon substrate at low scattering angles. Images of gold nanoparticles on lacey carbon TEM grids were acquired in an SEM (30 keV) with an ADF detector masked by an annular aperture to define inner and outer collection angles. We varied the distance between the sample and detector to systematically vary the ADF collection angles and found that the contrast of the gold nanoparticles decreased as a function of the ADF inner collection angle (Figure 1a). Nanoparticle contrast (Figure 1b) was lower overall for particles on the thicker lacey carbon support. To model the experimental contrast, we developed an analytical electron scattering model that includes collection angle-dependent elastic inelastic, and thermal diffuse electron scattering as well as geometric considerations for the ADF detector. We calculated the contrast by taking the difference between the number of electrons transmitted through the nanoparticle and carbon substrate to a solid angle defined by the inner and outer collection angles. The theoretical contrast was compared to the experimental images using the carbon support thickness as an adjustable parameter, as we could not accurately determine this thickness. With this fitting parameter, there was good agreement between the model predictions and experimental contrast (Figure 1c). The model indicated that the decrease in contrast with decreasing collection angle was due to increased thickness contrast from the carbon support, relative to the atomic number contrast of the gold nanoparticles. The discrepancy in the carbon support thickness Microsc. Microanal. 21 (Suppl 3), 2015 1264 is likely due to multiple scattering in the nanoparticles. We expect this analytical scattering model will be important for determining artifact-free imaging conditions for ADF t-SEM imaging [5]. References: [1] N. Brodusch, H. Demers, R. Gauvin, Microscopy and Microanalysis, 19 (2013) 1688-1697. [2] V. Morandi, P.G. Merli, Journal of Applied Physics 101 (2007), 114917. [3] B. Patel, M. Watanabe, Microscopy and Microanalysis 20 (2014), 124-132. [4] T. Klein, E. Buhr, C.G. Frase, in: P.W. Hawkes (Ed.) Advances in Imaging and Electron Physics, 171 (2012), 297-356. [5] T. Woehl and J. Holm acknowledge funding from the National Research Council Postdoctoral Research Associateship Program. Contribution of the U.S. Department of Commerce; not subject to copyright in the United States. Figure 1. (a) Experimental ADF t-SEM images of 30 nm gold nanoparticles on a lacey carbon TEM grid at various inner collection angles. (b) Method for experimental measurement of the contrast of gold nanoparticles on the TEM grid. The contrast is equal to the difference of the mean intensity inside the full width half maximum ( of the nanoparticle and the background intensity, divided by the total intensity. (c) Experimental and theoretical contrast for nanoparticles on thin and lacey carbon. Each data point is the mean contrast of 5 nanoparticles. Error bars represent two standard deviations of the mean. The black lines are the theoretically determined contrast using substrate thicknesses of 85 and 300 nm for the thin and lacey carbon, respectively.");sQ1[632]=new Array("../7337/1265.pdf","Detecting Localized Variation of Chemistry via Atomic-Resolution Secondary Electron Imaging","","1265 doi:10.1017/S1431927615007114 Paper No. 0632 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Detecting Localized Variation of Chemistry via Atomic-Resolution Secondary Electron Imaging Jane Howe1, Yoichiro Hashimoto2, Xue Wang3, Madeline Vara3, David Hoyle1, Tom Schamp4, Younan Xia3, and David Joy5,6 1. 2. Hitachi High-Technologies Canada Inc, Toronto, Canada. Science System Design Division, Hitachi High-Technologies Co., Hitachinaka, Ibaraki, Japan. 3. The School of Chemistry and Biochemistry, Georgia Institute of Technology, Atlanta, USA. 4. Hitachi High-Technologies America Inc, Gaithersburg, USA. 5. Department of Materials Science and Engineering, University of Tennessee, Knoxville, USA. 6. Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, USA. Secondary electron microscopy (SEM) has been widely used for characterization of a broad range of materials. Conventional SEM operates at an accelerating voltage ranging from a few hundred to 30 kV, commonly using secondary electrons (SE) for imaging. Secondary electrons, defined with an energy of less than 50 eV, are assumed to result from inelastic scattering between electrons. The energy loss is small and the SE could be generated a few nanometers away from the interaction. And both the generation and emission of SE are considered delocalized events [2-3]. In 1980, Pennycook and Howie reported that a high fraction of the energy loss in thin foils occurred close to atomic sites and resulted in the production of secondary electrons [4]. They further predicted that "ejected secondary electrons would show a localized or crystallographic effect". SE imaging at high voltages up to 300 kV can also be obtained in a scanning transmission electron microscope (STEM) equipped with an SE detector. Thanks to recent advances in aberration corrected STEM, Zhu and colleagues demonstrated atomicresolution SE imaging of crystals and single atoms at 200 kV [1]. This result supports Pennycook and Howie's earlier predication. Based on the current understanding that generation and emission of secondary electrons is both a localized and delocalized event, David Joy proposed the hypothesis that layer-by-layer SE imaging would reveal localized variation of chemistry and other defects in a crystal [5]. In order to verify Joy's hypothesis, we carry out SE imaging of a series of nanocrystals with well-defined chemical variation and defects. The atomic-resolution SE imaging has been undertaken using a Cs-corrected Hitachi HD2700A STEM equipped with a secondary electron detector. In Fig. 1, we present a set of SE and annular dark field (ADF) images taken from Pd@Pt core-shell cubes at 200 kV. The outer ~3 atomic layers of these 18-nm cubes are Pt, as revealed by the contrast as well as the EDS mapping. Comparing the SE image of Fig. 1b and the ADF image of Fig. 1c, the SE image revealed less contrast difference of Pt versus Pd than that of the ADF. In bright-field TEM study, image contrast changes as a function of focal length and Cs. In this study, we strive to carry out a through-Cs series of SE imaging. Particularly, we are going to explore whether we can obtain atomic-resolution SE images under negative Cs. As Zhu et al. suggested that secondary electrons produced by energy-loss events in the several keV range, such as innershell ionizations, as being the enabling step for localization [1], it is reasonable to speculate that localized SE event would be diminished if the primary beam energy is too low to produce energy-loss events in the several keV range. What is the cut-off energy of a certain element? Let us see if we can find it out. Microsc. Microanal. 21 (Suppl 3), 2015 1266 References: [1] Y Zhu et al, Nature Materials 8 (2009) 808. [2] H Salow Phys. Z. 41 (1940) 434. [3] H Bethe Phys. Rev 59 (1941) 940. [4] S J Pennycook and A Howie, Phil. Mag. Lett. 41 (1980) 809. [5] D C Joy, Nature Materials 8 (2009) 776. Fig. 1. SE and ADF images of Pd@Pt core-shell particles imaged at 200 kV. Scale bar for (b) and (c) is 5 nm.");sQ1[633]=new Array("../7337/1267.pdf","Sub-surface Serial Block Face Scanning Electron Microscopy","","1267 doi:10.1017/S1431927615007126 Paper No. 0633 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Sub-surface Serial Block Face Scanning Electron Microscopy Q. He1, M.A. Aronova1, D.C. Joy2,3, G. Zhang1, R.D. Leapman1 1 2 National Institute of Biomedical Imaging and Bioengineering, NIH, Bethesda, MD 20892; Department of Materials Science and Engineering, University of Tennessee, Knoxville, TN 37996; 3 Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, TN 37831 Serial block face (SBF) imaging in the scanning electron microscope (SEM) provides nanoscale 3D ultrastructure of biological samples up to several hundred micrometers in size. In SBF-SEM, an ultramicrotome built into the SEM specimen stage successively removes thin sections from a plasticembedded, heavy metal-stained specimen. After each cut, the freshly exposed block face is imaged at a low incident electron energy using the backscattered electron signal, which is sensitive to heavy atoms in the sample. Although the x-y resolution in the plane of the block face is approximately 5 nm, the resolution along the z-axis in SBF-SEM is limited by the minimum slice thickness of around 25 nm. We have explored the feasibility of improving the z-resolution in SBF-SEM by recording images at more than one primary beam energy, thus sampling different depths below the block surface [1, 2]. An illustration of the relationship between primary beam energy and penetration depth is shown in Fig. 1A. We used Monte Carlo simulations of SEM images from an epoxy block containing 5-nm diameter carbon spheres stained with 14% lead positioned at different depths, as a model for small biological structures within cells (Fig. 1B) [3]. A linear relationship was found between the depth of the spheres and the ratio of backscattered images recorded at primary beam energies of 1.4 keV and 6.8 keV, which allowed us to determine 3D structure within a 25-nm surface layer with a depth resolution of around 5 nm. The model obtained from the simulation was tested experimentally on a specimen consisting of 5-nm gold spheres embedded in a multilayer carbon film using a Zeiss Sigma-VP SEM equipped with a Gatan 3View SBF system. BSE images recorded at primary beam energies of 2.4 keV and 6.8 keV are shown in Fig. 2A and 2B. The computed 3D model is visualized in Figs. 2C and 2D, along with 5 nm calculated slices at depths of 0 nm, 5 nm, 10 nm, 15 nm, 20 nm, and 25 nm (Figs. D-I). Experiments were also performed on embedded blocks of stained biological tissues to demonstrate that 3D structure could be determined from 25-nm thick sub-surface layer as shown in Fig. 3, where the heavy-atom stain appears bright in the BSE images. Although damage of the block under electron irradiation limits the signal to noise ratio, sub-surface serial block face SEM using multiple primary beam energies provides the possibility of obtaining nearly isotropic 3D spatial resolution [4]. References [1] P Hennig and W Denk, Journal of Applied Physics 102 (2007), p.123101-1. [2] F Boughorbel et al, FEI Compnay, US Patent (#US8,232,523 B2) 2011. [3] H Demers et al, Scanning 33 (2011), p.135. [4] This research was supported by the intramural program of the National Institute of Biomedical Imaging and Bioengineering, NIH. Microsc. Microanal. 21 (Suppl 3), 2015 1268 Fig. 1. (A) Relationship between primary beam energy and penetration depth in a resin-embedded block of stained cell: higher beam energy E3 leads to deeper penetration than lower energies of E2 and E1. (B) Monte Carlo simulation of stained spheres containing 14% Pb in an epoxy resin matrix; upper image, 1.4 keV beam energy; lower image, 6.8 keV beam energy. Fig. 2. BSE images of specimen consisting of 5-nm diameter Au nanoparticles embedded in carbon film, acquired at primary beam energies of (A) 2.4 keV and (B) 6.8 keV; (C) calculated 3D model in top view, and (D) in side view. Panels E-I are computed 5-nm thick slices at depths of (E) 0-5 nm, (F) 5-10 nm, (G) 10-15 nm, (H) 15-20 nm and (I) 20-25 nm. Fig. 3. BSE images of sub-region of beta cell in pancreatic islet of Langerhans acquired at different primary energies of (A) 1.4 keV, (B) 2.4 keV, and (C) 3.4 keV. Differences in appearance of membrane structure (in box) are due to variations in stain as a function of depth.");sQ1[634]=new Array("../7337/1269.pdf","Applications of energy dependent backscatter yield variations at low voltage.","","1269 doi:10.1017/S1431927615007138 Paper No. 0634 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Applications of energy dependent backscatter yield variations at low voltage. Markus Boese1, Kerstin Sempf2, Fang Zhou1 and Alexander Thesen1 1. 2. Zeiss Microscopy, SEM Department, Oberkochen, Germany. Fraunhofer- IKTS, Ceramography and Phase Analysis, Dresden, Germany For imaging of nano-materials and composites it is often necessary to detect smallest compositional differences in these materials. Very often the nanostructure needs to be distinguishable from matrix or substrates in order to image it. The main advantage of material analysis at low Voltages is due to the reduced interaction volume of the primary beam. Benefit is here a higher resolution and the ability to gain surface sensitivity for better imaging conditions. It is known, that under certain low Voltage conditions a contrast reversal occurs, showing high Z materials with dark contrast and light materials with bright contrast in the BSE image. In this study we investigate, how the change of the backscatter coefficient with changing primary energy can be utilized for gaining information in SEM backscattered imaging. In particular, if this can be used to gain material contrast for low voltage backscattered imaging compared to imaging conditions above 1-2 kV. The primary energy dependency of the backscatter coefficient was described by Cazaux [1] using the empirical expression: = a (1 � e-b Eo) The sign is + for low Z and � for the high Z elements. The parameters "a" and "b" were fitted to experimental values and a resulting characteristic is given for a range of different elements Z range from 4 to 80. The different slope for different elements is resulting from the different b values, whereas the a value is corresponding to the asymptotic value for higher Eo (>2-4 kV) [1]. Figure 1 shows the energy dependency of the yield, plotted for high Z and low Z elements. In general backscattering yield becomes lower for heavier elements going to low voltage and higher for lighter elements. The shape of the curve depends strongly on "b". The discontinuous change of this parameter (as shown in Fig.2) enables us to produce some extra contrast for certain element combinations under low voltage imaging conditions. We can expect a raise in contrast up to 40% for imaging Ag and Zn at 300V (Fig. 1) compared to 10% contrast for primary energies higher than 2kV. (Note at high energy the backscatter yield is solely dependent on the "a" parameter) For low Z the parameter "b" changes as well for different elements. But the contrast (difference between the Al and O yield in Fig. 1) is not changing as much. Here the advantage of low voltage applications is the higher signal itself due to the higher yields at low Voltage. Experiments are carried out in a ZEISS Ultra 55 SEM equipped with an EsB detector capable of BSE detection in a sub 1kV regime. The GEMINI lens design enables the detection of backscattered electrons passing through the objective lens and then hit the EsB Detector positioned above the Inlens SE detector. A grid repels secondary electrons and can be used for energy filtered SEM Imaging [2]. Fig. 3 illustrates the low voltage imaging capabilities of this system. Microsc. Microanal. 21 (Suppl 3), 2015 1270 Another example is given by a ceramic composite sample consisting of high Z and low Z phases. As shown in Fig 4 the low voltage BSE image is suitable to differentiate 5 different phases due to the leveling of the individual backscattering signal for high and low Z. Further applications for the contrast enhancement at low voltage could be done on material systems like InGaAs, and Pb free solder containing different phases of Cu, Ag and Sn. References: [1] J. Cazaux, J. Appl. Phys 112 (2012), p. 84905. [2] H. Jaksch, EMC 2008 proceedings (2008), 1, 555 Figure 1. Energy dependency of backscatter yield for high and low Z materials. The different slope of the curves can be utilized to gain contrast at low voltage imaging conditions Figure 2. Empirical parameter "b" plotted for different elements. The strong variation here causes the different slopes in Fig. 1 Figure 3. For different primary energies (left image Eo=1600V and right image 500V) a contrast inversion is observed. Sample: tungsten carbide with cobalt sintering agent (in between the grains). The slight contrast variation within the tungsten carbide is due to channeling. Figure 4. 5 different phases are distinguishable under low voltage conditions. Phases are: SiC, Si3N4, Y-Al-Si-oxinitride, MoSi2, Mo5Si2");sQ1[635]=new Array("../7337/1271.pdf","Correlative Ultrastructural Analysis of Functionally Modulated Synapses Using Automatic Tape-Collecting Ultramicrotome - SEM Array Tomography.","","1271 doi:10.1017/S143192761500714X Paper No. 0635 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Correlative Ultrastructural Analysis of Functionally Modulated Synapses Using Automatic Tape-Collecting Ultramicrotome - SEM Array Tomography. Naomi Kamasawa1, Ye Sun2, Takayasu Mikuni2, Debbie Guerrero-Given1, Ryohei Yasuda2 1 2 Electron Microscopy Core Facility, Max Planck Florida Institute for Neuroscience, Jupiter, USA Neuronal Signal Transduction, Max Planck Florida Institute for Neuroscience, Jupiter, USA To understand synaptic plasticity, correlative microscopic approaches using functional imaging by light microscopy combined with electron microscopy (EM) are necessary. Ultrastructure of synapses can be visualized only by EM, but re-finding the same synapses that were manipulated and imaged by light microscope in the EM is a very laborious and time-consuming task when performing the conventional serial ultrathin-sectioning and observation method with transmission EM. Recently, new imaging applications to obtain volume data sets from brain tissues were introduced using an automatic tapecollecting microtome (ATUM) [1] or a focused ion beam [2] with scanning electron microscopy (SEM). We used an ATUMtome (RMC/Boeckeler) and automatic array tomography software, Atlas AT5 (Zeiss) and established a workflow to investigate ultrastructural changes of synapses. We first performed glutamate uncaging by 2-photon laser microscope to induce/modify spines, and obtained the whole cell image with a confocal microscope. Then the tissues are prepared with conventional pre-embedding immuno-EM using nanogold labeling for EM visualization as explained below. 2-photon glutamate uncaging and confocal laser scanning microscopy (CLSM) imaging: We uncaged MNI-glutamate with 4 mW (720 nm, 2 photon), 2 Hz X 40 stimulation on smooth dendrites of pyramidal neurons expressing EGFP in mouse hippocampal organotypic slice cultures. After the new spine formation visually confirmed, tissue slices were fixed in a mixture of 4% paraformaldehyde and 0.5-1% glutaraldehyde. CLSM Z-stack images were captured with a lower magnification to overview the whole tissue slice containing GFP fluorescent cells. Once the target neuron was identified, it was imaged to visualize its dendritic arbor and to define the position of the newly developed spines. Immuno-staining and resin embedding: Slices were cryo-protected with sucrose, permeabilized by liquid nitrogen, blocked and immunostained with anti-GFP antibody followed by nanogold conjugated secondary antibody. Slices were then silver enhanced, post-fixed with OsO4, en-block stained with uranyl acetate, dehydrated with a series of ethanol and acetone, and embedded in Durcupan resin. ATUM and SEM imaging with Atlas AT5/ Merlin VP Compact (Zeiss): The area containing the target cell was trimmed based on the CLSM image to a size of 1 x 1.5 mm. The sample was sectioned at 60-70 nm thickness with an ATUMtome and collected on Kapton tape automatically, then the kapton tape was placed on silicon wafers. The overview image of the Si wafer was captured using a digital camera, the image was loaded into the Atlas AT5 and aligned to the physical location of the wafer in the SEM using mesh grids as fiducials on the wafer. Low-resolution SE2 images for all sections were first captured with a 2 �m/pixel xy-resolution protocol (4 mm2 imaging area) and overlaid on the overview image. An approximate area of the target cell was demarcated on the serial sections, and montage images were captured with a 30 nm pixel resolution protocol using BSE. After finding the target cell, a rough 3D reconstruction was performed to see the orientation of the target dendrite and spine. The high-resolution images of the target spine were captured with a 4-7 nm pixel resolution protocol using an InlensDuo detector in the BSE mode. Microsc. Microanal. 21 (Suppl 3), 2015 1272 Summary: This workflow allowed analyzing morphologies of synaptic plasticity from 2-photon microscope to EM. By using a CLSM fluorescent Z-stack image as a bridge, immunogold labeled profiles that were functionally modulated and imaged by 2-photon microscope were relatively easily relocated without adding a fiducial- marking within the tissue. Three different detectors were used to fit to the best performance of imaging time and quality at each step of the workflow. It is still timeconsuming, but could be handled routinely and useful to obtain quantitative data sets. Figure 1. Correlative imaging workflow, A) 2-photon microscope image of a newly synthesized spine (de novo)(arrow) by glutamate uncaging, B) CLSM image of the cultured hippocampal slice containing GFP expressing neurons, C) enlarged image of a pyramidal neuron in the boxed area of (B), D) the same neuron immuno-stained, silver enhanced and embedded in a resin, E) ultrathin sectioning in ATUMtome collecting sections on the tape, F) a screen shot of the Atlas AT5. The first set of low resolution SE2 images were overlaid on the Si wafer image, G) 3x3 tiling BSE images with a 30 nm/pixel resolution to find target profiles, H) enlarged image of the boxed area of (G) showing immunogold labeled neuronal soma and a part of its dendrites, I) roughly reconstructed 3D image of the target neuron, J) example of an immunogold-labeled spine taken by the InlensDuo detector with a 5 nm/pixel resolution. References: [1] KJ Hayworth, JL Morgan, R Schalek, DR Berger, DG Hildebrand and JW Lichtman, Front Neural Circuit. 8 (2014), p. 1. [2] B Maco, A Holtmaat, A Jorstad, P Fua, and GW Knott in "Correlative light and electron microscopy II", ed. T Muller-Reichert and P Verkade, (Elsevier), p. 339 [3] The authors thank the specialists of Applications and Services from RMC/Boeckeler, Fibics and Carl Zeiss Microscopy for their insensitive supports.");sQ1[636]=new Array("../7337/1273.pdf","Conjugate Immunofluorescence � SEM Array Tomography for Studying Synapses and Other Subcellular Structures in the Brain.","","1273 doi:10.1017/S1431927615007151 Paper No. 0636 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Conjugate Immunofluorescence � SEM Array Tomography for Studying Synapses and Other Subcellular Structures in the Brain. Kristina D. Micheva1 1. Stanford University School of Medicine, Department of Molecular and Cellular Physiology, Stanford, CA, USA Most mammalian synapses are small structures, less than 1 �m in diameter, they are tightly packed in the brain and have diverse molecular composition. This makes them a difficult target for studying with a single imaging modality. Electron microscopy is the method of choice for identifying individual synapses and observing their ultrastructure, however it can provide only limited molecular information. Immunofluorescence has the advantage of multiple antibodies use, but cannot reliably distinguish individual synapses within brain tissue. We developed conjugate immunofluorescence � SEM array tomography which allows the imaging of both the molecular content and the ultrastructure of tissues. The method is based on physical ultrathin serial sectioning, immunostaining and acquiring fluorescence and electron microscopy images of resin embedded tissues, followed by computational volume reconstruction and analysis. Tissue preparation and imaging. Adult mice were perfusion fixed with 2% glutaraldehyde and 2% formaldehyde in phosphate buffer. Vibratome sections (200 �m) were cryoprotected with glycerol, freeze-substituted and embedded in Lowicryl HM-20 [1]. Serial 70 nm sections were collected on carbon-coated glass coverslips. Sections were processed for standard indirect immunofluorescence and imaged on an automated epi-fluorescent microscope (Zeiss AxioImager Z1) using a 63x Plan Apochromat 1.4 N.A. oil objective [2]. After immunofluorescence imaging, ribbons were poststained with uranyl acetate and lead citrate and imaged on a Zeiss Sigma Field Emission SEM using the backscatter detector. Image registration (between different fluorescence imaging sessions, and between fluorescent and electron microscopy mages) and alignment (within the stack of serial images) was done using the open source software FIJI. Fluorescence intensity measurements were performed using FIJI. Results. Osmium-free freeze-substitution embedding in Lowicryl HM-20 provides excellent ultrastructure, allowing the identification of synapses based on the presence of synaptic vesicles and postsynaptic densities. The majority of antibodies previously validated on formaldehyde-fixed, LR White embedded tissue, can be successfully used. Furthermore, additional antibodies, for example against small molecules (GABA), requiring the presence of glutaraldehyde in the fixative, can also be applied. Summary. Cryoembedding in Lowicryl HM-20 preserves both antigenicity and ultrastructure of brain tissue, enabling conjugate array tomography. Using fluorescence microscopy, dozens of different antibodies and other fluorescent markers can be imaged at synaptic resolution within large volumes of tissue. The same arrays can also be viewed using field emission SEM, which allows for the classic benefits of electron microscopy. Such integration of different imaging modalities can substantially enrich our understanding of the brain. Microsc. Microanal. 21 (Suppl 3), 2015 1274 Figure 1. Immunofluorescence and ultrastructure of a GABAergic and glutamatergic synapse in adult mouse neocortex. GABA, blue, gephyrin, yellow, VGlut1, cyan and PSD95, red. Scale bar, 0.2 �m. Figure 2. A myelinated GABAergic axon making a synaptic contact on a neuronal cell body in the adult mouse neocortex. The two images on the left are volume reconstructions from 40 serial sections, 70 nm each. GABA, red, myelin basic protein (MBP), white, tubulin, green, and DAPI, blue. Scale bar, 5 �m. The SEM image to the right shows the same axon making a synaptic contact on the cell body. GABA, red. Scale bar, 0.5 �m. References: [1] JD Ding, MB Kennedy and RJ Weinberg, J. Comp. Neurol. 521 (2013), p.3570. [2] KD Micheva, B Busse, NC Weiler, N O'Rourke and SJ Smith, Neuron 68 (2010), p.639. [3] The author wishes to thank Stephen J Smith, Richard J. Weinberg, JoAnn Buchanan and Forrest Collman for their invaluable scientific input and continuing support, and Kristen Phend and Nafisa Ghori for tissue preparation. This work was supported by grants from the National Institutes of Health (R21MH099797, R01NS075252, R01NS077601 to SJS, R01NS039444 to RJW).");sQ1[637]=new Array("../7337/1275.pdf","Protocol Development for 3D Reconstructions: Combining In Vivo Imaging, APEX2 and En Bloc Staining of Mouse Visual Cortex.","","1275 doi:10.1017/S1431927615007163 Paper No. 0637 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Protocol Development for 3D Reconstructions: Combining In Vivo Imaging, APEX2 and En Bloc Staining of Mouse Visual Cortex. JoAnn Buchanan1, Marc Takeno1, Tanya Daigle1, Adam Bleckert1 Daniel Bumbarger1, Agnes Bodor1, Derrick Brittain1, Ali Cetin1 and Nuno Ma�arico da Costa1. 1. Allen Institute for Brain Science, Seattle, WA, USA. The emerging discipline of connectomics has fueled interest in the development of methods for high throughput 3D electron microscopy methods for large volumes of brain. These methods require tissue processing methods that produce sharply delineated membranes with sufficient contrast to identify them in consecutive serial sections and allow manual and automated segmentation of the data. In order to facilitate this reconstruction and aid the identification of different cell types, we are using the genetically encoded EM markers APEX and APEX2 [1,2], to trace individual neurons, in combination with deep penetration en bloc staining to allow reconstruction of entire circuits. These aims are directed at reaching our final goal of combining multiphoton in vivo imaging of Ca2+ activity from the cortex of an awake mouse, followed by large-scale serial sectioning transmission electron microscopy (ssTEM) reconstruction of 1 mm3 from the same imaged area. To achieve this goal, we will begin with a sparse reconstruction of APEX2 [1] labeled neurons. We have expressed this engineered peroxidase label in vivo in a Cre-dependent manner to allow correlated LM and EM. Following in vivo imaging, the tissue is fixed and the APEX2 reacted with DAB and H2O2 to produce an electron-dense precipitate that can be seen in vibratome sections. The sections are processed for EM and 40 nm serial sections cut and imaged. The APEX2 label facilitates serial section image registration and alignment. Figure 1 shows correlative imaging of virus-mediated direct expression of APEX peroxidase. Figure 2 shows virus-mediated recombinant APEX2 peroxidase expression in a transgenic (Rorb-IRES2-Cre) Cre-dependent mouse line. Our long-term goal is to prepare EM samples for a large-scale, dense reconstruction of a block of the mouse visual cortex, where activity was previously recorded by multiphoton Ca2+ imaging in an awake, behaving mouse. This requires tissue with high en bloc contrast and tissue preservation throughout the depth of the block. We are testing protocols that include optimizing reduced osmium (RO) and osmium bridging (OTO) for contrast enhancement with techniques to improve penetration in brain samples up to 1.5 mm3. In our hands, the reduced osmium OTO (osmium ferricyanide-thiocarbohydrazide-osmium ferricyanide) provides good contrast, but leads to uneven penetration and infiltration problems. To circumvent these issues, we are examining resin formulations suggested by a previous study of whole mouse brain [3] as well as implementing both ultrasonic and microwave assisted osmium staining. Because we will cut up to 30,000 ultrathin serial sections, it is critical that sections have adequate contrast and that resin be uniformly polymerized throughout to facilitate serial sectioning. [1] S. Lam et al., Nature Methods , 12 (2015) p. 51. [2] J. Martell et al., Nature Biotechnology, 30 (2012) p. 1243. [3] S. Mikula, J. Binding, and W. Denk. Nature Methods, 9 (2012) p. 1198. Microsc. Microanal. 21 (Suppl 3), 2015 1276 [4] The authors are grateful to the Allen Institute founders, P.G. Allen and J. Allen, for their vision, encouragement, and support. Figure 1. APEX labelled neuron in layer 4 of mouse visual cortex. (A) Rectangle outlines the neuron and APEX reaction product filing the cell (arrows) in a 60 �m vibratome resin embedded slice by LM. Diamond boxed area shown in panel B. Blood vessel and pial surface of the brain are evident. (B) Electron micrograph showing the cell body (CB) and apical dendrite, filled with electron-dense reaction product. Arrow points to boxed area shown in C. (C) High magnification shows APEX labelled dendrite (arrow) making a synapse. Figure 2. Cre-dependent expression of APEX2 in the transgenic line Rorb-IRES2-Cre in (A) dendritic spines, and (B) boutons.");sQ1[638]=new Array("../7337/1277.pdf","Advanced substrate holder and multi-axis manipulation tool for ultramicrotomy","","1277 doi:10.1017/S1431927615007175 Paper No. 0638 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Advanced substrate holder and multi-axis manipulation tool for ultramicrotomy Waldemar Spomer1,2, Andreas Hofmann1,2, Irene Wacker2,3, Len Ness4, Pat Brey4, Rasmus R. Schr�der2,5 and Ulrich Gengenbach1,2 Institute for Applied Computer Science, Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany HEiKA, Heidelberg Karlsruhe Research Partnership, Heidelberg, Karlsruhe, Germany 3 Cryo EM, Centre for Advanced Materials, Universit�t Heidelberg, Heidelberg, Germany 4 RMC Boeckeler, Tucson, Arizona, USA 5 Cryo EM, CellNetworks, BioQuant, Universit�tsklinikum Heidelberg, Heidelberg, Germany 2 1 Array tomography (AT) allows reconstructing ultrastructure of large volumes at nano-scale resolution. This nondestructive method preserves sectioned samples and thus enables correlative imaging: Different imaging techniques, such as light microscopy (LM) and electron microscopy (EM), are applied on the same sample [1]. Samples are generated by cutting ultrathin sections and collecting them as ribbons swimming on the surface of a water filled knife trough. The substrate needs to be conductive (for EM) and transparent (for LM) such as indium tin oxide (ITO) coated glass coverslips. While sectioning is done by the ultramicrotome the collecting step is still a tedious, manual handling process and requires a human operator. Typically the substrate is held in one hand while the other manipulates the ribbon. Holding the substrate by hand may be affected by tremor � a major obstacle for perfectly aligned ribbons. Furthermore it is nearly impossible to deposit more than one ribbon on the substrate. There are several concepts to support the operator in this task. Beginning from the 1960s [2] until now [3] devices holding the substrate were presented from time to time. However they do not offer the required degrees of freedom to fit all desired knife orientations. The block-face of the sample mounted in the sample holder is very rarely in parallel with the knife edge. That is a reason why nowadays microtomes offer a yaw axis for knife rotation to produce usable sections right from the beginning of the sectioning process. While the knife is rotated a substrate held by a fixed holder remains in its orientation with respect to the microtome. Hence substrates using the full width of the knife trough may collide with the side of the trough if the knife is being realigned. Thus an adaptation for rotated knifes is a desirable feature of a substrate holder. Another important aspect is the lift-up movement of the substrate from the trough once the sections have been pinned to its surface. Depending on the type of sections and the substrate surface properties a lift-up trajectory involving several axes simultaneously reduces ribbon damage. The substrate holders proposed so far fall short on these two aspects. To increase reliability and flexibility in section collection we developed a novel, multi-axis device, which supports the operator in transferring the sections onto a solid substrate such as ITO glass or silicon wafer respectively. With its seven degrees of freedom (Fig. 1A) it enables positioning the substrate in different orientations (lower axes) as well as different lift-up movements (upper axes, Fig. 1B). Moreover, lift-up involving two axes at the same time enables smooth movements with minimal water turbulence and thus avoids ribbon drift and ribbon breaking. The substrate holder provides two axes (x/y) for coarse adjustment and a rotary stage for the rotation about the z-axis to fit rotated knifes. For finer positioning and height adjustment there are micro-positioning stages. With the device presented here thousands of ribbons have been collected even on challenging substrates with rough surface such as ITO-coated glass (Fig. 1C, cf. Wacker et al., this conference). Microsc. Microanal. 21 (Suppl 3), 2015 1278 References: [1] KD Micheva, DJ Smith, Neuron 55 (2007), p. 25-36. [2] O Behnke, J Rostgaard, Biotechnic & Histochemistry 39 (1964), p. 205-208 [3] H Horstmann et al., PLoS ONE 7 (2012), p. e35172 [4] The authors acknowledge C Bartels, L Veith and R Scharnowell for technical support, as well as the Heidelberg Karlsruhe Research Partnership (HEiKA) for initial funding. A B C 10 mm Figure 1: Substrate holder mounted in front of an ultramicrotome with seven adjustable axes for positioning (blue arrows) and lift up (red arrows) (A); Side view: Variable lift-up movements (B); Six ribbons deposited on an ITO-coated glass coverslip using the device presented here (cf. Wacker et al., this conference) (C)");sQ1[639]=new Array("../7337/1279.pdf","Complementary Approaches to Dissecting Mechanisms of Protein-mediated Membrane Fusion","","1279 doi:10.1017/S1431927615007187 Paper No. 0639 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Complementary Approaches to Dissecting Mechanisms of Protein-mediated Membrane Fusion Long Gui1,2, Natalie K. Garcia1, Miklos Guttman1, Jamie L. Ebner1, and Kelly K. Lee 1,2 1 2 Department of Medicinal Chemistry, University of Washington, Seattle, WA, USA Biological Physics Structure and Design Program, University of Washington, Seattle, WA, USA Enveloped viruses use specialized protein machinery to attach to host cells and fuse their membrane with the cellular membrane in order to deliver their genetic material for replication. In influenza virus, the trimeric hemagglutinin (HA) glycoprotein spike is responsible for attachment and mediating membrane fusion following endocytosis. While structures of a subset of HA conformations and parts of the fusion machinery have been characterized, the states that drive the fusion process have proven to be refractory to classical structure determination. In addition, the nature of membrane deformations during fusion has eluded characterization. We employ a combination of structural and biophysical methods to study the process of HA-mediated membrane fusion. Electron tomography is ideally suited for characterizing enveloped virus ultrastructure and the interaction of virus with membranes during fusion. At a resolution of ~2 nm, cryo-electron tomography (cryo-ET) provides the ability to image individual viral glycoprotein spikes and the leaflets of lipid bilayers as well as details of virus ultrastructure. In order to characterize structural changes in the fusion protein itself in more detail, hydrogen/deuteriumexchange mass spectroscopy (HDX-MS) is used to monitor the structural dynamic changes in the HA complex during acid-induced fusion activation. Previous cryo-ET studies demonstrated that during early stages of fusion between influenza virions and liposomal target membranes, membrane remodeling is generally focused on the target membrane while the virus envelope remains unaltered (1). With pure phosphatidylcholine (PC) liposomes, in the early stages of fusion, HA induces highly localized target membrane deformations and open-mouthed dimples consistent with a high degree of content leakage, Cryo-ET and fluorescence spectroscopy experiments are now being used to test the effect of two additional lipid components, cholesterol and lysobisphosphatidic acid (LBPA), on the fusion process. Cholesterol is a major constituent of plasma membranes while LBPA is enriched in late endosomes. Addition of either of these components to PCbased liposomes makes the lipid mixing stage of membrane fusion more efficient as monitored by fluorescence spectroscopy; cholesterol also reduces content leakage, while LBPA does not have a major effect on the degree of liposomal content leakage. 3-D imaging by cryo-ET reveals that the liposomes with 25-50% cholesterol or LBPA exhibit a greater proportion of closely apposed, extended contacts between virus and target membranes, in striking contrast to the more localized, punctate deformations observed with pure PC liposomes (Figure 1). We hypothesize that the closely apposed, extended contacts represent a key intermediate leading to efficient membrane merging. In order to characterize the nature of conformational changes in HA during fusion-activation, in a recent study, HDX-MS was also used to examine the HA ectodomain at pH values approaching fusiontriggering endosomal conditions (pH 6.0-5.5) (2). HA was found to exhibit increased dynamics at the fusion peptide and associated regions, while the interface between receptor-binding subunits (HA1) becomes bolstered (Figure 2). In contrast to many activation models, the HDX-MS data suggest that HA responds to endosomal acidification by releasing the fusion peptide prior to HA1 uncaging and springloaded refolding of HA2. A similar model has been proposed by Fontana et al., based upon cryo-ET data, although in that case, the sub-tomogram averaged models were not of sufficient resolution to identify the protein segments involved in the rearrangement (3). Microsc. Microanal. 21 (Suppl 3), 2015 1280 1274 Taken together, these complementary techniques are starting to reveal the staged, sequential nature of fusion protein activation and the resulting remodeling of target and viral membranes that take place during influenza HA-mediated fusion. This work was supported by NIH R01-GM099989, R00-GM080352, T32-GM007750 (N.K.G.), F32GM097805 (M.G.) and the Hope Barns Fellowship (N.K.G.). REFERENCES 1. Lee KK. 2010. Architecture of a nascent viral fusion pore. The EMBO Journal 29:1299-1311. 2. Garcia NK, Guttman M, Ebner JL, Lee KK. 2015. Dynamic changes during acid-induced activation of the influenza hemagglutinin fusion glycoprotein. Structure in press. 3. Fontana J, Cardone G, Heymann JB, Winkler DC, Steven AC. 2012. Structural Changes in Influenza Virus at Low pH Characterized by Cryo-Electron Tomography. J Virol 86:2919-2929. FIGURE 1. Population distribution of virus-target membrane contacts observed by cryo-ET (left). Addition of cholesterol or LBPA enhances fusion efficiency and is associated with increased formation of extended interfaces (right). This may be due to the lower cost of dehydrating the target leaflet with Chol or LBPA present. We propose that the extended, closely apposed contacts are key intermediates leading to fusion pore formation. FIGURE 2. HDX-MS analysis of influenza HA activation reveals the dynamic profile of a prefusion intermediate. The heat map shows the regions including fusion peptide that become increasingly dynamic as activation pH is approached. We hypothesize that this intermediate allows the fusion peptides to be deployed and presented by the intact trimer cage prior to HA2 hairpin formation; such staging of conformational changes allows for the efficient target membrane engagement followed by fusion. Garcia et al. (2015).");sQ1[640]=new Array("../7337/1281.pdf","Characterization of Outer Membrane Vesicle Release in Vibrio vulnificus.","","1281 doi:10.1017/S1431927615007199 Paper No. 0640 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of Outer Membrane Vesicle Release in Vibrio vulnificus. Cheri Hampton1, Ricardo C. Guerrero-Ferreira2, Hong Yi3 and Elizabeth R. Wright1,3 1. Emory University School of Medicine, Division of Pediatric Infectious Disease, and Children's Healthcare of Atlanta, Atlanta, Georgia 2. Laboratory of Structural Biology and Biophysics, Institute of Physics of Biological Systems, �cole Polytechnique F�d�rale de Lausanne, Switzerland. 3. Robert P. Apkarian Integrated Electron Microscopy Core, Emory University, Atlanta, Georgia The production of outer membrane vesicles (OMVs) by Gram-negative bacteria is an active process that takes place when a section of the outer membrane bulges off to form periplasm-containing vesicles without causing bacterial lysis. OMVs directly influence bacteria-host interaction and pathogenesis through their ability to modulate immune responses, contribute to biofilm formation, and deliver toxins and other virulence factors to host cells [1]. The ability of OMVs to bind to eukaryotic membranes makes them potential delivery vehicles for antibiotics and candidate vaccines against pathogens such as Vibrio cholerae, indicating their significant biotechnological potential [2]. OMVs have also been recognized to be important components of bacterial biofilms. V. vulnificus is a Gram-negative bacterium which colonizes filter-feeding shellfish. This opportunistic pathogen is responsible for 95% of seafoodrelated deaths in the United States and is categorized as a category B biodefense priority pathogen by the National Institutes of Allergies and Infectious Diseases [3]. Virulence in V. vulnificus is regulated by structural components such as cellular appendages (flagella and pili) and the capsular polysaccharide. Even though their contribution to bacterial pathogenesis is widely recognized, the mechanism of OMV production is not yet understood. Here, we use multiple TEM imaging techniques to characterize the unusual secretion and spatial arrangement of OMVs. We present further cryo-electron tomography data [4] (Fig. 1) of OMV release and distribution in Vibrio vulnificus and several of its mutant forms in order to link these results to known differences in virulence. Because of the low contrast of cryo TEM, the capsular polysaccharide remained obscure. In order to visualize the role of the capsular polysaccharide in providing this unique spatial arrangement we employ a high pressure freezing technique, self pressurized rapid freezing (SPRF) [5]. For cryo EM and cryo ET, V. vulnificus was grown in LB broth at 30 degrees C with shaking at 250 rpm overnight, then diluted into fresh LB and grown and additional 4 hours. Bacteria were applied to copper 200 mesh grids with R2/1 Quantifoil carbon supports and plunge frozen into liquid ethane. For SPRF, colonies from a fresh LB plate were scraped into a paste and thinned with enough LB broth to facilitate pipetting into a 16mm long and 0.65 mm o.d. copper tube. After crimping the ends closed, the tube was manually plunged into liquid propane. Tubes were sectioned and freezesubstituted. Both cryo EM and cryo ET reveal a regular arrangement of OMVs spaced approximately 100 nm from the outer membrane and in successive rings with 200 nm spacing. We do not observe this discrete spacing in reduced virulence mutants lacking the capsular polysaccharide. We were able to visualize the capsular polysaccharide using the SPRF technique, and observe that the capsule extends in a radius around the outer membrane a distance of ~100 nm, and that the individual OMVs retained these 100 nm polysaccharide chains which remain joined at the distal end, creating 200 nm spacing between the outer membrane and OMVs, as well as between rows of OMVs (Fig. 2). We hypothesize that the capsular polysaccharide has a definitive role in OMV distribution and virulence in V. vulnificus. Microsc. Microanal. 21 (Suppl 3), 2015 1282 References: [1] A Kulp and MJ Kuehn. Annual Review of Microbiology 64 (2010), p. 163-184. [2] S Schild, EJ Nelson, AL Bishop, and A Camilli. Infection and Immunity 77 (2009), p. 472-484. [3] MS Strom and RN Paranjpye. Microbes Infection 2 (2000), p. 177-188. [4] R Guerrero-Ferreira, G Williams, and ER Wright. Microscopy and Microanalysis 17 (2011), p. 142143. [5] JLM Leunissen and H Yi. Journal of Microscopy 235, pt 1 (2009), p. 25-35. [6] This research was supported by funds from Emory University, Children's Healthcare of Atlanta, the Emory Center for AIDS Research, the Georgia Research Alliance, Human Frontiers Science Program, National Institutes of Health (1R01GM104540), and the National Science Foundation (0923395) to E.R.W, and S10 RR025679 to P.W.S. All EM data was collected at the Emory University Robert P. Apkarian Integrated Electron Microscopy Core. Figure 1. Cryo ET slice of plunge-frozen V. vulnificus producing outer membrane vesicles. Scale bar 200 nm. Figure 2. SPRF, freeze-substituted section of V. vulnificus reveals the capsular polysaccharide chains extending from the outer membrane and attached to surrounding OMVs. Scale bar 200 nm.");sQ1[641]=new Array("../7337/1283.pdf","Molecular Characterization of Leading Edge Protrusions in the Absence of Arp2/3 Complex.","","1283 doi:10.1017/S1431927615007205 Paper No. 0641 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Molecular Characterization of Leading Edge Protrusions in the Absence of Arp2/3 Complex. Karen L Anderson1, Christopher Page1, Praveen Suraneni2, Rong Li2, Niels Volkmann1 and Dorit Hanein1 1 Bioinformatics and Systems Biology Program, Sanford-Burnham Medical Research Institute, La Jolla, CA 92037 Stowers Institute for Medical Research, Kansas City, MO 64110 2 Cells employ protrusive leading edges to navigate and promote their migration in diverse physiological environments. Classical models of leading edge protrusion rely on a treadmilling dendritic actin network that undergoes continuous assembly nucleated by the Arp2/3 complex, forming ruffling lamellipodia. Although the dendritic nucleation model has been rigorously evaluated in several computational studies, experimental evidence demonstrating a critical role for Arp2/3 in the generation of protrusive actin structures and cell motility has been far from clear. Most components of the pathway have been probed for their relevance by RNA interference or dominant-negative constructs. However, given that the Arp2/3 complex nucleates actin at nanomolar concentrations, even a dramatic knockdown could still leave behind a level sufficient to fully or partially support Arp2/3 complex-dependent functions. Our recent work renders the characterization of newly developed fibroblasts cells completely lacking the Arp2/3 complex. Characterization of the impact of the absence of Arp2/3 complex on these genetically matched cells with and without Arp2/3 complex included single cell spreading assays, wound healing assays, long-time single cell motility tracking, chemotaxis assays, fluorescence staining imaging with confocal or structured illumination microscopy [1.2]. In the absence of Arp2/3 complex, the fibroblasts were unable to extend lamellipodia but generated dynamic leading edges composed primarily of filopodia-like protrusions (FLPs), with formin proteins (mDia1 and mDia2) concentrated near their tips. ARPC3-/- fibroblasts maintained an ability to move but exhibited a strong defect in persistent directional migration in both wound healing and chemotaxis assays [1,2]. In the absence of the Arp2/3 complex in ARPC3-/- fibroblasts, formins are required for the extension of FLPs but they are insufficient to produce a continuous leading edge. We hypothesize that the leading edge protrusions of these motile cells are driven by local coordinated actions of neighboring FLPs and myosin-II-mediated contractility localized in the arcs in between the two respective FLPs. Our cryoEM studies provided direct evidence for the existence of short isotropic actin filaments in the arcs regions. Here, we will highlight our advances on determining the molecular-level organization of these actin networks, through an integrated approach that employs electron cryo-tomography of whole mammalian cells in conjunction with correlative light microscopy. This work is supported by NIGMS grant P01 GM066311. Microsc. Microanal. 21 (Suppl 3), 2015 1284 1278 References: [1] Suraneni P, Rubinstein B, Unruh JR, Durnin M, Hanein, D, and Li R (2012). The Arp2/3 complex is required for lamellipodia extension and directional fibroblast cell migration. J Cell Biol. 197, 239. PMC3328382 [2] Suraneni P, Fogelson B, Rubinstein B, Noguera P, Volkmann N, Hanein D, Mogilner A, Li R (2015). A mechanism of leading edge protrusion in the absence of Arp2/3 complex. Mol Biol Cell. Jan 7 [Epub ahead of print]. Figure 1: Actin organization at the leading edge in wt and ARPC3-/- fibroblasts. (A) wt and ARPC3-/- cells were stained with fluorescent phalloidin and imaged by SIM [2]. The white boxes represent the zoomed regions in the corresponding images. Scale bars: (A) wt 2 �m, 1.5 �m (zoom) and ARPC3-/- 2.5 �m, 1.5 �m. The corresponding molecular actin organization of these cells will be the center of this study.");sQ1[642]=new Array("../7337/1285.pdf","-Synuclein Amyloid Fibrils Formed of Two Protofibrils.","","1285 doi:10.1017/S1431927615007217 Paper No. 0642 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 -Synuclein Amyloid Fibrils Formed of Two Protofibrils. Altaira D. Dearborn1, Naiqian Cheng1, J. Bernard Heymann1, Joseph S. Wall2, Jobin Varkey3,4, Ralf Langen3, and Alasdair C. Steven1. 1. Laboratory of Structural Biology Research, National Institute of Arthritis and Musculoskeletal and Skin Diseases, National Institutes of Health, Bethesda, USA. 2. Department of Biology, Brookhaven National Laboratory, Upton, USA. 3. Zilkha Neurogenetic Institute, University of Southern California, Los Angeles, USA. 4. Centre for Converging Technologies, University of Rajasthan, Jaipur, India. -Synuclein (S) is a natively unstructured, 140 amino acid protein, expressed in the cytosol of neurons. Normally, it binds to and remodels membranes at the presynaptic terminal, becoming -helical in the process [1 and 2]. Following a nucleation event, the protein polymerizes into amyloid fibrils that are the main component of intraneuronal protein aggregates called Lewy Bodies. The presence and location of Lewy Bodies correlates with the presence and clinical presentation of a series of chronic, neurodegenerative diseases including Parkinson's disease [3]. Due to the lack of prominent surface features and sample heterogeneity, structural information has been limited. Recombinant S was expressed, purified and assembled into fibrils, which were observed by cryo-EM on a Philips CM200-FEG microscope and by dark-field STEM of unstained frozen-dried specimens at the Brookhaven STEM facility. From the resulting cryo-EM images and STEM micrographs, segmentaverages and 2D cross-sections were calculated, using Bsoft [4]. Mass-per-length measurements were made from the STEM data, using PCmass32. In cryo-EM (Figure 1A), the majority of fibrils have a periodic undulation in width, 8.2 - 11 nm, over an axial repeat distance of approximately 77 nm, which we take to represent a rotation through 180�. In STEM (Figure 1B), fibrils are seen to have a pair of thin threads of heightened density that we interpret to be tightly bound metal ions scavenged from buffer during the protein purification. These threads oscillate between a maximum center-to-center separation of 5 nm and a minimum separation of 0 nm (crossover points, Figure 1B inset), with the same periodicity as was observed in cryo-EM. 2D reconstruction of fibrils from cryo-EM revealed two protofibrils with elongated cross-sections, each protofibril being ~ 7.5 nm by ~ 2.5 nm (Figure 2A). The protofibril can be further divided into two ellipsoidal sub-domains (Figure 2D). The protofibrils are asymmetrically disposed about the fibril axis, giving a fibril cross-section shaped like the Greek letter Nu (Figure 2A), implying a non-equivalent mode of association. Reconstruction from the STEM data showed the two thread-like features symmetrically disposed at a radius ~ 2.5 nm from the fibril axis (Figure 2B). After the two reconstructions were aligned, the threads localized near the middle of each protofibril (Figure 2C), at the interface of the two subdomains (Figure 2D). The STEM mass-per-length measurements yielded a mean linear density corresponding to two subunits per 4.7 � axial rise, which indicates that each protofibril has one subunit per 4.7 � axial rise (the inter-strand spacing in cross- structures). This value is consistent with a parallel superpleated -structure [6]. These data, taken in conjunction with pre-existing data from other sources, form a basis for formulating specific cross- models for these amyloid fibrils. [7] Microsc. Microanal. 21 (Suppl 3), 2015 1286 References: [1] N Mizuno et. al, J Biol Chem. 287 (2012), p. 29301-29311. [2] CC Jao et. al, PNAS 101 (2004), p. 8331-8336. [3] H Braak et. al. Neurobiol Aging 24 (2003), p. 197-211. [4] JB Heymann and DM Belnap J. Struct. Biol. 157 (2007), p. 3-18. [5] Z Qin et. al. Biochemistry 46 (2007), p.13322-13330. [6]AV Kajava et. al. PNAS 101 (2004), p. 7885-7890. [7] This research is supported by the Intramural Research Program of the National Institute of Arthritis and Musculoskeletal and Skin Diseases of the National Institutes of Health. Figure 1: Micrographs of Synuclein fibrils. Cryo-EM (A) and STEM (B). (B) Inset, magnified region with entwined thread-like feature. Scale bars are 30 nm. Figure 2: Cross-sectional density maps of Synuclein fibrils. Cryo-EM (A) and STEM (B) 2D reconstructions, showing -shaped cross-section (A) and high-density thread positions (B), respectively. (C) Localization of high-density thread positions (yellow ovals) in the fibril cross-section. (D) Interpretive diagram of fibril cross-section formed of two protofibrils (black rectangles), each with two subdomains (orange ovals) and a high-density thread (yellow ovals). Scale bar is 4 nm.");sQ1[643]=new Array("../7337/1287.pdf","Bright-Field STEM Tomography of Blood Platelets in Thick Sections","","1287 doi:10.1017/S1431927615007229 Paper No. 0643 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Bright-Field STEM Tomography of Blood Platelets in Thick Sections R.D. Leapman1, M.A. Aronova1, J.D. Hoyne1, G.N. Calco1, B.C. Kuo1, Q. He1, I.D. Pokrovskaya2, L.J. MacDonald2, A.A. Prince2, B. Storrie2 1 2 National Institute of Biomedical Imaging and Bioengineering, NIH, Bethesda, MD 20892; Department of Physiology and Biophysics, University of Arkansas for Medical Sciences, Little Rock, AR 72205 By performing electron tomography in the scanning transmission electron microscope (STEM), it is possible to obtain 3D reconstructions at a resolution less than ~10 nm from stained plastic-embedded sections of eukaryotic cells in 1�2 �m thick sections. This is achievable because there are no imaging lenses after the specimen when the electron microscope is operated in STEM mode, so that chromatic aberration of the objective lens does not compromise the spatial resolution when there is strong multiple inelastic scattering [1-4]. However, to image thick sections it is necessary to avoid a second source of resolution loss due to geometrical broadening of the probe, which can be mitigated by selecting a small probe convergence angle of ~1 mrad. Previous work has also shown that by using an axial bright-field detector instead of a standard high-angle annular dark-field detector, it is possible to reduce a third source of resolution loss caused by multiple elastic scattering in the lower part of the specimen [2-4]. Here, we have applied axial bright-field STEM tomography to determine the 3D ultrastructure of human blood platelets, which are small anucleate blood cells that aggregate to seal leaks at sites of vascular injury and are important in the pathology of atherosclerosis and other diseases. Of particular interest are the morphological changes that occur in -granules, which contain important proteins released when platelets are activated [5]. Due to difficulty in controlling the physiological state of platelets, structural changes that occur in the early stages of -granule activation are not yet fully understood. Electron tomograms were acquired using an FEI Tecnai TF30 transmission electron microscope equipped with a field-emission gun and operating at an acceleration voltage of 300 kV. The instrument was equipped with a Gatan bright-field STEM detector. Blood drawn under IRB procedures was incubated for 5 minutes and treated with prostacyclin and apyrase to suppress further activation before being high-pressure frozen at 2100 bar. Samples were freeze-substituted with osmium tetroxide and glutaraldehyde, dehydrated with acetone and embedded in epon. Sections were cut to a thickness of 1.5 �m and stained with uranyl acetate and lead citrate, before being coated with carbon and 20-nm gold nanoparticles, which served as fiducial markers. Dual axis bright-field STEM tilt series were acquired over an angular tilt range of �68� with a 2� tilt increment. Tomograms were reconstructed using the IMOD program [6] and surface rendered using FEI Amira 3D software. An orthoslice through a STEM tomogram from a 1500 nm thick section of a blood platelet in early stage of activation in Fig. 1A shows tubules extending from a decondensing -granule to the plasma membrane, whereas other -granules remain in their condensed state, as seen in the 3D visualization in Fig. 1B. The internal arrangement of organelles is shown in more detail in Fig. 2A and plasma membrane in Fig. 2B reveals numerous pores from the tubular extensions of the -granules [7, 8]. Microsc. Microanal. 21 (Suppl 3), 2015 1288 References [1] A.E. Yakushevska et al, J. Struct. Biol. 159 (2007) p. 381. [2] M.F. Hohmann-Marriott et al, Nature Methods 6 (2009) p. 729. [3] A.A. Sousa et al, Ultramicroscopy 109 (2009) p. 213. [4] A.A. Sousa et al, J. Struct. Biol. 174 (2011) p. 107. [5] J. Kamykowski et al, Blood 118 (2011) p. 1370. [6] J.R. Kremer, D.N. Mastronarde, J. Struct. Biol. 116 (1996) p. 71. [7] L.J. MacDonald et al, (2015) in press. [8] Research supported by the intramural program of the National Institute of Biomedical Imaging and Bioengineering, NIH. Work in the Storrie laboratory was supported in part by NIH grant R01 HL119393. Fig. 1. STEM tomography of 1500 nm thick section of plastic-embedded, frozen and freezesubstituted human blood platelet in early stage of activation: (A) orthoslice showing tubular extensions from decondensing -granule (red arrow); (B) 3D model showing condensed unactivated granules (blue) and decondensing activated granules (brown). Adapted from reference [7]. Fig. 2. Surface rendered models of platelet in early stage of activation: (A) internal structure of organelles showing canalicular system (yellow), condensed -granules (blue), decondensing granules (brown), dense granules (red), and mitochondria (purple); (B) plasma membrane (green) showing openings of the internal tubular extensions originating from decondensing granules (red arrow). Adapted from reference [7].");sQ1[644]=new Array("../7337/1289.pdf","Using Graphene Liquid Cells for High-resolution Chemical Analysis of Nanoparticle Reactions","","1289 doi:10.1017/S1431927615007230 Paper No. 0644 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Using Graphene Liquid Cells for High-resolution Chemical Analysis of Nanoparticle Reactions Canhui Wang1, Tolou Shokuhfar1, 2 and Robert F Klie1 1. 2. University of Illinois at Chicago, Department of Physics, Chicago, IL 60607 Michigan Technological University, Department of Mechanical Engineering, Dept. of Biomedical Engineering, Houghton, MI, USA 49931 Radiolysis has been intensely studied due to its importance in areas, such as medicine, atmospheric science and nuclear energy production. In transmission electron microscopy, however, where the dose rate of a 200 keV electron beam is 7 orders of magnitude greater that dose rates commonly generated by other radiation sources, radiolysis has mostly been discarded as an unwanted side-effect. For in-situ TEM experiments, the creation of unwanted free radicals, such as hydroxyl and hydrogen, or hydrated electrons, as the result of water dissociation by a high energy electron beam, can have a significant effects on the stability of a suspended nano-particle or on the redox conditions in an electro-chemical experiment. Only recently was it realized that a fundamental understanding and control of radiolysis is required in order to reliably study chemical reactions in an electron microscope. To date, a few experiments [1, 2] and theoretical studies have been reported in the literature,[3] addressing the fundamental issues of radiolysis. Therefore, even a basic understanding of the observed effects of radiolysis is still lacking. Direct imaging of electron-liquid interaction has become possible at the nanometer scale with the widespread availability of commercial liquid cells, where a small volume of liquid is sealed between two electron transparent layers. While this enables thin liquid layers, and structures suspended in such layers, to be studied at atmospheric pressures in processes such as electroplating, nucleation of nanoparticles or bubbles, and biological reaction, the imaging resolution and chemical quantification capabilities are severely limited by the existing holder designs. Moreover, the electron dose rate required for atomicresolution imaging is believed to be significantly higher than the damage threshold in conventional liquid-cell experiments. We have recently developed a novel approach towards in-situ electron microscopy that allows the effects of radiolysis to be examined on the atomic-level scale and provides the required control over these effects to minimize beam-induced radiation damage. More specifically, we utilize a liquid cell, consisting of two graphene monolayers to encapsulate a small volume of liquid, as shown in Figure 1.[4] Such an approach allow us to image particles, such as the 12 nm diameter iron core of the protein ferritin with atomic resolution in a liquid environment. Moreover, electron energy-loss spectroscopy and spectrum imaging of the core-loss edges is now also possible for samples in a graphene liquid cell (Figure 2) , since the window layers consist of only a monolayer of carbon instead of >20 nm of SiN. In addition to the high spatial and spectra resolution that can be achieved using this new approach, we also discuss the effects of electron dose on the GLCs. We observed the generation of bubbles inside the GLC at sufficiently high electron dose rates. However, GLCs do not generate bubbles under even hours of electron beam irradiation at low magnification, indicating accumulation of total electron dose is not the reason for bubble formation in this system. This will be critical for our understanding of radiolysis Microsc. Microanal. 21 (Suppl 3), 2015 1290 and will be discussed in the context of the recent modeling study by Schneider et al. [3] Applications of this method to catalysts nanoparticles and battery cathode materials will be discussed. [5] References: [1] Grogan, J.M. et al, Nano Letters 14(1) (2014), p. 359. [2] Zheng, H., et al, Nano Letters 9(6) (2009), p. 2460. [3] Schneider, N.M., et al, Journal of Heat Transfer-Transactions of the Asme 136(8) (2014) [4] Wang, C., et al, Advanced Materials 26 (2014), p. 3410. [5] This research was supported by Michigan Technological University and Research Resource Center, University of Illinois at Chicago. Figure 1: Schematic diagram (A), as well as STEM images (B) and (C) of ferritin molecules in GLCs and graphene sandwiches. (B) is an annular bright field (ABF) image showing ferritin molecules encapsulated in both a GLC and graphene layers. The edges of the GLC is indicated by dash lines. (C) is a high angle annular dark field (HAADF) image showing atomic-resolution image of a sandwiched ferritin molecule, with the 12nm in diameter protein shell and individual Fe atoms resolved in light contrast. A GLC B Bubble Figure 2: (A) HAADF image of GLC containing ferritin molecules as well as a gas bubble in the center. (B) EELS spectra of the N K-, O K- end Fe L-edges taken from ferritin core and the protein shell in water as well as from regions without water.");sQ1[645]=new Array("../7337/1291.pdf","Liquid In Situ Analytical Electron Microscopy: Examining SCC Precursor Events for Type 304 Stainless Steel in H2O","","1291 doi:10.1017/S1431927615007242 Paper No. 0645 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Liquid In Situ Analytical Electron Microscopy: Examining SCC Precursor Events for Type 304 Stainless Steel in H2O S. Schilling1, A. Janssen1, X.L. Zhong1, N.J. Zaluzec1,2 and M.G. Burke1 1. Materials Performance Centre and Electron Microscopy Centre, The University of Manchester, Manchester M13 9PL UK 2. Electron Microscopy Center, NST Division, Argonne National Laboratory, Argonne, IL 60439 USA In situ transmission electron microscopy (TEM) has become an increasingly important research area in materials science with the advent of unique microscope platforms and a range of specialized in situ specimen holders. The ability to image and perform x-ray energy dispersive spectroscopy (XEDS) analyses of metals in liquids are particularly important for detailed study of the metal-environment interactions with specific microstructural features. In particular, these capabilities now make it feasible to explore what has been termed "SCC precursor phenomena" � that is, those sub-micron scale reactions between an alloy of interest and the environment. This topic is especially timely for this Swann Memorial Symposium, as Peter Swann's activities in this area, particularly in the 1970's concerning the initiation and early stages of transgranular SCC in austenitic stainless steels, were clearly prescient. In this study, we have used liquid cell TEM with XEDS to explore the "precursor phenomena" that can promote the development of defect initiation in Type 304 austenitic stainless steel. For this work, the FIB lift-out technique was used to extract specimens to be studied in the liquid cell TEM specimen holder. This technique has been applied to examine the localised dissolution of MnS inclusions, which can lead to pit initiation. FIB sections containing MnS inclusions were prepared from a 0.3 wt.% S Type 304 stainless steel for study in a Protochips Poseidon P210 liquid cell and P510 electrochemical cell in situ specimen holders with a 500 nm gap between the amorphous SiNx windows. Based on initial tests, the electrochemical Echip configuration was modified to optimize it for electrochemical measurements. The in situ experiments were performed in an FEI Tecnai T20 analytical electron microscope operated at 200 kV and equipped with an Oxford Instruments X-MaxN 80TLE Silicon Drift Detector (SDD) for spectrum imaging and analysis. A Zeiss Merlin FEG-SEM equipped with two Oxford Instruments X-MaxN 150 SDDs was used to analyse the bulk samples from the ex situ tests. MnS XED spectrum images in 1 bar air and after 24 h in deionised H2O in the P210 are shown in Fig. 1, and the XED sum spectrum obtained after dissolution is presented in Fig.2. The relevance of the in situ observations were confirmed by ex situ exposure of a bulk sample containing a Pt line sputtered next to MnS inclusions (to simulate the Ptcoated FIB section) since Pt would support galvanic corrosion, and possibly accelerate the dissolution rate [1]. XED spectrum images obtained from a bulk specimen containing MnS inclusions are shown in Fig. 3a; the same area after 8 days in deionised H2O is shown in Fig. 3b. Both ex situ (Fig. 4) and in situ electrochemical polarisation data will be discussed with respect to localised corrosion. Dissolution of MnS inclusions in low alloy steels is associated with Environmentally-Assisted Cracking [2]. Recent research has also indicated that MnS inclusions may also be associated with SCC as well as low crack growth rates in austenitic stainless steels during certain conditions of corrosion fatigue in primary water environments in light water reactors. References: [1] Zhang, X. G. (2011). Galvanic corrosion. Uhlig's Corrosion Handbook, 51,123. [2] H. H�nninen et al. (1990) Effects of MnS Inclusion Dissolution on Environmentally-Assisted Cracking in Low-Alloy and Carbon Steels. Corrosion: July 1990, Vol. 46, No. 7, pp. 563-573 Microsc. Microanal. 21 (Suppl 3), 2015 1292 (EP/ 1003290/1), also supported in part by the U.S. DOE, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357 at the Electron Microscopy Center, NST Division of ANL. [3] The authors acknowledge the support of Dr. M.A. Kulzick and BP, and EPSRC PROMINENT programmes 1 atm air S 24h in H2O Fig. 1: BF-STEM and XED spectrum images of MnS Inclusion in E-cell: a) 1 bar air; b) after 24 h in H 2O showing dissolution of MnS during in situ exposure in H2O. Fig. 2: XED Sum Spectrum from region containing the dissolved MnS inclusion after exposure in H 2O. Fig. 3: Ex situ observations: SE and XEDS spectrum images of MnS inclusions: a) 0 days; b) 8 days in H2O. Note the dissolution of the large MnS inclusions above the Pt strip whereas the inclusions covered by sputtered Pt remain intact. Fig. 4: Polarisation data for Type 304 stainless steel in H2O using conventional bulk samples.");sQ1[646]=new Array("../7337/1293.pdf","Design of a Heated Liquid Cell for in-situ Transmission Electron Microscopy","","1293 doi:10.1017/S1431927615007254 Paper No. 0646 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Design of a Heated Liquid Cell for in-situ Transmission Electron Microscopy Andrew J. Leenheer, Katherine L. Jungjohann and C. Thomas Harris Sandia National Laboratories, Center for Integrated Nanotechnologies, Albuquerque, NM 87123 USA. Microfabricated, silicon-based chips developed for advanced capabilities in sample holders have recently led to a broad expansion of in-situ transmission electron microscopy (TEM) experiments involving materials' responses to increased temperature, electrical bias, or mechanical stress. By including freestanding, electron-transparent silicon nitride membranes, a thin environmental chamber for gases or liquids can be created in the TEM allowing new insights into the nanoscale processes involved in electrochemical, catalytic, or biological systems. Heated gas environments in the TEM have been previously demonstrated with microfabricated chips [1-2], but little work has been done with heated liquid environments. With control over the thermal environment, systems that activate at increased temperature (e.g. nanoparticle growth, protein denaturation, corrosion) or systems that degrade with temperature cycling (e.g. battery materials) can be studied. We have developed a custom TEM liquid cell optimized for quantitative control of pA-level electrical currents [3]. The chip design is shown in Figure 1. The bottom chip contains ten insulated electrical leads converging to the center over a 40-�m diameter, 50-nm thick silicon nitride membrane seen in Figure 1(c). The lid chip likewise contains a central window as well as etched-through fluid fill ports. After adding the materials and electrolyte of interest, the assembly is hermetically sealed with epoxy and connected to the TEM holder stub as shown in Figure 1(b). The fluid chamber thickness is laterally uniform and has been measured to be 100-200 nm thick by electron energy loss spectroscopy; the small window size and relatively thick nitride membranes reduce the bowing commonly seen in TEM liquid cells. Therefore, there are no changes in the background contrast when moving around the liquid cell, and optimal imaging can be done anywhere within the viewing window. The bottom chip is easily customized using electron-beam lithography. Here we demonstrate a resistive heater near the center of the silicon nitride membrane. As seen in Figure 2(a), a bottom chip with widely separated electrode tips was patterned to create two 50-nm thick resistors out of Al metal. One serpentine resistor acts as a heater, while the straight-line resistor measures the nearby temperature. The thermometer's temperature dependence over the relevant 20-100 �C range was calibrated in air in an oven as shown in Figure 2(b) using a four-point resistance measurement. As expected, resistance is a linear function of temperature. Additionally, the power input to the serpentine heater required for a local temperature rise at the heater is shown in Figure 2(c). This localized sample heating draws little power and thus ensures fast response and minimal spatial drift due to thermal expansion. Because the bottom chip contains up to 10 independent electrodes, many electrodes are still available for electrical biasing after the heater has been fabricated. The remaining electrodes were used for electrochemical experiments; for example, thermal runaway in Li-ion batteries is related to decomposition at solid-liquid interfacial layers at elevated temperatures. Additionally, electrochemical nanowire growth is often enhanced at elevated temperatures. The voltage drop across the serpentine heater to achieve 100 �C is less than 0.5 V, and the heater can additionally be electrically isolated by masking with an insulating layer if needed. Microsc. Microanal. 21 (Suppl 3), 2015 1294 Temperature control in a TEM liquid cell will open up new insights into solid-liquid interfaces, allowing in-situ determination of activation barriers of the processes imaged. The custom design shown here can be adapted to a wide variety of geometries depending on the requirements of the electrochemical, biological, or chemical system under study [4]. References [1] J F Creemer et al, J. Microelectromech. Syst. 19 (2010), p. 254. [2] L F Allard et al, Microsc. Microanal. 18 (2012), p. 656. [3] A J Leenheer et al, J. Microelectromech. (2015), in press: DOI 10.1109/JMEMS.2014.2380771. [4] This work was performed at the Center for Integrated Nanotechnologies, an Office of Science User Facility operated for the U.S. Department of Energy (DOE) Office of Science. Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000. We thank Dr. J. Sullivan and M. Shaw for design and fabrication discussions and contributions. Figure 1. (a) TEM liquid cell layout, (b) assembled and sealed chip at the TEM holder tip, and (c) optical micrograph of the electrode tips converging over the circular silicon nitride membrane. Figure 2. (a) Scanning electron microscope image of a serpentine heater (top) and thermometer (bottom) patterned with 50-nm thick Al. (b) Thermometer resistance as a function of temperature, where the line is a linear fit to the experimental data points. (c) Temperature rise at the heater calculated as a function of serpentine heater current assuming cylindrical geometry and a liquid/membrane total thickness of 200 nm with composite thermal conductivity of 1.5 W/(m K).");sQ1[647]=new Array("../7337/1295.pdf","In situ Liquid TEM Study for Degradation Mechanisms of Fuel Cell Catalysts during Potential Cycling Test","","1295 doi:10.1017/S1431927615007266 Paper No. 0647 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In situ Liquid TEM Study for Degradation Mechanisms of Fuel Cell Catalysts during Potential Cycling Test Shinya Nagashima1,5, Kenta Yoshida2,3, Tomoki Hiroyama2, Kun Liu2, Yipu Kang2, Toshihiro Ikai4, Hisao Kato4, Tetsuo Nagami4 and Keisuke Kishita5 1. Materials Research and Development Laboratory, Japan Fine Ceramics Center, Atsuta-ku, Nagoya, 456-8587, Japan 2. Nanostructures Research Laboratory, Japan Fine Ceramics Center, Atsuta-ku, Nagoya, 456-8587, Japan 3. Institute for Advanced Research, Nagoya University, Chikusa-ku, Nagoya, 464-8603, Japan 4. Catalyst Design Department, Material Engineering Division, Toyota Motor Corporation, Toyota-cho, Toyota, Aichi, 471-8572, Japan 5. Material Analysis Department, Material Development Division, Toyota Motor Corporation, Toyotacho, Toyota, Aichi, 471-8572, Japan The polymer electrolyte fuel cell (PEFC) is a promising energy source for fuel cell vehicles. The typical electrocatalyst used in PEFCs consists of platinum nanoparticles on carbon black (Pt/C). The development of advanced electrocatalysts for PEFCs requires reduction of Pt usage and enhancement of durability [1-3]. To achieve further design improvements for Pt/C electrocatalysts, it is essential to understand the degradation mechanisms in real space for Pt/C electrocatalysts. We applied in situ liquid TEM observation technique using a liquid flow cell TEM holder with electrical biasing capabilities (Poseidon, Protochips Inc.) into the differential pumping environmental TEM (Titan ETEM, FEI Company) for direct observation of structural changes of Pt/C electrocatalysts. The electrochemical cell simulating environment of activated PEMFC was comprised of deposited Pt/C electrocatalyst onto an electrode on a MEMS chip and flowing electrolytic solution of 0.1 M and 0.2 M HClO4. We obtained cyclic voltammetry (CV) curves by electrochemical measurement and dynamic TEM movies simultaneously. Figure 1-3 are selected area captured (SAC) images of movies obtained from Pt/C electrocatalysts in HClO4 solution during potential cycling tests. We succeeded in direct observation of degradation behaviour of Pt/C electrocatalysts such as dissolution of Pt nanoparticles (Figure 1), detachment of a Pt nanoparticle from carbon support (Figure 2) and aggregation via Pt nanoparticles migration (Figure 3) which had been suggested [1-3]. In conclusion, in situ liquid TEM observation during potential cycling tests is a powerful tool for understanding of electrochemical behaviour of electrocatalysts of PEMFCs in nanometer scale. References: [1] Y. Shao-Horn, W. C. Sheng, S. Chen, P. J. Ferreira, E. F. Holby and D. Morgan, Top catal, 46, (2007), p. 285-305. [2] X. Yu, S. Ye, Journal of Power Sources, 172, (2007), p. 145-154. [3] J. C. Meier, C. Galeano, I. Katsounaros, A. A. Topalov, A. Kostka, F. Sch�th and K. J. J. Mayrhofer, ACS Catal 2, (2012), p. 832-843. Microsc. Microanal. 21 (Suppl 3), 2015 1296 Figure 1. SAC images obtained from the Pt/C electrocatalysts over the potential range from -0.75 V to 0.30 V (vs. Pt) at a scan rate of 50mV/s in 0.1 M HClO4 at room temperature. Figure 2. SAC images obtained from the Pt/C electrocatalysts over the potential range from -0.70 V to 0.45 V (vs. Pt) at a scan rate of 50mV/s in 0.2 M HClO4 at room temperature. Figure 3. SAC images obtained from the Pt/C electrocatalysts over the potential range from -0.70 V to 0.45 V (vs. Pt) at a scan rate of 500mV/s in 0.2 M HClO4 at room temperature.");sQ1[648]=new Array("../7337/1297.pdf","In situ TEM imaging of Nanoparticles interacting with Glioblastoma Stem Cells","","1297 doi:10.1017/S1431927615007278 Paper No. 0648 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In situ TEM imaging of Nanoparticles interacting with Glioblastoma Stem Cells Elliot S. Pohlmann1, Kaya Patel1, Sujuan Guo1, Madeline J. Dukes2, Zhi Sheng1, and Deborah F. Kelly1 1. 2. Virginia Tech Carilion School of Medicine and Research Institute, Roanoke, VA 24016, USA. Applications Science, Protochips, Inc., Raleigh, NC, 97606, USA. Modern cancer research is embracing alternative drug delivery vehicles to treat human tumors that are difficult to manage by conventional therapies. These systems often include, but are not limited to liposomal formulations, polymer coatings, gold nanoparticles (NPs), or combinatorial reagents. Gold NPs present an attractive platform for both therapeutic and medical imaging purposes as they have high contrast and small dimensions (typically < 100 nm) [1,2]. As NP-based therapies are currently being used in clinical trials to treat solid tumors [3], this creates a major impetus to closely monitor their use and effectiveness. Current measurements are limited to the level of tissue accumulation while information concerning NPs interacting with individual cells remains missing from these analyses. As such, the development of new technologies to visualize and directly assess the behavior of NP-based therapies at the molecular level may contribute essential information toward evaluating therapeutic delivery, uptake, and efficacy. To address this issue, we have recently engineered a new system to perform real-time visual assessments of NPs interacting with individual cells while contained in liquid. Here, we present the first real-time information of gold nanorods interacting with glioblastoma stem cells (GSCs). We elected to use these cells because (1) glioblastomas are a lethal threat to all patients diagnosed with these tumors; (2) malignant brain tumors are difficult to treat by conventional therapies, due to the GSCs that populate the tumors [4]; and (3) NP-based assessment for therapeutic purposes are currently ill-defined. First, we enriched for GSCs from a mixed population of glioblastoma cells (GS9.6 line) based on the presence of the NOTCH1 receptor. GS9-6/NOTCH1+ cells were then tethered to silicon nitride microchips containing integrated microwells and coated with antibodies against the NOTCH1 receptor (Figure 1). The tethered cells were treated with PVP-coated gold nanorods (Nanopartz, Inc.) or aqueous solution lacking nanorods as a negative control. The samples were sealed in a Poseidon specimen holder (Protochips, Inc.) and imaged using in situ Transmission Electron Microscopy (TEM) protocols. Realtime recordings of the gold nanorods interacting at the cell membrane and within the cells (Figure 2) were acquired under low-dose conditions at 120 kV using TEM. Contour plots of individual images acquired at various time intervals illustrated the dynamic movements of the nanorods with respect to the cell membranes. Therefore, our work shows the first real-time movements of gold NPs interacting with GSCs in a liquid environment using In Situ TEM. These advancements provide a new basis for directly evaluating the effects of NP therapeutics on individual cancer cells at the molecular level. References: [1] Chow, E. K., and D. Ho, Science Trans. Med. 5 (2013), pp. 216rv214. [2] Rink, J. S., Plebanek, M. P., Tripathy, S., and C. S. Thaxton, Current Opin. in Onco. 25 (2012), pp. 646-651. [3] Yasun, E., Kang, H., Erdal, H., Cansiz, S., Ocsoy, I., Huang, Y. F., and W. Tan, Interface Focus 3 (2013), pp. 20130006. [4] Stopschinski, B. E., Beier, C. P., and D. Beier, Cancer Letters 338 (2013), pp. 32-40. Microsc. Microanal. 21 (Suppl 3), 2015 1298 A Top microchip Microwells Base microchip 10 �m Electron beam Glioma Stem Cell 300 �m 0.15 - 2 �m 50 nm 300 �m B Fluidic Chamber Nanorods Edge of Cell Figure 1. (A) Schematic to demonstrate a cross-sectional view of the microfluidic system containing GS9-6/NOTCH1+ glioma cells while positioned in the TEM. (B) Microchips with integrated microwells were used to specifically tether GS9-6/NOTCH1+ GSCs treated with gold nanorods within the fluidic chamber of the Poseidon specimen holder. Scale bar is 500 nm. Figure 2. GS9-6/NOTCH1+ cells interacted with gold nanorods within 20-minutes of TEM imaging. Serial images were acquired for 30-second intervals within selected regions of interest (red square). Contour plots were generated for the initial and final time points, illustrating movements in the nanorods and solution (white dashed circles) within the tumor cells. Scale bars in side panels are 50 nm.");sQ1[649]=new Array("../7337/1299.pdf","Atomic study of Fe3O4/SrTiO3 Interface","","1299 doi:10.1017/S143192761500728X Paper No. 0649 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic study of Fe3O4/SrTiO3 Interface D. Gilks1, D.M. Kepaptsoglou2, K. McKenna1, L. Lari1,3, Q.M. Ramasse2, K. Matsuzaki4, T. Susaki4 and V.K. Lazarov1 1 2 Department of Physics, University of York, Heslington, York, YO10 5DD, UK, SuperSTEM Laboratory, STFC Daresbury Campus, Warrington, WA4 4AD, UK 3 York JEOL Nanocentre, University of York, Heslington, York, YO10 5BR, UK 4 Tokyo Institute of Technology, 4259 Nagatsuta, Midori-ku, Yokohama 226-8503, Japan Complex oxide heterostructures are of interest for developing new functional materials due to their rich physical properties [1]. Many oxides are insulators, but also wide band gap semiconductors, superconductor, ionic conductors, ferrimagnetic and antiferromagnetic materials. Hence oxide heterostructures provide a broad platform for devices with multifunctional properties. The functionality of those devices strongly depends on the atomic scale structural and electronic discontinuities across the multilayer heterostructural interfaces; therefore the need for atomic scale understanding of their properties. In this work we focus on spinel-perovskite heterostructures that have potential for the development of multiferroic devices by using Fe3O4/StTiO3 as a model system [2,3]. This heterostructure is of importance for spintronic applications but also it represents a class of heterostructure that can host ferrimagnetic and ferroelectric properties which can be coupled through suitable interface engineering. The Fe3O4 films were grown on SrTiO3 (STO) substrate by pulse laser deposition techniques. A postannealing has been undertaken in order to improve films structure and stoichiometry [4]. The overall structure of the films was determined by X-Ray diffractometry. Atomic and electronic study of the Fe3O4/ StTiO3 interface was performed using scanning transmission electron microscopy and electron energy loss spectroscopy with a JEOL 2200 FS TEM/STEM and a Neon Ultrastem 100. First principle calculations were also performed in order to determine the spin/electronic properties in the near interface region of the Fe3O4/ STO from the several model interfaces. Fig. 1 shows that the Fe3O4 film has a uniform thickness and single crystal structure. The epitaxial relation between the film and substrate is the standard cube on cube as following: Fe3O4(111)||STO(111) and Fe3O4(11-2)||STO(11-2). The annealing of the film has improved the structural ordering as indicated in Fig.2. Moreover the annealing significantly reduces the number of antiphase domain boundaries in the film (not shown here). The atomically resolved electron energy loss spectroscopy shows that the film/substrate interface is atomically sharp, and that no atomic mixing is present. The lattice mismatch between the film and substrate introduces an interfacial strain which is relieved by formation of interfacial periodic misfit dislocations. The region between the dislocations has well defined atomic structure (Fig.2). The most favourable model that fits the best the STEM-HAADF data is determined by ../Ti/O/FeB/.. atomic planes, where FeB represent the Fe octahedral planes in magnetite. The first principle calculations has found that atomic mixing at the interface is not favourable. In addition the calculations show that electronic structure of this, polar in nature, interface is very sensitive to the atomic structure; with the band offsets and Fermi level position different and uniquely determinant for the all abrupt interface models considered. Microsc. Microanal. 21 (Suppl 3), 2015 1300 References [1] R. Ramesh et al., MRS Bull., 2008. 33(11): p. 1006-1014 [2] K. Ghosh, et al., Appl. Phys. Lett., 73(5): p. 689-691. (1998). [3] J.G. Zheng, et al., Jour. Vac. Sci.Tech. B, 2007. 25(4): p. 1520-1523. [4] M. Kosuke, et al., Jour. Phys. D: Appl. Phys. 2013. 46(2): p. 022001. [5]The authors acknowledge funding from EPSRC via research grants EP/K013114/1 and EP/K032852/1. SuperSTEM is the U.K. National Facility for Aberration-Corrected STEM funded by the EPSRC. Part of this work was performed at BeyondNano CNR-IMM, supported by MIUR under the project Beyond-Nano (PON a3_00363) Fig. 1: Fe3O4/SrTiO3(111) interface. a) -2 x-ray diffraction shows single phase Fe3O4 film with parallel (111) planes shared between the film and the substrate in the growth direction. b) Low magnification image of the Fe3O4/SrTiO3. Fig.2: Atomically resolved HAADF-STEM image of the Fe3O4/SrTiO3 interface in the [1-10] zone axis.");sQ1[650]=new Array("../7337/1301.pdf","Interfacial Structure in Epitaxial Perovskite Oxides on (001) Ge Crystal","","1301 doi:10.1017/S1431927615007291 Paper No. 0650 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Interfacial Structure in Epitaxial Perovskite Oxides on (001) Ge Crystal Xuan Shen,1,2 K. Ahmadi-Majlan,3 Joseph H. Ngai,3 Di Wu,2 and Dong Su1 Center for Functional Nanomaterials, Brookhaven National Laboratory, Upton, NY 11973, USA National Laboratory of Solid State Microstructures and Department of Materials Science and Engineering, Nanjing University, Nanjing, 210093, China 3. Department of Physics, University of Texas at Arlington, 502 Yates Street, Arlington, TX 76019, USA 2. 1. Integration of high-quality epitaxial perovskite oxides on semiconductors have been achieved due to the recently development of Molecular Beam Epitaxy (MBE).[1] For the hetero-epitaxial thin film system, as the film grows up to a certain thickness, interfacial dislocations are expected to form to relax the strain.[2] Intensive studies, both experimentally and theoretically, have been reported that these interfacial dislocations are crucial for some physical properties, such as dielectricity, polarization, and magnetization. Previous work have been focused on perovskite-on-perovskite system, while the nature of interfacial dislocations in the perovskite-on-semiconductor system has never been reported. In this work, we have applied aberration corrected transmission electron microscopy (TEM) and scanning TEM (STEM) techniques to investigate the interfacial structure of SrZr0.68Ti0.32O3 (SZTO) perovskite thin films on (001) Ge single crystal wafers, The atomic resolution STEM image and selected area electron diffraction (SAED) pattern indicated the epitaxial crystallographic relation of (001)[100]SZTO//(001)[110]Ge with abrupt interface. The misfit dislocation with Burgers vector of �a [011]SZTO was then observed at the SZTO film close to the interface, coupled with a terrace step of Ge. Using the STEM-EELS technique, we found the Ti diffusion into substrate at the dislocation core [Fig. 1]. However, this �a <011> type dislocation could be the projection of <111> direction which will be determined from the perpendicular direction (plane-view) sample. In the plane-view characterization, threading dislocations were observed with Burgers vectors of types a <100> and �a <110>. Considering the misfit dislocation with a Burgers vector of �a [011], the �a [011] misfit dislocation can be 1] dislocation at the (100) plane. We also interpreted as the projection of the in-plane segment of �a [11 found the dislocation reaction from two partial dislocations to one perfect dislocation �a [111] + �a ] = a [100], which leads to the formation of threading dislocation with Burgers vector of a [100] [111 [Fig. 2]. In addition, we found the coupling of dislocation half-loop with anti-phase boundary (APB, shift vector of �a <111>) induced by lattice terrace of Ge and they can decouple after annealing. The possible models based on the half-loop theory are proposed for the dislocation reaction and the coupling behavior. We have used combined (S)TEM and modeling analysis to solve the nature of the interfacial dislocation in perovskite-on-semiconductor system, which gives insight into the accommodation of the misfit strain from the view of TEM [3]. References [1] R. A. McKee, F. J. Walker, and M. F. Chisholm, Science 293, 468 (2001). [2] J. W. Matthews, and A. E. Blakeslee, J. Cryst. Growth 27, 118 (1974). [3] X. Shen et al., Appl. Phys. Lett. 106, 032903 (2015). [4] Electron microscopy studies were performed at Center for Functional Nanomaterials, Brookhaven National Laboratory supported by the U.S. Department of Energy, Office of Basic Energy Sciences, under contract no. DEAC02-98CH10886. X. S. is grateful for the financial support of the China Scholarship Council and Brookhaven National Laboratory for his exchange program. K. A. -M. and J. H. Microsc. Microanal. 21 (Suppl 3), 2015 1302 N. acknowledge the support of The University of Texas at Arlington. D. W. thanks National Key Basic Research Program of China (2015CB921203) for support. The authors thank Dr. Woodhead for the proof reading. Figure 1. (a) Cross-sectional STEM image with Burgers vector along [011] direction. The Burgers circuit is drawn around the dislocation core. Schematic diagram of step and terrace structure is shown in the inset. Elemental maps (50 pixels by 50 pixels) from STEM EELS of Ti (red), O (green), Ge (blue) and their RGB mixture are shown in (b), (c), (d), and (e), respectively. Figure 2. (a) Plane-view HRTEM image of SZTO thin films. HR-STEM images of the plane view are shown in (b), (c), and (d). The APBs are indicated by white dashed arrows. Schematic diagrams of dislocation dissociation and coupling behavior between APBs and dislocations are shown in (e) and (f), respectively.");sQ1[651]=new Array("../7337/1303.pdf","Imaging Local Polarization and Domain Boundaries in Multiferroic (LuFeO3)m/(LuFe2O4)n Superlattices","","1303 doi:10.1017/S1431927615007308 Paper No. 0651 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Imaging Local Polarization and Domain Boundaries in Multiferroic (LuFeO3)m/(LuFe2O4)n Superlattices Megan E. Holtz1, Julia A. Mundy1, Jarrett A. Moyer2, Charles M. Brooks3, Hena Das1, Alejandro F. Rebola1, Robert Hovden1, Craig J. Fennie1, Peter Schiffer2, Darrell G. Schlom3,4, David A. Muller1,4 1. 2. School of Applied and Engineering Physics, Cornell University, Ithaca, NY, USA Department of Physics and Materials Research Institute, University of Illinois at Urbana-Champaign, Urbana, IL, USA 3. Department of Materials Science and Engineering, Cornell University, Ithaca, NY, USA 4. Kavli Institute at Cornell for Nanoscale Science, Ithaca, NY, USA Materials that couple strong ferroelectric and ferromagnetic order hold tremendous promise for nextgeneration memory devices. However, many so-called multiferroics have properties that are either weak, emerge well below room temperature, and/or lack coupling between the electric and magnetic domains, stymieing technological applications. The atomic-scale design of new multiferroics, usually realized as heterostructures or interface phases, requires a local probe of physical properties and structure inside the material. This atomic-scale feedback on ferroelectric polarization and domain structure has helped lead us to a new strong ferrimagnet-ferroelectric with the highest known simultaneous transition temperatures. These (LuFeO3)m(LuFe2O4)n superlattices (Fig 1) are constructed by integrating the ferroelectric, antiferromagnetic LuFeO3 and paraelectric, ferrimagnetic LuFe2O4. The hexagonal LuFeO3 is an improper ferroelectric where the Lu-O buckles into a polar structure [1]. We quantify the polar structure--manifest as a displacement of the lutetium atoms--with atomic precision for different superlattice layerings. Our ferroelectric domain measurements show large polarization and regular domain walls correlate with improved magnetic moment and critical temperature, TC. Aberration-corrected scanning transmission electron microscopy (STEM) can provide atomic column positions with sub-picometer precision [2]. Here we apply atomic mapping with picometer precision to find local polarization and domain structure in (LuFeO3)m(LuFe2O4)n as a function of m,n in a 100 keV NION Ultra-STEM. Each image was formed from many (>20) cross-correlated fast acquisitions. 2D Gaussian fitting located the center of the lutetium columns, from which the displacement, and thereby polarization, was measured (Fig 2ab). After analyzing over 100 images, or > 4000 nm2 of the material, we find that the (LuFeO3)m(LuFe2O4)1 superlattices with m 2 display a ferroelectric structure. The polarization is damped for lutetium rows adjacent to the LuFe2O4 layers for 2 m 5, thus we would expect net polarization to grow with layer width. Indeed, the polarization of the parent-compound LuFeO3 is reached for m 3 multilayers and surprisingly is exceeded for m = 7,8 (Fig 2c). We collected domain sizes and boundary type statistics from the local polarization data. We found that domain size grows with m, increasing the width in the growth plane but limited in height by the LuFe2O4 layers, for up to m=7. At m=7, the domain height stabilizes at half the height of the (LuFeO3)7 layer, and forms regular domains with tail-to-tail walls at the LuFe2O4 layer and head-to-head walls in the middle of the (LuFeO3)7 layer. First-principle calculations suggest that the doping associated with this particular domain configuration leads to the increase in magnetic moment, which we have observed (Fig 3). TC is also seen to increase, approaching room temperature. STEM analysis combined with first-principles results suggests pursuing strongly ferroelectric domains with regular domain boundaries will lead to enhanced magnetization and TC in these multiferroics. [3] Microsc. Microanal. 21 (Suppl 3), 2015 1304 References: [1] H Das, et al, Nat. Comm. 5 (2014) 2889. [2] A Yankovich et al, Nat. Comm. 5 (2014) 4155. [3] Work supported by the U.S. DOE BES, Award #DE-SC0002334. EM Facility support from the NSF MRSEC program (DMR 1120296). Figure 1. Diagram of the Lu-O buckling in LuFeO3 (a), which leads to the ferroelectric structure [1]. ADF-STEM of the (LuFeO3)m (LuFe2O4)n viewed along the [110] zone axis, the (LuFeO3)n blocks highlighted in blue, for the m, n = 1, 1 structure (b), 2, 1 (c), 3, 1 (d), 4, 1 (e) and 5, 1 (f). Figure 2. (a) STEM image of (LuFeO3)m (LuFe2O4)n with color overlays for the magnitude and direction of the local ferroelectric displacements. (b) r.m.s. displacement (square) for each lutetium row. Color indicates mean polarization, with lengths corresponding to 20-80% of the distribution. (c) r.m.s displacement averaged over many images for varying (m,n), distinguishing between layers on the edge bordering the double iron layer and in the middle. Blue line is for LuFeO3. Figure 3. Magnetization and critical temperature for the (LuFeO3)m (LuFe2O4)n, plotted alongside ferroelectric domain area. Larger domain area correlates to higher magnetization.");sQ1[652]=new Array("../7337/1305.pdf","Cross-Sectional Characterization of SrTiO3/Si(001) Interfaces using Aberration-Corrected STEM","","1305 doi:10.1017/S143192761500731X Paper No. 0652 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Cross-Sectional Characterization of SrTiO3/Si(001) Interfaces using AberrationCorrected STEM HsinWei Wu,1 Toshihiro Aoki, 2 Agham B. Posadas,3 Alexander A. Demkov, 3 and David J. Smith,4 1. School of Engineering for Matter, Transport, and Energy, Arizona State University, Tempe, AZ 85287 LeRoy Eyring Center for Solid State Science, Arizona State University, Tempe, AZ, 85287 3. Department of Physics, The University of Texas at Austin, Austin, TX 78712 4. Department of Physics, Arizona State University, Tempe, AZ, 85287 2. SrTiO3 (STO) is the only perovskite oxide that can be epitaxially grown directly on Si making it an excellent intermediate layer for the integration of other functional perovskite oxides on Si. STO grows such that there is a 45� relative rotation with the underlying Si resulting in a small lattice mismatch of 1.7%, with the STO under compression [1]. To date, the exact atomic structure of the interface is still not settled and this study can shed more light about the physical structure of the STO/Si(001) interface. In this work, 5 monolayers of single-crystal STO were grown directly on Si(001) by molecular beam epitaxy (MBE) with a variant of the Motorola-developed process. After cleaning and desorbing the native oxide layer, a half monolayer of Sr was deposited at 550 C on a clean Si(001) surface with 2 1 reconstruction. Sr and Ti were then co-deposited at 200 �C to the desired thickness with oxygen ramping. The STO layer was then crystallized by annealing for 5 minutes under vacuum at 550 C [2]. Samples for TEM observation were prepared via standard cross-section method with ion milling. Aberrationcorrected STEM images to characterize the interface were recorded with a JEOL ARM 200F operated at 200 keV. Figures 1(a) and (b) are BF and HAADF STEM images showing the sample structure with a thin STO film on top of Si substrate. The projection orientation is [100] for the STO layer and [110] for the Si. A short STO section of different <110> orientation, as arrowed in figure 1(a), as well as slight crystal rotations are visible. These defects may be formed to release strain within the STO layer because of the lattice mismatch with Si. A vertical offset in the STO layer can also be observed, as arrowed in figure 1(b), with a magnified image of the marked area in figure 1(c). The line profile shown in figure 1(d) indicates that an Sr atom column is adjacent to a Ti atom column to the right of the arrow. In figure 2, two different Si terminating surfaces are observed at the interface. Half Si dimers are visible at the STO/Si interface in figures 2(a) and (b). Figure 2(e) is the line profile after Gaussian blur of figure 2(b) showing Si dimers in the substrate and a half dimer at the interface with adjacent Ti. The distance between the Si dimer and the first Ti atom column is ~0.384 nm. On the other hand, full Si dimers are visible continuously from the substrate to the interface in figures 2(c) and (d). The same result can be seen in the line profile, as shown in figure 2(f) after Gaussian blur. The distance between the last Si dimer and the first Ti atom column is ~0.386 nm [3]. References: [1] J.H. Hao, et al., Appl. Phys. Lett. 87 (2005) 131908. [2] L. Ji, et al., Nat. Nanotechnol. 10 (2015) 84. [3] This work was supported by AFOSR Contract FA 9550-12-10494. We gratefully acknowledge the use of facilities within the John M. Cowley Center for High Resolution Electron Microscopy at Arizona State University. Microsc. Microanal. 21 (Suppl 3), 2015 1306 (a) STO (b) Si (c) (d) Fig. 1. (a) BF, and (b) HAADF, images showing a short section of different STO orientation (arrowed in the BF image) and a vertical offset in the STO unit cell (arrowed in the HAADF image); (c) Magnified image of area marked in (b); (d) Line profiles of (c) showing the vertical intensity offset. (a) (b) (c) (d) STO STO Si Si (e) (f) Fig. 2. (a) BF, and (b) HAADF, images of SrTiO3/Si interface with half Si dimers. (c) BF, and (d) HAADF, images showing full Si dimers at the SrTiO3/Si interface. (e) Line profile of (b) after Gaussian blur showing half Si dimer at the interface. (f) Line profile of (d) after Gaussian blur showing full Si dimer at the interface.");sQ1[653]=new Array("../7337/1307.pdf","Atomic-scale Mechanisms of Defect-Induced Retention Failure in Ferroelectric Materials","","1307 doi:10.1017/S1431927615007321 Paper No. 0653 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic-scale Mechanisms of Defect-Induced Retention Failure in Ferroelectric Materials L. Li1, L. Xie1, Y. Zhang1, J. R. Jokisaari1, H.M. Christen2, and X. Q. Pan3,* 1 Department of Materials Science and Engineering, University of Michigan, Ann Arbor, MI 48109, United States 2 Materials Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831, United States 3 Department of Chemical Engineering and Materials Science and Department of Physics and Astronomy, University of California - Irvine, Irvine, CA 92697, US The ease of reversibly switching the spontaneous polarization of a ferroelectric with an applied electric field has made this material attractive for application of high-density nonvolatile memories. One of the major challenges impeding this application, however, has been the socalled "retention failure" phenomenon � a self back-switching process of the written polarization that can lead to data loss. An understanding of the atomic-scale mechanism of the retention failure process is thus necessary to engineer reliable ferroelectric devices. Here, using in situ transmission electron microscopy (TEM), we report direct observation of polarization backswitching induced by non-stoichiometric defects that commonly exist in ferroelectrics. Our results of atomic-resolution scanning transmission electron microscopy (STEM) show a novel mechanism of the retention failure process, revealing that the process is induced by the strong atomic interaction between the defects and the surrounding domains. BiFeO3 (BFO) thin film of 50 nm in thickness with 8 nm thick La0.7Sr0.3MnO3 (LSMO) bottom electrode was grown on LaAlO3 substrate by pulsed laser deposition. In this film, BFO possesses a tetragonal-like (T-like) structure (Fig. 1a) in most regions, and a large density of nonstoichiometric defects is observed within the 15 nm thick region above the BFO/LSMO interface. Local switching was performed upon applying a bias between a tungsten surface probe and the LSMO bottom electrode (Fig. 1b). The bias was linearly increased from 0 to 10 V during 20s and then removed. A plot of the measured switched domain area versus time during the switching and back-switching process is shown in Fig. 1c. Corresponding selected TEM images in Fig. 1d show the formation of a large "up" polarized domain in the "down" polarized matrix during the initial 0 to -10 V ramp and back-switching to a much smaller stable domain pinned by defects after the removal of the bias. To understand the underlying mechanism of the backswitching process, atomic resolution high-angle annular dark-field (HAADF) STEM was performed to image a polarized BFO region interacting with a defect, as shown in Fig 2a. Atomic displacements of Fe cations from the center of four Bi neighbors for BFO, which are proportional to the polarization, [1] are measured and shown in Fig. 2b. Due to a strong atomic interaction between the BFO and the defect, significantly enhanced out-of-plane polarizations were observed for the first BFO lattice layer that is in direct contact with the defect. Such strong interaction would cause a large downward built-in field pointing to the defect and thus destabilize the "up" polarized written domain and cause retention failure. In conclusion, using in situ TEM combined with atomic resolution STEM, we have shown a strong interaction between ferroelectric polarizations and the non-stoichiometric defects, which Microsc. Microanal. 21 (Suppl 3), 2015 1308 can cause a large built-in field, leading to local polarization enhancement and destabilizing written domains. Our results provide new insights into the critical role of defects on domain configurations and dynamics and suggest a new route to design ferroelectric devices through defect engineering. References: [1]. Li, L., et al., Nano Letters, 2013. 13(11) 5218-5223. [2]. The authors gratefully acknowledge the financial support through DOE grant DoE/BES DEFG02-07ER46416. Applied voltage a Tetragonal-like structure of BiFeO3 Bi O Fe b Mobile probe c Switched domain area (nm ) (0V 16000 12000 8000 4000 0 0 10 10 V ) ( 0 V 0V) V Planar defects BiFeO3 P[001] La0.7Sr0.3MnO3 LaAlO3 LSMO 2 20 Time (s) 30 d 1) 0v 50 nm P P 2) -4.0 V 3) -10.0 V 4) 0V BiFeO3 La0.7Sr0.3MnO3 LaAlO3 P P P P Figure 1. (a) Atomic models of the tetragonal-like structure BiFeO3. (b) Schematic of in situ TEM experimental set-up. (c) Plot of the measured switched domain area versus time extracted from an in situ TEM video. (d) A chronological TEM image series showing the evolution of a written "up" polarized domain in a "down" polarized matrix. a b Defect BiFeO3 80 dz dx Atomic displament (pm) 60 40 20 BiFeO3 1 nm Defect 0 0 2 4 Unit cell 6 8 10 12 Figure 2. (a) A HADDF STEM image of a polarized BFO region that is in contact with a planar non-stoichiometric defect. (b) Plot of the measured atomic displacements (dz: out-of-plane atomic displacement, dx: in-plane atomic displacement) of Fe cations from the center of four Bi neighbors for BFO atomic structures shown in (a).");sQ1[654]=new Array("../7337/1309.pdf","Characterization of Two-Dimensional Electron Gas at the -AlO/SrTiO Interface","","1309 doi:10.1017/S1431927615007333 Paper No. 0654 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of Two-Dimensional Electron Gas at the -AlO/SrTiO Interface Sirong Lu1, Kristy J. Kormondy2, Thong Q. Ngo3, Toshihiro Aoki4. Agham Posadas2, John G. Ekerdt3, Alexander A. Demkov2, Martha R. McCartney5, and David J. Smith5 1. 2. 3. 4. 5. School of Engineering for Matter, Transport and Energy, Arizona State University, Tempe, Arizona 85287-6106, USA Department of Physics, University of Texas at Austin, Austin, Texas 78712, USA Department of Chemical Engineering, University of Texas at Austin, Austin, Texas 78712, USA Le-Roy Eyring Center for Solid State Science, Arizona State University, Tempe 85287-1704, USA Department of Physics, Arizona State University, Tempe, Arizona 85287-1504, USA The interfaces between certain insulating transition-metal oxides exhibit a board range of properties, such as ferromagnetism, magnetoresistance, conductivity and superconductivity, making it possible to design all-oxide electronic devices [1]. The crystalline -Al2O3/SrTiO3 forms a highly conductive layer (two-dimensional electron gas - 2DEG), attributed to oxygen vacancies, with a maximum roomtemperature sheet-carrier density as high as ~7x1013 cm-2 [2]. Here we use high-resolution TEM negative-Cs imaging (NCSI), high-angle annular-dark-field (HAADF) STEM imaging, energy-loss near-edge structure (ELNES) analysis, and off-axis electron holography, to characterize the nature of Al2O3/SrTiO3 interfaces, as grown by atomic layer deposition (ALD) and molecular beam epitaxy (MBE) at different temperatures. The effects of post-deposition annealing with or without the presence of oxygen, was also investigated. Figures 1(a) and (b) show NCSI and HAADF images of a 2.1-nm-thick Al2O3 layer grown by ALD on a TiO2-terminated STO (001) substrate at 345�C: this sample has a 2DEG at the interface. The NCSI (a) is taken at the [110] zone axis. The Sr and O mixed atomic columns in the substrate are the strong bright spots, the O atomic columns are the weak spots, and the Ti atomic column intensity is in between. The HAADF image (b) is taken at the [100] zone axis. The Sr atomic columns are bright spots, the Ti atomic columns are weaker, the Al atomic columns that form a square lattice are very weak, and the O atomic columns are not visible. From both images, it is clear that the interface at the substrate side is TiO2terminated, and the symmetry of the film is consistent with -Al2O3. The plots below the images are the auto-correlation results calculated from the images of the terminating TiO2 layer, showing that the repeating patterns correspond to TiO2. Figures 1(c), (d) are NCSI and HAADF images showing the sample grown by MBE at 600�C and annealed in air for an hour at 400�C. The 2DEG was not present in this sample after annealing. From the NCSI, the TiO2-terminated layer at the interface is different from (a). The auto-correlation results also show that the repeating patterns of the TiO2 surface layer have disappeared. From the HAADF image, the Ti contrast in the same layer has reduced, and the position of the invisible O atomic column shows some contrast. Both results indicate that Ti has diffused away from this layer, and that Al has diffused into this layer, thus suggesting the possibility that the disappearance of the 2DEG after annealing is related to the diffusion of Al and Ti atoms near the interface. Figure 2 shows the ELNES scan across the interface of the sample shown in Figs. 1(a) and (b). Although extracting conclusive evidence of oxygen vacancies from imaging is not possible in these images, because the signal from random and/or low concentration of oxygen vacancies is submerged into the noise as well as unevenness in sample thickness, the ELNES signal clearly shows that the Ti oxidation state near the interface has changed, which could be taken as indirect evidence of oxygen vacancies. The Microsc. Microanal. 21 (Suppl 3), 2015 1310 Ti4+ signals (with 4 peaks) in Figs. 2 (b) and (c) are very obvious in the substrate, while the Ti3+ signal (with 2 peaks) in Fig. 2 (a) shows up in a region less than one unit cell from the interface on the STO side. Figure 3(a) is a reconstructed thickness line profile from the sample shown in Figs. 1(a), (b), and Fig. 3(b) shows the reconstructed phase image. The thickness profiles of the substrate and the film (except near the surface) are almost constant. The arrow pointing at the positive curvature in the phase image indicates the position of the possible 2DEG. Further quantitative analysis and simulations are still in progress to better understand and interpret these results [3]. References: [1] Y.Z. Chen, et al., Nature Communications 4 (2013) 1371. [2] K.J. Kormondy, et al., J. Applied Physics 117 (2015) in press. [3] This work was supported by AFOSR Contract FA 9550-12-10494. We gratefully acknowledge the use of facilities within the John M. Cowley Center for HREM at Arizona State University. (a) (b) (c) (d) Fig.1. Images of -Al2O3/SrTiO3 (001) interface. Sample with 2DEG: (a) NCSI; (b) HAADF. Sample without 2DEG: (c) NCSI; (d) HAADF. STO unit cell structure is superimposed on the image. Arrows point to the terminating TiO2 layer. The plot below each image is the calculated auto-correlation result for the terminating TiO2 layer. Fig.3. Reconstructed thickness (a), and phase (b), (a) (b) (c) line profiles obtained from off-axis electron Fig.2. ELNES spectra near the interface at the STO holography. White arrow points to the positive side. Distance to interface: (a) within 1 unit cell; (b) curvature. 2 unit cells; (c) 3 unit cells.");sQ1[655]=new Array("../7337/1311.pdf","Atomic Scale Analysis of Terrestrial and Extra-Terrestrial Geomaterials Using Atom Probe Tomography","","1311 doi:10.1017/S1431927615007345 Paper No. 0655 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic Scale Analysis of Terrestrial and Extra-Terrestrial Geomaterials Using Atom Probe Tomography Brian P. Gorman1, David R. Diercks1, Stephen Parman2, and Reid Cooper2 1. Department of Metallurgical and Materials Engineering, Colorado School of Mines, Golden, CO, USA 80401. 2. Department of Earth, Environmental, and Planetary Sciences, Brown University, Providence, RI, USA 02912. From the electron microprobe to the secondary ion microprobe to laser-ablation ICP-MS, steady improvements in the spatial resolution and chemical detection limits of geochemical micro-analysis have been central to generating new discoveries. Continual improvements in instrumentation and experimental technique have now allowed Atom Probe Tomography (APT), and particularly laser assisted field evaporation, to open new areas of nanoscale analysis in inorganic geomaterials. Specifically, APT provides sub-nm scale spatial resolution in three dimensions with ppm level detection limits and typically less than one Da mass resolution [1]. Despite these improvements in APT, silicates and other metal-oxide materials prevalent in geomaterials still present some analytical challenges due to their electrical resistance, low thermal conductivity, and strong metal-oxygen bonds. As seen previously, oxide materials tend to field evaporate as cluster ions containing cations bonded to oxygen [2,3]. In previous investigations, we have successfully analyzed a range of olivine compositions from Fo0 to Fo90, however, the nature of oxide field evaporation does not typically allow for atomic scale resolution reconstructions as was observed in platinum group alloys [4]. In this work, we show the first APT analysis of an extra-terrestrial silicate glass bead acquired during the Apollo 15 space mission. Astronauts Dave Scott and Jim Irwin acquired the low-Ti green glass sphere on the Apennine Front. SIMS analyses of these glass beads has recently provided the first evidence for high volatile contents in the lunar interior [5]. APT analysis was conducted on a the surface of a single bead after FIB specimen preparation (Figure 1, [6]) using a laser pulse energy of 10 pJ and a 500 kHz repetition rate. A representative mass spectrum is shown in Figure 2. Peaks are resolved with a mass resolution better than 0.25 Da. Similar to other silicate chemistries analyzed, the glass bead tends to field evaporate as a mixture of Si, Mg, Fe, Ca, Al and O, as well as their associated monoxide clusters. Substantial mass interferences occur at m/q of 28 (Si+, Fe++, CaO++), 56 (2Si+, Fe+, Ca+), 40 (Ca+, MgO+) and 20 (Ca++, MgO++). At this point, these interferences cannot be resolved. Al has little intereferences and simple counting of Al in the analyses (7.45 at% Al) agrees well with published microprobe analyses (8.39 at%). The elements appear to be distributed randomly in x-y-z space (Figure 3). Nearest neighbor analysis from all of the species illustrates some Mg segregation at less than 0.5 nm scales, at the expense of Si species (Figure 4). All of these results suggest that the glass was well homogenized prior to ejection but the solidification rate was intermediate, allowing some Mg segregation, which is again not unexpected based on the high Mg concentration in the melt [7]. References: [1] M. K. Miller, Atom Probe Tomography: the Local Electrode Atom Probe, Springer (2014). [2] R. Kirchhofer, et.al., J. Am. Cer. Soc., 97 (2014) 2677. Microsc. Microanal. 21 (Suppl 3), 2015 1312 [3] R. Kirchhofer, et.al., J. Nuc. Mat., 436 (2013) 23. [4] S. W. Parman, et.al., Am. Mineralogist, in print (2015). [5] A. E. Saal, et al., Nature, 454 (2008) 192. K. Thompson, et.al., Ultramicroscopy, 107 (2007) 131. [6] The APT instrumentation used in this research was enabled via NSF grant DMR-1040456 Mg Si Figure 1. FIB induced secondary electron image of the APT specimen preparation, showing a section lifted out from the lunar glass bead. Fe Al Figure 3. Atom probe tomography reconstruction (same volume) illustrating the distribution of Mg, Si, Fe, and Al species. Visual inspection implies a homogeneous distribution. Figure 2. Mass spectrum of a lunar glass bead specimen acquired using laser assisted APT. Mass resolution better than 0.25 Da is illustrated. Figure 4. Nearest neighbor analysis of the different ionic species. Mg appears to be slightly segregated within 0.5 nm, which is compensated by a decrease in Si species.");sQ1[656]=new Array("../7337/1313.pdf","Correlative Transmission Electron Microscopy and Atom-Probe Tomography of an Iron Meteorite.","","1313 doi:10.1017/S1431927615007357 Paper No. 0656 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Correlative Transmission Electron Microscopy and Atom-Probe Tomography of an Iron Meteorite. Surya S. Rout1,2, Philipp R. Heck1,2, Dieter Isheim4, Thomas Stephan2,3, Andrew M. Davis2,3, and David N. Seidman4. 1 Robert A. Pritzker Center for Meteoritics and Polar Studies, The Field Museum of Natural History, 1400 S Lake Shore Drive, Chicago IL 60605, USA 2 Chicago Center for Cosmochemistry, Chicago IL 60637, USA 3 Department of Geophysical Sciences, The University of Chicago, 5734 S Ellis Avenue, Chicago IL 60637, USA 4 Northwestern University Center for Atom-Probe Tomography, and Department of Materials Science & Engineering, Northwestern University, 2220 Campus Drive, Evanston, IL 60208, USA Correlative use of transmission electron microscopy (TEM) and atom-probe tomography (APT) provides complementary structural and chemical information with a unique combination of atomic-level spatial resolution and single-atom analytical sensitivity [1,2]. Correlated TEM/APT studies are state of the art in material science [3]. In our effort to broaden the application of APT in cosmochemistry [4], we expand our range of projects from meteoritic nanodiamonds [5] and presolar silicon carbide [6,7] to other samples. Here, we present new data from a correlated TEM/APT study of an iron meteorite [8]. Iron meteorites are Fe-Ni alloys with minor amounts of Co, P, S, and C. They crystallized with extremely slow cooling rates (0.2 to 6000 K/Ma) within the metallic cores of asteroids during which the characteristic Widmanst�tten pattern forms by nucleation and growth of Ni-poor kamacite (-bcc, ferrite) from taenite (-fcc, austenite). The phase transformation occurring below 400�C involves a eutectoid reaction that results in formation of tetrataenite (FeNi) at the kamacite-taenite (K-T)_interface. The microstructure of the K-T interface and tetrataenite has been studied in detail using TEM [9]. One of our motivations is to take advantage of the superior spatial resolution and sensitivity of APT to study the composition of the K-T region on the nanoscale within fast cooled iron meteorites. We prepared a polished section from the Bristol IVA iron meteorite (Field Museum specimen ME2248),. Bristol cooled at a relatively fast rate (~250 K/Ma; [10]) and experienced comparably low shock pressures (13 GPa; [10]). We used a Zeiss EVO 60 SEM equipped with an Oxford AZtec SDD EDS system to image the polished section. A Zeiss 1540 XB FIB-SEM was utilized to prepare sharp nanotips for APT from a 105 �m2 lamella perpendicular to the K-T interface. The nanotips were attached to a copper half-grid with five presharpened posts. The grid was attached to a tomographic tip, fitted to a tomographic TEM holder (Hummingbird Scientific). TEM analysis was performed using a FEI Tecnai F20ST TEM prior to APT. APT was performed using a Cameca LEAP 4000XSi. The TEM image of a tip (Fig. 1a) shows the presence of a clear interface, which was identified as the kamacite/taenite interface with EDS and electron diffraction. The 3-D tomographic reconstruction (Fig. 1b) shows isoconcentration surfaces for Ni and Co; 3-D surfaces delineating regions with a concentration greater than a given threshold value for each element. The reconstruction shows clearly that Fe-rich (magenta dots) and Ni-rich (green) isoconcentration surface regions are separated by an interface, which represents the K-T interface, also seen in the TEM image (Fig. 1a). In a concentration profile obtained from the atom-probe data perpendicular to the K-T interface, the Fe concentration Microsc. Microanal. 21 (Suppl 3), 2015 1314 decreases from ~95.6 to 45.2 wt%, and the Ni concentration increases from 5 to 54.7 wt%. The high Ni concentrations of 54.7 wt% had not been detected before along a K-T interface within a IVA iron meteorite. Co is concentrated in kamacite and decreases from 0.5 wt% in kamacite to 0.02�0.2 wt% in taenite. High Ni concentration was measured in the Santa Clara IVB iron meteorite using atom probe field ion microscopy [11] � a non-tomographic technique � and also within a mesosiderite (65.5�3 wt%) using TEM [12]. We also observed variations in Ni and Fe concentrations within the tetrataenite suggesting the presence of inhomogeneities or nanostructures within it. The high Ni concentrations we measured in tetrataenite confirm predictions by the Fe-Ni phase diagram. C Figure 1. (a) TEM image of a nanotip prepared from the K-T interface. The region of APT reconstruction is shown with dashed lines. Corresponding APT reconstruction (b) with isoconcentration surfaces of Ni (green), and cobalt (dark blue) along with the distribution of Fe atoms (magenta dots) is shown. (c) Concentration profile along the long axis of the prism (cyan) shown in (b). References [1] Seidman D. N. and Stiller K. MRS Bulletin 34 (2009), p. 717�724. [2] Kelly T. F. and Larson D. J. Annual Review of Material Science 42 (2012), p. 1�31. [3] Moutanabbir O. et al, Nature 496 (2013), p. 78�82. [4] Heck P. R. et al, 45th Lunar and Planetary Science Conference (2014), #1811. [5] Heck P. R. et al, Meteoritics & Planetary Science 49 (2014), p. 453�467. [6] Heck P. R. et al, 41st Lunar and Planetary Science Conference (2010), #2112. [7] Stadermann F. J. et al, 41st Lunar and Planetary Science Conference (2010), #2134. [8] Rout S. S. et al, 46th Lunar and Planetary Science Conference (2015), #2938. [9] Goldstein J. I. et al, Chemie der Erde 69 (2009), p. 293�325. [10] Goldstein J. I. et al, Meteoritics & Planetary Science 44 (2009), p. 343�358. [11] Miller M. K. and Russell K. F. Surface Science 226 (1992), p. 441�445. [12] Reuter K. B. et al, Geochimica et Cosmochimica Acta 61 (1988), p. 2943�2956. [13] Acknowledgements: We thank D. Miller and N. Zaluzec for their help with FIB and TEM at the Electron Microscopy Center, Argonne National Laboratory. Use of the Center for Nanoscale Materials was supported by the U. S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357. LEAP was done at NUCAPT, a shared facility at the Materials Research Center of Northwestern University. The LEAP was purchased and upgraded with funding from NSF-MRI and ONR-DURIP. PRH acknowledges funding from the Tawani Foundation and from W. H. Ganz III.");sQ1[657]=new Array("../7337/1315.pdf","Atomic Scale Investigation of Orthopyroxene and Olivine Grain Boundaries by Atom Probe Tomography","","1315 doi:10.1017/S1431927615007369 Paper No. 0657 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic Scale Investigation of Orthopyroxene and Olivine Grain Boundaries by Atom Probe Tomography Mukesh Bachhav1, Yan Dong1, Philip Skemer2, Emmanuelle A. Marquis1 1 Department of Materials Science and Engineering, University of Michigan, Ann Arbor, MI 48109, United States 2 Department of Earth and Planetary Sciences, Washington University, One Brookings Drive, Saint Louis, Missouri, 63130, United States Olivine (Mg1.8Fe0.2)SiO4 and orthopyroxene (Mg0.9Fe0.1)SiO3 are the most abundant minerals in Earth's upper mantle and represent important geochemical reservoirs in Earth. In addition, the rheological properties of these two minerals exert first order control on mantle convection and the dynamics of Earth's tectonic plates [1, 2]. Both olivine and orthopyroxene have nominal compositions that include Mg, Fe, Si, and O. However due to extensive solubility of non-stoichiometric elements, both minerals may contain a number of impurities, including Ca, Al, Mn, Cr. During deformation, dynamic recrystallization generates new grain-boundaries, which then migrate due to differences in dislocation strain energy between adjacent grains. It has been shown that this grain boundary migration will cause impurities to become segregated along new grain boundaries. Thus, grain boundaries may represent important additional geochemical reservoirs in Earth. Accurate chemical analyses at grain boundaries using standard microscopic techniques are challenging, especially for poor conducting geological samples [3] . Atom probe tomography (APT) is unique technique to elucidate chemistry and 3-D distribution of elements from sample at nanometer length scale. With advances in laser and sample preparation techniques in last decade, APT is now successfully applied to wide range of insulating materials like metal oxides, ceramics, and more recently biological minerals [4]. Site-specific specimens for APT were prepared from experimentally recrystallized orthopyroxene and olivine using a dual beam focused ion beam (FIB) lift-out technique. Analyses were performed in laser mode with laser energy of 50 pJ/pulse, repetition rate of 200 kHz and detection rate of 1 %. 3D distribution of Fe, Al, and Ca is shown in Figure 1 for both phases; a 1D distribution profile across grain boundary for Mg, Si, and O along with several low-abundance impurities is shown in Figure 2. It is clear from figure 2, that orthopyroxene shows extensive segregation of Fe, Al and Ca along with a depletion of Mg, Si and O at grain boundary. Similar segregation of impurities is observed at grain boundary for olivine phase with no significant change in composition for Mg, Si and O. The effective grain boundary width is determined to be ~4 nm, which is a factor of two more narrow than detected using other microscopic techniques. These results are instrumental in understanding the geochemistry of the mantle, and may have a significant impact on deformation, recrystallization, and microstructural evolution. References: [1] G. Hirth, D. Kohlstedt, Rheology of the Upper Mantle and the Mantle Wedge: A View from the Experimentalists, in: Inside the Subduction Factory, American Geophysical Union, 2013, pp. 83-105. [2] S.-I. KARATO, P. Wu, Science, 260 (1993) 771-778. [3] T. Hiraga, I.M. Anderson, D.L. Kohlstedt, Nature, 427 (2004) 699-703. [4] M.N. Bachhav, S.-R. Chang, A. McFarland, E.A. Marquis, B. Clarkson, Microscopy and Microanalysis, 19 (2013) 184185. Microsc. Microanal. 21 (Suppl 3), 2015 1316 30 nm Ortho-pyroxene Grain boundary Fe Al Ca Olivine 40 nm Figure 1: 3D distribution of Fe, Al, Ca in orthopyroxene and olivine samples at grain boundaries. (a) 60 58 Ortho-pyroxene (b) 56 54 Olivine Concentration (%) Concentration (%) 56 54 52 52 50 28 24 20 16 12 Grain boundary Mg Si O Mg Si O 20 Grain boundary 16 0 4.8 4.2 2 4 6 8 10 12 14 16 18 20 0 4.8 4.4 4.0 3.6 0.4 2 4 6 8 10 12 14 16 18 20 Distance (nm) Distance (nm) Fe Mn Ca Al Cr Grain boundary Concentration (%) 3.6 3.0 2.4 1.8 1.2 0.6 0.0 0 2 4 6 8 10 12 14 16 18 20 Concentration (%) Grain boundary Fe Mn Ca Al 0.2 0.0 0 2 4 6 8 10 12 14 16 18 20 Distance (nm) Distance (nm) Figure 2: 1D distribution profile of Mg, Si, O, Fe, Ca, Mn, Al, Cr along grain boundary for (a) Orthopyroxene and (b) Olivine.");sQ1[658]=new Array("../7337/1317.pdf","Quantitative microstructural analysis of geological materials by atom probe: understanding the mechano-chemical behaviour of zircon","","1317 doi:10.1017/S1431927615007370 Paper No. 0658 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantitative microstructural analysis of geological materials by atom probe: understanding the mechano-chemical behaviour of zircon Julie M. Cairney1, Alexandre La Fontaine1, Patrick Trimby1, Limei Yang1, Sandra Piazolo2, Australian Centre for Microscopy & Microanalysis, University of Sydney, Australia Australian Research Council Centre of Excellence for Core to Crust Fluid Systems/GEMOC, Department of Earth and Planetary Sciences, Macquarie University, Australia 2 1 The mineral zircon (ZrSiO4) is ideally suited for radiogenic dating of rocks. Not only does it contain trace amounts of uranium and thorium, enabling dating using radioactive decay in the U-Pb system, but it is also a very robust mineral. It is generally believed to survive a range of geological processes such as erosion, deformation and high-grade metamorphism (up to 900 �C). The spatial resolution for zircon dating using conventional techniques (e.g. sensitive high resolution ion microprobe (SIMS)) is in the range of 10-20 �m, however several recent studies have suggested that zircons may not be as chemically robust as once believed, especially on the micron and sub-micron scale. Atom probe tomography (APT) is a powerful microscopy technique that can provide 3D maps showing the position and atomic mass of individual atoms with sub-nanometre resolution [1]. It is ideal for studying the spatial distribution of atoms across small volumes (the total sample area is normally only a few hundred nanometres in size). Isotopic sensitivity is also an advantage for its application in the study of geological materials. It has recently been applied for the first time to U/Pb isotope dating [2], where the results were found to agree well with SIMS. Here, we will provide an overview of the technique of atom probe and present novel APT analyses of deformed zircons for comparison with previous studies of this material by electron backscatter diffraction (EBSD), cathodoluminescence (CL) and radiogenic U-Pb dating studies. EBSD analysis has revealed that these zircons show evidence of severe intracrystalline deformation, fracturing and grain size reduction as well as a large spread in U-Pb ages. Low angle boundary networks that form up to 100 �m wide deformation zones exhibit significant disturbances of the otherwise homogeneous CL signature. These regions of low angle grain boundary networks show moderate chemical resetting of the original ca. 900 Ma old zircons. [3]. Correlation of the disturbance of the CL signature and U-Pb derived ages with lattice distortions and grain boundaries has been interpreted to originate from accelerated lattice diffusion through the highly distorted crystal lattice. Such enhanced pipe diffusion would proceed at rates 1-3 orders of magnitude faster than the lattice diffusion in an undeformed lattice. Alternatively, fluid-mediated coupled dissolution and precipitation [4] or just precipitation of zircon in chemical equilibrium with a fluid at the time of deformation at low and high angle boundaries may result in presence of zircon with different chemistry. In the latter case, changes in chemistry and therefore time signature would occur instantaneously (in geological timescales). Our atom probe analysis from distorted zircons shows the decoration of dislocations by Al suggesting that the chemical signature in zircon is indeed influenced by defects (Fig. 1b). Furthermore, APT reveals nanoscale zircon particles with distinctly different chemical signature in terms of Al and H content at high angle boundaries. The presence of such Microsc. Microanal. 21 (Suppl 3), 2015 1318 particles suggests rapid changes of zircon chemistry by fluid mediated precipitation and/or dissolution and precipitation. In our paper, we will also present the results of ongoing studies of older samples, to discover whether lead displays similar segregation behavior, shedding new light onto the robustness of zircons for dating. We will also discuss the strengths and limitations of the technique of atom probe for the study of geological materials, describe our approach to the challenging issue of site-specific specimen preparation, and show the results of experiments designed to optimize the quality of the data obtained from zircon. References: [1] Gault, Moody, Cairney, Ringer, "Atom probe microscopy", (Springer). [2] Kelly and Larson, Ann. Rev. Mat. Sci., 2012, 42, p. 1. [3] Piazolo, Austrheim, Whitehouse, American Mineralogist, 2012, 97, p. 1544. [4] Putnis & Putnis, 2007, J. Solid State Chem., 180, p. 1783. Figure 1. Atom probe tomography (APT) reconstructed volume of a distorted Zircon with a sub-grain boundary decorated with Aluminium with a 0.5 at.% Al iso-concentration map. The dataset is shown from two different angles, so that the low angle boundary can be distinguished from the dislocation.");sQ1[659]=new Array("../7337/1319.pdf","Electron Probe Microanalysis and Microscopy: Focused Electron Beam Techniques and Applications in Characterization of Mineral Inclusions in Diamond","","1319 doi:10.1017/S1431927615007382 Paper No. 0659 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electron Probe Microanalysis and Microscopy: Focused Electron Beam Techniques and Applications in Characterization of Mineral Inclusions in Diamond Donggao Zhao Department of Geological Sciences, Jackson School of Geosciences, University of Texas at Austin, Austin, TX 78712, USA. Electromagnetic lens-focused electron beam, commonly called electron probe, is widely used in microanalysis and microscopy [1-6]. For example, electron probe is exclusively used in a scanning electron microscope (SEM) to acquire information of samples on images at micro- to nano-scales (BSE, SE, CL, etc.), chemistry (EDS, WDS, etc.), and crystal structure (EBSD). Electron probe, not parallel electron beam, is also used in STEM mode in a transmission electron microscope (TEM) to acquire similar information of samples (bright field, dark field, HAADF, EFTEM, EDS, EELS, electron diffraction, etc.) at nano- to pico-scales. All these techniques in microanalysis and microscopy use an electron probe as primary input signal to bombard a sample; and the signals detected can come from above the sample surface (SEM) or below the sample if an electron probe penetrates through the sample (STEM). Therefore, the techniques mentioned above all belong to electron probe microanalysis and microscopy or EPMM in short. In EPMM techniques, the electron probe scans across the sample and this dynamic process is different from the static process in the parallel electron beam techniques (TEM mode). Use EPMM techniques, a variety of silicate, carbonate and sulfide inclusions were recovered in the diamond from the No. 50 kimberlite diatreme of Liaoning Province, China (Fig. 1) [7, 8]. These inclusions include, in order of abundance, olivine, chromite, garnet, orthopyroxene, Ca carbonate, magnesite, dolomite, norsethite, pyrrhotite, pentlandite, troilite, a member of the linnaeite group, an unknown hydrous magnesium silicate and an iron-rich phase. Abundance and composition of the mineral inclusions in diamond indicate that they belong to peridotitic suite and are mainly harzburgitic. No eclogitic mineral inclusions were found in the diamond. The slightly lower Mg # of the olivine inclusions (peak at 93) than that of harzburgitic olivine inclusions worldwide (Mg # peak at 94), the higher Ni content (0.25-0.45 wt %) of the olivine inclusions than those of olivine inclusions worldwide, the higher Ti contents (up to 0.79 wt %) in some chromite inclusions in diamond than those in chromite inclusions worldwide, the existence of carbonate inclusions in diamond, and the possible presence of hydrous silicate phases in diamond all indicate a metasomatic enrichment event in the source region of diamond beneath the North China craton. Sulfide inclusions in diamond often coexist with chromite and olivine or are rich in Ni content, indicating that the sulfide inclusions belong to the peridotitic suite. From the chemical compositions, most sulfide inclusions in diamond from the No. 50 kimberlite were probably trapped as monosulfide crystals, although some may have been entrapped as melts. The pressure and temperature obtained from applicable thermobarometers indicate that the diamond crystallized in the range of 980�C and 45 kbar to 1220�C and 69 kbar, corresponding to a depth of 140 to 200 km. References: [1] J. Goldstein et al., Scanning Electron Microscopy and X-Ray Microanalysis (Third Edition). Kluwer Academic Publishers (2003). Microsc. Microanal. 21 (Suppl 3), 2015 1320 [2] S.J.B. Reed (2005) Electron Microprobe Analysis and Scanning Electron Microscopy in Geology (Second Edition). Cambridge University Press (2005). [3] D.B. Williams et al., Transmission Electron Microscopy � A Textbook for materials Science (Second Edition). 822 pages. Plenum Press (2009). [4] D.J. Cherniak et al., (2010) Analytical Methods in Diffusion Studies. Reviews in Mineralogy & Geochemistry, vol. 72 (2010), 107�170. [5] D. Zhao et al., Electron Probe Microanalysis and Microscopy: Principles and Applications in Characterization of Mineral Inclusions in Chromite from Diamond Deposit. Ore Geology Reviews 65 (2015), 733�748. [6] S.N. Frelinger et al., Scanning electron microscopy cathodoluminescence of quartz: principles, techniques and applications in ore geology. Ore Geology Reviews 65 (2015), 840�852. [7] Y. Huang et al., (1992) Kimberlites and Diamonds in the North China Craton (in Chinese), Geological Publishing House, Beijing (1992). [8] D. Zhao, Kimberlite, diamond and mantle xenolith from Northwest Territories, Canada and North China. Ph.D. Thesis, University of Michigan. Fig. 1. BSE images of diamond hosts and their olivine mineral inclusions. (a) Diamond LN50D03 and an elongated olivine inclusion. (b) A second polished surface of diamond LN50D03 with two euhedral olivine inclusions. (c) Two olivine inclusions and one chromite inclusion at the upper right corner in diamond LN50D04. (d) The euhedral olivine inclusion at the center of diamond LN50D04. (e) Diamond LN50D39 with an olivine inclusion. (f) One anhedral olivine inclusion in diamond LN50D44. (g) One euhedral olivine inclusion in diamond LN50D45. Note the bent feature of the inclusion that was constrained by the crystal form of diamond host. (h) One subhedral olivine inclusion in diamond LN50D55. (i) One elongated olivine inclusion in diamond LN50D68.");sQ1[660]=new Array("../7337/1321.pdf","Controlling Beam-Sample Interaction in Low Dimensional Materials by Low Dose Rate Electron Microscopy","","1321 doi:10.1017/S1431927615007394 Paper No. 0660 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Controlling Beam-Sample Interaction in Low Dimensional Materials by Low Dose Rate Electron Microscopy Christian Kisielowski1, S. Helveg2, L. Hansen2, P. Specht3 1) The Molecular Foundry and Joint Center for Artificial Photosynthesis, Lawrence Berkeley National Laboratory, One Cyclotron Rd., Berkeley CA 94720, USA 2) Haldor Tops�e A/S, Nym�llevej 55, DK-2800 Kgs. Lyngby, Denmark. 3) Dep. of Materials Science and Engineering, UC Berkeley, Berkeley CA 94720, USA In recent years it became possible to image the atomic structure of low dimensional materials with single atom sensitivity. Clearly, the development of atomic resolution aberration-corrected electron microscopy using low voltages below 100 kV plays a key role in this achievement [1] since it relaxes limitations that are set by beam-sample interactions. It is commonly argued for graphene and similar materials that voltage-dependent threshold values for the displacement of single atoms dominantly contribute to sample degradation. However, our understanding of beamsample interactions will remain partial if it is ignored that electron beam-induced sample excitations reach far beyond the considered atom displacements and that they are commonly reversible [2]. In fact, the energy deposited by the electron beam onto a sample is greatly current dependent, too, and can be varied by more than six orders of magnitude [2]. In two-dimensional materials, however, beam currents larger than 10000 e/�2s must be used to detect single light atoms in single images with exposure times around one second. The Figure 1 gives an example for the detection of a boron vacancy (VB) in a BN double layer [1]. In this reference it is shown that displacement damage does not limit the imaging process of this point defect at all. Instead, at least 2 vacancy configurations coexist in the images with formation energies that differ by ~ 400 meV. If one adopts the view that there is a temperature equivalent for each energy, as schematized in Figure 1, the local temperature during observation would exceed several thousand Kelvin. It can be greatly reduced if the beam current is lowered, which necessarily degrades resolution and sensitivity. Low dose rate in-line holography [3] at variable voltages allows us to overcome the dilemma how to maintain atomic resolution and single atom sensitivity while addressing beam sample interactions. The method exploits best practices developed for the imaging of radiation sensitive, biological objects and reversible object excitations. In Figure 2 it is compared with the established way of acquiring single high resolution images. Note that the low dose rate approach greatly reduces object motion (compare Fig. 2 a-c with Fig. 2g where 50 images were merged), allows for testing if beam-induced object alterations occur (Fig. 2e,f), preserves the structure of small particles (Fig. 2e,f) and creates a similar contrasts with a lower total dose (compare Fig. 2a - c with Fig. 2g). This contribution highlights investigations of graphene, BN, MoS2 and other materials with low dose rate in-line holography using acceleration voltages between 80 kV and 300 kV. They show that the approach is generally applicable and maintains the genuine structure of low dimensional material, small nanoparticles and even molecules to an end that is currently explored. In particular, it is of great benefit to experiments performed at elevated pressure and temperature [4].[5] [1] N. Alem, O.V. Yazyev, C. Kisielowski, P. Denes et al. PRL 106, 126102 (2011) [2] C. Kisielowski, L.-W. Wang, Petra Specht et al. Phys Rev. B 88, 024305 (2013) [3] C. Kisielowski, P. Specht, S.M. Gygax, B. Barton, H.A. Calderon, Micron 68, 186 (2015) Microsc. Microanal. 21 (Suppl 3), 2015 1322 [4] S. Helveg, C.F. Kisielowski, J.R. Jinschek, P. Specht et al., Micron 68, 176 (2015) [5] Electron Microscopy at the Molecular Foundry was supported by the Office of Science, Office of Basic Energy Sciences, of the U.S. Department of Energy under Contract No. DE-AC0205CH11231 Figure 1: A schematic representation of two VB configurations on equivalent temperature and energy scales. If large beam currents are applied, the two structurally different configurations coexist in the images. Their formation energies differ by 0.4 eV, which corresponds to an equivalent temperature of ~ 4600K. Figure 2: The traditional approach to acquire single high resolution images (a - c, 1 sec exposure time) is compared with low dose rate in-line holography (dg) [3]. Gold particles on an amorphous carbon support are shown. For details see text.");sQ1[661]=new Array("../7337/1323.pdf","Structural Rearrangement of 2-D Zeolite Nanosheets under Electron Beam","","1323 doi:10.1017/S1431927615007400 Paper No. 0661 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Structural Rearrangement of 2-D Zeolite Nanosheets under Electron Beam Prashant Kumar,1 Michael L. Odlyzko,1 Neel Rangnekar,1 Michael Tsapatsis,1 K. Andre Mkhoyan1 1 Department of Chemical Engineering & Materials Science, University of Minnesota, Minneapolis, MN 55455. Two-dimensional (2-D) zeolites and zeolite nanosheets are porous silicate frameworks desirable for catalytic uses involving bulky molecules [1], thin film separation membranes [2], and low-k dielectric materials [3]. Functionality of such zeolites is highly dependent on their crystal structure, thickness and pore dimensions. Low-dose transmission electron microscopy (TEM) studies at an optimum accelerating voltage have proven particularly useful in crystallographic structure determination of these electron beam sensitive materials [4]. However, it is known that zeolites undergo amorphization through radiolysis at low accelerating voltages, as well as both sputtering and amorphization through knock-on at high accelerating voltages [5]. Being inherently destructive, the examination of zeolites in TEM cannot perfectly reveal as-synthesized structure. Here, we investigate the structural evolution of MFI-zeolite nanosheets over time upon electron beam exposure in the TEM. A time series of selected area electron diffraction (SAED) patterns was acquired using a FEI Tecnai G2 F30 (S)TEM with a TWIN pole piece at 300 kV accelerating voltage. Dose was limited to 1.9 e-/�2/s on the specimen. MFI-nanosheets containing (i) 9 wt% and (ii) 26-28 wt% organic structure directing agent (OSDA) were synthesized using a method described by Varoon et al. [6]. The crystal structure of these nanosheets was confirmed to be MFI-zeolite type within 2 minutes of electron beam exposure, using high-resolution bright-field conventional TEM (BF-CTEM) imaging, high angle annular dark field scanning TEM (HAADF-STEM) imaging, and [010] zone axis SAED (Figure 1a-f). The thickness of these nanosheets was determined to be 1.5 unit cells (or 3.2 nm) by mapping a rel-rod through a tilt series of SAED patterns (Figure 1g-h). The evolution of diffraction spot intensities under continuous electron beam exposure showed that even though there was evidence of amorphization, some spots increased in intensity over time, confirming that other structural rearrangements occur within these nanosheets (Figure 2a-d). The observed increases in spot intensity during TEM illumination are explained by the removal of OSDA upon beam exposure: mass loss of OSDA causes the nanosheets to reduce in volume, as well as allows the crystal structure to relax from monoclinic to orthorhombic type. We relate the observed structural changes to the OSDA fraction in these nanosheets by carefully analyzing the diffraction spot intensities and in-plane lattice parameters. References: [1] A. Corma et al, Nature 396 (1998), p. 353. [2] K. Varoon et al, Science 334 (2011), p. 72. [3] C.M. Lew et al, Acc. Chem. Res. 43 (2010), p. 210. [4] M.M.J. Treacy, J.M. Newsam, Ultramicroscopy 23 (1987), p. 411. [5] O. Ugurlu et al, Phys. Rev. B 83 (2011), p. 113408. [6] K. Varoon et al, AIChE. J. 59 (2013), p. 3458. Microsc. Microanal. 21 (Suppl 3), 2015 1324 [7] This work was supported as part of the Catalysis Center for Energy Innovation, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences under Award DE-SC0001004. Figure 1. Top and bottom rows correspond to MFI nanosheets containing 9 wt% OSDA and 27 wt% OSDA respectively. (a),(b) HAADF-STEM image showing uniformly thick MFI nanosheets. (c),(d) boriented HR-CTEM image confirming MFI structure type. (e),(f) [010] zone axis SAED patterns of MFI nanosheets. (g),(h) Modulation of encircled diffraction spots in (e),(f) with tilt angle. Solid lines represent multislice simulations of spot intensity modulation for 0.5, 1.0, 1.5 and 2.0 unit cell thick nanosheets. Black dots represent experimental data. Figure 2. Top and bottom rows correspond to MFI nanosheets containing 9 wt% OSDA and 27 wt% OSDA respectively. (a),(b) Normalized SAED peak intensities after 6 hrs of electron beam exposure. Horizontal line represents the spot intensity upon initial exposure to the beam. (c),(d) Evolution of peak intensities shown in (a) and (b) with time. Increasing time from 0-6 hrs is indicated by color gradients fading to black. {101} spot intensity increases for nanosheet containing 27 wt% OSDA while {501} spot intensities decay with time for both the samples.");sQ1[662]=new Array("../7337/1325.pdf","Imaging of Quantum Materials","","1325 doi:10.1017/S1431927615007412 Paper No. 0662 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Imaging of Quantum Materials Felix von Cube1, Estelle Kalfon-Cohen1, Yachin Ivry2, Andrea Kn�ller3,4, Tatiana Webb5, Dennis Huang5, Jenny Hoffman5, Tina Brower-Thomas6, and David C. Bell1 1. 2. School of Engineering and Applied Sciences, Harvard University, Cambridge, MA 02138. Quantum Nanostructures and Nanofabrication, Massachusetts Institute of Technology (MIT), Cambridge, MA 02139. 3. Institut f�r Materialwissenschaften, Lehrstuhl 3, Universit�t Stuttgart, Stuttgart, Germany 4. Max Planck Institute for Condensed Matter Science, Stuttgart, Germany 5. Department of Physics, Harvard University, Cambridge, MA 02138 6. College of Engineering Architecture and Computer Science, Howard University, Washington, D.C. 20059. A fascinating class of quantum materials is atomically layered materials such as graphene or hexagonal boron nitride (h-BN). The properties of such materials differ strongly from those of their threedimensional bulk state. Depending on the composition, quantum materials may act as conductors, insulators, semiconductors or even as superconductors. Combinations of multiple quantum materials are of high interest to explore new phenomena and to build the foundation for future electronic devices at the nanometer scale. We report on the imaging and characterization of several unique quantum materials systems, reaching from defect formation in graphene to the characterization of hybrid quantum materials. We use a Cs corrected Zeiss Libra TEM to investigate chemical vapor deposition (CVD) graphene with added copper and mercury defects. With TEM we examine the positioning of the Hg and Co atoms on the graphene lattice. At the same time, we observe the effect of the copper and mercury on the pi electrons in graphene with Raman spectroscopy. Furthermore, we examine graphene based hybrid structures, such as graphene oxide embedded in a vanadium pentoxide nanofiber matrix (Fig. 1). The graphene sheets and the nanofibers have approximately the same thickness, leading to a material with enhanced mechanical performance in comparison to pure vanadium pentoxide and pure graphene oxide sheets. We also investigate quantum materials with superconducting properties, such as iron selenide (FeSe). As a bulk crystal, the superconducting critical temperature (Tc) of FeSe is a mere 8 K. However, recent reports showed that single-layer FeSe films grown on strontium titanate (SrTiO3) may exhibit superconductivity at temperatures above 100 K [1, 2]. We combine STM and TEM measurements to investigate single unit cell FeSe grown by in situ molecular beam epitaxy on SrTiO3. TEM is used to characterize the detailed structure of the FeSe and underlying SrTiO3 interface, while STM is used to probe the electronic states, both below and above the Fermi level. Niobium-nitride (NbN) on top of graphene is a hybrid quantum material that is of great interest for its controllable superconductivity. In thick films, which essentially behave like a bulk material, this hetero structure is well characterized [3] and it has been shown that the charge density in the graphene can be controlled with a gate electrode, allowing the tuning of the superconducting behavior. In the present work we concentrate on the fabrication and characterization of a low dimensional version of a NbNgraphene hetero structure (Fig. 2), where the NbN has a thickness of only a few nanometers. We have performed high-resolution electron microscopy in combination with STM and Raman spectroscopy, to investigate quantum materials. This combination proved to be an excellent tool to Microsc. Microanal. 21 (Suppl 3), 2015 1326 understand the astonishing mechanical and electronic properties. References: [1] Q.-Y. Wang et al, Chin. Phys. Lett. 29, 037402 (2012). [2] J.-F. Ge et al, Nat. Mat., doi:10.1038/nmat4153 (2014). [3] H. Heersche et al, Nature 446, 56-59 (2007). [4] This work was supported by the STC Center for Integrated Quantum Materials, NSF Grant No. DMR-1231319 Figure 1. The image shows the structure of a free-standing thin-film composed of vanadium pentoxide nanofibers and graphene oxide nanosheets. Fibers and sheets both have an average thickness of 1 nm and oxygencontaining functional groups which promote the interaction between both components, leading to a material with enhanced mechanical performance in comparison to pure vanadium pentoxide and pure graphene oxide sheets. Figure 2. Niobium nitride (NbN) layer on CVD graphene. During the growth process, crystalline islands are formed, which merge to a continuous film.");sQ1[663]=new Array("../7337/1327.pdf","Quantitative High Resolution Chemical Analysis of the (PbxSn1-xSe)1+TiSe2 Intergrowth System.","","1327 doi:10.1017/S1431927615007424 Paper No. 0663 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantitative High Resolution Chemical Analysis of the (PbxSn1-xSe)1+TiSe2 Intergrowth System. Jeffrey Ditto1, Devin Merrill1, Douglas L. Medlin2, and David C. Johnson1 1. 2. University of Oregon, Department of Chemistry, Eugene, OR, USA Sandia National Laboratories, Energy Nanomaterials Department, Livermore, CA, USA Tuning the properties of materials is often achieved through chemical substitution. Chalcogenide based misfit layer compounds offer a promising class of tunable materials but have been limited by a lack of synthetic control of thermodynamic products. The modulated elemental reactant (MER) method provides a versatile diffusion limited synthesis approach for self-assembly of targeted kinetically stable products [1]. It has been shown that the nanostructure of the deposited precursor is preserved in the final products [2, 3, 4]. The added ability to form solid solutions within only the transition metal dichalcogenide constituent suggests promise for controlling the material properties on an even finer scale [5]. In this presentation, we discuss our HAADF-STEM and STEM-EDX investigations of the structure and compositional distributions in a series of (PbxSn1-xSe)1+TiSe2 films grown using the MER method. These films consist of alternating layers of the rocksalt-structured (PbxSn1-xSe)1+ compound interleaved with single layers of the dichalcogenide compound TiSe2. We systematically varied the Pb to Sn ratio, to investigate the effects of alloying in the rock-salt layer on electrical transport properties in the films. This analysis required detailed understanding of the macroscopic and localized alloying processes. Thus, macroscopic measurements of the structure and composition of the films, conducted through XRD and EPMA were compared against our STEM, providing further confirmation that the targeted products were synthesized. In particular, our STEM-EDX mapping provided explanations for deviations from stoichiometric composition by pointing to the formation of localized layer defects and segregation of Sn and Pb within theses defects (Figure 1). As we will discuss, the ability to form targeted compounds using the MER technique also provides an approach for fabricating structurally controlled compositional standards for EDX quantification. We have thus employed the MER method to make binary films of SnSe and PbSe and are employing these to evaluate the level of absorption and fluorescence effect present in a series of intergrown (PbxSn1xSe)1+TiSe2 films of known average composition. This data is then used to calculate the local concentration of Pb and Sn in nominally segregated films, (PbSe)1+TiSe2(SnSe)1+TiSe2. These results are compared to calculated average concentrations obtained from x-ray diffraction methods. References: [1] Fister, L. & Johnson, D. C., J. Am. Chem. Soc 114 (1992) p. 4639�4644 [2] Atkins, R.; Wilson, J.; Zschack, P.; Grosse, C.; Neumann, W.; Johnson, D. C. Chem. Mater. 24 (2012) p. 4594�4599. [3] Heideman, C.; Nyugen, N.; Hanni, J.; Lin, Q.; Duncombe, S.; Johnson, D. C.; Zschack, P. J. Solid State Chem. 181 (2008) p. 1701�1706 [4] Heideman, C., Tepfer, S., Lin, Q., Rostek, R., Zschack, P., Anderson, M., Anderson, I., Johnson, D., J. Am. Chem. Soc. 135 (2013) p. 11055�11062. Microsc. Microanal. 21 (Suppl 3), 2015 1328 [5] Westover, R.; Atkins, R.; Ditto, J. J.; Johnson, D. C. Chem. Mater. 26 (2014) p. 3443�3449 [10] The authors acknowledge funding from National Science Foundation under grant DMR-1266217. Coauthors MF and SB acknowledge support from the National Science Foundation through CCI grant number CHE-1102637. Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-AC0494AL85000. a b c Figure 1. (a) A HAADF-STEM image showing resolved atomic planes of a (PbxSn1-xSe)1+TiSe2 intergrowth and a 2 nm scale bar, (b) the corresponding composite EDX map showing distributions of Ti (blue), Pb (red), Sn (green), and Se (cyan), and (c) a cartoon model of the image demonstrating turbostratic disorder. Sn and Pb are both represented by the red positions while Se (cyan) and Ti (blue) correspond to the colors in (b). The [110] orientation of (PbxSn1-xSe) is seen in the HAADF while the rest of the layers exhibit off of zone axis rotation.");sQ1[664]=new Array("../7337/1329.pdf","Long Range Order and Atomic Connectivity in Two-Dimensional Square PbSe Nanocrystal Superlattices","","1329 doi:10.1017/S1431927615007436 Paper No. 0664 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Long Range Order and Atomic Connectivity in Two-Dimensional Square PbSe Nanocrystal Superlattices Benjamin H. Savitzky1, Robert Hovden2, Kevin Whitham3, Tobias Hanrath4 and Lena F. Kourkoutis2,5 1. 2. Department of Physics, Cornell University, Ithaca, NY 14853, USA. School of Applied and Engineering Physics, Cornell University, Ithaca, NY 14853, USA. 3. Department of Materials Science and Engineering, Cornell University, Ithaca, NY 14853, USA. 4. School of Chemical and Biomolecular Engineering, Cornell University, Ithaca, NY 14853, USA. 5. Kavli Institute for Nanoscale Science, Cornell University, Ithaca, NY 14853, USA. Two-dimensional materials have drawn significant attention due largely to their novel electronic band structures, yielding a wealth of potential applications in electronics and optoelectronics. Twodimensional materials based on superlattices (SLs) of semiconductor nanocrystals (NCs) have been recently demonstrated, and are particularly promising systems for band structure manipulation due to the number of independently tunable degrees of freedom, including nanocrystal size, shape, connectivity, and composition [1]. Inexpensive solution based synthesis make these systems viable candidates for scalable, efficient photovoltaics and other energy materials. Long range order (LRO) as well as atomically coherent interfaces between NCs are key to bringing twodimensional NC SL materials to maturity [2]. Here we study a square SL of 6-7 nm PbSe NCs with grains of up to 3�m, fabricated via self-assembly at a liquid-liquid interface. By combining atomicresolution aberration-corrected STEM data with large fields of view spanning entire NC SL grains (Fig. 1a) we have analyzed order at both length scales. Quantitative analysis of LRO was achieved using the pair correlation function g(r), which describes the probability of finding two NC centers at a distance r apart, normalized such that g(r) = 1 represents no correlation between particle positions. For SLs ~6 NC layers thick, we find oscillations in g(r) of �10% about unity out to ~250 nm, representing LRO in the sense that the NC positions remain statistically correlated out to distances of 250 nm. For NC SL's only ~2-3 layers thick g(r) decays much faster, indicating faster loss of LRO (Fig. 1b). In contrast to a perfect crystal lattice, in which all lattice sites are specified exactly by the appropriate basis vectors, we find that these SLs behave as paracrystals, in which disorder between NCs is allowed to propagate through the lattice. We show that the structure of g(r) closely matches a paracrystal model with an uncertainty in the inter-NC spacing of =2.2� for the 6 layer SL (Fig. 1c), while a paracrystal with =3.5� matches the 23 layer SL. In both cases, the decay in LRO appears to result from the propagation of atomic-scale disorder in the inter-NC spacing, and a greater disorder in NC-NC spacing causes faster loss of LRO in thinner SLs. We analyzed single layer PbSe NC SLs both to understand the source of the short-range NC-NC positional disorder, and to push the system to its 2D limit. Monolayer SLs grew smaller grains, and displayed greater disorder in inter-NC spacing, consistent with the prior conclusion that thinner samples result in less robust crystallinity. Using statistical analysis we further show that the presence or absence of a continuous atomic lattice connecting two adjacent NCs directly impacts the NC-NC spacing, with unconnected adjacent NCs 8.0� farther away from each other on average than connected NCs (Fig. 2). Interestingly, the disorder in inter-NC spacing for adjacent, connected NCs in the monolayer (3.9�) was similar to that in the ~2-3 layer SL (3.5�). However, the overall disorder in inter-NC spacing for the monolayer SL is significantly larger (5.3�) due to the presence of unconnected NCs. Our data suggest that the loss of large grains and LRO in monolayer SLs is a result of reduced NC-NC connectivity. Microsc. Microanal. 21 (Suppl 3), 2015 1330 [1] T. Mokari, M. Zhang, and P. Yang, J. Am. Chem. Soc. 129 (2007), 9864-9865 [2] M.P. Boneschanscher et al., Science 344 (2014), 1377-1380 [3] BHS acknowledges support from the NSF IGERT (DGE-0903653). This work was supported by the Cornell Center for Materials Research with funding from the NSF MRSEC program (DMR-1120296) (a) 30 nm 500 nm (b) (c) Figure 1. Long range order (LRO) in PbSe nanocrystal (NC) superlattices (SLs) (a) is analyzed using the pair correlation function, g(r). Good LRO in SLs ~6 NC layers thick is evident from continued oscillations of �10% about unity (yellow bar) at distances of ~250nm, while SLs ~2-3 NC layers thick have decreased LRO, shown by the faster decay of g(r) (b). g(r) closely matches a paracrystalline model, indicating that unlike in perfect crystals, NC positional disorder propagates through the SL (c). (a) (b) 2 nm NC-NC Spacing (nm) Figure 2. Statistical analysis of atomic-scale connectivity in a one-layer thick PbSe nanocrystal superlattice. Disorder in inter-NC spacing in monolayers is due in part to variation in the NC-NC connectivity (a). Adjacent, connected NCs (green) have a mean spacing of 7.0�0.39nm, while adjacent, unconnected NCs (blue) have an increased mean spacing of 7.8�0.48nm (b). Counts");sQ1[665]=new Array("../7337/1331.pdf","Defect Microstructure in Irradiated Silicon Carbide","","1331 doi:10.1017/S1431927615007448 Paper No. 0665 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Defect Microstructure in Irradiated Silicon Carbide 1. 2. Sosuke Kondo1, Yutai Katoh2, Lance L. Snead2, Tatsuya Hinoki1 Institute of Advanced Energy, Kyoto University, Uji, Japan. Materials Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, USA. Recent interest in silicon carbide (SiC) and its composites (SiC/SiC composites) has been motivated historically by its possible utilization as a structural and functional material in fusion blankets in 1970s [1]. They are now attracting attention as alternative materials for fuel cladding for operating fission reactors (LWR) because of the conceivable better chemical stability and strength under LOCA or beyond design basis conditions in addition to the fusion and gas cooled reactors [2]. The advantages include perceived radiation stability at temperatures <1000 �C that comes from retention of radiation produced nano-structured defects up to high fluences. The knowledge regarding the microstructural change in SiC during irradiation has thus far been mainly limited to the low-to-intermediate temperature regime, and has hampered the determination of upper temperature and fluence limits for the severe use conditions now being considered. Important rapid property changes are anticipated at very high temperatures, where irradiation induced microstructural defects undergo unstable growth. This paper will review our recent results on the evolutions of dislocation and void microstructures in ion- or neutron-irradiated -SiC, especially for the high temperatures beyond the upper service temperature limits for metallic heat resistant structural alloys. Some of advanced techniques on the defect analysis will also be touched on. The literature offers only limited microstructural information on -SiC subjected to high temperature neutron irradiation until late 1990. Price (1973) first observed irradiation induced void formation in SiC deposited on graphite discs, the irradiation condition of void formation was neutron fluence of 4.3 � 1025 n/m2 (E>0.18 MeV) and an irradiation temperature of 1250 �C or higher [3]. Price (1969) and Blackstone (1971) et al. reported macroscopic volumetric expansion in neutron irradiated SiC at high temperatures (~1500�C) [5]. No information on the relation between microstructure and macroscopic length change in irradiated -SiC, however, has been available until not long ago. Several self-ion irradiation experiments also reported cavity formation in SiC, where cavities were observed at >1000 �C [4]. Recently, Snead et al. reported the temperature and fluence dependent swelling in very high purity SiC (produced through chemical vapor deposition (CVD)) in the high temperature regime [6], and formation of voids was considered the probable cause of the swelling. Our work clarified that cavities were predominantly spherical in shape below 1300 �C. Additional voids faceted with {111} planes were dominating at 1460 �C, which were basically tetrahedral truncated at the corners with {111}. The tetrahedral shape was unexpected as the surface-to-volume ratio is larger than the alternative {111} octahedral void common in both metals and ceramics. From a geometric viewpoint, all faces of the observed voids are either Si- or C-terminated surfaces. By comparing the surface area with the octahedral void (composed of the both Si- and C-surfaces) of the same volume, the considerable difference in surface energy between the Si(111) and C(-1-1-1) was implicated [7]. One can conclude now that those are responsible for the high temperature swelling in SiC. Other microstructural features in neutron-irradiated SiC are black spot defects and/or small dislocation loops (r~3nm) after irradiation at relatively low fluences between 300 and 1100 �C. The small dislocation loops were identified as lying on {111} lattice planes, and have been tentatively identified as Frank loops without Burgers vector analysis. Recently, Katoh et al. (2006) suggested that these were a Microsc. Microanal. 21 (Suppl 3), 2015 1332 mixture of Frank loops and other type defect clusters, for example having Burgers vector of Shockley partial, by means of Burgers vector analysis for CVD SiC irradiated to 7.7 � 1025 n/m2 (E>0.1 MeV) at 800�C. No evidence was reported because of their small size, but these defects were believed to be interstitial type based on difference in the mobilities of interstitials and vacancies in SiC. A highresolution electron microscope image of a relatively larger planner defect on (111) in heavy-neutron irradiated SiC has been estimated to be an interstitial-type Frank loop by image simulation [8]. In our work, large dislocation loops were identified as interstitial-type, a0/3<111> Frank loops using the inside/outside contrast method. It would appear that small loops at 1130 �C were also Frank loops: at least we confirmed the habit plane was {111} and their Burgers vector was parallel to <111>. In order to obtain the information about the interstitial motion in SiC, the temperature-dependent width of the interstitial loop denuded zone (DZ) formed along the random grain boundaries was evaluated [10]. The quantitative analysis showed a positive temperature dependence of the DZ width, where the smallest DZ width of 8.9 nm was observed in SiC irradiated at 1010 �C and the largest of 57 nm was observed at 1380 �C. Significant populations of small TEM invisible voids (r = 0.2�0.7 nm) were theoretically found to be formed in specimens irradiated below 1130 �C based on a simple reactiondiffusion equation, which were supposed to be limiting the interstitial motion at lower temperatures. The temperature-dependent diffusion coefficient estimated from the loop-denuded width showed an activation energy of interstitial migration of 1.5 � 0.1 eV in SiC, which is likely associated with the slower-moving species of Si interstitials. Finally, we concluded that the excellent irradiation stability of SiC is attributed to its stable microstructural defects due to the very high sink strength of vacancies even at high temperature and fluence. Advanced analytical technique in an electron microscopy may be best for understanding the ceramic specific features of the irradiation defects in a covalent bonding SiC, such as the stoichiometry constraint of the loops and the being of charged defects. References: [1] R.E. Riley, T.C. Wallace, J.M. Dickinson, J. Nucl. Mater. 85&86 (1979) 221�224. [2] S.J. Zinkle, K.A. Terrani, J.C. Gehin, L.J. Ott, L.L. Snead, "Accident tolerant fuels for LWRs: A perspective," J. Nucl. Mater., 448, 374�9 (2014). [3] R.J. Price, J. Nucl. Mater. 48 (1973) 47-57. [4] S. Kondo, T. Hinoki, A. Kohyama, Mater. Trans. 46 (2005) 1923-1927. [5] R.J. Price, J. Nucl. Mater. 33 (1969) 17-22. [6] L.L. Snead, T. Nozawa, Y. Katoh, T.S. Byun, S. Kondo, D.A. Petti, J. Nucl. Mater. 371 (2007) 329377. [7] S. Kondo, L.L. Snead, Y. Katoh, Appl. Phys. Lett. 93 (2008) 163110. [8] T. Yano, H. Miyazaki, M. Akiyoshi, T. Iseki, J. Nucl. Mater. 253 (1998) 78-86. [9] Y. Katoh, N. Hashimoto, S. Kondo, L.L. Snead and A. Kohyama, J. Nucl. Mater. 351 (2006) 228240. [10] S. Kondo, L.L. Snead, Y. Katoh, Phys. Rev. B 83, 075202 (2011)");sQ1[666]=new Array("../7337/1333.pdf","Short-Range Atomic Ordering in Amorphous Ion-Tracks in Pyrochlores","","1333 doi:10.1017/S143192761500745X Paper No. 0666 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Short-Range Atomic Ordering in Amorphous Ion-Tracks in Pyrochlores R. Sachan1, B. Liu2, D. Aidhy1, Y. Zhang1,2, M. F. Chisholm1 and W. J. Weber2,1 1 Materials Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831 2 Department of Materials Science and Engineering, University of Tennessee, Knoxville, TN 37996 Pyrochlore structured complex oxides (A2B2O7) have attracted attention due to its unique physical, chemical and electrical properties. Due to its radiation tolerance, pyrochlores are seen as a candidate for application such as inert matrix fuels, nuclear waste immobilization and geological age-dating.[1,2] However, radiation under such extreme conditions introduces phase transformation in the ordered pyrochlore structure, which is critical to understand for such applications. High energy ion irradiation (MeV to GeV) on pyrochlores is seen as an important experiment to understand the phase transformations which take place due to localized (nanoscale) energy transfer to the lattice in extremely short time period (~ns).[3] In the present work, Gd2Ti2O7 is irradiated with high energy ions to form nanoscale amorphous tracks along the ion penetration path. It has been previously reported that a concentric shell of disordered crystalline phase of defect fluorite is formed around the amorphous ion track in Gd2ZrxTi(1-x)O7 pyrochlores.[4] In the case of Gd2Ti2O7, there are studies reporting similar formation of a concentric defect-fluorite phase. However, atomic scale characterization and its effect on electronic structure of material has not yet been understood which is important for performance of such materials. Here, we report the detailed study of concentric defect fluorite phase existing around amorphous ion track by using high-angle annular dark field (HAADF) imaging and electron energy-loss spectroscopy (EELS) in an aberration corrected scanning transmission electron microscope (STEM). Here, a (HAADF) image of an ion track in Gd2Ti2O7 pyrochlore lattice oriented in [110] zone axis is shown in Fig. 1(a). The ion track was created by 2.2 GeV Au ion irradiation. The image analysis based on the atomic column contrast reveals that a distinct crystalline phase with thickness of 2-3 atomic planes (0.5-1 nm) forms at the periphery of the ion track. This phase is identified as defect fluorite phase which is a disordered crystalline derivative of the ordered pyrochlore phase. The contrast variation seen in amorphous region is due to tilting of ion track from perfect normal view. To understand the electronic structure, electron energy-loss spectroscopy (EELS) is performed. Fig. 1(b) and (c) show the Ti L-edge (456 eV) and O K-edge (532 eV) respectively from ordered crystalline and amorphous region. In an important observation, the Ti L-edge in the amorphous ion track features crystal field splitting similar to the one from crystalline region. This result indicates the presence of short-range ordering in the Ti-O octahedra in amorphous phase of Gd2Ti2O7. In addition to that, the shift in the eg and t2g peaks in both L3 and L2 shows the evidence of Ti-O octahedral distortion in the amorphous phase. Fig. 1(c) shows that the pre-peak merges with the main peak and results in a significant Microsc. Microanal. 21 (Suppl 3), 2015 1334 broadening of the O K-edge. These results are consistent with density functional theory (DFT) calculations that show overlapping of the density of states for the amorphous state. References: 1. K. E. Sickafus et al., Science 289 (2000) 748. 2. J. Zhang et al., J. Mater. Res., 25 (2010) 1344. 3. J. Wang, J. Phys.: Condens. Matter, 25 (2013) 135001. 4. M. Lang et al., Nat. mater. 8 (2009) 793. 5. This work was sponsored by US Department of Energy, Office of Science, Basic Energy Sciences, Materials Sciences and Engineering Division. (a) (b) (c) Fig. 1 (a) A high angle annular dark field (HAADF) image of an ion-track in Gd2Ti2O7 matrix showing amorphous, disordered crystalline and ordered crystalline region. (b) Ti L-edge from amorphous and ordered region, featuring presence of crystal field splitting, thus short-range ordering in amorphous phase. (c) O K-edge from amorphous and crystalline region.");sQ1[667]=new Array("../7337/1335.pdf","TEM Characterization of ion radiation damage in Ca(1-x)La2x/3TiO3 Perovskites","","1335 doi:10.1017/S1431927615007461 Paper No. 0667 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 TEM Characterization of ion radiation damage in Ca(1-x)La2x/3TiO3 Perovskites Mohsen Danaie1, Stella Pedrazzini1, Neil P. Young1, Paul A. J. Bagot1, Karl R. Whittle2 and Philip D. Edmondson3 1. 2. University of Oxford, Department of Materials, Oxford, UK Department of Engineering Materials, University of Sheffield, Sheffield, UK 3. Materials Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, USA Perovskite phases- generic formula ABO3, with both A and B denoting cation species- can find diverse applications both in fission and fusion nuclear energy production. Perovskites have been considered as potential matrix candidates for inert matrix fuel designs [1] and also, among other ceramics, as a more suitable encapsulation media for high-level nuclear waste disposal [1, 2]. A number of high-temperature superconducting Perovskites have also been proposed in the design of the magnetic containment of plasma in fusion energy reactors [3]. In all of the applications above, the Perovskite phase would need to withstand high doses of radiation exposure and potential incorporation of inert gases, e.g. Xe and Kr, produced during various nuclear reactions. Upon accumulation of a critical dose of radiation damage, the structure would inevitably undergo a crystalline-to-amorphous phase transition. Understanding the characteristics of this phase transformation is the key to the successful engineering application of Perovskites, as amorphisation is normally accompanied by volume change, cracking, and reduced thermodynamic stability. The main importance of the Perovskite crystal structure lies in the tunability of this system, both with regards to introducing point defects, e.g. by having partially vacant cationic A sites, and also in the ability in changing the crystal structure by varying composition or temperature. The presence of vacancies plays an important role in the amorphisation process of irradiated Peroskites. Presence of large population of vacancies, given the high mobility of these point defects, can result in structural healing of the atomic displacements caused by the knock-on damage, as was reported in LaxSr(1-3x/2)TiO3 for x 0.2 [4]. Pellets of Ca(1-x)La2x/3TiO3 (CLTO) were synthesized via the standard solid-state ceramic routes, yielding compositions of x=0.1, 0.5, 0.7, and 0.8. Half of the pellets were irradiated with Xe+ ions with the energy of 400 keV with 1015 ions/cm2 fluence at room temperature (University of Surrey ion beam centre). Transmission electron microscopy (TEM) samples were prepared by tripod polishing followed by ion milling. A JEOL 2100 microscope, operated at 200 kV, equipped with an X-ray energydispersive spectrometer (XEDS), was used for microstructure characterization. The pristine state of the CLTO phases was characterized using both XEDS and selected area diffraction (SAD). Figure 1 shows that the distribution of the constituent elements is uniform across bulk of the specimen, with compositions close to the nominal values. Electron diffraction pattern from the same area in Figure 1 can be closely indexed as the Orthorhombic Cmmm (65) crystal structure reported in [5] (Figure 2(A)). This was confirmed by examining various zone axes, one case shown in Figure 2(B). Irradiated samples showed evidence of amorphisation, with the depth of damage providing close match with the SRIM predictions. Mechanisms of crystalline-to-amorphous phase transition in CLTO will be further discussed. Microsc. Microanal. 21 (Suppl 3), 2015 1336 References: [1] RC Ewing, Progress in Nuclear Energy 49 (2007), p.635. [2] WJ Weber, RC Ewing, CRA Catlow, et al., Journal of Materials Research 13 (1998), p.1434. [3] R Fuger, M Eisterer, HW Weber, IEEE Transactions on Applied Superconductivity 19 (2009), p. 1532. [4] KL Smith, GR Lumpkin, MG Blackford, et al., Journal of Applied Physics 103 (2008), 083531. [5] Z Zhang, GR Lumpkin, CJ Howard, et al., Journal of Solid State Chemistry 180 (2007), p. 1083. Figure 1. CLTO-X=0.7 pristine sample: (A) STEM-ADF image, (B) XEDS elemental maps of the area marked in (A). Figure 2. CLTO-X=0.7 pristine sample: (A) Same area as observed in Figure 1: (1) bright-field image, (2) SAD pattern, and (3) SAD with simulation. (B) Another area in the same sample, (1) through (3) same as panels in (A).");sQ1[668]=new Array("../7337/1337.pdf","Atomic Resolution Imaging of Black Spot Defects in Ion Irradiated Silicon Carbide","","1337 doi:10.1017/S1431927615007473 Paper No. 0668 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic Resolution Imaging of Black Spot Defects in Ion Irradiated Silicon Carbide Li He1, Hao Jiang1, Yizhang Zhai1, Cheng Liu2, Izabela Szlufarska1, Beata Tyburska-P�schel2, Kumar Sridharan2 and Paul Voyles1 1. Department of Materials Science and Engineering, University of Wisconsin-Madison, Madison, Wisconsin, U.S.A. 2. Department of Engineering Physics, University of Wisconsin-Madison, Madison, Wisconsin, U.S.A. Silicon carbide is of great interest as a nuclear fuel cladding material. At relatively low irradiation temperatures (< 1000 C) and doses (< 10 dpa or displacements per atom), the major irradiation induced defects are black spot defects (BSD), which appear as nanometer scale black spots in bright field transmission electron microscopy (TEM) images [1,2]. BSDs are associated with radiation-induced swelling [1]. The detailed internal structure of BSD is unknown. We are working towards understanding the structures and evolution of BSD by combining high resolution scanning transmission electron microscopy (STEM) and defect structure modeling [3]. We have irradiated single-crystal 4H-SiC and polycrystalline 3C-SiC with two ion species. One is 3.15 MeV carbon ions to a dose of 5.14 � 1016 at/cm2 at 870 K, 1070 K, and 1250 K. The other is 1 MeV krypton ions to a dose of 3 � 1014 at/cm2 and 6 � 1014 at/cm2 at 870 K, 1070 K. The corresponding damage levels are 0.4 dpa for carbon irradiation at 1 m implantation depth, and 0.4 dpa / 0.8 dpa for krypton irradiation at 0.3 m depth. TEM observation was performed at 300 kV in a FEI Tecnai TF30 and STEM observation was performed at 200 kV in a probe Cs-corrected FEI Titan. Low-angle annular dark field (LAADF) STEM imaging are used to obtain atomic resolution images of BSD strain fields [2]. Figure 1a and 1b are two STEM images of a 4H-SiC sample irradiated with 1 MeV Kr ions at 0.4 dpa, 870 K, with 17.5 mrad semi-convergence angle and 29 � 144 mrad collection angle. They were acquired 24 s apart under 200 keV, 3.0 � 107 electrons/nm2 continuous irradiation. Non-rigid registration eliminated sample drift and other instabilities, and aligned the images in figures 1a and 1b [4]. Many BSDs moved through the crystal and changed shape in two dimensional projection under 200 keV electrons, as shown by arrows pointing to a BSD in Figures 1a and 1b. Tracking through all the intermediate frames confirms that this is the same defect. Figure 1c shows the mean square displacement of 5 BSDs, each tracked for at least 20 s within a 430 s, 5.56 � 108 e-/nm2 STEM image series. In general, BSDs cannot be tracked for longer times as they either leave the image field or change in contrast too much, potentially due to motion in the depth of the TEM specimen as well as in the plane. Irradiation-induced diffusion of clusters of high intrinsic migration barrier, as opposed to irradiation releasing clusters from dislocations or other defect "traps", has not been reported before. We have estimated the migration barrier using density functional theory calculations for a small interstitial cluster in SiC [3]. Figure 1d presents the minimum barrier migration path of carbon tri-interstitials in SiC. Within experimental uncertainty, the predicted beam-induced diffusivity of the BSDs is in reasonable agreement within predictions based on the barrier in Figure 1d. Combining HRSTEM study and theoretical structural models will provide valuable information about the structure of BSD [5]. Microsc. Microanal. 21 (Suppl 3), 2015 1338 References: [1] Lance L. Snead et al, Journal of Nuclear Materials 371 (2007), p. 329. [2] Li He et al, Microsc. Microanal. 20 (Suppl 3) (2014), p. 1824. [3] Chao Jiang, Dane Morgan, and Izabela Szlufarska, Physical Review B 86 (2012), 144118. [4] Andrew B. Yankovich, et al, Nature Communications 5 (2014), p. 4155. [5] This research is being performed using funding received from the DOE Office of Nuclear Energy's Nuclear Energy University Programs under contract number CFP-12-3357. Figure 1. (a) LAADF STEM image of a 4H-SiC sample irradiated with 1 MeV Kr ions at 0.4 dpa, 870 K. BSDs are circled. The sample is normal to [1120] . (b) The same area of (a), after 23.5 s continuous 200 keV, 3.0 � 107 electrons/nm2 irradiation. (a), (b) are aligned against drifting with non-rigid registration [4]. Arrows indicate displacement of an individual BSD, labeled as # 3 in (a) and # 9 in (b). (a) and (b) have been Gaussian smoothed to reduce noise. (c) Mean square displacement of 5 BSDs. (d) Migration trajectory of a carbon tri-interstitial cluster under electron radiation by ab initio molecular dynamic simulation, assuming 10 eV is transferred to the cluster at each step.");sQ1[669]=new Array("../7337/1339.pdf","Application of Quantitative Metallography to Cast Nickel-Based Superalloys","","1339 doi:10.1017/S1431927615007485 Paper No. 0669 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Application of Quantitative Metallography to Cast Nickel-Based Superalloys Agnieszka Szczotok1, Jan Cwajna1 1. Silesian University of Technology, Faculty of Materials Engineering and Metallurgy, Department of Materials Science, Katowice, Poland Quantitative stereology, or metallography, is concerned with the quantitative characterization of microstructures in terms of their point, lineal, areal, and volume elements. By means of suitable twodimensional measurements on the plane(s) of polish, statistically exact information can be obtained about the microstructural features in the three-dimensional space occupied by the alloy. These geometrical relationships have great generality and apply equally well to minerals, ceramics, plants, and metals [1]. In other words, quantitative stereology is a good method for estimating 3D structures, quantifying them through the measurement and calculation of parameters of 2D structures (twodimensional sections). Stereology has had a great impact in astronomy, geology, metallurgy, biology, medicine and materials science since 1961 when International Society for Stereology was founded. The presented review is a result of many years of experience achieved during the research on production technologies and microstructures of cast nickel-based superalloys. The data presented in the work cases concerned investigations on microstructure of metallic materials employed in a hot section of the aircraft engines. Their microstructure provides the unusual properties making possible utilization at high temperatures and under load bearing conditions. The materials research were carried out according to the authors' proposed scheme of a complex procedure for quantitative description of the selected, investigated elements of the microstructure, e.g., grains, carbides, pores, etc. The suggested procedure consists of many steps. All of them are important and indispensable. Even their sequence can influence obtaining proper results. The experiences obtained showed that only by application of a complex research procedure, including every step of knowledge beginning with the technological history of the element, the research aims, sampling strategy, specimen preparation, through selection of observation method, image acquisition and analysis and finally determination of measurement errors and carrying out statistical analysis of the results, enables one to obtain precise, repeatable results. The examples presented were selected to review the studies performed and to emphasize the need to focus attention upon certain aspects of the research conducted; for example, sampling strategy and its affect on comparison of results, selection of the appropriate etchant and condition of etching to reveal the macro- and microstructure or the state of the material. The appropriate sampling strategy enables one to obtain information about diversification of microstructure in the whole cast element. Sometimes you can fail to obtain similar results conducting grain observation and their measurements on the surface of the turbine blade compare to on its cross-sections [2, 3]. In some cases, finding and applying the proper etchant for revealing microstructure of the superalloy is tedious. There are many known etchants used for superalloys, but not all are useful for microstructural investigations of a specific superalloy. The etchants which are good for light microscopy (LM) observation of the superalloy microstructure usually are completely useless for scanning electron microscopy (SEM) observations. On the other hand, if you are looking long enough you can match the etchant to the superalloy to obtain great effects with microstructure observations using SEM and LM. Some etchants seem to be appropriate for qualitative description rather than quantitative estimation of microstructure elements. Technological history of the cast element is very important for a quantitative estimation of specific structural elements. For example microstructure of nickel based superalloys highly depend on the heat treatment used. The ' phase is the primary strengthening phase in nickel-based superalloys. The Microsc. Microanal. 21 (Suppl 3), 2015 1340 mechanical properties of these materials strongly depend on the alloy microstructure, i.e., mainly on the chemical composition, the volume fraction and morphology of ' particles. That is why a careful and precise quantitative description of phase precipitation is so significant. Illustration of quantitative measurement of primary ' phase precipitation with a highly diversified size and shape is presented in Fig. 1. The measurements performed and estimation of ' phase volume fraction showed that, in the case of that microstructure and measurements of a series of randomly selected images gave completely different results than in the case of analyzing ' phase precipitates with diversified size and shape in three groups. Figure 1. Steps of the quantitative evaluation of primary ' phase precipitates with highly diversified size and shape in CMSX-4 single crystal superalloy The microstructure images were captured by scanning electron microscopy (SEM) and then analyzed by means of the image analysis program. Quantitative evaluation of the investigated precipitates was complicated because of their size and shape diversification. The volume fraction of the ' phase precipitates was determined by area of measurements and three types of the ' phase precipitates morphology; therefore, the formula for the total volume fraction of the ' phase in the CMSX-4 SC superalloy was determined. The other proposal of the quantitative evaluation of ' phase precipitates taking into account secondary and tertiary ' phase was described in the work [4]. Precision in estimation of the total volume fraction of ' phase and the precipitate shape is needed to properly assess the quality of the applied production technology. The quantitative description by means of standard metallography now can be supplemented by other techniques, such as 3D imaging. References: [1] E. E. Underwood in "Microstructural Analysis, Tools and Techniques," J. L. McCall and W. M. Mueller (editors), (Plenum Press, New York-London) p. 35. [2] J. Adamiec et al., Proceedings of 9th European Congress on Stereology and Image Analysis and 7th International Conference on Stereology and Image Analysis in Materials Science STERMAT, Zakopane, Poland, May 10-13 (2005) p. 197. [3] J. Chmiela et al., Arch. Foundry Eng., Vol. 11, No. 4 (2011), p. 19. [4] A. Szczotok et al., Materials Characterization., Vol. 60, No. 10 (2009), p. 1114. [5] The authors acknowledge funding from the United Federation of Planets, X File Department, Grant the Structural Funds in the Operational Programme � Innovative Economy (IE OP) financed from the European Regional Development Fund � Project No POIG.0101.02-00-015/08.");sQ1[670]=new Array("../7337/1341.pdf","Quantitative Nano-Analysis of Superconducting Materials via SEM-FIB 3D-EDS Tomography.","","1341 doi:10.1017/S1431927615007497 Paper No. 0670 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantitative Nano-Analysis of Superconducting Materials via SEM-FIB 3D-EDS Tomography. Giuseppe Pavia1, Martin Kienle1, Ingo Schulmeyer1, Frank Bauer2, Marco Cantoni3, and Ken Lagarec4 1. 2. Carl Zeiss Microscopy GmbH, Product Management Materials, Oberkochen, Germany. Oxford Instruments GmbH, Wiesbaden, Germany. 3. EPFL, CIME, Lausanne, Switzerland. 4. FIBICS Incorporated, Ottawa, Canada. The final goal of any investigation about the world surrounding us, animated or unanimated, is to have a better understanding of it, about its structure, composition, and behavior. The present investigation target is pushed down to nanoscale features, where the assumption of homogeneity fails dramatically, both for living and inanimate structures. 3-dimensional (3D) investigation is henceforth more and more necessary and required. The advent of FIB-SEM microscopes has been crucial for providing large scale availability of 3D tomography at the nanoscale level. Energy Dispersive X-ray microanalysis (EDX) has been since long time a principal investigation method due to the ability to determine composition with a good accuracy. Latest large area Silicon Drift Detectors (SDD) has drastically increased the acquisition speed and with that the ability to analyze bigger portions of a sample. The excellent resolution of modern FIB-SEM instruments also at high currents contribute also to faster 3D EDX analysis We present here our studies about the structure of a superconductor cable. It has a multifilament structure with about fourteen thousands of Nb3Sn filaments, with an average diameter of about five micrometers each, embedded in a bronze matrix. The sample was first investigated at the macro scale, with optical instrumentation, from macro photography to optical microscopy. The sample surface and cross section show features helping to identify suitable regions of interests, that are selected for being observed in more detail and with different contrast mechanisms with SEM imaging. During this process of progressive targeted reduction of the investigated area size, it is possible and useful to spatially correlate large area scale images with images taken at larger magnification and resolution, and even with the results of microanalysis investigation. For the acquisition of the 3D EDX dataset we use a Zeiss Crossbeam 540, equipped with an Oxford Instruments X-Max 100 EDS SDD. Goal of this study is to get a better understanding of the sample by correlating optical images with electron microscopy images and the information provided by the chemical EDX analysis. The 3D dataset is acquired by serial sectioning using the ion beam and secondary electron and backscattered imaging combined with EDX analysis of the cross-sections. The setup allows to individually set up the parameters for all relevant milling, imaging and EDX parameters. Microsc. Microanal. 21 (Suppl 3), 2015 1342 10 �m Figure 1. Left to right: optical image and SEM image of the sample. EDX e-beam I-beam Figure 2. Acquisition Geometry Cu Nb Figure 3. Left to right: 3D maps of copper, niobium, and tin; image. Sn composite map superposed to SEM");sQ1[671]=new Array("../7337/1343.pdf","Microstructural characterization of alnico 9 alloy","","1343 doi:10.1017/S1431927615007503 Paper No. 0671 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microstructural characterization of alnico 9 alloy Lin Zhou1, M. Miller2, D. A. Cullen2, Ping Lu3, R. W. McCallum1, I. E. Anderson1, S. Constantinides 4 , M. J. Kramer1 1 Ames Lab, Ames, IA 50011 2 Oak Ridge National Laboratory, Oak Ridge, TN 37831 3 Sandia National Laboratories, PO Box 5800, MS 1411, Albuquerque, NM 87185 4 Arnold Magnetic Technologies Corp., 770 Linden Avenue, Rochester, NY 14625 Permanent magnets (PM) with rear-earth (RE) elements, such as dysprosium (Dy) and neodymium (Nd), have been widely used in motors and generators for hybrid electronic vehicles and wind turbines.[1] Concern over supply and price of the RE alloys has stimulated the search for alternative PMs.[1,2] One of the attractive non-RE PMs is alnico, a family of magnetic alloys composed primarily of Al, Ni, Co and Fe, with excellent magnetic stability at high temperature. The magnetic properties of alnico alloys are closely related to the control of the spinodal decomposition (SD) into an FeCo-rich (1 phase) hard magnetic phase and a non-magnetic NiAlrich phase (2 phase). Improving alnico will require subtle changes in chemistry and processing to reduce the diameter of the magnetic phase while maintaining its volume fraction. Alnico 9 is the current available commercial alnico alloy with the highest energy product (BH)max, which is both grain aligned and spinodally decomposed with an applied magnetic field.[1] Needed improvements can only be achieved through a better understanding of nanostructuring during SD. This study focuses on structural characterization of alnico 9 alloy from Arnold Magnetic Technologies. Atom-probe tomography (APT) and a combination of TEM techniques, including diffraction contrast TEM, high resolution transmission electron microscopy (HREM), high-angle annular-dark-field (HAADF) scanning transmission electron microscopy (STEM), energy dispersive X-ray spectroscopy, and Lorentz microscopy, were used. The interpenetrating nature of the 1 and 2 phase of alnico 9 is shown in Fig. 1. The 2 phase is continuous, but whether the 1 phase consists offully isolated particles or has some degree of interconnectivity cannot be established definitively, because the volume sampled by the APT is too small relative to the size of the 1 phase. Fine Cu-enriched rod-shaped particles with either a cylindrical or elliptical cross section were also observed at the corners of adjacent cuboidal 1 phase, and they occupy ~4% volume fraction of the alloy. Moreover, a Ni-enriched rod-shape phase parallel to the primary 1 /2 interface in the NiAlTi phase was detected. A nanometer scale mosaic structure formed by SD in alnico 9 (viewed along the transverse direction) was shown in Fig. 2 (a). The red bricks with a size of ~35nm are 1 precipitates, while the dark blue mold with similar size is 2 matrix. The 1 /2 interface facets along {110} planes as well as {100} planes. Cu rods (light blue, ~5nm) sit in the 2 phase at the corner of two {110} 1 facets. A Ni-rich (green) shell was observed in the 2 phase at the 1 /2 boundary. Ti partitioned to the 2 phase. Selected area diffraction pattern taken along [110] zone axis confirms the 2phase has a L21 ordered structure. Aberration-corrected STEM images of the Cu-rich phase indicates that when the Cu-rich phase has a small diameter (cylindrical shape, Fig. 2 (b)), it is forms a coherent interface with the 2-phase. For Cu-enriched phase with larger diameter, either intrinsic or extrinsic stacking faults were introduced. The Cu-rich phase forms an elliptical rod shape with distorted fcc structure (Fig. 2(c)). Further, Ni-enriched particles were observed in the Microsc. Microanal. 21 (Suppl 3), 2015 1344 2 phase when the 1 phase is faceted on {100} planes. Observation of alnico 9 along the longitudinal direction (Fig. 2(d)) showed that the 1 precipitates were very long (>400nm) and generally had tapered ends with an aspect ratio >10. Branching was commonly observed in the 1 rods. An in situ Lorentz microscopy study was also performed to analyze the movement of magnetic domains with an external magnetic field. Research was supported by U.S. DOE, Office of Energy Efficiency and Renewable Energy (EERE), under its Vehicle Technologies Program, through the Ames Laboratory, Iowa State University under contract DE-AC02-07CH11358. Atom-probe tomography research (MKM) and aberration corrected TEM imaging was supported through a user project supported by ORNL's Center for Nanophase Materials Sciences (CNMS), which is a DOE Office of Science User Facility. [1] M. J. Kramer, R. W. McCallum, I. A. Anderson, and S. Constantinides, JOM , 64, 752, (2012). [2] Lin Zhou , M. K. Miller, Ping Lu, Liqin Ke, R. Skomski, M. McCartney, D. Smith, H. Dillon, Q. Xing, A. Palasyuk, S. Constantinides, R.W. McCallum, I. E. Anderson, V. Antropov, and M. J. Krame, Acta Materilia, 74, (2014), 224. Figure 1. Isoconcentration surfaces: 10% Cu to show the Cu (brown), 10% Ni to show the outline of the NiAlTi phase/FeCo phases (red) and 30% Ni to show the high Ni regions (blue). Cu-enriched rods are at the corners of the FeCo phase and the NiAl region is contiguous. Figure 2. (a) Color composite energy-dispersive-X-ray map of alnico 9 taken along [001] crystal direction; (b,c) aberration corrected HAADF STEM images of alnico 9 taken under [100] zone axis along transverse direction. (d) HAADF STEM image of alnico 9 along longitudinal direction.");sQ1[672]=new Array("../7337/1345.pdf","Automated Correlative Tomography of an Aluminum 7075 Alloy Spanning Length Scales and Modalities","","1345 doi:10.1017/S1431927615007515 Paper No. 0672 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Automated Correlative Tomography of an Aluminum 7075 Alloy Spanning Length Scales and Modalities Arno P. Merkle1, Lorenz Lechner1, Luke Hunter1, Jeff Gelb1, S. S. Singh2, N. Chawla2 1 2 Carl Zeiss X-ray Microscopy, Inc. 4385 Hopyard Road, Pleasanton, CA, USA Arizona State University, 1151 S. Forest Ave., Tempe, AZ, USA The recent confluence of advanced computing power with new 3D characterization approaches has yielded great enthusiasm in the materials science community to pursue new scientific pathways, supported by the `materials by design' approach. It has become clear that one technique by itself cannot span all of the necessary length-scales in materials characterization nor provide a complete set of modalities. There is need in several areas such as energy materials, electronics, and metals, where the combination of two or more techniques is required to obtain a fundamental understanding of the microstructure [1]. Here we detail a study on light-weight structural alloys, which has benefitted from characterization of the same sample volume across multiple length scales[2]. A new, efficient 3D correlative microscopy workflow is presented, that utilizes a non-destructive submicrometer tomographic imaging approach[3, 4], 3D X-ray microscopy (XRM), to guide focused ion beam and scanning electron microscopy (FIB-SEM), in order to reveal targeted sub-surface regions of interest at high resolution. This approach, enabled by the emergence of an integrated, modern workflow environment (ZEISS Atlas 5), points to the future of efficient correlation in 3D across modalities and length scales. An aluminum 7075 alloy was investigated, exhibiting hierarchical structures that required characterization in 3D across multiple length scales, modalities and instruments. Al 7075 alloys are used extensively in structural applications due to their high strength-to-weight ratio. The microstructure of the alloy contains precipitates, constituent particles (also called inclusions), and pores associated with the inclusions. The precipitates are intentionally formed through heat treatment to provide strength to the alloy, whereas constituent particles are undesirable and form during alloy casting. The size, shape, and distribution of these microstructural features in the alloy are known to affect the corrosion and mechanical behavior (fatigue, tensile, and stress corrosion cracking) [5]. Therefore, it is necessary to obtain combined information of the size, shape, and distribution of the precipitates, constituent particles, and pores, in three dimensions (3D) and from the same volume of interest. Here we present the experimental methodology of combining tomographic length scales, as well as the approach to extracting key 3D microstructural quantification. By utilizing XRM to automate FIB-SEM tomography acquisition at specific locations via Atlas 5, we demonstrate the ability to efficiently seek and acquire volumes of interest, greatly increasing the utilization potential of the FIB-SEM. Microsc. Microanal. 21 (Suppl 3), 2015 1346 References: [1] Burnett T., et. al. Scientific Reports, (2014) [2] A. P. Merkle et al., Microscopy and Analysis, 28 (2014), p. S10-S13 [3] A. P. Merkle and J. Gelb, Microscopy Today, 21 (2013), p. 10 [4] E Maire and P Withers, International Materials Review, 59 (2014), p. 1 [5] Singh, S. et al., Mater. Res. Lett. 2 (2014), p. 217 Figure 1: A 3D automated correlative workflow demonstrated on an Aluminum 7075 alloy to access information about inclusions, voids, precipitates and the Al matrix grain structure. Figure 2 � 2D SEM image from the 3D FIB-SEM tomography acquisition, confirming the location of both a Si-bearing inclusion and void were identified, in the center of the region of interest. Multiple such sites may be programmed for automated FIB-SEM acquisition.");sQ1[673]=new Array("../7337/1347.pdf","Toward Deterministic Switching in Ferroelectric Systems: Insight Gained from In Situ TEM","","1347 doi:10.1017/S1431927615007527 Paper No. 0673 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Toward Deterministic Switching in Ferroelectric Systems: Insight Gained from In Situ TEM James Hart1, Michael Jablonski1, Andrew Lang 1, Anoop Damadoran2, Shi Liu 3, Miryam Arredondo4, Lane W. Martin2, Andrew Rappe 3, and Mitra L. Taheri1 1. 2. Drexel University, Department of Materials Science & Engineering, Philadelphia, PA, USA University of California-Berkeley, Department of Materials Science & Engineering, Berkeley, CA, USA 3. University of Pennsylvania, Department of Chemistry, Philadelphia, PA, USA 4. Queens University, Centre for Nanostructured Media (CNM), Belfast, United Kingdom To gain a greater understanding of the mechanisms that control material properties, researchers often turn to in situ TEM. This technique provides insight into many processes that are otherwise unclear in static experiments. Dynamic microscopy can potentially fill in gaps in the current understanding interfacial phenomena in a wide variety of materials. In this talk, the exploration of ferroelectric domain behavior in select oxide structures is presented [1-3]. Utilization of ferroelectrics for device applications requires precise control of domain structure. To facilitate device integration, an understanding of the microstructural factors that affect ferroelectric domain switching, and in most cases, ferroelastic relaxation, must be developed. In-situ transmission electron microscopy is an ideal tool for studying domain dynamics due to its inherent high spatial and temporal resolution. Specifically, we present quantitative dynamic studies of ferroelectric domain motion in two systems: a uniaxial ferroelectric, RbKTiOPO4 (RKTP), and a multiferroic material, BiFeO3 (BFO), which exhibits both ferroelectric and magnetic order. In situ studies were performed using a JEOL 2100 LaB6 TEM operated at 200 keV and a Hummingbird in situ biasing holder. In RKTP, we show that by manipulating the electron beam, we can reverse the direction of domain propagation, and by using a condensed probe we can locally nucleate domains; this process is dependent on both the sample geometry and electron beam condition. In BFO, the evolution of ferroelastically switched ferroelectric domains during many switching cycles is investigated, and the role of local defects and other extrinsic factors on the reversibility of domains during cycling is discussed. The results of these time-resolved biasing experiments provide a real time view of the complex dynamics of domain switching and complement scanning probe techniques and are critical to the development of improved ferroelectric devices. References: [1] Winkler, C.R. et al, Nano Letters 14.6 (2014), p. 3617. [2] Winkler, C.R. et al, Journal of Applied Physics 112 (2012), p. 052013. [3] Winkler, C.R. et al, Micron 43 (2012), p. 1121. [4] The authors acknowledge funding from the Office of Naval Research under contract number N000141410058. Microsc. Microanal. 21 (Suppl 3), 2015 1348");sQ1[674]=new Array("../7337/1349.pdf","Chemical Homogeneity in Entropy-Stabilized Complex Metal Oxides","","1349 doi:10.1017/S1431927615007539 Paper No. 0674 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Chemical Homogeneity in Entropy-Stabilized Complex Metal Oxides Ali Moballegh, Christina M. Rost, Jon-Paul Maria, Elizabeth C. Dickey Department of Materials Science and Engineering, North Carolina State University, Raleigh, NC, USA Innovation in new mixtures of constituents can lead to discover exciting new materials with unexpected properties and revolutionary applications [1,2]. It is known, the Gibbs energy needs to be minimized, as the main requirement, to achieve a stable single phase compound. Conventional approach to minimize the total energy of system is searching for a large and negative enthalpy. However, in this work, we show that the phase stability can be reached where the configurational entropy is maximized with mixing as many diverse elements as possible. In this work, five binary metal oxides, MgO, CaO, NiO, CuO, and ZnO, were chosen considering favored coordination, ionic radii, and diversity in crystal structures. An equimolar mixture of the constituent metaloxide powders were mixed and pressed into ceramic pellets. The pellets were subsequently annealed in an air until equilibrium was achieved and then quenched to the room temperature. A series of samples was prepared and annealed at temperatures ranging from 700�C to 1100�C to monitor the phase evolution. In Figure 1, X-ray diffraction (XRD) patterns show two distinguishable phases, rocksalt and wurtzite, are present up to 700�C, but are fully converted to single-phase rocksalt between 850�C and 900�C. The multiphase state reappears when the fully uniform single-phase sample was reannealed at 700�C for 2 hours, which is consistent with an entropy-driven mechanism. Microstructure and microchemistry of the phase-pure sample was studied by using a variety of electron microscopy and spectroscopy techniques. An Allied Multiprep polishing system was utilized to prepare a cross-sectional electron microscopy sample by wedge polishing technique [3]. An aberration corrected FEI Titan G2 60-300 kV S/TEM equipped with an X-FEG source and an advanced Super-XTM EDS detector system was used to analyze the structure and chemistry of the sample. The microscope was operated at 200 kV for high-angle annular dark-field (HAADF) scanning transmission electron microscope (STEM) imaging and energy dispersive x-ray spectroscopy (EDS) mapping with a convergence semi-angle of 15 mrad. STEM and EDS maps obtained at a variety of length scales reveal chemical and structural homogeneity of the cation mixture within the sample. To further understand the phase stability and decomposition, the XRD studies (Fig. 1) are reproduced via in-situ heating experiments performed in Protochips AdoroTM heating holder in a JEOL 2010F operated at 200kV. The phase evolution mechanisms and kinetics are studied by a combination of energy-filtered TEM, HAADF imaging and electron diffraction Microsc. Microanal. 21 (Suppl 3), 2015 1350 References: [1] Gludovatz, B. et al. Science 345, 1153�1158 (2014). [2] Gali, A. & George, E. P. Intermetallics 39, 74�78 (2013). [3] P. Voyles, J. Grazul, and D. Muller, Ultramicroscopy 96, 251 (2003). [4] This work is supported by ARO under contract W911NF-14-0285 Figure 1. X-ray diffraction patterns of a single pellet during the annealing process. The temperature spanned a range from 700�C to 1100�C, in 50 �C increments. Figure 2. (a) Low magnification HAADF image of single phase rocksalt. (b) EDS map indicates uniform distribution of Zn, Ni, Cu, Mg and Co. (c) HAADF image at higher magnification taken along <110> direction and (d) EDS map at higher magnification shows distribution of Zn, Ni, Cu, Mg and Co at cation sublattice.");sQ1[675]=new Array("../7337/1351.pdf","Microstructure evolution of a Cu and -Al2O3 composite observed by aberration corrected HAADF-STEM","","1351 doi:10.1017/S1431927615007540 Paper No. 0675 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microstructure evolution of a Cu and -Al2O3 composite observed by aberration corrected HAADF-STEM Zhiyang Yu1, Michael Kracum1, Animesh Kundu1, Helen M. Chan1, Martin P. Harmer1 1. Center for Advanced Materials and Nanotechnology, Department of Materials Science and Engineering, Lehigh University, Bethlehem, PA, 18015. Delafossite-structured CuAlO2 is a p-type transparent semiconductor that is of interest for optoelectronic applications [1]. The stability of thick film [2] and bulk [3] CuAlO2 has been examined in varying atmospheres where novel metal-ceramic composite microstructures were observed to form during reduction. These composite microstructures, consisting of nano-scale metallic copper and alumina are expected to provide interesting combinations of mechanical and electrical properties. The object of this work was to understand the microstructural evolution of the nano Cu/-Al2O3 composite during the reduction of delafossite CuAlO2. Bulk samples of CuAlO2 were prepared using the methods described in ref. 3. The samples were then subjected to a reduction anneal of 3 h at 1000 oC (pO2 < 1020 ). Focused ion beam milling (FIB) was used to extract thin foil samples from regions at the reduction front that contained both reduced and unreduced CuAlO2. TEM characterization was conducted using an aberration corrected STEM (JEOL 200CF), enabling sub-angstrom imaging of the structure before, during, and after reduction. Figure 1 shows the microstructure of the reduced CuAlO2 at different length scales. Figure 1 (a) depicts the region of the sample containing the reduction front. At low magnification, coarse bands of metallic Cu are clearly visible within the microstructure of the reduced CuAlO 2. Observation at higher magnification reveals a nano-scale two-phase structure of -Al2O3 and Cu (Figure 1(c-d)). Note that under BSE (back-scattered electron) imaging, the Cu exhibits brighter contrast relative to the -Al2O3. SEM examination of the partially reduced samples revealed that the transformation nucleates at the CuAlO2 grain boundaries/edges, with copper/ -Al2O3 laths growing inwards and consuming the grain body (Figure 1 (b), Figure 2(a)). The CuAlO2 structure consists of planar arrays of Cu+ ions alternating with layers of edge-sharing AlO6 octahedra; the plane normal is [0001] [4]. For the atomic resolution images depicted in Figure 2 (c), the CuAlO2 phase is oriented to a [1010] direction; hence the (0003) Cu+ atomic planes are aligned edgeon and are readily distinguishable in HAADF imaging due to their bright contrast. At the phase boundary of the CuAlO2, these planes terminate in a series of ledges. It was consistently observed that for the copper atomic planes, there was a gradual reduction in contrast in the ledge region, which indicates a lower concentration of copper within the atomic columns (see Figure 2 (b-c)). There was no discernable contrast variation, however, in the adjacent Al-O layers. These observations strongly suggest that during the reduction transformation, the copper atomic planes retract by the sequential outward diffusion of copper atoms, with detachment occurring at the plane edges. It is suggested that the remnant Al-O layers undergo a slight rearrangement to form -Al2O3. Clearly there must also be concurrent outward diffusion of oxygen, but unfortunately, imaging of the oxygen ions at the requisite concentration levels is beyond the capability of the microscope. Careful study also revealed that an epitaxial orientation relationship exists between the -Al2O3 and the parent CuAlO2 phase, as well as In summary, aberration corrected between the nano-copper regions and the parent CuAlO2 phase. HAADF imaging has provided valuable insight into the mechanism by which delafossite CuAlO2 Microsc. Microanal. 21 (Suppl 3), 2015 1352 transforms to Cu and -Al2O3 during reduction. Characterization of the mechanical and electrical properties of the composite is ongoing. References: [1] Hiroshi K., Yasukawa M., Hyodo H., Kurita M., Yanagi H., Hosono H., Nature 389 (1997), p. 939. [2] Byrne D., Cowley A., McNally P., McGlinn E., Crystal Engineering Communication 15 (2013), p. 6144. [3] Kracum M., Kundu A., Harmer M.P., Chan H.M., Journal of Material Science 50 (2015), p. 1818. [4] Meagen M., Ashmore N., Cann D., Thin Solid Films 496 (2006), p. 146. [5] The authors acknowledge funding support from ONR-MURI, Grant N00014-11-1-0678, monitored by Dr. D. Shifler. Figure 1. (a-d) Microstructure of the Cu/-Al2O3 composite at different scales. (a) The position where the FIB sample was extracted is marked. (b) The dotted lines delineate the boundaries of the grain of interest. (SEM, BSE) Figure 2. (a-b) The reduction front of delafossite CuAlO2 at different magnifications. (c) High resolution HAADF images clearly showing the decrease in copper concentration in the vicinity of the plane ledges.");sQ1[676]=new Array("../7337/1353.pdf","Absence of phase separation in nano-chessboard super-lattices in A-site deficient Ca-stabilized Nd2/3TiO3","","1353 doi:10.1017/S1431927615007552 Paper No. 0676 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Absence of phase separation in nano-chessboard super-lattices in A-site deficient Ca-stabilized Nd2/3TiO3 Feridoon Azough1, Demie Kepaptsoglou2, Quentin M. Ramasse2, Bernhard Schaffer2 and Robert Freer1 1. School of Materials, Materials Science Centre, University of Manchester, Manchester M1 7HS, United Kindgom. 2. SuperSTEM Laboratory, STFC Daresbury Campus, Keckwick Lane, Warrington WA4 4AD, United Kingdom. A-site deficient perovskites form a class of functional oxides of particular interest because of their attractive properties, such as ionic conductivity [1], dielectric behaviour [2] and transport properties [3]. A number of these ceramics show `cross-type' satellite reflections in their [001] diffraction patterns indicating the presence of a two-dimensional superstructure, that has been attributed to a micro domain model comprising of a system of periodically tilted oxygen octahedra [4]. Intriguingly several such compounds exhibit a peculiar contrast in their [001] High Resolution Transmission Electron Microscopy images (HRTEM), resembling a `nano-chessboard'. The origin of this contrast has been the object of debate, with two main models put forward: the first is based on a chemical phase separation into `chessboard' domains [5] and the second one attributes the origin of observed contrast to strain arising from a network of incommensurately titled oxygen octahedra [6]. Here, we report on a nano-chessboard structure in the A-site deficient Nd0.6Ca0.10.3TiO3 ceramic (where denotes vacancies). Using Electron Energy Loss Spectroscopy (EELS) in the UltraSTEM 100 aberration corrected, dedicated Scanning Transmission Electron Microscope (STEM) , we demonstrate beyond any doubt that the observed nano-chessboard contrast (Figure 1) does not originate from chemical phase separation into nano-domains [7]. Instead, closer inspection of High Angle Annular Dark Field (HAADF) STEM images and atomically resolved electron energy loss spectroscopy (EELS) chemical maps in two orthogonal directions suggest that, in the Nd0.6Ca0.10.3TiO3 system, Ca predominantly occupies Nd-vacancy shared sites, creating locally a higher occupation of the site and thus promoting vacancy-cation ordering in both a and b lattice directions. These observations corroborate previous studies [6], which suggest that the observed contrast in electron micrographs is a result of strain originating in intricately-modulated octahedral tilting distortions of the O sub-lattice combined with local cation-vacancy pairing. References: [1] J. L. Fourquet, H. Duroy, and M. P. Crosnier-Lopez, J Solid State Chem 127, 283 (1996). [2] D. Suvorov, M. Valant, S. Skapin, and D. Kolar, J Mater Sci 33, 85 (1998). [3] I.-S. Kim, T. Nakamura, Y. Inaguma, and M. Itoh, J Solid State Chem 113, 281 (1994). [4] M. Labeau, I. E. Grey, J. C. Joubert, H. Vincent, and M. A. Alario-Franco, Acta Crystallogr Sect A 38, 753 (1982). [5] B. S. Guiton and P. K. Davies, Nat Mater 6, 586 (2007). [6] A. M. Abakumov, R. Erni, A. A. Tsirlin, M. D. Rossell, D. Batuk, G. N�nert, and G. Van Tendeloo, Chem Mater 25, 2670 (2013). [7] F. Azough, D. Kepaptsoglou, Q. M. Ramasse, B. Schaffer, and R. Freer, Chem Mater 27, 497 (2015). Microsc. Microanal. 21 (Suppl 3), 2015 1354 [8] We acknowledge the financial support of EPSRC through EP/H043462, EP/I036230 and R113738. The SuperSTEM Laboratory is the U.K. National Facility for Aberration-Corrected STEM, funded by the EPSRC. Figure 1. [001] STEM images (a) Bright Field, (b) High High Angle Annular Dark Field (HAADF) and (c) Medium Angle Annular Dark Field (MAADF) of the A-site deficient Nd0.6Ca0.10.3TiO3 ceramic, showing a nanochessboard contrast. While the contrast is weak in the HAADF images it is significantly enhanced in the more strain-sensitive MAADF images. Figure 2. (a): MAADF STEM image of [001] the A-site deficient Nd0.6Ca0.10.3TiO3 ceramic acquired during 2D EELS mapping. (b) MAADF intensity profile averaged over the blue the line indicated in (a) clearly the nano-chessboard intensity modulations. The integrated signals of the Ca L2,3, Ti L2,3 and Nd M4,5 edges are also shown and by contrast show no appreciable modulations across the domains. (c) corresponding Ca L2,3, Ti L2,3 and Nd M4,5 2-D maps.");sQ1[677]=new Array("../7337/1355.pdf","Quantitative Analysis of a Lithium Ion Battery Cathode Material with X-ray Photoelectron Spectroscopy and Auger Electron Spectroscopy","","1355 doi:10.1017/S1431927615007564 Paper No. 0677 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantitative Analysis of a Lithium Ion Battery Cathode Material with X-ray Photoelectron Spectroscopy and Auger Electron Spectroscopy M. Shima, K. Tsutsumi, A. Tanaka, H. Onodera and T. Tazawa JEOL Ltd., 1-2 Musashino 3-Chome Akishima Tokyo Japan Research of Li-ion batteries has been intensively carried out to improve its performance. XPS (X-ray Photoelectron Spectroscopy) and AES (Auger Electron Spectroscopy) are typical analytical techniques for direct detection of Lithium. But these instruments have quite different feature such as analysis area, escape depth of Lithium signal and so on. For the XPS case, the x-ray source of a laboratory system is usually Alk (which is about 1.5 keV) and MgK (which is about 1.3 keV). The binding energy of Li1s orbital is around 50 eV, so the Kinetic energy of Li1s photoelectron is at least 1 keV. XPS has two advantages to analyze Li; the photoelectron with energy over 1 keV has a long mean free path, and the region of the Li 1s spectra has very low background [1]. These are the reasons why Lithium is detected easily by XPS. Thus, XPS can provide useful information for Li-ion battery materials. In the charging and discharging process of Li-ion battery, Li ions move between positive and negative electrodes. The movement of Li ions causes the change of the chemical bonding states of the transition metal in order to retain the charge balance of positive electrode. For these reason, XPS is very useful to analyze the materials of Li-ion batteries. An XPS system for laboratory use has been developed for micro area analysis (it is several tens micrometer) [2]. The micro area analysis is suitable for semiconductor samples. But the difficulty of the XPS micro area analysis is to precisely determine the analytical position. Of course it is easy to determine the analytical position of the sample with a clear structure like printed board and an area of discoloration. In the case of powder samples, it is very difficult to determine its analytical position because most powder samples have no specific feature. But in many cases, powder samples have the compositional differences. At this study, XPS analysis of a Li-ion battery cathode material was performed. The equipment is a JPS9200 (JEOL) which has a combined electromagnetic and electrostatic input lens system and two apertures inside the input lens. This system makes it possible to change the diameter of the analytical area from 30 �m to 3mm. Fig. 1 shows the result of qualitative and quantitative analysis of an XPS measurement at three different points with a 0.1 mm diameter. Fig. 2 shows the results of qualitative and quantitative analysis of the XPS measurement at three different points with a 3mm diameter. The quantitative data in Fig. 1 (0.1 mm diameter) shows a considerable dispersion, whereas that in Fig. 2 (3mm diameter) shows no such dispersion. Recently it was showed that AES has the great potential to analyze Lithium ion battery materials because AES can detect Lithium directly and AES can analyze chemical state of transition metal elements [3]. And off course AES has the function to measure the back scattered electron image (Fig. 3) not only secondary electron image. The high spatial resolution image contains information about compositional differences, which is quite useful to determine the analytical position even in powder samples. In the present study we will show the secondary electron image and back scattering electron image using SEM (Scanning Electron Microscopy) and the auger spectra measured with a JAMP-9510F (JEOL). We discuss the dependence of XPS, SEM and AES data on analytical positions. We propose the creatively use the XPS and AES analysis for Li-ion battery materials. [1] S Tanuma et al, Surf. Interface Anal., 21, 165 (1994) [2] H. Iwai et al, Journal of the Surface Science Society of Japan, Vol. 16, No. 9, pp592-597 (1995) [3] K.Tsutsumi et al, Journal of the surface science society of Japan Vol. 33, No8, pp.431-436 (2012) Microsc. Microanal. 21 (Suppl 3), 2015 1356 Intensity(arb. unit) Ni Co Mn O Cr Point A C Li Point B Point C 1000 800 600 400 200 0 Binding energy(eV) Figure 1 Qualitative and quantitative analysis with a 0.1 mm diameter with XPS Ni Co Intensity(arb. unit) Mn Cr O C Point A Li Point B Point C 1000 800 600 400 200 0 Binding energy(eV) Figure 2 Qualitative and quantitative analysis with a 3 mm diameter with XPS Figure 3 secondary electron image(left)and back scattered electron image(right) of LiB cathode material observed with JAMP-9510F 5000 Intensity [dN(E)/dE] 4000 3000 2000 1000 0 -1000 -2000 -3000 -4000 Li Plot_B Plot_C K O K O Mn Ni Ni Ni Mn Ni Ni Ni 100 200 300 400 500 600 700 800 900 1000 Electron energy [eV] Figure 4 Auger spectra of LIB with JAMP-9510F at dark and bright area in fig. 3");sQ1[678]=new Array("../7337/1357.pdf","Microstructure and Strain Hardening in Tensile-Tested Fe-Mn-Al-Si Steels","","1357 doi:10.1017/S1431927615007576 Paper No. 0678 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microstructure and Strain Hardening in Tensile-Tested Fe-Mn-Al-Si Steels D. T. Pierce,1,2 J. A. Jim�nez,3 J. Bentley,4 D. Raabe,5 and J. E. Wittig,1 1 Interdisciplinary Materials Science, Vanderbilt University, PMB 351683, Nashville, TN 37232, USA 2 Now at: Adv. Steel Processing & Products Res. Ctr., Colorado School of Mines, Golden, CO 80401, USA 3 Centro Nacional de Investigaciones Metal�rgicas (CSIC), Avda Gregorio del Amo 8, 28040-Madrid, Spain 4 Microscopy and Microanalytical Sciences, PO Box 7103, Oak Ridge, TN 37831-7103, USA 5 Max-Planck-Institut f�r Eisenforschung, Max-Planck-Strae 1, D-40237 D�sseldorf, Germany The exceptional combination of strength, ductility and strain hardening of high-Mn transformation- and twinning-induced plasticity (TRIP/TWIP) steels makes them appealing for automotive applications (e.g. vehicle weight reductions through down-gauging and room-temperature (RT) forming of complex shaped parts). The present study uses three Fe-22/25/28Mn-3Al-3Si alloys to investigate the effect of changes in stacking-fault energy (SFE) on the evolution of microstructure and mechanical properties during RT tensile deformation. The SFEs were previously measured by analysis of partial-dislocation separations using weak-beam dark-field TEM [1-4] that ultimately [1] incorporated single-crystal elastic constants measured on polycrystalline specimens by a novel nano-indentation method [5,6]. The RT SFEs of the Fe-22/25/28Mn-3Al-3Si alloys are 15�3, 21�3, and 39�5 mJm-2, respectively. Details of alloy and specimen preparation, tensile testing (see Figure 1), and specimen preparation for transmission electron microscopy (TEM) have been described elsewhere [1-4]. Microstructural characterization included optical microscopy, X-ray diffraction and TEM (performed at 200 kV with a Philips CM20T). The following important conclusions were drawn from this work: (i) A SFE of 15 mJm-2 (Fe-22Mn-3Al3Si) resulted in a deformation microstructure dominated by highly planar slip, suppression of dislocation cross-slip, and bcc/hcp-martensite transformation as the dominant secondary deformation mechanism (see Figure 2). The onset of grain refinement due to the formation of multiple variants of hcp-martensite within any given grain occurs from the beginning of plastic deformation and provides superior work hardening at low and intermediate strains (0-0.34 true strain), and the highest strength (687�7 MPa) but lowest elongation (85�3%) of the three alloys. (ii) A SFE of 21 mJm-2 (Fe-25Mn-3Al-3Si) resulted in a dislocation structure that exhibits both planar and wavy characteristics. The formation of both hcpmartensite and mechanical twinning (see Figure 3) results in excellent strain hardening in the initial, intermediate and final stages of deformation, along with the largest elongation (91�1%) of the three alloys, albeit with intermediate strength (642�7 MPa). (iii) At low strains (0 to 0.1 true strain), a SFE of 39 mJm-2 (Fe-28Mn-3Al-3Si) facilitates greater dislocation cross slip and mobility resulting in the formation of a dislocation cell structure (see Figure 4a) and reduced strain hardening compared to that of lower SFE alloys. Formation of hcp-martensite is completely suppressed, but mechanical twinning (see Figure 4b) enhances the strain hardening from ~0.1 true strain to failure, resulting in excellent ductility (87�2%) but the lowest strength (631�5 MPa) of the three alloys. (iv) The range of SFE from 15 to 39 mJm-2 results in an excellent product of strength and elongation (55-58 GPa%) with only small variations in strength and ductility, despite the transitioning of the steels from TRIP- to TWIPdominated behavior. Comparisons with literature data indicate that strength and ductility decrease significantly above a SFE of ~40 mJm-2, corresponding to a reduction in mechanical twinning [7]. References: 1. D T Pierce, J A Jim�nez, J Bentley, D Raabe, C Oskay, and J E Wittig, Acta Mater 68(2014)238-53 2. D T Pierce, J Bentley, J A Jim�nez and J E Wittig, Scripta Mater 66(2012)753-6 3. D T Pierce, J Bentley, J A Jim�nez and J E Wittig, Microsc Microanal 17-Suppl 2(2011)1888-9 4. D T Pierce, J Bentley, J A Jim�nez and J E Wittig, Microsc Microanal 18-Suppl 2(2012)1894-5 5. D T Pierce, K Nowag, A Montagne, J A Jim�nez, J E Wittig and R Ghisleni, Mater Sci Eng A578(2013)134-9 6. D T Pierce, K Nowag, A Montagne, J A Jim�nez, J E Wittig and R Ghisleni, Microsc Microanal 19-Suppl 2(2013)1052-3 7. This work was sponsored by the US National Science Foundation Division of Materials Research, under grant DMR0805295 and by the Ministry of Science and Innovation of Spain, under Grant MAT2012-39124. DTP acknowledges support for extended visits to CENIM, Madrid and MPI, D�sseldorf. JB acknowledges his appointment as Adjoint Professor of Materials Science at Vanderbilt University. Microsc. Microanal. 21 (Suppl 3), 2015 1358 Figure 1. RT tensile data (3 tests for each alloy) at 4 x 10-4 s-1 using subsized flat specimens with 20-mm gauge length, 5-mm width and 1.5mm thickness. (a) True stress vs true strain. (b) Strain-hardening rate, normalized by the experimental shear modulus (G = 69 GPa), vs true strain. Data in (b) are derivatives of 9th order polynomial fits of data in (a). 4 stages (plus 3 sub-stages for the 22%Mn alloy) of strain hardening are labeled. Figure 2. TEM BF images of 22%Mn alloy after 0.1 plastic true strain. (a) High density of overlapping SFs (inclined hcp-martensite laths) and (b) grain with 2 variants of edge-on hcpmartensite laths oriented with (111)||(0001)/[1-10]||[1-210] where indicates the austenite matrix. SAD pattern (inset) was recorded at a <110> zone whereas the BF image was recorded slightly off the zone axis in a two-beam condition. Arrows indicate lath intersections (black) or terminations (white). Figure 3. 25%Mn alloy deformed to 0.1 true strain. BF images of (a) mechanical twinning and (b) fine hcpmartensite lath structure. The SAD patterns (inset) were recorded at <110> zones whereas the BF images were recorded a few degrees off axis in two beam conditions. SAD patterns show twin reflections at 1/3 positions along <111> rows except through the central spot or hcp-martensite reflections also along <111> rows but based on a rectangular net with (0001) at ~1/2<111> position. Figure 4. 28%Mn alloy deformed to 0.1 true strain. (a) BF image of grain with dislocation cell structure. (b) DF image of mechanical twins using a {111} twin refection. The SAD pattern and BF image (insets) were recorded at a <011> zone and slightly off axis in a two-beam condition, respectively. 25 and 100% of grains contain mechanical twins for true strains of 0.10 and 0.18, respectively. High densities of dislocations are present in inter-twin regions, especially near twin/matrix interfaces.");sQ1[679]=new Array("../7337/1359.pdf","Stability and Transformation of Quasicrystalline Phase in Transition Metal Modified Al-(Mn-Fe)-based Alloys.","","1359 doi:10.1017/S1431927615007588 Paper No. 0679 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Stability and Transformation of Quasicrystalline Phase in Transition Metal Modified Al-(Mn-Fe)-based Alloys. K. Stan-Glowinska1, L. Litynska-Dobrzynska1, J. Dutkiewicz1, M. A. Gordillo2, J. M. Wiezorek2 1. 2. Institute of Metallurgy and Materials Science, Polish Academy of Sciences, Krakow, Poland. Department of Mechanical Engineering and Materials Science, University of Pittsburgh, USA. Due to environmental and economic concerns significant interest has arisen in the development of higher strength low-density Al-based alloys for aerospace and automobile applications. One notable Alalloy strengthening approach uses fine dispersions of quasicrystalline (icosahedral) particles. These metastable phases can be formed through rapid solidification methods (e.g. melt spinning, suction casting, gas atomization), and lead to enhanced hardness (> 450 HV) and tensile strength (>1000 MPa) [1] of the obtained material. However, the high cooling rates necessary for icosahedral phase (I-phase) formation limits the final dimensions of castings to a few millimeters. Fabrication of bulk forms of quasicrystal dispersion strengthened Al alloys typically necessitates hot-compaction of melt-spun ribbon, for instance. Thus, the thermal stability of the quasicrystalline phase in these microstructures is important to retain the enhanced alloy properties after warm-extrusion or -compaction. Ternary Al91Mn7Fe2 alloys exhibit one of the highest tensile strengths values (1250 MPa) among the group of quasicrystal strengthened Al-Transition Metal (TM) systems [2]. Unfortunately, their thermal stability is amongst the lowest. Since the thermal stability of the icosahedral phase has been shown to relate directly to its chemical composition [3], alloying with transition metal elements characterized by high melting temperature and low diffusivity in Al, such as Mo, W, and V may offer potential for enhanced thermal stability. The main objective of this work is the determination of the influence of the alloying TM additions on the I-phase decomposition under thermal stimuli. Al91Mn7Fe2 and Al91Mn6Fe2X1 (X1 stands for 1 at% Mo, W or V) ribbons were prepared by meltspinning under argon atmosphere. A combination of electron microscopy, x-ray diffraction, and thermal analysis techniques were used to investigate the thermal stability of the I-phase in melt-spun ribbon and hot-compacted conditions. The microstructure in the ribbons consisted of I-phase dispersoids and FCC Al solid solution (Fig. 1a). The Mo, W and V additions are incorporated preferentially into I-phase (Fig. 1b). Differential scanning calorimetry (DSC) measurements revealed a shift of peak temperature for the exothermic peak of the I-phase decomposition towards higher temperatures (60-80 K) for the TM modified samples. The activation energies for decomposition determined using the Kissinger method exhibited an increase from 192 kJ/mol (Al91Mn7Fe2) to 227 kJ/mol (Al91Mn6Fe2V1). After short term annealing near the exothermic peaks observed by DSC (Fig.2a), the I-phase decomposed into the following product phases: i) Al91Mn7Fe2, orthorhombic Al6(Mn,Fe); ii) Al91Mn6Fe2Mo1, orthorhombic Al6(Mn, Fe) and cubic Al12(Mn, Mo); iii) Al91Mn6Fe2V1, monoclinic Al45(Mn,Fe,V)7 and orthorhombic Al6(Mn,Fe) (Fig. 2b). In all cases, the decomposition products formed at the interface between Almatrix and I-phase, appearing to grow into the latter. The distributions of the alloy additions after annealing suggest that Mn and Fe diffuse to the I-phase/Al matrix interface to facilitate the decomposition (Fig. 3). Although the hot-compaction process results in a complete loss of I-phase for the Al91Mn7Fe2 alloy, the Mo and V modified alloys still retained a significant fraction of the I-phase after consolidation. A discussion of the difference in the diffusivity of constituent elements in the quasiperiodic lattice and its effect on the thermal stability will be presented. Microsc. Microanal. 21 (Suppl 3), 2015 1360 References: [1] A. Inoue, H. Kimura, Sh. Yamaura, Metals and Materials International 9 (2003), p. 527. [2] A. Inoue, H. Kimura, K. Sasamori, T. Masumoto, Mater. T JIM 37 (1996), p. 1287. [3] R. A. Dunlap, K. Dini, J. Phys. F. 16 (1986) p. 11 Work of K.S.G. was supported by National Science Center Poland, based on decision No. 2013/08/T/ST8/00222 Figure 1. a) TEM bright field image of as spun Al91Mn6Fe2V1 ribbon with inserted an electron diffraction pattern form one of particles, b) STEM image with distribution of elements in the I-phase. Figure 2. a) DSC curve for V modified alloy ribbon; b) TEM bright field image of sample annealed to 823K (e.g. see arrows in a)) with I-phase partially decomposed into two types of intermetallics. Figure 3. STEM image of Al91Mn6Fe2V1 ribbon after annealing at 773K for 30 minutes showing distribution of elements in both I-phase and orthorhombic Al6(Mn, Fe).");sQ1[680]=new Array("../7337/1361.pdf","An Overview Of How Microscopy Is Employed By the Oregon State Police Forensic Services Division.","","1361 doi:10.1017/S143192761500759X Paper No. 0680 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 An Overview Of How Microscopy Is Employed By the Oregon State Police Forensic Services Division. Celeste M. Grover1 1 Oregon State Police Forensic Services Division, Clackamas, Oregon Microscopy is used in some capacity in many of the forensic science disciplines found within a typical full service forensic science laboratory system. The information gathered from the observations made possible by microscopes is critical in the development of conclusions made about evidence in a crime. The following is a general overview of how microscopy is employed in some of the forensic disciplines of the Oregon State Police Forensic Services Division. The low magnification stereoscope can be utilized in any section examining physical evidence. It can aid in the ability for the forensic scientist to see potential trace and/or biological evidence not visible to the unaided eye. In the biology section compound microscopes are used to determine if spermatozoa are present on evidence typically related to cases involving an alleged sexual assault. An extract of suspected material is heat fixed to a slide and stained with Christmas Tree stain (i.e., nuclear fast red/picroindigocarmine). The slide is then examined for the presence of spermatozoa via bright field, phase contrast, and/or oil immersion. Confirmation of any sperm is done at 400X magnification or greater. (Figure 1) Both stereomicroscopy and polarized light microscopy are utilized by forensic scientists in the controlled substances section. Observations of plant morphology by stereomicroscopy are crucial to the identification of marijuana. Suspected plant material is examined for the presence of a combination of identifying features including: cystolith hairs, cover hairs, styles/stigmas, glandular hairs, serrated leaves, and mottled seeds. Low magnification stereomicroscopes can also aid in the determination of worn pharmaceutical and clandestine tablet imprints. Additionally, polarized light microscopy (PLM) can be used to observe microcrystalline tests specific for certain drugs or other compounds of interest. (Figure 2) The stereomicroscope plays a key role in the firearms and toolmarks section of the laboratory. In order to determine if bullets or cartridge cases could have been fired from a particular firearm, the microscopic marks imparted by the barrel or firing pin are examined and compared to a bullet or cartridge case known to have been fired from the firearm in question (test fire). The questioned bullet/cartridge case is compared side by side with the test fire using a comparison microscope (two stereomicroscopes attached with an optical bridge) at 6-70x magnification. Toolmark comparisons are done in a similar fashion comparing marks in question to a particular implement. Additionally, stereomicroscopes are used to aid in observation and examination of gun powder particles and obliterated serial numbers. (Figure 3) Microscopy is a cornerstone in the examination of trace evidence. Trace evidence comes in many forms, and as such a variety of microscopes and microanalytical techniques are used. The initial exam of most items of evidence for trace evidence involves a screening step utilizing the stereomicroscope (5120x). Targeted particles can include: glass, paint, hairs, fibers, low explosive residues, or other miscellaneous debris of interest. Once collected, the particles may then be examined further using Microsc. Microanal. 21 (Suppl 3), 2015 1362 additional microscope-influenced techniques at higher magnification (50-400x) such as PLM, Fourier Transform Infrared microspectroscopy, microspectrophotometry (MSP), fluorescence microscopy, microcrystalline tests, and refractive index measurement. When applicable, the morphological characteristics and instrumental data of the unknown are compared to a known standard. In the case of hair and fiber evidence, the microscopic characteristics of the unknown sample can be compared side by side to those of a sample from a known source using a compound comparison microscope. (Figure 4) Figure 1. Christmas Tree stain in: brightfeld (left); phase contrast (right) Figure 2. marijuana leaf (left); gold bromide microcrystalline test for heroin (right) Figure 3. bullet comparison (left); shot pellet in bone (right) Figure 4. camel hair (left); paint chip cross section (center); silver nitrate microcrystalline test for potassium sulfate (right)");sQ1[681]=new Array("../7337/1363.pdf","Firearm Serial Number Restoration with Electron Backscatter Diffraction","","1363 doi:10.1017/S1431927615007606 Paper No. 0681 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Firearm Serial Number Restoration with Electron Backscatter Diffraction Ryan M. White1, Robert R. Keller1 1. National Institute of Standards and Technology, Boulder, CO USA. Serial numbers are the primary means of identifying and tracking firearms, and it is common for serial numbers to be destroyed in an effort to mask the identity of the specific firearm used in a crime. Fortunately, the initial stamping of the serial number results in a deformation field much larger than the mark itself. In some cases, defaced serial numbers can be restored via acid etching techniques, which etch deformed and undeformed materials at different rates [1]. This restoration technique, however, is not always reliable, and is somewhat easily defeated via overstamping or heat treating. A more sensitive reconstruction technique may allow for reconstruction of even severely defaced serial numbers. The basis of serial number restoration is the detection of sub-surface microstructural change imparted by the marking tool. There are SEM imaging modes, including forward scattered (FS) imaging and electron backscatter diffraction (EBSD), which probe material microstructure and can be extremely sensitive to localized changes the crystal structure of a material. In fact, the use of EBSD in serial number restoration has been proposed previously [2], but a full proof-of-concept has not been demonstrated. In this work, the letter "X" was die-stamped into a polished piece of 316L stainless steel (figure 1a) and then polished away such that the imprint was no longer visible (figure 1b). After the surface was polished, the sample was imaged with the FS and EBSD imaging modes of the SEM. A tessellation of forward scattered images produced a faint reconstruction of the stamped imprint, outlined by arrows in figure 2a. While the change in microstructure is apparent in the FS image, it is difficult to discern the "X" shape without previous knowledge of the imprinted shape. Conversely, EBSD pattern quality maps, which can provide qualitative information about the localized deformation produce a clear and unambiguous reconstruction of the original imprint, shown in figure 2b. Further, pattern quality mapping of a cross-section of a stamped imprint indicated that damage detectable via EBSD extends approximately 520 �m beneath the imprint (or approximately 760 �m beneath the surface of the sample). It is yet unclear how this depth sensitivity compares to the traditional acid-etching-based methods of serial number reconstruction. This specific comparison between EBSD and acid etching restoration is the basis of future research on the topic. Unfortunately, with current detector hardware, the EBSD image acquisition is time consuming. With hardware-based limits of EBSD pixel sizes (6.72 �m), reconstruction of a single character requires up to nine hours of microscope time. Undersampling of the data in figure 2b provides a virtual pixel size increase, indicating that an unambiguous reconstruction can be performed with pixel sizes as large as 67.2 �m, decreasing acquisition time to approximately 5.4 minutes for a single character, and less than 1 hour for an 8-character serial number.[3-4] Microsc. Microanal. 21 (Suppl 3), 2015 1364 [1] J. I. Thornton and P. J. Cashman, Journal of the Forensic Science Society 16(1976), p. 69-71 [2] C. Necker and R. Forsyth, Microscopy and Microanalysis 16 (2010), p. 1579-9 [3] R. White acknowledges funding from the National Research Council Postdoctoral Research Associateship Program. [4] Contribution of the U.S. Department of Commerce; not subject to copyright in the United States. Figure 1. The letter "X" was a) die stamped into a piece of 316L stainless steel and b) ground/polished away to simulate destruction of a firearm serial number. The scale shown in the images is in millimeters. Figure 2. The imprinted/polished area visualized with tessellations of a) forward scattered electron images and b) EBSD pattern quality maps. While the "X" imprint is lightly visible in the forward scattered images, the EBSD pattern quality images produce an unambiguous reconstruction.");sQ1[682]=new Array("../7337/1365.pdf","New Possibilities of Using Microscopic Techniques in Forensic Field","","1365 doi:10.1017/S1431927615007618 Paper No. 0682 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 New Possibilities of Using Microscopic Techniques in Forensic Field Marek Kotrl�*1,2 1. 2. Institute of Criminalistics Prague (ICP), Czech Republic; Charles University in Prague, Faculty of Science, Praha, Czech Republic. Introduction Techniques of electron microscopy are widely applied in a forensic field, both for an initial screening and a final expert assessment. The set of materials that are delivered to a forensic lab is very extensive, practically any material produced by human and nature activities relating to the case that is examined can be encountered (ranging from the fragment of an ancient vessel, data records, documents, to hightech semiconductors). Therefore, materials of organic origin, plants and animal fragments, etc., are investigated as well. Methodology Electron microscopy combined with EDS/WDS is currently employed e.g. for analyses of these materials: a) unknown samples (including powders from extortionate letters, etc.); b) mineralogical, petrological and gemmological objects (mineral relics, soils, precious stones and their imitations, etc.); c) gunshot residues (GSR); d) explosives, propellants and fulminating compounds; e) post-blast residues (PBR), and other thermogenetic particles; f) fillers and additives of paper and plastics; g) pigments and paint systems, including colour layers of artworks; h) cosmetic and pharmaceutical products (surfaces and coating layers of tablets, granular composition, phase analysis); i) morphological structures of textile materials; j) determination of sorts of damages to fibres (smelting, fibre rupture, tear/cut, fracture, etc.); k) expert examinations of biological materials - trichological material and its damage, shells of soil microorganisms for pedological forensic investigations, insect eggs to determine post mortem interval, etc.); l) intersecting lines of writing and print tools (superposition - writing tool x print toner); m) glass; n) fragments of building materials; o) fractures of materials (determination of the character of the fracture area); p) toolmark slipped impressions (forensic technical examinations, ballistics, unlike classic optical light microscopy facilitates more detailed comparisons); Recently, dual systems with focused ion beam have considerably extended possibilities of electron microscopy. These systems are predominantly applied in the study of the inner structure of micro-and nanoparticles, layers and composites (intersecting lines in forensic graphical examinations, analyses of functional glass layers, colour variable pigments, etc.), the study of alloys microdefects, creating 3D models of particles and aggregates, and the possibility to study the inner structure of thermogenetic particles is also of key importance, etc. Automated mineralogical analyses are a great asset of the system for the analysis of mineral phases, particularly soils. These are modified SEM systems that are equipped with 2-4 EDS detectors. Classification of minerals into individual classes is carried out on the basis of BSE signal and chemical composition. These systems are standardly used for polished this sections. The experiments that were conducted at the ICP proved that it is possible to work directly with untreated grains of mineral phases. Grain preparations from real soil samples, mineral fractions (approx. 200-600 �m) were employed for testing. Grains were attached through a double- sided stub with carbon tape 26 mm in diameter, namely in one layer. The actual data collection was carried on mineral analyzer TIMA, on platform MIRA, using field-emission source of electrons. The device was equipped with two opposite fitted EDS Microsc. Microanal. 21 (Suppl 3), 2015 1366 detectors with an active area of 30 mm2. The data collection was undertaken in accelerating voltage of 25 kV and current of 5nA. The working distance amounted to 15 mm and the size of the microscope field of view was set at the value of 1450 �m. The threshold to differentiate mineral particles from the background in imaging by backscattered electrons was set to at level of 15%. Measurements were carried out with the step of 10 �m. The spectrum of 1000 counts was collected from each point analyzed. Percentage rate of the unidentified phase was in all conducted experiments about 5-16%, which is a satisfactory result for the majority of comparative tasks in a forensic field. The obtained quantitative data are subsequently processed through statistical methods. Other identification options for differentiation of mineral phases hardly distinguishable by different analyses are provided by cathode luminescence, mainly colour one and an accurate quantitative measurement of their spectral characteristics using microspectrofotometer. The measured spectra usually show considerable variability in the intensity of luminescence (reflect differences in chemistry of phases, but also variable conditions of measurement). However, after standardization of the spectra, the differences within one sample are only minimal and it is possible to apply them for differentiation of otherwise identical phases. Until recently, the analysis of the organic phase has brought big problems in multicomponent particles and composites by applying SEM. This question has been in principle unsolvable for e.g. analysis of postblast residues, when retrieval of the particles of 10 �m in heavily contaminated sample by techniques that would allow the analysis of its organic component is not feasible. One of the very interesting options appears to be the Raman spectrometry technique, which is nowadays available as a supplement to SEM EDX. Newly available is the device that is fully confocal, SEM keeps full functionality and scan range, very high resolution (for green laser resolution 360nm FWHD; 430nm Rayleigh), it is fitted with high quality objective lens, allows mapping through Raman spectrometry in a volume 250�m x 250�m x 250�m by piezo driven scanner (capacitive feedback linearized) and obtaining high quality white light image (250�m x 250�m) immediately in the SEM chamber. This technique is currently undergoing intensive testing and it seems that the method could significantly help to solve problems with the analysis of organic phases in electron microscopy, not only in the case of post-blast residues and explosives. Summary The application of new techniques significantly extends possibilities of classic microscopic procedures in a forensic field and allows acquisition of needed quantitative data in the forensic analysis of pedological phases, discrimination of mineral phases or the option of analysis of the organic phase directly in the SEM chamber. References: [1] Kotrl� M., Turkov� I.: Proc. SPIE 9073, 2014, SPIE, 90730U-1 � 10, Vol. 9073, 2014. [2] Jiruse J. et al.: Microsc. Microanal. 20 (Suppl 3), 990-991, 2014. [3] Acknowledgements - microanalytical methods at ICP were supported by projects: VD20062008B10, VD20072010B15, VG20102015065, VF20112015016, VF20122015027.");sQ1[683]=new Array("../7337/1367.pdf","Progress toward Understanding Lithiation Mechanisms of TiO2 Via In-situ TEM","","1367 doi:10.1017/S143192761500762X Paper No. 0683 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Progress toward Understanding Lithiation Mechanisms of TiO2 Via In-situ TEM X.Q. Pan1, S.J. Kim2, K. Zhang2, J. R. Jokisaari2, A. Kargar3, D. Wang3, G. W. Graham2 1 2 Dept. of Materials Science, University of California - Irvine, Irvine, CA 92697 USA Dept. of Materials Science and Eng., University of Michigan, Ann Arbor, MI 48109 USA 3 Dept. of Electrical & Computer Eng., University of California-San Diego, La Jolla, CA 92093 USA Owing to eventual depletion of fossil fuels and increasing demand for renewable energy resources, lithium ion batteries (LIBs) have attracted great interest for green energy storage application in hybridelectric vehicles (HEVs) and portable electronics. TiO2 has been considered for LIB anodes due to its chemical stability, non-toxicity, and abundance. The two widely studied TiO2 polymorphs are rutile and bronze, known for good capacity and cyclability, especially in nanostructured forms [1]. Despite their optimized performances, understanding of lithiation mechanisms in these polymorphs is still controversial, since most studies have been performed using bulk characterization techniques like X-ray diffraction and X-ray absorption near edge structure, which do not show local morphology. Here, we employ in-situ transmission electron microscopy (TEM) and aberration-corrected scanning TEM (STEM) to perform nano-scale structural studies of TiO2 upon cycling. We first study the lithiation mechanism of rutile TiO2 in the form of a nanowire (NW), especially focusing on its structural transformation [2]. Rutile TiO2 NWs were grown on Ti foil via a hydrothermal method before scraping a single NW onto a Cu rod for in-situ TEM. Our nano-scale electrochemical cells were assembled inside the TEM by connecting a TiO2 electrode to a tungsten tip scraped in Li metal that serves as the counter-electrode. Li2O naturally formed on the Li metal due to air exposure acts as a solid electrolyte. After a couple of electrochemical cycles, the NW underwent irreversible phase transformation from rutile to a monoclinic phase, accompanied by a large anisotropic volumetric expansion of approximately 120% (Fig. 1a and b). At this stage, no further morphological change was found to occur, even after 20 cycles, indicating remarkable structural stability of the NW. More injection of Li ions into the NW, accomplished by sustained bias at a constant lithiation potential of xx V, however, induced bubble-like dilation and further phase transformation to another phase, with rock-salt structure (Fig. 1c), thus completing a two-step lithiation process (Fig. 1d). Bronze TiO2 (TiO2-B), known to have the highest charge capacity among all TiO2 polymorphs, was synthesized as a thin film oriented along [001] by pulsed laser deposition [3]. To better explore the atomic-scale phenomena induced by Li ion insertion, both in-situ TEM and ex-situ high-resolution STEM were employed. Upon applying a bias through the c-plane of the film, Li ions immediately wetted the film surface before they found the most favorable spots for insertion into the TiO2-B lattice (Fig. 2a and b). During this c-axis lithiation, Li ions first diffuse into energetically favorable interstitial sites of the top TiO2-B layer before inducing a low-energy shear that allows continued lithiation deeper into the lattice (Fig. 2c). Interestingly, a high concentration of Li ions at the top surface also interacts with nearby TiO2-B layers to form another crystalline phase. In conclusion, we have directly probed the lithiation mechanism of two well-known TiO2 polymorphs, rutile and bronze. We expect our findings to contribute to the further understanding of Li ion insertion chemistry of these promising anode materials. Microsc. Microanal. 21 (Suppl 3), 2015 1368 References: [1] Arico, A. S. et al., Nature 4 (2005) 366-377; Deng, D. et al., Energy Environ. Sci. 2 (2009) 818-837 [2] Kim, S. J. et al., Chem. Commun. 50 (2014) 9932 [3] Zhang, K. et al., Adv. Mater. 26 (2014) 7365-7370 [4] This work was supported by the Center for Solar and Thermal Energy Conversion, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences under Award Number DE-SC0000957. Fig. 1. (a, b, c) Sequential TEM images showing the two-step structural transformation of a rutile TiO2 NW at different stages of lithiation and (d) schematics illustrating phase transformation corresponding to these stages. Fig. 2. TEM micrographs of a TiO2-B thin film at (a) the earlier and (b) the later stage of lithiation. (c) High-resolution STEM showing a half unit-cell shear induced by Li ion insertion.");sQ1[684]=new Array("../7337/1369.pdf","Revealing Near-Surface to Interior Redox upon Lithiation in Conversion Electrode Materials Using Electron Microscopy","","1369 doi:10.1017/S1431927615007631 Paper No. 0684 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Revealing Near-Surface to Interior Redox upon Lithiation in Conversion Electrode Materials Using Electron Microscopy Kai He1, Huolin L. Xin1, Jing Li1, 2, Eric A. Stach1, and Dong Su1,2. 1. 2. Center for Functional Nanomaterials, Brookhaven National Laboratory, Upton, New York, USA Department of Materials Science and Engineering, Stony Brook University, New York, USA. Lithium-ion batteries and supercapacitors both rely on electrochemical redox processes, although different mechanisms determine their relative energy and power densities [1]. For nanostructured electrodes of lithium-ion batteries, the capacity contains contributions from redox reactions that occur in both the interior (I) and near-surface (NS) regions. It is believed that the interior redox reactions contribute more to the overall battery capacity, but these take a longer time to be activated. In contrast, redox reactions in the near-surface reaction may exhibit a supercapacitor-like behavior (i.e. a high power density) because of the short transport paths for ions and electrons. Thus, an understanding of the kinetics of the transition from NS-redox to I-redox is critical to determining the rate capability of a lithium ion battery. In this work, we performed TEM experiments using two setups: (I) TEM "grid-in-a-coin-cell" using liquid electrolyte, and (II) in situ lithiation using Li2O solid electrolyte. The in situ measurements and tomography were performed on a JEOL 2100F TEM operated at 200 kV. The high-resolution imaging and analytical EELS were conducted on a Hitachi HD2700C STEM operated at 200 kV and equipped with a probe aberration corrector. Using combined electron microscopy approaches (in situ TEM/STEM, diffraction, tomography and STEM-EELS(Fig.1)), we observe the heterogeneous lithiation pathways that occur in NiO electrodes in real time (Fig.2). We find that the near-surface electroactive (Ni2+Ni0) sites saturated very quickly, and then encounter unexpected difficulty in propagating the phase transition into the electrode (referred to as a "shrinking-core" mode). However, the interior capacity for Ni2+Ni0 can be accessed efficiently following the nucleation of lithiation "fingers" which propagate into the sample bulk, but only after a certain incubation time. The reaction timescale and patterns we discovered from in situ TEM correlate with the ultimate rate performance of large-format batteries and are further supported by ex situ TEM and X-ray spectroscopies. We believe such heterogeneous transition mechanisms from NS-redox to I-redox may be generic and transferrable to a large class of conversion nano-electrode materials [2]. References: [1] P Simon, Y Gogotsi, B Dunn, Science 343, (2014), p1210-1211 [2] K He et al, Nano Letters, (2015) DOI: 10.1021/nl5049884 [3] The TEM work was carried out at the Center for Functional Nanomaterials, Brookhaven National Laboratory, which is supported by the U.S. Department of Energy (DOE), Office of Basic Energy Sciences, under Contract No. DE-AC02- 98CH10886 and DE-SC-00112704. We acknowledge the collaborations from Kejie Zhao and Ju Li(MIT), Xiqian Yu and Xiaoqing Yang(Brookhaven National Lab) Dennis Nordlundand Tsu-Chien Weng (SLAC National Accelerator Lab) Yi Jiang(Cornell U), Christopher A. Cadigan and Ryan M. Richards(Colorado School of Mines) Marca M. Doeff and Feng Lin(Lawrence Berkeley National Laboratory) Microsc. Microanal. 21 (Suppl 3), 2015 1370 Figure 1. Tomography and EELS mapping of lithiated NiO fingers. (a) A series of reconstructed 3D tomograms from an in situ lithiated NiO nanosheet. (b), (e), and (h) are zoom-in ADF-STEM images corresponding to the labeled regions in (a). (c) and (f) show EELS mapping of Ni2+ (green) and Li++Ni0 (red) for areas in (b) and (e) using the low-loss spectra components shown in (g). (d) Atomicallyresolved STEM image showing a crack region coherently bounded with unlithiated NiO in (b). (i) EELS charge mapping of Ni2+ (green) and Ni0 (red) for the area in (h), a series of EELS spectra and Ni2+/0 concentration profiles along the arrow are shown in (j) and (k), respectively. Scale bars, 20 nm (a), 10 nm (b), 2 nm (d), 20 nm (e), 10 nm (h). Figure 2. Structural evolution during in situ lithiation of NiO nanosheets. (a) Schematic illustration of in situ setup. Time-sequenced S/TEM snapshots (b) shrinking-core mode, and (c) finger mode. (d) Schematic cartoon showing heterogeneous pathways. (e) The velocity of reaction front propagation. (f) Histogram showing statistics of incubation time. Scale bars, 20 nm.");sQ1[685]=new Array("../7337/1371.pdf","In situ TEM Observation of Lithiation and Sodiation Process of ZnO Nanowire","","1371 doi:10.1017/S1431927615007643 Paper No. 0685 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In situ TEM Observation of Lithiation and Sodiation Process of ZnO Nanowire Hasti Asayesh-Ardakani1, 3, Anmin Nie1, 3, Farzad Mashayek2, Robert F. Klie3 and Reza Shahbazian-Yassar1, 2, 3 1 Department of Mechanical Engineering-Engineering Mechanics, Michigan Technological University, Houghton, MI 49933-1295, USA 2 Department of Mechanical and Industrial Engineering, University of Illinois at Chicago, Chicago, IL60607-7059, USA 3 Department of Physics, University of Illinois at Chicago, Chicago, IL60607-7059, USA The limited source of Li and over-growing demands of high capacity storage for application in electrical vehicles and green power backup energy makes the Li-ion battery an expensive choice for future. This need triggers the new battery concept search beyond Li-ion. Natural abounding of Na on earth makes it more affordable and eco-friendly choice. Another considerable issues in batteries are performance and cyclability of batteries. The anode materials usually experience large volume changes through the ion insertion and extraction. This volume change and lithium embrittlement usually causes cracks and loss of contact in the anode material, which ultimately causes the failure of battery. In this work, we investigated and compared the structural and mechanical changes of ZnO nanowires during sodiation and lithiation process by using aberration corrected scanning transition electron microscopy and in situ TEM (Figure 1 shows a pristine ZnO nanowire and corresponding SAED pattern before and after lithiation). The cracks were created upon the first lithiation process of single crystalline ZnO nanowire. The lithiated ZnO nanowire shows multiple glassy domains, which has low strength and ductility. This results in poor cyclability of battery. On the other hand, ZnO nanowire after sodiation show dislocations on the surface of nanowire that results in more ductility of sodiated nanowire rather than lithiated one. Another important issue is the control of reaction rate. In the fast reaction rate, the large Zn nanocrystals can grow and suppress the formation of Zn alloys and dramatically reduce the capacity of battery. This direct comparison demonstrates the critical role of anode material's mechanical properties and reaction rates on failure mechanism and cyclability of Li/Na-ion batteries. Microsc. Microanal. 21 (Suppl 3), 2015 1372 Figure 1. In situ lithiation of single crystalline ZnO nanowire. (a) pristine ZnO nanowire and (b) selected area electron diffraction pattern of pristine nanowire. (c) The selected area diffraction pattern of lithiated nanowire");sQ1[686]=new Array("../7337/1373.pdf","Microstructural Characterization of Air Electrode Architectures in Lithium-Oxygen Batteries","","1373 doi:10.1017/S1431927615007655 Paper No. 0686 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microstructural Characterization of Air Electrode Architectures in Lithium-Oxygen Batteries Jianguo Wen1, Jun Lu2, D.J. Miller1, Xiangyi Luo2, Hsien-Hau Wang3, Kah Chun Lau3, Eric Tyo3, Stefan Vajda3, Larry A. Curtiss3 & Khalil Amine2 1 2 Electron Microscopy Center � Center for Nanoscale Materials, Argonne National Laboratory, USA Chemical Sciences and Engineering Division, Argonne National Laboratory, USA 3 Materials Science Division, Argonne National Laboratory, USA The aprotic Lithium-Oxygen (Li-O2) batteries, consisting of lithium metal and porous air electrode separated by electrolyte, have the potential needed for long-range electric vehicles. The architectures of air electrodes are found to be of paramount importance to achieve good electronic and ionic conductivity, fast oxygen diffusion, and stable integrity for high-performance Li-O2 batteries [1]. In this work, we first characterized an air electrode showing a dramatic reduction in charge overpotential [1]. We found that the nanostructured cathode architecture with an Al2O3 coating and Pd nanoparticles led to a nanocrystalline lithium peroxide (Li2O2) discharge product that contributes to the low overpotential. As shown in Fig. 1a, the pristine carbon (super P) surface has almost no amorphous phase. The carbon surface was partially coated with 3 cycles of Al2O3 (Fig. 1b) by atomic layer deposition to passivate carbon defect sites. The protective Al2O3 coating on the air cathode prevents electrolyte decomposition on carbon defect sites, which can increase the charge potential. Fig.1c shows the Pd nanoparticles (2-6 nm) are often directly attached to the carbon support. The architecture of cathode promotes growth of a nanocrystalline form of discharged product Li2O2 as shown in Fig. 1d. Density functional theory calculations show that amorphous Li2O2 may have a metal-like density of states, in contrast to the poor electronic conductivity of crystalline Li2O2 [1]. The amorphous Li2O2 in the grain boundaries improves electronic transport properties that are needed to lower the charge potential. Fig. 1e shows a dramatic reduction in overpotential of ~0.2V during charge using the nanostructured cathode architecture. To further understand the effect of metal nanoparticle on the growth of Li2O2, subnanometer silver clusters of defined size and number of atoms were deposited on passivated carbon [2]. Fig. 2a shows a TEM image of a 15 atom Ag cluster (Ag15) on an Al2O3 coated carbon particle. In most cases, discrete Ag15 clusters are observed. Occasionally, agglomeration of the Ag15 clusters (circled in Fig. 2a) is observed. Fig. 2b shows the atomic structure in an Ag15 cluster from a video (1 frame per second). The atomic arrangement of the clusters changes quickly under the electron beam, so very low-dose imaging techniques were required to obtain these images. SEM observation indicates that dramatically different morphologies of the discharged product Li2O2 is dependent on the size of the Ag clusters [3]. The Li2O2 product using Ag3 clusters as a catalyst is film-like. In contrast, the discharge product using Ag9 clusters is largely toroid-like with rough surfaces, while Li2O2 toroids obtained using the Ag15-based cathode have smooth surfaces. The discharge capacity (~3500 mAhg-1) for the first cycle of the cell with the Ag15-based cathode is much larger than that using the Ag3 and Ag9 clusters (~2400 mAhg-1). The results of this study indicate that precise control of subnanometer size of catalytic nanoparticles on air electrodes can be used as a means to improve the performance of lithium�oxygen cells [3,4]. Microsc. Microanal. 21 (Suppl 3), 2015 1374 References: [1] Jun Lu, et al, Nature Communications, DOI:10.1038/ncomms3383 (2013). [2] S. Vajda, et al. Nature Materials, 8, 213�216 (2009). [3] Jun Lu, et al, Nature Communications, DOI:10.1038/ncomms5895 (2014). [4] Research at the Electron Microscopy Center � Center for Nanoscale Materials at Argonne National Laboratory is supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357. 1 nm Fig. 1. HRTEM images of the air electrode architectures resulting in a low charge overpotential. (a) Pristine carbon, (b) carbon surface coated with 3 cycles of Al2O3, (c) further coated with 3 cycles of Pd, (d) nanocrystalline Li2O2 for electronic conductivity, (e) voltage profile during discharge-charge of cells with different air electrode architectures. Fig. 2. HRTEM images of the air electrode architectures leading to the control of Li2O2 morphologies. (a) TEM image of Ag15 clusters (arrows) on an Al2O3 coated carbon particle. Occasionally, agglomeration of Ag15 clusters (circled) is observed. (b) HRTEM image of an Ag15 cluster on an Al2O3 coated carbon surface.");sQ1[687]=new Array("../7337/1375.pdf","Understanding the Surface Structure of LiMn2O4 Spinel Cathodes with Aberration-Corrected HAADF STEM and EELS","","1375 doi:10.1017/S1431927615007667 Paper No. 0687 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Understanding the Surface Structure of LiMn2O4 Spinel Cathodes with AberrationCorrected HAADF STEM and EELS C. Amos1, M.A. Roldan2,3, M. Varela2,3, J.B. Goodenough1, P.J. Ferreira1 1. 2. Materials Science and Engineering Program, The University of Texas at Austin, Austin, TX, USA. The STEM Group, Materials Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN, USA. 3. Universidad Complutense de Madrid, Madrid, Spain. As global energy consumption continues to rise, the importance of energy storage becomes increasingly important. Energy density, rate-capability, and cyclability must continually improve. This constant struggle for advancement is seen most easily in the high-density, electrical energy storage market, which is dominated by lithium-ion batteries. One of the most promising chemistries in lithium-ion batteries is LiMn2O4 (LMO), a spinel cathode material which has the advantage of both a high energy density and a high rate capability, but this chemistry is plagued with cyclability problems. In the LMO system the main contributor to cycling degradation is the Mn disproportionation reaction (2Mn3+ = Mn2+ + Mn4+) which creates soluble Mn2+ that is lost to solution. This loss of active material from LMO leads to capacity degradation. In order to understand exactly how LMO loses active material from its surface, it is crucial to determine the surface's atomic structure. This is because the surface structure dictates how the electrolyte will interact with the cathode. In this paper, we use a combination of high-angle annular dark-field (HAADF) aberration-corrected scanning transmission electron microscopy (STEM) and electron energy loss spectroscopy (EELS) to identify the atomic surface structure and composition of LMO. We confirm the underlying spinel structure and for the first time we find, in as-processed LMO, a surface structure composed of Mn3O4 and a lithium-rich Li1+xMn2O4 subsurface layer which occurs as a result of the surface reconstruction. These conclusions are reached based on Mn oxidation state and oxygen content mapping with EELS as well as structural identification with HAADF STEM and STEM simulations. The presence of Mn2+ at the surface provides direct evidence of the Mn disproportionation reaction which has only been indirectly proven, until now. We have also identified that oxygen loss is the mechanism by which the surface reconstruction occurs. This solid understanding of the interface between the cathode and the electrolyte gives us insight into mitigating the undesired surface reactions that occur in the LMO system. Through these pioneering experiments we have gained a deeper understanding of interfacial phenomena in LMO which will be of central importance for all types of lithium-ion chemistries and continue to push the evolution of portable, electrical energy storage. [1] This work was supported by a NASA Office of the Chief Technologist's Space Technology Research Fellowship. We acknowledge the use of the aberration-corrected ARM 200F STEM at the University of Texas at San Antonio, a facility supported by a grant from the National Institute on Minority Health and Health Disparities (G12MD007591) from the National Institutes of Health. J.B. Goodenough thanks the Robert M. Welch Foundation Grant F-1066 for financial support. Research at ORNL (MV) supported by the US Department of Energy, Office of Science, Basic Energy Sciences, Materials Sciences and Engineering Division and through a user project supported by ORNL's Center for Nanophase Materials Sciences (CNMS), which is sponsored by the Scientific User Facilities Microsc. Microanal. 21 (Suppl 3), 2015 1376 Division, Office of Basic Energy Sciences, U.S. Department of Energy. Research at UCM (MR) was supported by the ERC starting Investigator Award, grant #239739 STEMOX. Figure 1. HAADF STEM image of LMO viewed along the [110] zone axis. LMO diamonds (blue) are found in the bulk while Mn3O4 (red diamond) is visible at the surface. The insets in the image show a higher magnification of the bulk (top right) and surface (top left). An FFT of the full image (top center) is included to indicate the crystal orientation. Figure 2. HAADF STEM image of an LMO particle (top). The green rectangle in the STEM image shows the area from where an EELS spectrum image was acquired. The colored maps (red, yellow, and green) and corresponding colored spectra below each image, represent the location of different Mn valence states within the nanoparticle. The oxygen content relative to Mn extracted from the O-K and Mn L2,3 edges is shown (middle right). The black arrow shows where the relative oxygen percentage (bottom right) was extracted. The ratio is determined by the number of oxygen atoms to the total number of atoms in the sample.");sQ1[688]=new Array("../7337/1377.pdf","Optimizing Workflows in Correlative Light and Electron Microscopy.","","1377 doi:10.1017/S1431927615007679 Paper No. 0688 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Optimizing Workflows in Correlative Light and Electron Microscopy. Kirk Czymmek1, Jeff Caplan2, Cherish Warner3, Janine Sherrier3 and Alexandra Elli4 Carl Zeiss Microcopy, LLC, One Zeiss Drive, Thornwood, NY 10594, USA 15 Innovation Way, Delaware Biotechnology Institute, University of Delaware, Newark DE 19711, USA 3. 15 Innovation Way, Delaware Biotechnology Institute, Department of Plant and Soil Sciences, University of Delaware, Newark DE 19711, USA 4. Carl Zeiss Microscopy, Carl-Zeiss-Strasse 22, 73447 Oberkochen, Germany 2. 1. Correlative microscopy allows researchers to look at the same exact structure of interest using typically separate, but powerful microscopy platforms commonly including light and electron microscopy [1]. By combining the unique imaging modalities to study singular biological events, more detailed investigations of structure-function relationships can be conducted that greatly facilitate our understanding of cellular phenomenon. With photon-based microscopy, rare structures or targeted phenomenon in cells and tissues can be readily located and imaged due to the large field-of-view and ease of sample screening. Conventional light (resolution ~0.2 �m) and super-resolution (resolution up to ~20nm) microscopies are well-positioned to benefit from the high resolution subcellular or ultrastructural details inherent to electron microscopy for correlating the spatial distribution of macromolecules at nanoscale resolutions. In one example provided here, we used correlative microscopy combining immunofluorescence and light microscopy on epoxy resin sections followed by field emission scanning electron microscopy to provide key structural data at high-resolution over large areas (Fig 1A-D.). This work sought to understand the spatial symbiotic relationships between nitrogen-fixing bacteria rhizobia and the host legume plant, Medicago truncatula in root nodules. Samples were fixed with 1% glutaraldehyde and 4% paraformaldehyde in 100mM PIPES and then post-fixed in 1% osmium tetroxide (aq). Subsequently, nodules were dehydrated in acetone, infiltrated and embedded in an Epon-Araldite epoxy resin and serial sections acquired and dried down on 22 x 22mm indium tin oxide coated-coverslips with fiducials marks. Sections were further processed for immunofluorescence using a callose primary antibody and AlexaFluor� 647 secondary antibody which labeled plasmodesmata traversing the cell walls (Fig 1 A,B & D). Light microscopy was performed using Shuttle & Find on a ZEISS AxioImager and then samples were stained with lead citrate and uranyl acetate and then relocated and imaged with a backscatter detector on a ZEISS SIGMA HD VP FESEM at 8kV and at 10nm pixel resolution. This presentation will focus on a variety of contemporary methods, probes and workflow for correlative light and electron microscopy (CLEM) and will demonstrate with real world examples of animals, plants and microbes, strategies to facilitate relocation of microscopic data for correlative experiments including automation and 3-D approaches [2]. Microsc. Microanal. 21 (Suppl 3), 2015 1378 References: [1] J. Caplan, M. Niethammer, R. Taylor and K Czymmek, Current Opinion in Structural Biology 21 (2011), p. 686-693. [2] The authors acknowledge funding from NSF�IOS�1127155 to D.J.S. and NSF�IOS�0923668 to D.J.S Figure 1. CLEM of nitrogen-fixing bacteria rhizobia and the host legume plant, Medicago truncatula in root nodules. 1A. Low magnification overview of the section showed immunolabeled callose (red) localized along the cell walls. Scale bar = 25um. 1B. Closer inspection of the sample dataset showed abundant rhizobia within individual plant cells and a large central vacuole (V) and small puncta (red) localized along the cell walls. (N) nucleus. Scale bar = 10um. 1C. At higher magnification, small channels representing plasmodesmata could be observed traversing the cell wall (arrows) in the FESEM image. Scale bar = 3um. 1D. A correlative overlay of the fluorescence channel (red) localizing to callose confirmed the antibody localized to plasmodesmata (arrows). Scale bar = 3um.");sQ1[689]=new Array("../7337/1379.pdf","Correlative Light and Electron Microscopy (CLEM) on Biological Samples Using Immuno Electron Microscopy","","1379 doi:10.1017/S1431927615007680 Paper No. 0689 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Correlative Light and Electron Microscopy (CLEM) on Biological Samples Using Immuno Electron Microscopy C. ten Brink1, V. Oorschot1,2, J. Klumperman1 1. Dept of Cell Biology, Cell Microscopy Core, University Medical Center Utrecht, Heidelberglaan 100, Utrecht, The Netherlands. 2. Current address: Monash University, Melbourne, Australia Correlative light and electron microscopy (CLEM) techniques integrate light microscopy (LM) and electron microscopy (EM) on a single sample. Application of CLEM on fixed samples often involves the use of fluorescent microscopy to define regions-of-interest that are subsequently traced back in the EM to provide subcellular context information (e.g. membrane organization, non-labeled surroundings. CLEM of live samples is very powerful since it infers dynamic information to static EM pictures. In addition, CLEM can be used to identify transient or rare events in live cells for high resolution analysis. By merging LM and EM an integrated set of information is obtained that cannot - or not easily - be achieved when using separate images of related events. However, imaging requirements are intrinsically different between LM and EM which is why creating conditions that are ideal for both modalities can be tedious and laborintensive. Our lab is specialized in the development of new probes and CLEM pipelines. We focus on methods that allow us to define the membrane compartments or membrane sub-domains that contain fluorescently-tagged proteins previously localized in fixed or living cells. We apply our technologies to study the cellular pathways and mechanisms that control the cell's digestive systems � i.e. the endo-lysosomal system - in health and disease conditions. Our main CLEM technology is based on the use of immunogold labeling of ultrathin cryosections (Tokuyasu technique) [1]. Essential for CLEM of live cells is to prepare sections perpendicular to the plane of growth/filming. To this end, we developed a specialized `flat embedding' protocol for Tokuyasu sections[2], which we have adapted to live cells[3]. Here we apply these techniques to perform CLEM. An example of CLEM of live cells using Tokuyasu sections is shown in Figure 1. HeLa cells containing GFP-tagged LAMP1, a lysosomal membrane protein, were imaged alive and subsequently fixed, cryosectioned and labelled with antiGFP and protein A conjugated to 10 nm gold particles. Hence, membrane structures containing LAMP1-GFP were first followed in live cells, after which their ultrastructure was obtained by immunoEM. We currently use this technique to make a comprehensive overview of the dynamics of distinct endo-lysosomal intermediates. Figure 2 shows CLEM of fixed HeLa cells containing LAMP1-GFP. Ultrathin cryosections were prepared and labelled with antiGFP that was marked with both Alexa488conjugated secondary antibody as well as with protein A-10 nm gold [4]. CLEM allows us to select by LM cell profiles of interest in ultrathin sections after which detailed information is obtained by EM. CLEM using immunoEM is a powerful method, but in order to increase the throughput of this methodology improvements in efficiency and accuracy are highly warranted. One restriction of current CLEM technologies is the limited resolution of LM since most set-ups use conventional LM. To circumvent these limitations we are currently adapting the CLEM approach shown in Figure 2 to super-resolution microscopy. Microsc. Microanal. 21 (Suppl 3), 2015 1380 Figure 1. CLEM of live cells. A, A': Stills from movie of HepG2 cells transfected with mGFP.LAMP-1. Distinct LAMP-1 positive endo-lysosomal intermediates were followed over time by LM. B. EM picture of compartment 5 shown in A/A'. Cells were fixed in 2% formaldehyde, prepared for ultrathin cyrosectioning and immunogold labeled (10 nm gold) for GFP. The high resolution morphology of compartment 5 is typical for autolysosomes. Figure 2. CLEM of fixed cells. HeLa cells transfected with mGFP.LAMP-1 were prepared for ultrathin cryosections and labeled with rabbit antiGFP, goat anti rabbit Alexa488 and protein A 10 nm gold. A. LM image of ultrathin cryosection. B. EM image of yellow Square in A. The two bright fluorescent spots are both endosomes (E). In addition, small vesicles positive for LAMP-1 (arrows) are seen by EM but not LM. PM = plasma membrane. References 1. Slot, J.W. and H.J. Geuze, Cryosectioning and immunolabeling. Nature protocols, 2007. 2(10): p. 2480-2491. 2. Oorschot, V., et al., A novel flat-embedding method to prepare ultrathin cryosections from cultured cells in their in situ orientation. J Histochem Cytochem, 2002. 50(8): p. 1067-80. 3. van Rijnsoever, C., V. Oorschot, and J. Klumperman, Correlative light-electron microscopy (CLEM) combining live-cell imaging and immunolabeling of ultrathin cryosections. Nat Methods, 2008. 5(11): p. 973-80. 4. Cortese, K., et al., High data output method for 3-D correlative light-electron microscopy using ultrathin cryosections. Methods Mol Biol, 2013. 950: p. 417-37.");sQ1[690]=new Array("../7337/1381.pdf","Correlative Microscopy using Serial Blockface Scanning EM","","1381 doi:10.1017/S1431927615007692 Paper No. 0690 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Correlative Microscopy using Serial Blockface Scanning EM Thomas Deerinck1, Eric Bushong1 and Mark Ellisman1,2 National Center for Microscopy and Imaging Research and the Center for Research in Biological Systems, 2Department of Neurosciences, University of California, San Diego, La Jolla, CA, USA. Correlative microscopy aims to combine observations made with disparate techniques, most commonly light and electron microscopy (referred to as CLEM), with the goal of combining the strengths of each imaging approach. For light microscopy, it is the ability to chronicle dynamic cellular processes in living systems or to visualize multiple specific labels over large fields of view, while the electron microscope offers the spatial resolution needed to determine precise subcellular distributions in the context of cellular ultrastructure. The technique of serial blockface scanning EM (SBEM) is revolutionizing biological electron microscopy by offering unprecedented 3-D views of whole cells and tissues at near nanometer-scale resolution [1]. SBEM employs a miniature ultramicrotome fitted inside an SEM, and instead of imaging sections, it uses backscattered electrons (BSEs) to image the blockface in between repetitive cutting cycles that remove a thin layer of material from the block surface. Continued advancements in instrumentation and specimen preparation protocols have improved the spatial resolution achievable by SBEM such that most organelles and cellular constituents can readily be resolved, making it an ideal approach for CLEM. Specimen preparation for SBEM differs from conventional TEM in that the sample, whether it is a cell culture monolayer or complex tissue, requires much more substantial heavy metal staining than is normally employed for TEM [2]. The intense metal staining reduces specimen charging and improves the BSE yield and image resolution. However, this staining process renders most tissues completely opaque to light, requiring special procedures such as the use of reference fiducial markers and/or X-ray microscopy in order to track and image a precise subregion of a sample by SBEM [3]. SBEM also places several constraints on the labeling approaches that can be used, since the label must be introduced into a bulk specimen. We have focused our efforts on developing molecular-genetic and chemical-labeling approaches designed to facilitate both SBEM and CLEM. These include the tetracysteine/biarsenical labeling system [4], the recombinant fluorescent protein MiniSOG [5], APEX2 (a small and versatile genetically modified ascorbate peroxidase [6]), and a variety of chemical labels that can be used for fluorescence photooxidation of diaminobenzidine. The localization of proteins, macromolecules and organelles using these probes have a number of key advantages over other methods including: 1) excellent preservation of cell ultrastructure, since conventional EM fixation methods can be employed and no permeablizing detergents such as Triton or saponin are required; 2) uniform 3-D labeling can be achieved throughout relatively large volumes of tissue for 3-D light 1The Microsc. Microanal. 21 (Suppl 3), 2015 1382 and EM; 3) high-resolution labeling. Recent developments with these probes and others will be discussed in the context of their application towards SBEM and CLEM. References: [1] W Denk and H Horstmann PLoS Biol. 2004 Nov; 2(11). [2] TJ Deerinck et al, Microsc. Microanal. 2010 16 (suppl 2), 1138-1139. [3] EA Bushong et al, Microsc Microanal. 2014 Nov 13:1-8. [4] G Gaietta et al, Science 2002 Apr 19;296 (5567):503-7. [5] X Shu et al, PLoS Biol. 2011 Apr; 9(4):e1001041. [6] SS Lam et al, Nat Methods. 2015 Jan;12(1):51-4. Figure 1. 3-D reconstruction of a dividing cell imaged by SBEM. The cell was transfected to express the fluorescent protein MiniSOG (molecule enlarged and shown in schematic at upper left) fused to Histone 2B to label chromosomes (shown in yellow). The outline of the cell is shown in blue. Scale bar = 5 microns.");sQ1[691]=new Array("../7337/1383.pdf","Improving Our Vision of Nanobiology","","1383 doi:10.1017/S1431927615007709 Paper No. 0691 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Improving Our Vision of Nanobiology Deborah F. Kelly1 1. Virginia Tech Carilion School of Medicine and Research Institute, Roanoke, VA 24016, USA. Understanding the properties of molecular machines is a common goal of biologists and engineers. As such, transmission electron microscopes (TEMs) are often used to directly view the intricate details of molecular entities at the nanoscale. Biological assemblies that are preserved under frozen-hydrated conditions permit us to peer into the nanoworld in a near-native environment. This form of specimen preservation can be accomplished by plunge-freezing samples at high velocity into liquid ethane, thereby embedding the starting material in thin films of vitreous ice (Figure 1A). Conventionally, amorphous carbon is the most popular support material used to prepare biological specimens for TEM imaging. Micron-sized holes engineered into these carbon films provide a suitably transparent background to visualize biological complexes while embedded in vitreous ice (Figure 1B). However, there remain many limiting factors in high-resolution imaging, such as specimen charging, beaminduced movements, and other noise-producing artifacts. The revolution in EM phase-plates, direct electron detectors, and in-column energy filters offers premiere technology to record pristine images of weak-phase objects. Correspondingly, the next generation of specimen support materials must also be developed to best utilize these new tools. New materials are being produced worldwide, which present a prime opportunity to test alternative substrates for EM support films. Such alternative substrates include, but are not limited to, conductive titaniumsilicon metal glass (Ti88Si12) [1], silicon carbide (cryomesh) [2], graphene [3], and silicon nitride (cryo-SiN) [4]. One recent example is the use of cryo-SiN to tether active viral assemblies while preserving them for cryo-EM imaging (Figure 1C) [5]. Another major benefit of using alternative substrates, such as silicon nitride, is their versatile surface properties. By decorating microchips or other substrates with specific adaptor molecules, we can create new tunable devices. Tunable microchips have been recently used to capture and visualize native protein assemblies from the nuclear material of patient-derived cancer cells (Figure 2). Moreover, with these new tools in hand, the field is uniquely poised to elucidate molecular processes in the context of human disease. In this tutorial, I will discuss a variety of considerations toward optimizing the preparation of biological macromolecules, ranging from the biochemical to the specimen preservation level. References: [1] Rhinow, D. and W. Kuhlbrandt. Ultramicroscopy. 108 (2008), pp. 698�705. [2] Yoshioka, C., Carragher, B., and C. S. Potter. Microsc. Microanal. 16 (2010), pp. 43�53. [3] Russo, C. J., and L. A. Passmore. Nat. Meth. 11 (2014), pp. 649-652. [4] Tanner, J. R., Demmert, A. C., Dukes, M. J., Melanson, L. A., McDonald, S. M., and D. F. Kelly. J. Anal. Mol. Tech. 1 (2014), pp. 1-6. [5] Kam J. A., Demmert, A. C., Tanner, J. R., McDonald, S. M., and D. F. Kelly. Technology. 2 (2014), pp. 1-5. [6] Scheres, S. H. J. Mol. Biol. 415 (2012), pp. 406�418. Microsc. Microanal. 21 (Suppl 3), 2015 1384 Figure 1. (A) Biological specimens are blotted and plunge-frozen into liquid ethane for cryopreservation and TEM imaging. A comparison of active rotavirus assemblies imaged over holey carbon support film (B) and optimized on cryo-SiN (C). Scale bar is 100 nm. Figure 2. Native protein assemblies can be directly isolated upon tunable microchips from the nuclear material of patient-derived cancer cells. Captured assemblies were imaged using cryo-EM and reconstructed using 3D computing routines in the RELION software package [6].");sQ1[692]=new Array("../7337/1385.pdf","Electron Beam-Induced Charging and Modifications of Thin Films","","1385 doi:10.1017/S1431927615007710 Paper No. 0692 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electron Beam-Induced Charging and Modifications of Thin Films Marek Malac1,2, Marco Beleggia5,7, Teddy Rowan1,3, Ray Egerton1,2, Masahiro Kawasaki4, Yoshio Okura4, Robert A. McLeod6 1 2 National Institute of Nanotechnology, 11421 Saskatchewan Drive, Edmonton, Canada Department of Physics, University of Alberta, T6G 2E1, Edmonton, Canada 3 University of British Columbia. 2329 West Mall, Vancouver, V6T 1Z4, Canada 4 JEOL Ltd. 1-2 Musashino 3 chome, Akishima, Tokyo 198-8558, Japan 5 Centre of Electron Nanoscopy, Denmark Technical University, Denmark 6 Fondation Nanosciences, 23 rue des Martyrs, 38000 Grenoble, France 7 Helmholtz-Zentrum Berlin fuer Materialen und Energie, 14109 Berlin, Germany A thin film irradiated by high-energy primary electrons (PE) emits secondary electrons (SE). The SE are either emitted from the sample (SE) or travel within the sample (SE) [1]. Figure 1 illustrates the various types of SEs. The PEs, SE and the SE affect, by modifying the film properties, the rate at which the irradiated sample area reaches a steady state. Both PE and SE can cause radiation damage [2] and electron beam induced conductivity (EBIC) [1]. The discussed phenomena are relevant to sample charging, damage and to various implementations of hole-free phase plate (HFPP) [3, 4]. The beam-induced modifications can result in a spatiallydependent surface potential (r), generating an electric field EV(r,z) that extends in the vacuum surrounding the sample [3]. An irradiated thin film tends toward a steady state, as indicated by the time dependence of SE current and the evelution of the contrast transfer function (CTF) [3,4]. Figure 2 shows SE signal (a) and, (b), the position of extrema in a CTF as a function of time. Both SE and CTF time evolution can be fitted to: Y(t) = Y0 + A0 exp(-t/t0) Eq. (1) Here Y(t) and Y0 are the dependent variable (SEa signal or CTF extrema position) and their initial values respectively, and t0 is the exponential characteristic time, which we take as a measure of the settling time. The values of t0 obtained from SE and CTF correlate; they are a few seconds to a few tens of seconds for all studied materials and conditions. According to Eq. (1), the film state is within 2% of its asymptotic steady state after tS = 4t0. A steady state is achieved when a balance is reached between the SE leaving the sample and the current from ground electrode IG. At steady state, the electric field outside sample EV(r,z) reduces the SE until it is fully compensated by the IG. When multiple processes control the settling the Eq. (1) includes more than one term. The settling time tS is then determined by the slowest process. The A0 > 0 for SE decay and for Thon rings that decrease diameter with time (i.e. the defocus and charging term in a CTF contribute phase shift with the same sign). The A0 < 0 for Thon rings when defocus and charging contribute opposite sign of phase shift. Microsc. Microanal. 21 (Suppl 3), 2015 1386 The irradiated area appears to remain close to the steady state after the PE beam is switched off [3,4,5]. This can be explained by existence of a charge-trapping layer on the film surface with electron tunneling governing the recombination charge transport. At steady state, the emitted SE current density jSE(r) equals to a tunneling current jT(r) at every location r of the film, as graphically depicted in Figure 3 for Fowler-Nordheim (FN) and Poole-Frenkel (PF) tunneling mechanisms. The tunneling current exponentially depends on electric field ET(r,z) inside the charge trapping layer which is related to the jSE(r). When the PE irradiation ceases and the jSE(r) goes to zero, an exponentially small decrease in ET(r,z) switches off the jT(r) effectively maintaining the steady state charge distribution. Figure 3 indicates that both FN and PF mechanisms achieve steady state for moderate surface potential values. The SE escape depth is an important parameter in models capturing the behavior of irradiated films. Figure 4 shows the dependence of SE signal and sample thickness for an insitu drilled amorphous carbon film at 600�C. Because the SE escape from both surfaces, the must be less than � (6.8 nm) = 3.4 nm. Since the escape depth is large in insulators and small in conductors [1], amorphous carbon at T = 600�C behaves as a good conductor. The relative importance of EBIC, SE emission and charge tunneling jT(r) is determined by mobility of charge carriers in the film, PE current density, by SE yield and its dependence on bias and thickness and dielectric properties of any surface layers on the film including contamination. The contribution of SE to EBIC generation is small compared to PE [2]. 1387 Microsc. Microanal. 21 (Suppl 3), 2015 References [1] L. Reimer, Scanning Electron Microscopy 2nd ed., Springer, 1998. [2] RF Egerton, M Malac Microscopy and Microanalysis 10 (S02) (2004), p.1382. [3] M. Malac et. al., Ultramicroscopy 118 (2012), p. 77. [4] R. Danev et. al. PNAS, 111 (2015), p. 15635. [5] J. Berriman and K. L. Leonard, Ultramicroscopy 19 (1986), p. 349. [6] Support of NINT, NSERC, JEOL USA Inc. and JEOL Ltd. is gratefully acknowledged. Fig. 1. Primary electrons (PE) incident on a thin film generates SE that escape to vacuum from a depth , and SE that travel within the specimen. The PE and SE generate e-h pairs that affect sample conductivity. Fig. 3. At a steady state a balance between jSE(r) and jT(r) is maintained. An intersection of I-V curves corresponds to a balance. The FN tunneling (blue) has much steeper IV dependence than the PF tunneling (green). The jSE(r) (red) is flat at typical values of surface potential. Fig. 2. Time traces of SE current and CTF extrema from a 15 nm thick carbon film at room temperature and PE beam current density jPE = 0.1 A/ cm2. Both the SE time traces and the position of extrema in CTF obey an exponential decay in Eq. (1): SE ~ 4s, CTF ~ 8 to 10 s. Fig. 4. SE current and film thickness of an in-situ drilled carbon film at 600�C. The SEa escape depth is less than � 6.8 nm, i.e. less than ~30 atomic layers. The jPE = 0.5 A/cm2. Microsc. Microanal. 21 (Suppl 3), 2015 1388");sQ1[693]=new Array("../7337/1389.pdf","Comparison of Cryo TEM Images Obtained with Zernike and Hole-Free Phase Plates","","1389 doi:10.1017/S1431927615007722 Paper No. 0693 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Comparison of Cryo TEM Images Obtained with Zernike and Hole-Free Phase Plates Naoki Hosogi1 ,Anindito Sen1 and Hirohumi Iijima1 1. EM Business Unit, JEOL Ltd., Akishima, Japan Cryo transmission electron microscopy (TEM) provides structural information of a specimen close to its natural state without any disturbance due to the specimen preparation process. The cryo specimens exhibit low contrast in TEM images even using large defocus phase contrast imaging. Cryo TEM is more advantageous when it is combined with a technique of contrast enhancement using a phase plate. Several applications of a Zernike phase plate (ZPP) were already reported [1, 2]. The ZPP is made of a thin amorphous carbon film with a center hole whose diameter is less than 1 m. The contrast enhancement by the ZPP can be achieved by phase shifting the scattered beam by half with respect to the unscattered direct beam. The half phase shift is given to the diffracted beams by the thin film of the phase plate placed at the back focal plane, while the direct unscattered electrons just pass through the center hole with no phase shift. However, characteristic fringes appeare around edges of high-contrast objects such as high density particles in the images with a ZPP [3, 4]. The fringe artifact results from the sudden jump of the contrast transfer function due to the sudden cut-on frequency of the phase plate hole. A ``hole-free phase plate'' (HFPP) [5] is considered to reduce the fringe artifact. The HFPP is a thin amorphous carbon film phase plate without hole. In the HFPP, the charging at a local area in the thin film, which is induced by the direct beam, acts as a tiny sized phase shifter to achieve a phase contrast. Diameter of this charging area is estimated to be several tens of nm, which is pretty smaller than the center hole of the ZPP, when we use FEG for the electron source. Therefore, the fringe effect of HFPP is less than that of ZPP. In the present study, we tried cryo TEM observations for some biological specimens with the ZPP and the HFPP. The ZPP and HFPP images (Fig. 1b and c) with a near-to-focus condition exhibited highly improved contrast compared to the largely defocused phase contrast image (Fig. 1a). Reduced fringe contrast is observed in a HFPP image shown in Fig. 1c, compared with a ZPC image shown in Fig. 1b. In addition to the reduction of the fringe contrast, the HFPP does not require any alignment procedure due to lack of a center hole. Thus, the data acquisition with HFPP could be done easier than that with ZPP. The high contrast HFPP images with the reduced fringe effect should lead to a higher-resolution observation of cryo specimens. Microsc. Microanal. 21 (Suppl 3), 2015 1390 References [1] K Nagayama et al, Microsc. Today 18 (2010), p. 10. [2] W Dai et al, Nature 502 (2013), p. 707. [3] R Danev and K Nagayama, Biophysics 2 (2006), p. 35. [4] Y Fukuda et al, J. Struct. Biol. 168 (2009), p. 476. [5] M Malac et al, Ultramicroscopy 118 (2012), p. 77. [6] Mr. N Tominaga (National Cancer Center Research Institute and TheUniversity of Tokyo) is thanked for kindly providing exosomes. a b c Figure 1. Cryo TEM images of ice-embedded exosomes acquired with a few microns of defocus (conventional) (a), ZPP (b) and HFPP (c). The image in B and C was taken at a near-to-focus condition. Scale bar: 200 nm.");sQ1[694]=new Array("../7337/1391.pdf","Practical Aspects and Usage Tips for the Volta Phase Plate","","1391 doi:10.1017/S1431927615007734 Paper No. 0694 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Practical Aspects and Usage Tips for the Volta Phase Plate Radostin Danev1, Bart Buijsse2, Maryam Khoshouei1, Yoshiyuki Fukuda1, Wolfgang Baumeister1 1. 2. Max Planck Institute of Biochemistry, Martinsried, Germany. FEI, Eindhoven, The Netherlands. Compared to other technical advances in transmission electron microscopy (TEM), such as brighter electron sources, energy filters, better optics and direct detection cameras, phase plates have lagged behind in both development and applications. The main reason for that is the difficulty in solving practical issues, such as beam-induced electrostatic charging, which cannot be reliably predicted or avoided based on theoretical research. Another reason is that phase plates add more complexity to the operation of the microscope and until not so long ago were not usable for automated data acquisition. The recently developed Volta phase plate (VPP) [1] solves most of the practical issues and is relatively easy to use and automate. It is already being routinely used in cryo-tomography applications where it has demonstrated remarkable improvements in contrast and visibility of features [2,3]. In single particle applications the VPP has shown some promise but the resolutions achieved thus have been limited to about 8 �. In this paper we present a few examples of the use of VPP in both cryo-tomography and single particle applications. In addition to the application results we will discuss the practical aspects of using a VPP. The most important microscope alignments for proper VPP operation are the on-plane condition and the beam shift pivots (Fig. 1). The on-plane condition (Figs. 1a, b) concerns the position of the central beam crossover at the back focal plane relative to the phase plate. The VPP operates best on-plane, which provides a proper phase shift of ~ /2 (Fig. 2) and the lowest possible cut-on frequency. In suitably designed microscopes the onplane condition for the phase plate corresponds to parallel illumination on the specimen. The beam shift pivot point setting (Figs. 1c, d) is responsible for the lateral stability of the beam crossover on the phase plate. When the pivot point is properly set the beam crossover remains stationary at a fixed point on the phase plate irrespective of the beam movements at the specimen plane. This setting is especially important for low-dose operation where beam shifts combined with image shifts are used to offset the observation area on the specimen for e.g. focusing. In addition to the on-plane and beam shift pivot point alignments the third important factor for a successful VPP observation is proper conditioning of the phase plate. The phase shift of the VPP is created by the central beam and builds up gradually with the total dose on the phase plate. The blue curve in Fig. 2 can be used to calculate the pre-irradiation time required for creating a desired phase shift. The time in seconds is equal to the dose in nC (from the horizontal axis in Fig. 2) divided by the total beam current in nA. [1] R Danev et al, PNAS 111 (2014) p. 15635. [2] S Asano et al, Science 23 (2015) p. 439. [3] Y Fukuda et al, Journal of Structural Biology (2015) in press. Microsc. Microanal. 21 (Suppl 3), 2015 1392 Figure 1. (a, b) On-plane and off-plane conditions. (c, d) Correct and incorrect setting of the beam shift pivot points. Figure 2. Phase shift vs. dose for the VPP on-plane (blue curve) and off-plane (red curve).");sQ1[695]=new Array("../7337/1393.pdf","Combination of Different Techniques in Cryo-Electron Tomography with a Volta Phase Plate","","1393 doi:10.1017/S1431927615007746 Paper No. 0695 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Combination of Different Techniques in Cryo-Electron Tomography with a Volta Phase Plate Maryam Khoshouei1, G�nther Gerisch2, Maria Ecke2, Julio Ortiz1, J�rgen M. Plitzko1, Radostin Danev1, Wolfgang Baumeister1 1. Department of Molecular Structural Biology, Max-Planck Institute for Biochemistry, Am Klopferspitz 18, D-82152, Martinsried, Germany. 2. Department of Cell Dynamics, Max-Planck Institute for Biochemistry, Am Klopferspitz 18, D-82152, Martinsried, Germany. Due to the sensitivity of the biological samples to beam irradiation resulting from induced changes to their structure, the limited allowed dose to frozen-hydrated biological specimens leads cryo images typically noisy and low in contrast. One way to improve the contrast is by applying few microns of defocus which is a well-known conventional technique since many years ago. The drawback of this technique is resolution deterioration caused by applied defocus that suppresses high-frequency information due to its sine type CTF. As opposed to the convectional technique, another way of increasing contrast is using a phase plate. Phase plate is located on the back-focal plane of the TEM. It changes the sine type CTF to a cosine type and the object information in the low spatial frequency domain is transferred well into the image information. There are various types of phase plates such as electrostatic, magnetic, anamorphotic, tulip type and etc. Nevertheless, none of them is as applicable as the thin film phase plate in terms of difficulties in fabrication and usage [1]. Among all different types of thin film phase plate, Zenrike phase plate has been the most successful one until the proposed novel phase plate. In the last two years, Volta phase plate (VPP) was developed and it improves on the performance of the Zernike phase plate by having a longer life time, overall better performance, simplicity in use and no fringe artefacts. The performance of the VPP is based on volta potential resulting from the incident electron beam and physical and chemical changes on the surface of the carbon [2]. Moreover, one of the model systems to evaluate of the performance of the VPP is an earth worm sperm. The worm sperm was obtained from the seminal receptacles bags of the Lumbricus terretris. Understanding the morphological and ultrastructural information play a significant role in taxonomic and phylogenetic studies. However, this study could not be done unless applying the combination of cryo-electron tomography (CET) techniques and VPP. This huge cell comprises different compartments such as acrosome, nucleus, midpiece and flagellum. In this work, the combination of batch CET, single axis CET and dual axis CET with VPP are shown to observe different segments of this specimen. This is a first time to observe such a complicated vitrified biological system with a preserved structure and more importantly in the physiological status. In contrast, from the pertinent researches this sample was investigated based on tissue fixation and plastic sectioning with less visible structural details [3]. In the current work, the improved contrast makes the structural details more visible. These structural details could be critical factors in the cell motility, signaling of the sperm cell towards an ovum and fertilization process. Microsc. Microanal. 21 (Suppl 3), 2015 1394 [1] R Danev et al, Journal of Structural Biology 171 (2010), p. 174. [2] R Danev et al, Proceedings of the National Academy of Sciences of the United States of America (2014). [3] B. G. M Jamieson, Journal of Zoology 26 (1978), p. 225. (a) (b) (c) Figure 1. Slices of CET with the VPP (a) Single axis CET with the VPP from the end of the acrosome, scale bar: 400nm (b) Dual axis CET with the VPP from the midpiece. Visibility of a transition gate (left inset image) and triplet microtubules (right inset image), scale bar: 400nm. (c) Single axis CET with the VPP from the flagellum. Visibility of the doublet microtubules (upper inset image) and two singlets (lower inset image), scale bar: 400nm. [Titan Krios 300kV, energy filter, direct detector, defocus: 500nm, Magnification: 33000, 26000; Pixel size: 0.421 and 0.531 nm respectively]. Figure 2. One projection of batch CET with the VPP. 15 single axis CET with a complete overlap from a worm sperm cell Slices of CET with the VPP. [Titan Krios 300kV, energy filter, direct detector, defocus: -500nm, Magnification: 19500, Pixel size: 0.71 nm]. Scale bar: 3�m");sQ1[696]=new Array("../7337/1395.pdf","Electron Holography for Measuring Electrostatic Potentials and Strain Fields","","1395 doi:10.1017/S1431927615007758 Paper No. 0696 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electron Holography for Measuring Electrostatic Potentials and Strain Fields Michael Lehmann1 1 Technische Universit�t Berlin, Institut f�r Optik und Atomare Physik, Berlin, Germany Today's latest high-resolution transmission electron microscopes (HRTEM) with image correctors are fantastic instruments easily allowing the imaging of atomic structures with resolutions better than 100 pm and precision of atomic column positions of a few pm. This is amongst other things the result of better control of residual aberrations and the tremendous reduction of delocalization leading to a strongly improved signal-to-noise ratio at atomic positions. However at the same time, it deteriorates the transfer of large area phase contrast represented as low spatial frequencies. Consequently, an image corrected HRTEM is a band-pass filter for phase information, good for atomic resolution, bad for measuring electrostatic potentials due to e.g. p-n junctions. This phase problem of imaging is solved by off-axis electron holography. By means of a M�llenstedt biprism, which is only a minor supplement to a FEG-based HRTEM, amplitude and phase of the electron wave with all spatial frequencies up to the information limit are recorded within an interference experiment. Based on the fundamental considerations by Lichte [1] and Harada [2], latest holographic microscopes with at least two biprisms offer a higher experimental flexibility for adjusting width and fringe spacing of holograms [3]. In combination with improved instrumental stability, less Fresnel diffraction, reduced vignetting effect, and the acquisition of hologram series consisting of 20 or more holograms and their subsequent reconstruction considering the drifts of specimen and biprism over the series, a full reconstruction of a GaN crystal's object exit-wave up the information limit of 75 pm of the instrument without transfer gaps has been demonstrated with a high signal-to-noise ratio allowing a precise comparison with simulations [4]. This flexible experimental setup also enables tackling fundamental questions arising from applying electron holography to real world problems: Electron holography is often used for measuring the built-in voltage of p-n junctions in doped semiconductors. However, the measured value is always smaller than the expected one as calculated from the dopant levels. It is reported that at least for Silicon providing a conductive path to ground by coating with Carbon the discrepancy is strongly reduced. Up to now, an effect only partially considered is the influence of the electron beam illumination producing electron hole pairs so that the p-n junction acts like a solar-cell illuminated in the TEM instead of light with electrons. Additionally, the high-energy electron beam leads to the generation of secondary electrons as observed as positive charging of samples, which are badly connected to ground. Using needle-shaped GaN p-n junctions prepared by FIB, careful holographic experiments (figure 1) by reducing the electron dose rate over three orders of magnitude, but acquiring hologram series with an accumulated exposure time up to 1000 s enabling low dose rate electron holography, have shown that indeed electron-hole pair generation plays a significant role in explaining the discrepancy whereas the generation of secondary electrons can be neglected since they do not produce a net current over the p-n junction. Of overall importance for defined experimental conditions is a small shunt, which in the Silicon system can easily be obtained by Carbon coating, whereas in the GaN system a much larger interface resistance between bulk and conducting surface layers must exist [5]. An interesting derivative of off-axis electron holography is dark-field electron holography for measuring Microsc. Microanal. 21 (Suppl 3), 2015 1396 strain fields as developed by Hytch [6]. Here, a diffracted beam of an unstrained crystalline reference area is brought to interference with the corresponding reflection stemming from a strained area. The resulting phase gradient is a measure for local variations of geometrical phase. By repetition of the experiment with a second noncollinear reflection, the full 2D strain field can easily be evaluated by simple matrix algebra. Also dark-field electron holography benefits from the flexible holographic setup by strongly reducing Fresnel diffraction at the biprism filament and the recording of holographic series yielding improved signal-to-noise ratio. This is demonstrated at a GaAs mesa structure with a buried AlOx current aperture (figure 2) where the in-plane tensile strain promotes the aimed nucleation of InAs quantum dots in the middle of the aperture [7]. Quantitative comparison with corresponding calculations based on linear elasticity theory clearly shows that the strain relaxation in a TEM lamella even at a thickness of a few 100 nm has to be considered in the simulation. [8] References: [1] H. Lichte, Ultramicroscopy 64 (1996) 79. [2] K. Harada et al., Appl. Phys. Lett. 84 (1994) 3229. [3] F. Genz et al., Ultramicroscopy 147 (2014) 33. [4] T. Niermann and M. Lehmann, 63 (2014) 28. [5] Jae Bum Park et al., Appl. Phys. Lett. 105 (2014) 094102. [6] M. Hytch et al., Nature 453 (2008) 1086. [7] F. Kie�ling et al., Phys Rev. B 91 (2015) 075306. [8] The results represent the joint efforts of the workgroup at TU Berlin especially by Tore Niermann, Felix Kie�ling, Jae Bum Park, and Florian Genz. The financial support through the DFG (CRC 787, TP A4) is gratefully acknowledged. Figure 1. Equivalent circuit of the p-n junction under electron illumination. The conductive surface is altered by plasma and KOH etching. The measured built-in voltage is calculated for an active specimen thickness of 320 nm. Figure 2. Left: Bright-field TEM of cross-sectional mesa structure showing clearly the extension of the AlOx current aperture. Right: Strain components as determined from dark-field electron holography.");sQ1[697]=new Array("../7337/1397.pdf","Characterization of Trapped Charge in Ge/LixGe Core/Shell Structure during Lithiation using Off-axis Electron Holography","","1397 doi:10.1017/S143192761500776X Paper No. 0697 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of Trapped Charge in Ge/LixGe Core/Shell Structure during Lithiation using Off-axis Electron Holography Z. Gan1, M. Gu2, J. Tang3, C. Y. Wang4, K. L. Wang3, C. M. Wang2, D. J. Smith1, and M. R. McCartney1 Department of Physics, Arizona State University, Tempe, AZ 85287. Environmental Molecular Sciences Lab, Pacific Northwest National Laboratory, Richland, WA 99352. 3. Device Research Lab., Dept. Electrical Engineering, University California, Los Angeles, CA 90095. 4. Department of Materials Science and Engineering, National Taiwan University of Science and Technology, Taipei City, Taiwan, 10607, Republic of China. 2. 1. Lithium ion batteries (LIBs) have important applications as energy-storage systems for portable electronics, electric vehicles, and sources of renewable energy [1,2]. Ge is being considered as a possible alternative to graphite as anode for its theoretically high capacity and high Li+ diffusivity, despite of its large volume change upon lithiation/delithiation [3]. Knowledge of the charge distribution during lithiation is important to better understand the lithiation mechanism and the associated electrochemistry. Off-axis electron holography (EH) is an interference technique that can effectively measure electrostatic fields with nanoscale spatial resolution [4]. After hologram reconstruction, the projected potential distribution of the object along the electron beam direction can be retrieved from the phase image and compared with calculations. Here, EH has been used to characterize the charge distribution across Ge/LixGe core/shell nanowires (NWs) during lithiation. The Ge NWs were grown along [111] directions using the vapor-liquid-solid (VLS) method on Si substrates. The NW was mounted on a Nanofactory holder and attached to Li metal for in-situ lithiation and EH measurement. Figures 1a, 1d and 1g show sequential holograms of Ge NW taken during lithiation. In Figure 1a and 1d, Ge/LixGe core/shell structure was observed, as shown by the darker contrast in the NW center. The Ge core was reduced from 45nm to a faceted one of 16nm in radius, while the outer shell increased from 88nm to 110nm in radius. Li EELS mapping and HAADF image (not shown) taken after Figure 1a confirm the Ge/LixGe core/shell structure. In Figure 1g, the Ge core disappeared and the NW radius increased to 125nm, corresponding to a chemical composition of LixGe (x~3.75). The reconstructed phase images are shown in Figures 1b, 1e and 1h, respectively, in which pseudo-color is used to show the phase change. Phase profiles along the white arrows are shown on the right. The center part of the NW has higher phase compared to the outer part, due to the greater thickness and higher mean inner potential (MIP) of crystalline Ge, compared to LixGe. Both the core and shell parts mimic a cylindrical NW shape. To interpret phase profiles, it is first assumed that there are no trapped charges in the NW and that the phase shift is only due to MIP Vshell. Assuming the NW has a cylindrical shape, the phase shift due to the shell can be calculated and compared with the experimental results, as shown by the red curves in the phase profiles. As lithiation continued, Vshell changed from 7.6V�0.1V to 6.4V�0.1V and then 5.1V�0.1V. This drop indicates that the Li component has increased during lithiation, because Li is a lighter element and has smaller MIP, compared to Ge. The bias applied to the NW should not affect this conclusion since the bias was fixed at -2V. Similar fitting was applied for the core part in Figure 1c, as shown by the blue curve. However, the best fitted Vcore is only 10.6�0.1V, compared to the value of 14.3V for crystalline Ge [5]. This difference suggests that charges are trapped in the NW during lithiation. A proposed model, as shown in Figure 2a, assumed that Li+ ions accumulated in the Li2O layer on the NW surface, while electrons were uniformly trapped in the Ge NW core. The fitted result is shown in Figure 2b, where the charge density was calculated to be 3�1018 electrons/cm3 and the MIP for the shell was 8.4V. Further investigations are in progress [6]. Microsc. Microanal. 21 (Suppl 3), 2015 1398 References: [1] M. Armand and J. M. Tarascon, Nature 451 (2008) 652. [2] A. S. Arico, et al., Nature Materials 4 (2005) 366. [3] D. Larcher, et al., Journal of Materials Chemistry 17 (2007) 3759. [4] M. R. McCartney and D. J. Smith, Ann. Rev. Mater. Res. 37 (2007) 729. [5] J. Li, et al., Acta. Crystal. Sec. A 55 (1999) 652. [6] The electron holography studies have been supported by DoE Grant DE-FG02-04ER46168. We also acknowledge use of the EMSL user facilities. Figure 1. (a), (d) and (g) Holograms of Ge/LixGe core/shell NW structure during lithiation; (b), (e) and (h) Corresponding reconstructed phase images, shown in pseudo-color (scale bar shown at top right is in units of radian); (c), (f) and (i) Phase profiles along the white arrows in (b), (e) and (h), respectively. Figure 2. (a) Schematic diagram of the model for trapped charges in Ge/LixGe core/shell NW structure; (b) Experimental data (black) and best fitted results (red).");sQ1[698]=new Array("../7337/1399.pdf","Aberration Corrected Off-Axis Electron Holography of Layered Transition Metal Dichalcogenides","","1399 doi:10.1017/S1431927615007771 Paper No. 0698 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Aberration Corrected Off-Axis Electron Holography of Layered Transition Metal Dichalcogenides Amir H. Tavabi1, Florian Winkler1, Yung-Chang Lin2, Kazu Suenaga2, Emrah Yucelen3, Rafal E. Dunin-Borkowski1, Beata E. Kardynal4 1 2 3 4 Ernst Ruska- Centre for Microscopy and Spectroscopy with Electrons and Peter Gr�nberg Institute 5, Forschungszentrum J�lich, Germany National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba, Japan FEI Company, Eindhoven, The Netherlands Peter Gr�nberg Institute 9, Forschungszentrum J�lich, Germany Two-dimensional transition metal dichalcogenides (TMDs) have attracted great attention for device applications in the last decade, in part due to their direct band gaps [1, 2]. Off-axis electron holography can provide unique information about the properties of TMDs that is complementary to the information provided by more conventional transmission electron microscopy (TEM) techniques. TEM samples of TMDs were prepared by cleaving, followed by transfer onto an elastomer gel film. Subsequently, the flakes were transferred onto gold-coated holey SiN membranes without the use of wet chemicals. Before TEM measurements, each specimen was annealed at 85 �C overnight in a vacuum furnace (< 1�10-5 mbar). Medium and high resolution off-axis electron holograms of a wide range of TMDs with nominal composition MX2 (M: Mo, W, Re; X: S, Se) were acquired at accelerating voltages of 50, 60 and 80 kV using spherical and chromatic aberration corrected TEMs. Figure 1 shows representative results acquired at 80 kV using off-axis electron holography from an MoS2 flake. The holographic interference fringe spacing was 30 pm and a mask corresponding to a spatial resolution of 90 pm was applied to the sideband (Fig. 1c) before reconstructing the phase image. Figure 1d shows part of a phase image reconstructed from the hologram, showing a two and three monolayer region of the flake and the presence of some contamination at the specimen edge. Phase images such as that shown in Fig. 1d were used to determine the mean inner potential, the number of monolayers and the number of atoms in each atomic column across the field of view. Challenges in such measurements were found to arise from the stability of the specimen and the instrument, the presence of residual aberrations, contamination on the specimen surfaces and electronbeam-induced charging and damage of the specimen. Charging effects are visible in the line profile shown in Fig. 1e, in the form of a slight slope in the phase in a region of uniform thickness inside the specimen when compared with the slope of the phase in vacuum. Approaches for overcoming such problems will be discussed. References: [1] Radisavljevic, B. et al, Nature Nanotech. 6 (2011), p. 147. [2] Wang, Q. et al, Nature Nanotech. 7 (2012), p. 699. Microsc. Microanal. 21 (Suppl 3), 2015 1400 Figure 1. a) Off-axis electron hologram of an MoS2 flake recorded at 80 kV. b) Magnified region of electron hologram showing fine shifts of interference fringes in a three-layer flake. c) Fast Fourier transform of the hologram, showing information at 84 pm. d) Part of a reconstructed phase image of two and three layer MoS2, showing the presence of contamination at the specimen edge. e) Phase shift profile extracted from the black rectangle marked in the phase image.");sQ1[699]=new Array("../7337/1401.pdf","In Situ Electron Holography of Ferroelectric Thin Films","","1401 doi:10.1017/S1431927615007783 Paper No. 0699 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In Situ Electron Holography of Ferroelectric Thin Films Myung-Geun Han1, Matthew S. J. Marshall2, Lijun Wu1, Frederick J. Walker3, Charles H. Ahn3 and Yimei Zhu1 1. 2. Condensed Matter Physics & Materials Science, Brookhaven National Laboratory, Upton, NY, USA Department of Applied Physics and Center for Research on Interface Structures and Phenomena, Yale University, New Haven, CT, USA Ferroelectric thin films have received intense research interests because of their switchable spontaneous polarization that can be utilized in technologically important applications, such as nonvolatile data storage and solar energy harvest [1]. However, size and shape of domains strongly depend on electrical and mechanical boundary conditions and are difficult to predict by general thermodynamic theory [2]. The domain structures and their switching properties can be best understood by directly visualizing domains (walls) during switching and characterizing the electrostatic potentials. We have utilized various electron microcopy techniques combined with in situ electrical biasing to study polarization reversal in epitaxial Pb(Zr0.2Ti08)O3 (PZT) thin films grown on the (001) Nb-doped SrTiO3 (Nb-STO) substrate [3]. TEM samples in a capacitor form are prepared by FIB lift-out and milling. Electrical contacts have been made using piezo-controlled metallic probes (Nanofactory Instruments, AB). We have directly observed ferroelectric domain switching process under various external biases using dark-field TEM imaging (Fig. 1a-c) by tilting the sample to a two-beam condition in order to maximize the 180o domain contrast resulted from the violation of Friedel's law [4]. Surprisingly, head-to-head domain walls were found in as-grown samples and the domains near the PZT/Nb-STO interface were not switchable. Off-axis electron holography results showed unidirectional potential barrier and electric fields (Fig. 1d and e) across the PZT/Nb-STO interface. These were confirmed by EBIC showing strong electric fields (Fig. 1k-m) found near the interface under positive external biases applied to the Nb-STO substrate. The observed unidirectional electric fields can be explained by considering band bending and corresponding charge distributions near the PZT/Nb-STO interface. The presence of polarization charge will influence the zero-bias band bending (Fig. 1f) model. Conceptually, the polarization charge can be treated as an external bias to induce the same amount of charge within the ferroelectric. For the negative polarization charge near the PZT/Nb-STO interface can be considered as a forward biasing the heterojunction, thus reducing the band bending (Fig. 1g). This scenario of flattening band bending by polarization implies that the PZT/Nb-STO interface behaves as a pn junction. When a forward bias is applied, the band bending is further reduced, yielding no significant electric fields that can initiate domain switching. As a result, in the course of domain switching, energetically unfavorable charged domain walls with head-to-head configuration are induced in the PZT thin films. The formation of charged domain walls reported in this study can be an origin for imprint and retention loss in ferroelectric thin films, major obstacles for practical applications. References: [1] Setter, N. et al, Journal of Applied Physics, 100 (2006). [2] Catalan, G. et al, Review of Modern Physics, 84 (2012). Microsc. Microanal. 21 (Suppl 3), 2015 1402 [3] Han, M.-G. et al, Nature Communications, 5 (2014). [4] Blank, H. et al, Applied Physics Letters 2, 140 (1963). [5] The authors acknowledge funding from Division of Materials Science and Engineering, Office of Basic Energy Science, the U.S. Department of Energy, under Contract number DE-AC02-98CH10886 and NSF, MRSEC under DMR11926 (CRISP) and DMR 1309868. Dr. Toshihiro Aoki and Dr. Ray Twesten are acknowledged for their useful discussion and helps on electron microscopy. Figure 1. Dark-field TEM images showing ferroelectric domain switching in PZT thin film: as grown (a), after 10 V (b), and after -10 V, respectively. Electrostatic potential line profiles (d) and electric fields along the c axis (e) obtained by electron holography. Schematic band diagrams without (f) and with (g) negative polarization charge (-Ps) at the PZT/Nb-STO interface and corresponding charge distributions (h). Schematic (i) of simultaneous acquisition of electron-beam-induced-current (EBIC) images (k ~ m) with HAADF STEM image (j). Line profiles of EBIC (n) showing electric field distributions that are consistent with holography data (e).");sQ1[700]=new Array("../7337/1403.pdf","The Perfect Cut: Focused Ion Beam Preparation for In Situ TEM","","1403 doi:10.1017/S1431927615007795 Paper No. 0700 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Perfect Cut: Focused Ion Beam Preparation for In Situ TEM Andrew Lang 1, Wayne Harlow 1, Michael Jablonski1, James Hart1, Christopher Barr1, Hessam Ghassemi1, Osman El-Atwani1, and Mitra L. Taheri1 1. Drexel University, Department of Materials Science & Engineering, Philadelphia, PA, USA In situ TEM techniques have improved considerably in recent years with respect to their ability to understand materials behavior with high temporal and spatial resolution. While significant advances have been made in elucidating atomic-scale mechanisms that control properties of materials for a wide range of applications, geometric compromises made to accommodate in situ TEM experiments could play a detrimental role in the ability to apply data to "real-life" structures or devices. In recent years, focused ion beam (FIB) preparation has gained a stronghold in the in situ TEM community as it allows for complex architectures to be prepared for specific holders. Because of the requirements and limitations of many in situ experiment geometries, the FIB provides a necessary platform toward achieving the proper lay-out, ranging from connecting electrical contacts to fabricating nanopillars. The overarching drawbacks of ion beam damage and preparation time, however, remain a challenge. This talk examines the interplay between the need to mimic a true application and the need to access microstructural features at the nanoscale, and how this can be achieved using FIB lift-out procedures. Case studies of high electron mobility transistors [1], ferroelectric memory devices [2-4], and alloys for nuclear energy applications will be reviewed. Specifically, the preparation of gated device structures for electrical degradation measurements and the analysis of microstructure-specific corrosion in Zircaloy will be presented. Microsc. Microanal. 21 (Suppl 3), 2015 1404 Figure 1. (a) SEM image of a typical AlGaN/GaN HEMT in the pristine (unbiased) state, (b) Crosssection view of the lift-out indicates the features of the device as well as the Pt protective layers. (c) and (d) represent weak-beam dark-field TEM images of drain-gate region at two different diffraction conditions of g � �0002 and g � �1120, respectively. Scale bars are 200 nm and the threading dislocations are labeled as 1�5, from [1] Figure 2. Schematic diagrams of an in situ biasing geometry for BFO (ferroelectric) domain switching experiments, revealing: (a) a plan-view of BFO device structure with SRO electrodes, (b) a side-view of FIB-thinned region of STO substrate with an enlarged schematic showing the planar applied voltage between the SRO electrodes, from [3] References [1] Ghassemi, H., A. Lang, C. Johnson, R. Wang, B. Song, P. Phillips, Q. Qiao, R.F. Klie, H.G. Xing, and M.L. Taheri. "Evolution of strain in aluminum gallium nitride/gallium nitride high electron mobility transistors under on-state bias." Journal of Applied Physics 114 (2013): 064507. [2] Winkler, C.R., M.L. Jablonski, A.R. Damodaran, K. Jambunathan, L.W. Martin, and M.L. Taheri. "Accessing intermediate ferroelectric switching regimes with time-resolved transmission electron microscopy." Journal of Applied Physics 112 (2012): 052013. [3] Winkler, C.R., A.R. Damodaran, J. Karthik, L.W. Martin, and M.L. Taheri. "Direct observation of ferroelectric domain switching in varying electric field regimes using in situ TEM." Micron 43 (2012):1121-1126. [4] Winkler, C.R., Jablonski, M.L., Ashraf, K., Damodaran, A.R., Jambunathan, K., Hart, J.L., Wen, J., Miller, D.J., Martin, L.W., Salahuddin, S., and M.L. Taheri. "Real-time observation of local strain effects on non-volatile ferroelectric memory storage mechanisms." Nano Letters 14.6 (2014): 36173622. [5] The authors acknowledge funding from the Office of Naval Research under contract number N000141410058, funding from the United States Department of Energy, Basic Energy Sciences under the Early Career program through contract DE-SC0008274, funding from the National Science Foundation's Faculty Early Career Program under contract #1150807, funding from the Office of Naval Research through contract N00014-1101-0296, and funding from the Department of Energy's Nuclear Energy University Program under contract NE0000315.");sQ1[701]=new Array("../7337/1405.pdf","In-Situ EDS Characterization of TEM Lamellae Created by Xe Plasma FIB","","1405 doi:10.1017/S1431927615007801 Paper No. 0701 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In-Situ EDS Characterization of TEM Lamellae Created by Xe Plasma FIB Tom�s Hrnc�1, Christian Lang2, Jozef Vincenc Oboa1 and Tom�s Bart�k3 1. 2. TESCAN, Brno, Czech Republic Oxford Instruments Nanoanalysis, High Wycombe, UK 3. Oxford Instruments, Prague, Czech Republic As a result of the rapidly progressing miniaturization in microelectronics, the critical dimensions of semiconductor devices are now so small that only TEM analysis can give conclusive answers with regard to fault isolation and failure analysis. In order to investigate structures on the nanometer and subnanometer scale, ultrathin TEM lamellae have to be prepared that contain the fault in question. As site specific preparation is crucial and samples are precious, FIB-SEM has become the standard tool for the preparation of TEM lamellae for failure analysis. However, with increasing integration of devices and three dimensional packaging starting to be common now, it is becoming more difficult to rapidly locate the position of a fault with sufficient accuracy to prepare a TEM lamella using a conventional Ga+ FIB in a reasonable time frame. The milling rates achievable in most Ga+ FIBs restrict the maximum lamella dimensions in both width and height to tens of microns. This necessitates time consuming additional investigations to pinpoint the location of a defect before putting the sample into the FIB-SEM. Novel plasma FIB-SEMs employs Xe+ beam which enables milling rates more than 50 times faster than conventional Ga+ FIBs [1]. Therefore lamellae can be prepared rapidly with dimensions of hundreds of microns [2], saving time in isolating defects. While a large lamella opens up the possibility to speed up failure analysis, the lamella quality is nevertheless as critical as ever in order to obtain good quality TEM images. Preparation artifacts need to be understood and eliminated as far as possible. Here we investigate lamellae prepared in TESCAN FERA3 FIB-SEM. All the process steps were done at 30 kV acceleration voltage starting with rough milling at 1 �A beam current and final polish at 1 nA. In order to accurately characterize the lamella thickness and any implantation of ions during the preparation process, we use in-situ EDS analysis in combination with a special software to calculate the lamella thickness (AZtec LayerProbe). This method has been shown to give accurate results for Ga+ prepared lamellae of a range of materials [3]. The LayerProbe software refines a starting model of the sample structure against the EDS spectra to calculate the film thickness and composition of the layers. The starting model comprises the layer sequence in the sample and a substrate material. As the TEM lamella is a free-standing layer the substrate is defined as comprising an element that is not contained in the lamella and only weakly scatters electrons such as Beryllium. The first layer is defined as the material comprising the lamella and the element implanted by the ion beam. The thickness of that layer corresponds to the lamella thickness. Figure 1 shows an electron image with an overlayed X-ray map of the whole lamella. There are three regions of different thickness, varying from around 2 micron down to 0.2 micron (cf. figure 2). The EDS map clearly indicates a Pt rich layer on top of the lamella which was deposited to protect the lamella during the Xe milling process. A map of the Xe-L line indicates that this top layer also contains significant amounts of Xe ions whereas the rest of the lamella shows very little Xe implantation. This difference in the Xe content between the Pt layer and the rest of the lamella can be explained by the fact that to deposit Pt, (CH3)3(CH3C5H4)Pt gas was injected and decomposed by a vertically incident Xe beam. Microsc. Microanal. 21 (Suppl 3), 2015 1406 However, during the preparation of the lamella, the Xe beam was nearly parallel to the lamella and Xe ions were deflected off the sides rather than being implanted. To quantify the amount of Xe in the Pt rich top layer and the Si below two X-ray spectra were taken and the Xe composition was calculated by assuming a homogeneous distribution of Xe throughout the material (figure 3). The results show that the Pt rich layer contains around 3 at% Xe and 79 at% C. This can be compared to previous studies on Ga ion beam induced deposition [4]. The Pt content is similar whether Xe or Ga ions are used, the carbon content is higher when depositing with Xe but the implanted ion content lower. The levels of Xe implanted in the lamella itself are below the detection limit of EDS technique (figure 3). Xe seems not to be as readily implanted as Ga which may be due to differences in the atomic radius of the ions. Further studies are in progress to corroborate these results in the TEM. References: [1] T. Hrnc� et al., 38th ISTFA Conference Proceedings (2012), p. 26. [2] A. Delobbe et al., Microscopy and Microanalysis 20 (2014), p. 298. [3] C. Lang et al., Microelectronics Reliability 54 (2014), p. 1790. [4] C. Lang et al., Microscopy and Microanalysis 18 (2012), p. 620. Figure 1 shows an electron image and X-ray map of the lamella prepared in Xe plasma FIB. The inset shows the Xe map. The large rectangle indicates the area over which the thickness was mapped (figure 2). The small squares indicate area from which spectra were taken for the Xe implantation measurements (figure 3). Figure 2 shows a thickness map of part of the lamella Figure 3 shows two X-ray spectra (1) and (2). as indicated in figure 1. (1) is from the Pt rich top layer and (2) from an area just below containing only Si. The Cu and Al peaks are stray X-rays from the sample holder and grid post.");sQ1[702]=new Array("../7337/1407.pdf","Vacuum Assisted ex situ Lift Out of FIB Prepared Specimens","doi:10.1017/S1431927615007813","1407 doi:10.1017/S1431927615007813 Paper No. 0702 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Vacuum Assisted ex situ Lift Out of FIB Prepared Specimens Lucille A. Giannuzzi1, Dustin Hess2 and Trevor Clark2 1. 2. EXpressLO LLC, Lehigh Acres, FL 33971 USA. The Pennsylvania State University, University Park, PA 16802 USA Conventional ex situ lift out (EXLO) relies on adhesion forces to pick up a specimen with a solid probe tip and place it on a suitable carrier [1-3]. The primary adhesion forces at work are Van der Waals, capillary, and electrostatic forces [3]. New developments in ex situ lift out include a new grid carrier design and methods which allow fast and easy manipulation of specimens outside of the costly FIB [38]. Once manipulated to this new grid design, the specimen may be further processed via FIB, broad beam ion milling, or plasma cleaning. Specimens can be easily positioned into a backside orientation and then FIB polished, reducing curtaining artifacts [5,6]. EXLO is also amenable to very large specimens routinely available from plasma FIB instruments [7,8]. A single ex situ lift out system can support multiple FIB instruments, and when coupled with its speed and ease of use, reinforces its cost effectiveness. Vacuum micropipetting techniques are well known methods used in cell physiology and micro-robotics [9-11]. The shape of these vacuum grippers may be beveled, bent, or angled to optimize and accommodate the target surface of interest for both the pick and the place step [9-11]. In this paper, we combine a vacuum micropipetting module with an ex situ lift out station, making use of both suction vacuum forces and adhesion forces for the Pick&PlaceTM of a FIB milled specimen onto an EXpressLOTM grid. First, a hollow glass tube is pulled to a fine point (~ 1 m) and then beveled to an angle, , with an opening of ~ 2-4 m, such that plus the probe attack angle, = 90o as depicted in the schematic diagram in Figure 1. A completely FIB milled free specimen is prepared as in conventional EXLO. As the shaped hollow probe nears the freed specimen, a valve is opened and vacuum is pulled through the probe (Figure 2a) causing the specimen to stick to the probe tip via suction forces (Figure 2b). The probe tip and specimen is then manipulated to an EXpressLOTM grid. The vacuum valve is closed as the probe tip approaches the grid, using just adhesion forces to hold the specimen. The probe slides through the open EXpressLOTM grid slot (Figure 2c), allowing the specimen to rest on the grid surface in a backside orientation (Figure 2d). The specimen can then be further FIB thinned as before [3-6]. Vacuum pick of specimens, combined with the EXpressLOTM place method, increases specimen manipulation accuracy and positioning reliability to EXpressLOTM grids which are uniquely designed for direct analysis or return to the FIB for further processing. References: [1] L.A. Giannuzzi et al., Mat. Res. Soc. Symp. Proc. Vol. 480 (1997), MRS, 19-27. [2] L.A. Giannuzzi and F.A. Stevie (eds.) Introduction to Focused Ion Beams, (2005) Springer. [3] L.A. Giannuzzi et al., Microsc. Microanal., in review (2015). [4] L.A. Giannuzzi, Microsc. Microanal. 18 (S2) (2012) p. 632. [5] L.A. Giannuzzi, Proc. ISTFA (2012) p. 388. [6] L.A. Giannuzzi, Microsc. Microanal. 19 (S2) (2013) p. 906. [7] L.A. Giannuzzi and N.S. Smith, Microsc. Microanal. 20 (S3) (2014) p. 340. Microsc. Microanal. 21 (Suppl 3), 2015 1408 [8] L. Chan et al., Proc. ISTFA (2014) p. 283. [9] K.T. Brown and D.G. Flaming D.G., (eds) (1986) Advanced Micropipette Techniques for Cell Physiology, Chichester, John Wiley & Sons, Ltd. [10] W. Zesch et al., IEEE Intl. Conf. on Robotics and Automation, (1997) 2, p. 1761. [11] D. Petrovic et al., 23rd Intl. Conf. on Microelectronics, IEEE, (2002) p. 247. Figure 1. A schematic diagram of the hollow glass tube geometry. Figure 2. Light optical micrographs of EXpressLOTM Pick&PlaceTM lift out and manipulation with a vacuum micropipette. (a) The vacuum valve is opened as the hollow probe nears the FIB milled free specimen. (b) The specimen pick is performed. (c) The vacuum valve is closed and the specimen is centered and placed on the EXpressLOTM grid while sliding the probe through the open slot. (d) The specimen rests on the EXpressLOTM grid in a backside orientation.");sQ1[703]=new Array("../7337/1409.pdf","Focused Ne+ Ion Beams for Final Polishing of TEM Lamella Prepared Through Ga-FIB Systems","","1409 doi:10.1017/S1431927615007825 Paper No. 0703 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Focused Ne+ Ion Beams for Final Polishing of TEM Lamella Prepared Through GaFIB Systems Doug Wei1, Chuong Huynh1 and Alexander Ribbe2 1. 2. Ion Microscopy Innovation Center, Carl Zeiss Microscopy, LLC. Peabody, Massachusetts, USA Polymer Science and Engineering, University of Massachusetts, Amherst, Massachusetts, USA TEM specimen preparation using in-situ and ex-situ liftout techniques on Ga+ ion beam based FIB systems has long been realized and become a very important routine in semiconductor industry as well as other material research laboratories [1]. However, more and more data have shown that Ga-FIB processed TEM samples not only contain structural damages due to high energy Ga+ ion bombarding but chemical contamination caused by Ga+ ion implantation as well [2]. To reduce structural damage, recent efforts have been made by using low energy Ga+ ion beam (down to <1 keV) for lamellae final polishing. This low energy Ga+ ion beam polishing has shown significant reduction in the amorphized layer on lamellae surfaces. Other techniques (for instance, low energy Ar+ beam polishing, the NanoMill system) have also been applied after routine TEM sample preparation in order to reduce both structural damage and Ga+ ion contamination. The gas field ion source (GFIS) based helium ion microscopy has been received a lot attention since the first commercially available Orion system was introduced to the market. Initial applications were focused on imaging. Later, more and more work was conducted for materials processing and fabrication. TEM sample polishing using He+ ion beam has also been investigated by several research groups [3, 4]. The limited preliminary results with He+ polishing indicated no obvious benefit in reducing amorphous layers on TEM lamellae. However all the work was done with 30+ keV He+ beams. With the development of the second generation GFIS technology, neon ion beam was made commercially available on the Orion NanoFab systems. Ne+ ion has an atomic mass around 20 much heavier than He+ (4) and way lighter than Ga+ (69.7). The sputtering rate of Ne+ is moderate when milling materials. This is an ion species that is inert, leaving no residues in material after processing (no contamination) and can still introduce significant sputtering to be used for materials processing, but not as damaging as Ga+ ion does. Recent work in nano fabrications with Ne+ beams has shown great value of this newly available focused ion beam in processing materials in sub-micron to nano scales. To investigate the benefit of Ne+ beam in polishing TEM liftout samples, we designed a mechanism to realize TEM sample preparation on an Orion NanoFab multibeam focused ion microscope. With a micromanipulator and a GIS installed onto a NanoFab system that contains two columns, Ga+ column and He+/Ne+ column (He+/Ne+ beams are switchable). A Fin FET DRAM chip was used for TEM sample preparation and then TEM imaging. As in routine liftout TEM sample preparation, we deposited a Pt layer for sample protection and used Ga+ ion beam for major cutting. A lamella was lifted out and welded onto an Omni grid. He+ ion beam was used for imaging purposes in order to monitor the process. After welding the bulk lamella to the Omni grid, initial polishing was still conducted using 30 keV Ga+ beam for speed. After reducing the lamella thickness to about 600 nm, Ne+ beam milling takes over. At beginning of this study, only 25 keV Ne+ beam was used for polishing. Figure 1 shows a thin window (about 50 nm) that was prepared Microsc. Microanal. 21 (Suppl 3), 2015 1410 on the lamella liftout from the Fin FET chip, and Figure 2 shows a bright field TEM image taken from the thin window area. Since we just began this study, our plan is to investigate the effect of Ne+ beam energy and incident angle on the quality of TEM sample including benefit in reducing structural damaging and chemical contamination. Compared to the conventional Ga+ column, the performance of the GFIS column for He+/Ne+ beams is superior, especially at low energy range, meaning that making precise cut and polishing at micron and submicron range using low energy beams become much easier. We will also explore this direction in this study. References: [1] F.A. Stevie, C.B. Vartuli, L.A. Giannuzzi, et al, Surface Interface Analysis, 31 (2001), 345-351. [2] Joachim Mayer, Lucille Giannuzzi, Takeo Kamino, Joseph Michael, MRS Bulletin, 32 (2007), 400407. [3] Daniel Fox, et al., Beilstein J. Nanotechnol. 3 (2012), 579�585. [4] L. Giannuzzi, et al., AVS 58th International Symposium (2011), Nashville. Figure 1. A Fin FET DRAM liftout TEM lamella sample, a small window thinned with Ne+ beam. Figure 2. A bright field TEM image from the thin window area. Figure 1 Figure 1 Figure 2");sQ1[704]=new Array("../7337/1411.pdf","Examining Foil Sidewall Damage During TEM Sample Preparation Using Gallium FIB and Needle Geometries","","1411 doi:10.1017/S1431927615007837 Paper No. 0704 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Examining Foil Sidewall Damage During TEM Sample Preparation Using Gallium FIB and Needle Geometries Michael Presley1, Dan Huber1, Hamish Fraser1 1 Center for the Accelerated Maturation of Materials, The Ohio State University, Columbus OH The focused ion beam (FIB) has proved itself an invaluable tool for both microscopy and sample preparation. FIBs may be used to create rapidly site specific thin foils that are both highly parallel and thin enough for high resolution electron microscopy. However like all methods for creating thin foils, FIB processing results in artifacts that may obscure features of interest. Of particular interest for electron microscopists is the tendency for thin amorphous layers to develop on FIB produced foils [1], which scatter the incoming electrons, reducing clarity. As these amorphous layers run to the edge of the foil, several authors have had great success showing that a combination of cleaning cross sections, decreasing accelerating voltage, and broad beam argon milling may be used to reduce the degree of amorphization and increase image clarity [2,3]. Unfortunately while the extent of amorphous damage may be reflected in the degree of edge damage, the geometry of a foil precludes direct measurement of the actual thickness of the layers. The amorphous layers lie normal to both the beam and thinnest direction, making it impossible to simply tilt the foil to evaluate the damage. In order to circumvent this obstacle, many authors have resorted to encapsulation in order to measure directly the thickness of the amorphous layer. A trench is made at the locations of interest and subsequently coated with an ex-situ protective layer. The resulting sample must be placed back in the FIB to allow a second foil to be pulled perpendicular to the first. While this method is capable of preserving the damage layers it has two major drawbacks, the first being that the protective layer by its nature modifies the damage surface, and the second being that the method itself is time intensive. An alternative method for directly measuring the layer thickness is to use a needle geometry. Needles are already being produced by FIB processing for tomography, atom probe, and micro-mechanical testing. Needles and foils both maintain the same orientation with the beam during processing, with the only difference being the curvature of the path of the cleaning cross section. However while the geometry of thin foils precludes direct examination of the damage layer, a needle may be made sufficiently small to be fully electron transparent. The resulting damage layer should be consistent around the entire needle regardless of orientation. Since the surface is unmodified by subsequent coatings and the through thickness view of the needle approximates what is seen in thin foils, a needle may provide a way to directly link the damage layer properties and thickness to image quality. The needles used in this study were produced using a FEI DualBeamTM FIB at 30 kV using annular ring patterns with an inner diameter of 150 nm. The needles were then imaged using FEI's Image-Corrected TitanTM G2 60-300 S/TEM with ChemiSTEMTM technology. The first needle was extracted from a gold coated single crystal silicon wafer. In Figure 1 the tip of the needle can be clearly seen. The dark region appears fully crystalline in HRTEM while the amorphous layers are both continuous and of equal thickness, confirming the equivalence of the layers around the needle. The amorphous layer measured ~15 nm along the entire length of the needle, falling well within the range of experimental and modeling work reported for Si [4]. In HAADF-STEM imaging, the atomic lattice could be detected up to the region of dark contrast in Figure 1, being also 15 nm in width. By contrast a needle produced from pure aluminum can be seen in Figure 2, where the amorphous layer ranges between 1-3.5 nm in width in both Microsc. Microanal. 21 (Suppl 3), 2015 1412 HRTEM and HAADF-STEM. Whereas the damage layer in silicon is extremely regular and smooth, the amorphous layer in aluminum appears to be more variable. In each case lattice could be resolved in both HRTEM and HAADF-STEM, providing experimental measurements to be compared with quantitative modeling. 20 nm 4 nm 8 nm Figure 1. Silicon Needle produced at 30 kV: needle tip HRTEM, amorphous damage layer HRTEM, amorphous damage layer HAADF-STEM 4 nm 2 nm Figure 2. Aluminum Needle produced at 30 kV: amorphous damage layer HRTEM, amorphous damage layer HAADF-STEM [1] J.P. McCaffrey, M.W. Phaneuf, Ultramicroscopy 87 (2001) 97-104 [2] A. Genc, R.E.A. Williams, D. Huber, H.L. Fraser, Microsc Microanal 13Suppl 2 (2007) [3] Z. Huang, Journal of Microscopy 215, Pt 3 (2004) 219-223 [4] N. Kato, Journal of Electron Microscopy 53 (2004) 451-458");sQ1[705]=new Array("../7337/1413.pdf","Nanoscale Probes in Ultrafast Transmission Electron Microscopy","","1413 doi:10.1017/S1431927615007849 Paper No. 0705 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Nanoscale Probes in Ultrafast Transmission Electron Microscopy Aycan Yurtsever* * Centre Energie, Mat�riaux et T�l�communications, Institut National de la Recherche Scientifique (INRS), 1650 Boulevard Lionel-Boulet, Varennes, QC, J3X 1S2, Canada Capturing ultrafast dynamics at the microscopic level with combined spatial and temporal resolutions requires probes that have sufficient electron counts, coherency and temporal length. Pulsed electrons in an ultrafast transmission electron microscope (UTEM) exhibit such properties, which allows us to capture dynamic events in real space, diffraction domain and energy-loss/gain spectroscopy [1]. In a UTEM, an event is initiated by a femtosecond optical pulse and the ensuing change in the specimen is imaged by pulsed electrons. Images are formed in a stroboscopic fashion [2], which enables superior signal-to-noise levels with combined spatial, temporal and spectral resolutions. Recently, 6 nm spatial, less than 1 ps temporal and 1.4 eV energy resolutions were demonstrated simultaneously in energy-filtered real-space imaging [3]. In this particular contribution, we emphasize the probe forming capability of UTEM. Traditionally, focused electron probes provided one of the most powerful means to access many material properties with rich analytical information. Scanning transmission electron microscopy and spectrum imaging are two examples whose fundamental component is an atomic scale probe. It follows that, the capability to focus pulsed electrons on a nanoscale area, by retaining their femto/picosecond temporal characteristic, would open new avenues in this newly emerging field. In a UTEM, such probes are formed by using the condenser elements of the microscope. Fig. 1a, shows a frame of convergent beam electron diffraction (CBED) pattern from a nanoscale volume of a silicon crystal, acquired stroboscopically. The pattern is captured at negative time that is when electron pulse arrives on the specimen before the photon pulse. The diffraction image contains all the well-known features of continuous-beam CBED with similar signal-to-noise levels. With such high quality frames, it is possible to investigate one, or all, of these features for dynamic behavior. The temporal evolution of two Kikuchi bands are shown in Fig. 1b. In the data, it is evident that one of the bands start to oscillate (shift its position) at time zero, before which it is stationary. The oscillation has a pulse-like profile with a period of 30 ps and an envelope width of 140 ps. We attributed this behavior to a laser induced, propagating strain wave caused by inhomogeneous sample thickness [4]. Another study where ultrafast electron probes are utilized is displayed in Fig. 2. In this example, plasmonic fields of a nanoparticle are mapped by rastering the probe in the area of interest. At every probe position and time delay, an electron energy gain/loss spectrum is acquired. The real-space intensity distribution of an inelastic scattering peak is reconstructed across the particle for every time delay. Here, only time zero and +200 fs are shown. For details of this work see Ref. [5]. Currently, UTEM is the only approach that can simultaneously reach the sub-10 nm and sub-1 ps domains in real space. With further improvements to electron counts and coherencies, perhaps it will be possible to reach the atomic resolution with ultrafast times [6]. Microsc. Microanal. 21 (Suppl 3), 2015 1414 References [1] A.H. Zewail & J.M. Thomas, "4D Electron Microscopy", Imperial College Press, London [2] In stroboscopic acquisition, many pulses (up to a million) are used to form an image, a diffraction pattern or a spectrum at a given time delay. [3] A. Yurtsever, J. S. Baskin and A. H. Zewail, Nano Letters 12, 5027 (2012) [4] A. Yurtsever, S. Schaefer & A.H. Zewail, Nano Letters, 12, 3772 (2012). [5] A. Yurtsever, R.M. van der Veen & A.H. Zewail, Science, 335, 59 (2012). [6] Experiments were conducted in the group of Prof. Ahmed Zewail at the California Institute of Technology. The original work was published in [4] and [5]. FIG. 1 a) A CBED frame taken at negative time, along the [-114] crystallographic direction of silicon. b) The temporal evolution of the two observed Kikuchi bands. The cross-section of each band is evaluated at the rectangular region indicated in a); they are plotted consecutively as a function of the time delay in b). The KB1 band shows strong oscillations (note the edges of the band), in the form of band shift, and exhibits a period of 30 ps; KB3 exhibits no oscillations. The sample was excited with green femtosecond laser pulse at 520 nm wavelength; repetition rate of the laser was 100 KHz. FIG. 2 Ultrafast nanoscale electron probes utilized for plasmonic near-field imaging. The plasmonic response of a triangular particle to the excitation laser is mapped in a way similar to spectrum imaging. Here are shown the strength of the fields at time zero and +200 fs. The full-with-at-halfmaxima of the temporal response is less than 1 ps and the probe diameter is 10 nm.");sQ1[706]=new Array("../7337/1415.pdf","Infrared Pump-Probe Spectroscopy of Plasmons in Graphene and Semiconductors","","1415 doi:10.1017/S1431927615007850 Paper No. 0706 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Infrared Pump-Probe Spectroscopy of Plasmons in Graphene and Semiconductors M. Wagner1, Z. Fei1, A. S. McLeod1, S. J. Maddox2, A. S. Rodin1, W. Bao3, E. G. Iwinski1, Z. Zhao4, M. Goldflam1, M. Liu1, G. Dominguez5,6, M. Thiemens6, M. M. Fogler1, A. H. Castro-Neto7, C. N. Lau4, S. Amarie8, F. Keilmann9, S. R. Bank2, R. D. Averitt1, and D. N. Basov1 University of California San Diego, Department of Physics, La Jolla, California 92093. The University of Texas at Austin, Microelectronics Research Center, Austin, TX 78758. 3. University of Maryland, Materials Research Science and Engineering Center, College Park, Maryland 20742. 4. University of California, Department of Physics and Astronomy, Riverside, California 92521. 5. California State University, Department of Physics, San Marcos, San Marcos, California 92096. 6. University of California San Diego, Department of Chemistry and Biochemistry, La Jolla, California 92093. 7. Graphene Research Centre and Department of Physics, National University of Singapore, 117542, Singapore. 8. Neaspec GmbH, Bunsenstr. 5, 82152 Martinsried, M�nchen, Germany. 9. Ludwig-Maximilians-Universit�t and Center for Nanoscience, 80539 M�nchen, Germany. 2. 1. Unraveling exciting new physics in complex novel materials requires access to both material excitations and their dynamics, thus continuously pushing ultrafast pump-probe spectroscopy to its limits. However, most of the materials whose dynamics are at the center of current attention are also known to be inhomogeneous at the nanoscale. Hence, diffraction-limited optical techniques with their inherent areaaveraging character inhibit access to characteristic time scales of nanoscopic, heterogeneous systems. Circumventing the diffraction limit, scattering scanning near-field optical microscopy (s-SNOM) is a well-established technique that enables broad-band infrared spectroscopy with the nanoscale spatial resolution. In s-SNOM backscattered light from an atomic force microscope (AFM) tip reveals the local dielectric function of a sample [1]. Previous infrared s-SNOM studies were static, utilizing primarily continuous wave (CW) laser sources. Here, we extend s-SNOM by merging nano-spectroscopy with ultrafast pump-probe techniques and exemplify new capabilities with the time-resolved control of the plasmonic response of graphene and the semiconductor InAs. The goal of plasmonics is to utilize electromagnetic energy on a sub-wavelength scale in form of collective surface charge oscillations. Amongst various candidates for plasmonic media graphene stands out due to its most favorable properties: ultimate energy confinement in a monoatomic layer along with easy control over its charge density via electrostatic fields. Recently, the strong coupling between Dirac plasmons and the AFM tip of a near-field microscope has proven ideal to investigate their characteristics, so far utilized for their infrared spectroscopy [2] and real space visualization [3,4]. Another interesting material class for plasmonics are semiconductors due to mature processing technologies and carrier density control via doping. In our experiment we combine mid-infrared s-SNOM and ultrafast 100 fs, near-infrared laser excitation (Fig. 1a)) to study the time-dependent behavior of graphene [5] and InAs plasmons [6] at the nanoscale in infrared spectroscopy. For graphene we find strong pump-induced spectral changes in the infrared plasmonic response (Fig. 1b), top) around the SiO2 substrate phonon resonances at 800 and 1125 cm-1 (Fig. 1b), bottom panel). Modeling reveals that pump-induced heating of carriers up to a temperature of 2100 K is the dominant effect. It results in an increase in Drude weight that s-SNOM is sensitive to. In Microsc. Microanal. 21 (Suppl 3), 2015 1416 terms of its efficiency, optical control of graphene plasmons challenges conventional electrostatic gating (dots in Fig. 1b)), but occurs on a much faster, sub-picosecond time scale as deduced from the pumpprobe time-delay curves in Fig. 1c). Remarkably, modifying graphene's plasmonic behavior requires two orders of magnitude less pulse energy than for metal-based plasmonic structures, which can be easily achieved with a standard fiber laser at telecom wavelengths of 1.5 �m, as in our case. In addition, we discuss data obtained on the prototypical semiconductor InAs where the photoinduced change in the infrared originating from plasmons is an order of magnitude larger at very low control pulse fluence compared to any previous plasmonic modulator scheme. Figure 1d) shows the infrared response for different pump-probe time delays together with a Drude model that nicely fits our data. After photoexcitation a broad, low-frequency peak appears that shifts to lower frequencies as the carriers relax on a time scale of several picoseconds. For both materials, graphene and InAs, the pump-probe spectroscopy results reveal ultrafast optical modulation with high efficiency opening thus the gate to ultrafast plasmonic devices. References: [1] J M Atkin et al., Adv. Phys. 61 (2012), p. 745. [2] Z Fei et al., Nano Lett. 11 (2011), p. 4701. [3] Z Fei et al., Nature 487 (2012), p. 82. [4] J Chen et al., Nature 487 (2012), p. 77. [5] M Wagner et al., Nano Lett. 14 (2014), p. 894. [6] M Wagner et al., Nano Lett. 14 (2014), p. 4529. a) Mid-IR probe NIR pump b) s/s() (%) Integral s/s (%) 20 15 10 5 10 mW 5 mW 2 mW c) 20 d) Layers 1 2 3 12 8 -0.2ps 0ps 0.2ps 0.8ps 1ps 2.1ps 3.1ps 11ps 31ps 15 10 5 s() Gating 4 0 s() Graphene or InAs 0 1 800 1000 -1 Frequency (cm ) 1200 0 -1 0 1 2 3 4 5 Time delay (ps) 600 800 1000 -1 Frequency (cm ) Fig 1. a) Sketch of the ultrafast NIR pump mid-IR probe experiment on exfoliated graphene or InAs. b) Relative pump-induced spectral changes for different NIR pump powers in graphene compared to electrostatic gating. The panel at the bottom shows the absolute signal without pump revealing the hybrid graphene plasmon SiO2 substrate phonon modes at 800 and 1125 cm-1. c) Ultrafast decay of the spectral changes shown in b) for single-, bi- and trilayer graphene together with biexponential decay fits (red curves). d) Photoexcited plasmons in InAs for different pump-probe delays showing ultrafast decay. Experimental data (dots) are plotted together with theory curves from a Drude model (solid lines).");sQ1[707]=new Array("../7337/1417.pdf","Ultrafast Plasmonic Forces Imposed by Fast Electrons on Metal Particles","","1417 doi:10.1017/S1431927615007862 Paper No. 0707 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Ultrafast Plasmonic Forces Imposed by Fast Electrons on Metal Particles M. J. Lagos1,2,3, A. Reyes-Coronado4, A. Konecna5, P. M. Echenique5, J. Aizpurua5, P. E. Batson1,2,3 1 2 Inst. for Advanced Materials, Devices and Nanotechnology, Rutgers University, Piscataway, NJ USA. Department of Physics and Astronomy, Rutgers University, Piscataway New Jersey USA. 3 Department of Materials Science and Engineering, Rutgers University, Piscataway New Jersey USA. 4 Departamento de F�sica, Facultad de Ciencias, Universidad Aut�noma Nacional de Mexico, Mexico. 5 Donostia International Physics Center, and Materials Physics, CSIC-UPV/EHU, San Sebastian Spain. Recently, several groups reported movement of nano-objects apparently driven by electron beams [1-3]; thus pointing out the possibility of using electron-beams for effective manipulation of nanostructures. Among the different types of behavior reported -- such as pushing and pulling of single nanoparticles (NP's), rotation, and NP coalescence -- the observation of both attractive and repulsive forces driving movement of gold NP's has been the most difficult to understand. Attractive forces between an electron and a NP seem entirely reasonable, simply by imagining a positive image charge within the NP, induced by the negative charge on the swift electron [4]. But understanding of a repulsive force still remains elusive. In this work, we have studied the temporal evolution of forces acting on a metal sphere induced by a swift electron; in particular aiming to explore the dynamics of plasmon excitations and their role in the generation of ultra-fast Lorentz forces. We modeled theoretically the time-varying electromagnetic forces acting on a spherical 1nm-radius Au NP, imposed by a 120KV- relativistic electron travelling in a non-intersecting geometry (Fig.1). We found fascinating results which bring more elements to our understanding of the attractive and repulsive forces in the context of plasmonic response: (i) surface plasmons at optical frequencies (~ 2.5 eV) produce oscillatory forces at femtosecond times which are likely mediated by the emission of photons into the electromagnetic fields during the plasmon decay. At frequencies around 25 eV (deep ultraviolet range), we noticed a second plasmonic instability that apparently originates with the excitation of 5d electrons in gold [5]. Interestingly, these two plasmonic modes occur at quite distinct times, following the passage of the swift electron. We think that these two modes are weakly coupled, leading to formation of a beating pattern in the response forces, in spite of their large energy mismatch. These plasmonic features produce very weak oscillatory forces at moderate distances, so their contribution to the total momentum transfer to the NP is quite small. (ii) At higher energies, we notice a strongly confined surface wake pattern on the sphere, having a wavelength that is significantly shorter than the particle diameter. This wake pattern lags the passing electron and is composed of positively charged regions (holes) between large regions of negative charge. It induces, on average, a repulsive force in the spherical NP (Fig. 2). (iii) During the close approach at atto-second times, external electric and magnetic fields imposed by the swift electron interact with induced charges and currents within the sphere to produce strong attractive and repulsive forces. These forces compete one against the other, resulting in a net force which is primarily a dielectric attraction during the approach of the swift electron and a diamagnetic repulsion as the electron leaves. These attosecond forces contribute most of the total momentum transfer from the electron to the NP. Our results provide progress in understanding the physical origin of the repulsion behavior of NP's driven by swift electrons, pointing out the possibility of wide exploration of ultrafast phenomena in nano-sized systems. Microsc. Microanal. 21 (Suppl 3), 2015 1418 References: [1] P. E. Batson, et al, Nano Letters 11 (2011) 3388. [2] O. Cretu, et al, Carbon 50 (2012) 259. [3] H. Zheng, et al, Nano Letters 12 (2012) 5644. [4] A. Reyes-Coronado, et al, Physical Review B 82 (2010) 235429. [5] I. G. Guturbay, et al, Computational Materials Science 22 (2001) 123. [6] Acknowledgements: This research was supported by DOE project #DE-SC0005132. Fig. 1. Modeling geometry for the electromagnetic interaction between an electron and metallic nanosphere in an aloof geometry: the electron travels in the +z direction, with speed v along the trajectory r = (R+b, 0, vt), where R is the radius of the sphere and b is the impact parameter. Fig. 2. Lorentz forces resulting from the interaction between an induced surface wake and a swift electron (red dot) during attosecond times. The broad accumulation of negative charge induces repulsive electric forces which lead to weakening of the attractive electric contributions. Note the forces (and the charge density fluctuations) display a cone-like pattern spreading over the surface in the direction which is transverse to the electron trajectory.");sQ1[708]=new Array("../7337/1419.pdf","Photonic Processes Visualized with Electrons in Photoemission Electron Microscopy (PEEM)","","1419 doi:10.1017/S1431927615007874 Paper No. 0708 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Photonic Processes Visualized with Electrons in Photoemission Electron Microscopy (PEEM) Rolf K�nenkamp1, Robert C. Word1 and Joseph P. S. Fitzgerald1 1 Physics Department, Portland State University, Portland, OR 97201, USA Photoemission electron microscopy (PEEM) combines in a unique way light and electron optics: As photo-emitted electrons are used in the imaging process, light-optical phenomena can be imaged with a resolution significantly below the light-optical diffraction limit. In addition the light-optical excitation offers non-destructive operation conditions, an inherent sensitivity to the surface-near region and extremely short observation times [1]. We use aberration-corrected PEEM to directly visualize with electrons the propagation of light at solid surfaces [2]. Aberration-correction is achieved with an electrostatic mirror that allows simultaneous correction of chromatic and spherical aberrations [3]. The typical spatial resolution in this work is 5 to 40nm. From the PEEM micrographs we can quantitatively determine local optical material properties, phase shifts as well as coupling and scattering probabilities etc. In many cases a detailed quantitative understanding of dynamic optical processes can be achieved. PEEM imaging also allows the direct observation of optical processes in devices such as optical filters, light-routers, plasmonic routers, sensors, etc. [4, 5]. Figure 1 shows the propagation of light waves in a thin, transparent indium tin oxide (ITO) film. Coherent plane wave light from a 100-fs pulsed laser is coupled into the ITO film at a semi-circular arrangement of air holes. The shown image represents an interference pattern resulting from a diffracted light wave in the film and an un-diffracted plane reference wave [6]. As the two waves originate from a pulsed source, the number of interference fringes is finite and the extent and duration of the image are limited: Overall about 100 interference fringes are generated corresponding to a time interval of less than a picosecond. The interaction that gives rise to this electron micrograph is a non-linear 2-photon photoemission process. Careful analysis of this process allows a reconstruction of the image from fundamental optics principles. A two-dimensional image calculation based on Kirchhoff integration is shown in Figure 1c. The analytical image calculation reproduces the experimental features with many details [7]. Beatings between different modes as well as non-linear interactions are seen. Combining the information provided by experiment and simulation allows an accurate determination of film thickness, refractive index, first and secondorder absorption coefficients etc. The high spatial resolution of the electron micrographs also lends itself to a detailed near-field analysis. For example, the polarization dependence of the diffraction process can quantitatively be analyzed in the same microscopy approach [8]. Furthermore, the analysis is not restricted to guided and or confined light modes. Figure 2 shows a photoemission micrograph obtained at a micro-structured silicon surface. A detailed analysis shows that in this case the interferences are generated by diffraction at surface features of the silicon. However, only vacuum modes contribute to the photoelectron emission. We therefore conclude that grazing optical vacuum modes near solid surfaces can produce photoelectron emission and can thus also be used in PEEM for imaging surface properties. Finally it is shown that the same theoretical image analysis can be applied to PEEM micrographs of surface plasmon polariton modes at metal surfaces [9]. Microsc. Microanal. 21 (Suppl 3), 2015 1420 We conclude that multiphoton-PEEM provides an excellent means to study optical and photonic processes in waveguides and at surfaces. In 2- and 3-photon processes all of the visible range becomes available for photoelectron emission. Spatial resolution significantly below the optical wavelength can be achieved and near-field studies with high time-resolution are feasible. [1] Ernst Bauer in "Surface Microscopy with Low Energy Electrons", Springer (2014) [2] Robert C. Word et al., Ultramicroscopy 110, 741 (2010). [3] Joseph P. S. Fitzgerald et al., Ultramicroscopy 115, 35 (2012) [4] Rolf K�nenkamp et al., Appl. Phys. Lett. 101, 141114 (2012) [5] Robert C. Word et al., Appl. Phys. Lett. 105, 111114 (2014) [6] Livio I. Chelaru and F.-J. Meyer zu Heringdorf, Appl. Phys. Lett. 89, 241908 (2006) [7] J.P.S. Fitzgerald et al., Phys. Rev. B 89, 195129 (2014) [8] Theodore Stenmark et al., this conference [9] Niemma M. Buckanie et al., Ultramicroscopy 130, 49 (2013) (a) (b) (c) Figure 1: (a) Experimental PEEM image of a guided optical wave in a thin ITO film. The wave is excited in the 15 holes arranged in a semicircle. A gold nanowire is seen in the lower half of the image. (b) Close-up from the experimental results shown in part (a). (c) Calculated image of the optical system shown in (a). The gold nanowire is not considered in this calculation. Figure 2: Experimental PEEM image of diffracted vacuum modes generated at a micro-structured silicon surface.");sQ1[709]=new Array("../7337/1421.pdf","Global Analysis Peak Fitting Applied to EELS Images","","1421 doi:10.1017/S1431927615007886 Paper No. 0709 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Global Analysis Peak Fitting Applied to EELS Images M.H. Van Benthem, P.G. Kotula, M. Marinella, W. Mook, and K.L. Jungjohann Sandia National Laboratories, PO Box 5800, MS 0886, Albuquerque, NM 87185-0886, USA Spectroscopic transitions often result in features that lend themselves nicely to peak fitting techniques.[1-3] However, just as often, neighboring peaks or background signals obstruct those peaks. Hyperspectral imaging, as opposed to single spectrum collection, may provide some relief to this problem by isolating sources of signal within the image. Employing the common practice of summing or integrating the image[3], or areas of the image, may sacrifice the advantages gained from imaging. Analyzing all of the pixels within the image simultaneously, or globally, permits one to retain the benefits imaging provides throughout the process of analysis and interpretation. In this presentation, we will demonstrate how to perform global peak fitting on STEM-EELS data in an attempt to elucidate the behavior of a Ta/TaOx/Pt stack sputter deposited on a polished W surface, Fig. 1, when challenged with low voltage biasing. We imaged the sample before, during and after the challenge. We employed a FEI Company Titan G2 80-200 TEM/STEM operated at 200 kV and equipped with a high-brightness Schottky field emission electron source, a spherical aberration corrector on the probe forming optics (AC-STEM) and a Gatan Quantum 963 EELS. The STEM acquired full EELS spectral images (SI) of the low-loss energy region (-20 eV to 180 eV), including the zero-loss peak, plasmons and Ta O2,3 edge, with a 50 msec dwell time. EELS-SI data preprocessing involved aligning the energy baseline to a common axis, correcting for small instabilities in the primary beam energy and drift/instability of power supplies. We aligned the peaks by first restricting the data set to -3 eV to +3 eV (the region of the zero loss peaks). Then we upsampled data by fifty times to an effective channel size of 2 meV, using the Piecewise Cubic Hermite Interpolating Polynomial (PCHIP) function in Matlab[4]. Next, we computed the cross-correlation coefficients across the zero loss peaks for each pixel spectrum relative to the first row, first column pixel of the before-challenge sample. Choosing the index of the maximum cross-correlation coefficient as the zero loss value, we shifted all the spectral to a common energy-loss baseline (Fig 2(l)). The concept of global analysis[5] for these data seeks to fit a set of peaks or transitions using nonlinear optimization to determine parameters such as center and width, then using linear least squares to find the linear amplitude terms. Thus, the data in n-pixel by m-spectral channel matrix D follows the model = where A is the n by p-peak matrix of amplitudes or abundances and S is the m by p matrix of unit height peaks. Here the model estimates the columns of S in a nonlinear sense and linear least square provides the amplitudes as = ( )- We employed global analysis on the energy-aligned data fitting five Voigt peaks, two asymmetric Voigt peaks, two edge transitions, and an offset or bias. One of the fit peaks and its spatial extent for the three Microsc. Microanal. 21 (Suppl 3), 2015 1422 challenge states is displayed in Fig 2(r). References: [1] St�hr, J. "NEXAFS Spectroscopy." Berlin: Springer, (1992). [2] Reed, B.W., et al. Phys. Rev. B 60(8), (1999). 5641-5652 [3] Varela, M., et al. Phys. Rev. B 79(8), (2009). 085117. [4] MATLAB. Natick, MA: The MathWorks, Inc. (ver 2014a). [5] Beechem, J.M. Global Analysis of Biochemical and Biophysical Data. In Numerical Computer Methods, L. Brand and M. L. Johnson (Eds.), pp. 37-54. San Diego: Academic Press, (1992). [6] Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the US Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000. Figure 1. Ta/TaOx/Pt stack on W surface prepared for STEM-EELS imaging. Figure 2. Left: Effects of alignment procedure on EELS-SI zero loss peaks for a small subset of data. Right: Global analysis Voigt peak fit representing notional Ta energy loss from Ta/TaOx/Pt stack.");sQ1[710]=new Array("../7337/1423.pdf","Revisiting Noise Scaling for Multivariate Statistical Analysis","doi:10.1017/S1431927615007898","1423 doi:10.1017/S1431927615007898 Paper No. 0710 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Revisiting Noise Scaling for Multivariate Statistical Analysis P.G. Kotula and M.H. Van Benthem Sandia National Laboratories, PO Box 5800, MS 0886, Albuquerque, NM 87185-0886, USA It has been demonstrated that scaling of spectral image data, such as those containing X-ray spectra or other count data, for non-uniform noise is an essential part of multivariate statistical analysis. [1-3] Multivariate statistical analysis algorithms, such principal components analysis (PCA), rank the respective components by variance. Variance can arise due to chemical signals (e.g., peaks) and also due to noise. The goal of the scaling is to weight the data such that variance due to noise in high signal regions of the spectral image is down weighted. The method described by Keenan and Kotula for optimal scaling [2,3] scales the spectral image data in D using the column and row space means (1) where G is a diagonal matrix whose diagonal elements are the inverse of the square roots of the row means (mean image) and H is a diagonal matrix whose diagonal elements are the inverse of the square roots of the column means (mean spectrum). While this has been an extremely powerful normalization approach, for very sparse data sets this might not be optimal. Today with the advent of large-area silicon-drift detectors (SDD) in analytical electron microscopes (AEM), there is a temptation to increase the number of pixels in a spectral image beyond what the nominal signal levels can support. Alternatively specimen damage can limit acquisition time. This, in principle, poses no significant problems as we can post process the data via geometrical binning for example, to improve the signal levels. Nanoparticles on a thin support are likely to generate few X-ray counts under imaging conditions which do not damage the specimen quickly. For the subsequent example, an FEI Company Titan G2 80-200, operated at 200kV and equipped with a high-brightness Schottky emitter, spherical aberration corrector on the probe forming optics, and 4 SDDs with a combined solid angle of collection of 0.7sr. The probe had a measured diameter of ~0.15nm with 200pA and a convergence angle of 18 mrad. Higher currents could have been used but the specimen damaged too quickly. A spectral image was acquired with 0.5nm pixels, 200 pixels by 200 pixels (100 nm by 100 nm field of view) by 2048 channels (10 eV/channel) in 1900 seconds. The specimen consisted of yttria particles decorated with smaller Pd particles on a thin carbon support. Figure 1 shows the results of a multivariate statistical analysis (multivariate curve resolution (MCR) [1]) of the data as acquired. This results in one reasonable looking factor with a number of non-physical factors seemingly comprised of random noise. The spectral image data were 99.8% sparse containing a total of 150k counts in almost 80 million data elements. Given this overall sparsity and the fact that the sample is one central yttria particle decorated with Pd particles, it is also very spatially sparse. Figure 2 shows the results of an analysis of the same spectral image where we have binned spatially 4x4, spectrally by a factor of two and modified the normalization method in Eq. 1 so that we are not performing spatial weighting. The equation thus becomes (2) indicating that we have weighted only in the spectral domain. The same MCR calculation as before now results in only three factors all with physical significance. As sparse data may be more the rule than Microsc. Microanal. 21 (Suppl 3), 2015 1424 exception in the AEM, this combined approach of binning of the data and a modified noise normalization approach may make sense for many analyses. [1] Kotula, P.G., Keenan, M.R. & Michael, J.R. (2003). Microsc Microanal 9, 1�17. [2] M.R. Keenan & P.G. Kotula, (2004) Appl. Surf. Sci., 231/232 240-244. [3] Keenan, M.R. & Kotula, P.G. (2004) Surf Int Anal 36, 203�212. [4] Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the US Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000. The authors would like to thank Prof. Barry Carter at the University of Connecticut for provision of the specimen. Figure 1. MCR of the raw spectral image showing the first six factors of which only one is physically reasonable. Figure 2. MCR analysis of the binned spectral image normalized for noise in only the spectral domain showing three physically reasonable factors.");sQ1[711]=new Array("../7337/1425.pdf","When will low-contrast features be visible in a STEM X-ray spectrum image?","","1425 doi:10.1017/S1431927615007904 Paper No. 0711 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 When will low-contrast features be visible in a STEM X-ray spectrum image? Chad M. Parish1 1. Materials Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN USA Under what circumstances will a low-contrast feature, such as a nanoparticle embedded in a foil prepared for transmission electron microscopy (TEM), be visible in an X-ray mapping experiment? This broad question does not have a general answer, but here I present a simplified model of oxide nanoclusters (NCs) embedded in a metallic matrix. This model allows an a priori prediction of what features will be visible under given experimental conditions (specimen structure, microscope X-ray collection efficiency, beam current and pixel dwell time, etc.). The ability to estimate what combination of X-ray collection efficiency (as solid angle ), probe current (ib), and pixel dwell time (), is necessary for given features to become visible in an X-ray map. First, NCs in the matrix are simulated (Fig. 1a) and their relative densities of Fe and Ti projected down the beam direction (Z-axis)(Figure 1b-c). Application of the fundamental X-ray detection equation [1] and parameters [2,3] allows calculation of the anticipated Fe and Ti X-ray maps (Fig. 2a-b), which can incorporate finite spot size and beam broadening [4]; no bremsstrahlung contribution is (yet) present in the calculation. DTSA-II [5] is used to calculate ideal Fe85Cr14W1 and Y2Ti2O7 spectra. By scaling the spectra relative to the specific Ti and Fe counts calculated from the X-ray detection equation, each individual pixel's spectrum can be calculated (Fig. 3). Because the time-consuming Monte Carlo X-ray simulation is only performed twice (for the idealized matrix and NC spectra), it is very computationally efficient to populate the entire spectrum image with individualize pixel spectra. Poisson noise is then used to produce the noisy point spectra; Fig. 3d). By generating full spectrum images (SIs), multivariate statistical analysis (MVSA) methods can be applied for datamining [6]. Comparisons to X-ray spectrum images taken on a Philips CM200, Hitachi HF3300, and FEI Titan G2 with ChemiSTEM are favorable when the input parameters to the model are well-matched to the experimental case. Fig. 5 compares Titan G2 data from simulation and experiment; the 4-detector SuperX system is well-suited to mapping small NCs embedded in a metallic matrix. Varying parameters such as SI size, probe current, pixel pitch, and pixel dwell time, all modify the visibility of the particles in a map, and are amenable to rapid screening by these calculations [7, 8]. [1] NJ Zaluzec, Micros. Microan. V15(Supp. 2) (2009) P. 458. [2] MR Talukder et a., Int. J. Mass Spect. V269 (2008) P. 118-130; V309 (2012) P. 118. [3] DB Williams & CB Carter," Transmission Electron Microscopy (2 ed.)," (Springer, Belin, 2009). [4] JR Michael et al., J. Microsc., V160 (1990) P. 41. [5] NWM Ritchie, Mirosc. Microan., V15(5) (2009) P. 454. [6] PG Kotula et al., Mirosc. Microan., V9(1) (2003) P. 1. [7] CM Parish, Mirosc. Microan., accepted. [8] This work was supported by the U.S. Department of Energy, Office of Science, Basic Energy Sciences, Materials Sciences and Engineering Division. A portion of the Microscopy conducted as part of a user proposal at ORNL's Center for Nanophase Materials Sciences, which is an Office of Science User Facility. I acknowledge the use of the Analytical Instrumentation Facility (AIF) at North Carolina State University, which is supported by the State of North Carolina and the National Science Foundation. Microsc. Microanal. 21 (Suppl 3), 2015 1426 Fig. 1: (a) Simulated structure. Red are oxides, matrix invisible. (b-c) Z-projected thicknesses assuming Fe85Cr14W1 and Y2Ti2O7 stoichiometry, respectively. Fig. 2: Anticipated noisefree characteristic X-rays for (a) Fe (b) Ti. Fig. 3: (a) Details of simulated count maps. (b) Idealized DTSAII calculated spectra for matrix and precipitates. (c) Summed noise-free point spectrum. (d) With Poisson noise added. Fig. 4: Top row experiment (NCSU Titan G2 with ChemiSTEM) and bottom row simulation, showing similar signals and visibilities. MVSA comp#2 denotes MVSA score image for precipiates. Inset value (i.e., 256�256) is binned pixel size for MVSA of original 512�512 map.");sQ1[712]=new Array("../7337/1427.pdf","Name that Atom in 60 Seconds or Less: Energy Dispersive X-Ray Spectroscopy of Individual Heteroatoms in Low Dimensional Materials.","","1427 doi:10.1017/S1431927615007916 Paper No. 0712 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Name that Atom in 60 Seconds or Less: Energy Dispersive X-Ray Spectroscopy of Individual Heteroatoms in Low Dimensional Materials. R. M. Stroud1, T. C. Lovejoy2, M. Falke3, N. D. Bassim1, G.J. Corbin2, N. Dellby2 P. Hrncirik2, A. Kaeppel3, M. Rohde3, and O. Krivanek2. 1. 2. Materials Science and Technology Division, Naval Research Lab., Washington, DC 20375 USA. Nion Co., 11511 NE 118th St. Kirkland, WA 98034, USA. 3. Bruker Nano GmbH, Schwarzschildstr. 12, 12489 Berlin, Germany. The feasibility of identification of individual Si atoms on single-layer graphene on the basis of energy dispersive x-ray spectroscopy (EDXS) in a scanning transmission electron microscope (STEM) has been demonstrated [1]. However, the employed measurement conditions, i.e., manual tracking of single Si atoms on single carbon sheets over collections time of 4 to 6 minutes greatly limited the applicability of the method for more wide-spread use. Under such conditions, there is no clear advantage of EDXS over the use of quantitative annular dark field image analysis, or electron energy loss spectroscopy (EELS). To be truly practical for samples such as supported catalyst nanoparticles or other low dimensional materials, single atom EDXS measurement conditions must allow for robust atom identification in seconds, in samples with variable thickness that contain combinations of multiple, unknown heteroatom species. The collection efficiency of EDXS measurements, which is critical to the practicality of single-atom sensitivity measurements, is a function of multiple instrumental parameters, including: the geometrical solid angle of the detector, the detector window material, the quantum efficiency of the detection electronics, the sample holder material and geometry, and the stability of the host microscope. For our experiments, we used a Bruker XFlash6:100, 100mm2 elliptical, windowless silicon drift detector, mounted in a Nion UltraSTEM 200 UHV cold field emission STEM at the Naval Research Laboratory. The calculated solid angle of the detector is ~ 0.7sr, and measurements of NiO standards yield similar values. Additional design considerations of the detector-microscope combination are discussed in more detail elsewhere [2]. We operated the UltraSTEM at 60 keV with ~120 pm probe and a beam current of 50 to 150 pA. The sample investigated was a nanodiamond-amorphous carbon phase mixture extracted from the Murchison meteorite by acid dissolution, and drop cast onto a lacey carbon support film on Cu TEM grid [3]. Multiple methods for atom identification were tested. Regions of interest were first identified as amorphous carbon or nanodiamond through imaging and C K-edge EELS. Heteroatoms of interest associated with either phase were then selected in annular dark field STEM images. These atoms were not robustly incorporated in the carbon lattice, but instead able to diffuse quickly over the surfaces of the sample. Thus, the EDXS signal was collected as either individual spectra during the manual tracking the heteroatom in a live scanning image window (~ 0.25 nm2) in Digital Micrograph, or as long-dwell time single, or short-dwell time, multipass spectrum images with the Bruker Esprit software. In the case of the manual tracking, the live image window was recorded as an image series. For spectrum images, annular dark field images were recorded before and after the spectrum image for single pass acquisition, and with each frame, for multiframe acquisition. In all cases, the acquisitions were terminated when the heteroatom moved out of the imaging area. Microsc. Microanal. 21 (Suppl 3), 2015 1428 1418 Figure 1 illustrates the identification of a Si atom on a few-layer region of amorphous carbon by the manual tracking method. The total acquisition livetime for this spectrum was 55.5 sec at a probe current of 120 pA. The net count yields were 1856 in C and 48 in Si. A second tracking experiment on the same atom yielded similar results, i.e., of order 50 counts in Si in 55 seconds. We estimate that minimum time for reasonable confidence in identifying the heteroatom under these conditions is ~ 10 to 15 sec, limited in part by the amount of overscan area needed to ensure that the atom remained in the field of view. As the atom occupied ~ 1/10 the scan area, an immobile atom should be identifiable in ~ 1 sec. Additional experiments revealed heteroatoms of Ca, S, and O, all with total collection times under 60 sec, and ~ 5 net counts/sec of on-atom beam exposure time. As shown in Fig. 2, it is also possible to identify individual heteroatoms in hyperspectral images. The motion of the atom, in this case S, was recorded as x-ray signal originating from different spatial locations during the acquisition. The 122 � 96 px map was obtained in total of 1.72 sec by integrating five frames with a 32 �s/px dwell time. Longer dwell-time single pass spectrum images also enabled identification of heteroatoms, but the interpretation was more difficult, because the atoms moved faster than the beam, which resulted in streaked images. References: [1] T.C. Lovejoy et al., Appl. Phys. Lett. 100 154101 (2012). [2] T. C. Lovejoy et al., Proceedings of M&M 2015 (2015), this volume. [3] R. M. Stroud et al., Astrophys. J. Lett. 738 L27 (2011). [4] The authors thank C. M. O'D. Alexander for providing the nanodiamond residue. RMS gratefully acknowledges financial support from the NASA LARS program and the NRL CPP program. Figure 1. Energy dispersive x-ray spectrum of a single Si atom on amorphous carbon. The green box on the dark field image indicates the area of the tracking window. Figure 2. Energy dispersive x-ray spectrum imaging of a S atom on fewlayer amorphous carbon. The inset dark field images show the area of interest before and after the spectral acquisition. The green box indicates the ~0.5 nm � 0.4 nm mapped region. The C and S maps are overlayed in blue and red, respectively.");sQ1[713]=new Array("../7337/1429.pdf","Single Atom Detection Through HAADF-STEM and EELS/EDX Characterization of Fluorophore Ru(bpy)32+ for Optical DNA-Chip Applications","","1429 doi:10.1017/S1431927615007928 Paper No. 0713 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Single Atom Detection Through HAADF-STEM and EELS/EDX Characterization of Fluorophore Ru(bpy)32+ for Optical DNA-Chip Applications G. Nicotra1, E.L. Sciuto1, M.F. Santangelo2, G. Villaggio3, F. Sinatra4, C. Bongiorno1, S. Libertino1 1 2 CNR-IMM, Strada VIII, 5, 95121 Catania, Italy. Distretto Tecnologico Sicilia Micro e Nano Sistemi, Catania, Italy 3 Dipartimento G. F. Ingrassia, Universit� degli Studi di Catania, Catania, Italy 4 Dipartimento di Scienze Biomediche, Universit� degli Studi di Catania, Catania, Italy Optical DNA-chip is widely used to study genome, gene expression and genetic diseases [1] We can divide it into three basic components: the sensing element or probe specific for target gene (single strand DNA); the labeling system, provided by conventional fluorophore CY5 (indodicarbocyanine) [2]; the optical detector, i.e. imagers or scanners. The labeling organic fluorophores exhibiting absorption/emission peaks very close, ~ 20 nm apart. Unfortunately, their prolonged exposure to laser light causes photo bleaching [3]. Moreover, their short lifetime, below 3 ns [4], could be an obstacle for the design of a miniaturized device, since a complex and fast (in the GHz range) electronic management system may be needed. In order to achieve the goal of a miniaturized, cheap and "simple" optical biosensor, we characterized an organometallic fluorophore, the tris(2,2'bipyridyl)ruthenium(II) (Ru(bpy)32+). Our studies revealed that Ru(bpy)32+ emission properties (peak and lifetime) strongly depend from the chemical and physical environment. In particular we found an additional emission peak at 590 nm which dominates in the dried form regardless of the surface used to deposit the sample, while the lifetime value increases of a factor ten to ~30 ns. In order to understand the relation between the fluorophore form and its optical properties we firstly performed a full structural characterization through Low-Energy Scanning Transmission Electron Microscopy (STEM), operated at 60keV, by using a Cs corrected TEM JEOL ARM 200F. STEM analysis, of pure drop casted water solution containing Ru(bpy)32+ Thanks the very high sensitiveness of HAADF no negative staining, such as heavy uranyl acetate, was necessary. STEM observation in conjunction with chemical mapping performed by Electron Energy Loss Spectroscopy (EELS) and Energy Dispersive X-Ray Analysis (EDX), allowed us to identify fluorophore clusters of few nanometers, Figure 1, and to find a possible explanation for optical properties of the dried form. We than moved at the 200keV of beam energy, and for the first time we observed the single atoms of Ru, which are the direct evidence of the actual presence and the amount of molecules and their distribution on the substrate, Figure 2,. Motion of such single atoms has been also observed and studied under the electron beam radiance. For this study, a state-of-the-art aberration-corrected microscope installation at Beyond-Nano sub-�ngstom Lab, in Catania, Sicily, Italy, has been used. This consists of a probe corrected STEM microscope equipped with a C-FEG and a fully loaded GIF Quantum ER as EELS spectrometer. This particular installation is capable to deliver a probe size of 0.68 � at 200kV, and 1.1 � at 60kV. Chemical mapping were extracted from the simultaneous acquired EELS and EDX spectra, by using the Spectrum Imaging mode. Microsc. Microanal. 21 (Suppl 3), 2015 1430 References [1] R. W. Ye, T. Wang, L. Bedzyk and K. M. Croker, "Applications of DNA microarrays in microbial systems", J. Microbiol. Methods 47-257, 2001.. [2] M. Schena, D. Shalon, R. W. Davis and P. O. Brown, "Quantitative monitoring of gene expression patterns with a complementary DNA microarray", Science 270, 467-470, 1995. [3] J. R. Lakowicz, "Radiative Decay Engineering: Biophysical and BiomedicalApplications", REVIEW Analytical Biochemistry Vol. 298, pp. 1�24, 2001. [4] M. F. Santangelo, R. Pagano, S. Lombardo, E. L. Sciuto, D. Sanfilippo, G. Fallica, F. Sinatra, A. C. Busacca, S. Libertino, "SiPM as novel Optical Biosensor: transduction and applications", submitted for publication IEEEXplore, 2014. [5] This work was performed at Beyondnano CNR-IMM, which is supported by the Italian Ministry of Education and Research (MIUR) under project Beyond-Nano (PON a3_00363). Figure 1. a) Z-Contrast HAADF STEM image with no negative contrast. b) EELS map of N K1 edge; c) EDX chemical map of Ru L edge; Chemical maps shows that Nitrogen and Ruthenium are localized on the region occupied by the bright contrast. d) HAADF showing the presence fluorophore clusters of few nanometers Figure 2. a) Plain carbon supporting film gold grid and corresponding EDX spectrum b); c) carbon supporting film gold grid containing the fluorophore Ru(bpy)32+, bright spot are single Ru atoms, and corresponding EDX spectrum d)");sQ1[714]=new Array("../7337/1431.pdf","SFM Assisted In-Situ by ToF-SIMS: Accessing Chemical Information in True Three Dimensions","","1431 doi:10.1017/S143192761500793X Paper No. 0714 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 SFM Assisted In-Situ by ToF-SIMS: Accessing Chemical Information in True Three Dimensions Rapha�lle Dianoux1, Adi Scheidemann1 Ewald Niehuis2, Rudolf M�llers2, Felix Kollmer2, Henrik Arlinghaus2 Hans-Josef Hug3, Laetitia Bernard3, Sasa Vranjkovic3 1 2 NanoScan AG, Duebendorf, Switerzland ION-TOF GmbH, Muenster, Germany 3 Empa, Duebendorf, Switzerland Scanning Force Microscopy (SFM) is a well-established tool for surface analysis in research and industry around the world nowadays. Originally developed to image atoms, the range of applications has since been widely deployed to make it not only an imaging technique but also a sensor of local forces of the surface and a manipulator of atoms and molecules. Very sharp, functionalized tips mounted on micro-levers constitute extremely sensitive sensors to the local force field emanating from the surface and thus make the SFM a general tool for the analysis of physical properties at the nanoscale. Operated in dynamic, multi-frequency mode, the SFM can discern between concurring forces. For instance, in Magnetic Force Microscopy (MFM), magnetic properties can be recorded simultaneously to topographical data, with a lateral resolution down to 10 nm [1]. In Kelvin probe Force Microscopy (KPFM), the contact potential difference between tip and sample is compensated while simultaneously measuring an artifact-free topography. Vacuum operation enhances sensitivity of dynamic SFM by a factor of ca. 20, as the quality factor of the cantilever is increased by a factor of 100 or more. SFM is a non-destructive method yielding calibrated forces in the pico to the nano-Newton range, with a lateral resolution ranging from atomic scale to the nanometer and with a sub-nanometer vertical range. However, except in rare cases [2], it cannot identify unequivocally the chemical origin of the surface, nor can it give insight into the depth of the sample. Time-of-Flight Secondary Ion Mass Spectrometry (ToF-SIMS) is the method of choice to remedy these limitations. ToF-SIMS provides chemical information at the elemental as well as the molecular level with a sensitivity up to the ppb range and a lateral resolution down to 20 nm. For depth profiling applications, a second, dedicated sputtering beam is used to carry out sample erosion. In this so-called dual beam mode, a depth resolution below 1 nm can be achieved [3]. We have combined the techniques of SFM and ToF-SIMS in one UHV analysis chamber. This instrument was conjointly developed within the framework of a FP7 research project [4]. One of the key components of the new instrument is a piezo driven XYZRT-stage which moves the sample between the TOF-SIMS and the SFM analysis site with sub-�m precision and high speed. This allows analyzing exactly the same surface area with both techniques. The SFM unit is cantilever-based and equipped with a 4-quadrant Optical Beam Deflection System (OBDS). The flexure stage is a fully-linearized scanner with 80x80x10 �m3 range driven by a 20-bit digital control. Scanner and piezo stage movements are correlated to allow for the measurement of large-area images surpassing the scan range, or of long profiles across sputter craters. Crater depth is of crucial importance to localize vertically successive ToF-SIMS images. The SFM controller is further Microsc. Microanal. 21 (Suppl 3), 2015 1432 equipped with two Phase-Lock-Loops (PLL) which allow for dual-frequency demodulation in dynamic mode. It also features a KPFM module. The TOF-SIMS analysis is performed using a new bismuth liquid metal cluster ion gun that can achieve a beam size for Bi3 cluster primary ions down to 20 nm [5]. These heavy projectiles exhibit very high secondary ion yields in particular for organic materials. For sputtering, the instrument is equipped with both an oxygen and a cesium beams. At low sputter energies of a few hundred eV a depth resolution of about 1 nm is achieved for inorganic thin films. In this work, we aim at showing how ToF-SIMS is complementary to SFM on the two aspects of chemical information and of 3-dimensional analysis by presenting results on reference samples. First, an application of KPFM is shown on a patterned sample consisting of layers of silicon, aluminum and copper (see Figure 1). Contact potential differences (CPD) are clearly identified for the 3 layers (Fig. 1b), however they cannot be assigned to a specific element. Only with the ToF-SIMS analysis is the identification unequivocal (Fig. 1c). Second, the in-depth magnetism of a conventional hard disk is imaged using MFM. As layers are being sputtered away, the magnetic contrast becomes first stronger due to the removal of the lubricating layer, but then disappears as the magnetic layer is progressively removed. Finally, an OLED unit cell is analyzed in-depth by successive ToF-SIMS and SFM topographical analyses of the sputtered slices. On such a complex structure, different sputtering rates lead to a depth differing widely between homogeneous regions. SFM is thus essential to reconstruct the correct volume of the cell. a) b) c) Al CPD [V] Z [nm] Intensity 2 0 0 .5 0 Intensity Si 0 .4 5 0 0 .4 0 -2 0 0 .3 5 0 1 0 2 0 D is t a n c e [ � m ] 0 1 0 2 0 D is t a n c e [ � m ] Figure 1: a) 32x10 �m2 SFM topography of a patterned Al/Cu/Si surface. b) Simultaneous KPFM signal measuring the contact potential difference. c) ToF-SIMS images of aluminum and silicon. References: [1] B.C. Stipe et al. Nature Photonics 4 (2010), p. 484 [2] Y. Sugimoto et al. Nature 446 (2007), p. 64 [3] K. Iktgen et al. in "Secondary Ion Mass Spectrometry", SIMS X., ed. A. Bennighoven, B. Hagenhoff and H.W. Werner, (John Wiley & Son) [4] The authors acknowledge funding from the European Commission (FP7, Grant number 200613). [5] F. Kollmer et al. , Surf. Interface Anal. 45 (2013), p. 312");sQ1[715]=new Array("../7337/1433.pdf","AFM Integrated with SEM/FIB for Complete 3D Metrology Measurements","","1433 1433 doi:10.1017/S1431927615007941 doi:10.1017/S1431927615007941 Paper No. 0715 Microsc. Microanal. 21 (Suppl 3), 2015 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 � Microscopy Society of America 2015 AFM Integrated with SEM/FIB for Complete 3D Metrology Measurements Aaron Lewis1, A. Komissar2, A. Ignatov2, Oleg Fedoroyov2, and E. Maayan2 1. The Department of Applied Physics, The Hebrew University of Jerusalem 2. Nanonics Imaging Ltd, 19 Hartum St., BYNET Bldg, Jerusalem, Israel Scanning electron microscopy (SEM) and ion beam milling (FIB) techniques are mature nanoscale measurement technologies, while Atomic Force Microscopy (AFM) is a developing technology with intense interest in the scientific community for basic research and development. We discuss the recent integration of these techniques into a single instrument enabled by technological innovation in AFM instrument design, and applications of this instrument for characterization of FIB milled trenches. An AFM integrated into the SEM/FIB has specific design constraints that need to be met involving the AFM scanner, probe design, and feedback loop so that it geometrically fits into the chamber without interfering with the electron and ion beams. An ultraflat scanning stage was thus developed for precise sample motion that enable large ranges (85 microns or greater) in x,y, z. SPM probes that do not interfere with the electron beam were manufactured from fused silica in a cantilevered geometry (see Fig 1) where there is full visualization of the tip, without any loss of SPM functionality. These probes are combined with a tuning fork based feedback mechanism to maintain accurate tip-sample control; this mechanism does not involve lasers or optical beam deflection that interferes with the electron and ion beams. The many benefits of this combined approach to three-dimensional nanoscale characterization are illustrated in the imaging of a FIB milled trench. Figure 2a shows an AFM image profiling a 25-�m-deep trench milled in silicon with FIB imaged with the large Z range of the system. In this measurement, the FIB beam was used to mill this feature, and it was followed immediately by AFM imaging without having to remove the specimen from the FIB chamber and search in the AFM. Furthermore, the true depth of the trench was measured by the AFM in a crosssectional profile (Figure 2b) while a 3D reconstruction revealed the geometry of the sidewall (Figure 2c). Sidewall imaging with such long and exposed tips now are feasible with this instrument. The red arrow in this image clearly shows a Pt decoration on this structure, and the white arrow shows an undercut. Additionally, the ability of AFM to quickly and easily follow FIB milling with high Z resolution provides a straightforward and convenient method to check on FIB milling results. This capability is demonstrated in Figure 3 where several FIB-milled features in Si were monitored during the milling process. The grayscale SEM image in Figure 3a can easily locate these features and measure their widths but cannot provide direct information on their depths. The AFM delivers a 3D reconstruction (Figure 3b), and a cross sectional profile that together provide a true topographic map of specimen features. These depth profiles demonstrate the linearity of the etching process with time. This instrument incorporates important innovations in the design and technology of atomic force microscopy enabling its integration with SEM and FIB into a single powerful capability. Microsc. Microanal. 21 (Suppl 3), 2015 Microsc. Microanal. 21 (Suppl 3), 2015 1434 1424 Figure. 1 Figure 2a 2b 2c Figure 3a 3b References (1) Maier, S.; Kik, P.; Atwater, H. A.; Meltzer, S.; Harel, E.; Koel, B. E.; Requicha, A. g. Local detection of electromagnetic energy transport below the diffraction limit in metal nanoparticle plasmon waveguides. Nature Materials 2003, 2, 229. (2) Prikulis, J.; K.Murty; Olin, H.; Kall, M. Large area topography analysis and near field Raman spectroscopy using bent fibre probes. Journal of Microscopy 2003, 210, 269-273. (3) Raghu, D.; Hoffman, J. A.; Garnari, B.; Reeves, M. E. Near-field ablation threshold of cellular samples in the mid-infrared wavelength region. Appl Phys Lett 2012, 100, 203703. (4) Dekhter, R.; Khachatryan, E.; Kokotov, Y.; Lewis, A. Investigating material and functional properties of static random access memories using cantilever glass multiple-wire force-sensing thermal probes. Appl Phys Lett 2000, 77, 4425. (5) Ren, X.; Liu, A.; Zou, C.; wang, L.; Cai, Y.; Sun, F.; Guo, G.; Guo, G. Interference of surface plasmon polaritons from a "point" source. appl Phys Lett 2011, 98, 201113. (6) Lewis, A.; Kheifetz, Y.; Shambrodt, E.; Radko, A.; Khatchatryan, E.; Sukenik, C. Fountain pen nanochemistry: atomic force control of chrome etching. Appl Phys Lett 1999, 75, 2689. (7) Ternes, M.; Lutz, C. P.; Hirjibehedin, C. F.; Giessibl, F. J.; Heinrich, A. J. The force needed to move an atom on a surface. Science 2008, 319, 1066. (8) Kohlgraf-Owens, D.; Sukhov, S.; Dogariu, A. Mapping the Mechanical Action of Light. Phys Review A 2011, 84, 011807. (9) Little, A.; Hoffman, A.; Haegel, N. M. Optical attenuation coefficient in individual ZnO nanowires. Optics Express 2013, 21, 6321.");sQ1[716]=new Array("../7337/1435.pdf","Integrating Atomic Force Microscopy in Scanning Electron Microscopy.","","1435 doi:10.1017/S1431927615007953 Paper No. 0716 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Integrating Atomic Force Microscopy in Scanning Electron Microscopy. Andrew J. Smith1, Klaus Schock1, Gregor Renka1, Andreas Lieb2, Massoud Dadras3, and Stephan Kleindiek1 1. 2. Kleindiek Nanotechnik, Reutlingen, Germany Nanosurf AG, Gr�ubernstrasse 12�14, 4410 Liestal , Switzerland 3. Centre Suisse d'�lectronique et de Microtechnique, CSEM SA, Neuch�tel, Switzerland Combining Atomic Force Microscopy (AFM) with Scanning Electron Microscopy (SEM) opens up new possibilities in 3D imaging. The SEM has the ability to quickly generate images of the sample using a wide array of detectors and analysis methods (SE, EBSD, EDX, ...). These methods can be utilized to locate an area of interest. With an in situ AFM, the located site can be imaged using force microscopy without the need for complicated registration or cumbersome relocating of the area of interest ex situ. In this setup, the AFM cantilever is mounted to a micromanipulator while the sample is mounted to a sample scanner. Thus, once the area of interest has been located as described above, the cantilever can be brought into position and landed on that site. Subsequently, the sample is scanned under the cantilever tip in order to generate a topography image. With the large amount of current research and development focused on nano wires, carbon nano tubes, and other nano scale materials, imaging these materials has become a large part of the challenges involved. The two most prominent methods for imaging at the nano scale are Scanning Electron Microscopy and Atomic Force Microscopy. These complimentary methods utilize fundamentally different principles for generating imagery - SEM exploits the interaction of electrons with matter, while AFM is based on physical interaction of a sharp tip with the sample surface. Both approaches have strengths and weaknesses. The SEM's strength is to quickly generate images with a large range of magnifications, making it easy to locate the area of interest. However, it doesn't yield 3D information, e.g. "invisible" contamination layers. The AFM's main advantage is its ability to obtain 3D information, the downsides are that it is hard to find the target area and image generation is slow. Combining these two tools into one setup - putting an AFM inside an SEM - gives quick access to a more complete data set. Additionally, FIB-milled or FIB-deposited structures can be characterized using this combination of tools in a FIB/SEM system. The utility of this combination of tools is demonstrated with several examples where locating the area of interest purely by AFM or light microscopy would have been highly impractical. Microsc. Microanal. 21 (Suppl 3), 2015 1436 Figure 1. Addressing the target area for AFM inspection is fast and easy using the SEM. Figure 2. Si surface observation by SEM and AFM characterization of defects. The defects have a height of about 12nm.");sQ1[717]=new Array("../7337/1437.pdf","Integrated SIMS-AFM Instrument for Accurate High-Sensitivity and High-Resolution Chemical 3D Analysis","","1437 doi:10.1017/S1431927615007965 Paper No. 0717 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Integrated SIMS-AFM Instrument for Accurate High-Sensitivity and HighResolution Chemical 3D Analysis Y. Fleming, T. Wirtz, S. Eswara Moorthy Advanced Instrumentation for Ion Nano-Analytics (AINA), Luxembourg Institute of Science and Technology, 41 rue du Brill, L-4422 Belvaux, Luxembourg With the progress of miniaturization, driven by future needs in various fields in materials and life science, the 3D analysis of devices and material structures becomes increasingly challenging. As a consequence, the interest for performing bimodal or even multimodal nano-analysis has increased during the last decade. In particular, nano-analytical techniques and instruments providing both excellent spatial resolution and high-sensitivity chemical information are of utmost importance for investigations at the nanoscale. Secondary Ion Mass Spectrometry (SIMS) is a method of choice for high sensitivity analysis. State-ofthe-art SIMS imaging instruments can provide chemical 2D and 3D maps with a lateral resolution of around 50nm. Other advantages of SIMS include excellent dynamic range and the ability to differentiate between isotopes and thus to map isotopic ratios. However, several important artifacts arise from the fact that the 3D mappings do not take into account the sample's surface topography. While the 3D reconstruction protocols and software assume that the initial sample surface is flat and the analyzed volume is cuboid, "real samples" present a surface topography, which furthermore changes during the ion bombardment as the local sputter yields depend on parameters such as the local angle of incidence of the ion beam and the crystal orientation. In addition, the situation is worsened if the sample is constituted of different materials due to preferential sputtering phenomena. As a consequence, the produced 3D images are affected by a more or less important uncertainty on the depth scale and can be distorted. In order to obtain real high-resolution SIMS 3D analyses without being prone to the aforementioned artifacts, we developed a Scanning Probe Microscopy (SPM) module which we integrated into the Cameca NanoSIMS50 [1-4]. This in-situ combination between SIMS and SPM avoids the artifacts (e.g. different environment conditions, topography changes due to surface diffusion and reaction of reactive species used as primary ion sin SIMS) occurring when an ex-situ combination between these same techniques is used. In addition, this integrated instrument allows a combination of SIMS images with KPFM (Kelvin Probe Force Microscopy) data recorded in-situ in order to provide an extended picture of the sample under study. The known information channels of SIMS, AFM and KPFM are thus combined in one analytical and structural tool, enabling new multi-channel nanoanalytical experiments. This opens the pathway to new types of information about the investigated nanomaterials. In this paper, we will illustrate the analytical potential of the combined SIMS-SPM approach by presenting several applications in the fields of materials science and life science. References: [1] Y. Fleming, T. Wirtz, U. Gysin, T. Glatzel, U. Wegmann, E. Meyer, U. Maier, J. Rychen, Appl. Surf. Sci. 258(4), 1322-1327 (2011). Microsc. Microanal. 21 (Suppl 3), 2015 1438 [2] T. Wirtz, Y. Fleming, U. Gysin, T. Glatzel, U Wegmann, E. Meyer, U. Maier, and J. Rychen, Surf Interface Anal. 45 (1), 513-516 (2013). [3] T.Wirtz, Y. Fleming, M. Gerard, U. Gysin, T. Glatzel, E. Meyer, U. Wegmann, U. Maier, A. H. Odriozola and D. Uehli, Rev. Sci. Instrum. 83, 063702 (2012). [4] C. L. Nguyen, T. Wirtz, Y. Fleming and J. B. Metson, Appl. Surf. Sci. 265, 489-494 (2013). Figure 1. PVP/PS polymer blend after Cs+ bombardment of 1016 ions/cm2: The SIMS recorded secondary ion intensity and the AFM recorded topography of the area of interest are superposed and compiled into a 3D surface mapping. Figure 2. 27Al16O- (left) and 52Cr16O- (right) 3D SIMS-AFM images of a nickel-based super-alloy. The 3D images nicely show the correlations between the chemical composition and the topography.");sQ1[718]=new Array("../7337/1439.pdf","Quantification of Nano-inclusions by EPMA Using Conventional Accelerating Voltages","","1439 doi:10.1017/S1431927615007977 Paper No. 0718 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantification of Nano-inclusions by EPMA Using Conventional Accelerating Voltages. Claude Merlet GM, CNRS, Universit� de Montpellier, Place E. Bataillon, 34095 Montpellier, France. For more than four decades, the electron probe microanalysis (EPMA) has been routinely used to determine the chemical composition of materials on a micrometer scale. For analysis, typical operating conditions use beam energies in a range of 15�20 keV, which give spatial resolutions of about 2�4 �m for a material of mass density of 5 g/cm3. The volume of interaction can be reducing by decreasing the beam energy, but the large increase in spot size with tungsten gun EPMA at low beam energy limits the lateral resolution to a minimum of about 0.8 �m at 7 keV. The Schottky emitter field-emission gun (FEG) EPMA now offer the possibility of obtaining a focused electron beam of about 100 nm with a current of 10 nA at a beam energy of 5 keV. At this energy, the penetration range of incident electrons drops to 0.3 �m, and leads to a significant improvement in the lateral resolution and surface sensitivity, but quantification, at low beam energy is accompanied by experimental and analytical problems which may affect the accuracy of results [1]. Low beam energies require the use of low-energies X-ray lines which are often affected by spectroscopic difficulties and larger uncertainties in the associated massattenuation coefficients (MACs). Moreover, the fluorescence yields are generally lower than for the conventional K-lines, and as a result the detection limit worsens. At low beam energy, as the ratio of beam energy to excitation energy of the lines (overvoltage) is often low, the X-ray intensity decreases significantly. As the X-ray generation is significantly reduced at low beam energy it is necessary to increase the beam current to give a sufficient count rate that can induce damage to the sample. Finally, at low beam energy, a surface layer of several nanometers in thickness represents a much larger fraction of the sample and, therefore, the influence of carbon contamination, surface oxidation, the quality of the sample polish or a conductive coating become more significant. To reduce these difficulties, another option to quantify small inclusions is to use conventional beam energies in a range of 8-15 keV, and to extract by calculation the signal resulting of the inclusion embedded in the surrounding material (matrix). This approach requires firstly that the X-ray intensities coming from the inclusion and the matrix can be calculated and secondly that the surrounding material can be accurately determined. X-ray intensities can be calculated using Monte Carlo simulation [2] or specific analytical program [3]. In this work, to quantify the inclusions, an analytical X-ray distribution in three dimensions was developed. The three dimensions X-ray distribution ((x, y, z)) use the product of an X-ray lateral distribution ( ( x, y, z ) ) and the well known double Gaussian X-ray depth distribution ( (z ) ) model, whose accuracy has been largely demonstrated for bulk [4] and thin layers [5]. As the number of ionization is included in the X-ray depth distribution model, the X-ray lateral distribution has to be normalized. In the proposed analytical (x, y, z) model, the X-ray lateral distribution use a two-dimensional Gaussian distribution with a width depending of the depth. As a consequence the geometry of the X-ray distribution is approaching a truncated sphere (Fig. 1). For a given inclusion geometry and by using this analytical (x, y, z) model, the X-ray intensities of the inclusion and of the matrix can be computed numerically [6]. Microsc. Microanal. 21 (Suppl 3), 2015 1440 As, in practice the geometry of the inclusion is unknown, an alternative is to approximate the sample as a cylindrical thin layer with same thickness and radius having the composition of the inclusion in a substrate having the composition of the matrix. The method is an improvement of the multilayer calculation [5], which is done in three dimensions by taking into account the X-ray lateral emission. The algorithm computes by an iterative procedure or by a trial and error approach the emergent intensities corresponding to the various elements of the combination inclusion-matrix and the inclusion size. For the consistency of the results, it is important to have a reasonable convergence between calculation and measurement of the elements of the inclusion but also of the matrix. To assess the validity of this 3D layer approximation, the method was tested on inclusions embedded in super-alloys, minerals and pure metals for diameters ranging from 100nm to 2�m. For this study, some samples were built from bulk standards which were crushed to small diameter sizes and embedded in soft metals. Since the compositions of these inclusions are known, these systems were chosen to validate this quantitative method. Some experimental results obtained are presented to demonstrate the validity of this method to perform quantitative analysis of small inclusions. As example, figure 2A shows a BSE map of a geological sample, which contains a large distribution in size of Chromium-spinel inclusion in feldspar [7]. Figure 2B gives the mass concentration of the Chromium-spinel inclusion obtained with the 3D layer approximation in function of the inclusion size; the horizontal lines are the concentrations obtained on sizes larger than 10 microns. The deviation in the concentration is at the maximum 10% for the magnesium and less that 6% for the other elements. References [1] C. Merlet and X. Llovet, IOP Conf. Ser. Mater. Sci. Eng. 32 (2012) p. 012016. [2] R. Gauvin, P. Hovington and D. Drouin, Scanning 17 (1995) p. 202. [3] O. Arnould and F. Hild, X-Ray Spectrometry 32 (2003) p. 345. [4] C. Merlet, Mikrochimica Acta, 114/115 (1994) p. 363. [5] X. Llovet and C. Merlet, Micros. Microanal. 16 (2010) p. 21. [6] J. T. Armstrong and P. R. Buseck, Anal. Chem. 47 (1975) p. 2148. [7] F. Kalfoun, D. Ionov and C. Merlet. Earth Planet. Sci. Lett. 199 (2002) p. 49. 30 25 weight fraction % Al 20 15 10 5 0 0 0.2 0.4 0.6 0.8 1 Inclusion size (micron) Mg Cr Fe Figure 1. Analytical three-dimensional X-ray distribution (x, y, z). B A Figure 2. (A) BSE map of Chromium-spinel inclusion in feldspar. (B) Weight fraction of inclusions obtained by using the 3D layer approximation.");sQ1[719]=new Array("../7337/1441.pdf","Multi-beam energy acquisition in FE-EPMA","","1441 doi:10.1017/S1431927615007989 Paper No. 0719 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Multi-beam energy acquisition in FE-EPMA Silvia Richter and Philippe T. Pinard Central Facility for Electron Microscopy, RWTH Aachen University, Aachen, Germany. Varying the beam energy is a common method used for the analysis of laterally homogeneous multilayered structures in EPMA. With increasing beam energy the information depth is increasing as well. Thus, the multi beam energy acquisition delivers a possibility for depth profiling [1]. On the other side, cross sections measurements are often performed at one single beam energy. This guarantees stable conditions like well-adjusted focus, astigmatism and gun alignment. Furthermore, available quantification software is mainly designed for single beam energy acquisition and for the accurate determination of the lateral resolution the lateral ionisation probability is less studied and proven than the well-known (z)-function. With the introduction of the field emission electron microprobes the analysis of submicron features have been becoming more and more of interest due to better focusing capabilities. Since the spatial resolution is then mainly determined by the broadening of the beam due to scattering events within the specimen, a common solution is to apply low voltage analysis, i.e. the use of one single low beam energy. However, as shown recently [2] the low voltage analysis has its limitations, if soft X-ray lines, i.e. L lines of the transmission elements, are used. Thus, the goal of our work is to present a new technique, where the acquisition is performed at multiple beam energies using well-established X-ray lines and almost the same overvoltage ratios. For the proper selection of the beam energies the lateral intensity distribution was studied by means of Monte Carlo simulations in MONACO [3]. The features of interest are needle-like carbides in high manganese steel (Fig. 1) containing the binding elements Fe, Mn and Al. A minimum beam energy of 9 keV is needed to excite the highest energetic line: Fe K . However, at this beam energy the soft X-ray lines, Al K and C K , limit the lateral resolution to about 600 nm (see Fig. 2). Lowering the beam energy down to 4 keV and keeping in mind from previous work [4], that the beam diameter is worsen at lower beam energy, the lateral resolution of Al K and C K was adapted to the resolution of the other ones (see Fig. 3). At this well-chosen set of conditions the measurement was performed by scanning the beam across the precipitate. In between the two conditions the beam was focused and positioned manually. After background subtraction and calibration k-ratios were extracted from the maximum of the line profile and converted into elemental concentration using a special matrix correction procedure (see Table 1). One step forward an automotive procedure was developed for a series of high spatial resolution measurements where each position, i.e. precipitate, on the sample is analyzed under different conditions. The challenges are related to the stage reproducibility, beam drift, focusing capabilities and positioning of the beam [4]. Especially, changing the beam energy results in a large shift of the beam of up to several �m (Fig. 4). A shift correction procedure is proposed based on image registration techniques. The basic principles are shown in Fig. 5, where a reference image is compared by a crosscorrelation function with images taken at 4 and 10 keV. The goal of the analyses was to quantify the chemical composition of small Nb-enriched precipitates in a low alloyed steel matrix (see Table 2). Fe K and Cr K were analysed at 10 keV and the other X-ray lines at 4 keV. Since contrast and brightness values of backscattered electron images are highly sensitive to condition changes, limits of the new measurement strategy are also presented. Microsc. Microanal. 21 (Suppl 3), 2015 1442 1432 [1] P. Willich, Mikrochim. Acta 12 (1992), pp.1�17. [2] P. Pinard et al, Micros. Microanal. 20 S3 (2014), pp. 700-701. [3] N. Ammann, P. Karduck, Microbeam Analysis (1990) (J. R. Michael, P. Ingram, eds.), San Francisco Press, San Francisco, p. 150. [4] S. Richter et al, Microsc. Microanal. 20 S2 (2014), pp. 680-681. Figure 1: BSE-image of high Mn steel with carbides. Figure 2: Lateral intensity distributions at 9 keV at% Fe 55.8 Mn 18.2 Figure 3: Lateral intensity distributions at 4 keV Al 12.5 C 13.5 RMS(kexp,kth) 1.03% Table 1: Composition of carbide shown in Fig. 1and root mean square deviation between measured and simulated k-ratios. Figure 4 (left): Shift of the beam position as a function of the beam energy at different beam currents wt% Si 3.7 Cr 5.3 Fe 50.9 Nb 30.9 Mo 8.6 Total 99.2 Table 2: Composition of Nb-enriched precipitate Figure 5 : Principle of image registration technique using a reference image (left) for comparison with images acquired at 4 keV (middle) and 10 keV (right).");sQ1[720]=new Array("../7337/1443.pdf","Improvements in EPMA: Spatial Resolution and Analytical Accuracy","","1443 doi:10.1017/S1431927615007990 Paper No. 0720 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Improvements in EPMA: Spatial Resolution and Analytical Accuracy P. K. Carpenter1 and B. L. Jolliff2 1 Dept. of Earth and Planetary Sciences, Washington University, St. Louis, MO 63130, USA 2 Dept. of Earth and Planetary Sciences and the McDonnell Center for the Space Sciences, Washington University, St. Louis, MO 63130, USA In electron-probe microanalysis (EPMA), the analyst must balance the desire for high spatial resolution with practical aspects of quantitative analysis. A reduction in accelerating voltage and probe current improves spatial resolution and image quality, but forces the selection of X-ray lines with low excitation energy (e.g., L and M-family X-ray lines vs. K and L-family lines). Measurement of L-family X-rays is complicated by low fluorescent yield, an increase in X-ray absorption, and numerous X-ray interferences from other elements in the sample. The use of transition element L lines generally results in both poor sensitivity and accuracy. Lunar agglutinates are rock and mineral soil materials included in glass formed by micrometeorite impact events and exposure to solar wind processes [1]. Agglutinates contain sub-micron spherical droplets of Fe metal that form initially on the surface of grains during exposure to energetic particles. Impact processes rework the regolith, resulting in agglutinates containing mineral and rock fragments with melt and quench textures and pervasive metal spherules. Micron sized spherules are variably derived from melted minerals plus a meteoritic component, and may contain Ni, P, and Co, but nanophase Fe is essentially pure Fe, and the relationship between spherule size and chemistry has not been extensively studied. The fine scale of these spherules requires microanalysis at high spatial resolution as well as consideration of sampling, measurement, and accuracy. An Apollo 17 agglutinate 76503,7020 from the study by Jolliff et. al. [2] has been selected to evaluate spatial resolution and accuracy of EPMA. Gopon et. al. have analyzed submicron Fe-Si droplets from Apollo 16 soils at low voltage, using the Fe Ll X-ray line rather than the Fe L line due to accuracy issues with the latter [3]. The chemistry of Fe spherules is an indicator of oxygen fugacity and source material, and thus provides insight into both the conditions and source materials which form agglutinates. The current study will present aspects of microanalysis and the relative merits of multiple kV, reduced kV, and low-kV measurement methodology to improve spatial sampling coupled with an assessment of analytical accuracy. Agglutinate 76503,7020 consists of glass with inclusions of regolith materials such as rock fragments and mineral grains with textures indicative of impact melting events (Fig. 1). Submicron Fe spherules are pervasive and abundant at grain boundaries and FeTi-rich zones in the agglutinate glass. Preliminary EPMA of micron-sized Fe spherules indicates that they contain minor Ni and P with trace Co, thus suggesting a meteoritic contribution. The fundamental problem with microanalysis of the spherules is to discriminate between elements in the spherule vs. elements in the matrix. It is necessary to quantitatively determine the electron and X-ray primary analytical volume and degree of sampling of adjacent matrix by secondary fluorescence. Primary electron scattering through the spherule results in preferential excitation of low Z matrix elements such as Mg and Al, which are not likely contained in the spherule, but also Si, which may well be. Secondary fluorescence by characteristic and continuum X-rays from spherule elements (e.g., Fe, Ni, and P) may excite Ca and Ti (by Fe K and Ni K) and Si (by P K). Monte Carlo simulations using DTSA-II (version Halley) have been made for a cubic inclusion in a matrix of glass, using EPMA data from larger spherules and adjacent glass [4]. These simulations are a valuable tool to use in combination with calculated electron and X-ray ranges in these materials. The results are shown in Fig. 2 which illustrates a factor of 2 reduction in scattering volume with decrease in accelerating potential from 15 to 10 kV. The excitation volume for Fe and Ni K at 10 kV is modeled to be within a 0.5 x 0.5 m cube approximating a Fe spherule (Wt.% element: Fe 96.4, Ni 2.3, P 0.43) embedded in a glass of measured composition. These simulations were run with both characteristic and continuum fluorescence options enabled, which significantly expands the sampled volume. The simulated EDS spectra at 15 vs. 10 kV show a significant reduction in the matrix elements Mg, Al, and Si at 10 kV which indicates these elements are Microsc. Microanal. 21 (Suppl 3), 2015 1444 present in the matrix glass only. In practice the Al/Fe ratio of Fe spherules is a sensitive indicator of matrix glass sampling by side-scattered electrons. The accuracy of EPMA at reduced overvoltage has not been well studied. Strong curvature in the X-ray ionization cross-section at overvoltage below ~ 2 and poor agreement of algorithms results in the EPMA rule to "avoid overvoltage less than 2" in analytical setups. The excitation efficiency also decreases (e.g., Ni peak intensity at 10 kV in Fig 2), so both replicate analysis and detection limit will suffer at reduced overvoltage. A summary of measurements made on a suite of standards at reduced overvoltage will be discussed to provide a framework for evaluation of the benefits obtained by the improved spatial resolution. References [1] Lucey, P., New Views of the Moon, pp 83-220, Mineralogical Society of America, 2006. [2] Jolliff et. al., Met. Planet. Sci. 31, 116-145, 1996. [3] Gopon et. al., Microsc. Microanal. 19, 1698-1708, 2013. [4] http://www.cstl.nist.gov/div837/837.02/epq/dtsa2/index.html A Apollo 17 Agglutinate 76503,7020 B C Figure 1 Apollo 17 agglutinate 76503,7020, BSE images of agglutinate and increased magnification of area outlined in red. Fig 1A, Overview image of agglutinate with glass, mineral, and lithic inclusions. Fig 1B, Enlarged area from Fig 1A of FeTi-rich glass and brighter Fe melt spheres. Fig 1C, Enlarged area from Fig 1B illustrating range of nano- and micron-sized Fe melt spheres pervasively distributed in the agglutinate glass. 76503,7020 Cube: 0.5 m Fe(Ni,P) spherule Matrix: agglutinate glass Si K 15 kV Si K 10 kV 0.5 m Fe(Ni,P) inclusion in agglutinate glass Red 15 kV, Blue 10 kV Fe K 15 kV Fe K 10 kV Figure 2 DTSA-II simulations of 0.5 m cube of Fe(Ni,P) outlined in blue, in agglutinate matrix glass. Left, Monte Carlo scattering simulation for Si K and Fe K at 15 vs. 10 kV. Note ~ 2x reduction in analytical volume at 10 vs. 15 kV. Right, Simulated EDS spectra compared, note significant excitation of matrix elements at 15 kV vs. near elimination of same at 10 kV, but reduction in excitation of Fe and Ni K at 10 kV due to low overvoltage.");sQ1[721]=new Array("../7337/1445.pdf","Improving Trace Element Analysis Precision By Not Using Off-Peak Measurements","","1445 doi:10.1017/S1431927615008004 Paper No. 0721 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Improving Trace Element Analysis Precision By Not Using Off-Peak Measurements John J. Donovan1 and John T. Armstrong2 1. 2. CAMCOR, University of Oregon, Eugene, OR, 97403 Carnegie Institution for Science, Geophysical Lab, Washington, DC, NW, 20015-1305 It is well known that trace element sensitivity in Electron Probe Micro Analysis (EPMA) is limited by intrinsic random variations in the x-ray continuum background produced by the deceleration of the electron beam by the Coulombic field of the specimen atoms. Typically, this continuum must be characterized with sufficient precision (along with the peak intensity of the emission line of interest), in order to attain reasonable sensitivity for the elements of interest. Generally we characterize these intensities by measuring on either side of the emission line and interpolate the intensity underneath the peak to obtain the net intensity. However, due to off-peak interferences from secondary emission lines, "holes" in the continuum from secondary Bragg diffraction, non-linear curvature of the wavelength dispersive spectrometer (WDS) continuum and other continuum characterization issues, this off-peak method often requires careful study of a wide swath of the emitted continuum spectrum by means of high precision WDS scans which can be quite time consuming. This is true when acquiring WDS scans with sufficient precision to detect not only the trace element emission lines of interest, but also secondary emission lines from other elements in order to avoid them when selecting off-peak measurement positions. We will demonstrate that, at least for materials with a relatively simple matrix such as SiO2, TiO2 or ZrSiO4 where one may obtain suitably well characterized standards for use in the so called "blank correction" [1], we can take advantage of the Mean Atomic Number (MAN) background correction method. This technique was originally published for major and minor element characterization [2], but we will demonstrate in this paper that the MAN background method can also be utilized to obtain high precision trace element characterization without off-peak measurement by modeling the continuum absorption [3]. Trace element accuracy can be further assured by use of the previously mentioned "blank correction" technique, so that one may obtain similar accuracy with improved precision and in approximately half the acquisition time of typical off-peak trace element measurements. This MAN background method applies both to point analyses and quantitative x-ray mapping where the time savings are particularly significant and improvements in precision are especially noticeable. Background intensities for a synthetic zircon (ZrSiO4) material calculated from traditional off-peak linear interpolation are shown fig 1, while the lower-variance backgrounds in fig 2 are from the averageZ-based MAN background calibration curve from measurements on a number of oxide standards. [1] JJ Donovan, HA Lowers, BG Rusk (2011) Improved Electron Probe Microanalysis of Trace Elements in Quartz, American Mineralogist, 96, 274-282 [2] JJ Donovan and T Tingle (1996), An Improved Mean Atomic Number Background Correction for Quantitative Microanalysis, Jour. of Micros. Microanal., 1-7 [3] JT Armstrong (1988) Quantitative Analysis of Silicate and Oxide Materials: Comparison of Monte Carlo, ZAF, and Procedures, Microbeam Analysis, 239-246 Microsc. Microanal. 21 (Suppl 3), 2015 1446 Fig.1. Calculated background intensities in a synthetic zircon using a linear interpolation from the measured off-peak intensities for Hf M, U,M, P K and Th M. Conditions were 20 keV, 100 nA, 3000 msec on-peak, 1500 msec off-peak (x2). Note that the calculated background intensities show the expected variance from the off-peak measurement uncertainties. Synthetic Zircon (Hanchar) MAN, Bgd Intensities Hf cps/1nA U cps/1nA -550 -550 -600 -600 -650 19650 19700 X (um) [127 um] -650 19650 19700 X (um) [127 um] P cps/1nA Th cps/1nA -550 -550 -600 -600 -650 19650 19700 X (um) [127 um] -650 19650 19700 X (um) [127 um] Fig.2. Calculated background intensities using a linear regression curve from the measured on-peak intensities for a number of standard materials which do not contain the elements of interest. Hf M, U M, P K and Th M, at 20 keV, 100 nA, 3000 msec on-peak only. Note that the calculated MAN background intensities show a much smaller degree of variance than the off-peak background intensities in Fig 1.");sQ1[722]=new Array("../7337/1447.pdf","Using EPMA, Raman LS, Hyperspectral CL, SIMS, and EBSD to Study Impact- Melt-Induced Decomposition of Zircon","","1447 doi:10.1017/S1431927615008016 Paper No. 0722 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Using EPMA, Raman LS, Hyperspectral CL, SIMS, and EBSD to Study ImpactMelt-Induced Decomposition of Zircon M. Zanetti1, A. Wittmann1, P. K. Carpenter1, B. L. Jolliff1, E. Vicenzi2, A. Nemchin3, N. Timms3 1 Earth and Planetary Sciences and McDonnell Center for Space Sciences, Washington University in St. Louis, St. Louis, MO 2 Smithsonian Institution, Museum Conservation Institute, Suitland, MD 3 Curtin University, Perth, Australia Using an integrated multi-instrument approach we investigate two exceptionally well-preserved zircon grains in a obsidian-like (holohyaline) impact melt glass from the 28 km diameter Mistastin Lake Impact Structure (Labrador, Canada). The zircon grains (Fig 1) contain an un-shocked relict core and a 20�50 micron thick quenched decomposition rim, formed by the dissolution reaction of zircon to ZrO2 + SiO2 phases when the grains were entrained in impact melt during the impact event ~36 Ma [1, 2]. Using electron-probe microanalysis (EPMA), laser Raman spectroscopy (LRS), hyperspectral imaging cathodoluminescence (CL), secondary ion mass spectrometry (SIMS), and electron backscatter diffraction (EBSD), we determined the composition and phases present in the zircon core, decomposition rim, and surrounding glass, and we investigated the rim-core dissolution interface, correlating REE zoning with hyperspectral CL in the zircon core. The zircon grains were identified in thin-section. EPMA analysis included backscattered electron (BSE) imaging, compositional x-ray mapping, wavelength-dispersive (WDS) spot and traverse analyses, including selected REEs, using a JEOL JXA-8200 at Washington University in St Louis. Mineral phases in the core and decomposition rim were identified and mapped using a Renishaw In-Via LRS with 0.6 �m pixel resolution. CL data were collected with a Gatan MonoCL4 Elite system. Ion mapping (U-Th-Pb and Y, Dy, Er, and Yb) was conducted for comparison to EPMA analyses and CL zoning, and spot measurements were made to determine the age of the zircon core using the NORDSIM Cameca 1280 in Stockholm, Sweden. LRS data show that the decomposition rim is a predominantly vermicular intergrowth of baddeleyite and amorphous glass (sourced from impact melt infiltrating the rim during decomposition) and confirm the presence of an unaltered zircon core. LRS data also reveal relict blebs of poorly crystalline tetragonal ZrO2 associated with baddeleyite intergrowths, suggesting a minimum impact melt temperature of 1676�C (and possibly much greater [e.g., 3]). EPMA x-ray maps show that interstitial material in the rim has nearly the same composition as the surrounding glass and that some components diffused into, and others out of, the rim zone (i.e, Al and other impact melt components moving into the grain, and Zr and other zircon components moving out to the surrounding melt), suggesting that surrounding melt infiltrated the grain as the heat of the melt decomposed the zircon crystal rim. CL data exhibit broad spectra with superimposed sharp peaks interpreted to represent specific REE and spatial patterns related to magmatic zoning and zircon decomposition (Fig 2).. Within the zircon core, crystallization zoning patterns in CL display broad spectral features, in addition to high-frequency sharp peaks interpreted as resulting from HREEs (Tm3+, Er3+, Dy3+) based on correlation with the CSIRO luminescence spectral database (http://www.csiro.au/luminescence/). High resolution WDS trace-element analyses (traverse seen in Fig 1b; 2 �m beam, 15 kV, 150 nA) indicate an inverse correlation between Y concentrations and CL intensity (Fig 1b). Dark areas in CL within the core contain low concentrations (measureable, but at detection limits) of Er (20-30 ppm) and Dy (~10 ppm), and relatively higher concentrations of Y (~1500 ppm). It is known that REE (including Sc and Y) are important activator elements, and that Dy3+ is the main CL activator in zircon by virtue of more efficient excitation in the zircon lattice ([4, 5, 6]. We see Y3+ quenching in the brightest areas of the zircon core (Fig 1b), but a measureable concentrations of Th4+ and Er3+ in bright areas. It may be that Y3+ quenching dominates the CL signal in dark areas, and activators (Dy3+ , Er3+, and Th4+) are enough to cause the bright CL, even in very low concentrations (< 100 ppm). Distinctive blue CL is observed along the core-rim interface and along cracks in the zircon grain, and must be related to the dissociation of zircon to ZrO2 + SiO2. Although not well understood, blue CL in zircon has been attributed to a delocalized electron on the [SiO4] group [5, 6], suggesting the interface records the initial stage of breakdown. Blue CL along fractures in the grain may be related to the infiltration of melt, which imparted heat energy into the lattice, delocalizing electrons in the zircon. If left undisturbed it is likely that zircon along these fractures would have begun decomposing, but the process was halted by the quenching of the impact melt. As with the LRS data, we do not observe CL features in the zircon core that indicate structural lattice changes due to shock processes, suggesting Microsc. Microanal. 21 (Suppl 3), 2015 1448 that the grains were only affected by the high heat of impact melt immersion. The lack of shock alteration in the core is also supported by the lack of petrographic shock features and SIMS data that show no resetting of the age of the grains (original crystallization age of 1417 � 22 Ma (within analytical uncertainty of 1451 � 12 Ma; [2]). Age dating of the zircon core and recently completed EBSD work will be discussed at the meeting. References: [1] El Goresy (1965) Baddeleyite and its significance in impact glasses. J. Geophys. Res. 70, 3453�3456. [2] Marion and Sylvester (2010) Composition of anorthositic impact melt at Mistastin Lake Crater. P&SS, 58, 4, 552-573. [3] Butterman and Foster (1967) Zircon stability and the ZrO 2-SiO2 phase diagram. Amer. Mineral. 52, 880-885. [4] McFarlane (2014) Trace element characterization of CL response in zircon by LA-ICP-MS, in Cathodoluminescence Application to Geoscience, Mineral. Assoc. Canada, 45, 169-182. [5] Kempe et al. (2000) Relevance of CL for the interpretation of U-Pb zircon ages, In Cathodoluminescence in Geosciences; Springer; 415-455. [6] Nasdala et al. (2003) Spectroscopic methods applied to zircon; in Rev. Mineralogy & Geochemistry, 53; 427-467. Figure 1: (left) False color RGB CL image overlain on a BSE image of partially decomposed zircon in impact melt glass. (Right) WDS spot-analysis traverse (yellow points) showing concentrations of selected REE CL activator and quenching elements. Figure 2: Hyperspectral CL data collected from various spot analyses. (left column) BSE images with spot analyses (yellow dots). Top row: decomposition interface (blue rim in Fig 1a); middle row: bright CL (white in Fig 1a); bottom row: intermediate CL (salmon color in Fig 1a). (center column) deconvolved spectra for spot analyses as a function of energy (eV). Broad humps may represent "intrinsic" CL, and sharp peaks are related to specific REE activation. (left column): same spot spectra as a function of wavelength.");sQ1[723]=new Array("../7337/1449.pdf","Quantitative Electron Microscopy and the Application by Single Electron Signals","","1449 doi:10.1017/S1431927615008028 Paper No. 0723 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantitative Electron Microscopy and the Application by Single Electron Signals Ryo Ishikawa1, Andrew. R. Lupini2, Scott D. Findlay3, Takashi Taniguchi4, and Stephen J. Pennycook4 1. 2. Institute of Engineering Innovation, University of Tokyo, Bunkyo, Tokyo, Japan. Materials Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN, USA. 3. School of Physics, Monash University, Victoria 3800, Australia. 4. Advanced Key Technologies Division, National Institute for Materials Science, Tsukuba, Japan. 5. Department of Materials Science and Engineering, University of Tennessee, Knoxville, TN, USA. Annular dark field (ADF) imaging in scanning transmission electron microscopy (STEM) is an ideal mode to quantitatively interpret images: the ADF intensity monotonically increases as a function of specimen thickness owing to the incoherent nature. ADF detector calibration combined with frozen phonon image simulations has demonstrate excellent agreement between experiment and simulation [1], which is now actively used in quantitative analysis. Recently we also developed our high-accurate quantitative method by using single electron signals, enabling us to measure some basic physical quantity such as the number of atom per the column [2]. Moreover, it becomes possible to identify the three-dimensional location of a single dopant in bulk materials on the basis of intensity comparison between experiment and simulations [3]. To establish high accurate quantitative measurement, it is of critical importance to understand possible sources of the background signals. Electronic noise can be measureable by acquiring images with the electron beam off. When we turn on the electron beam in a vacuum, distinct increment of the signals is observed in background images, suggesting that a few stray or `accidental' electrons are impinging on the ADF detector. In addition to a single peak related to DC offset, new discrete peaks are observed with a constant interval in background images (beam on). As a systematic analysis, we developed the method to correct background signals and moreover it becomes possible to estimate the signal level of a single electron. Figure 1 shows the histogram of corrected background image, and discrete peaks appear at the integer number of electrons. Therefore it becomes possible to convert the ADF image with the electron number unit. On the basis of this analysis, we measured specimen thickness with high accuracy (� 1 atom precision for thin samples) and also determined the inelastic mean free path combined with electron energy-loss spectroscopy. Once we know the number of atoms per the column, we may also determine the other physical quantities such as absolute defocus value and an effective source size. Fig. 2(a) show ADF STEM image of w-AlN viewed along the [1120] direction, where the column contains 15 atoms. To compensate the spatial incoherence of the probe, we assume a Gaussian source and the effective source size is estimated to be 0.62 � through chi-square test along X-X', shown in Fig. 2(b). By convolving the source size with simulated image (Fig. 2(c)), the experimental image is well reproduced, as most evident in Z-contrast profile of Fig. 2(d). Using those quantities, it allows us to analyze image shape of the atom and becomes possible to access three-dimensional information such as the location of a single dopant. Microsc. Microanal. 21 (Suppl 3), 2015 1450 References [1] J.M. LeBeau et al, Phys. Rev. Lett. 100 (2008) 206101. [2] R. Ishikawa et al, Microsc. Microanal. 20 (2014) 99. [3] R. Ishikawa et al, Nano Lett. 14 (2014) 1903. [4] R.I. acknowledges support from JSPS Postdoctoral Fellowship for Research Abroad and Prof. Naoyoa Shibata and Prof. Yuichi Ikuhara (University of Tokyo) for helpful discussions. A.R.L. acknowledges support by the U.S. Department of Energy, Basic Energy Sciences, Materials Sciences and Engineering Division. S.D.F. acknowledges support under the Discovery Projects funding scheme of the Australian Research Council. R.I. and T.T. acknowledge support by a Grant-in-Aid for Scientific Research on Innovative Areas "Nano Informatics" (Grant 25106006) from JSPS. integration G = 218, T = 32 zeroth peak single electron double electron triple electron Figure 1. The histogram of the corrected background image, where the horizontal axis is the number of electrons. The overlaid profiles are estimated based on Poisson distribution for the discrete events of single, double and triple electrons. 2% a experiment b Al simulation X N 0 X X' 0.5 X' experiment simulation c Fractional ADF signal (%) d 0.8 0.4 Chi square (%) 0.6 0.3 0.2 0.4 0.1 0.2 0.0 0.3 0.4 0.5 0.6 0.7 0.8 Effective source size (�) 0.9 0.0 X 0.0 0.5 1.0 1.5 Distance (�) X' 2.0 Figure 2. (a) Atomic-resolution ADF STEM image of w-AlN viewed along a-axis (15 atom thickness), where the intensity range is 0 � 2 % of that in the incident electron beam. (b) Simulated ADF STEM image convoluted with a 0.62 � Gaussian profile. (c) Chi-square test profile as a function of FWHM Gaussian source size. (d) Z-contrast profiles along X-X' for experiment (circles) and simulation (solid line). The scale bar in (a) is 2 �.");sQ1[724]=new Array("../7337/1451.pdf","Large-Scale Molecular Dynamics and High-Resolution Transmission Electron Microscopy Study of Graphene Grain Boundaries","","1451 doi:10.1017/S143192761500803X Paper No. 0724 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Large-Scale Molecular Dynamics and High-Resolution Transmission Electron Microscopy Study of Graphene Grain Boundaries Colin Ophus1, Haider Rasool2,3, Alex Zettl2,3, and Ashivni Shekhawat3 1. National Center for Electron Microscopy, Molecular Foundry, Lawrence Berkeley National Laboratory, Berkeley, USA 2. Department of Physics, University of California Berkeley, Berkeley, USA 3. Materials Science Department, Lawrence Berkeley National Laboratory, Berkeley, USA Graphene is a promising material for various technological applications, due to its excellent electrical and [1] mechanical properties [2]. New graphene deposition methods are continually increasing the maximum fabricated sheet size of a single crystallographic orientation, but most deposition methods still produce polycrystalline sheets, containing grain boundaries (GBs) [3]. Graphene GBs are very interesting scientifically due to the two-dimensional nature of graphene. Bulk three-dimensional materials require 5 angles to characterize the macroscopic degrees of freedom, while two-dimensional materials require only 2 angles: the misorientation angle M, defined as the angle between the unit cell vectors of each grain, and the boundary line direction L, defined as the angle between the boundary vector and the symmetric tilt boundary vector. Thus the parameter space for 2D GBs is far smaller than that of 3D grain boundaries. With modern computer simulation methods, it is now possible to enumerate this entire parameter space in a very fine-grained manner. In this study, we have combined phase-contrast high resolution transmission electron microscopy (HRTEM) observations of free-standing graphene GBs with molecular dynamics (MD) simulations spanning the entire orientation parameter space of graphene GBs. We have collected hundreds of observations of GB structure using HRTEM images, and dozens of higher-precision measurements using complex exit wave reconstruction (EWR) of HRTEM focal series [4]. Figure 1 shows lattice strain measurements of a graphene GB performed on an EWR dataset taken from our study in [5]. The experimentally-determined structures are compared to over 40 000 MD simulations of graphene GB structures, with steps of 0.5� over M and L. The structures were generated by fist constructing rational approximates of tiled graphene cells with the desired grain angles, periodic along the boundary line. Next the atomic coordinates were updated using a simple geometric algorithm to generate plausible sp2bonded structures. Finally the structures were relaxed using molecular dynamics. Figure 2 shows a subset of the calculated structures. From these simulated boundaries, we can evaluate many physical and energetic parameters; some examples are given in Figure 3. We can see that the GB enthalpy is very closely related to the dislocation density of the GB. In this talk we will examine various measurable parameters and compare them directly to experiment, with the aim of creating a predictive model of graphene GB structure, stability and resistance to deformation or fracture. References [1] OV Yazyev and SG Louie, Nature Materials 9 (2010) p. 806 [2] Y Wei, J Wu, H Yin, X Shi, R Yang and M Dresselhaus, Nature Materials 11 (2012) p. 759 [3] Haider deposition paper. [4] H Rasool, C Ophus, WS Klug, A Zettl and JK Gimzewski, Nature Comm. 4 (2013) p. 2811 [5] HI Rasool, C Ophus, Z Xiang, MF Crommie, BI Yakobson, and A Zettl, Nanoletters 14 (2014) p. 7057 [6] The authors acknowledge the financial support of the Office of Science, Office of Basic Energy Sciences of the US Department of Energy under contract number De-AC02-05CH11231. Microsc. Microanal. 21 (Suppl 3), 2015 1452 Figure 1. EWR image reconstructed from HRTEM focal series (A), showing the structure of a high angle graphene GB (B). Strain along x-axis shown from -1% (blue) to +1% (red) plotted in (C). Scale bar length is 1 nm, data and analysis in this figure is from [5]. Figure 2. Graphene grain boundaries computed with molecular dynamics for misorientation angles of 4, 8, 16, 27, and 30, and boundary line angles of 0, 5, 10, 15, 25 and 30. Red pentagons and blue heptagons show 5-atom and 7-atom carbon rings respectively, black line traces boundary. Figure 3. Measurements of graphene GB enthalpy, Wulff shape, excess length, and dislocation density simulated with MD. Enthalpy vs dislocation density and excess length are compared.");sQ1[725]=new Array("../7337/1453.pdf","Applications of Bicrystallography: Revealing Generic Similarities in Coincidence Site Lattice Boundaries of all Holohedral Cubic Materials and Facilitating the Design of 3D Printed Models of such Grain Boundaries","","1453 doi:10.1017/S1431927615008041 Paper No. 0725 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Applications of Bicrystallography: Revealing Generic Similarities in Coincidence Site Lattice Boundaries of all Holohedral Cubic Materials and Facilitating the Design of 3D Printed Models of such Grain Boundaries Peter Moeck1, Andrew Maas1, Jennifer Stone-Sundberg1,2, Bryant York3, Trevor Snyder1,4, Werner Kaminsky5, and Nigel Browning6 1. 2. Nano-Crystallography Group, Department of Physics, Portland State University, Portland, OR 97207-0751 Crystal Solutions, LLC, Portland, OR 97205 3. Department of Computer Science, Portland State University, Portland, OR 97207-0751 4. 3D Systems Corporation, Wilsonville, OR 97070 5. Department of Chemistry, University of Washington, Seattle, WA 98195 6. Chemical and Materials Sciences Division, Pacific Northwest National Laboratory, Richland, WA 99352 The application of bicrystallography [1-3] makes structural units [4-6] and dislocations [7] superfluous as descriptors of grain boundaries. All there is around an ideal (un-relaxed) coincidence site lattice (CSL) grain boundary are predicable atomic positions with certain black-white symmetries. Free energy minimization driven relaxations of such a hypothetical grain boundary structure may lead to the breaking of some or all of the black-white symmetries, just as monochrome space group symmetries may be broken locally in the two real single crystals that make up the bicrystal. Any relaxed (real) atomic position will, however, be very close to the predicted un-relaxed (ideal) atomic position as long as that position remains occupied. Grain boundary structures with nine degrees of freedom as predicted by bicrystallography in three dimensions (3D) are, therefore, ideal as atomic level starting structures for free energy minimization calculations! Bicrystallography in two dimensions (2D) allows for both visualizations of edge-on projections of CSL tilt boundaries [3] and the extension of the atomic column indexing procedure of ref. [8] to all experimental 2D images of such boundaries. In order to illustrate the predictive power of bicrystallography in 2D briefly, we reproduce below in Figure 1 migration related segments of translation averaged Z-STEM images of a 13a (510) tilt boundary in SrTiO3 in [001] projection from ref. [3] (with permission from the publisher). A range of CSL [001] tilt boundaries in CeO2 were imaged with an aberration corrected Z-contrast Scanning Transmission Electron Microscope (Z-STEM) and striking visual similarities of their (projected) structural units to those of (analogous orientation relationship) grain boundaries in face-center cubic metals, yttriastabilized zirconia, and SrTiO3 (Fig. 1) were recently noted in ref. [4]. There is, however, some arbitrariness in the choosing of structural units [7], despite their emergence in connection with free energy minimization calculation [5] and their identification as cores of grain boundary dislocations [6]. In essence, they are arrangements of atoms from both sides of a grain boundary with geometries that can be chosen "simply for convenience" [7]. The predictive power of structural units is, accordingly, limited to series of grain boundaries in the same material when only one of the five macroscopic degrees of freedom is varied [7]. For predictions of the structures of tilt boundaries on the basis of structural units, the parameter that varies in a series is most often the tilt angle, e.g. see ref. [4]. Our bicrystallography analysis, on the other hand, shows that these structural similarities are simply byproducts of the edge-on projection of the black-white layer groups of the bicrystals in these materials [3]. In short, whenever B�rnighausen trees [9] reveal structural relations (i.e. similarities) between the space groups of different materials, there will (per bicrystallography) also be predicable structural similarities in their analogous orientation relationship grain boundaries. The bicrystal layer groups of all CSL [001] tilt boundaries in the above mentioned (cubic holohedral) materials project along [001] to black-white frieze symmetries p11m' (Figs. 1a and 1b), p11g' (Figs. 1c and 1d), and p111' [3] so that we can use simple sketches Microsc. Microanal. 21 (Suppl 3), 2015 1454 of 2D projections of CSL [001] tilt boundaries to reveal the above mentioned structural similarities visually. Such sketches have been produced with the popular drawing program GIMP on the basis of both our 2D bicrystallography procedure [3] and supporting MATLAB code outputs that generate [001] projections of the structure of the single crystal components of cubic holohedral bicrystals [10]. We also produced 3D printed hand-held models of CSL [001] tilt and twist boundaries in some of the above mentioned materials in order to demonstrate their generic structural similarities at the atomic level. The Cif2VRML program [11] was used for the design of these models. This program reads atomic coordinates of grain boundary structures in the very well documented Crystallographic Information File (CIF) format [12] and converts them directly into 3D print file formats. In order to provide a concise structure description in the corresponding CIFs, the bicrystal layer group symmetry of a grain boundary can stand in for the space group symmetry of the component crystals and the CSL parameters can stand in for these crystals' unit cell parameters. In the CIFs that encode the structure of the highest symmetric grain boundary models, only one eighth of the atoms that make up the 3D printed models (i.e. the content of the "asymmetric black-white unit" of the CSL unit cell) need then be included explicitly. Due to the inherent conciseness of these kinds of "bicrystal layer group/CSL parameter" CIFs, we propose to make them the new standard for the encoding of atomic level starting structures for free energy minimization calculations of grain boundaries [13]. [1] R. C. Pond and W. Bollmann, Proc. Royal Soc. London A 292 (1979) 449. [2] R. C. Pond and D. S. Vlachavas, Proc. Royal Soc. London A 386 (1983) 95. [3] P. Moeck et al, Cryst. Res. Technol. 49 (2014) 708. [4] W. Tong et al, Acta Mater. 61 (2013) 3392. [5] A. P. Sutton and V. Vitek, Phil. Trans. Royal Soc. London A 309 (1983) 1, 37, 55. [6] N. D. Browning et al, In: Modeling Nanoscale Imaging in Electron Microscopy, eds. T. Vogt et al, (Springer, 2012). [7] L. Priester, Grain Boundaries From Theory to Engineering, (Springer, 2013). [8] X. Sang et al, Microsc. Microanal. 20 (2014) 1764. [9] U. M�ller, Symmetry Relations between Crystal Structures, (Oxford University Press, 2013). [10] A. Maas, Capstone project, http://nanocrystallography.research.pdx.edu/media/andrew-capstone.pdf. [11] W. Kaminsky et al, Powder Diffraction 29 (2014) S42. The program is freely downloadable at http://cad4.cpac.washington.edu/Cif2VRMLHome/Cif2VRML.htm. [12] http://www.iucr.org/resources/cif [13] The authors acknowledge support from the National Science Foundation (grant EEC-1242197), 3D Systems Corporation, and the U.S. National Committee for Crystallography. Figure 1. Edge-on projections of migration related segments of a 13a (510) tilt boundary in SrTiO3 (after ref. [3]). a and c, as predicted by bicrystallography in 2D, b and d as experimentally observed by Z-STEM imaging and translation averaging. The marked mixed Ti/O and Sr columns in a and c must be in energetically unfavorable positions because their counterparts are (in essence) not occupied in b and d. Note that the black-white symmetries in the bicrystallography predictions (a,c) are not broken (at least on average) by atomic level free energy minimization effects that show up in the experiments (b,d). 2 nm c modified after P. Moeck et al., Cryst. Res. Technol. 49 (2014) 708 d");sQ1[726]=new Array("../7337/1455.pdf","Defect Character at Grain Boundary Facet Junctions: A Combined HRSTEM and Atomistic Modeling Study of a =5 Grain Boundary in Fe","","1455 doi:10.1017/S1431927615008053 Paper No. 0726 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Defect Character at Grain Boundary Facet Junctions: A Combined HRSTEM and Atomistic Modeling Study of a =5 Grain Boundary in Fe D.L. Medlin1, K. Hattar2, J. Zimmerman1, F. Abdeljawad2, and S.M. Foiles2 1 Sandia National Laboratories, Livermore, California 94551, USA 2 Sandia National Laboratories, Albuquerque, New Mexico 87185, USA Electron microscopy plays an important role in motivating and testing our theoretical understanding of grain boundary structure and behavior across the atomistic and continuum length scales. One of the key foundational challenges in grain boundary science is to establish meaningful links between atomic-scale interfacial configurations, which can often be described in terms of sets of characteristic structural units [1], and more macro-scale descriptors of the interfacial geometry, such as grain misorientation and boundary inclination. A useful approach for establishing such linkages is to identify and to characterize the sets of interfacial line defects that accommodate departures from low-energy singular reference configurations [2]. In this presentation, we consider such approaches in the context of atomic resolution experimental observations and theoretical calculations of =5 <001> tilt boundaries in BCC iron. We focus in particular on the analysis of facet junctions and grain boundary dislocations. Figure 1a shows a high angle annular dark field (HAADF) scanning transmission electron micrograph (STEM) of such a boundary observed in an annealed pulsed-laser-deposited thin film of Fe. The boundary has formed nanoscale facets lying on {310} and {210} planes, which correspond to the two types of symmetric inclination that are geometrically possible for =5 <001> tilt boundaries. We have conducted atomistic simulations for these inclinations using several interatomic potentials and density functional theory and have compared these calculations with the observed structures. Examples showing structures calculated using the potential of Mendelev et al. [3] are shown in Figures 1b and 1c. Our calculations predict these inclinations to be cusps in the energy versus inclination landscape, which is consistent with our observation of faceting on these planes. Measurements of the boundary misorientation show a small deviation (2.4�) from the exact =5 alignment. By analyzing the boundary dislocation content (Figure 2) we have shown that this angular deviation is accommodated by an array of (1/5)[3,1,0] and (1/5)[1,2,0] dislocations distributed at the facet junctions in a manner consistent with the observed average inclination. These atomistic observations help in challenging our thinking about the interplay between dislocation and boundary faceting energetics. To illustrate, we will discuss how this analysis is being used to motivate and to refine a continuum phase-field approach for modeling the effects of junction dislocation character on the energetics and kinetics of grain boundary facet coarsening. [1] A.P. Sutton and V. Vitek, Phil. Trans. Royal Society A 309 (1983) 1-68. [2] R.C. Pond, in Dislocations in Solids (Ch. 38) (Ed. F.R.N. Nabarro, Elsevier, 1989). [3] M.I. Mendelev, et al., Philosophical Magazine 83 (35) (2003) 3977-3994. [4] C.L. Kelchner, S.J.Plimpton, J.C. Hamilton, Phys. Rev. B 58 (1998) 11085. [5] Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000. Microsc. Microanal. 21 (Suppl 3), 2015 1456 Figure 1. (a) HAADF-STEM image of a =5 [001] boundary in Fe. The boundary has formed nanoscale facets on {310} and {210} type planes. Figures (b) and (c) show atomic structures for the ideal =5 {310} and {210} boundaries as calculated using the Mendelev potential [3] and shaded by the centro-symmetry parameter [4]. Figure 2. (a) The dislocation content at the facet junctions was determined by circuit mapping. Here the circuit is separated into two paths, C! and C! . The Burgers vector is calculated from = -(C! + C! ), where is a matrix transformation re-expressing path C! into the crystal coordinates of crystal . The analysis identifies grain boundary dislocations that are located at the facet junctions and possess Burgers vectors of type (1/5)[3,1,0] (b) and (1/5)[1,2,0] (c). These Burgers vectors are plotted here on the =5 dichromatic pattern, which shows the superposition of the two BCC crystal lattices.");sQ1[727]=new Array("../7337/1457.pdf","Texture and Phase Analysis in Nanocrystalline Ni Thin Films by Precession Electron Diffraction Microscopy","","1457 doi:10.1017/S1431927615008065 Paper No. 0727 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Texture and Phase Analysis in Nanocrystalline Ni Thin Films by Precession Electron Diffraction Microscopy Szu-Tung Hu1, Lauren Morganti1, Shreyas Rajasekhara2*, Khalid Hattar2, Paulo Ferreira1 Materials Science and Engineering, University of Texas at Austin, Austin TX, USA Sandia National Laboratories, Albuquerque, NM, USA *Now at Intel, Hillsboro, OR, USA 2 1 Grain refinement is a powerful method to obtain high strength in metals and alloys. However, in materials with nano grain sizes, rapid grain growth may occur due to the high curvature of grain boundaries In addition, as grain growth occurs, the local texture may evolve, which affects the mechanical properties. Therefore, to achieve nanocrystalline metal films with both high strength and ductility it is imperative to correlate the film thickness with grain size and local texture as a function of annealing. In this work, nanocrystalline Ni thin films with thicknesses of 30 nm and 120 nm were subjected to an annealing treatment at 350 C for 20-80 minutes. To identify the average grain size, grain orientation and phase fraction, precession electron diffraction was used. However, the reliability of the precession analysis is sometimes low due to the contribution of various grains to the diffraction pattern (DP). To improve the analysis, a two-step method was used where 1) the DP is filtered for noise threshold, spot enhance loop, gamma, spot radius and softening loop [1], and 2) the indexed DP is subjected to a reliability threshold of 15%. In this fashion, the reliability was improved by 5-10%. The results show the presence of fcc and hcp phases (Fig. 1) in all samples [2]. In terms of grain orientation, the annealing treatment promotes the formation of [100] texture in the 30nm films, and a near [111] texture for the 120nm film (Fig. 2). In regards to grain size, the annealing treatment leads to grain growth in all samples, although pockets of very small grains remain stable and do not grow (Fig. 3). References [1]A. Kobler et al. Ultramicroscopy, 128 (2013), p.68-81 [2] S. Rajasekhara et al. Scripta Materialia, 67. (2012), p.189-192 [3] Acknowledgments: This work was fully supported by the Division of Materials Science and Engineering, Office of Basic Energy Sciences, US Department of Energy. Sandia National Laboratories is a multi-program laboratory operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Company, for the US Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000. Microsc. Microanal. 21 (Suppl 3), 2015 1458 (a) (b) Fig. 1 Phase maps combined with reliability maps for samples of thickness (a) 30nm and (b) 120nm. The presence of fcc and hcp phases in both samples is evident. (a) (b) (c) (d) (e) (f) (g) (h) Fig. 2 Orientation maps combined with reliability maps for 30nm films annealed for (a) 20min, (b) 40 min, (c) 60 min and (d) 80min and 120nm film annealed for (e) 20min (f) 40 min, (g) 60min and (h) 80 min. (a) (b) Fig. 3 TEM bright field images showing grain growth and pockets of small grains in the 30nm thin films after annealing for (a) 20min and (b) 60min.");sQ1[728]=new Array("../7337/1459.pdf","Application of EBSD and Precession-Enhanced Diffraction (PED) to Study Crystallography of -Titanium Alloy During Transformation under Severe Hot Plastic Deformation","doi:10.1017/S1431927615008077","1459 doi:10.1017/S1431927615008077 Paper No. 0728 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Application of EBSD and Precession-Enhanced Diffraction (PED) to Study Crystallography of -Titanium Alloy During Transformation under Severe Hot Plastic Deformation S.V.Prikhodko1, P.E.Markovsky2, S.D.Sitzman3, M.A.Gordillo4, J.M.K.Wiezorek4, O.M.Ivasishin2 1 Department of Materials Science and Engineering, University of California Los Angeles, Los Angeles, CA 90095, USA 2 G.V. Kurdyumov Institute for Metal Physics, National Academy of Science of Ukraine, 36, Vernadsky Blvd., 03142, Kiev, Ukraine 3 Oxford Instruments America, Concord, MA, 01742, USA 4 Department of Mechanical Engineering and Materials Science, University of Pittsburgh, Pittsburgh, PA 15261, USA Over the past two decades EBSD has become one of the most powerful and widely used techniques for statistically proven microstructural characterization, including for grain size and crystallographic orientation. Since EBSD is a bulk-sample, SEM-based technique, it can analyze sample areas at the centimeter scale, yet is capable of mapping grains as small as a few 10s of nanometers. Still finer grained materials are analyzable with diffraction-based mapping techniques on thinned samples, including by SEM/EBSD-based Transmission Kikuchi Diffraction (TKD) and by TEM-based Precession-Enhanced Diffraction (PED). PED uses spot patterns under pseudo-kinematical diffraction conditions, and has successfully demonstrated phase and orientation mapping in a variety of materials at resolutions down to 3-5nm, allowing analysis of previously inaccessible grains in nanostructured materials [1]. The improved resolution of the transmission techniques also makes them more amenable to higher strain materials, since in general fewer dislocations are present in the interaction volume at any given point; however, they are limited in coverage to the electron transparent region of a thinned sample. Thus, for some materials a combination of methods such as PED and EBSD is especially useful for analysis of characteristics at different length scales. Here we present our recent results on the microstructure and texture of a typical titanium-based metastable -alloy, VT22 (Ti-5.0Al-4.79Mo-4.70V-0.97Fe-0.71Cr, wt. %), processed by severe hot (below 750) plastic deformation. This type of deformation results in a microstructure comprised of elongated, un-recrystallized -phase grains having a sharp axial texture, and fine intragranular precipitates. Experimental details are reported elsewhere [2]. EBSD and PED were used in this study for analysis of the phases involved in the transformation. Both techniques are in good agreement, even over very different sample areas. The Burgers orientation relationship (OR) is fulfilled; however, the realization probability of some OR variants within individual -grains is not evenly distributed between all possible variants. This can be attributed to the stress accommodation taking place during transformation under conditions of severe plastic deformation. [1] J.G.Brons, G.B.Thompson, JOM (2013) DOI: 10.1007/s11837-013-0799-5. [2] O.Ivasishin et al., ASME 2013 IMECE 11/2013; 2A:V02AT02A042 IMECE2013-63767, (2013) 1-10. Microsc. Microanal. 21 (Suppl 3), 2015 1460 A B 001 111 C 0001 1010 -2110 110 Figure 1. PED results on the crystal structure characterization of the phases involved during transformation of VT22 alloy after severe hot plastic deformation. The e-beam is parallel to the rolling direction. Scale bar is 800 nm. A. Virtual bright-field image. B. Texture map for -Ti shows essentially a single orientation for the -phase parent grain, close to {111}. C. Texture map for - Ti shows a variety of grain orientations. A A B B C 1 4 2 6 3 5 Figure 2. Pole figures obtained using PED of the area on the sample shown in Figure 1. A. {110} of -Ti shows predominantly one orientation of -phase parent grain. B. {0002} of - Ti shows that the vast majority of the -crystals have {0001} parallel to {110}. C. Computer simulated pole figure of {110} of -Ti shows that the realization probability of some OR variants within individual -grains is not evenly distributed between all possible variants: variants 1 and 2 are the strongest, 3, 4 and 5 are medium and 6 practically not realized. AA B C Figure 3. Example of EBSD results. A. Orientation map subset of one -grain. Blue is for -Ti and red is for -Ti. The plane of the image is perpendicular to the rolling direction. Scale bar is 10 �m. B. {110} pole figure of -Ti. C. {0001} pole figure of -Ti. {0001} is parallel to {110}. Two of 6 possible orientations of - Ti are not realized strongly.");sQ1[729]=new Array("../7337/1461.pdf","Precession Electron Diffraction and Orientation Phase Mapping of Assembled Ag/ ZnO Nanoantennas.","","1461 doi:10.1017/S1431927615008089 Paper No. 0729 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Precession Electron Diffraction and Orientation Phase Mapping of Assembled Ag/ ZnO Nanoantennas. John E. Sanchez1, Francisco Ruiz Zepeda1, Miguel Jos� Yacaman1, and Arturo Ponce1 1. Department of Physics & Astronomy - The University of Texas at San Antonio, San Antonio, TX - US The manipulation of the geometrical and structural arrangement of the constituent's elements on devices at nanoscale level is highly desirable for a precise monitoring of the opto-electrical properties exhibited for these nanomaterials. In fact, a great effort has being made to understand the coupling mechanisms on metal-semiconductors systems, most precisely at interfaces nanoscale level. For instance, it is well known that the multidirectional radiation pattern generated by the active elements on nanoantenna applications is highly dependent on both the structural and orientation distribution of the receiver elements as well as the passive element on the nanoscale device. For example, Wang, et al [1] have synthesized high order nanostructures in a hierarchical configuration to study the photo-induced optical properties of these systems in function of the ZnO concentration distributed along the silver nanowires. However, few is known about the structural coupling mechanisms between this metal-semiconductor heterojunctions. Thus, to understand the dynamic coupling at the interface level in the Ag/ZnO metalsemiconductor heterojunctions we report the epitaxial growing of zinc oxide nanorods on the pentagonal exposes faces of Ag nanowires resembling a hierarchal nanoantenna. Moreover, the studied of the growth mechanism in the active/contact faces of the metal-semiconductor heterojunction has been done by mapping simultaneously the dynamical electron diffraction pattern under in-situ precession electron diffraction at the heterojunction interface Ag/ ZnO nanosystem. Indeed, by indexing the dynamical diffraction patterns using orientational/phase mapping from the precessed electron diffraction data collected an orientational mapping has been retrieved showing the interfacial growing polar planes (0002) of ZnO nanorods on the pentagonal planes of silver nanowires with a mismatch between planes along the coupling interface. For completeness, grazing angle x-ray diffraction measurements on prepared substrates Ag/ZnO systems shown well-defined peaks associated to the main phases of ZnO nanorods and Ag nanowires respectively. A full understanding of the fit faces mechanism between Ag/ZnO along the mismatch direction undoubtedly will allow elucidating the mechanism through which the contact metal-semiconductor behaves at the heterojunction interface. The figure 1A shows Ag/ZnO metal-semiconductor systems grown by microwave irradiation process as described by Sanchez et al [2]. Two main features could be listed in the micrograph; first a constant distribution of ZnO nanorods covering completely the surface area along the silver nanowire and secondly the multi-pentagonal arrangement of ZnO nanorods running through the long axis of the silver nanowire. In fact, figure 1B and 1C SEM reveal the epitaxial distribution more clearly for a lateral view and plan view respectively. Because of the epitaxial distribution, it was indexed the main planes of the crystalline structures at the interface of Ag/ZnO metal-semiconductor nanosystem using precession electron diffraction. Indeed, figure1 (right) shows a virtual bright field image, for a localized zone at the interface, obtained after applied precession-assisted crystal orientation mapping technique ASTAR [3]. Moreover, using DiffGen tool, the localized precession diffraction patterns were indexed along the interface Ag/ZnO for phase identification and map orientation distribution [4]. Microsc. Microanal. 21 (Suppl 3), 2015 1462 Figure1 (left). A multi-pentagonal arrangement showing ZnO nanorods growing perpendicular to the silver nanowires facets. A) and B) low and high magnification micrographs of an experimental nanosystem as obtained using an STEM microscope. C) oriented Ag/ZnO nanostructure. Figure1 (right) ASTAR orientation map of an isolated Ag/ZnO nanostructure, the corresponding electron diffraction patterns are marked along the interface zone. There, it was inferred a misorientation between (0 -2 0) planes and (0002) at the silver nanowires and ZnO nanorods respectively. Figure 2. Shows the misorientation between Ag nanowires and ZnO nanorods as a function of the distance along the perpendicular direction to the silver wire on the multi-pentagonal distribution. A) lateral view of the Ag/ZnO nanosystem, three zones (green, wine and blue) could be identified clearly on the virtual bright field image on the silver nanowire. B) and C) misorientation at the interface on the pentagonal faces of silver nanowire. D) misorientation at the Ag/ZnO interface of an isolated Ag/ZnO nanosystem. E) Line profile across the whole system of the Ag/ZnO system. References [1] Shaowu Wang, Yang Yu, Yuanhui Zuo, Chunzhong Li, Jinhu Yang and Conghua Lu, Nanoscale 4 (2012), p. 5763-6074. [2] J. E. Sanchez, F. Mendoza Santoyo, J. Cantu Valle, J. Velazquez-Salazar, M. Jose Yacaman, F. J. Gonzalez, R. Diaz de Leon and A. Ponce. J. Appl. Phys. 117, (2015), p. 0343606-2 [3] D. Viladot, M. Veron, M. Gemmi, F. Peiro, J . Portillo, S. Estrade, J . Mendoza, N. Llorca-Isern and S. Nicolopoulos, Journal of Microscopy. 252 (2013), p.23 [4] The authors acknowledge funding from: NSF PREM # DMR 0934218. Scholarship Francisco Jose de Caldas 512 COLCIENCIAS. National Institute on Minority Health and Health Disparities (G12MD007591). Department of Defense #64756-RT-REP and the Welch Foundation Grant Award No. # AX-1615.");sQ1[730]=new Array("../7337/1463.pdf","A Precession Electron Diffraction and EELS Study of Beta-phase Evolution in Nano-crystalline Mg-9 wt.% Al Thin Films during Heat Treatment","","1463 doi:10.1017/S1431927615008090 Paper No. 0730 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A Precession Electron Diffraction and EELS Study of Beta-phase Evolution in Nano-crystalline Mg-9 wt.% Al Thin Films during Heat Treatment Karen Kruska1, Daniel J Edwards1, Rama S Vemuri1, Libor Kovarik1, Aashish Rohatgi1, Nigel D Browning1 1 Pacific Northwest National Laboratory, {FCSD, EED, EMSL}, Richland, WA, United States. Modern transmission electron microscopy (TEM) and spectroscopy techniques routinely probe the microstructure and local compositions in nanocrystalline materials using various analytical techniques such as electron energy-loss spectroscopy (EELS) or energy dispersive x-ray spectroscopy (EDX) maps. Electron diffraction is another technique the can help to identify phases in materials, but in applying such a technique, it is cumbersome to obtain statistically relevant information from nanomaterials. With the development of precession electron diffraction (PED) mapping and the use of image correlation for pattern identification, novel information can be obtained relatively rapidly that covers a much larger number of grains [1]. Combined with elemental mapping, this provides a powerful technique to characterize the distribution of phases and chemical distribution in nanomaterials. The goal of a broader study at PNNL is to understand the evolution of Mg17Al12 -phase precipitates as a function of heat-treatment in an automotive Mg AZ91 (Mg-9 wt.% Al-~1 wt.% Zn) alloy. In this work, EELS and PED on a cold-FEG JEOL ARM200CF TEM were used to investigate the evolution of the phase precipitates in a binary Mg-9 wt.% Al thin film subjected to a heat treatment at 300C for 30 min. 80 nm Mg-9Al thin films were deposited on a 25 nm amorphous Si3N4 support membrane. Heating was conducted in a Gatan single-tilt heating stage. EELS acquisition was performed with a convergence halfangle of 34.4 mrad and a collection half-angle of 82.6 mrad with a dispersion of 0.25 eV. Multivariate statistical analysis (MSA) was used to reduce statistical noise in EELS data [2]. PED data was collected in nanoprobe mode using a probe size of 1.2 nm at 50Hz, a precession angle of 0.69 deg and a step size of 4 to 5 nm. The PED data in Figure 1 shows the -Mg-Al (red), and the -Mg17Al12 (green) that nucleated and grew upon heating from an as-sputtered -Mg-9 wt.% Al film. The black pixels in Figures 1a and b show pixels of high uncertainty [1]. Much higher reliability could be achieved when precession was enabled during mapping. The apparent area fraction of the -phase in Figure 1b is ~8 %. In comparison, the EELS relative composition maps in Figures 2b and c show ~26 % "Al-rich" grains of which majority have an apparent Al concentration between 10-30 %. The stoichiometric concentration of Al in the precipitates of 41 % was only measured in few grains. The maximum solubility of Al in Mg is ~12%, and existence of grains with 12 % < Al < 41 % is not predicted by the equilibrium Mg-Al phase diagram. Given the moderate cooling rate after heat treatment, such non-equilibrium conditions seem unlikely. An alternative interpretation is the formation of -precipitates that do not stretch over the entire thickness of the thin film. Thus, while PED only identifies the strongest diffraction pattern from the two overlapping phases in the beam path, the 2D EELS projection leads to a perceived Al concentration between 12 % and 41 %. Implications of these findings for the treatment of PED data in this set-up will be discussed [3]. References [1] E.F. Rauch, M. V�ron, Materials Characterization 98 (2014), p. 1-9 Microsc. Microanal. 21 (Suppl 3), 2015 1464 [2] [3] G. Lucas, P. Burdet, M. Cantoni, C. Hebert, Micron 52-53 (2013), p. 49-56 This work was sponsored by the Vehicle Technologies Office of the U.S. Department of Energy. The utilized PED system was transferred from Portland State University. A portion of the research was performed using EMSL, a national scientific user facility sponsored by the Department of Energy's Office of Biological and Environmental Research and located at Pacific Northwest National Laboratory. Figure 1: PED data showing the phases in Mg-9Al thin film. a) Phase map acquired without precession. b) Phase map acquired with precession. Black pixels indicate less than 15% reliability in the indexing. Figure2: EELS data showing the phases in Mg-9Al thin film. a) STEM HAADF image showing the analyzed area b) Mg relative composition map for data reconstructed with 6 Eigenvalues. c) Respective Al relative composition map. d) EELS spectrum showing the poor signal:noise during mapping.");sQ1[731]=new Array("../7337/1465.pdf","Quantitative Phase Analysis of Rapid Solidification Products in Al-Cu Alloys by Automated Crystal Orientation Mapping in the TEM.","","1465 doi:10.1017/S1431927615008107 Paper No. 0731 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantitative Phase Analysis of Rapid Solidification Products in Al-Cu Alloys by Automated Crystal Orientation Mapping in the TEM. K. W. Zweiacker 1, M. A. Gordillo 1, C. Liu 1, J. T. McKeown 2, G. H. Campbell 2, T. LaGrange 3, B. W. Reed 3, J. M. Wiezorek 1 1. 2. University of Pittsburgh, Pittsburgh, PA, USA Lawrence Livermore National Laboratory, Livermore, CA, USA 3. Integrated Dynamic Electron Solution, Inc, Pleasanton, CA, USA There has been significant interest in laser-based-processing methods for the manufacturing of complex components (e.g. additive manufacturing) or post-processing/repair work (e.g. laser welding). Since laser processing tends to result in the formation of rapidly solidified microstructures, it becomes increasingly important to understand the microstructural development under such non-equilibrium conditions. Pulsed-laser-based-processing methods have been foci of investigations on rapidly solidified microstructures. This method has been shown to produce unique microstructures and micro-constituents in metallic thin films at the nanometer-scale [1]. Recently, in-situ dynamical transmission electron microscopy (DTEM) studies on the rapid solidification of hypo-eutectic Al-Cu alloys have identified a modulation of the microstructural features as a function of solid-liquid interface motion [2]. Elucidating crystallographic dependencies during rapidly solidifying crystal growth in multi-phase alloys systems is of fundamental importance in understanding laser-assisted solidification processes [3-5]. However, due to the nm-scale of the resulting microstructural features, acquiring data sets with the appropriate spatial-resolution is challenging with electron backscatter diffraction (EBSD) based orientation image mapping (OIM) in scanning electron microscopes. A promising alternative analytical technique, which offers nm-spatial resolution, is precession electron diffraction (PED) based automated crystal orientation mapping (ACOM) in the TEM. This technique provides statistically significant and representative data sets for mapping of crystal orientations and spatial distribution of crystalline phases for instance. Here we report an investigation of the microstructural development in a pulsed-laser melted rapidly solidified thin film alloy of Al-3at. % Cu. Under equilibrium conditions this hypoeutectic Al-alloy contains the face-centered cubic -Al matrix and tetragonal -Al2Cu precipitate phases. Solid-liquid interface velocities obtained via in-situ DTEM experiments as high as 1.5 m/s are reached during solidification. These high interface velocities produce microstructures that deviate from those expected under more conventional conditions. Figure 1a is a collection of bright field TEM images of a representative melt-pool produced from laser irradiation exhibiting the different morphological zones arising during rapid solidification, e.g. Heat-affected zone, Columnar grain growth, Banded microstructure. Figure 1b is an expanded view of the melt-pool edge region, which shows the columnar growth zone with fine-scale copper rich precipitates, identified as the metastable Al2Cu (') phase. The crystallographic orientation relationships (OR) of Al matrix and precipitates were determined by ACOM (e.g. Figure 2). The Al grains are aligned such that the <100> direction is parallel to the crystal growth , i.e. solidification, direction. A large proportion of the precipitates adopt an OR of {001}' || {110} & <110>' || <100> (Fig. 2b), and the minority adopt the expected `cube' OR of {001}' || {100} & <100>' || <100> (Fig. 2c). However, many precipitates do not exhibit any low-index OR with the Al matrix. The statistically significant analysis shows clearly deviations from the ORs characteristic of Microsc. Microanal. 21 (Suppl 3), 2015 1466 solid-state transformation products for this system, which are strongly affected by interfacial strain energy, implying here that matrix and precipitate crystals formed concomitantly from the liquid. References: [1] Kline J.E., Leonard J.P., Applied physics letters 86 (2005), p.201902 [2] McKeown J.T. et al., Acta Materialia 65 (2014), p.56 [3] Gill S.C. et al., Acta Metallurgica Materiala 40(1992), p.2895. [4] Gill S.C., Kurz W., Acta Metallurgica Materiala 41(1993), p.3563. [5] Kurz W, Gilgien P., Materials Science and Engineering A 178(1994), p.171. [6] Work was performed under the auspices of the U.S. Department of Energy, Office of Basic Energy Sciences, Division of Materials Sciences and Engineering for FWP SCW0974 by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. The research activities at the University of Pittsburgh received support from the National Science Foundation, Division of Materials Research, Metals & Metallic Nanostructures program through Grant No. DMR 1105757. Figure 1: a) BF TEM images of a representative melt-pool produced by laser irradiation, b) HAADF STEM image obtained from the melt-pool edge showing the finely separated 'Al2Cu precipitates (bright) within the columnar Al grains (gray) formed during rapid crystal interface movement. -Al a b '-Al2Cu 800 nm c Figure 2: Representative PED OIM data obtained from an Al-3at.% Cu alloy: a) phase map revealing the distribution of precipitates, red and green denote the Al matrix and ' phase, respectively; (b) and (c) selected pole figures centred on a (001)' pole from the right and left Al grains, respectively.");sQ1[732]=new Array("../7337/1467.pdf","Electron Energy-Loss Spectroscopy of Organic Photovoltaics.","","1467 doi:10.1017/S1431927615008119 Paper No. 0732 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electron Energy-Loss Spectroscopy of Organic Photovoltaics. Frank J. Scheltens1, Lawrence F. Drummy2, Michael F. Durstock2, James B. Gilchrist3, Sandrine Heutz3 Jessica A. Alexander1, and David W. McComb1 1. Center for Electron Microscopy and Analysis, Department of Materials Science and Engineering, The Ohio State University, Kinnear Road, Columbus OH 43212, United States 2. Materials and Manufacturing Directorate, Air Force Research Laboratory, WPAFB, Ohio 45433, United States 3. Department of Materials, Imperial College London, SW7 2AZ, United Kingdom Advances in organic photovoltaic (OPV) based solar cell device technology have increased power conversion efficiencies (PCE) beyond 11% [1], pushing flexible architecture OPV devices closer to being a viable low-cost, environmentally friendly alternative to contemporary inorganic based solar cells [2, 3]. Extending OPV performance beyond this limit is a critical challenge that requires better understanding of the PCE limiting processes. Since only photo-generated excitons that diffuse to the interface between electron donor and acceptor materials can dissociate into holes and electrons, understanding the chemistry and molecular structure of this interface is critical to identifying and mitigating these limitations. Other factors such as the amount of light absorption, efficiency of photogeneration of electrons and holes, and their collection efficiency at the respective electrodes must also be optimized in order to improve the device PCE. Electron energy-loss spectroscopy (EELS) is an extremely useful tool that can be used to probe the nature and structure of these interfaces and further the understanding of processes that occur there. In this contribution we describe results from 30 nm thick films of copper phthalocyanine (CuPc), templated by a 5 nm layer of 3,4,9,10-perylene tetracarboxylic dianhydride (PTCDA) grown on potassium chloride (KCL) substrates by evaporation in a Kurt J. Lesker high vacuum chamber with a base pressure about 10-8 mbar and a growth rate of 1 �s-1. The CuPc films were floated off the KCl substrates with deionized water and collected onto lacey carbon coated grids. Monochromated low loss EELS was performed on these films using an FEI Titan 60-300 transmission electron microscope operating in scanning transmission electron microscopy (STEM) mode at 60 kV. EELS data was collected using a Gatan GIF Model 966 Quantum ERS imaging filter and then processed and analyzed using the Kramers-Kronig implementation within Gatan Digital Micrograph software. A monochromated low loss EELS spectrum collected from a plan view CuPc thin film was processed to remove the zero loss peak (ZLP) and then corrected for plural scattering. The resulting single scattering distribution, S(E), shown in Figure 1a, displays a number of features between the energy loss values of 1 eV to 10 eV, corresponding to the optical range of the solar spectrum. The extraction of the S(E) from the collected valence loss data is not a trivial process. Substantial issues exist that are related to the proper removal of the monochromated ZLP from the low loss EELS spectra. After performing a Kramers-Kronig analysis of this S(E), the resulting imaginary part of the complex dielectric function, shown in Figure 1b, reveals single electron transitions and collective excitations of the CuPc in the same solar energy range. This nano-localized EELS based imaginary dielectric function measurement correlates well with bulk light optical based ellipsometry measurements shown in Figure 1c. Microsc. Microanal. 21 (Suppl 3), 2015 1468 References: [1] Chen, C.-C. et al., Advanced Materials 26 (2014), 5670. [2] C.J. Brabec, N.S. Sariciftci, et al., Advanced Functional Materials 11 (2001), 15. [3] N.S. Lewis, Science 315 (2007), 798. Single)sca<ering)distribu7on)S(E)) Intensity)(a.u.)) a) Imaginary)(2))part)of)the)dielectric)func7on)) Imaginary)Dielectric) b) 14" Imaginary"Dielectric" 12" 10" 8" 6" 4" 2" 0" 0" 5" Ellipsometry)Data) Measured" Da>a"(2008)" c) Energy"(eV)" 10" 15" 20" Figure 1. Low loss EELS of evaporated CuPc film: a) Single scattering distribution S(E) ; b) Imaginary (2) part of the dielectric function calculated from S(E); c) Imaginary (2) part of the dielectric function from ellipsometry.");sQ1[733]=new Array("../7337/1469.pdf","Vibrational phonon spectroscopy of boron nitride polymorphs: a comparison between theory and experiment","","1469 doi:10.1017/S1431927615008120 Paper No. 0733 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Vibrational phonon spectroscopy of boron nitride polymorphs: a comparison between theory and experiment R. Nicholls1, F.S. Hage2, J. Yates1, D. McCulloch3, D.M. Kepaptsoglou2, T.C. Lovejoy4, N. Dellby4, O.L. Krivanek4, K. Refson5,6 and Q.M. Ramasse2 1 2 3 4 5 6 Department of Materials, University of Oxford, Parks Road, Oxford OX1 3PH, U.K. SuperSTEM Laboratory, STFC Daresbury Campus, Keckwick Lane, Daresbury WA4 4AD, U.K. School of Applied Sciences, RMIT University, Melbourne VIC 3001, Australia Nion Company, 11511 NE 118th St., Kirkland, WA 98034, U.S.A. STFC Rutherford Appleton Lab., Harwell Science and Innovation Campus, Didcot OX11 0QX, U.K. Department of Physics, Royal Holloway, University of London, Egham TW20 0EX, U.K. The increased energy resolution of a new generation of electron microscopes has made exploring the phonon region of the EELS spectrum possible [1]. To fully interpret the extra information available it is important to be able to compare the data to theoretical calculations so that we can extract physically meaningful results. Here we use different polymorphs of boron nitride and in particular the cubic (c-BN) and hexagonal (hBN) phases, as well as other BN-containing complex engineering materials, to compare experimental phonon spectroscopy results in the scanning transmission electron microscope to theoretical calculations. Both cubic and hexagonal boron nitride samples were prepared by simple drop casting onto lacey carbon grids. Imaging and EELS core loss and phonon spectrum imaging were carried out on the different BN polymorphs using a monochromated Nion UltraSTEM100MC `HERMES' microscope [2] installed at the SuperSTEM Laboratory. Atomic resolution images and core loss chemical maps of the regions of interest were obtained in order to fully characterize them prior to phonon spectroscopy. For the high energy resolution experiments, the instrument was subsequently adjusted to provide a zero-loss peak (ZLP) full width at half-maximum (FWHM) of approximately 15 meV for a probe size of 2.5 . The microscope was operated at 60 kV in order to minimize any damage to the samples. Typical phonon spectra are shown in figure 1: all spectra were normalized to the maximum of the zeroloss peak (ZLP) to account for total signal variations as the probe was moved into the rather thick BN flakes. Several very striking features can be observed. For h-BN, the main phonon peak appears at 183 meV (note that the difference with previously reported values [1] could be due flake size effects or to slight energy calibration differences). Interestingly, a small energy shift is observed as the probe moves from aloof geometry into the sample. The FWHM of the peak also increases within the sample. By contrast, the main phonon for c-BN appears at a lower energy of 147 meV. A small shoulder at ~100 meV appears only when the beam goes through the sample: its origin is being investigated. Phonon modes are a property of the material and several techniques exist to calculate them including finite-displacement methods and perturbation theory. The intensity with which the modes appear in an experimental spectrum depends on the type of experiment and experimental set up. Several ab initio codes are able to compute phonon modes in solids together with IR and Raman intensities. We use the density functional theory code CASTEP [3] to calculate the phonon modes for the different BN polymorphs using a linear response method [4] and compare these to the experimentally observed spectra. One of the in-plane optical modes of h-BN is shown in Figure 2. Microsc. Microanal. 21 (Suppl 3), 2015 1470 1460 References [1] O.L. Krivanek, T.C. Lovejoy, N. Dellby et al., Nature 514 (2014), pp. 209-212. [2] O.L. Krivanek, J.P. Ursin, N.J. Bacon et al., Phil. Trans. Roy. Soc. 367 (2009), pp. 3683-3697. [3] S.J. Clark, M.D. Segall, C.J. Pickard et al., Z. Kristallog. 220 (2005), pp. 567-570. [4] K. Refson, S.J. Clark and P.R. Tulip, Phys. Rev. B 73 (2006), 155114. [5] SuperSTEM is funded by the UK Engineering and Physical Sciences Research Council (EPSRC). Figure 1. Position-resolved low loss EEL spectra taken at the edge and into (a) c-BN and (b) h-BN flakes. HAADF images, inset, show the exact location of the EELS acquisition with respect to the flake edge. 100 frames of 8ms (resp. 100 frames of 5ms) exposure each were averaged to improve the signalto-noise for the c-BN (resp. h-BN), after careful alignment to their maximum. The full ZLP is shown for comparison, scaled by a factor of 1000 (a) (resp. 500, (b)). Figure 2. Simulated in-plane optical phonon mode for h-BN. Each column of atoms moves as a whole and, within each layer, the carbon-nitrogen bond oscillates.");sQ1[734]=new Array("../7337/1471.pdf","Variable Angle Spectroscopic Ellipsometry and Electron Energy-Loss Spectroscopy","","1471 doi:10.1017/S1431927615008132 Paper No. 0734 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Variable Angle Spectroscopic Ellipsometry and Electron Energy-Loss Spectroscopy Jessica A. Alexander1, Frank J. Scheltens1, Lawrence F. Drummy2, Michael F. Durstock2, James B. Gilchrist3, Sandrine E.M. Heutz3, and David W. McComb1 1. Center for Electron Microscopy and Analysis, Department of Materials Science and Engineering, The Ohio State University, College Road, Columbus OH 43210, United States 2. Materials and Manufacturing Directorate, Air Force Research Laboratory, WPAFB, Ohio 45433, United States 3. Department of Materials, Imperial College London, SW7 2AZ, United Kingdom Electron energy-loss spectroscopy (EELS) in the scanning transmission electron microscope (STEM) has considerable potential for investigation of interfaces in organic photovoltaic (OPV) devices. In particular, the low-loss region of the EEL spectrum can be used to obtain the complex dielectric function of the material. The complex dielectric function in turn allows us to distinguish between single electron transitions and collective excitations. Spatial mapping of the single electron transitions can then be used to learn about the chemistry and bonding in the vicinity of interfaces between the acceptor and donor interfaces in the OPV samples [1]. In order to validate EELS measurements of organic samples, the technique of variable angle spectroscopic ellipsometry (VASE) has been used. In VASE, a light beam is directed onto the sample, and information related to phase shifts and amplitude changes of the incident beam upon reflection off of the sample is collected and analyzed [2]. Similar to EELS, it is possible to determine the energydependent refractive index and extinction coefficient, as well as the energy-dependent real and imaginary dielectric components of the complex dielectric function. Since the source is a light beam, it is possible to assume that the sample undergoes little or no beam damage. As a result, the data collected using VASE can be compared to the data collected using EELS in order to substantiate the results from EELS so that it is possible to determine if the organic samples have undergone any damage in the electron microscope. This analysis has been done for four different organic materials of interest: copper phthalocyanine (CuPc), fullerene-C60, poly(3-hexylthiophene) (P3HT), and [6,6] phenyl C61 butyric acid methyl ester (PCBM). Thin films of CuPc, C60, and PCBM were prepared by thermal evaporation of the organic material onto freshly cleaved rock salt substrates (KCl for CuPc and PCBM; NaCl for C60) using an organic evaporation chamber in an argon-filled glove box. The deposited films were of thicknesses of 50 nm or less so that they would be suitable for imaging and spectroscopy in the scanning transmission electron microscope (STEM). For ellipsometry measurements, these films were left as prepared on the rock salt substrates, but for EELS measurements, the thin films were floated off by dissolution of the substrates in distilled water and collected on lacey carbon coated TEM grids. Figure 1 shows TEM images of CuPc and C60 thin films prepared in this manner. Lattice fringes in the TEM image of CuPc confirm that the deposited film has crystalline structure. Thin films of P3HT were prepared by spincoating a solution of P3HT and dichlorobenzene onto KCl substrates. For example, a comparison of the real dielectric functions obtained from VASE and EELS measurements on CuPc is shown in Figure 2. One sample (denoted as CuPc/PTCDA) consisted of evaporated CuPc on KCl with an interlayer of PTCDA and a second sample (denoted as CuPc) consisted Microsc. Microanal. 21 (Suppl 3), 2015 1472 1462 of evaporated CuPc on KCl. The datasets obtained were in good agreement with previously reported data for evaporated CuPc on Si [3]. The EELS data obtained from the CuPc and CuPc/PTCDA samples are also in good agreement with the VASE data. This correlation validates the use of STEM-EELS measurements for OPV studies and confirms that such data can be obtained without electron beam damage of the molecular structure. References [1] R.F. Egerton in "Electron Energy-Loss Spectroscopy in the Electron Microscope, Third Edition," (Springer, New York). [2] H.G. Tompkins & E.A. Irene in "Handbook of Ellipsometry," (Springer, 2005). [3] D. Datta, V. Tripathi, et al., Thin Solid Films 516 (2008), 7237. [4] The authors acknowledge the Air Force Office of Scientific Research and the Air Force Research Laboratory (AFRL) Materials and Manufacturing Directorate for funding, Kurt Eynik and Luke Bissel from AFRL for technical support, and The Ohio State University Center for Electron Microscopy and Analysis for technical support. (a) (b) <100> 1.3 nm <001> 1.2 nm 1.2 nm 20 nm 10 nm 1.3 n Figure 1. (a) HAADF STEM image of evaporated CuPc film. (b) indexed HRTEM image of evaporated CuPc film. Ellipsometry - Da;a (2008) Ellipsometry - Da;a (2008) Ellipsometry - CuPc/PTCDA Ellipsometry - CuPc EELS - CuPc EELS - CuPc/ PTCDA Real Dielectric Ellipsometry - CuPc/PTCDA Ellipsometry - CuPc EELS - CuPc EELS - CuPc/ PTCDA -3 Imaginary Dielectric -1 0.0 1.0 2.0 Energy (eV) 3.0 4.0 5.0 6.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 Energy (eV) Figure 2. Comparisons of the real and imaginary dielectric functions for CuPc.");sQ1[735]=new Array("../7337/1473.pdf","Detection and Characterization of OH Vibrational Modes using High Energy Resolution EELS","","1473 doi:10.1017/S1431927615008144 Paper No. 0735 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Detection and Characterization of OH Vibrational Modes using High Energy Resolution EELS Peter A. Crozier1, Toshihiro Aoki2, Qianlang Liu1 and Liuxian Zhang1 1.School for the Engineering of Matter, Transport and Energy, Arizona State University, Tempe Arizona 85287-6106 2. LeRoy Eyring Center for Solid State Science, Arizona State University, Tempe, Arizona 85287, USA The recent detection of vibrational excitations in monochromated electron energy-loss spectroscopy recorded from scanning transmission electron microscopes has opened up new opportunities for nanoscale materials characterization [1]. The enhanced energy resolution has the greatest impact on the low-loss EELS and it is now possible to probe vibrational and electronic excitations at the nanometer level. For example, localized bandgap mapping and detection of interband states is now possible providing a new tool to correlate optical properties with atomic structure [2,3]. Vibrational spectroscopy allows hydrogen containing species to be identified and correlated with materials structure. Detection of water and OH species on nanoparticle surfaces is important for developing a fundamental understanding of solar water splitting catalysts. The delocalized nature of the low-loss spectrum also makes it possible to use the aloof beam spectral acquisition mode (i.e. with the electron probe positioned outside the sample) dramatically reducing electron beam damage. To investigate the feasibility of OH detection, a series of hydroxide and hydrates have been investigated. Samples of Ca(OH)2, H3BO3 and Ni(NO3)26H2O were prepared for TEM by crushing the powders and dispersing over holey carbon films. Spectra were recorded from these samples using a Nion UltraSTEM fitted with a monochromator, an aberration corrector and a Gatan Enfinium spectrometer. The microscope was operated at 60 kV and the typical energy resolution was between 15 � 30 meV (the larger resolution was sometimes employed to increase the signal to noise in the spectrum) with a typical dwell time of 30 � 60 s. Spectra were acquired in both transmission mode and in aloof beam mode. In some cases, linescans were recorded in which a series of spectra were recorded as the beam was moved from the vacuum into the bulk. The OH stretch mode at 452 � 2 meV was easily detected from Ca(OH)2 using aloof beam EELS as shown in Figure 1a. The peak full width half maximum (FWHM) was around 24 meV and the linescan (Figure 1b and c) shows that the OH signal is maximized when the probe is positioned just outside the sample and falls to less than 10% of its maximum when the probe is 35 nm away from the surface. The vibrational signal also falls off rapidly as the probe moves into the sample due mostly to electrons being elastically scattered outside the spectrometer entrance aperture and possibly radiation damage. The OH stretch was also observed in H3BO3 (Figure 1d) at an energy of about 453 � 2 meV with a FWHM of 20 meV. (A smaller peak at around 350 meV may be due to hydrocarbon contamination). The aloof beam mode is also useful for minimizing radiation damage for valence-loss spectroscopy. Figure 2a shows the low-loss spectrum from the Ca(OH)2 recorded in aloof mode (beam 6 nm outside sample) and in transmission mode (beam 10 nm inside sample). Both spectra show a bandgap of approximately 6.5 eV but the aloof mode spectrum shows prominent peaks at 7.3 and 8.4 eV which do not show up in the transmission spectrum. The leading tail on the edge of the conduction band in the transmission spectrum is consistent with a large number of defects introduced as a result of beam damage. Microsc. Microanal. 21 (Suppl 3), 2015 1474 Figure 2b shows the OH vibrational fingerprint for Ni(NO3)26H2O. The peak center at around 430 meV is almost 25 meV below the corresponding stretch for the hydroxides. It is also asymmetric and much broader with a FWHM of around 80 meV suggesting a variation in the molecular environment around the OH species. The sample was in the microscope high vacuum system of 10-9 Torr for several days and there may have been selective loss of water from the surface region probed which is primarily probed by aloof beam EELS. During the presentation other factors influencing the detection of OH species especially on oxide surfaces will be discussed. The ultimate goal is to detect hydroxide species locally on the surface of water splitting catalysts. References [1] O.L. Krivanek et al. Nature 514, 209-212 (2014). [2] W.J. Bowman et al, these proceedings [3] Q. Liu et al, these proceedings [4] The support from US Department of Energy (DE-SC0004954) and the use of ETEM at John M. Cowley Center for HR Microscopy at Arizona State University is gratefully acknowledged. a) b) Energy Loss (meV) d) Aloof mode c) Energy Loss (meV) nm Figure 1: a) Aloof mode spectrum showing OH vibrational peak at 454 meV from Ca(OH)2. b) Zcontrast image and c) OH peak intensity as a function of position from Ca(OH)2. c) OH peak recorded in aloof beam mode from H3BO3. a) b) Energy Loss (eV) Energy Loss (meV) Figure 2: a) Low-loss spectrum from Ca(OH)2 recorded 6 nm outside the sample (blue) and in transmission mode (red). b) OH vibrational peak from Ni(NO3)26H2O.");sQ1[736]=new Array("../7337/1475.pdf","Study Using Low-loss EELS to Compare Properties of TMDs Produced by Mechanical and Liquid Phase Exfoliation","","1475 doi:10.1017/S1431927615008156 Paper No. 0736 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Study Using Low-loss EELS to Compare Properties of TMDs Produced by Mechanical and Liquid Phase Exfoliation Hannah C Nerl1, Fredrik S Hage2, Lothar Houben3, Quentin M Ramasse2 and Valeria Nicolosi1,4 1. CRANN & AMBER, Trinity College Dublin, Dublin 2, Ireland and School of Physics, Trinity College Dublin, Dublin 2, Ireland 2 SuperSTEM Laboratory, STFC Daresbury Campus, Keckwick Lane, Daresbury WA4 4AD, UK 3 Ernst Ruska Center for Microscopy and Spectroscopy with Electrons, Research Center J�lich, Germany 4. School of Chemistry, Trinity College Dublin, Dublin 2, Ireland In recent years, methods for the dispersion and exfoliation of 2D nanostructures of a range of nanomaterials have been successfully developed [1-8], opening up numerous possibilities for a range of innovative technologies [4, 6-10]. As opposed to mechanically cleaving, liquid phase exfoliation can produce large quantities of the material, but to make real applications of liquid phase exfoliated materials feasible there is a need to fully characterize and understand the impact the production route has on the properties of the nanostructures. In addition, very little is known about the effect of flake edges or the presence of surface contaminants on the properties of the materials. Due to the recent improvements in energy resolution of scanning transmission electron microscopy electron energy-loss spectroscopy (STEM EELS) [11,12] it is now possible to access new information such as the near-infrared/visible/ultraviolet spectral range using this technique. When compared with conventional techniques for measuring optical properties, STEM EELS offers the unique combination of high spatial as well as energy resolution opening up new possibilities for studying properties in a localized manner at an unprecedented energy resolution. For this study we used low-loss STEM EELS using the Nion UltraSTEM100 (SuperSTEM, UK) and the FEI PICO (J�lich, Germany) to compare the optical properties of MoS2 and other 2D materials produced by mechanical exfoliation and liquid phase exfoliation. Particular attention was being paid to changes in the very low loss EELS (energy losses <10eV) and to relate these to changes in the optical properties when going from multi- to single layered material as well as effects of flake edges [13]. In addition, we studied the effect of surface contamination and orientation dependence of the low loss EELS. To compare mechanical and liquid exfoliation routes we first analysed MoS2 produced via both production routes by STEM imaging and STEM EELS analysis of the low-loss region. Overall, we found no difference between the peak positions and their spatial variations between the two materials. The high angle annular dark field dark (HAADF) STEM image of the edge region of a MoS2 nanosheet produced by mechanical exfoliation is shown in figure 1, A. A STEM EELS map was acquired over the boxed region shown in figure 1, B and the corresponding spectra are shown in figure 1, C. The excitonic lines at 1.9 and 2.1eV [14] are visible in region in figure 1, C. These peaks have not previously been unambiguously identified using EELS. The peak in region in figure 1, C was the only peak visible in the spectra acquired over or close to the nanosheet edge and it was found to be just below 3eV. In addition, it appeared to slightly shift towards higher energy-losses in spectra acquired over regions further inwards, away from the nanosheet edge. The peak in region in figure 1, C was found to increase with increasing layers of material and has previously been associated with interlayer bonding and structure variations [14]. Microsc. Microanal. 21 (Suppl 3), 2015 1476 References: [1] AK Geim and KS Novoselov Nature Materials 6 (2007) p.183 [2] SD Bergin et al. Advanced Materials 20, 10 (2008) p.1876 [3] Y Hernandez et al. Nat. Nanotechnol. 3, (2008) p.563 [4] JN Coleman et al. Science 331, (2011) p.568 [5] M Chhowalla et al. Nat. Chem. 5, (2013) p.263 [6] V Nicolosi et al. Science 340, (2013) p.1226419 [7] AK Geim Science 324, (2009) p.1530 [8] KS Novoselov et al. Nature 490, (2012) p.192 [9] QH Wang et al. Nat. Nanotechnol. 7, (2012) p.699 [10] M Osada & T Sasaki J. Mater. Chem. 19, (2009) p.2503 [11] OL Krivanek et al. Nature 464, (2010) p.571 [12] OL Krivanek et al. Microscopy (Oxf). 62 (1), (2013) p.3 [13] C Backes et al., Nat. Comm. 5, (2014) p.4576 [14] M.M. Disko et al. Ultramicroscopy 23, (1987) p.313 [15] SuperSTEM is the UK National Facility for aberration-corrected STEM, funded by the UK EPSRC. Figure 1. A) HAADF STEM image (Nion, SuperSTEM, Daresbury, UK) of the edge region of a MoS2 nanosheet produced by mechanical exfoliation. B) HAADF STEM Survey image C) Summed EEL spectra, normalized to the zero-loss peak and de-noised with mild Principal Component Analysis using 80 components. The spectra were acquired over the regions marked in B going from vacuum (spectra 1-3), over the nanosheet edge region (spectra 4-5), over nanosheet steps (spectra 9-12 and spectra 16-19) into the nanosheet with increased thickness (spectra 22-24). The EEL spectra exhibited several changes when moving from the edge to the center, most prominently in energy-loss regions marked -.");sQ1[737]=new Array("../7337/1477.pdf","Standardization and Metrology for Efficiency and Reliability in Microbeam Analysis � No pain, no gain","","1477 doi:10.1017/S1431927615008168 Paper No. 0737 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Standardization and Metrology for Efficiency and Reliability in Microbeam Analysis � No pain, no gain Vasile-Dan Hodoroaba1, Ryna B. Marinenko,2 Mike Matthews3 and Jiang Zhao4 1. BAM Federal Institute for Materials Research and Testing, Division 6.8 Surface Analysis and Interfacial Chemistry, 12200 Berlin, Germany. 2. National Institute of Standards and Technology (NIST), Materials Measurement Science Division, Microanalysis Research Group, Gaithersburg, MD 20899, USA. 3. 17 Circuit Lane, RG30 3HB Reading, Great Britain. 4. Institute of Chemistry, SAC, #2 Zhong Guan Cun Bei Yi Jie Beijing, 100190, China. Standardization and metrology are two terms which are used rather rarely at major conferences as well as in scientific publications in the field of microbeam analysis. For laboratories operating under an accreditation scheme the operator of the microscope/microprobe must have available internal, national or international written standards which should be applicable to any quantitative analysis. Hence, requirements, specifications, guidelines or characteristics of methods, instruments or samples are provided with the final goal that these can be used consistently. In this way it is ensured that microbeam analyses results are reliable and meet quality-management requirements [1]. Standardization organizations operating at national level such as ANSI (USA), BSI (GB), DIN (Germany), SAC (China) or JISC (Japan) synchronize their activities to the international organization for standardization, ISO. In the field of microbeam analysis the ISO technical committee TC 202 is in charge with the harmonization of technical specifications of products and services making industry more efficient and breaking down barriers to international trade. Since the TC 202 establishment in 1991 eighteen ISO standards including updates have been published. The current list and the projects in development is updated on the ISO webpage [2], see Fig. 1. A main and continuous task of TC 202 and its subcommittees is to identify and evaluate feasible projects/proposals needed to be developed into new international standards. A short overview of the ongoing projects will be given. As regards the metrological aspects, The International Bureau of Weights and Measures (BIPM) through the Consultative Committee for the Amount of Substance (CCQM) is concerned with the metrological aspect in Chemistry and Biology. The development of the technical and organizational infrastructure of the International System of Units (SI) as the basis for the world-wide traceability of measurement results occurred here [3]. In particular, the Surface Analysis Working Group (SAWG) assists in identifying and establishing inter-laboratory work to test the consistency as well as to improve the traceability of spatially resolved chemical surface analysis at the micro and nanoscale. Examples of recent and ongoing projects related to microbeam analysis will be presented. The relevance of the evaluation of the complete chain of measurement uncertainties towards ensuring the traceability and comparability of the results will be emphasized through examples [4]. Thereby the compliance of the results of the microbeam analysis with the ISO document "Guide to the expression of uncertainty in measurement (GUM)" [5] will be explained. Another international platform in the frame of which pre-standardization work can be organized is VAMAS (Versailles Project on Advanced Materials and Standards) [6]. International collaborative projects involving mainly national metrological institutes aim at providing the technical basis for Microsc. Microanal. 21 (Suppl 3), 2015 1478 harmonized measurements, testing, specifications, and standards. One key point of VAMAS activities is constituted by inter-laboratory comparisons. In the field of microbeam analysis the technical working area (TWA) 37 Quantitative Microstructural Analysis as well as TWA 34 Nanoparticle Populations deal with corresponding projects. The contribution is meant not only to give a short overview of the existing standardisation and metrology activities in the field of microbeam analysis, but to bring possibilities and needs closer to the users. Especially younger personnel should be aware of these rather disliked parts of the microbeam analysis. At the same time motivation of conference participants to generate potential new projects is a hidden scope of the present contribution. References: [1] ISO/IEC 17025 (2005) General requirements for the competence of testing and calibration laboratories. ISO, Geneve. [2] http://www.iso.org (Standard developments / Technical Committees / ISO/TC 202 Microbeam analysis). [3] http://www.bipm.org/en/committees/cc/ccqm/ [4] V-D Hodoroaba et al., Microsc. Microanal. 20 (Suppl 3), 2014, p. 730. [5] ISO/IEC Guide 98-3:2008 Uncertainty of measurement Part 3: Guide to the expression of uncertainty in measurement (GUM:1995), Geneva: Internat. Org. Stds, 101 pp. [6] http://www.vamas.org/ Figure 1. Screenshots with standards published by ISO/TC 202 (left) and with standards under development (right) [2].");sQ1[738]=new Array("../7337/1479.pdf","Characterizing the Geometric Detection Efficiency of EDX Detectors","","1479 doi:10.1017/S143192761500817X Paper No. 0738 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterizing the Geometric Detection Efficiency of EDX Detectors Frederick H. Schamber Retired: 5095 Cherry Drive, Murrysville, PA 15668, frederick.schamber@gmail.com Energy Dispersive X-ray analysis (EDX) has been a powerful tool for microanalysis since the early 1970's. However, as SDD (Silicon Drift Detector) technology has supplanted the older Si(Li) (Lithium drifted Silicon) detector it has created the opportunity for a wide range of detector sizes and configurations that were not previously practical. Whereas sensor area was previously a crude but reasonably reliable means for predicting the relative collection efficiency of EDX detectors, such is no longer the case and sensor area is today a very inadequate predictor of actual performance, as is illustrated in Fig. 1. Geometric collection efficiency (GCE) is defined as the probability that an x-ray emitted from a given point on the specimen will be incident on the active area of the sensor, subject to geometric considerations only. In other words, this definition is intended to be independent of efficiency factors such as energy-dependent window absorption, sensor stopping power, count rate dependence, analyzer system settings, etc. The relative solid angle ( /4) is, of course, the definitive mathematical expression of GCE and Zaluzec has provided comprehensive aids for performing accurate calculations of solid angle () from known detector dimensions [1,2,3]. However, solid angle is not a quantity that can be directly measured, and must be computed from dimensions that are not readily available to an end-user. Further, a true assessment of effective solid angle must necessarily take into account such things as occlusion by the support grid of the detector window and vignetting by the electron trap � factors that become increasingly problematic when the x-rays do not originate on the detector's axis. Thus, a calculation of solid angle, though a useful statement of how a detector should perform under specific idealized conditions, is not necessarily representative of how the detector might actually be performing in a given configuration. And finally, the mathematical abstraction of a solid angle value expressed in steradians does not have obvious intuitive meaning for most individuals. Microscopists experience the effects of geometric collection efficiency in terms of count rate realized for a given beam current and do quickly develop an intuitive sense for this relationship. Thus the ratio "counts/sec-nA" is a natural empirical metric. Such measurements are extremely easy to make with only minimal equipment and provide a measure of detector performance that naturally extrapolates to other situations. However, without a consistent methodology, such measurements remain subjective assessments that cannot readily be compared between different users and laboratories nor readily related to formal solid angle calculations. The present proposal is to employ the effective solid angle ( ) as the standard theoretical/mathematical expression of GCE, where incorporates the effect of all opaque obstructions such as window support grid, specimen holder, and electron-trap collimation effects. Because of the complexity of quantifying such obstructions for off-axis x-rays, will typically be computed for the ideal case where the beam is incident at a point on the detector axis. It is then further proposed that a standard empirical metric of GCE be defined as the kilocounts/second/nanoamp for the measured K series x-rays, corrected for loss fraction (i.e. "dead time"), Microsc. Microanal. 21 (Suppl 3), 2015 1480 on a smooth pure copper specimen excited at normal incidence with a 20 KV electron beam. This metric, expressed as KCPS/nA (CuK @ 20KV) is proposed because: (a) it is easily measured with basic instrumentation; (b) it is intuitively meaningful; (c) it is insensitive to the settings (e.g., time constants or thresholds) of the x-ray analyzer; and (d) it can be directly related to by the relationship: where is the beam current incident on the specimen and ICR is the true "input count rate" of CuK events presented to the x-ray analyzer system. (X-ray analyzers conventionally infer ICR by dividing the actual measured counts by an effective "live time".) The constant () is the probability that a copper K or K x-ray is emitted with a takeoff angle of when a 20 KV electron is normally incident on a flat copper specimen. A table of () values therefore permits easy conversion between the empirical measure of GCE and its formal expression in solid angle. This proposal is being submitted for consideration by the USA Technical Advisory Group of ISO/TC202 dealing with standard practices for microbeam analysis. References: [1] N.J. Zaluzec, "Calculating the Detector Solid Angle in X-ray Energy Dispersive Spectroscopy", Microscopy and Microanalysis, 15[Suppl. 2] (2009) 520-521. [2] N.J. Zaluzec, "Analytical Formulae for Calculation of X-Ray Detector Solid Angles in the Scanning and Scanning Transmission Analytical Electron Microscope", Microscopy and Microanalysis, 20[4] (2014) 1318-1326. [3] N.J. Zaluzec, "XEDS Tools: Solid Angle Calculator", Website: http://tpm.amc.anl.gov/NJZTools/XEDSSolidAngle.html = () Figure 1 � Sensitivity comparison of two detector types for copper K x-rays. Type A has twice the peak sensitivity even though the sensor is only 2/3 the area of Type B. (Both 20 KV on pure Cu.)");sQ1[739]=new Array("../7337/1481.pdf","Determination of the Effective EDS Detector Area Using Experimental and Theoretical X-ray Emission Yields","","1481 doi:10.1017/S1431927615008181 Paper No. 0739 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Determination of the Effective EDS Detector Area Using Experimental and Theoretical X-ray Emission Yields Mathias Procop1,2, Vasile-Dan Hodoroaba1 and Ralf Terborg3 BAM Federal Institute for Materials Research and Testing, Division 6.8 Surface Analysis and Interfacial Chemistry, Berlin, Germany. 2. present address: IfG-Institute for Scientific Instruments, 12489 Berlin, Germany. 3. Bruker Nano GmbH, 12489 Berlin, Germany. An energy dispersive X-ray spectrometer operating with a semiconductor detector should be specified in compliance with the ISO standard 15632 [1]. Requirements for specification are: a) a general description of the spectrometer to evaluate its performance, b) the energy resolution with corresponding dead time, c) the P/B ratio in the Fe55 spectrum and, finally, d) the L/K intensity ratio in a Ni or Cu spectrum to estimate spectrometer efficiency at low energies. Items b) to d) can be easily checked by the user. Related procedures are recommended in the annexes of the standard. The solid angle is another important parameter as this defines the collection efficiency. It is defined as: = F / d� with F being the active area of a spherical detector and d being the distance between the radiation origin and the center of the detector. is not an intrinsic spectrometer property and can only be defined in combination with a specific SEM/EDX system. The detector area is one of the most interesting spectrometer properties since large area detectors have been available. Commonly, this area is given with the general description, but it is not specified. In the present paper a procedure was developed to check the detector area. The procedure is simple and can be applied by any spectrometer user. Measured intensity (count rate in cps) I of any X-ray line depends on the X-ray yield Y defined as the probability of emission of a photon per incident electron and per steradian: I = jeYA = F jeYA d2 1. (1) denotes the spectrometer efficiency (counts per incident photon), je the beam current (electrons per s), and A the absorption correction to be applied for the given take-off angle. Y was determined by various authors for selected elements [2-5]. Equation (1) enables the determination of the solid angle from count rate measurements, known efficiency and absorption correction as well as published or calculated X-ray yields. The detector-specimen distance d can be found from a simple experiment. Count rates have to be measured at different detector positions d+di, whereat di being the retraction distance. A plot of 1 = Ii 1 (d + di ) F dI i d (2) vs. (d+di) gives the distance d. Fig. 1 shows an example. Knowing d, the detector area F can be calculated from equation (1). The test procedure has been applied to 3 SDD and 1 Si(Li) spectrometers from different manufacturers. Selected specimens were Cu, Ti and Si, for which X-ray yields have been published [2-5]. Spectra were measured at 20, 10 and 5 kV SEM high voltages, respectively. Cu has the advantage of lowest uncertainty of the spectrometer efficiency as well as published X-ray yields Y. Beam currents were measured with a calibrated Keithley Amperemeter. Table 1 gives the results for the spectrometers tested until now. Determined areas are distinctly smaller than given in the spectrometer description. The reason for the differences is likely the aperture in front of the detector Microsc. Microanal. 21 (Suppl 3), 2015 1482 crystal shadowing its outer region. F determined according to equation (1) is hence the "active" or "effective" area of the detector. For three spectrometers the size of the aperture was known enabling a cross checking of the procedures accuracy. Determined areas F are within the �5% range for all three elements when experimentally determined yield data from ref. [5] and in case of Cu also from ref. [4] are used. Calculated areas F agree at best for Ti when theoretical X-ray yields are used. Because absorption and fluorescence corrections are small for the high voltages used, the selected model should not influence the result. However, calculated areas F depend on the selected model for the stopping factor, especially on the selected cross section. The new theoretical cross sections [6] were tested, but best results could be achieved with cross sections calculated with Casnati's empirical formula [7]. In summary it was demonstrated that the effective detector area F can be determined by the described procedure. Associated measurement uncertainty is below 10%, if published experimental yields [5] for Si-K, Ti-K and Cu-K or calculated yields with Casnati cross section are used. References: [1] ISO 15632:2012, ISO: Geneva. [2] M Green M and V E Cosslet, Br. J. Appl. Phys. (J. Phys. D) 2(1) (1968), p. 425. [3] E Lifshin, M F Ciccarelli and R Bolon 1980 in: Proceedings of 8th ICXOM (Boston, August 18-24) (Pendell, Midland, MI, USA), p. 141. [4] D C Joy, J. Microsc. 191(1) (1998), p. 74. [5] M Procop, Microsc. Microanal. 10 (2004), p. 481. [6] D Bote, F Salvat, A Jablonski and C J Powell, At. Data and Nucl. Data Tables 96 871 and ibid. 97 186 , NIST Standard Reference Database 164 (2009 & 2011). [7] E Casnati, A Tartari and C Baraldi, J Phys B 15 (1982), p. 155. Figure 1. Determination of the detector specimen distance. At detector position 40 mm the distance amounts to 40+9.2=49.2 mm. Table 1: F in mm� for tested spectrometers Detector SDD 10 mm� SDD 30 mm� SDD 100 mm� Si(Li) 10 mm� Si exp. Y 8.5 21.0 77 9.2 Si theor. Y 8.8 25.5 79 9.4 Ti exp. Y 8.7 22.7 75 Ti theor. Y 8.5 23.3 76 Cu exp. Y 8.8 22.0 79 10.1 Cu theor. Y 8.1 20.3 73 10.8");sQ1[740]=new Array("../7337/1483.pdf","Energy Resolution of Modern EDS Spectrometers: Is the current ISO standard definition outdated?","","1483 doi:10.1017/S1431927615008193 Paper No. 0740 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Energy Resolution of Modern EDS Spectrometers: Is the current ISO standard definition outdated? 1 Ralf Terborg1 and Meiken Falke1 Bruker Nano GmbH, Am Studio 2D, 12489 Berlin, Germany It is important to define characteristic parameters for energy dispersive X-ray spectrometers (EDS) in order to compare their performance. The ISO 15632 standard [1] defines various types of specifications, e.g. intensity ratios of certain L and K lines (L/K ratio) or the peak-to-background ratio. In the mentioned standard special interest is put into the measurement of the line widths (full-widthat-half-maximum, FWHM) to determine the energy resolution of the spectrometer. The method defined in that standard is a simple procedure: An EDX spectrum of a specific sample should be recorded, e.g. Fe55 or manganese, CaF2, carbon, but also an EDS-TM001 or EDS-TM002 [2] sample can be used. After a linear background subtraction the channel with the highest intensity (peak center) needs to be found. Then the two neighboring channels on each side of the peak (lowand high-energy side) with counts just above and below half of the highest peak intensity channel should be selected. After a linear interpolation of the two channel pairs the (fractional) number of channels, multiplied by the channel widths gives the energy resolution of the line, Fig. 1. While this procedure can be easily completed with a piece of paper and a pencil, it is just an estimation. Additionally, the procedure originates from the first edition of this standard from 2002. At this time the work was focused on Si(Li) detectors with energy resolution around 130eV for MnK and around 60eV for C-K lines. Since then silicon drift detectors (SDDs) have become the standard detector type, which have much better energy resolutions down to 121eV and 38eV for MnK and C-K, respectively which is very close to the theoretical limit of about 118eV for Mn-K and offers much better performance in the low energy region. Furthermore, the standard seems to focus on spectra with a channel width of 10 eV although spectrometer systems with smaller channel widths, e.g. 2.5 eV, are meanwhile available. In order to test if the proposed method can still be applied to SDDs with high energy resolution 130 spectra of the EDS-TM002 reference material [2,3,4] have been recorded with a Mn-K peak intensity of around 40000 counts using an XFlash 6|10 spectrometer. The Mn-K peak has been evaluated according to the method proposed in the ISO standard and also with other peak fitting methods using different types of background correction, Fig. 2. The determination using the ISO standard method shows large fluctuations of around +/-5eV and mostly underestimates the energy resolution while other peak fitting algorithms show a better agreement. The results will be further compared and discussed. References: [1] ISO 15632:2012 [2] M Alvisi et al., Micosc. Microanal. 12 (2006) p. 406 [3] http://www.rm-certificates.bam.de/en/certificates/layered_and_surface_materials/index.htm [4] V-D Hodoroaba, M Procop, Microsc. Microanal. 20 (2014) p. 1556 Microsc. Microanal. 21 (Suppl 3), 2015 1484 Fig. 1. Manganese K peak of a spectrum of an EDS-TM002 sample showing height and width of the peak. Fig. 2. Plot of the determined energy resolution of test spectra recorded with an XFlash 6|10 SDD using Gaussian peak fits with different background correction and the method proposed in the ISO standard.");sQ1[741]=new Array("../7337/1485.pdf","Further Differences in Biochemical Composition of Roots of Ni-Hyperaccumulating and Non-Hyperaccumulating Genotypes of Senecio coronatus","","1485 doi:10.1017/S143192761500820X Paper No. 0741 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Further Differences in Biochemical Composition of Roots of Ni-Hyperaccumulating and Non-Hyperaccumulating Genotypes of Senecio coronatus J. Mesjasz-Przybylowicz�, A.D. Barnabas�, I. Yousef�, P. Dumas�, F. Jamme�, T.P. Sechogela�, W.J. Przybylowicz1,3 1. 2. iThemba LABS, National Research Foundation, PO Box 722, 7129 Somerset West, South Africa Soleil Synchrotron, BP 48 F91192 Gif sur Yvette Cedex, France 3. AGH University of Science and Technology, Faculty of Physics & Applied Computer Science, al. A. Mickiewicza 30, 30-059 Krakow, Poland Hyperaccumulation is an unusual plant response to metaliferous soils. Such soils like those derived from ultramafic rocks, have elevated concentrations of heavy metals, mainly Ni, Cr, Zn and Cd. Most plants growing on these soils exclude metals from their shoots as excessive accumulation of heavy metals is toxic to the majority of plants. However, about 2% of plants growing on metaliferous soils take up and accumulate large quantities of metals in their shoots: a phenomenon known as hyperaccumulation. Senecio coronatus (Thunb) Harv, a widespread South African plant, also occurs on ultramafic outcrops where two genotypes have been identified: a Ni-Hyperaccumulator (H) and a Non-Hyperaccumulator (NH).In previous studies [1,2] some aspects of differences in cytology and chemical composition of the roots of the genotypes were reported. The present study details further biochemical differences between the roots and the possible relationship of these to the differential uptake of Ni. Chemical composition information was obtained using high spatial resolution synchrotron infrared microspectroscopy on thin vibratomed root sections 50 mm from the root tips, at the SOLEIL Synchrotron Facility in France (SMIS beam line). As described in a previous study [1], distinct groups of cells were present in the inner cortex of the H on the same radius as the xylem and phloem elements (Fig.1A, arrow) but the equivalent cells in the same region of the cortex of the NH were not as distinct (Fig.1E, arrow). To obtain more information about these cell groups, their biochemical composition was investigated. Spectra were recorded within the rectangular areas demarcated by red lines in the optical images of the H (Fig.1B) and the NH (Fig.1F). Comparison of the average normalized spectra of the H (Fig.1HS, blue spectrum) and the NH (Fig.1HS, red spectrum) showed that a significant difference between them was in the C-H stretching region (3100-2800 cm-1) often attributed to lipids. The strongest absorbance bands in the H appeared at 2920 and 2850 cm-1 which can be assigned to asymmetrical and symmetrical C-H stretching vibrations respectively. A major difference therefore between the roots was the increased lipid concentration in the H. Chemical profiles confined to the rectangular areas of the optical images of the genotypes (Fig.1B,F) were also constructed using the integrated absorbance of the CH stretching region between 3100-2800 cm-1. Comparison of these chemical profiles clearly show many more strong absorbance areas in the cell group of the H (Fig.1C) compared to that of the NH (Fig.1 G) confirming the higher concentration of lipids in the former. Another difference between spectra of the genotypes was the presence of a small band at 3010 cm-1 in the C-H stretching region of the H (Fig.1HS). This band is attributed to H-C=C unsaturated C-H stretching, possibly from unsaturated lipids. Chemical imaging of this band in the frequency region 3030-2990 cm-1 showed that these unsaturated vibrational bands were superimposed on the lipid Microsc. Microanal. 21 (Suppl 3), 2015 1486 localization sites of the H (Fig.1D). A third difference between the spectra of the genotypes was the detection of a slight shoulder at 1465 cm-1 in the spectrum of the H (Fig.1HS) assigned to CH2 bending, probably from lipids [3].No differences were found in spectra of the polysaccharide region between 1200-800 cm-1. Principal component analysis (PCA) on the full range (4000-2800 +1800-800) as well as PCA loadings plots of the PC1 component were also carried out and these confirmed the spectral differences observed in the genotypes. The significance of increased lipid levels and possibly also of different lipid types in roots of the H, compared to those of the NH, is presently not clear but may be a reflection of stress caused by the presence of higher Ni concentrations in the roots of the former. [1] Mesjasz-Przybylowicz J, Barnabas A, Przybylowicz W, Plant Soil 293(2007):61-78. [2] Mesjasz-Przybylowicz J, et al, Microscopy and Microanalysis 20(2014):1276-1277. [3] Komal Kumar J, Devi Prasad A G, Romanian Journal of Biophysics 21 (2011):63-71. Fig.1 A-HS: Optical images (A,B,E,F) and chemical profiles (C,D,G) of inner cortical cell groups in roots of Senecio coronatus genotypes.The warmer colours in C,D and G (towards the red end of the colour scale) indicate higher absorbance. Average normalized spectra (HS) of Ni- Hyperaccumulator (blue spectrum) and Non-Hyperaccumulator (red spectrum). H, Ni-Hyperaccumulator; NH, NonHyperaccumulator; X, xylem.");sQ1[742]=new Array("../7337/1487.pdf","The Use of Fluorescence In Situ Hybridization in Bacterial Detection in Mycorrhizal Networks","","1487 doi:10.1017/S1431927615008211 Paper No. 0742 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Use of Fluorescence In Situ Hybridization in Bacterial Detection in Mycorrhizal Networks Will Chrisler1 , Gayla Orr2, Alice Dohnalkova2, Ljiljana Pasa-Toli2, Gary Stacey3, Chanlan Chun4 Michael Sadowsky4 1 2 Health Impacts & Exposure Sciences Division, Pacific Northwest National Laboratory, Richland, WA Environmental Molecular Sciences Laboratory, Pacific Northwest National Laboratory, Richland, WA 3 Division of Plant Sciences, National Center for Soybean Biotechnology, Christopher S. Bond Life Sciences Center, University of Missouri, Columbia, Missouri 4 BioTechnology Institute, University of Minnesota Rhizosphere, the belowground biogeochemical hotspot of plant roots intimately associated with soil bacteria and plant fungi (mycorrhizae), is a highly interactive microcosm where fundamental processes such as carbon fixation, nutrient acquisition and nitrogen cycling occur. The detailed exploration of these micro-scale rhizosphere processes is essential to our better understanding of large-scale processes in soils, and determination what drives source-sink relationships that control the cycling of plant nutrients belowground. To study the nutrient partitioning into roots and their exchange with the soil microbial community, a multidisciplinary, multi-institutional research project was designed to address the complexity of the plant-soil interface, using the means of genomics, and correlated analytical tools including methods of mass spectroscopy (analyzing proteomic, metabolomic and transcriptomic components) and high resolution microscopy. Arabidopsis thaliana wild-type and SEX (=starch excess) mutants were grown hydroponically or in artificial potting mix, and rhizosphere population arising from inoculation with soil extract was established and analyzed by a variety of tools to view the live interactions of plant roots with the soil microbial population. The SEX mutants lack the ability to allocate starch (stored carbon) from their leaves to the rest of the plant during the dark period. Research has shown that the growth differs significantly in SEX mutants from the wild-type in carbon allocation to the roots [1]. Our focus will be to develop a method to examine how the mutated genome affects the spatial and temporal differences in colonization between the bacteria and roots of A. thaliana. The challenge to this complex system is to differentiate the bacteria associated with the plant root fro the root itself. Many dyes are not specific to a single organism and will not stain one community member independently. Using a combination of species specific fluorescent stains, non-specific fluorescent stains, and FISH probes, the spatial distribution of the bacteria-root environment can be investigated. Propidium iodide allows the detection of DNA in both bacteria and the A. thaliana roots. Using Calcofluor White, the cellulose in A. thaliana root can be specifically stained to determine the root boundary. The probe EUB 338, specific for the domain Bacteria, was used to identify bacteria. A. thaliana grown with/without bacterial inoculant derived from soil extract are harvested prior to the beginning of light and dark cycles. Roots are washed in PBS and then fixed in 4% paraformaldehyde for 30 minutes at room temperature. The samples were then washed in PBS and kept in 4 �C. Segments of the roots were dissected from A. thaliana wild-type and Sex-like mutants, and processed for Fluorescence in situ hybridization (FISH) with rRNA-targeted probes following a slightly modified protocol [2]. The FISH labeled samples are then imaged by confocal laser scanning microscopy. Microsc. Microanal. 21 (Suppl 3), 2015 1488 1478 Additional morphological determination of the microbial community was done by scanning electron microscopy (SEM). [1] Weber A. et al. The Plant Cell (13), 1907�1918. ( 2001) [2] Hugenholtz P. et al. Methods in Molecular Biology (179), 2002, pp 29-42 [3] This research was performed at the W. R. Wiley Environmental Molecular Sciences Laboratory, a national scientific user facility sponsored by the U.S. DOE, located at PNNL. A B Figure 1. (A) Confocal laser scanning microscopy (CLSM) micrographs showing the bacterial colonization of A. thaliana roots stained by fluorescence in situ hybridization (FISH). (green, EUB-338 AlexaFluor488; blue, root fluorescence; red, DNA). (B) Increased magnification of region indicated by box in (A). Scale Bar 20�m");sQ1[743]=new Array("../7337/1489.pdf","A Multiscale Approach to Understanding Calcium Toxicity in Australian Proteaceae","","1489 doi:10.1017/S1431927615008223 Paper No. 0743 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A Multiscale Approach to Understanding Calcium Toxicity in Australian Proteaceae Peta L. Clode1, Patrick Hayes1,2, Nicolas Honvault1,2,3, and Hans Lambers2 1. Centre for Microscopy, Characterisation & Analysis, The University of Western Australia. Crawley, WA 6009 Australia. 2. School of Plant Biology, The University of Western Australia. Crawley, WA 6009 Australia. 3. Agriculture, Institut Polytechnique LaSalle Beauvais. Beauvais Cedex, 60026 France. The Proteaceae are a family of plants predominantly distributed within the Southern hemisphere, with >600 species in West Australia alone. They display staggering diversity and endemism but are highly restricted in their distibution by soil quality and type. In order to understand the role of calcium in influencing distribution patterns, we are sampling plant species that are soil-indifferent (few, grow across all environments) and calcifuge (common, grow in acidic, nutrient poor soils). From this, the distribution, form, and amount of calcium in leaves is being investigated at the cellular level using a variety of correlative techniques, including optical-based microscopies, Raman spectroscopy, X-ray microscopy, and quantitative EDS X-ray microanalysis. For optical based imaging and analysis, chemically fixed samples are either sectioned (100 um thickness) using a vibratome or embedded in ultra low viscosity resin and microtomed (1 um thickness). Samples are subsequently imaged using brightfield and ultraviolet techniques, and analysed via Raman spectroscopy (WITec alpha 300RA+). For X-ray microscopy (Xradia Versa XRM-520), chemically fixed samples are incrementally scanned over 360 degrees to produce 3-dimensional data sets, which are then reconstructured and quantitatively analysed using a variety of software packages. For quantitative EDS X-ray microanalytical mapping, samples are rapid frozen on pins and cryoplaned to produce a flat, transverse cross section. Element distributions are mapped using a cryoSEM (Zeiss Supra FESEM) fitted with an 80 mm EDS detector and quantified by AZtec Software (Oxford Instruments). Paired optical and ultraviolet imaging allows for the determination of the overall structure of the leaf, including visualizing distinct tissue layers, stomates, and crypts (Figs. 1 and 2). Imaging of resin embedded sections reveals a higher level of structural detail at the cellular level, including observation of intracellular crystals within mesophyll cells (Fig. 3), which can then be characterized using Raman spectroscopy. X-ray microscopy allows for 3-dimensional visualization and quantification of these crystals within a large area of the leaf (Fig. 4). The presence, abundance, size, shape, and composition (calcium or silicon based) of these crystals varies extensively between plant species and with regard to their location within the leaf tissues. Elemental mapping (Figs. 5 and 6) and quantitation is providing further insight into the cellular distribution of calcium within leaves. The arrangement and concentration of calcium within different cell layers also varies immensely between plant species. These data suggest that calcium regulation, storage, and toxicity in the Proteaceae is not a conserved, family trait. References: [1] The authors acknowledge funding from the ARC Discovery Program to PLC and HL (DP130100005). PH is the recipient of an APA Scholarship, NH's internship at UWA is supported by the Kwongan Foundation. The authors acknowledge use of the CMCA, a facility funded by the University, State and Commonwealth Governments. Microsc. Microanal. 21 (Suppl 3), 2015 1490 1 2 3 4 5 6 Figures 1, 2. Paired optical micrographs of a vibratome section from a Banksia prionotes leaf, imaged using 1) brightfield, and 2) ultraviolet techniques. Scale bars = 100 um. Figure 3. Optical micrograph of a resin-embedded section from a Banksia menziesii leaf, highlighting cellular structure and the presence of calcium oxalate crystals (red arrows). Scale bar = 20 um. Figure 4. Single image from a 3-D tomogram, showing an array of crystals distributed throughout the inner tissues of a Petrophile macrostachya leaf. Scale bar = 250 um. Figures 5, 6. Paired qualitative EDS X-ray maps from a Persoonia comata leaf, showing 5) oxygen, and 6) calcium distributions. Scale bars = 100 um.");sQ1[744]=new Array("../7337/1491.pdf","Investigation of Structure-Function Relationship of Long-Distance Transport in Plants: New Imaging Tools to Answer Old Questions","","1491 doi:10.1017/S1431927615008235 Paper No. 0744 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Investigation of Structure-Function Relationship of Long-Distance Transport in Plants: New Imaging Tools to Answer Old Questions Daniel L Mullendore, Daniel R Froelich, Sierra Beecher, Tim J Ross-Elliott, Jan Knoblauch, Michael Knoblauch Plant Cell Biology Lab, School of Biological Sciences, Washington State University, Pullman WA The phloem is a plant tissue that connects distant plant parts via the sieve tube system. It is a major player to maintain organismal integrity due to its activity in translocation of photoassimilates and distribution of long distance signals [1, 2]. Almost all food consumed by humans, or biomass produced for bioenergy, has at one point been translocated through the phloem. While major research efforts focus on optimization of photosynthesis to tackle biomass and food production, phloem loading and transport is poorly understood. Photosynthetic efficiency is, however, dependent on phloem transport and is down-regulated if production exceeds export significantly. Therefore, a better understanding of processes dictating long distance transport is essential. One of the major reasons for our poor understanding of phloem structure-function relations is that preparation induces immediate artefacts [3, 4]. The preparation for electron optic observation requires small sample sizes. While cutting tissue samples before fixation has only minor effects on standard cell types such as parenchyma-, mesophyll- or epidermis cells, sample preparation of sieve elements usually leads to massive artifact generation due to the high conductivity combined with high turgor in sieve tubes. Cutting the tube system results in an instant pressure release followed by dislocation and structural alteration of subcellular components [5]. A large group of investigators tackled this problem from the 1960's to the 1980's. Hundreds of publications later, the results were still confusing [1]. One of the major problems was that investigators had no in vivo reference. While electron microscopy provides unprecedented resolution, samples often contain severe preparation and dehydration induced artefacts. Without a (low resolution) reference of the location and structure of components in vivo, it is very difficult to interpret and to optimize preparation protocols. Even though some of the micrographs might have shown nearly natural conditions, it is unclear which ones. Over the last years we have developed a series of methods for in situ studies of sieve tubes by confocal and super resolution microscopy (6,7,8). Based on our observations, we have developed a preparation method to maintain the subcellular structure of sieve tubes close to the natural state. We developed microscopy rhizosphere chambers (Micro-ROCs) which are a square plant pot with one wall consisting of a #1.5 coverslip. A micro-pore fabric is attached flat against the coverslip from the inside of the pot. The pot is filled with soil and a seed is placed between the meshwork and the coverslip. The mesh forces the germinating root to grow two dimensionally along the coverslip, while the root hairs are in contact with the soil medium to ensure natural growth conditions. For observation, the whole pot is put under the microscope to ensure minimum disturbance. The plants can be grown for weeks through their entire life cycle. We investigated the phloem in Arabidopsis roots in situ and were able to study the dynamics of phloem transport and phloem unloading. The generation of transgenic plants expressing fluorescent proteins tagged to different structures and organelles allowed us to study the precise location of the structures [8]. Microsc. Microanal. 21 (Suppl 3), 2015 1492 1482 Our investigations generated with micro-ROCs revealed structures that had never been described before. To gain more insight by higher resolution, we attempted to generate a preparation protocol for electron microscopy. Since cutting inevitably generates artefacts, we used whole young plants for tissue preparation and embedding. Standard preparation methods failed to preserve structures observed in situ. Therefore we attempted to explore freezing techniques. The phloem is at least 3-4 cell layers deep inside the tissue which calls for high pressure freezing. Unfortunately the sample size did not allow making use of this technology. Therefore we plunge froze whole seedlings in slushy nitrogen followed by freeze substitution. While the depth of the phloem inside the tissue reduces heat transfer rates, the high sugar content reduces water crystal formation. It turned out that neighboring cells like parenchyma and mesophyll cells showed severe freezing artifacts, but sieve elements and their adjacent companion cells were nicely preserved. Most importantly, however, we were able to image structures that had never been described in sieve elements and match perfectly the location and shape of structures found in in situ observations[8]. We now have all the tools necessary to investigate the structure and physiology of phloem transport and close a major gap in our understanding of plant function. Besides phloem studies, micro-ROCs are suitable to study root physiology and root soil communities in situ. [1] ed HD Behnke, RD Sjolund (1990) Sieve elements (Springer, Berlin, New York) [2] M Knoblauch, WS Peters, Photosynthesis Research 117 (2013) 189-196 [3] M Knoblauch, KJ Oparka, The Plant Journal 70 (2012) 147-156 [4] M Knoblauch, WS Peters, Plant, Cell, Environment 33 (2010) 1439 � 1452 [5] M Knoblauch, AJE van Bel, The Plant Cell 10 (1998) 35-50 [6] DL Mullendore et al., The Plant Cell 22 (2010) 579-593 [7] DF Froelich et al., The Plant Cell 23 (2011) 4428-4445 [8] Anstead et al., Plant Cell Physiology 53 (2012) 1033-1042 Figure 1. Imaging of P-proteins taggd with YFP in an intact sieve tube of a transgenic Arabidospsis plant. Bar = 10 �m. Figure 2. TEM micrograph of Arabidopsis sieve elements after plunge freezing and freeze substitution. The sieve elements contain well preserved mitochondria (solid arrow) and P-protein filaments (dashed arrow). Bar = 1 �m.");sQ1[745]=new Array("../7337/1493.pdf","Dynamics of Thrombus Formation in Mouse Testicular Surface Vein Visualized by Newly Devised "Vascular Mapping" Method for Live-CLEM Imaging in vivo.","","1493 doi:10.1017/S1431927615008247 Paper No. 0745 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Dynamics of Thrombus Formation in Mouse Testicular Surface Vein Visualized by Newly Devised "Vascular Mapping" Method for Live-CLEM Imaging in vivo. Akira Sawaguchi1, Satoshi Nishimura2, 3, 4 1. Department of Anatomy, Ultrastructural Cell Biology, Faculty of Medicine, University of Miyazaki, Miyazaki, Japan. 2. Department of Cardiovascular Medicine, The University of Tokyo, Tokyo, Japan 3. Translational Systems Biology and Medicine Initiative, The University of Tokyo, Tokyo, Japan 4. Center for Molecular Medicine, Jichi Medical University, Tochigi, Japan One of the goals of biomedical microscopy is to elucidate a functional morphology in vivo, leading to associated pathophysiology and further clinical investigation. The dynamics of thrombus formation has yet to be elucidated to prevent human diseases, such as infarction in the heart and the brain, etc. Most recent progress of in vivo multi-photon laser scanning microscopy (LSM) enables us to visualize experimentally induced vascular damage followed by thrombus formation in blood vessels of mice [1-3]. However, a new reliable method remains to be developed to acquire a transmission electron microscopic (TEM) image of the experimentally formed thrombus in situ, hampered by difficulties in locating and processing the thrombus to be cut into ultrathin sections. To address this problem, we newly developed a "Vascular Mapping" method for correlative light and electron microscopy (CLEM) of thrombus formation in vivo. Briefly, after anesthesia, Texas-Red� dextran was injected to visualize blood flow, and then the mouse testicular surface vein was exposed onto a multi-photon LSM. Following a capture of complex vascular pattern at lower magnification (Fig 1), the focused region was damaged by laser irradiation to obtain sequential images of thrombus formation in situ (Fig 2). For TEM image preparation, the testis was excised and immersed into half-strength Karnovsky fixative. Then, the focused region was excised referring the captured image of complex vascular "map" (Fig 3), and processed to be embedded into epoxy resin. Next, semi-thin sections (3 �m in thickness) were sequentially cut tracing upstream of the damaged vein to reach the thrombus (asterisk in Fig 4A). The obtained semi-thin section of thrombus was re-embedded into epoxy resin to be cut into ultrathin sections (70 nm in thickness) using capsule-supporting ring [4]. As results, the present "vascular mapping" approach succeeded in TEM observation of the thrombus as shown in Fig 4B and 4C, yielding live-CLEM imaging in vivo. The preliminary observation demonstrated a noteworthy attachment of neutrophils (Fig 4B) onto the thrombus as well as fine structure of the aggregated platelets (Fig 4C) to arrest the bleeding. It is highly anticipated that further application will clarify not only thrombus formation but also the following fibrinogenolysis and blood vessel repair, leading to a goal of biomedical microscopy for further clinical investigation such as antithrombotic treatment under "live" CLEM imaging. We demonstrate step-by-step procedures from "live" to "TEM" imaging of thrombus formation in vivo by newly devised "Vascular mapping" method. Microsc. Microanal. 21 (Suppl 3), 2015 1494 References: [1] S Nishimura et al, Blood 119 (2012) p45-56. [2] MJ Kuijpers and JW Heemsker, Methods Mol Biol 788 (2012), p3-19. [3] MM Kamocka et al, J Biomed Opt 15 (2010), p016020. doi: 10.1117/1.3322676. [4] A Sawaguchi et al, J Microsc 234 (2009), p113-7. Figure 1. Representative vascular map on the mouse testicular surface vein. Boxed region corresponds to Fig 2. Figure 2. Sequential images of laser irradiation (framed region), and thrombus formation (asterisk in B) in the boxed region of Fig 1. Figure 3. Excised piece of the testis (compare to Fig 1) and a semi-thin section showing the branch indicated with blue arrow in B. Figure 4. Semi-thin section light microscopic image of the thrombus (asterisk in A), and TEM image (B and C) obtained by re-embedding of the semi-thin section. Arrow indicate attached neutrophils.");sQ1[746]=new Array("../7337/1495.pdf","Cryo-Planing of Frozen Hydrated Samples by Triple Ion Gun Milling (CryoTIGM)","","1495 doi:10.1017/S1431927615008259 Paper No. 0746 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Cryo-Planing of Frozen Hydrated Samples by Triple Ion Gun Milling (CryoTIGM) Irene Y.-T. Chang1, Derk Joester1 1. Materials Science and Engineering, Northwestern University, Evanston, IL 60208, USA Cryo-fixation, i.e. the rapid vitrification of hydrated samples, omits the artifact-prone steps of chemical fixation, embedding, and dehydration, and consequently offers improved preservation of biological ultrastructure [1, 2]. Imaging of cryo-fixed samples by cryo-SEM is particularly efficient since timeconsuming steps such as embedding and preparation of thin sections is not required. However, a fresh interior surface has to be prepared, typically by freeze fracture. The random nature of fracture does not ensure the passage of the fracture plane through the regions of interest [1]. Cryo-planing by cryo-FIB allows positional control of structural investigation, and smaller structures like single cells can be selected with surgical precision [3]. However, in many situations it would be advantageous to be able to cryo-plane larger areas of the frozen-hydrated samples. Herein, we describe a cryogenic sample preparation workflow for cryo-SEM based on broad ion-beam milling that addresses this issue. The central innovation of the workflow of the cryo triple ion gun milling (CryoTIGM) method is a custom-built tool based on the Leica TIC3X slope cutter. Specifically, a TIC3X unit was fitted with a vacuum load lock that allows cryo-transfer of a vitrified sample. Samples were high pressure-frozen between aluminum planchettes and trimmed using a custom-built cryo-saw. The cryo-saw consists of a liquid nitrogen reservoir, a sample compartment, a diamond blade, and a VCT-docking port (Fig. 1A). Trimming was performed under liquid nitrogen, and samples were then positioned in a sample holder next to a milling mask (Fig. 1B). The sample was transferred to the CryoTIGM tool (Fig. 1C), where three broad Ar+ beams converge at the mask shielding the trimmed sample edge (Fig. 1D). Material above the mask was removed, creating a cross-section in the sample at the level of the mask (Fig. 1E). The ion-milled sample was subsequently freeze-etched and coated with Pt to increase contrast. We optimized operating parameters for CryoTIGM for a range of samples, including yeast suspensions, mouse liver biopsies, and suspensions of whole sea urchin embryos. Irrespective of the sample type, we find that ion milling with Ar+ at an acceleration voltage of 3.0 kV, a current of 1.0 mA/gun, a base temperature of -120oC, and for 2 h results in very smooth cryo-planed area of ~700,000 �m2 (Fig. 1E). Analysis of cryo-planed surfaces after freeze-etching and coating indicates that CryoTIGM does not induce crystallization of vitreous ice in well-frozen samples. A direct comparison of samples prepared by conventional freeze fracture and CryoTIGM revealed that 1) surfaces prepared by CryoTIGM are much smoother; 2) cellular and organellar details are observed at comparable high resolutions with good contrast; 3) most importantly, additional ultrastructural features can be revealed (Fig. 2). For example, in CryoTIGM-prepared sea urchin embryo samples, tight contacts among neighboring ectodermal cells, suggesting the presence of intercellular junctions, can be identified (Fig. 2B) Membrane tethers extended from ectodermal cells to the hyaline layer are visible. The hyaline layer is resolved to consist of two layers, with the space between them occupied by some kind of vesicles or granules. References: [1] Studer, Humbel and Chiquet, Histochemistry and Cell Biology 130 (2008), p. 877. [2] Gilkey and Staehelin, Journal of Electron Microscopy Technique 3 (1986), p. 177. Microsc. Microanal. 21 (Suppl 3), 2015 1496 [3] Heymann et al, Journal of Structural Biology 155 (2006), p. 63. [4] The authors acknowledge funding from the NSF Major Research Instrumentation program (NSF MRI-1229693), the Northwestern University Materials Research Center (DMR-1121262), and the NSF Biomaterials program (DMR-1106208). Figure 1. Experimental setup of CryoTIGM. (A) The cryo-saw used for trimming sample carriers. DS = diamond saw, LR = LN reservoir, SC = sample compartment, VL = VCT loading dock. (B) Top view of the sample compartment. A sample carrier is trimmed (along the dashed line) and transferred to the sample holder. (C) CryoTIGM tool with attached cryo/vacuum-transfer-shuttle. (D) Schematic drawing of CryoTIGM milling process. (E) A frozen hydrated sample after milling shows a triangular cryoplaned area of 700,000 �m2 (). Scale bar represents 500 �m. Figure 2. Cryo-SEM of high pressure frozen sea urchin embryos prepared by freeze fracture (A) and CryoTIGM (B). hv = hyaline layer vesicle, ic = intercellular contact, n = nucleus, t= tether, v = vesicle. Ice segregation is visible in the extracellular space and blastocoel. Segregation was observed in both freeze-fracture and CryoTIGM samples and is therefore an indication of poor vitrification, likely due to the high salt content (3.5 wt%) of the medium. Scale bar represents 5 �m.");sQ1[747]=new Array("../7337/1497.pdf","Correlative iPALM and Platinum Replica Electron Tomography to Highlight Single Molecules on Clathrin Endocytic Structures in 3D.","","1497 doi:10.1017/S1431927615008260 Paper No. 0747 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Correlative iPALM and Platinum Replica Electron Tomography to Highlight Single Molecules on Clathrin Endocytic Structures in 3D. Kem Sochacki1, Gleb Shtengel2, Andrea Dickey1, Harald Hess2, and Justin Taraska1. 1. 2. National Heart Lung and Blood Institute, Bethesda, MD, USA. Janelia Research Campus, Howard Hughes Medical Institute, Ashburn, VA, USA. Clathrin mediated endocytosis is a ubiquitous process used by all eukaryotic cells to internalize material. This is an integral process for neural and endocrine signaling as well as cellular homeostasis. While severe mutations to this process are lethal, minor mutations are associated with several human maladies including cancer and Alzheimer's disease [1]. There are well over 30 known proteins regulating the entire endocytic process including initiation, membrane curvature, pit size, cargo recruitment, and fission. These proteins are well biochemically characterized, but there appears to be a large amount of redundancy which has made it difficult to parse out their specific roles in endocytosis. The spatial organization of clathrin associated proteins within single clathrin structures is not well studied and may give important structural insight into how this process is regulated. Here, we develop a correlative method that combines 3D interferometric photoactivated localization fluorescence microscopy (iPALM) and platinum replica electron microscopy (PREM) to localize clathrin associated proteins on the topography of mammalian cell membranes [2]. Platinum replicas of mammalian cell cortices viewed by transmission electron microscopy have been used for decades to survey the 3D shape of single clathrin structures across the membrane [3] (Figure 1). However, identifying proteins of interest in these images with immunogold requires large metal particles (15 nm) that not only label sparsely but also obstruct the image beneath. iPALM localizes fluorescently labeled proteins to a precision of better than 20 nm in the membrane plane (XY) and 10 nm in the axial (Z) plane [4]. Our iPALM/PREM correlation method allows us to correlate clathrin fluorescence with clathrin viewed with EM, to within 20 nm across a 20 �m cell membrane. We then use this method to find the position of the clathrin adapter protein, epsin 1, with respect to the shape of clathrin structures (Figure 2). We unexpectedly find that epsin 1 is located along the edge of the clathrin structures in XY but along entire height of clathrin pits. While there are several models of how epsin functions in mammalian cells, this localization is most consistent with the recent model that epsin coordinates actin at clathrin sites. As we continue to use this method to study other endocytic proteins, we are finding that different isoforms are localized very differently within cells and that protein localization can be very different in different mammalian cell lines. Additionally, this method is poised to answer many questions about other important processes on the plasma membrane like clathrin-independent endocytosis, exocytosis, cell adhesion, and cell motility. Microsc. Microanal. 21 (Suppl 3), 2015 1498 References: [1] McMahon, Harvey T., and Emmanuel Boucrot. "Molecular mechanism and physiological functions of clathrin-mediated endocytosis." Nature reviews Molecular cell biology 12.8 (2011): 517-533. [2] Sochacki, Kem A., et al. "Correlative super-resolution fluorescence and metal-replica transmission electron microscopy." Nature methods 11.3 (2014): 305-308. [3] Heuser, John. "Three-dimensional visualization of coated vesicle formation in fibroblasts." The Journal of cell biology 84.3 (1980): 560-583. [4] Shtengel, Gleb, et al. "Interferometric fluorescent super-resolution microscopy resolves 3D cellular ultrastructure." Proceedings of the National Academy of Sciences 106.9 (2009): 3125-3130. Figure 1. Platinum Replica of U87 Glioblastoma cell line. Note the different sizes and shapes of clathrin structures marked with arrows. Scale Bar 0.5 �m. Figure 2. Correlative iPALM/PREM of Epsin1 on a clathrin coated pit. Image is 400 nm x 400 nm.");sQ1[748]=new Array("../7337/1499.pdf","Correlative Electron and Fluorescence Microscopy of Magnetotactic Bacteria in Liquid: Toward In Vivo Imaging","","1499 1499 doi:10.1017/S1431927615008272 doi:10.1017/S1431927615008272 Paper No. 0748 Microsc. Microanal. 21 (Suppl 3), 2015 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 � Microscopy Society of America 2015 Correlative Electron and Fluorescence Microscopy of Magnetotactic Bacteria in Liquid: Toward In Vivo Imaging Taylor J. Woehl1, Sanjay Kashyap1, Emre Firlar1, Teresa Perez-Gonzalez2, Damien Faivre2, Denis Trubitsyn3, Dennis A. Bazylinski3, and Tanya Prozorov1 1 Emergent Atomic and Magnetic Structures, Division of Materials Sciences and Engineering, Ames Laboratory, Ames, IA 50011, USA. 2 Department of Biomaterials, Max Planck Institute of Colloids and Interfaces, Science Park Golm, 14424 Potsdam, Germany. 3 School of Life Sciences, University of Nevada at Las Vegas, Las Vegas, NV 89154, USA. Magnetotactic bacteria, present in many natural aquatic environments, biomineralize ordered chains of uniform magnetite or greigite nanocrystals, also known as magnetosomes [1]. These nanoparticles exhibit nearly perfect crystal structures, faceting, and consistent species-specific morphologies, leading to well-defined magnetic properties. As a result, magnetotactic bacteria can serve as a model system for the study of the molecular mechanisms for magnetite biomineralization [2]. However, little is known about their complex formation mechanism. Transmission electron microscopy (TEM) can provide critical information about the organization of the magnetosomes and the growth mechanisms by revealing the nanoparticle structure on the atomic level [3]. Conventional TEM traditionally does not allow imaging in native liquid or atmospheric environments because of the high vacuum of the specimen enclosure. The vitreous ice environment used during cryo-TEM imaging of bacteria [4] is not entirely representative of the native hydrated cellular state either, and may introduce artifacts due to specimen preparation and electron radiation damage [5]. Therefore direct observation of the time-dependent growth of magnetosomes in their native cellular environment remains a significant challenge. Here we utilize correlative fluorescence and fluid cell scanning TEM (STEM) imaging to visualize magnetotactic bacteria containing nanoscale biomineralized magnetosomes [6]. Fluorescently labeled cells of Magnetospirillum magneticum strain AMB-1 were immobilized on microchip window surfaces and visualized in a fluid cell with high angle annular dark field (HAADF) STEM, followed by correlative fluorescence imaging (Figure 1). The confined environment of the fluid cell caused a rapid decrease in bacteria viability over a time of two hours, necessitating the use of post-STEM fluorescence microscopy to verify bacteria membrane integrity. Notably, the post-STEM fluorescence imaging indicated that the bacterial cell wall membrane did not sustain radiation damage during STEM imaging at low electron dose conditions ( 0.2 ). Correlative STEM and fluorescence imaging of magnetotactic bacteria is a first step in observing biomineralization of magnetite nanocrystals in vivo, and the described approach is expected to be applicable to a broad range of microorganisms that biomineralize various nanomaterials [7]. Microsc. Microanal. 21 (Suppl 3), 2015 Microsc. Microanal. 21 (Suppl 3), 2015 1500 1490 References: [1] D.A. Bazylinski, R.B. Frankel, Nature Reviews Microbiology, 2 (2004) 217-230. [2] D. Faivre, D. Sch�ler, Chem. Rev., 108 (2008) 4875�4898. [3] R.E. Dunin-Borkowski, M.R. McCartney, R.B. Frankel, D.A. Bazylinski, M. Posfai, P.R. Boseck, Science, 282 (1998) 1868-1870. [4] A. Komeili, H. Vali, T.J. Beveridge, D.K. Newman, Proceeding of the National Academy of Sciences of the United States of America, 101 (2004) 3839-3844. [5] J. Dubochet, M. Adrian, J.J. Chang, J.C. Homo, J. Lepault, A.W. McDowall, P. Schultz, Quarterly Review of Biophysics, 21 (1988) 129-228. [6] T.J. Woehl, S. Kashyap, E. Firlar, T. Perez-Gonzalez, D. Faivre, D. Trubitsyn, D.A. Bazylinski, T. Prozorov, Scientific Reports, 4 (2014). [7] This work was supported by the U.S. Department of Energy, Office of Basic Energy Science, Division of Materials Sciences and Engineering. The research was performed at the Ames Laboratory, which is operated for the U.S. Department of Energy by Iowa State University under Contract No. DE-AC02-07CH11358. T.P. acknowledges support from the Department of Energy Office of Science Early Career Research Award, Biomolecular Materials Program. The authors would like to thank C. L. Mosher, M. J. Kramer, and T. M. Pepper for informative discussions. This research was supported in D.F's laboratory by the Max Planck Society, and a starting grant from the ERC (Project MB2, no. 256915). Figure 1. Correlated liquid cell STEM and fluorescence images of magnetotactic bacteria with intact membranes. (a) False colored, background subtracted HAADF-STEM image of two bacterial cells near the corner of the SiN window, the magnetosome chains appear in purple and are denoted with white arrows. (b) Post-STEM composite fluorescence image of the same fluid cell sample. (c) Correlated STEM and composite fluorescence image of the bacterial cells highlighted in the red box in (b). The scale bar is 1 m in (a) and (c) and 10 m in (b).");sQ1[749]=new Array("../7337/1501.pdf","Indium-Tin-Oxide (ITO) as Stable and Effective Coating Material for Correlative Confocal and Immuno-Scanning Electron Microscopy Studies.","","1501 doi:10.1017/S1431927615008284 Paper No. 0749 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Indium-Tin-Oxide (ITO) as Stable and Effective Coating Material for Correlative Confocal and Immuno-Scanning Electron Microscopy Studies. Andrea Falqui,1a Simona Rodighiero,2 Elisa Sogne,2,1a Bruno Torre,1b Roberta Ruffilli,3 Maura Francolini,2,4 Cinzia Cagnoli,2 Enzo di Fabrizio 1b King Abdullah University for Science and Technology (KAUST), Thuwal 23955-6900, Kingdom of Saudi Arabia: a Biological and Environmental Sciences and Engineering Division; b Physical Sciences and Engineering Division 2 Fondazione Filarete, Viale Ortles 22/4, 20139 Milano, Italy 3 CEMES/CNRS, 29 Rue Jeanne Marvig BP 94347, 31055 Toulouse Cedex 4, France 4 Universit� degli Studi di Milano, Via Vanvitelli 32, 20129 Milano, Italy Correlating Confocal Microscopy (CM) and Electron Microscopy (EM) imaging of cells and tissues is a well-known method to understand the relations occurring between cellular structure and function. Conventional CM is capable to visualize the presence of either specific antigens by the use of immunofluorescent labeling or fluorescent proteins (FP), with resolution of few hundreds of nanometers. On the other hand, EM is capable to image the cellular ultrastructure down to nanometer scale. Putting together the information given by the two techniques on the same area of the specimen allows then to determine the antigen location on the cellular ultrastructure. EM imaging could be carried out on biological specimens both in transmission (TEM) and in scanning (SEM) mode. In both the EM approaches, to get information on antigen/protein distribution, cells can be labeled with antibodies conjugated with small (<20 nm) gold particles. In the case of SEM the secondary electrons (SE) are used to image the specimen surface morphology, whilst compositional contrast obtained by collecting backscattered electrons (BSE) allows to simultaneously localize the gold particles that labeled a cellular surface antigen. A key point in the observation of the cellular ultrastructure is the preparation protocol followed. In particular, in the case of SEM imaging with surface immuno-labeling, the specimen has to be made electrically conductive, while preserving the compositional contrast needed to localize the gold nanoparticles acting as immuno-markers. This need brings to inevitably exclude heavy metals as coating agents, as they would completely mask the BSE signal coming from the nanogold immuno-markers. Enough recent literature indicates two possible approaches to face such a problem. The first one consisted in using substrates covered by Indium Tin Oxide (ITO), being ITO a well known both optically transparent and stably conductive material [1]. In this approach the cells are not coated with any kind of conductive film since the ITO-covered substrates allow to not get specimen charging, but the overall quality of the cellular imaging so obtained is enough limited. Conversely, in the second approach the cells surface was coated with a very thin layer (< 5nm) of chromium, but with the main limitation due to the lack of conductivity because of chromium's fast oxidation that occurs if it is deposed under low vacuum condition or exposed to air [2]. We report here the correlative CM and immuno-SEM studies of HeLa cells and neurons grown on specific Ti-patterned glass substrates [3]. We studied if an optically transparent thin layer of ITO deposed by ion sputtering on the samples surface could be used in order to overcome the major restrictions resulting from the approaches reported in [1-2]. With this aim, we have first studied the optimal ITO layer thickness needed to obtain both good specimen conductivity and preservation of the BSE signal coming from gold nanoparticles used as immuno-markers. Second, we have estimated by 1 Microsc. Microanal. 21 (Suppl 3), 2015 1502 Atomic Force Microscopy (AFM) the actual cells thickness range. That parameter was used to quantitatively determine by Monte Carlo simulations the most appropriate electron beam acceleration voltage to use for collecting BSE signal mainly coming from the immuno-labeled cells covered with the thinnest and conductive ITO coating layer. We have then found that such a coating film is stable over time, and capable to provide both suitable electrical conductivity, good SE production and preservation of the BSE signal coming from the gold immuno-markers. As a consequence, we show how it allows to easily perform correlative CM and immuno-SEM microscopy studies. References: [1] H. Pluk et al., J. Microsc. 233 (2009), p. 353. [2] M. W. Goldberg, Methods Cell. Biol. 88 (2008), p. 109. [3] L. Benedetti et al., Sci. Rep. 4 (2014), p. 7033. Figure 1. (a) Confocal Microscopy Images of primary cortical neurons where the subunit alfa1 of the GABAA receptor was tagged with ATTO 488 fuorophore. (b) SEM SE Low magnification image of the rectangular area reported in panel (a), after deposition of an ITO coating film of 20 nm. Scale bar: 5 �m Figure 2. SEM imaging of primary cortical neurons where the subunit alfa1 of the GABAA was tagged with 12-nm nanogold markers and coated with an ITO film of 20 nm in thickness. (a) High resolution SE Images of the rectangular area reported in Figure 1 (b). (b): High resolution BSE image corresponding to panel (a): the gold immuno-markers can be easily observed. Scale bar: 50 nm");sQ1[750]=new Array("../7337/1503.pdf","Wide Area Observation of Fully Hydrophilic Tissue Achieved by Sliding It on the Dish of the of the Atmospheric Scanning Electron Microscope (ASEM)","","1503 doi:10.1017/S1431927615008296 Paper No. 0750 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Wide Area Observation of Fully Hydrophilic Tissue Achieved by Sliding It on the Dish of the Atmospheric Scanning Electron Microscope (ASEM) Chikara Sato1, 2 Mari Sato1, Tatsuhiko Ebihara1, Hiditoshi Nishiyama3, Mitsuo Suga3, Nassirhadjy Memtily1, 2, 4 Biomedical Research Institute, National Institute of Advanced Industrial Science and Technology (AIST), Higashi 1-1-1, Tsukuba, Ibaraki 305-8566, Japan. 2. Graduate School of Comprehensive Human Sciences, University of Tsukuba. 1-1-1 Tennodai, Tsukuba, Ibaraki Prefecture 305-0006, Japan. 3. Advanced Technology Division, JEOL Ltd., Musashino 3-1-2, Akishima, Tokyo 196-8558, Japan. 4. Traditional Uyghur Medicine Institute of Xinjiang Medical University, 393 Xinyi Rd, Urumqi, Xinjiang Uyghur Autonomous Region, 830011 China. Correspondence should be addressed to Chikara Sato (ti-sato@aist.go.jp). The ASEM has an inverted SEM configuration, and was developed to realize SEM observation of a sample in aqueous liquid in a readily-accessible, open container (ASEM dish) [1]. An optical microscope (OM) positioned above the dish can be used to observe wide areas of the sample, and smaller regions can be observed by the SEM through a thin silicon nitride (SiN) film in the base of the ASEM dish (Fig. 1A). The optical axes of both microscopes are aligned and fixed to ensure that correlative images are recorded, and the specimen stage can move two-dimensionally (x-y) for targeting. From below the ASEM dish, the inverted SEM projects an electron beam up through the SiN film to the sample. Backscattered electrons are captured for ASEM imaging by a disk-shaped detector located just beneath the SiN film. The specimen depth observable by ASEM is 2-3 m at 30 kV, and the resolution is 8 nm near the SiN membrane [1-2]. ASEM was widely applied to observe various primary cells, tissues, bacteria and protein crystals [3-7]. The OM makes the handling of specimens easy, which is exploited here for the wide area observation necessary for intra-operative cancer diagnosis. For wide area observation using ASEM, a fixed tissue slab in a radical scavenger (glucose) solution is repeatedly pushed slightly to the side with tweezers under monitoring by OM from above, and imaged by SEM from below (Fig. 1A) [8]. This can be done using a standard ASEM dish with a single SiN window (Fig. 1B) or a newly developed 8-window ASEM dish (Fig. 1C). In the example shown, a slab of spinal cord tissue stained with PTA was placed in an 8-windowed ASEM dish, and repeatedly induced to slide very slightly across the thin membrane windows and imaged. Although the observable area in each ASEM imaging was restricted to the field of the 8 windows, two sequential images partly overlapped and could be merged, covering a wider area of the spinal cord (Fig. 1D, E). The observable area of the multi-windowed dish was thus successfully extended. The shift of the tissue on the flat bottom surface of ASEM dish (Fig. 1A) was precisely monitored by OM from above. ASEM realize high throughput observation of wet tissues, and could improve intraoperative cancer diagnosis because ASEM does not require the cryo-thin-sectioned OM which takes about 15-30 minutes for each sample. Skeletal muscle fibers were clearly visible when PTA-stained tissue slabs of gastrocnemius muscle were observed by ASEM (Fig. 2). In myocytes, I-bands and A-bands were visualized as dark zones and broader bright zones, respectively, and Z-lines were distinguished as a fine thin line in the center of the dark I-bands (Fig. 2B, C). These results suggest that protein complexes in natural aqueous liquid environment can be studied at high resolution in combination with tissue engineering using the ASEM. 1. Microsc. Microanal. 21 (Suppl 3), 2015 1504 References [1] H Nishiyama et al, J Struct Biol. 172 (2010), p. 191-202. [2] Y Maruyama et al, J Struct Biol. 180 (2012), p. 259-270. [3] C Sato, JEOL news 46(1) (2011), p.17-22. [4] K Hirano et al, Ultramicroscopy143 (2014), p. 52-66. [5] T Kinoshita et al, Microsc Microanal 20(2) (2014), 469-483. [6] C Sato et al, Biochem Biophys Res Commun 417(2012), p1213-1218. [7] Y Maruyama et al, Int J Mol Sci 13(8) (2012), p.10553-10567. [8] N Memtily et al, International J Oncology, in press. Figure 1. Wide area imaging by shifting a tissue on the ASEM dish. (A) Schematic diagram of the procedure. The shift caused by pushing with the tweezers is precisely controlled by low magnification monitoring using the upper OM. (B) Schematic diagrams of a one-window and an 8-window ASEM dish (C). All windows are 250 x 250 �m. (D) Spinal cord image initially recorded from window b and the image recorded in the same window after the tissue has been pushed in one direction causing it to slip across the SiN-film window, b'. (E) Merged image of windows b and b'. Figure 2. ASEM images of gastrocnemius muscle stained with PTA. (A) Low magnification image. Filaments run along striated muscle fibers. (B, C) Higher magnification image of the white rectangle in the preceding panel. The muscle fiber has A-bands, evident as broad bright zones (A), and I-bands, evident as dark zones (I). Z-lines, look like a faint thin white line in the middle of the I-bands (I). Scale bar 10 m in A, 5 m in B and 1 m in C.");sQ1[751]=new Array("../7337/1505.pdf","Building and Imaging Silicide Nanostructures in Nanowires","","1505 doi:10.1017/S1431927615008302 Paper No. 0751 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Building and Imaging Silicide Nanostructures in Nanowires F. Panciera, Y.-C. Chou2,3,4, M. C. Reuter2, D. Zakharov4, E. A. Stach4, E. Jensen5, K. M�lhave5, S. Hofmann1 and F. M. Ross2 1. 2. Department of Engineering, University of Cambridge, 9 J. J. Thomson Avenue, Cambridge, UK. IBM Research Division, T. J. Watson Research Center, Yorktown Heights, NY, USA 3. Department of Electrophysics, National Chiao Tung University, 1001 University Road, Hsinchu city, Taiwan. 4. Center for Functional Nanomaterials, Brookhaven National Laboratory, Upton, NY, USA. 5. DTU-Nanotech, Department of Nano- and Microtechnology, Technical University of Denmark, Kgs. Lyngby, Denmark. The vapour-liquid-solid (VLS) growth process allows the fabrication of semiconductor nanowires with a remarkable level of structural control. This structural versatility has contributed to the tremendous impact of VLS-grown nanowires. By tuning growth parameters and catalyst composition, we can control the composition, diameter and growth direction of the nanowires, and form branches, kinks, periodically arranged twins, and even polytype superlattices, leading to applications in low power electronics and optoelectronics. Transmission electron microscopy (TEM) has been an essential tool for imaging structures after growth, while in situ TEM has provided insights into growth mechanisms, helping to drive the development of more complex structures [1]. Here we describe a variation on the conventional VLS growth concept, developed through in situ TEM observations, which provides opportunities for building new types of nanowire-based structures. The principle is to first grow silicon nanowires using liquid AuSi catalyst droplets, then interrupt growth and supply a new species, such as a metal, that can react with the material in the droplet. In situ ultra-high vacuum TEM [1] visualizes the nature and kinetics of these droplet reactions. As one might expect, species like Ni or Co nucleate silicide phases. But, surprisingly, the phases form as faceted nanocrystals that float within the liquid droplet. Eventually each nanocrystal makes contact with its nanowire-catalyst growth interface and becomes attached. Further growth of Si allows the nanocrystal to become incorporated into the nanowire. In Figures 1 and 2, the formation and incorporation steps are imaged in situ using ultra-high vacuum TEM. Figure 3 shows a nanowire imaged after carrying out the incorporation step in an aberration-corrected environmental TEM (ETEM). The outcome can be nearperfect endotaxial incorporation of a nanocrystal of controlled structure, size and location, with the whole process repeatable to incorporate multiple nanocrystals in a single nanowire. Potential applications may include single-electron transistors, high-density memories, semiconductor lasers, and tunnel diodes. Ultra high vacuum TEM provides reaction kinetics under well-controlled growth conditions, while aberration-corrected ETEM provides improved spatial and temporal resolution, revealing details of the phase sequence and step flow growth. However, imaging phase nucleation and growth is challenging in aberration-corrected ETEM due to its non-UHV environment and the strict stability requirement on heating. To address both issues, we describe a low-drift sample heater based on a resistively heated cantilever geometry [2], and we discuss a protocol to reduce background contamination during growth, Microsc. Microanal. 21 (Suppl 3), 2015 1506 based on operating the ETEM cold finger at a temperature tuned to condense water vapour but not the growth species, disilane [3]. We will show how these experimental modifications can lead to quantitative information on nanowire growth, silicide nucleation and phase transformations with high spatial and temporal resolution [4]. References: [1] F. M. Ross, Reports on Progress in Physics 73 (2010), p. 114501. [2] C. Kalles�e, et al., Nano Lett. 12 (2012), p. 2965. [3] Y.-C. Chou, et al., in preparation (2015). [4] We acknowledge the National Science Foundation (Grants No. DMR-0606395 and 0907483), ERC Grant 279342: InSituNANO, the National Science Council of Taiwan (Grant NSC-101-2112-M-009021-MY3), the Center for Interdisciplinary Science (MOE-ATU project for NCTU) and the Center for Functional Nanomaterials, Brookhaven National Laboratory, which is supported by the U.S. Department of Energy, Office of Basic Energy Sciences, under contract DE-AC02-98CH10886; C. Czarnik for assistance with image processing and A. W. Ellis for technical support. Figure 1. Formation and epitaxial contact of a NiSi2 nanocrystal in a Si nanowire after 1nm Ni deposition, oxidation to prevent sidewall reactions, then temperature increase to 500�C and flow of 5�10-5 Torr Si2H6. A Below the eutectic temperature of 370�C the Au is solid. B As we exceed the eutectic temperature, AuSi liquid and Ni-silicide nanocrystals form. C The silicide agglomerates into an octahedral nanocrystal. D The nanocrystal attaches at the Si interface. Figure 3. A nanowire with embedded NiSi2 nanocrystal, imaged after Si growth in the ETEM. Figure 2. Continued growth of the nanowire in Figure 1 at 500�C and 5�10-5 Torr Si2H6, showing the NiSi2 nanocrystal becoming embedded in the Si nanowire. Time shown in minutes since the final image of Figure 1.");sQ1[752]=new Array("../7337/1507.pdf","Understanding nanomaterial synthesis with in situ transmission electron microscopy","","1507 doi:10.1017/S1431927615008314 Paper No. 0752 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Understanding nanomaterial synthesis with in situ transmission electron microscopy Bethany M. Hudak1, Yao-Jen Chang1, Lei Yu1, Guohua Li2, Danielle N. Edwards1, Matthew E. Park1, and Beth S. Guiton1,3 1. 2. Department of Chemistry, University of Kentucky, Lexington, KY 40506 Department of Electrical Engineering, Anhui University, Hefei, 230601, China 3. Materials Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831 The vapor-liquid-solid (VLS) nanowire growth technique is a synthesis method widely used to grow high-quality, single-crystalline semiconductor nanowires [1-3]. First introduced in 1964 by Wagner and Ellis to grow silicon nanowires, this method has evolved to utilize many different metal catalyst materials to grow a wide variety of inorganic nanowires with facile control of diameter, length, and dopant concentration [1,4]. These inorganic nanowires have many applications, such as for Li-ion battery electrodes, gas sensors, and solar cell components [5-7]. While VLS is a ubiquitous growth method, understanding of the growth kinetics is limited, especially for binary and ternary crystal systems. Theoretical predictions suggest that the growth of such nanowires is governed by steady-state kinetics, and that the crystal chemistry of the reverse process may be different from that which governs the nanowire growth [8]. The use of in situ techniques has advanced the understanding of the VLS process and the kinetics of VLS growth [9]. By use of heating in a transmission electron microscope (TEM), we have developed a method to observe the Au-catalyzed VLS growth of metal oxide nanowires occurring in reverse; this nanowire dissolution is dubbed the solid-liquid-vapor (SLV) process. All nanowires used in this study were grown via VLS synthesis using gold metal as the catalyst material. The as-grown nanowires were analyzed by powder x-diffraction (XRD) to determine crystal structure. The nanowires were then dispersed in high-quality methanol, and dropcast onto Protochips thermal Echips for heating in the TEM. Nanowires of like-composition were heated using a consistent heating profile. Due to the ultrafast heating and cooling capabilities of the Protochips in situ TEM holder, nanowire dissolution can be quenched periodically throughout the SLV process in order to track the material content of the Au catalyst particle using energy dispersive x-ray spectroscopy (EDS), as the nanowire dissolves at the solid-liquid interface and evaporates at the liquid-vapor interface. Highresolution imaging is used to determine nanowire growth direction and its role in SLV dissolution. This method of observing the reverse of the nanowire growth process should provide an experimental platform to explore features relevant to the VLS growth mechanism, such as saturation concentration of a reactant within a VLS catalyst droplet and the use of VLS catalyst metals for controlled etching of semiconducting materials. In addition, the use of in situ heating offers the ability to study nanowire heterostructure formation, such as evolution from a homogeneous nanowire to core-shell or peapod type nanowire. In the absence of advanced in situ TEM techniques, these processes would likely remain unstudied [10]. Microsc. Microanal. 21 (Suppl 3), 2015 1508 [1] R.S. Wagner and W.C. Ellis, Appl. Phys. Lett. 4 (1964), p. 89. [2] Y. Wu and P. Yang, J. Am. Chem. Soc. 123 (2001), p. 3165. [3] Y. Cui et al, Science 293 (2001), p. 1289. [4] M. Law et al, Ann. Rev. Mater. Res. 34 (2004) p. 83. [5] Y.-D. Ko et al, Nanotechnology 20 (2009), p. 455701. [6] A. Kolmakov et al, Adv. Mater. 15 (2003), p. 997. [7] J.B. Baxter and E.S. Aydil, Appl. Phys. Lett. 86 (2005), p. 053114. [8] S. Ryu and W. Cai, J. Mater. Res. 26 (2011), p. 2199. [9] S. Kodambaka et al, Phys. Rev. Lett. 96 (2006), p. 096105. [10] The authors acknowledge funding from the Kentucky NFS EPSCoR program (Y.-J.C., B.M.H.), NASA Kentucky under NASA award No: NNX10AL96H (B.M.H), Office of Basic Energy Sciences, Materials Sciences and Engineering Division, U.S. Department of Energy (B.S.G.), and in part by ORNL's Center for Nanophase Materials Sciences (CNMS), which is sponsored by the Scientific User Facilities Division, Office of Basic Energy Sciences, U.S. Department of Energy. Figure 1. (a) Cartoon showing the relation between VLS nanowire growth and SLV dissolution of a SnO2 nanowire. (b) Snapshots from a movie recorded during heating of a SnO2 nanowire showing dissolution occurring at elevated temperature. Scale bar equals 0.2 �m.");sQ1[753]=new Array("../7337/1509.pdf","In Situ TEM for Quantitative Electrochemistry of Energy Systems","","1509 doi:10.1017/S1431927615008326 Paper No. 0753 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In Situ TEM for Quantitative Electrochemistry of Energy Systems Megan E. Holtz1, Yingchao Yu2, Johary Rivera2, H�ctor D. Abru�a2, and David A. Muller1,3 1. School of Applied and Engineering Physics, Cornell University, Ithaca NY 14853. 2. Department of Chemistry and Chemical Biology, Cornell University, Ithaca NY 14853. 3. Kavli Institute at Cornell for Nanoscale Science, Ithaca, NY, USA In 2010, we began designing a broadly applicable in situ TEM liquid cell chip to enable studies that correlate quantitative electrochemistry with the microstructure of the active material in the TEM. Protochips fabricated the chip we designed, which has demonstrated reliable electrochemical performance even for surface-sensitive measurements such as cyclic voltammetry of fuel cell catalysts, and has become the leading electrochemical cell chip used for quantitative in situ TEM studies today [13]. Here we discuss critical factors behind chip performance, and illustrate applications for alkaline fuel cell studies with an ion exchange membrane, and for lithium ion battery materials. Through careful materials choices, the in situ chip we designed reproduces the performance of a traditional electrochemical cell (Fig 1). The working electrode is patterned onto the silicon nitride viewing membrane of a liquid flow cell holder (Fig 1ab). For quantitative electrochemistry, the chip must not introduce extraneous electrochemical signals so the process of interest may be studied. We use an electron-transparent glassy carbon working electrode which offers significantly lower scattering than previous metal electrodes. Electrochemically it is cleaner, with a featureless background signal from the capacitive response of the liquid (Fig 1c). The working electrode is entirely in the viewing window, and electrical leads are covered in the photoresist SU8 to prevent additional electrochemical response. We used Ti instead of Cr for adhesion layers, to prevent Cr diffusion that can dominate the electrochemistry. The reference electrode is near the working electrode to minimize uncompensated resistance, and the counter electrode is large and far away to provide ample current and prevent species migration. Classical test cases for catalyst electrochemistry are quantitatively reproduced in situ, as shown by cyclic voltammetry in 0.1M H2SO4 of platinum nanoparticles on Vulcan deposited on the working electrode and measured in the TEM (Fig 1c). This represents a rigorous test case for quantitative electrochemistry, since the features are surface effects that are sensitive to contaminants at the submonolayer level, including hydrogen adsorption and desorption and oxide formation and reduction on the platinum surface. The in situ electrochemistry reproduced the characteristic voltametric profile of the platinum nanoparticles at an appropriate current scale, while a chip with no platinum nanoparticles exhibited only a background current associated with the double layer capacitance of carbon. Figure 2 shows in situ STEM of platinum nanoparticles separated from the counter electrode by a phosphonium alkaline anion exchange membrane. After flowing in methanol, the cyclic voltammogram displays a methanol oxidation process (Fig 2a). During methanol oxidation, we see formation of particles that are likely carbonates (Fig 2bc), which block pores and poison the fuel cell. The particles are generated by the electrochemistry but assisted in agglomeration by the electron beam - small particles appear over the entire electrode, and only while performing methanol oxidation. Real time spectroscopic imaging is also possible using valence EFTEM. Fig 3 shows the cycling of a LiFePO4 battery electrode material, where EELS reveals the lithiated and delithiated regions, providing dynamic information about Li-ion transport and degradation of the cathode material [1,4]. References: Microsc. Microanal. 21 (Suppl 3), 2015 1510 [1] M Holtz et al, NanoLetters 14 (2014) p. 1453. [2] R Unocic et al, Microscopy and Microanalysis 20 (2014) p. 452. [3] G.Z. Zhu et al, J. Phys. Chem. C 118 (2014) p. 22111. [4] Work supported by the Energy Materials Center at Cornell, DOE EFRC BES (DE-SC0001086). EM Facility support from the NSF MRSEC program (DMR 1120296). Figure 1. In situ electrochemistry TEM holder and electrochemical data. (a) Cross-section of the chips. (b) Schematic of the top chip, with a carbon working electrode (WE) on the membrane, Pt reference electrode (RE) and counter electrode (CE). The chips exhibited electrochemical activity qualitatively similar to an ex situ microelectrode, as shown for the platinum nanoparticle cyclic voltammetry (CV) in (c). The chip alone with no Pt deposited shows a minimal electrochemical response. Figure 2. (a) Cyclic voltammogram in 0.1M NaOH and 0.1M methanol with a hydrated phosphonium alkaline anion exchange membrane and Pt catalyst particles in a Tecnai F20 at 200 keV. STEM images of (b) before and (c) after methanol oxidation on the membrane. What appear to be carbonate particles appear after cycling. The particles are easily moved and agglomerated under the electron beam. Figure 3. (a) Charge/discharge of the battery cathode material LiFePO4. 5 eV EFTEM images of the lithiated (b,d) and delithiated (c) LiFePO4, highlighting regions of FePO4, at the times marked in (a).");sQ1[754]=new Array("../7337/1511.pdf","Effects of non-contact electric fields on consolidation behavior of agglomerated yttria-stablized zirconia","","1511 doi:10.1017/S1431927615008338 Paper No. 0754 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Effects of non-contact electric fields on consolidation behavior of agglomerated yttria-stablized zirconia Hasti Majidi and Klaus van Benthem Department of Chemical Engineering and Materials Science, University of California, Davis, CA, 95616, USA Electric field-assisted sintering (EFAS) techniques, which include spark plasma sintering and "flash sintering", have demonstrated the potential for enhanced densification at lower temperatures compared to conventional methods of sintering.[1-2] However, the fundamental mechanisms governing the enhancements of the densification processes are debated in the literature. The separate effects of the applied field and/or the resulting current remain unknown.[3] Studying the mechanisms of EFAS therefore requires separating the effects of the electric field from those of the electric current. Here, we report, for the first time, in situ scanning transmission electron microscopy (STEM) observations and quantitative shrinkage analysis of 3 mol% yttria-stablized zirconia (3YSZ) powder agglomerates during consolidation in the presence of a non-contacting externally applied electric field. 3YSZ particle agglomerates were dispersed in ethanol and drop-casted onto MEMS devices (Protochips, Inc.), which are capable of simultaneous heating and biasing. A schematic of an electrothermal MEMS device is illustrated in Figure 1. Agglomerates are supported by an amorphous carbon layer which covers the holey SiC membrane serving as a resistive heating element. A noncontacting electric field was applied by placing the 3YSZ agglomerates in between two parallel W electrodes under bias. Real-time STEM observations allow monitoring the evolution of particle agglomerates during in situ consolidation with and without applying the electric field. Figure 2 shows microstructural changes of two 3YSZ agglomerates during in situ isothermal heating at 900 �C in the absence (Figure 2(a-c)) and presence of an electric field (Figure 2(d-f)). In the absence of the electric field, the morphology of the agglomerate does not change significantly within 106 min at 900 �C. However, when 3YSZ agglomerates are exposed to 500V/cm, neck formation and growth, particle coalescence and pore shrinkage is observed within only 4 min at 900 �C. The shrinkage analysis of the agglomerate during in situ heating was performed using a recently developed image processing tool,[4] where the projected area of the agglomerate, obtained from STEM micrographs, is plotted as a function of time and electric field strength. While heating in the absence of the electric field slowly reduces the projected area of the agglomerate to 97% after 106 min, the combination of an external electric field with sample heating leads to a sudden shrinkage of 93% only after 4 min at 900 �C. Achieving similar magnitude of shrinkage without applied electric field requires higher temperatures and longer times (subsequent heating at 1000 �C for 28 min). While the exposure of particle agglomerates to an electric field in the absence of current flow through the sample results in sudden consolidation, the magnitude of shrinkage is small compared to that observed during, e.g., flash sintering experiments. Hence, Joule heating due to raising electric currents must contribute significantly to densification during EFAS while effects of electric fields are appreciable. Future studies require the evaluation of defect formation as a function of electric field strengths to identify mechanisms of field-assisted sintering. Microsc. Microanal. 21 (Suppl 3), 2015 1512 [1] U. Anselmi-Tamburini et al, J. Mater. Res. 19 (2011), P. 3255. [2] M. Cologna, A.L.G. Prette, and R. Raj, J. Am. Ceram. Soc. 94 (2011), P. 316. [3] J.E. Garay, Annu. Rev. Mater. Res. 40 (2010), P. 445. [4] H. Majidi, T. Holland, and K. van Benthem, Ultramicroscopy 152 (2015), P. 35. [5] H. Majidi, and K. van Benthem (submitted for publication). [6] This work was supported by the Army Research Office under award #W911NF121049. The authors are grateful to Dr. John Damiano for useful discussions regarding Protochips MEMS devices. Figure 1: a schematic of electrothermal MEMS device enabling simultaneous heating and biasing to monitor particle agglomerates during in situ consolidation in the presence of non-contact electric field Figure 2: STEM micrographs of 3YSZ agglomerates during in situ heating to 900 �C (a-c) in the absence and (d-f) presence of non-contact electric field. Data reproduced from [5].");sQ1[755]=new Array("../7337/1513.pdf","In-Situ Analytical Transmission Electron Microscopy Study of Electrochemical Lithiation of a Sulfur � Carbon Nanotube Composite Cathode","","1513 doi:10.1017/S143192761500834X Paper No. 0755 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In-Situ Analytical Transmission Electron Microscopy Study of Electrochemical Lithiation of a Sulfur � Carbon Nanotube Composite Cathode Jeremy Ticey1, Vladimir Oleshko1,3, Yujie Zhu2, Chunsheng Wang 2, and John Cumings1 1. 2. Department of Materials Science and Engineering, University of Maryland, College Park, MD Department of Chemical and Biomolecular Engineering, University of Maryland, College Park, MD 3. Materials Science and Engineering Division, Material Measurement Laboratory, National Institute of Standards and Technology, Gaithersburg, MD Rechargeable Li-S batteries have the potential to meet the high power demands of next generation lightweight, low-cost, and environmentally friendly batteries useful in both small-scale portable devices and large-scale applications such as electric vehicles. Lithium-sulfur (Li-S) batteries offer a high theoretical capacity of 1,672mAh/g and a theoretical energy density of 2567Wh/kg, roughly five times larger than that of currently utilized carbon-based Li-ion batteries. [1] In addition, sulfur is light weight, earthly abundant, and nontoxic. Despite its promise, Li-S batteries still suffer from poor cycling performance caused by the polysulfide shuttle process which occurs during the multistep reduction of sulfur to Li2S. Due to the insulating nature of sulfur, this reaction relies on the incorporation of sulfur into a conductive carbon host structure. Various carbons serve as both an electrically conductive pathway, as well as a structural network to accommodate the volumetric expansions associated with cycling. Carbon nanotubes (CNTs), possessing a high electrical conductivity and mechanical strength, are considered a prospective material to serve this role. In spite of extensive efforts, significant gaps still remain in the understanding of the behavior of lithium and reduced sulfur species at the carbon-sulfur interface during working conditions. Therefore a deeper understanding of the fundamental electrochemical reaction mechanisms and kinetics of this system is required to develop the next generation of ultrafast, long-life, high-energy density Li-S batteries. [2] We employ in-situ transmission electron microscopy (TEM) which offers the opportunity to investigate complex reactions in real time while obtaining valuable information on the structural, chemical, and electrical properties during these processes. Using energy-filtering TEM (EFTEM) and scanning TEM (STEM) imaging modes, as well as various analytical methods, such as electron energy loss spectroscopy (EELS), energy-dispersive X-ray spectroscopy (EDXS), and electron diffraction, we have monitored the evolution of these complex structures starting from pristine composites (FIG. 1). Our insitu work utilizes a customized TEM sample holder. The holder has a small piece of metallic lithium attached to a movable piezo-driven STM tip, which served as the anode and reference electrode. A thin layer of Li2O formed during loading, was used as a solid electrolyte. The Li/Li2O tip is then contacted with the S-CNT cathode and cycled through the application of an external bias. In-situ observations indicate complex morphological and compositional changes of both the Li anode and S-CNT cathode during cycling. (FIG. 2) References: [1] P.G. Bruce, et al, Nature Materials 11 (2012), p. 19. [2] V.P. Oleshko, et al, Microsc Microanal 15 (2009), p. 1398. Microsc. Microanal. 21 (Suppl 3), 2015 1514 (a) (b) (c)) Figure 1. (a) Zero-loss energy filtered TEM image of a thick pristine S-MWCNT composite and corresponding SAED pattern of the composite (right inset). (b) HRTEM image showing nanoscale morphologies of the S-MWCNT composites. (c) Overlaid electron spectroscopic S L2,3- map (orange) and C K- map (green) of the S-MWCNT composites (a) (b) (c) Figure 2. In situ TEM frames acquired during lithiation of S-MWCNT composites at negative 3 V relative to the Li electrode. (a) after initial contact of Li/Li2O (left) and S-MWCNT composite (right); (b) after 50 minutes of discharge; (c) after 100 minutes of discharge. All scale bars are 1�m in length.");sQ1[756]=new Array("../7337/1515.pdf","Materials Development Aided by Atomic-Resolution Electron Microscopy","","1515 doi:10.1017/S1431927615008351 Paper No. 0756 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Materials Development Aided by Atomic-Resolution Electron Microscopy Randi Holmestad1, Sigurd Wenner1, Magnus Nord1, Per Erik Vullum1,2 and Calin Marioara2 Department of Physics, Norwegian University of Science and Technology (NTNU), 7491 Trondheim, Norway 2 SINTEF Materials and Chemistry, 7034 Trondheim, Norway In the end of 2013 the TEM Gemini Centre at NTNU and SINTEF, Trondheim, Norway, installed a double corrected JEOL ARM200CF equipped with a large angle EDS detector and a GIF Quantum with Dual-EELS capabilities as a part of the NORTEM investment [1]. This presentation will focus on how we use the high angle annular dark field scanning transmission electron microscopy (HAADF-STEM) technique to get useful information at the atomic scale from different materials. The talk will concentrate on Al alloys as it is an important research activity in the Gemini Centre and supported by Norwegian Aluminum industry. Also a few other examples from functional materials will be shown. Age hardenable aluminum alloys such as 2xxx, 6xxx and 7xxx alloys are important as structural materials for automotive and aerospace applications due to properties like high strength/weight ratio, good formability and weldability, often combined with good corrosion resistance. The alloys acquire strength through the formation of nanoscale, metastable precipitate phases during heat treatment. Our objective is to understand more of the fundamental physics going on at the atomic scale, which governs nucleation, phase stabilization and in general precipitation in these alloys. Strengthening properties of age-hardening precipitates depend on their atomic structure, morphology and coherency with the Al matrix, which is influenced by the alloy composition and the thermo-mechanical history of the material. Therefore, studying precipitates in detail is fundamental in alloy development. Some elements affect the precipitation significantly at low concentration levels, e.g. 0.01%. In recycled Al alloys several such impurity elements can be both beneficial and detrimental to the properties. Through micro-alloying, certain elements can also be added on purpose to improve material properties. The main hardening precipitates in Al�Mg�Si (6xxx) alloys form and grow as needles along <100>Al, and are observed in cross-section when the material is tilted to a <100>Al zone axis. HAADF-STEM is an excellent technique for studying the distribution of heavier elements such as Ag, Cu and Zn in the precipitates [2]. Figure 1 illustrates how aberration corrected STEM gives both high resolution and compositional information of a Q' type precipitate containing Cu. Comparing a high resolution, an uncorrected and a corrected HAADF-STEM image, we see that in the latter every atomic column in the precipitate is resolved, and the distribution of the Cu-containing columns is clear. When Cu, Zn or Ag is added to 6xxx alloys, the precipitate structures often become disordered. Although the disordered precipitates lack periodicity in their cross-sectional planes, they all contain an ordered network of Si atomic columns having a projected hexagonal symmetry. The disordered precipitates consist of fragments of known phases in the Al-Mg-Si(-Cu) alloy system connected through the common Si-network [3]. Other examples are from studies of interfaces in perovskite materials. HAADF-STEM is used to ensure the quality and measure the strain in the epitaxial interface between La0.7Sr0.3MnO3 (LSMO) thin films and SrTiO3 (STO) substrates [4]. Lead free piezoelectric materials are a challenge. A very simple spin coating technique can be used to synthesize piezoelectric K0.5Na0.5NbO3 (KNN) films on top of STO with various substrate orientations (i.e. (001), (011) and (111) STO wafer orientations). The interface between the film and the substrate is one important parameter for the functional properties of the film. 1 Microsc. Microanal. 21 (Suppl 3), 2015 1516 Here we demonstrate how HAADF-STEM can be used to study the interface structure, as shown in figure 2. The presentation will show how advanced TEM methods, and in particular aberration probe corrected HAADF-STEM are used to acquire useful atomic information from our samples [5]. References: [1] http://www.nortem.no/ [2] T Saito, CD Marioara, SJ Andersen, W Lefebvre and R Holmestad, Phil. Mag., 94 (2014) 520. [3] FJH Ehlers, S Wenner, SJ Andersen, CD Marioara, W Lefebvre, C Boothroyd, R Holmestad, J. Mater. Sci. 49 (2014), 6413. [4] M Nord, PE Vullum, M Moreau, JE Boschker, SM Selbach, R Holmestad, and T Tybell, Appl. Phys. Lett. 106 (2015) 041604. [5] The TEM work was carried out on the NORTEM facility. The authors acknowledge Profs. M-A Einarsrud from Materials Science and Engineering and T Tybell from Electrical Engineering, both at NTNU, for providing the perovskite materials. 1 nm 1 nm 1 nm a) b) c) Figure 1. Example of a) HRTEM (JEOL2010F), b) uncorrected HAADF-STEM (JEOL2010F) and c) aberration corrected HAADF-STEM (JEOL ARM200CF) techniques used to image the Q'-phase precipitate structure in an Al-Mg-Si-Cu alloy. a) b) c) Figure 2. HAADF-STEM images of the interface between STO and KNN. (a) The KNN film grows epitaxially on top of the (001)STO substrate. It can be seen that the STO substrate terminates with a SrO layer and the KNN film starts with a NbO2 layer. In (b) and (c) the KNN film is grown on top of (011)STO. The difference between (b) and (c), except that they are acquired at slightly different interface locations, is the dark field STEM collection angles. In (b) the inner and outer angles are 43 and 170 mrad, while in (c) they are reduced to cover the range 37 � 146 mrad. The decrease in collection angles makes the light weight A cations (K and Na) visible as very weak bright dots.");sQ1[757]=new Array("../7337/1517.pdf","Characterization of Various Interfaces Structure in a Titanium Alloy Using Aberration-Corrected Scanning Transmission Electron Microscope","","1517 doi:10.1017/S1431927615008363 Paper No. 0757 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of Various Interfaces Structure in a Titanium Alloy Using Aberration-Corrected Scanning Transmission Electron Microscope Y. Zheng1, R. E. A. Williams1, H. L. Fraser1 1. Department of Materials Science and Engineering, The Ohio State University The chemical and structural interface between phases plays an important role in metallic materials, not only by influencing microstructural evolution due to extra energy arising from lattice distortion near the interface, as well as diffusion of multiple solute elements across the interface, but also by affecting the performance of materials due to interaction with defects. Therefore, it is critical to obtain detailed structural and chemical information about interfaces in order to better manipulate the microstructural evolution and therefore improve the overall performance. In beta titanium alloys, the hexagonal closed pack (hcp) structure, alpha phase, is the most common stable phase. The omega phase is a metastable phase with hexagonal structure found in the body centered cubic (bcc) structure beta phase matrix. The anisotropy present in the alpha/beta interfacial structure influences the morphology of alpha phase precipitate and structural defects, such as dislocations and ledges present at alpha/beta interface, may also alter the precipitate-dislocation interaction and change the deformation mechanism [1]. The structural defects present at the omega/beta interface have been reported to influence the subsequent alpha phase precipitation by means of providing extra driving force [2] and the misfit at the omega/beta interface can also influence the morphology of omega precipitates to adopt an ellipsoidal or cuboidal morphology. Many efforts have been made to characterize these events using conventional transmission electron microscopy (TEM), such as systematic dislocation analysis using diffraction contrast to study alpha/beta interface structure and dark field imaging to study the correlation between pre-formed omega phase and subsequent formed alpha phase. With the development of aberration-corrected scanning transmission electron microscope, atom columns near various interfaces have been characterized directly and the structure of various interfaces in beta titanium alloys has been analyzed. In the first part of this current work, the crystallography and the structure of the habit plane of coarse and refined alpha precipitates in the beta matrix of Ti5553 were studied using probe corrected scanning transmission electron microscope (FEI Titan 80-300). For the first time, in z-contrast HAADFHRSTEM atomic terrace and ledge structures on the habit plane of coarse and refined alpha precipitates were observed directly along [011] beta direction (shown in Fig 1(a)). The lengths of terrace, or the spacing of ledges, are not uniform while conversely the height of ledge is always twice of {112} atom plane spacing. The crystallography analysis indicates that the Burgers vector of disconnection on the habit plane is �[3 -3-5]. It has been determined on coarse alpha precipitate interfaces that misfit dislocations lie on the habit plane, but no misfit dislocation has ever been observed on the habit plane of refined alpha precipitates (shown in Fig 1(b) and 1(c)). The correlation between {0001} atom planes in hcp structure and {011} atom planes in bcc structure and the misfit of two sets of lattice along [0001]alpha/[011]beta direction are different for coarse alpha precipitates and refined alpha precipitates, which could be the reason why two different morphologies and size scale alpha precipitates are formed in Ti5553 [3]. The second part of the work to be presented focused on the structure of the omega/beta interface and its influence on subsequent alpha precipitation in Ti5553. This phenomenon was investigated using probe corrected scanning transmission electron microscope (FEI TitanTM 80-300). As shown in Fig 2(a), from Microsc. Microanal. 21 (Suppl 3), 2015 1518 the center of omega particle to the omega/beta interface, the degree of {222} beta atom planes collapse alters from complete collapse to partial collapse. Atomic terrace and ledge structure are characterized at the omega/beta interface (highlighted by blue dashed line in Fig 2(a)). The presence of ledges along the omega/beta interface is to increase the coherency between hexagonal and bcc phases. These ledges could act as a favorable nucleation site for subsequent alpha precipitates. Such a case is shown in Fig 2(b), one super-refined alpha lath nucleates at the omega/beta interface and grows into beta matrix. This is the first time the nucleation site of alpha precipitation is clearly characterized indicating the influence of pre-formed omega particles [4]. References: [1] R. Shi, et al., Acta Materialia 60 (2012), p. 4172-4184. [2] S. Nag, et al., Acta Materialia 60 (2012), p. 6247-6256 [3] Y. Zheng, et al., Acta Materialia, in preparation [4] Y. Zheng., et al., Acta Materialia, in preparation (a) (b) (c) Figure 1. Filtered HAADF-HRSTEM images showing the alpha/beta interface structure from (a) [011]beta/[0001]alpha direction and (b), (c) [111]beta/ [11-20]alpha direction (a) (b) Figure 2. Filtered HAADF-HRSTEM image showing the structure of (a) omega/beta interface (b) omega/alpha interface");sQ1[758]=new Array("../7337/1519.pdf","Direct Observation of Chemical Pressure in Intermetallic Alloys by Scanning Transmission Electron Microscopy","","1519 doi:10.1017/S1431927615008375 Paper No. 0758 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Direct Observation of Chemical Pressure in Intermetallic Alloys by Scanning Transmission Electron Microscopy Adedapo A. Oni1, Xiahan Sang1, Aakash Kumar2, Susan B. Sinnott2, James M. LeBeau1 1. 2. North Carolina State University, Materials Science and Engineering, Raleigh, USA. University of Florida, Materials Science and Engineering, Gainesville, USA. As solute atoms are added to intermetallic compounds, local chemically induced pressure develops and results in atomic displacements. These distortions play a critical role in defining the mechanical behavior of these materials, for example, by impeding dislocation motion. While indirect methods, such as diffraction determined pair distribution functions, can access the average structure, spatially resolved information is lost. Scanning transmission electron microscopy (STEM), on the other hand, possess excellent spatial resolution, but image distortions often limit the ability to accurately and precisely measure projected crystal structure. In this talk, we will discuss combining high-angle annular dark-field (HAADF) STEM and density functional theory (DFT) calculations to investigate the correlation between the atom column chemistry and lattice distortion observed in a Ni-based superalloy intermetallic phase. We select a model system, � precipitates in a Ni-Al-Cr superalloy, which adopts the L12 structure (space group Pm-3m), see Figure 1(a). Using a probe corrected FEI Titan G2 S/TEM, we apply revolving STEM (RevSTEM) to accurately and precisely remove drift distortion [1]. Atomic resolution energy dispersive X-ray spectroscopy (EDS) reveals that Cr preferentially occupies the Al sub-lattice (Figure 1(b)) [3]. Moreover, we will show that lattice strain in Figure 1(c), and crystal distortion (tetragonality), is likely connected to the random fluctuation of Cr concentration in the Al sub-lattice of the projected STEM images. To further elucidate the mechanism for relative lattice displacement, relaxation of pure Ni3Al and Ni3(Al,Cr) supercells are calculated using DFT. These results indicate a relative displacement of the near neighbor Ni when Cr is substituted for Al atoms, as depicted in Figure 2(a). HAADF images are simulated while implementing the static displacement derived from the DFT calculations. The Ni3(Al,Cr) supercell was generated by randomly distributing Cr on Al sub-lattices, ensuring that the overall concentration of Cr is approximately the same as the sample. Subsequently, the average Al-Al nearest-like-neighbor (NLN) distances were measured and outlined on the Al sub-lattice in Figure 2(b). From inspection, Ni3(Al,Cr) displays a larger distribution of the NLN distances compared to the pure Ni3Al of the same thickness. Note, the Cr-rich columns (high intensity) shows larger average NLN distances and the Al rich columns (low intensity) have lower average NLN distances. The average Al-Al NLN distance plotted against the Al sub-lattice intensity (Figure 2(c)) shows a correlation between the distance and atom column intensity only for Ni3(Al,Cr). Within this context, we will examine the correlation between atom column chemistry and displacement for experimental results as a function of solute species and sample thickness. Further, we will discuss how these results can help bridge theoretical and experimental approaches toward understanding the structural chemistry of intermetallic compound alloys [4]. Microsc. Microanal. 21 (Suppl 3), 2015 1520 1510 References: [1] X Sang and J. M. LeBeau, Ultramicroscopy 138 (2014) 28-35. [2] X. Sang, A. A. Oni, J. M. LeBeau, Microsc. Microanal. 20 (2014) 1764. [3] A. A. Oni et al, Appl. Phys. Lett., 106 (2015) 011601. [4] The authors acknowledge support from the Air Force Office of Scientific Research (Grant No. FA9550-12-1-0456). We acknowledge the use of the Analytical Instrumentation Facility (AIF) at North Carolina State University, which was supported by the State of North Carolina and the National Science Foundation. Figure 1. (a) RevSTEM image of � Ni3(Al,Cr) along <001> with inset showing projected unit cell. (b) Atomic resolution energy dispersive X-ray spectroscopy (EDS) maps. (c) xx and yy strain maps. The scale bars represent 1 nm. Figure 2. (a) A 2 x 2 x 2 relaxed DFT supercell of the Ni3(Al,Cr) structure illustrating the chemical pressure effect of Cr atoms on the near neighbor Ni atoms. (b) HAADF-STEM simulations of pure Ni3Al and Ni3(Al,Cr) with the average NLN distance outlined around each Al sub-lattice atom column. The intensity of the Ni sub-lattice was clipped to highlight the intensity of the Al sub-lattice. The scale bar represent 500 pm. (c) Average Al-Al NLN distances versus atom column intensity from simulated images.");sQ1[759]=new Array("../7337/1521.pdf","Physical and Practical Challenges in Analytical Electron Tomography of Aluminum Alloys","","1521 doi:10.1017/S1431927615008387 Paper No. 0759 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Physical and Practical Challenges in Analytical Electron Tomography of Aluminum Alloys A. Orthacker1,2, G. Haberfehlner1, J. T�ndl3, M.C. Poletti3, L. J. Allen4 and G. Kothleitner1,2 1. Graz Centre for Electron Microscopy, Graz, Austria 2. Institute for Electron Microscopy and Nanoanalysis, Graz University of Technology, Graz, Austria 3. Institute of Materials Science and Welding, Graz University of Technology, Graz, Austria 4. School of Physics, University of Melbourne, Parkville, Victoria 3010, Australia For a thorough understanding of a material, investigations at the nanoscale are often essential. Analytical techniques like electron energy loss spectroscopy (EELS) and energy dispersive X-ray spectroscopy (EDXS) in probe corrected scanning transmission electron microscopy (STEM) can reveal important chemical information necessary for the development and improvement of high-tech materials. The integrative character of the signal acquired through transmission, however, might hide important structural details of the material, relevant for its properties. Those details can be revealed through electron tomography, where the data is acquired at different tilt angles and, after alignment, reconstructed to form a full 3D model of the material under investigation. The combination of both techniques, analytical STEM and tomography, gives full insight into structure and composition of a material [1]. This novel technique is not yet well established and various experimental challenges remain, especially if a quantitative analysis of the material is the ultimate goal. From previous investigations at the atomic scale it is well known that channeling effects have a great influence, not only on STEM image intensities, but also on the analytical signals detected [2]. The subject of this investigation are bulk metallic samples. Even if the tomographic reconstructions of the material of interest are not carried out at atomic resolution, but rather at the nanoscale, the question of the influence of channeling on the detected intensities remains. In this study an aluminum alloy containing scandium (Sc) and zirconium (Zr) rich nano-precipitates was investigated at different stages of ageing. High resolution STEM and analytical EELS and EDX tomographic investigations were carried out. For the latter, problems such as sample damage, noisy spectra and the difficult detection of very low Zr concentrations were successfully tackled. The resulting 3D elemental reconstructions deliver otherwise inaccessible information on the sample's chemistry and structure. To approach quantitative analytical tomography the influence of channeling on the detected signal intensities needs to be understood, to make sure its effect on elemental quantification results is either incorporated in the analysis or the experimental conditions are chosen to minimize such effects. Therefore a comparative study is performed on a rod shaped sample with an off-axis nano - precipitate. Reconstructions from different sets of tilt series performed using different experimental conditions such as tilt angles, convergence angles and defocus values are conducted and compared to investigate the reproducibility of analytical results under different channeling conditions. [1] G. Haberfehlner et al, Nanoscale 6 (2014), p. 14563. [2] G. Kothleitner et al, PRL 112, (2014), p. 85501 Microsc. Microanal. 21 (Suppl 3), 2015 1522 [3] The authors thank the Austrian Cooperative Research Facility, the European Union (7th Framework Programme: ESTEEM2) and the Austrian Research Promotion Agency FFG (TAKE OFF project 839002) for funding. [4] Special thanks go to Prof. Matthew Weyland and Dr. Scott Findlay (Monash University, Clayton, Victoria 3800, Australia) for their support regarding quantitative ADF-STEM imaging and simulations. Figure 1. Left and right top: STEM HAADF images of the aluminum alloy and nano-precipitates. Bottom right: EDX spectra from a precipitate (grey) and the aluminum matrix (black). Scandium and zirconium peaks suggest that those two elements are found at the sites of the precipitates. Figure 2. Segmented volumes from reconstructions of EDX signals of an aluminum alloy with nanoprecipitates after ageing for 72h at 500�C. The reconstructions were performed using a total variation minimization algorithm after processing X-ray intensities, which are subject to channeling effects.");sQ1[760]=new Array("../7337/1523.pdf","Implementation of Atomic Resolution Electron Tomography of a Needle Sample.","","1523 doi:10.1017/S1431927615008399 Paper No. 0760 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Implementation of Atomic Resolution Electron Tomography of a Needle Sample. M.C. Scott1, W. Theis2, Rui Xu1, Li Wu1, Chien-Chun Chen1, Colin Ophus3, Peter Ercius3 and Jianwei Miao1 1. Department of Physics & Astronomy and California NanoSystems Institute, and University of California, Los Angeles, CA 90095, USA. 2. Nanoscale Physics Research Laboratory, School of Physics and Astronomy, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK. 3. National Center for Electron Microscopy, Molecular Foundry, Lawrence Berkeley National Laboratory, Berkeley, California 94720, USA. Electron tomography has become a ubiquitous technique for three-dimensional (3D) characterization in the physical and life sciences. In materials science, improved stability and resolution in scanning transmission electron microscopy (STEM) have made this imaging configuration preferable for tomography of nanocrystals. Recently, lattice and crystalline grain structure [1] and atomic resolution defect structures [2] of nanoparticles have been reconstructed in 3D from a tilt series of annular dark field (ADF)-STEM images obtained on an uncorrected Titan STEM. 3D resolution can be further improved by use of an aberration corrected microscope, as was recently demonstrated in the atomic resolution reconstruction of FePt nanoparticles [3]. Reconstructions of FePt particles from data obtained on the TEAM I microscope at Lawrence Berkeley National lab not only resolved most of the atoms in the particles, but also differentiated between the two atomic species in the particles [3]. So far, these high resolution 3D reconstructions of nanoparticles, reconstructed from data with a single tilt axis. Experimental difficulties that limit resolution in single tilt tomography include the so-called "missing wedge"- a wedge of missing information that arises from the fact that the sample cannot be tilted a full �90� due to the geometry of most sample holders. Additionally, to image nanoparticles as described above, the sample is suspended on a substrate, which contributes additional signal that negatively impacts signal to noise ratio and image resolution, especially at high tilt angles. Equally sloped tomography (EST) is an iterative, Fourier-based algorithm that uses equally sloped angular increments, rather than equal angle as is more common in tilt axis tomography. EST, which was originally conceived for phase retrieval in coherent X-ray tomography, can recover missing information, and can also accurately arrive at a solution even with noisy input data [1, 2, 4]. EST makes no assumptions about a sample's crystalline structure or symmetry, and therefore it is possible obtain local atomic resolution information in 3D with this technique. Although the effects of the missing wedge of information and loss of image quality due to the substrate can be alleviated by advanced reconstruction algorithms such as EST, these two major drawbacks will always negatively affect resolution in tilt axis tomography. However, difficulties can be removed by changing the sample geometry to a needle shape and rotating about the axis of the needle. By employing the TEAM stage, a custom piezo-controlled 5-axis (x, y, z, and ) tilt stage, a needle sample can be rotated a full �180�, and by varying both and , an arbitrary rotation axis can be chosen. The TEAM stage, therefore, can deliver not only superior vibrational stability, but also can access tilt geometries Microsc. Microanal. 21 (Suppl 3), 2015 1524 unavailable to traditional single tilt sample holders [5]. A tungsten needle sample with tip diameter of <10nm was prepared by electrochemical etching and then mounted in a holder appropriate for the TEAM stage. The tip was imaged in ADF-STEM mode in the TEAM I. To collect the data, custom scripting software was used to calculate the appropriate set of tilt angles to rotate about the desired axis, and to roughly center and focus while delivering minimal electron dose. A set of equally sloped angles was calculated such that the tip was rotated about the [011] direction during the dataset. At each angle, two images were taken. Linear sample drift and scan distortion were corrected before reconstruction via EST. Fig 1 shows the 0� image of the dataset. By removing the missing wedge and background contribution from the substrate, reconstructions of the tungsten needle show improvement over similar datasets obtained in a single tilt geometry. Although the tungsten tip was reconstructed from only 62 projections, the resolution and contrast of the resultant reconstruction are greatly improved, even compared to reconstructions from single tilt tomography of nanoparticles with more projections and a higher electron dose. Additionally, the sample geometry in this experiment is similar to that of atom probe tomography, and therefore this technique can be applied to a wide variety of materials. [1] M.C. Scott et al, Nature 483 (2012), p. 444-447 [2] C.-C. Chen et al, Nature 496 (2013), p. 74-77 [3] M.C. Scott et al, Microscopy and Microanalysis 20 (2014), p. 804-805 [4] C. Zhu et al, Phys. Rev. B 88 (2013), p. 100201 [5] P. Ercius et al, Microscopy and Microanalysis 18 (2012), p. 676-683 [6] This work was mainly supported by the U.S. Department of Energy (Grant No. DEFG02-13ER46943). This work was also partially supported by NSF (DMR-1437263) and Office of Naval Research (N00014-14-1-0675). Experiments were performed at the Molecular Foundry, which is supported by the Office of Science, Office of Basic Energy Sciences of the U.S. Department of Energy under Contract No. DE-AC02--05CH11231. Figure 1. ADF-STEM image of a sharp tungsten tip, imaged along the [111] direction.");sQ1[761]=new Array("../7337/1525.pdf","The Oxidation State of Nanophase Fe Particles Produced by Space Weathering as Revealed Through Aberration-Corrected Transmission Electron Microscopy","","1525 doi:10.1017/S1431927615008405 Paper No. 0761 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Oxidation State of Nanophase Fe Particles Produced by Space Weathering as Revealed Through Aberration-Corrected Transmission Electron Microscopy Michelle S. Thompson and Thomas J. Zega Lunar and Planetary Laboratory, Department of Planetary Sciences, University of Arizona, 1629 E University Blvd, Tucson, AZ, 85721 Mineral grains on the surfaces of planetary bodies such as the Moon and near-Earth asteroids are continuously modified by collisional processes such as micrometeorite impacts and radiation processing from solar energetic ions. These processes are collectively known as space-weathering and they alter the microstructural and microchemical characteristics of minerals on planetary surfaces [1]. These features were first identified in lunar soils returned by the Apollo missions e.g., [2]. Modification of crystal chemistry and structure results in changes to optical properties of materials. Space weathering has occurred over the history of the solar system and so has progressively altered surface optical properties of the moon and other airless bodies. Understanding the nature of such alteration and how it occurs is important for determining the mineralogy of planetary bodies from remote sensing data and understanding the processes operating on their surfaces [3]. A primary characteristic of space weathering is the development of iron nanoparticles, termed npFe. The npFe particles are typically <10 nm in size and occur in the amorphous rims of mineral grains and within glassy agglutinate grain interiors. Prior studies have shown that the glassy agglutinates contain Fe in multiple oxidation states [4], but the individual nanoparticles have been assumed to contain only metallic Fe, i.e., Fe0. Here we use electron energy-loss spectroscopy (EELS) coupled to a monochromated and aberrationcorrected transmission electron microscope (TEM) to analyze the oxidation state of npFe particles in several lunar soils. Understanding the oxidation state of Fe in the npFe can provide insight into the formation conditions of the nanoparticles and the evolution of space weathering processes on airless body surfaces. We embedded <1 mm grains of three Apollo lunar soils in low-viscosity epoxy and prepared them for TEM analysis via ultramicrotomy. EELS analysis was performed with the 100 keV monochromated and aberration-corrected Nion UltraSTEM at Arizona State University, equipped with a Gatan Enfinium spectrometer. We used a convergence semi-angle of 30 mrad, a 0.2 nm probe size, 3 mm spectrometer entrance aperture, and a collection angle of 45 mrad. With the monochromator on, we achieved an energy resolution between 200 to 300 meV. We measured individual nanoparticles via line profiles with a 0.1 eV/channel dispersion and dwell times between 0.01s/px to 0.5 s/px. EELS spectra were collected at the Fe L2,3 core-loss edge and were compared to metal (Fe0), w�stite (FeO), and hematite (Fe2O3) standards for reference to Fe0, Fe2+, and Fe3+, respectively. We quantified the oxidation state of Fe in the npFe particles using a flux-weighted linear, least-squares fitting routine programmed in MatLab. Fitting over the L3 edge, which exhibits edge-onset and multiplet structure that is indicative of oxidation state, e.g., [5], the routine generated Fe0/Fetotal, Fe2+/Fetotal, and Fe3+/Fetotal ratios for each measured spectrum. Microsc. Microanal. 21 (Suppl 3), 2015 1526 We acquired line profiles from >120 individual particles, over 40 from each of our three samples, as shown in Figure 1. An example of an extracted spectrum is shown in Figure 2. Spectral processing included removal of background and deconvolution of the zero loss peak using the Fourier-Ratio method. Once the core-loss edge was isolated from the background continuum (using a straight line from 702 to 725 eV), the Matlab code was used to simulate the experimental spectra from the standards. We quantified each fit by assessing the residual intensity, expressed as R2, after subtraction of the simulation from the experimental spectrum. Fits having an R2 >0.9 are presented here. The EELS data reveal that Fe occurs in multiple oxidation states in these lunar soils. The immature soil, which is the sample with the shortest exposure time to interplanetary space, contains nanoparticles composed primarily of Fe0. The submature soil, which is the sample with intermediate exposure time, contains nanoparticles with mixtures of Fe0 and Fe2+ in varied ratios. Moreover, the EELS spectrum images show that several individual particles in this submature soil contain an oxidized (Fe2+) rim surrounding a reduced Fe0 core. In comparison, the most mature sample contains nanoparticles composed of mixtures of Fe2+ and Fe3+. Structural measurements were used to corroborate the EELS data, e.g., Fig. 3. These results indicate there is increasing oxidation of Fe in the nanoparticles in soils with increasing maturity. This trend suggests that space weathering products such as Fe nanoparticles are evolving microchemically and microstructurally after their formation. The mechanisms for oxidation of the nanoparticles and the implications for reflectance spectroscopy will be discussed at the meeting. References: [1] Hapke B. Journal of Geophysical Research- Planets 106 (2001), p. 10,039. [2] Keller L. P. and McKay D. S. Geochimica et Cosmochimica Acta 61 (1997) p. 2331. [3] Pieters C. M. et al, Meteoritics and Planetary Science 35 (2000), p. 1101. [4] Keller L. P and Clemett S. J. Lunar and Planetary Science Conference Proceedings (2001), p. 2097. Figure 2: EELS spectrum of a nanoparticle [5] Zega T. J. et al, American Mineralogist 88 (2003), p. from mature soil sample 79221 showing Fe 1169. L2,3 edge. Shown are the measured sample (blue), the simulated best fit (red), the residuals (black), and the components fitting the spectrum, in this case Fe2+ (gold) and Fe3+ (green) are shown. This particle is composed of 77% Fe2+ and 23% Fe3+. Figure 1: High-angle annular dark field (HAADF) image of bright npFe particles in a darker glassy matrix. The green dashed line represents the locations where matrix spectra were collected; orange shows those of the nanoparticle spectra. Figure 3: HRTEM image of a nanoparticle in lunar soil sample 79221. Spacing measurements and possible identifications are shown for two directions.");sQ1[762]=new Array("../7337/1527.pdf","Aberration-Corrected STEM-EELS Measurements in Fe-bearing Silicate Glasses","","1527 doi:10.1017/S1431927615008417 Paper No. 0762 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Aberration-Corrected STEM-EELS Measurements in Fe-bearing Silicate Glasses K. D. Burgess1, B. T. De Gregorio2, M. D. Dyar3, M. C. McCanta4 and R. M. Stroud5 ASEE Postdoc, Naval Research Laboratory, Washington, DC 20375 USA (kate.burgess.ctr@nrl.navy.mil) Nova Research, Inc., Alexandria, VA 22308 USA 3 Mount Holyoke College, South Hadley, MA 01075 USA 4 Tufts University, Medford, MA 02155 USA 5 Naval Research Laboratory, Washington, DC 20375 USA 2 1 Space-weathered materials exhibit a wide range of complex nanometer-scale features caused by solar wind irradiation and micrometeorite impacts, such as nanophase iron metal particles (npFe0) and amorphous rims [1,2]. The high brightness and focused probe of the aberration-corrected STEM enable fast acquisition of data and low detection limits in EELS, and aberration-corrected STEM measurements of the oxidation state of individual nanoparticles have been reported [3]. However, the highly focused beam can cause significant damage to sensitive samples, including breaking bonds, which can change valence or coordination states of atoms, or cause loss of material [4]. Returned space-weathered planetary samples are of limited availability. Thus, prior determination, on analog samples, of the best experimental conditions, such as accelerating voltage, beam current, scan speed, and exposure time, is necessary for obtaining high-quality data with minimal beam damage to the returned samples. We have a large number of homogenous, well-characterized (e.g., using microprobe, x-ray diffraction, M�ssbauer, x-ray absorption spectroscopy), synthetic glasses prepared for EELS measurement. The glasses range in composition from komatiite to rhyolite (43-78 wt% SiO2) and have been equilibrated in atmospheres buffered at iron-w�stite (IW), quartz-fayalite-magnetite (QFM), in air, and in CO2. Small pieces of each sample have been embedded in epoxy and microtomed, then placed on Quantifoil carbon support film TEM grids. The thinnest regions (usually less than ~50 nm) of the glass shards are used in the measurements. To collect EELS data, we use PRISM, the NION UltraSTEM at the Naval Research Laboratory equipped with a Gatan Enfinium ER EEL spectrometer (0.3 eV energy resolution) at a range of conditions including 60 kV, 100 kV and 200 kV, and 0.01-1.5 nA. The glass samples measured here are easily affected by the electron beam, both from knock-on damage and radiolysis; for measurements at 200 kV, knock-on damage and field-induced migration and phase separation occur quickly [5]. Scans shown here were done using a 40 pA probe current and dwell times between 0.001 and 0.1 s dwell for core-loss measurements. Depending on total electron dose and doserate, an O pre-peak at 528 eV appears, then disappears at higher dose (Fig. 1a). Sequential scans (0.01 s dwell) over the same region causes the Fe L2 peak to shift from 706 eV to slightly higher energy, indicating oxidation from Fe2+ to Fe3+ (Fig. 1b). A peak at ~5 eV related to the presence of C * transitions [6] in the sample decreases (Fig. 1c), and loss of C could cause the oxidation of the Fe. Fig. 2 illustrates the changes over regions with different acquisition parameters; integrated signals for the final scan at 0.001 s/pixel show correlations between O pre-peak intensity, Fe concentration and low-loss C * intensity. Additional measurements at lower accelerating voltages (60 kV and 100 kV), where radiolysis should dominate over knock-on damage and carbon losses may be minimized, are planned for comparison. References: Microsc. Microanal. 21 (Suppl 3), 2015 1528 1518 [1] Keller L. P. and McKay D. S. (1997) Geochim. Cosmochim. Acta, 61, 2331-2341. [2] Pieters C. M. et al. (2000) Meteorit. Planet. Sci., 35, 1101-1107. [3] Thompson M. S. and Zega T. J. (2014) LPS XLV, Abstract #2834. [4] Egerton R. F. (2011) 3rd Ed. p4. [5] Jiang et al. (2003) Phys. Rev. B, 68, 064207. [6] Egerton R. F. (2011) 3rd Ed. p135. [7] The authors acknowledge funding from NASA SSERVI RIS4E. Figure 1. (a) O K-edge spectra in silicate glass showing the appearance and then decrease in the 528 eV pre-peak associated with O2 * bonds indicative of damage to the sample. (b) Fe L2,3 edges showing oxidation of the sample with increased dose and differences between regions with different thickness at the same electron dose. (c) Low-loss region of spectrum before and after Fe L-edge acquisition. The peak at ~5 eV and the Fe M2,3 edge both decrease with dose. Energy resolution is 0.4 eV. Probe current was 40 pA. Figure 2. (a) HAADF image of the silicate glass sample showing Spectrum Imaging regions. (b) Integrated line profiles for the Fe L2 edge (705-708 eV), C c (4-7 eV), and the O K-edge pre-peak at 528 eV. The final scan was done with a dwell of 0.01 s/px. The small regions between the scanned areas have increased Fe and a larger * peak. The region scanned at 0.02 s/px also has a larger * peak prior to the thickness change. The HAADF image to the right is from the yellow box in (a) and shows phase separation in the regions with the highest electron doses.");sQ1[763]=new Array("../7337/1529.pdf","Ti Valence States in Al,Ti-rich Pyroxenes from Comet Particles: Anchoring Ti ELNES Spectra with Vanadium Thin-films","","1529 doi:10.1017/S1431927615008429 Paper No. 0763 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Ti Valence States in Al,Ti-rich Pyroxenes from Comet Particles: Anchoring Ti ELNES Spectra with Vanadium Thin-films Dave Joswiak and Don Brownlee Department of Astronomy, University of Washington, Seattle, WA USA 98195 Detailed investigations of particles captured from comet Wild 2 by the Stardust spacecraft and a giant cluster interplanetary dust particle (IDP) of probable cometary origin have shown that high temperature, moderately refractory materials are present in outer Solar System bodies where Kuiper Belt objects accreted [1,2]. Unraveling the physical and chemical processes of how these particles formed is vital toward understanding comets. The mineral pyroxene, a common phase in these refractory grains and believed to be a nebular condensation product, often contains significant amounts of the transition element Ti which can adopt different oxidation states, thus pyroxenes can provide evidence of redox conditions during formation of their host grains. Refractory minerals in comet samples are typically micron to submicron in size and electron energy loss near-edge spectroscopy (ELNES) combined with transmission electron microscopy (TEM) is an effective technique to measure the valence states of Ti in the pyroxenes. With electron energy loss spectroscopy (EELS), this is possible because of the chemical shift that occurs in the Ti core-loss edges with oxidation number, typically 2 eV per oxidation state [3]. Interpretation of ELNES spectra, however, can be complex and difficulties are manifest in a least two ways. First, differences in the fine-structure of Ti core-loss spectra due to differences in coordination, octahedral volume, polyhedron distortions and site symmetries [3] complicate ELNES spectra comparisons. Secondly, instrument instabilities and local magnetic field perturbations induce energy drift in the EELS spectra during acquisition obscuring the absolute energy loss positions of ELNES spectra. It is this latter category that we address here with a technique that allows direct comparison between Ti ELNES spectra in both standards and examined pyroxenes. The technique uses the element V to fix the Ti ELNES peaks on the energy loss axis. Because the V peaks have fixed energy positions and their core-loss edges occur above Ti (~58eV) they can be used to precisely position the Ti peaks whose energy positions vary with valence. This method removes the uncertainties due to energy drift during EELS spectrum acquisitions and anchors the Ti core-loss ELNES spectra on the energy-loss axis thus all measured spectra can be directly compared to one another. To implement the technique, a V-coated carbon film (V < 10 nm) on a separate TEM grid is placed directly below the TEM grid containing the pyroxene sample. To estimate the Ti valence states in the pyroxenes we use the "white-line" method where Ti L2/L3 core-loss ratios are measured and applied to a L2/L3 vs Ti4+/(Ti4+ + Ti3+) calibration curve established from Ti oxides with known oxidation states (Fig. 1). One eV-wide regions-of-interest (ROIs), chosen to maximize the L2/L3 ratios from the Ti oxide standards, are used for the measurements. Because the local energy environments in the pyroxenes are different than those in the Ti oxide standards which were used to generate the calibration curve, the measured Ti4+/(Ti4+ + Ti3+) ratios in the pyroxenes are considered semi-quantitative. Microsc. Microanal. 21 (Suppl 3), 2015 1530 We have applied this technique to Ti-rich pyroxenes on <10 �m particles from comet Wild 2 and an IDP of probable cometary origin (Fig. 2). In Fig. 3 detailed Ti ELNES spectra are compared to one another and are shown with the Ti oxide standards TiO2 (100% Ti4+) and Ti2O3 (100% Ti3+). We measured Ti4+/(Ti4+ + Ti3+) ratios of 0.39 and 0.55 in the comet Wild 2 sample and 0.27 � 0.89 in pyroxenes in the giant cluster IDP. The data indicate wide-ranging oxidation states in pyroxenes in refractory materials from comets. Because the pyroxene measurements were obtained from grains in close physical proximity to one another (within 10 �m particles), this suggests that refractory-rich materials in comets formed in the solar nebula in a region with significant spatial and/or temporal variations in oxygen abundance. Variable oxidation states in pyroxenes in refractory inclusions were obtained in some carbonaceous chondrite meteorites [5]. References: [1] S. Simon et al. (2008) Meteoritics and Planetary Science 43: 1861-1877. [2] D. J. Joswiak and D. E. Brownlee (2014) 45th Lunar and Planet. Sci. Conf., abstract #2282. [3] E. Stoyanov, F. Langenhorst and G. Steinle-Neumann (2007) Amer. Mineral. 92: 577-586. [4] H. Tan, J. Verbreeck, A. Abakumov and G. Van Tendeloo (2012) Ultramicro.. 116: 24-33. [5] K. A. Dyl, J. I. Simon and E. D. Young (2011) Geochim. et Cosmochim. Acta 75: 937-949. 15 10 L2/L3 5 0 0 0.2 0.4 Ti4+/(Ti3+ + 0.6 Ti4+) 0.8 1 Fig 1 Ti2O3 Fig 3 Ti3O5 TiO2 TiO2 IDP IDP IDP Fig 2 TiL3 TiL2 TiO2 std Comet Wild 2 IDP VL3 VL2 IDP Wild 2 Wild 2 Ti2O3 stds Fig. 1) "White-line" calibration curve used to estimate Ti4+/(Ti4+ + Ti3+) ratios in pyroxenes in comet samples. Fig. 2) Cometary pyroxene and Ti oxide (standards) spectra showing both Ti and V peaks. Vertical line shows VL3 peak anchored to 514.4 eV. Fig. 3) Ti ELNES spectra of cometary pyroxenes (green, red) and Ti oxide standards (black). ROIs shown with dotted lines.");sQ1[764]=new Array("../7337/1531.pdf","Insights Into Regolith Evolution from TEM Studies of Space Weathering of Itokawa Particles.","","1531 doi:10.1017/S1431927615008430 Paper No. 0764 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Insights Into Regolith Evolution from TEM Studies of Space Weathering of Itokawa Particles. Eve L. Berger 1, Lindsay P. Keller 2 1 2 GeoControl Systems, Inc. - Jacobs JETS Contract - NASA Johnson Space Center, Houston, TX USA NASA Johnson Space Center, Houston, TX USA Exposure to solar wind irradiation and micrometeorite impacts alter the properties of regolith materials exposed on airless bodies [1]. However, estimates of space weathering rates for asteroid regoliths span many orders of magnitude [e.g., 2, 3]. Timescales for space weathering processes on airless bodies can be anchored by analyzing surface samples returned by JAXA's Hayabusa mission to asteroid 25143 Itokawa. Constraints on timescales of solar flare particle track accumulation [4, 5] and formation of solar wind produced ion-damaged rims [6] yield information on regolith dynamics. Multiple electron transparent thin sections of Itokawa particles RAQD02-0211 (0211), RA-QD02-0125 (0125), and RA-QD02-0192 (0192) were prepared using a hybrid ultramicrotomy-focused ion beam (FIB) technique [7] on a Leica EM UC6 ultramicrotome and an FEI Quanta 3D dual beam FIB-SEM. This technique results in whole slices of particles - preserving both edge and interior features (e.g., Fig. 1a). Transmission electron microscope (TEM) analyses of the FIB sections, which allow for accurate determination of solar flare particle track densities and rim characteristics (e.g., width, crystallinity), were done on the JEOL 2500SE 200kV field emission STEM. All instruments are housed at NASA JSC. All three particles are olivine-rich (Fo70) with minor sulfides and show features of space weathering: adhering mineral grains and melt particles, solar flare particle tracks, and continuous solar wind damaged rims. The rims are compositionally similar to the cores of the grains, structurally disordered, and nanocrystalline [8]. The rim thickness varies and track density gradients (fig. 2) are observed across the particles. The track density gradient across particle 0211 correlates with the rim thickness (Figs. 1b). The highest track density (3.4�109 tracks/cm2) is on the side of the particle with the thickest rim (~80nm), while the lowest track density (9.2�108 tracks/cm2) correlates with the thinnest rim (~40nm). Particle 0192 also shows a track density gradient (2.9�109 to 1.1�109 tracks/cm2) and has comparable rim widths to particle 0211. Exposure ages, based on the track production rate of 4.1�1.2�104 tracks/cm2/year at 1AU [4] are: ~80,000 years for 0211, ~70,000 years for 0192, and ~24,000 years for 0125. The heterogeneous distribution of the space weathering effects on two Itokawa particles is consistent with both particles maintaining a relatively fixed orientation in the Itokawa regolith throughout the time they were being irradiated by incoming solar flare particles. 1.) Cosmic ray exposure ages for other Itokawa particles are relatively young (1-1.5 Ma) [9-11] when compared to other LL chondrites (8-50Ma) [12]. The CRE age indicates that the regolith was stable at meter depths over ~106 years. 2.) Based on track production rates in olivine at 1AU [4], Itokawa particles recorded solar flare tracks over timescales of <105 years. The track gradient indicates that over this timescale, the particles maintained a relatively stable orientation at mm to cm depths. 3.) Solar wind produced ion damaged rims are predicted to become amorphous and to reach thicknesses of 100nm within 103-104 years [6]. The rims on the Itokawa particles are not amorphous and have thicknesses of <60-70nm, suggesting a residence time on the surface (with direct exposure to the solar wind) of less than ~103 years. The continuous rims found on these grains, which have varying thicknesses, indicate that all sides of the Microsc. Microanal. 21 (Suppl 3), 2015 1532 particles have had direct exposure to the solar wind. The uppermost surface of the Itokawa regolith was sufficiently dynamic that while grain rotation must have occurred, the particles were not lost to space. The Itokawa particles were shielded from direct exposure to the solar wind, at mm to cm depths, over timescales of 104-105 years. The track gradients in these particles suggest that the regolith in the MusesC region of Itokawa experienced little overturn; rather, it was relatively stable at these depths. However, late in their history, over <103 years, all sides of the particles were directly exposed to the solar wind (as evidenced by continuous ion-damaged rims), which requires grain rotation on Itokawa's surface. [1] B Hapke, Journal of Geophysical Research 106 (2001), p. 10039. [2] M Willman et al., Icarus 208 (2010), p. 758. [3] P Vernazza et al., Nature 458 (2009), p. 993. [4] E L Berger and L P Keller, LPSC XLVI (2015), p. 1543. [5] E L Berger and L P Keller, LPSC XLVI (2015), p. 2351. [6] R Christoffersen and L P Keller, LPSC XLVI (2015), p. 2084 [7] E L Berger and L P Keller, Microscopy Today 23 (2015), p. 18. [8] L P Keller and E L Berger, Earth, Planets & Space 66 (2014), p. 71. [9] M M M Meier et al., LPSC XLV (2014), p. 1247. [10] K Nagao et al., LPSC XLIV (2013), p. 1976. [11] K Nagao et al., Science 333 (2011), p. 1128. [12] T Graf and K Marti, Meteoritics 29 (1994), p. 643 1a Fig. 1a. Bright field image (BFI) of Itokawa 0211, FIB section 1. 1b. BFI mosiac from 0211 (red arrow in fig. 1a). The thickest rim and highest track density are at the top, while the thinnest rim and lowest track density are at the bottom. Tracks are highlighted in red for ease of viewing. 1b Fig. 2. Track density vs. depth. The trend line slopes are consistent with the Itokawa particles experiencing only mild tumbling while exposed on the surface of the asteroid; see [4, 5].");sQ1[765]=new Array("../7337/1533.pdf","Novel Microtomy Methods For Small, Hard And Precious Extraterrestrial And Terrestrial Rocky Materials","","1533 doi:10.1017/S1431927615008442 Paper No. 0765 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Novel Microtomy Methods For Small, Hard And Precious Extraterrestrial And Terrestrial Rocky Materials D.E. Brownlee1 and D. Joswiak1 1. Dept of Astronomy, Univ. of Washington, Seattle, WA, USA The use of Focused Ion Beam (FIB) methods has provided revolutionary capabilities for making thin-sections for TEM, ion probe and beam-line microanalyses of extraterrestrial materials. Although FIB methods have been a remarkable enabling-technology in the study of extraterrestrial materials [1,2], the production of thin-sections by diamond knife ultramicrotomy of hard rocky materials [3] still plays important roles. One advantage of microtomy is that it can efficiently, inexpensively and quickly provide large numbers of slices from small precious samples. For example, a single 5�m diameter particle can be cut into as many as one hundred 50 nm serial sections, ideally preserving all of a complex sample. Adjacent slices can be distributed to different investigators and analyzed by different microbeam analytical methods. This capability is particularly important for complex, valuable samples such as small particles returned to Earth by comet and asteroid sample return missions and also for small complex samples such as presolar grains found in meteorites and interplanetary dust. The ability to use adjacent slices provides the capability to do coordinated isotopic mapping with the ion probe and detailed TEM analyses of submicron components. Similar coordination can also be done with slices and the remaining bulk sample (potted butt) and this is invaluable for certain analyses such as precision isotopic analyses that require greater sample thickness than normal FIB or microtome sections provide. Additional benefits of microtomy are the lack of radiation damage or beam deposition effects that can occur in FIB samples. We will describe three microtomy methods that we use for cutting small hard extraterrestrial samples and mounting them on TEM grids. Intact sections of hard materials: When brittle materials such as silicate minerals are cut with diamond knives, they fracture because of the sharp bend at the knife-edge and >micron brittle components often fracture into a ordered fabric of elongated micron-sized shards. TEM analyses of microtomed rocky materials are usually done on shards and some shards experience substantial movement that disturbs original textural relations. In some cases, shards are displaced beyond the original particle perimeter. Shard formation in hard phases cannot be prevented, even with a 35� diamond knife, but we have developed a simple method that keeps the shards in their original positions and the sample sections remain intact, resembling the craquelure (crazing crack texture) on old paintings. This process is done by putting a thin plastic film on the surface of the sample block just before each slice is cut and then dissolving the film off afterwards. Small samples are mounted in acrylic or epoxy bullets that are trimmed so that the cut-face is a 100�m x 200�m trapezoid mesa with near vertical sides about 30�m high. Usually the mesa contains a single 3-10�m sample. Just before each cut is made, we manually touch the mesa with the tip of a millimeter-sized triangular piece of filter paper that has just been dipped in a 0.1% solution of collodion in amyl acetate. The microtome is stopped and the coating is done under microscopic observation and left to dry, all in only a few Microsc. Microanal. 21 (Suppl 3), 2015 1534 seconds time. The collodion film is later removed by dissolution with condensed chloroform vapor in a special sub-boiling grid-washing apparatus. The entire process is actually very simple, it works extremely well and provides excellent sections even of hard silicate minerals such as olivine that are at least as large as 50�m. The sections are excellent for microanalysis but the shards, while preserved in place, are tilted relative to each other and this affects large-scale HAADF and conventional imaging where thickness or crystal orientation contributes to image contrast. FIB sections provide better textural information on sizes larger that a few microns, but this microtome method provides generally excellent slices of hard materials that are often difficult or impossible to section by conventional microtomy. Although we have not tried it, it likely that this method could be used on samples that are much larger that can be sectioned with the FIB. Alignment: A major issue with microtomy of small priceless samples has always been the challenge of aligning individual or serially sectioned samples in the center of grid square or bars that hold carbon support films. One method that we have found to be very effective is to pick sections off the microtome boat with a loop (the Diatome Perfect Loop) and then use a kolinksy sable hair to force the sections to their desired location as the water dries on the grid. We pick up the slices with the loop, then pick up a carboncoated grid and place the loop holder on a movable magnetic clamp to center the suspended loop under a stereomicroscope. We then use a fine sable hair mounted on a needle, controlled by a simple glide-plate micromanipulator, to press down on the section or sections so that they can be aligned as the water evaporates. When the water has dried to only a few microns thick, the section can be precisely positioned above the grid support film, before the water evaporates. We speed up the drying process using filter paper wicks and by heating by a focused 10w halogen lamp that also plays a critical role in illuminating floating sections that can be <50 nm thick and sometimes difficult to see. Concentrating submicron particles: A unique role of microtomy is the ability to make sections of large arrays of micron-sized particles all mounted on a flat plane. This might have many applications but we developed the "Rubber Stamp" method to find rare submicron presolar grains that are contained in primitive interplanetary dust particles. A single 10�m dust particle is a loose aggregate that might contain 105 submicron grains, only a tiny fraction of which are presolar. The particle can be gently broken down into its diverse components and dispersed over a wide flat area. With the rubber stamp we concentrate component grains onto a 50�m square flat silicone rubber mesa, formed by casting clear silicone rubber into a flat-bottomed 50 �m hole. Under a stereo microscope, the rubber mesa is repeatedly pushed against the dispersed grains. Particles adhere to the rubber analogous to using tape to remove cat hair from a shirt. When the desired density of particles is obtained, the mesa with its monolayer of particles is embedded in Weld-On 40 acrylic or epoxy and sectioned. With proper alignment, the result is sections of large numbers of spatially concentrated micron and smaller particles all lying in the same plane. This method could also be used to concentrate very rare small components that are chemically isolated from a larger sample and left in a dispersed state on a flat surface. [1] R. Stroud, Cometary Dust in Astrophysics (2003), p6011. [2] G. Graham et al. Meteoritics and Planetary Science 43 (2008), p561. [3] J. Bradley and D. Brownlee, Science 231 (1986), p1542.");sQ1[766]=new Array("../7337/1535.pdf","Microscopy and Chemical Analysis of Topological Insulator Bi2Se3 and Topological Crystalline Insulator SnTe Nanostructures","","1535 doi:10.1017/S1431927615008454 Paper No. 0766 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microscopy and Chemical Analysis of Topological Insulator Bi2Se3 and Topological Crystalline Insulator SnTe Nanostructures Judy J. Cha1,2 1. 2. Department of Mechanical Engineering & Materials Science, Yale University, New Haven, CT, USA Energy Sciences Institute, Yale West Campus, West Haven, CT, USA Topological insulators and topological crystalline insulators are a new quantum matter whose band structure belongs to a different topology compared to a regular, trivial insulator [1, 2]. The physical consequence of this is manifestation of conducting surface or edge states that are spin-polarized and Dirac-dispersive. The unique electronic property of the surface states opens opportunities to study previously inaccessible fundamental condensed matter phenomena such as Majorana fermions [3] and magnetic monopoles [4] and to design future spin-based device applications. A plethora of experiments have been carried out to investigate the exotic electronic properties of the surface states in topological insulator Bi2Se3 and topological crystalline insulator SnTe. Nanostructured Bi2Se3 and SnTe are ideal to study and exploit the properties of the surface states due to the large surface-to-volume ratios, which enhance the surface effects. In addition, the nanostructure morphology can play an important role in confining the surface electrons to follow well-defined paths for interference experiments. Magnetic and charge-compensation doping has also shown to be important for controlling the surface state property. This highlights the critical need to carefully characterize the morphology, atomic structure, and chemical composition of topological insulator and topological crystalline insulator nanostructures. In this talk, I will present structural and chemical characterizations of Bi2Se3 nanoribbons and SnTe nanoplates using scanning transmission electron microscopy and showcase how their structures determine their electronic properties. For chemical analysis, X-ray energy dispersive spectroscopy (EDX) was proven very useful due to the high atomic number elements such as Bi and Te that make up topological insulators and topological crystalline insulators. Monochromated electron energy-loss spectroscopy (EELS) was used to probe plasmonic and optical properties of these materials in thin platelet forms. A few examples I will discuss are 1) analysis of chemical compositions of Bi2(SexTe1-x)3 nanoribbons and nanoplates and resulting changes in the lattice constant, plasmon excitations, and electron carrier density (Fig. 1) [5], 2) Electron energy-loss spectroscopy investigation of plasmonic and optical property changes due to dielectric molecule intercalation into Bi2Se3 nanoribbons (Fig. 2) [6-8], and 3) characterization of Indium doping concentrations in SnTe nanostructures, which induces superconductivity (Fig. 3) [9, 10]. The structure-property relation of topological insulator and topological crystalline insulator nanostructures, elucidated by analytical electron microscopy studies, provides an essential guide to improve the material quality for enhanced electrical transport properties of these materials. Microsc. Microanal. 21 (Suppl 3), 2015 1536 [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] A X.-L. Qi and S.-C. Zhang, Reviews of Modern Physics 83, 1057 (2011). L. Fu, Phys. Rev. Lett. 106, 106802 (2011). F. Wilczek, Nature Physics 5, 614 (2009). X.-L. Qi, R. Li, J. Zang, and S.-C. Zhang, Science 323, 1184 (2009). J. J. Cha, D. Kong, S.-S. Hong, J. G. Analytis, K. Lai, and Y. Cui, Nano Letters 12, 1107 (2012). K. J. Koski et al, J. Am. Chem. Soc. 134, 7584 (2012). J. J. Cha et al, Nano Letters 13, 5913 (2013). J. Yao et al, Nature Communications 5, 5670 (2014). J. Shen, Y. Jung, A. S. Disa, F. J. Walker, C. H. Ahn, and J. J. Cha, Nano Lett. 14, 4183 (2014). J. Shen, Y. Xin, and J. J. Cha, arXiv:1410.4244 (2014). B C D E Figure 1. (A) Chemical analysis of Bi2(SexTe1-x)3 nanoplates by EDX. (B) TEM image of Bi2Se3 (top) and Bi2Te3 (bottom). (C) EELS spectra showing the surface plasmons of Bi2(SexTe1-x)3 nanoplates. (D) Surface plasmon peak linearly scales with the Se/Te ratio. (E) Carrier density scales with the Se/Te ratio. A B C Figure 2. (A) Intercalation into the van der Waals gap of Bi2Se3. (B) Plasmon propagation in a Bi2Se3 nanoplate by EELS mapping. (C) Optical property change by pyridine intercalation into Bi2Se3. A HAADF B 2.4 C 2.4 0T 0.3 T R () 1.8 1.5 5m 2m In Sn Te 0 100 T (K) 200 300 R () 2.1 2.1 1.8 1.5 1.5 2.0 2.5 3.0 T (K) Figure 3. (A) Chemical analysis of In-doped SnTe by EDX. (B) Superconductivity induced by the In doping. (C) Field-cooled resistivity measurement confirms the superconductivity.");sQ1[767]=new Array("../7337/1537.pdf","Characterization of the Oxide-Semiconductor Interface in 4H-SiC/SiO2 Structures using TEM and XPS","","1537 doi:10.1017/S1431927615008466 Paper No. 0767 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of the Oxide-Semiconductor Interface in 4H-SiC/SiO2 Structures using TEM and XPS Joshua Taillon1, Karen Gaskell2, Gang Liu3, Leonard Feldman3, Sarit Dahr4, Tsvetanka Zheleva5, Aivars Lelis5, and Lourdes Salamanca-Riba1 1. 2. University of Maryland, Materials Science and Engineering, College Park, MD, USA. University of Maryland, Chemistry and Biochemistry, College Park, MD, USA. 3. Rutgers University, Institute for Advanced Materials, New Brunswick, NJ, USA. 4. Auburn University, Physics, Auburn, AL, USA. 5. US Army Research Laboratory, Sensors and Electron Devices Directorate, Adelphi, MD, USA. Silicon carbide (SiC) is a very promising wide bandgap material for high power and high-temperature applications due to its native SiO2 oxide, high thermal conductivity, and high bulk electron mobility [1]. Performance and reliability in these devices is limited however, by their low carrier mobility within the FET channel caused by electrically active defects at the oxide interface. A number of methods have been developed for improving the mobilities in fabricated devices, and the most prevalent of these is incorporation of nitrogen at the oxide interface through a post-oxidation nitric oxide (NO) anneal (POA) [2]. While the mechanisms behind the resulting device improvement and the interfacial structure are a matter of active research, the true nature of the interface is still not totally clear. Distinct transition layers at the interface between SiC and SiO2 have been previously observed by electron energy loss spectroscopy (TEM-EELS) and an inverse relationship was observed between the length of NO annealing time and the width of the transition region (wTL) [3]. In order to gain a more thorough understanding of the chemical and electronic structure at this interface, devices in a variety of orientations and both with and without a POA have been investigated in this work using high resolution TEM (HRTEM), high-angle annular dark field imaging (HAADF-STEM), EELS, and angle resolved xray photoemission spectroscopy (AR-XPS). The aim of this study is to analyze how both device orientation (leading to atomic scale roughness) and post-oxidation processing affect various atomic scale properties, including the transition layers' composition and strain, as well as the electronic configuration and contributions from possible interfacial states. HRTEM inspection and EELS measurements were made on a series of six samples, with varying orientations and processing. Oxide layers were grown on the epitaxial layer of 4H-SiC wafers with one of three orientations: Si-face (Si-terminated [0001] surface normal) on-axis, Si-face miscut by 4� (the orientation used in commercial applications), and a-face ([1120] surface normal). For each of these � orientations, one sample was analyzed prior to any NO POA process, and another that had received a 2hr POA. wTL was measured by monitoring the edge onset energy of the Si-L2,3 EELS edge across the interface, as described in [3]. As shown in Figure 1a, the wTL values on the Si-face were practically identical regardless of NO anneal, contradicting expectations. For all the Si-face samples, wTL values were significantly lower than those measured previously on similar devices, indicating an overall improvement and that current mobility-limiting defects in state of the art devices may not be accurately probed through observation of the Si-L2,3 edge. The a-face devices did have slightly lower wTL values, indicating that roughness (Figure 1b-g) at the interface may be a limiting factor in these devices. To further probe the nature of the interfacial transition region, chemical depth profile experiments using AR-XPS have been performed. Using a spin-etch technique refined from previous work on the Si/SiO2 interface [4], layers of oxide can be removed with sub-nm precision to carefully approach the SiC/SiO2 interface with very little damage or adverse chemical modification (verified by identical XPS spectra Microsc. Microanal. 21 (Suppl 3), 2015 1538 a after etches on thicker oxide layers This proc o s). cess was us to comp sed pare the nea ar-interface r region of o oxides grown on Si-face 4H-SiC, bot with and without an a n th w additional NO POA. O D Detailed AR R-XPS scans were perfor rmed on eac element i question t investigat difference arising ch in to te es f from the PO process and from cha OA a anges in oxi thickness or interfac strain. T Si-2p (Fi ide s cial The igure 1h) s signal revealed three di istinct states at the inte s erface, corre esponding to "bulk" SiO2 and SiC and an o C, interfacial st tate. An add ditional state was presen in sample that were fully etched at an energy level e nt es d, s slightly lowe than the bulk oxide, but this was not observ in any sa er b b s ved amples that retained the original e o oxide, indica ating that besides the int terfacial stat all Si ato in the o tes, oms oxide are ful complem lly mented by O atoms. It is suspected that these lo i ower energy states result from a mon t nolayer of oxygen that c cannot be r removed by HF etching, though this state emer rged at an en nergy level about 1eV higher than previous r reports [7]. E Examining the N-1s sca (Figure 1i), 4 distin compone t ans nct ents were ob bserved, and the N con d ntent was f found to be localized ent l tirely near th interface (not remove d by the etch he ( hing process in a silicon nitrides) n l like configur ration. Furth hermore, ana alysis of the C-1s comp e ponents reve ealed signific cant C-O bo onding at t interface, which appe to be red the ears duced upon NO POA. [8 8] F Figure 1: (a) Measureme of wTL at the SiC/SiO2 interface obtained by analysis of Si-L2,3 EELS line ) ent t O y S s scans perpen ndicular to th interface (error bars re he ( epresent 0.95 confidence intervals). (b-g) Representative 5 e H HRTEM ima ages of each interface. Conditions an orientatio are indicat in the fig C nd on ted gure. Where known, r representativ peak field effect mobi ve d ility () valu are given for similar samples [5-6 (h) AR-X ues n 6]. XPS ( (scan was 70 relative to normal) res 0� sults showing spin-orbit split components of the Si-2p orbita (i) Ng al. 1 componen present in an NO POA sample. 1s nts n A [ J Cooper et al, IEEE Trans. Elect [1] tron Devices 49 (2002), p. 658. s [ P Jamet, S Dimitrijev and P Tann J. Appl. Phys. 90 (20 [2] v ner, 001), p. 5058. [ JA Taillo et al, J. Ap Phys. 11 (2013), p. 044517. [3] on ppl. 13 p [ DB Fenn DK Bieg [4] ner, gelsen and RD Bringans, J. Appl. Ph 66 (1989 p. 419. R , hys. 9), [ G Liu et al, IEEE Ele [5] ectron Devic Lett. 34 (2 ce 2013), p. 181 1. [ J Senzaki et al, IEEE Electron De [6] i E evice Lett. 23 (2002), p. 13. 2 [ S Dhar et al, J. Amer [7] t rican Chemic Society 131 (2009), p 16808. cal 1 p. [ The autho gratefully acknowled funding from ARL c [8] ors dge contracts W9 911NF-11-2 2-0044 and W W911NF0 07-2-0046. JAT addition J nally acknow wledges fund ding through the NSF GR RFP, grant D DGE 132210 06.");sQ1[768]=new Array("../7337/1539.pdf","Hybrid Calcium Phosphate Neuron-Like Structures under the Microscope","","1539 doi:10.1017/S1431927615008478 Paper No. 0768 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Hybrid Calcium Phosphate Neuron-Like Structures under the Microscope Vesna Srot1, Montserrat Espanol2,3,4, Zhitong Zhao2,3, Peter A. van Aken1 and Maria-Pau Ginebra2,3,4 1. 2. Max Planck Institute for Intelligent Systems, Stuttgart Center for Electron Microscopy, Stuttgart, Germany Biomaterials, Biomechanics and Tissue Engineering Group, Department of Materials Science and Metallurgical Engineering, Technical University of Catalonia, Barcelona, Spain 3. Centre for Research in Nanoengineering, Technical University of Catalonia, Barcelona, Spain 4. Biomedical Research Networking Center in Bioengineering, Biomaterials, and Nanomedicine, Barcelona, Spain Many biominerals display unique morphologies where inorganic building blocks interplay with organic matter. Organic molecules dictate the complex and sophisticated shapes of biominerals which take place at ambient conditions [1]. The close interplay between organic and inorganic components endows biominerals with unique architectures and exceptional properties [2,3]. The calcium phosphate system is of special interest due to its widespread appearance in biomineralization, especially in the formation of teeth and bones. However several fundamental aspects are still not clearly understood. Amorphous calcium phosphate as a precursor for biomineralisation is still being under consideration. Detecting early stages of biomineralisation is a real challenge. Due to unstable nature of amorphous calcium phosphate phases no unambiguous proof of its presence was shown until now. For our study, different organic molecules were used to produce stabilized very unique calcium phosphate neuron-like structures [4]. Electron energy-loss spectroscopy (EELS) and energy-dispersive X-ray spectroscopy (EDX) combined with scanning transmission electron microscopy (STEM) imaging at high spatial and high energy resolution, as well as energy-filtered TEM (EFTEM) were used to characterize these structures using Zeiss SESAM and JEOL ARM200F microscopes at different accelerating voltages. Lower magnification annular dark-field (ADF)-STEM images of typical calcium phosphate neuron-like structures are shown in Figure 1. The dense core of the structure and the filaments are clearly visible. According to our high-resolution TEM and electron diffraction data the structures are amorphous. EFTEM experiments using the low-loss EELS region were conducted on several neuron-like structures. A bright-field (BF)-TEM image and a corresponding calcium (Ca) map are shown in Figure 2. The intensity of the background corrected Ca-M2,3 edges in the energy range between 32 and 40 eV was used to form the Ca map. In the course of our work also low-loss and core-loss EELS experiments were performed on these neuron-like structures as well as on standard materials and will be discussed. References: [1] S Weiner and PM Dove in "An Overview of Biomineralisation Processes and the Problem of the Vital Effects" (2003), Rev Mineral Geochem 54, 1-29. [2] UGK Wegst and MF Ashby, Philos Mag 84 (2004), 2167. [3] AP Jackson and JFV Vincent, J Mater Sci 25 (1990), 3173. [4] M Espanol, ZT Zhao, J Almunia and M-P Ginebra, J Mater Chem B 2 (2014), 2020. Microsc. Microanal. 21 (Suppl 3), 2015 1540 Figure 1. (a, b) ADF-STEM images of the calcium phosphate neuron-like structures. Figure 2. (a) BF-TEM image of neuron-like structures with (b) corresponding Ca map obtained by using the background subtracted intensity of Ca-M2,3 edges.");sQ1[769]=new Array("../7337/1541.pdf","Exploring Lithium-ion Battery Performance through in situ Characterization","","1541 doi:10.1017/S143192761500848X Paper No. 0769 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Exploring Lithium-ion Battery Performance through in situ Characterization Dean J. Miller1, Arnaud Demortiere1, Lifen Wang1, Jianguo Wen1, Jun Lu2, Khalil Amine2 1. 2. Electron Microscopy Center - Center for Nanoscale Materials, Argonne National Laboratory. Chemical Sciences and Engineering Division, Argonne National Laboratory. We have developed an approach for in situ and in operando characterization of single Li-ion battery cathode particles through which a single particle can be characterized in detail during electrochemical cycling, providing a better correlation between performance and microstructural evolution. In this work, we focused on high capacity cathode materials that have a graded composition, with an Nirich core and an Mn-rich periphery. The concept for these materials with a general composition of LiNi1-x-yCoyMnxO2 is that the Ni-rich core provides high capacity while the Mn-rich periphery minimizes detrimental interaction with the electrolyte. [1,2] In addition to high capacity, these materials exhibit better long-term performance with less "fade" in capacity over many cycles compared to, for example, LiNi0.8Co0.15Al0.05O2 ("NCA"), which shows much more significant capacity fade. Our in situ single particle studies suggest one of the mechanisms for capacity fade in NCA is particle fracture that occurs during cycling. [3] In this work, we applied our approach to graded cathode materials to see if this was an important factor in their improved performance. An example of the electrochemical performance of a single LiNi1-x-yCoyMnxO2 compositionally graded particle (hereafter referred to a "gradient" material) is shown in Figure 1. In this experiment, the single particle was cycled in situ, and the microstructure was evaluated at various points during cycling. The voltage versus discharge capacity plot shows an increase in capacity for the first 13 cycles, but then a decrease for the 14th cycle. The microstructure of the particle after the 14th cycle, shown in the inset, reveals a moderate degree of cracking (grain-to-grain separations). This microstructure is generally similar to that observed in NCA, but with a notable difference in that the particle cracking tends to be radial rather than isotropic, as is the case for NCA. This difference is illustrated more definitively in Figure 2, which shows a comparison between gradient material and NCA material for both in situ cycled single particles and for particles harvested from cycled coin cells. This data confirms the aggressive, isotropic fracture mode in NCA that can lead to electrical isolation of individual grains. In contrast, the gradient material exhibits a much more fractureresistant microstructure with anisotropic fracture that should be less deleterious to performance. This comparison also confirms that the behavior we observe in our single particle measurements is indeed representative of larger scale "real world" battery systems. We suggest that the microstructural basis for the radial fracture in gradient materials is based on their grain structure and morphology. The gradient materials exhibit a radial grain structure and refined grain sizes that are more resistant to fracture. Furthermore, the fractures that do eventually appear are less detrimental to performance because they do not lead to isolated fragments that can lose electrical connectivity, as is the case for NCA. References: [1] Y.-K. Sun, et al., Adv. Funct. Mater. 20, 485-491 (2010) [2] Y.-K. Sun, et al., Nat. Mater. 11, 942-947 (2012). [3] D.J. Miller, et al., Adv. Energy Mater. 3 (8) 1098-1103 (2013) F ig ur e 1: Sc he m at ic ill us tr at io n of co m po sit io na lly gr ad ed Li Ni 1x- Microsc. Microanal. 21 (Suppl 3), 2015 1542 [4] This work was carried out in the Electron Microscopy Center-Center for Nanoscale Materials, which is supported by the U. S. Department of Energy, Office of Science under Contract No. DE-AC0206CH11357 Cycle"#" 1" 5" 10" 14" 13" Figure 1. Voltage vs capacity plot measured for a single LiNi1-x-yCoyMnxO2 gradient particle and scanning electron microscope image of the particle structure after 14 cycles. Figure 2. Scanning electron microscope images of NCA and gradient material after cycling, comparing material cycled 50 times in a coin cell with a corresponding single particle cycling measurement for the number of cycles as indicated.");sQ1[770]=new Array("../7337/1543.pdf","Dynamic Study of Sodiation Process in Single Crystalline -MnO2 Nanowires","","1543 doi:10.1017/S1431927615008491 Paper No. 0770 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Dynamic Study of Sodiation Process in Single Crystalline -MnO2 Nanowires Yifei Yuan1, Anmin Nie2, Wentao Yao2, Reza Shahbazian-Yassar2 Department of Materials Science and Engineering, Michigan Technological University, 1400 Townsend Dive, Houghton, Michigan 49931 USA 2. Department of Mechanical Engineering-Engineering Mechanics, Michigan Technological University, 1400 Townsend Dive, Houghton, Michigan 49931, USA -MnO2 is widely applied as an energy storage electrode in rechargeable batteries due to its unique 2�2 tunneled structure that facilitates diffusion of charge carriers [1]. By now, it is unclear how the intercalated charge carriers such as Li+, Na+ and Mg2+ interact with the tunnel-based host due to the lack of atomic scale understanding of the tunnel configuration and the complicated effect from generally existing tunnel stabilizers (like K+). In this paper, using aberration-corrected scanning transmission electron microscopy (ACSTEM) to cross sectioned K+-stabilized -MnO2 nanowires, the 1�1 and 2�2 tunneled structures as well as defective 2�3 and 2�4 tunnels are clearly demonstrated at atomic level. An open cell design in TEM for dynamic study of -MnO2's sodiation process confirms that an intermediate phase NaxMnO2 will first appear upon sodiation and finally the tunneled structure will totally collapse, generating Mn2O3 polycrystals embedded in Na2O matrix. The originally existing tunnel stabilizer K+ will be partially removed upon sodiation, as shown in Figure 1. It also shows that defective 2�3 and 2�4 tunnels function as the fast sodiation path during initial Na+ intercalation stage. This study provides fundamental understanding of the tunnel-charge carrier interaction and reveals the structural evolution mechanism of sodiation in -MnO2. The key role of 2�3 and 2�4 tunnels on increasing the discharge rate is also demonstrated, shedding light on potential tunnellevel modification for improving the overall performance of tunnel-based electrodes. Reference: [1] Yuliang Cao et al, Advanced Materials 23 (2011), p. 3155 1. Microsc. Microanal. 21 (Suppl 3), 2015 1544 Figure 1 (a) in-situ TEM image of one -MnO2 nanowire being sodiated with four areas circled as b, c, d and e; (b-e) corresponding selected area diffraction patterns from areas b, c, d and e as indicated in (a); (f) EDS mappings of Mn, O, Na and K inside one partially sodiated K+stabilized -MnO2 nanowire.");sQ1[771]=new Array("../7337/1545.pdf","TEM Study of Heavily Twinned Cu3Pt Nanoparticles","","1545 doi:10.1017/S1431927615008508 Paper No. 0771 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 TEM Study of Heavily Twinned Cu3Pt Nanoparticles Goran Drazi, Marjan Bele, Andraz Pavlisic, Primoz Jovanovic, Milena Zorko, Barbara Jozinovic, Miran Gaberscek Laboratory for Materials Chemistry, National Institute of Chemistry, Ljubljana, Slovenia The efficiency of proton exchange membrane fuel cells (PEM-FC) is mainly limited by the activity of the cathode catalyst for oxygen reducing reaction (ORR). Various materials based on Pt � transition metal alloys are used for such application [1, 2] where it was found that electrocatalytic activity can be substantially increased through the formation of ordered intermetallic compound (like Cu3Pt) and the formation of core-shell structure [3]. The physicochemical properties of metal nanoparticles strongly depend on their size, shape and the internal crystal structure. Nanocrystals with twinned structures can exhibit different properties comparing to single-crystal counterparts, due to a large defect-to-volume ratio. The lattice strain caused by twin defects could have a significant impact on the electronic structure of metal nanocrystals influencing the interatomic distances and thus the energy levels of bonding electrons, which in turn determines the catalytic, electrical and optical properties [4]. It was reported, that in the case of the parallel twined structure of platinum nanoparticles, the conductivity is increased significantly, up to 6 times compared to pure (defect free) Pt. [5] Some particles could exist as single twin (just one twin defect) or as multiple twin particles. Two types of multiple (repeated) twinning are known in metal nanoparticles: lamellar and cyclic. Lamellar twinning is characterized by parallel contact twins repeating continuously one after another while cyclic twinning requires nonparallel coplanar composition planes forming decahedron and icosahedron morphologies. Using a novel, modified sol-gel method the ordered (Pm3-m) intermetallic Cu3Pt nanoparticles for catalytic oxygen reduction reaction applications were prepared. Varying specific parameters like chemical composition, temperature profile and atmosphere of synthesis the degree of ordering, presence of Pt rich layer (skin) at the surface and the amount of particles with lamellar twins could be tailored. The material obtained exhibits up to 5-fold improvement of mass activity and a 9-fold improvement of specific activity compared to the Pt/C benchmark. In this work we report the structure and properties of Cu3Pt nanoparticles with high amount of lamellar twins, investigated with a Cs probe-corrected JEOL ARM 200 CF microscope. In Fig. 1 LAADF (lowangle annular dark-field) and HAADF (high-angle annular dark field) micrographs of nanoparticles with (111) twin boundaries are displayed. With LAADF technique the local strain field can be detected through its distortion of adjacent atomic sites and subsequent dechanneling of the electron beam from the atomic columns [6]. We can see that at the twin boundaries the contrast is much higher indicating large local strain field. From XRD measurement it was found that beside ordered (Pm-3m) Cu3Pt and disordered Cu3-xPt (Fm-3m) phases also some other phase in small amount is present. In Fig. 2 the XRD diffractogram splitting of (011) peak is visible. Using selected area electron diffraction we found that spots with identical d-values (2,55 � and 2,72 �) are always present in pair on the same reciprocal vector g, indicating they are in close crystallographic relationship. Fig. 2b, c represents SAED patterns where two pairs of spots and bright and dark-field images of the twined particles are displayed. Those findings could be explained with rhombohedral distortion of the Cu3Pt structure near the (111) twin boundary. The influence of twin boundaries on electrocatalytic properties will be discussed in details. Microsc. Microanal. 21 (Suppl 3), 2015 1546 References [1] HA. Gasteiger, et al, Appl. Catal., B56 (2005), p. 9. [2] IEL. Stephens, et al, J. Am. Chem. Soc., 133 (2011), p. 5485. [3] D. Wang, et al, Nature Mater., 12 (2013), p. 81. [4] Y. Xia, et al, Angew. Chem. Int. Ed. 48, (2009), p. 60. [5] R. Esparza, et al, Revista Mexicana de Physica, 55, (2009), p. 339. [6] P.J. Phillips, et al, Ultramicroscopy 116 (2012), p. 47. a. b. c. Figure 1. a. � LAADF micrograph of Cu3Pt nanoparticle, b. � LAADF micrograph of the (111) twin boundary and c. � HAADF micrograph of the same twin boundary 2 theta d values 24,07� 3,69 � 33,20� 2,71 � 34,30� 2,62 � 35,50� 2,53 � 42,25� 2,14 � hkl 001 x 011 x 111 d1 = 2,55 � d2 = 2,72 � a. b. c. Figure 2. a. XRD pattern and listing of d-values (inset) of heavily twined sample, b. selected area electron diffraction pattern of Cu3Pt heavily twined particles, c. � dark-field, bright-field micrograph of twined particle using lower pair of the labeled diffraction spots in b.");sQ1[772]=new Array("../7337/1547.pdf","The Role of Aberration-Corrected STEM in the Characterization of Oxide Cathode Materials","","1547 doi:10.1017/S143192761500851X Paper No. 0772 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Role of Aberration-Corrected STEM in the Characterization of Oxide Cathode Materials P.J. Phillips1 , D.P. Abraham2 , J. Bareno2 , C. Kim3 , T. Yi3 , J. Cabana3 , R.F. Klie1 ~ 1 Department of Physics, University of Illinois at Chicago, Chicago IL 60607 2 Chemical Sciences & Engineering, Argonne National Laboratory, Argonne, IL 60439 3 Department of Chemistry, University of Illinois at Chicago, Chicago, IL 60607 The development of next-generation batteries for use in advanced applications (such as hybrid electric vehicles) continues to be a sought-after goal and thus a relevant research topic, in particular with respect to Li-ion and multivalent-ion batteries. In the case of Li-ion, their full potential has yet to be reached given their inherent cycling performance issues, such as voltage instability and capacity fading. On the other hand, by extending battery technology to include multivalent ions, it is proposed that such systems can surpass the energy storage capabilities of Li-ion technology, hence the impetus to identify materials which can accommodate (de)intercalation of such multivalents. The present contribution will discuss a variety of oxide cathode materials in the context of scanning transmission electron microscopy (STEM) analysis. Specifically, these materials require characterization of chemical compositions/gradients, electronic structure (ion valence), structural disorder/rearrangements, and in the largely-unknown realm of multivalent ions: whether or not intercalation has actually occurred. Thus, STEM-based methods are quickly becoming the most promising characterization tools, with techniques available which include the direct imaging of both heavy and light elements, and both energy-dispersive X-ray (EDX) and electron energy loss (EEL) spectroscopies. Two Li-based cathode materials will be discussed: a Mn-rich spinel and the monoclinic Li2MnO3. Nanocrystalline spinel particles were engineered with an Al-rich passivating epitaxial shell [1] to curb the common issue of degradation at the electrolyte/electrode interface; as such, the STEM will focus on directly imaging the shell in low angle annular dark field (LAADF) to confirm its epitaxial nature and composition analysis via EDX, as presented in Fig. 1. Li2MnO3 will be discussed in its pristine, electrochemically cycled, and in-situ electron beam irradiated states [2]; the latter allows for single particle tracking of the dynamic processes occurring upon Li and O loss from the material. In this case, the focus will be on tracking structural evolution (and hence the use of annular bright field (ABF) STEM) as well as the Mn valence and O loss via high-resolution EELS (Fig. 2). Additionally, multivalent-based cathodes will be discussed. For these cases, generally a combinatorial approach is required, consisting of LAADF/ABF imaging, EELS to determine valence, and highspatial-resolution EDX spectroscopy to determine the nature of intercalation; furthermore, in some cases, the imaging voltage may change between 80/200 kV, depending on the desired output. It is important to note that much of the above analysis would not be possible without the use of advanced STEM instruments, outfitted with both EEL and EDX capabilities. In the present case, the UIC � aberration-corrected JEOL JEM-ARM 200CF STEM instrument, capable of 0.73 A spatial and 0.35 eV energy resolution, equipped with a large angle silicon drift EDX detector, was employed [3]. References [1] C. Kim, P.J. Phillips, L. Xu, A. Dong, R. Buonsanti, R.F. Klie, J. Cabana, Chem. Mater. 27 (2015) 394�399. [2] P.J. Phillips, H. Iddir, D.P. Abraham, R.F. Klie, Appl. Phys. Lett. 105 (2014) 113905. [3] Supported by an MRI-R2 grant from the National Science Foundation (Grant No. DMR-0959470). [4] Support is acknowledged from the Joint Center for Energy Storage Research, an Energy Innovation Hub funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences. Microsc. Microanal. 21 (Suppl 3), 2015 1548 Figure 1: LAADF STEM image of the core-shell nanocrystals (left), along with a higher magnification view of the epitaxial shell (right); the Mn atomic positions are indicated with solid red spheres, while O and Li atomic positions are represented by hollow blue and yellow spheres, respectively, in the spinel structure. The Li site would not show intensity in this LAADF view, indicative that there are heavier elements present (Al). Finally, an EDX map and line scan of a core-shell nanoparticle, showing Al enrichment at the surface (shell). Figure 2: a) LAADF image showing two pockets of damage formed during electron beam irradiation; LAADF (b) and ABF (c) of a damaged region, showing anomalous intensity in what would be Li positions in the pristine oxide (arrowed); this is indicative of Mn moving into the Li sites. EELS O K- and Mn L-edge results form a moderately damaged (blue) and a severely damaged (red) region, with severely damaged material showing a decrease in the Mn valence and the presence of O vacancies; the table at right summarizes EELS results from various stages of damage, I being the pristine material, V being the most severely damaged (red EELS curve).");sQ1[773]=new Array("../7337/1549.pdf","Challenges for ABF-STEM characterization of Li battery materials.","","1549 doi:10.1017/S1431927615008521 Paper No. 0773 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Challenges for ABF-STEM characterization of Li battery materials. Barnaby D.A. Levin1 and David A. Muller1,2 1. 2. Department of Applied and Engineering Physics, Cornell University, Ithaca, NY, USA, 14853. Kavli Institute for Nanoscale Science, Cornell University Annular Bright Field (ABF) STEM has attracted considerable interest for its potential to directly image light atomic columns within a crystal structure [1]. One promising application for ABF is imaging lithium columns in the crystalline intercalation electrodes of lithium-ion batteries, such as LiFePO4, at various states of charge. However, low energy barriers for vacancy diffusion (< 0.2 eV), which are required by design in order to transport and store Li ions reversibly [2], make intercalation electrodes vulnerable to vacancy-enhanced displacement (VED) [3], a form of knockon damage that can lead to rapid redistribution of Li by the electron beam. High electron doses are needed for high resolution imaging of Li, but this also causes high VED. Here we assess the limits placed on the quantification of Li column occupancy in LiFePO4 by the competition between obtaining a sufficient dose for imaging, and the displacement of imaged atoms. Our methodology applies to other Li-ion cathode materials, and other energy materials containing light atoms. From Poisson statistics, the dose-limited resolution [4] scales as the square root of the maximum allowable dose, and inversely with Li column contrast. The contrast depends on, inter alia, beam voltage, spot size, column occupancy, and sample thickness. For most imaging modes such as EELS or HAADF, higher beam voltages lead to lower contrast as scattering cross-sections are reduced. However, multi-slice simulations of LiFePO4 (Fig. 1) show that for a fixed probe size, higher beam voltage leads to better Li contrast in ABF. This is a channeling effect for samples thicker than about 10 nm. The increased contrast and reduced damage cross section (Fig 2a) with beam voltage favor a higher beam voltage for imaging Li in ABF. Fig 2a shows that the knock-on damage cross section [5] for VED, which redistributes Li in the sample, is ~10x larger than the cross section for surface sputtering, which removes Li from the sample. Fig 2b shows the intersection of the family of curves for dose needed to resolve a given contrast change vs the dose required to displace a given fraction of Li atoms. Multi-slice simulations indicate that a resolution of better than ~0.95 � is needed to resolve Li columns in LiFePO4 along the [001] axis. Reading off intersections from the 0.95 � line, we find that the smallest change in Li concentration that can be determined without a comparable amount of Li being displaced to neighboring columns by VED is 38%. If one ignored column to column variations, and only sought to measure the average Li concentration, ABF-STEM could be used to determine average Li content in a partially charged LiFePO4 sample to ~6% accuracy, but local ordering patterns in column occupancy are limited to 38% uncertainty before damage artifacts dominate. Microsc. Microanal. 21 (Suppl 3), 2015 1550 References: [1] S.D.Findlay et. al. Appl. Phys. Lett. 95, (2009) 191913. [2] Zhang et. al. PNS MI, 23, 3, (2013) p. 256. [3] D.A. Muller & J. Silcox, Phil. Mag. A., 71, 6, (1995) p.1375-1387 [4] B.E.H. Saxberg & Saxton, Ultramicroscopy 6, 85�90 (1981) [5] W.A. McKinley, Jr.& H. Feshbach, Phys. Rev. 74, (1948) p. 1759-1763. [6] C. R. Bradley, "Calculations of Atomic Sputtering and Displacement Cross-Sections in Solid Elements by Electrons with Energies from Threshold to 1.5MV", Argonne National Lab: 1988. [7] Work supported by the Energy Materials Center at Cornell (DOE BES DE-SC0001086). Figure 1. a) Multislice simulation of ABF image of LiFePO4, 10 nm thickness viewed down caxis. Field of view ~ 1 nm x 1 nm. 300kV beam voltage, probe size 0.7 �. 0% column vacancies, defocus 0 nm. b) Contrast vs beam voltage over a Li column (along red line in a), no vacancies for 300 kV, 200 kV, and 100 kV, half occupancy for 300 kV. c) Contrast vs Occupancy as % of vacancies for Li columns at 300 kV, probe size 0.7 �, defocus 0 nm. Figure 2. a) Surface sputtering cross section for Li metal [6] and the VED cross section for Li in LiFePO4 vs beam voltage. b) Dose vs resolution for Li columns in 10 nm thick LiFePO4, 300kV beam voltage, probe size 0.7 �, defocus 0 nm imaged down c-axis. With Li columns resolvable at 0.95 � resolution, 6% of Li atoms would be sputtered at the dose needed to detect 6% occupancy difference between columns. In LiFePO4 sample with vacancies, 38% of Li would be displaced by VED for a dose needed to detect 38% occupancy differences between columns.");sQ1[774]=new Array("../7337/1551.pdf","Artefacts Induced on Soft Layer of Hybrid Metallic Nanoparticles in TEM","","1551 doi:10.1017/S1431927615008533 Paper No. 0774 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Artefacts Induced on Soft Layer of Hybrid Metallic Nanoparticles in TEM Hui Qian1, Lars Laurentius2 and Ray F. Egerton3 1. 2. National Institute for Nanotechnology, NRC, Edmonton, AB, T6G 2M9, Canada Nano Institute of Utah, University of Utah, Salt Lake City, UT 84112, USA 3. Department of Physics, University of Alberta, Edmonton, AB, T6G 2E1, Canada Hybrid organic shell - inorganic core nanoparticles (NPs) are of interest in a wide range of applications, such as drug delivery and nanosensors. The structural and chemical arrangement of organic layers plays a crucial role in the function of hybrid NPs, therefore physicochemical characterization can help us to understand and control the synthesis processes. Transmission electron microscopy (TEM) has been applied to the visualization and measurement of organic shell thickness of hybrid NPs [1]. However, there are some challenges to be overcome, including the awareness of artefacts when applying TEM to characterize organic layers. One typical challenge concerns the electron beam induced hydrocarbon contamination (HC), which predominates when examining NPs deposited from solution and could cause the improper analysis of soft layers of hybrid metallic NPs. HC sources, HC observation in EM and mitigation approaches have been discussed and reported in previous work [2]. However, in order to visualize and analyze the soft layer of hybrid NPs made in solution, most HC mitigation approaches cannot be applied. For instance, UV cleaning, plasma cleaning and heating in vacuum may damage and remove the soft layers or increase the layer thickness before TEM analysis. Here we show evidence for a thickness increase of the soft layer and investigate its cause. Proper mitigation approaches are then recommended for analysis of such hybrid NPs in TEM. The citrate-capped 40 � 3.2 nm gold NPs were purchased from BBInternational, with a concentration of 9 x 1010 particles/ml. AuNPs were modified with 4-Nitroazobenzene (dNAB) using the method introduced by Laurentius et al. [3]. The modified Au/dNAB NPs were separated from solution by centrifugation at 8000 rpm for 10 minutes in a microcentrifuge. Next the NP pellet was redispersed in deionized water. A second centrifugation step was added to wash the particles further. The suspended modified particles are stable and were stored at 7oC before use. A 5 �l droplet of Au/dNAB solution was placed on TEM grids. Excess solution was blotted after one minute and then grids with NPs on were air-dried at room temperature for half to one hour before imaging in TEM. All zero-loss bright field (BF) images were recorded in a JEM2200FS TEM at 200 kV with a 10 eV energy-filter slit. The HC was severe for the Au/dNAB NPs specimen on holey carbon film, as shown in Fig.1. The holey carbon film was used as provided commercially. The dark ring (red arrow marked) represents HC built up around the edge of illumination area. Meanwhile, it was observed that the soft layers of AuNPs near to or even about 300 nm away from the illumination area became asymmetrical. The thickness of the soft layer is larger on the side facing towards the illuminated area, compared to the other side of the NPs, whereas the soft layer of Au NPs on carbon film within the illuminated area is symmetrical. The AuNPs near the edge of holes (Fig. 1b) have asymmetrical layers inside the illuminated area and their thickness observed to increase during irradiation. To mitigate the HC source, the holey carbon film and SiN membrane were cleaned in a UV TEM Zone cleaner and evaluated in TEM before applying solution, in which case there was no HC built up. However, the same phenomenon of asymmetrical thickness of the soft layer was observed, as shown in Fig.2, although a lighter Microsc. Microanal. 21 (Suppl 3), 2015 1552 HC ring built up compared with the sample used in Fig.1. In order to properly measure the thickness of the soft layer in the TEM, heating, cryogenic imaging and beam shower approaches were carried out. The asymmetrical thickness phenomenon disappeared when the specimen was heated but the layer thickness increased. The most effective method of avoiding HC contamination was imaging under cryogenic conditions, as shown in Fig. 3, but with a longer setting-up time and the possibility of ice interference. The beam-shower method was also evaluated for measuring the layer thickness and proved also to be effective. As a result, we successfully measured the layer thickness of different functionalized AuNPs, for comparison with theoretical values. In summary, the possibility of artefacts in thickness measurement or composition analysis of soft layer in hybrid NPs or nanowires needs to be recognized. An increase of layer thickness can be caused by hydrocarbon contamination from grid supporting films or the organic residuals from solution. References: [1] D. Li et al, Advances in Colloid and Interface Science 149 (2009), 28�38. [2] R. F. Egerton et al, Micron 35 (2004), 399-409. [3] L. Laurentius et al, ACSNANO, Vol.5, No.5 (2011), 4219-4227. [4] The authors acknowledge Masahiro Kawasaki from JEOL USA, Inc. for his useful discussion. a b b Figure 1. Zero � loss BF TEM image of Au/dNAB NPs on holey carbon grid: a) taken at lower magnification to include previous illuminated areas at higher magnification; b) taken near the edge of the hole of carbon film. Figure 2. BF TEM images of Au/dNAB NPs on UV cleaned SiN membrane. Figure 3. BF TEM image of Au/dNAB on UV cleaned carbon film and imaged at LN2 temperature.");sQ1[775]=new Array("../7337/1553.pdf","Serial block face SEM and TEM imaging for quantitative measurement of cellular uptake of semiconductor quantum dot nanoparticles","","1553 doi:10.1017/S1431927615008545 Paper No. 0775 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Serial block face SEM and TEM imaging for quantitative measurement of cellular uptake of semiconductor quantum dot nanoparticles Nicole Hondow1, M. Rowan Brown2, Tobias Starborg3, Alexander G. Montieth4, Rik Brydson1, Huw D. Summers2, Paul Rees2 and Andy Brown1 1. Institute for Materials Research, School of Chemical and Process Engineering, University of Leeds, Leeds LS2 9JT, UK. 2. Centre for Nanohealth, College of Engineering, Swansea University, Singleton Park, Swansea SA2 8PP, UK. 3. Wellcome Centre for Cell Matrix Research, University of Manchester, Manchester M13 9PT, UK. 4. Gatan UK, 25 Nuffield Way, Abingdon OX14 1RL, UK. There are an increasing number of potential applications for nanoparticles in clinical medicine, including targeted drug delivery and contrast agents for biomedical imaging, which promise faster, less invasive and more precise treatments than those currently available [1]. Current in vitro studies are concerned with the biological impact of nanoparticles, with electron microscopy commonly employed to image the intracellular location. It is critical to quantify the absolute nanoparticle dose received in a given exposure, and to understand the factors which affect this. This is difficult, with the complex and varied mechanisms of nanoparticle interactions with cells. Our aim is to develop a full quantitative description of nanoparticle uptake by an in vitro cell line. Imaging flow cytometry is a high-throughput, low-resolution technique useful for measuring the cellular uptake of fluorescent nanoparticles [2], but it cannot measure the dose in terms of a number of particles. TEM of thin cell sections has the required spatial resolution to provide the location and number of cellular vesicles per 2-D cell slice plus the number of nanoparticles per vesicle [3,4]. However this is limited by both the nature of the 2-D thin section, with only a small amount of the cell analyzed, and the time-intensive nature of TEM imaging. Serial sectioning can provide information across a whole cell, and the increased use of serial block face scanning electron microscopy (SBF SEM) opens avenues to analysis of much larger volumes without the labour- and time-intensive nature of examining serial sections in the TEM [5]. The reduced resolution of SBF SEM as compared to TEM limits examination to nanoparticle filled endosomes rather than individual nanoparticles, but the size, shape and location of these endosomes can be quantified in whole cell volumes. We will show results from studies where commercially available Qtracker 705 quantum dot nanoparticles were loaded into human osteosarcoma (U-2 OS) cells and at certain time points were fixed and resin-embedded for quantitative electron microscopy analysis [3,4]. The same resin-embedded sample was used for both the production of the TEM thin sections and for SBF SEM. No post-fixation heavy metal staining was required for either analysis as the electron-dense quantum dots are easily identifiable from the cellular features which are still visible due to staining by the osmium fixative. SBF SEM analysis using a Gatan 3-View system allowed collection of a data set containing 3-D information from numerous cells. Microsc. Microanal. 21 (Suppl 3), 2015 1554 SBF SEM data has been used to correlate higher resolution 2-D TEM data to high throughput, low resolution optical imaging of quantum dot nanoparticle loaded cells [3]. After 1 hour of exposure to Qtracker 705 quantum dots, internalized nanoparticles can be identified in both TEM and SBF SEM. The location of quantum dot endosomes can be identified in SBF SEM, and it was found that they are distributed evenly throughout the cell after the short exposure time. This indicates that any one thin section of a cell is potentially representative of the whole cell, allowing for conversion of the quantitative 2-D TEM data to 3-D. This results in the determination of a calibration factor to transform flow cytometry fluorescence intensity data to a nanoparticle dose distribution, in terms of the fundamental unit, the number of nanoparticles internalized per cell [3]. We will also show how the distribution of the endosomal load within cells develops over a further 24 hour period [4]. References: [1] TL Doane and C Burda, Chem. Soc. Rev. 41 (2012), p.2885. [2] HD Summers, P Rees, MD Holton, MR Brown, SC Chappell, PJ Smith and RJ Errington, Nat. Nano. 6 (2011), p. 170. [3] HD Summers, MR Brown, MD Holton, JA Tonkin, N Hondow, AP Brown, R Brydson and P Rees, ACS Nano 7 (2013), p. 6129. [4] N Hondow, MR Brown, T Starborg, AG Montieth, R Brydson, HD Summers, P Rees and A Brown, J. Microsc. Accepted (2015). [5] W Denk and H Horstmann, PLoS Biol 2 (2004), p. e329. Figure 1. SBF SEM of a U-2 OS cell exposed to Qtracker 705 quantum dots. (a) and (b) Stills from a reconstruction of a cell, (c) contrast inverted SBF SEM image and (d) segmented version of (c).");sQ1[776]=new Array("../7337/1555.pdf","Thickness mapping of freestanding Ionic Liquid films using Electron Energy Loss Spectroscopy in the TEM.","","1555 doi:10.1017/S1431927615008557 Paper No. 0776 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Thickness mapping of freestanding Ionic Liquid films using Electron Energy Loss Spectroscopy in the TEM. A.E. Ribbe, P. Kim, D. Hoagland, T.P. Russell Department for Polymer Science & Engineering, University of Massachusetts Amherst, Amherst, Massachusetts 01003 USA One of the biggest challenges in Electron Microscopy in fluids is to overcoming the limitation of the high vacuum requirements. To observe the dynamics of fluid samples, sophisticated fluid cells have been developed that typically use 25 nm Silicon Nitride membranes with a separation of ~ 200 nm [1]. Recently it was demonstrated that graphene can be used to trap liquids for TEM observation [2] which improves resolution substantially due to the negligible scattering contribution of the graphene layers. However, sample sizes are quite small, ~ 100 nm in diameter. Ionic Liquids (IL) [ 3 ] recently found application in Cryo-TEM for visualization of polymers morphologies [4] or nanoparticle assemblies [5]. The ability to perform room temperature TEM [6] measurements is enabled by the very low vapor pressure. Recently HRTEM studies showing a microstructure in ILs have been shown [7]. We here describe the shape of IL films that can be used as a medium to investigate nanoparticle dynamics n TEM and SEM by using Electron Energy Loss Spectroscopy (EELS) thickness mapping. Lacey carbon, which has an average hole diameter of ~2 m. and a thickness of 20-30 nm [8] provided the most reliable support for IL films. The rather large hole size distribution allows for the random generation of an array of different IL film geometries i.e. thickness, diameter and surface curvature. IL (TEGO1) thickness maps where acquired using a JEOL JEM-2200FS EFTEM with in-line omega-type electron energy loss filter. First, the local thickness of the IL film was determined using EELS with a small filter entrance aperture at various locations, as marked in Figure 1. Absolute thicknesses were calculated using the log-ratio method: t/ = ln(It/I0) where It and I0 are the total spectrum the zero-loss intensities, t is the film thickness and is the mean free path of the IL. Subsequently, an uncalibrated thickness map was acquired in energy filtered (EFTEM) imaging mode by calculation of the log-ratio-image of an unfiltered and zero-loss filtered bright field image. The absolute thicknesses determined by EELS were then used to calculate an absolute thickness map using a 2D gray value profile of the uncalibrated thickness map, as illustrated in Figure 2. The shape of the individual free-standing IL films depends mainly on the thickness of the respective hole. While films < 100 nm appear to have convex shape with thicknesses of smaller than 20 nm in the center, films with t> 100 nm are generally concave, with up to 300 nm thickness in the center. Recently we described particle motion in IL films on solid substrates by SEM [9]. This work Microsc. Microanal. 21 (Suppl 3), 2015 1556 expands our experiments to free standing IL films and we will demonstrate that IL films can be utilized in both SEM and TEM to study particle dynamics as for example Brownian motion. [1] de Jonge, N., Ross, M.F. "Electron microscopy of specimens in liquid" Nature Nanotechnology", 6, 695-704 (2011). [2] Yuk, J.M., Park, J., Ercius, P., Kim, K., Hellebusch, D.J. "High-Resolution EM of Colloidal Nanocrystal Growth Using Graphene Liquid Cells" Science, 336, 61-64 (2012). Michael F. Crommie,1,2,6 Jeong Yong Lee,3 A. Zettl,1,2,6 A. Paul Alivisatos2,4 [3] Welton, T., "Room-Temperature Ionic Liquids. Solvents for Synthesis and Catalysis", Chem. Rev. 99, 2071-2083 (1999). [4] Simone, M., Lodge, T.P., "Micellization of PS-PMMA Diblock Copolymers in an Ionic Liquid" Macromol. Chem. Phys., 208, 339�348, (2007). [5] Dalaver, H. et al. "Cryo-transmission electron microscopy of Ag nanoparticles, grown on an ionic liquid substrate" Mater. Res., 25(7), 1264-1271 (2010). [6] Maddikeri, R., Colak, S., Gido, S.P., Tew, G.N. "Zwitterionic Polymersomes in an Ionic Liquid: Room Temperature TEM Characterization" Biomacromolecules, 12(10), 3412�3417 (2011). [7] Chen, S., Kobayashi, K., Kitaura, R., Miyata, Y., Shinohara, H. "Direct HRTEM Observation of Ultrathin Freestanding Ionic Liquid Film on Carbon Nanotube Grid" ACS Nano, 5(6), 4902 (2011). [8] Karlsson, G. "Thickness measurements of lacey carbon films", Journal of Microscopy, 203/3, 326-328 (2001). [9] Ribbe, A.E., Kim, P., Hoagland, D., Russell, T.P. "Diffusion Processes in Ionic Liquids Observed via low voltage Scanning Electron" Microscopy and Microanalysis, 18(S2), 1618-1619 (2012). Figure 1: Zero-loss filtered bright field Figure 2: Absolute thickness map of TEGO1 image of TEGO1 and filter entrance suspended on lacey carbon. Color scale in nm; aperture locations for EELS thickness image size 11m x 11m. determination.");sQ1[777]=new Array("../7337/1557.pdf","PSRT: Progressive Stochastic Reconstruction Technique for Cryo Electron Tomography","","1557 doi:10.1017/S1431927615008569 Paper No. 0777 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 PSRT: Progressive Stochastic Reconstruction Technique for Cryo Electron Tomography Beata Turoov�1,2, Lukas Marsalek1,3,4, Tom�s Davidovic1,5 and Philipp Slusallek1,3,5 1. 2. Saarland University, Campus E 1.1, 66123 Saarbr�cken, Germany IMPRS-CS, Max-Planck Institute for Informatics, Campus E 1.4, 66123 Saarbr�cken, Germany 3. Agents and Simulated Reality Group, DFKI GmbH, Campus E 3.4, 66123 Saarbr�cken, Germany 4. Eyen SE, Na Niv�ch 1043/16, 14100 Prague, Czech Republic 5. Intel VCI, Campus E 1.1, 66123 Saarbr�cken, Germany Cryo Electron Tomography (cryoET) is one of the essential techniques in Structural Biology, as it allows us to study the structure of macromolecular complexes in their native environment in situ. The tomographic reconstruction in cryoET is a particularly challenging task as the input data suffers from very low contrast, high noise, and limited tilt range. Moreover, the scanned specimen is larger than the detector, introducing the interior problem into the reconstruction process, which causes vignetting artifacts on the edges of the reconstructions. To alleviate some of these limitations, high-resolution protocols such as Subtomogram Averaging (SA) are applied to obtain structures of individual macromolecular complexes from a tomogram. Results of these protocols are highly dependent on the quality of the reconstruction. Current state-of-the-art methods such as Weighted Back Projection (WBP) or Simultaneous Algebraic Reconstruction Technique (SART) deliver noisy and low-contrast reconstructions and thus manual intervention is often needed during SA. We present a novel iterative approach to tomographic reconstruction in cryoET called Progressive Stochastic Reconstruction Technique (PSRT), which is designed to improve the reliability and decrease the need for manual intervention in the high-resolution protocols [1]. The approach is based on a different mathematical framework than the existing techniques. It uses Monte Carlo random walks to reconstruct the volume by gradually placing spherical elements, called samples, of a given size and intensity into the volume (Fig. 1). The position of each sample is generated randomly and the sample is accepted only if the improvement of the current volume estimate is sufficient with respect to a selected error metric. This is guided by a sampling strategy similar to Metropolis-Hastings, where the areas with higher acceptance potential are sampled more densely, thus speeding up the convergence of the method. During the reconstruction process we progressively decrease both the size and the intensity of the samples, performing a coarse-to-fine reconstruction. Furthermore, we can design additional importance sampling that allows us to focus on specific parts of the volume. This is of special importance in SA, where one is interested in individual structures distributed within the volume which, however, represent only a small part of it. The current state-of-the-art methods lack this ability and reconstruct every part of the volume with the same procedure. PSRT also provides a memory efficient solution to specimen-level interior problem and removes all associated artifacts. We compare the method to the state-of-the-art techniques, validate it on synthetic data and show on experimental data of known biological systems that it is able to reconstruct the correct high-resolution structures. Moreover, PSRT delivers smoother reconstructions with enhanced contrast and fewer artifacts. This improves template-based localization and thereby enables better automation in SA (Fig. 2). Microsc. Microanal. 21 (Suppl 3), 2015 1558 References: [1] B Turoov� et al, J Struct Biol (2015) doi: 10.1016/j.jsb.2015.01.011. [2] We thank Massimiliano Maletta from NKI AVL and Peter J. Peters from Department of Health Medicine and Life Science at Maastricht University for providing us with the experimental data of the 70S ribosome. Figure 1. Scheme of one iteration of PSRT. In each iteration we generate starting samples, called seeds, with given size and intensity and for each seed we perform a random walk. The iteration has completed once all seeds and their corresponding random walks were processed. The next iteration starts with seeds of smaller size and intensity. Figure 2. Comparison of template-based localization for (a) WBP, (b) SART (with relaxation parameter 0.1 and one iteration), and (c) PSRT. The enlargements (2x) demonstrate contrast properties of the reconstructions. Black arrows indicate false positives in the localization for WBP and SART. In the PSRT reconstruction, only ribosomes were localized and no manual correction of the localization results was therefore necessary.");sQ1[778]=new Array("../7337/1559.pdf","First Results of Integrating a Compression/Tension Load Cell with Nano-scale X-ray Transmission Microscopy","","1559 doi:10.1017/S1431927615008570 Paper No. 0778 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 First Results of Integrating a Compression/Tension Load Cell with Nano-scale X-ray Transmission Microscopy Nikolaus L. Cordes1, Kevin Henderson1, Benjamin Hornberger2, Brian M. Patterson1 1 Polymers and Coatings Group, Materials Science and Technology Division, Los Alamos National Laboratory, Los Alamos, NM USA 2 Carl Zeiss X-ray Microscopy Inc., Pleasanton, CA USA Understanding mechanical failure, crack propagation, and compressive behavior at the submicrometer scale is essential for tailoring material properties for structural performance. Typically, tension or compression loading is needed to understand these processes. Here we demonstrate the coupling of a custom compression/tension load cell with a nano-scale X-ray transmission microscope. The load cell, capable of 9 N of force, is custom designed to fit into an Xradia (now Carl Zeiss X-ray Microscopy Inc., Pleasanton, CA) UltraXRM-L200 nano-scale X-ray transmission microscope. Driven by a piezoelectric motor, the load cell has a total displacement of 500 m and can be operated in either uniaxial tension or uniaxial compression configurations. In the latter configuration, the bottom platen is driven upwards into a stationary top platen. Figure 1 (left) presents a photograph of the custom load cell. A magnified view (Fig. 1, right) shows the cell in uniaxial compression mode, before alignment of the bottom platen with the stationary top platen. A variety of samples will be described, such as the compression of silicone polymer foam ligaments (Fig. 2A-D) and hollow glass microspheres (Fig. 3A-B). The microspheres were acquired from 3MTM (S22 Glass Bubble Series) and vary in diameter, ranging from ~20 m to ~65 m, with a median diameter of 35 m [1]. The representative crush strength of the microsphere is quoted as 2758 kPa. In addition, we will present results of tensile strain of Al-Cu eutectics (data not shown). Figure 2A-D displays nano-scale X-ray transmission radiographs of the compression of a single silicone polymer ligament, imaged in Zernike phase contrast mode. Edges are visible inside the ligament, corresponding to the diatomaceous earth filler used in the silicone polymer synthesis. The ligament can be seen bending in Fig. 2B and 2C, leading to densification in Fig. 2D. Figure 3A-B highlights the uniaxial compression of a hollow glass microsphere, used in the synthesis of syntactic (i.e., glass microsphere-templated) polymer foams. Several microspheres of varying diameters can be seen on the bottom platen (Fig. 3A); however, the large sphere (diameter = 22.7 m) is aligned so that it is the only microsphere being compressed. Figure 3A shows this microsphere in contact with the top platen, which is fabricated from diamond, thus allowing a sufficient number of X-ray photons to transmit through the platen to the detector. Figure 3B shows the compressed microsphere at 19% strain, as the bottom platen is being driven upwards. The software which drives the load cell is capable of acquiring and recording force data. By measuring the diameter of the sphere from the radiograph and by averaging the acquired force data, it is possible to construct a stress-strain curve of the uniaxial compression (Fig. 3C). From the manufacturer, the representative crush strength of the microspheres is 2758 kPa; here, we find the crush strength of one microsphere to be 1338 � 127 kPa at a strain of 44%. Microsc. Microanal. 21 (Suppl 3), 2015 1560 From these first results, we show the feasibility of coupling a custom load cell with nano-scale X-ray transmission microscopy. Radiographs acquired through this coupling reveal the damage of materials due to compressive or tensile strains, with sub-micrometer resolution. References [1] 3MTM Microspheres Selection Guide. http://multimedia.3m.com/mws/media/130063O/3mtmglass-bubbles-selection-guide.pdf?fn=MicroSelectGuide_Celum_9841447.p (02/12/2015). Figure 1. Optical picture of the custom load cell (Left) and a magnified view of the load cell in uniaxial compression mode before fine alignment. Figure 2. Nano-scale X-ray transmission radiographs, acquired in Zernike phase contrast mode, of a polymer foam ligament (A-D) under uniaxial compression. The displacement of the compression cell was 0 m (A), 10 m (B), 20 m (C), and 30 m (D). Figure 3. Nano-scale X-ray transmission radiographs, acquired in Zernike phase contrast mode, of a glass microsphere before (A) and during (B) uniaxial compression. C) The stress-strain curve of a glass microsphere, obtained while undergoing uniaxial compression in the custom load cell. Data points are the average stress and error bars correspond to the standard deviation of the measured stress.");sQ1[779]=new Array("../7337/1561.pdf","Non-local Prior Modeling for Tomographic Reconstruction of Bright Field Transmission Electron Microscopy Images.","doi:10.1017/S1431927615008582","1561 doi:10.1017/S1431927615008582 Paper No. 0779 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Non-local Prior Modeling for Tomographic Reconstruction of Bright Field Transmission Electron Microscopy Images. Suhas Sreehari1, S. V. Venkatakrishnan2, Jeffrey P. Simmons3, Lawrence F. Drummy3 and Charles A. Bouman1. 1. 2. Electrical and Computer Engineering, Purdue University, West Lafayette, IN, USA. Lawrence Berkeley National Laboratory, Berkeley, CA, USA. 3. Air Force Research Laboratory, Dayton, OH, USA. Many important imaging problems in materials and biological sciences involve reconstruction of images that contain several repeating non-local structures. The presence of similar/identical particles or structures in images gives rise to enormous redundancy. Model-based iterative reconstruction (MBIR) is a powerful reconstruction framework that could � in principle � exploit such redundancies [1]. This is normally done through the selection of a log prior probability term that is part of the cost function used to compute the maximum a posteriori (MAP) estimate of the unknown. However, in practice, determining such a log prior probability term that accounts for the similarity between non-local particles or structures in the image is quite a challenging task. The open question of how to capture this non-local redundancy brings our attention to the advances made in the area of denoising algorithms. Non-local, patch-based denoising algorithms like BM3D and non-local means (NLM) are known to capture non-local similarities in images. However, since these denoising algorithms are not explicitly formulated as solutions to optimization problems, it is unclear as to how to use them in the MBIR framework. In this paper, we formulate a solution to the tomographic reconstruction of bright field transmission electron microscopy (TEM) images. Bright field electron tomography has a reliable solution within the MBIR framework [2], however it does not support non-local prior modeling. To open up MBIR to advanced prior modeling techniques, we make use of a novel framework we call "plug&play priors" [3]. The plug&play technique is an application of the alternating directions method of multipliers (ADMM) [4], wherein we split the state variable into two variables, and solve two smaller optimization problems with the additional constraint that the two split variables be equal to each other. In doing so, plug&play effectively decouples the log likelihood term (based on the data) from the log prior probability term. We then define two operators � the inversion operator (based just on the data) and the denoising operator (based just on the prior model) � that are applied sequentially until convergence. We specifically use 3D non-local means (NLM) as the prior model in the plug&play framework, and we showcase high quality tomographic reconstructions of two real datasets of aluminum spheres and ferritin structures. In all the results, we observe that smear artifacts are visibly suppressed, and that edges are preserved. NLM-based reconstructions are also sharper than reconstructions that use the standard qGGMRF [1] as prior model [5]. References: Microsc. Microanal. 21 (Suppl 3), 2015 1562 [1] C. A. Bouman, "Model based image processing." [2] S.V. Venkatakrishnan et al, "Model-Based Iterative Reconstruction for Bright-Field Electron Tomography," accepted to the IEEE Transactions on Computational Imaging. [3] S. V. Venkatakrishnan et al, "Plug-and-play priors for model based reconstruction," IEEE Global Conference on Signal and Information Processing, 2013. [4] S. Boyd et al, "Distributed optimization and statistical learning via the alternating direction method of multipliers," Foundations and Trends in Machine Learning, 2011. [5] This work was supported by AFOSR/MURI grant #FA9550-12-1-0458, by UES Inc. Figure 1. (Clockwise from left) 0� tilt of the aluminum spheres bright field dataset; filtered backprojection reconstruction; qGGMRF reconstruction; 3D NLM reconstruction � all of slice #307 of the dataset. Figure 2. (Clockwise from left) Contrast-stretched version of the 0� tilt of the ferritin structures bright field dataset; filtered backprojection reconstruction; qGGMRF reconstruction; 3D NLM reconstruction � all of slice #1149 of the dataset.");sQ1[780]=new Array("../7337/1563.pdf","A Modern Correlative Workflow Environment to Master the Multi-scale Challenge","","1563 doi:10.1017/S1431927615008594 Paper No. 0780 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A Modern Correlative Workflow Environment to Master the Multi-scale Challenge Arno P. Merkle1, Lorenz Lechner1, Andy Steinbach1 1 Carl Zeiss X-ray Microscopy, Inc. 4385 Hopyard Road, Pleasanton, CA, USA In most fields of study, it is imperative to understand the behavior of a system across several length scales in three dimensions in order to properly address the structural parameters that govern its performance. In order to characterize a system, be it the brain, a structural metal alloy, or a porous rock reservoir, multiple microscopic methods have evolved to specialize in capturing a relatively well defined window of scales, modalities or dimensions of information. Examples of this include medical-CT, confocal light microscopy, X-ray tomography, FIBSEM tomography, serial block face SEM, TEM tomography, atom probe tomography, and more. As these techniques have progressed individually, a clear challenge that has emerged has been how to intelligently navigate to and acquire 3D volumes of interest (from centimeter to nanometer), and, subsequently, to fuse multi-scale and multi-modality datasets in such a way that leaves the microscopist in control as recently published in the context of a corrosion study [1]. In this forum, we will discuss the emergence of two major technologies that address this challenge. The first is the increased utilization of laboratory X-ray microscopy (XRM), a powerful technique to perform nondestructive 3D imaging on samples over a range of length scales and material types [2]. XRM has shown the ability to provide valuable context, timelapse, and navigational information in the correlative microscopy environment, complementing information from light and electron microscopy instruments [3]. The second major development enabling correlation, is the recent development of a modern microscopy workflow environment (ZEISS Atlas 5), which acts as the glue between experiments obtained on multiple platforms (SEM, LM, FIB-SEM, XRM, etc.). Atlas 5 automates several advanced SEM and FIB-SEM acquisition tasks, but also provides a visualization environment to co-locate and register multiple datasets from multiple instruments in one place. Going further, this environment extends beyond the conventional 2D correlation approach, by incorporating 3D datasets such as those obtained by XRM or FIB-SEM tomography. We will review both developments in the context of three examples, from Materials Research, Life Science and Geoscience. In Materials Science, XRM tomograms collected on an Al 7075 aluminum alloy (Figure 1) are used to locate inclusions and pores within the interior of the microstructure, which are then selected and examined at higher resolution by targeted FIB-SEM serial sectioning at defined regions of interest [4]. In Life Sciences, XRM presents a unique opportunity to bridge the length scales between light and electron microscopy (Figure 2), easing the `needle in a haystack' navigation problem for locating the same region of interest using multiple microscopes [5]. In Geosciences, we demonstrate the great 2D and 3D multi-scale challenge, and how Atlas 5 has been successfully used to enable the efficient study of such porous media systems. Microsc. Microanal. 21 (Suppl 3), 2015 1564 References: [1] Burnett T., et. al. Correlative Tomography, Scientific Reports, 2014 [2] A. P. Merkle and J. Gelb, Ascent of 3D X-ray Microscopy in the Laboratory, Microscopy Today, 21 (2013), p. 10 [3] E Maire and P Withers, Quantitative X-ray tomography, International Materials Review, 59 (2014), p. 1 [4] A. P. Merkle et al., Automated Correlative Tomography Using XRM and FIB-SEM to Span Length Scales and Modalities in 3D Materials, Microscopy and Analysis, 28 (2014), p. S10-S13 [5] E. Bushong et al., X-Ray Microscopy as an Approach to Increasing Accuracy and Efficiency of Serial Block-Face Imaging for Correlated Light and Electron Microscopy of Biological Specimens, Microscopy and Microanalysis (2014). Figure 1: A 3D automated correlative workflow demonstrated on an Aluminum 7075 alloy to access information about inclusions, voids, precipitates and the Al matrix grain structure. In collaboration with N. Chawla and S. Singh of Arizona State University. Figure 2: XRM dataset of stained (for EM) mammalian brain tissue, used to navigate to specific subsurface volumes of interest quickly, thereby multiplying the efficiency of 3D EM techniques. In collaboration with NCMIR at the University of California San Diego.");sQ1[781]=new Array("../7337/1565.pdf","Cryo-Correlative Light and Electron Microscopy (Cryo-CLEM): Specimen Workflow Paths and Recent Instrument Developments","","1565 doi:10.1017/S1431927615008600 Paper No. 0781 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Cryo-Correlative Light and Electron Microscopy (Cryo-CLEM): Specimen Workflow Paths and Recent Instrument Developments Michael Schwertner1, Duncan Stacey1 1. Linkam Scientific Instruments, 8 Epsom Downs, Tadworth, Surrey, KT20 5LR, United Kingdom. Corresponding author: MichaelSchwertner@linkam.co.uk Cryo - Correlative Light and Electron Microscopy (cryo-CLEM) is an exciting recent technique that benefits from combining the complementing advantages from Electron Microscopy (EM) and Fluorescence (Light) Microscopy (FM) [1,2,3,4]. While EM gives superior spatial resolution beyond the limit of light microscopy it mainly delivers structural information. EM has limitations in terms of the biological and chemical sensing of events. Fluorescence Microscopy (FM) on the other hand can reveal information about chemical, biological and even genetic processes and is compatible with living cells. FM markers, such as Green Fluorescent Protein (GFP), can be tailored to answer specific biological questions and are widely established. The downside of FM is that spatial resolution is severely limited compared to EM. Cryo-CLEM brings it all together by sequentially imaging the same specimen location using both complimentary EM and FM techniques and merging the datasets. Genetic and biological events can be identified and pinpointed in FM before "zooming in" and studying details in very high resolution in EM. An elegant way to implement CLEM is to prepare biological samples in frozen-hydrated "vitrified" ice under cryo-conditions. In this state ice-crystal formation is suppressed, the biological sample can be observed in its near-native state. This has two unique advantages: First, fluorescence signals in FM are very strong under cryo-conditions, where reduced bleaching rates lead to superior signal to noise compared to imaging at room temperature. Second, frozen-hydrated samples are directly compatible with the vacuum requirements of the EM and the ultrastructural preservation of vitrified cryo-samples is considered the "gold standard". Screening in FM does not cause electron beam damage and preserves the sample for HR-EM In this paper we will give a brief overview on the different sample preparation options, imaging workflows and recent developments in the instrumentation for cryo-CLEM. A cryo-CLEM setup and workflow needs to fulfill the following conditions: a) The sample has to remain vitrified, e.g. kept below the critical temperature of approx. -140 �C b) Coordinate recording and mapping is required to identify and record the sample locations of interest with both FM and EM imaging techniques c) Ice contamination which can be caused by condensation of humidity on the sample has to be prevented in all steps of sample handling and transfer. An instrument that fulfils all of the above requirements is the CMS196 cryo-correlative stage from Linkam Scientific, shown in Figure 1, inset. It uses liquid nitrogen to maintain the samples below the vitrification temperature. Up to three EM grids can be mounted and screened in one sample cassette, which is magnetically self-aligning and mounted inside the sample chamber. Several sample mount adapters and holders are available for a range of grid types and sample carriers. The sample transfer process uses a pre-cooled closed cryotransfer container and special manipulation tools. An LN2 auto-refill system allows independent operation of the system for a full day, without ice contamination. The cryo-stage is mounted on a light microscope equipped for fluorescence imaging and long working distance lenses (dry, up to NA 0.9). Microsc. Microanal. 21 (Suppl 3), 2015 1566 The integrated transmitted light condenser also allows standard optical contrasttechniques such as PlasDIC and / or phase contrast to be used. A high resolution motorised XY sample positioning system with optical encoders provides a calibrated sample coordinate system and enables fully automated EM grid mapping in fluorescence mode. The user can then review large area tiled / stitched fluorescence image data. The raw coordinate correlation accuracy between EM and FM in the order of one micron can be achieved, while much higher correlation accuracy below 50 nm can be achieved by the use of fiducial markers [5]. Figure 1: Flow chart with workflow options for cryo-fixation, sample processing and imaging. Inset top right: cryo-correlative stage CMS196 from Linkam Scientific for the implementation of cryo-CLEM. The flowchart in figure 1 outlines the most common sample workflows. FM can be performed at several stages in the process. For example, cells or sub-regions can be identified during live cell imaging before cryo-fixation of the sample. Instead of EM, x-ray imaging, tomography or EM-tomography modalities can also be used for correlation with FM. It is anticipated that this young family of correlative techniques will be an indispensable tool for studying the processes of life, beyond the capabilities of FM or EM alone. References [1] Van Driel et al.," Tools for correlative cryofluorescence microscopy and cryoelectron tomography applied to whole mitochondria in human endothelial cells", EJCB No 88, Vol. 11, 621710 [2] Briegel et al., "Correlated Light and Electron CryoMicroscopy", Methods in Enzymology, Vol. 481, ISSN 00766879 [3] Methods in Cell Biology, Volume 111, Pages 1404 (2012), Correlative Light and Electron Microscopy, Ed.Thomas M�llerReichert & Paul Verkade, ISBN: 9780124160262, Academic Press [4] Methods in Cell Biology, Volume 124, Correlative Light and Electron Microscopy II, 1st Edition, (2014), Ed. Thomas M�llerReichert and Paul Verkade, page 1 � 417, ISBN 9780128010754 [5] Schorb et al., "Correlated cryofluorescence and cryoelectronmicroscopy with high spatial precision and improved sensitivity", Ultramicroscopy(2013)");sQ1[782]=new Array("../7337/1567.pdf","Correlative Array Tomography � from 2D towards 3D","","1567 doi:10.1017/S1431927615008612 Paper No. 0782 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Correlative Array Tomography � from 2D towards 3D Alexandra F. Elli1, Robert Kirmse1, Jeff Caplan2, Cherish Warner3, Janine Sherrier3 and Kirk Czymmek4 Carl Zeiss Microscopy GmbH, Carl-Zeiss-Promenade 10, 07745 Jena, Germany 15 Innovation Way, Delaware Biotechnology Institute, University of Delaware, Newark DE 19711, USA 3. 15 Innovation Way, Delaware Biotechnology Institute, Department of Plant and Soil Sciences, University of Delaware, Newark DE 19711, USA 4. Carl Zeiss Microscopy, LLC, One Zeiss Drive, Thornwood, NY 10594, USA 2. 1. Correlative microscopy combines information from light and electron microscopy (LM and EM) into one comprehensive dataset. The "Shuttle & Find" interface from Zeiss was the first easy to use solution for imaging one and the same sample region in different microscopes, i.e. from widefield or confocal LSMs to environmental or field emission SEMs (FE-SEM). However, Shuttle & Find is limited to 2D applications. Correlative Array Tomography (CAT) is a correlative volumetric microscopy method on a high-throughput basis. Ordered arrays of ultrathin, resin-embedded serial sections can be imaged with different microscopical modalities. Serial sections are prepared using an ultramicrotome and for correlative array tomography, sections were stained for imaging with a widefield microscope and a scanning electron microscope. CAT is a non-destructive method that allows multiple investigations of the same sample and even multiple successive fluorescent staining procedures [1]. Multiple staining cycles of a large number of antigens or the use of different fluorescent proteins followed by the ultrastructural investigation in a SEM enables the analysis of functional and structural information in the same context. The software module " ZEN Correlative Array Tomography" enables an easy and efficient workflow from LM (e.g. fluorescence) to FE-SEM or vice versa. The first step after inserting the sample into one microscope (LM or SEM) is to perform a quick calibration to establish a coordinate system. Subsequently the software guides the user through an automated process to recognize the ribbons of serial section. Regions of interest (ROI) can then be selected in one section and these ROIs are then automatically duplicated to all other recognized section. Once this process is completed image acquisition of all ROIs is performed automatically by the software thus generating a 3D volume stack from the serial sections. In the second step the sample along with the acquired data is transferred to next microscope (LM to SEM or vice versa). After the initial calibration of the coordinates the subsequent information on section and ROI position can be loaded and re-used for a fast set-up of the imaging parameters. Again, the microscope will then automatically acquire all images at the same regions that were already investigated. Finally a correlative 3D dataset can be created by first aligning both LM and SEM image stack separately. Afterwards the user correlates both volumes by identifying the same four reference points in both dataset thus obtaining a full 3D correlative volume. Microsc. Microanal. 21 (Suppl 3), 2015 1568 [1] K. D. Micheva and S. J. Smith, Neuron 55:25-36, 2007 Figure 1: Overview image of individual ribbons on a glass slide detected with a light microscope. Reconstruction of 3D datasets of fluorescence images and images showing the ultrastructure of nodules. The purpose of the method is to reveal the precise localization of proteins in an ultrastructural context in 3D through correlated light and electron microscopy (CLEM) of zstacks.");sQ1[783]=new Array("../7337/1569.pdf","Micro Computed Tomographic X-ray Imaging (Micro CT): A Versatile and Non-Destructive Method for Biological Specimens.","","1569 doi:10.1017/S1431927615008624 Paper No. 0783 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Micro Computed Tomographic X-ray Imaging (Micro CT): A Versatile and NonDestructive Method for Biological Specimens. James C. Williams, Jr.1 1. Department of Anatomy and Cell Biology, Indiana University School of Medicine, Indianapolis, IN, USA. Micro CT utilizes multiple x-ray images of a specimen to reconstruct the three-dimensional structure as visible by differences in x-ray attenuation. As such, it provides a useful tool for many biological specimens. The method is non-destructive, and thus opens the possibility of a specimen being imaged using micro CT and then further analyzed using other methods. The essence of micro CT is similar to computed tomographic imaging used clinically, but with several significant differences. While in clinical CT the x-ray source and detector are spun around the patient, in micro CT the specimen is rotated within the stationary x-ray beam. This rotation is done in discrete steps, making it more like the old `step-and-shoot' CT systems than the newer helical CT technologies. The size of the specimen is limited to a few cm in micro CT, and to obtain the high resolution characteristic of micro CT, the intensities of x-rays are high and the exposure times long, making this kind of micro CT inappropriate for living things (although some intermediate-resolution forms of micro CT are designed for scanning laboratory animals). For the images shown below, scanning time was typical for our specimens, which require scans from 10-60 minutes in length, typically for the collection of 450 x ray images, with the specimen rotated 0.4� between each image. The `tomographic' part of micro CT involves taking the series of x-ray images of the specimen and calculating what arrangement of x-ray-absorbing materials would yield such a series. The most commonly used method for making this calculation is that developed by Feldkamp. [1] Improvements in computing power mean that this calculation--or `reconstruction'--can be done during the time the next specimen is being scanned. Our Skyscan 1172 system utilizes the graphics processor in the PC to reduce the time of this reconstruction to typically on a few minutes. We have pioneered the use of micro CT in the study of urinary stones. Micro CT is able to reveal structure within stones to a resolution of just a few micrometers, and a number of mineral types can be accurately distinguished from one another with high spatial resolution. [2] In addition, micro CT allows the screening of large numbers of fragments from a patient so that fragments of distinctive composition can be identified much more easily than by visual inspection. Identification of distinctive regions of a single stone specimen can be beneficial for dissection purposes. For either a group of fragments or for dissection of a single stone, micro CT thus can provide guidance for collecting specimens for further analysis by other means such as Fourier-transform infrared spectroscopy (FT-IR). Finally, we have also shown that micro CT can provide structural/compositional information which can be used to understand the etiologies of stone formation. [3] The use of micro CT for imaging soft-tissue specimens stained with heavy metals is presently being refined in several laboratories [4] Given that many traditional en bloc histologic stains contain . metals in them, these also may be useful to enhance contrast for micro CT imaging of tissues specimens. In this way, methods of the past can be used to provide three-dimensional structural information in Microsc. Microanal. 21 (Suppl 3), 2015 1570 1560 specimens of tissue. The plan is to highlight some examples of this application of older technology to micro CT in the talk. References: [1] Feldkamp, L.A., L.C. Davis, and J.W. Kress, J Opt Soc Am A 1 (1984), p. 612-619. [2] Zarse, C.A., J.A. McAteer, A.J. Sommer, S.C. Kim, E.K. Hatt, J.E. Lingeman, A.P. Evan and J.C. Williams. BMC Urology 4 (2004), p. 15. [3] Williams, J.C., J.E. Lingeman, F.L. Coe, E.M. Worcester and A.P. Evan. Urolithiasis 43 (2015), p. 13-17. [4] Khonsari, R.H., C. Healy, A. Ohazama, P.T. Sharpe, H. Dutel, C. Charles, L. Viriot and P. Tafforeau. The Anatomical Record 297 (2014), p. 1803-1807. Figure 1. Micro CT imaging of a urinary stone. A. X-ray image taken in the machine. Over 400 such images were collected as the specimen was rotated 0.4� between each. B. Tomographic reconstruction of a slice through the stone, showing three different mineral types: Calcium oxalate monohydrate (COM), calcium oxalate dihydrate (COD), and calcium phosphate in the mineral form of apatite can clearly be distinguished by x-ray attenuation values (grayscale in the image). C. Surface rendering of the reconstruction, colored to show COD (yellow) and COM (orange).");sQ1[784]=new Array("../7337/1571.pdf","Optimization of the Excitation Light Sheet in Selective Plane Illumination Microscopy","","1571 doi:10.1017/S1431927615008636 Paper No. 0784 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Optimization of the Excitation Light Sheet in Selective Plane Illumination Microscopy Liang Gao1,2 1. 2. Department of Chemistry, Stony Brook University, Stony Brook, NY, USA Department of Biochemistry and Cell Biology, Stony Brook University, Stony Brook, NY, USA Selective plane illumination microscopy (SPIM) is a powerful technique used for 3D live fluorescence imaging in biological research [1-6]. Unlike conventional fluorescence imaging techniques based on the epi-illumination configuration, SPIM uses two objectives with optical axis orthogonal to each other for sample excitation and fluorescence detection separately. By confining the excitation light near the detection focal plane, SPIM allows high speed 3D imaging with high 3D spatial resolution, good optical sectioning capability, and minimal photobleaching and phototoxicity. In order to maximize the benefit of SPIM, a key problem is how to create a thin and uniform excitation light sheet to cover the region of interest and confine the excitation light near the detection focal plane as much as possible. Different methods have been developed to create the excitation light sheet in SPIM, including the Gaussian light sheet created by cylindrical lenses or scanning Gaussian beams [1-3, 5, 6], the Bessel light sheet created by scanning Bessel beams [4, 7, 8], the Airy light sheet created by scanning Airy beams [9], and the optical lattice light sheet developed more recently created by dithering nondiffracting optical lattice patterns [10]. However, each of these methods has its own strengths and weaknesses, and it is very difficult to obtain a light sheet with thin thickness, large size and tight excitation light confinement at the same time. Therefore, tradeoffs must be made among these factors based on the desired performance and the sample to be imaged no matter what method is to be used to create the excitation light sheet. The selection of the SPIM excitation light sheet and the optimization of its geometry are complicated. It involves the implementation of the appropriate method to create the light sheet and the tuning of the light sheet geometry based on the application and the specimen to reach the optimal balance between spatial resolution, optical sectioning capability and field of view (FOV). Here, we present a strategy to select, optimize and estimate the linear excitation light sheet in SPIM using the Gaussian light sheet, the Bessel light sheet, and the lattice light sheet as examples. We show that the spatial resolution of SPIM is determined by both the detection NA and the thickness of the excitation light sheet. However, whether the theoretical resolution can be obtained practically is determined by the optical sectioning capability of SPIM. We also show that the excitation light sheet should always stay in focus in SPIM, and the off focus excitation light should be reduced as much as possible. The alignment of SPIM using higher detection NA and thinner excitation light sheet is more critical, and the optical sectioning capability is worse with higher detection NA than that with lower detection NA, although the spatial resolution is generally higher. By using a detection NA of 1.0 or above, the Gaussian light sheet satisfies general requirements for spatial resolution, FOV and optical sectioning capability when the required SPIM axial resolution is above a micron, and the lattice light sheet should be used when a submicron axial resolution is required because it is thinner than the Gaussian light sheet of the same length and confines excitation light better than the Bessel light sheet of the same thickness. Meanwhile, the thickest and shortest light sheet that is able to give the required axial Microsc. Microanal. 21 (Suppl 3), 2015 1572 resolution and FOV should always be used for all types of the excitation light sheet to minimize the off focus excitation light in SPIM. References: [1] J. Huisken et al., "Optical Sectioning Deep Inside Live Embryos by Selective Plane Illumination Microscopy," Science 305, 1007-1009 (2004). [2] P. J. Keller et al., "Reconstruction of zebrafish early embryonic development by scanned light sheet microscopy," Science 322, 1065-1069 (2008). [3] U. Krzic et al., "Multiview light-sheet microscope for rapid in toto imaging," Nat. Methods 9, 730733 (2012). [4] T. A. Planchon et al., "Rapid three-dimensional isotropic imaging of living cells using Bessel beam plane illumination," Nat. Methods 8, 417-423 (2011). [5] R. Tomer et al., "Quantitative high-speed imaging of entire developing embryos with simultaneous multiview light-sheet microscopy," Nat. Methods 9, 755-763 (2012). [6] Y. Wu, et al., "Inverted selective plane illumination microscopy (iSPIM) enables coupled cell identity lineaging and neurodevelopmental imaging in Caenorhabditis elegans," Proc. Natl. Acad. Sci. U.S.A. 108, 17708-17713 (2011). [7] L. Gao, et al., "3D live fluorescence imaging of cellular dynamics using Bessel beam plane illumination microscopy," Nat. Protocols 9, 1083-1011 (2014). [8] L. Gao, et al., "Noninvasive imaging beyond the diffraction limit of 3D dynamics in thickly fluorescent specimens," Cell 151, 1370-1385 (2012). [9] T. Vettenburg, et al., "Light-sheet microscopy using an Airy beam," Nat. Methods 11, 541-544 (2014). [10] B.-C. Chen, et al., "Lattice light-sheet microscopy: Imaging molecules to embryos at high spatiotemporal resolution," Science 346, 1257998 (2014).");sQ1[785]=new Array("../7337/1573.pdf","Development of Amorphous Carbon Thin Film Phase Plate","","1573 doi:10.1017/S1431927615008648 Paper No. 0785 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Development of Amorphous Carbon Thin Film Phase Plate Y. Konyuba1, H. Iijima1, N. Hosogi1, Y. abe2, I. Ishikawa1, and Y. Ohkura1 1. 2. JEOL Ltd., 3-1-2 Musashino, Akishima, Tokyo, 196-8558, Japan. Yamagata Research Institute of Technology, Matsuei 2-2-1, Yamagata, 990-2473, Japan. Recent years, in transmission electron microscopy (TEM), the phase plate enhances phase contrast of a specimen image. Naturally, we expect that TEM with the phase plate enables us to obtain high contrast images of the samples, which are composed of light elements such as biological samples and polymer samples. So far, many types of phase plates for electron microscopy have been proposed. The most productive type of the phase plate is the thin film. Above all thin film, Zernike phase plate has been producing promising results [1-2]. However, thin film Zernike phase plate had some problems, those are, their reliability, lifetime (due to charging and aging) and cost (due to craft production including hole forming by a focused ion beam). To solve these problems, we have been challenging to fabricate several kinds of the thin film phase plates with various materials and structures by a high throughput fabrication method utilizing a micro electro mechanical systems (MEMS) technology. As a first trial, we have fabricated titanium (Ti) / silicon nitride (SiN) / Ti sandwich type thin film Zernike phase plates [3] and we improved the manufacturing yield. However, we could not achieve sufficient stability, due to charging of the Ti/SiN/Ti thin film Zernike phase plate. Next, we tried to fabricate the amorphous carbon thin film Zernike phase plate. The fabrication method for this phase plate is similar to one for the Ti/SiN/Ti thin film phase plate [4]. To improve the characteristics for the thin film phase plate: crystallization of the amorphous film, electro resistibility, thickness controllability and others, we adjusted the deposition condition for amorphous carbon thin film Zernike phase plates. Finally, we have succeeded to fabricate an amorphous carbon thin film phase plate having sufficient characteristics. Figures 1(a) and 1(b) show SEM images of top and bottom surfaces of amorphous carbon thin film Zernike phase plate. As shown in the figures, the amorphous carbon thin films are clean. The thickness of the amorphous carbon film was fabricated to be approximately 30 nm that gave /2 phase shift for 200 kV electrons. In the center of each square window, there is a small hole where the unscattered electrons pass through. The diameter of the hole was fabricated to be approximately 0.7 m as shown in Figure. 2. The stability of thus-fabricated phase plate in TEM imaging was good enough for sample observation. It is worthy to say that a hole-free phase plate [5-6] fabricated with the same method was also confirmed to show good performance. References: [1] R Danev et al, Ultramicroscopy, 88, (2001), p. 243. [2] W Dai et al, Nature protocols 9, (2014), p. 2630. [3] Y Konyuba et al, Microscopy and Microanalysis, 20.s3, (2014), p. 222. Microsc. Microanal. 21 (Suppl 3), 2015 1574 [4] H Iijima, Y Konyuba, US Patent 8,829,436. EP 2750160. JP 2014-130715. [5] M Malac et al, Ultramicroscopy 118 (2012), p. 77. [6] M Malac et al, U.S. Patent US 8,785,850. Figure 1. SEM images of the amorphous carbon thin film Zernike phase plate array. (a) top side (b) bottom side. Figure 2. TEM image of the center holes in the amorphous carbon thin film Zernike phase plate. The hole diameter is approximately 0.7 m.");sQ1[786]=new Array("../7337/1575.pdf","Thin-Film-Based Phase Plates for Transmission Electron Microscopy Fabricated From Metallic Glasses","","1575 doi:10.1017/S143192761500865X Paper No. 0786 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Thin-Film-Based Phase Plates for Transmission Electron Microscopy Fabricated From Metallic Glasses M. Dries1, S. Hettler1, T. Schulze1, W. Send1, E. M�ller1, R. Schneider1, D. Gerthsen1, Y. Luo2 and K. Samwer2 1. 2. Laboratorium f�r Elektronenmikroskopie, Karlsruher Institut f�r Technologie, Karlsruhe, Germany I. Physikalisches Institut, Universit�t G�ttingen, G�ttingen, Germany Intense development of physical phase plates (PPs) in the past decade led to substantial improvements in transmission electron microscopy (TEM) imaging of weak-phase objects. Research has focused on thinfilm PPs, which are typically fabricated from amorphous carbon (aC)-films [1]. Amorphous carbon has two important properties, which are essential for phase-contrast TEM: A sufficiently high electrical conductivity and an amorphous structure to avoid Bragg diffraction in the PP material. Thin-film PPs based on aC-films have already become widely accepted to enhance the contrast of weak-phase objects in TEM [2]. However, the irradiation with high-energy electrons initiates a steady, irreversible degeneration of the aC-film, which reduces the lifetime of aC-film-based PPs. Therefore, recent investigations have focused on the search of alternative materials with an improved material stability [3,4]. This study, for the first time, presents thin-film PPs fabricated from a metallic glass alloy. Metallic glasses are characterized by a high electrical conductivity and an amorphous structure. Moreover, structural degradation under the intense electron-beam is not expected if the crystallization temperature is high enough. Zr65.0Al7.5Cu27.5 (ZAC) was chosen for its favorable properties and its high crystallization temperature of 437 �C [5]. We have applied Hilbert PPs (HPPs) in this work, which consist of a microstructured thin film located in the back focal plane of the objective lens [6]. The film thickness is adjusted in such a way, that a phase shift of is imposed on the electrons in one half of the diffraction pattern except for the zero-order beam. This yields an overall phase shift of /2 for spatial frequencies above the cut-on frequency. The ZAC-film was sputtered on a cleaved mica-substrate and floated on a Cu-grid. Using a focused ionbeam system, rectangular windows were structured into the ZAC-film, which yields HPPs in several meshes of the Cu-grid. The Cu-grid was mounted in an objective aperture stripe and implemented in the back focal plane of a Philips CM200 FEG/ST. At an acceleration voltage of 200 kV, the ZAC-film of 24 nm thickness induces a phase shift close to . Fig. 1a shows a cross-section TEM image of a ZAC-film sputtered on a Si-substrate. Even short periods at ambient air lead to an oxide layer of 4 nm thickness, which appears with intermediate gray contrast in Fig. 1a. The oxygen is also visible in the composition profile shown in Fig. 1b, which was obtained by energy dispersive X-ray spectroscopy (EDXS). The electrically insulating oxide layer causes electrostatic charging of the ZAC-film, which affects its phase shifting behavior. Therefore, a thin aC-coating was applied to the ZAC-film. Fig. 2a depicts the power spectrum of an amorphous test object, which demonstrates the desired phase shifting properties. The power spectrum is subdivided in a central stripe (red) and outer areas (green) with Thon-rings shifted by /2. The complementary behavior in the two regions is also demonstrated by the azimuthally averaged intensity profiles shown in Fig. 2b. Although amorphous carbon was not fully removed from the PP-production process, the properties of Microsc. Microanal. 21 (Suppl 3), 2015 1576 metallic glasses are promising to improve the applicability of thin-film PPs for phase-contrast TEM [7]. References: [1] R Danev and K Nagayama, Ultramicroscopy 88 (2001), p. 243-252. [2] R Danev et al, PNAS 111 (2014), p. 15635-15640. [3] M Marko et al, J. Struct. Biol. 184 (2013), p. 237-244. [4] M Dries et al, Ultramicroscopy 139 (2014), p. 29-37. [5] R Rambousky, M Moske and K Samwer, Z. Phys. B 99 (1996), p. 387-391. [6] R Danev and K Nagayama, J. Phys. Soc. Jpn. 73 (2004), p. 2718-2724. [7] Financial support by the Deutsche Forschungsgemeinschaft (DFG). Figure 1. Formation of oxide layers at the surface of ZAC-films. (a) Cross-section TEM image of an oxidized ZAC-film with a Pt/C-protection layer on top. (b) Composition profile along the white arrow in (a) obtained by EDXS measurements. Figure 2. Phase shifting behavior of the aC-coated ZAC-film-based HPP. (a) Power spectrum of a phase-contrast TEM image of an amorphous test object. (b) Azimuthally averaged intensity profiles taken from the Thon-rings in the red and green regions of (a).");sQ1[787]=new Array("../7337/1577.pdf","Optimization of JEM2200FS for Zernike Phase Contrast Cryo-EM","","1577 doi:10.1017/S1431927615008661 Paper No. 0787 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Optimization of JEM2200FS for Zernike Phase Contrast Cryo-EM Htet A. Khant1, Caroline Fu1, Sohei Motoki2, Matthew H. Sullivan3, Guy DeRose3 and Wah Chiu1 1 National Center for Macromolecular Imaging, Verna and Marrs McLean Department of Biochemistry and Molecular Biology, Baylor College of Medicine, Houston, TX 77030,USA 2 JEOL USA, Inc., 11 Dearborn Road, Peabody, MA 01960, USA 3 Kavli Nanoscience Institute, California Institute of Technology, Pasadena, CA 91125 USA Zernike phase contrast optics has been demonstrated to be effective in enhancing image contrast for biological specimens. The Zernike phase plate is a thin carbon film with a central hole positioned at the back focal plane of the objective lens (OL) which shifts the modulation of the contrast transfer function from a sine to a cosine function, thereby enhancing the image contrast at low spatial frequencies while maintaining the high resolution information. Cryo-electron microscopy (cryo-EM) uses a transmission electron microscope (TEM) to study macromolecular complexes in their native state. It can also reveal small molecular components in a large macromolecular assembly or identify various subcellular components in a cryo-electron tomogram of cells [1,2]. In order to study complexes that are either small, conformationally heterogeneous or that reside in a crowded cellular environment, the overall system must be able to provide high resolution images with adequate signal-to-noise ratio. The set up of robust TEM and fabrication of enduring phase plates have been challenging. It is a major challenge to build a fully integrated, high performance TEM that is both user-friendly and robust enough to study various biological systems. Progress has been made by different commercial sources using various methods to make the Zernike phase plate and custom configuration of both hardware and electron optics of JEM2200FS. Over the past several years, we have been optimizing our JEM2200FS for high resolution Zernike phase contrast cryo-EM. An important consideration in configuring such a system depends on a number of parameters including the objective lens configuration, anti-contaminator design and types of cryo-specimen tilt holders. Figure 1 shows two configurations in JEM2200FS electron microscope. Their primary differences lie on the design of the anti-contaminator and the focal length of the objective lens. Due to the limited life span of phase plates, we installed an airlock system to allow for the rapid exchange of phase plates without breaking the microscope vacuum. However, the pole piece had a relatively long focal length as well as a weak prefield, which required a long exposure time to achieve a typical single particle cryo-EM dose rate. We then installed a new pole piece with a shorter focal length that has a smaller illumination area, which increased the dose rate to an optimal value (Fig. 1). However, the degree of contrast improvement from the Zernike phase plate depends on its cut-on frequency, which is inversely proportional to the diameter of the central hole and directly proportional to the effective focal length of the lens. Therefore, to recover the contrast lost with the shorter focal length pole piece, we fabricated smaller diameter holes (Fig. 2). As an alternate strategy, we installed another airlock phase plate system at the selected area (SA) aperture, which is at back focal plane of objective mini lens (conjugate to the back focal plane of the OL). This configuration increased the effective focal length (Fig. 3) approximately two times. Using the phase plate in the SA position, we can now also use an objective aperture at the OL back focal plane, which has the added benefit of reduced specimen charging. However, SA phase plate configuration has a maximum microscope magnification of 25,000x. We also evaluated the performance of our system with a series of experiments. Image distortion for SA phase plate configuration did not appear to be significantly greater than that from standard imaging mode based upon catalase crystal diffraction standards. Our ongoing tests show the graphitized carbon lattice spacing (3.4 �) with the OL phase plate and 12.35� diffraction spots from a frozen hydrated Microsc. Microanal. 21 (Suppl 3), 2015 1578 catalase crystal with the SA phase plate (Fig. 4). The alignment and low dose operation of SA phase plate is being further optimized to improve resolution and stability. References: [1] Danev, R & Nagayama, K (2008) Single particle analysis based on Zernike phase contrast transmission electron microscopy. J Struct Biol 161(2):211-218. [2] Dai, W, Fu, C, Raytcheva, D, Flanagan, J, Khant, HA, Liu, X, Rochat, RH, Haase-Pettingell, C, Piret, J, Ludtke, SJ, Nagayama, K, Schmid, MF, King, JA, & Chiu, W (2013) Visualizing virus assembly intermediates inside marine cyanobacteria. Nature 502(7473):707-710. Figure 1. A comparison of the illumination area of the original pole piece (left) and new pole piece with a shorter focal length. Figure 2. The effect of the cut-on frequency of Zernike phase plate on image contrast. Left, a defocus contrast image of a latex sphere on a line grating grid. Middle image was acquired from a 0.7�m diameter phase plate hole, whereas right image was obtained from a 0.35 �m diameter phase plate hole. Figure 3. Electron optical ray diagram for phase plates located at objective aperture (left) and selected area aperture (right). Notice the objective aperture cannot be inserted in the left diagram since the space is occupied by the OL phase plate. Acknowledgement: This research has been supported by a NIH grant (P41GM103832). Figure 4. Graphitized carbon image (left) obtained from the OL phase plate; inset is the Fourier transform showing the 3.4 � lattice spacing (Gatan US4000 CCD at 1.1 �/pix). Frozen hydrated catalase crystal image (right) obtained from the SA phase plate showing diffraction spots at 12 � resolution (Direct Electron DE12 at 2.0 �/pix).");sQ1[788]=new Array("../7337/1579.pdf","Initial Experience with the Volta Phase Plate","doi:10.1017/S1431927615008673","1579 doi:10.1017/S1431927615008673 Paper No. 0788 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Initial Experience with the Volta Phase Plate Michael Marko1, and Chyongere Hsieh1 1. Wadsworth Center, New York State Department of Health, Empire State Plaza, Albany, NY 12201 Specimens prepared solely by vitreous freezing facilitate high-resolution imaging of macromolecules in-situ and in a near-native state. However, such samples are weakly scattering, suffering from low contrast and requiring phase-contrast imaging. In addition, high-resolution information is limited by electron-radiation damage. Cryo-TEM technology has steadily been improving so that improved image contrast can be obtained without increased electron dose. After cryo-TEM specimen stages became common, the next advance was implementation of zero-loss energy filtering, approximately doubling contrast without increasing the electron dose [1]. Starting in 2000 with the renaissance of phase-plate imaging, the oscillating contrast-transfer function of traditional defocus-phasecontrast imaging could be avoided, providing near-uniform transfer of information over a wide range of spatial frequencies [2]. Very recently, practical implementation of direct-electron detectors have led to great advances [3,4], due to their improved efficiency over previous cameras or photographic film, and also due to a rapid frame rate, such that alignment of a stack of frames can correct for specimen movement during exposure [5,6,7]. We are concerned with optimizing phase-contrast imaging. While there are many means to do so, as reviewed by Glaeser [8], we are currently concentrating on phase plates. The Zernike phase plate (ZPP) proved itself in early biological applications [9,10] and has become the standard for practical application [11]. However, the introduction of the Volta phase plate (VPP) [12] promises to replace the Zernike for routine applications [e.g. 13]. A Volta phase plate has the same configuration as a "hole-free phase plate" (HFPP) [14], and can be easily constructed in any TEM lab by simply placing a thin (5-15 nm) carbon film over an objective aperture disc. Unlike the ZPP, there is no requirement to create, and then accurately center, a small central hole. A VPP must be operated above 200�C, while the HFPP operates at room temperature. Images made using a VPP or HFPP do not contain the fringe artifacts, which are a result of the abrupt edge of the central hole in the ZPP. We have been using ZPPs in our JEOL JEM-3200FSC/PP cryo-TEM for some time. This TEM is equipped with JEOL's airlock-type heated phase-plate holders [15] in both objective aperture and selected-area (SA) diffraction-plane positions, and it has a special transfer lens to effect a three-times enlargement of the objective-lens back-focal plane at the SA position. For cryo-TEM low-dose imaging, our TEM has a coarse stage piezo that can move the specimen along the tilt axis to reach the low-dose focus/tracking position. We realized that it would be simple to use a VPP or HFPP in our TEM, and indeed the results so far with the VPP are excellent (Fig. 1). We collect tomographic tilt series with a modified version of SerialEM [16] that employs our special stage piezo. However, a simple firmware modification by JEOL allows access to the focus/tracking position by means of image shift and beam shift, while maintaining parallel illumination (i.e. the phase-plate "on plane" condition), but without any beam tilt that would cause displacement of the beam from the active center-point on the phase plate. We use JEOL's standard phase-plate positioning system [15] to select and recall the best VPPs on our multi-hole objective aperture. Tilt-series collection proceeds automatically, without the need to re-center the VPP active area. Our TEM is ideally configured to test a variety of phase-shifting devices. We will compare thin-film, holefree phase plates at both room temperature and above 200�C (an HFPP / VPP comparison). We have also started to test ZPPs in the "transfer mode" (SA position) to investigate the effects of a lower cut-on frequency Microsc. Microanal. 21 (Suppl 3), 2015 1580 afforded by the longer effective objective-lens focal length (Fig. 2). While low-frequency contrast is clearly greater than from a VPP at the normal objective aperture position, the typical artifacts due to the abrupt cut-on of the ZPP are apparent, although they could be reduced by filtering [17,18]. We have also found that a VPP functions in "transfer mode", and we are investigating whether or not this results in improved contrast at very low spatial frequency. [19] References [1] R Grimm et al, J. Microsc. 190 (1998), p. 339. [2] R Danev et al, Ultramicroscopy 109 (2009), p. 312. [3] N Grigorieff eLife 2 (2013), p. e00563. [4] W K�hlbrandt, eLife 3 (2014), p. e03678. [5] B Bammes et al, J. Struct. Biol. 177 (2012), p. 589 [6] X Li et al., Nat. Methods 10 (2013), p. 584. [7] G. McMullan et al, Ultramicroscopy 109 (2009), p. 1126 [8] RM Glaeser, Rev. Sci. Instrum. 84 (2013), p. 111101. [9] R Rochat et al, J. Virol. 85 (2011), p. 1871. [10] K Murata et al, Structure 18 (2010), p. 903. [11] M Marko et al, J. Struct. Biol. 184 (2013), p. 237. [12] R Danev et al, PNAS USA 111 (2014), p. 15635. [13] S Asano et al, Science 347 (2015), p. 439. [14] M. Malac et al, Ultramicroscopy 118 (2012), p. 77. [15] W Dai et al, Nat. Protocols 9 (2014), p. 2630. [16] http://bio3d.colorado.edu/SerialEM/ [17] H Sui et al, Microsc. Microanal 20 (2014), p. 234. [18] R Danev and K. Nagayama Ultramicroscopy 111 (2011) p. 1305. [19] Thanks to S. Motoki of JEOL USA for TEM support, D. Mastronarde of UC Boulder for SerialEM support, and F. Maley of Wadsworth Center for T4 phage. Supported by NIH grant GM103555 (to M. Marko). Figure 1. T4 phage with 10-nm Au particles. VPP Cryo-TEM at 300 keV with in-column energy filter and K2 Summit camera in counting mode. Figure 2. T4 phage imaged as in Fig 1, but with ZPP at 3X greater focal length. Hole size 700 nm. Note increased contrast of tails, and "fringe" artifacts on the left.");sQ1[789]=new Array("../7337/1581.pdf","High-Resolution Transmission Electron Microscopy With Zach Phase Plate","","1581 doi:10.1017/S1431927615008685 Paper No. 0789 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 High-Resolution Transmission Electron Microscopy With Zach Phase Plate Simon Hettler1, Manuel Dries1, Tina Schulze1, Marco Oster2, Christian Wacker2, Rasmus R. Schr�der2 and Dagmar Gerthsen1 1. 2. Karlsruher Institut f�r Technologie, Laboratorium f�r Elektronenmikroskopie, Karlsruhe, Germany. BioQuant CellNetworks, Universit�t Heidelberg, INF 267, D-69120 Heidelberg, Germany. Physical phase plates (PP) for transmission electron microscopy (TEM) have gained tremendous attention in the past decade. PPs introduce a relative phase shift between scattered and unscattered electrons and thus enhance the contrast of weak-phase objects in TEM. Several approaches to PP TEM exist [1] with thin-film PPs being the most frequently used concept up to now [2]. A major disadvantage of thin-film PPs is scattering of electrons by the PP itself which induces a damping of the phase contrast transfer and reduces the resolution of the microscope. The electrostatic Zach-PP uses a strongly localized electrostatic potential to induce the phase shift and is not affected by this effect [3]. In this work, the application of a Zach-PP for high-resolution (HR) TEM is investigated by studying a crystalline Si-sample. The Zach-PP (Fig. 1) consists of a single rod with an Au electrode which is surrounded by insulating and shielding electrically conductive layers. If a voltage is applied to the electrode, an inhomogeneous potential emerges at the PP tip. The PP was implemented in the back-focal plane of a Zeiss 923 transmission electron microscope equipped with a field-emission gun operated at 200 kV and a 4k TVIPS CCD-camera. For PP TEM imaging, the tip is usually positioned close to the zero-order beam to minimize the cut-on frequency and enhance the contrast of weak-phase objects up to 10 nm size [4]. For HR PP TEM in this work, the cut-on frequency is of minor importance and the PP was positioned at an increased distance to the zero-order beam to reduce charging effects. Fig. 2 presents results obtained with a single-crystalline Si-sample in [110] zone-axis orientation. Fig. 2a shows a HR PP TEM image in an overview perspective which was acquired without applied voltage. A relatively thick amorphous region is observed which results from Si-oxide formation and amorphization during sample preparation. The marked area is displayed enlarged in Fig. 2b where (111)-type lattice fringes are clearly resolved. The power spectrum of the HR PP TEM image (Fig. 2a) is shown in Fig. 2c and reveals the four (111)-type and the (002)-reflections as well as the PP rod. The influence of the phase shift on the reflection intensities is analyzed by acquiring an image series with varying the PP voltage from -5 V to + 5 V. The reflection intensity of the (111)-reflections in the corresponding power spectra is plotted in Fig. 2d, where the blue and red crosses are the averaged intensities from the reflections marked + )| as predicted by theory. The phase in Fig. 2c. The data is well fitted by a function ~|cos ( comprises the wave aberration function and the phases of the zero-order beam and the corresponding (111)-type reflection. The results show, that the electrostatic Zach-PP does not reduce the resolution of the microscope and is well suited for HR TEM applications [5]. References: [1] R M Glaeser, Rev. Sci. Instrum. 84 (2013), 111101. [2] R Danev, K Nagayama, Ultramicroscopy 88 (2001), p. 243-252. [3] K Schultheiss et al, Microsc. Microanal. 16 (2010), p. 785�794. [4] N Frindt et al, Microsc. Microanal. 20 (2014), p. 175�183. [5] Financial support by German Research Foundation (DFG) under contract Ge 841/16 and Sch 424/11. Microsc. Microanal. 21 (Suppl 3), 2015 1582 Figure 1. Scanning electron microscopy images of the Zach PP used in the experiments. (a) Image of the PP tip shows the layer system of central electrode, surrounding insulating layers and shielding metal layers. (b) The PP aperture has a diameter of 90 �m. Figure 2. HR PP TEM images and analysis. (a) Overview HR PP TEM image of the Si-sample in [110] zone-axis orientation. (b) Detail image of the area marked by a black frame in (a). (c) Power spectrum of the image in (a) with (002)-reflections and (111)-type reflections marked by the red and blue circles. (d) Plot of the averaged reflection intensity of the four (111)-type reflections marked blue and red in (c) as a + )| behavior. function of the applied voltage. Solid lines represent fits with a |cos (");sQ1[790]=new Array("../7337/1583.pdf","TEM Video Compressive Sensing","","1583 doi:10.1017/S1431927615008697 Paper No. 0790 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 TEM Video Compressive Sensing Andrew Stevens1,2 , Libor Kovarik1 , Patricia Abellan1 , Xin Yuan2 , Lawrence Carin2 , and Nigel D. Browning1 1. 2. Pacific Northwest National Laboratory, {NSD, EMSL, EMSL, FCSD}, Richland USA Duke University, ECE, Durham USA One of the main limitations of imaging at high spatial and temporal resolution during in-situ TEM experiments is the frame rate of the camera being used to image the dynamic process. While the recent development of direct detectors has provided the hardware to achieve frame rates approaching 0.1ms, the cameras are expensive and must replace existing detectors. In this paper, we examine the use of coded aperture compressive sensing methods [1, 2, 3, 4] to increase the framerate of any camera with simple, low-cost hardware modifications. The coded aperture approach allows multiple sub-frames to be coded and integrated into a single camera frame during the acquisition process, and then extracted upon readout using statistical compressive sensing inversion. Our simulations show that it should be possible to increase the speed of any camera by at least an order of magnitude. Compressive Sensing (CS) combines sensing and compression in one operation, and thus provides an approach that could further improve the temporal resolution while correspondingly reducing the electron dose rate. Because the signal is measured in a compressive manner, fewer total measurements are required. When applied to TEM video capture, compressive imaging could improve acquisition speed and reduce the electron dose rate. CS is a recent concept, and has come to the forefront due the seminal work of Cand�s [5]. Since the publication of Cand�s, there has been enormous growth in the application of CS and development of CS variants. For electron microscopy applications, the concept of CS has also been recently applied to electron tomography [6], and reduction of electron dose in scanning transmission electron microscopy (STEM) imaging [7]. To demonstrate the applicability of coded aperture CS video reconstruction for atomic level imaging, we simulate compressive sensing on observations of Pd nanoparticles and Ag nanoparticles during exposure to high temperatures and other environmental conditions. Figure 1 highlights the results from the Pd nanoparticle experiment. On the left, 10 frames are reconstructed from a single coded frame--the original frames are shown for comparison. On the right a selection of three frames are shown from reconstructions at compression levels 10, 20, 30. The reconstructions, which are not post-processed, are true to the original and degrade in a straightforward manner. The final choice of compression level will obviously depend on both the temporal and spatial resolution required for a specific imaging task, but the results indicate that an increase in speed of better than an order of magnitude should be possible for all experiments [8]. References: P Llull, X Liao, X Yuan et al. Optics Express 21(9), (2013), p. 10526. X Yuan, J Yang, P Llull et al. In ICIP 2013 (IEEE), p. 14. J Yang, X Yuan, X Liao et al. Image Processing, IEEE Transactions on 23(11), (2014), p. 4863. X Yuan, P Llull, X Liao et al. In CVPR 2014 (IEEE), p. 3318. EJ Cand�s, J Romberg and T Tao. Information Theory, IEEE Transactions on 52(2), (2006), p. 489. P Binev, W Dahmen, R DeVore et al. In Modeling Nanoscale Imaging in Electron Microscopy, eds. T Vogt, W Dahmen and P Binev (Springer US), Nanostructure Science and Technology (2012). p. 73. [7] A Stevens, H Yang, L Carin et al. Microscopy 63(1), (2014), pp. 41. [1] [2] [3] [4] [5] [6] Microsc. Microanal. 21 (Suppl 3), 2015 1584 [8] This work was supported in part by US DOE Grant No. DE-FG02-03ER46057. This research is also part of the Chemical Imaging & Signature Discovery Initiatives conducted under the LDRD Program at Pacific Northwest National Laboratory (PNNL) and used EMSL, a national scientific user facility sponsored by the DOE Office of Bio & Env. Research at PNNL. PNNL is a multi-program national lab operated by Battelle Memorial Institute under Contract DE-AC05-76RL01830 for the US DOE. Figure 1: Left: An illustration of CS inversion from 10 frames compressed into 1. The top left image shows the compressed frame, the middle column of images shows the reconstructed frames, and the right column shows the original frames. During the sequence a peak atop the nanoparticle forms. Even though the peak is not visible in the compressed data it is accurately reconstructed. Right: From top to bottom original, 10�, 20�, and 30� compressed reconstruction; from left to right frames 109, 127, and 157. There is a significant denoising effect in the reconstructed images. The salient features remain, but contrast reduces as compression increases. The width of the imaged region is 26.67nm. Full resolution visible with zoom.");sQ1[791]=new Array("../7337/1585.pdf","Imaging At the Timescale Of Micro- and Milliseconds With the pnCCD (S)TEM Camera","","1585 doi:10.1017/S1431927615008703 Paper No. 0791 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Imaging At the Timescale Of Micro- and Milliseconds With the pnCCD (S)TEM Camera H. Ryll1, M. Simson2, M. Den Hertog3, R. Dunin-Borkowski4, K. El Hajraoui3, R. Hartmann1, M. Huth1, S. Ihle1, V. Migunov4, J. Schmidt3, H. Soltau2 and L. Str�der1 1. 2. PNSensor GmbH, Otto-Hahn-Ring 6, 81739 Munich, Germany PNDetector GmbH, Otto-Hahn-Ring 6, 81739 Munich, Germany 3. Institut N�el-CNRS, 25 Rue des Martyrs, 38042 Grenoble, France 4. Ernst Ruska-Centre , Peter Gr�nberg Institute, Forschungszentrum J�lich, Leo-Brandt-Stra�e, 52425 Juelich, Germany Dynamic processes that can, in principle, be observed in TEM imaging can be too fast to be resolved with conventional cameras which are typically running below 40 frames per second (fps). The pnCCD (S)TEM camera is routinely running at 1000 fps in full frame mode [1]. This camera uses a direct detecting, radiation hard pnCCD with 264x264 pixels and features binning and windowing modes which substantially increase the frame rate. For example, 4-fold binning in one direction, i.e. 66x264 pixels, yields a readout speed of 4000 fps. In windowing modes up to 20000 fps are possible. The propagation of a metal-semiconductor phase inside a nanowire was observed in-situ with the pnCCD (S)TEM camera. A silicon nanowire was contacted with Pt strips at each end in an electrical biasing holder. By flowing a current through a Pt strip the nanowire's temperature was controlled by Joule heating which excited the propagation of a Pt-Si phase into the nanowire [2]. Images of the nanowire during the growth process were recorded continuously with 1000 fps for over 30 minutes. The resulting high speed movie shows the growth process with millisecond time resolution. Single frames from the movie are shown in Figure 1. Various abrupt propagation steps could be observed in the movie. This shows the potential of the pnCCD (S)TEM camera for imaging of dynamic processes at the millisecond timescale. The flexible configuration of the camera readout permitted a mode with even shorter integration times down to 25 �s at a readout speed of 1850 images per second. Such a short integration time was required for imaging of oscillations of a CdS nanowire. The experimental setup consisted of a free standing nanowire [3] that was brought close to a probe tip. A sine voltage was applied to the probe tip that induced resonant vibrations of the nanowire. Part of the oscillating nanowire's shadow image was then recorded with the camera. The readout of the camera was triggered with a variable delay with respect to the zero crossing of the sine function (see Figure 2a). In this way, it was possible to sample the position of the nanowire at the corresponding value of the sine voltage. The decisive configuration was to decrease the integration time of each single image to 25 �s, otherwise the high velocity movement of the nanowire would blur the image. The excitation frequency was 1850 Hz, the camera was read out with the same speed. In order to shorten the integration time, only a part of the image area was read out. The readout sequence starting at the trigger signal consisted of three steps: 1) clear pixel of signal charge, 2) integrate for a set time (here 25 �s), 3) readout of charge collected in the pixel. In the analysis of the measurement, for each delay, the center of gravity of the intensity distribution was calculated for each image as the position of the nanowire. The dependency of the position of the nanowire on the delay and hence the exciting voltage reveals that the resonant frequency of the nanowire Microsc. Microanal. 21 (Suppl 3), 2015 1586 is approximately 4 times the excitation frequency (Figure 2b). Elastic properties of the nanowire could then be deduced through further analysis. In conclusion, the imaging capabilities of the pnCCD (S)TEM camera at millisecond time scales and below have been shown. Live, high-speed movies of platinum growth in nanowires were recorded with 1000 frames per second. Integration times of single images down to 25 �s were possible with flexible readout modes. [1] H. Ryll et al., Microscopy and Microanalysis 19 (2013), p.1160-1161 [2] M. Mongillo et al., ACS Nano 5 (2011), p. 7117-7123 [3] R. Liu et al., Nano Letters 13 (2013), p. 2997-3001 Figure 1. Examples of single raw images from a movie sequence in which a Pt-Si phase is propagating in a silicon nanowire. Notice the area of growth indicated by the circle in the first image. Figure 2. Overview image of nanowire and probe tip position in a). A sine voltage was applied to the probe tip inducing oscillations in the nanowire. The shadow of the vibrating nanowire was imaged with an integration time of 25 �s. The center of gravity of these images was calculated and plotted against the trigger delay in b). Single images of the readout area are shown in c) for three different delays.");sQ1[792]=new Array("../7337/1587.pdf","NO2 Hydrogenation over Pt and Rh Catalysts: a Study at The Atomic Level by Field Emission Microscopy","","1587 doi:10.1017/S1431927615008715 Paper No. 0792 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 NO2 Hydrogenation over Pt and Rh Catalysts: a Study at The Atomic Level by Field Emission Microscopy C�dric Barroo1,2,3, Yannick De Decker2,3, Thierry Visart de Bocarm�2,3 and Norbert Kruse2,5 School of Engineering and Applied Sciences, Harvard University, Cambridge, MA 02138, USA (current affiliation) 2 Chemical Physics of Materials, Universit� libre de Bruxelles (ULB), CP243, 1050 Brussels, Belgium 3 Interdisciplinary Center for Nonlinear Phenomena and Complex Systems (CENOLI), Universit� libre de Bruxelles (ULB), 1050 Brussels, Belgium 4 Non Linear Physical Chemistry Unit, Universit� libre de Bruxelles (ULB), CP231, 1050 Brussels, Belgium 5 Department of Chemical Engineering and Bioengineering, Washington State University, Pullman, WA 99163, USA This study aims at investigating the catalytic hydrogenation of nitrogen dioxide (NO2) on Pt, Rh and PtRh alloy catalysts. Nitrogen oxides (NOx) are produced during the combustion of gasoline in the leanburn regime, and their emission remains an issue especially for Diesel-type engines. Platinum-GroupMetals are used as active components of the catalytic converter. They turn toxic gases into harm less compounds, for example, NOx into nitrogen. The catalyst is dispersed in the form of metallic nanoparticles of approximately 5 to 10 nm over a support of high specific surface area. In most studies of heterogeneous catalysis, the structure of the supported catalyst is determined before and after reaction. Unfortunately, very little information is available about the possible local morphological/structural changes that might occur during the reaction. Moreover, from a dynamical point of view, the kinetic parameters only reflect the global kinetics of the reaction taking place on the surface a large ensemble of nanoparticles with specific size and shape distribution. Thus, ensembleaveraging takes place eventually hiding more complex dynamics, such as periodic oscillations. A better understanding of the ongoing catalytic reaction would lead to a better reproducibility, predictability and control of this reaction. In this view, a study of the hydrogenation of NO2 gas over Pt, Rh and Pt-Rh nanocrystals is undertaken at the molecular level by means of field emission methods [1,2]. Experiments were carried out in a stainless steel field ion/electron emission microscope. Field emitter tips, used as model catalysts, were electrochemically etched, and then cleaned in situ. After characterization of the tips by Field Ion Microscopy (FIM), the samples were heated and a field of 5 V/nm was applied using Field Electron Microscopy (FEM). A mixture of NO2 and H2 was then admitted in the chamber, and the reaction was followed by recording the FEM patterns appearing dynamically on the screen of the microscope. Time series were extracted from videos (at 50 frames per second - fps) and established by probing the mean brightness over regions of interest (Fig1a). The dynamic of the process was analyzed by measuring the variations of grey levels with time, image by image. Transient phenomena were also monitored with an acquisition rate of 10,000 fps. The catalytic hydrogenation of NO2 was monitored in real time. On Pt at 390K, amongst several nonlinear behaviors, self-sustained periodic oscillations were observed (Fig1b) [3,4]. Fourier transform analyses and temporal autocorrelation functions were used to characterize the dynamics and to quantify the robustness of the kinetic oscillations. NO2 hydrogenation was also studied on Rh at 450K and found to present many similarities by with platinum. However, the details of the dynamics reveal significant differences, such as the emergence of oscillations via a different bifurcation, the robustness of the 1 Microsc. Microanal. 21 (Suppl 3), 2015 1588 system, as well as the pressure range for the occurrence of non-linear behaviors. Finally, the NO2+H2 reaction was studied on a Pt-17.4at.%Rh alloy catalyst of similar composition -that isused in catalytic converters. Surface explosions were observed in the temperature range of 390-515K. At 425K, periodic oscillations are observed with features lying between those observed on pure Pt and pure Rh samples. This observation implies the existence of synergistic effects between the metals. In the case of Pt, high-speed experiments highlighted the presence of diffusing processes down to the nanoscale between active facets. The propagation of chemical waves on a single facet of the nanocrystal was also observed. This proves the robustness of dissipative, nonlinear behaviors down to the nanoscale. The velocity of the observed waves is of the order of a few m/s, which is in good agreement with previous studies of catalytic reactions at the mesoscale. Field ion microscopy and field emission microscopy are powerful techniques in materials science and chemistry. By exploiting the nanoscale resolution of the techniques, the results obtained allow for a better understanding of catalytic systems at the molecular level. The experiments prove that robust nonlinear behaviors can be observed down to the nanoscale, without a significant loss of correlation due to fluctuations inherent to small chemical systems. These results shed a new light on the conditions under which collective order can emerge at the nanoscale [5]. References: [1] C. Barroo et al., New J. Chem. 38 (2014) 2090 [2] C. Barroo et al., Appl. Surf. Sci. 304 (2014) 2 [3] J.-S. McEwen et al., Langmuir 26 (2010) 16381 [4] C. Barroo et al., J. Phys. Chem. C 118 (2014) 6839 [5] C.B. thanks the Fonds de la Recherche Scientifique (F.R.S.-FNRS) for financial support (PhD grant). The authors gratefully thank the Van-Buuren Foundation for a financial support for the acquisition of equipment and the Wallonia-Brussels Federation (Action de Recherches Concert�es n�AUWB 20102015/ULB15). Figure 1. a) NO2-H2 reaction over platinum imaged by field emission microscopy (Reproduced from C. Barroo et al., New J. Chem. 38 (2014) 2090 with permission from the Centre National de la Recherche Scientifique (CNRS) and The Royal Society of Chemistry.) � b) Experimental time series over the (012) facet during self-sustained periodic oscillations");sQ1[793]=new Array("../7337/1589.pdf","Characterization of Dielectric Waveguides Through Photoemission Electron Microscopy (PEEM) in the Infrared Regime","","1589 doi:10.1017/S1431927615008727 Paper No. 0793 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of Dielectric Waveguides Through Photoemission Electron Microscopy (PEEM) in the Infrared Regime T. A. Stenmark1, Robert C. Word1 and R. K�nenkamp1 1 Department of Physics, Portland State University, Portland, OR 97201, USA Photoemission Electron Microscopy (PEEM) provides a powerful tool for imaging electric fields and plasmon excitation. PEEM relies on the photoelectric effect with the photoemission yield directly dependent on the intensity of the electric field at the surface of the object being imaged. These properties provide valuable information about diffraction and interference as well as surface plasmon propagation and photo emission thresholds. High power pulse lasers allow for multiphoton PEEM where multiple photons are required in order to emit a single electron. These techniques produce images with resolution down to around 5nm in the visible and infrared regime [1]. It is therefore possible to directly image the near-field region of the electromagnetic field [2]. We report here, for the first time, the direct observation polarization dependent phase shifts in the in-coupling of light in the 780nm regime at the edge of a linear dielectric waveguide. Using finite element techniques we show that these phase shifts are correctly predicted. Furthermore the obtained photoemission micrographs allow a determination of optical properties on the nanometer scale, i.e. optical near-field distributions can be quantified and energy transfer and loss process can be analyzed. This opens the doors to sub-wavelength and dynamic imaging in metamaterial devices. In our experimental setup the substrate is a 0.2mm thick glass platelet covered by an indium-tin-oxide (ITO) layer approximately 290 nm thick. The greater refractive index of the ITO layer creates a twodimensional waveguide. An FEI Strata 237 focused ion beam was used to define a strip waveguide of 2 um width and milled 50nm into the ITO layer. In this strip waveguide only discrete set of modes can be excited. These modes create an interference pattern with the incident light. For the in-coupling of light into the guide a rectangular slit was milled at the edge of the strip waveguide in a direction perpendicular to the excitation light, as shown in Fig. (1) A general analysis of the waveguide performance has been carried out recently [3]. These results are extended here to include the nearinfrared region, where a 3-photon process is needed to generate a free photoelectron. By taking the Fourier transform of the intensity distribution along the length of the waveguide and extracting the periodic information (fig 2) we are able to determine distinct single modes for both TE and TM polarizations for 780nm illumination and calculate the effective indexes of refraction for these modes. The resulting effective indexes are 1.596 for TE and 1.544 for TM which translates to a thickness of approximately 250nm according to numerical calculations [4] (fig 3). Additional simulations are highly consistent with numerical calculations matching effective indexes to 0.2% error. Our new results demonstrate the possibility of obtaining images with nanometer resolution for optical phenomena across the entire visible spectrum. These possibilities show PEEM to be an efficient tool for analyzing and modeling photon [3] and plasmon [5] dynamics and interactions at solid surfaces and devices. Microsc. Microanal. 21 (Suppl 3), 2015 1590 References [1] JPS Fitzgerald, RC Word, R K�nenkamp, Phys. Rev. B 89 (2014), 195129. [2] (6) Livio I. Chelaru and F.-J. Meyer zu Heringdorf, Appl. Phys. Lett. 89, 241908 (2006) [3]JPS Fitzgerald, RC Word, R K�nenkamp, Phys. Rev. B87 (2013), 205419. [4]A. Yariv, Optical Electronics (Saunders College Publishing, Philadelphia, 1991) [5] Niemma M. Buckanie et al., Ultramicroscopy 130, 49 (2013) Figure 1: Colorized PEEM images. 410nm illumination on the left and 780nm on the right. Note the noisier image due to reduced signal at the lower energy wavelength. Figure 2: Periodic data of the interference pattern. TE peak at 1.07um, TM 1.15um Figure 3: Numerical calculation of Effective index vs waveguide Thickness For ITO on a glass substrate at 780nm. Dots are where the experimental values for Neff fall on the graph.");sQ1[794]=new Array("../7337/1591.pdf","Characterization of Laser Ablation Dynamics for Nickel Thin Films on Silicon Using Movie Mode Dynamic TEM","","1591 doi:10.1017/S1431927615008739 Paper No. 0794 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of Laser Ablation Dynamics for Nickel Thin Films on Silicon Using Movie Mode Dynamic TEM Sahar Hihath1,2, Melissa K. Santala3, Geoffrey Campbell3 , Klaus van Benthem1 1 2 Department of Chemical Engineering and Materials Science, University of California, Davis, CA. Department of Physics, University of California, Davis, CA. 3 Condensed Matter and Materials Division, Lawrence Livermore National Laboratory, Livermore, CA Laser ablation processes, which involve the removal of a material from a substrate have important industrial applications in laser machining of structural components and micro-electronic devices. [1] Limited data is available in the literature about direct imaging of the laser ablation process that occurs for metal films with thicknesses below 100 nm since time resolutions in the nanosecond range or faster are required [1]. Dynamic TEM is an ideal tool to directly image laser ablation processes on the nanometer length scale, and provides a fundamental insight to the mechanisms that govern material removal from solid substrates during laser irradiation. Nickel thin films with a thickness of 64 nm were sputter deposited at room temperature on the (100) surface of silicon substrates that were covered with a native oxide layer. Plan-view TEM samples were prepared by conventional grinding, polishing, dimpling, and subsequent Ar+ ion-milling for 3 hours at 3.5 kV with a final ion beam cleaning for 30 minutes at 0.5 kV. [2] Real time observation of laser ablation was performed with a Dynamic Transmission Electron Microscope (DTEM) equipped with a pump-probe laser system. The DTEM is installed at Lawrence Livermore National Laboratory (LLNL) [3]. The experiments were carried out with a 532 nm wavelength pump laser and a 12 ns FWHM pulse duration. The laser beam had a Gaussian beam profile with a diameter of 135 � 5 �m, and an energy of 12 �J per pulse. Nine laser pulses were used to acquire a total of 9 consecutive micrographs of the thinnest, i.e., most electron transparent, area of the sample to capture the evolution of the thin nickel film. Prior to the DTEM experiments the geometry of each TEM sample was characterized by valence electron energy-loss spectroscopy using a JEOL-JEM 2100F/Cs operated at 200 keV. The temperature distributions, heat transfer and resulting mechanical stress due to laser irradiation was modeled for the sample geometry using spatio-temporal simulations with the COMSOL Multiphysics software. Figure 1 shows a series of bright field TEM images recorded by DTEM, and reveal the transient process of laser ablation caused by a laser power density below 108 W/cm2 [4]. Figure 1A shows that the asdeposited nickel thin film was continuous prior to laser irradiation. Figure 1B displays the surface reorganization that was observed 20 ns after the laser pulse. The image intensities representing a cellular structure of the nickel film indicate the onset of liquid dewetting of the nickel film in this area [5]. Figure 1C demonstrates the formation of round nanoparticles and substrate fracture 115 ns after laser irradiation. Figure 1D, taken minutes after laser irradiation, shows the substrate fractured in the thin region where nickel film originally dewetted (Figure 1B). It has been shown that laser ablation processes with low power density are accompanied by particle formation where round particles are liquid droplets ejected through hydrodynamic processes and irregularly-shaped particles with smaller sizes are formed via thermal-induced stresses and fracture [6]. Since direct temperature measurements are not possible during DTEM experiments, the temperature distribution in the sample was modeled as a function of time (see Figure 2). Figure 2 shows a series of time-temperature profiles where the Gaussian laser pulse is centered at the edge of the sample; the temperature profiles are given for different distances from the Microsc. Microanal. 21 (Suppl 3), 2015 1592 edge of the sample. The simulated temperature profiles indicate that, for the laser energy used in these experiments, the temperature exceeds the melting temperature of nickel thin film, which is ~1680 K. The dynamics of the low power density laser ablation was observed with high temporal and spatial resolution through capturing the early stages of liquid dewetting of the nickel film followed by nanosized particle formation and substrate fracture, which will be interpreted with thermal-induced stress simulation via COMSOL. References [1] K. Sugioka, et al."Laser Precision Microfabrication", (Springer, Berlin Heidelberg), 2010, p. 91-120. [2] A. Strecker, et al. Metallkd 94, (2003), p. 290-297. [3] T. LaGrange, et al. MRS Bulletin 40, (2015), p. 22-28. [4] S. Hihath, Melissa Santala, Geoffrey Campbell, Klaus van Benthem (submitted for publication). [5] J. Trice, et.al. Physical Review B 75, (2007), p. 1-15. [6] R. L. Webb, et al. Applied Spectroscopy 51, (1997), p. 707-717. [7] This work was supported by a University of California Laboratory Fee grant (#12-LR-238313). DTEM experiments at Lawrence Livermore National Laboratory were carried out under the auspices of the US Department of Energy, Office of Basic Energy Sciences under contract DE-AC52-07NA27344 A 2 �m B C D Figure 1. Bright field images from 12 ns electron pulses demonstrating the dynamics of laser ablation. A) as-deposited Ni film on a silicon substrate [4]. B) morphological instabilities developed 20 ns after laser heating showing the nickel film dewetting to the right of the dotted line [4]. C) nanoscale particle formation and substrate fracture captured at 115 ns [4]. D) the same region minutes afterwards [4]. Figure 2. Spatio-temporal COMSOL simulation of the temperature evolution as a function of time for a laser pulse centered at the edge of the wedge-shaped sample [4]. The distances are relative to sample edge.");sQ1[795]=new Array("../7337/1593.pdf","New Approach to Analysis of Noisy EELS Data","","1593 doi:10.1017/S1431927615008740 Paper No. 0795 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 New Approach to Analysis of Noisy EELS Data Emre Firlar1, Tanya Prozorov1 1 US DOE Ames Laboratory, Division of Materials Sciences and Engineering, Ames, IA USA Electron Energy Loss Spectroscopy (EELS) is used to probe the chemical environment of materials at the nanometer scale. This analysis becomes challenging when dealing with embedded nanostructured materials, where small regions of interest residing in thick matrices yield noisy data. The data analysis can benefit from modeling the core loss spectra with the available reference peaks, using the experimental peak positions and ratio of the fitted peak areas to gauge the chemical binding in a material. We applied this method for studying the early stages of magnetite biomineralization in magnetotactic bacteria. These microorganisms produce chains of magnetic nanocrystals via the uptake of soluble iron and iron oxide biomineralization in the intracellular membrane vesicles, called magnetosomes [1]. Even though this process has been studied for years, early stages of the magnetite magnetosome biomineralization pathway remains not fully understood [2]. Early stages of magnetosome magnetite biomineralization were investigated. To induce this process, iron-replete growth medium was added to the bacterial culture grown under the low iron conditions. We utilized Tecnai G2 F20 Scanning Transmission Electron Microscope (STEM) equipped with Gatan Image Filter (GIF) to monitor the induced bacterial magnetosome formation in samples harvested at 30 min intervals. The data acquired from smallest ( 10 nm) nanocrystals was compared to those acquired on larger (10 nm) magnetosomes. A representative High Angle Annular Dark Field (HAADF)-STEM image from the bacterium with magnetosomes of the sizes ranging from 6 to 51 nm from a culture harvested 30 minutes after the induction, is shown in Figure 1A. From the electron diffraction analysis, the mature magnetosomes (~ 45 nm) were found to be fully crystalline magnetite (Figure 1B), while the electron diffraction pattern of the particles with sizes smaller than 10 nm (Figure 1C, D) is indicative of the presence of both crystalline magnetite and amorphous ferric oxyhydroxide (i.e. ferrihydrite (xFe2O3.yH2O)) structures. After the initial characterization with the electron diffraction, core loss EELS was utilized to probe the oxidation state of iron in the biomineralized nanocrystals with varying sizes. Initial peak fitting of the EELS data was performed on a large magnetite magnetosome in the bacteria. Next, this approach was utilized to analyse smaller magnetosomes. Iron oxide particles have spectral signature around 709 eV corresponding to the Fe L3 edge. The EEL spectra acquired were modelled with L3 peaks of Fe2+ (707.8 eV), Fe3+ (709.5 eV) [3] and a satellite peak (~713.0 eV) [4]. From the EEL spectrum in Figure 2A, the roughly 2:1 ratio of the peak areas corresponding to Fe3+ and Fe2+, matches that expected for magnetite. Here the peak fitting was a straightforward process. On the other hand, the raw data acquired from a juvenile, poorly crystalline magnetosomes, were noisy, due to the small size of the particles residing in a thick bacterial cell, as shown in Figure 2B. To eliminate the ambiguity of the noisy spectrum, the data processed using the 5 and 20-point adjacent averaging smoothing methods were compared with the raw data. We have determined that excessive smoothing removes the small fingerprints, thus the averaging method should be evaluated in terms of both noise and the actual signal removal before the peak fitting is performed. Figure 2C and D show the peak fitting applied to the raw and processed data for the small Microsc. Microanal. 21 (Suppl 3), 2015 1594 magnetosome particle, respectively, using the previously determined peak positions. Based on the absence of 707.8 eV peak attributable to the Fe2+, the Fe2+ concentration in small magnetosomes is below the detection limit. Furthermore, the peak position of the L3 was found to be shifted toward the higher energy in the smaller particles (709.5 eV), consistent with the presence of only of Fe3+ in the smaller particles. EELS peak fitting and the electron diffraction data can be interpreted in terms of conversion of the as-formed ferric oxyhydroxide to magnetite occurring during the early stages of magnetosome formation and growth [5]. Further reduction of the ferric to ferrous iron during the crystallization and growth of these magnetosomes is also discussed. Current characterization methods include cryo-HAADF-STEM and in situ HAADF-STEM analysis. This conventional TEM characterization in conjunction with this new approach to EELS fitting analysis proved to be very informative with regard to the structural analysis and interpretation of chemical environment of the bacterial magnetosome nanocrystals. We expect this method to be applicable for the data analysis of poorly crystalline nanomaterias located in thick matrices. References: [1] Woehl, Taylor J. et al, Scientific Reports 4: 6854, (2014) [2] Bazylinski, Dennis A, Internatl Microbiol 2 (1999), 71-80 [3] Van Aken, P.A. et al, Phys Chem Minerals 25 (1998), 323-327 [4] De Groot, Frank M. F. et al, J. Phys. Chem. B 109 (2005), 20751-20762 [5] Baumgartner, Jens et al, Nature Materials 12 (2013), 310-314 A B C D Figure 1. A) HAADF STEM image of 30 minute - time lapse sample. Inset shows (ref: white rectangle in low mag image) four different magnetosomes with sizes ranging from 6 nm to 51 nm. Electron diffraction patterns and corresponding bright field TEM images of magnetosomes with sizes of B) 45 nm and C) 8 nm and D) 6 nm (2 particles). A B Fe3+ C D arbitrary units arbitrary units arbitrary units Fe3+ arbitrary units Fe3+ Fe2+ Raw data 5 points AA smoothing 20 points AA smoothing 706 708 710 eV 712 714 706 708 710 eV 712 714 706 708 710 eV 712 714 706 708 710 eV 712 714 Figure 2. EELS Fe L3 core loss data from A) a big (size ~ 48 nm) and B) a small (size ~ 10 nm) magnetosome. A, C, D) Experimental data, fitting (Fe2+: 707.8 eV, Fe3+: 709.5 eV and a satellite peak around 713.0 eV) peaks and their sum are given by dots, dash lines and solid line, respectively. B) Raw data (solid line with black square) and smoothed data (5 points: solid line with grey sphere, 20 points: solid line with empty square) for the small magnetosome are compared.");sQ1[796]=new Array("../7337/1595.pdf","Use of a hybrid silicon pixel (Medipix) detector as a STEM detector","","1595 doi:10.1017/S1431927615008752 Paper No. 0796 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Use of a hybrid silicon pixel (Medipix) detector as a STEM detector Damien McGrouther1, Matus Krajnak1, Ian MacLaren1, Dzmitry Maneuski1, Val O'Shea1, Peter D. Nellist2 1. 2. SUPA School of Physics and Astronomy, University of Glasgow, Glasgow G12 8QQ, UK Department of Materials, University of Oxford, 16 Parks Road, Oxford OX1 3PH, UK Scanning transmission electron microscopy has traditionally relied on the high angle annular dark field technique for imaging atoms [1,2], which provides a simple and easy to understand contrast, which is strongly related to atomic number. More recently, there has been a resurgence of interest in alternative imaging modes, including bright field [3,4] and annular bright field imaging [5,6]. Ultimately, however, the most flexible STEM experiment would be to record the entire back focal plane of the specimen onto a pixelated detector, and then post-process the dataset to access whichever features in the contrast are desired. This could produce all the above-mentioned signals, but many more besides, including ptychographic reconstruction of the exit wave [7,8]. In order to achieve this aim, we need to have a detector that is capable of recording single electron events at typical beam energies for a scanning TEM (e.g. 100 or 200 keV) with enough pixels to allow significant flexibility for performing different imaging modes, a readout speed far quicker than that available of previous pixelated detectors such as traditional scintillator-CCD devices (typically < 30 frames per second), and synchronization to the scan system. We have installed a Medipix-3 detector onto our JEOL ARM200F STEM (probe-corrected, cold FEG version), which is a silicon based hybrid pixel detector with a CMOS readout architecture and 256 x 256 pixels. The Medipix family of detectors are true counting detectors. Each of the 55�mx55�m pixels contains amplifiers and digitisation circuitry that determines whether energy deposited by the incident radiation lies within the range set by userdefined low and high thresholds, counting digitally those that do. Medipix2 detectors have been investigated for TEM by us [9,10] and others [11]. Presently we are exploring the capabilities of the third generation Medipix-3 detector coupled to a high-speed read-out system. The MERLIN read-out system [12] is capable of frame rates of 1200fps for durations that are limited only by the ability to transfer frame data to non-volatile storage. Constant frame rates of 100fps are guaranteed but we have found no problems running continuously at 500fps. Figure 1 shows that for 200keV electrons, single electron detection is achieved but the detector settings have a strong influence on the spatial frequency response of recorded images. Utilising an exposure time of 1�s and values of 70 and 240 for the pixel low energy threshold, it can be seen that the effects of charge-sharing between pixels can be drastically reduced, but at the expense of reducing the counting efficiency from 100% to ~50%. We are currently measuring the performance of the detector with respect to Detective Quantum Efficiency (DQE) and Modulation Transfer Function (MTF). An exciting aspect of the Medipix3 is its novel architecture, which allows the detection of charge-sharing and re-attribution back to a single pixel. For high-energy electrons we expect this to significantly improve MTF values. For STEM mode detection, we have achieved synchronisation of the beam scanning and detector acquisition. Figure 2 shows a sequence of images of the bright field disc resulting from scanning across a cross-grating replica and post-processing to produce bright field (sum) and phase contrast images. This dataset was recorded with beam dwell time of 2ms and scan resolution of 256x256 pixels. The detector Microsc. Microanal. 21 (Suppl 3), 2015 1596 operated at 500fps (total scan acquisition requiring ~130seconds) and yielded a 4-dimensional datavolume of size ~8Gb. References: [1] AV Crewe, J Wall and LM Welter: J. Appl. Phys. 39 (1968) p. 5861. [2] P Hartel, H Rose and C Dinges: Ultramicroscopy 63 (1996) pp. 93�114. [3] JM LeBeau et al., Phys. Rev. B, 80 (2009) 174106. [4] I MacLaren, et al., APL Mater. 1 (2013) 021102. [5] M Hammel and H Rose: Ultramicroscopy 58 (1995) pp. 403�415. [6] S D Findlay, et al., Ultramicroscopy 110 (2010) pp. 903�923. [7] TJ Pennycook et al., Ultramicroscopy (2015) in press (doi:10.1016/j.ultramic.2014.09.013). [8] H Yang et al., Ultramicroscopy (2015) in press (doi:10.1016/j.ultramic.2014.10.013) [9] A MacRaighne, et al., J. Instr. 6 (2011) C01047 [10] R Beacham, et al., J. Instr. 6 (2012) C12052 [11] G McMullan et al.,Ultramicroscopy 107 (2007) 401 [12] R Plackett et al., J. Instr. 8 (2013) C01038 [13] The authors gratefully acknowledge funding from the EPSRC under grant numbers EP/M009963/1 and EP/M010708/1. This work was supported by the EU FP7 Grant Agreement 312483 (ESTEEM2). Figure 1. (a) & (b) 1ms exposures showing single 200keV electron events and charge-sharing among pixels captured with pixel low threshold values = (a) 70 and (b) 240. (c) Bright field (Low Mag mode) TEM image of a cross-grating replica (2180 lines per mm). Figure 2. (a) & (b) images of the bright field STEM disc. Post-processing of the data-volume produced the STEM bright field image (c) and differential phase contrast images (d) & (e).");sQ1[797]=new Array("../7337/1597.pdf","Development of an Aberration Corrected 1.2-MV Field Emission Transmission Electron Microscope","","1597 doi:10.1017/S1431927615008764 Paper No. 0797 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Development of an Aberration Corrected 1.2-MV Field Emission Transmission Electron Microscope Hiroyuki Shinada1, Toshiaki Tanigaki1, Tetsuya Akashi1, Yoshio Takahashi1, Tadao Furutsu1, Tomokazu Shimakura1, Takeshi Kawasaki1, Keigo Kasuya1, Heiko M�ller2, Maximilian Haider2, Nobuyuki Osakabe1, and Akira Tonomura1, 3 1 2 Central Research Laboratory, Hitachi, Ltd., Hatoyama 350-0395, Japan Corrected Electron Optical Systems GmbH, Englerstr. 28, D-69126, Heidelberg, Germany 3 RIKEN Center for Emergent Matter Science (CEMS), Wako 351-0198, Japan A practical holography electron microscope was first developed in 1978 by the late Dr. Tonomura[1]. After that, we (Tonomura's group) developed bright and monochromatic field-emission electron beams over 35 years for observing quantum phenomena by utilizing the wave nature of electrons. As it turns out, every time we developed a brighter electron beam, electron interference experiments became easier to perform, and the precision in the phase measurements increased, thereby opening up new application fields. Atomic-resolution electromagnetic field analysis of non-periodic local structures such as interfaces is important in developing heterostructures and bulk materials with boundaries because their properties derive from local electromagnetic characteristics. For example, the development of rare-earth permanent magnets with high coercive force at high temperatures is required for the efficient motors used in hybrid or electric vehicles. Improvements in their coercive force are expected by controlling their grain boundary structures and their magnetic properties [2]. Thus, electron microscopes need to be developed for observing electromagnetic fields at atomic resolution. For this purpose, a 1.2-MV cold field-emission transmission electron microscope (TEM) equipped with a spherical-aberration corrector [3] has been developed (Fig. 1). The microscope has the following superior properties: stabilized accelerating voltage (stability: 0.3 ppm peak to peak), minimized electrical and mechanical fluctuation, and a field emission gun with high stability and brightness [4]. Information transfer of 43 pm was accomplished by using W{633} chromatic lattice fringes. When this developed FE-TEM was applied to observations of GaN [411] thin samples, the projected Ga atom positions were visualized with 44 pm separation--the smallest separation ever observed (Fig. 2). This resolution is an important base performance for effective electron holography observations. The microscope enables performing electromagnetic field observations at high-resolution that were not possible with 300-kV TEMs in various types of research. Microsc. Microanal. 21 (Suppl 3), 2015 1598 References: [1] A. Tonomura et al., Jpn. J. Appl. Phys. 18, 9 (1979). [2] H. Sepehri-Amin et al., Acta Mater. 61, 6622 (2013). [3] M. Haider et al., Nature 392, 768 (1998). [4] K. Kasuya et al., J. Vac. Sci. Technol. B 32, 031802 (2014). [5] This research was supported by a grant from the Japan Society for the Promotion of Science (JSPS) through the Funding Program for World-Leading Innovative R&D on Science and Technology (FIRST Program)," initiated by the Council for Science and Technology Policy (CSTP). Figure 1. Aberration corrected 1.2-MV cold field-emission transmission electron microscope. Figure 2. (a) High-resolution TEM image of GaN [411] thin sample. Projected Ga atom positions (white arrows) with 44-pm separation were clearly observed. Inset shows corresponding simulated image. (b) Corresponding Gaussian low-pass filtered image. (c) Line profiles of Ga atom pairs indicated by black rectangles 1 and 2 in (b).");sQ1[798]=new Array("../7337/1599.pdf","Aberration Corrected Electron Microscopy Enhanced for Lower Accelerating Voltages","","1599 doi:10.1017/S1431927615008776 Paper No. 0798 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Aberration Corrected Electron Microscopy Enhanced for Lower Accelerating Voltages Takaki Ishikawa1, Eiji Okunishi1, Toshikatsu Kaneyama1, Yukihito Kondo1, Syo Matsumura2 1. 2. JEOL Ltd., 3-1-2 Musashino Akishima Tokyo 196-8558 Japan. Research laboratory of high-voltage electron microscope, Kyushu University, Motooka 744 Nishi-ku Fukuoka 819-0395 Japan. Radiation damage by incident electron beam becomes a serious problem in chemical analysis by EDS and/or EELS, since the dose for these analysis requires more than that for imaging. Low acceleration voltage (here in after we call "low kV" in this paper) is effective to solve the problem, since it reduces the specimen damage of materials composed of light elements such as carbon materials. The dedicated low kV microscopes for the purpose have been developed on the special projects [1,2]. However, we can make an adequate tuning for a low kV such as 80, 60 and 30 kV on the aberration corrected 200 or 300kV microscopes (ex. JEOL JEM-ARM200F). In this paper, we explore the analytical capabilities of JEM-ARM200F for the performances in low kV. In low kV, an obstacle to make atomic resolution image or analysis is an image blur (), determined with a chromatic aberration coefficient (Cc), an energy spread of an electron source (dE) and primary electron energy (Eo), as = Cc*dE/Eo. As Eo decreases in low kV, the blur () becomes larger. Therefore, small Cc*dE is essential to make a small probe. The Cc also decreases as Eo decreases, since the saturation of the magnetic objective lens released as the required excitation of the lens reduces. With an electron source from cold field emission gun (CFEG), dE is 0.3-0.4 eV. Figure 1(a) and 1(b) plot the probe diameters (in D59) at 60 kV and 200 kV with thermal field emission gun (TFEG) and CFEG, and with A5 aberration correction or not, depending on the convergence angle. The D59 is a probe diameter which includes 59% of the total probe current. The total chromatic aberration coefficients (Cc) for this calculation were 2.36 mm for 200 kV and 1.41mm for 60 kV, which are calculated sum of Cc for the objective lens with HRP (analytical poles allowing highly sensitive dual EDS detectors system) and the aberration correction system optics. Since the specimen damage decreases in low kV, we can increase the allowable current of analytical probe, resulting in better analytical sensitivity. In addition to this, the ionization cross section increases in low kV. It also enhances the sensitivity. To obtain more probe-current keeping the probe size small, convergence angle must increase. Namely, it is essential to reduce the higher order geometrical aberration such as sixth fold astigmatism (A5). As shown in Fig. 1, without the correction of A5, the allowable aperture size starts to increase rapidly at 40 mrad. Therefore, it is effective to correct A5 when we use the CFEG. Figures 2(a) and 2(b) show Ronchigrams obtained with JEM-ARM200F equipped with CFEG and the higher order geometrical aberration (DCOR CEOS Gmbh). The flat phase regions, which are generally 25-30 mrad with no A5 correction, clearly increase to be 55-60 mrad in these Ronchigrams. Thus, the probe current can be increased. On the other hand, with highly sensitive detectors are also sought after for the analysis of beam sensitive materials. The dual SDD system, developed recently [3], dramatically raises the detection solid angle of fluorescence X-ray from a sample to be > 1.7 sr. Figure 3 shows the obtained elemental maps from SrTiO3 with JEM-ARM200F equipped with the dual SDD system, HR type objective lens and CFEG at 200 kV. The acquisition conditions for these maps are probe size = 90 pm, Microsc. Microanal. 21 (Suppl 3), 2015 1600 probe current = 24 pA, number of pixels = 128 x128, dwell time for each pixel = 10sec and number of integrated frames = 1000. Each element is clearly separated on these maps with rather short time and rather small probe current, which allows us to observe an atomic level high resolution STEM image. Thus, with employment of dual SDD system, low kV and higher order aberration correction system, the analytical sensitivity is greatly improved and it greatly lowers the threshold of the analysis of beam sensitive materials. References: [1] Suenaga project, "Low-voltage TEM/STEM for atomic level characterization of soft matters", CREST (2006-2012) Japan Science & Technology Agency. [2] Salve project, Sub-Angstrom Low-Voltage Electron Microscopy (SALVE) I-II project (2008~). [3] S Kawai et al, Microscopy and Microanalysis 20 (S3) (2014) p.1150. 800 (a) 60kV HRP 700 600 dE = 0.8eV no A5 correcion dE = 0.8eV with A5 correction dE = 0.4eV no A5 correction dE = 0.4eV with A5 correction Probe size D59 (pm) 500 400 300 200 100 0 10 20 30 40 50 60 70 Convergence semi angle (mrad) Figure 1 Probe diameter depending on convergence semi angle with CFEG, TFEG, A5 correction. (a) at 60 kV, (b) at 200 kV Figure 2 Observed Ronchigram from amorphous carbon film. The microscope equipped with higher order aberration corrector and CFEG. (a) at 60 kV, (b) at 200 kV Figure 3 Elemental maps with dual SDD system from SrTiO3. (a) Sr, (b) Ti, (c) O, (d) Sr+O, (e) Sr+Ti, (f) HAADF STEM image");sQ1[799]=new Array("../7337/1601.pdf","Ettention: Building Blocks for Iterative Reconstruction Algorithms","","1601 doi:10.1017/S1431927615008788 Paper No. 0799 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Ettention: Building Blocks for Iterative Reconstruction Algorithms Tim Dahmen1,3, Lukas Marsalek2,3*, Nico Marniok3,*, Beata Turoov�3,4, Sviatoslav Bogachev3, Patrick Trampert1, Stefan Nickels1 and Philipp Slusallek1,3 German Research Center for Artificial Intelligence, 66123 Saarbr�cken, Germany Eyen SE, Na Niv�ch 1043/16, 141 00 Praha 4, Czech Republic 3 Saarland University, 66123 Saarbr�cken, Germany 4 IMPRS-CS Max-Planck Institute for Informatics, 66123 Saarbr�cken, Germany 2 1 We present a novel software package for tomographic reconstruction in electron microscopy, named Ettention [1]. The software consists of a set of modular building-blocks for iterative reconstruction algorithms. Ettention simultaneously features (1) a modular, object-oriented software design, (2) optimized access to high-performance computing (HPC) platforms such as graphic processing units (GPU) or many-core architectures like Xeon Phi, and (3) accessibility to microscopy end-users via integration in the IMOD package and user interface. We provide developers with a clean application programming interface (API) that allows for extending the software easily and thus makes it an ideal platform for algorithmic research while hiding most of the technical details of high-performance computing. Several case studies are provided to demonstrate the feasibility of the concept [2]. Ettention is built on top of the OpenCL API. While this allows the software to run on many hardware architectures, the OpenCL programming model still exposes a number of technical properties, such as the need to very explicitly handle parallelism and memory management. Therefore, Ettention introduces the concepts of "building blocks" for iterative reconstruction algorithms. Those building blocks include forward projections, back projections, image manipulation, and input/output operations. They encapsulate memory management and parallelism and can be combined to reconstruction algorithms. As Ettention allows the reconstruction volume resolutions to exceed memory on the HPC device, most operations involve consecutive upload of parts of the volume to the GPU, the execution of a kernel that operates on parts of the volume, and the combination of partial results. The system maps memory objects, such as volumes or reconstruction images to configurable HPC representations, such as float buffer or a GPU texture. By tracking the up-to-date status of the different representations the system can detect at runtime when to transfer data to device memory or back. Compared to existing software packages for tomographic reconstruction in electron microscopy, Ettention fills a gap between IMOD and the ASTRA toolbox. For algorithmic research without the need for immediate application to high-resolution data, we recommend using the ASTRA toolbox because of the powerful MATLAB language binding. The limitation to the in-core case, i.e. to small reconstruction volumes, is less important in this case. For microscopy experimentalists, who require well-established reconstruction algorithms, we recommend using IMOD because of its highly-optimized reconstruction performance on high-resolution data. For projects that require both, algorithmic innovation and immediate application to high-resolution experimental data, we believe that the Ettention software package should be considered because it delivers a reasonable (though not optimal) performance even on high-resolution data and exposes a rich and well-structured API that allows for fast and efficient implementation of algorithmic innovations. Microsc. Microanal. 21 (Suppl 3), 2015 1602 References: [1] T Dahmen et al "The Ettention Software Package" under review Ultramicroscopy (2015). [2] T Dahmen et al Microsc. Microanal.(2014) vol. 20, pp. 548�60 Figure 1: A block diagram of the SART algorithm as an example for an iterative reconstruction algorithm implemented using the building blocks, also called "operators" in Ettention. CPU code objects including dataflow are shown on the left side, GPU kernels on the right side. Some operators like the residual computation corresponds 1:1 with calls to GPU kernels, while the forward and back operators, which work on volumes, require more than one call to the corresponding kernel.");sQ1[800]=new Array("../7337/1603.pdf","Analysis of TEM tomography artifacts with experiments on model specimens","","1603 doi:10.1017/S143192761500879X Paper No. 0800 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Analysis of TEM tomography artifacts with experiments on model specimens Jonathan Winterstein1, Joshua Schumacher1 and J. Alexander Liddle1 1. NIST Center for Nanoscale Science and Technology, Gaithersburg, MD 20899 TEM tomography is becoming a critical technique for the characterization of materials. Despite the rapid development of software and imaging methods for obtaining tomograms from a wide range of materials [1], important questions remain about the resolution in three dimensions and the uncertainties in quantitative measurements. Recently, Mezerji et al. reported a useful method for measuring the anisotropic resolution in the reconstructed volume using curve fitting to edge profiles, and derived values different from those predicted by the well-known Crowther criterion [2]. Perhaps more importantly, it is necessary to quantify the accuracy of dimensional measurements (e.g., particle volumes) which are in turn determined by the accuracy of segmentation or edge-finding methods and not directly related to resolution (the ability to distinguish two closely spaced objects). We will report measurements of 3D feature sizes and resolution in reconstructed volumes using several different reconstruction algorithms from different model samples: MgO smoke cubes, Au nanoparticles and FIB-prepared pillars. In the case of MgO smoke cubes and the FIB pillars it is possible to know the 3D sample shape a priori and therefore quantitatively to compare the accuracy of reconstruction and analysis methods. Unlike reconstructions with digital phantoms, these experiments also permit analysis of the effects of noise, beam damage and tilt-series alignment on the 3D reconstruction. Figure 1 shows the results of a reconstruction of a tilt series of ADF-STEM images of an MgO cube using a model-based reconstruction method [3]. The cubic shape is preserved in three dimensions and the reconstruction shows good fidelity. Some artifacts appear, possibly due to the effect of diffraction contrast in the STEM images or the missing wedge. A SIRT reconstruction (not shown) of the same dataset suffers from significantly worse artifacts due to the missing wedge. This particle presents a particularly difficult challenge for tomographic reconstruction due to the sharp, flat facets oriented perpendicular to the optical axis, and is thus an excellent test case. Volume measurements of Au nanoparticles also show significant differences between different reconstruction methods and thresholding procedures for individual particles and groups of particles. Measurement of the anisotropic resolution is straightforward with Au nanoparticles using the curve fitting method of [2] and is compared between different reconstructions in figure 2. Using samples with a priori known dimensions allows us to compare the fidelity of different segmentation methods as well. Comparisons of standard segmentation procedures will be presented and the possibility of fully automated, objective segmentation and quantification discussed. We will also discuss possible superior alternative model samples for characterization of TEM tomography reconstruction and analysis. References: [1] PA Midgley, M Weyland, Ultramicroscopy 96 (2003), p. 413. [2] H Heidari Mezerji, W Van den Broek, S Bals, Ultramicroscopy 111 (2011), p. 330. Microsc. Microanal. 21 (Suppl 3), 2015 1604 [3] SV Venkatakrishnan, LF Drummy, MA Jackson, M De Graef, J Simmons, CA Bouman, IEEE Transactions on Image Processing 22 (2013), p. 4532. (a) (b) (c) Figure 1. Tomographic reconstruction of a MgO smoke cube using the MBIR algorithm from [3]: a single slice in the X-Y plane (a), and in the X-Z plane (b) where the horizontal line in (a) indicates the position of the slice shown in (b), and a 3D perspective view of the particle after segmentation (c). (a) (b) Figure 2: Results of curve fitting to edge intensity profiles along the z direction (optical axis) from tomographic reconstructions of the same Au particle using two different reconstruction methods. 141 ADFSTEM images acquired at 1� tilt intervals from -70� to + 70� were used for the reconstruction.");sQ1[801]=new Array("../7337/1605.pdf","Rapid 3D Reconstruction in the EDS Tomography by using Iterative Series Reduction (ISER) Method","","1605 doi:10.1017/S1431927615008806 Paper No. 0801 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Rapid 3D Reconstruction in the EDS Tomography by using Iterative Series Reduction (ISER) Method. Yoshitaka Aoyama1, Hideo Nishioka1 and Yukihito Kondo1 1. JEOL Ltd., 1-2 Musashino 3-Chome Akishima Tokyo 196-8558 Japan Energy Dispersive X-ray Spectroscopy (EDS) is widely used to obtain the elemental maps of a sample. EDS Tomography is a method to reconstruct three-dimensional (3D) elemental maps from a set of twodimensional (2D) EDS elemental maps. The maps can be obtained by a transmission electron microscope (TEM) equipped with EDS detector. The high accelerating voltages of TEM is necessary to obtain the transparent image and elemental maps. On the other hand, many 2D elemental maps, forming a tilt series map, are necessary to reconstruct 3D elemental maps. Therefore, the electron beam damage is serious in EDS Tomography, and a method to reduce the number of elemental maps is sought-after. It is well known that there are two types of algorism to reconstruct the 3D structure. One is the analytical method like Filtered Back Projection method (FBP), and another is the algebraic method like Simultaneous Iterative Reconstruction Technique (SIRT). Iterative Series Reduction (ISER) is a new algebraic method for 3D reconstruction based on the sparse modeling technique, which is rapidly developing in the field of the image processing [1] [2]. By using ISER, 3D reconstruction can be obtained from only several projections. In this method, only the outlines of the objects in the projections are used in the 3D reconstruction to make the sparse model. The other regions are ignored in the 3D reconstruction. This reconstruction is achievable with small number of the variables. Therefore, by using ISER, small number of the sparse modeled projections is enough to reconstruct the 3D structure, if one only wants to know the outline of the objects. It is useful to determine the shape of the sample, even if the gradient of the density in the specimen cannot be reconstructed. In this study, we tried to combine EDS Tomography and ISER to shorten the acquisition time. This avoids the beam damage and contamination of the sample. Figures 1 show the BF-STEM image of a paint film and elemental maps of Titanium, Aluminum, Iron and Silicon. A tilt series of EDS elemental maps for the region of sample shown in Fig. 2 was collected with the tilting range of �56 degree. The interval of the tilt steps was 8 degree. The number of pixels for these maps was 256 x 256. The number of the elemental maps was only 15. Figure 2 shows the 3D elemental maps of the paint film reconstructed by ISER. Figure 2(a-c) shows the orthogonally-cut digital slices from the 3D reconstructed map of Titanium. The spurious streaks in the recorded tomograms (not shown here) due to the missing wedge of rotation were disappeared in digital slices shown in Fig. 2(a-c). Figure 2(d) shows the volumerendering image of the paint film. The brown, green, pink and yellow phases correspond to the Titanium, Silicon, Aluminum and Iron. In conclusion, using ISER for EDS Tomography reconstruction, we can reduce number of elemental maps to be 15, which is generally 50 by the conventional method per a tilt series for 3D reconstruction. This greatly reduces the beam irradiation damage of specimens and opens the possibility for 3D reconstruction of beam sensitive materials. References: [1] D L Donoho, IEEE Trans. Inf. Theory 52 (2006) p. 1289-1306. [2] E J Candes, J Romberg and T Tao, IEEE Trans. Inf. Theory 52 (2006) p. 489-509. Microsc. Microanal. 21 (Suppl 3), 2015 1606 Figure 1. BF-STEM image and elemental maps of the paint film. (a) BF-STEM image, (b) Titanium map, (c) Silicon map, (d) Aluminum map and (e) Iron map. Figure 2. (a) The 3D reconstruction of the Titanium from the paint film was obtained by ISER. The interval of tilt steps was 8 degree. The tilting angular range was �56 degree. (b) The volume-rendering image of the paint film. The brown, green, pink and yellow phases correspond to Titanium, Silicon, Aluminum and Iron.");sQ1[802]=new Array("../7337/1607.pdf","Moving atomic-resolution imaging into the age of deep data","","1607 doi:10.1017/S1431927615008818 Paper No. 0802 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Moving atomic-resolution imaging into the age of deep data Oleg S. Ovchinnikov1, S. Jesse2, A. Belianinov2, S. T. Pantelides1,2, S.V. Kalinin2, A.Y. Borisevich2 1 2 Department of Physics and Astronomy, Vanderbilt University, Nashville TN 37235 USA Oak Ridge National Laboratory, Oak Ridge, TN 37831 USA The current ability of electron microscopy to collect frames in ultra-quick succession allows for the examination of the material structure in a dynamic manner offering unique insights into what determines a material's properties. The ability of computer simulations based on density functional theory (DFT) to predict materials behavior has greatly increased as access to super computers has increased, making running simulations with many different starting conditions possible in a short amount of time. The combination of these two advances in fundamental science has been used on a case by case basis, which does not fully exploit the potential advantages. An approach that combines these advantages in a more general manner, allowing each to build off of the other, is an important step into the future. This can be accomplished through the use of a deep data approach that allows simulations to quickly search and analyze data for starting conditions to simulate theory and for experiments to quickly search completed simulations to pull out the underlying physics. A major bottleneck in this approach is the need to analyze and interpret multiple frames. The process of finding all of the atoms in a frame and running analysis on the frame is a lengthy process when done manually on a single frame, let alone a movie, in which the frames must also be co-registered together. Many current analysis methods are not scalable with the ability to acquire data. This causes a large amount of collected data to go unanalyzed and unused when it might contain important physical information. In order for a deep data approach to be fully utilized, all data must be analyzed and stored in an easily searchable manner. In this presentation, I will describe a method for atom-finding analysis of scanning transmission electron microscope (STEM) images that will allow a user to quickly analyze a single frame and then use this frame in order to analyze multi-frame movies. This will be accomplished through the use of a repository of filtering, atom finding, and data analyses (Figure 1). This ability, combined with central tracking of all operations performed on an image, will allow for both batch processing and databasing. Another advantage of such a program is to have a centralized ability to analyze atom positions and shapes using methods that have been shown to pull information out of images that would be impossible to detect with the human eye. Such methods include looking at the spatial distribution of principal component analysis (PCA) eigenvalues of the atom shape and spatial distribution of nearest neighbors (Figure 2).[1,2] Once analyzed in a uniform manner, all data can then be centrally stored for searching. Such searches could include comparing to simulations, finding similar physics across multiple materials, or any other number of reasons. Information analyses like these have great benefits for simulations such as DFT. An example of this would be using the information from the PCA of the nearest neighbors of many images of a material to narrow down the range of the initial conditions for the simulations. In order to accomplish this, ground work has been laid in the form of a test program that will combine many methods of filtering, atom finding, and data analyses into a single graphical user interface that also allows tracking of users' actions, allowing for repeatability in data analysis. This allows for data analysis to be performed on large amounts of data in a fraction of the time. This in turn allows for the ability to build a catalog of data for use in deep data analysis [3]. Microsc. Microanal. 21 (Suppl 3), 2015 1608 References [1] A. Borisevich, O.Ovchinnikov et al., ACS Nano 4, 6071-6079 (2010) [2] A. Belianinov et al.,, these proceedings [3] The research is sponsored by the ORNL's Laboratory Directed R&D Fund (OSO, SJ, SVK, AYB) and the Division of Materials Sciences and Engineering, Office of Basic Energy Sciences, U.S. Department of Energy under the grant DE-FG02-09ER46554 (STP). FIG 1. Example of ability to filter image, identify atom center, and filter into sub-lattices (a) original image (b) filtered with blind deconvolution, background removal, and Gaussian filter of FFT space (c) centers found and separated into two sub lattices with manually seeded points. FIG 2. Example of analysis on data set (a) first 6 Eigen-vectors of atom shapes (b) spatial distribution of first Eigen-vector of atom shape showing large defect and sub lattices (c) spatial distribution of second Eigen-vector of atom shape showing large defect and sub lattices (d) first 6 Eigenvectors of 6 nearest neighbors with 0.4 scaling factor for length (e) spatial distribution of first Eigenvector of nearest neighbors showing large defect and sub lattices (f) spatial distribution of third Eigenvector of nearest neighbors showing large defect and sub lattices, second Eigen-vector skipped due to showing little information.");sQ1[803]=new Array("../7337/1609.pdf","High-accuracy electron tomography of semiconductor devices.","","1609 doi:10.1017/S143192761500882X Paper No. 0803 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 High-accuracy electron tomography of semiconductor devices. Misa Hayashida1, Lina Gunawan2, Marek Malac1,3 , Chris Pawlowicz2, Martin Couillard4 1 2 National Institute for Nanotechnology, Edmonton, Alberta, Canada. TechInsights, Kanata, Ontario, Canada 3 Department of Physics, University of Alberta, Edmonton, Alberta, Canada. 4 National Research Council, Ottawa, Ontario, Canada. As semiconductor industries have implemented three dimensional (3-D) structures in order to improve device performance, i.e., to decrease power requirement and to lower current leakage, a conventional 2D TEM cross-section is no longer sufficient to visualize the structure of a device. Electron Tomography (ET) provides more comprehensive representation in comparison to conventional 2-D cross section TEM images. Electron tomography allows for object visualization of practical interest at sub-nm resolution in 3D. For undistorted rendering and accurate measurement of object dimensions the data must be acquired over the entire �90� tilt range and the accuracy of the alignment of the projected images must be better than the desired resolution of the reconstructed volume [1]. Here we discuss high resolution ET investigation of semiconductor devices using nano-dot fiducial markers on samples without a missing wedge. Figure 1 shows a) conventional cross sectional image, and b) a projected image of a rod-shaped sample from a 22-nm node computer processor chip. Answering the engineering questions requires high resolution reconstruction of a volume ~ (150 nm)3. The large investigated volume and the nature of the sample implicates that atomic column tomography is not suitable approach, but measurements need to be performed at nearly atomic resolution, i.e. well below 1 nm. The sample in Fig. 1b is composed of (from top) the circuit layer and the support silicon wafer with fiducial markers. The nano-dot fiducial markers for accurate alignment were fabricated using an electron beam induced deposition of tungsten in an SEM [1] The images were acquired in a TEM mode of a Hitachi HF-3300 TEM operated at 300 keV. The tilt range was �90� with 3� step. Collection semiangle of ~100 mrad was used to reduce the effect of diffraction contrast [2]. The accuracy of the alignment using the nano-dot fiducial markers is less than 1.9 pixels RMS at 0.67 nm/pix. Filtered back projection was used to reconstruct the 3D volume. Figure 2-4 shows slices extracted from the reconstruction. Three PMOS tri-gate transistors with tungsten contact to the Si-fin can be seen in Figure 2. The corners of Si-Ge diamond were etched between adjacent cells, eliminating a short between neighboring transistors as shown in fig. 3. Figure 4 shows section along lines A, B and C indicated in Figure 2. In summary, the current generation of computer processor chips was imaged by electron tomography utilizing rod-shaped sample without a missing wedge. Using fabricated nano-dot fiducial markers allowed to clearly image features within sub-10 nm objects. As more and more devices move from planar to 3-D design structure, for instance flash memory devices [3], electron tomography will become essential to provide a comprehensive structural representation of these innovative breakthroughs [4]. [1] M Hayashida, M Malac, M Bergen, P Li, Ultramicroscopy 144, 50-57 [2] Huai-Ruo Zhang, Ray F Egerton, Marek Malac, Micron 43 (2012), 8 � 15. [3] http://www.samsung.com/global/business/semiconductor/product/vnand/overview [4] Support of Alberta Innovates Technology Futures, visiting researcher grant is gratefully acknowledged. Microsc. Microanal. 21 (Suppl 3), 2015 1610 1600 This work was done when the first author worked at her earlier affiliation, AIST. Figure 1 cross sectional TEM image of 22 nm node computer processor chip. Figure 1 b) projected image from a tilt series of 22 nm node processor chip. It shows the carbon layer with fiducial markers, device layer and the Si wafer. Figure 2 X-slice showing Si-Ge diamond shape under tungsten contact right on top of the fin. Figure 3 Z-slice showing three PMOS transistors with one tungsten contact to Si-fin. Figure 4 Y-slice images of the gate along the sections indicated in Figure 3.");sQ1[804]=new Array("../7337/1611.pdf","Progress in Crystallographic Image Processing for Scanning Probe Microscopy","","1611 doi:10.1017/S1431927615008831 Paper No. 0804 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Progress in Crystallographic Image Processing for Scanning Probe Microscopy Peter Moeck and Jack Straton Nano-Crystallography Group, Department of Physics, Portland State University, Portland, OR 97207-0751 Crystallographic Image Processing (CIP) originated with the electron crystallography community. Nobel Laureate Sir Aaron Klug (OM, FRS) and coworkers pioneered the technique for the analysis of long-range ordered biological materials in parallel illumination Transmission Electron Microscopes (TEMs). Corrections for the effects of the TEM's phase contrast transfer function and for less than optimal imaging conditions are part of this kind of CIP. There are also "electron microscope independent" 2D crystallography foundations to this kind of image processing. Based on these foundations, we applied CIP to images of long-range ordered 2D periodic surface arrays that were recorded with different kinds of scanning probe microscopes (SPMs) [1,2]. We amended our method recently [3] to detect and correct frequently encountered artifacts in scanning probe microscopy, i.e. effects of multiple mini-tips that collectively result in a blunt tip [4,5]. Loosely speaking, our version of CIP has the effect of "sharpening up" a blunt scanning probe tip. This is achieved by the deconvolution of the prevailing microscope's point spread function from the SPM images. Although many scanning probe microscopists have so far been content with ignoring these kinds of artifacts, there are also highly credible reports on unambiguous observations on scanning probe tip changes during data recordings that led to blunt tip artifacts in SPM images [6,7]. One of these reports proposes that multiple mini-tips cannot affect the character of the observed translation symmetry in such an image while the 2D periodic motif may be smeared out [6]. Our theoretical analysis [4] confirms this idea so that one can confidently take "inconsistencies" between observed 2D translation and point symmetries in SPM images (Fig. 1) as the hallmark of multiple mini-tip artifacts. Our unambiguous determination of the underlying Bravais lattices of 2D periodic surface arrays [3] on the basis of a geometric Akaike Information Criterion (AIC) [8] achieves the detection of multiple minitip artifacts on a statistically sound basis. Figure 1 demonstrates the effectiveness of our version of CIP in removing blunt tip artifacts from a simulated scanning tunneling microscope (STM) image of long-range ordered 2D periodic arrays of cobalt-phthalocyanine molecules on a (001) oriented gold surface. Note that the central circular area in the left part of Fig. 1 illustrates both (i) an inconsistency between 2D translation and point symmetries and (ii) the fact that structural scanning probe tip changes during the recording of experimental data cannot change the character of the observed translation symmetry in the resulting SPM image. Our recently developed unambiguous translation symmetry determination procedure [3] also constitutes progress towards making CIP more objective in both electron crystallography by TEM and surface feature assessments by SPM. This is because not all plane symmetry groups are disjoint. As it is well known, all of the symmetry operations of a plane symmetry group are contained in its so called translationengleiche minimal supergroup [9]. The application of the traditional quantifiers of the deviations of experimental images from their plane symmetry enforced counterparts (as used in electron crystallography) can, therefore, never be completely objective [1,4]. Because the practitioners of electron crystallography have typically chemical intuition (and sometimes prior knowledge) about the atomic arrangement they are trying to solve and often work with multiple 2D projections of the same 3D crystals, this lack of complete objectivity is tolerable in that field. Scanning probe microscopists that are trying to utilize CIP in their field are less fortunate as they need to work from one 2D projection only. Geometric AICs [8] have specifically been developed to select the best model out of a set of non non-disjoint models so that their application in CIP for Microsc. Microanal. 21 (Suppl 3), 2015 1612 SPM is mandatory. Associated with the usage of geometric AICs in CIP [4,5] is, however, the requirement that systematic errors in 2D periodic images need to be corrected to such a level that they become small compared to random errors [10]. [1] P. Moeck in "Microscopy: Science, Technology, Applications and Education", eds. A. M�ndez-Vilas and J. D�az, (FORMATEX, 2010), p. 1951, open access: http://www.formatex.info/microscopy4/1951-1962.pdf. [2] B. Moon, Master of Science Thesis, Portland State University, 2011; open access: http://nanocrystallography.research.pdx.edu/media/thesis14acorr.pdf. [3] T. T. Bilyeu, Master of Science Thesis, Portland State University, 2013; open access: http://nanocrystallography.research.pdx.edu/media/cms_page_media/6/Taylor_thesis_final.pdf. [4] J. C. Straton et al, Cryst. Res. Technol. 49 (2014) 589. [5] J. C. Straton et al, J. Advanced Microsc. Res. 9 (2014) 206. [6] E. V. Iski et al, J. Vac. Sci. Technol. A 29 (2011) 040601. [7] A. Labuda et al, Microscopy and Analysis, SPM suppl., March/April 2014, p. 21. [8] K. Kanatani, Int. J. Computer Vision 26 (1998) 171. [9] T. Hahn (editor), Brief Teaching Edition of Volume A, Space-group symmetry of the International Tables for Crystallography, (Springer, 2005). [10] The first author thanks Portland State University's Venture Development Fund, Faculty Enhancement program, and Internationalization Council for financial support. Professor Kerry W. Hipps of Washington State University in Pullman is thanked for the STM image that we used for the creation of Figure 1. double tip # III sharpened-up tip double tip # II double tip # I single tip Figure 1. Demonstration of the CIP removal of double-tip artifacts in scanning probe microscopy. Left: simulated raw data. The very small stripe at the bottom of this image represents the sample as imaged with a single scanning probe tip. The three wider stripes represent the sample as imaged with double tips of different geometries. The inset represents 1.1 unit cells of the sample array as imaged with a single tip as contour plot with 32 gray scale levels. Middle: Classical Fourier filtering of the image (to the left) cannot remove double (and multiple) scanning probe mini-tip artifacts as it represents only translation symmetry averaging. Right: CIP result on the circled area in the raw data (i.e. the first image in the row). The insets in the last two images of this row represent 1.1 unit cells of the middle and right array as contour plots with 32 gray scale levels. Note the striking similarity of the left and right inset, which demonstrates the above mentioned sharpening-up of the scanning probe tip. The 2D periodic motif in all three insets is approximately 1.5 nm wide. Note also that 2D periodic motifs with point symmetry 1 (i.e. no other point symmetry than a rotation by 360�) in the first two images in this row are inconsistent with the clearly recognizable square Bravais lattice, because that kind of 2D translation symmetry requires as part of the plane symmetry groups p4, p4mm, and p4gm two Wyckoff positions that posses at least point symmetry 4.");sQ1[805]=new Array("../7337/1613.pdf","Conductive Atomic Force Microscopy Characterization of Ultra-Thin Diamond-Like Carbon Films on Magnetic Recording Heads","","1613 doi:10.1017/S1431927615008843 Paper No. 0805 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Conductive Atomic Force Microscopy Characterization of Ultra-Thin DiamondLike Carbon Films on Magnetic Recording Heads. Shirley Mejia1, Sean P. Leary1, Guilherme P. Souza1, and Kurt C. Ruthe1 1. Western Digital Corporation, Magnetic Heads Operations, Metrology & Materials Characterization, Ayutthaya, Thailand, 13160 Tetrahedral amorphous diamond-like carbon (DLC) has enjoyed a long tenure in the technology evolution of magnetic recording in rotating hard disk drives, where it serves dual functions as a barrier to corrosion of the ferrous recording elements and an interface for tribological wear resistance [1]. Incremental advances in areal storage density continue to rely in great part to decreasing the spacing between the recording head and media, both of which use some form of DLC which in turn can consume as much as 40% of the total spacing budget allowed by design. Thus, commensurate downward scaling of these overcoats is a critical technology enabler. With that scaling, the efficacy of the films to meet their core functional requirements as diffusion barriers and tribological wear protection becomes challenged as the overcoat thickness approaches the scale of the surface topography [2]. In this work, we discuss the application of conductive atomic force microscopy [3] (CAFM) for the characterization of DLC films in the 10-60 � thickness regime and new insights gained into the robustness of films to meet functional requirements as scaling progresses into the sub-20� regime. A conventional filtered cathodic arc (FCA) process [4] was used to deposit DLC films in the 10-60 � thickness range on magnetic recording heads, and all films included a Si-N adhesion layer on the order of 10 � deposited in situ under vacuum in the same deposition system used for the DLC films. Prior to deposition, samples received mechanical lapping, chemical cleaning, and an in situ sputter pre-etch. The nominal film thicknesses were controlled in situ by spectroscopic ellipsometer and verified by TEM. Surface topography and conductivity were measured simultaneously with a commercially available CAFM system. A variable gain current amplifier capable of 103-109 V/A provided probe current sensitivity in the 10-104 pA range. I-V curves of the films at various thicknesses were used to identify the optimal sample bias for high SNR below the noise and breakdown voltage thresholds of the films, but a fixed sample bias of 0.5 V, optimized for 20 � nominal films, was used throughout the studies. Subsequent TEM imaging and EELS were performed using a Schottky field emission gun TEM operating at 200 kV, and a post-column spectrometer with a ~50 mrad collection angle and energy resolution ~1.2 eV. TEM samples were prepared with a Cr2O3 capping layer to preserve the DLC integrity prior to FIB cross-sectioning. The RMS current detected as a function of DLC thickness, shown in Fig. 1, is observed to increase in the thickness range 10-20 � where is reaches a maximum and then rapidly decays as the thickness increases to 60 �. Additionally, the sample-to-sample dispersion in the measurements also scales inversely with the film thickness as a second-order quadratic. This behavior suggests two regimes can be defined, and it is proposed that the sub-20 � regime is dominated by oxide formed with the Si-based adhesion layer due to poor coverage and permeability of the DLC film, while the thick film regime (>20 �) is dominated by the resistance of the DLC film itself. This hypothesis was tested by partitioning the head fabrication process with CAFM. Fig. 2 shows that no significant current is detected from the surface after cleaning, etching, and deposition of the DLC adhesion seed layer, respectively, consistent with oxidation of the surface occurring between the processing and measurements. Conversely, a 20 � Microsc. Microanal. 21 (Suppl 3), 2015 1614 DLC film on a clean NiFe surface yields a strong response, but the inclusion of the seed layer in the same film weakens the response, consistent with the hypothesis that the seed layer plays a role. However, the sample-to-sample dispersion suggests that the variation may be dominated by seed-carbon inter-diffusion and oxidation rather than oxidation alone. EELS elemental profiles of full films including seed and DLC shown in Fig. 3 show that at 20 � DLC thickness oxidation of the seed layer begins to occur, and below 20 � Si diffuses into the DLC and oxidizes more fully, indicating that the effectiveness of the DLC as a diffusion barrier is compromised. In summary, we have shown CAFM to be a useful technique for characterization of DLC thin films on magnetic recording heads. It can be used to complement established techniques such as TEM as a proxy film thickness metrology; however, the ultra-thin sub-20� thickness regime reveals a highly variable conductivity mechanism which complicates interpretation of data without additional TEM analysis. These phenomena, resulting from marginality of the film's ability to completely protect the head from oxidation, potentially puts a fundamental physical limit on opportunity for further downward thickness scaling on magnetic recording heads without new process and materials breakthroughs. References: [1] AC Ferrari, Surf. Coat. Tech. 180 �181 (2004), p. 190. [2] A Herrera-Gomez et al, Surf. Interface Anal. 39 (2007), p. 904. [3] AT Rakhimov et al, Proceedings of 11th IVMC (1998) p. 225. [4] J Robertson, Mater. Sci. Eng. R37 (2002), p. 129. Figure 1. CAFM RMS current measured on NiFe as a function of film thickness. Figure 2. CAFM partitioning of process steps in DLC fabrication. Figure 3. EELS elemental profiles of various DLC thicknesses.");sQ1[806]=new Array("../7337/1615.pdf","Characterization of Nanocarrier Complexes with Plasmid DNA using SEM, TEM and AFM","","1615 doi:10.1017/S1431927615008855 Paper No. 0806 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of Nanocarrier Complexes with Plasmid DNA using SEM, TEM and AFM Jae-Young Cho1, Usha D. Hemraz2, Pankaj Bhowmik3, Goska Nowak3 and Patricia L. Polowick3 1. National Institute for Nanotechnology, National Research Council, 11421 Saskatchewan Drive, Edmonton, Alberta T6G 2M9, Canada 2. National Research Council, 6100 Royalmount Avenue, Montreal, H4P 2R2, Canada 3. National Research Council, 110 Gymnasium Place, Saskatoon, Saskatchewan S7N 0W9, Canada Transfection methods using nanomaterials have been studied intensively over the past several years as they show potential in biomedicine and therapeutic applications [1]. However, there are many barriers to the successful use of these new nanomaterials, the plant cell wall is especially troublesome. One technical hurdle is the development of an effective delivery system, in which the formation of nanocarrier complexes is the most critical factor. There are many different nanocarriers for complexation, such as nanotubes and nanoparticles. Among these, rosette nanotubes (RNTs) have been brought to our attention since they are biocompatible nanomaterials generated from the self-assembly of a bio-inspired bicycle that features the hydrogen bonding arrays of guanine and cytosine. These RNTs, which can grow up to several micrometers in length, are stable nanocarriers with a hollow core of 11 �. The dimensions and properties of these nanotubes can be tuned by modifying the building block or the functional groups attached to it [2-3]. In this study, biocompatible self-assembled RNTs functionalized with oligo-lysine side chains were used as cationic nanocarriers. Microscopy techniques, namely, scanning electron microscopy (SEM), transmission electron microscopy (TEM) and atomic force microscopy (AFM) were used to investigate the interaction between RNTs and plasmid DNAs under various conditions. The bio-science AFM, which uses fluorescent microscopy in conjunction with AFM, was used to identify the complexation. For the SEM and TEM studies, samples were prepared by depositing a droplet of solution on a carboncoated 400-mesh copper grid (Electron Microscopy Sciences), the excess solution was blotted after 10 s. The staining of samples for TEM was performed by depositing one droplet of a 2% uranyl acetate solution for 120 s. The grid was then blotted and dried. For tapping mode AFM imaging, samples are made by depositing 5�l solution on freshly cleaved mica, which were then air dried. For bio-science AFM imaging, 5�l samples were deposited in solution on a cover glass (0.17 mm thick-EMS) and airdried. The sample surface was then rinsed with 500 ml MilliQ water. SEM images were obtained without negative staining, at an accelerating voltage of 30 kV, 20 �A and a working distance of 5-8 mm on high resolution ultra-high resolution (0.4 nm) Hitachi S-5500 cold field emission SEM. TEM observation was carried out on JEOL 2200 FS TEM � 200 kV Schottky field emission instrument equipped with an in-column omega filter. Bright field TEM images were acquired using energy filtered zero loss beams (slit width 10ev). Tapping mode AFM images were obtained using a Digital Instruments/Veeco Instruments MultiMode Nanoscope IV AFM, equipped with an E scanner. For io-Science AFM, a combination of an inverted microscope (Nikon) and AFM (JPK NanoWizard II) were used with 2N/m, 70 kHz, tetrahedral tip. Microsc. Microanal. 21 (Suppl 3), 2015 1616 Figure 1 shows that plasmid DNA binds strongly with RNTs and produce complexes. The details of complexation were imaged, as shown in Figure 1b and c. Especially, many free plasmid DNAs are visible outside of complexes in Figure 1c. The inset shows the periodic relationship between plasmid DNAs and RNTs, which may indicate that the plasmid DNAs follow the surface characteristic (i.e. surface charge density) of RNTs during complexation. Figure 2 shows the complexes with fluorescein isothiocyanate (FITC) labelled RNTs with plasmid DNA. Comparing both fluorescent and AFM images reveals the presence of additional plasmid DNAs, which are not attached to the fluorescent nanocarriers. In summary, the complexation of plasmid DNA and RNTs was investigated by different microscopic techniques. In addition, the complexes of FITC attached RNTs with plasmid DNAs were visualized successfully by Bio-Science AFM, showing the fluorescent and AFM images simultaneously. Complexation with different ratios and in-solution AFM will be performed with the goal to evaluate uptake efficiency of the complex into plant cells in the near future. References: [1]. Mozafari, M. R. Nanocarrier Technologies: Frontiers of Nanotherapy, Dordrecht: Springer, 2006. [2]. Cho, J.Y., Fenniri, H., Imaging & Microscopy, 2010, 4, 33�35. [3]. Hemraz, U.D.; El-Bakkari, M.; Yamazaki, T.; Cho, J.Y.; Beingessner, R.; Fenniri, H. Nanoscale, 2014, Nanoscale, 2014, 6, 9421-9427. [4]. We thank National Research Council CANADA, University of Saskatchewan and Government of Saskatchewan for their support. a b c 400nm Figure 1. Microscopic images of complexation of plasmid DNA and RNT: (a) SEM, (b) TEM, (c) AFM. An inset shows the plasmid DNAs wrap the RNT periodically as indicated by arrows. a b c Figure 2. Bio-science AFM images: (a) JPK NanoWizard II (b) fluorescent image and (c) combined AFM image on the same area: the plasmid-DNA is only visible by AFM as indicated by white arrows. Scale bar is 20 um.");sQ1[807]=new Array("../7337/1617.pdf","Custom Modification of AFM Tips for Fast, High Force Resolution Single-Molecule Force Spectroscopy.","","1617 doi:10.1017/S1431927615008867 Paper No. 0807 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Custom Modification of AFM Tips for Fast, High Force Resolution Single-Molecule Force Spectroscopy. Aric W. Sanders1, Jaevyn K. Faulk2, Devin T. Edwards2 and Thomas T. Perkins2 1. National Institute of Standards and Technology, Boulder, Colorado, USA. 2. JILA, National Institute of Standards and Technology and University of Colorado, Boulder, Colorado 80309, United States In addition to providing the ability to image on the nanoscale, atomic force microscopy (AFM) has the ability to measure small (pN) forces. This ability has led to new insights into conformational changes in biological molecules; in particular, single-molecule force spectroscopy (SMFS) is a powerful tool to investigate folding in proteins. Ideally, one could observe folding in proteins at time scales in the microsecond range with both short-term force precision and long-term force stability. Recent work [1] has shown that to minimize force drift, one must use a soft AFM cantilever due to instrumental noise. Such soft, long cantilevers have poor temporal resolution. When high temporal resolution is required, one choses a shorter cantilever, which has a higher spring constant ( -3 ) and hence increased force drift. Recently, it has been shown that by modifying a commercially available cantilever with focused ion beam (FIB) milling, short AFM cantilevers (L~40 m) can be softened to reduce force drift, without sacrificing temporal resolution. The resulting cantilevers offer state-of-the-art force stability and precision, with temporal resolution ~70 s [2]. To further improve the performance of modified AFM cantilevers, FIB modification strategies need to be developed to extend the technique to ultra-small cantilevers (L~9 m) , which offer temporal resolution ~1 s [3].We present several modification strategies, each of which decrease the stiffness of the resultant cantilevers while retaining fast response times. Additionally, we explore techniques that improve the yield of the process. Our basic process of AFM cantilever modification relies on decreasing the stiffness of a commercially available cantilever, such as a Biolever Mini or Bio Lever Fast [4]. The reduction of stiffness is achieved primarily through the removal of material on the cantilever through Ga-ion-based FIB milling. We are also exploring He-ion milling. This milling is either used to thin the full cantilever (Figure 1) or to remove a rectangle of material and then to thin the remaining supports (Figure 2). In addition to lowering k, removing a cantilever's gold coating has been shown to reduce force drift [5]. This is achieved by protecting an area using electron beam induced deposition of a suitable etch mask, and then wet etching most of the gold away leaving a reflective surface to measure the cantilever, or FIB-milling of the gold. These methods have successfully produced FIB-modified, ultra-short cantilevers with pN integrated force noise (1-350 Hz) and s temporal resolution. Yet, the yield of the process is limited by the tendency of the cantilevers to bend while being modified. This bending can cause issues with milling. More importantly, if the cantilevers remain highly bent, they are not measurable by the optical deflection laser. The bending of cantilevers is seen to be a function of both electron and ion exposure and several strategies to minimize both have been developed. (see Figure 2). For sufficiently long cantilevers (L 40 m), exposure to electrons causes little to no bending, whereas ultra-small cantilevers bend quickly. We present a reproducible protocol to FIB-modify cantilevers of varying sizes without bending. Microsc. Microanal. 21 (Suppl 3), 2015 1618 References: [1] Churnside, A. B. and T. T. Perkins FEBS Letters 588 (2014), p 3621-3630. [2] Bull, M. S., et al. ACS Nano 8 (2014), p 4984-4995. [3] Rico et al., Science 342 (2013), p 741-743 [4] Certain commercial equipment, instruments, or materials are identified in this paper to foster understanding. Such identification does not imply recommendation or endorsement by the National Institute of Standards and Technology, nor does it imply that the materials or equipment identified are necessarily the best available for the purpose [5] Churnside, A. B., et al. Nano Letters 12 (2012), p 3557-3561. Figure 1. A modified cantilever, with thermal noise curve. This cantilever, a Biolever fast, has been thinned from the bottom, decreasing the spring constant from 187 pN/nm to 105 pN/nm. In this case the modified cantilever has higher force noise, an occasional undesirable side effect of modification. Figure 2. Modification of AFM cantilevers to reduce stiffness by removal of a central section. Top left, unmodified ultra-small cantilever. Top right, cantilever after modification imaged normal to cantilever. Bottom left, an unmodified small cantilever. Bottom right, a tilted image of the modified cantilever showing a large degree of bending.");sQ1[808]=new Array("../7337/1619.pdf","Comparison of Magnetic Domain Observation by Means of Magnetic Force Microscopy and Lorentz Transmission Electron Microscopy","","1619 doi:10.1017/S1431927615008879 Paper No. 0808 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Comparison of Magnetic Domain Observation by Means of Magnetic Force Microscopy and Lorentz Transmission Electron Microscopy S. Hua1 and M. De Graef1 1 Dept. of Materials Science and Engineering, Carnegie Mellon University, Pittsburgh PA 15213, USA With the emergence of scanning probe technology such as magnetic force microscopy (MFM), magnetic domain imaging has become relatively easy and fast compared to some of the more traditional observation techniques, such as Lorentz transmission electron microscopy (LTEM). MFM is based on a standard scanning probe microscope (SPM), but with an extra ferromagnetic layer deposited on the tip. The magnetostatic force or force gradient between a magnetic sample and a magnetic tip is recorded and the topographic background is subtracted out. The magnetic contrast obtained from MFM is representative of the stray magnetic field above the sample surface and is determined by the magnetization orientation of the tip. Thus, in order to fully describe the magnetic field distribution inside and around the sample, another domain observation technique must be combined with MFM. LTEM is a good candidate because it has similar spatial resolution [1] and is sensitive to the in-plane magnetic component inside the sample. With both MFM and LTEM results available on the same sample, quantitative magnetization profiles of a domain wall can be obtained in principle using a dedicated image processing approach. Then, quantitative magnetic information, such as the domain wall width (DWW), or potentially the more fundamental exchange parameter, can be determined from a combination of these two domain observation techniques. The observation of the same magnetic domain structure using both MFM and LTEM was achieved using a single crystal of Ni49.9 Mn28.3 Ga21.8 (AdaptaMat Ltd, Finland). The sample was heat-treated to achieve a thermally induced multi-variant martensitic state prior to electropolishing [2]. Fig. 1(a) shows a Fresnel image taken on a Tecnai F20 FEG-TEM operated at 200 kV in Lorentz observation mode. Two kinds of martensitic variants can be found in this area, and each has a distinct magnetic domain structure. In variant I, the easy axis (tetragonal [001]) is oriented out of the plane, nearly normal to the sample, which results in a fine-scale maze-like domain structure due to the strong magnetostatic interactions which force the magnetization to rapidly change over short distances in order to minimize the stray field energy. In comparison, straight and more widely spaced magnetic domain walls are observed in variant II. This domain structure indicates that the magnetization direction and the easy axis are in the foil plane, and hence there is a low stray field surrounding these variants. Fig. 1(b) shows a MFM image from an area close to that shown in Fig.1(a). The same wedge-shape martensitic variants are observed in the MFM phase image. The variants that display very little magnetic contrast correspond to variant II in the Lorentz image Fig. 1(a). This is because the magnetization of the MFM tip was perpendicular to the plane of the foil and the magnetization state inside variant II is mostly in-plane; thus, the out-of-plane component of the demagnetizing fields is very small and does not strongly affect the tip vibration state. In comparison, strong magnetic contrast is found in the other martensitic variant, which corresponds to variant I in Fig. 1(a). In these variants, because the magnetization in different domains is nearly perpendicular to the foil plane, the magnetostatic energy dominates and the flux closure forms a strong demagnetizing field above the variant. However, the domain structure in Fig. 1(b) is not as smoothly curving as the maze-like domain structure in Fig. 1(a). This likely indicates that the magnetization is not precisely perpendicular to the foil plane and the domain walls are likely tilted; in the LTEM images, this would only give rise to a slight broadening of the domain wall contrast. Microsc. Microanal. 21 (Suppl 3), 2015 1620 1610 For the LTEM observation mode, the electron phase shift due to the magnetic vector potential inside and surrounding the sample can be reconstructed from through-focus Fresnel images using the transport-ofintensity equation approach. This phase shift undergoes a linear change inside a uniformly magnetized domain and a non-linear change (with non-zero second derivative) across a domain wall, which we can use to determine the projected width of a domain wall. Using this approach, the average domain wall width in variant I of Fig. 1 was determined to be 17.5 � 3 nm. However, this method assumes that the domain walls are strictly perpendicular to the foil plane. In Fig. 1(b), the MFM image indicates that the magnetization in some domain regions may not be perpendicular to the foil plane, which results in a somewhat discontinuous maze-like domain structure. Further analysis, including micromagnetic modeling, of these domain configurations will be presented. In summary, we have managed to observe the same magnetic domain structure corresponding to different martensitic variants in a Ni -Mn-Ga alloy foil using both LTEM and MFM techniques. In our ongoing work, we are attempting to correlate the domain wall width measurements from the LTEM observartion mode with those from the MFM mode, with as a final goal the experimental determination of the magnetic exchange constant of this material. References: [1] A. Hubert in "Magnetic domains: the analysis of magnetic microstructures", (Springer, New York) [2] A. Budruk, C. Phatak, A.K. Petford-Long, M. De Graef, Acta Materialia 59 (2011), p.4895 [3] The authors acknowledge funding from the National Science Foundation, Grant NSF-DMR #1005330. Figure 1. (a) Fresnel image of magnetic domain structure in a thermally treated multi-martensitic state Ni Mn Gathin foil. (b) MFM image of the same magnetic domain structure in the area close to Fig.1(a).");sQ1[809]=new Array("../7337/1621.pdf","Use of the Unroofing Technique for AFM Direct Imaging of the Intra-Cellular Structure at High Resolution","","1621 doi:10.1017/S1431927615008880 Paper No. 0809 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Use of the Unroofing Technique for AFM Direct Imaging of the Intra-Cellular Structure at High Resolution Jiro Usukura1, Eiji Usukura1, Akihiro Narita1, Akira Yagi2, Shuichi Ito2 1. Graduate School of Science, Nagoya University, Nagoya, 464-8630 Japan 2. Olympus Corporation. Hachioji, Tokyo 192-8512 Japan Atomic force microscope (AFM) is capable of detecting topological surface structure of living cells in aqueous condition, but not able to describe inside of the cells directly. Recently, however, unroofing technique provided the way to investigate the intra cellular structure by AFM, since inside of cell was exposed by removing the cell membrane (unroofing) and subsequent washing the cytoplasm out (1). In addition to such improvement of preparation methods, recent function of AFM is advanced markedly, in particular, on resolution and scanning speed. Then we used AFM in order to investigate intracellular cytoskeleton in the buffer solution at high resolution comparable to electron microscopy. NRK or Vero cells were cultured on the slide glass with hydrophobic framework for a few days. Cultured cells were washed once with HEPE based Ringer's solution and soaked briefly in poly-L-lysine solution followed by washing three times with Ca free Ringer's solution. Subsequently, cells placed in KHMgE buffer (30 mM HEPES, pH 7.4, 70 mM KCl, 3 mM MgCl2, 1 mM EGTA, 1 mM DTT, 0.1 mM AEBSF (4-(2-aminoethyl) benzenesulfonyl fluoride hydrochloride)), and the dorsal cell membrane was removed (unroofing) by sonication induced bubble jet (2) at low power (27kHz, 0.5W). Cells unroofed were washed briefly in fresh KHMgE buffer, and then fixed with 2 % glutaraldehyde in the same buffer. After being washed twice with the same buffer, cells unroofed on above slide glass were brought into AFM (BIXAM, Olympus Corporation) for observation. We succeeded to view membrane cytoskeleton, clathrin coats and caveoli directly in buffer solution at high resolution similar to electron microscope. Spatial architecture of stress fibers formed of single actin filament extending from the cytoplasmic surface of the cell membrane was arrested in detail (Fig 1). In practice, several substances raping and thereby bundling actin filaments was recognized clearly on the stress fibers. AFM was also able to describe characteristic short periodicity of 5 nm as a striation on the single actin filaments branched from stress fibers. Furthermore, clathrin coated pits and caveoli in addition to peripheral membrane proteins were observed clearly. In common sense, it has been considered to be difficult so far that AFM displayed the surface of caveoli in detail, though characteristic vortex ridges (supposed to be caveolin 1) appeared on the surface of caveola were detected clearly rather than in freeze-etching replica and scanning electron microscopy (Fig. 2). This is due in part to loss of metal coating. In case of clathrin coats, frequently, tri-skelion of clathrin and adaptin were partly discernible (Fig. 3). Fig.4 shows higher magnification image of unfixed single actin filament with periodical striations of 5 nm that are derived from assembled G actins. Thus, isolated single actin fibers were observed well at high resolution under non-fixed condition. However, in the cells, it is difficult to observe the single actin filament at atomic resolution since unknown reason. Spatial architecture of cytoskeleton, in particular actin filaments crowing cytoplasmic membrane surface, clathrin coats and caveoli were successfully observed directly in buffer solution under AFM at high resolution comparable to electron microscope. Microsc. Microanal. 21 (Suppl 3), 2015 1622 References: [1] Usukura J et al. J. Electron Microsc. 61 321-326 2012 [2] Heuser, J E, Anderson, R G J. Cell Biol. 108 (1989), 389�400 [3] The author (J.U) acknowledges funding from the Japan Science and Technology Agency (JST), "Development of Systems and Technologies for Advanced Measurement and Analysis" Program. Fig.1: Membrane cytoskeleton showing stress fiber and branching actin filaments on the cell membrane. Fig. 2: Cluster of Caveola showing characteristic ridge structures on their surface. Fig.3: High power image of clathrin coat. Fig. 4: Molecular resolution image of an isolated actin filament.");sQ1[810]=new Array("../7337/1623.pdf","Lateral Force Microscopy using Nanomanipulation","","1623 doi:10.1017/S1431927615008892 Paper No. 0810 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Lateral Force Microscopy using Nanomanipulation Yen-Kai Hseu, Helen McNally1. 1. School of Engineering Technology, Birck Nanotechnology Center, Purdue University, West Lafayette, United Sates. The scanning probe microscope (SPM) is a high precision measurement research equipment that enables one to obtain images of a sample's surface topography at the nano-level using atomic force microscopy (AFM) techniques. In addition to its imaging capabilities, the SPM can also be used to obtain material characteristics such as surface electrical charges, magnetic properties, vertical/lateral force, and friction. Surface force data, represented through force curves, can be generated by different methods of SPM tip and surface interactions. The most common and well developed method of force measurement using SPM is vertical force microscopy. Force measurements in the lateral direction have also been conducted but results have mainly been qualitative. The scope of this research is to develop an SPM technique that would enable lateral forces to be quantified just like vertical forces. One motivation for this research involved incorporation of the technique with medical research. Complex structures contained within human cells are encapsulated within a lipid bi-layer known as the cell membrane. The cell membrane acts as a gateway between the internal cell structures and the external environment regulating what enters and exits the cell. Exposure of a cell to internal or external stimuli such as mutation or non-ideal physiological conditions can cause changes in the physical properties of the cell membrane. Past research in a similar area have shown that a direct correlation exists between a sample's surface vertical forces and its viscosity[1]. The Bruker Catalyst AFM system will be used to develop the multi-axial force measurement technique. The new lateral force technique incorporates a variety of the system's existing functions including the existing topographical scanning functions and an add-on nanomanipulation function known as the NanoMan. Two different methods have been proposed to conduct the force measurement in the lateral direction to generate lateral force curve data. The two methods involves pushing against the sample's internal and external structure to induce a twist in the cantilever that would mimic a lateral approach by the AFM tip. The two methods are presented in figure 1 and will be achieved through the Catalyst AFM's nanomanipulation NanoMan function. The multi-axial force microscopy technique will be tested on a variety of samples of different hardness consisting of hard, medium, and soft samples. The hard sample chosen was an AFM calibration grid with depth of 200nm and pitch of 10um. The pushing force method will be conducted on the sidewall of the pits. The soft sample consisted of DOPC (1,2-dioleoyl-sn-glycero-3-phosphocholine) lipids prepared with the method discussed by Johnson et al[2] to form liposome membranes. Liposomes have been used by a multitude of researchers, including Sch�nherr et al[3], to simulate a real cell membranes. The liposomes are laid down on a layer of muscovite MICA through natural attraction forces. Muscovite MICA was chosen for the material's relatively smooth surface. A topographical image and profile of the soft sample is presented in figure 2. The medium hardness sample to be tested is still to be determined. Over the last decade, researchers have well defined vertical force measurement techniques. Vertical force curves can be easily generated using built in functions in almost all SPM systems. The anticipated results Microsc. Microanal. 21 (Suppl 3), 2015 1624 of the research involves producing lateral force curves, using a combination of nanomanipulation and force measurement techniques, that will to a certain extent produce the same results produced by a lateral force curves generated on the same sample. Figure 1. Multi-axial force measurement lateral force measurement methods conception. The pushing force method (left) consists of the AFM tip pushing against the side of the sample and the scratching force method (right) consists of AFM tip puncturing into the sample surface and pushing left and right in the lateral direction. Figure 2. Soft sample AFM topographical scan (left) and profile (right). The soft sample consists of a MICA surface with DOPC liposome bilayers (seen in white) attached. The topographical profile (right) depicts the height and length measurements of one of the DOPC liposome bilayer seen on the topographical scan marked on the top right section of the image. References: [1] Mustata, Mirela, Ken Ritchie, and Helen A. Mcnally. "Neuronal Elasticity as Measured by Atomic Force Microscopy." Journal of Neuroscience Methods 186.1 (2010): 35-41. [2] Johnson, Joseph M., Taekjip Ha, Steve Chu, and Steven G. Boxer. "Early Steps of Supported Bilayer Formation Probed by Single Vesicle Fluorescence Assays." Biophysical Journal 83.6 (2002): 3371-379. [3] Sch�nherr, Holger, Joseph M. Johnson, Peter Lenz, Curtis W. Frank, and Steven G. Boxer. "Vesicle Adsorption and Lipid Bilayer Formation on Glass Studied by Atomic Force Microscopy." Langmuir 20.26 (2004): 11600-1606.");sQ1[811]=new Array("../7337/1625.pdf","Correlated Atomic Force Microscopy and Single Molecule Localization Microscopy","","1625 doi:10.1017/S1431927615008909 Paper No. 0811 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Correlated Atomic Force Microscopy and Single Molecule Localization Microscopy Pascal D. Odermatt1, Arun Shivanandan2, Hendrik Deschout2, Radek Jankele2, Adrian, P. Nievergelt1, Lely Feletti2, Michael W. Davidson3, Aleksandra Radenovic2 and Georg E. Fantner1 1 Laboratory for Bio- and Nano-Instrumentation, Institute of Bioengineering, School of Engineering, EPFL, 1015 Lausanne, Switzerland 2 Laboratory of Nanoscale Biology, Institute of Bioengineering, School of Engineering, EPFL, 1015 Lausanne, Switzerland 3 National High Magnetic Field Laboratory, Florida State University, Tallahassee, Florida, USA. Department of Biological Science, Florida State University, Tallahassee, Florida, USA. Nanoscale characterization of living samples has become essential for modern biology. Atomic Force Microscopy (AFM) creates topological images of fragile biological structures from biomolecules to living cells in aqueous environment [1]. However, correlating nanoscale structure to biological function of specific proteins can be challenging. Fluorescence microscopy on the other hand is capable of revealing specific biochemical structures with great specificity. The combination of fluorescence microscopy and atomic force microscopy has therefore long since been a very powerful tool. However, the resolution of the two techniques differs vastly, which has made correlated imaging somewhat ambiguous. With the rapid development of single molecule localization microscopy techniques [2] (SMLM), such as PALM (photo activated localization microscopy) and STORM (stochastic optical reconstruction microscopy), this gap in resolution is closing. These methods are therefore often called "super resolution" microscopy techniques. Correlated AFM/SMLM therefore hold tremendous potential for structural and cellular biology. To make full use of a combination of the two techniques it is essential that each technique retains its optimal performance, even in combination experiments. In this work we present a combination AFM/SMLM instrument that allows each of the two techniques to operate at its full potential. The special construction of the microscope allows for integrating an AFM with an optical microscope without an increase in vibrational noise. The AFM noise performance on the combined system was measured to be 0.24 Angstrom, which is equivalent to that of very low noise stand alone AFMs. The resolution in the SMLM image obtained with the combined system shows a mean localization precision of 12.5 nm, which is also equivalent to that of dedicated systems. AFM offers the unique capability of observing living cells in physiological solution with exceptional surface resolution, but internal cellular structures remain hidden. SMLM can reveal structures on the inside of cells. This is particularly interesting for microtubules or actin filaments, which play a big role in cell stability and intra-cellular transport. Actin filaments have been studied extensively both with atomic force microscopy as well as super-resolution optical microscopy. To compare the imaging resolution of atomic force microscopy and super resolution optical microscopy we used direct Stochastic Optical Reconstruction Microscopy (dSTORM)/AFM, to directly correlate and quantify the density of localizations using both imaging modalities along (F-) actin cytoskeletal filaments. After having established the correlation of the two techniques by measuring the same aspect of the same sample, we demonstrate the ability of measuring complementary aspects of a sample. Photo Activated Light Microscopy (PALM)/AFM, we provide correlative images of bacterial and mammalian cells in aqueous condition. The complementary information provided by the two techniques opens a new dimension for structural and functional nanoscale biology. [3] Microsc. Microanal. 21 (Suppl 3), 2015 1626 Figure 1: Comparison of AFM, dSTORM and TIRF imaging resolution from correlative imaging. ( a, c & e) AFM, dSTORM and TIRF images respectively (1.5 �m x 1.5 �m) of correlated filaments. Lines and colours indicate the location of profiles shown in (b, d & f). (g) Overlay of dSTORM probability map (blue) and 3D rendered AFM image (yellow-brown-). (h) Different F-actin arrangements suggested based on the AFM data. (i) Bar plot showing the number of localizations detected at a specific height measured by AFM divided by the number of pixels having that particular height. Bin size: 2 nm. Orange: Maximum height projection of the AFM section showed below the plot. Blue: Number of localizations detected in regions between the white lines per line scan of the AFM. (j) Correlative AFM/PALM image of E.coli bacteria expressing RNA polymerase-mEos2. [1] Henderson, E., Haydon, P. G., Sakaguchi, D. S. Science 257, (1992), p5078, 3. [2] Betzig, E., et al. Science 313 (2006), p. 1642-1645 [3] This work was financially supported by FNS grants No. 200021�125319, No. 20021�132206, No. 205321-134786 and No. 205320-152675 as well as the European Union's Seventh Framework Program FP7/2007-2011 under grant agreement 286146 and FP7/2007-2013/ERC grant agreement 307338. A. Shivanandan was funded by a PhD fellowship grant from NCCBI. We thank Dr. P. Annibale for initial help with the instrument. We thank Lina Carlini for discussion and Prof. Mike Heilemann on the generous gift of bacterial strain expressing mEos2-RNAP");sQ1[812]=new Array("../7337/1627.pdf","Pushing the Limits on SEM Quantification � Combined Quantification with SDD and Fully Focussing WD detectors","","1627 doi:10.1017/S1431927615008910 Paper No. 0812 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Pushing the Limits on SEM Quantification � Combined Quantification with SDD and Fully Focussing WD detectors C.L. Collins1, J. Holland1, S.R. Burgess1 1 Oxford Instruments Nanoanalysis, Halifax Road, High Wycombe, HP12 3SE, UK Wavelength Dispersive Spectroscopy (WDS) was, until recently, the only way to achieve accurate, repeatable minor (<1.0wt%) and trace (<0.1%) element analysis on the SEM. As a single spectrometer solution, however, it is limited by serial data acquisition and the need for high probe currents. Improvements in detector resolution and quantification software means that Energy Dispersive Spectroscopy (EDS) is now producing useful results in what was previously WD territory (i.e. overlap corrections and the quantification of minor elements with concentrations below 1%). Consequently, multi-element analysis with a single WDS spectrometer is becoming less likely, but overall, it still remains the best technique on the SEM for single element analysis, mapping and separating overlaps and measuring concentration at trace level. The combination of WDS with EDS utilises the strengths of both techniques offering simultaneous fast and accurate quantification down to less than 100ppm. The faster EDS is used for major elements while the minor and trace element analysis is done by WDS. Here we present results showing how combined ED-WD mapping and trace element quantification can be achieved even at high beam currents (>20nA), without any loss in accuracy of results. By introducing a variable collimator on the EDS detector (the Oxford Instruments X-Max+) simultaneous high beam current ED-WD acquisition is possible. Figure 1 shows the combined ED-WD quantification data from a verified glass standard (SRM 621). It contains both light and heavy elements at a variety of concentrations down to 80ppm of Ti. There is also the potential issue of overlaps in the sample between Ba and Ti, separated by 42eV. At these low concentration levels, this overlap would be extremely difficult, if not impossible, to solve with EDS alone. The data was collected using an Oxford Instruments Wave spectrometer at 20kV accelerating voltage and 20nA beam current in conjunction with an Oxford Instruments X-MaxN 150mm2 EDS detector fitted with an X-Max+. Higher beam currents can decrease WD analysis times, but must be balanced against damaging the sample. Minor and trace elements (Mg, S, Ti, Fe and As) were measured with WDS while all remaining elements were analysed with EDS. Five results were averaged and compared to the standard. In all cases, the calculated WD average was within 2 sigma of the published standard concentrations. At 20nA, there was evidence of Na migration within the glass. Figure 2 shows X-ray maps collected at 100nA on an experimental slag sample. In this sample the K lines from important elements (Mn, Fe and Cr) overlap. Without any further software correction, the pure intensity (windows integral) maps fail to separate the Mn K and Cr K contributions, resulting in a very misleading Mn K window map. When AZtec EDS TruMap is used to deconvolute the Microsc. Microanal. 21 (Suppl 3), 2015 1628 spectra at each pixel, the overlap is corrected successfully, but the data is very noisy and difficult to interpret. The WDS map has much less noise and provides an accurate representation of the Mn elemental distribution in this sample. Conclusions Improvements in EDS hardware and software, and the introduction of large area SDD detectors mean that EDS detectors are beginning to rival WDS for results when it comes to the quantification of multi element samples, even for minor elements. For individual elements (i.e. trace element (<0.1 wt%) quantification) and mapping in the presence of major elemental overlaps, however, WDS continues to offer the most accurate results and sensitive detection. Combined ED-WD analysis offers analysts the best combination, both in speed and accuracy and analysing the whole elemental range right down to less than 100ppm. Method Spectrum 1 Spectrum 2 Spectrum 3 Spectrum 4 Spectrum 5 AVERAGE: Sigma STANDARD O EDS 46.0600 46.1162 45.9721 45.7567 46.1661 46.0142 0.0189 46.1000 Na EDS 8.9824 8.8870 8.8007 8.8505 8.8001 8.8641 0.0530 9.4400 Mg WDS 0.1546 0.1531 0.1533 0.1553 0.1551 0.1543 0.0050 0.1600 Al EDS 1.4845 1.4840 1.4459 1.5287 1.4760 1.4838 0.0240 1.4600 Si EDS 33.1345 33.0679 33.0455 33.0079 33.0792 S WDS 0.0375 0.0408 0.0403 0.0385 0.0460 K EDS 1.7571 1.7715 1.7964 1.7230 1.7191 Ca EDS 7.7543 7.7198 7.7463 7.7468 7.7406 7.7416 0.0390 7.6400 Ti WDS 0.0059 0.0040 0.0082 0.0033 0.0107 0.0064 0.0030 0.0080 Fe WDS 0.0436 0.0332 0.0220 0.0305 0.0381 0.0335 0.0060 0.0300 As WDS 0.0282 0.0162 0.0319 0.0253 0.0191 0.0241 0.0040 Ba WDS 0.1122 0.1044 0.1068 0.1104 0.0845 TOTAL 99.5549 99.3980 99.1695 98.9768 99.3346 33.0670 0.0406 1.7534 0.0670 0.0030 0.0240 33.2200 0.0500 1.6700 0.1037 99.2868 0.0150 0.0200 0.1100 99.9080 All measurements in Wt% Figure 1 Combined ED-WD quantification results from a glass standard (SRM 621). The minor and trace elements were quantified by WDS and all others by EDS. The combined total measurement time was 24 minutes. (b) (a) Figure 2 High beam current (100nA) mapping of a experimental slag sample. (a) Spectrum showing the spectral overlap between CrK and MnK (b) Windows integral, TruMaps and WDS maps collected from the sample.");sQ1[813]=new Array("../7337/1629.pdf","Factors Affecting WDS Performance Superiority over EDS","","1629 doi:10.1017/S1431927615008922 Paper No. 0813 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Factors Affecting WDS Performance Superiority over EDS Patrick Camus EDAX Inc., A Division of Ametek, Materials Analysis Division, Mahwah, NJ; USA WDS has traditionally been credited with superior analytic performance compared to EDS. WDS's spectral resolution is far better than EDS, but uses a reduced energy window. WDS's better detection ability was always promoted claiming that the better signal and peak-tobackground (P/B) would always provide superior sensitivity. But WDS only gains that superiority when using high to extreme beam currents. The energy filtering of the spectrometer ignores 99% of the x-ray signal ensuring that the WDS electronics are not overwhelmed. The EDS has no energy filtering and must process all of the x-rays. Traditional SiLi EDS detectors and electronics could not process high input count rates and would become overwhelmed even at medium beam currents. However, modern high-throughput SDD and matched electronics do not have this same limitation and can provide very good spectroscopic performance, in fact rivaling WDS capabilities [1-2]. If EDS capabilities have progressed in general, it is very important to understand under what analysis conditions or energy range(s) WDS retains its dominance. A study is performed to measure spectroscopic performance of a modern SDD EDS system and a WDS system. The data are collected simultaneously on a JEOL JSM-7000 FESEM thus leading to identical operating conditions. EDS spectra were collected for 10 sec livetime and WDS spectra were collected at 1 sec per 1 eV step. Performance metrics that were measured directly from spectra include peak intensity (counts per second per nanoamp, cps/nA), FWHM spectral resolution (eV), and P/B measurements. Spectral resolutions were coarsely measured by counting spectral channels of 10 eV in EDS and 1 eV in WDS. Background counts are a linear interpolation to the peak energy. Beam current was measured using a Faraday cup into a picoammeter. A series of spectra from metals, compounds, and mineral standards were collected that had characteristic x-rays in the operating range of the WDS spectrometer (250 eV � 12 keV). The acquisitions were collected using an SEM beam voltage of 20 kV, beam current of 5 nA, an EDAX Octane 30mm2 EDS detector at full insertion, and a EDAX TEXS WDS spectrometer. The important spectral values were measured from these spectra. Ratio measurements of these values as a function of energy were plotted to compare one spectroscopy to the other. The advantage of using ratios is that the effects of some experimental conditions (for instance beam voltage and x-ray over-voltage) are removed and only spectroscopic measures are compared. One spectroscopic value not removed by using a ratio method is the solid-angle of the EDS detector. The EDS intensity values scale directly with solid angle, which is a function of both detector active area and insertion distance. This outcome can have a significant effect on the determination of the superior spectroscopy. As all WDS spectrometers are focusing, they have a fixed insertion geometry, but the solid angle varies within the spectrometer with other energy dependent terms. This means that in practice, the user has no direct ability to change the solid angle between experiments. Some spectroscopic ratio results are shown in Figures 1-2. Figure 1 shows the EDS/WDS FWHM spectral resolution ratio plotted as a function of energy. The EDS values vary Microsc. Microanal. 21 (Suppl 3), 2015 1630 monotonically with energy but the WDS values vary with angle for each diffractor. The WDS values are constant for SEM operating conditions, but the EDS values depend heavily on the electronics settings which affect the ultimate output count rate; these data were collected using an electronics shaping time which produces a Mn resolution of 130 eV. The ratios are above 1.0 for the whole energy range which shows the spectral resolution superiority of WDS for peak separation, as expected. Figure 2 shows the ratio of the WDS/EDS spectral peak intensity as a function of energy. The individual measurements depend on at least the x-ray over voltage, detector solid angle, and WDS diffractor efficiency at the Bragg angle. The x-ray over-voltage is the same for both techniques and is removed using the ratio of values. The diffractor efficiency produces a very complicated energy functionality. More importantly, ratio values above the threshold of 1.0 show the superior performance of the WDS spectrometer compared to the current EDS detector. Using the current EDS detector geometry, the WDS has superior intensity values only in the energy region below 2 keV while the current EDS detector has superior performance above. It should be noted that an EDS detector of 1/3 the solid angle would use a threshold of 0.3 in this plot for technique comparison. Using this small EDS detector, the superiority of WDS covers a much larger energy range up to 8 keV except for the small energy range of ~2000-3800 eV where EDS is superior. A benefit of the resultant plots is that a WDS spectrum can be modeled or simulated from a simple EDS spectrum. With ratio plots of resolution, intensity and P/B, transformation of an EDS spectrum to a representative WDS spectrum is possible. With this information, a user could determine over which energy ranges EDS or WDS would provide superior analytic capabilities. WDS has maintained its superiority over SDD EDS for spectral resolution. However, the energy dependence of WDS intensity values and the availability of various size SDDs has eroded the sensitivity advantage of WDS. References: [1] N. W.M. Ritchie, J. M. Davis and D. E. Newbury, Microsc Microanal 17 (2011), pp. 556-7. [2] N. W.M. Ritchie, D. E. Newbury and J. M. Davis, Microsc Microanal, 18, (2012), pp. 892904. Figure 1: A plot of EDS/WDS spectral Figure 2: A plot of WDS/EDS peak intensity resolution ratio as a function of peak energy. ratio as a function of peak energy.");sQ1[814]=new Array("../7337/1631.pdf","Enhanced Theoretical Model for Avoiding Mistakes in SEM-EDS Analysis","","1631 doi:10.1017/S1431927615008934 Paper No. 0814 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Enhanced Theoretical Model for Avoiding Mistakes in SEM-EDS Analysis P. J. Statham Oxford Instruments NanoAnalysis, High Wycombe, Bucks HP12 3SE, U.K. Under the right conditions, SEM-EDS can be fast and accurate but how can we tell if results are unrealistic? Most software provides a display to show if fitted peaks and background concur with the measured spectrum but goodness of fit of fitted peaks is insensitive to a number of factors that have a serious effect on the analytical result, for example specimen charging, surface layers, variation in composition within the excited volume and incorrect or missing elements. The oft-used normalisation simply transfers any inaccuracy in results for one element to all others. An un-normalised analytical total that is far from 100% reveals potential problems. When results are normalised for speed and convenience, an alternative "Check Total" can still be used [1] but neither measure points to the cause of a bad total. Many years ago, Duncumb suggested display of a synthesised spectrum might be helpful [2]. The potential to realise this principle came with subsequent development of a theoretical model optimised to match measured ratios of peak intensity to total background (P/Btot) [3]. In any attempt to develop parameterised expressions, a critical factor is the calculation of detector efficiency as a function of energy. Instead of a calculation, an efficiency characteristic has now been measured for a large solid angle SDD detector [4] and data acquired from an extensive set of standard materials at both 5kV and 20kV. The original theoretical model [3] has been enhanced to incorporate differential absorption and excitation of emission lines, including the effect of Coster-Kronig transitions. The parameterisation has been adjusted to fit measured characteristic intensities and both background shape and intensity at the same beam current. The theoretical spectrum for any analysis result can now be overlaid on the original spectrum and is sufficiently accurate to provide useful diagnostics. Fig.1 shows a spectrum of Nickel-based superalloy where the overlaid results of peak and background fitting are good. The detail in Fig.1(b) shows a small Zn K peak that has been chosen in preference to Re L. The analysis total in Table 1 is only slightly high, but the theoretical spectrum in Fig.2 shows that if Zn were present at 0.92%, there should be a Zn L peak at 1keV. When Zn and Sr are removed and replaced with Re, Fig.3 shows much better agreement with the theoretical spectrum and the analytical total in Table 2 is also improved. At low kV, some lines may not even be excited so that analysis results after normalisation can conceal gross errors. In Fig.4, a 4kV spectrum of Quartz (SiO2) gives a result where both the fitted spectrum and theoretical spectrum line up with the measurement. The spectrum of Fig.5 has similar well-fitted element peaks to Fig.4 and gives result O: 56% Si: 44% but when these values are used to calculate the theoretical spectrum, the inconsistent peak heights warn of a serious problem. The analysis is wrong because at 4kV the Ca K peak from Wollastonite(CaSiO3) is not even excited so Ca cannot be measured. In such cases, immediate access to the theoretical spectrum provides additional intelligence to judge if the analysis is valid and thus avoid mistakes and bad conclusions. References [1] P.J.Statham (2004) Microchim. Acta 145, 229-235 [2] P.Duncumb (1994) Mikrochim. Acta 114/115, 3-20 [3] P.Duncumb, I.R.Barkshire, P.J.Statham (2001) Microsc. Microanal 7, 341-355 [4] P.J.Statham (2010) Microscopy and Microanalysis 16 (Suppl 2), 1304-1305 Microsc. Microanal. 21 (Suppl 3), 2015 1632 FIG. 1(a) Spectrum from Nickel superalloy (yellow bars) overlaid with result of spectrum fitting (pink). Elmt Al S Ti Cr Co Ni Zn Sr Ta W Total Wt% 5.08 0.25 0.70 6.22 9.62 62.42 0.92 2.86 4.73 9.30 102.10 FIG. 1(b) Region showing small Zn peak Elmt Al S Ti Cr Co Ni Ta W Re Total Wt% 5.12 0.23 0.7 6.22 9.61 62.35 4.64 8.8 2.37 100.04 Table 1: Analysis result FIG. 2 Theoretical FIG. 3 Theoretical spectrum for Table 1 spectrum with Re instead composition overlaid on of Zn and Sr original spectrum. Note differences near ZnL, TaM Table 2: Results used for fig.3 FIG. 4. SiO2 4kV spectrum overlaid with fitted (pink) and theoretical spectrum (blue) FIG. 5. CaSiO3 4kV spectrum overlaid with fitted (pink) and theoretical spectrum (blue)");sQ1[815]=new Array("../7337/1633.pdf","Quantitative Analysis of Heterogenous Samples by SEM/EDS","","1633 doi:10.1017/S1431927615008946 Paper No. 0815 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantitative Analysis of Heterogenous Samples by SEM/EDS John Konopka Thermo Fisher Scientific It is well established that samples for quantitative analysis by SEM/EDS must be flat, conductive and homogeneous [1]. However, many analysts are often forced to estimate the composition of a complex sample and their only tool is the standardless quantitative analysis button. Also, one hears that a spectrum from a large field of view produces an average result. The point of this work is to document this problem and offer some suggestions and guidance. The algorithms for quantitative corrections assume a sample with a uniform composition. If the element distribution is not uniform then the wrong correction factors are calculated and applied resulting in a wrong result. Sampling errors arise as there may be subsurface objects not seen at the analyzed surface. Also, the distribution of materials may change depending on the area viewed causing the result to change depending on the field of view. There is no generic way to account for unseen or unanalyzed portions of the sample short of slicing the sample and fully analyzing each slice or milling the sample and pressing a pellet for analysis. A suggestion is to analyze a number of areas and then examine the variance of the results. If the results are not consistent then clearly this is a problem sample. A second suggestion is to parse the sample view into regions of homogenous composition and then quantify each area. This result is legitimate as each analysis of a uniform area adheres to the assumptions of the algorithms. Area fractions do not easily extend to bulk composition but at least this result is honest. Non-homogeneous samples were analyzed with a JEOL JSM-7001F FESEM and a Thermo Scientific NS7 EDS analyzer equipped with an UltraDry silicon drift detector. All measurements were conducted at 15 kV accelerating voltage. Corrections were by standardless Phi Rho Z and results are reported as weight percents. A steel standard was analyzed at seven different locations. The results are shown in Table 1. The results did not agree closely with the known standard composition. A meteorite sample was mapped with Spectral Imaging and the results were parsed with COMPASS2 and XPhase. The resulting nine phases were quantified by standardless analysis and the overall spectrum was quantified. The results are shown in Table 2. If forced to characterize a complex sample by SEM/EDS one should, at a minimum, analyze multiple areas to check for consistency. Second, parse the sample and report Microsc. Microanal. 21 (Suppl 3), 2015 1634 the constituents as area fractions. When forced to tackle a challenging sample be honest and report the manner in which the results were acquired. Do not report a result from a single field of view as an "average" result. [1] Beaman, D. R., Isasi, J. A., ASTM Special Tech. Publ. 506, American Society for Testing and Materials, Philadelphia, Pa., 1972, p. 51. [2] Advanced Materials and Processes, September 2002, p. 17. Si min max avg sd std val 0.9 1.0 1.0 0.01 0.84 S 0.2 1.9 0.7 0.55 0.33 Cr 13.9 14.8 14.2 0.31 Mn 0.0 1.8 0.8 0.55 Fe 79.7 83.7 82.5 1.40 84.1 Ni 0.3 0.4 0.4 0.03 0.42 Cu 0.1 0.1 0.1 0.01 0.10 Mo 0.2 0.4 0.3 0.05 0.20 13.28 0.41 Figure 1. Image of SS 416 steel sample Table 1. Summation of analyses of Brammer standard SS 416 at seven locations. Due to the variation in the density of inclusions from one view to the next the variance in the results is quite high. Note that the Mn result ranges from 0.0 to over four times the expected result. Results are weight percents. Figure 2. Image of meteorite sample. OK Phase 1 Phase 2 Phase 3 Phase 4 Phase 5 Phase 6 Phase 7 Phase 8 Phase 9 Field of View 41.9 8.8 27.0 2.6 44.0 35.7 39.7 4.4 45.9 22.1 Na K 0.0 0.0 0.0 0.0 5.4 0.0 0.0 0.0 0.0 0.0 Mg K Al K 15.9 1.4 1.7 0.6 4.4 4.6 20.7 0.8 10.1 15.2 0.0 0.1 2.5 0.2 7.1 0.5 0.1 0.0 1.1 0.4 Si K 27.0 1.8 0.9 0.4 28.3 7.8 17.4 1.0 23.9 20.4 SK 0.0 31.6 0.8 0.0 0.0 5.5 0.0 5.6 0.0 7.2 Ca K 0.6 0.0 0.0 0.7 2.6 0.5 0.0 0.0 10.4 0.6 Ti K 0.0 0.0 1.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Cr K 0.3 0.0 39.0 0.0 0.0 0.0 0.3 0.0 0.0 2.3 Mn K Fe K 0.4 0.0 0.0 0.0 0.0 0.0 0.3 0.0 0.0 0.4 14.0 55.2 26.5 79.0 7.6 30.1 21.4 51.2 8.6 29.2 Ni K 0.0 1.0 0.0 16.4 0.7 15.3 0.0 37.2 0.0 2.2 Area 0.38 0.16 0.04 0.03 0.05 0.02 0.29 0.01 0.02 Table 2. Weight percent results. Individual phases vs. one spectrum covering the field of view.");sQ1[816]=new Array("../7337/1635.pdf","Trace Element Analysis under 100 ppm and Chemical State Analysis in Small Area using Wavelength Dispersive Soft X-ray Emission Spectrometer in FE-SEM","","1635 doi:10.1017/S1431927615008958 Paper No. 0816 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Trace Element Analysis under 100 ppm and Chemical State Analysis in Small Area using Wavelength Dispersive Soft X-ray Emission Spectrometer in FE-SEM H. Takahashi1, S. Asahina2, 1. M. Terauchi3 , C. Nielsen4, P. McSwiggen4 Global business promotion division, JEOL Ltd., 13F, Otemachi Nomura Bld. 2-1-1 Otemachi, Chiyoda-ku, Tokyo, 100-0004 Japan 2. SM business unit JEOL Ltd., 1-2 Musashino, 3-chome, Akishima, Tokyo 196-8558, Japan. 3. Institute for Multidisciplinary Research for Advanced Materials, Tohoku University, Sendai 980-8577, Japan. 4 . JEOL USA, 11 Deaborn Rd. Peabody, MA 01960, ISA . A novel wavelength dispersive soft X-ray emission spectrometer (WD-SXES) has been developed. It covers nominally the X-ray energy range between 50 and 210 eV [1, 2] using two kinds of gratings. One of the characteristic features of the WD-SXES is its parallel detection of the signal. This allows it to be used like a conventional energy dispersive spectrometer. A second feature is its high energy resolution, which is about 0.2 eV for Al-L emission. This resolution is comparable to that of X-ray photoelectron spectroscopy or electron energy-loss spectroscopy. This enables us the ability to obtain important information about chemical bonding in bulk samples from observed spectra as a result of the very high- energy resolution. We have already reported a few examples obtained with the WD-SXES [3]. One important feature of the WD-SXES is that it can detect the Li-K emission spectrum. In the case of an anode electrode from a lithium ion battery (LIB), two types of lithium peaks are observed; one lower energy peak at 50eV, from the valence band, and the other higher energy peak at 54 eV, from the core loss. In addition, we have documented the ability to measure various kinds of X-ray spectra of K, L, M and N emission from Lithium to Uranium [4]. In the modern FE-SEM, a wide variety of attachments are used, including multi-SDD, EBSD, WDS, many kinds of electron signal detectors, sputtering tools, low vacuum devices, cathodeluminescence detectors, etc. These are very important for characterizing a wide range of materials. Now, due to its good peak to background ratio and high sensitive CCD camera, the WD-SXES would be an ideal attachment for trace element analysis. Combining the WD-SXES with a FE-SEM, the higher spatial resolution and high sensitivity can provide new insights in material science. For example, the SEM is very useful for observing the distribution of microstructures for metals and ceramics. We have also demonstrated that trace carbon contents can be detected in concentrations from 67 ppm to 8700 ppm using low alloy steel NBS (NIST) series from 1261a to 1265a respectively. Backscattered electron images of these materials, along with their trace carbon peaks are shown in Figure 1 (a) and (b). In addition, from the peak shape of the C-Kspectra, it is possible to distinguish a carbide grain from graphite, as shown in Figure 2. Using a FE-SEM, we can observe tiny grains or grain boundaries, and by combining it with the WD-SXES, we can analyze these features for trace carbon abundances. It is also possible to analyze for trace boron and nitrogen. In this presentation, we will discuss the possibilities for using the WD-SXES for trace elemental analyses and chemical state analyses. References: [1] M. Terauchi, et al., J. Electron Microscopy, 61, 1 (2012). [2] T. Imazono, et al., Appl. Opt. 51, 2351 (2012). [3] H. Takahashi, et al., Microscopy and Microanalysis, 19 (supple. 2), 1258 (2013). [4] H. Takahashi, et al., Microscopy and Microanalysis, 20 (supple. 3), 684 (2014) Microsc. Microanal. 21 (Suppl 3), 2015 1636 1626 (a) (b) Figure 1. Backscattered electron images and C-K peaks using JSM-7800FPrime+SXES for (a) 8700 ppm and (b) 67 ppm carbon in low alloy steel NBS (NIST) 1264a and 1265a Figure 2. Comparison of C-K spectra between Fe3C (Solid blue line) and graphite (Dash green line)");sQ1[817]=new Array("../7337/1637.pdf","Characterization and Comparison of Detector Systems for Large Area X-ray Imaging","","1637 doi:10.1017/S143192761500896X Paper No. 0817 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization and Comparison of Detector Systems for Large Area X-ray Imaging Jeffrey M. Davis1,* , Julia Schmidt2 , Martin Huth2 , Sebastian Ihle2 , Daniel Steigenh�fer2 , Peter Holl2 , Gerhard Lutz2 , Udo Weber1 , Adrian Niculae1 , Heike Soltau1 , Lothar Str�der2 1. 2. PNDetector GmbH, Otto-Hahn-Ring 6, 81739 M�nchen, Germany PNSensor GmbH, Otto-Hahn-Ring 6, 81739 M�nchen, Germany In the last 15 years, there has been a major shift in the type of detector used in X-ray microanalysis [1]. The silicon drift detector (SDD) is now the predominant energy dispersive X-ray spectrometer, having replaced the lithium drifted silicon detector. With high spectral resolution at input count rates over 100 000 counts/s, an SDD reduces the total time required to record a high quality X-ray image. In 2005, a multi-element SDD was also introduced, which combined multiple SDDs into a single detector [2]. These detectors combine large solid angles (for more efficient X-ray collection) with fast spectral processing through multiple pulse processors. Yet its primary drawback is that the SDD does not record the position of each X-ray event in addition to the energy. An array of SDDs could do this, but the position resolution would be poor. Otherwise, an X-ray scintillator combined with an ordinary charge coupled device (CCD) can record the location of the X-ray event, but the energy resolution is usually poor. An ideal imaging spectrometer would be one that can record both position and energy. In 2011, PNSensor introduced an X-ray pnCCD camera, an energy dispersive X-ray spectrometer developed in collaboration with PNDetector. The pnCCD is a unique CCD that combines the excellent position resolution of the CCD with the energy resolution of an SDD. The pnCCD is able to collect X-rays at high count rates with the spectral resolution of every pixel on par with high performing SDDs, because it records both the position of the X-ray event and the energy. This spectroscopic imaging capability opens up new opportunities for X-ray imaging and analysis, especially for large samples analyzed in the �XRF. Numerous experiments have been performed to test and quantify the spectroscopic qualities of the detector as they relate to high precision X-ray microanalysis. Functioning purely as a spectrometer, the pnCCD can count at a rate of 200 000 counts/s (or more) with a resolution of approximately 150 eV at Mn K-alpha. However, when a polycapillary optic is placed in front of the pnCCD to focus the incoming X-rays, the pnCCD can be used to produce X-ray images. This is particularly useful in �XRF where the pnCCD eliminates the requirement of scanning the sample under the X-ray beam by recording all 70 000 spectra at once. The imaging area of the pnCCD is 1.2 cm x 1.2 cm, with a pixel size of 48 �m (sub pixel resolution down to 10 �m possible). Although the electronics and data processing are different from the SDD, the end result of an experiment performed with the pnCCD is an X-ray Spectrum Image (a full X-ray spectrum stored at each pixel). As a large area imaging spectrometer, the pnCCD has proved extremely useful in diverse fields such as cultural heritage conservation and materials science, where efficient, large area imaging is necessary [4]. This work is primarily concerned with the advantages and opportunities provided by the pnCCD when used as an imaging spectrometer in a �XRF system. As shown in Figure 1, the spectral performance of the detector, while counting at a rate of over 200 000 counts/s is easily comparable to an SDD counting at a rate of approximately 20 000 counts/s. Figure 2 shows an image produced with the pnCCD. The image was collected in approximately 600 s, and it covers an area of 1.4 cm2 . To get the same data from an SDD based system, the stages in a conventional �XRF would have to move at a rate of 5 m/s Microsc. Microanal. 21 (Suppl 3), 2015 1638 and count at a rate of 115 000 counts/s. References: [1] Str�der, L., Fiorini, C., Gatti, E., et al, High resolution non dispersive x-ray spectroscopy with state of the art silicon detectors, Mikrochimica Acta, Supplement 15 (1998) 11 [2] Longoni, A., Fiorini, C., Guazzoni, C., et al, A novel high-resolution XRF spectrometer for elemental mapping based on a monolithic array of silicon drift detectors and on a polycapillary x-ray lens, X-ray Spectrometry, 34 (2005), p. 439:445. [3] Scharf, O., Ihle, S., Ordavo, I., et al, Compact pnCCD-nased X-ray camera with high spatial and energy resolution: A color X-ray camera, Analytical Chemistry, 83 (2011), p. 2532:2538. [4] Romano, F.P., Caliri, C, Cosentino, L., et al, Macro and micro full field X-ray fluorescence with an X-ray pinhole camera presenting high energy and high spatial resolution, Analytical Chemistry, 86 (2014), p.10892:10899. Figure 1: Spectral comparison of a spectrum from an SDD (black line) and a spectrum made with a pnCCD (red line). The pnCCD spectrum has an equivalent resolution of 152 eV at Mn K-alpha The performance of the spectrometer at a count rate of over 200 000 counts/s is very good for imaging purposes. Figure 2: The X-ray image of a section of an IBM circuit board has dimensions of 1.2 cm x 1.2 cm, and it was acquired in 600 s. Yet this image was produced with the pnCCD without moving the stage, fully resolving spatial features, such as the wires in the center, which are approximately 150 �m in diameter. Here the Cu X-rays are colored in red, the Au X-rays are in blue and the Ni X-rays are in green. The scale bar represents 1.4 mm on the image.");sQ1[818]=new Array("../7337/1639.pdf","Accurate EPMA Quantification of the First Series Transition Metals using Ll Lines","","1639 doi:10.1017/S1431927615008971 Paper No. 0818 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Accurate EPMA Quantification of the First Series Transition Metals using Ll Lines M. Moret, C. Hombourger, and M. Outrequin CAMECA, 29 Quai des Gr�sillons, 92622 Gennevilliers Cedex, France mona.moret@ametek.com In conventional EPMA, X-ray K lines are used for accurate quantification of the transition metal elements. However, at low beam accelerating voltages (i.e., < 5 keV), only low-energy X-ray lines are emitted, including K lines for Z 22 and L and M lines for Z > 22. Low beam energy operation offers several advantages: improvement of analytical spatial resolution and reduction of both secondary fluorescence and sample charging effects. The use of lowenergy X-ray lines for quantitative analysis does present new analytical challenges, though, because these lines are subject to larger peak shifts, more line overlaps and lower fluorescence yields, as compared to higher-energy K lines. The low yields also reduce the intensity of certain lines, as does the low overvoltage, U (defined as the ratio of beam energy to ionization energy for a given line), which lowers the ionization probability for the X-ray line. If we consider the example of Fe (Z=26) analyzed at 15 keV, many characteristic X-ray lines (K- and L-series) are produced from the atom. The X-ray lines traditionally used for quantification are the K line (transition from L3 sub-shell to K sub-shell) and L line (transition from M5 sub-shell to L3 sub-shell). Generation of the Fe L line involves valence electrons, which are affected by the chemical bonding of Fe in the target sample. Wavelength shifts, peak shape modifications, and increases or decreases in the relative intensities of the characteristic lines can be readily observed between different chemical types. In general, for elements of the first transition metal series (Sc to Zn), the pattern of L emission spectra varies according to the energy of the incident electrons (E0), as follows: E0 between the L3 and L2 threshold energies: intensity of the high energy satellite lines is reduced. E0 from the L3 sub-shell threshold energy up to 3 times this energy: excitation and development of the high energy satellites increases and the shape of the L line becomes increasingly distorted. E0 above 3 and 4 times the L3 sub-shell threshold energy: the absorption path of the generated X-rays increases and thus the fine structure on the high energy side "disappears" due to selfabsorption. From these spectroscopic observations, it becomes obvious that the shape and the peak position of the L emission band is strongly dependent on the incident electron energy and thus can lead to inaccuracies in quantification when using L lines. As an alternative, Fialin et al. [1] have suggested the use of the Ll line, instead of the more commonly used L or L lines. The Ll line has the advantage of being independent of the chemical state of the element, since the electrons of the outer shell involved in the Ll transition are core, not valence, electrons. One drawback is the low intensity of Ll, compared to the L line (typically 10 times less for Fe measured with a TAP crystal). Quantitative results achieved by using the Ll lines for the elements of the first transition series will be presented. � � � Microsc. Microanal. 21 (Suppl 3), 2015 1640 [ 1] Fialin M, Wagner C and Remond G, 1998 in: Proceedings, EMAS'98, 3rd Regional Workshop. Figure 1: Energy level diagram showing electron transitions producing Fe K and L X-rays.");sQ1[819]=new Array("../7337/1641.pdf","Multilayer Thin Films as Pseudo-Homogeneous EDX Standards","","1641 doi:10.1017/S1431927615008983 Paper No. 0819 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Multilayer Thin Films as Pseudo-Homogeneous EDX Standards Hendrik O. Colijn1, Jonathan Orsborn1, Daniel Chmielewski2, and David W. McComb1. 1. 2. Ohio State University, Center for Electron Microscopy & Analysis, Columbus, OH, USA. Ohio State University, Electrical & Computer Engineering, Columbus, OH, USA. Quantifying EDX data in the TEM/STEM is generally accomplished using either the Zeta-factor or the Cliff-Lorimer method (k-factors) [1]. The Zeta-factor method requires accurate measurement of the beam current which can be difficult to measure accurately on most microscopes. Consequently the Zeta-factor approach is not frequently used. The Cliff-Lorimer method is much more commonly used but requires accurate k-factors in the analysis. While theoretical k-factors are available there can easily be large variations in the values of the k-factor depending on the theoretical model chosen. These variations can lead to errors in the final composition values. Experimentally determined k-factors are generally more reliable for doing the analysis but require the availability of suitable experimental standards. Suitable standards can be difficult to obtain and issues such as inhomogeneities and surface films due to oxidation etc. can complicate the analyses. Advances in thin film growth techniques now allow the deposition of nanometer scale uniform thin films of known composition. High quality thin films can be grown by a number of techniques including Atomic Layer Deposition (ALD), Radio Frequency (RF) sputtering, and Molecular Beam Epitaxy (MBE). If the sample has sufficient layers, one can achieve a reasonable approximation of a homogeneous material. This does require that the exact thickness and composition of each layer be well known. Cross-section imaging of the multi-layers allows accurate thickness measurements to be made. In many cases, one can very accurately determine the layer thicknesses by counting the number of atomic planes. We have used multi-layer standards to measure k-factors for several elemental systems. We verified the layer thicknesses by preparing cross-sections using a FIB with subsequent TEM analysis. Multiple spectra were then taken at various thicknesses of the material. We employed several software packages such as FEI TIA and EDAX TEAM to determine the peak intensities. The intensity ratios from both packages were similar. A regression fit on these intensity ratios was used to extrapolate them to zero thickness using the van Capellen method [2]. The regression fit also provides an estimate of the uncertainly in the y-intercept value which gives an estimate of the bias error in the Cliff-Lorimer quantification method. Figure 1 shows a STEM image of an MBE grown multilayer of AlGaAs and InGaP used in this experiment. The layer thicknesses were confirmed by HRTEM. Figure 2 shows the intensity ratios for several elements that were in the multilayer film and the regression fit to find the zero-thickness intensity ratio. Using the layer thicknesses and the layer composition we calculated the average composition of the multilayer. The kab values were then determined by using the average composition and the y-intercept of the intensity ratio. [3] References [1] see e.g. D.B. Williams & C.B. Carter, Transmission Electron Microscopy: A Textbook for Materials Science, (Springer), 2009, chapter 35. [2] E. van Capellen, Microsc. Microanal. Microstruct. 1, (1990), pp. 1-22. Microsc. Microanal. 21 (Suppl 3), 2015 1642 [3] The authors acknowledge support from The Ohio State University and the Ohio Third Frontier Research Scholar program. 5nm In 0.48 0.3 Ga 0.52 P 5nm Al Ga As 0.7 10 cycles GaAs Figure 1. HAADF STEM image of MBE grown multilayer film and schematic of the layers. 1.0 0.8 0.6 0.4 0.2 0.0 00000 0.250 Al/In 0.200 0.150 0.100 0.050 Al/Ga 40000 80000 120000 160000 0.000 0 100000 200000 300000 1.000 0.900 0.800 0.700 0.600 0.500 0 200000 400000 600000 Wt. ratio Al/Ga Al/In As/Ga 0.096 0.096 0.884 0.366 0.657 Int. ratio 0.180 0.417 0.679 0.654 0.422 k-factor 0.531 (0.019) 0.229 (0.013) 1.303 (0.037) 0.559 (0.025) 1.557 (0.070) As/Ga P/Ga In/Ga Figure 2. Typical intensity ratio plots with regression fit and calculated zero thickness k-factors with the 1 uncertainty in parentheses.");sQ1[820]=new Array("../7337/1643.pdf","A Method of Component Extraction of EDS and EELS maps","","1643 doi:10.1017/S1431927615008995 Paper No. 0820 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A Method of Component Extraction of EDS and EELS maps Shixin Wang Micron Technology, Inc., 8000 S. Federal Way, Boise, ID 83707, USA In microanalysis, we generate elemental maps from analytical data, such as EDS and EELS, to show the spatial distribution of elements of interest. In either EDS or EELS, the intensity of a map pixel may have contributions from multiple components. For example, overlapping signals, x-ray fluorescence (for EDS), or spatial mixture can form mixture maps. To obtain a map for each individual component, we need to find a way to calculate each individual contribution. A commonly used method to separate overlapping signals is to deconvolute the source signal. For situation of x-ray fluorescence or spatial mixture, spectral deconvolution is not applicable. Here we demonstrate a component extraction method on EDS and EELS maps. Let I(x,y) be the intensity function of a map. Suppose I(x,y) has contributions from two components, A(x,y), and B(x,y), from signal a and b, respectively. Mapping is obtained through integration of a signal over a fixed window and integration is a linear operation, we thus have, I(x,y) = A(x,y) + B(x,y). If we know B(x,y), we get a pure A(x,y) through a simple subtraction, I(x,y) � B(x,y). The way to find B(x,y) is to form a map C(x,y) over the same region. C(x,y) is a map generated from signal c. The criteria of choosing c are: (1) the map C(x,y) has a region free of contribution of a, (2) c has no signal overlap with a, and (3) the intensity of c is linearly proportional to b. With C(x,y), we can strip the contribution of b, and obtaining the A(x,y) through the following equation, A(x,y)= I(x,y) �k�C(x,y) (1) Where, k is a constant independent of spatial coordinates and its value is B(x,y)/C(x,y). A(x,y) can be explicably solved since I(x,y) and C(x,y) are simply the pixel reading of maps. The key is finding the constant k. This is through the region free of contribution of a, in which B(x,y)/C(x,y) is a constant. By comparing the pixel intensity ratio, we compute the value k. With the value k being a constant, eq. (1) is simply image subtraction. For EDS or EELS, it is always possible to find such a component like c. For example, Fig. 1 (A) is "Si-k map". Because Si-k peak overlaps with W-M1 peak, the map contains W component. From examination of local EDS spectrum, we know the vertical bright structure is pure W. Thus we pick a W-L1 signal to be our c (Fig. 1B). The ratio k is found by comparing pixel intensities of W-L1 map and "Si-k map" at a Si-free region (such as the area marked by the box in Fig. 1(B). If the component c is not readily available on the sample, we can put a piece of material containing c onto the sample by FIB manipulation. There are two methods in choosing c: (1) c can be a different signal, such as different EDS peaks, EELS edges, or from different type of detector, of the same element; (2) c can be from a different element of a chemical compound. The process described by Eq. 1 can be readily extended to maps with more than two components. Fig. 2 shows the result of pure component maps involving signal overlaps and x-ray fluorescence. With iterative application of Eq. 1, a clean spatial distribution of each individual element is obtained (Fig. 2B). Through computer programming, the calculation of Eq. 1 can be easily automated. Fig. 3 shows a program written with Gatan� DigitalMicrograph scripting language. When a user selects a region for c, a subtraction image is formed by I � k�C, with k value selected such that the average intensity of the same selected area in the subtraction image is zero. The application of this method is limited to thin samples, when b:c linearity requirement is satisfied. For thick sample, this method losses its accuracy and eventually becomes not applicable when the linearity between b and c does not hold due to absorption. Microsc. Microanal. 21 (Suppl 3), 2015 1644 Fig. 1. EDS maps. (A) Si-K map, contaminated by W-M1 signal. (B) W-L1 map. (C) Si map stripped of W component. Fig. 2. EDS maps. (A) Original unprocessed maps. (B) pure component maps. (C) color overlays of true elemental maps. Fig. 3. A computer program to carry out the map component extractions interactively.");sQ1[821]=new Array("../7337/1645.pdf","Windowless, Silicon Nitride window and Polymer window EDS detectors: Changes in Sensitivity and Detectable Limits","","1645 doi:10.1017/S1431927615009009 Paper No. 0821 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Windowless, Silicon Nitride window and Polymer window EDS detectors: Changes in Sensitivity and Detectable Limits J. Rafaelsen1, T.Nylese1, M. Bolorizadeh1, V. Carlino2 1 EDAX Inc, A division of Ametek, Materials Analysis Division, 91 McKee Drive, Mahwah, NJ 07430 USA 2 ibss Group INC, 1559B Sloat Blvs., Suite 270, San Francisco, Ca 94132 USA With the introduction of Silicon Drift Detectors (SDD) and the development of fast and low noise pulse processors, Energy Dispersive Spectroscopy (EDS) analysis has seen remarkable increases in throughput and reliability in the last decade. But one often overlooked aspect of the detection technology is the x-ray window. While windowless detectors are becoming the standard for Transmission Electron Microscopes (TEM), they are rarely used on Scanning Electron Microscopes (SEM), primarily due to the risk of detector contamination when venting the vacuum chamber. By adding a sealed window in front of the detector module it is possible to keep the detector cooled and under vacuum at all times, eliminating the risk of detector contamination during venting cycles. A variety of window technologies are available including Beryllium, polymer films, and the most recent addition, silicon nitride. The different window materials will have an effect on the measured spectrum due to the x-ray absorption in the window, but also the support grid for the window influences the spectra. Figure 1 shows an SEM image of the window structure for silicon nitride and polymer windows. The hexagonal pattern of the support grid for the silicon nitride window covers roughly 18% of the total area while the venetian blind support structure for the polymer window covers roughly 23% of the total area. In this work, we characterize the differences between silicon nitride and polymer windows and compare those to the performance of a windowless detector. To this end we have collected a series of spectra from a sample block containing traceable standards using a windowless detector. We then mounted a cap in front of the electron trap containing either a silicon nitride or polymer window. By using a windowless detector and exchanging windows, we can use the same detector for all measurements and thus ensure that the results are not affected by detector characteristics. Though the addition of a cap containing a window will work as a limiting aperture and reduce the solid angle of the detector, comparable statics between windows were ensured by acquiring the same number of counts in the spectra. An example of the spectra can be seen in Figure 2. In order to quantify the differences between the windows, we calculated the Minimum Detection Limit (MDL) for spectra acquired from the same samples using the three different detector/window configurations. While there are several models for calculating the MDL (1) (2), we applied the criteria the peak height should be higher than the background with a 95.45% confidence level. Plasma cleaners have become common accessories to modern SEMs as they help reduce sample and chamber contamination, and they have even been shown to improve light element sensitivity in EDS measurements (3). However, there is still debate about to which extent cleaning is compatible with x-ray windows. We will present, to our knowledge, the first comparison Microsc. Microanal. 21 (Suppl 3), 2015 1646 between plasma cleaning of polymer and silicon nitride windows. The plasma cleaning experiments were conducted with a GV10x Asher from ibss Group Inc. in a dedicated chamber. Figure 1: SEM images showing the support grids for silicon nitride window (left) and polymer window (right). Figure 2: Si spectra for polymer and silicon nitride window. References 1. Precision and Sensitivity in Electron Microprobe Analysis. Ziebold, Thomas O. 8, 1967, Analytical Chemistry, Vol. 39, pp. 858-861. 2. Goldstein, et al. Scanning Electron Microscopy and X-Ray Microanalysis. s.l. : Plenum Press, New York, 1984. ISBN 0-306-40768-X. 3. Improved Carbon Analysis with Evactron Plasma Cleaning. Rolland, P., Carlino, V. and Vane, R. . S02, s.l. : Microscopy Society of America, 2004, Microscopy and Microanalysis, Vol. 10, pp. 964-965x.");sQ1[822]=new Array("../7337/1647.pdf","Transmission Electron Microscopic and First-principles Study of SrTiO3/GaAs Hetero-interfaces","","1647 doi:10.1017/S1431927615009010 Paper No. 0822 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Transmission Electron Microscopic and First-principles Study of SrTiO3/GaAs Hetero-interfaces Liang Hong, Serdar ��t and Robert Klie Department of Physics, University of Illinois at Chicago, Chicago, IL 60607 This SrTiO3/GaAs interface is primarily being studied due to the interest of using it in metal-oxidesemiconductor field-effect transistors (MOSFETs), where the GaAs substrate would act as the semiinsulating base material and the SrTiO3 would act as the barrier oxide layer between the GaAs and the gate material. Previously, SrTiO3 ultra-thin films on As-terminated GaAs substrate have been studied experimentally [1-3]. In this work, various configurations of the SrTiO3/GaAs interface are considered and characterized using transmission electron microscope (TEM), atomic-resolution Z-contrast imaging and Energy-dispersive X-ray spectroscopy (XEDS), along with first-principles-based density functional theory (DFT) calculations, to study the energetically most favorable configuration and obtain a deep understanding of the structural and electronic properties. The sample used in this work is grown using molecular beam epitaxy (MBE) method. A 3 nm SrTiO3 thin film is grown on Si substrate, and covered by a 1 �m thick GaAs layer. The cross-sectional atomicresolution Z-contrast image (Figure 1) of the SrTiO3/GaAs interface is obtained using JEOL JEM-3010 TEM at 300 kV. The chemical components are characterized by XEDS. In Figure 1, we can see the typical dumbbell-structure of the GaAs [110] and the SrTiO3 [100] thin film. A higher resolution Zcontrast image will be obtained from JEOL ARM200CF scanning transmission electron microscopy (STEM) along with XEDS mapping to know the atomic structure and chemical components of the interface in the future work. First-principles calculations are carried out within the framework of DFT using the projector augmented wave (PAW) method as implemented in the Vienna Ab initio Simulation Package (VASP) code with the generalized gradient approximation (GGA) in the scheme proposed by Perdew-Burke�Ernzerhof (PBE). The optimized lattice constant of bulk GaAs (Zinc-blende structure) and SrTiO3 (perovskite structure) are 5.76 � and 3. 94 �, respectively. For the thin SrTiO3 film on GaAs substrate, the lattice constant of SrTiO3 is set as 4.07 � to match the GaAs lattice parameters. Double-slab structural model with an 8 � vacuum in the middle is used to study the SrTiO3/GaAs hetero-interface (shown in Figure 2). With regard to stoichiometric and non-stoichiometric SrTiO3 terminated with either SrO or TiO2 layer in contact with either Ga or As surface, as well as different bonding directions, 32 different configurations of the SrTiO3/GaAs interface are considered. The formation energy of the interfacial configurations is calculated through the equation ! = !"#$ - !" !" - !" !" - !"# !"# - !"#$ !"#$ . The chemical potential for each component in the equation is constrained by their bulk chemical potentials correspondingly. By comparing the formation energy, non-stoichiometric SrTiO3 sandwiched by SrO layers are found to be favored in both Ga and As terminated situations. A phase diagram of formation energy comparison for all the interfacial configurations is shown in Figure 3. Nonstoichiometric Ga/SrO interface is found to be energetically most favorable since it occupies the largest area in the phase diagram. Microsc. Microanal. 21 (Suppl 3), 2015 1648 References: [1] R. F. Klie et al, Appl. Phys. Lett. 87 (2005) 143106. [2] Q. Qiao et al, Phys. Rev. B. 85 (2012) 165406. [3] Y. Liang et al, Appl. Phys. Lett. 85 (2004) 1217. [4] This work was supported by the National Science Foundation (Grant No. DMR-1408427). (a) (b) Figure 1 (a) Atomic-resolution Z-contrast image of the SrTiO3/GaAs hetero-interface. The GaAs is seen in [110] projection while SrTiO3 is in [100] projection. (b) Diffraction pattern on this condition. Figure 2 Double-slab structural model used in the DFT calculations. A non-stoichiometric SrTiO3 terminated with SrO monolayer on the Ga-terminated GaAs substrate is shown in the figure. Figure 3 Phase diagram of the formation energy for the interfacial configurations. GSS, GTS, GTT represent the non-stoichiometric Ga/SrO, stoichiometric Ga/TiO2 and non-stoichiometric Ga/TiO2 interfaces, respectively.");sQ1[823]=new Array("../7337/1649.pdf","Observation of Skyrmions at Room-temperature in Amorphous Fe/Gd Films","","1649 doi:10.1017/S1431927615009022 Paper No. 0823 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Observation of Skyrmions at Room-temperature in Amorphous Fe/Gd Films Jordan Chess1, Sergio Montoya2, James Lee1,3, Sujoy Roy3, Steven Kevan1,3, Eric Fullerton2, Ben McMorran1 1. Department of Physics, University of Oregon, Eugene 2. Center for Magnetic Recording Research, University of California, San Diego 3. Lawrence Berkeley National Laboratory We have employed Lorentz TEM (LTEM) and focal series reconstruction at multiple sample tilt angles and applied fields to investigate magnetic structures in amorphous Fe/Gd thin films. Magnetic bubbles, investigated in the 1970s as potential data storage media1, have seen a recent surge in scientific inquiry due to the their topological spin textures [1]. A specific type of bubble with whirling magnetic spins (see Fig. 1) called a skyrmion is the major driving force behind these investigations. Skyrmions are characterized by a non-zero chirality or topological charge, given by = ( � ), where m is the normalized in-plane magnetization = (, )/|(, )|. These spin textures were first observed in MnSi using neutron scattering [2-4]. Much of the excitement surrounding skyrmions is due to their high potential for application in spintronics [5,6]. Yu et al. observed skyrmion motion in FeGe using LTEM at current densities as low as 10 Am-2, ~106 orders of magnitude lower than that required to move domain walls in ferromagnetic devices making them promising for racetrack memory applications [5-7]. While previous magnetic skyrmion research has largely been focused on non-centrosymmetric crystals where the Dzyaloshinskii-Moriya interaction (DMI) stabilizes the skyrmion phase, here we present a LTEM study of an amorphous ferromagnetic material with perpendicular magnetic anisotropy, consisting of an Fe/Gd multilayer film. FeaGdb films with a = 3.4, 3.5, 3.6 � b = 4.0 � and 80 repeating layers were produced by sputter deposition onto a SiN window. Data was recorded in Lorentz mode using an FEI Titan equipped with an image aberration corrector. By preforming transport-of-intensity equation (TIE) analysis, the under- and over-focus LTEM images can be used to determine the in-plane magnetic induction of the sample. At zero applied field all samples have magnetic stripe/labyrinth domains shown in Figures 1b and 1c. Increasing the magnetic field applied out-of-plane causes the stripe or labyrinth domains to break up into magnetic bubble structures shown in Figure 2. These structures can be placed in three categories: type II magnetic bubbles, skyrmions, and biskyrmions, Figure 2 has examples of these three structures. Biskyrmions are a composite state formed by two skyrmions with opposite chirality for which the total topological charge is 2, and they have only been observed in La2-2xSr1+2xMn2O7 by Yu et al. [8]. As seen in Figures 2c and j the type II bubbles and biskyrmions have very similar spin textures. Additionally, both these structures form in lines where stripe domains previously resided, similar to what was observed in La2-2xSr1+2xMn2O7 [8]. The presence of biskyrmions and skyrmions with both helicities indicates that in this material these topological spin textures are stabilized by long range dipole interactions, as opposed to the Dzyaloshinskii-Moriya interaction responsible for stabilizing the skyrmion phase in non-centrosymmetric crystals such as MnSi and FeGe [9]. To further investigate the stability of these magnetic structures we have varied both the magnetic field strength and tilt angle of the samples. We found that the type of magnetic feature (stripe, labyrinth, skyrmions...) that appears is largely determined by the detailed field history of the sample. For example, Microsc. Microanal. 21 (Suppl 3), 2015 1650 tilting the sample to 30�, applying a saturating field, then reducing the field back to zero produces a nearly uniform stripe domain similar to that seen in Figure 1b. We also observe that uniform stripe domain patterns favor the creation of biskyrmions and type II bubbles when an out-of-plane field is applied, while the skyrmion state is favored when the film starts in a labyrinth pattern. These results give further evidence that these magnetic structures are stabilized by long range magnetic dipole interactions. (a) (b) (c) Figure 1. (a) Spin structure of a (d) Bloch type skyrmion, underfocus Fresnel contrast Lorentz microscopy images showing (b) stripe domain pattern, (c) labyrinth domain pattern (a) (d) (g) Figure 2. (a-c) under-focus Lorentz images showing (a) type II magnetic bubbles, (b) skyrmions with both helicities and (c) biskyrmions. (d-f) color representations of the in-plane magnetic induction calculated using TIE analysis hue and saturation indicate the direction and magnitude of the magnetic induction. (g-i) vectorial representations of the in-plane magnetic induction. (b) (e) (h) (c) (f) (i) References: [1] Malozemoff, A. & Slonczewski, J. "Magnetic Domain Walls in Bubble Materials: Advances in Materials and Device Research", ed. R Wolfe (Academic press, New York). [2] M�hlbauer, S. et al. Science 323, 915�919 (2009). [3] Neubauer, A. et al. Phys. Rev. Lett. 102, 186602 (2009). [4] Pappas, C. et al. Phys. Rev. Lett. 102, 197202 (2009). [5] Fert, A., Cros, V. & Sampaio, J. Nat. Nanotechnol. 8, 152�156 (2013). [6] Tomasello, R. et al. Sci. Rep. 4, (2014). [7] Yu, X. Z. et al. Nat. Commun. 3, 988 (2012). [8] Yu, X. Z. et al. Nat. Commun. 5, (2014). [9] Yu, X. et al. Proc. Natl. Acad. Sci. U. S. A. 109, 8856�8860 (2012).");sQ1[824]=new Array("../7337/1651.pdf","A Biomimetic-Computational Approach to Optimizing the Quantum Efficiency of Photovoltaics","","1651 doi:10.1017/S1431927615009034 Paper No. 0824 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A Biomimetic-Computational Approach to Optimizing the Quantum Efficiency of Photovoltaics Lisa M. Perez1 and Andreas Holzenburg2 Laboratory for Molecular Simulation, Texas A&M University, College Station, TX 778433012, U.S.A.; 2 Microscopy & Imaging Center, Department of Biology, Department of Biochemistry and Biophysics, Texas A&M University, College Station, TX 77843-2257, U.S.A. The most advanced low-cost organic photovoltaic cells have a quantum efficiency of ~10%. This is in stark contrast to plant/bacterial light-harvesting systems which offer quantum efficiencies close to unity. Of particular interest is the highly effective quantum coherence-enabled energy transfer. Noting that quantum coherence is promoted by charged residues and local dielectrics, classical atomistic simulations and Time-Dependent Density Functional Theory (TD-DFT) [1] can be used to identify charge/dielectric patterns and electronic coupling at exactly defined energy transfer interfaces incorporating structural information obtained on photosynthetic protein-pigment complexes. To this end, the project focuses on the first protein-pigment-redox carrier complex of the linear electron transport phosphorylation chain termed photosystem II [PSII] [2]. PSII contains more than 10 major polypeptides in addition to hundreds of pigment molecules amounting to a molecular mass in excess of 1 Mio Dalton. Owing to the complexity and fragility of PSII, this project bases the overall architecture of PSII on in situ EM data providing structural clues about the entire, unperturbed PSII complex [3, 4]. Albeit not to high resolution when compared to X-ray crystallography and NMR spectroscopy, the EM tomographic results and projection maps provide an accurate delineation of the native complex suitable for fitting high-resolution X-ray data of PSII subcomplexes (Fig. 1) [5- 8] towards an atomistic model of the entire PSII complex. This must also include the light-harvesting antennae, i.e. the light-harvesting chlorophyll (Chl) a/b protein complex [LHCII]. With respect to LHCII one should take into account positioning LHCII next to PSII as well as in a separate, complementary membrane thus permitting to test for both, horizontal (intramembrane) and vertical (intermembrane) energy transfer, respectively. The presence of LHCII in a membrane different from PSII is supported by strong biochemical evidence and tomographic data [3], and it has also been noted that the organization of LHCII may change in response to environmental conditions [9-11]. Theoretical investigations thus far have focused on UV-Vis spectra calculations for single Chla, Chlb, the strong coupling special Chla/Chla pair (D1/D2, the reaction center) and intermediate coupling Chlb/Chlb pairs in LHCII. As expected, the TD-DFT B97X-D [12] calculated spectra showed strong communication (strong red-shift) between the Chla/Chla pair that has a short MgMg distance (8 �) and varying degrees of weaker communication (little to no red-shift) for Chlb/Chlb pairs in the LHCII displaying Mg-Mg distances above 9 �. In order to move quantum-structured photovoltaics into the mainstream, it is important to identify dielectric patterns and changes therein at the quantum scale. With this strategy and the interdisciplinary combination of complementary fields of research depicted in Fig. 2, bandgap engineering of nanomaterials aimed at the replication of biosystems that have been optimized for millions of years together with the establishment of quantum coherence principles could constitute a perfect synergy towards the optimization of achievable target module efficiencies. 1 Microsc. Microanal. 21 (Suppl 3), 2015 1652 References: [1] Bauernschmitt, R., and Ahlrichs, R.: Chem. Phys. Lett. 256 (1996) p. 454. [2] Wydrzynski, T.; Satoh, Kimiyuki (Eds.) (2005) Photosystem II. Advances in Photosynthesis and Respiration, Vol. 22, Springer, 786 pp. [3] Ford R.C., Stoylova, S.S., and Holzenburg A.: Eur. J. Biochem. 269 (2002) p. 326. [4] Stoylova, S.S., Ford, and R.C., Holzenburg, A.: Ultramicroscopy 77 (1999) p. 113. [5] Umena, Y. et al.: Nature 473 (2011) p. 55. [6] Loll, B. et al.: Nature 438 (2005) p. 1040. [7] Liu, Z. et al.: Nature 428 (2004) p. 287. [8] Pan, X.W. et al.: Nat. Struct. Mol. Biol. 18 (2011) p. 309. [9] Ford, R.C., and Holzenburg, A.: Cryst. Res. Technol. 49 (2014) p. 637. [10] Simidjiev, I., et al.: Proc. Natl. Acad. Sci. U.S.A. 97 (2000) p. 1473. [11] Garab, G., et al.: Biochemistry 41 (2002) p. 15121. [12] Chai, J.D., and Head-Gordon, M.: J. Chem. Phys. 128 (2008) 084106. [13] The authors would like to acknowledge support by the Supercomputing Facility and the Division of Research (Texas A&M University). Figure 1. Synergy between X-ray and in situ electron crystallography. Conceptualizing the atomistic model of PSII. Subunit CP24, 26, and 29 positions are marked. One unit cell is shown with a x b = 15.5 x 23.1 nm and = 97.3� [3]. Figure 2. Interdisciplinary strategy critically involving complementary expertise.");sQ1[825]=new Array("../7337/1653.pdf","Combine Simulation and Experiment EELS to Characterize Ionomer Conformation","","1653 doi:10.1017/S1431927615009046 Paper No. 0825 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Combine Simulation and Experiment EELS to Characterize Ionomer Conformation Chen Wang1, Stephen J. Paddison2, and Gerd Duscher3 1. Division of Applied Research and Technology, National Institute for Occupational Safety and Health, Cincinnati, OH, 45226, USA 2. Department of Chemical and Biomolecular Engineering, University of Tennessee, Knoxville, TN 37996, USA 3. Department of Materials Science and Engineering, University of Tennessee, Knoxville, TN, 37996, USA The development of modern electron energy-loss spectroscopy (EELS) techniques allows one to probe both local chemical compositional information and structural properties of polymeric materials [1]. High resolution EEL spectra acquired by a monochromatic TEM/STEM may be analyzed and compared with simulated spectra from first principle based calculations. The specific features in the low-loss and Kshell core-loss spectra of carbon are useful in characterizing the complex backbone conformations of perfluorosulfonic acid (PFSA) ionomers, such as NafionTM and Aquivion� [2-3]. Chain conformations of PFSA ionomers were investigated with a monochromatic EELS on a 200 kV Zeiss Libra 200 TEM/STEM. Membrane samples were prepared by cryo-microtome and examined in the cryo environment to minimize sample damage due to electron beam exposure. The EEL spectra were then fitted with Gaussian and Lorentzian functions using the Quantifit program [4].To better understand the structural and optical properties of these materials, density function theory (DFT) calculations were undertaken to compute energy-loss spectra in both low-loss and core-loss regions. The simulations were carried out with the GGA-PBEsol exchange-correlation functional. The polarization of the electromagnetic field was taken into account by varying the directions of the electric field vector of the incident electrons. The contributions from three directional components were then averaged to simulate the optical response of the polycrystalline systems. Results using this approach to characterize the chain conformations of polytetrafluoroethylene (PTFE) and Nafion are shown in Figs. 1 and 2, respectively. The experimental spectra show several unique features in the low-loss region and the onset of the carbon K edge for PTFE. Similar features were observed for Nafion and further examined with DFT calculations. The spectral features including shape, width, and relative intensities of the characteristic peaks in the experimental spectra are well reproduced by the DFT calculations. By comparing the experimental and theoretical spectra, we have demonstrated an application of combining high resolution EELS and DFT calculations to discriminate changes of chain conformation and orientation in PFSA ionomers. References: [1] M.R. Libera and R.F. Egerton, Polymer Reviews, 50 (2010), 321. [2] C. Wang, G. Duscher and S.J. Paddison, Microscopy, 63 (2014), 73�83. [3] C. Wang, G. Duscher and S.J. Paddison, RSC Adv., 5 (2015), 2368-2373. [4] G. Duscher, M.E. Hmielewski and J.D.O. Oduor, Microsc. Microanal., 15 (2009), 446-447. Microsc. Microanal. 21 (Suppl 3), 2015 1654 Figure 1. Experimental EEL spectra of PTFE show features in low-loss and near C-K edge. Energy loss functions (ELF) and local density of states (LDOS) of carbon were calculated to investigate spectral properties as a function of chain conformation. Figure 2. Comparison of low-loss and C-K ELNES from experiment and simulation for Nafion TM. The gross features of experimental EELS can be reproduced from DFT calculations.");sQ1[826]=new Array("../7337/1655.pdf","Investigation of the Structural and Electronic Properties of Pt/-Al2O3, a Model Catalyst System","","1655 doi:10.1017/S1431927615009058 Paper No. 0826 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Investigation of the Structural and Electronic Properties of Pt/-Al2O3, a Model Catalyst System Cecile S. Bonifacio1, Qing Zhu1, Dong Su2, Fernando Vila3, Henry O. Ayoola1, Stephen D. House1, Josh Kas3, John J. Rehr3, Eric A. Stach2, Wissam A. Saidi4, and Judith C. Yang1 Department of Chemical and Petroleum Engineering, University of Pittsburgh, Pittsburgh, PA 15260 Center for Functional Nanomaterials, Brookhaven National Laboratory, Upton, NY 11973 3. Department of Physics, University of Washington, Seattle, WA 98195 4. Department of Mechanical Engineering and Materials Science, University of Pittsburgh, Pittsburgh, PA 15260 2. 1. Pt/ -Al2O3 is possibly the most important heterogeneous catalyst system based on its application in numerous technologically important processes, including oil refining, catalytic converters, and fuel cells [1,2], making it the subject of numerous experimental and theoretical studies. The bulk structure and surfaces of -Al2O3 under catalytic conditions are complex, thus the system remains poorly understood. For example, Digne et al. [3] argued that the defective spinel structure of -Al2O3 is not correct based on DFT calculations combined with thermodynamic models. To bridge the gap between experiment and theory, the nature of the binding sites (and the modification of these sites) of Pt nanoparticles (NPs) to the well-defined crystalline -Al2O3 surface will be investigated using electron energy loss spectroscopy (EELS) and theoretical simulations to understand the effect of the support on the catalytic properties of the system. The model -Al2O3 catalyst system was prepared by first growing the single crystalline -Al2O3 thin films through the controlled oxidation of NiAl (110) and then depositing Pt NPs using a UHV dual ebeam evaporator. Cross-section TEM samples of the Pt/-Al2O3 were prepared using the focused ion beam (FIB) lift-out technique. EELS data were acquired across the Pt NP and -Al2O3 (110) substrate interface and on the bulk -Al2O3 substrate at room (RT) and cryo temperatures. Simulated EELS O K edge signals on the two areas on the sample were computed based on plane wave DFT optimized for Digne's -Al2O3 structure using FEFF9 code. Scanning transmission electron microscopy (STEM) observations show that the cryo temperature conditions prevented issues related to electron-beam damage such that no holes and no growth of the EELS O K pre-peak were observed after EELS acquisition. The O K pre-peak (* in Figure 1) was consistently present in spectra taken at the Pt/-Al2O3 interface at cryo temperature but not in those acquired at room temperature (c.f., Figure 1d and 1b, respectively). The calculated EELS O K signals for the Pt on (110) -Al2O3 exhibited a similar pre-peak at 532 eV correlating to the experimental EELS O K data. The observed O K pre-peak can be attributed to surface defects, such as vacancies, that may pin the Pt NPs. This is similar to the O K pre-peak EELS results in the La0.5Sr0.5CoO3- structure, where the pre-peak was shown to be due to the ordering of high concentration of O vacancies [4,5]. Additional EELS experiments are underway to identify the regions and conditions (e.g., areas with twins on the -Al2O3, pristine -Al2O3 at cryo conditions) that produce the pre-peak to ultimately explain its origin. Likewise, EELS O K signal calculations using FEFF9 are concurrently in progress for comparison with the experimental data [6]. Microsc. Microanal. 21 (Suppl 3), 2015 1656 References: [1] G. Ertl, "The Handbook of Heterogeneous Catalysis", (Wiley�VCH, Weinheim), p. 2012 [2] P. Euzen, "Handbook of Porous Materials", ed. J. Weitkamp, (Wiley -VCH, Weinheim), p. 1591. [3] M. Digne et al,, Journal of Catalysis 226 (2004), p.54. [4] R.F. Klie et al,, Physics Review Letters 99 (2007),p. 047203. [5] J. Gazquez et al,, Nano Letters 11(2011), p. 973-976. [6] CSB, QZ, HOA, SDH and JCY acknowledge financial support by DOE Basic Energy Sciences (DE FG02-03ER15476) and the Center for Functional Nanomaterials (CFN) at Brookhaven National Lab supported by the Office of Basic Energy Sciences of the US Department of Energy Contract No. DESC0012704. Figure 1. High-angle annular dark-field (HAADF) images and the corresponding EELS O K signals from the Pt/-Al2O3 interface acquired at room temperature (a,b) and at cryo temperature (c,d). A prepeak (marked as *) in the O K signal was intermittently observed in the spectra from the Pt/-Al2O3 interface. No pre-peak was detected from Spectrum 3 in (d), which was acquired from the -Al2O3 substrate. Figure 2. Atomistic model and calculated EELS O K edge using the FEFF9 code of the of the bulk Al2O3 (a) and for Pt on -Al2O3 (110) (b). A similar O K pre-peak (marked as *) from experiment was observed in (b).");sQ1[827]=new Array("../7337/1657.pdf","Analysis of Surface Structures in Ru Nanocatalysts","","1657 doi:10.1017/S143192761500906X Paper No. 0827 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Analysis of Surface Structures in Ru Nanocatalysts N.P. Walker, B.K. Miller, and P.A. Crozier School for Engineering of Matter, Transport and Energy, Arizona State University, Tempe, AZ 85287-6106 Hydrogen fuel cells are poisoned by even small concentrations of carbon monoxide (CO). Ruthenium (Ru) nanocatalysts are a promising candidate for converting CO into carbon dioxide (CO2), in these fuel cell feedstocks. In order to have a full understanding of ruthenium's capabilities as a CO oxidation catalyst, the surface structure of particles in transmission electron microscope (TEM) images was analyzed via image simulations. Analyzing the surface structure begins with the fundamental principle of Wulff constructions. Wulff constructions are characteristic, three-dimensional shapes determined by the space group of the crystal and the surface energies [1]. The Wulff construction of a small Ru particle is shown in Figure 1. At elevated temperatures in the environmental TEM, we have found that most crystalline nanoparticles observed appear to match the known topology and structure of the Wulff construction. However, to identify individual surfaces, the orientation must also be found. Lattice spacings and angles were measured using a Fourier Transform. When this was performed on the experimental image in Figure 2, the beam direction was found to be [101]. This leads to the conclusion that the (111) and (110) surfaces are visible at the edges of the particle's image. After the particle shape and orientation were determined, a model of the Wulff construction, made in CrystalMaker, was used as the input to JEMS image simulation software to simulate TEM images of the nanoparticle. JEMS can also use the aberrations measured by the image correction software of the Titan ETEM to increase image accuracy. The simulation can be seen in Figure 2 to fit the particle with two exceptions: corners are rounded in the experimental image due to the high energy of the corner sites and part of the particle is obscured and cut off due to the amorphous silica support. More details on the technique and conditions used to obtain the experimental image are found in Miller and Crozier [2]. The surfaces identified in the experimental image can now be compared in detail to the simulation. As the simulated model does not does have any surface reconstruction, all comparisons will be with respect to the simulated bulk-terminated structure. From the measurements, it can be concluded that both the (111) and (110) surfaces experience lateral surface relaxation. We will present similar measurements of surface reconstructions while the Ru nanoparticles are reacting with various gaseous environments including combinations of CO, H2, and O2 relevant to CO oxidation. References: [1] R.V. Zucker, et. al. Journal of Materials Science 47, (2012). p. 8290�8302. [2] B.K. Miller, P.A. Crozier. Microscopy and Microanalysis (these proceedings) [3] J. Gavnholt. Thesis. Technical University of Denmark, 2009. [4] The support from National Science Foundation CBET-1134464 and the use of ETEM at John M. Cowley Center for HR Microscopy at Arizona State University is gratefully acknowledged. Microsc. Microanal. 21 (Suppl 3), 2015 1658 Figure 1: Wulff construction of a Ru particle which has a P63/mmc space group and surface energies of 0.176, 0.221, and 0.221 eV/� for (001), (111), and (110) respectively is made using WulffMaker [1,3]. Figure 2: Comparison between the averaged experimental TEM image and simulated Wulff construction. It can be seen that the corners are rounded due to the high energy of those sites. Table 1: Lattice Spacing Measurements in nm. (111) Experimental Simulation (110) Experimental Simulation d1 0.27 0.22 d1 0.26 0.25 d2 0.32 0.27 d2 0.35 0.29 d3 0.25 0.23 d3 0.30 0.27 Figure 3: Comparison between surfaces found in the averaged TEM image and the bulk-terminated surface of the Wulff construction simulation. Results of measurement can be found in Table 1.");sQ1[828]=new Array("../7337/1659.pdf","Medium-Range Structure of Zr-Cu-Al Bulk Metallic Glasses from Structural Refinement Based on Fluctuation Microscopy","","1659 doi:10.1017/S1431927615009071 Paper No. 0828 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Medium-Range Structure of Zr-Cu-Al Bulk Metallic Glasses from Structural Refinement Based on Fluctuation Microscopy Jason J. Maldonis1, Pei Zhang1, Matt Besser2, Matt Kramer2, Paul M. Voyles1 1 2 Department of Materials Science and Engineering, University of Wisconsin, Madison, WI, USA Ames Laboratory (DOE), Ames, Iowa, USA and Department of Materials Science and Engineering, Iowa State University, Ames, Iowa, USA Bulk metallic glasses (BMGs) show a variety of outstanding properties including high yield strength [1], high elastic limit [2], and excellent nanoprocessability [3]. However, applications of BMGs are limited by the high cooling rates required to avoid crystallization and create a glass. Enhancing the glassforming ability (GFA) and understanding the origin of good GFA are therefore important to further applications of these alloys. We have investigated the medium-range, nanometer-scale structure two Zr-Cu-Al alloys with different GFA by combining experimental data from fluctuation electron microscopy (FEM) [4] with simulated system energies from an empirical interatomic potential in a hybrid reverse Monte Carlo (HRMC) structural refinement [4,5]. FEM data constrain the medium-range order of the structure, and the potential constrains the short-range and chemical order. HRMC optimizes the structure using the Metropolis algorithm and random, single-atom moves. Fig. 1(a) shows FEM V(k) data and simulated V(k) from the HRMC structures for a Zr50Cu35Al15 BMG in the as-quenched state and after annealing at 0.83Tg for 10 min and 60 min, compared to previous results for Zr50Cu45Al5, an alloy with higher GFA [4]. Because FEM is a diffraction technique, the structure it imposes on the HRMC models is well-visualized in reciprocal space. Fig. 2 shows the computed 3D reciprocal space for the model of the 60 minute, 300 C annealed sample. Fig. 2(a) and (b) are isosurface images of the 3D reciprocal space, which identify volumes where significant diffraction occurs. Fig. 2(c) shows the most important atoms that contribute to the diffraction spots labeled in (a) and (b), identified by filtering and back transforming this pair of spots. This "sub-model" consists primarily of an ordered cluster of approximately 300 atoms near the center of the supercell along with some surrounding atoms. By rotating the sub-model with respect to the orientation of the vector g, we can identify exactly which set of planes give rise to the diffraction from a specific pair of spots. This planar structure is shown (with planes vertical) in Fig. 2(d). Models were also analyzed using radial distribution functions and Voronoi polyhedra (VP) indices [6]. VP indices were grouped into three categories: crystal-like, icosahedra-like, and mixed following Hwang et al. [4]. Analysis of the "sub-model" in Fig. 2(c) enables us to connect the feature in V(k) at the k = |g| = 0.42 �-1 from Fig. 2(a), to crystal-like structure in the model. Fig. 1(b) shows that the fraction of crystal-like VP increases as a function of annealing, while the number of icosahedral-like VP simultaneously decreases. This indicates that the planar structure in Zr50Cu35Al15 becomes more crystallike with annealing at the cost of icosahedral-like local structures, which is opposite to the trend found for the better glass former, Zr50Cu45Al5 [4]. For the poorer glass former, the crystal-like peak has larger V(k) while the icosahedral-like peak is smaller. Overall, results suggest that achieving better glass forming ability in this alloy system may depend on destabilizing crystal-like structures as much as enhancing non-crystalline ones. Microsc. Microanal. 21 (Suppl 3), 2015 1660 References [1] C. Hays et al, Phys. Rev. Lett. 84 (2000), p. 2901. [2] R. Vaidyanathan et al, Acta Mater. 49 (2001), p. 3781. [3] M. Carmo et al, ACS Nano 5 (2011), p. 2979. [4] J. Hwang et al, Phys. Rev. Lett. 108 (2012), p. 195505. [5] M. M. J. Treacy and K. B. Borisenko, Science 335 (2012), p. 950. [6] S. Hao et al, Phys. Rev. B 79 (2009), p. 104206. [7] This work was supported by the NSF (DMR-1205899 and CMMI-1232731, JJM, PZ, and PMV). The facilities and instrumentation for microscopy were supported by the University of Wisconsin Materials Research Science and Engineering Center (DMR-1121288). The synthesis of alloys was supported by U.S. DOE, Office of Basic Energy Sciences, through the Ames Laboratory (MFB and MJK), Iowa State University under contract DE-AC02-07CH11358. a b Figure 1. a) Experimental () (markers) and HRMC simulated () (lines) for two MGs. b) Histograms of VP categories for the Zr50Cu35Al15 MG HRMC models. From top to bottom: Al-centered (green), Cu-centered (orange), Zr-centered (gray) VPs. a b c d Zr Cu Al Figure 2. a-b) Two orientations of an isosurface plot of the 3D reciprocal space for Zr50Cu35Al15. c) The sub-model that results from the circled spots in (a) and (b). d) The sub-model in (c) reoriented corresponding to the direction of g.");sQ1[829]=new Array("../7337/1661.pdf","Orientation Mapping by Precession Transmission Electron Microscopy","","1661 doi:10.1017/S1431927615009083 Paper No. 0829 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Orientation Mapping by Precession Transmission Electron Microscopy Danqi Wang1, Amir Avishai1, Arthur Heuer1,2 1. Swagelok Center for Surface Analysis of Materials, Case Western Reserve University, Cleveland, OH, USA. 2. Department of Materials Science and Engineering, Case Western Reserve University, Cleveland, OH, USA. Precession transmission electron microscopy (PTEM) is currently a very "hot" topic [1]. One of its major applications is orientation mapping. While conventional electron backscattered diffraction (EBSD) and transmission Kikuchi diffraction (TKD) [2] rely on Kikuchi pattern analysis, PTEM (ASTAR system by Nanomegas) acquires micro-diffraction patterns for orientation information. Therefore, even though both methods work in most samples, it is possible that for certain samples PTEM may have an advantage over the EBSD/TKD technique. Low-temperature nitridation of stainless steels introduces several GPa residual stresses due to a "colossal" interstitial nitrogen supersaturation [3], which introduces a significant lattice deformation. While Kikuchi diffraction is sensitive to the lattice distortion, the ASTAR electron micro-diffraction approach focuses on image matching, and is not sensitive to such distortions. TKD was carried out on low-temperature nitrided ferrite in 17-7 PH stainless steel. As shown in Fig. 1, two ferrite grains were identified in the cross sectional TEM foil. As the sample free surface is to the left of the foil, the two ferrite grains are at different depths in the nitrided layer, and thus subject to different amount of deformation. For the ferrite grain deeper in the layer (on the right), which is subject to a lower deformation, TKD could identify almost the entire ferrite grain. However, the ferrite grain closer to the free surface (on the left) cannot be identified at all, due to a much higher lattice deformation. In the highly deformed ferrite grain, the Kikuchi pattern from the grain was completely smeared out and thus could not be indexed by the software. This suggests the limitation of TKD on the highly deformed microstructurse. Such deformation is more related to the microscopic lattice deformation, rather than the macroscopic deformation, such as White Etched Areas observed in low-carbon steels after dynamic loading conditions [5]. The same TEM foil was characterized by PTEM. Fig. 2 shows that both of the ferrite grains were successfully indexed. There are still regions in the ferrite cannot be indexed due to the distortion in diffraction patterns from the deformed ferrite grain. This is illustrated in the reliability map (Fig. 2b); the ferrite grain with higher deformation has a lower reliability compared to that of ferrite deeper in the material. Therefore, in the case of heavy deformed microstructures, PTEM has an advantage over TKD. References: [1] Brons JG, et al., JOM, 66 (2014), p. 165. [2] Wu D, et. al., Acta Mater, 79 (2014), p. 339. [3] Wang D, et al., Acta Mater, 86 (2015), p. 193. [4] Keller RR, et al., J Micro 245 (2012), p. 245. [5] Avishai et al., submitted to M&M 2015. Microsc. Microanal. 21 (Suppl 3), 2015 1662 a. b. Figure 1. Data acquired by TKD from cross-sectional nitrided 17-7 PH stainless steel with free surface on the left. (a) Phase map and (b) Band contrast. a. b. c. Figure 2. Data acquired by precessed electron beam in TEM of the same TEM foil in (a). (a) Phase map, (b) reliability map, and (c) orientation map. Scale bars in all images are 1 �m.");sQ1[830]=new Array("../7337/1663.pdf","Determination of Reliable Grain Boundary Orientation using Automated Crystallographic Orientation Mapping in the Transmission Electron Microscope","","1663 doi:10.1017/S1431927615009095 Paper No. 0830 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Determination of Reliable Grain Boundary Orientation using Automated Crystallographic Orientation Mapping in the Transmission Electron Microscope Xinming Zhang1, Jorgen F. Rufner1, Thomas LaGrange2,3, , Ricardo H.R. Castro1, Julie M. Schoenung1, Geoffrey H. Campell2, Klaus van Benthem1 1 Department of Chemical Engineering and Materials Science, University of California, Davis, CA 95616 USA 2 Physical and Life Sciences Directory, Lawrence Livermore National Laboratory, 7000 East Ave, Livermore, CA 94550 USA 3 now at: Ecole Polytechnique Federale de Lausanne, Interdisciplinary Center of Electron Microscopy, 1015 Lausanne, Switzerland Precession assisted automated crystallographic orientation mapping (PA-ACOM) in TEM allows automated collection and indexing of electron nanodiffraction patterns from an area scanned by the electron beam. Acquisition and subsequent indexing of a large number of diffraction patterns obtained from the scanned area provides capabilities to generate orientation maps similar to electron backscattered diffraction. PA-ACOM bears significant advantages over equivalent scanning electron microscopy techniques for the determination of grain boundary orientations in nanogranular materials due to the fact that orientation is determined using transmission electron diffraction rather than backscattered Kikuchi patterns. This study demonstrates quantitatively that the compensation of second order aberrations introduced by high-angle precession of the electron beam is critical for the reliable determination of grain boundary orientations. Two different types of materials were investigated to evaluate PA-ACOM based grain boundary analysis: a polycrystalline MgAl2O4 ceramic and a 5083 aluminum alloy composite. Nanocrystalline MgAl2O4 with a final average grain size of 150nm was isothermally sintered at 1300 �C for 40 minutes in air. The 5083 aluminum alloy composite, consisting of coarse-grained aluminum (grain size 600-2600 nm) and ultrafine grained 5083 aluminum (average grain size around 200 nm), was fabricated via cryomilling and subsequent hot isostatic pressing. More details regarding the synthesis of these two different types of materials are described elsewhere [1�3]. Table 1 summarizes average confidence indices (CI) for orientation maps obtained from both materials. The data for MgAl2O4 demonstrate that compensation of aberrations through the alignment procedures for beam precession improves confidence indices during ACOM. Highest average CI values are obtained using the `full matching' procedure for data analysis. Although it was expected that smallest CI values are obtained without beam precession, the lowest average CI value was observed for experiments with misaligned beam precession and subsequent analysis using the `fast matching' procedure, i.e. only a subset of diffraction pattern templates was considered for automated indexing [4]. ACOM is sensitive to accurate template matching, and improper compensation of lens aberrations that result from beam precession can lead to less reliable results than data obtained in the absence of precession. To illustrate the effects of precession alignment on grain boundary analysis, CI data across a single grain boundary were plotted in Fig.1. Microsc. Microanal. 21 (Suppl 3), 2015 1664 While with and without aberration compensation a confidence index of 20 was obtained at the center of the grain boundary, the improper alignment led to lower confidence indices extending at least 40 nm in each direction. After proper alignment, however, confidence indices only dropped significantly within 10 nm of each side of the grain boundary core. Reference [1] Rufner J, Anderson D, van Benthem K, Castro RHR. J Am Ceram Soc 2013;96:2077. [2] Vogt R. Ultrafine-Grained Aluminm and Boron Carbide Metal Matrix Composites. [S.l.]: Proquest, Umi Dissertatio; 2011. [3] Vogt R, Zhang Z, Topping TD, Lavernia EJ, Schoenung JM. J Mater Process Technol 2009;209:5046. [4] Rufner, J.F., Zhang, X., LaGrange, T., Castro, R.H.R., Schoenung, J.M., Campell, G.H., van Benthem, K., submitted (2015) Table 1 � Average CI values for CI maps for each scan performed under the three conditions: no precession, precession not aligned, and precession aligned. [4] Properly Aligned with Precession MgAl2O4 Spinel 5083 Aluminum Alloy 26.8 � 20.6 `fast Match': 46.6 � 23.5 `full match': 46.5 � 22.6 Not Aligned with Precession `fast Match': 18.3 � 17.9 `full match': 25.4 � 20.9 24.6 � 17.8 No Precession 21.5 � 20.5 33.3 � 21.1 Fig.1 The histogram shows CI values for an intensity line scan of a grain boundary with precession on but not aligned (black line), and aligned (red line). Reproduced with permission form [4].");sQ1[831]=new Array("../7337/1665.pdf","The Presence of Higher-Order Laue Zone Intensities and the Relrod Effect in Cubic Metals in the Transmission Electron Microscope","","1665 doi:10.1017/S1431927615009101 Paper No. 0831 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Presence of Higher-Order Laue Zone Intensities and the Relrod Effect in Cubic Metals in the Transmission Electron Microscope Cody Miller, Robert Field and Michael Kaufman Colorado School of Mines, Golden, CO, USA The presence of 1/3{422} reflections in <111> selected area diffraction patterns (SADP) has been attributed to short-range order (SRO) leading to diminished stress corrosion cracking resistance in IN690 after extended aging1. In this study, it is shown that these diffuse intensities in IN690 are likely the result of scattering from higher-order Laue zone (HOLZ) reflections into the positions observed in the zero order Laue zone (ZOLZ). To support this hypothesis, it is shown that several fcc structures, such as IN600, IN945, commercially pure (CP) nickel, Co-28.5Cr-6Mo and Ni-33at%Cr as well as diamond cubic structures such as Si and zinc-blende structures (CdTe) also give rise to intensities in these positions. This hypothesis explains the presence of diffuse scattering down the <111>, in addition to several other higher index zone axes. The diffuse intensities originally investigated in IN690 (Figure 1) have been confirmed in the above mentioned systems. Centered dark-field imaging shows no indication of ordered domains. A focused ion beam (FIB) liftout of CP nickel was examined to ensure this diffuse scattering was not an artifact of sample preparation or oxidation. Cryo-stage TEM was performed on IN690 to rule out thermal diffuse scattering (TDS) as a possible cause. Additionally, similar results have been observed in silicon nanowires, as shown by Bell and coworkers, which the authors attribute to HOLZ diffraction via the relrod effect.1 To further explain this phenomenon, an experiment was conducted on a Co-28.5Cr-6Mo alloy, shown in Figure 2. Tilting from an adjacent [112] zone axis, intensity from the (111) reflections is seen to be sustained into the 1/3{422} position within the [111] zone. Further examining the remaining two [112] zone axes near the [111], we can explain all diffuse intensities present in the [111] zone axis via HOLZ diffraction. Furthermore, simulations via JEMS crystallography software demonstrate that HOLZ effects can account for all anomalous intensities in the SADP, as shown in Figure 3.2 The concept of HOLZ diffraction in this context requires that this effect occur from layers both above below the zero-order Laue zone (ZOLZ). Furthermore, a full understanding of the Weiss zone law is required. For zone axes where U+V+W sums to an odd value, first-order Laue zone (FOLZ) diffraction occurs. Conversely, for the case when U+V+W sums to an even value, second-order Laue zone (SOLZ) diffraction occurs. Geometrical considerations of both the Ewald's sphere and HOLZ reflections within the reciprocal lattice show that these events are likely to result in diffuse intensity projected into the ZOLZ layer, with modeling of this diffraction event currently in progress. [1] Y. S. Kim, "Effect of short-range ordering on stress corrosion cracking susceptibility of Alloy 600 studied by electron and neutron diffraction," Acta Mater., vol. 83, pp. 507�515, 2015. [2] D. C. Bell, Y. Wu, C. J. Barrelet, S. Gradecak, J. Xiang, B. P. Timko, and C. M. Lieber, "Imaging and analysis of nanowires.," Microsc. Res. Tech., vol. 64, no. 5�6, pp. 373�89, Aug. 2004. [3] P. Stadelmann, "JEMS EMS Java Version V4," 2014. [Online]. Available: http://cimewww.epfl.ch/people/stadelmann/jemsWebSite/jems.html. Microsc. Microanal. 21 (Suppl 3), 2015 1666 Figure 1. SADP's of IN690, showing additional diffuse intensities down multiple zone axes, all of which are attributable to first- and second-order Laue zone diffraction. Figure 2. SADP's of Co-28.5Cr-6Mo tilting experiment, showing incomplete cancellation of {111} reflections within the <111> zone axis. Figure 3. JEMS simulation of HOLZ reflections in the 111 ZA (left), overlaid on a [111] SADP (right).");sQ1[832]=new Array("../7337/1667.pdf","Semi-Automated DigitalMicrograph Routine for Real-Time Phase Identification","","1667 doi:10.1017/S1431927615009113 Paper No. 0832 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Semi-Automated DigitalMicrograph Routine for Real-Time Phase Identification Edward R. White1, Jonathan Weiner1, Andr�s Garc�a-Trenco1, Sebastian D. Pike1, Charlotte K. Williams1, and Milo S. P. Shaffer1 1. Department of Chemistry, Imperial College London, South Kensington, London, SW7 2AZ, UK. Individual nanocrystallite composition can be determined by comparing high resolution transmission electron microscopy (HRTEM) measurements of lattice spacings to the known lattice spacings of materials [1]. Without the aid of expensive software, however, this process can be overwhelmingly time consuming for even a single image containing multiple crystals [2]. Here, we describe a semi-automated routine that enables the user to rapidly identify the phases present in HRTEM images using the native acquisition software on most commercially available TEMs. The algorithm, described below, has two main functions. The first is to overlay rings corresponding to some or all known lattice spacings of a chosen material in an electron diffractogram. The second is to filter the electron diffractogram at the selected spacings and then show in real space where each phase is located. The routine produces an intuitive graphical user interface (GUI) directly in DigitalMicrograph, which allows image processing without data type conversion or third-party software processing [3]. When first running the routine, the user inputs a text file containing the lattice spacings and corresponding hkl values for each possible phase present. The routine parses output text files from a standard X-ray diffraction database; adaptation to other formats is straightforward. As shown in Figure 1, the program loads a GUI containing lists of several different phases and their corresponding hkl values. The example shown in Figure 2 is a post-catalysis mixture of zinc and copper nanoparticles, where the exact composition is unknown. Here, we have chosen to examine the possible presence of zinc oxide, metallic copper, copper oxides, copper hydride, brass, metallic zinc, and zinc stearate. Pressing `Generate FFT' in the GUI computes the fast-Fourier-transform of the HRTEM image and displays the resulting electron diffractogram. Selecting a material in the GUI then overlays rings on the electron diffractogram at each material's known lattice spacings. As shown in Figure 2c, rings at individual spacings are displayed by selecting only those corresponding hkl values. In this manner, determining which phases are present is done rapidly by inspection; in this case both zinc oxide and metallic copper are identified. Once the user identifies all present phases, the contribution of each to the original HRTEM image is highlighted. Selecting `IFFT on selected' in the GUI places ring-shaped masks of adjustable width and Gaussian tail length on the electron diffractogram at each selected spacing, generating an additional electron diffractogram for each masked region. The inverse-fast-Fourier-transform is then computed for each new electron diffractogram. As shown in Figure 2d-e, this routine produces real space images localizing each phase. The contribution from each hkl selection can be displayed separately to indicate specific lattice orientations if desired (not shown). Finally, the resultant real space images are colored and overlaid on the original HRTEM image. Figure 2f shows such an image, in this case the HRTEM image contains a cluster of ZnO nanoparticles surrounding a metallic Cu nanoparticle. The ability to quickly identify regions containing an interface between catalyst species should lead to a better understanding of the active reaction sites in nanoparticle catalytic systems [4]. Microsc. Microanal. 21 (Suppl 3), 2015 1668 References: [1] DB Williams and CB Carter in "Transmission Electron Microscopy" (Springer, New York) p. 283. [2] EF Rauch and M V�ron, Materials Characterization 98 (2014) p. 1. [3] DRG Mitchell, Microscopy Research and Technique 71 (2008) p. 588. [4] The authors acknowledge funding from EPSRC grant number EP/K035274/1. Dr. Bernhard Schaffer is thanked for his advice with scripting DigitalMicrograph. Figure 1. Graphical user interface for real-time phase identification DigitalMicrograph routine. Figure 2. Example of routine in operation. (a) HRTEM image of nanoparticle cluster. (b) Electron diffractogram computed from (a). (c) Phase identification by selecting the hkl values shown in Figure 1. (d) Sum of all IFFTs with corresponding filters placed at ZnO spacings. (e) IFFT with a filter placed at the Cu(111) spacing. (f) Overlay of images (d) and (e) colored green and red respectively, on (a).");sQ1[833]=new Array("../7337/1669.pdf","Improving Spatial Detection of Twins Achieved by Measuring Individual Kikuchi Band Intensity in EBSD Patterns","","1669 doi:10.1017/S1431927615009125 Paper No. 0833 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Improving Spatial Detection of Twins Achieved by Measuring Individual Kikuchi Band Intensity in EBSD Patterns Travis Rampton1, David Fullwood2, Stuart Wright3. 1. 2. EDAX, Mahwah, NJ, USA Department of Mechanical Engineering, Brigham Young University, Provo, UT, USA 3. EDAX, Draper, UT, USA Twin boundaries in a crystalline material can be defined by a particular rotation angle about a particular access or a mirrored crystal orientation about a particular plane [1]. For example, copper twins are typically defined by a 60o rotation about <111> where the associated twin plane is of the {111} family. One critical area of twin research looks at deformation twinning as the limiting factor for formability of Mg alloys, such as AZ31 [2]. In AZ31 there are two basic twin modes: compression twinning and tension twinning. The latter phenomenon forms fairly large, easy to detect twinned regions within parent grains, whereas the former tends to form extremely thin twins that are on the order of 100 nm wide. Additionally, the copper which is frequently seen in many microelectronics contains twins on the order of 10 nm [3]. In both cases these features are within the detectable limits for a modern scanning electron microscope (SEM). However, identifying these twins via crystal orientation relations with electron backscatter diffraction (EBSD) in the SEM relies on a larger spatial resolution which makes detecting these twins from crystallographic information difficult in the SEM. This study presents a method whereby improved spatial resolution of thin twins can be achieved with EBSD. EBSD data is produced when the electron beam of an SEM diffracts off the surface of a crystalline material and forms a pattern of Kikuchi bands on the EBSD detector. The captured EBSD pattern is assigned an orientation, or indexed, based on the pattern of Kikuchi bands which correspond to crystal planes. This diffraction information emanates from the interaction volume of the electron beam which is inversely proportional to the density of the material and thus less dense materials typically end up having lower spatial resolution. The physical interaction volume size can result in a weak EBSD pattern containing data from both the twin and its surrounding parent crystal. When diffraction signals mix in EBSD the result in low image quality (IQ) which is a measure of diffraction strength [4]. Drops in IQ are quite typical near grain boundaries, twin boundaries, and surface defects. For this study the data extraction will proceed in the following manner: Areas of low IQ will be used to identify potential sites of twin boundaries in a microstructure. Individual Kikuchi band intensities will be measured from EBSD patterns in the regions of interest. Kikuchi band intensities will be analyzed for consistent intensity. Bands of consistent intensity will be correlated to crystal planes to check if they match on of the crystal planes related to twinning. Following these steps proof of concept will also be demonstrated on Inconel 600 and Ta with large, easily detectable twins. Figure 1 shows a set of Kikuchi band intensity profiles taken across an Inconel 600 sample. This study will then identify thin twins in Mg and Cu where traditional EBSD could not due to spatial limitations. Figure 2 shows an example area of a twin only detected after applying the diffraction band intensity method. Microsc. Microanal. 21 (Suppl 3), 2015 1670 References [1] J. W. Christian et al, Progress in materials science, 39 (1995), pp. 1-157. [2] J. Scott et al, Met Trans A, 2013. [3] D. Chen et al, J Microsc, 236 (2009), pp. 44-51. [4] S. I. Wright et al, Microscopy and Microanalysis, 12 (2006), pp. 72-84. Figure 1. (Left) Inverse pole figure and (right) IQ and vertical Kikuchi band intensity across twin boundary. Crystal planes associated with each Kikuchi band are labeled. Figure 2. IQ map and twin/parent map. Parent grains detected by standard EBSD are blue and the associated twins are red. A few twins detected by the new band intensity method are circled.");sQ1[834]=new Array("../7337/1671.pdf","Transmission Electron Backscatter Diffraction (tEBSD) analysis of Au Thin Films","","1671 doi:10.1017/S1431927615009137 Paper No. 0834 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Transmission Electron Backscatter Diffraction (tEBSD) analysis of Au Thin Films Eliot Estrine1, Nicholas Seaton2, Prabesh Dulal1 and Bethanie Stadler1 1. 2. ECE Department, University of Minnesota, Minneapolis, MN Characterization Facility, University of Minnesota, Minneapolis, MN Identifying and characterizing nanoscale crystallographic features is vital for understanding deformation in plasmonic devices and developing more robust plasmonic structures. Here, advanced characterization processes using transmission Electron Back Scatter Diffraction (tEBSD) were developed to analyze thin film crystal structure and its role in plasmonic device stability. The specific crystalline features of interest are film texture, grain boundary configuration and grain size. These results will hopefully contribute to the successful implementation of Heat Assisted Magnetic Recording (HAMR) as well as improve plasmonic device reliability in other applications.[1] Nanoscale deformation has significant differences from macroscale melting. Higher energy gold {110} facets have higher surface energy than either {111} or {100} facets and have been seen to reconfigure into a lower energy configuration when exposed to large temperatures or energies.[2] The presence of low energy grain boundaries, including low angle and coincidence site lattice (CSL) boundaries can also result in more robust nanostructures.[3] EBSD is a promising technique for studying such grain boundaries, but the resolution of EBSD is limited by the interaction volume of the electrons within the sample. Recently, improvements in EBSD resolution have been obtained using thin specimens or nanoparticles in a transmission configuration.[4] In transmission mode, the electron interaction volume is much smaller, and there is no scattering from the substrate. This relatively new characterization technique offers a way of directly observing morphology and failure mechanisms in films and nanostructures with small grain sizes. For this work, sputtered Au films were studied by both conventional (reflection) and transmission EBSD. The Monte Carlo simulator CASINO was used to calculate electron trajectories within the sample, Figure 1. Simulations showed that a transmission configuration reduces the lateral spread of backscattered electrons to 4 nm (Figure 1A) compared with approximately 100 nm using conventional EBSD (Figure 1B). Samples were prepared by sputtering 25 nm of Au on commercial TEM windows made of thin Si3N4 membranes. Although designed for TEM imaging, these substrates also work exceptionally well for tEBSD since the thin Si3N4 is almost entirely transparent to the electrons. A customized sample holder was fabricated to keep a thin sample at approximately 10 degrees off normal during imaging. Using tEBSD, 80 � 90% grain indexing was achieved over a 1 m by 1 m area, Figure 2. It is possible to increase indexing to greater than 90% by applying a post processing noise reduction algorithm. Texture information can be visualized using an Inverse Pole Figure (IPF) plot, Figure 2A. The results show that the films have a strong <111> texture perpendicular to the plane of the film, matching our XRD measurements and confirming the validity of these measurements. It is also possible to observe twinning in several of the grains. Crystallographic information can also be used to derive the grain boundary angle between two adjacent grains, Figure 2B. High angle grain boundaries are shown in red while low angle grain boundaries are shown in yellow. CSL boundaries are labelled according to the Microsc. Microanal. 21 (Suppl 3), 2015 1672 inverse density of their common lattice points. In summary, EBSD is useful technology for characterizing the crystallographic structure of a variety of samples. However, until recently it has been restricted to low resolution applications due to high electron interaction volume. A dramatic increase in resolution with tEBSD configurations allows us to apply this powerful characterization technique to the nanoscale. References: [1] T. Rausch, A. S. Chu, P. Lu, S. Puranam, D. Nagulapally, and J. Dykes, Recording Performance of a Pulsed HAMR Architecture, (2014). [2] S. Link, Z. L. Wang, and M. A. El-Sayed, The Journal of Physical Chemistry B 104, 33 (2000). [3] D. Brandon, Acta Metallurgica 14, 11 (1966). [4] R. Keller, and R. Geiss, J. Microsc. 245, 3 (2012). A B Figure 1. Electron trajectories simulated using CASINO demonstrating the distribution of electrons during conventional (A) and transmission mode (B) EBSD. A B Figure 2. Inverted pole figure diagrams showing film texture in the X, Y, and Z (out of plane) directions (A). Grain boundary information showing grain boundary angle and coincidence site lattice boundaries (B).");sQ1[835]=new Array("../7337/1673.pdf","Automated Image Alignment and Distortion Removal for 3-D Serial Sectioning with","","1673 doi:10.1017/S1431927615009149 Paper No. 0835 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Automated Image Alignment and Distortion Removal for 3-D Serial Sectioning with Electron Backscatter Diffraction Amanda J. Levinson1, David J. Rowenhorst1, Alexis C. Lewis2 1 2 Naval Research Laboratory, 4555 Overlook Avenue SW, Washington, DC 20375 Formerly Naval Research Laboratory, 4555 Overlook Avenue SW, Washington, DC 20375 Three-dimensional characterization is required for measuring the true shape, connectivity, and spatial distribution of microstructural features. For accurate property or processing correlations with 3D microstructures, a statistically significant volume of data is required. Serial sectioning has emerged as pertinent method to obtain large volumes of 3D structural information while maintaining a relatively high spatial resolution [1]. For this process, a large series of 2D images are collected on the same area in successive parallel planes then reconstructed to form the 3D volume. Focused Ion Beam (FIB) sectioning in conjunction with SEM imaging excels for the characterization of fine features (<5�m), but lacks the milling rate to collect large statistically relevant volumes of material. Serial sectioning, using mechanical polishing or alternative milling methods [2] for material removal is capable of collecting much larger volumes, with fields of view up to a several mm2 while imaging with resolutions better than 1�m, allowing it to operate on a length scale relevant for a large number of material problems. The biggest limitation to any 3D characterization is typically the image processing and segmentation of relevant structures. However, Electron backscatter diffraction (EBSD) is a powerful characterization tool that maps the orientation on the surface of a crystalline material and is an attractive imaging tool for serial sectioning polycrystalline samples. EBSD data is based on a physical attribute of the material, the crystallographic orientation at every point, thus eliminating lengthy semi-automated data segmentation procedures associated with other imaging. However, the experimental setup requiring a high sample tilt angle of 70o often results in significant geometric image distortions [3], which can be variable from section to section due to small sample orientation changes when the sample is loaded and unloaded from the SEM chamber between each section. Furthermore, these distortions can be exacerbated as the scanning area increases at low magnifications. In this investigation a large volume of 316L stainless steel has been collected via serial sectioning using ex-situ mechanical polishing with EBSD imaging to evaluate grain boundary networks as a function of processing. Each 2D section is composed of a grid of overlapping EBSD images that span a total area of 1 mm2 with an in-plane resolution of 0.5 �m and a section-to-section spacing of approximately 1 �m. After montaging the EBSD images for each section, large and inconsistent image distortions are present between sections, as shown in Figure 1(a) by the difference image of two sequential sections. It was found that the distortion in each section is unique due to the changing experimental parameters, such as sample placement and SEM conditions from the ex-situ nature of the technique. An automated procedure has been developed for eliminating the distortions within each 2D section individually without requiring a distortion calibration for every possible imaging condition a priori. The procedure utilizes another imaging technique that is undistorted, but lacks the microstructural information needed for reconstruction as a "ground truth" image for the geometry in each section. In this case optical microscopy images of fiducial marks on the perimeter of the analysis area were used as the optical micrographs were found to have virtually no significant distortions over a wide field-of-view. Microsc. Microanal. 21 (Suppl 3), 2015 1674 Each EBSD section was undistorted to align the fiducuial marks in the EBSD image with its corresponding optical image using a Nelder-Mead optimization technique with mutual information (MI) as the figure of merit. MI is a quantitative measure of the information two variables share; this value is maximized when the EBSD and optical images are perfectly aligned. MI has two benefits. First, it can be used to align multimodal data, i.e. images collected using different techniques. Second, it has a high tolerance for convergence, ensuring that the optimization has reached a local minimum solution. After this step, all undistorted EBSD sections are aligned in the same manner to each other. Figure 1(b) shows the difference image of the same two sections in 1(a) after final alignment. This automated process was applied to the reconstruction of 46 sections of 316L composing a volume of approximately 825 x 825 x 46 �m, containing just over 20,000 grains (Figure 2 shows a tenth of the reconstructed volume). By comparison, this volume is about 100 times larger than a typical volume obtained via FIB sectioning. The major benefit of utilizing these tools for the reconstruction of large 3D volumes is that image distortions do not need to be measured directly, but are accounted for through image correlation techniques. [4] [1] M.D. Uchic, M.A. Groeber, and A.D. Rollett, JOM 63 (2011) p. 25. [2] M.P. Echlin, A. Mottura, C.J. Torbet and T.M. Pollock, Review of Scientific Instruments 83 (2012) p. 023701-1. [3] G. Nolze, Ultramicroscopy 107 (2007) p. 172. [4] This work was funded by the Naval Research Laboratory under the auspices of the Office of Naval Research and from the Structural Metallics program of ONR. Figure 1. Difference map between two consecutive EBSD sections with (a) after stitching and (b) after alignment. White indicates minimal difference between the two images. Figure 2. Reconstructed volume taken from the white square inset in the 2D section on right.");sQ1[836]=new Array("../7337/1675.pdf","Growth and Characterization of (110) InAs Quantum Well Heterostructures by","","1675 doi:10.1017/S1431927615009150 Paper No. 0836 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Growth and Characterization of (110) InAs Quantum Well Heterostructures by Transmission Electron Microscopy and Electron Channeling Contrast Imaging. Michael B. Katz1,2, Mark E. Twigg1, Adrian A. Podpirka1,2, Mike Hernandez3, Shawn Mack1, and Brian R. Bennett1 1. 2. Division of Electronic Science & Technology, US Naval Research Laboratory, Washington, DC, USA Post-Doctoral Research Associate, National Research Council, Washington, DC, USA 3. Nanotechnology Systems Division, Hitachi High Technologies America, Inc., Clarksburg, MD, USA InAs quantum wells grown in the (110) direction are an important materials system in the field of spin systems. The major hurdle in realizing such devices is structural defects arising from the large mismatch, 7.2%, between the InAs film and its GaAs substrate. Thus, we embark on a study utilizing transmission electron microscopy (TEM) and electron channeling contrast imaging (ECCI) to understand the density and nature of these defects. Films were grown by molecular beam epitaxy on a GaAs (110) substrate, with a 15 nm InAs quantum well between two layers of GaAlSb. Electron diffraction and x-ray reciprocal space mapping reveal that the film stack is tilted about 2.1� about the in-plane [0-11] axis with respect to the substrate. The defect most likely responsible for this tilt is the (a/2)[110]-type Lomer edge dislocation, which, as a misfit dislocation, provides the primary mechanism for interfacial strain relief. Enough of these parallel Lomer edge dislocations of the same polarity would account for the film tilt, as is the case in a low-angle tilt grain boundary. Cross-sectional TEM images, as shown in Figure 1, confirm the presence of such misfit dislocations at the necessary density corresponding to a separation of 5-10 nm. In contrast, images along the orthogonal [100] zone axis, also in Figure 1, show no such population of misfit dislocations, confirming that the strain relief mechanism is highly anisotropic with respect to the growth plane. A population of inclined microtwin-like planar defects along the (-111) plane are also present in the film. They are variously 1-3 nm thick and originate from the film-substrate interface, implying that they are seeded at the very early stages of growth. TEM images of typical examples are shown in Figure 2. Since they almost universally belong to one set of (111)-type planes in the regions examined, we surmise that these inclined microtwin arrays contribute to both the strain relief and the tilting of film relative to the substrate. [1, 2] We also utilized ECCI in order to get a larger view of the dislocations in the heterostructure. This technique allows the imaging of perturbations from the nominal crystal structure over a large set of length scales. Both backscatter and forescatter geometries were used in the ECCI acquisition, and are shown in Figure 3, in which microtwins were starkly visible in the ECC images. Indeed, they ran solely along the [011] direction. Through ECCI, a threading dislocation density of approximately 109 cm-2 could be measured. [1] R. Lihl et al., J. Microsc. 118 (1980), p. 89 [2] F. Ernst and P. Pirouz, J. Mater. Res. 4 (1989), p. 834. Microsc. Microanal. 21 (Suppl 3), 2015 1676");sQ1[837]=new Array("../7337/1677.pdf","Techniques for Transmission EBSD Mapping of Atom Probe Specimens","","1677 doi:10.1017/S1431927615009162 Paper No. 0837 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Techniques for Transmission EBSD Mapping of Atom Probe Specimens Katherine P. Rice1, Yimeng Chen1, Ty J. Prosa1, David J. Larson1, Matt Nowell2 and Mark P. Stoykovich3 CAMECA Instruments Inc., Madison, WI 53711 USA EDAX, Draper, UT 84020 USA 3. Department of Chemical and Biological Engineering, University of Colorado, Boulder, CO 80309 USA 2. 1. Transmission Electron Back Scatter Diffraction (t-EBSD)/Transmission Kikuchi Diffraction (TKD) enables crystallographic identification in samples of much smaller size, and with improved spatial resolution, as compared to conventional reflection EBSD. In t-EBSD, the elimination of a severe tilt angle and the collection of primarily low-loss electrons keeps the probe small enough for mapping resolution down to approximately 2 nm [1]. Atom probe tomography (APT) relies on needle shaped specimens with an apex radius of ~50 nm in order to produce the critical electric field for evaporating ions. Such needle shapes are then ideally sized for characterization with t-EBSD. Atom probe data can provide information about segregation at grain boundaries. Combination of this chemical information with crystallographic information from t-EBSD provides a more complete picture of the chemical composition and the microstructure of the sample [2]. Here we present strategies to optimize t-EBSD for atom probe applications, including sample holding fixtures and microscope parameters that enable the mapping of multiple specimens without the need to break vacuum. Strategies to avoid contamination of the sample during the mapping process will be presented and optimal map conditions for atom probe specimens will be discussed. Moreover, we will discuss challenges with the application of t-EBSD to the characterization of the inherently threedimensional specimens of interest for APT. Unlike conventional reflection EBSD, the mapped surface in transmission is the bottom surface [3], which can present unique difficulties in a specimen with multiple or overlapping grains that can cause overlapping or blurring of the Kikuchi patterns. We demonstrate this particular phenomenon in an APT specimen by mapping a sample with Atom Probe AssistTM mode with side-by-side grains and introducing a 90 degree rotation, such that only one pattern is visible. One critical element in successfully mapping an APT specimen is overcoming the FIBinduced Ga ion implantation which can cause loss of crystallinity in the region of interest [4]. Cleaning the surface with a low-energy ion beam [5] is demonstrated to significantly improve the image quality of the Kikuchi patterns. Figure 1 shows a typical t-EBSD map obtained from a nickel APT specimen. The inset demonstrates a typical Kikuchi pattern obtained from the specimen. In t-EBSD, like all transmission imaging and diffraction techniques, the sample affects the achievable spatial resolution because the beam broadens and loses energy as the electrons undergo multiple scattering events to reach the exit surface. Beam broadening is particularly important when mapping atom probe specimens, where sample thicknesses may range from tens to a few hundred nanometers. Beam size changes across the cone-shaped APT specimen can be modeled to a first approximation with Monte Carlo simulations of electron trajectories. Our models [3] provide estimates of beam size and effective resolution at each point. Figure 2 shows the electron distribution at the exit surface for two different cases: (a) the beam impinging at the center of the tip 100 nm down the shank from the apex; and (c) the beam impinging at the center of the tip 500 nm down the shank from the apex. The models Microsc. Microanal. 21 (Suppl 3), 2015 1678 show that at a distance of 500 nm away from the apex, the distribution of electrons covers a considerable portion of the tip surface. This suggests that there are enough electrons to create a diffraction pattern from the entire bottom surface of the specimen at this thickness, and it may be difficult to isolate single grains during the mapping process. This conclusion is supported by the data in Figure 1, where there are non-indexed points (black regions) in the center of the specimen at the thicker portions of the tip. References [1] P. W. Trimby et al., Ultramicroscopy 120 (2012), p. 16. [2] K. Babinsky et al., Ultramicroscopy 144 (2014), p. 9. [3] K. Rice et al., J. Microscopy 254 (2014), p. 129. [4] J. Mayer et al., MRS Bulletin 32 (2007), p. 400. [5] K. Thompson et al., Microscopy and Microanalysis 12(S2) (2006), p. 1736. Figure 1: Orientation map with minimal cleanup of a nickel APT specimen showing a twin grain boundary. Inset shows an example Kikuchi pattern obtained from the specimen. Figure 2 (a) and (c): Schematics of incident electron beam impinging on two locations along an APT specimen (a) 100 nm from the tip apex, and (c) 500 nm from the apex. Figure 2 (b) and (d): Electron scattering simulations showing the location and energy of electrons exiting the bottom surface of the cone-shaped APT specimen. The electrons are colored by the percentage of incident energy retained at the exit surface. One thousand electrons were used in each simulation.");sQ1[838]=new Array("../7337/1679.pdf","Characterization of Sulfonated Polysulfone Polymers by EELS","","1679 doi:10.1017/S1431927615009174 Paper No. 0838 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of Sulfonated Polysulfone Polymers by EELS Chen Wang1, Stephen J. Paddison2, John R. Dunlap3, and Gerd Duscher4 1. Division of Applied Research and Technology, National Institute for Occupational Safety and Health, Cincinnati, OH, 45226, USA 2. Department of Chemical and Biomolecular Engineering, University of Tennessee, Knoxville, TN 37996, USA 3. Advanced Microscopy and Imaging Center, University of Tennessee, Knoxville, TN 37996, USA 4. Department of Materials Science and Engineering, University of Tennessee, Knoxville, TN, 37996, USA A novel class of polysulfone ionomers have recently been synthesized and proposed as candidate electrolyte membrane materials for fuel cells.[1] These materials show superior proton conductivity and thermal stability due to their unique backbone structures consisting of sulfonated aromatic rings and sulfone units (-SO2-). The local hydration and proton conductivity are closely related to the degree of backbone sulfonation and the spacing of the sulfone units. Electron energy-loss spectroscopy (EELS) has been undertaken to understand the conformational changes in the backbone of various perfluorosulfonic acid (PFSA) ionomers.[2] The combination of spectroscopy and simulation has successfully revealed the conformational dependence of the EEL spectra for PFSA ionomers. In the present work, the specific features in the low-loss and core-loss spectra of light elements (e.g., C and S) were investigated to understand the backbone chemistry of polysulfone ionomers. We have performed EELS with a 200 kV Zeiss Libra 200 TEM/STEM equipped with a monochromator to investigate the spectral characteristics of three types of sulfur-containing aromatic polymers: poly(1,4-phenylene ether-ether-sulfone) (PEES); poly-(phenylene sulfide) (PPS) and sulfonated poly(phenylene sulfone) (sPSO2), as seen in Fig.1. Both low-loss EEL spectra and the energy-loss near-edge structure (ELNES) have been acquired with a high energy resolution of 0.15 eV. The thin sections of the samples were prepared by cryo-microtome and examined in the cryo environment to minimize sample damage due to electron beam exposure. The spectral dependence of the different aromatic backbone structures were investigated by low-loss spectra, C K-edges and S-L2,3 edges as shown in Fig. 2. The spectra show distinct features in the lowloss region and the onset of the C K-edges for PEES, PPS, and sPSO2. The addition of sulfonic acid groups (�SO3H) or ether linkages (-O-) in backbones can both alter the intensities and shapes of peaks in C K-edges and near edge structures. A strong singlet peak in the S-L2,3 ELNES was observed in sPSO2 ionomers, which can be used to characterize the sulfonate groups directly attached to the aromatic backbones. The fine structures of the S-L2,3 ELNES of PPS and PEES were also identified and possessed features similar to the previously reported X-ray absorption spectra of sodium sulfate.[3] References: [1] M. Schuster et al., Macromolecules, 42 (2009), 3129-3137. [2] C. Wang et al., RSC Adv., 5 (2015), 2368-2373. [3] F. Jalilehvand, Chem. Soc. Rev., 35 (2006), 1256�1268. Microsc. Microanal. 21 (Suppl 3), 2015 1680 Figure 1. Chemical structures of repeat units of poly-(1,4-phenylene ether-ether-sulfone) (PEES); poly(phenylene sulfide) (PPS) and sulfonated poly-(phenylene sulfone) (sPSO2). Figure 2. Measured EEL spectra of PEES, PPS, and sPSO2: (a) low-loss (0�40 eV) after the removal of zero-loss (logarithmic function) and plural scattering (Fourier-log deconvolution); (b) C-K ELNES (280-310 eV) and (c) S-L2,3 ELNES (160-210 eV) after background subtractions (power-law function) and single scattered removal (Fourier-ratio deconvolution).");sQ1[839]=new Array("../7337/1681.pdf","Structural and Optical Properties of AlGaN MQWs Grown by MOCVD Using One and Two TMG Sources","","1681 doi:10.1017/S1431927615009186 Paper No. 0839 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Structural and Optical Properties of AlGaN MQWs Grown by MOCVD Using One and Two TMG Sources Shuo Wang1, Hongen Xie1, Yong O. Wei1, Alec M. Fischer1, Fernando A. Ponce1, M. Moseley2, B. Gunning2, and W. A. Doolittle2 1. 2. Department of Physics, Arizona State University, Tempe AZ 85287-1504, U.S.A Advanced Semiconductor Technology Facility, School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332, U.S.A. Quantum wells are used to confine the electron-hole pairs and thus increase the quantum efficiency. However, the growth of a good-quality quantum well (QW) is challenging. Many factors such as choice of substrate, growth temperature, and number of sources affect the structure of the QWs. In this study, QWs are grown by metal-organic chemical vapor deposition (MOCVD) at a temperature of 1155 oC, with one and two trimethylgallium (TMG) sources. The differences between one and two TMG sources are schematically shown in Figure 1. In one TMG configuration, the growth conditions between a quantum well and a quantum barrier (QB) requires the TMG flow rate to be interrupted and changed. During this change, there is no Ga flowing onto the sample's surface. In two TMG configuration, QW and QB are grown by two TMG sources with different flow rate. The active region growth switches between QW and QB without interruption. The aluminum content in the QWs and the QBs in both samples (using one and two TMG) are targeted to be 0.6 and 0.75, respectively. Their structures are studied by scanning transmission electron microscope (STEM) and the optical properties by cathodoluminescence (CL). Figure 2(a) shows a high-angle annular dark-field (HAADF) image of the QWs grown with one TMG source. The contrast in HAADF is proportional to Zn, where Z is the atomic number and n = 1.7-2 [1]. There are dark layers at the interfaces between the QWs and the QBs. QWs have similar brightness as QBs. We can see three atomic layers in each dark layer from Fig. 2(b). Figure 2(c) shows a HAADF image of QWs grown with two TMG sources. There is no dark layers at the interfaces and the contrast difference between QWs and QBs is sharp, as shown in Fig. 2(d). Figure 3 shows the CL spectra of the two samples under the same excitation conditions. The sample grown by two TMG sources exhibits a CL emission intensity 7 times higher than the one grown by one TMG source. The emission peak position also changes from 263 nm (4.71 eV) to 257 nm (4.82 eV). In the sample grown by one TMG source, the Al atoms are mostly gathered at the interfaces (dark layer in Fig. 2), which leads to a reduced quantum confinement as the width of the barriers is actually smaller than expected. This explains the lower efficiency of one TMG sample. The QWs grown with one TMG have a lower Ga content than 40% (less contrast in Fig. 2) which together with the Al-rich dark layers shift the ground state in the QW, leading to the observed emission peak blue shift [2]. References: [1] P. D. Nellist and S. J. Pennycook, Adv. Imag. Electr. Phys. 113 (2000), p. 147. [2] The authors gratefully acknowledge the use of facilities within the LeRoy Eyring Center for Solid State Science at Arizona State University. Microsc. Microanal. 21 (Suppl 3), 2015 1682 (a) (b) Figure 1. Schematic diagram of (a) one TMG source and (b) two TMG sources configuration. (a) (b) (c) (d) Figure 2. HAADF images of the QW region with (a) one TMG source and (c) two TMG sources. Higher magnification images are shown in (b) and (d), respectively. Figure 3. CL spectra of the QW emission by one TMG source (lower curve) and two TMG sources (upper curve).");sQ1[840]=new Array("../7337/1683.pdf","Probing Plasmons in Three Dimensions within Random Morphology Nanostructures","","1683 doi:10.1017/S1431927615009198 Paper No. 0840 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Probing Plasmons in Three Dimensions within Random Morphology Nanostructures Jordan Hachtel1,2, Anas Mouti2, Daniel Mayo3,4, Claire E. Marvinney3, Richard Mu4, Stephen J. Pennycook5, Andrew R. Lupini2, Matthew F. Chisholm2, Richard F. Haglund1,3, Sokrates T. Pantelides1,2,6 1. Vanderbilt University, Department of Physics and Astronomy, Nashville, TN USA Oak Ridge National Laboratory, Materials Science and Technology Division, Oak Ridge, TN USA 3. Vanderbilt University, Interdisciplinary Materials Science Program, Nashville TN, USA 4. Fisk University, Department of Physics and Astronomy, Nashville, TN USA 5. National University of Singapore, Department of Materials Science and Engineering, Singapore 6. Vanderbilt University, Department of Electrical Engineering and Computer Science, Nashville, TN USA 2. Surface plasmon resonances (SPRs) are highly tunable and versatile in device applications, but also vulnerable to inhomogeneities and fabrication variances that can negatively influence the overall device performance. As a result, methods to study and analyze the plasmonic response in complex nanostructures with nanometer precision are needed. Scanning transmission electron microscopy (STEM) and STEM-based cathodoluminescence (CL) and electron energy loss spectroscopy (EELS) have proven to be valuable tools in nanoplasmonic analysis [1�3]. Both types of spectroscopy are useful in their own regard, but each suffers from its own set of limitations, restricting the possible analyses. However, the strengths and weaknesses of CL and EELS are complementary to one another, EELS is a measure of excitation, while CL is a measure of radiative decay. By comparing the two spectroscopies to one another, while keeping this distinction in mind, new information is available that is unavailable from the spectroscopies performed individually [4]. So far STEM-based investigations of SPRS have been limited to ordered structures and symmetric particles, or have relied on simulations to extract information from the data. Here we show that the combination of spatially-resolved STEM-CL and STEM-EELS analysis can be used to probe the plasmonic response of an irregularly shaped nanoparticle, randomly oriented relative to the electron beam, in three dimensions. The distinctive differences between the CL and EELS profiles allow us to observe and identify the dominant plasmon modes, and account both for the variations in intensity and for spatial distribution of the SPRs without the assistance of simulations. Here we present a combined analysis for a single Ag nanoparticle on the surface of a ZnO/MgO core/shell nanowire. The nanoparticle is selected for having high aspect ratios, thus generating nonoverlapping surface plasmon modes, as well as having a thick MgO barrier between the nanoparticle and nanowire to isolate the SPRs from charge-transfer interactions with the optically active ZnO. A high angle annular dark field (HAADF) image of the nanoparticle can be seen in Figure 1a, and next to it the CL and EELS maps of the three dominant longitudinal plasmon modes (Figure 1b-g). The differences between them are immediately apparent, but by combining the CL and EELS maps we can form an understanding of each mode. In Figure 1b and 1c we see the CL and EELS profile of the 2.0 eV peak. By comparing the two, we infer that the 2.0-eV feature is a long-axis plasmon mode that is most strongly excited at the top of the nanoparticle. Similarly the CL and EELS profiles of the 2.8-eV feature (Figure 1d and 1e) reveal it to be a short-axis plasmon mode most strongly excited at the bottom of the nanoparticle. Microsc. Microanal. 21 (Suppl 3), 2015 1684 The 2.4-eV feature is more complex, exhibiting a unique spatial distribution from the CL (Figure 1f), while the EELS profile (Figure 1g) is barely distinguishable from the EELS profile of the short-axis plasmon mode shown in Figure 1e. The similarity arises because the EELS signal at 2.4 eV is dominated by the tail of the short-axis plasmon, indicating no distinct feature is present at 2.4 eV in the EEL spectrum, while a distinct feature at 2.4 eV is present in the CL spectrum. We now invoke the complementarity of EELS and CL, excitation and decay. The two spectroscopies react oppositely in the case of thick samples. Since EELS measures excitation, thicker samples result in increased excitation, and as a result the EEL spectrum of a thick sample is dominated by bulk effects that scale with sample thickness, as opposed to surface effects whose cross-sections remains unchanged. Conversely, the CL is insensitive to bulk effects. For light in the visible range, the skin depth of silver is less than 15 nm. Hence, any emission coming from the bulk is reabsorbed and not detected efficiently. So for all samples, regardless of thickness, the CL spectrum is dominated by surface effects. The presence of the 2.4-eV peak in the CL, coupled with its absence in EELS, assures us that the feature is a surface effect localized above the body of the nanoparticle. Since the peak shows no components along the edges of the nanoparticle, where EELS would detect it easily, we conclude that it is an out-ofplane longitudinal plasmon mode, oscillating parallel to the electron beam With only a CL or EELS analysis, the full plasmonic behavior of the nanoparticle would not be fully understood, and simulations based on estimations of its precise morphology would be needed in order to determine the full plasmonic response of the nanoparticle. However, with the combination of CL and STEM a determination of the LSPR response of a complex nanostructure in three dimensions can be achieved using solely experimental data [5]. References: [1] C. E. Hoffman et al, Nano Lett. 7 (2007), p. 3612. [2] J. Nelayah et al, Nat. Phys. 3 (2007), p. 348. [3] E. M. Pressai et al, Nano Lett. 10 (2010), p. 2097. [4] A. Losquin et al, Nano Lett. Online Early Access (Published online: Jan. 20, 2015). [5] This work was funded by NSF-EPS-1004083, NSF-TN-SCORE, DOD-W911NF-11-1-0156, DODW911NF-13-1-0153, DE-FG02-09ER46554, DE-FG02-01ER45916, and the DOE Office of Science BES Materials Science and Engineering Division. Figure 1. Plasmon mapping with complementary CL and EELS. (a) ADF image of a nanoparticle with a random shape and orientation. (b,d,f) CL and (c,e,g) EEL spectrum images of the dominant plasmon peaks. By using CL and EELS together we determine the type of mode and where it is most strongly active. (b) and (c) 2.0 eV: Long-axis plasmon, strongest at the top of the nanoparticle. (d) and (e) 2.8 eV: Short-axis plasmon strongest at the bottom. (f) and (g) 2.4 eV: Out-of-plane plasmon strongest in the middle of the nanoparticle.");sQ1[841]=new Array("../7337/1685.pdf","Strong Coupling between ZnO Exciton and Localized Surface Plasmon in Ag Nanoparticles Studied by STEM-EELS","","1685 doi:10.1017/S1431927615009204 Paper No. 0841 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Strong Coupling between ZnO Exciton and Localized Surface Plasmon in Ag Nanoparticles Studied by STEM-EELS Jiake Wei1,2, Jia Xu3, Xuedong Bai2, Jingyue Liu1 1. 2. Department of Physics, Arizona State University, Tempe, Arizona 85287, USA Beijing National Laboratory for Condensed Matter Physics, Institute of Physics, Chinese Academy of Sciences, Beijing, 100190, China 3. School for Engineering of Matter, Transport and Energy, Arizona State University, Tempe, Arizona 85287, USA Metallic nanoparticles (NPs) in combination with semiconductor nanostructures have been intensively investigated recently. The complementary optical properties of two composite units, with long-live excitonic excitations in semiconductor nanostructures and localized electromagnetic modes in metal NPs, provide the possibility to modify and design specific optical responses and to observe new phenomena based on exciton-plasmon coupling [1]. For example, recent studies show enhanced emission, luminescence emission wavelength shift and the nonlinear Fano effect in the semiconductor-metal nanostructures [2]. We report here scanning transmission electron microscopy-electron energy loss spectroscopy (STEM-EELS) study on the coherent coupling between excitons in ZnO nanowires (NWs) and localized surface plasmons (LSPs) in Ag NPs. The Ag/ZnO NWs were synthesized by a modified evaporation deposition method. The EELS experiments were conducted on a Nion UltraSTEMTM 100 equipped with a monochromator, a C3/C5 aberration corrector, and a Gatan Enfina electron energy-loss spectrometer. The monochromated STEMEELS system can achieve an energy resolution < 15 meV as well as a spatial resolution < 0.1 nm at 60 KV [3]. Figure 1a shows the high-angle annular dark-field (HAADF) image of Ag nanoparticles supported on a ZnO NW. Figure 1b shows the EELS spectrum of a pure ZnO NW (black curve) and an individual Ag particle with a diameter of ~15 nm (shown in the inset), supported on a 10 nm thick Si3N4 substrate (red curve). The excitonic peak of pure ZnO NW, located at approximately 3.42 eV, is very weak due to the phonon broadening effect. The LSP resonance energy of the Ag particle is 3.29 eV, attributable to the dipole mode. For the Ag/ZnO composite system, coupling of the ZnO excitons with the Ag LSPs leads to the formation of two new exciton-plasmon polaritons (upper polariton (UP) and lower polartiton (LP), shown in Fig. 1c). When the electron beam was positioned on the edge of the NP (position 1 in the inset), both UP and LP were excited. However, when the electron beam was moved to position 2, only UP was highly excited. The peak at ~ 2.75 arises from the interface plasmon of Ag and ZnO. Figure 2 shows a series of EELS spectrum for different sizes of Ag NPs varying from 4.4 nm to 42.5 nm. Figure 2a (2b) was acquired when the beam was put on location 1(2) in the inset of Fig. 1c. For pure individual Ag NPs, a significant blueshift of the LSP resonance energy from 3.2 eV to 3.7 eV has been reported when the size of the Ag NPs decreases from about 30 nm to 2-3 nm [4]. In our Ag/ZnO system, the EELS spectra show a clear anti-crossing behavior of the UP (dotted red arrow in Fig. 2a) and LP (dotted red arrow in Fig. 2b) at the energy of the ZnO exciton (~3.42eV), which indicated by the dotted black line in Fig. 2a and 2b. The interface plasmon peaks (2.60-2.75 eV) in Fig. 2b are sensitive to the specific structure of the interfacial region. This work demonstrates that the monochromated STEM-EELS is a Microsc. Microanal. 21 (Suppl 3), 2015 1686 powerful tool for studying the plasmon-exciton coupling in metal/semiconductor composite systems on a nanometer scale [5]. References: [1] M. Achermann, J. Phys. Chem. Lett. 1 (2010), p.2837. [2] W. Zhang, A. O. Govorov and G.W. Bryant, Phys. Rev. Lett. 97 (2006), p. 146804. [3] O. L. Krivanek et al., Nature, 514 (2014), p. 209. [4] J. A. Scholl, A. L. Koh and J. A. Dionne, Nature 483 (2012), p. 421. [5] The authors acknowledge the College of Liberal Arts and Sciences of Arizona State University for funding and the use of the John M. Cowley CHREM facilities at Arizona State University. (b) Pure ZnO Pure Ag(15nm) LSP 5 n m 3.5 band gap Counts 3.5 Exciton LP UP (c) 1 2 1 2 2.5 3.0 3.5 4.0 4.5 5.0 Energy Loss(eV) Figure 1. (a) HAADF image of the Ag/ZnO. (b) The EELS spectrum of a pure ZnO NW black) and Ag nanoparticle (red) with a diameter of 15 nm (inset). (c) The EELS spectra of the Ag/ZnO NW (inset) from different positions. The size of the Ag particle is similar to the one in (b). All spectra were normalized by the zero loss peak intensity. (a) 4.4 nm (b) 6.4 nm 7.5 nm 11.6 nm 18.7 nm 22.5 nm 42.5 nm 2.5 3.0 3.5 4.0 2.5 3.0 3.5 4.0 Figure 2. (a) A series of EELS spectrum obtained with different sizes of Ag NPs when the electron beam was located at position 1 in Fig. 1c. (b) EELS spectra obtained from the position 2 in Fig. 1c. The dotted black line indicates the energy of the ZnO exciton. The dotted red arrow indicates the blueshift of the LP and UP energy. Energy Loss(eV) Counts Energy Loss(eV)");sQ1[842]=new Array("../7337/1687.pdf","Advances in Scanning Transmission Electron Microscope Cathodoluminescence","","16871687 doi:10.1017/S1431927615009216 doi:10.1017/S1431927615009216 Paper No. 0842 Microsc. Microanal. 21 (Suppl 3), Microsc. Microanal. 21 (Suppl 3), 20152015 � Microscopy Society of America � Microscopy Society of America 20152015 Advances in Scanning Transmission Electron Microscope Cathodoluminescence S. Meuret1, L.H.G Tizei1, A. Losquin1, R. Bourrellier1, L.F. Zagonel1, M. Tenc�1, A. Zobelli1, O. St�phan1, M. Kociak1 1. Laboratoire de Physique des Solides, University Paris-Sud, Orsay France The typical sizes at which confinement effects become predominant in optical properties range from few angstroms or nanometer (for excitons) to tens or hundred of nanometers (for plasmons). As in addition the confinement leads to optical properties intimately related to the size and geometry of the objects, it is thus important to have tools able to probe optical, morphological and structural properties at these scales. Optical microscopies and spectroscopies can hardly deliver such spatial resolutions. Recently, electron spectroscopies such as Electron Energy Loss Spectroscopy (EELS) and Cathodoluminescence (CL) used in a Scanning Electron Microscope (STEM) have shown some successes in addressing this issue. We will thus present how recent technical and conceptual developments in EELS and CL have allowed exploring various aspects of plasmonics and quantum optics at the scale relevant for plasmons and quantum emitters. In order to investigate the physics of plasmons in small metallic nanoparticles, we have probed plasmon modes of the exact same individual particles through combined EELS and CL measurements [1]. CL only probes the radiative modes, in contrast to EELS, which additionally reveals dark modes. The combination of both techniques on the same particles demonstrates that although the radiative modes give rise to very similar spatial distributions when probed by EELS or CL, their resonant energies appear to be different [1]. We trace this phenomenon back to plasmon dissipation. It agrees with electromagnetic numerical simulations and therefore demonstrates that CL and EELS are closely related to optical scattering and extinction, respectively. Exploring the optical properties of quantum confined objects requires yet another step in instrumentation. Indeed, in those cases, not only the spectral variations arise at scale much smaller than in for plasmons (1 nanometers compared to 10-100 nm), but also information on the quantum states of light or the emitters' lifetimes is crucial. We have shown that CL-spectral imaging in the STEM is a very relevant tool to measure and disentangle individual quantum wells emission [2]. A step further has been the introduction of a Hanbury Brown-Twiss intensity interferometer set up fitting our CL-STEM system. This allowed to probe the quantum character of single photon emitters (SPE) with deep subwavelength resolution [3] thanks to the recording of the autocorrelation function (g(2)CL()). We have applied this method to prove that a well-known defect in h-BN [4] is indeed a SPE [5], making the h-BN a promising platform for quantum nano-optics. In these experiments, the detection of a dip in g(2)CL() at small time delays clearly demonstrated anti-bunching and thus the creation of nonclassical light states. At first sight, this seems like a clear proof of the similarity between CL and photoluminescence (PL) when exciting a unique SPE, as anti-bunching from single photon emitters is a well-known effect in PL. However, we also measured the g(2)CL() resulting from the excitation of several of these defects. In this case, we show that the g(2)CL() of multiple defect centers is dominated by a large nanosecond zero-delay bunching (g(2)(0) > 20), in stark contrast with its flat photoluminescence (g(2)PL()=1) function. We developed a model showing that this bunching can be attributed to the synchronized emission of several defect centers [6]. We also show that such phenomenon makes it a very promising method of lifetime measurement at the nanometer scale [7]. [1] A. Losquin and al, Nano Letters (2015) [2] L. Zagonel and al, Nano Letters 11, 568�573 (2011) [3] L. Tizei and al, PRL 110, 153604 (2013) [4] L. Museur and al, PRB 78, 155204 [5] R. Bourrellier and al, in preparation iomaGUNE, Paseo de Miramon 182, 20009 Donostia-San Sebastian, Spain Microsc. Microanal. 21 (Suppl 3), 2015 2015 Microsc. Microanal. 21 (Suppl 3), , Mediterranean Technology Park, 08860 Castelldefels (Barcelona), Spain cience, 48013 Bilbao, Spain a i Estudis Avancats, Passeig Llu�s Companys, 23, 08010 Barcelona, Spain [6] S. Meuret and al, submitted [7] S. Meuret and al, in preparation 1678 1688 he exact same individual hrough combined nanopectroscopy (EELS) and ments. We show that CL contrast to EELS, which he combination of both provides complementaryCombined HAADF imaging (top), EELS (middle), and CL (bottom) spatially resolved Figure 1. data sets. Total acquisition times are similar CL and 16 s for EELS. In probed hat although the radiative modes give rise to very157 s forspatial distributions when the case of EELS, the spectra are first deconvolved and normalized. The to and EELS images are which ies appear to be different. We trace this phenomenon back CL plasmon dissipation,generated by coloring on signatures each filtered maps of the data sets according to its energy, weighing each pixel with maps by its probed by these techniques. Our experiments are in agreement of the intensity and summing all the resulting images. This simplified representation of the EELS and CL s and can be furthersets straightforwardly shows that the EELS data exhibit both dipolar and higher order modes, interpreted within the framework of a quasistatic analytical model. We data ELS are closely whereas to optical scattering and extinction, respectively, with thesuperimposed on the maps related the CL data exhibit mainly the dipolar mode. The black lines addition of indicate the prism shape as obtained from the HAADF image [1]. , electron energy loss spectroscopy, cathodoluminescence, bright and dark modes, light scattering a) b) d) optical states (LDOS)9 or alternative descriptions in terms of impressive advances in noobjects with new and modal decompositions10 have become relevant for understrongly dependent on standing nanoscale-resolved experiments, including scanning and local environment, c) near-field optical microscopy,11,12 thermal radiation scanning tive field of nanooptics. tunnelling microscopy,13 electron energy-loss spectroscopy ave been largely fuelled (EELS),12,14-20 and cathodoluminescence (CL).16,17,21,22 dress optical properties Nevertheless, the physics embodied in optical extinction, new theoretical and absorption, and scattering phenomena should hold, to some ed fundamental issues extent, at the nanometer scale. In other words, we still need to cal optical concepts of understand how a nanostructure absorbs and scatters electroe no longer sufficient to magnetic subwavelength scales. nanoscale. Instead, the a) HADF and b) waves at map of the defects in an An experimental and of defects emission h-BN flex. The number Figure 2. otonic eigenmodes have the same time is much larger than one. c) Emission spectrum of defects, taken on the white excited at square of role when interpreting the HADF image. The blue square2014 Received: November 14, indicates the filter used for the HBT experiment. d) signal taken on the4, 2015square of the HADF image a). We can see a huge g(2)() of the CLRevised: numerous near-field January white bunching effect (g(2)(0) > 50), for an excitation current of I = 2.8 pA. We retrieve the lifetime of the as the local density of (2) h-BN defect thought an exponential fit of the g () function and find in this case = 1.3 ns [6]. A DOI: 10.1021/nl5043775 Nano Lett. XXXX, XXX, XXX-XXX an Chemical Society");sQ1[843]=new Array("../7337/1689.pdf","Modeling Ion Beam Induced Secondary Electrons","","1689 doi:10.1017/S1431927615009228 Paper No. 0843 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Modeling Ion Beam Induced Secondary Electrons U. Huh1, V. Iberi1, W. Cho2, R. Ramachandra3, and D. C. Joy1, 4 1 2 Biochemistry and Cellular and Molecular Biology, University of Tennessee, Knoxville, TN 37996 Electrical and Computer Engineering, University of Tennessee, Knoxville, TN 37996 3 National Center for Microscopy and Imaging Research, University of California San Diego, La Jolla, CA 92093 4 Center for Nanophase Materials Science, Oak Ridge National Laboratory, Oak Ridge, TN 37831 In the Helium Ion Microscope the incident ion beam interacts with the specimen to generate secondary electrons which then carry the desired image information to the detector for viewing. An essential requirement is to be able to determine the ion generated secondary electron (iSE) yield from the sample as a function of the incident ion energy and the choice of the material. Presently experimental iSE yield data for helium ions is limited to just 40 elements or so and a few binary compounds [1]. There is therefore a need for a detailed predictive model which can provide iSE yield data for a wide range of beam energies and for both pure elements and for compounds, which so far has not been possible. In the published first generation version of this program the ion stopping power as a function of ion atomic weight was taken from published data and numerically stored for use. Instead of using multiple individual data tables of the stopping power we will exploit the "ASTAR" program, recently developed by Berger et al [2] at NIST, which computes ion stopping power data for arbitrary energies and target material compositions. This code, freely available at the NIST web site and can be run on any standard computer, is able to compute the stopping power profiles for any ion impinging on an arbitrary target material, across five orders of magnitude of the beam energy as shown in Figure 1 for He+ ions. As a starting point we have chosen to use the averaged value of the ASTAR plot for some 80 elements in the periodic table, and a number of binary compounds in the energy range 1 to 100 keV from Figure 1. The ion energy range of interest presently extends from about 10 keV to 100 keV if ASTAR is used to compute stopping power data for all elements and other common materials. Then, if these plots are super-imposed one on other, it immediately gives a hint that there is very little difference in generic shape of the stopping power curve and absolute intensity within the energy range from 10 keV to 100 keV and the overlaid plots indicates all the predicted data falls closely around the same for images up to energies in excess of 1MeV. Figure 2 shows stopping power data for carbon, silicon, carbon dioxide, and silicon dioxide are in good agreement to the previously discussed aspects. 1st approximation of the overlaid stopping power curves for He+ ions within the range of interest, effective range of the beam in use, produces a "universal" stopping power curve for helium ion interactions with any given element yet compounds and at any energies as shown in Figure 3. With this stopping power data it is then possible for the IONiSE program to reliably predict parameters such as the beam range, and the secondary electron yield from not only pure elements but also compound materials, which so far has not been possible. This approximation approach fits the presently available published data very well. iSE data for compounds is very scarce but the few published examples are also in close agreement to this ASTAR approach. This result is not being proposed as a perfect solution to the problem of simulating ion beam induced images but, given the almost total absence of high quality experimental data, this approach is a big step forward. Microsc. Microanal. 21 (Suppl 3), 2015 1690 Figure 1. Composite plot of the variation in stopping power (eV/cm2/1015) predicted by ASTAR for a He+ ion beam source as a function of its beam energy (keV) and of the target material. The black line shows the averaged stopping power for all the materials tested and as a function of ion energy [2, 3] Figure 3. 1st approximation of overlays of stopping power curves for He+ ions in range of interest Figure 2. Stopping power of C, CO2, Si and SiO2 for He+ ions [2] References [1] R. Ramachandra, B. Griffin, and D. Joy, "A model of secondary electron imaging in the helium ion scanning microscope," Ultramicroscopy, vol. 109, May 2009. [2] J. S. C. M.J. Berger, M.A. Zucker, J. Chang. (2011). Stopping-Power and Range Tables for Electrons, Protons, and Helium Ions. Available: http://www.nist.gov/pml/data/star/index.cfm [3] J. F. Ziegler. (2013). PARTICLE INTERACTIONS WITH MATTER. Available: http://www.srim.org/ [4] This works was partially supported by the Biochemistry and Cellular and Molecular Biology at The University of Tennessee.");sQ1[844]=new Array("../7337/1691.pdf","Scanning Helium Ion Microscopy-Induced Secondary Electron Yields of Composite Materials","","1691 doi:10.1017/S143192761500923X Paper No. 0844 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Scanning Helium Ion Microscopy-Induced Secondary Electron Yields of Composite Materials Vighter Iberi1,2, Uk Huh2, Yueying Wu2, Philip D. Rack1,2, Adam J. Rondinone1, David C. Joy1,2 1. 2. Center for Nanophase Materials Science, Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA Department of Materials Science and Engineering, University of Tennessee, Knoxville, TN 37996, USA This project attempts to quantify the ion-generated secondary electron (iSE) yields of pure elements and composite materials generated by scanning helium ion microscopy. Additionally, the effects of surface roughness and purity on the experimental iSE yields of materials are investigated.[1] Secondary electron imaging has been the most robust method of obtaining high-resolution images and is at the core of imaging and signal analysis in scanning electron and ion microscopy. Briefly, stainless steel, copper and silicon samples were obtained and used without further purification. Gold (Au) and silver (Ag) were co-sputtered using radio frequency (RF) magnetron sources to form a ~200 nm Au-Ag alloy film with a gold composition ranging from ~25-75 at% along the diameter of a 100 mm silicon (100) wafer. The measured sputtering rates at the center of the substrate was ~3 nm/min for both gold and silver under the following conditions � 5 cm diameter sputter targets and a RF power of 20 W for both targets, and a processing pressure of 25 sccm Ar at 5 mTorr. Scanning helium ion microscopy was performed using a Carl Zeiss ORION Nanofab microscope, operating at a fixed extraction voltage of 30 kV while the accelerating voltages were set to 15 kV, 25 kV and 29 kV for each sample. The helium gas pressure in the column was maintained at 2 � 10-6 Torr and a probe current of ~1.3-1.6 pA, measured from the beam blanker, was achieved using a 20 � aperture and a spot control of 7. Secondary electrons ejected from each material were collected using an Everhart-Thornley (ET) detector without applying any bias to the sample. This was done in order to avoid the introduction of electric fields that may affect the landing energy of the ion beam. Other possible effects occurring within the specimen chamber were taken into account by using silicon as a reference material throughout the experiment. The experimental iSE yield of each material was obtained by performing a histogram analysis routine in ImageJ (available as public domain software from http://rsb.info.nih.gov/ij/download.html) on each image. References: [1] D. C. Joy in "Helium Ion Microscopy, Principles and Applications", Springer, (New York) p. 27. [2] Scanning Helium ion experiments were conducted at the Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, which is a DOE Office of Science User Facility. Gold and silver composite materials were prepared at the Department of Materials Science and Engineering, University of Tennessee Knoxville. Microsc. Microanal. 21 (Suppl 3), 2015 1692");sQ1[845]=new Array("../7337/1693.pdf","Study of SEM images obtained with an electron energy and take-off angle (E-) selective detector","","1693 doi:10.1017/S1431927615009241 Paper No. 0845 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Study of SEM images obtained with an electron energy and take-off angle (E-) selective detector Takeshi Otsuka1, Motohiro Nakamura1, Ken-ichi Yamashita1, Masaya Hara1, Felix Timischl2, Kazuhiro Honda1, Masato Kudo2 and Shin-ichi Kitamura1 1. 2. JEOL Ltd., 1-2 Musashino, 3-Chome, Akishima, Tokyo, 196-8558, Japan JEOL Technics Ltd., 6-38 Musashino 2-chome, Akishima, Tokyo, 196-0021, Japan Scanning electron microscopes (SEMs) are usually equipped with different types of electron detectors. This allows acquisition of images showing diverse contrast, which is caused by the range of detectable energy and take-off angle, respectively. However, the phenomenon is not fully understood yet. We have manufactured an E- detector, which can detect electrons emitted from a sample with a selectable range of both energy and take-off angle [1]. Figure 1 shows a schematic diagram of the E- detector. The operation is as follows: 1. The aperture plate acts as a selector of take-off angle. The plate has two types of apertures; one for high angle range detection and one for low angle range detection. 2. Electron energy selection is carried out by applying voltages of opposite polarity to the inner and outer electrodes as shown in Figs. 1(a) and (b). The polarity of the applied voltages is reversed in the two cases of high angle and low angle range detection, so that the deflection direction of the electrons is reversed. Consequently, each detection element detects a different electron energy range. 3. We use a circular Si-photodiode (SiPD) as the electron detection device, which is separated into concentric elements. A high voltage is applied between the metal mesh and the detection elements to accelerate the incoming electrons. Figure 2 shows a set of images from the same area of a solder sample with high and low angle ranges and high and low energy ranges. The high energy and high take-off angle image B shows composition contrast. The high energy and low take-off angle image D shows a clear shadow effect. Edge effect can be seen in the low energy images A and C, however it is more pronounced in the low angle image C than in the high angle image A. Figure 3 shows a set of images of spherical polystyrene particles with a diameter of 1 m taken at 10 kV accelerating voltage. The low energy range images A and C, and the high energy and high take-off angle range image B show charging effects, proving that different take-off angle and energy ranges contribute to contrast corruption by the charging mechanism. The high energy and low take-off angle range image D is not affected by charging. These images show clear differences, which indicate the usefulness of the E- detector. We are now manufacturing a smaller E- detector for use at short working distances. In the presentation, we will show a series of high magnification images obtained with different ranges of energy and take-off angle. References: [1] T. Otsuka et al, Microscopy and Microanalysis (2014) p. 36. Microsc. Microanal. 21 (Suppl 3), 2015 1694 High Energy Electrons Low Energy Electrons Objective lens Cage 17mm Electron detection device Metal mesh Outer electrode Inner electrode 45mm (a) High angle range detection setup Sample Aperture plate (b) Low angle range detection setup Figure 1. Schematic diagram of the E- detector A B High angle C D Low angle 10m Low energy High energy Figure 2. Topographic and composite contrast images of a solder sample taken at 10 kV. A High angle B C Low angle D 10m Low energy High energy Figure 3. Charging contrast images of spherical polystyrene particles with diameter of 1 m taken at 10 kV.");sQ1[846]=new Array("../7337/1695.pdf","Repetitive Observation of Coniferous Samples in ESEM and SEM.","","1695 doi:10.1017/S1431927615009253 Paper No. 0846 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Repetitive Observation of Coniferous Samples in ESEM and SEM. E. Tihla�kov�, V. Nedla Environmental Electron Microscopy Group, Institute of Scientific Instruments of ASCR, Kr�lovopolsk� 147, 61200 Brno, Czech Republic Environmental scanning electron microscopy (ESEM) is well known for its ability to observe nonconductive moist or wet samples in optional conditions including the possibility for in-situ study of their dynamical changes. ESEM allows the observation of a wide range of biological samples from different kinds of fully hydrated fixed cells to live animals, as well as plant samples in their native state, free of any treatment [1]. Nevertheless, observation of wet susceptible biological samples is burdened with the low possibility for repetitive imaging of samples due to sample collapse and relatively low resolution in comparison with scanning electron microscopy (SEM). The observations of wet samples in ESEM are usually performed in high pressure water vapor conditions with cooling of the sample to a temperature slightly higher than 0 �C. Our newly published methodological study of early somatic embryos (ESEs) of conifer [2] pointed out high sample stability and increased resistance to beam damage in nonstandard conditions (low temperature around -20 �C and 400 Pa air environment). The low-temperature (LT) method allows the observation of samples in conditions of reduced gas pressure and relative humidity, hence with higher resolution. The aim of this paper is to introduce and discuss the utilization of the LT method for ESEM [2] as an alternative method for the preparation and repetitive observation of suitable plant samples. The method allows the study of the same plant surface microstructure in a fully hydrated, freeze-dried or sputter coated state and making a compromise between adequate advantages and disadvantages (naturally wet structure and in-situ study of dynamical changes vs. low resolution, the highest beam sensitivity and impact of free radicals in ESEM; chemical free freeze-dried surface structure with higher resolution, lower beam sensitivity and possibility of repetitive observation in SEM/ESEM vs. skills in LT method and necessity of precise control of the environment in the ESEM specimen chamber; expensive and time-consuming sample preparation, shape modification and susceptible microstructure damage vs. the highest resolution, the lowest beam sensitivity and repetitive observation in SEM). The embryogenic cultures of Picea abies were obtained from the collection of Mendel University (Brno). Firstly, the ESEs were observed in their native state according to the LT method protocol using a non-commercial ESEM AQUASEM II (Figure 1A) and dried by controlled pressure decreasing to 10 Pa (Figure 1B). Due to the process of freeze substitution, water from the sample surface as well as from the inner structure was dried; while natural surface structure including specific morphological features of ESEs [3] (Figure 1C-light microscopy) was preserved. When the dried samples were stabilized and the absorption of environmental air moisture was prevented, the sample was prepared for transportation, storage or sputter coating and repetitive observation, see Figure 2. For the possibility of comparison, the dry sample was firstly observed in a low vacuum (10 Pa of air) in ESEM (Figure 2A), and then gold coated and observed in SEM (Figures 2 B and C). Image comparison shows minimum morphological changes but also the possibility to observe the susceptible extracellular matrix in SEM (detail from the white box is visible in Figure 2C). Shrinkage visible in Figure 2B (white arrow) was caused during sputter coating. Microsc. Microanal. 21 (Suppl 3), 2015 1696 In our experiment we observed that the possibility for the accurate setting of the working environment in the specimen chamber of the ESEM can be utilized for fast and cheap sample preparation and for repeatable morphological studies of the ESEs of conifer in SEM. The quality of results is strongly dependent on the process conditions; however high variability of ESEM parameters, such as appropriate temperature, freezing velocity and humidity, can be found and set. Despite the fact that this alternative method has many limitations, the surface microstructure is well preserved with minimum artifacts and without expensive and time-consuming chemical treatments. Figure 1: A,B Process of freeze substitution of the ESEs in the ESEM AQUASEM II. The wet mucous layer on the sample surface (A) is slowly removed from the sample until the surface is exposed (B). The light microscopy of the ESEs shows specific structures in their native state (C). Figure 2: A) The dry sample of ESEs prepared by the LT method observed in low vacuum mode of the ESEM AQUASEM II (10 Pa). (B, C) After gold coating ESEs were repetitively observed in SEM JEOL 6700F in high vacuum 1.5 e-5 Pa. High stability of the sample during observation in different vacuum conditions and microscopes is evident. The white arrow points out shrinkage caused by sputter coating. References: [1] E Tihla�kov�, V Nedla and M Shiojiri, Microscopy and Microanalysis 19(2013, p 914. [2] V Nedla, E Tihla�kov� and J Hib, Microscopy Research and Technique 78 (2015), p 13. [3] V Samaj et al., Plant Cell Rep 27 (2008), p 221. [4] This work was supported by the Grant Agency of the Czech Republic: grant No. GA 14-22777S and LO1212 together with the European Commission (ALISI No. CZ.1.05/2.1.00/01.0017).");sQ1[847]=new Array("../7337/1697.pdf","The Size and Morphological Study of Spherical Polyelectrolyte Complex Beads Using Environmental Scanning Electron Microscopy.","","1697 doi:10.1017/S1431927615009265 Paper No. 0847 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Size and Morphological Study of Spherical Polyelectrolyte Complex Beads Using Environmental Scanning Electron Microscopy. V. Nedla 1, M. Bucko 2, E. Tihla�kov� 1, T. Krajcovic 2, P. Gemeiner 2 Environmental electron microscopy group, Institute of Scientific Instruments of ASCR, Kr�lovopolsk� 147, 61264 Brno, Czech Republic 2 Institute of Chemistry, Centre for Glycomics, Slovak Academy of Sciences, D�bravsk� cesta 9, 84538 Bratislava, Slovakia Development of new polyelectrolyte complex (PEC) capsules/beads for biotechnological applications such as the immunoisolation of Langerhans islets for the treatment of diabetes and the stabilization and reuse of enzymes and bacterial cells as biocatalysts is very important [1]. Morphological characterization and study of PEC beads properties represents important challenge for electron microscopy. These very beam-sensitive bio-polymer capsules used for immobilization of cells are laboratory produced as a uniform with a controlled shape, size, membrane thickness, permeability and mechanical resistance [1]. Owing to importance to study above mentioned parameters, samples must be inspected in their fully native and functional state. It means free of freezing, chemical contamination or preparation, shape distortion and in thermodynamically stabile and fully wet state, precisely reached after very slow changing of conditions in the specimen chamber of ESEM. Violation of these conditions leads to deformation and burst of thin semipermeable membrane surrounding liquid core, containing live bacterial cells, for example E. coli. The aim of this work is to prove ability to use our non-commercial ESEM AQUASEM II for inspection and developmental support of this type of samples and demonstrate their sensitivity on two types of PEC beads prepared from alternative materials. PEC beads has been produced by air-stripping nozzle via polyelectrolyte complexation (20 min) of sodium alginate and cellulose sulphate (CS) as polyanions, poly(methylene-co-guanidine) as a polycation, CaCl2 as a gelling agent and NaCl as an antigelling agent [1] without the use of a multiloop reactor. Two different sources of CS have been tested for preparation of PEC beads (CS from Across Organics, N.J., USA; Fig. 1) and alternative PEC beads (tailor-made CS from Senova, Weimar, Germany; Fig 2). Due to the relatively big size of samples (800m in diameter) and their beam sensitivity, a combination of our newly published method [2] and special improvement of our ionization detector of SE were used. The gentle and slow sample chamber pumping procedure [2] and our ionization detector of SEs [3] (beam current up to 40 pA) enhanced for larger field of view (850 m) were combined. Samples were observed in small droplet (approximately 10 l) of distilled water. Both types of matrices were observed at sample to second pressure limiting aperture distance 4 mm. For thermodynamic equilibrium adjustment control, Pfeiffer pressure gauges CMR 261 and CMR 263, a custom build Peltier stage and a hydration system were used. Fully wet and well preserved membrane free PEC beads with visible surface microstructure are presented in Fig. 1B and Fig. 2B. Especially PEC beads made from alternative material, see Fig. 2 are very sensitive to beam and environmental impact. The extent of sample damages due to ESEM observation can be evaluated by the juxtaposition of images from the light microscope (Figs.1A, 2A) and ESEM (Figs.1B, 2B). Long-term effect of primary electron beam impact on the sample is obvious in Figs. 1 C and 2C. Sample shrinkage throughout the volume, see Fig. 2C, or deformation of the outer spherical segment, see Fig. 1C is evident. New methods for study of PEC beads at lower beam 1 Microsc. Microanal. 21 (Suppl 3), 2015 1698 accelerating voltage (3keV), beam current bellow 20 pA and using new detection systems are in process of our research. Acknowledgments go the grants [4]. Figure 1: PEC beads with immobilized recombinant cells E. coli (10% of wet cells) overproducing enzyme cyclopentanone monooxygenase. A) light microscope; B) AQUASEM II, ionization detector, acc. voltage 20 kV, 4 min. from start of observation, beam current 35 pA, vapor pressure 684 Pa, stage temperature 2�C, humidity 97%. C) the same as B but 12 min. from start of observation, vapor pressure 671 Pa, humidity 95%. Bar is 200 m. Figure 2: Alternative PEC beads prepared using tailor-made CS with immobilized recombinant cells E. coli (10% of wet cells) overproducing enzyme cyclopentanone monooxygenase. PEC beads compared to Fig.1 (more beam and environmental sensitive material). A) light microscope; B) ESEM AQUASEM II, ionization detector, acc. voltage 20 kV, 4 min. from start of observation, beam current 35 pA, vapor pressure 661 Pa, stage temperature 2�C, humidity 94%. C) the same as B but 12 min. from start of observation, vapor pressure 657 Pa, humidity 93%. Bar is 200 m. References: [1] A Schenkmayerov� et al., Applied Biochemistry and Biotechnology 174 (5) (2014), p. 1834. [2] V Nedla, et al., Nuclear Instrumentation and Methodology A 645 (2011), p. 79. [3] E Tihla�kov�, V Nedla and M Shiojiri, Microscopy and Microanalysis 19 (2013), p. 914. [4] This work was supported by the Grant Agency of the Czech Republic: grant No. GA 14-22777S and Slovak Grant Agency for Science VEGA 1/0229/12");sQ1[848]=new Array("../7337/1699.pdf","Morphological and Chemical Analysis of Impurities in Ice Using the Environmental Scanning Electron Microscopy and Fluorescence Spectroscopy.","","1699 doi:10.1017/S1431927615009277 Paper No. 0848 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Morphological and Chemical Analysis of Impurities in Ice Using the Environmental Scanning Electron Microscopy and Fluorescence Spectroscopy. V. Nedla 1, J. Runstuk 1, J. Krausko 2, P. Kl�n 2, D. Heger 2 Environmental electron microscopy group, Institute of Scientific Instruments of ASCR, Kr�lovopolsk� 147, 61200 Brno, Czech Republic 2 Department of Chemistry and RECETOX, Faculty of Science, Masaryk University, Kamenice 5, 62500 Brno, Czech Republic Accumulated and concentrated impurities can be stored in natural ice or snow. They are found to be rejected from the freezing solution to the ice grain boundaries, free ice surfaces or liquid/brine inclusions. Information about compartmentation and phase speciation in ice is thus essential for the assessment of their fate. The location of impurities and their interactions with the water molecules of ice, still not sufficiently clarified, must be studied at low temperatures because thawing smears the information out. When the impurities keep their location while the surrounding ice sublimes, a 3D morphology of the ice boundaries is revealed. Environmental scanning electron microscopy (ESEM) is one of the few methods allowing direct observation of ice bulk sample with location and compartmentation impurities in dynamically changing conditions of relatively high pressure of gas and stable temperature of cooled sample holder. For this work, aqueous (aq) samples were frozen under atmospheric pressure on a silicon plate cooled by the Peltier stage. The initial sample holder temperature was above �1 �C, and a droplet of pure water or the uranyl nitrate solution placed on it was frozen. A uranyl nitrate solution (0.01 M) acidified by perchloric acid to pH = 1 was used because hydrolysis of UO22+ is suppressed, and only a single species (i.e., a hydrated uranyl ion) is present under these conditions. ESEM AQUASEM II equipped with a YAG:Ce3+ BSEs detector, an ionization detector of SEs, a special hydration system and a Peltier cooled stage were used [1]. The pressures between 400-700 Pa, 50% water-vapor saturation, and the temperatures above 250 K were utilized in the experiments. The phenomena of etching and subsequent stripping of impurities are largely suppressed at these conditions. In order to get information about the phase speciation of the uranyl ion and its microenvironment in the ice samples, the corresponding frozen aq solutions were subjected to a luminescence analysis. Crystalline uranyl nitrate hexahydrate provided a luminescence emission spectrum with the emission band maxima located at 488, 509, 533, 559, and 587 nm at 293 K (Fig. 1a). A hydrated uranyl ion in a solution exhibited one additional band at 473 nm besides those observed in the emission spectrum of a crystal. The luminescence lifetime of crystalline uranyl nitrate is known to depend considerably on the degree of its hydration. Uranyl perchlorate was prepared from solid uranyl nitrate by repeated cycles of dissolution in perchloric acid (70%) and evaporation. The luminescence lifetime of uranyl perchlorate crystals was found to be (283 � 10) s. In addition, the dependence of the uranyl ion luminescence lifetime on the perchloric acid concentration has been reported to be nearly linear at the concentrations between 0.3 and 10 M (0.1 M aq HClO4: = 2 s; 11 M aq HClO4: = 65 s) [2]. We utilized the linear regression equation to estimate the perchloric acid concentration in the brine. In this work, the mono-exponential lifetime of uranyl aq solutions frozen at 267 K was (12.3 � 0.3 s) (n = 5). Such a slow freezing apparently caused rejection of the solute solution into the veins and on the surface of a polycrystalline ice matrix along with a substantial increase of the perchloric acid local 1 Microsc. Microanal. 21 (Suppl 3), 2015 1700 concentration. Based on the reported linear dependence of the uranyl luminescence lifetime on the perchloric ion concentration, we estimate that the final perchlorate concentration was 1.5 M, see Fig.1b white areas [3]. Figure 1c shows an ESEM image of the ice sample prepared by freezing of pure water under atmospheric pressure inside the specimen chamber. Different shapes and sizes (30�200 �m) of the ice grains can be distinguished. Due to the detection of SEs, sensitive mostly to the surface topography, the ice grain boundaries are visible as black lines with a bright halo. Since emission of BSEs is related to the atomic number of the present elements, 92U-rich regions appear brighter, whereas the regions consisting of water molecules remain dark, see Figure 1b. The difference between pure ice and the frozen uranyl solution is largely manifested in the channels and pools of concentrated UO22+ solutions (bright) along with the individual ice grains (black). Pools are usually the largest at the triple junctions, although some may also be present on the ice surface. A liquid layer containing UO22+ was expected to be considerably more concentrated than the parent solution due to the freezing concentration effect. Figure 1: An ESEM image of ice prepared by freezing of a solution inside ESEM chamber: a) the luminescence emission spectra of uranyl nitrate (dotted lines) and uranyl perchlorate (dashed lines) crystals at 296 K (bottom) and 77 K(top); b) uranyl salt solution; c = 10�2 M, BSE YAG detector, 254.65 K, 280 Pa (3.9 torr), c) pure water; ionization SE detector; 270 K, 695 Pa (5.2torr). The spectrum of 0.01 M uranyl nitrate solution in perchloric acid at 296 K (pH = 1; solid black lines) is given for comparison. Bar is 100 �m. Our ESEM and fluorescence analyses thus provided unequivocal evidence that freezing of the uranyl salt aq solutions causes rejection of a solute solution to the ice grain boundaries to form a more concentrated brine layer at temperatures above the eutectic temperature, regardless of the rate and method of freezing (at 270 or 77 K). However, uranyl ion speciation was largely dependent on the experimental conditions [3]. Acknowledgments go the grants [4]. References: [1] V Nedla, Journal of Microscopy-Oxford, 237 (2010) p. 7. [2] M Bouby et al, Chem. Phys., 240 (1999) p. 353. [3] J Krausko et al, Langmuir, 30 (19) (2014) p. 5441. [4] This work was supported by the Grant Agency of the Czech Republic: grant No. GA 14-22777S.");sQ1[849]=new Array("../7337/1701.pdf","Dynamic Gas Environmental System Development for in situ Real-time SEM Imaging under Atmospheric Pressure","","1701 doi:10.1017/S1431927615009289 Paper No. 0849 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Dynamic Gas Environmental System Development for in situ Real-time SEM Imaging under Atmospheric Pressure Takeshi Daio1,2,3 , Hai-Wen Li 1,4 , Takashi Gondo 5,7, Hiroya Miyazaki7 , Tatsuya Ikuta6, Takashi Nishiyama6, Koji Takahashi4,6,Yasuyuki Takata4,6,Stephen Matthew Lyth4 , and Kazunari Sasaki1,2,3 1. 2. International Research Center for Hydrogen Energy, Next-Generation Fuel Cell Research Center (NEXT-FC), 3. Faculty of Engineering, Department of Hydrogen Energy Systems, 4. International Institute for Carbon-Neutral Energy Research (WPI-I2CNER), 5. Department of Material Science and Engineering, 6. Department of Aeronautics and Astronautics, Kyushu University, Motooka 744, Fukuoka, Japan 7. Mel-Build Corporation, 3-1-15 Shimoyamato, Nishi-ku, Fukuoka 819-0052, Japan In situ scanning electron microscopy (SEM) has become an increasingly important tool for materials characterization. It provides key information on the structural dynamics of materials technologies, including automotive exhaust catalytic converters, fuel cell degradation due to humidified air and micro structural evolution of hydrogen storage materials. To date, environmental SEM and environmental static cell techniques have been applied to these fields. However, pure hydrogen gas requires a special pumping system due to its fast gas diffusion. In addition, in situ quantitative experiments by welldefined quantitative gases are challenging in the case of gas injection systems without a cell. A closed cell system would allow us to use pure hydrogen atmosphere and liquid observation. Hence dynamical studies on gas species dependency are structurally difficult. [1] Here, we demonstrate a dynamic study in gaseous environment at atmospheric pressure. The use of MEMS technology and a sophiscated gas in/outlet system allowed us to perform a quantitative and dynamic gas study at atmospheric pressure. In addition, this experimental platform allows us to use a conventional SEM (SEM /FIB) system via a simple modification. It is therefore known that MEMS membranes can be used for liquid observation [2] and composition analysis [3], this experimental platform provides future opportunities of dynamical insitu composition analysis under high pressure gas / liquid atmosphere. In this report, a brief in situ SEM observation was performed under argon / hydrogen gas atmosphere at atmospheric pressure. Figure 1 shows an overview and a schematic diagram of the bespoke SEM stage. Two gas lines underneath the stage are connected using smart tube connectors. This is essential in order to ensure stable performance at high- resolution without vibration from the gas line. These also aid installation into a conventional SEM specimen chamber, which generally allows only 25 mm height and 50 mm diameter (e.g. FEI, Holland). The introduced gas species is enclosed by a Si3N4 membrane, which is penetrated by the electron beam. The specimen is attached to a membrane aligned inside the cell. The window size of the membrane is 50 m square. The specimen used in this proof-of-concept experiment is a hydrogen storage material, LaNi5-based alloy with the commercially used composition MmNi3.6Al0.3Co0.7Mn0.4 (Mm denotes ``misch metal'', which is composed of some rare earth elements) was prepared in a purified argon atmosphere by induction melting from element constituents (99.9 % purities). By uncoupling the gas inlets, specimen was enclosed in the cell with Ar gas as reference state. This was then replaced by H2 gas Figure 2 shows an environmental SEM image under (a) Ar gas (b) H2 gas at 100 k Pa (implication of atmospheric pressure) . By measuring these particle sizes, 9 % volume expansion is confirmed. The previous reports and estimated value on maximum hydrogen storage capacity suggests ca. 20 % of volume expansion. [4] Microsc. Microanal. 21 (Suppl 3), 2015 1702 However, this difference can possibly imply grain size and surface oxidization dependence which couldn't studied by microscopic method. Grain size expansion in the presence of H2 is successfully revealed without significant morphological change. This kind of hydrogen storing is sensitive to the surface state. In our previous experiments without first performing bake-out, and after exposure to humidity in the lab didn't show any expansion. In practice, our enclosable cell system with gas in and outlet would provide easy access to the surface sensitive materials. In conclusion, we have developed a gas environment system with dynamically switchable gas atmosphere. This in situ microscopy can allow observation at high pressure of 100 k-Pa. We demonstrate its unique capability by using conventional SEM to study the micro scale mechanism of hydrogen uptake behavior at high pressure. Expansion was observed in the hydrogen storage material to the hydrogen intercalation. This experiment suggests broad opportunities for liquid-based electron microscopy studies relevant to energy storage. In particular, the possibility of measuring the dynamic behaviour under operating gas atmosphere, or even 100 k-Pa on conventional SEM, is now anticipated. References: [1] Niels de Jonge, Frances M. Ross, Nature Nanotechnology 6 (2011), p. 695. [2] Lawrence F. Allarda, Steven H. Overburya, Wilbur C. Bigelowa, Michael B. Katza, David P. Nackashia and John Damiano, Microscopy and Microanalysis 18 (2012), p. 656. [3] Lawrence F. Allarda, Steven H. Overburya, Wilbur C. Bigelowa, Michael B. Katza, David P. Nackashia and John Damiano, Microscopy and Microanalysis 20 (2014), p. 616. [4]Y.Oosumi, Agune technique center. 3 (2008), p.65 [5] The authors acknowledge funding from the Fukuoka Strategy Conference for Hydrogen Energy and suggestion by S.Miyazaki (FEI Japan) Figure 1 (a) Photograph and (b) schematic diagram of the specially designed SEM stage. Figure 2 In situ observation SEM photograph under (a) Ar gas and (b) H2 gas conditions. The gas pressure is 100 k-Pa.");sQ1[850]=new Array("../7337/1703.pdf","Determining the Causes of Scanning Distortions in SEM and FIB","","1703 doi:10.1017/S1431927615009290 Paper No. 0850 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Determining the Causes of Scanning Distortions in SEM and FIB Mariusz Pluska1, Andrzej Czerwinski1, Marek Wzorek1, Marcin Juchniewicz1, Jerzy Ktcki1 1. Institute of Electron Technology, Al. Lotnikow 32/46, Warsaw 02-668, Poland Scanning distortions are a well-known issue in the scanning electron microscopy (SEM) [1, 2], and the problem is related to the focused ion beam (FIB) instrumentation as well [3]. The characteristic jagging of vertical edges in SEM images may disturb observations and impede imaging. The same effect occurring during Focused Ion Beam (FIB) operations causes imperfect patterning of designed shapes and affects the FIB micro/nano machining. In effect, it physically harms or damages patterned structures or devices. The image distortions are attributed most frequently to the presence of alternating magnetic field [4] (or broadly speaking to electromagnetic interference - EMI). Their acoustic origin has been also considered [5]. Accurate determining the real causes of observed imaging and patterning distortions is necessary for their removal. The presented studies showed that although both influences are probable, there is a way to distinguish which one (EMI or acoustic) is the dominant cause of distortions. Such knowledge can prove to be useful. The most popular solutions to the distortion problem are the magnetic field cancellation systems, which installation may be however expensive and uncomfortable. Moreover, it may not help, what makes matters worse. Therefore, it is recommended to check with the use of the proposed methods, whether the main problem is not caused e.g. by a noisy air conditioner or an equipment cooling fan. The influences of EMI and the acoustic noise are distinguishable because they act on the electron/ion beam in different ways. The method to separate them is based on the measurement of the level of the image distortion for several working distances. The peak-to-peak value of edge deformation is a good measure for the level of distortion (inset in Fig. 1a). EMI acts on the electron or ion beam directly through the Lorentz force The impact of the external electromagnetic field is usually reduced by the SEM/FIB chassis. However, especially for low frequencies (e.g. 50-60 Hz), it is impossible to remove EMI impact completely [6]. The effect of distorting electromagnetic field accumulates along the entire path of a charged particle. Therefore, the longer working distance we use, the stronger distortions we observe. The path of the electron or ion under the influence of the uniformly distributed electromagnetic field is a quadratic curve similar to the trajectory in the gravitational field. If the peak-to-peak values of the deformation of a specimen edge are measured for a number of working distances and the plotted points fit the quadratic function, then it may be concluded that the distortions are caused by the electromagnetic field (Fig. 1a). The plot shown in figure 1a was registered in SEM for electron energy of 5 keV and intentionally generated distortions: magnetic field B = 220 nT, frequency f = 50 Hz. The acoustic noise acts on the mechanical components of the instrument. The acoustic wave propagates within the chassis of the instrument and if it matches the resonance frequency of any component (e.g. a piezo-driven specimen stage), then the vibration occurs and gains in intensity. Therefore, the frequency response of the SEM or FIB instrument to the acoustic distortions consists of several narrow peaks (inset in Fig 1b). The present investigation showed that the ambient noise in the instrument environment may contain resonance frequencies of its particular components and therefore it can evoke strong vibrations Microsc. Microanal. 21 (Suppl 3), 2015 1704 visible in SEM/FIB images. It was found that for the distortions caused by the acoustic noise there is no predictable dependence between the working distance and the level of the image deformation (Fig. 1b). The measurement of the peak-to-peak value of specimen edge deformation versus working distance in case when it is caused by the noise, does not match the quadratic function. The plot shown in figure 1a was registered in SEM for electron energy of 5 keV and intentionally generated acoustic distortions: SPL (sound pressure level) = 75 dB, f = 216 Hz (which is one of the resonance frequencies). In conclusion, the presented study describes a simple and useful method for determining if the observed distortions are caused by EMI or by acoustic noise. This is an important guideline when considering methods aimed at SEM/FIB images improvement and FIB micro/nano machining refinement [7]. [1] D.H. Kim, S.J. Kim and S.K. Oh, Nuclear Instruments and Methods in Physics Research A 620.2(2010), p. 112. [2] M. Pluska et al, Journal of Microscopy 224.1 (2006), p. 89. [3] M. Pluska et al, Nuclear Instruments and Methods in Physics Research B (2014) doi:10.1016/j.nimb.2014.11.020. [4] A.E. Vladar, "Scanning electron microscopy in real world environment" (2003), www.nanobuildings.com. [5] K.O. Jung, S.J. Kim and D.H. Kim, Nuclear Instruments and Methods in Physics Research A 676 (2012), p. 5. [6] C.S. Cheung, "Shielding Effectiveness of Superalloy, Aluminum, and Mumetal Shielding Tapes." (2009), Master's Theses and Project Reports, p. 126. [7] The research was partially supported by the grant LIDER/26/169/L-3/11/NCBR/2012 of National Centre for Research and Development, Poland. a) b) Figure 1. Plots of the peak-to-peak value dp-p of the edge deformation versus the working distance (wd) for intentionally generated magnetic (a) and acoustic (b) distortions in SEM. Inset in fig. 1a shows the method to measure the distortion level. Inset in fig. 1b shows measured frequency response of exemplary instrument (Helios NanoLab 600 Dual Beam FIB) to the acoustic vibration.");sQ1[851]=new Array("../7337/1705.pdf","Secondary Electron Yield at High Voltages up to 300 keV","","1705 doi:10.1017/S1431927615009307 Paper No. 0851 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Secondary Electron Yield at High Voltages up to 300 keV Jane Howe1, David Hoyle1, Kota Ueda1, Stas Dogel1, Hooman Hosseinkhannazer2, Matthew Reynolds2, Ren� Veillette3, Michel L. Trudeau3, and David Joy4,5 1. 2. Hitachi High-Technologies Canada Inc, Toronto, Canada. Norcada Inc, Edmonton, Canada. 3. Institut de recherche d'Hydro-Qu�bec, Varennes, Canada 4. Department of Materials Science and Engineering, University of Tennessee, Knoxville, USA. 5. Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, USA. Secondary electrons (SE), emitted from bombardment of a matter with energetic particles such as electrons, ions, or protons, are conventionally defined as those with a kinetic energy of 50 eV or less. The parameter describing SE emission is called SE yield, , which defines as the number of SEs produced by each incident particle. The SE micrograph, which reveals the surface of materials, is the most widely used imaging mode for scanning electron microscopy (SEM). Recently, SE detectors have been installed on scanning transmission electron microscopes (STEM), as it augments the ability of imaging surface features to a bright-field or dark-field STEM image, which essentially reveal bulk information of a very thin specimen. The benefit of simultaneous imaging of surface (SE) and transmitted electrons (TE) at atomic-scale has been demonstrated in materials science research [1], such as learning the degradation mechanism of battery materials [2] and morphological evolution of catalysts during in situ heating [3]. What is the SE yield under high accelerating voltages, such as ranging from 60 to 300 keV in a STEM environment? This is a fundamental question that is yet to be answered. Knowing the SE yield is important to quantitative SE imaging analysis. To date, the majority of the SE yield data have been obtained in a conventional SEM environment at voltages up to 30 keV on bulk samples (which are too thick to emit transmitted electrons) [4-6]. Reimer and Drescher reported the only study of SE yield in a TEM up to 100 keV on thin Al and Au films [7]. We hereby carry out a systematic study of SE yield of selected thin film materials in a Hitachi HF-3300 TEM/STEM at accelerating voltages from 60 to 300 keV. We have selected TEM window chips (Norcada Inc., Edmonton, Canada) as our materials of interest, because this type of MEMS-based chips are flat, with uniform thickness, and widely used in the microscopy and microanalysis community. As shown in Figure 1, the 3mm-diameter chip is a 200-�m-thick silicon wafer with a 250x 250 �m2 thin window at the center. The window materials are SiN, and gold and tungsten films deposited on the SiN layer with varied thickness (Table I). It covers medium to high atomic number (Z) materials with thickness ranging from 10 to 400 nm. We can also focus the beam on the 200-�m-thick silicon wafer, which enable us to measure SE yield of layered bulk composite. The SE yield is evaluated from the equation, = (Iscb - Isc)/Ib, whereas Isc is the specimen current, Iscb is the 50 V-biased specimen current, and Ib the incident beam current. Other than challenges such as high leakage current and surface contamination, measurement errors from possible high SE2 and SE3 emission from a STEM column with narrow beam path has to be taken into consideration. Nonetheless, we expect that this first study of SE yield up to 300 keV in STEM will open a new interesting topical area [8]. Microsc. Microanal. 21 (Suppl 3), 2015 1706 References: [1] Y Zhu et al, Nature Materials 8 (2009) 808. [2] JY Howe et al, Journal of Power Source 221(2013), 455. [3] JY Howe et al, Nano Research Letters 9 (2014). [4] H Seiler, J. Appl. Phys. 54 (1983), R1. [5] Y Lin and DC Joy, Surf. Inter. Analy. 37 (2005), 895. [6] Y Lin, A study of the secondary electrons, Ph.D. Thesis (2007), University of Tennessee Knoxville. [7] H Reimer and H Drescher, J. Phys. D: Appl. Phys. 10 (1977), 805. [8] We thank the insightful discussions with Dr. Donovan Leonard and Dr. Martha McCartney. Figure 1. Norcada MEMS chips used for measuring SE yield. A-D: SiN window chips with varied thickness; E) tungsten coated SiN window; and F) gold-coated SiN window. Table I. List of materials and thickness for the measurement of SE yield Materials SiN films Au on SiN W on SiN 10 10 + 10 30 30 + 30 30 + 30 Thickness (nm) 50 50 + 50 50 + 50 100 100 + 100 100 + 100 200 200 + 200 200 + 200");sQ1[852]=new Array("../7337/1707.pdf","Thin Film Porosity Determined by X-Rays at SEM","","1707 doi:10.1017/S1431927615009319 Paper No. 0852 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Thin Film Porosity Determined by X-Rays at SEM E. Ortel1, R. Kraehnert2, F. Galbert2 and V.-D. Hodoroaba1 1 2 BAM Federal Institute for Materials Research and Testing, Berlin, Germany Technical University of Berlin, Berlin, Germany Various applications such as automotive catalysts, photocatalysis and sensors rely on thin coatings with high active surface area. The development of new and improved coatings requires in-depth understanding of the film morphology and texture. Especially the film porosity is a key parameter to identify structure-property relationships in applications. However, an accurate porosity determination of thin porous coatings is a challenging. In this contribution we present and discuss a new approach for determining the porosity of films by x-rays at SEM. The mass coverage of layers can be calculated by the film analysis software STRATAGem [1] from EDX determined k-values (the ratio of normalized intensity between unknown sample and standard) based on the well known Pouchou and Pichoir approach.[2] The high accuracy of STRATAGem results has been proven for various layered materials.[3-5] We present an approach for determination of the porosity by a combined SEM/EDX/STRATAGem analysis. From the STRATAGem-derived mass coverage and the coating thickness as determined by cross-section SEM micrographs, the average density of the film can be obtained. In a second step, the porosity can be calculated from the measured porous film density and the literature value for the pore walls bulk density. The complete approach is sketched in Figure 1. The general procedure to determine the porosity is demonstrated on thin mesoporous TiO2 films on Si wafer substrates. The TiO2 films were synthesized by a template-assisted synthesis route, which provides a scalable model system with tunable porosity, see Figure 2.[6] EDX spectra of these films were measured at beam voltages of 7, 10, 15, 20, and 30 kV by using a high-throughput SDD EDS detector which enables a time-efficient analysis. Polished bulk of pure Ti and Si served as standard references. Oxygen was quantified by stoichiometry as well as by using a pure SiO2 reference. The resulting k-values were verified by WDX analysis. The calculated mass coverage was independently validated by weighing and by ICP-OES of dissolved coatings. To determine the film thickness, the TiO2 films were cleaved and the resulting cross-section was imaged with a SEM. The measured k-values for Ti K and Si K in dependence on the beam voltage and the corresponding fitted curves calculated by STRATAGem are shown in Figure 3a for a porous 180 nm thick TiO2 film. Figure 3b visualizes the resulting mass coverage values for all beam voltages. The scattering of the individual mass coverage values at different voltages reflects the robustness of the model. An excellent agreement was obtained between the mass coverage calculated by STRATAGem and the mass coverage determined by weighing and ICP-OES of the same film. The increasing porosity of the thin TiO2 films illustrated in Figure 2 could be calculated quantitatively by the new approach. The validation and traceability of the porosity results will be discussed. These combined SEM/EDX/STRATAGem analyses demonstrate a unique tool for non-destructive examination of the porosity of thin porous films. Microsc. Microanal. 21 (Suppl 3), 2015 1708 References: [1] Stratagem version 2.6, SAMx, 4, rue Galil�e, 78280 Guyancourt, France [2] J.-L. Pouchou, Mikrochim Acta 138 (2002), p. 133. [3] V.-D. Hodoroaba et al., Surf. Interface Anal. 44 (2012), p. 1459. [4] M. Procop et al., Anal. Bioanal. Chem. 374 (2002) p. 631. [5] F. Galbert, Microsc. Microanal. 13 Suppl. 3 (2007) 96. [6] E. Ortel et al., Small 8 (2012), p. 298. EDX / STRATAGem mass coverage film density SEM cross section film thickness bulk density POROSITY Figure 1. Illustration of the approach for the determination of the porosity of thin films by EDX/SEM analysis. 100 nm 100 nm 100 nm Figure 2. Top-view SEM micrographs of TiO2 films synthesized with increased amount of a mesopore template, which results in an increase of the void fraction, i.e. film porosity (from left to right). a) b) Figure 3. a) k-values versus beam voltage for Ti K (red) and Si K (black); the markers represent the measured values, the curves are fitted to the k-values. b) k-values of Ti K and Si K versus calculated mass coverage at each beam voltage.");sQ1[853]=new Array("../7337/1709.pdf","Specimen Preparation Method for Size Distribution Measurements of Nanomaterials by Scanning Electron Microscopy - Fixing of Nano-particles on a Substrate with Adhesive Coating","","1709 doi:10.1017/S1431927615009320 Paper No. 0853 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Specimen Preparation Method for Size Distribution Measurements of Nanomaterials by Scanning Electron Microscopy - Fixing of Nano-particles on a Substrate with Adhesive Coating Kazuhiro KUMAGAI1 and Akira KUROKAWA1 1. Surface and Nano-Analysis Section, Nanomaterial Characterization Division, National Metrology Institute of Japan, National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba JAPAN In response to the regulations on manufactured nano-materials (NMs), which started in several EU countries based on the definition of NM by European Commission [1], there are increasing discussions to find the solution for the size distribution measurements of NMs to test if a product meets the definition or not [2]. As one of approaches to the regulation, electron microscopies (EMs) such as transmission electron microscopy and scanning electron microscopy (SEM) attract attentions as NM measuring tools, due to their capability to easily and quickly observe individual nano-particles (NPs) [3]. For size distribution measurements by EMs, specimen preparation is one of the most important processes. To achieve reliable measurements, the particles in a micrograph should be clearly visualized and easy to analyze. To minimize ambiguity, particles should be (1) uniformly placed over a specimen holder with no localization in particle size and number, or (2) placed only in a small area that allow us to count all particles in the area. For both of these cases, particles are expected not to overlap each other. Furthermore, it would be helpful, if there are neither agglomerating nor aggregating particles. It is, however, not easy to make such specimens, because NMs, which are usually supplied as suspension, often get localized and aggregated in drying process. In this paper, we present a preparative method by using polycationic adhesive coating on a substrate to make specimens that satisfy the above condition (1). The approach with adhesive coating has been used in the research field of biology to fix samples such as cell onto a substrate for SEM observations [4,5]. In this study, we applied this technique to important and basic NMs: Au, silica and polystyrene NPs. To form a polycationic adhesive layer, mirror polished Si substrates were dipped in 0.01 % Poly-L- lysine (PLL) solution (Trevigen Inc.) for 2 hours, then rinsed in ultrapure water. These PLL coated Si substrates were dipped into NP suspension to adsorb NPs for 10, 30, 60 and 90 minutes, and then rinsed in ultrapure water. Specimens without PLL coating were also prepared in the same manner. For SEM image observation, we employed a SEM equipped with a filed-emission gun (JSM-7100F, Jeol Ltd.). Figures 1a-1d show secondary electron (SE) images of Au NPs (30 nm in diameter) fixed on PLL coated Si substrate at four different dipping times. These SE images show that Au particles are randomly and individually fixed on the substrate with no overlapping or agglomeration. On the other hand, the Si substrate without PLL layer hardly adsorbs Au NPs even after 90 minutes dipping (Fig. 1e) that corroborates the effect of PLL layer. The variation of coverage of Au NPs against dipping time is plotted in Fig. 1f. The coverage, namely number of particles, linearly increases with dipping time, which can realize easy tuning of particle density on the substrate. This method is applicable to not only Au NPs, but also silica NPs and polystyrene NPs as shown in Figure 2. We demonstrated that PLL coating useful to fix NPs on Si substrate uniformly and individually. Our Microsc. Microanal. 21 (Suppl 3), 2015 1710 method could be effective to a wide range of NMs that are negatively charged in suspension, and offers easy specimen preparation for size distribution measurement of NMs by SEM. References: [1] European Commission, Official Journal of the European Union L 275, 20.10.2011 (2011) p. 38. [2] A Lopez-Serrano et al, Analytical Methods 6 (2014) p. 38. [3] EFSA Scientific Committee, EFSA Journal 9 (2011) p. 2140. [4] SK Sanders et al, The Journal of Cell Biology 67 (1975) p. 476. [5] K Tsutsui et al, J. Electron Microsc. 25 (1976) p. 163. [6] The authors thank Dr. H. Kato in AIST for kind help in preparation of the silica NPs. This work was performed as one of the studies conducted by the consortium for measurement solutions for industrial use of nanomaterials (COMS-NANO) in Japan. Figure 1. (a)-(d) SE images of Au NPs on PLL coated Si substrate. The dipping time was 10, 30, 60 and 90 minutes, respectively. (e) The same as (d) but without PLL coating. Primary beam energy was 30 keV for all SE images. (f) A plot of coverage of Au NPs against dipping time. Figure 2. SE images of NPs on PLL coated Si substrate. (a) Silica NPs. Primary beam energy was 3 keV. (b) Polystyrene NPs. Primary beam energy was 4 keV. The average particle diameter of both of the materials is 50 nm.");sQ1[854]=new Array("../7337/1711.pdf","Quantitative Measurement of Resolution as a Function of Defocus in Different Microscopy Modalities Using a Simplified Technique","","1711 doi:10.1017/S1431927615009332 Paper No. 0854 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantitative Measurement of Resolution as a Function of Defocus in Different Microscopy Modalities Using a Simplified Technique Aric W. Sanders1, Alexandra E. Curtin1 and Ryan Skinner2 1. 2. National Institute of Standards and Technology, Boulder Colorado, USA. University of Colorado, Boulder Colorado, USA. The distance over which the resolution of a microscope image changes appreciably, related to the depth of field, is an important parameter. This value determines the height of an object that can be imaged and said to be "in focus". Although this distance is important in microscopy, it is typically quoted from calculations based on simplified assumptions (such as perfect optics) [1] or presented qualitatively [2]. In order to measure the effects of defocussing in different microscopies, we use the logistic equation to determine image resolution from light-to-dark edge transitions [3]. The logistic equation is given as: S ( x) a 1 e ( x x 0 ) / b c where the relevant fit parameter is b, and resolution is defined to be 3.33b. The characteristic distance over which the light-to-dark transition occurs has been related to the resolution of a spatially calibrated scanning transmission electron image of the Si<110> lattice. An example of an image analyzed by this method is presented in Figure 1. Images like this provide many different light-to-dark transitions that provide a robust measurement of resolution. Here we use this definition of resolution to measure resolution at different values of defocus in different types of microscopy. Our method compares changes in resolution with changes in focus for different instruments, different imaging modes, and for the same instrument in different configurations. For example, characterizing optical microscope objectives of different magnifications and numerical apertures is accomplished by repeatedly imaging while changing the sample-objective distance. An identical region of these images containing many light-to-dark transitions is then statistically described at each stage height by the mean and standard deviation of the resolutions derived from each line fit in the region of interest (see Figure 2). This method can also be used to relate image resolution to defocus in electron and ion micrographs. In electron and ion microscopy, the focus of the objective lens is varied providing defocused images with varied reported working distance. A region of interest is then fit in an identical manner, allowing us to determine the effective depth of field for a specific configuration of apertures, detectors, and lens values (see Figure 2). References: [1] "Scanning electron microscopy and X-ray microanalysis. A text for biologists, materials scientists, and geologists." Goldstein, J. I.; Newbury, D. E.; Echlin, P.; Joy, D. C.; Fiori, C.; Lifshin, E., (Plenum Publishing Corporation, New York, New York). p 4 [2] E. Hecht in "Optics" (Addison and Wesley Inc., New York, New York) [3]AE Curtin, R Skinner, AW Sanders, Microscopy and Microanalysis 20 (2014), p 984 � 985 Microsc. Microanal. 21 (Suppl 3), 2015 1712 Figure 1. Measurement of resolution using a sigmoidal fit. (a) Spatially calibrated optical bright field image. This region of interest includes 248 light-to-dark transitions, each of which is fit separately resulting in the histogram in (b). (c) A single fit, showing the intensity profile (black), the resultant fit (blue), the tangent line at midpoint (green dashed line), and the residuals of the fit (red) . Figure 2. Resolution versus defocus in optical and scanning electron micrographs. Left, resolution as function of defocus for multiple optical microscope objectives in bright field reflection imaging. Right, histograms of analyzed field emission SEM images showing the change in resolution of an edge imaged at 5 kV with a 30 �m aperture, using an Everhart-Thornley detector at a working distance of 5 mm.");sQ1[855]=new Array("../7337/1713.pdf","Improved Spatial Resolution of EDX/SEM for the Elemental Analysis of Nanoparticles.","","1713 doi:10.1017/S1431927615009344 Paper No. 0855 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Improved Spatial Resolution of EDX/SEM for the Elemental Analysis of Nanoparticles. Johannes Mielke1, Steffi Rades1, Erik Ortel1, Tobias Salge2 and Vasile-Dan Hodoroaba1 1. BAM Federal Institute for Materials Research and Testing, Division 6.8 Surface Analysis and Interfacial Chemistry, 12200 Berlin, Germany. 2. Natural History Museum, Core Research Laboratories, Cromwell Road, SW7 5BD London, United Kingdom. The interest in nanoparticles remains at a high level in fundamental research since many years and increasingly, nanoparticles are incorporated into consumer products to enhance their performance. Consequently, the accurate and rapid characterization of nanoparticles is more and more demanded. Electron microscopy (SEM, TSEM and TEM) is one of the few techniques which are able to image individual nanoparticles. It was demonstrated recently that the transmission electron microscopy at a SEM can successfully be applied as a standard method to characterize accurately the size (distribution) and shape of nanoparticles down to less than 10 nm [1, 2]. In many cases a (nano)material consists of particles having a similar size (and form) with varying elemental composition. In this contribution, some case studies are presented, demonstrating the chemical sensitivity of modern EDX systems at the nano-scale, enabling a distinction between nanoparticles of different chemical composition. In order to reach the necessary high spatial resolution of the EDX signals, two prerequisites should be fulfilled: the operation of the transmission mode at SEM, i.e. preparation of the (nano)material on TEM grids, as well as the use of highly-sensitive EDS detectors with increased solid angle to compensate for the low signal-to-noise EDX spectra emitted by nanoparticles. Besides pure nanoparticles with a size well below 100 nm, and nanoparticle mixtures, also nanoparticles with a more complicated form and chemical composition such as core-shell nanoparticles are investigated by high-resolution TSEM/EDX [3]. Also the detection of contaminants/impurities becomes possible up to a certain scale. In order to check the attended EDX spatial resolution by using various high-sensitive EDS systems in conjunction with T-SEM, dedicated experiments have been carried out. The lack of reference nanomaterials with certified size and chemical composition is well-known. One of the few examples is the reference sample BAM-L200, a material with stripe structures of certified dimensions in the nano-range, specially designed for testing the spatial resolution of imaging micro/nano-probe techniques. This material has been prepared as a FIB lamella [4] and it was demonstrated that spatial resolutions of the EDX signals in the range of a few nm can be achieved. The need for and the availability of automatic image acquisition (including spectral information) with the goal of automatic chemical classification will be discussed. The coupled automatic data evaluation will be also addressed by some practical examples. Microsc. Microanal. 21 (Suppl 3), 2015 1714 References: [1] E Buhr, N Senftleben, T Klein et al., Meas. Sci. Technol. 20 (2009), 084025 (9 pp.). [2] V-D Hodoroaba, C Motzkus, T Mac� et al., Microsc. Microanal. 20 (2014), p. 602. [3] S Rades, V-D Hodoroaba, T Salge et al., RSC Adv. 4 (2014), p. 49577. [4] M Senoner, A Maa�dorf, H Rooch et al., Anal Bioanal Chem, DOI 10.1007/s00216-014-8135-7, in press. [5] The research leading to these results has received funding from the European Union's Seventh Framework Programme (FP7/2007-2013) under grant agreements n� 604347 (NanoDefine) and n� 263147 (NanoValid). Figure 1. Comparison between the EDX mapping capabilities (Si K: green, Ti K: red) of an SEM for a mixture of SiO2 and TiO2 nanoparticles on bulk graphite (a, b) and on a thin, electron transparent TEM grid (c, d). On the bulk substrate, the different chemical composition of the individual nanoparticles cannot be resolved due to the intense substrate background, whereas on the TEM grid, the EDX mapping (with the same EDS detector) shows a clear distinction between the different nanoparticle types, which is possible due to the largely reduced background intensity.");sQ1[856]=new Array("../7337/1715.pdf","What about Ionic Liquids as a "hot " Certified Reference Material Candidate to check Your EDS below 1 keV?","","1715 doi:10.1017/S1431927615009356 Paper No. 0856 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 What about Ionic Liquids as a "hot " Certified Reference Material Candidate to check Your EDS below 1 keV? Markus Holzweber1, Vasile-Dan Hodoroaba1 and Wolfgang E. S. Unger1 1. BAM Federal Institute for Materials Research and Testing, Division 6.8 Surface Analysis and Interfacial Chemistry, Unter den Eichen 44, 12200 Berlin, Germany. Energy-dispersive X-ray spectrometry (EDX) is one of the most applied methods used for the analysis of the chemical composition of solids and thin films. Recent progress in EDS (Energy Dispersive X-ray Spectrometer) technology has increased the general performance also in the energy range < 1 keV addressing low Z elements. Suitable test materials to be employed especially to check the low-energy EDS performance � also in line with ISO 15632 [1] - are rather limited and mainly based on C K and F K lines [2]. In order to obtain valid results in laboratories accredited in compliance with ISO/IEC 17025 [3] it is necessary to periodically check the instrument performance. Room temperature ionic liquids (RTIL) are composed of ions only and are defined as molten salts with a melting point below 100�C. [4] They have numerous interesting physical properties such as high thermal stability and electrical conductivity. Besides their low melting points they have also a very low vapor pressure, which enables their analysis even in ultra -high vacuum devices. In this contribution we explore the applicability of RTIL for the routine test check of the energy scale and the energy resolution. RTILs fulfil in principle the general requirements to qualify them as a reference material: RTILs can easily be prepared reproducibly to exhibit a smooth and flat surface, see Figure 1, show excellent lateral and in-depth homogeneity, are electrically conductive in the liquid state and are long-term stable at specific storage conditions (e.g. low temperature, moisture free). Additionally, the surface is regenerating itself. By varying the alkyl side chain length different relative elemental concentration can be obtained easily. As a proof�of-principle we investigated RTILs of the imidazolium bis(trifluoromethyl)sulfonylimide class with variable carbon content (i.e. different side chain length). We choose this class of RTILs since they are readily available in high purity and contain light elements (C, N, O, F) which offer several Xray characteristic K line peaks below 1 keV besides the S K line above 1 keV. The capability of an EDS system to detect S L series at 148.7 eV can also be evaluated. Examples of EDX spectra taken from an RTIL reference material candidate are shown in Figure 1. The differences in the energy resolution of the two EDS systems used here in the low-energy range are expressed by different half widths (FWHM) of the lines. This has an effect on the separation of the N K lines. Furthermore, direct comparisons of EDX spectra taken under identical conditions by EDS with different performances in the low energy range can quickly distinguish differences not only in the energy resolution but also in the spectrometer efficiency. At a later stage, this new candidate can be used as a sensitive test material to validate the quantification of X-ray lines below 1 keV. Microsc. Microanal. 21 (Suppl 3), 2015 1716 References: [1] ISO 15632 (2012) Microbeam analysis � Selected instrumental performance parameters for the specification and checking of energy-dispersive X-ray spectrometersEDX for use in electron probe microanalysisEPMA. ISO, Geneva. [2] V Rackwitz, M Krumrey, C Laubis et al., Anal Bioanal Chem, in press, DOI 10.1007/s00216-0148242-5. [3] ISO/IEC 17025 (2005) General requirements for the competence of testing and calibration laboratories. ISO, Geneva. [4] P Wasserscheid and T Welton, Ionic Liquids in Synthesis (2002) (Wiley-VCH, Weinheim). [5] This work is funded by the European Union through the European Metrology Research Program (EMRP). The EMRP is jointly funded by the EMRP participating countries within EURAMET and the European Union. M.H. is grateful for financial support by the Austrian Science Found (FWF) through the Erwin-Schr�dinger fellowship program (project number J 3471-N28). EDS #1 EDS #2 Figure 1: Example of a RTIL test material candidate measured with two different SDD EDS spectrometers under similar conditions. Note the superior energy resolution of the second EDS system (see the better separation of N K line). The excellent surface homogeneity can be taken from the SEM micrograph.");sQ1[857]=new Array("../7337/1717.pdf","`Rupture' Type Thrombopoiesis from Bone Megakaryocyte is Regulated by IL-1Alpha: Visualization by Two Photon Microscopy and Software Analysis","","1717 doi:10.1017/S1431927615009368 Paper No. 0857 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 `Rupture' Type Thrombopoiesis from Bone Megakaryocyte is Regulated by IL1Alpha: Visualization by Two Photon Microscopy and Software Analysis Satoshi Nishimura1,2,3 Akira Sawaguchi4 1. 2. Center for Molecular Medicine, Jichi Medical University, Tochigi, Japan Department of Cardiovascular Medicine, the University of Tokyo, Japan 3. Translational Systems Biology and Medicine Initiative, the University of Tokyo, Japan 4. Dept of Anatomy, Ultrastructural Cell Biology, Faculty of Medicine, University of Miyazaki, Japan snishi-tky@umin.ac.jp Blood platelets are generated in the bone marrow (BM) from their precursors, megakaryocytes (MK). Although we know that MKs produced many platelets throughout life, precisely how platelets are produced and how circulating numbers are tightly regulated in vivo remain uncertain, largely because of the rarity of MKs in the BM and the lack an adequate visualization technique. Thus, we utilized highspeed XYZT two photon imaging using GaAs detectors, and rapid resonance scanning mirrors for intravital visualization of BM MKs. This system enabled us to evaluate the detailed thrombopoietic processes from BM MKs. By visualizing living bone marrow in vivo, we identified that a second thrombopoietic process, `rupture'-like MK fragmentation, can be ongoing simultaneously with previously identified `proplatelet formation' in the same mouse BM. Rupture was identified as megakaryocyte death with small plateletlike particles into vessels. We defined these two types by quantifying three dimensional deformation and perimeter changes (Fig). It was also revealed that the `rupture' type produced 40 platelets from one MK in one minute, which was larger than that by proplatelet thrombopoiesis (only 2 from one MK). Large number platelet release processes in rupturing MKs take only few minutes, and easily escaped from conventional morphological evaluation methods. Careful examinations using image analysis software elucidated that platelets are released preferentially into vessel lumens during `rupture' type thrombopoiesis. Proplatelets were dominant thrombopoietic processes under steady state, which was regulated by thrombopoietin (TPO). Conversely, following blood loss, 5-FU administration, antibody-based platelet depletion or acute inflammation, there was accelerated release of large number platelets from MKs mediated by novel `rupture' type MK behaviors. Rupture was regulated by the interleukin-1 (IL-1)alphatype1 IL-1 receptor axis and ERK-dependent MK atypical apoptosis. It is known that proper microtubule assembly is vital for proplatelet formations in TPO-stimulated MKs, but did not take place in IL-1alpha-stimulated MKs due to uncoordinated expression tubulin. We also revealed the functional instability of plasma membrane after IL-1alpha treatment in MKs using FRAP analysis. AFM showed that IL-1alpha treated MK showed decreased "stiffness" and "contractile force", which reflect the tendency to rupture type thrombopoiesis. These findings support the ideas that IL-1alpha acts acutely as a platelet releasing factor, coordinating with TPO to dynamically modulate the cellular programming of MKs that regulates platelet counts. Microsc. Microanal. 21 (Suppl 3), 2015 1718 1706 Figure 1 References: [1] Nishimura S, et al. Cell Metabolism. 2013, 18, 759-766 [2] Nakamura S, Takayama N, ,,,, Nishimura S, Eto K Cell Stem Cell, 2014 in publication. [3] Nishimura S, et al. Blood. 2012 ;119(8):e45-56. [4] Takayama N, Nishimura S, et al. J Exp Med, 2010; 207(13):2817-2830 [5] Takizawa H*, Nishimura S*, et al. (*equal contribution) J Clin Invest., 2010, 120(1): 179-190. [6] Nishimura S, et al. Nature Medicine, 2009, 15:8, 914-920. [7] Nishimura S, S. et al. Progress in Biophysics and Molecular Biology, 2008;97:282-297. [8] Nishimura S, et al. J Clin Invest. 2008, 118(2): 710-721. [9] Nishimura S, et al. Diabetes. 2007,56:1517-1526.");sQ1[858]=new Array("../7337/1719.pdf","In Vivo Imaging and Software Analysis Revealed the Contribution of Endothelial Damage to Thrombus Development Processes","","1719 1719 doi:10.1017/S143192761500937X doi:10.1017/S143192761500937X Paper No. 0858 Microsc. Microanal. 21 (Suppl 3), 2015 Microsc. Microanal. 21 (Suppl 3), 2015 � � Microscopy Society of America 2015 Microscopy Society of America 2015 In Vivo Imaging and Software Analysis Revealed the Contribution of Endothelial Damage to Thrombus Development Processes Satoshi Nishimura1,2,3 Akira Sawaguchi4 Center for Molecular Medicine, Jichi Medical University, Tochigi, Japan Department of Cardiovascular Medicine, the University of Tokyo, Tokyo, Japan 3. Translational Systems Biology and Medicine Initiative, the University of Tokyo, Tokyo, Japan 4. Department of Anatomy, Ultrastructural Cell Biology, Faculty of Medicine, University of Miyazaki, Miyazaki, Japan 2. 1. The cellular mechanisms associated with cardiovascular events remains unclear, largely because of an inability to directly visualize and evaluate thrombus formation in living body. We developed in vivo imaging technique based on multi-photon microscopy and light-manipulation technique to reveal the multicellular processes during thrombus development. Additionally we developed software analysis system which can evaluate thrombotic processes from XYZT visuals, with high reproducibility. The software did not use `parameter' adjustment by observers, and completely free from researchers bias. We visualized the cell dynamics in single platelet levels, and assessed thrombus formation processes using three animal models. First, we induced rapidly developing thrombi composed of discoid platelets, which was triggered by ROS stimulation by photo-chemical reactions. In this model, thrombus development was mainly dependent on P-selectin and GPIIbIIIa activations. The functional properties including activation, aggregation, and fibrinogen binding were different between `newer' and `aged' platelets in developing thrombus. In the second model, thrombus formation was induced by endothelical cell disruption by laser irradiations. With the rapid recruitment of inflammatory leukocytes into damaged area, fibrin net formation and tissue regenerative changes were also observed. TLR4 signaling contributed to these steps, and pretreatmet of LPS markedly enhanced inflammatory reactions. In last model, spontaneous platelet aggregations were induced by transient ischemia and reperfusions. We revealed that the endothelial cell damage levels strictly determine the entire course of thrombotic processes, and our imaging system can evaluate the therapeutic strategies against them. Automatic software analysis can be powerful tools to give statistical evidence from XYZT visuals. Microsc. Microanal. 2121 (Suppl 3), 2015 Microsc. Microanal. (Suppl 3), 2015 1720 1708 Figure 1 References: [1] Nishimura S, et al. Cell Metabolism. 2013, 18, 759-766 [2] Nakamura S, Takayama N, ,,,, Nishimura S, Eto K Cell Stem Cell, 2014 in publication. [3] Nishimura S, et al. Blood. 2012 ;119(8):e45-56. [4] Takayama N, Nishimura S, et al. J Exp Med, 2010; 207(13):2817-2830 [5] Takizawa H*, Nishimura S*, et al. (*equal contribution) J Clin Invest., 2010, 120(1): 179-190. [6] Nishimura S, et al. Nature Medicine, 2009, 15:8, 914-920. [7] Nishimura S, S. et al. Progress in Biophysics and Molecular Biology, 2008;97:282-297. [8] Nishimura S, et al. J Clin Invest. 2008, 118(2): 710-721. [9] Nishimura S, et al. Diabetes. 2007,56:1517-1526.");sQ1[859]=new Array("../7337/1721.pdf","In vivo Three Photon Imaging of Neuronal Activities from Hippocampus in Intact Mouse Brain","","1721 doi:10.1017/S1431927615009381 Paper No. 0859 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In vivo Three Photon Imaging of Neuronal Activities from Hippocampus in Intact Mouse Brain Tianyu Wang1, Dimitre Ouzounov1, Nicholas G. Horton1, Jean C. Cruz Hern�ndez2, Danielle Feng2, Nozomi Nishimura2, and Chris Xu1 1. 2. School of Applied and Engineering Physics, Cornell University, Ithaca, NY, USA Department of Biomedical Engineering, Cornell University, Ithaca, NY, USA Optical imaging has enabled chronic observation of living mouse brain structure and function with single cell resolution [1]. Multiphoton microscopy (MPM) has significantly extended imaging depth to subcortical layers in highly scattering mouse brain [2-3]. Combined with genetically encoded calcium indicator GCaMP6s, we demonstrated that 3PM with 1350 nm excitation is capable of simultaneous recording of calcium transients in a neuron population in stratum pyramidal (SP) layer of cornus ammonis (CA1) region of hippocampus in intact mouse brain. The excitation source for the custom-built multiphoton microscope was an optical parametric amplifier (OPA) operated at 1350 nm and a repetition rate of 250 kHz. After dispersion compensation, laser pulse width was ~ 65 fs after the objective. We imaged brain of mouse under anesthesia through cranial window two weeks after injection of AAV2/1 encapsulated GCaMP6s. The imaging site was approximately 2 mm posterior and 2.5 mm lateral to the bregma point. With all overlaying cortical tissue intact, we imaged tissue structure in a stack from 700 m to 1250 m below brain surface. Fluorescence and third harmonic generation (THG) were simultaneously recorded. The THG signal delineates boundaries between deep cortical neurons, external capsule (EC), and hippocampus (Figure 1a). The maximum average power used to image SP layer in CA1 region of hippocampus was ~ 36 mW. We further recorded calcium activities from 25 SP layer neurons simultaneously for 8 minutes (Figure 1b). Neuronal activities were continuously recorded with 128x128 pixel frames at a frame rate of 6.1 Hz. The time resolution was 164 ms, which is adequate for GCaMP6s with a half decay time of 1.8 s for 10 action potentials (AP) [4]. Observed spike duration varies from l s to ~20 s. Relative fluorescence change (F/F0) of the spikes was as high as 700% (Figure 1c). The method demonstrated here enables in vivo functional imaging of subcortical neuron population with single cell resolution within intact mouse brain. We anticipate that this technology will have an impact on minimally invasive investigation of neuronal circuit deep within a mouse brain. References [1] J. N. Kerr and W. Denk, Nature Reviews. Neuroscience 9 (2008), p. 195-205. [2] N. G. Horton, et al., Nature Photonics 7, (2013), p. 205-209. [3] D. A. Dombeck, et al., Nature Neuroscience 13 (2010), p. 1433-1440. [4] D. S. Kim, et al., Neuroscience 2014 Short Course I: Advances in Multi-neuronal Monitoring of Brain Activity (2014) p. 12. Microsc. Microanal. 21 (Suppl 3), 2015 1722 Figure 1. (a) Image stack of GCaMP6s labeled neurons spanning from cortex to hippocampus (fluorescence colored in bright grey, THG in dark purple, 175 m x 175 m field-of-view (FOV), taken with increments of 2 m in depth, scale bar 50 m) (b) Neuronal population in the CA1 region of 1121m depth (0.21 frames/s, 20 averages, 512x512 pixels, scale bar 20 m) (c) Spontaneous Ca2+ transients recorded from the somata indicated in (b)");sQ1[860]=new Array("../7337/1723.pdf","Airy Beams for Light-sheet Microscopy","","1723 doi:10.1017/S1431927615009393 Paper No. 0860 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Airy Beams for Light-sheet Microscopy Jonathan Nylk1,2, Zhengyi Yang1, Miguel Preciado1, Michael Mazilu1, Tom Vettenburg3, Clara CollLlad�2, David E. K. Ferrier2, Tom�s Cizm�r4, Frank. J. Gunn-Moore2, Kishan Dholakia1 1. 2. SUPA, School of Physics and Astronomy, University of St Andrews, St Andrews, UK School of Biology, University of St Andrews, St Andrews, UK 3. Present address: Department of Bioengineering and Aerospace Engineering, Universidad Carlos III de Madrid, Madrid, Spain 4. Present address: School of Engineering, Physics and Mathematics, University of Dundee, Dundee, UK Light-sheet microscopy (LSM) is a promising technique for live imaging as it facilitates fast, highcontrast imaging of large volumes with minimal phototoxicity [1]. Variations on this imaging method are required to give high-resolution over large fields-of-view and to enable imaging at greater depths into specimens. A fundamental limitation of LSM with Gaussian beam illumination is the rapid divergence of the beam which prevents a uniformly thin light-sheet, thus impeding high-resolution imaging over a large volume. The use of Bessel beams for light-sheet illumination has grown in popularity because the inherent propagation-invariance of the Bessel beam can be utilized to produce uniformly thin light-sheets over an extended field-of-view [2] however, the extended transverse structure of the Bessel beam lowers the achievable axial resolution and must be combined with additional techniques such as multi-photon excitation, confocal scanning, or structured illumination (see, for example, [3]) to recover highresolution images. We present a number of innovations using the Airy beam as an alternative propagation-invariant beam for single-photon excitation LSM. The Airy beam also has an extended transverse structure but, rather than detract from image quality, this transverse structure facilitates high-contrast, high-resolution imaging when combined with a simple, one-dimensional deconvolution algorithm [4]. This technique allows for isotropic high-resolution imaging over a ten-fold larger field-of-view compared to a Gaussian light-sheet, and without additional, unnecessary irradiation of the specimen. Figure 1 compares images obtained using Gaussian, Bessel, and Airy light-sheets. Additionally, the broad distribution of energy across the Airy beam lowers peak power and reduces phototoxicity. Unlike the Bessel light-sheet, which results from an intrinsically two-dimensional pupil modulation and requires the Bessel beam to be scanned to make a light-sheet, an Airy light-sheet can be formed from a one-dimensional cubic phase modulation with a tilted cylindrical lens. This can lead to an inexpensive and accessible Airy light-sheet microscope [5]. An exponentially decreasing signal across the image can result from absorption and scattering of the light-sheet in large biological samples. Our latest approach shows that an attenuation compensating Airy beam can counteract the effects of absorption [6]. We apply this as a novel single-photon excitation approach to deliver a light-sheet deep into specimens without increasing the unnecessary irradiation of other parts of the specimen. The cylindrical pupil function required to produce an Airy light-sheet which compensates for linear attenuation is P(u,0)=exp(uu)*exp(2iu3), where u is the normalised pupil coordinate orthogonal to the light-sheet, determines the propagation-invariant length of the Airy light- Microsc. Microanal. 21 (Suppl 3), 2015 1724 sheet, and u is the strength of attenuation compensation. Figure 2 compares the image quality obtained in a strongly absorbing medium with and without attenuation compensated Airy light-sheets. In summary, we have shown the benefits of Airy beams for LSM. Airy beams can enable highresolution imaging over a large field-of-view and their broad transverse structure ensures low phototoxicity. The Airy beam can be easily generated from a tilted cylindrical lens, thus bringing the benefits of Airy beam LSM to a greater number of end users. The use of attenuation compensated Airy beams is also presented as a novel approach to obviate the effects of absorption and scattering in the sample. References: [1] J. Huisken, and D. Y. R. Stainer, Development 136 (2009), p. 1963-1975. [2] F. O. Fahrbach and A. Rohrbach, Opt. Express 18 (2010), p. 24229-24244. [3] L. Gao et al, Cell 151(6) (2012), p. 1370-1385. [4] T. Vettenburg et al, Nat. Methods 11 (2014), p. 541-544. [5] Z. Yang et al, Biomed. Opt. Express 5(10) (2014), p. 3434-3442. [6] M. A. Preciado, K. Dholakia, and M. Mazilu, Opt. Lett 39(16) (2014), p. 4950-4953. Figure 1. Comparison of various light sheet types imaging the dorsal end of the notochord in a fixed amphioxus (Branchiostoma lanceolatum). Nuclei have been fluorescently labelled with propidium iodide. y-axis and x-axis projections, respectively for the case of (a,b) Gaussian, (c,d) Bessel10, (e,f) Bessel5, and (g,h) Airy beam illumination. Bessel10 and Bessel5 denote Bessel beams with a propagation invariance of �21 and �42m from focus of a Gaussian beam respectively. Figure 2. Comparison of attenuation compensated Airy beam light-sheets. y-axis projections are shown for fluorescent microspheres embedded in an absorbing medium.");sQ1[861]=new Array("../7337/1725.pdf","Effect of Interfacial Structure on Gold-assisted Growth of Oxides","","1725 doi:10.1017/S143192761500940X Paper No. 0861 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Effect of Interfacial Structure on Gold-assisted Growth of Oxides Xin Li1, Wei Zhou1, Fang Liu1 and Guo-zhen Zhu1 1. State Key Laboratory of Metal Matrix Composites, School of Materials Science and Engineering, Shanghai Jiao Tong University, 800 Dongchuan Rd., Shanghai, 200240, P. R. China. Heterogeneous catalysts, e.g. Au/Fe2O3, Au/TiO2, and Au/MgAl2O4, have profound significance for many energy-related reactions such as low-temperature CO oxidation and selective oxidation of alcohols. [1-2] In spite of their technological and scientific importance, the underlying catalytic mechanism remains largely unclear in literature. Many plausible hypotheses are related to the Au/oxide interface, which either plays a role in controlling the morphology of Au nanoparticles or has a direct contribution to their catalytic behaviors. [3-5] Nevertheless, it is crucial to fully clarify the interfacial structure, including the atomic arrangement and chemical bonding nature. In order to facilitate structural characterization, we synthesized model systems with Au nanoparticles on a single-crystal oxide support by the presence of an Au overlayer and the application of heat. A new phenomenon, we have discovered recently in those systems, involves the dewetted gold nanoparticle acting as a seed causing crystalline substrates to spontaneously grow underneath the gold nanoparticle.[5] Fig.1 shows intricate nano-structures including dewetted gold nanoparticles supporting by the regrowth oxide bases. As shown in the HAADF (Z-contrast) images, gold nanoparticles, with well-defined facets on crystalline bases, share the same contrast with the substrate. The composition and the crystallographic orientation of the bases are further confirmed by the electron energy-loss spectroscopy and diffraction techniques, respectively. The formation of such intricate nanostructures depends on a few parameters such as the thickness of the Au overlayer and the heating treatment. Limited to the Au-TiO2 and Au-MgAl2O4 systems we have studied so far, the intricate nano-structures include: a). gold nanoparticles, normally close to the Wulff equilibrium shape, have a few preferential crystallographic orientations with respect to the single-crystal substrates; b). regrowth oxide bases, with a well-defined shape, are epitaxially aligned with the single-crystal substrates; and c). interfacial monolayers, which have completely different atomic arrangement, extends both the gold and oxide lattices. The formation of such interfacial monolayers always accompanies with the regrowth of the substrates. Those unique interfacial structures may be responsible for the epitaxial growth of the substrate under self-organized gold nanoparticles. The interfacial monolayers referred as rearranged Au atoms between the gold and oxide lattices. We detected an interfacial bilayer between the gold nanoparticles and spinel bases, labeled by the red and blue arrows in Fig.2. On top of monolayer A with Au atoms extending the spinel lattice, the monolayer B exhibits an oscillating contrast in the HAADF image when viewed from [110]MgAl2O4. In the case of Au-TiO2 with much large lattice mismatch, different interfacial structures have been observed with different crystallographic orientations between the gold and rutile lattice. Fig.3 shows the interfacial monolayers varies from single monolayer to a few monolayers viewed from [001]TiO2. The detailed atomic arrangement and chemical bonding nature of these interfacial monolayers are currently being investigated by a scanning TEM. The clarification of such structures provides deep insights in the understanding of the abnormal growth of the substrate. We believe that such phenomenon Microsc. Microanal. 21 (Suppl 3), 2015 1726 may shed the light on understanding the catalytic behavior of gold-oxide heterogeneous systems. References: [1] M. Haruta, S. Tsubota et. al., J. Catal. 144 (1993) p.175. [2] C.H. Christensen, B. J�rgensen et al., Angew. Chem. Int. Ed. 45 (2006) p.4648. [3] N. Lopez and J. K. N�rskov, J. Am. Chem. Soc. 124 (2002) p.11262. [4] A.A. Herzing, C.J. Kiely, et. al., Science, 321 (2008) p.1331. [5] J.A. Rodriguez, S. Ma, et. al., Science, 318 (2007) p.1757. [6] G.-z. Zhu, T. Majdi, et. al., Appl. Phys. Lett. 105(2014) p.231607 [7] The authors acknowledge funding from the National Natural Science Foundation of China. Part of the microscopy work was carried out at Canadian Centre for Electron Microscopy (CCEM). Figure 1. Intricately shaped gold-oxide nanostructures. (a) The SEM image of Au-TiO2 nanostructure. The red arrows show the crystalline oxide bases supporting Au nanoparticles. (b) and (c) are the HAADF (Z-contrast) images presenting the cross-sectional view of the intricate nanostructure in AuMgAl2O4 and Au-TiO2 systems, respectively. Figure 2. The Au-MgAl2O4 interfacial bilayer. Figure 3. The Au-TiO2 interfacial bilayer. (a) and (b) show interfacial monolayers with different gold nanoparticles, respectively.");sQ1[862]=new Array("../7337/1727.pdf","Quantifying and Correlating the Composition and Conductivity of Grain Boundaries in Ca-doped CeO2 Electrolytes, an AC-STEM EELS Study","","1727 doi:10.1017/S1431927615009411 Paper No. 0862 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantifying and Correlating the Composition and Conductivity of Grain Boundaries in Ca-doped CeO2 Electrolytes, an AC-STEM EELS Study W.J. Bowman, C.A. Hernandez, K. McGuinness, P.A. Crozier School for the Engineering of Matter, Transport and Energy, Arizona State University, 501 E. Tyler Mall, Tempe Arizona 85287-6106 Grain boundaries in polycrystalline oxygen ion conductors such as Y-stabilized ZrO2 (YSZ) and doped CeO2 are highly resistive at low and intermediate temperatures (<550 �C), which degrades the total ionic conductivity of the bulk material [1-3]. High resistivity relative to grain interiors is widely believed to be the result of an intrinsic space charge potential barrier which emanates from grain boundary cores several nanometers into grains, creating a region of charge carrier depletion--in this case oxygen vacancies [1,2]. The magnitude of the potential barrier, and thus the width of the affected space charge zone are thought to be closely related to the atomic structure and elemental composition within nanometers of the boundary core. However, it is well known that there is significant diversity in the interfacial atomic structure of grain boundaries in a given material, and recent AC-STEM EELS measurements by our group suggest that there is also considerable variation in boundary core composition. Here, we employ ACSTEM imaging and nanospectroscopy to quantify elemental composition of grain boundaries in a series of Ca-doped cerias, and correlate these findings with electrical conductivity data. Imaging and nanospectroscopy were performed using a probe-corrected JEOL ARM200F S/TEM. Atomic resolution ADF STEM images were correlated with core-loss EELS near-edge fine structure to characterize electronic structure and to quantify the local ionic concentrations in the vicinity of grain boundaries in a series of CaxCe1-xO2- with 0.02 x 0.10. Ceria nanopowders--prepared in the authors' lab using a spray drying (rapid solution evaporation) technique, underwent traditional ceramic processing to yield bulk polycrystalline samples from which electrical conductivity data and TEM specimens could be extracted. Quantitative nanospectroscopy was aided by the use of standard specimens used to experimentally determine proper Ca:Ce and O:Ce k-factors (scattering cross-section ratios, [4]). For instance, a Ca:Ce EELS k-factor was determined from simultaneous EELS/EDX measurements whereby the Ca:Ce EDX k-factor (derived from Ca:Ce standard, see fig. A) was used to identify the Ca:Ce molar concentration ratio. To address plural scattering, EELS spectral integration windows of 80 eV were used. AC-STEM data were correlated with macroscopic electrical conductivity measured using AC impedance spectroscopy, which discerns grain interior and boundary conductivities. As seen in fig. B, energy-loss spectra acquired at grain boundary cores differed significantly from those acquired away from the boundary in grain interiors. The boundary core Ca concentration was typically greater than twice that of the grain interior, indicating Ca segregation to grain boundaries which was accompanied by Ce depletion. Figures C and D show representative 2D elemental maps derived from EELS spectrum images, and illustrate the very significant variability in cation concentration in the vicinity of the boundary. The relative intensity of the Ce-M4,5 white lines was also found to vary within several nanometers of boundary cores, indicating local reduction of Ce4+ to Ce3+. The O-K near edge fine structure was also variable with distance and contained features indicative of the presence of oxygen vacancies in and around boundary cores. These electronic structure data, the quantification of ion concentrations, and correlation with grain boundary electrical conductivity will be discussed. Microsc. Microanal. 21 (Suppl 3), 2015 1728 References and acknowledgements 1. X. Guo et al. Electrochem. Soc. 148 (3) E121-E126 (2001) 2. H.J. Avila-Paredes et al. Sol. State. Ionics. 177 3075-3080 (2006). 3. W.J. Bowman et al. Sol. State Ionics. 272 9-17 (2015). 4. R.F. Egerton. 3rd Ed. Springer. 5. C.A.H. and K.M. wish to thank the Fulton Undergraduate Research Initiative at ASU for generous financial support throughout this work. W.J.B. would like to acknowledge the National Science Foundation's Graduate Research Fellowship (DGE-1211230) for continued financial support. Finally, we gratefully acknowledge support of NSF grant DMR-1308085 and ASU's John M. Cowley Center for High Resolution Electron Microscopy. Fig. A: Background-subtracted EDX spectra acquired from Ca:Ce standard and 10 mol% Ca with highlighted integration windows. Fig. B: Unprocessed EELS data acquired on and off a grain boundary in 2 mol% Ca. Inset are corresponding low-loss plasmon signals. Fig. C & D: AC-STEM ADF images of grain boundaries in 5 mol% and 10 mol% specimens with overlaid cation maps derived from 2D EELS spectrum images. Fig. C inset shows the Ca signal profile across the boundary.");sQ1[863]=new Array("../7337/1729.pdf","Aberration-corrected STEM of Highly Loaded Pt1/NiO Single-atom Catalysts: Structural and Catalytic Stability","","1729 doi:10.1017/S1431927615009423 Paper No. 0863 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Aberration-corrected STEM of Highly Loaded Pt1/NiO Single-atom Catalysts: Structural and Catalytic Stability S. B. Duan1, 2 R. M. Wang2 and J. Y. Liu1 1. 2. Department of Physics, Arizona State University, Tempe, Arizona 85287, USA Department of Physics, Beihang University, Beijing 100191, China Supported precious metal catalysts play a significant role in many chemical transformations including energy generation and control of toxic emissions. Precious metals are, however, expensive and thus efficient usage of precious metals is of critical importance for practical applications. The development of single-atom catalysts (SACs), which maximize the efficiency of costly active components of supported metal catalysts, is important for both fundamental studies and industrial applications [1-3]. The issue with the current SACs is that they usually consist of extremely low loading levels of metal, resulting in low specific activity and conversion rate. High loading levels of isolated single metal atoms are not stable and sinter to larger particles during catalytic reactions. Stabilization of single metal atoms by anchoring sites, for example, surface/subsurface defects, hydroxyl groups, and structural stabilizers, becomes imperative for further development of practical SACs. The high loading Pt SACs were synthesized by a modified adsorption method. Briefly, H2PtCl6 solution was used as the Pt precursor and was mixed with the pre-formed NiO nanocrystallites. The nominal loading of the Pt metal was 2.0 wt%. The resultant precipitate was filtered, washed, dried and calcined at 400�C for 5 hours (denoted as 2Pt1/NiO). The catalytic performance of the synthesized catalysts for CO oxidation was evaluated in a fixed-bed reactor. The feed gas composition was 1vol% CO + 1vol% O2 and balance He with a flow rate of 33 ml/min, and 80 mg catalyst was directly used without reduction. The outlet gas compositions were on-line analyzed by a gas chromatograph and the CO conversion was calculated based on the inlet and outlet CO concentrations. Aberration-corrected high-angle annular dark-field (AC-HAADF) microscopy was used to characterize the synthesized and used catalysts. The low magnification image of Figure 1a clearly shows that there were no Pt particles in the assynthesized 2Pt1/NiO catalyst. The atomic resolution image of Figure 1b reveals only isolated Pt single atoms uniformly dispersed onto the surfaces of NiO nanocrystallites. By analyzing many such low and high magnification images of various regions of the as-synthesized catalyst, we unambiguously concluded that the as-synthesized 2Pt1/NiO catalyst contained only isolated single Pt atoms. Figure 2 displays CO conversion versus temperature and time-on-stream for the 2Pt1/NiO catalyst. Except the initial increase of the CO conversion (Figure 2b), the 2Pt1/NiO did not deactivate at all during the 1,000 minutes test for CO oxidation reaction at 500K. Figure 1c and 1d display the AC-HAADF images of the used catalyst, which contained only individually isolated Pt single atoms. Thus we have developed a stable, high number density Pt1/NiO SAC for CO oxidation. Similar synthesis strategies will be used for developing other types of SACs for various types of catalytic reactions. References: [1] [2] [3] [4] Qiao, B. et al, Nature Chem. 3, (2011), p. 634. Moses-DeBusk, M., et al, J Am. Chem. Soc. 135, (2013), p. 12634. Yang, M., et al, Science 346, (2014), p. 1498. The authors acknowledge the College of Liberal Arts and Sciences of Arizona State University for funding and the use of the John M. Cowley CHREM facilities at Arizona State University. Microsc. Microanal. 21 (Suppl 3), 2015 1730 Figure 1. Low and high magnification AC-HAADF images of (a, b) fresh 2Pt1/NiO SAC and (c, d) after CO oxidation for 1,000 minutes at 500K, revealing that the 2Pt1/NiO catalyst consisted of only isolated single Pt atoms which were stable during the CO oxidation reaction. 2Pt1/NiO WHSV = 24750 mlh-1 g-1 cat 2Pt1/NiO WHSV = 24750 mlh-1 g-1 cat T = 500K Figure 2. CO conversion vs (a) temperature and (b) time-on-stream (T = 500K) for the 2Pt1/NiO SAC, clearly demonstrating the stability of the as-synthesized 2Pt1/NiO SAC.");sQ1[864]=new Array("../7337/1731.pdf","Aberration-Corrected STEM and Tomography of Pd-Pt Nanoparticles: Core-Shell Cubic and Core-Frame Concave Structures","","1731 doi:10.1017/S1431927615009435 Paper No. 0864 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Aberration-Corrected STEM and Tomography of Pd-Pt Nanoparticles: Core-Shell Cubic and Core-Frame Concave Structures Ning Lu1, Jinguo Wang1, Shuifen Xie2, Jacob Brink3, Kevin McIlwrath3, Younan Xia2,4 and Moon J. Kim1 1. Department of Materials Science and Engineering, The University of Texas at Dallas, Richardson, Texas 75080, United States 2. The Wallace H. Coulter Department of Biomedical Engineering, Georgia Institute of Technology and Emory University, Atlanta, Georgia 30332, United States 3. JEOL USA, Inc., Peabody, Massachusetts 01960, United States 4. School of Chemical Biochemistry, Georgia Institute of Technology, Atlanta, Georgia 30332, United States Nanosize bimetallic particles have long been recognized as important heterogeneous catalytic materials. Platinum (Pt) is an inestimable material due to its excellent catalytic performance. To reduce the use of Pt and lower overall cost, depositing a thin layer of Pt on the surfaces of less expensive metal cores, such as Palladium (Pd), is an effective approach. Controlling the shape, morphology, and chemistry of the outer Pt layer has been widely accepted as a critical factor to realize the specific applications. Our studies show that the growth shape of nanocrystals can be controlled by adjusting the ratio between the rates of atom deposition and surface diffusion (V deposition/Vdiffusion).[1] Herein, we provide a comprehensive study of the atomic structures, composition, and 3D morphology of the Pt shell layer in Pd-Pt nanoparticles with core-shell cubic and core-frame concave structures by aberration (Cs) corrected high angle angular dark field (HAADF) - scanning transmission electron microscopy (STEM), energy dispersive X-ray spectroscopy (EDS), and HAADF tomography. Core-shell cubic and core-frame concave Pd-Pt nanoparticles were synthesized by site-specific growth.[2] TEM, HAADF-STEM, EDS mapping, and HAADF tomography were performed in a JEOL ARM200F with a STEM Cs corrector operated at 200 kV. A Fischione 2030 ultra-narrow gap tomography holder was used in the tilt range of � 72� with 2� step. A low dose mode at 1 M� magnification was used via SerialEM software. The 3D reconstructions were achieved by simultaneous iterative reconstruction algorithm (SIRT) of 2D projections in IMOD. Visualization was performed with AMIRA 3.1. Atomic resolution Cs-corrected HAADF imaging combined with EDS line scanning can identify the atomic level differences in layer-by-layer thickness.[3] Figure 1 shows the results of EDS line scanning indentifies Pt atomic layers in Pd-Pt core-shell nanocubes. Figure 1a is an atomic HAADF image of a Pd-Pt core-shell nanocube, where the red arrow indicates the direction of EDS line scanning. The intensity profiles (Figure 1b) shows outer 7 atomic layers have higher intensity than inner layers, implying that the 7 atomic layers are Pt, based on HAADF imaging contrast theory. The corresponding line profile of EDS scanning (Figure 1c) confirms the structure of Pt shell/Pd core. The distance between half-intensity of Pt EDS signal is about 1.25 nm, which is in good agreement with the measured peak distance between the 7 Pt layers (1.22 nm) in Figure 1b. It is worth noting that the most outset surface Pt layer is not fully covered, while the interfacial layer is a mixed Pt and Pd layer at the atomic scale due to the termination of growth steps under a two-dimensional growth mechanism. 3D electron tomography Microsc. Microanal. 21 (Suppl 3), 2015 1732 1720 confirmed that the Pd-Pt core-frame concave nanoparticle has high index crystallography facets on its surfaces, as shown in Figure 1d. In this work, we investigated the structural and chemical characteristics of the Pt shell layer in the coreshell cubic and core-frame concave Pd-Pt nanoparticles at the atomic scale. Our results reveal the deposition and growth mechanism of Pt ad-atoms onto Pd cubic seeds under various growth conditions. The characterization methodology developed in this work provides important insights that will serve to guide the analyses of nano and atomic structures, and the design of Pd-Pt nanoparticles for specific catalytic functionality.[4] References: [1] X. H. Xia, S. F. Xie, M. C. Liu, H. C. Peng, N. Lu, J. G. Wang, M. J. Kim and Y. N. Xia, Proceedings of the National Academy of Sciences USA 110 (2013), 6669�6673. [2] N. Lu, J. G. Wang, S. F. Xie, J. Brink, K. McIlwrath, Y. N. Xia and M. J. Kim, The Journal of Physical Chemistry C 118 (2014), 28876�28882. [3] S. F. Xie, S. I. Choi, N. Lu, L. T. Roling, J. A. Herron, L. Zhang, J. Park, J. G. Wang, M. J. Kim, Z. Xie, M. Mavrikakis and Y. N. Xia, Nano letters 14 (2014), 3570-3576. [4] This work was supported in part by Louis Beecherl, Jr. endowment funds. The synthesis work was supported in part by a grant from the NSF (DMR-1215034) and start-up funds from Georgia Institute of Technology. Figure 1. EDS line scanning indentifies Pt atomic layers in Pd-Pt core-shell nanocube (a-c) and 3D tomography of Pd-Pt core-frame concave nanoparticle (d). (a) Atomic HAADF image. (b) Intensity profiles along the red lines indicated in (a). (c) EDS line scan profiles of Pt and Pd acquired along the red lines indicated in (a). (d) 3D tomography of Pd-Pt core-frame concave nanoparticle.");sQ1[865]=new Array("../7337/1733.pdf","Aberration-corrected STEM Study of Atomically Dispersed Pt1/FeOx Catalyst with High Loading of Pt","","1733 doi:10.1017/S1431927615009447 Paper No. 0865 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Aberration-corrected STEM Study of Atomically Dispersed Pt1/FeOx Catalyst with High Loading of Pt Botao Qiao1,2, Aiqin Wang2, Tao Zhang2 and Jingyue Liu1 1. 2. Department of Physics, Arizona State University, Tempe 85281, USA. State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Chinese Academy Sciences, Dalian 116023, China Single-atom catalysis is emerging as a new frontier in the field of catalysis [1]. It not only maximizes the atom efficiency of the metal (especially the expensive and scarce noble metal) in supported catalysts but also possesses the potential to bridge heterogeneous and homogenous catalysis. Furthermore, with single atoms as active centers, theoretical predictions and correlations to experimental data become possible. Single-atom catalyst (SAC) systems have been successfully developed and proved to be effective for various reactions [1-3]. However, most of these SACs were loaded with very small amount of metal since high-loading levels usually result in the formation of metal nanoclusters/particles. Fabrication of SACs with high metal loading remains a challenge. The fundamental study of the synthesis processes of SACs and the effects of the post-treatment on the properties of SACs are both of fundamental interest and practical significance. We report here the synthesis of high metal loading Pt1/FeOx catalyst and how the post-treatment affected the Pt dispersion. In our previous work [4-5], we found that for the Pt/FeOx catalyst with same preparation method (coprecipitation (CP) method) and post-treatment procedure (calcination at 400 oC for 5 hours and reduction at 200 oC for 0.5 hours), the 0.17 wt% of Pt loading resulted in SAC while a ~2.0 wt% loading of Pt resulted in the formation of nanoparticles with diameters of 1-2 nm and subnano clusters with sizes < 1 nm. Clearly, the high Pt loading facilitates the formation of larger particles or clusters. However, it is not clear how these cluster/particle species are formed: They could form during the co-precipitation process or during the calcination process since it is generally accepted that calcination could result in sintering of highly dispersed nano species [6]. The metal particles could also form during the activation processes of the catalysts since it has been reported that Pt could sinter under reduction while re-disperse under calcination [7]. In this work, we performed a systematic study, by aberration-corrected STEM, of a 2.0 wt% Pt/FeOx catalyst and examined how the different post-treatments affect the individually isolated Pt atoms supported on high-surface-area FeOx. The preliminary results showed that Pt existed as isolated single atoms under calcination at elevated temperatures as high as 600 oC. The reduction treatment, however, resulted in the movement of isolated single Pt atoms to form subnano clusters and nanoparticles. The 2.0 wt% Pt/FeOx was synthesized by the CP method as previously reported [4, 5]. Typically, a suitable amount of H2PtCl4� 2O and Fe(NO3)3� 2O mixture solution was dropwise added into 6H 9H o Na2CO3 solution at 50 C. After being stirred and aged for 2 hours, respectively, the sample was filtered and washed with deionized water for several times to remove the residual Cl -. The resultant precipitate was dried at 60 oC for 5 hours and the catalyst was denoted as 2Pt/FeOx. The 2Pt/FeOx catalyst was further calcined at 600 oC for 5 hours (denoted as 2Pt/Fe2O3-C600) or directly reduced at 200 oC for 1 hour under 5 vol% H2/He (denoted as 2Pt/FeOx-R200). Figure 1 displays representative STEM highangle annual dark-field (HAADF) images of the 2Pt/FeOx catalyst, clearly revealing isolated Pt atoms dispersed on/in the FeOx support with high number density (Figure 1a and 1b). The low magnification Microsc. Microanal. 21 (Suppl 3), 2015 1734 1722 image (Figure 1c) does not show any Pt nanoparticles/clusters, suggesting that even with 2.0 wt% Pt loading it is possible to form atomically dispersed SAC. The high magnification image (Figure 1d) shows that Pt atoms occupy the locations of the Fe sites. After being calcined at 600 oC, the Pt still existed as isolated single atoms (Figure 2a and b), suggesting that the SAC is extremely stable under oxidation conditions. However, when the as-synthesized 2Pt/FeOx was reduced at 200 oC for 30 minutes most isolated Pt single atoms disappeared and small Pt clusters were formed (Figure 2c and 2d), suggesting that under reduction environment the atomically dispersed Pt single atoms sintered, presumably due to the change of the surface chemistry and/or structure of the FeOx support. Reference: [1] [2] [3] [4] [5] [6] [7] [8] X.-F. Yang et al., Acc. Chem. Res. 46 (2013), p. 1740. M. Flytzani-Stephanopoulos and B.C. Gates, Annu. Rev. Chem. Biomol. Eng. 3 (2012), p. 545. M. Flytzani-Stephanopoulos, Acc. Chem. Res. 47 (2013), p. 783. B. Qiao et al., ACS Catal. 4 (2014), p. 2113. B. Qiao et al., Nat. Chem. 3 (2011), p. 634. W.Z. Li et al., Nat. Commun. 4 (2013), p. 2481. Y. Nagai et al., Angew. Chem. Int. Ed. 47 (2008), p. 9303. This research was funded by Arizona State University. We gratefully acknowledge the use of facilities in the John M. Cowley Center for High Resolution Electron Microscopy at Arizona State University. a b c d 5 nm 5 nm 20 nm 2 nm Figure 1. STEM-HAADF images of the as-synthesized 2Pt/FeOx catalyst show uniform distribution of isolated Pt atoms (a, b), no presence of Pt nanoparticles (c) and Pt occupancy of the Fe sites (d). a b c d 2 nm 5 nm 10 nm 5 nm Figure 2. STEM-HAADF images of 2Pt/Fe2O3-C600 (a, b) and 2Pt/FeOx-R200 (c, d) show isolated Pt atoms at elevated temperatures under calcination condition (a, b) and formation of Pt clusters/particles after reduction treatment (c, d).");sQ1[866]=new Array("../7337/1735.pdf","Multislice HAADF STEM Image Simulation of Complex Oxidation Catalyst "M1".","","1735 doi:10.1017/S1431927615009459 Paper No. 0866 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Multislice HAADF STEM Image Simulation of Complex Oxidation Catalyst "M1". Douglas A. Blom1 NanoCenter and College of Engineering and Computing, University of South Carolina, Columbia, SC, USA Annual worldwide acrylonitrile production is nearly 1 kg/person [1]. Currently, production is via the "SOHIO process" originally developed in the 1950s [2]. Propylene precursor is reacted over a multiphase catalyst in the presence of ammonia. For both economic and energy use reasons, there is a strong desire to produce acrylonitrile using propane as the starting material. This requires the development of a new catalyst. The most promising catalyst for the direct ammoxidation of propane to acrylonitrile is a quatenary Mo-V-Nb-Te oxide phase originally reported by the Mitsubishi Chemical Co. and therefore typically referred in the literature as "M1". The crystal structure was originally solved with the combined Rietveld refinement of synchrotron X-ray and neutron powder diffraction data [3]. More recently, the model was improved due to input from HAADF STEM observations [4]. Figure 1 is the structural model of the cations in this phase in a {001} orientation. The unit cell consists of a series of pentagonal, hexagonal and heptagonal rings of metal-oxygen octahedra. The center of the pentagonal ring is populated by the Nb in the crystal. Te and oxygen chains are present in both the hexagonal and heptagonal channels. The other 10 crystallographically distinct cation sites are populated by various mixtures of Mo and V. Previously, we have published the results of multi-slice ADF STEM image simulations on this material [5]. The simulations were performed for the older structural model from [3]. In addition, the virtual crystal approximation (VCA) was used in the published simulations. The projected potential of the cation sites with partial substitution of Mo and V in the VCA is the weighted average of the potentials of Mo and V. For STEM, the ADF signal is sensitive to not only the average mass of the cations, but also the location of the cations along the beam propagation direction [6]. In this work, we report the results of multi-slice ADF STEM image simulations for the improved structural model [4] without the VCA. Frozen phonon simulations using the multislice code of Kirkland [7] were performed. An accelerating potential of 200kV, convergence semi-angle of 17.3 mrad, Cs 3�m, C5 0, defocus 2nm were the common microscope parameters. The ADF detector in the simulation spanned 100-425 mrad. 64 phonon passes were used. The simulations were carried out for an area slightly larger than � of the unit cell which contained all the cation sites for the structure (see Fig. 1). 30 unit cells along the c axis (12nm) was the simulated thickness. The atomic coordinates, thermal parameters, and occupancy were taken from [4]. Several simulations with the same occupancy of Mo and V but different configurations along the beam propagation direction were run. A random solid solution was assumed. Figure 2 is a graphical representation of the cation distribution along the beam direction was for 13 independent cation sites for a single configuration. The intensity is equal to the atomic number of the cation for all 30 possible locations The last two columns are the Te containing sites that are either empty or Te. Integrated intensities for the ADF simulation of the S1 site (30% V) is shown in Fig. 3 for 7 different configurations. The image intensity varies by 9% relative to the mean depending on the cation distribution [8]. References: 1. Microsc. Microanal. 21 (Suppl 3), 2015 1736 [1] http://www.prweb.com/releases/2014/04/prweb11740156.htm retrieved 02/19/2015. [2] http://www.acs.org/content/acs/en/education/whatischemistry/landmarks/acrylonitrile.html retrieved 02/19/2015. [3] P. DeSanto et al., Z Kristallogr. 219 (2004), p.152. [4] X. Li et al., Top Catal. 54 (2011), p.614. [5] D. A. Blom, Ultramicroscopy, 112 (2012), p.69. [6] E. Carlino and V. Grillo, Phys Rev B, 71 (2005), p. 235. [7] E.J. Kirkland, Advanced Computing in Electron Microscopy, 2nd ed. Springer, New York, 2010.. [8] The author acknowledges funding from the National Academies Keck Future Initiative. This work used the Extreme Science and Engineering Discovery Environment (XSEDE), which is supported by National Science Foundation grant number ACI-1053575 Figure 1. Cation positions in the 001 projection of the Mo-V-Nb-Te oxide M1 catalyst. Color correspond to crystallographically distinct atom positions. The box shows the area of the image simulations. Figure 3. Integrated ADF intensity for cation site 1 for 7 different orderings of V and Mo along the beam direction. The composition is constant, but the ADF intensity varies by 9%. Figure 2. Graphical representation of the cation distribution along the electron propagation direction. Each column corresponds to one of the unique cation sites in the unit cell. The intensity is equal to the atomic number of the cation.");sQ1[867]=new Array("../7337/1737.pdf","Structural and Optical Characterization of Biosynthesized CdS Quantum Dots","","1737 doi:10.1017/S1431927615009460 Paper No. 0867 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Structural and Optical Characterization of Biosynthesized CdS Quantum Dots Zhou Yang1, Li Lu2, Victoria F. Berard1, Christopher J. Kiely2, Steven McIntosh1 and Bryan W. Berger1 1. Department of Chemical and Biomolecular Engineering, Lehigh University, Bethlehem, PA 18015, USA. 2. Department of Materials Science and Engineering, Lehigh University, Bethlehem, PA 18015, USA. Quantum dots (QDs) are semiconductor nanocrystals that possess size-dependent photoluminescent properties due to quantum confinement effects. They have potential applications in many fields, including cancer cell tracking, LED lighting, and solar cells. Conventionally, QDs are producedin organic solvents via multi-step, batch processes at high temperatures and pressures, which results in limited production scale and high cost [1].We have developed a novel, scalable approach for the biosynthesis of water-soluble CdS QDs using a strain of Stenotrophomonas maltophilia (SMCD1) to control particle size and hence QD photoluminescence through variation of growth time.Aberration-corrected scanning transmission electron microscopy (AC-STEM) and high-resolution electron microscopy (HREM) have provided us with valuable insight into the crystal structureand particle size of the biosynthesized CdS QDs. The ability of prokaryotes like S. maltophilia to transform toxic levels of heavy metals (e.g.cadmium) into insoluble metal precipitates has inspired the pursuit of a range of biological approaches to synthesize QDs. S. maltophilia was isolated from soil and iteratively selected for variants in culture that were tolerant to cadmium acetate at concentrations in excess of 1 mmol [2]. The selected cadmium-tolerant colonies were cultured in M9 minimal media containing L-cysteine and cadmium acetate. From the measured photoluminescence and absorption spectra, we confirmed the existence of fluorescent CdS QDs with a well-defined first excitonic peak. In addition, the reproducibility of the biosynthetic protocol was confirmed as different batches of growth culture under the same growth conditionsresulted in almost identical QD optical properties. Photoluminescence was retained in the culture supernatants after removal of the cells by centrifugation, indicating extracellular production of CdS QDs. We also found that the optical properties of the QDs correlated with growth time in culture (Fig. 1). The peaks of the absorption spectra (Fig. 1(b)) and the fluorescence spectra (Fig. 1(c)) both shifted to higher wavelengths with increasing growth time. This trend impliesa gradual increase in the mean QD size with growth time. These results are most consistent with a sequential mechanism involving an initial rapid QD nucleation stage, followed by continuous particle growth in culture. AC-STEM was utilized to identify the crystal structure of the biosynthesized CdS QDs, since X-ray and electron diffraction were unable to distinguish the two possible polymorphs of CdS (i.e. zinc-blende and wurtzite) due to the similar lattice spacings of the two structures and the nanoscopic nature of the QDs.Highlylocalized crystallographic information from individual particles could be obtained via high-angle annular dark field (HAADF) imaging in STEM mode, enabling us to measure the d-spacings and interplanar angles and compare them against the two CdS polymorphs. Fig. 2 shows examples of two CdS QDs whose lattice images match with the zinc-blende (Fig. 2(a)) and wurtzite (Fig. 2(b)) structures respectively. Thus we conclude that both zinc-blende and wurtzite structures co-exist for CdS QDs produced via strain SMCD1. To directlydetermine particle size distributions of the QDs, Microsc. Microanal. 21 (Suppl 3), 2015 1738 HREMimaging was used to preferentially pick up the periodicities in nanocrystals with well-defined particle edges. The mean particle size was found to increase with growth time, and could be correlated with the evolution of the optical spectra with increasing growth time [3]. References: [1] K. Sanderson, Nature 459 (2009), p. 760. [2] C. Bollet, A. Davin-Regli and P. De Micco, Appl. Environ. Microbiol. 61 (1995), p. 1653. [3] The authors acknowledge funding from National Science Foundation. Figure 1.Optical properties of purified, biosynthetic CdS QDs of different sizes. (a) Photograph of the culture supernatants from strain SMCD1 collected at various growth times (30-360 min) when illuminated under UV light; (b) UV-vis absorption spectra of CdS QDs as a function of culture growth time; (c) Fluorescence emission spectra using 350 nm excitation of CdS QDs as a function of culture growth time. Figure 2.HAADF images of individual CdS particles after 60 min of growth exhibiting (a) zinc-blende and (b) wurtzite structures respectively.");sQ1[868]=new Array("../7337/1739.pdf","Aberration-corrected STEM of Cross-sectional View of Core-shell Nanowires Prepared by Ultramicrotomy","","1739 doi:10.1017/S1431927615009472 Paper No. 0868 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Aberration-corrected STEM of Cross-sectional View of Core-shell Nanowires Prepared by Ultramicrotomy J. Xu1 and J. Y. Liu2 1. School for Engineering of Matter, Transport and Energy, Arizona State University, Tempe, Arizona 85287, USA 2. Department of Physics, Arizona State University, Tempe, Arizona 85287, USA The ultramicrotome is used extensively for preparing cross-sectional samples for optical and/or electron microscopy [1]. Ultramicrotomed thin sections usually have a uniform thickness and retain the original elemental distribution in phases of the sample [2]. Generally speaking, thin nanowires (NWs) are ideal for electron microscopy observations. However, it is extremely difficult, if not impossible, to examine the cross-sectional view of long NWs. Such situation arises if one wants to examine the interfacial atomic structures of core-shell nanowires or other more complicated configurations. In these cases, embedding the NWs into resins and ultramicrotoming the cured resin/NW composite provide a method to extract information about interfacial atomic structures of core-shell NWs or other types of nanostructures. We report here, via the use of ultramicrotome to prepare cross-sectional samples, the investigation of the interfacial structures of the -Bi2O3 epitaxially grown onto the selected facets of pure wurzite ZnO NWs. Bi2O3/ZnO composite nanostructures were embedded into capsule modes and polymerized for 24 h at 70�C. The cured block was then trimmed to dimensions of 0.5mm�0.5mm and mounted on the ultramicrotome (Leica Ultracut R microtome) and sectioned with a diamond knife with a wedge angle of 45�. Thin slices of the sample with a thickness of 30-50nm were produced. Lacy-carbon coated TEM copper grids were used to collect the ultrathin sections. A thin layer of carbon was coated onto the ultramicrotomed thin sections prior to TEM/STEM observation. Aberration-corrected STEM, in the highangle annular dark-field (HAADF) imaging mode, was used to examine the cross-section core-shell NWs. Figure 1a and 1b show low and high magnification HAADF images of a cross-section of a -Bi2O3/ZnO NW with the electron beam parallel to the ZnO [0001] zone axis. The side facets of the ZnO NWs consist of {10-10} and {11-20} surfaces and the -Bi2O3 grew selectively only on the {11-20} facets. Our previous work demonstrated that the -Bi2O3 {100}facets grew epitaxially onto the {11-20} facets of the ZnO NWs [3]. With the analysis of the atomic arrangements of the Zn and Bi atoms as revealed in Fig. 1b, it is proposed that the oxygen layer of -Bi2O3 may be in direct contact with the ZnO (11-20) surface. The detailed arrangement of the interfacial oxygen atoms is still unclear and is an active area of investigation. The wetting behavior between the -Bi2O3 {100} and the ZnO {11-20}, however, suggests strong interfacial reactions. A schematic illustration of the proposed interfacial structure is shown in Fig. 1c. Schematic diagrams in Figure 1d show the atomic arrangement of the reconstructed -Bi2O3 (100) and ZnO (11-20). The interfacial Bi atoms are strained to accommodate the epitaxial relationship, and a good dimensional match exists between the ZnO {11-20} and the -Bi2O3 {100} planes. The fact that -Bi2O3 has a high concentration of oxygen vacancies may facilitate to form a strong bonded interface [4]. Detailed analyses of the interfacial structures and the growth mechanisms will be discussed [5]. References: [1] H. Plummer, Microsc. Microanal. 3 (1997), p.239. [2] L.Y. Wei and T. Li, Microsc. Res. Tech. 36 (1997), p.380. Microsc. Microanal. 21 (Suppl 3), 2015 1740 [3] J. Xu, J. Liu, Microsc. Microanal. 19 (2013), p.1516. [4] J. Luo, J. Am. Ceram. Soc. 95 (2012), p.2358. [5] This research was funded by Arizona State University. We gratefully acknowledge the use of facilities within the LeRoy Eyring Center for Solid State Science at Arizona State University. Figure 1 (a-b) Low and high magnification HAADF images of the cross-sectional view of a typical Bi2O3 decorated ZnO nanowire. Epitaxial layers of -Bi2O3 were deposited selectively onto ZnO{11-20} facets. The reconstruction of the interfacial Bi atoms is clearly revealed in Figure 1b. The proposed atomic structure of the interfacial region is schematically illustrated in Figure 1c. Figure 1d shows the proposed positions of the interfacial atoms on the -Bi2O3 (100) and ZnO (11-20) planes. To accommodate the epitaxial relationship, one Bi atom (indicated by the red arrow in Figure 1d) has to be shifted by 0.09 nm to bond directly with the Zn atom.");sQ1[869]=new Array("../7337/1741.pdf","Microscopy Guided Design of Radial p-n Junction in Single TiO2 Nanotubes","","1741 doi:10.1017/S1431927615009484 Paper No. 0869 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microscopy Guided Design of Radial p-n Junction in Single TiO2 Nanotubes Miaofang Chi,1 Mina Yoon,1 and Parans Paranthaman2 1. 2. Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, TN 37831. Chemical Sciences Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831. Engineering one-dimensional nanostructures in the axial direction is of great importance to researchers in energy harvesting fields, as utilizing materials with such a unique configuration could potentially introduce significant efficiency improvements for use in future solar energy conversion, storage, and nanoelectronic devices [1-3]. In 2007, the Lewis group proposed an interesting solar cell configuration consisting of radial p-n junction nanorods that exhibited a considerably improved cell efficiency compared to traditional planar solar cells [4]. Such a configuration decouples the direction of incident light and the proximity of generated charge carriers to the p-n junction. Vertical alignment of many cylindrical junctions facilitates the transport of photon-generated minority carriers to the junctions, and therefore, greatly increases carrier collection efficiency, as well as enhanced tolerance to radiation damage, defects, and impurities. Further advantages are anticipated with nanotube structures for solar energy conversion based on recent theoretical calculations, which showed that nanohole arrays (aligned nanochannels in a Si matrix) provide better light absorption compared to nanowire arrays due to the high density of waveguide modes [5]. Anodic TiO2 nanotube arrays (NTA) have been studied extensively such that their radial physical properties can be tailored by incorporating an additional layer having different physical properties. These layers are either incorporated on the inner or outer walls of the nanotubes [6,7]; however, the synthesis of well-controlled and reproducible axial p-n junctions with desirable properties at a relatively low cost for NTAs remains a challenge. Previous reports of the microscopy characterization of TiO2 nanotubes summarized observations regarding general morphological parameters, such as crystallinity/phase, size/diameter, length, and wall thickness. A detailed analysis of the surface atomic structure, however, has not been studied in detail since it has been assumed that TiO2 nanotubes have the same rolled structure as observed for carbon nanotubes [8]. Current electron microscopy work has revealed that anodic TiO2 nanotubes are single crystal tubes with clear surface facets. The TiO2 outer walls are consistent with (110) and (100) surface facets, as shown in Figure 1, e.g., four {110} and two {100} outer wall surfaces comprise a nearhexagonal morphology. In contrast, the inner surface/wall is characterized by a relatively smooth, curved surface that has essentially formed by many adjoining small facets (1-2 atomic layers). Internal strain in the radial direction was determined by analyzing a series of high-resolution STEM images using geometrical phase analysis (GPA) analysis. The inner walls are under a compressive strain that gradually transitions to a tensile strain at the outer tube walls. Furthermore, the atomic termination and electronic structures at the inner and outer wall surfaces/facets were different, as evidenced by analysis of the EELS Energy Loss Near Edge Structure (ELNES), as shown in Figure 2. For example, the first pre-peak of the Ti-L3 edge (labeled "A" in Figure 2b, which arises from the 2p3/2 Ti to the t2g molecular orbital MO transition), is more pronounced in the spectrum from the outer wall compared to that from the inner wall. Although the corresponding pre-peak splitting of the O-K edge is not clearly visible, peak "b" of the O-K edge, which can be directly correlated to the hybridization of O 2p orbitals and Ti 4s/4p orbitals, is notably different [9]. These electronic structural differences will be correlated to their atomic structures and be discussed in detail in our presentation. Furthermore, distinctive chemical activities are associated with different surface terminations, as predicted by first principle theoretical calculations. Microsc. Microanal. 21 (Suppl 3), 2015 1742 Results from microscopy studies can offer direct structural input for designing novel coaxial p-n junctions that can exploit the unique chemical properties provided by tailored inner and outer wall structures and chemistry. This concept was recently validated experimentally where it was shown that elements such as Nb, Sr, and Cr, can be used to preferentially dope specific locations and establish internal junctions. [10] References: [1] LJ Lauhon et al, Nature 420 (2002) p. 57. [2] BZ Tian et al, Nature 449 (2007) p. 885. [3] LF Shen et al, Journal of Physical Chemistry Letters 2 (2011) p. 3096. [4] BM Kayes, HA Atwater, and NS Lewis, Journal of Applied Physics 97 (2005) p. 1678. [5] SE Han and G Chen, Nano Letters 10 (2010) p. 4692. [6] DA Wang et al, Journal of Materials Chemistry 20 (2010) p. 6910. [7] SP Albu et al, Advanced Materials 20 (2008) p. 4135. [8] SM Liu et al, Chemistry of Matererials 14 (2012) p.1391. [9] E. Stoyanov et al, American Mineralogist 92 (2007) p. 577. [10] Research supported by the Laboratory Directed Research and Development (LDRD) program at Oak Ridge National Laboratory (ORNL). Microscopy was conducted at ORNL's Center for Nanophase Materials Sciences (CNMS), which is a U.S. Department of Energy, Office of Science User Facility. Figure 1. (a) SEM image of an aligned array of anodic TiO2 nanotubes. (b) Schematic figure of the morphology of TiO2 nanotubes exhibiting {100} and {1-10} surface facets. (c) High-resolution TEM image showing different surface atomic structure at inner and outer walls. Figure 2. (a) STEM image of a cross-sectioned TiO2 nanotube; (b) comparison of the EELS spectra acquired from inner (blue) and outer (red) nanotube walls.");sQ1[870]=new Array("../7337/1743.pdf","Direct TEM/STEM Observation and Analysis of Bismuth Segregation to Grain Boundary Dislocation Cores in Bismuth Embrittled Copper Bicrystals","","1743 doi:10.1017/S1431927615009496 Paper No. 0870 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Direct TEM/STEM Observation and Analysis of Bismuth Segregation to Grain Boundary Dislocation Cores in Bismuth Embrittled Copper Bicrystals C. A. Wade, M. Watanabe 1. Dept. of Materials Science and Engineering, Lehigh University, Bethlehem, PA Bismuth (Bi) has long been known to cause dramatic embrittlement of Copper (Cu) grain boundaries (GBs). When the relatively large Bi atoms are introduced to the relatively small Cu atoms the Bi nearly completely segregates to the Cu GBs. Once on the Cu GBs, Bi causes a reduction in the fracture toughness of the Cu GBs by almost completely preventing the GB from any stress relieving plastic deformation. The suppression of this plastic deformation results in rapid intergranular fracture once the elastic limit of the GB has been reached. Recently, it has been shown through micro-mechanical tensile tests of certain tilt bicrystal GBs that there may be a degree of plastic deformation possible in an embrittled GB that still shows a marked reduction in fracture toughness [1]. This is in contrast to most previous studies which focused on either specific tilt Cu GBs or general GBs found in polycrystalline samples [2]. This study investigates the structure of Bi doped Cu tilt GB and the relationship that this structure may have with the mechanical properties of the GBs. In this study of Bi doped Cu [001]/33� twist bicrystals, the burgers vector of GB dislocations was analysed using g�b analysis via two-beam dark-field transmission electron microscope (TEM) imaging. The occurrence and spacing of GB dislocations was also investigated by high-resolution TEM (HRTEM) imaging. Figure 1a shows a bright-field (BF) TEM image clearly exhibiting the GB dislocations threading through the bicrystal specimen along the GB. Figure 1b shows a HRTEM image of the GB dislocations at a higher magnification, in both Figure 1a and 1b the spacing of dislocations was measured to be approximately 2 nm. To directly image the occurrence of Bi at the GB and any correlation it may have with the GB dislocations, aberration-corrected BF and high-angle annular darkfield (HAADF) STEM imaging techniques were used. BF STEM imaging has the advantage of both showing lattice strain from dislocations present at the GB allowing the dislocations to be visible in images, as well as being weakly sensitive to atomic mass changes allowing Bi atoms to be directly imaged. HAADF STEM imaging is strongly sensitive to atomic mass allowing both individual atoms of Bi to be distinguished from the bulk Cu and allowing areas with a higher composition on heavy dopant elements to be imaged. Figure 2a shows a BF STEM image where Bi atoms can clearly be seen on the left of the figure. Across the GB, bright and dark areas can also be seen in this image, corresponding to strain fields created by dislocations as would be seen in a BF or DF TEM image. Figure 2b exhibits a HAADF-STEM image of the same field of view in the vicinity of the GB showing the Bi atoms on the left side of the boundary and bright areas crossing the boundary through the specimen. The bright areas in this HAADF-STEM image indicate that the boundary contains a greater amount of higher atomic number elements (i.e. Bi dopant atoms) without any dislocation strain contrast present as HAADFSTEM images are not formed using any diffraction information. X-ray energy dispersive spectrometry (XEDS) was also used to verify the presence of Bi at the GB and on GB dislocations. This preferential segregation of Bi to different character GB dislocations may explain both the presence of plasticity in Bi doped specimens and the complicated relationship between the occurrence of Bi induced embrittlement and GB structure. Microsc. Microanal. 21 (Suppl 3), 2015 1744 References: [1] M. McLean, C.A. Wade, R.P. Vinci and M. Watanabe, Exp Mech 54 (2013), p. 685-688. [2] D.E.J. Armstrong, A.J. Wilkinson and S.G. Roberts, Phil Mag Lett 91 (2011), p. 394-400. [3] The authors acknowledge support NSF through grants DMR-0804528 and DMR-1040229 (a) (b) Figure 1. TEM images of a Bi doped [001]/33� twist Cu GB showing dislocations along the GB threading through the specimen, (a) shows a BF image of the GB while (b) shows a higher magnification high resolution TEM image of the boundary. The spacing between dislocation cores in both images is approximately 2 nm. (a) (b) Figure 2. Aberration-Corrected STEM (a) BF and (b) HAADF images of the same Bi doped [001]/33� twist Cu GB imaged in Figure 1 with Bi atoms and dislocation spacing marked.");sQ1[871]=new Array("../7337/1745.pdf","Fe-25Mn-3Al-3Si TWIP�TRIP Steel Deformed at High Strain-Rates","","1745 doi:10.1017/S1431927615009502 Paper No. 0871 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Fe-25Mn-3Al-3Si TWIP�TRIP Steel Deformed at High Strain-Rates J. T. Benzing1, J. E. Wittig1, T. M. Smith2, M. J. Mills2, J. R. Johnson2, G. S. Daehn2, J. Bentley3, D. Raabe4, C. Ophus5 1. 2. Interdisciplinary Materials Science, Vanderbilt University, Nashville TN 37232, USA Materials Science and Engineering, The Ohio State University, Columbus, OH 43210, USA 3. Microscopy and Microanalytical Sciences, PO Box 7103, Oak Ridge, TN 37831-7103, USA 4. Max-Planck-Institut f�r Eisenforschung, Max-Planck-Strae 1, D-40237 D�sseldorf, Germany 5. National Center for Electron Microscopy, Molecular Foundry, Lawrence Berkeley National Lab, Berkeley, CA 94720, USA High-Mn steels exhibit high strength, superior strain hardening and high formability, which are attractive to the automotive industry for reduced weight, crashworthiness and cold stamping of complex parts [1]. When Fe is alloyed with 25% Mn, 3% Al and 3% Si (wt%), these substitutional elements stabilize the gamma phase, enhance the solid-solution strengthening effect, and reduce the stacking fault energy (SFE). Strain hardening in this alloy involves the development of secondary deformation mechanisms such as twinning-induced plasticity (TWIP) and transformation-induced plasticity (TRIP), that allow for uniform elongation up to 80% in a room-temperature (RT) tensile test at quasi-static strain rates (~10-4 s-1). A SFE of 21�3 mJm-2, which was experimentally determined [2] for this alloy using weak-beam dark-field transmission electron microscopy (TEM) and incorporated elastic constants measured by a novel combination of orientation imaging microscopy and nano-indentation [3], promotes TWIP as the dominant mode of RT secondary deformation with some evidence of TRIP in the form of hexagonal -martensite. In the current work, uniaxial ring expansion that utilized a shock wave from a vaporized wire provides strain rates from ~103 to 104 s-1. Since adiabatic heating is a product of higher strain rates and SFE is sensitive to temperature, understanding the deformation mechanisms as a function of strain rate requires quantitative characterization of the microstructural evolution [4]. The strain rate from the ring expansion can be controlled by ring diameter and strain by input energy. With radius-to-thickness ratios held constant, inner-ring diameters of 51, 25, & 16 mm resulted in strain rates of 1100, 3000 & 9000 s-1 and input energies of 8, 4 & 2 kJ produced total strains true = 20, 14 & 12%, respectively, for each size. TEM samples were sectioned orthogonal to the ring outer radius with electro-discharge machining to produce 3mm diameter discs, which were electropolished to electron transparency with a Struers twin-jet system (20% HNO3/80% CH3OH, -30 �C, 15 V). A Philips CM20 was used to survey the samples in order to locate grains with secondary deformation habit planes parallel to the <110> beam direction. As was observed for RT deformation at a quasi-static strain rate, many grains were characterized by multiple twin/martensite variants as well as termination of stacking faults at twin/martensite interfaces. Grains selected for high-resolution (HR) imaging met the criteria of minimal alpha/beta tilt, so as to minimize effective sample thickness. Figure 1 shows results from the 50 mm dia. ring expanded to true = 20% at =1100 s-1, where selected area electron diffraction (SAED) patterns indicate the presence of either TRIP or TWIP in separate grains of the same TEM disc. These two grains were imaged in high-angle annular dark-field (HAADF) STEM mode with a probe-corrected FEI Titan and in HRTEM mode with an image-corrected FEI Titan. The HAADF-STEM images provided the most useful data as the HRTEM phase-contrast was complicated by variations in sample thickness, crystal orientation due to bending of the foil and objective-lens defocus. Although the SAED patterns in Figure 1 indicated either TRIP or TWIP, the HAADF images in Figure 2a-b show that both twinning and hexagonal stacking are present together in the same grain with widths on the order of a few nanometers, regardless of mechanism that dominated the SAED pattern. Preliminary results from strain mapping of the HR-STEM images using a real-space lattice method that combines images taken at 0 and 90�, with respect to the slow-scan direction, reveal significant shear strains at twin and martensite interfaces. Quantitative strain measurements at twin and martensite interfaces may provide insight into the role of secondary deformation mechanisms on the strain hardening behavior of these high-Mn austenitic steels [5]. Microsc. Microanal. 21 (Suppl 3), 2015 1746 [1] O Bouaziz et al," Curr. Opin. Solid State Mater. Sci., (2011). [2] D T Pierce, J A Jim�nez, J Bentley, D Raabe, C Oskay and J.E. Wittig, Acta Mater 68(2014)238-53 [3] D T Pierce, K Nowag, A Montagne, J A Jim�nez, J E Wittig and R Ghisleni, Mater Sci Eng A578(2013)134-9 [4] Pierce, Dean Thomas, Ph.D dissertation, Vanderbilt University, (2014). [5] This work was funded by the US National Science Foundation Division of Materials Research under grant DMR1309258, the MPIE in D�sseldorf and the OSU in Columbus (MSE dept. and CEMAS). Work at the Molecular Foundry was supported by the Office of Science, Office of Basic Energy Sciences, of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231. Figure 1: BF TEM images of true = 20%, = 1100 s-1. The inset SAED patterns, recorded at <110> zones, indicate two modes of secondary deformation: a) TRIP b) TWIP. Figure 2: High-resolution HAADF-STEM images recorded using a probe-corrected FEI Titan from the respective grains in Figure 1, where regions of matrix (M), -martensite platelets (bracket), twinning with respect to matrix (T) and twin boundaries (arrow) are marked: a) SAED with martensite reflections (TRIP) b) SAED with twin reflections (TWIP).");sQ1[872]=new Array("../7337/1747.pdf","Tracking the Evolution of in-situ Radiochemistry with Transmission Electron Microscopy","","1747 1747 doi:10.1017/S1431927615009514 doi:10.1017/S1431927615009514 Paper No. 0872 Microsc. Microanal. 21 (Suppl 3), 2015 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 � Microscopy Society of America 2015 Tracking the Evolution of in-situ Radiochemistry with Transmission Electron Microscopy J. A. Aguiar*, L. Wolfsberg, W. Taylor, B. P. Uberuaga, B. Scott, C.R. Stanek Los Alamos National Laboratory, Los Alamos, NM, 80465 * Now at, National Renewable Energy Laboratory, Golden, CO 80401 Recent developments within the nuclear materials community have lead researchers to predict nanocrystalline materials can plausibly address concerns regarding nucleation, growth, migration, waste storage, and fission product evolution at higher temperatures and radiation environments. From the work so far in the community, it is not clear how these phases of material impact waste form evolution and behavior. Advanced microscopy allows for unprecedented insights into the underlying physical mechanisms responsible for the improved performance of these materials, especially through the use and study of evolving decaying isotopes. To extend our understanding of nuclear materials after service, in particular waste forms, we have studied the fundamental evolution of these materials induced by the decay of radioisotopes by coupling the use of accelerated aging and analytical microscopy. The effect of transmutation of radionuclides, especially "short-lived" Sr-90 and Cs-137, to chemically distinct daughter products (Zr and Ba respectively) will impact nuclear waste form stability. Due to the technical challenges associated with studying this problem, the topic of transmutation has received limited attention during the past 30 years of waste form development. In order to develop a predictive capability to design radiation tolerant and chemically robust nuclear waste forms, we must first address a fundamental materials science question, namely what is the impact of daughter product formation on the stability of solids comprised of radioactive isotopes. To address this question, a multidisciplinary approach integrating first principles modeling with the synthesis and characterization of small, highly radioactive surrogate samples has been instigated. We present the details of this approach as well as recent results for a range of materials system, including 177Lu2-xHfxO3. Here, Lu-177 is chosen as a surrogate for the radionuclides present in the actual waste because of its very short half-life (6.65 days), after which it decays to Hf. In this presentation, we will examine both the evolution of structure and chemistry in connection with decaying radioactive isotopes with aberration corrected transmission electron microscopy. In particular we will focus on the results of transmission electron microscopy coupled with diffraction and related spectroscopies (EDS & EELS) over a continuous time-series of data collected more than two half-lives. This talk highlights how we can apply transmission electron microscopy to study highly radioactive isotopes in small quantities. Microsc. Microanal. 21 (Suppl 3), 2015 Microsc. Microanal. 21 (Suppl 3), 2015 1748 1736 This work was supported by the laboratory directed research program at Los Alamos National Laboratory. Figure 1: Above is an overview of the collected diffraction profiles tracking the decay of 177Lu2-xHfxO3. Collected diffraction profiles span the course of more than twenty days of transmuting Lu2O3 highlighting the structural evolution of transmuting oxide within an aberration corrected FEI Titan. These results highlight the potential use of in-situ based microscopy to study the structural decay and effects transmuting such as radioactive waste forms, such as those found at current nuclear repositories.");sQ1[873]=new Array("../7337/1749.pdf","Nonmetallic Inclusions and Acicular Ferrite in Arc Welds of Pipeline Steels","","1749 doi:10.1017/S1431927615009526 Paper No. 0873 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Nonmetallic Inclusions and Acicular Ferrite in Arc Welds of Pipeline Steels Alexander A. Kazakov1, Daria Lyubochko1 St. Petersburg State Polytechnical University/Metallurgical Technologies Department, St. Petersburg, Russian Federation The purpose of this work is to study the influence of nonmetallic inclusions on structure formation of acicular ferrite in longitudinal submerged arc welds using light optical microscopy, scanning and transmission electron microscopy, and X-ray microanalysis. Acicular ferrite is a desirable structure in welded pipeline steels, because it provides an optimal combination of both high strength and good toughness due to the fine "basket-weave" microstructure. The weld metal of the samples investigated has a structure of acicular ferrite both in the center and at the periphery up to the point of the weld-fusion line (Fig.1). Acicular ferrite forms in consequence of the rapid cooling of steel from the austenitic state, when formation of allotriomorphic, Widmanst�tten and massive ferrites are suppressed. Acicular ferrite plates are formed on ultrafine nucleation centers based on titanium oxides evolved during the deoxidation processes. Particles of MnS, namely Mn-depleted zones formed around the particles, may also be sites for acicular ferrite nucleation. To ensure the formation of titanium oxides, the amount of aluminum added to the metal should be minimal (0,005%), because of its strong affinity for oxygen. Boron segregates to austenite grain boundaries and reduces grain boundary energy and the driving force for ferrite transformation in solid solution. It inhibits the formation of grain boundary ferrite and promotes the formation of intragranular acicular ferrite [1]. There were nonmetallic inclusions of different sizes inside the weld metal of all investigated specimens. The quantity of nonmetallic inclusions (Fig. 2a) varies from tens to hundreds for different samples, and depends on the explored place in the weld (inside weld, outside weld, central fusion zone). The composition of the inclusions corresponds to two major multi-component systems: Al2O3-CaO-SiO2MnO-TixOy and Al2O3-CaO-SiO2-MnO-MgO-TixOy-MnS (Fig. 2b). The ultrafine nonmetallic inclusions create conditions for acicular ferrite formation in a weld. The presence of those kinds of inclusions is caused by technological modes of welding, and also wire and flux composition for welding. These inclusions may serve as substrates for formation of an acicular ferritic structure in weld metal [2]. It provides an optimum combination of strength and plastic properties of weld metal. Moreover, this structure and technology ensures achievement of full strength welds, because the curve of microhardness changes monotonically over the cross section of the welded joint, including the heat-affected zone. Data for the size and composition of nonmetallic inclusions are confirmed by studies carried out by transmission microscopy (Fig. 2c). The study of the nature and distribution of nonmetallic inclusions in a cross section of a welded joint makes it possible to estimate their contribution to the structure formation in the joint. With knowledge of the thermal and physico-chemical characteristics of nonmetallic inclusions formation it becomes possible to correct welding technology in such a way as to obtain the desired weldment microstructure. 1. Microsc. Microanal. 21 (Suppl 3), 2015 1750 References: [1] D.S. Sarma et al., ISIJ International, 49 (2009), p.1063. [2] M.V. Karasev et al., Welding and Diagnostics, 6 (2014), p. 45. Figure 1. Panoramic image of the weld joint, �50, �1000 ) b) c) Figure 2. a) The distribution of inclusions by size; b) and c) typical inclusion, found in the weld: (b) scanning electron microscopy; (c) transmission electron microscopy images");sQ1[874]=new Array("../7337/1751.pdf","Distribution Pattern of Nonmetallic Inclusions on Cross Section of Continuous-Cast Steel Billets For Rails","","1751 doi:10.1017/S1431927615009538 Paper No. 0874 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Distribution Pattern of Nonmetallic Inclusions on Cross Section of ContinuousCast Steel Billets For Rails Alexander A. Kazakov1, Andrey Zhitenev1, Pavel Kovalev1 1. St.Petersburg State Polytechnical University/Metallurgical Technologies Department, St.Petersburg, Russian Federation Nonmetallic inclusions are formed at all stages of the steelmaking process of melting and secondary treatment before casting and crystallization. Redistribution of impurities and alloying elements between solid and liquid phases during crystallization occurs [1], forming chemical and structural inhomogeneity on a cross section of the billet. The formation of inclusions in a solidifying steel occurs simulteniously with formation of dendritic structure. All of these interconnected processes determine distribution pattern of non-metallic inclusions in billet's cross section [2]. Dendritic structure was revealed by hot etching in 50% water solution of HCl for 20 min. After this a specimen was swabbed in nital to remove a sludge and immersed by cc. H3PO4 for blackening a surface. Next the sample was polished by abrasive paper P1200 to remove blackening from dendrite arms. For a quantitative determination of a dendritic structure the secondary dendritic arm spaces (SDAS) were evaluated. To obtain reliable results panoramic images of a dendritic structure were created by means of Image Analyzer ThixometProTM. An estimation of nonmetallic inclusions were performed in accordance with the standard ASTM E 1245. Fig.1 shows an panoramic image of an identified dendritic structure. All structural zones formed during crystallization [1] were distinguished: outer equiaxed zone, columnar zone, inner equiaxed zone. Columnar crystals zone are more extended from a side of the inner radius of continuous casting machine because of a greater cooling rate. The distribution pattern of nonmetallic inclusions on the cross section of continuous-cast steel billets for rails in consideration of dendritic parameters in all crystal structure zones were studied. Maximum nonmetallic impurity rating was detected at the boundary intersection of adjacent crystal structure zones. The secondary dendrite arm spacing SDAS (2), which characterizes the size of dendritic cell, was examined on the cross section of a billet. For the rail continuous-cast steel billet there is 2 increase only for the columnar crystal zone while for the equiaxed zone it is unaffected (fig.1). Local areas with increased SDAS exist due to maximum solidification local time. Areas of local thermal centers are located irregularly and asymmetrically with respect to the geometric center of the billet. It was that determined effects of secondary dendrite arm spacing upon the dimentions of tertiary nonmetallic inclusions appeared during solidification. That is for rail continuous-cast steel billet such a relationship was found only for a columnar crystal zone when in the region of equiaxed zone a size of sulfides varies stochastically (fig.2) due to the macrosegregation phenomena. Method of sampling for the control of nonmetallic inclusions in continuous-cast steel billets for rails was developed [2] based upon revealed relationships between nonmetallic inclusions formation and dendritic structure formation which correlates good with a sampling for end products (rails). Microsc. Microanal. 21 (Suppl 3), 2015 1752 1740 References: [1] Kurz W., Fisher D.J. Fundamentals of solidification. � Trans. Tech. Publ., 1998.-305 p. [2] Kazakov A.A., et al., Chernye Metally (Ferrous Metalls), 2014, 4 (988), p. 31. Figure 1. Macrostructure of a billet's specimen and measured distribution of SDAS on a cross section of the billet Figure 2. Relationship between sulfide's size and secondary dendrite arm spaces");sQ1[875]=new Array("../7337/1753.pdf","As-Cast Structure and Metallurgical Inheritance of High Nitrogen Austenitic Stainless Steel","","1753 doi:10.1017/S143192761500954X Paper No. 0875 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 As-Cast Structure and Metallurgical Inheritance of High Nitrogen Austenitic Stainless Steel Alexander A. Kazakov 1, Eduard Kolpishon2, Aleksey Shakhmatov3 and Robert Badrak4 St. Petersburg State Polytechnical University/Metallurgical Technologies Department, St. Petersburg, Russian Federation 2. NPO TSNIITMASH, Moscow, Russian Federation 3. Weatherford, St. Petersburg, Russian Federation 4. Weatherford, Houston, Texas, USA High nitrogen steels belong to special class of iron-based multicomponent alloys. Nitrogen is a strong austenite stabilizer and improves the mechanical and corrosion resistance properties. Nitrogen is also a relatively low cost alloying element for producing stainless steel. The final stainless steel properties depend on the microstructure, the formation of which takes place at all stages of production, from the melting process, casting and solidification to plastic deformation and heat treatment. With that, the microstructure of stainless steel depends on chemical composition and evolves at each subsequent technological stage based on metallurgical heritage of structure formed during the solidification and it is known that austenitic steels could solidify by different modes and phase formation, including ferrite [1, 2]. The purpose of this work is observation and characterization of the as-cast structure of high nitrogen steels using optical metallography and thermodynamic simulation. Specimens of high nitrogen steels were obtained by the method of fractional casting [3] related to making of small ingots with different chemical compositions (table 1). As-cast samples were studied in after annealing at 1060�C for 1 h. Microstructures were revealed by tint etching using reagent with the following composition: 10ml HCl+90ml H2O+1g Na2S2O5. Tint etching was performed using polarized light and filters that allows emphasizing structural and isolated fragments with dendritic arms which have the similar crystallographic orientation. Microstructural analysis was carried out with using a motorized light microscope Zeiss Axiovert 200 MAT, powered by image analyzer Thixomet. Software package FactSage with SGTE databases were used for thermodynamic simulation. Fig. 1a shows an austenitic structure of specimen 1 stainless steel, where all the austenite grains are formed by clear primary and secondary dendritic arms with the common crystallographic orientation. In figure 1b, the austenite structure of specimen 2 does not contain ferrite, where austenite grains formed by primary and secondary dendritic arms, but some of them have a "blurring effect" where some grains have a blurred dendritic pattern. Figure 1c shows a duplex structure of specimen 3 with about 5% � 10% ferrite phase and in this case dendritic arms in austenite grains are not observed. According to the results of simulation (fig.2) specimen 1 solidifies with only primary austenite phase and ferrite phase did not form in the region of annealing temperature 1060�C. Specimen 2 shows no ferrite at the annealing temperature, and the solidification process was accompanied by the formation of primary austenite with formation of ferrite phase. Specimen 3 was solidified with primary ferrite and secondary austenite. Also, ferrite phase does exist (about 5%) at the annealing temperature. 1. Microsc. Microanal. 21 (Suppl 3), 2015 1754 By summarizing we can conclude that the ferrite formed during the solidification process, and further it's polymorphic transformation leads to a concentration gradient due to the different solubility of alloying and impurity elements in the ferrite and austenite. Because of this gradient, the elements diffuse to the depletion region of austenite that was formed from ferrite. The solidification with austenite only fixes chemical inhomogeneity, whereas solidification with ferrite, and the further transformation of ferrite to austenite leads to a "blurring" of a dendritic structure and homogenizing of chemical composition while cooling the solidified steel and in the process of heat treatment. References: [1] K. Rajasekhar et al, Materials Characterization, 38 (1997), Issue 2, p. 53 [2] M. A. Martorano et al, ISIJ International, 52 (2012), . 6, p. 1054 [3] A.V. Shahmatov et al, Tyazheloe Mashinostroenie Journal, . 2-3 (2014), p.3 Specimen C Si Cr Mn 1 0.06 0.48 20.56 24.32 2 0.06 0.44 20.59 23.94 3 0.06 0.36 22.78 25.98 Table 1. Chemical compositions of investigated specimens, % Ni 1.50 2.19 2.65 Mo 1.23 1.11 0.35 N 1.20 1.09 0.99 Figure 1. As-cast structure of steel specimens: a, b, c � specimen 1, 2, 3, tint etched, polarized light, sensitive tint, �50 Figure 2. Sequence of phase formation in investigated specimens: 1(1), 2(2), 3(3)");sQ1[876]=new Array("../7337/1755.pdf","Research on the Origin of Nonmetallic Inclusions in High-Strength Low-Alloy Steel Using Automated Feature Analysis","","1755 1755 doi:10.1017/S1431927615009551 doi:10.1017/S1431927615009551 Paper No. 0876 Microsc. Microanal. 21 (Suppl 3), Microsc. Microanal. 21 (Suppl 3), 2015 2015 � Microscopy of America 2015 � Microscopy Society Society of America 2015 Research on the Origin of Nonmetallic Inclusions in High-Strength Low-Alloy Steel Using Automated Feature Analysis Alexander A. Kazakov1, Sergei Ryaboshuk1, Daria Lyubochko1, Lev Chigintsev1 1. St. Petersburg State Polytechnical University/Metallurgical Technologies Department, St. Petersburg, Russian Federation The aim of this paper is an investigation of thermo-temporal nature of nonmetallic inclusions, which were found in samples taken from high-strength low-alloy (HSLA) steel forgings and plates, treated with silico-calcium. For an objective assessment of the chemistry of non-metallic inclusions, a SEM-EDS microanalysis method called automated feature analysis (AFA) was used. This technique makes it possible to examine large quantities of inclusions only comparable to optical microscopy methods, but AFA provides also a compositional data along with conventional metrics [1]. By chemical composition, the inclusions in the acquired database mostly belong to the Al-Mg-Ca-S-O system. Note that the composition of these inclusions also contains small concentrations of manganese and silicon. All inclusions are divided into four main groups (Fig.1): I) inclusions of the ternary system Al2O3-MgO-CaO, which belong to the region of coexistence of solid magnesium spinel and liquid oxide melt; II) inclusions of the same system, which belong to the region of liquid oxide melt; III) calcium aluminates of the binary system Al2O3-CaO; and IV) Al2O3-MgO system compounds. It should be noted that almost all detected inclusions contain sulfur, wherein the mass of sulfides increases with increasing magnesium content in the inclusions. In the presence of magnesium, calcium has more opportunities to interact with sulfur. Next, we consider in detail the thermo-temporal nature of the first and second group inclusions, as the most complex. Inclusions of the first group in cast metal have rounded shape and size up to 15-30 m (Fig.2a). Initially, primary and secondary solid magnesium spinel inclusions and liquid oxy-sulfides formed in the melt (Fig.2b). During crystallization of the oxy-sulphide melt (CaO42%; MgO28%; Al2O323%; SiO26%; S0.12%), the next compounds form sequentially: first MgOAl2O3 and MgO, which continue to grow on the surface of spinel already existed in the steel melt, then 2CaOSiO2, 3CaOMgO2Al2O3 and finally, CaS (Fig.2c). Inclusions of the second group (Fig.3a) differ from the first group with a higher content of calcium, which in some cases exceeds 45%. The thermo-temporal nature of the nucleation of these inclusions is that only liquid oxy-sulphide melt (Al2O325%; CaO35%; MgO32%; SiO27%; S0.05%) solidify until the temperature drops to 1555�C in the steel melt (Fig. 3b). Solid oxides like CaO�Al2O3, CaO�6Al2O3 and Al2O3 form sequentially at lower temperatures in molten steel and later form solid oxy-sulphides. Solidification begins with MgO and MgOAl2O3 formation, particles of which have correct facets due to conditions of free growth in liquid steel (Fig. 3c). Upon further cooling, calcium silicates, CaO3MgO2Al2O3 and finally CaS form. Thermodynamic calculations [2] of nonmetallic inclusion formation in the liquid and solidifying steel melt, including solidification processes of forming oxy-sulfide liquid inclusions, permit explanation of the variety of phases in complex multiphase inclusions. Microsc. Microanal. 21 (Suppl 3), 2015 2015 Microsc. Microanal. 21 (Suppl 3), 1744 1756 References: [1] T Taniguchi et al, ISIJ International 51 (2011) p. 1957. [2] C Bale et al, Calphad, 26 (2002) p.189 Figure 1. Chemical composition of inclusions and phase diagram of Al2O3-MgO-CaO system at 1550 a) b) c) Figure 2. First group: (a) distinctive inclusion and its phase composition; (b) thermodynamic modeling of phase formation during cooling and solidification of the liquid steel; (c) crystallization of liquid oxysulphide inclusions a) b) c) Figure 3. Second group: (a) distinctive inclusion and its phase composition; (b) thermodynamic modeling of phase formation during cooling and solidification of the liquid steel; (c) crystallization of liquid oxysulphide inclusions");sQ1[877]=new Array("../7337/1757.pdf","Influence of Additional Annealing on Properties of Ni-Mn-In-Co Heusler Alloy.","","1757 doi:10.1017/S1431927615009563 Paper No. 0877 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Influence of Additional Annealing on Properties of Ni-Mn-In-Co Heusler Alloy. Alexander Kamantsev1, Elvina Dilmieva1, Alexey Mashirov1, Victor Koledov1, Vladimir Shavrov1, Vladimir Khovaylo2, Maria Lyange2, Sergey Konoplyuk3, Vladimir Kokorin3, Rostislav Grechishkin4, Pnina Ari-Gur5. 1. 2. Kotelnikov Institute of Radio-engineering and Electronics of RAS, Moscow, Russia. National University of Science and Technology "MISIS", Moscow, Russia. 3. Institute of Magnetism of NASU and MESU, Kyiv, Ukraine. 4. Tver State University, Tver, Russia. 5. Western Michigan University, Kalamazoo, MI, USA. A promising class of solid materials for magnetic cooling at room temperatures is that in which a first order metamagnetostructural phase transition (PT) is induced by the magnetic field [1]. In this case, socalled inverse magnetocaloric effect (MCE) originates from a structural PT from the paramagnetic or antiferromagnetic martensite phase to the ferromagnetic austenite phase on the application of a magnetic field. Recently, much interest is attracted to Ni-Mn-In-Co alloys due to large magnetic-field-induced strains [2] and giant inverse MCE [3,4]. We created the new series of Ni-Mn-In-Co alloys with 43 at. % of Ni and 7 at. % of Co. The samples from this series were prepared by arc melting under an argon atmosphere with subsequent homogenizing annealing during 48 hours at 900 0C. Metamagnetic alloy Ni43Mn37.8In12.2Co7 was chosen for further research. We investigated the properties of this alloy by electrical resistance measurements (ERM), differential scanning calorimetry (DSC) and energydispersive X-ray spectroscopy (EDX). After that, samples of this alloy were exposed to additional annealing during 50 hours at 750 0C and all measuring procedures were repeated. It was determined, that electrical resistance of annealed samples is less on 30% than before annealing in martensite and austenite state too (see Fig. 1). In addition we observe a narrowing of hysteresis curve almost on 40% (from Thyst = 400 downto 250) and a shift of curve on 150 to higher temperatures. The martensitic transformation temperatures and the latent heat during the PT were determined by DSC at heating and cooling rates of 10 K/min. As seen from Fig. 1, DSC scans of the sample demonstrate exothermic and endothermic peaks which are associated with the martensitic PT occurring in the sample. The characteristic transition temperatures Ms, Mf and As, Af corresponding to start and finish temperature of direct and reverse martensitic transformation, respectively, before and after additional annealing. The transition temperatures were determined as a crossing point between the extrapolation lines of the peaks and the base line. For our alloy transformation's temperatures before annealing were found: martensite start Ms = 6 0C, martensite finish Mf = - 10 0C and austenite start As = 32 0C, austenite finish Af = 45 0C. Transformation's temperatures after annealing: MsA = 26 0C, MfA = 14 0C and AsA = 39 0C, AfA = 54 0C. The Curie temperature of austenite state does not depend on heat treatment: TC = 150 0C. Calculated from the DSC data the latent heat upon direct (cooling) and reverse (heating) PT are LC = + 3.8 J/g and LH = - 4.3 J/g in both cases (differences ~ 10% - in limits of error). The EDX analysis of samples before and after annealing was conducted; SEM micro-photos of investigated fields are presented on Fig. 2. The black areas are flaws. The white areas (martensite plates) have the following average composition in at. %: Ni42.4Mn37.2In13.9Co6.5. The grey area (main body) has the similar composition. The significant difference in composition observed in dark-grey areas (the second phase): Ni37.4Mn40.0In1.0Co21.6. These areas have little 1% of In and a lot 21.6% of Co. One can see that annealing increases grain size and helps to reveal additional information about inner structure. Microsc. Microanal. 21 (Suppl 3), 2015 1758 [1] T. Krenke, E. Duman, M. Acet, et al., Nature Materials 4, (2005), p. 450. [2] R. Kainuma, Y. Imano, W. Ito, et al., Nature 439, (2006), p. 957. [3] J. Liu, T. Gottschall, K.P. Skokov, et al., Nature Materials 11, (2012), p. 620. [4] A. Kamantsev, V. Koledov, E. Dilmieva, et al., EPJ Web of Conferences 75, (2014), p. 04008. [5] The authors acknowledge funding from the Russian Sciences Foundation, Grant 14-22-00279. Figure 1. ERM and DSC investigations of metamagnetic Ni43Mn37.8In12.2Co7 alloy before and after additional annealing during 50 hours at 750 0C. Figure 2. SEM micro-photos of region of EDX analyses of metamagnetic Ni43Mn37.8In12.2Co7 alloy before and after additional annealing during 50 hours at 750 0C.");sQ1[878]=new Array("../7337/1759.pdf","A microstructure comparison of Iron borides formed on AISI 1040 and D2 steels","","1759 doi:10.1017/S1431927615009575 Paper No. 0878 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A microstructure comparison of Iron borides formed on AISI 1040 and D2 steels J. L. Bernal-Ponce1, A. Irvin-Mart�nez1, E. Vera-C�rdenas1, A. Garc�a-Barrientos2, A. Medina-Flores3, L. B�jar-G�mez3, A. Juanico4, S. Borjas-Garc�a5, 1. 2. Automotive Mechanics Department, Universidad Polit�cnica de Pachuca, Hidalgo, M�xico. Electronics Department, Universidad Aut�noma del Estado de Hidalgo, Hidalgo, M�xico. 3. Metallurgical Research Institute, UMSNH, Michoac�n, M�xico. 4. Industrial and Materials Department, Universidad Polit�cnica del Valle de M�xico, Tultitl�n, M�xico. 5. Institute of Mathematical and Physical, UMSNH, Michoac�n, M�xico. Due to its special properties, boron compounds are developed by numerous ways, .i.e. boronizing gas, molten salt boronizing, with and without electrolysis and pack boronizing. Pack boronizing have attracted increasing interest from the technological and scientific point of view, further it has the advantage of simplicity and cost-effectiveness with other boronizing techniques [1]. Wearing resistance, toughness and corrosion, oxidation resistance, and hardness are some of the properties that are significantly enhanced by this thermochemical diffusion process [2-4]. These interesting properties are related with the microstructure that make these coatings excellent candidates to use in applications involving sliding contact and abrasion wear situations [5]. The phase formed in the substrate, its thickness and morphology have interest for study because they are related to their properties. In the present study, the microstructural characterization of iron borides developed at different temperatures by pack-boronizing has been carried out in a AISI 1040 carbon steel and a AISI D2 alloy steel. The AISI 1040 and AISI D2 were borided. The samples had a disc shape with a diameter of 18 mm and a thickness of 3 mm. Prior to the boriding process; the samples were polished, ultrasonically cleaned in an alcohol solution and deionized water for 15 minutes at room temperature, and dried and stored under clean-room conditions. Then the samples were embedded in a closed cylindrical case, containing a fresh Durborid powder mixture. The active boron is then supplied by the powder quantity placed over and around the material surface. The powder-pack boriding process was performed in a conventional furnace under a pure argon atmosphere. The boriding process was carried out at two different temperatures 1220 K and 1320 K for a time of 8 h. The boriding temperatures were selected according to the position of the solidus line in the Fe-B phase diagram. Once the treatment was completed, the container was removed from the furnace and slowly cooled to room temperature. Finally, the presence of borides formed on the surface of AISI 1040 and AISI D2 steels were confirmed by Scanning Electron Microscopy (SEM), Energy Dispersive Spectroscopy (EDS) and X-Ray Diffraction (XRD) techniques. Figure 1 shows (SEM) images of boron deposited on the surface of AISI D2 and AISI 1040 steels, the XRD scans (Figure 2) of the borided samples confirmed the presence of a single FeB phase formed in the D2 steel at 1220 K (Fig. 2a) and 1320 K (Fig. 2c). A double phase borided layer FeB+Fe2B was confirmed by XRD in the AISI 1040 samples treated at 1220 K (Fig. 2b) and 1320 K (Fig. 2d). Fe2B layers has a columnar morphology and dense structure. The depth of borides formed on AISI D2 is lower than that of AISI 1040 as expected. The chemical composition of borided samples are given in Table 1 showing that all samples contain deposited boron on its surface. These results demonstrate that the process of boron powder-pack applied to AISI 1040 and AISI D2 steels formed superficial layer FeB or Fe+Fe2B depending on the heat treatment and the chemical composition. Microsc. Microanal. 21 (Suppl 3), 2015 1760 [1] Keddam, Chentouf SM. A diffusion model for describing the bilayer growth (FeB/Fe2B) during the iron powder-pack boriding. Appl Surf Sci. 252 (2005), p. 393 [2] Bektes M, Uzun O, Akturk A, Ekinci AE, Ucar N. Vickers microhardness studies of Fe-Mn binary alloys. Chin J Phys 42 (2004), p. 733. [3] Xu CH, Xi JK, Gao W. Improving the mechanical properties of boronized layers by superplastic boronizing. J Mater Process Technol 65 (1997), p. 94 [4] Ucisik AH, Zeytin S, Bindal C. Boride coating on iron based alloys. J Aust Ceram Soc 37(1) (2001), p. 83. [5] M.A. B�jar, E. Moreno. Abrasive wear resistance of boronized carbon and low alloy steels. Journal of Materials Processing Technology 173 (2006), p. 352 Fig. 1 Fig. 2 Figure 1. SEM cross-sectional micrograph, and XRD (Fig. 2) diffraction patterns of borided samples: (a) AISI D2 at 1220 K, (b) AISI 1040 at 1220 K, (c) AISI D2 at 1320 K, (d) AISI 1040 at 1320 K. Borided sample AISI D2 1220 �K AISI 1040 1220 �K AISI D2 1320 �K AISI 1040 1320 �K C 10.25 19.63 7.81 19.07 Cr 4.73 0 4.22 0 Mo 0 0 0 0 V 1.17 0 1.04 0 Si 0.68 0 0.13 0 Ni 0 0 0 0 Mn 0 0.31 0.20 1.06 S 0 0.01 0 0.02 P 0 0.02 0.03 0 B 11.34 10.32 15.76 17.15 Fe 71.82 69.71 70.81 62.70 Table 1. The chemical composition of borided samples [wt. %]");sQ1[879]=new Array("../7337/1761.pdf","Effect of Preheating Temperature on the Microstructural Features of Welded Rail Head","","1761 doi:10.1017/S1431927615009587 Paper No. 0879 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Effect of Preheating Temperature on the Microstructural Features of Welded Rail Head K. R. Prayakarao1 and H. A. Aglan1 1. College of Engineering, Tuskegee University, Tuskegee, AL 36088 USA Preheating can be defined as heating the base metal(s) to a certain temperature before welding. Preheating serves four major purposes: (a) to decrease cooling rate, which produces ductile microstructure that increases resistance to cracking and helps in diffusion of hydrogen, (b) to reduce shrinkage stress between the weld zone and base metal, (c) to prevent chilling effect (`cold start') and ensure proper fusion, and (d) to eliminate moisture from the sample surface [1, 2]. However, excessive preheating is expensive and yields defects such as thermal distortion in the welded component [3]. On the other hand, inadequate preheating results in different types of cracking, insufficient fusion, and penetration [4]. Carbon equivalent (CE) is used as a tool for approximating proper preheats [2]. Besides CE, optimum preheating temperature also depends on section thickness, restraint, ambient temperature, filler metal hydrogen content, and previous cracking problems [1]. The focus of this research was to evaluate the microstructure of welded pearlitic rail steels preheated at different temperatures. Multipass gas metal arc welding (GMAW) was used to weld the rail steels. Four different preheating temperatures were used (200oC, 300oC, 350oC and 400oC). Mechanical testing, metallographic analysis, and fracture behavior analysis were carried out on the welded samples. The optimum preheat temperature has been identified in view of the welding efficiency and the fracture resistance of the welded rail head. The microstructures of the weld zones for all the welded samples with different preheat temperatures were similar and mostly evident of a mixture of acicular ferrite and bainite. However, a different weld microstructure, which is a mixture of martensite and bainite, was observed for the samples welded with no preheat (RT). Figures 1 and 2 represent micrographs taken at the fusion and weld zones for both room temperature and 200 oC preheated samples. In the case of dissimilar filler welding, a small portion of the base metal melts and mixes with the filler material to form a weldment. The percentage of the weld that comes from base metal during welding is known as the dilution percentage. In the present work, since there is a large difference in the chemistries of the parent rail and the filler material (especially wt% C), the weld composition might vary depending on the percentage of dilution. Considering the base metal dilution, the carbon percentage of the weld for different preheating conditions was determined by weld cross-section and the composition calculation. It is clear that the wt% C for the weld increases with increasing preheat temperature. Another important factor of preheating is that it decreases the cooling rate. Since an increasing preheat temperature will increase the wt% C in the weld, the CCT diagram will shift to right. In contrast, a higher preheat temperature will reduce the cooling rate and will also shift the cooling curve to the right. Due to the above mentioned combined effects, the cooling curves for different preheat temperatures most likely cut the CCT diagram in the same region. Hence, we can expect a similar weld microstructure for different preheat temperatures. However, for welded sample with no preheat (RT), the cooling rate will be much higher when compared to other preheated samples, as well as to the parent rail, and will act as a heat sink. It is believed that due to this very high cooling rate, martensite and bainite were formed in their microstructures. Since a higher preheat temperature involves extra energy to achieve, 200o C can be a recommended preheat for welding of pearlitic rail steel. Microsc. Microanal. 21 (Suppl 3), 2015 1762 References [1] R.S. Funderburk, `Talking your weld's temperature', Proc. on `North American steel construction conference', Las Vegas, USA, February 2000, Modern Steel Construction. [2] BOC Library: AU: IPRM:2007: Section 8: Consumable, p. 326-329, available in http://www.bocworldofwelding.com.au/media/pdf/file/library/WOWLibraryPreheating%20of%20materials-Consumables.pdf (accessed 28th May' 2013). [3] T. Kasuya and N. Yurioka, Determination of necessary preheat temperature to avoid cold cracking under varying ambient temperature, ISIJ International, 1995, 35(10), p. 1183-1189. [4] L. Baughurst and G. Voznaks, `Welding defects, causes and correction', Australian Bulk Handling Review, July/August 2009. [5] This work was supported by the Federal Railroad Administration (FRA), DOT. The guidance of Mr. M. Fateh, the technical monitor is appreciated. 20 m (a) (b) 20 m Figure 1. Micrographs of welded sample with no preheat, (a) near fusion (b) weld 20 m 20 m (a) (b) Figure 2. Micrographs of welded sample at 200oC preheat, (a) near fusion (b) weld 20 m");sQ1[880]=new Array("../7337/1763.pdf","Study of precipitation along a concentration gradient","","1763 doi:10.1017/S1431927615009599 Paper No. 0880 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Study of precipitation along a concentration gradient C. G. Garay-Reyes1, I. Estrada-Guel1, J.L. Hern�ndez-Rivera2, J. M. Mendoza-Duarte1, E. CuadrosLugo1, S. E. Hernandez-Martinez2, J .J Cruz-Rivera2 and R. Mart�nez-S�nchez1. Centro de Investigaci�n en Materiales Avanzados (CIMAV) Miguel de Cervantes No. 120, 31109, Chihuahua, Chih., M�xico. 2. Universidad Aut�noma de San Luis Potos�, Instituto de Metalurgia, Sierra leona 550, Col. Lomas 2 secci�n, 78210, S.L.P, M�xico. The present work is based on the microstructural characterization method proposed by T. Miyazaki [13], the so-called Macroscopic Composition Gradient (MCG) method. This technique allows the investigation of phase transformations in a single specimen and helps to evaluate the mechanical properties for different alloy compositions. It is based on the microstructural observation of a continuous concentration gradient, which can be generated by several methods, for instance, diffusion coupling, imperfect homogenization of coarse discontinuous precipitates, etc. Buttons of Ni�11.5 wt. % Ti alloy and pure Ni were melted in an electric-arc furnace under an argon atmosphere using pure elements (99.9 %). An assembly consisting of the buttons was placed into an austenitic stainless steel holder with two screws, encapsulated into a quartz tube under an argon atmosphere and heat treated at 1200 �C for 28 h to promote the diffusion and generate the concentration gradient in the diffusion couple, subsequently, the diffusion couple was isothermally aged at 850, 750 and 650 �C for different times. Microstructural characterization was carried out by High Resolution Scanning Electron Microscopy (HR-SEM) using a JSM-7401F microscope with Energy Dispersive Spectroscope (EDS). The diffusion process that occurs during annealing and aging treatments produces a characteristic microstructure in the diffusion couple, where the Kirkendall effect and a mixture of phases are evidenced. The Fig. 1a shows the microstructure at the interface of the Ni�13.75 Ti (at. %)/Ni diffusion couple after annealing at 1200 �C. The variation of Ti concentration as a function of distance is also shown in this figure evidencing the concentration gradient at the diffusion couple interface. A region of about 140 m that goes from the interface to the Ti-rich side, delimited by the solvus line, exhibits the presence of voids, which are formed due to the different diffusion rates of the diffusing elements. As reported elsewhere [4], the Ni diffusion rate is higher than that of Ti. Figs. 1b and 1c, show the solvus line and the precipitation boundary of the ' phase in samples thermal aged at 850 �C. The phases observed in these figures with cuboidal-shaped morphology correspond to ' phase and those with plateshaped morphology to Ni3Ti precipitates (-D024). The solvus and the precipitation boundary of the ' phase determined experimentally by EDS were found at 9.16 and 9.92 Ti (at. %), respectively. These values are close to the corresponding values in the Ni-Ti phase diagram [5]. The variation in Vickers hardness (HVN) as function of aging time in Ni-rich Ni-Ti alloys with different Ti concentration is shown in Fig. 2. The maximum hardness observed (under all temperatures) is related with the presence of ' precipitates. In addition, it is observed that as aging temperature decreases, the fv of precipitates and the HVN increase, but at concentrations less 6 at. % Ti there is not precipitation hardening at 3 aging temperatures studied. 1. Microsc. Microanal. 21 (Suppl 3), 2015 1764 References [1] T. Miyazaki, T. Koyama, S. Kobayashi, Metall Mater Trans 27A (1996). p. 949-954. [2] T. Miyazaki, S. Kobayashi, T. Koyama, Metall Mater Trans 30A (1999), p. 2783-2789. [3] T. Miyazaki, Prog Mater Sci 57 (2012), p. 1010-1060. [4] S. Hinotani, Y Ohmori, J Jpn Inst Metals 29 (1988), p. 116-124. [5] P. Vyskocil, et al, Acta Mater 45 (1997), p. 3311-3318. Figure 1. a) Optical micrograph of Ni�13.75 Ti (at. %)/Ni diffusion couple and Ti concentration profile, b) and c) FE-SEM images indicating the solvus line (-�-�-) and precipitation boundary of the ' phase (��-��-). Figure 2. Age-hardening curves obtained as a function of Ti concentration at 850, 750 and 650 �C.");sQ1[881]=new Array("../7337/1765.pdf","Evolution of microstructure in an Al-Si system modified with the transition element addition and its effect on mechanicals properties.","","1765 doi:10.1017/S1431927615009605 Paper No. 0881 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Evolution of microstructure in an Al-Si system modified with the transition element addition and its effect on mechanicals properties. H. M. Medrano-Prieto, C.G. Garay-Reyes, C.D. G�mez-Esparza, R. Mart�nez-S�nchez. Centro de Investigaci�n en Materiales Avanzados (CIMAV), Laboratorio Nacional de Nanotecnolog�a, Miguel de Cervantes No. 120, C.P. 31109, Chihuahua, Chih., M�xico. The effect of transition element addition and solution treatment time on the microstructure and hardness of the Al-Si alloy were studied by Vickers microhardness, Rockwell B Hardness, X Ray Diffraction (XRD), Optical Microscopy (OM) and Scanning Electron Microscopy (SEM). The A319 alloy and the A319 alloys modified with Ni were solution treatment at 495 �C for 5 and 7h, quenching in water at 60 �C and aged at 170 �C for 0.5, 3, 5, 10 and 96 h. The Ni addition between 1 and 2 wt. % to the A319 alloy have a direct effect on the microstructure, the morphology, size and distribution of precipitates during aging heat treatment, as well as, favors the formation of the Al-Fe-Ni, Al-Ni-Cu- and Al3Ni2 intermetallic phases, additionally, the Ni presence reduce the Cu content in the matrix for the Al2Cu formation due formation of Al-Ni-Cu phases. The A319 alloy is commonly used in the automotive industry as a material for engine blocks and cylinder heads. In order to improve the mechanical properties of such components they are frequently T6 heat treatment. In aluminum alloys, some transition metals like Ni and Fe, and some rare earths like Ce, which the main characteristic is their low solubility in Al (maximum of 0.01% to 0.03%), are employed mainly to reduce the coefficient of thermal expansion [1]. Additionally, the Ni is commonly used in aluminum alloys to improve the mechanical properties at elevated temperatures. For example, additions of 1 to 2% Ni to 2xxx and 3xxx series alloys enhance hardness and tensile properties at elevated temperatures [2-3]. Hayajneh et al. [4] studied the effect of Ni additions in the mechanical response in Al-Cu alloys; they reported that the presence of the Al3Ni, Al3(CuNi)2 and Al7Cu4Ni intermetallic compounds have a direct relationship with mechanical properties. Higher amounts of dispersed intermetallic compounds higher hardness. Others investigations have reported the presence of Al9FeNi Intermetallic compound [5]. The figure 1, show the microstructure of as-cast conditions in A319 alloy modified with Ni additions, where is observed a change in morphology of dendrites and interdendritic phase (fig. 1a), as well as the presence of platelets-like phases (fig. 1b). When Ni is added to the Al-Si system the eutectic transformation is characterized by a simultaneous formation of eutectic Si and Al3Ni phase. In figure 2, it is observed during the solution treatment as the eutectic Si platelets spheroidize and their aspect ratio decreases, which result in a loss of interconnectivity of the eutectic phases, which has been widely reported [6-7]. Additionally, the Al-Ni-Cu phases with Chinese script type and plate morphologies can be observed. Microsc. Microanal. 21 (Suppl 3), 2015 1766 Fig. 1 OM micrographs of A319 alloy modified with Ni additions in the as-cast condition: a) 1 wt. % Ni and b) 2 wt. % Ni. Fig. 2 SEM micrograph and elemental mapping of the A319 alloy with 1 wt. % Ni after solution treatment for 5 h. References. [1] Jack W Bray in "Properties and Selection: Non Ferrous Alloys and Special Purpose Materials" , 10 ed. (ASM International, USA) p. 165-166. [2] W. S. Miller, L. Zhuang, J. Bottema, A. J. Wittebrood. P. D. Smet, A. Haszler, A. Vieregge, Mater. Sci. Eng 280 (2000) p. 37�49. [3] R. S. Rana, Rajesh Purohit, and S Das, Int. J. Sci. Res. Pub 2 (2012) p. 1-7. [4] Mohammed T. Hayajneh, Adel Mahamood Hassan, Younis Mohammad Jaradat, Jordan Univ. Sci. Technol 141 (2007) p. 1-5. [5] J. K. Wessel in "Handbook of advanced materials, enabling new designs", 1 ed. (John Wiley & Sons, New Jersey) p. 185-193. [6] M. A. Moustafa, F. H. Samuel, H. W. Doty, J. Mater. Sci 38 (2003) p. 4523� 4534. [7] Kyu-Sik Kim, Si-Young Sung, Bum-Suck Han, Chang-Yeol Jung, and Kee-Ahn Lee, Metall. Mater. Int 20 (2014) p. 243-248.");sQ1[882]=new Array("../7337/1767.pdf","Characterization of Microstructural Inhomogeneities in Rolled Microalloyed Steels Based on EBSD Data.","","1767 doi:10.1017/S1431927615009617 Paper No. 0882 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of Microstructural Inhomogeneities in Rolled Microalloyed Steels Based on EBSD Data. Nikolay Y. Zolotorevsky1, Sergey N. Petrov2, Sergey N. Panpurin1, Alexander A. Kazakov3 and Olga Pakhomova3 1. 2. Department of Mechanics and Control Processes, Polytechnical University, Saint-Petersburg, Russia. Central Research Institute of Structural Materials "Prometey", Saint-Petersburg, Russia. 3. Metallurgical Technologies Department, Polytechnical University, Saint-Petersburg, Russia. The method of quantitative metallography has been developed recently making it possible to evalutate bainitic microstructure of modern pipeline steels 1. When using this method, the microstructural inhomogeneities identified as the blocks of bainite with lath-like morphology was revealed among different morphological forms elongated along the rolling direction. Moreover, these blocks were shown to reduce the essential mechanical properties of the plate. In the present study, the crystallographic features of such inhomogeneities was examined by means of the SEM using EBSD analysis in order to characterize them more rigorously and clarify their nature. Industrially processed low-carbon microalloyed steel was investigated. Steel plate was hot rolled from a 300-mm slab to a 27.7-mm plate under industrial conditions. Last passes were conducted at nonrecrystallization temperatures 750�40�C, and a true strain accumulated therewith was about 1.6. After rolling, the steel was cooled with a rate about 10 K/s. The samples for SEM examination were prepared from a central region of the steel plate. EBSD analysis was carried out by SEM Quanta 200 3D FEG using the EDAX Pegasus system with a step size 0.5 m. The orientation relationship (OR) between and phases, the statistics of crystal misorientations and the orientations of the former austenite were determined for comprehensive characterization of the microstructure. Fig. 1a presents an example of the microstructure, where the coarse-grained region corresponds to the inhomogeneity found previously by optical metallography. The differentiation of fine-grained and coarse-grained regions is supported in Fig. 1b by grain average misorientation (GAM) mapping. The scale of color differentiation used here was suggested recently to distinguish various -phase species 2. One can see that the fine-grained regions are colored partially by blue, green and yellow. These colors indicate a relatively low level of crystal curvature. In particular, the blue color corresponds to the polygonal ferrite 2. Alternatively, coarse-grained regions are colored in red (high level of crystal curvature). This corresponds to species formed by a displacive transformation and thus confirms that we are dealing with bainite. Multiple variants of -phase lattice orientations may be produced within a single former austenite grain due to crystal symmetry. However, the so called "variant selection" leads to the predominance of certain inter-variant misorientations. Different kinds of bainite were shown to have different spectrums of the inter-variant boundaries [3-5]. The spectrum obtained in the specimen examined is shown in Fig. 2a (V1...V24 indicate possible inter-variant misorientations). It corresponds to the spectrums observed earlier for upper bainite 3-5. This is supported by the ORs determined on the base of inter-variant misorientations for three regions (Fig. 2b). The ORs are shown here together with the ORs reported by Takayama et al. [3]. The latter ones were obtained for a bainite in low-carbon steel formed during isothermal holding (the holding temperatures are indicated at the plot). One can conclude from the Microsc. Microanal. 21 (Suppl 3), 2015 1768 comparison that we have the bainite formed at relatively high temperature. Orientations of former austenite grains have been determined within the regions having microstructures of various types. They were found to correspond well to a rolling texture of FCC metals; most of them belong to the -fiber. Therewith, the austenite, from which relatively fine-grained microstructure is produced, has orientations distributed uniformly within the -fiber. Alternatively, the austenite, which produces the coarse-grained microstructural inhomogeneities, has orientations concentrated around the "brass component" {110} 112 . This observation can be explained by an orientation dependency of austenite deformation behavior, since peculiarities of the brass orientation � lower level of misorientations and greater homogeneity [6] � lead to a more strict variant selection during phase transformation, which as a result promotes formation of larger bainite packets as well as formation of bainite blocks with low misorientations. References: [1] A A Kazakov et al., CIS Iron and Steel Review, 2012, p. 4. [2] A A Zisman, S N Petrov and A V Ptashnik, Metallurgist, 58 (2015), in press. [3] N Takayama, G Miyamoto and T Furuhara, Acta Mater, 60 (2012), p. 2387. [4] S N Panpurin et al., Materials Science Forum, 762 (2013), p. 110. [5] N Y Zolotorevsky et al., Metal Sci. and Heat Treatment, 55 (2014), p. 550. 6 L Delannay et al., Acta mater. 49 (2001), p. 2441 (a) (b) Figure 1. Boundary (a) and GAM (b) maps of specimen examined. The boundaries are drawn by green (2<<5), blue (5<<15) and black(>15) lines. 6 Angle between CP directions 5 4 3 2 450�C G-T Miyamoto Fish Greninger-Trojano 1 0 500�C 580�C Takayama 0 1 2 3 (b) (a) Figure 2. Length fractions of inter-variant boundaries (a) and the ORs presented by angles between planes {111} and {110} as a function of angles between directions <110> and <111> (black triangles) together with Greninger-Troiano (G-T) OR and the ORs reported by Takayama et all [3] (b). Angle between CP planes");sQ1[883]=new Array("../7337/1769.pdf","Effect of Cu on Hydrogen Permeation for the Low Carbon Steel under Mildly Sour Environments","","1769 doi:10.1017/S1431927615009629 Paper No. 0883 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Effect of Cu on Hydrogen Permeation for the Low Carbon Steel under Mildly Sour Environments Haruo Nakamichi1, Kazuhiko Baba1, Daisuke Mizuno2, Kyono Yasuda3 and Nobuyuki Ishikawa3 1. 2. Analysis & Characterization Res. Dept., Steel Res. Lab., JFE Steel Corp., Kawasaki, Japan. Material Surface & Interface Science Res. Dept., Steel Res. Lab., JFE Steel Corp., Kawasaki, Japan. 3. Steel Products Res. Dept. Steel Res. Lab., JFE Steel Corp., Fukuyama, Japan. Hydrogen Induced Cracking (HIC) and Sulfide Stress Cracking (SSC) are the major issue for low carbon alloyed steels applied to sour line pipes at oil and natural gas fileds. It is caused by the presence of hydrogen sulfide, which enhances hydrogen entry into steels and causes the environmental cracking. It is well known that Cu is effective element for preventing HIC in mildly sour environments in which the solution pH is more than 5.0[1]. However, the mechanism of Cu rolls on HIC prevention has not been fully clarified. It was found that Cu addition remarkably decreased hydrogen entry into steel[2] and Cu was enriched in the corrosion product based on EPMA micron order analysis [3]. In present paper, corrosion test under H2S environments are carried out with measuring the amount of hydrogen permeation. Microstructural morphologies of corrosion products, especially Cu distribution and morphologies are investigated through FIB/SEM and STEM analysis for understanding its rolls on preventing the corrosions. Two samples, concentration of 0.06C-1.7Mn-0.1Si with and without 0.3wt%Cu, were prepared through a vacuum melting. Corrosion tests were performed under 100%H2S gas environment in 5%NaCl+1N(CH3COOH+CH3COONa) solution for 96hrs. The hydrogen permeation was measuring during corrosion test. For evaluating effects of solution environment on corrosion product, two pH conditions of 5.0 and 5.3 were performed. After corrosion test, samples were picked up from the solution and SEM and TEM observations were carried out. Cross section samples of SEM and TEM were prepared by FIB techniques. Distribution of alloy elements in the corrosion product and steel were measured through EDS analysis equipped with SEM and TEM Fig.1 shows plane view SEM images of corrosion product of Cu added steel immersed in the pH 5.3 and 5.0 solutions. It is observed that the surface morphology of corrosion products formed in high pH solution were smooth compared with that of in low pH solution. Micrographs of SEM cross sectioning are show in Fig. 2, together with the hydrogen permeation value at 90 hours of corrosion test. Several micro-meter thickness of corrosion product is formed during immersion test and there are roughly two layers in that product. Different microstructure is recognized dependent on the Cu concentration and pH value. The upper layer formed on the Cu added steel in high pH solution is uniform even though it structure is porous. On the other hand, that of formed on non-Cu added steel in low pH solution is very rough and has many large cracks. The bottom layer formed on the Cu added steel in high pH solution has dense structure and no cracks. The hydrogen permeated value is increasing when the corrosion layer has large cracks. It is supposed that adding Cu is very useful for forming uniform tight corrosion layer without large cracks for preventing hydrogen permeation. For evaluating the Cu distribution in the corrosion film, cross sectional TEM analysis is carried out and the STEM images shown in Fig.3. It is found that the small crystal size of FeS (less than 10nm) is formed at the bottom layer of corrosion products from electron diffraction analysis. The upper layer of corrosion products has bigger crystal Microsc. Microanal. 21 (Suppl 3), 2015 1770 size and has many pores. EDS mapping results of Cu in the corrosion layer is shown in Fig.4. Thickness of Cu enriched layer is about 100nm. This tiny Cu layer would prevent the hydrogen permeation and form a tighten corrosion layer in pH 5.3 solution. Based on SEM cross section result (Fig.2), this Cu layer effectiveness for preventing hydrogen permeation is decreasing under pH 5.0 solution because of less of uniformity of bottom corrosion layer in this case. It is suggested that the formation of this tight layer is very important for preventing HIC for preventing hydrogen immersion and Cu adding is enhancing to form this layer. References: [1] Y. Inohara et. al., Proceedings of 13th ISOPE, USA (2003) paper no. Symp-05. [2] Y. Nakai, H. Kurahashi, N. Totsuka and Y. Uesugi, Corrosion/82, (1982) paper no.132 [3] T. Hara and H. Asahi, Proc. Pipeline Technology Conf., (2004) Cu free-pH5.0 Cu added-pH5.0 Cu free-pH5.3 Cu added-pH5.3 m m m m Figure 1. SEM plan view of corrosion product C layer Cu free pH 5.0 1.4 mA/cm C layer Rust layer Cu added 1.0 mA/cm Rust layer Steel C layer Rust layer Steel 10 m 0.5 mA/cm Steel C layer Rust layer 10 m 0.1 mA/cm pH 5.3 10 m Steel 10 m Figure 2. SEM cross sectional view of corrosion product Porous layer Dense layer 500nm 500nm Figure 4. STEM-EDS mapping of Cu in the indicated area of figure 3 Figure 3. STEM image of corrosion product of Cu added steel in pH5.3 sample");sQ1[884]=new Array("../7337/1771.pdf","Optical and structural properties of rare earths-doped ZnO nanorods synthetized by sonochemical method.","","1771 doi:10.1017/S1431927615009630 Paper No. 0884 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Optical and structural properties of rare earths-doped ZnO nanorods synthetized by sonochemical method. R. Sanchez-Zeferino1, M.E. Alvares-Ramos1, G. Moreno-Corella2 and E. U. Le�n-Salguero2 1. 2. Departamento de F�sica, Universidad de Sonora, Hermosillo, Sonora, M�xico. Posgrado en nanotecnolog�a, Universidad de Sonora, Hermosillo, Sonora, M�xico. Nanostructured materials and semiconductors specifically seem to be important and promising in the development of next-generation electronic and optoelectronic devices [1]. In recent years, synthesis and properties of semiconductors such as TiO2, SnO2, ZnO, etc. have been investigated due to their potential and important applications in catalysis, bioimagining, wavelength tunable laser, solar cell [2, 3]. On the other hand, doping is a way to modify electronic, optical, and magnetic properties of bulk semiconductors. Dopants can strongly influence optical behavior. The undoped nanostructures are highly luminescent and emit light with wavelength which depends on size and defects, however, lasers based on this emission are intrinsically inefficient. A possible solution to this situation is adding dopants that provide carriers. Doping of nanomaterials lead to new phenomena not found in bulk because their electronic states are confined to a small volume. Luminescence phenomena are being investigated extensively in rare earths (RE) doped nanoparticles, due to RE- doped can emit light from the RE ions give rise to sharp emission through electrical injection. The intra-4f shell transitions of the RE ions give rise to sharp emission lines whose wavelengths are largely independent of both the host materials and temperature. This stability occurs because the filled outer 5s and 5p electron shells screen the transitions within the inner 4f electron shell from the interaction with the host [4]. RE-doped zinc oxide nanostructures have been synthetized using ultrasonic technique. It is a simple and cheap synthesis method which is based on acoustic cavitation phenomena. Zinc [Zn (NO3)] and Na (NaOH) salts were used as precursor. Then 1.0 mol % Dy(NO3)3, Er(NO3)3, Nd(NO3)3 and Eu(NO3)3 were added in each zinc solutions for obtaining various samples. After to prepare the solutions with pH controlled (10), these were carried out a sonication process (200 W, 20 kHz) for 1h. Selected ZnO REdoped samples were thermally annealed at 500 �C in air. Figure 1 present typical SEM images of RE-doped ZnO which shown particles agglomerated. It also shows the beginning of the formation of the clusters of nanorods. The nanorods formed have wide size distribution with diameter of 100-200 nm and lengths of 1-2 �m. The morphology did not change with others RE dopants. It is important to note that the morphology ZnO is directly related with sonication time. As this time increases, ZnO nanorods appeared well-defined. However, the aim is to determine the best conditions to grow ZnO using the least amount of energy in the synthesis. On the other hand, figure 2 present Raman spectra of RE-doped ZnO which consist of four peaks located at about 96, 380, 437, and 580 cm-1, which correspond to the E2L, A1(TO), E2H, and A1(LO) fundamental phonon modes of hexagonal ZnO, respectively. The Raman peak located at about 200 and 1150 cm-1 were assigned to the 2E2L and 2A1(LO) second-order phonon modes, respectively. Other Raman peak located at about 331 cm-1 could be assigned to the E2H -E2L multiphonon scattering modes, respectively. The Raman peak appeared at about 140 cm-1 has been assigned to B1L silent phonon mode of ZnO. However, its origin is not yet clear. In summary, RE-doped ZnO nanostructures have been successfully synthetized via an ultrasonic technique. The structure, morphology, luminescent properties and growth mechanism of nanostructures Microsc. Microanal. 21 (Suppl 3), 2015 1772 are investigated in detail. The experimental part demonstrated that the RE-doped ZnO has a similar morphology which is independent of the element dopant. RE-doped ZnO nanostructures have an excellent structural properties and this research may provide guidance for synthesis of nanomaterials by a simple method. References: [1] M. Ahmad and J. Zhu, Journal of Materials Chemistry 21 (2011) p. 599. [2] P. Kumar et al, Proceedings of the National Conference on `Advances in Basic & Applied Sciences' (ABAS-2014) p. 44. [3] C. Wang, C. Shao, X. Zhang, and Y. Liu, Inorganic Chemistry 48 (2009) p. 7261. [4] Y. Fujiwara et al, CS MANTECH Conference (2010) p. 215. Undoped-ZnO Dy-doped ZnO Eu-doped ZnO Figure 1. SEM images RE-doped ZnO nanostructures synthetized by ultrasonic method. 14000 E2H- E2L E2L Intensity (a. u.) 10000 8000 6000 4000 2000 0 2E2L A1(TO) 12000 A1(LO) E2H ZnO:Nd ZnO:Eu ZnO:Er ZnO:Dy ZnO 2A1(LO) 200 400 600 800 -1 1000 1200 Raman shift (cm ) Figure 2. Raman spectra of RE-doped ZnO nanostructures synthetized by ultrasonic method.");sQ1[885]=new Array("../7337/1773.pdf","Light Microscopic Analysis for Bone Responses to Implant Surfaces.","","1773 doi:10.1017/S1431927615009642 Paper No. 0885 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Light Microscopic Analysis for Bone Responses to Implant Surfaces. In-Sung Yeo1, Jae-Hyuk Sim2 and Taek-Ka Kwon3 1. Department of Prosthodontics, School of Dentistry and Dental Research Institute, Seoul National University, Seoul, Korea. 2. Department of Prosthodontics, Seoul National University School of Dentistry, Seoul, Korea. 3. Department of Dentistry, St. Catholic Hospital, Catholic University of Korea, Suwon, Korea. Clinically, dental implants are an excellent treatment option for missing teeth. Implant surfaces have been modified in various ways in order to accelerate biologic bone responses around the implants, resulting in the decrease of patients' edentulous periods. Histomorphometry by light microscope is known to be the most reliable method to evaluate the early biologic response at the interface between the bone and the implant surface. The purpose of the study was to investigate the early bone response to several modified implant surfaces using histomorphometric analysis by light microscope. The tibiae of New Zealand white rabbits received the modified surfaced screw-shaped titanium implants; two implants in each tibia. The first experiment compared among the calcium phosphate coated, anodized, and sandblasted surfaces; the second between the calcium phosphate and hydroxyapatite coated surfaces; the third between fluoride-treated and anodized surfaces; the last among calcium phosphate coated/anodized (CPA), sandblasted/acid-etched (SA), and anodized surfaces. Commercially pure titanium surfaced implants served as control. The rabbits were sacrificed after 2 weeks of the implant insertion. The tibial bone blocks were sectioned and processed in the undemineralized states for the microscopic analyses. Two histomorphometric quantities were measured; bone-to-implant contact (BIC), which is defined as ratio of the bone length contacted with the implant surface over the total length of the surface, and bone area (BA), which is defined as ratio of the bone formed between the implant threads over the total area between the threads. The modified surfaces investigated in the study showed higher BIC and BA than the control (P < 0.05). The CPA surface suggested higher BA than the anodized surface (P < 0.05). There were no significant differences in the rest of comparisons among the modified surfaces. In conclusion, modified implant surfaces demonstrate the accelerated bone responses from the results of microscopic histomorphometry, compared to the commercially pure titanium surface. These microscopic analyses may be adequate tools to evaluate early bone responses to implant surfaces. However, such analyses have the limit of 2-dimensional measurement. This field requires an improved micro-analysis technique for 3-dimensional evaluation. Microsc. Microanal. 21 (Suppl 3), 2015 1774 Figure 1. The measurements of the histomorphometric quantities. (A) The parts of the implant that the bone attaches with are designated as 1 (red lines), and those of the implant without bone contact are designated as 2 (blue lines). BIC is calculated as (the sum of 1s) / (the sum of 1s and 2s). (B) The total area between the implant threads is designated as 1 (in the red line polygon), and the areas without bone are designated as 2. BA is calculated as {(the area of 1) � (the sum of 2s)} / (the area of 1). References: [1] Yeo IS, et al, Journal of Biomedical Materials Research Part B: Applied Biomaterials, 87 (2008), p. 303. [2] Yeo IS, et al, Journal of the Korean Physical Society, 57 (2010), p. 1717. [3] Choi JY, et al, Implant Dentistry, 21 (2012), p. 124. [4] Koh JW, et al, The International Journal of Oral & Maxillofacial Implants, 28 (2013), p. 790.");sQ1[886]=new Array("../7337/1775.pdf","SEM and Raman Spectroscopy Applied to Biomass Analysis for Application in the Field of Biofuels and Food Industry","","1775 doi:10.1017/S1431927615009654 Paper No. 0886 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 SEM and Raman Spectroscopy Applied to Biomass Analysis for Application in the Field of Biofuels and Food Industry O. Samek1, A. Haronikova2, N. Vaskovicova1, K. Hrubanova1, J. Jezek1, I. Marova2, V. Krzyzanek1, P. Zemanek1 1. 2. Institute of Scientific Instruments of the ASCR, v.v.i., Brno, Czech Republic. Centre for Material Research, Brno University of Technology, Brno, Czech Republic. A biomass of algal (Trachydiscus minutus, Botryococcus sudeticus, and Chlamydomonas sp.) and red yeast strains (Rhodotorula spp., Cystofilobasidium spp. and Sporobolomyces spp.) has been studied due to their potential applications in the field of biofuel generation and food industry [1-2]. In order to utilize biomass for efficient industrial production, the optimal cultivation parameters have to be determined which in turn lead to high production of desired substances such as oil and carotenoids in the selected cell line [2]. Main aim of our investigations was to study � using scanning electron microscopy (SEM) and Raman spectroscopy techniques � how different cultivation conditions influence production of oil and carotenoids. Raman spectroscopy can be used for the determination of the oil present in the biomass and also for the determination of carotenoids as the intensity ratios of specific, selected Raman bands [1]. SEM uses electron beams to gain information about morphology of cells (biomass structure) which is very important factor to study cells response on the applied stress. In our experiments we have observed different morphology when employing SEM to study cells of Cystofilobasidium capitatum cultivating in the two media with different carbon to nitrogen (C/N) ratio (Figure 1). C/N ratio of cultivation media has an influence to the production of yeast lipids or carotenoids [3]. First growth medium (medium 1) indicates positive results to carotenoid production and second medium (medium 2) with high C/N ratio leads to an increased lipid production which can be indicated using Raman spectroscopy (Figure 2). Samples for SEM imaging were prepared in the following way: (i) after thawing suspension of Cystofilobasidium capitatum (cultivated aerobically at 28�C on rotary incubator at 90 rpm for 144 hours), the samples were further cultivated in the two different media for 24h at temperature 37�C, (ii) cells suspensions were processed by chemical fixation � 2 hours in 2.5 % glutaraldehyde in PBS and 30 min in 1 % OsO4, dehydrated by acetone series and dried in HDMS on the glass slides. Both images of prepared samples were scanned without any metal coating at electron beam energy 1 keV and beam current 3.1 pA in SEM Magellan (FEI). Figure 1a shows ,,clean/smooth" surface of the cells (medium 1) contrary to the surface of the cells exposed to medium 2 where cells are most likely covered by lipid crystals. This could be explained by higher osmotic pressure introduced by medium 2. Here we combined information about morphology of the sample with Raman fingerprint by which lipid and carotenoids molecules can be identified. Thus, all matrix changes within the studied cells introduced by stress response mechanisms can be visualized (SEM) and chemically characterize (Raman). However, systematic studies are still required to investigate cell stress response mechanisms in more details. Such studies are currently under way in our laboratories, exploiting combination of SEM and Raman spectroscopy approaches. Microsc. Microanal. 21 (Suppl 3), 2015 1776 References: [1] O. Samek et al, Sensors 10 (2010), p. 8635. [2] I. Marova et al, J Environ Manag 95 (2012), p. S338. [3] T. Braunwald et al, Appl Microbiol Biotechnol 97 (2013), p. 6581. [4] This work received support from the Ministry of Health, Ministry of Education, Youth and Sports of the Czech Republic (LO1212) together with the European Commission and the Czech Science Foundation (ALISI No. CZ.1.05/2.1.00/01.0017) and the Grant Agency of the Czech Republic (GA14-20012S). a b Figure 1. a) SEM image of Cystofilobasidium capitatum (medium 1); b) SEM image of Cystofilobasidium capitatum (medium 2). Figure 2. Raman spectra of Cystofilobasidium capitatum cultivated in medium 2 (increased lipid production in medium with high C/N ration is clearly visible).");sQ1[887]=new Array("../7337/1777.pdf","Addressing Catalytic Material Problems using Automated Feature Analysis","","1777 doi:10.1017/S1431927615009666 Paper No. 0887 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Addressing Catalytic Material Problems using Automated Feature Analysis David Pollick1 and Patricia Nelson1 1 BASF Corporation, Catalyst Division, Iselin, NJ, USA Electron microscopy and microanalysis have always been indispensable tools for characterizing material properties, which are strongly correlated to the overall, catalytic performance of a finished commercial product. Historically, most of the feature analysis relies heavily upon visual judgment and rigorous measurements of an electron micrograph by the microscope operator. However, such processes suffer from inconsistencies due to human errors, especially when analysis was performed by different operators. Furthermore, the rigorous measurements are time-consuming and provide poor statistical results. In a commercial setting, where analysis throughput, consistency and efficiency are of essence, a more robust approach is needed. Automated image analysis provides a quantitative means for objectively differentiating among similar materials with improved speed, essential statistics and consistency. In this paper, we describe feature analysis performed on catalytic materials using a software suite provided by Bruker-AXS within their Quantax ESPRIT imaging and spectroscopy system [1]. High resolution scanning electron microscopy (SEM) images were first acquired, followed by image-filtering and grey-level `thresholding' to enhance and/or isolate specific attribute(s) of interest. Specifically, the software provides flexibility for establishing either a binary or multiphase histogram for control and fine tuning of phase separations. Automation-related errors can be further minimized through selection rules. For example, features at the image borders can be excluded; image noise and fine contaminants can be filtered out by ignoring features smaller than a user-defined critical size. Once the feature analysis `macro' is defined, the software allows for 30 physical properties such as, volume, phase, roughness, length etc. to be rapidly determined. The feature analysis routine can be saved and reused for consistency. Results derived from each micrograph can be appended for larger sampling and better statistics. Combining energy-dispersive X-ray spectroscopy (EDS) with automated feature analysis transforms the technique into a powerful tool for solving technical problems by adding another dimension to the characterization. Chemical composition, phase analysis and physical attributes can be correlated, again, through automated algorithms. The synergistic approach allows one to find answers to questions like: `how many 30�m particles have 3% alumina?' or `what is the average diameter of particles with less than 85% silica?' Through automated image and phase analysis, rapid characterization of physical attributes and chemistry to gain insights in the performance of commercial catalyst materials is made possible. The flexibility of the software allows screening of complex materials to be achievable. Results can be accurate, consistent and independent, which is important in a high-throughput environment. [1] Bruker-Axis Quantax, User Manual, (2011) Microsc. Microanal. 21 (Suppl 3), 2015 1778 Particle Avg StdDev Min Max Area 30.9 34.5 0.8 321.4 Average Diameter 6.16 2.90 1.25 20.53 Aspect Ratio 1.36 0.35 1.02 6.57 Shape Factor 0.62 0.21 0.12 0.99 Circularity 1.69 0.63 1.06 7.56 Roughness 1.16 0.17 0.95 2.15 Figure 1. SE Image, Processed Image, Binned Histogram, and Resulting Table of Powdered Catalyst Figure 2. Variation in Precious Metal Average Diameter after Aging Protocol");sQ1[888]=new Array("../7337/1779.pdf","Coloration Defects in Industrial-Grade Zirconium Sponge","","1779 doi:10.1017/S1431927615009678 Paper No. 0888 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 447,943010.9833/:897, 7,/07.43:2$5430 .,,892,3,3/:3,4 !479,3/$9,90&3;0789 05,79203941!8.8 !479,3/ # &$ !479,3/$9,90&3;0789 05,792039410.,3.,,3/,907,8330073 !479,3/ # &$ 8,/0,55.,-0,43 2,907, 7.43:2 8,3088039,5,7941 2,3 3/:89708 ,07485,.0 3:.0,70330073 570.8432,.33 09. ,3/841903/02,3/0/35:7988902,9.89:/ ,890701470,:3.0/347/0794.42-,9,97,.0 0;0.447,943/010.9/8.4;070/33/:897, 7,/0 7.43:2 85430%0/0 ;,70941 247544., 10,9:708,3/.447,9430110.9857080390/ -90 /010.9 128 7/08.039 -:0 5:750 2488 4;0 7003 -743 -7. 70/ 4/03 897, 09. . ,70 4-807;0/.43.:770392,0897,/943,2.748.45,3/850.9748.45,3/0,,;03:01473;089,93 90.42548943,3/1472,94320.,38284190128 $.,33300.97432.748.45 $ 2,38489,990128:71,.08,7094547,5.,3/ ,990 82,07 8.,0 .439,3 2.745474:8 10,9:708 94: 90 8:71,.0 ,9 98 8.,0 ,84 ,8 .0,7;8-0.789,301,.098 908:71,.0,8,70#,2,33,.9;084,90/;,.,3.08419012 41903 0-90/ 850.97, .,7,.90789. 41 2434.3. 7.43:2 4/0 . 4..:78 3,9;0 ,9 7442 902507,9:70 3,//943 34:.3080704-807;0/033;089,90/900.9743-,. 8.,9907 /117,.943 $ ":,399,9;0 0307 /85078;0 7, 850.9748.45 $ 8:0898 9,9 90 12 8:71,.08 ,;0 7 7 ,3/ 70 ,942. 7,948 41 ,3/ 70850.9;0 43;0780 7,948 41 ,3/ 070 4-807;0/ 3 90 7 85430 -: 147 ,3/ 0 70850.9;0 -:4037,948,700774304:8/:0949002-0//320/,:80/3.7488 80.9438,250 5705,7,943 7,549400.9743850.97, !$ .40.90/174290 12 8:71,.0 8487/,3/ 850,8 .,703/.,9;0417.43:24/0 (%00550,88:089,0 470 20/;,03. ( 40;07 /090723,943 41 5488-0 .4730 .439,33 .4254:3/8 8:. ,8 7.43:2 47 743 4.47/0702,38,2-:4:8/:094904;07,541905-3/3030730,7 0' 347/07940;,:,9090/897-:94341908025:790838/090785430 .7488 80.9438,2508070 5705,70/-708302-0//3 9380.9433 5483 ,3/.,7-43.4,93-0;,547,943 32 $ 2,553 41 .7488 80.943 8,2508 70;0,8 80;07, 573.5, .4770,9438 3 90 -: 41 90 2,907, 30,7 90 12 8:71,.0 , 5990/ 7043 -038 94 ,780 30,7 90 12 8:71,.0 .4730 .439,33570.59,908-0.420 1307 380,3/ 24700;03/897-:90/ 743 7.-:70438/4 349 503097,90 394 9080 /850780/ .4730 70438 403 .439039 2, ,84 82,7 /0.70,80 30,7987043%043437080,7. 3903/894;0719080 13/3843, ,707 8.,0 ,3/ /0391,3.,:8,70,943858-0900390804-807;,9438,3//010.9121472,943 #010703.08 (47,3909, $:71,.0$.03.0 5 (,770.,09, $:71,.0$.03.0$50.97, 5 ( :,3/74,, 55!8 5 (%85740.9,88:554790/35,79-90 704309,839,9;0 Microsc. Microanal. 21 (Suppl 3), 2015 1780 1768 :70$2,0841128:71,.08 , '/010/41;0,3/ - '3,77410/41;0 :70!$850.97,41/010.9123/.,9390 , 7/,3/-,.74:3/ 8:-97,.90/ - 8,3/ . 0550,8 :70 , $2,04102-0//0/.7488 80.9438,250,3/.47708543/3$2,5841 - , . , ,3/ / 0,50,8 ,.40.90/,9' 128:71,.0894,7/890019410,.2,0");sQ1[889]=new Array("../7337/1781.pdf","Synthesis of Mesoporous Zirconia by Using CTAB as Template","","1781 doi:10.1017/S143192761500968X Paper No. 0889 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Synthesis of Mesoporous Zirconia by Using CTAB as Template Salom�n E. Borjas-Garc�a1, Ariosto Medina-Flores2, L. B�jar2, C. Aguilar3, J. L. Bernal4. 1 Instituto de F�sica y Matematicas, Universidad Michoacana de San Nicol�s de Hidalgo, Morelia, Michoac�n, M�xico 2 Instituto de Investigaciones Metal�rgicas, Universidad Michoacana de San Nicol�s de Hidalgo, Morelia, Michoac�n, M�xico 3 Depertamento de Ingenier�a Metal�rgia y Materiales. Universidad t�cnica Federico Santa Mar�a. Valparaiso, Chile. 4 Automotive Mechanics Department. Universidad Polit�cnica de Pachuca. Zempoala, Hidalgo. M�xico A mesoporous material is a material which contains pores in a range between 2 and 50 nanometers [1]. This type of material can be used as catalyst for bulky reactant molecules due their large pore size. The first publication about synthesis of mesoporous materials by using a surfactant template was in 1992 by Mobil scientists [2]. After that, several methods have been develop for the synthesis of mesoporous metal oxides using cetyltrimethylammonium bromide (CTAB) [1,3] or triblock copolymer non ionic surfactant as a template [1,4]. However, there is not enough research about the synthesis of mesoporous zirconia by using CTAB as template or the methods used are very complex. Mesoporous zirconia was synthesized using both Sol � Gel method with a hydrothermal soft treatment. The mesopores was prepared by using zirconium oxide chloride octahydrate (Sigma-Aldrich, purity 99.5%), Sodium hydroxide (J.T.Baker, purity 97%) and hexadecyltrimethyl ammonium bromide, CTAB (Sigma, purity 99%) as source, alkaline material and template, respectively. In a first step, two solutions were prepared. The first one was obtained by dissolving 1.611 g of ZrOCl2*8H2O and 3.645 of CTAB in 20 g of distilled water. For the second solution, 0.4 g of NaOH was dissolved in 10 g of distilled water. In a second step, the Na-solution was added slowly (drop by drop) to Zr-solution and stirred. The final solution was stirred and heat in a hot stir plate at 90 �C to get a material with a molar ratio of ZrOCl2*8H2O:CTAB:NaOH:H2O equal to 1:2:8:75. The gel obtained was aged in a polypropylene bottle at 80 �C for 1 day. After the hydrothermal treatment, the sample was washed with 100 ml of distilled water and centrifuged at 4000 rpm. After that, the material was dried at 80 �C for 1 day. The powder X-ray diffraction patterns were collected with a siemens D5000 X-ray diffractometer equipped with graphite monochromatized high�intensity CuK (=1.54178 �). The Bragg angle 2 ranges from 1� to 8� at a scanning rate of 0.02�/0.6 s. The surface morphology images of the samples were analyzed by using a scanning electron microscopy FEG-SEM JEOL JSM 7600. Figure 1 shows XRD pattern of zirconia synthesized. This figure shows a characteristic peak about the presence of mesopores in the material around 1.04 degrees. Figure 2a shows an image of zirconia material which has mesopores with a size around 28 nm. Figure 2b shows the EDS spectrum taken over material analyzed. The results showed that the molar relation between template and zirconium could be crucial in the formation of mesopores. However, by soft hydrothermal treatment and low amount of water could help the formation of mesoporos in the material let us developed an easy procedure for the synthesis, which can be applied at different materials. Microsc. Microanal. 21 (Suppl 3), 2015 1782 References [1] D.W. Bruce, D. O�Hare and R.I. Walton, "Porous materials", 1st ed. (Wiley, United Kingdom, 2010) p. 1. [2] C. T. Kresge et al, Nature Volume 359 (1992) p. 710. [3] Z. R. Tian et al, Science Volume 276 (1997) p. 926. [4] P. Yang et al, Nature Volume 396 (1998) p. 152. [5] The authors acknowledge funding from H. Consejo T�cnico of Institute of Physics and Mathematics, and Consejo Nacional de Ciencia y Tecnolog�a (CONACyT), M�xico. Figure 1. XRD pattern of mesoporous zirconia. Figure 2. a) SEM image of mesoporous zirconia after synthesis (40000x of magnification), b) EDS spectrum taken from over one particle.");sQ1[890]=new Array("../7337/1783.pdf","High Resolution TEM Observation of Nanocrystalline Silicon Fabricated by High Pressure Torsion (HPT)","","1783 doi:10.1017/S1431927615009691 Paper No. 0890 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 High Resolution TEM Observation of Nanocrystalline Silicon Fabricated by High Pressure Torsion (HPT) Yuta Fukushima1,2, Kaveh Edalati1,3, Yoshifumi Ikoma1, Zenji Horita1,3, and David J. Smith2 Department of Materials Science & Engineering, Kyushu University, Fukuoka 819-0395, Japan Department of Physics, Arizona State University, Tempe, Arizona 85287, USA 3. WPI, International Institute for Carbon-Neutral Energy Research (WPI-I2CNER), Kyushu University, Fukuoka 819-0395, Japan 2. 1. Silicon has 12 different crystal structures under different pressure [1]. Si has the cubic diamond (cd) structure (Si-I) under pressure up to ~11.7 GPa. Si-I transforms to the metallic -tin structure (Si-II) above ~11.7 GPa. Si-II transforms to the R8 structure (Si-XII) at ~8 GPa to Si-I when pressure is released slowly. Si-XII may transform subsequently to the BC8 structure (Si-III) upon releasing the pressure. Si-XII and Si-III transform to the hexagonal diamond (hd) phase (Si-IV) by subsequent annealing at moderate temperature [1,2]. Si-IV also can transforms to Si-I by prolonged annealing. In this study, the effect of severe plastic deformation (SPD) on pressure-induced phase transformations is investigated using the high-pressure torsion (HPT) method, which is originally used to improve the mechanical properties of alloys by producing ultrafine-grained (UFG) structures [3]. The HPT facility used in this study consisted of two anvils with a hole (5 mm diameter and 0.25 mm thickness) on the surface of each anvil. Two Si (100) wafers were placed on the hole of the lower anvil, and then slowly compressed under a pressure of 24 GPa. After compression, the lower anvil was rotated with respect to the upper anvil for 20 turns using a rotation speed of 1 rpm at room temperature. After HPT processing, some samples were annealed at 873 K for 1, 2, 6, 12 hours under N2 flow (0.1 L/m). For transmission electron microscopy (TEM), samples were prepared by crushing in isopropanol and deposition onto holey carbon films. Nanostructures were examined by phase-contrast high-resolution TEM using a JEOL JEM-4000EX operated at 400 keV and a FEI-Philips CM200-FEG operated at 200 keV. High-resolution images were analyzed using Gatan DigitalMicrograph software. Examination of TEM images indicates that multiple metastable high-pressure phases are formed by HPT processing. Figure 1(a) shows the presence of Si-XII phase with orientations of [112] and [0-1-1]. Other diffraction spots in the diffractogram of Fig. 1(a) correspond to Si-I phase. Figure 1(b) suggests the coexistence of Si-I and Si-IV with orientations of [110] and [100], respectively. The presence of many spots in the diffractogram of Fig. 1(a) and (b) indicates the existence of small nanograins with sizes in the nanometer scale. The grain boundaries are shown more clearly in Fig. 1(b). After annealing, Si-I is the only detected phase, consistent with an earlier report [4]. The grain sizes remain in the nanometer range after annealing and many dislocations, as shown in Fig. 2(a), and many nanotwins with a typical lamella width in the range of 2-3 {111} atomic fringes, as shown in Fig. 2(b), are formed. These lattice defects also form after HPT processing, but their fraction increases after annealing [5]. References: [1] B. D. Malone, J. D. Sau, and M. L. Cohen, Phys. Rev. B78, 035210 (2008). [2] M. Cohen and B. Malone, J. Appl. Phys. 109, 102402 (2011). [3] R. Z. Valiev, A.V. Korznikov, R. R. Mulyukov, Mater. Sci. Eng. A168, 99 (1993). [4] Y. Ikoma, et al, J. Mater. Sci. 49, 19 (2014). Microsc. Microanal. 21 (Suppl 3), 2015 1784 [5] This work was supported in part by a Grant-in-Aid for Scientific Research from the MEXT Japan, in Innovative Areas "Bulk Nanostructured Metals" (Nos. 22102004, 25102708). The authors gratefully acknowledge use of facilities in the John M. Cowley Center for High Resolution Microscopy at Arizona State University. (a) (b) Figure 1. TEM lattice images (left), and corresponding diffractograms (right), for Si after HPT processing. Formation of metastable Si-XII and Si-IV are shown in (a) and (b), respectively. (a) (b) (c) Figure 2. (a) Lattice image of {111} dislocation and (b) corresponding inverse FFT constructed using [111] diffraction spots. (c) Lattice image of Si nanotwins present after annealing.");sQ1[891]=new Array("../7337/1785.pdf","ENERGY-DISPERSIVE X-RAY SPECTRUM SIMULATION AND EMPRICAL OBSERVATION OF 22NM NODE HIGH-K METAL GATE STRUCTURE","","1785 doi:10.1017/S1431927615009708 Paper No. 0891 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 ENERGY-DISPERSIVE X-RAY SPECTRUM SIMULATION AND EMPRICAL OBSERVATION OF 22NM NODE HIGH-K METAL GATE STRUCTURE Imen Rezadad1, 2, Brenda Prenitzer2, Stephen Schwarz2, Brian Kempshall2, Robert Peale1 1. Physics Department, University of Central Florida, Orlando, Florida, USA. 2. NanoSpective, Inc., Orlando, Florida, USA. Characterization of advanced technology node integrated circuits is an ever-increasing challenge. In order for analytical capabilities to keep pace with technology it is important to carefully consider all aspects in the design and execution of experiments. It is essential to be able to distinguish artifact from meaningful data and to recognize the limitations as well as the capabilities of a given technique. Some critical considerations include specimen preparation, physical limitations of the instrumentation, alignment and calibration as well as user selected acquisition parameters. The effects of sampling size, signal to noise ratio, specimen geometry, instrumentation and accelerating voltage on the microanalysis of multilayered metal gate structures created using atomic layer deposition (ALD) are presented. Empirical results are compared with modeled data for energy dispersive spectroscopy (EDS) line profiles through two different ALD thin film stacks routinely used in high-k metal gates (HKMG) at the 22nm node. The question of whether and under what conditions do EDS line profiles have adequate lateral spatial resolution to sufficiently differentiate diffusion at monolayer length scales is investigated. Analytical calculations using the Jones matrix formalism showing the degree of beam spreading will be presented for two different TEM configurations. [1] Electron trajectories through the specimen thickness are also simulated using Casino.[2] EDS simulations are done using MC X-Ray [3] to represent the compositional analysis of two different thin film stacks containing slices of Si, TiN, HfO2 and SiO2 and TiAl, TiNTa, Ta, TiN, HfO2, SiO2 and Si. Simulation results are correlated with empirical measurements performed on gates extracted from two different devices. Spectra from analyses performed on an FEI Tecnai F30 TEM with an EDAX Sapphire Si(Li) EDS detector will be contrasted with spectra obtained on a probe corrected FEI Titan 80-300 using a Bruker SuperX4 silicon drift detector (SDD) EDS system. Figure 1 shows an example of an analysis performed on the Titan. The simulations are performed on atomically smooth interfaces to illustrate the profound effect that user defined parameters can have on data acquisition and interpretation. Figure 2 shows the effect that undersampling can have on the shape of a line profile. A region may be undersampled if the step size is too large with respect to the feature size. A similar loss in lateral spatial resolution will be observed even if the step size is small enough but the probe diameter is too large with respect to the feature size. Figure 4 shows the effect that probe position can have on the shape of a line profile. The apparent location of the interface can be displaced by a distance as large as the step size. Similar to undersampling, the effects of a misplaced sampling probe can be minimized by selecting the appropriate step size with respect to the feature of interest and the resolution requirements for your analysis. References [1] David B. Williams, "Transmission Electron Microscopy", (Springer, New York), P. 665. [2] D. Drouin et al., Scanning 29 (2007), 92-101. [3] http://montecarlomodeling.mcgill.ca/ Microsc. Microanal. 21 (Suppl 3), 2015 1786 Fig. 1. STEM image and EDS line profile acquired with adequate sampling through the HKMG stack of an NMOS transistor in an advanced technology node device Fig. 2. The effect of undersampling on profile shape. The skewed shape erroneously resembles a diffusion profile. Fig. 3. Simulated EDS spectrum for a structure and sampling corresponding to that shown in figure 1. Compared to the empirical spectrum, the interfaces show steeper slopes suggesting that the profiles observed during the experiment are attributable to actual diffusion. Fig. 4. The effect of probe position with respect to interface. The apparent position of the interface can be displaced by a distance as large as the step size.");sQ1[892]=new Array("../7337/1787.pdf","Na-induced Structural Unit in 3 [-110]/(-1-11) Grain Boundaries of Si","","1787 doi:10.1017/S143192761500971X Paper No. 0892 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Na-induced Structural Unit in 3 [-110]/(-1-11) Grain Boundaries of Si M. Shamsuzzoha School of Mines and Energy Development, University of Alabama, Tuscaloosa, AL 35487, USA The incorporation of impurity atoms at grain boundaries has been suggested to produce changing of interface bonds that result in reduction of material cohesion [1]. Structural studies using HREM taken from a =5[001]/(310) grain boundary of aluminium present in an Al-5% Mg alloys revealed that the incorporation of Mg into the grain boundary alters the structure in some of the basic kite-like structural units of interface and induces such a change in interface bonding [2]. The basic kite-like structural unit of this interface contains seven Al atoms in its structure with one Al atom at the center of the unit. The Mg-incorporated structural unit at the interface contains eight atoms and has two atomic sites made of a coplanar Mg-Mg atom pair at the central atomic site, allowing the unit to be comprised of Al and Mg atoms. Elsewhere [3], calcium segregation at a MgO grain boundary as determined by a combination of atomic-resolution Z-contrast imaging, first-principles densityfunctional calculations and EELS studies yielded a similar type impurity-induced structural transformation of the interface structure. This paper presents an HREM study that reveals the incorporation of sodium atoms at some 3 [-110]/(-1-11) boundaries of Si in a Na-doped Al-12 wt.% Si cast alloy results in structural transformation for the interface structure. Figure 1a shows a HREM image containing a number of 3 [-110]/(-1-11) grain boundaries in a silicon crystal taken from a sodium-doped Al-12 wt.% Si cast alloy. Visual inspection of the image reveals the presence of defect-free (A) and impurity-affected (B) boundaries. Auer electron spectroscopy of the alloy reveals that the Na is mainly adsorbed on (111) surfaces of such twin boundaries Figure 1b is a magnified image of a segment of a defect-free boundary located at A of Figure 1a. The schematic of the orientation relationships of twin projected on the (-110) as drawn from the HREM image of Figure1b is shown as an inset in the image. It should be noted that limited resolution of the 400 keV microscope prohibited the resolution of Si dumb-bells, and the [-110] projected black atomic columns in the images were considered of a FCC crystal. The structure of this segment of the boundary can be described by periodic occurrences of a kite-like structural unit (x) shown as abcd in the inset of Figure 1b containing five atomic columns. The structural unit, also is drawn on the image of Figure 1b, contains only one atomic column at its center and has the width of the CSL. This structural unit is generally considered as correct structural unit of a 3 [-110]/(-1-11) boundary of Si. Figure 1c also shows a high magnification view of a segment of the impurity-affected boundary located at B of Figure 1a.The orientation relationships of twin as drawn from the corresponding HREM image is schematically shown in the inset. The boundary in this segment exhibits a periodic arrangement of another kite-like structural, unit (y) shown as efgh in the inset of Figure 1c, but contains six atomic columns. This kite-like unit shows two atomic columns along the direction of electron beam at the center and has the same type of atomic arrangement described for the structural unit x. The two central atomic columns of this structural unit (also drawn on the image of Figure 1c) lay along the length of the twin boundary. Simple comparison of the HREM images of these two kite-like structural units suggests that the center of the y structural unit does not provide sufficient space to occupy two atomic columns of Si dumb-bells. Also based upon similarity in the size of the dark atomic columns that exist at the center of the structural unit y, it seems unlikely that one of the atomic columns belongs to Si and other to Na. It seems to suggest that both atomic columns existing at the center of this unit belong to Na atoms. A literature review revealed that the bonding energy of Na-Na and Si-Si atom pair is 73.60(20) and 326.6(10.0) kJ mole-1 respectively [4]. This implies that the structural unit containing lower bonding energy Na atomic pair is expected to experience weaker interaction than that which contains Si-Si atom pair and suggests that the destabilization brought about by the Na-Na pair is weak enough to allow the structural unit y to remain stable at the boundary as observed presently. It is believed that during solidification of the Na doped Al-12 wt.% Si alloys Na atoms of the melt poison the Si growth by incorporating itself into twinned growth tip and retained into the twin boundary as members of an ordered structural unit. References: [1] P. Lejcek, Grain Boundary Segregation in Metals; Springer: Heidelberg, Germany, 2010. [2] M Shamsuzzoha, I Vazquez, P. A. Deymier, David. J. Smith, Interface Science.3 (1996), 227. Microsc. Microanal. 21 (Suppl 3), 2015 1788 [3] Y. Yan, M. F. Chisholm, G. Duscher, A. Maiti, S. J. Pennycook, and S. T. Pantelides, Phys. Review Letters, 81(17), (1998), 3675. [4] R. L David, CRC Hand Book of Chemistry and Physics, (CRC Press Inc 1993-1994), pp. 9-123,128. (a) (b) (c) Figure 1: (a) HREM image taken with electron beam parallel to [-110] of a silicon crystal taken from a Na doped Al-12wt.% Si alloy. It exhibits a number of 3 boundaries. (b) Magnified image of the location A in (a). A CSL and two structural units (x) are outlined at the twin boundary. (c) Magnified image of the location B in (a). A CSL and two structural units (y) are outlined at the twin boundary.");sQ1[893]=new Array("../7337/1789.pdf","Towards a Comprehensive TEM Toolbox for Complex Molecular Fluids","","1789 doi:10.1017/S1431927615009721 Paper No. 0893 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Towards a Comprehensive TEM Toolbox for Complex Molecular Fluids Min Gao Liquid Crystal Institute, Kent State University, Kent, Ohio, USA. In this paper, we summarize our ongoing efforts to set up a comprehensive yet easily accessible toolbox for TEM observation of complex molecular fluids (CMFs) and discuss the future possibilities that TEM community can develop for these fascinating but challenging materials. Many simple molecular fluids can be described using hard-sphere models, and have direct phase transitions from isotropic liquid phases to crystalline phases. Some nonspherical molecules can form complex structures, for example, micelles, or more interestingly, intermediate phases (the so-called liquid crystals, LCs) with orientational order only (nematic mesophases) or both orientational and 1D (or 2D in some cases) positional orders (smectic mesophases). The CMFs have tremendous impacts (e.g., liquid crystal display or LCD, detergents, and crude oil) on our lives. However, the understanding of their detailed structures has been surprisingly limited due to their complicated structure and the lack of effective structural probe at subnanometer scale. For example, though initial direct TEM observation on the much studied CMF group � liquid crystals demonstrated promising results in late 1980's, the TEM studies were soon dominated by a replica TEM technique, namely freeze-fracture TEM (FFTEM). FFTEM effectively avoids two of the challenges for direct TEM observation on CMFs: i) difficulty in the preparation of thin TEM specimens to preserve the native structure partly due to the strong surface anchoring effect; ii) severe radiation damage. However, the relatively low resolution (a few nanometers) of FFTEM has hindered the understanding of CMF structures at the molecular scale. In this study, a series of specimen preparation routines have been established for CMFs (Fig. 1), which have been applied to a variety of LC materials, including thermotropic (consisting of single- or multiplecomponent complex molecules) and lyotropic (complex molecules dissolved in certain solvents, for example, water) LCs [1]. Some new concepts include a simple procedure to produce suspended thermotropic LC thin films, combination of high-pressure freezing and cryo-ultramicrotomy for lyotropic LCs, and complementary applications of the direct cryo-TEM and FFTEM. We demonstrate that a widely available cryo-TEM, a high sensitivity CCD camera and a modified low-dose imaging procedure have made a ready combination for subnanometer resolution imaging of many challenging CMFs. We also explored the employment of diffraction, STEM Z-contrast, EDS, EELS and in situ observation of dynamic process. The significance of subnanometer resolution cryo-TEM observation is demonstrated in a few important issues in LC studies, including: i) revealing the existence of nanoscale layered smectic domains in nematic bent-core thermotropic LCs (Fig. 2) [1,2]; ii) providing evidence for the existence of twist-bend nematic phase with a complex director structure that follows the geometry of an oblique helicoid with a nanoscale pitch [1,3]; iii) probing the nematic packing of columnar aggregates in lyotropic chromonic liquid crystals (Fig. 2) [1,4]. More importantly, direct TEM observation opens ways to a variety of TEM techniques, suggesting that TEM (replica, cryo, and in situ techniques), in general, may be a promising part of the solution to the lack of effective structural probe at the molecular scale in CMF studies [5]. References: Microsc. Microanal. 21 (Suppl 3), 2015 1790 [1] M Gao et al, Microsc Res Tech 77 (2014), 754. [2] C Zhang et al, Phys Rev Lett 109 (2012), 107802. [3] V Borshch et al, Nature Communications 4 (2013), 2635. [4] D Pucci et al, J Mater Chem C 2 (2014), 8780. [5] The TEM-related experiments were carried out at the cryo-TEM lab of the Liquid Crystal Institute, Kent State University. The author thanks Dr. Oleg D. Lavrentovich, Dr. Georg H. Mehl, and Dr. Wolfgang Weissflog for providing the samples. Figure 1. TEM specimen preparation routines for thermotropic liquid crystals (TLC) (a-c) and lyotropic liquid crystals (LLC) (d-f). (a) Plunge freezing of supported and suspended TLC thin films for cryoTEM. (b) Cryo-sectioning of plunge frozen "bulk" TLC for cryo-TEM. (c) Freeze fracture of plunge frozen "bulk" TLC for FFTEM. (d) Plunge freezing of LLC thin film for cryo-TEM. (e) Vitreous sectioning of high pressure frozen LLC for cryo-TEM. (f) Freeze fracture of frozen (plunge freezing or high pressure freezing) LLC for replica TEM. The scale bars in the schematics roughly demonstrate the feature sizes. Figure 2. (a) Cryo-TEM image and corresponding FFT pattern showing layered smectic nanoclusters in a 3-ringe bent-core nematic liquid crystal (3RBC-N, the molecular structure is shown as an inset). The narrow solid arrows point out a few overlapping clusters, while the hollow arrows denote assemblies of side-by-side clusters. The lower insets are the magnified views of the 30 nm � 30 nm junction areas marked by squares. An extra half layer is pointed out by a solid arrow. (b) � (d) Magnified images of representative complex assemblies of the layered smectic nanoclusters in 3RBC-N. The junctions are marked by hollow arrows, and the extra half layers by solid arrows. (e) � (f) Cryo-TEM images of the elongated aggregates in a silver(I) complex (20%wt in water) with chromonic lyotropic LC behavior: side-view (e) and top-view (f). The insets in (e) are the molecule structure and the corresponding FFT. The inset in (f) shows a magnified image of the marked area (30 nm � 30 nm) with a dashed square.");sQ1[894]=new Array("../7337/1791.pdf","In Situ LC-TEM Studies of Corrosion of Metal Thin Films in Aqueous Solutions","","1791 doi:10.1017/S1431927615009733 Paper No. 0894 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In Situ LC-TEM Studies of Corrosion of Metal Thin Films in Aqueous Solutions Jeung Hun Park1,2, See Wee Chee3, Suneel Kodambaka1, and Frances M. Ross2 1 Department of Materials Science and Engineering, University of California Los Angeles, 410 Westwood Plaza, Los Angeles, CA 90095, USA 2 IBM T. J. Watson Research Center, 1101 Kitchawan Road, Yorktown Heights, NY 10598, USA 3 Centre of Bioimaging Sciences, Department of Biological Sciences, National University of Singapore, 14 Science Drive 4, Singapore 117557 Chemical reactions with the environment result in corrosion or degradation of the properties of materials [1]. Metals in contact with aqueous solutions, depending on their surface morphology, undergo different modes of corrosion [2]: general, pitting, crevice, inter-granular, galvanic, and erosion corrosion, environmentally induced fracture, and dealloying. In spite of its importance, many aspects of localized corrosion of metals are not well understood [3], probably because it is often difficult to monitor the corrosion processes at high spatial resolution under a liquid layer. Liquid cell transmission electron microscopy (LC-TEM), coupled with electrochemical control and analytic functions, enables the observation of structural changes and chemical processes in liquid phases. It provides a combination of temporal and spatial resolution that is difficult to achieve using other characterization techniques [4], and has yielded detailed information on corrosion kinetics in liquid solutions [3,4]. In this work, we studied corrosion of metal films in aqueous solutions using in situ LC-TEM. We examined general, pitting, and galvanic corrosion in several metals (Au, Al, Ni, Cu, and Zn) under different aqueous solutions. The samples are 20 - 50 nm thick metal films deposited by electron beam evaporation over a 50 nm thick silicon nitride membrane of liquid cell window chips. The experiments were carried out in a FEI CM30 TEM, operated at 300 kV, using a continuous flow Hummingbird Scientific LC-TEM holder. Deionized water was first introduced into the liquid cell to completely remove air bubbles inside the cell. Then the liquid, either deionized water or a solution containing HCl, H2SO4, CuSO4, PbSO4, NaCl, ZnCl2, and ZnSO4, was introduced. We used low dose imaging protocol and verified that these solutions can be examined for long periods (many hours) without visible beam effects in the liquid. Figure 1 presents corrosion morphologies of a Ni film in deionized water (pH = 7). This corrosion process took place under a constant electron beam illumination and a flow rate of 5 �l/min. Bright field TEM images showed that dissolution of Ni films occurs via the general corrosion mode in which chemical or electrochemical reactions proceed over the entire exposed surface at approximately the same rate of dissolution. The electron beam accelerates the process, presumably via reactive species created by beam-induced radiolysis of water [5,6]. Figure 2 shows corrosion morphologies of an Au film in 0.1M HCl (pH = 1.26). This corrosion took place during linear sweep voltammetry. Image analysis showed that the corrosion of Au films occurs via dissolution of individual grains. We will describe how measurements of the local dissolution kinetics - morphology, composition, and the etching rate of these metal films as a function of solution composition, pH, and metal thickness provide insights into corrosion processes with good spatial and temporal resolutions. References: [1] B. A. Shaw and R. G. Kelly, The Electrochemical Society Interface Spring (2006) 24. [2] H.-H. Strehblow et al. in "Corrosion Mechanisms in Theory and Practice (3rd Ed., 2012)", ed. P. Marcus, (CRC Press , New York) 1 - 572. Microsc. Microanal. 21 (Suppl 3), 2015 1792 [3] S. W. Chee et al., Chemical Communications 51 (2015) 168. [4] N. de Jonge and F. M. Ross, Nature Nanotechnology 6 (2011) 695. [5] J. H. Park et al., Microscopy and Microanalysis 19 Suppl. 2 (2012) 1098. [6] J. H. Park et al., Microscopy and Microanalysis 18 Suppl. 2 (2013) 486. [7] We gratefully acknowledge funding from the National Science Foundation (NSF-GOALI: DMR1310639), and technical assistance of M. C. Reuter and A. W. Ellis. Figure 1. Dissolution of a Ni film: Images extracted from a bright field TEM movie recorded while flowing deionized water at a rate of 5 �l/min. Scale bar is 500 nm. Figure 2. Dissolution of an Au film on silicon nitride under applied voltage: (Left) Images extracted from a bright field TEM movie obtained during electrochemical etching of Au film in 0.1M HCl solution. In the final two images, beam-induced redeposition of Au is visible as darker features in bare regions of the silicon nitride window. Scale bar is 200 nm. (Right) Linear sweep voltammetry data are recorded simultaneously. The voltage sweeps from +0.35 V to -0.8 V at 1 mV/s. Current recorded over the whole electrode is also shown. The top graph shows the average image brightness, indicating the approximate Au thickness and onset of dissolution in the field of view.");sQ1[895]=new Array("../7337/1793.pdf","Multi-modal Microscopy of Almost Nothing � Aerogels","","1793 doi:10.1017/S1431927615009745 Paper No. 0895 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Multi-modal Microscopy of Almost Nothing � Aerogels Beer G.A.1, Olin A.1, Humphrey E.C.2 1 Department of Physics and Astronomy/TRIUMF, University of Victoria, Victoria, Canada 2 Advanced Microscopy Laboratory, University of Victoria, Victoria, Canada Aerogels, consisting mostly of air caught in a matrix of low-density strands of Silicon Dioxide, were first made by supercritical drying of gels in the 1930s. When viewed on a white background they almost disappear, and on a black background they appear blue due to Rayleigh scattering. There is essentially no boundary scattering of light off the solid surface. While excellent thermal insulators, they have a multitude of other uses. For purposes outlined below, we required macro, micro, and nano 2D and 3D images of aerogel sheets of densities >27 mg/cc (20X air at STP) with modified surfaces. Critical to a proposal to measure the anomalous magnetic moment of the muon, (a "heavy electron" which is produced at accelerator laboratories), our group had need of a solid, windowless muonium production target [1] similar to powder. A solid aerogel layer should efficiently slow positive muons within the layer where they then capture electrons to form neutral muonium atoms which thermalize and diffuse out of the solid surface into vacuum. A range of densities of aerogel sheets was manufactured in Japan by our collaborators [2] and early experimental measurements at TRIUMF of the muonium yield in vacuum from these solid sheets showed that significantly less muonium escaped from the solid aerogel surface than was observed from silicon dioxide (Cabosil) powders [3]. Simulations of muonium diffusing from aerogels suggested that modification of the aerogel surface at a scale of hundreds of micrometers might improve the yield. Several surface modification techniques were tried and we settled on using a pulsed laser to "drill" 160 micrometer diameter holes into, but not through, the aerogel surface in a grid pattern. Weight measurements suggested that little of the solid material was lost when the holes were drilled. To characterize the microstructure of the aerogel surfaces, a Hitachi S-4800 SEM was used with varying degrees of success. Problems encountered and overcome resulted from the tenuous nature of the structure of the aerogel samples. First, our Japanese colleagues and then we prepared early samples with several-nanometer-layers of osmium or carbon to reduce charging with the result that the apparent size of the Silicon Dioxide particles was about 20-40 nm, in disagreement with expectations and with our TEM images of 5-6 nm. Imaging uncoated surfaces rectified this error in image size. However charging of thicker SEM samples caused the surface to go in and out of focus as the target "breathed" due to electrostatic repulsion. This, and obvious charging of the target surface, was reduced or eliminated by choosing thinner samples and a very low electron beam energy of between 500 and 700 eV along with an adjusted condenser setting. Bonding of the ~10 mg samples to stubs with colloidal carbon was facilitated using vacuum tweezers and a grounded mat with connecting wrist straps to eliminate electrostatic interactions while handling the aerogel near metal surfaces. Optical imaging of patterns of 160 um diameter holes in 30 and 50 mg/cc aerogel sheets 5 to 7 mm thick (with optical refractive indices of the order of 1.01) show the cylindrical aerogel surface created by the laser interaction. (This laser interaction creates the optical boundary that enables optical imaging). In 2D, it was difficult to determine that they were holes and not pillars pointing out from the surface Fig. 1. Using the Helicon Focus image processing program on a stack of ~40 optical images taken with an Microsc. Microanal. 21 (Suppl 3), 2015 1794 Olympus SZX 16 microscope with a DP71 camera, we obtained in-focus images of arrays of the several-mm-deep holes and created a rotating image built from the image stack to see the 3D effect. To clarify, a calibrated rotating target holder was built, with the possibility to light the sample at a constant angle from above or below, and in-focus images were created at a series of angles from flat to ~ 25 degrees on a black background. These were used to create stereo and rocking movie images. Images in 2D and 3D of a sample sliced along the axis of the holes showed further details of the hole shape. These optical images were supplemented by auto-fluorescence Confocal images taken with an Olympus BX61 Wi (with Fluoview). Images taken looking straight down the holes showed the surface clearly but, when computer-rotated through angles up to 90 degrees to show hole shape, the resultant image of the deeper region did not agree with images taken with the target placed at 90 degrees Fig. 2. To obtain comparative micro surface details, a sheet of 30 mg/cc-drilled aerogel was sliced parallel to the axis of the hole and 3D SEM images were taken of the cylindrical hole surface, the un-drilled aerogel between the holes, and the laser-interaction region Fig. 3. Clear differences are seen in the surface appearance and roughness of these three regions. In conclusion, optical, confocal, HR-TEM and uncoated SEM surface macro, micro and nano images have been obtained of aerogels, mostly in 3D, to study laser-drilled holes in their surfaces. These images can be used both to understand the effect of surface structure on muonium yields into vacuum and to provide quality control of the fabricated aerogel targets. [1] Beer G. et al, Prog. Theor. Exp. Phys. 091C01 (2014) 7 pages [2] Tabata M et al, Nucl. Inst. and Meth. A 668 (2012) 64-70 [3] Janissen A Phys. Rev. A 42 (1990) 161-169 Fig. 1. Optical surface Fig. 2. Confocal slice Fig 3. SEM 3D images of hole surface and surrounding aerogel");sQ1[896]=new Array("../7337/1795.pdf","Studying Mass Transfer in Glasses using Cathodoluminescence.","","1795 doi:10.1017/S1431927615009757 Paper No. 0896 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Studying Mass Transfer in Glasses using Cathodoluminescence. Peter McSwiggen1 and Colin M. MacRae2 1 2 McSwiggen & Associates / JEOL USA, Inc., St. Anthony, MN, U.S.A. Microbeam Laboratory, CSIRO Minerals, Clayton, Victoria 3168, Australia Optical fibers are formed through a complex manufacturing process. Many types of optical fibers are manufactured. The details depend on their designed task. Some are designed for simple telecommunication functions, others act as amplifiers, and some are designed to maintain the polarization of the light as it passes along its length. There are some common features, though. Optical fibers are designed such that light photons traveling down the core of the fiber will reflection off the interface between the core and cladding, keeping the light contained within the core. The core and cladding are glasses of two different compositions, with the core having the higher refractive index. This difference in refractive index causes the total reflection at the interface. Surrounding the working portion of the fiber, the remainder of the fiber is typically pure SiO2 glass. This material started out as a hollow glass tube, during the initial stage of manufacturing. Using a chemical vapor deposition process, the working components of the desired fiber are deposited sequentially onto the inner surface of the hollow tube. This is done by passing the desired chemicals, in a vapor form, though the tube. While the tube is slowly rotated horizontally on a lathe, the external surface is heated to 1600�C (3000�F). The chemicals react with the oxygen in the tube, and are deposited as sooty, oxide particles. Once the required layers have been deposited, the hollow tube is collapsed into a solid glass rod, called the preform. This process is accomplished by concentrating the heat, for a longer period, on the external surface of the tube, while it continues to rotate. The tube slowly redistributed its mass, while its diameter decreases and the thickness of the walls increase. Eventually it becomes one solid glass rod. Intuitively one might think that this process of redistributing the mass within the walls of the glass tube would be a chaotic, disorganized process. However, observations show that the exact opposite is true. Cathodoluminescence (CL) imaging of the outer glass shows that it has separated into layers, and each layer has segregated into elliptical, concentric zones (Fig. 1). The number of concentric zones is always six per layer. The dimensions of these concentric zones vary dramatically, but there are always six elliptical rings per layer. This zoning can only be seen in the CL signal. Conventional, X-ray elemental mapping shows no variation in the composition of the glass, since it was made from high purity SiO2. However, the variation in the CL intensity might be the result of trace compositional differences that are in concentrations below the detection limits of electron probe microanalyzers (EPMA). Full spectrum mapping of the preforms show that each of the different components of the glass rod has a very different CL spectrum (Fig. 2). The concentric, elliptical zones must represent the method by which the mass of the glass is transferred from the outer regions of the glass tubing inwards as the diameter collapses. It is not known whether outward layering previously existed in the original glass tubes, or whether they also developed during the collapse of the tube. In either case, CL imaging gives dramatic insights into the mechanism by which the mass of the tubing redistributed itself while it is compressed into a glass rod. Samplings and CL imaging at various stages of this transition could explain in better detail this mass transfer mechanism. Microsc. Microanal. 21 (Suppl 3), 2015 1796");sQ1[897]=new Array("../7337/1797.pdf","Structure and electrical property changes of ZnO:Al films, prepared by radio frequency magnetron sputtering, by thermal annealing","","1797 doi:10.1017/S1431927615009769 Paper No. 0897 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Structure and electrical property changes of ZnO:Al films, prepared by radio frequency magnetron sputtering, by thermal annealing Yi-Yan Chen1, Po-Yu Chen1, Shao-Liang Cheng2, Bo-Ming Huang1, Jer-Ren Yang1, Masahiro Kawasaki3, and Makoto Shiojiri4* Department of Materials Science and Engineering, National Taiwan University, Taipei, Taiwan, ROC. Department of Chemical and Materials Science Engineering, National Central University, Taoyuan, Taiwan, ROC. 3 JEOL USA Inc., 11 Dearborn Road, Peabody, MA 01960, USA. 4. Kyoto Institute of Technology, Kyoto 606-8585, Japan. * Present address: 1-297 Wakiyama, Kyoto 618-0091, Japan. 2 1 Among transparent conducting oxide films that are widely applied in optical and electronic devices such as solar cells, thin film transistors and light emitting diodes, ZnO:Al has attracted much attention because it is an abundant, inexpensive, non-toxic and environmentally friendly raw material with high crystallinity and good conductivity, and can be easily prepared. Magnetron sputtering is one of the preferred techniques to prepare the ZnO:Al films because it produces films with a composition close to that of the source material. Here, we report changes of structure and electrical properties by thermal annealing of ZnO:Al films sputtering-deposited on glass substrates ZnO:Al thin films with a thickness of 650 nm were deposited on Corning Eagle2000 glass substrates at a deposition rate of about 2.2 nm/min from a ZnO/2 wt.%Al2O3 target. After deposition, the ZnO:Al films were annealed at 300, 400, and 500oC for 1 h in vacuum (~0.133 Pa) in a quartz-tube furnace. Optical transmittance was measured with a spectrophotometer. Electric resistivity was measured using the four-point probe method. X-ray diffraction (XRD) and electron microscopy observation were performed. Figure 1 shows a cross-section TEM image and an ED pattern of the as-deposited film. It consisted of columnar grains perpendicular to the glass substrate surface, with the highly preferred orientation along the [0001] axis. Figure 2a reveals that the average optical transmittance of the as-deposited and post-annealed ZnO:Al films were all over 80% in a 400-800 nm wavelength range. The oscillation can be ascribed to the multiple beam interference between the top and bottom surfaces of the film, which causes the minimum transmittance through reflection of the incident light beam. So, the difference in minimum position between these curves corresponds to a small difference in thickness of the film, which was caused by the heat treatment. Plots of (h)2 versus h in Figure 2b indicate that the optical band-gap energy estimated from the intercept was significantly higher for the annealed films than the as-deposited ZnO:Al film. This blue shift of the band-gap can be attributed to the Burstein-Moss effect. Figure 3 shows the electrical resistivity as well as the optical transmittance over the 400-800 nm wavelength range and the estimated optical band-gap energy for these films. Due to the activity of the doped-Al elements and oxygen vacancies, the resistivity of the annealed ZnO:Al films significantly decreased, and the lowest resistivity occurred with annealing at 400oC. X-ray diffraction patterns in Fig. 4a indicate the preferential (0001) orientation of the ZnO:Al film. Since the solid solubility limit of Al in ZnO:Al is 2 mol% [1] and no peaks from other crystals such as ZnAl2O4 appeared, the composition of the present films is probably approximately ZnO/1.6 mol.% Al2O3. Details of the 0002 peaks in Fig. 4b indicate that the (0001) lattice spacing decreased with annealing temperature until 400oC as shown in Fig. 4c. This observed shrinking of the unit cell is ascribed to the substitution of Al3+ (radius 76.5 pm) for Zn2+ (radius 88 pm) and the formation of oxygen vacancies during the activation annealing. HRTEM Microsc. Microanal. 21 (Suppl 3), 2015 1798 images in Figure 5 reveal that the angle between the [000 1 ] and [01 1 0] axes or the lean of the c axis increased with annealing temperature and reached a maximum of about 4o (=93.7o-90o) at an annealing temperature of 400oC (Fig. 4c). The hexagonal cell was hence deformed to a triclinic cell, leaning the c axis and shrinking the volume, due to the formation of point defects during the annealing. At an annealing temperature of 500oC, the lean of the c axis was smaller than that at 400oC. This is closely related with optical band-gap energy and electrical resistivity. Since the optical and electrical properties are greatly influenced by point defects, this could be ascribed to the difference of the density of point defects that are the substituted Al3+ ions and O2- vacancies. At 400oC, the point defects might be saturated in the ZnO:Al film in the most favorable state. One possible explanation for the higher resistivity and lower optical band-gap energy caused by annealing at 500oC may be introduced oxygen atoms from the ambient atmosphere at elevated temperature, whereby the oxygen atoms occupy the vacancy sites without appreciably changing the unit cell volume and reducing the lean of the c axis. [1] M.H. Yoon et al, J. Mat. Sci. Lett. 21, 1073 (2002). Fig. 1 (top left). TEM and ED pattern of as-deposited ZnO:Al film. Fig. 2 (top right). (a) Optical transmittance spectra. (b) (h)2/h relation. is the optical absorption coefficient. Fig. 3 (middle left). Optical transmittance and band-gap energy, and electrical resistivity. Fig. 4 (middle right). (a) X�ray diffraction patterns of these films. (b) the details of the 0001 peaks. (c) The (0001) spacing and the lean of the c axis. Fig. 5 (bottom). HR-TEM lattice images and the corresponding FFT diagrams, showing the angle between the [000 2 ] and [01 1 0] axes or the lean of the c axis.");sQ1[898]=new Array("../7337/1799.pdf","Aluminum and Tantalum Doping of Sputtered TiO2 Thin Films","","1799 doi:10.1017/S1431927615009770 Paper No. 0898 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Aluminum and Tantalum Doping of Sputtered TiO2 Thin Films Inga Goetz1, Rong Liu2, Richard Wuhrer3 and Leigh Sheppard4 1 2 Flensburg University of Applied Sciences, Germany Advanced Materials Characterisation Facility, University of Western Sydney, Australia 3 SIMS Facility, University of Western Sydney, Australia 4 School of Computing, Engineering, and Mathematics, University of Western Sydney, Australia TiO2 is a promising material for the generation of hydrogen fuel from the photoelectrochemical splitting of water using sunlight [1]. In this application, performance is determined by the ability to generate electron-hole pairs during illumination, and keep them separated long enough to do useful work. A TiO2-based homojunction, consisting of a graded composition is considered promising for this role [2]. The use of magnetron sputtering is considered promising for the fabrication of such a material. The aim of this project is to explore the deposition of Al and Ta doped TiO2 films and identify deposition parameters that facilitate control of film crystallinity, control of doping level and control film thickness. Meeting these aims will provide the basis for fabricating TiO2-based homojunctions. All films were prepared using the UWS Reactive Magnetron Sputtering Facility. A number of deposition parameters were tested including Ar:O2 ratio, deposition pressure, source power and types (DC or RF), RF substrate bias voltage, and substrate temperature. The films were deposited onto substrates of titanium foil, silicon wafer (001), and glass, after initial evacuation to < 5 x 10-6 Torr. The films were then characterised using a Jeol 7001F Scanning Electron Microscope (SEM) with Energy Dispersive X-ray Spectroscopy (EDS) and a Bruker D8 Advance X-ray Diffraction (XRD). A Secondary Ion Mass Spectrometer (Cameca IMS 5FE7 � SIMS) was applied to determine the doping gradient of two graded films. Tantalum, in contrast to both titanium and aluminium, showed very high deposition rates. The power levels for the initial films were set at 200W DC, 100W RF, and 50W DC for Ti, Al, and Ta respectively. 10 and 1.5 mTorr were chosen as deposition pressures to reflect the poisoning regime. The initial Al-doped films exhibited a bimodal particle size distribution and increased surface roughness when deposited at 10 mTorr in comparison to the 1.5 mTorr films. Al-doping levels obtained from EDS were around 20% lower at 1.5 mTorr. The initial Ta-doped films showed even higher dopant concentrations and consisted predominantly of Tantalum. All of the initial films were found to be amorphous. To control and reduce doping levels, power levels and types were adjusted for all elements. Ta doping levels were significantly reduced, even though a range of lower amounts would be desirable. Al concentrations also decreased, still a significant uncertainty persists concerning the actual values, as even an undoped sample showed significant amounts of Al in the EDS. In comparison with the non-heated substrates, the heated samples with reduced dopant concentration exhibit a rougher surface structure and lower doping levels in the EDS analysis. Both the films grown on heated and non-heated substrates were found to consist of mixed amorphous, anatase and rutile phases. By varying the dopant power during deposition, two graded films were obtained, which exhibited the desired gradient in Ta concentration as determined by SIMS. These films demonstrate good control over thickness and composition, but further structural control is needed. Microsc. Microanal. 21 (Suppl 3), 2015 1800 Figure 1: The initial set of 8 films (P(Ti) = 200W DC, no Bias, no heating) [Al] = (Al*100%)/(Al+Ti), [Ta] = (Ta*100%)/(Ta+Ti) (Al, Ta, Ti in at.%). Figure 2: Films with lower dopant powers (P(Ti) = 2*200W DC, P(Ta) = 35W RF, P(Al) = 25W RF (30W RF for Al 2.24), Ar:O2 = 10:5). Figure 3: SIMS Results. Ta-doped graded films (P(Ta) = 35-100W RF, P(Ti) = 2*200W DC, 1.5 mTorr, Ar:O2 = 10:5, Bias: 100V RF, T = 400�C). Al and Ta doped TiO2 films have been successfully deposited using magnetron sputtering but with limited control. In particular, Ta doping at low levels (< 1 at.%) appears challenging. Al doping at low levels appears easier. Crystallinity is favoured by the application of substrate bias and heating but can yield a mix of anatase, rutile and amorphous phases. Control of film thickness was reliably obtained. Further investigations of the effect of Ar:O2 and deposition pressure are required to tune the sputter yield of dopant sources. References: [1] Sheppard, L. R.; Nowotny, J. Materials for Photoelectrochemical Energy Conversion. Adv. Appl. Ceram. 2007, 106, 9-20 [2] Sheppard, L. R.; Dittrich, T.; Nowotny, M. K.; The Impact of Niobium Surface Segregation on Charge Separation in Niobium-Doped Titanium Dioxide, J. Phys. Chem. C, 2012, 116, 2092320929");sQ1[899]=new Array("../7337/1801.pdf","Microstructure and Dislocation Analysis of L10 FePt for HAMR Media Application","","1801 doi:10.1017/S1431927615009782 Paper No. 0899 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microstructure and Dislocation Analysis of L10 FePt for HAMR Media Application Yang Yi, Hu Jiangfeng, Kelvin Cher Kiat Min, Wong Seng Kai, Hnin Yu Yu Ko, Nelson Lim Chee Beng, Serene Ng Lay Geok, and Shi Jianzhong Data Storage Institute, A*STAR (Agency for Science, Technology and Research), Singapore 117608 L10 ordered FePt thin films have very high magnetic anisotropy energy (Ku), which ensures the thermal stability as the bit size and grain size become smaller. The particular property is one of the key reasons that granular L10-FePt thin films are considered as the most promising candidate for high-density heat assisted magnetic recording (HAMR), a technology geared for future hard disk drives (HDDs). From application viewpoint, FePt based granular film with small well-isolated columnar grains is required. The microstructure control of the FePt based granular film has been reported [1-3]. In general, grain isolation was achieved by fabricating FePt based composite films with a variety of additives such as SiO2, MgO, ZrO, C, etc. In this study, we have investigated the microstructure and dislocations of L10-FePt granular film by high-resolution transmission electron microscopy (HRTEM) and correlated the results to the thermal conditions and magnetic properties. The FePt based granular films were prepared by industry compatible sputter system where the duration of the heating and deposition processes were within several seconds. Post annealing and a two-step heating processes were applied to enhance the chemical ordering of FePt, in comparison to the one-step heating process. The M-H loop measurement shows FePt with active thermal annealing has a higher in-plane and out-of-plane coercivity. The plane-view TEM images of FePt media fabricated by one-step, post annealing and two-step heating were shown in Figures 1(a-c), respectively. The microstructure of the one-step heating sample shows large grain size and lateral interconnection of the grains indicated by arrows. A simple post annealing process doesn't vary the grain size significantly by comparing Figures 1 (a) and (b). In comparison, the two-step heating sample shows much smaller grain size and better grain isolation, although there are still grains interconnected. The annealing effect is further investigated by looking at the FePt/MgO interface. Figures 2(a-c) shows the cross-sectional HRTEM analysis of the NiTa/MgO/FePt-C/FePt-SiO2 fabricated by one-step, post annealing, and two-step heating, respectively. The grain size of FePt is usually significantly larger than MgO buffer layer, as indicated by dotted line in MgO, which is usually a result of grain segregation. As can be seen in Figure 2(a), with one-step heating, the grain boundary in MgO can largely affect the crystalline quality of FePt by inducing {111} plane dislocations originated from FePt/MgO interface. After simple post annealing, FePt/MgO interface is similar to Figure 2(a), while only some dislocation can still be observed in FePt grain. On the contrary, as shown in Figure 2(c), with a two-step active heating process, MgO has been pushed upward into the spacing of FePt grains to become part of spacing oxide, and smaller grains formed with less grain segregation. It's obvious that MgO played as isolation oxide before introducing SiO2 in the two-step heating sample. The FePt grain is around or less than the size of MgO and dislocations in FePt by the two-step heating are not significant in most of small grains. Microsc. Microanal. 21 (Suppl 3), 2015 1802 Inter-layer diffusion was further studied by EDX to understand the thermal annealing effect on FePt/MgO interface. References [1] T. O. Seki, Y. K. Takahashi, K. Hono, "Microstructure and magnetic properties of FePt-SiO2 granular films with Ag addition", J. Appl. Phys., vol. 103, pp. 023910, 2008. [2] G. S�fr�n, T. Suzuki, K. Ouchi, P. B. Barna, G. Radn�czi, "Nano-structure formation of Fe-Pt perpendicular magnetic recording media co-deposited with MgO, Al2O3 and SiO2 additives", Thin solid films, vol. 496, pp. 580, 2006. [3] M. Watanabe, T. Masumoto, et al., "Microstructure and magnetic properties of FePt-Al-O granular thin films", Appl. Phys. Lett. vol. 76, pp. 3971, 2000. Figure 1 In-plane TEM images of FePt based HAMR media fabricated by (a) one-step heating, (b) post annealing, and (c) two-step heating. Figure 2. HRTEM analysis of the NiTa/MgO/TiN/FePt-C/FePt-SiO2 fabricated by (a) one-step heating, (b) post annealing, and (c) two-step heating. The MgO spacing oxide induced grain boundaries are indicated in (c).");sQ1[900]=new Array("../7337/1803.pdf","TEM Analysis of Defects in AlGaN Heterostructures Grown on c-Al2O3 by Plasma Assisted Molecular Beam Epitaxy","","1803 doi:10.1017/S1431927615009794 Paper No. 0900 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 TEM Analysis of Defects in AlGaN Heterostructures Grown on c-Al2O3 by Plasma Assisted Molecular Beam Epitaxy Sergei Rouvimov1,2, V. N. Jmerik3, D.V. Nechaev3, and S.V. Ivanov3 1. 2. Department of Electrical Engineering, University of Notre Dame, Notre Dame, Indiana 46556, USA Notre Dame Integrated Imaging Facility, University of Notre Dame, Notre Dame, Indiana 46556, USA 3. A. F. Ioffe Physical-Technical Institute, 194021, 26 Politekhnicheskaya str., St.-Petersburg, Russia This paper addresses the defect formation in AlGaN-based heterostructures with high Al-content (x>0.4) grown by plasma assisted molecular beam epitaxy (PA MBE). The growth of AlGaN epitaxial layers with reduced defect density is a challenging technological task that requires understanding of defect generation and propagation. Low defect density in AlGaN-based heterostructures is critical for fabrication of high efficiency optoelectronics devices such as UV- emitters with wavelength below 300 nm. The reduction of threading dislocation (TDs) densities allows for better control of the carrier's transport and residual stress in the optically-pumped UV-laser structures grown over standard c-sapphire substrates by using PA MBE. Here we review defect generation in AlGaN heterostructures to understand the mechanisms of their formation especially at initial growth stages. The samples were analyzed by TEM using both high resolution TEM (HRTEM) and High Angle Annual Dark Field (HAADF) scanning TEM (STEM) modes at FEI Titan 80-300 electron microscope. The microscope was operated at 300 keV and equipped with an Oxford Inca EDX detector. TEM crosssectional samples were prepared by Focus Ion Beam (FIB) using FEI Helios SEM/FIB dual beam equipment. It was demonstrated that migration-enhanced epitaxy (MEE) of the AlN buffer layer and 3D growth of ultra-thin GaN interlayers in AlN layers result in substantial reduction of dislocation density in AlGaN-based heterostructures. Figure 1 shows typical Bright Field (BF) TEM images of AlN layers demonstrating that the threading dislocations are mainly of edge type (Fig. 1a). The enlarged BF TEM image in Fig. 1c shows the formation of voids at AlN-sapphire interface. One can speculate that the lateral overgrowth of separated AlN islands may reduce strain at the initial growth stage while increasing the size of the islands, and, as a result, leading to subsequent reduction of TD generation at epi-substrate interface. The other approach for the reduction of the dislocation density includes ultra-thin GaN interlayers (Fig. 1) in AlN layers that lead to abrupt stress at AlN/GaN interface during the growth and, as a result, to dislocation bending into the interface with their further annihilation ("blocking" effect). This approach was shown to result in the reduction of dislocation density down to 108-109cm-2 in the top (active) part of (2-3)-m-thick heterostructures. In addition, the precise control of the stress in AlGaN layers was achieved by a digital alloying technique and with optimization of growth parameters. HAADF STEM imaging (Fig. 3) evidences the presence of aperiodic superlattices (alternating Al- and Ga-rich AlGaN monolayers) formed in AlGaN cladding and waveguide layers grown under strong metal-enriched stoichimetrical conditions. Analysis of HAADF STEM images of AlGaN waveguide layers (Fig. 3) revealed the presence of 1-ML-thick Ga-enriched disks distributed in the nominally 2nm-thick AlxGa1-xN/AlyGa1-yN(x=0.5-0.7, y-x=0.1) SQW-structures in accordance with the growth cycle sequences in the sub-monolayer digital alloying technique. The mechanisms of defect generation in such hetero-epitaxial structures are discussed. Microsc. Microanal. 21 (Suppl 3), 2015 1804 References: [1] S. V. Ivanov, D. V. Nechaev, A. A. Sitnikova, V.V. Ratnikov, M.A. Yagovkina, N. V. Rzheutskii, E. V. Lutsenko, and V. N. Jmerik, Semicond.Sci.Technol. 29, 084008 (2014). Figure 1 BF TEM images (a, b) of AlN layers with GaN inserts taken at two perpendicular g vectors to compare the dislocation density with edge (a) and screw (b) components. Image (c) is enlarged BF TEM image of MEE AlN buffer layer showing voids at the AlN-sapphire interface. Figure 2 The HAADF STEM (a, b) and HRTEM (c) images of MEE grown AlN buffer layer and AlN-sapphire interfaces demonstrating formation of voids at the interfaces and lateral overgrowth phenomena. Figure 3. HAADF STEM images of Al0.6Ga0.4N waveguide layer with 2-nm-thick SQW showing formation of spontaneous super lattice in the waveguide layer.");sQ1[901]=new Array("../7337/1805.pdf","Study of the phase equilibria in the L-alanine � sodium nitrate system by optical microscopy and X-ray powder diffraction.","","1805 doi:10.1017/S1431927615009800 Paper No. 0901 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Study of the phase equilibria in the L-alanine � sodium nitrate system by optical microscopy and X-ray powder diffraction. Denniz M�rquez-Ruiz1, Javier Hern�ndez-Paredes1, Gemma Moreno-Corella1, Hilda E. Esparza-Ponce2, Ofelia Hern�ndez-Negrete1 and Mario Enrique �lvarez-Ramos1 Departamento de F�sica, Universidad de Sonora (UNISON), Blvd. Luis Encinas y Rosales S/N, Col. Centro, Hermosillo, cp 83000, Sonora, M�xico. 2. Departamento de F�sica, Centro de Investigaci�n en Materiales Avanzados S.C. (CIMAV), Miguel de Cervantes 120, Complejo Industrial, cp 31109, Chihuahua, M�xico. There is a continuous interest on controlling the final crystalline form of molecular materials. Among the organic materials, amino acids are of particular interest since they have relevant implications for biological systems [1]. In addition, they can form semi-organic (hybrid) materials in combination with inorganic salts [2,3]. A key to optimizing their physical and chemical properties is to control their final crystalline form. In the present work, we focused on determining the phases present in the L-alanine � sodium nitrate system (LASN). To achieve this objective, a series of crystallization experiments were set up in the laboratory followed by characterization techniques such as optical microscopy (OM) and X-ray powder diffraction (XRPD) to determine the crystal morphology and the crystalline structure, respectively. The study will help to understand the behavior of the system at different pH (3, 7 and 11), which is important in the development of these kind of materials. The crystallization experiments were carried out by the slow evaporation technique. The starting materials were purchased from Sigma Aldrich. Mixtures of a (1:1) molar ratio L-alanine/sodium nitrate were prepared and dissolved in double distilled water. Nitric acid (70%) was added to prepare the solution at pH 3 and ammonium hydroxide solution (28-30 %) for pH 11. OM images were taken using and Olympus MIC-D digital microscope. XRPD experiments were carried out in a Panalytical X-Pert PRO diffractometer equipped with an X�Celerator detector, using Nifiltered CuK radiation. The equipment was operated at 40 kV and 30 mA and the XRPD patterns were acquired from 5� to 80� 2 with a step size of 0.017�. Figure 1 displays the typical growth morphology of the starting materials grown from solution at room temperature for comparison. Figure 2 displays the products of the crystallization experiments in solution identified by crystal growth habit along with the XRPD of LASN at different pH carried out after complete evaporation. At pH=3, the Poly[2-L-alanine-3-nitrato-sodium (I), the L-alanine alaninium nitrate along with the precursor materials L-alanine and sodium nitrate were in equilibrium in solution, the XRPD data confirmed the observed materials. At pH=7, the Poly[2-L-alanine-3-nitrato-sodium (I), the L-alanine and sodium nitrate were in equilibrium in the solution. And at pH=11, only the Poly[2-L-alanine-3-nitrato-sodium (I) and the L-alanine were in equilibrium in the solution. References: [1] A. M. Alabanza and K. Aslan, Cryst. Growth Des. 11, (2011), 4300. [2] M. Ramos Silva, J. A. Paix�o, A. Matos Beja and L. Alte da Veiga, Acta Cryst. C57, (2001), 838. [3] K. Van Hecke, E. Cartuyvels, T. N. Parac-Vogt, C. G�rller-Walrand and L. Van Meervelt, Acta Cryst. E63, (2007), m2354. 1. Microsc. Microanal. 21 (Suppl 3), 2015 1806 [4] The authors acknowledge funding from the Mexican Government (CONACyT) through project No. 132856. a) b) c) Figure 1. Optical microscopy images of the growth morphology of: a) L-alanine; b) D-alanine and c) Sodium nitrate. 100000 a) 1 CPS 80000 (1) L-Alanine (2) Poly[l2-L-alanine-l3-nitrato-sodium(I)] (3) Sodium nitrate (4) L-alanine alaninium nitrate 3 4 60000 40000 2 b) 20000 60000 (1) L-Alanine (2) Poly[l2-L-alanine-l3-nitrato-sodium(I)] (3) Sodium nitrate 50000 40000 2 1 c) CPS 30000 3 2 20000 10000 120000 (1) L-Alanine (2) Poly[l2-L-alanine-l3-nitrato-sodium(I)] 100000 80000 CPS 60000 40000 1 20000 10 15 20 25 30 35 40 45 50 2 Figure 2. Optical microscopy images along with the XRPD data of the crystallization experiments at: a) pH=3; b) pH=7 and c) pH=11. (1) L-alanine; (2) Poly[2-L-alanine-3-nitrato-sodium (I); (3) Sodium nitrate and (4) L-alanine alaninium nitrate. (0,2,0)");sQ1[902]=new Array("../7337/1807.pdf","Microscopy Study of Morphological and Optical Changes in ZnO Nanostructures Induced by Pulsed Optical Excitation","","1807 doi:10.1017/S1431927615009812 Paper No. 0902 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microscopy Study of Morphological and Optical Changes in ZnO Nanostructures Induced by Pulsed Optical Excitation Zachariah M. Peterson, Robert C. Word and Rolf K�nenkamp Portland State University, Department of Physics, Portland, OR, USA The photoluminescence intensity and morphology of zinc oxide (ZnO) nanowires have shown inherent instability under high light intensities [1, 2]. We present scanning electron microscope (SEM) results that indicate accumulation of morphological changes under strong light exposure. In ZnO nanostructures, one typically finds a strong enhancement of the luminescence intensity up to some optimum annealing time, followed by a decrease as illumination continues [1, 2]. Here we show that a similar metastable behavior applies to the threshold and intensity of random lasing in ZnO nanoparticle films. In combination with a detailed analysis of the lasing behavior we conclude that local melting affects the electronic and optical properties of the material and can give rise to large lasing threshold and efficiency changes. Improvement as well as deterioration of the optical performance can be observed. Random lasers are optical devices that utilize random scattering in micrometer volumes to establish optical feedback and coherent laser emission. Since the optical feedback is not optimized, high pump intensities provided in short pulses are usually needed to achieve lasing. ZnO nanoparticle assemblies were among the first solid state random lasers to show promise for applications where small lasers sources for blue and ultraviolet light are needed [3]. Here we show that in simple ZnO random lasers the lasing threshold undergoes significant changes under typical pumping conditions. Among the various possible mechanisms we identify light induced melting and movement of individual nanoparticle scatterers during pumping. Our lasing samples were pumped and annealed with nanosecond UV laser pulses. Before imaging, we initially burn a reference spot into the sample with a high intensity laser pulse. The sample is then translated horizontally a specified distance and a reference area is imaged in SEM. Between annealing steps, the sample was again imaged in SEM to monitor morphology changes and the lasing properties of the sample were measured. When pumped with low intensity UV light, ZnO nanoparticles on Si substrate show strong excitonic photoluminescence with FWHM of ~15 nm. Steady improvement of the photoluminescence over time can also be observed, similar to previous reports [1, 2]. At high pump intensity we see a clear narrowing of the linewidth to ~3 nm. We also observed the characteristic fluctuations in the peak emission wavelength and intensity from shot-to-shot due to nanosecond pumping [3]. We show that adding scattering magnesium oxide (MgO) nanoparticles without gain to a ZnO nanoparticle random laser also decreases the threshold for lasing. The sample containing 30% MgO by weight had the lowest threshold and highest probability for lasing when compared to the other samples. Replacing MgO with an equal volume fraction of titanium dioxide (TiO2) had the opposite effect: the threshold was greatly increased. Pumping with single shots at high pump energies (141 mJ/cm2) did not show any visible changes in the film morphology. After 550 mJ of energy is supplied to the sample by laser annealing, however, accumulated damage to the nanostructure can be seen in SEM. The morphology changes due to melting that are visible in our nanoparticle samples are similar to the changes observed in ZnO nanowires that Microsc. Microanal. 21 (Suppl 3), 2015 1808 were laser annealed at exposures of ~160 mJ/cm2 (equivalent total energy of 624 mJ) with the same laser we used for pumping and annealing [1, 2]. Another area of the sample was annealed with 60 mJ/cm2 pumping energy with a total annealing energy of 1700 mJ. Similar melting could also be seen due to high annealing energies at moderate intensity. We have presented experimental results for threshold changes in a ZnO nanoparticle random laser on Si substrate. Adding 30% MgO (by weight) to ZnO showed the optimal decrease in the lasing threshold and corresponding increase in lasing probability and intensity. The threshold was also shown to change due to annealing by the pump laser. By extrapolating the rate of emission intensity change down to zero, our ZnO random lasers will have a stable morphology and average intensity when pumped with nanosecond pulse energies below ~20 mJ/cm2. Our data indicate that significant morphology changes can only be observed after accumulated exposure to the pump laser. These results motivate research for further methods for reducing the lasing threshold in ZnO random lasers. References: [1] Nadarajah, A., et. al. 11th IEEE International Conference on Nanotechnology, 2011. [2] Nadarajah, A., and K�nenkamp, R. Nanotechnology, 22, 025205, 2011. [3] Fallert, J., et. al. Journal of Luminescence, 129 (2009) 1685-1688. Figure 1. SEM micrographs before (left) and after laser annealing (right) with visible melting after annealing with energy density of 141 mJ/cm2 (total energy = 550 mJ). Yellow dots mark the same location in each image. Figure 2. (Left) Amplification thresholds for random lasing in ZnO (yellow), ZnO/30% MgO (blue), and ZnO/35% TiO2 (green). (Center) Emission intensity improvement due to various laser annealing intensities. The pumping energy density for each data point is 60 mJ/cm2. Dashed lines mark the standard deviation in the intensity fluctuations of single-shot spectra. (Right) Initial slopes for each curve in the center figure versus the corresponding annealing intensity.");sQ1[903]=new Array("../7337/1809.pdf","TEM Characterization of a Mg2Si0.5Sn0.5 Solid Solution for High-Performance Thermoelectrics","","1809 doi:10.1017/S1431927615009824 Paper No. 0903 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 TEM Characterization of a Mg2Si0.5Sn0.5 Solid Solution for High-Performance Thermoelectrics Minghui Song,1 Ji-Wei Liu,1 Masaki Takeguchi,1 Naohito Tsujii,2 and Yukihiro Isoda3 Transmission Electron Microscopy Station, National Institute for Materials Science (NIMS), 1-2-1 Sengen, Tsukuba, Ibaraki 3050047 Japan. 2. Quantum Beam Unit, NIMS, 1-2-1 Sengen, Tsukuba, Ibaraki 3050047 Japan. 3. Battery Materials Unit, NIMS, 1-2-1 Sengen, Tsukuba, Ibaraki 3050047 Japan. Intermetallic compounds of Mg2X (X = Si, Sn or Ge) are promising thermoelectric materials in the intermediate temperature range (400-800 K) because of their good electrical properties, low lattice thermal conductivity, and potentially high thermoelectric performance [1]. Studies have demonstrated that in the Mg2Si1-xSnx system, x 0.5 is one of the best choices for achieving a high thermoelectric performance of the material, because the large atomic mass difference between Si and Sn dramatically reduces the thermal conductivity of the solid solutions, which leads to a high value of figure of merit, ZT, of about 1.1 near 800 K [2]. The thermoelectric properties of these materials strongly depend on the local microstructure. Therefore, precisely understanding of the microstructure and composition of the material is essentially important for the development of high performance materials. In the present work, a Mg2Si0.5Sn0.5 solid solution was characterized with TEM and other related methods, in scales from nanometer to sub-millimeter. A Mg2Si0.5Sn0.5 solid solution was prepared by mixing Mg2Si and Mg2Sn powders and hot-pressing the mixture [3]. The starting materials were powder of Mg (purity 99.9%), Si (99.9999%), and Sn (99.999%). Mg2Si and Mg2Sn alloys were synthesized by the liquid-solid reaction and the melting reaction methods, respectively. The obtained Mg2Si and Mg2Sn ingots were ground, mixed, and hotpressed under Ar atmosphere at 80 MPa and 1068 K for 50 h. Structural and compositional characterization was carried out with XRD, SEM, electron probe microanalysis (EPMA), and TEM. The TEM observation and analysis were performed on a TEM, JEM-2100F (JEOL Co. Ltd.) operated at 200 kV. TEM specimens were prepared in two ways. One was the FIB method using a 30 keV Ga+ ion beam (JEOL, JIB-4000). Another way was to crush bulk materials to powders in sub-micrometer size and load the powder on polymer microgrids. XRD confirmed the successful synthesis of the Mg2Si0.5Sn0.5 solid solution which was in an antifluorite structure. The samples exhibited a much lower thermal conductivity (1.92 W m-1 K-1 at 300 K) than the parent Mg2Si (8.75 W m-1 K-1) and Mg2Sn compounds (6.28 W m-1 K-1). The low thermal conductivity of Mg2Si0.5Sn0.5 resulted in a relatively high ZT = 0.0132 at 300 K. SEM revealed that the grains were mainly 10-20 m in size and had clean grain boundaries without obvious inclusions and precipitates. EPMA analysis at points and for elemental mappings revealed that the relative differences in composition between grains could reach several percents, and that there were differences in composition at grain boundaries and inside the grains. Figure1 shows the EPMA elemental mappings. TEM and STEM observations revealed that there were typical two kinds of grains, structurally homogeneous and heterogeneous ones. Almost no large inclusions were observed in the former grains, while inclusions of Si and Sn in size from tens to hundreds of nanometers were observed in the later grains. Futhermore, in both kinds of grains, high density MgO nanoparticles 10-20 nm in size were observed which dispersed evenly in the grains. Figure 2 shows the TEM images, and the selected area diffraction patterns (SADP) 1. Microsc. Microanal. 21 (Suppl 3), 2015 1810 of a specimen prepared with FIB from a homogeneous grain. When the specimen was observed in a direction out of a low index zone axis, diffraction rings were observed which were from tiny MgO particles. The bright dots observed in the dark field image were MgO nano particles which distributed evenly in the specimen. The Oxygen and/or MgO may be originated from the starting materials or the atmesphere during the synthesis process. Dispersed MgO nano particles could decrease the thermal concuctivity since they scatter phonons, therefore, they might be benifical to the thermoelectric performance of the material. The reasons for the MgO nano particles to distribute dispersely in grains may be that the alloing during the hot-press was processed partially in solid state. The existance of the unreacted Si and Sn inclusions and the heterogeneous in elemental distribution convinced that there is still a space to improve the thermoelectric performance of the materials by optimizing the synthesis procedure, alloy composition, and doping level, etc. [4] References: [1] W. Liu, et al, Chem. Mater. 23 (2011), p. 5256. [2] V. K. Zaitsev, et al, Phys. Rev. B 74 (2006), p. 045207. [3] J. W. Liu, et al, J. Electronic Mater. 44 (2015), p. 407. [4] Thanks to Mitsuaki Nishio in NIMS for his help and discussion on the EPMA analysis. Figure 1. EPMA (electron probe microanalysis) elemental mappings of a Mg2Si0.5Sn0.5 specimen. (a) Backscattered electron image; (b-e) Mappings of Mg, Si, Sn, and O, respectively; (f) Intensity scales of the mappings. Figure 2. TEM images of a Mg2Si0.5Sn0.5 specimen. (a-c) The BF (bright field), the SADP (selected area diffraction pattern), and high resolution transmission electron microscopy images observed in direction of [110]; (d-f) The BF, SADP, and dark field images observed in a direction out of a low index zone axis.");sQ1[904]=new Array("../7337/1811.pdf","Real-time Observation of Electrochemical Sodiation of Co3O4/CNTs by in-situ Transmission Electron Microscopy","","1811 doi:10.1017/S1431927615009836 Paper No. 0904 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Real-time Observation of Electrochemical Sodiation of Co3O4/CNTs by in-situ Transmission Electron Microscopy Qianqian Li, Jinsong Wu, Vinayak P Dravid Department of Materials Science and Engineering, NUANCE Center, Northwestern University, Evanston, Illinois 60208, United States In recent years in-situ transmission electron microscopy (TEM) has become a unique and effective technique in revealing the underlying mechanisms and fundamental science of charge and ion transport, such as in lithium-ion batteries (LIBs) [1-7] and sodium-ion batteries (NIBs) [8-11] as well as potassium-ion batteries (KIBs) [11]. Such in-situ or socalled in-operando observations enable direct monitoring of the microstructural evolution during the electrochemical reaction process. Sodium ion battery as inexpensive and reliable energy storage setup attracts more and more researchers' attention. In this work, the electrochemically driven sodiation process and microstructural evolution of Co3O4/carbon nanotube nanocomposites are directly visualized in real-space and investigated close to atomic resolution by in-situ TEM under applied electrical bias. Uniform Co3O4 nanocubes with a regular cuboidal shape and size of ~5 nm anchored on the surface of carbon nanotube were synthesized by a simple hydrothermal method [12]. Carbon nanotube can act as both, the substrate to support active materials (Co3O4 nanoparticles) and a good conductor of ions and electrons. Upon the first sodiation reaction as shown in Fig.1, sodium ions firstly diffuse along the surface of carbon nanotube and then get inserted into Co3O4 crystalline lattice leading to a collapse of the Co3O4 crystalline lattice and formation of Na-Co-O amorphous clusters, a process which is reminiscent of solid amorphization reaction. In situ diffraction patterns indicate that the amorphous phase then gradually transforms into to a composite of Na2O and Co phase with continuous Na+ insertion in the sodiation process. There is about two-fold volume expansion of one nanotube after full sodiation, which is smaller than that in lithiation of Co3O4. Both discharging rate and reaction depth of sodiation reaction are much lower than that of lithiation due to larger ionic size of Na+. Thus, the sodiation dynamics is quite different from that of lithiation. In summary, we have observed the electrochemical reaction and microstructure evolution of Co3O4/CNTs nanocomposites during sodiation process, confirming that some of the 3d transition metal oxides have potential as electrode materials in sodium ion batteries [13, 14]. These observations may help understand the sodiation mechanism and aid in designing new advanced electrodes for high-performance NIB battery in the future. Reference: [1] J.Y. Huang, et al. Science, 330, (2010), p.1515. [2] C.M. Wang, et al. Nano Letters, 12, (2012), p.1624. [3] X.H. Liu, et al. Energy & Environmental Science, 4, (2011), p.3844. [4] Q. Li, et al. Chemistry of Materials, 26, (2014), p.4102. [5] Q. Li, et al. Journal of Materials Chemistry A, 2, (2014), p.4192. Microsc. Microanal. 21 (Suppl 3), 2015 1812 [6] Q. Li, et al. Carbon, 80, (2014), p.793. [7] L. Luo, et al. ACS Nano, 8, (2014), p.11560. [8] J.W. Wang, et al. Nano Letters, 12, (2012), p.5897. [9] K. He, et al. ACS Nano, 8, (2014), p.7251. [10] Q. Su, et al. ACS Nano, 8, (2014), p.3620. [11] Y. Liu, et al. Nano Letters, 14, (2014), p.3445. [12] J. Xu, et al. Journal of Power Sources, 274, (2015), p.816. [13] Z. Jian, et al. Chemical Communications, 50, (2014), p.1215. [14] Z. Jian, et al. Journal of Materials Chemistry A, 2, (2014), p.13805. [15] This work was supported as part of the Center for Electrochemical Energy Science, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences under Award # DEAC02-06CH11357. This work was also supported by the NUANCE Center new initiatives, and made use of the EPIC facility (NUANCE Center-Northwestern University), which has received support from the MRSEC program (NSF DMR-1121262) at the Materials Research Center, The Nanoscale Science and Engineering Center (EEC-0118025/003), both programs of the National Science Foundation; the State of Illinois; and Northwestern University. Figure 1. (a)-(l) The typical morphological evolution of one Co3O4 nanocube on the edge of CNT during sodiation process. Regular cube (a) has gradually changed to a smoothly expanded oval shape (l) (indicated by the red ovals in (a) and (l), with the (111) crystalline lattice of Co3O4 disappearing from the surface in the sodiation process. The (111) lattice finges of Co3O4 are shown by the yellow arrowheads.");sQ1[905]=new Array("../7337/1813.pdf","In Situ TEM Characterization of Nanostructured Dielectrics","","1813 doi:10.1017/S1431927615009848 Paper No. 0905 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In Situ TEM Characterization of Nanostructured Dielectrics Ming-Siao Hsiao1,4, Yifei Yuan2,3, Christopher Grabowski1,4, Anmin Nie2,3, Reza Shabazian-Yassar2,3, Lawrence F. Drummy1 1 2 Materials and Manufacturing Directorate, Air Force Research Laboratories, WPAFB, OH 45433 Department of Mechanical Engineering, Michigan Technological University, Houghton, MI 49931 3 Physics Department, University of Illinois at Chicago, Chicago, IL 60607 4 UES, Inc., Dayton, OH 45432 Polymer nanocomposites are being considered as potential materials for high energy capacitor due to the possibility of obtaining high dielectric breakdown strength characteristic of the organic matrix and large dielectric permittivity from inorganic filler [1,2]. However in most case the increase in the content of inorganic ceramic filler in polymer nanocomposite inevitably causes a decrease in the dielectric breakdown strength due to the agglomeration of fillers, the existence of defects, and inorganic-organic interfacial effects. Many studies have attempted to improve the compatibility of organic and inorganic, however in the large majority the dielectric breakdown strength is poor and the dielectric loss is high. Additionally, as nanostructured dielectrics are a relatively new class of materials for this application, reports of investigating fundamental mechanisms of dielectric breakdown for the dielectric nanocomposite on the nanoscale have been limited. Recently the dielectric performance of novel hairy polymer grafted nanoparticle assemblies has been compared to that of blended polymer nanocomposites; these two particular materials systems possessed well-dispersed morphology on the naonscale and had low dielectric contrast. It was shown that "hairy" polymer grafted nanoparticle assemblies showed substantial improvement in reducing dielectric loss and maintaining charge/discharge efficiency [3]. Here we presented in situ TEM observation of these hairy polystyrene grafted SiO2 nanoparticle assemblies (PS @ SiO2 NPs, aHNP-PS-1) during pre-breakdown/breakdown process using a TEM MEMS device and STM-TEM holders [4]. Fig. 1 shows the deposition of thin PS @ SiO2 NPs film, prepared by cryo-microtoming, across two electrodes on rectangular Si3N4 window in TEM MEMS device for in situ TEM experiment. HAADF-STEM images show the PS @ SiO2 nanoparticles are well dispersed. Fig. 2 shows in situ TEM of a PS @ SiO2 NPs film during a continuous voltage ramp within a STM-TEM holder, and significant microstructural morphology changes of the film at different electric fields were evident. Fig. 3 shows both morphology and electrical properties of PS @ SiO2 NPs assemblies before and after breakdown, and it was found that a highly conductive, tree-like structure formed after breakdown. Several additional phenomena including nanoparticle motion and electrostriction were observed. Surprisingly, the measured breakdown strength of PS @ SiO2 NPs film from in situ characterization can reach more than 900 V/�m when the assemblies on the nanoscale approach defect-free state [5]. References: [1] Whittingham, M. S. MRS Bull. vol. 33 (2008), p. 411 [2] Barber, P.; Loye H.-C. Z. et al Materials vol. 2 (2009), p. 1697 [3] Grabowski, C. A.; Vaia, R. A. et al ACS Appl. Mater. Interfaces vol. 6 (2014) p. 21500. Microsc. Microanal. 21 (Suppl 3), 2015 1814 [4] Hsiao, M.-S.; Drummy, L. F. et al In preparation [5] The authors acknowledge support from the Air Force Research Laboratory Fig. 1 SEM image of a cryo-microtoming PS @ SiO2 NPs film depositing across two platinum electrodes on rectangular silicon nitrile window in the TEM MEMS holder. HAADF-STEM image shows silica nanoparticles disperse well in the polystyrene. Fig. 2 In situ TEM observation of a PS @ SiO2 NPs film at different electric field. Fig. 3 Morphology and conducting behavior of a PS @ SiO2 NPs film before and after breakdown, respectively.");sQ1[906]=new Array("../7337/1815.pdf","Preparation of Electron and X-Ray Transparent Inorganic Particles for Analytical","","1815 doi:10.1017/S143192761500985X Paper No. 0906 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Preparation of Electron and X-Ray Transparent Inorganic Particles for Analytical Microscopy Using the Ultramicrotome Mark Homer1, Patrick Cappillino2, Farid El Gabaly1, Helmut Gnaegi3, Dave Robinson1, and Josh Sugar1 Sandia National Laboratories, Livermore, CA, USA. University of Massachusetts Dartmouth, Chemistry and Biochemistry Department, North Dartmouth, MA, USA. 3 Diatome Ltd., Biel-Bienne, Switzerland. 2 1 Many advanced materials characterization techniques that perform at the highest spatial and energy resolutions have strict dimensional requirements on acceptable samples. For example, the transmission electron microscope (TEM) generally requires samples to be electron transparent, which means samples must be <100 nm thick and they must fit in a holder that typically holds 3 mm-diameter discs. Electron energy loss spectroscopy (EELS) in the TEM may place further restrictions on sample thickness (e.g. < 50 nm). Likewise, scanning transmission x-ray microscopy (STXM) requires samples be x-ray transparent for x-ray absorption spectroscopy (XAS) and typically < 500 nm thick. Common sample preparation and thinning techniques include 1) dimpling and Ar-ion milling, or 2) focused Ga-ion beam (FIB) thinning. However, for samples that are made up of a collection of 100 nm � 5 �m diameter particles, this can be challenging. For example, preserving the arrangement of particles in an electrochemical experiment for ex situ microscopy is difficult. In addition, artifacts caused by redeposition and beam damage from ion beam removal techniques can be difficult to prevent. Here, we demonstrate the successful use of the ultramicrotome [1], usually used for soft and biological materials, for the preparation of thin inorganic particle samples. We will show examples of how the ultramicrotome can prepare thin sections of inorganic powder samples used in energy storage applications. Li-ion battery electrodes generally consist of inorganic oxide particles mixed with an organic binder. By mapping the spatial distribution of ion state of charge in the electrodes particle-by-particle, it is possible to understand the mechanisms of the charge/discharge reaction and degradation in these materials [2-4]. In our first example, the ultramicrotome provides a way to make thin samples from battery electrodes in which the electrode particle arrangement is preserved and particles can be imaged with nanometer spatial resolution. Our second example will show how the ultramicrotome can be used to prepare thin sections of metallic Pd-alloy based powders for hydrogen storage. Surface modification of these powders can have a large impact on their thermal stability, hydrogen storage properties, and kinetics of hydrogen uptake and release [5]. The ultramicrotome provides a way to make samples such that we can image surface modifications that range from tens of nanometers to sub-nanometer dimensions. References [1] D. Studer and H. Gnaegi, Journal of Microscopy-Oxford 197 (2000) [2] J.D. Sugar, et al., Journal of Power Sources 246 (2014) [3] Y. Li, et al., Nat Mater 13 (2014) [4] W.C. Chueh, et al., Nano Lett 13 (2013) [5] P.J. Cappillino, et al., Langmuir 30 (2014) [6] Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000. Microsc. Microanal. 21 (Suppl 3), 2015 1816 Figure 1: (a) Optical micrograph of LiFePO4 battery electrode sectioned with the ultramicrotome and attached to a TEM grid. The full section of the electrode is intact and the particle arrangement is preserved relative to the current collector. A lower magnification bright-field TEM image (b) shows that the cross section is electron transparent and a fraction of the particles have fallen off of the section. A STXM Fe state of charge map is overlaid a bright-field TEM image in (c) where the state of charge of the Fe cations is determined from the reference spectra shown in (d). It is possible to map the state of charge particle-by-particle in these Li-ion battery electrodes. Figure 2: (a) Lower magnification HAADF STEM image of Pd-alloy particles used for hydrogen storage. A higher magnification image near a particle surface is shown in (b) where lattice fringes are visible, demonstrating the high quality of this thin sample. Compositional analysis from EDS is shown in (c) where a <1 nm-thick surface modification layer enriched in Rh (by atomic layer electroless deposition) is visible.");sQ1[907]=new Array("../7337/1817.pdf","Development of TEM techniques dedicated for characterization of energy related composites and its application","","1817 doi:10.1017/S1431927615009861 Paper No. 0907 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Development of TEM techniques dedicated for characterization of energy related composites and its application Toshie Yaguchi1, Keiji Tamura1, Takashi Kubo1, Masaki Kondo1, Hiroaki Matsumoto2, Takahiro Shimizu3 and Takeo Kamino3 1. 2. Electron Microscope Systems Design 2nd Dept., Hitachi High-Technologies Corp., Ibaraki, Japan Application Development Dept., Hitachi High-Technologies Corp., Ibaraki, Japan 3. Japan Automobile Research Institute, Ibaraki, Japan In the field of energy-related composites, TEM techniques allowing characterizations of initial structures of electron beam sensitive nano-composites and their dynamics during degradation or synthesis are strongly required [1]. In this paper, recent progress in the improvement of TEM techniques based on a 40-120 kV high resolution analytical TEM and some application to the catalysts are discussed. The spherical aberration coefficient and TEM image resolution of high resolution objective lens at the accelerating voltage of 120 kV are 1.1 mm and 0.2 nm respectively. The high resolution objective lens pole-piece has been newly designed to accommodate a high solid angle silicon drift detector of an energy dispersive X-ray analyzer. A FIB fabricated apertures with the smallest diameter of 1m are equipped as the field limiting aperture for structural analysis of individual nanometer sized crystalline. The external view of the microscope (HT7700) is shown in Figure 1. A high-speed turbo molecular pump with the pumping speed of 250 l/s, oil free scroll pump and safety equipment such as gun-air lock valve linking of the gun vacuum enabled high temperature-high resolution in situ TEM observation of solid-gas reactions routinely and safely. Depending on the required gas pressure level at the specimen area, the electron source can be chosen either a LaB6 single crystal emitter or a tungsten hair-pin filament. A gas pressure in the specimen area can be raised up to 0.1 Pa when the tungsten hair filament is employed. A compact gas supply system containing three 4.5 litter standard gas cans, giving responsive controlling of specimen atmosphere, has been developed for in situ TEM experiment of gas-solid reaction [2-3]. The flow rates of three gases can be individually and precisely controlled before introduction to the specimen chamber. Figure 2 shows external view of the gas supply system installed on the operation board of a TEM. The length of gas path between the gas supply system and a specimen heating holder equipped with a gas injection nozzle is 20 cm at minimum. Combination of the short gas path and the local gas injection method to the specimen made time of gas replacement much shorter than that of conventional environmental TEM. An in situ sequence, taken from a video recording, of the degradation of a Pt/CB electrocatalyst of a fuel cell is shown in Figure 3. Accelerating voltage and electron beam density were 120 kV and 28 mA/cm2, respectively. The specimen was heated to approx. 200 oC in the air atmosphere of 0.1 Pa. Structural changes of both carbon support and Pt nano particles are clearly demonstrated. References: [1] Kamino T, et al, J. Electron Microsc.54, (2005) pp.497�503. Microsc. Microanal. 21 (Suppl 3), 2015 1818 [2] Yaguchi T, et al, J. Electron Microsc. 60, (2011) pp.217-225. [3] Yaguchi T, et al, Proc.of IMC 18, Prague, Czech Republic (2014) ID-12-P-1550. Figure 1. External view of 40-120 kV high resolution analytical TEM. Figure 2. External view of the gas supply system installed on the operation board of a TEM. Figure 3. In situ sequence, taken at 120 kV, of the degradation of a Pt/CB electrocatalyst of a fuel cell. The specimen was heated to approx. 200 oC in the air atmosphere of 0.1 Pa.");sQ1[908]=new Array("../7337/1819.pdf","Investigation of Li ion and Multivalent Battery Systems Using In situ TEM and High Resolution EELS","","1819 doi:10.1017/S1431927615009873 Paper No. 0908 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Investigation of Li ion and Multivalent Battery Systems Using In situ TEM and High Resolution EELS Arijita Mukherjee1, Hasti Asayesh Ardakani2, Patrick J. Phillips1, Reza S.Yassar2, Robert F. Klie 1 1. Department Of Physics, University Of Illinois At Chicago, 845 West Taylor Street,Chicago,Illinois 60607,United States 2. Department of Mechanical Engineering-- Engineering Mechanics, Michigan Technological University,1400 Townsend Drive, Houghton, Michigan 49931,United States Energy storage research is of utmost relevance today with demands for higher energy density, faster cycling times, safer and more cost effective options. Research is underway both for exploring novel cathode materials for traditional Li ion batteries and also multivalent elements such as Mg, Ca (both divalent) or trivalent Al to replace Li ion for next generation batteries. V2O5 has been proposed as a promising cathode host for multiple reasons: its layered structure can potentially intercalate Li or other multivalent ions (e.g., Mg); a shorter Li diffusion path and high Li ion mobility has been reported for nanostructured V2O5; its theoretical energy density is higher than conventional Li ion battery electrodes; and, its low cost. [1] The following contribution will focus on various in-situ transmission electron microscopy (TEM) experiments using a biasing electrochemical holder, with a Li metal anode and V2O5 nanowires as the cathode, over a variable bias window within an open cell battery approach. This approach has shown considerable success for in-situ experiments involving Li-ions and SnO2 nanowire cathodes. [2] Electron diffraction patterns of the V2O5 nanowire cathode collected before and after bias application have shown significant change from the crystalline phase before to some polycrystalline, possibly intercalated phase afterwards. Electron energy-loss spectroscopy (EELS) analysis of the nanowires both before and after bias application also manifested peak shifts indicating a change in the valence state of Vanadium (initially V5+). The primary goal is to identify the dominant lithiated phase upon in-situ cycling. Preliminary indexing of polycrystalline diffraction pattern indicates 2 2 5 as the dominant phase. EELS analysis will be used to verify this. [3]. Figure 1 shows EELS spectra of the cathode collected before and after bias application at around -2.5 V for in-situ lithiation; the peak shifts indicate a definite change in the Vanadium valence state following the bias application. Figure 2 presents selected area electron diffraction patterns, captured from the nanowire tip which was in contact with Li; the phase change from crystalline (before bias, pristine V2O5) to a polycrystalline state (after bias) also confirms Li intercalation. The polycrystalline rings have been indexed to 2 5 . (JCPDS card 48-0076). Results will also be discussed regarding multivalent systems such as Ca or Mg, using a similar approach as outlined above, and an appropriate ionic liquid electrolyte. Preliminary results have shown some changes in the diffraction pattern following attempted Ca intercalation and further experiments are planned. Also, in operando TEM will be discussed using electrochemical liquid cells and appropriate electrolyte and cathode configuration. Microsc. Microanal. 21 (Suppl 3), 2015 1820 References [1] Zhiguo Wang et al, Phys.Chem. Chem. Phys., 15 (2013), p.8705 [2] Anmin Nie et al, Atomic-Scale Observation of Lithiation Reaction Front in Nanoscale SnO2 Materials, ACS Nano, 7 (2013),p. 6203 [3] Candace K Chan et al, Fast, Completely Reversible Li Insertion in Vanadium Pentoxide Nanoribbons, Nano Letters,7(2)(2007),p.490 [4] This work is supported by the Joint Center for Energy Storage Research (JCESR), an Energy Innovation Hub funded by the U.S. Department of Energy (DOE), Office of Science, Basic Energy Sciences. a) b) Figure 1. (a) EELS spectra of V2O5 nanowire before and (b) after bias application for in situ lithiation experiment Figure 2. (a) Electron Diffraction Pattern of V2O5 nanowire before and (b) after bias application for in situ lithiation experiment");sQ1[909]=new Array("../7337/1821.pdf","Phase Transformation of Molybdenum Carbide in Carburization of MoS2 Studied by in-situ Environmental TEM","","1821 doi:10.1017/S1431927615009885 Paper No. 0909 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Phase Transformation of Molybdenum Carbide in Carburization of MoS2 Studied by in-situ Environmental TEM Jian Chen1, Qiang Wei2, Jinwen Chen2 National Institute for Nanotechnology (NINT), National Research Council Canada, 11421 Saskatchewan Drive, Edmonton, Alberta T6G 2M9, Canada. 2. CanmetENERGY-Devon, Natural Resources Canada, 1 Oil Patch, Devon, Alberta T9G 1A8, Canada. The catalytic behaviour similarity of molybdenum carbides to the noble metals has brought great interest in academia and in industry. However, most molybdenum carbides are produced via high temperature reaction between molybdenum oxide and a carburizing gas. This leads to low surface area, which brings negative impact on ultimate catalytic performance of molybdenum carbides. To date, techniques for synthesizing MoS2 with high surface area are well developed [1]. This provides a possibility to synthesize molybdenum carbides with high surface area through carburization of MoS2. In the current study, the in-situ carburization of MoS2 was performed in the Hitachi H-9500 environmental TEM. Acetylene/H2 mixture was employed in the experiment. MoS2 sample was first heated up to 900oC, and then H2 and acetylene was delivered in sequence to the microscope column part which is separated from the rest part of column by differential pumping apertures. As shown in Figure 1, nano particles (NPs) emerged at the surfaces of MoS2 slabs after exposure to acetylene for five minutes. Three NPs marked by a square are displayed in Figure 1(A). Their enlarged image is shown as an inset in the figure as well. Figure 1(B) shows another region where a particle with well-developed crystalline lattice (see inset in the figure) and its difractrogram. The simulation for the diffractrogram suggests a cubic -MoC1-x with facecentered cubic (fcc) structure form. The similarity between diffractrogram in Figure 1(A) and (B) suggests the NPs studied are all -MoC1-x. Judging from their sizes, we postulate that -MoC1-x is the first phase nucleates and grows under the experimental condition. Studies suggest that there is a relationship between the formed type of molybdenum carbide and carburizing gas. For example, -MoC1-x usually forms when carburising with a mixture of butane and hydrogen while hexagonal closed packed (hcp) -Mo2C comes out when changing the carburizing agent to a mixture of methane and hydrogen [2]. Molybdenum carbide can occur in a variety of crystal structure, but is commonly seen in two types, fcc -MoC1-x and hcp -Mo2C, which possess different catalytic property [3]. Therefore studying the phase transformation mechanism between -MoC1-x and -Mo2C is beneficial for tailoring catalytic properties. MoC1-x has an ABCABC packing of molybdenum atoms, whereas -Mo2C exhibits ABAB stacking sequence. The phase transformation from fcc to hcp structure can be realized by the passage of 1/6 <112> Shockley partial dislocation along alternate (111) plane in fcc structure. Figure 2 shows a HRTEM image of such a region. The indexation of the diffractrograms for 1. Microsc. Microanal. 21 (Suppl 3), 2015 1822 regions A, B and C reveals -MoC1-x forms in A and B while -Mo2C develops in C. -MoC1-x in A and B forms a twin structure. The twinning plane is marked by Line 1. A Burgers circuit around the plane marked by Line 2 reveals closure failure in the <112> direction, which indicates the existence of Shockley partial dislocation (b=1/6 <112>) at the position indicated by arrow. This implies that Shockley partial dislocation induces fcc to hcp transformation in molybdenum atoms. A prominent structural feature of carbon in molybdenum carbide with fcc structure is to occupy octahedral sites sharing common edges. When fcc structure transfers to hcp structure, the connection of octahedral site changes from edge-sharing to face-sharing. A strongly repulsive interaction between carbon atoms would happen if carbon atoms occupy simultaneously two octahedral sites sharing a common face. Therefore, vacancies need to be introduced to carbon sublattice to accommodate such repulsive interaction. Obviously, large deviation from the stoichiometric composition in -MoC1-x provides such accommodation for the arrangement of carbon atoms in octahedral sites when phase transformation from fcc to hcp occurs in molybdenum atom arrangement. References: [1] N Liu, P Kim et al, ACS Nano 8 (2014) 6902. [2] T C Xiao, A P E York et al, J. Mater. Chem 11 (2001) 3094. [3] W Q Xu, P J Ramirez et al, Catal. Lett. 144 (2014) 1418. 1 (A) (B) 2 A B C Figure 1 Bright field images of formed -MoC1-x NPs on MoS2 slabs. Figure 2 HRTEM image of a region where MoC1-x and -Mo2C form. A Burgers circuit around the plane marked by line 2 reveals closure failures in the <112> direction.");sQ1[910]=new Array("../7337/1823.pdf","Real Structure and Structural Changes of Functional Tellurides","","1823 doi:10.1017/S1431927615009897 Paper No. 0910 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Real Structure and Structural Changes of Functional Tellurides Lorenz Kienle1, Torben Dankwort1 Jan K�nig2, Markus Winkler2, Anna Hansen3, Christine Koch3, Jeffrey Ditto4, Dave C. Johnson4 and Wolfgang Bensch3 1. 2. Institute for Materials Science, University of Kiel, Kaiserstrasse 2, 24143 Kiel, Germany Fraunhofer Institute IPM, Heidenhofstrasse 8, 79110 Freiburg, Germany 3. Institute for Inorganic Chemistry, University of Kiel, Max-Eyth-Strasse 2, 24118 Kiel, Germany 4 Department of Chemistry and Materials Science Institute, University of Oregon, 373 Klamath Hall, Eugene, OR 97403, USA Nowadays, functional tellurides are widely used as bulk- and nanomaterials for thermoelectric power generators and phase-change based applications, e.g. optical data storage. The function and performance of the materials strongly depend on their unique nanostructural properties and their structural evolution upon operation. Both features can be fully characterized in-situ, ex-situ and on a broad range of length scales by combining diverse characterization techniques with transmission electron microscopy. Consequently, essential information about real-structure property relations and fatigue mechanisms can be determined enabling first steps to a knowledge-based tailoring of the materials. Crystalline phase change materials with general formula (GeTe)a(Sb2Te3)b, are based on a rocksalt-type structure. The cation sites of the parent structure are occupied by Ge and Sb atoms as well as a molar fraction of c = b � a structural vacancies per formula unit. Distinct arrangements of the structural vacancies are observed in the course of thermal operation. Starting with amorphous thin films in-situ observations prove a two-step mechanism for the crystallization of most of the GeTe-rich (a > b) phases [1]. Firstly crystalline but metastable nanoparticles with random distribution of structural vacancies are formed; secondly these vacancies are combined to layered aggregates, the so called vacancy layers. The distances in-between the vacancy layers scales with the GeTe content (a). This crystalline state with trigonal average structure remains stable upon long-term annealing. In order to derive information about the intermediate state in-between random disorder and layered aggregation of structural vacancies we applied in-situ microscopy on chemically modified phase change materials. In case of Sn-doped thin films we firstly observed the evolution of diffuse scattering by means of in-situ heating experiments. Such indicator for pronounced short-range ordering of the structure is assumed to be interconnected with a correlated arrangement of the structural vacancies. Telluride bulk- and thin-film materials were also regarded promising candidates for thermoelectric applications, particularly when introducing chemical and/or structural segregation (so called nanostructuring [2]) of the materials to reduce the thermal conductivity, and thus enhancing thermoelectric efficiency. However, nanostructuring as metastable feature of materials is frequently degrading in the course of thermoelectric application. Recently, we probed the thermoelectric efficiency and stability of doped phase change materials by introducing them into thermoelectric power generators [3]. Such optimized benchmarking of thermoelectric materials is not restricted to the determination of the intrinsically imprecise thermoelectric figure of merit (ZT value), and gives information about the microstructural and performance changes during thermal operation of the materials. Consequently, failure mechanisms can be determined and the resistance of the materials against functional and microstructural fatigue can be specified. In case of a low doping level of the phase change material the microstructure of the as-prepared material remains unchanged upon thermal cycling, consequently, Microsc. Microanal. 21 (Suppl 3), 2015 1824 excellent and stable thermoelectric performance was obtained. For higher doping level we observed strain phenomena for the as-prepared state which are released by the formation of planar defects after thermal cycling. Consequently, the thermoelectric performance degrades significantly upon repeated thermal cycling. In a second set of experiments, we analyzed the thermal stability of chemically segregated (Bi2Te3)a(Sb2Te3)b superlattices [4,5] which were firstly prepared by molecular beam epitaxy. Starting at T = 200 �C, the superlattice structures are gradually transformed into the homogeneous alloy. In-situ microscopy proves that the interdiffusion preferably starts next to chemical defects of the superlattice while areas with low density of defects remain stable up to 300 �C. These observations suggest that the structural integrity of the superlattices critically depends on the defect density of the material as further supported by recent in situ- and ex situ observations. References: [1] J. Tomforde, W. Bensch, L. Kienle, V. Duppel, P. Merkelbach and M. Wuttig, Chem. Mater., 2011, 23, 3871. [2] K. Biswas, J. He, I. D. Blum, C.-I Wu, T. P. Hogan, D. N. Seidman, V. P. Dravid and M. G. Kanatzidis, Nature, 2012, 489, 414. [3] J. Koenig, M. Winkler, T. Dankwort, A.-L. Hansen, H.-F. Pernau, V. Duppel, M. Jaegle, K. Bartholom�, L. Kienle and W. Bensch, Dalton Trans., 2015, 44, 2835. [4] A.-L. Hansen, T. Dankwort, M. Winkler, J. Ditto, D. C. Johnson, J. D. Koenig, K. Bartholom�, L. Kienle and W. Bensch, Chem. Mater., 2014, 26, 22, 6518. [5] U. Sch�rmann, Markus Winkler, J. D. Koenig, X. Liu, V. Duppel, W. Bensch, H. B�ttner and L. Kienle, Adv. Eng. Mater., 2012, 14, 3, 139. The authors acknowledge funding from the German Research Foundation within priority program 1386");sQ1[911]=new Array("../7337/1825.pdf","In Situ Electrochemical Deposition of Poly(3,4-ethylenedioxythiophene) (PEDOT)","","1825 doi:10.1017/S1431927615009903 Paper No. 0911 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In Situ Electrochemical Deposition of Poly(3,4-ethylenedioxythiophene) (PEDOT) Jinglin Liu1, Bin Wei1, Jennifer Sloppy1, Liangqi Ouyang1, Chaoying Ni1 and David C. Martin1,2 Department of Materials Science and Engineering, The University of Delaware, Newark, DE 19716, USA. 2. Department of Biomedical Engineering, The University of Delaware, Newark, DE 19716, USA. Conjugated polymers are widely used in organic solar cells, chemical sensors and biomedical devices because of their relatively high conductivity and soft mechanical properties [1,2,3]. Electrochemical deposition has long been an important method of fabricating conjugated polymer thin films for various applications. The morphology of the films, and the corresponding performance of the devices is particularly sensitive to the detailed fabrication conditions. Previous efforts have investigated the nucleation and growth mechanisms of electropolymerized conjugated polymers, however the results are usually obtained from in-direct and lower resolution methods like UV-vis and AFM [4,5]. More detailed information with ultra-high resolution is still needed. Electrochemical liquid electron microscopy makes it possible to analyze this initial nucleation and growth process in direct imaging method in a TEM with nanometer scale resolution. This method has proven effective for the analysis of electrochemical deposition mechanisms in copper and lead [6,7]. No previous studies have examined the in situ TEM electrochemical deposition of conjugated polymers, mostly probably due to the fact that polymers are highly electron beam sensitive. This high sensitivity makes the in situ TEM imaging of conjugated polymer electrodeposition much more experimentally difficult than for inorganic materials. Of all the conjugated polymers, poly(3,4-ethylenedioxythiophene) (PEDOT) has been extensively studied for its high conductivity and stability, and has been utilized in various applications. PEDOT was chosen here as a model system to demonstrate the feasibility of performing electrochemical deposition of conjugated polymers within the TEM [8]. The in situ imaging of electrochemical deposition of PEDOT was conducted on a 300 kV JEOL 3010 electron microscope with a Protochips Poseidon 500 electrochemical liquid flow holder. Ethylenedioxythiophene (EDOT) monomer solution with counter-ion electrolyte (LiClO4) was used in continuous flow mode, where the liquid layer was sealed between two microfabricated silicon chips. Thin silicon nitride windows (~50 nm) on the chips allow the electron beam to pass through for TEM imaging. The liquid layer thickness was defined by the spacers on bottom chips. Electrochemical deposition and measurements were carried out with a Gamry Reference 600 potentiostat/galvanostat. A 20 �m wide glassy carbon working electrode was in the middle of the membrane with a platinum reference electrode around working electrode and remotely counter electrode (Figure 1a). Effective electrical contact was tested using cyclic voltammetry (CV) and successful PEDOT deposition from both CV and constant voltage (potentiostatic) methods were confirmed from TEM imaging. PEDOT was found to nucleate first near the edges of the working electrode (Figure 1b). With further deposition, the initial clusters increased in both size and thickness, along with additional nucleation sites seen in the middle of the working electrode. Liquid-like fluctuating domains were observed during deposition and of a size and rounded shape consistent with the bumpy surface structure seen in SEM images of electrochemically polymerized PEDOT thin films. After the experiment, the electrochemical chip was taken out for correlative optical microscopy (OM) and scanning electron microscopy (SEM) 1. Microsc. Microanal. 21 (Suppl 3), 2015 1826 imaging. The dark color under OM, the bumpy surface structure seen in the SEM, and the strong sulphur signal from EDS elemental mapping all confirmed the successful deposition of PEDOT (Figure 1c). In conclusion, the in situ electrochemical deposition of conjugated polymers was successfully conducted and imaged under TEM with an electrochemical liquid flow cell for the first time. This new technique provides an unparalleled method for directly investigating the early stage nucleation and growth mechanisms of electropolymerized conjugated polymer thin films. This information will be of great value in understanding the performance of conjugated polymer thin films and in optimizing processing conditions for devices with enhanced properties [9]. References: [1] S. G�nes, H. Neugebauer and N. S. Sariciftci, Chemical reviews, volume 107, no. 4 (2007), p. 1324� 1338. [2] K. Lee, J. M. Rouillard, T. Pham, E. Gulari and J. Kim, Angewandte Chemie, volume 46, no. 25 (2007), p. 4667�4670. [3] X. Cui, V. A. Lee, Y. Raphael, J. A. Wiler, J. F. Hetke, D. J. Anderson and D. C. Martin, Journal of biomedical materials research, volume 56, no. 2 (2001), p. 261�272. [4] A. R. Hillman, and E. F. Mallen, Journal of Electroanalytical Chemistry, volume 243 (1988), p. 403�417. [5] M. Innocenti, F. Loglio, L. Pigani, R. Seeber, F. Terzi and R. Udisti, Electrochimica Acta, volume 50 (2005), p. 1497�1503. [6] A. Radisic, P. M. Vereecken, J. B. Hannon, P. C. Searson and F. M. Ross, Nano Letters, volume 6, no. 2 (2006), p. 238�242. [7] E. R. White, S. B. Singer, V. Augustyn, W. A. Hubbard, M. Mecklenburg, B. Dunn and B. C. Regan, ACS Nano, volume 6, no. 7 (2012), p. 6308�6317. [8] J. Liu, B. Wei, J. Sloppy, L. Ouyang, C. Ni and D. C. Martin, in preparation. [9] The authors acknowledge funding from the National Science Foundation, Grant Number 1103027 and the University of Delaware. a Pt reference electrode Pt counter electrode Silicon nitride window b c Sulphur Glassy carbon working electrode 200 �m 20 �m 50 �m Figure 1. a: Scanning electron micrograph of the electrochemical top chip showing the middle silicon nitride window for imaging with glassy carbon working electrode, Pt reference and counter electrode for electrochemical deposition and measurements. b: Transmission electron micrograph showing early stage PEDOT deposition around the edges of glassy carbon working electrode. c: Elemental map for sulphur signal confirming the successful deposition of PEDOT.");sQ1[912]=new Array("../7337/1827.pdf","Cryo-STEM Reveals Humidity-Controlled Shape Change in Silica Nanoparticles","","1827 doi:10.1017/S1431927615009915 Paper No. 0912 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Cryo-STEM Reveals Humidity-Controlled Shape Change in Silica Nanoparticles Katherine A. Spoth1, Yao Sun2, Ulrich Wiesner2 and Lena F. Kourkoutis1,3 1. 2. Department of Applied and Engineering Physics, Cornell University, Ithaca, New York, USA. Department of Materials Science and Engineering, Cornell University, Ithaca, New York, USA. 3. Kavli Institute at Cornell for Nanoscale Science, Ithaca, New York, USA. Cryo-electron microscopy allows the direct observation of cryo-immobilized structures in solution. It therefore is a powerful tool for visualizing formation processes in liquids for which resolution, contrast, or beam damage prevents real-time liquid cell TEM observation. Here, we use cryo-STEM to image at high-resolution hybrid mesoporous silica nanoparticles with structure dictated by the interaction between inorganic silica and an ordered organic template. We demonstrate that surprisingly these silica particles behave as shape change materials (SCMs). Shape changes can be induced via the presence of water, e.g. by controlling the humidity during particle drying, suggesting potential applications of these silica-based nanoparticles as humidity-responsive material. Mesoporous silica nanomaterials show great potential for applications in a variety of fields. Single-pore particles have been investigated for drug-delivery purposes, having a sufficiently small diameter for removal by the kidneys to avoid toxic effects [1]. Larger particles with high surface area due to their interconnected pore structure are promising for energy and sensor applications [2]. While porous silica materials have been extensively studied, their formation is not well understood. Cryo-TEM allows study of these solution-based processes as demonstrated recently for assembly of monodisperse silica nanoparticles [3]. Typically, however, these materials are characterized by conventional electron microscopy in which the drying process can alter the materials' structure. To reveal the structure of mesoporous silica nanoparticles in solution (80% ethanol and 20% water), we vitrified the sample by plunge freezing using a liquefied ethane/propane mixture and imaged it with cryo-STEM. Figure 1 shows the particles' regular hexagonal shape and symmetric pore structure, visible in both front and side views of single particles. The average pore spacing determined from the power spectrum of the image is ~3.6 nm. The image resolution is at least 1.8 nm as indicated by clear spots at that spacing. In contrast, when the sample is dried directly on the grid for traditional TEM the structure shrinks and deforms, resulting in a star-shaped cross-section. To understand these structural modifications, we have studied the effect of the drying environment on the particle shape. Starting with the known structure imaged by cryo-STEM (Fig. 2a, d), the sample was dried in the vacuum of the microscope by warming it to room temperature, allowing the ice to sublimate during the process. Surprisingly, the particle shape remained close to hexagonal (Fig. 2e), however, its size shrank by ~10%, from 176 nm to 159 nm (corner to corner distance). Subsequently, the sample was stored in a humid environment and re-imaged after 12 days. The particles' shape changed significantly with some pores visible but irregular (Fig. 2c, f); in addition the size of the particle further decreased to 137 nm. The effect of water on the particle shape was confirmed by imaging particles synthesized in pure water. These particles exhibited star-shaped cross sections even when imaged by cryo-STEM [4]. [1] K Ma et al, Chem. Mater., 25 (2013), p. 677-691. [2] F Hoffmann et al, Angew. Chem. Int. Ed. 45 (2006), p. 3216 � 3251. [3] C Carcou�t et al, Nano Letters, 14 (2014), p. 1433-1438. [4] This work was supported by the Cornell Center for Materials Research with funding from the NSF MRSEC program (DMR-1120296). Microsc. Microanal. 21 (Suppl 3), 2015 1828 50 nm 1.8 nm Figure 1. Cryo-STEM reveals symmetric pore structure of silica nanoparticles embedded in vitreous ice. Front (left) and side (right) orientations show average pore spacing of ~3.6 nm confirmed by FFT with resolution better 1.8 nm. (a) (b) (c) 500 nm (d) (e) (f) 50 nm Figure 2. Shape change behavior of silica nanoparticles in response to humidity during drying process. Cryo-STEM shows regular, symmetric hexagonal structure of material (a, d). After vacuum drying in the microscope column, the same particles exhibit minor changes in shape (b, e). Exposing the sample to humid air causes complete deformation of the structure (c, f). A single particle shows size decrease when dried (d-f). Starting at 176 nm in solution, vacuum drying decreases the corner-to-corner distance to 159 nm; further shrinkage to 137 nm occurs upon exposure to humidity.");sQ1[913]=new Array("../7337/1829.pdf","Direct Conversion Biogas to Multiwall Carbon Nanotubes and Syngas over Starch","","1829 doi:10.1017/S1431927615009927 Paper No. 0913 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Direct Conversion Biogas to Multiwall Carbon Nanotubes and Syngas over Starch Derived Ni@C Nanoparticles. Qiangu Yan1, Fei Yu1, I-Wei Chu2, Amanda Lawrence2 1. Department of Agricultural and Biological Engineering, Mississippi State University, Mississippi State, MS, USA 2. Institute for Imaging and Analytical Technologies, Mississippi State University, Mississippi State, MS, USA Biogas is mainly composed of CH4 and CO2 [1]. The primary usage of biogas is as an internal combustion engine (ICE) fuel. Biogas is also studied to produce hydrogen as fuel cell feedstocks. An exciting alternative is to catalytic convert biogas to syngas which can be further upgraded to liquid fuels and chemicals through Fischer�Tropsch process. During CH4-CO2 dry reforming process, the most challenging problem is the catalyst deactivation by carbon deposition. This concern is even significant if using biogas directly, since the higher CH4/CO2 ratio (up to 1.5~1.6 for biogas) will end with higher and faster carbon formation over the catalyst that rapidly deactivates it. A new approach has been proposed to overcome this catalyst deactivation by converting the deposited carbon to a value-added product while avoiding the catalyst deactivation. The key issue of this procedure is to find a suitable catalyst to implement both the reforming reaction and the nano carbon structure formation since nano-carbons have been widely applied in material, electronic, energy storage, biomedical and environmental area. Nickelbased catalysts are widely used for both methane reforming and nano carbon formation processes. They promote the methane reforming reaction and effectively convert methane into highly valued products like carbon nanotubes, graphene and carbon nanofibers. This reactivity makes nickel-based catalysts an excellent choice to convert biogas into hydrogen-rich syngas and carbon nanostructures since biogas have relatively high CH4:CO2 ratios. The principal objective of this work was to produce and characterize multiwall carbon nanotubes (MWCNTs) and hydrogen-rich syngas from biogas over a highly efficient carbon-encapsulated metal nanoparticle catalyst. Carbon-encapsulated nickel-core nanoparticles were synthesized hydrothermally by the reduction of Ni2+ from a starch solution. Six grams of Ni(NO3)2�6H2O,30 g starch and 250 ml DI water were stirred for 30 min then transferred into a 500-mL Parr reactor; the reactor was heated up to 180 �C and kept for 8 h. After reaction the resulting black product was collected and washed three times with DI water and ethanol. The final product was oven-dried at 110 �C overnight. The dried carbon-encapsulated nickel nanoparticles were calcined at 900 C under a continuous nitrogen flow (100mL/min) for one hour. Catalytic conversion of biogas was carried out in a stainless steel tubular reactor (0.0127 m O.D.) with 3g of calcined nickel nanoparticles. The catalyst was first reduced at 400 �C under H2/N2 (1:1) gas flow for 3 h, then increasing the temperature to 850 �C feeding with CH4/CO2 gases (100 mL/min) at a ratio of 3/2 for one hour. The solid sample was collected and boiling in 0.1M HNO3 solution for one hour. The prepared samples were characterized using microscopy to investigate the morphology, particle size and fine crystal structures. Samples for SEM were prepared by sprinkling the dried material on to double sided carbon tape affixed to AL stubs and coating with 10 nm PT (EMS 150 T coater). Samples were examined using a JEOL JSM 6500F microscope at 10 kv. For TEM the dried samples were mixed with 100 % ETOH and sonicated for 20 min. A drop of the liquid was placed on a formvar/carbon 300 mesh copper grid for one hour then gently removed with filter paper. Grids were examined with a JEOL JSM 2100 microscope at 200kv. X-ray powder diffraction (XRD) patterns of the synthesized samples were obtained using a Rigaku Ultima III X-ray Diffraction. Microsc. Microanal. 21 (Suppl 3), 2015 1830 SEM shows that the fresh nickel catalyst sample is spherical in shape with particle size between 1 and 2 �m (Fig.1a) whereas the micro-sphere calcined at 900 C exhibits a porous structure (Fig.1b). After catalytic conversion biogas at 850�C for 1 hour Fig.1c shows a large amount of MWCNTs formed in the spheres. The purified sample (Fig.1d) clearly shows the MWCNTs with much higher purity with diameters ranging from ~ 30 to 100 nm. TEM shows MWCNT with a nickel particle (arrow) on its top (Fig. 2a). The diameter of the nickel particle was about 30 nm. TEM image of MWCNTs indicates 50 to 100 carbon planes with the inter-layer spacing of 0.34 nm which corresponds to its graphite structure (Fig. 2b). MWCNTs formation is also confirmed by XRD ,with the typical peaks for MWCNTs at 26.22� and 53.97� 2 which corresponds to the (002) and (004) planes of graphite 2H. SEM, TEM and XRD results showed that biogas (CH4/CO2 mixture) can be efficiently converted to MWCNTs and syngas over starch derived Ni@C nanoparticles. References: [1] S Rasi et al, Energy, 32 (8) (2007), p. 1375. Figure 1. SEM images of the carbon-encapsulated nickel nanoparticles. a: the dried sample; b: the calcined sample; c: the catalytic conversion sample; d. purified sample. Figure 2. TEM images of MWCNTs formed during biogas conversion. a: MWCNTs with nickel nanoparticle on the top (arrow); b. the graphite plane structure of the MWCNT wall.");sQ1[914]=new Array("../7337/1831.pdf","The Biomineral-Cell Interface in the Sea Urchin Embryo","","1831 doi:10.1017/S1431927615009939 Paper No. 0914 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Biomineral-Cell Interface in the Sea Urchin Embryo Irene Y.-T. Chang1, Derk Joester1 1. Materials Science and Engineering, Northwestern University, Evanston, IL 60208, USA As man-made materials become more similar to the biological structures that inspire them, they increasingly combine nano-sized hard and soft, synthetic and biological components. This creates new challenges for characterization, especially in those materials where water is an integral part of the structure. Cryogenic sample preparation and imaging is often necessary for such specimens. Imaging of cryo-fixed, freeze-fractured samples by cryo-SEM is particularly efficient. However, the propagation of the fracture plane is unpredictable at best, and frequently the fracture surface fails to reveal the interface of interest [1]. This can be a major complicating factor, for example in the analysis of the interaction of cells and the endoskeleton in the sea urchin embryo [2]. Herein, we describe a cryogenic sample preparation workflow for cryo-planing and imaging large areas of frozen-hydrated samples, using whole sea urchin embryos as an example for a hybrid material with large hardness contrast between the organic and biomineralized tissues. The central innovation of the cryo triple ion gun milling (CryoTIGM) method is a custom-built tool based on ion mill slope cutter. Specifically, a Leica TIC3X unit was fitted with a vacuum load lock that allows cryo-transfer of a vitrified sample. Sea urchin embryo suspensions were high pressure-frozen between aluminum planchettes and trimmed using a custom-built cryo-saw. The cryo-saw consists of a liquid nitrogen reservoir, a sample compartment, a diamond blade, and a VCT-docking port (Fig. 1A). Trimming was performed under liquid nitrogen, and samples were then positioned in a sample holder next to a milling mask (Fig. 1B). The sample was transferred to the CryoTIGM tool (Fig. 1C), where three broad Ar+ beams converge at the mask shielding the trimmed sample edge (Fig. 1D). Material above the mask was removed, creating a cross-section in the sample at the level of the mask (Fig. 1E). The ion-milled sample was subsequently freeze-etched and coated with Pt to increase contrast. For whole, frozen-hydrated sea urchin embryos, we find that ion milling with Ar+ at an acceleration voltage of 3.0 kV, a current of 1.0 mA/gun, a base temperature of -120oC, and for 2 h results in very smooth cryo-planed area of ~700,000 �m2 (Fig. 1E). Sections clearly revealed cell-endoskeleton interfaces (Fig. 2A). The membranes of the syncytium that envelopes the endoskeleton appear welldefined (Fig. 2B). Numerous organelles are observed within the syncytial compartment, indicating excellent preservation of cellular ultrastructure. These results suggest that CryoTIGM is a promising new tool for interfacial studies of hybrid hard/soft materials. Given the large and smooth cryo-planed surface, microanalysis by cryo-SEM-EDS appears particularly promising. In this context, I will discuss recent attempts to identify vesicles that store and/or transport biomineral precursors in the sea urchin embryo. References: [1] Studer, Humbel and Chiquet, Histochemistry and Cell Biology 130 (2008), p. 877. [2] Vidavsky et al, Proceedings of the National Academy of Science of the United States of America 111 (2014), p. 39. [3] The authors acknowledge funding from the NSF (NSF MRI-1229693, DMR-1106208), the Northwestern University Materials Research Center (DMR-1121262). Microsc. Microanal. 21 (Suppl 3), 2015 1832 Figure 1. Experimental setup of CryoTIGM. (A) The cryo-saw used for trimming sample carriers. DS = diamond saw, LR = LN reservoir, SC = sample compartment, VL = VCT loading dock. (B) Top view of the sample compartment. A sample carrier is trimmed (along the dashed line) and transferred to the sample holder. (C) CryoTIGM tool with attached cryo/vacuum-transfer-shuttle. (D) Schematic drawing of CryoTIGM milling process. (E) A frozen hydrated sample after milling shows a triangular cryoplaned area of 700,000 �m2 (). Scale bar represents 500 �m. Figure 2. Cryo-SEM of sectioned endoskeleton of sea urchin embryos prepared by CryoTIGM. (A) Plane of intercept of the ion beams and the endoskeleton. Milling marks are visible in the embryonic matrix (m) on the two sides of the endoskeleton (s). (B) The cross-section of a syncytium enveloping the endoskeleton. Various organelles are observed within it. The terraces in the mineral region are likely milling artifacts. The inset shows an ion-milled whole embryo, whose endoskeleton is enclosed by a rectangle and shown close-up in panel A. Scale bar represents 10 �m in A and 500 nm in B.");sQ1[915]=new Array("../7337/1833.pdf","Automated cryo electron tomography and sub-tomogram averaging with the FEI","","1833 doi:10.1017/S1431927615009940 Paper No. 0915 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Automated cryo electron tomography and sub-tomogram averaging with the FEI Volta phase plate Wim J. H. Hagen1, William Wan1, John A. G. Briggs1 1. Structural and Computational Biology Unit, European Molecular Biology Laboratory, 69117 Heidelberg, Germany. Optimized automated cryo-electron tomography combined with sub-tomogram averaging currently can achieve sub-nanometer resolution using a post-column energy filter with a standard charged coupled device camera [1]. Since low frequency information is crucial for sub-tomogram alignment, we explored the use of a prototype FEI Volta phase plate [2] with our automated data collection setup to further enhance low frequency contrast. Cryo-EM data is typically collected with defocus, to be CTF-corrected post-acquisition. At subnanometer resolution, CTF-correction becomes crucial for any data-averaging scheme [3]. Postacquisition contrast transfer function correction relies on defocus determination for each individual raw image. This is challenging for close-to-focus tomographic data, and the need to also determine the phase shift of the Volta phase plate complicates things further. We explore optimization of microscope alignments and calibrations to meet the increased stability demands needed for in-focus phase plate data acquisition. Absolute focus and beam tilt calibrations on the microscope are crucial. Autofocus routines have been optimized to establish stable absolute focus, rather than stable relative focus, during automated data acquisition, as well as for determining and logging phase-shift. These routines facilitate collecting data at a predetermined focus and phase shift, as well as determining and correcting the resulting CTF for structure determination by sub-tomogram averaging. [1] Schur, F. K. M. et al, Nature, doi:10.1038/nature13838 [2] Danev, R. et al, Proceedings of the National Academy of Sciences. doi:10.1073/pnas.1418377111 [3] Schur, F. K. M. et al, Journal of Structural Biology, doi:10.1016/j.jsb.2013.10.015 Microsc. Microanal. 21 (Suppl 3), 2015 1834 Figure 1. In-focus zero-tilt energy-filtered image of Marburg virus at 200KV, 1.1 electrons per square Angstrom dose using the Volta phase plate. Figure 2. Ten slices through an in-focus tomogram of Marburg virus at 200KV, tilt range +/- 60 degrees, tilt step 3 degrees, total dose 70 electrons per square Angstrom dose using the Volta phase plate.");sQ1[916]=new Array("../7337/1835.pdf","In situ studies of cellular architecture by Electron Cryo-Tomography with Volta Phase Plate","","1835 doi:10.1017/S1431927615009952 Paper No. 0916 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In situ studies of cellular architecture by Electron Cryo-Tomography with Volta Phase Plate Yoshiyuki Fukuda1, Shoh Asano1, Ulrike Laugks1, Florian Beck1, Antje Aufderheide1, Friedrich F�rster1, Vladan Luci1, Wolfgang Baumeister1, Radostin Danev1. 1. Department of Molecular Structural Biology, Max Planck Institute of Biochemistry, Martinsried, Germany. Studies of molecular sociology of cells and in situ studies of macro molecular assemblies are critical for our understanding of cellular function. Electron cryo-tomography (ECT) of vitrified, frozen-hydrated cells provides a means of studying the three dimensional structure of pleomorphic objects, such as organelles or cells preserved in their natural, cellular environment, with a resolution of 1 to 3 nm range [1]. However, low signal-to-noise ratio in image is a drawback of ECT. In order to improve the image contrast of frozen-hydrated specimens, several kinds of phase plates have been developed [2]. Among them, Zernike type phase plate has been used for studies of frozen-hydrated biological specimens in previous studies [3, 4]. Although Zernike type phase plate is possible to enhance image contrast but it also generates fringes around structures which make interpretation of structure difficult [5]. Recently, new type of phase plate for TEM named Volta phase plate (VPP) was developed [6]. In this study, we applied VPP to ECT of frozen-hydrated cells such as magnetotactic bacteria and primary cultured neuronal cells for visualizing details of molecular architecture in situ. In order to investigate the performance of the VPP for thick specimens we observed plunge-frozen magnetotactic bacteria. Four tomographic tilt-series were acquired with different combinations of an energy filter and a VPP. Energy filtered images show stronger contrast for large objects, such as the poly--hydroxybutyrate (PHB) granules, and less background noise in the thicker parts of the sample than the non-filtered images. In comparison, the VPP improves the contrast of all visible features. As a next step, we compared the quality of reconstructed tomograms. Similarly to the projection images, these slice images show that the energy filter improves the contrast of large objects, such as the PHBs. The contrast of inner and outer membranes was also slightly improved by the energy filtering. Surprisingly, the effect of zero-loss filtering on the overall contrast in reconstructed tomograms is diminished compared to that in the tilt-series images. Comparing the CTEM slices with the VPP ones the visibility and contrast of many intracellular structures and macromolecules was significantly improved by the VPP. For example, the chemoreceptor arrays are easier to detect in both the unfiltered and filtered VPP tomograms than in the energy filtered CTEM tomogram. Therefore, the VPP produces a significant increase in contrast, in addition to that provided by energy filtering. In order to quantitatively evaluate the contrast of the inner and outer membranes for the different acquisition conditions we calculated intensity profiles across the membranes. To make the comparison as unbiased as practically possible, considering the tomograms were acquired at different specimen areas, the profiles were calculated at a position which showed approximately the same cell. The energy filter provided ~16% contrast increase without a VPP and ~22% with a VPP. On the other hand the VPP provided ~61% contrast improvement without and ~68% with an energy filter. Microsc. Microanal. 21 (Suppl 3), 2015 1836 Due to the improvement of visibility provided by VPP, 26S proteasome particles could be visualized in ECT of primary cultured neuronal cells. For the in situ studies of macro molecular structure, we attempted template matching using a 3D model of single-capped 26S proteasome and sub-tomogram averaging of matched particles. Two of major conformational states "ground state" and "substrateprocessing state" are obtained by classification. Thus, VPP provides ECT of frozen-hydrated specimens a significant improvement of contrast for visualization of cellular architecture and macro molecular structure in situ. We expect that VPP will become a standard element of the electron cryo-tomography workflow. References: [1] V Luci, A Rigort and W Baumeister, J. Cell Biol., 202 (2013), p. 407. [2] R M Glaeser, Rev. Sci. Instrum., 84 (2013), 111101. [3] R Danev, R M Glaeser, K Nagayama, Ultramicroscopy, 109 (2009), p. 312. [4] W Dai, et al, Nature, 502 (2013), p. 707. [5] R Danev and K Nagayama, Ultramicroscopy, 111 (2011), p. 1305. [6] R Danev et al, Proc. Natl. Acad. Sci. USA., 111 (2014), p. 15635. Figure 1. Comparison of projection images of plunge-frozen Magnetotactic bacteria between CTEM with energy filter and VPP without energy filter. Scale bars: 500 nm. Figure 2. Comparison of tomographic slice images of plunge-frozen Magnetotactic bacteria between CTEM with energy filter and VPP without energy filter. Scale bars: 500 nm.");sQ1[917]=new Array("../7337/1837.pdf","Applications and New Investigations of the Volta Phase Plate","","1837 doi:10.1017/S1431927615009964 Paper No. 0917 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Applications and New Investigations of the Volta Phase Plate Kasim Sader1,Bart Buijsse1, Ilaria Peschiera2,Ilaria Ferlenghi2 1. 2. FEI, Achtseweg Noord 5, 5651 GG Eindhoven, The Netherlands Novartis Vaccines, Research Centre, Structural Microscopy, Via Fiorentina 1, 53100 Siena, Italy The use of the Volta Phase Plate for cryoTEM has increased significantly in the last year. Recently two important papers described the novelty of the technique [1] and highlighted the importance of the technique in making new biological discoveries [2]. Here we present our work on both new applications of the Volta Phase Plate to improve imaging of small protein complexes, and new explorations to enhance the information transfer of low coherency sources through the ability to work in focus. The use of Phase Plates improves the low resolution information content in cryoTEM images, thus enabling the 3D reconstruction of small protein complexes through improved alignment of single particles, similar and synergistic to the improvements obtained by using a counting direct electron detector [3]. The technique has been successfully used to image the NadAV3 (Neisserial adhesion variant 3) homo-trimeric protein and the complex formed by binding to one of its neutralizing antibodies (Figure 1). The NadA protein is one of the components of the recently approved Meningitis B vaccine 4CMenB [4], with the variant 3 being the most important out of all the strains worldwide because it causes the strongest immune response. While the structure of a similar homotrimer from a different variant (NadAV5) has been obtained by X-ray crystallography [4], the structure of NadAV3 has never been determined due to its high flexibility. To increase its low molecular weight (105kDa) the homotrimer has been complexed with Fabs binding to the head region. We used a Titan Krios (300kV) with an FEI Falcon 2/3, Volta Phase Plate, and high dose, and find the contrast to be better than most negative stain images obtained (Figure 1). The structure appears as an elongated rod with one end possibly decorated by 2/3 copies of Fabs. Optimization of the freezing procedure by pre-binding the complex to the carbon, followed by washing and addition of new sample is ongoing. As an application on the technical side, we have been exploring the use of the Volta Phase Plate to overcome poorly coherent sources (ie. LaB6). Much of the early high resolution unstained biological TEM work [5] was performed with an electron source other than a field emission gun (FEG). This was primarily applied to 2D and helical arrays, as the orientation of the individual subunits of the arrays did not need to be determined relative to one another, and therefore images could be taken very close to focus. This avoided the damping of the CTF due to defocus and the lower spatial coherence of the electron beam. For single particle work, where low resolution information is required for determining orientations of each particle, a substantial defocus (-2�m) is needed to generate the required low resolution frequencies. This normally necessitates a highly coherent electron beam to avoid damping of the CTF, and therefore a FEG. As Phase Plates allow operation in focus, the damping of the CTF is avoided and we are therefore exploring the use of the Volta Phase Plate for single particle analysis (Figure 2). The main challenge is that the on-plane beam covers a large area on the phase plate due to the lower coherency, and it will be more challenging to maintain a constant phase shift close to /2 for the duration of the experiment. In summary, we describe two important aspects of the use of the Volta Phase Plate: firstly the visualization of small proteins, and in particular antibody-antigen complexes, to aid in the screening of Microsc. Microanal. 21 (Suppl 3), 2015 1838 possible vaccine candidates. The Volta Phase Plate, together with improvements in direct electron detectors and increased doses, provides a significant improvement in the low resolution signal. Secondly we present our exploration of a method to enable the use of lower coherency sources combined with the Volta Phase Plate for single particle (or tomography) work at medium resolution. References: [1] Danev, R. et al. Proceedings of the National Academy of Sciences 111 (2014), 15635-15640. [2] Asano, S. et al. Science 347 (2015), 439-442. [3] Lu, P. et al. Nature 512 (2014), 166-170 [4] Malito, E. et al. Proceedings of the National Academy of Sciences 111 (2014), 17128-17133 [5] Henderson, R., Unwin, P.N.T., Nature 257(1975), 28-32. Figure 1. 300kV Krios, Falcon 2, ~100 e-/�2, Volta Phase Plate images of the NadAV3-Fab complex with selected molecules magnified. Figure 2. Frequency dependent signal-to-noise ratios derived from experimental data on amorphous carbon images from a LaB6 T20 (30 �m C2), ~10 e-/�2/s with and without a Volta Phase Plate (PP).");sQ1[918]=new Array("../7337/1839.pdf","A Study of Gallium FIB induced Silicon Amorphization using TEM, APT and BCA Simulation","","1839 doi:10.1017/S1431927615009976 Paper No. 0918 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A Study of Gallium FIB induced Silicon Amorphization using TEM, APT and BCA Simulation Jin Huang1, Markus Loeffler1, Uwe Muehle2, Wolfhard Moeller3, Hans Mulders4, Laurens Kwakman4 and Ehrenfried Zschech1, 2 1. 2. Dresden Center for Nanoanalysis, Technische Universitaet Dresden, Dresden, Germany. Fraunhofer Institute for Ceramic Technologies and Systems, Dresden, Germany. 3. Helmholtz-Zentrum Dresden-Rossendorf, Dresden, Germany. 4. FEI Company, Eindhoven, the Netherlands. Crystalline silicon (c-Si) is partially amorphized in Focused Ion Beam (FIB) TEM lamella preparation. A 30kV Ga+ beam with a small glancing incident results in a 20-30nm amorphous layer [1, 2]. The precise mechanisms remain uncertain and a damage prediction can hardly be made. In this study, a Binary Collision Approximation (BCA) software is employed to simulate c-Si amorphization in various Ga+-FIB conditions. These results are compared with experimental data from transmission electron microscopy (TEM) and atom probe tomography (APT) of Si samples prepared by a FEI Helios 660 FIB/SEM system. BCA simulation is widely used for ion-solid interactions, however, simulation of the FIB processes requires taking dynamic phenomena into account. The TRIDYN code treats the deceleration of ions in solids and the associated formation of recoil atom cascades in the BCA model [3]. In contrast to the TRIM code [4], it considers dynamic alterations of the local composition which arise from the ion implantation process. In this way, phenomena such as sputter erosion, atomic mixing and recoil implantation can be simulated with improved accuracy. Due to the limitation of the BCA model, TRIDYN is unable to simulate any crystalline-to-amorphous transition. It was demonstrated by Monte Carlo simulation and experimental studies that Si amorphization can be described by average displacement of host-atoms, where the critical displacement is 5� [5]. It was also shown that, in terms of atom displacement, the Molecular Dynamics method is more accurate, but does not have a significant difference to BCA modeling for >5� atom displacement [6]. In this context, TRIDYN can be employed for the simulation of Si amorphization during FIB under the mentioned assumptions. Figure 1 shows the amorphous layer thickness as a function of the Ga+ ion energy at 3 different ion impingement angles: 3�, 5� and 7�. These curves behave differently in lower and higher Ga+ energy ranges, which may indicate that the mechanisms are different. TEM images of c-Si cross-sections after Ga+ FIB preparation at 5 different ion energies are shown in Figure 2(a)-(e). C-Si samples are prepared as wedge lamellae using a standard 30kV Ga+ FIB preparation method. Subsequently, a Ga+ beam of 0.5, 1, 2, 5 and 30kV, with a 5� glancing angle is applied to the side of the wedge. Typically, when removing material, the energy-dependent damage and implantation layer that extends into the sample as material is removed. The experiment was designed in such a way that sufficient amorphous and crystalline material is removed to reach a steady state and to retain only an amorphous layer of this energy-dependent thickness. For comparison, the measured values are plotted in Figure 1. All experimental data correlate well with the model-based predictions, showing that TRIDYN is a valid and well adapted model to estimate amorphous layer thicknesses. Implanted Ga ion distributions have been modelled with TRIDYN and experimentally measured with APT as well: for this, c-Si samples are prepared with a specific cone geometry for APT measurements. In the last step of the preparation, a 5 kV Ga+ beam is applied collinearly to the tip direction, which results in a final cone angle of ~22�. The Ga concentration along a radius on the axial cross-section is plotted in Figure 3. Note that the strong scatter Microsc. Microanal. 21 (Suppl 3), 2015 1840 at the shallow surface is caused by the insufficient number of detected atoms. A TRIDYN simulation of the Ga distribution with the same beam conditions is also shown for comparison. Although the exact relationship between Ga concentration and Si amorphization is not yet established, the highly matched data suggests TRIDYN can be considered as a reliable reference in future studies. It is concluded that the FIB milling process induced damage and implant can be well described with the TRIDYN simulation. Optimized recipes for ultra-thin lamella preparation can be now engineered using this simulation model. For such critical lamellae, multiple ion energies have to be used, each for an optimized time to achieve the final lamella thickness in shortest time and with the minimum damage. References: [1] HJ Engelmann et al, Microscopy Today 11 (2003), 22-24. [2] LA Giannuzzi, R Geurts and J Ringnalda, Microsc Microanal 11 (2005), 828-829. [3] W Moeller, W Eckstein and JP Biersack, Computer Physics Communications 51 (1988), 355-368. [4] JP Biersack and LG Haggmark, Nuclear Instruments and Methods 174 (1980), 257-269. [5] LJ Gracia et at, Journal of Applied Physics 109 (2011), 123507. [6] L Bukonte et al, Nuclear Instruments and Methods in Physics Research B 297 (2013), 23-28. [7] The authors acknowledge the project funding from FEI and Makizu project funding from BMBF. Figure 1. TRIDYN Simulated amorphous Si layer thickness as a function of ion energy for 3�, 5�and 7� ion impingement angle. Experimental TEM data are also depicted for comparison. (a) (b) (c) (d) (e) Figure 2. TEM images of a c-Si cross-section after a 0.5/1/2/5/30kV Ga FIB process show amorphous layer thicknesses of 1.5/2.3/3.1/5.0/30.5nm, from (a) to (e) respectively. Figure 3. APT determined Ga concentration profiles from three APT samples, with corresponding TRIDYN simulation (left); illustration of APT measurement (Ga: black and Si: red) (right).");sQ1[919]=new Array("../7337/1841.pdf","Probe Optimization studies For High current Focused Ion Beam Instruments","","1841 doi:10.1017/S1431927615009988 Paper No. 0919 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Probe Optimization studies For High current Focused Ion Beam Instruments Srinivas Subramaniam1 , John Richards1 and Kevin Johnson1 1 Intel Corporation, Hillsboro, USA. Xenon plasma focused ion beam (PFIB) instruments can operate in beam currents ranging from ~20pA at the low end to ~2�A at the high end of the current spectrum allowing for versatile use of the probe in imaging and milling applications. This is made possible by use of current limiting apertures of varying diameters resulting in a wide range of probe diameters and profiles. At the high current range (>300nA), probe profiles exhibit large probe diameters and extended beam tails. The tail regions representing a small fraction of the primary probe current can still have significant effects in causing over-milling and degradation of the samples during milling and trenching operations. Focused probe conditions at high currents can exhibit significant tails with gross degradation of the sample during milling, while defocused probe conditions show improved milling characteristics with reduced over mill beyond the target raster region [1]. Quantitative approaches to characterizing the probe profiles at high current settings can provide valuable insight in optimizing beam conditions. Previously researchers have used theoretical and experimental studies characterizing gallium (Ga) ion beams, employing a variety of methodologies in probe current distributions and modeling the primary and secondary profiles of the ion beam [2,3]. In this study we have used a combination of techniques in characterizing the Xenon (Xe) ion probe for high current applications. The 1�A beam current setting on a Tescan FERA Xe PFIB was used to explore means of optimizing the probe for milling applications in the high current regime. In addition to empirical work done on characterizing probe conditions and performing milling comparisons using the different parameters we have also used AFM profilometry to measure the profile of a high current Xe ion beam to analyze the current distribution and profiles of the ion beam under varying conditions. A new aperture with standard alignments and centering was used for the experiments. Beam burns were performed over a range of over-focus, focus and under-focus conditions. A mild over-focus condition of 17.10KV was used for comparisons with the standard focus condition at 17.26KV. Top down SEM measurements of the probe profiles and beam tail extension were complemented with AFM profiles of the probes to quantify the Z depth corresponding to the amount of material removed in the tail. Probe profile measurements were performed documenting the changes in probe profiles and extent of beam tail spread under various focus conditions. Figures 1(top and bottom) shows SEM and AFM profilometry measurements correlated to the probe burns at the over-focus and focus conditions respectively. Figure 1B shows the extent of material removal and tapering in the tail region of the probe which can be clearly seen with the extended tail of the focus condition resulting in a 45�m radius compared to only 26�m for the over-focus condition shown in Figure 1A. Comparisons of FIB milled regions using over-focus and focus settings confirm these results with the focus condition (Figure 2B) showing significant over mill with much of the forward region of the sample being already milled away. In comparison, in the over focus condition (Figure 2A), the over mill appears to be in good control and should be negated once lower current final polishing of the sample has been performed. The authors would like to acknowledge Liza L. Ross with Materials Analysis Labs at Intel for assistance with AFM measurements. References [1] Srinivas Subramaniam and Kevin Johnson, Microsc. And Microanal. 20(S3) (2014) 296. [2] Shida Tan, Richard Livengood, Yuval Greenzweig, Yariv Drezner and Darryl Shima, J.Vac.Sci.Technol. B B 30, (2012) 06F606-1. Microsc. Microanal. 21 (Suppl 3), 2015 1842 [3] G.Ben Assayag, C.Vieu, J. Gieerak, P. Sudraud and A. Corbin, J.Vac. Sci.Technol. B11(6), (1993) 2420. Figure 1A Figure 1B Over- Focus Focus R1=26 �m R2=12 �m R1=45�m R2=6�m Over-focus Focus Figure 1A Top: AFM profile on top shows primary probe profile and extent of beam tail for under-focus condition. Bottom: SEM measurement of 1�A probe with 17.10KV objective voltage over focus setting. Primary probe radius measures 12�m with beam tail extending out to 26�m. Figure 1B Top: AFM profile on top shows primary probe profile and extent of beam tail for focus condition. Bottom: SEM measurement of 1�A probe with 17.25KV objective voltage setting. Primary probe radius is smaller and measures 6�m but the beam tail is significantly worse extending out to 45�m. Figure 2A Figure 2B Figure 2: Shows comparison of locations milled using 1umA beam current at the focus setting of 17.25KV (2A left image) versus over-focus setting of 17.10KV (2B right image). Figure 2A shows the effect of extended beam tail with excessive material removal in the advance regions showing as compared with the more benign over focus setting.");sQ1[920]=new Array("../7337/1843.pdf","Methodology for Studying Nanoscale Details of Focused Ion Beam Gas-Assisted Etching and Deposition by TEM and Numerical Modeling","","1843 doi:10.1017/S143192761500999X Paper No. 0920 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Methodology for Studying Nanoscale Details of Focused Ion Beam Gas-Assisted Etching and Deposition by TEM and Numerical Modeling. Valery Ray1,4, Eddie Chang2, Kevin Toula3, Sz-Chian Liou1, and Wen-An Chiou1 1. 2. NISP Lab, NanoCenter, University of Maryland, College Park, MD 20742, USA Department of Physics, University of Maryland, College Park, MD 20742, USA 3. Department of Material Science and Engineering, University of Maryland, College Park, MD USA 4. PBS&T, MEO Engineering Co., Inc. Methuen, MA 01844, USA Understanding nanoscale details of Focused Ion Beam (FIB) sputtering, Gas-Assisted Etching (GAE), and beam-induced deposition is critical for a wide range of applications, from patterning thin films to prototyping integrated circuits. Most of research in the field of beam-induced chemistry was historically based on macroscale experiments and observations [1]. While recently developed methodology for investigating details of FIB GAE within a single beam profile [2] relies on analysis of primary ion beam current density profile using multiple HRTEM lattice images, it is difficult for practical routine when resources are limited. This study presents a simplified approach for first-approximation reconstruction of the ion beam characteristics from information available on a single TEM micrograph, and describing nanoscale details of FIB sputtering, Ga implantation, and beam-induced deposition observed during analysis of single-line etching and deposition experiments. Etching and deposition was carried out in a 30kV Helios 660 FIB/SEM on fused silica (SiO2) substrate coated by ~24nm of evaporated Al to prevent charging. Three series of straight lines were made with identical ion beam parameters: first etched by physical sputtering, second etched with XeF2 precursor, and third deposited from W precursor. Ion dose for lines within each series was set to 0.2nC/�m2, 0.6nC/�m2, and 1.8nC/�m2 and repeated for ion beam currents of 1pA, 7.7pA, and 24pA with dwell of 25nSec and overlap of -20%. Lines created by etching and deposition were protected by a layer of ebeam deposited Pt, followed by FIB Pt deposition. TEM lamella positioned across etched lines was prepared by FIB, and FIB cross-section was made across the lines deposited from W precursor. TEM micrographs of single-line etching profiles were acquired on a JEOL 2100 LaB6 TEM and revealed nanoscale details of sputtering (Fig. 1). Ion implantation into SiO2 substrate and agglomeration of the implanted atoms with formation of distinct variable-density zones can be observed within the implanted layer. Lines etched with GAE exhibited dramatic reduction of the apparent implantation and enhanced etching in peripheral areas. Cross-sectional SEM imaging of single-line deposition series not only reveals the fine details of FIB deposition process, but also allows direct observation of Ga damage under the deposited layer. Observations of cross-section images suggest that apparent damage of the substrate material by primary ion beam may continue even after the initial deposition layer has formed. Further study of FIB deposition dynamics using TEM is needed to investigate possible defects and migration of substrate material during the deposition. Effective diameter and current density distribution within focused ion beam are important characteristics for evaluating beam-induced processes. This information, however, is scarcely available from OEMs of FIB equipment; a practical method for deducing beam characteristics is needed. For practical purposes effective beam diameter is assumed equivalent to the largest width of Ga damage profile measured on corresponding TEM micrographs. To simplify beam current profile, the low-current-density linear beam Microsc. Microanal. 21 (Suppl 3), 2015 1844 tails were disregarded and bi-Gaussian beam containing wide beam tails with low current density, and a main peak with relatively high current density was modelled. Calculation of the substrate material sputtering assumed that etching profiles with aspect ratio of depth to width 3:1 have linear dependency of depth on the ion dose. To concurrently simulate implantation and low-aspect-ratio sputtering MATLAB scripts were developed and integrated with SRIM Supporting Software Module (SSSM) and Transport of Ions in Matter (TRIM) packages. Additional MATLAB scripts were created to visualize simulated implantation and etching for comparison with experimental profiles (Fig. 2). Series of beam profile simulations with varied parameters were carried out, and characteristics resulting in simulated profiles with the closest match to experiment were assumed to describe the primary beam. The results showed that implantation of Ga is lowest and efficiency of FIB GAE is highest in the low-current-density areas of the beam. This in turn suggests that ion beams with current density that is significantly lower than the current Ga FIB technology would be optimal for GAE processes. References: [1] Ivo Utke et al, JVST B 26(4) (2008) p. 1197. [2] Shida Tan et al, JVST B 30(6) (2012) p. 06F606. [3] Research was partially supported by NSF-MRSEC (DMR 05-20471) and UMD. Oleg Sidorov of FEI contributed with outstanding execution of single-line etching and deposition design of experiment. Figure 1. TEM micrographs show profile of physical sputtering and GAE by XeF2 using 7.7pA 30kV Ga+ FIB. Cross-sectional SEM image depicts W deposition with the same primary ion beam. Figure 2. Simulated implantation and sputtering profile compared with matched to scale TEM micrograph, and plot of etching efficiency as function of current density for 7.7pA 30kV Ga+ FIB");sQ1[921]=new Array("../7337/1845.pdf","Using 3D Nanotomography to Visualize Defects in the Fabrication of Superconducting Electronics.","","1845 doi:10.1017/S1431927615010004 Paper No. 0921 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Using 3D Nanotomography to Visualize Defects in the Fabrication of Superconducting Electronics. Aric W. Sanders1, Anna E. Fox1 and Paul D. Dresselhaus1 1. National Institute of Standards and Technology, Boulder Colorado, USA. Superconducting electronics is an established technological field for sensors, quantum computation and quantum-based standards and is emerging as an important low-power alternative to semiconductors. As with any electronics fabrication, the production of complex circuits requires many iterations and stringent quality control. In particular, post-fabrication metrology of buried circuit elements to diagnose device failures is of utmost importance to increase the yield [1] and complexity of superconducting based electronics. To fill this need, we have developed methods for the 3D visualization of superconducting electronic components based on focused ion beam � scanning electron (FIB-SEM) nanotomography. Unlike in the semiconductor industry, which has long used FIB-SEM methods to investigate and edit circuits [2], superconducting electronics have not benefited from this technique. Since the basic structure of superconducting electronics is different than that of semiconducting electronics, device specific optimization has been required [3]. This optimization has led to the ability to produce tomograms of buried features at high speeds and visualize defects that are common in the fabrication of superconducting electronic devices. To build these 3D tomograms, device structures of interest are selected and then milled at the highest rate that provides the needed resolution. As these structures are milled, SEM images are acquired at regular intervals. The resulting images are then corrected for drift, thresholded by relative brightness for the material of interest, and selectively cropped to produce a series of aligned images of the region and material of interest. This series of images is then viewed using the appropriate 3D visualization software [4]. In this way we create a 3D representation of fabricated structures that are used to explore both functioning and faulty devices allowing feedback to improve the fabrication of superconducting electronics. Variations in the development and treatment of photoresist based etch masks can cause important variations in the geometry of devices. In Figure 1 A, the presence of residual resist has caused a sloped etch profile in the device layer. As with any multi-level electronics microfabrication process, each layer has the potential to introduce defects and reduced device yield in a multitude of ways. The layer to layer alignment or registration is a critical parameter. Misalignments of layers can cause shorts, improper etch masking and other device failures. In figure 1 B, a 3D rendering of a short is shown, along with a typical 2D electron micrograph. This short is caused by a registration error and effectively bypasses the circuit's active element, a Josephson junction. If the material coverage and edge profiles of each layer are not optimal underlying layers can be attacked by subsequent etching steps. In figure 2, the first and last image frames of a tomographic series are shown. In this series, extended defects at the interface of Nb, SiO2 and Si are highlighted in 3D reconstructions by isolating the material of interest through brightness and repeatedly reconstructing the data to show the effect of the defects on different layers. In summary, we will present the methods and results of using FIB nanotomography in superconducting electronics for post fabrication 3D visualization. Microsc. Microanal. 21 (Suppl 3), 2015 1846 References: [1] AE Fox, PD Dresselhaus, A Rufenacht, A Sanders, SP Benz, IEEE Transactions on Applied Superconductivity, 99 (2014), DOI 10.1109 [2] Steve, R. and P. Robert, Journal of Micromechanics and Microengineering 11 (2001), p 287 [3] A Sanders, A Fox, P Dresselhaus and A Curtin, Microscopy and Microanalysis 20 (2014), p 772773. [4] Schmid B, Schindelin J, Cardona A, Longair M, Heisenberg M, BMC Bioinformatics 11 (2010), p 274. Figure 1. Sloped etch profile and short visualized by FIB nanotomography A) A raw slice illustrating a sloped device wall resulting from photoresist etch mask residue (red arrow), and the 3D tomogram (top) showing that it propagates along the device. B) A raw slice with an area circled for reference and the 3D tomogram (top) showing a buried short (red arrow). Figure 2. Extended defects at complex interfaces highlighted by brightness based material selection. Left, the first and last raw image processed to create the 3D tomogram. A void left by etching at the interface of a metal device layer, an oxide layer and the Si substrate (red arrow) is seen to extend into the device structure (top right). A similar etch defect is seen on the interface of only the oxide and metal (blue arrows) as shown in the 3D tomogram on the bottom right.");sQ1[922]=new Array("../7337/1847.pdf","In situ Femtosecond Laser and Argon Ion Beams for 3D Microanalysis using the TriBeam","","1847 doi:10.1017/S1431927615010016 Paper No. 0922 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In situ Femtosecond Laser and Argon Ion Beams for 3D Microanalysis using the TriBeam M.P. Echlin1, W.C. Lenthe1, J. Douglas1, M. Titus1, R. Guerts2, M. Straw2, T.M. Pollock1 1 2 Materials Department, University California at Santa Barbara, Santa Barbara, USA FEI Company Advanced engineering materials require microstructural characterization in 3D across lengthscales, motivating the development of new tomography techniques and coupling with existing capabilities. The acquisition of 3D datasets with structural and chemical information at lengthscales between those accessible using Ga+ and Xeon FIB SEMs and those of X-ray tomography techniques is still challenging, particularly for dense multiphase materials. Femtosecond lasers have been employed for low damage [1,2] material removal, in tomography applications, over mm3 regions in situ in a FIB SEM as shown in Figure 1. FIB cross sections investigated by TEM have shown that dislocations can be injected to microns in depth in some materials [3], but are primarily confined to less than 100s of nanometers of the surface in the low fluence ablation regime. Parametric studies of laser fluence and beam scanning conditions in silicon in the TriBeam show that, when the propagating laser beam is scanned parallel with the sample surface, the damage is exclusively limited to that of the low fluence ablation regime. Laser ablation studies have also shown the ability to resolve surface sensitive EBSD maps from the ablated surface of many metals and/or alloys primarily containing magnesium, titanium, nickel, steel, copper, tungsten, tin and niobium. Recently, a broad beam argon ion source capable of microamp currents has been integrated into the TriBeam for the further reduction in surface damage by glancing angle milling. Previous in situ Ar+ ion beam experiments show ion induced amorphization in Si can be reduced to 10s of nm at low accelerating voltages [4]. Femtosecond laser ablated silicon samples were investigated, as well as samples that had been subsequently Ga+ ion milled, as shown in Figure 1 (right). TEM lamellae were extracted to determine the changes in surface structure and amorphization depths as a function of types of beam exposure. A brightfield TEM image in Figure 2 demonstrates amorphization depths of 20 nm as resulting from laser machining with fluences at 20x the ablation threshold, which were scanned in a surface parallel to the beam orientation. Currently an optimal balance for microanalaytical 3D dataset acquisition (EBSD and EDS), dictated by detector speed and sensitivity, must be struck between in plane resolution and experiment duration. Enhancements in 3D EBSD collection speed and pattern quality will be discussed for femtosecond laser ablated and Ar+ ion cleaned surfaces in the context of multimodal data collection of NiTiSn alloys for thermoelectric applications. Microsc. Microanal. 21 (Suppl 3), 2015 1848 References: [1] S. Ma et al, Met. Mat. Trans. A 38 (2007) p. 2349-2357. [2] Q. Feng et al, Scripta Mat. 53(5) (2005) p. 511-516. [3] M.P. Echlin et al, Materials Characterization: Tutorial Review 100 (2015) p. 1-12. [4] L.A. Giannuzzi et al, Microscopy and Microanalysis 11 (2005) p. 828-829. Figure 1. (left) The TriBeam tomography FIB-SEM vacuum chamber interior. (right) A silicon sample that was femtosecond laser ablated and then Ga+ ion milled at glancing angle (top half of micrograph). The location of a TEM lamella liftout is visible as a raised platinum deposited area. Figure 2. A brightfield TEM image of a lamella extracted from a silicon sample that was femtosecond laser machined at a glancing angle showing approximately 20nm of amorphization depth.");sQ1[923]=new Array("../7337/1849.pdf","Using Ultrafast X-ray Lasers to Image Structure & Dynamics","","1849 doi:10.1017/S1431927615010028 Paper No. 0923 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Using Ultrafast X-ray Lasers to Image Structure & Dynamics S�bastien Boutet1 1. Linac Coherent Light Source, SLAC National Accelerator Laboratory, Menlo Park, CA, USA X-ray lasers have recently become a reality, with the Linac Coherent Light Source (LCLS) becoming the first ever X-ray Free Electron Laser (XFEL) in operation. The capabilities of XFELs are nothing short of revolutionary, with the production of coherent pulses of x-rays on the few femtosecond to few tens of femtosecond pulse duration regime, at wavelengths capable of achieving atomic resolution. These capabilities are opening the door to access new regimes in structural and dynamics studies with unprecedented simultaneous high temporal and spatial resolution. Biological samples are known to be very sensitive to absorbed radiation which causes structural changes or damage to the sample during the exposure required to generate observable quantities about the sample. X-rays are typically absorbed more than they are scattered by roughly one order of magnitude. This implies that for techniques relying on the scattering of x-rays such as many imaging techniques and crystal diffraction, the sample is inevitably damaged by the deposition of energy during the time needed for accumulation of the signal. However, with ultrashort pulses of x-ray, it is in principle possible to overcome the radiation damage limitations caused by long, integrating measurements. If the pulses are short enough to allow all incident x-rays to pass through the sample faster than the response of the atoms to the absorbed energy, it is possible to obtain structural information free of damage [1]. This technique is now known as "diffraction-then-destruction", where significantly larger doses can be absorbed by the sample than what is possible with continuous sources of x-rays, while still producing interpretable signal, due to the fact that the inertia in the sample causes the radiation induced changes in the sample to propagate slower than the pulse duration. These assumptions and damage models can now be tested at LCLS. The diffraction-before-destruction technique using x-rays has been demonstrated to be a valuable tool to study a variety of samples using a few different techniques, including crystallography [2-4], diffractive imaging [5] and various spectroscopic techniques [6]. Serial Femtosecond Crystallography (SFX) [2], a new technique which uses a constant flow of small protein crystals inside a liquid jet, has been demonstrated to yield damage-free high-resolution structures. It has since been utilized to study increasingly challenging samples that have escaped structural determination using other techniques. New biological information can now be obtained using this technique [3, 4]. The typical experimental geometry is shown on Figure 1. Snapshots of the samples are taken with the short x-ray pulses without any sample motion during the pulse. This allows the study of dynamics if a structural change can be initiated in the sample using some trigger with a tunable delay. The addition of a laser illuminating the sample prior to the arrival of x-rays can be used as one such trigger and makes accessible ultrafast dynamics. The technique of time-resolved Serial Femtosecond Crystallography has been demonstrated [7] and opens the door to making molecular movies of important reactions such as those involved in photosynthesis. Coherent Diffractive Imaging (CDI) is a technique which measures the continuous diffraction patterns Microsc. Microanal. 21 (Suppl 3), 2015 1850 of single objects and utilizes computational phase retrieval methods to recover an image of the sample. The technique has been under development for about 15 years and holds the possibility of studying single particles at high resolution without the need to crystallize them when used in combination with an X-ray Free Electron Laser like LCLS. Single particle imaging has been demonstrated with limited spatial resolution [5] and still requires development work to become a useful tool for scientific discovery but if successful, would open the doors to unique possibilities to study structural dynamics of single molecules. Overall, structural and dynamics studies enabled by LCLS are numerous in a verity of fields. Structural biology has been an early area of broad activity and exciting results have been and will continue to be obtained. References: [1] Neutze, R., et al., Nature 406 (2000), p. 752-757. [2] Boutet, S., et al., Science 337 (2012), p. 362-364. [3] Liu, W., et al., Science 342 (2013), p. 1521-1524. [4] Redecke, L., et al., Science 339 (2013), p. 227-230. [5] Seibert, M.M., et al., Nature 470 (2011), p. 78-U86. [6] Alonso-Mori, R., et al., PNAS 109 (2012), p. 19103-19107. [7] Tenboer, J., et al., Science 346 (2014), p. 1242-1246. Figure 1. Experimental geometry for serial femtosecond crystallography (SFX).");sQ1[924]=new Array("../7337/1851.pdf","Studying Shocked Material Dynamics with Ultrafast X-rays","","1851 doi:10.1017/S143192761501003X Paper No. 0924 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Studying Shocked Material Dynamics with Ultrafast X-rays R.L. Sandberg,1 C. Bolme,1 K. Ramos,1 Q.McCulloch,1 R. Martinez,1 V. Hamilton,1 T. Pierce,1 M. Greenfield,1 S.McGrane,1 J.L. Barber,1 B. Abbey,2 A. Schropp,3 F. Seiboth,3 P. Heiman,4 B. Nagler,4 E. Galtier,4 E. Granados4 1. Los Alamos National Laboratory, PO Box 1663, Los Alamos, NM 87545 Department of Physics, University of La Trobe, Rm 419, PS1, Melbourne, Victoria, Australia 3. Institute of Structural Physics, Technische Universitat Dresden, D-01062 Dresden, Germany 4. SLAC National Accelerator Laboratory, 2575 Sand Hill Road, Menlo Park, CA 94025, USA Author e-mail address: sandberg@lanl.gov 2. The response of micron-scale inhomogeneities dictates the overall dynamic, structural and chemical response of many materials. Of particular interest is the response of micron scale voids. It is believed that such micron scale voids are responsible for the nucleation of damage leading to structural failure in metals and to initiation of detonation in explosive material under high strain-rates. A critical step towards developing safer, stronger, and longer lasting materials in a range of applications from energy to defense requires understanding the dynamic response of these inhomogeneties on the micron-scale. X-rays are particularly well suited for micron-scale materials studies due to their short wavelength and ability to penetrate bulk materials. Combining this ability with the intense ultrafast pulses from X-ray free electron lasers such as SLAC's Linac Coherent Light Source (LCLS) provides an ideal system for studying material response under extreme conditions such as impact, high load, or other non-equilibrium processes. Here we demonstrate this groundbreaking ability by showing sub-micron single-shot X-ray imaging of laser-shocked materials performed at the Materials in Extreme Conditions (MEC) hutch at the LCLS in January 2014. We imaged shock wave interactions with 10-micron voids in two single crystal materials: lithium floride (LiF) and the explosive pentaerythritol tetranitrate (PETN) with near 100 nm resolution. Shock wave interactions with voids in explosive are of particular interest due to the prevailing theory that detonation initiation is caused by void collapse. Fig. 1a shows the experimental setup at the MEC hutch for these studies. The LCLS beam was tuned to 6 keV and focused through a set of beryllium compound refractive lenses (CRLs)[1] to focus the beam down to near 100 nm as verified through ptychographic imaging of the focused beam as demonstrated previously [2,3]. The ~150 micron thick single crystal samples (either LiF or PETN) were placed at a variable distance (about 5 cm) behind the focus which allowed for an adjustable spot size on the sample (2-100 microns) depending on the focus to sample distance. Several detectors were used in these experiments to capture either the low angle coherent scattering (coherent diffractive imaging or Gabor holography) or the higher angle Bragg X-ray diffraction (XRD). For the low angle scatter, several detectors were available in order to optimize dynamic range, sensitivity, and pixel size including an Andor direct detection CCD, an FLI phosphor coupled CCD, and a 1x2 CS-PAD. These detectors were placed 4.1 m behind the samples. For the higher angle XRD a 4x4 array of CS-PAD detectors were used. Due to the brilliant (1012 photons) coherent pulses with 50 fs pulse duration, the laser shock dynamics were captured with no blur in a single pulse. The shock was driven perpendicular to the XFEL beam by focusing down the two arms of the 10 Hz glass drive laser (up to 35 J per pulse at 532 nm) into a thin ablation layer consisting of a polyimide (25 microns) covered with ~100 nm of aluminum. Shock pressures ranged from a few GPA up to 10 GPA as measured with line resolved laser Microsc. Microanal. 21 (Suppl 3), 2015 1852 velocity interferometer system for any reflector (VISAR). A silicon channel cut monochromator was also used on some shots to narrow the XFEL bandwidth. Fig1 b and c show respectively a Gabor hologram of a collapsing void in PETN and the simultaneous dynamic XRD. These images show the two-wave structure of the shock (plastic and elastic wave) and Bragg shifting and broadening due to lattice distortion. Fig. 2 shows a series of four voids before and during the shock loading with increasing time delay between the drive laser and XFEL image. References: [1] F. Seiboth, et al., J. Phys. Conf. Ser., vol. 499, no. 1, p. 012004, Apr. 2014. [2] A. Schropp, et al., Sci. Rep., vol. 3, Apr. 2013. [3] A. Schropp, et al., SPIE Optical Engineering + Applications, 2013, p. 88490R. Figure 1. a) Schematic of the experimental setup showing the focused 6keV LCLS beam illuminating the laser-machined voids in the single crystal sample with the 20 ns glass laser (532 nm) beam driving a shock wave perpendicular to the XFEL beam; b) single-shot X-ray image of a the two-wave shock structure interacting with the 10 micron void in a PETN single crystal; c) Dynamic X-ray diffraction pattern taken simultaneously with the image showing Bragg peak shifting and broadening due to lattice compression. Figure 2. A montage of preliminary images comparing the static (top) and dynamic (bottom) single pulse images of four 10 micron diameter voids in 150 nm thick PETN at different delays with respect to the laser drive. The shock drive surface is at the top with the shock wave propagating towards the bottom. These images were taken on the FLI detector with the detector placed 4.1 m behind the samples.");sQ1[925]=new Array("../7337/1853.pdf","Imaging Irreversible Transformations with Movie-Mode Dynamic Transmission Electron Microscopy","","1853 doi:10.1017/S1431927615010041 Paper No. 0925 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Imaging Irreversible Transformations with Movie-Mode Dynamic Transmission Electron Microscopy Joseph T. McKeown, Melissa K. Santala, Tian Li, John D. Roehling, and Geoffrey H. Campbell Materials Science Division, Lawrence Livermore National Laboratory, Livermore, CA, USA In situ transmission electron microscopy (TEM) has been utilized for decades to image materials processes at high spatial resolution. Yet the relevant dynamics of many of these processes often remain elusive, as they unfold too rapidly to discern at small spatial scales using conventional TEM imaging conditions. For example, consider a transformation front moving with a relatively low velocity of 0.1 mm/s. In situ TEM imaging conducted with conventional acquisition rates of 30 frames/s corresponds to a temporal resolution of ~33 ms and, limited by motion blur, provides a spatial resolution of only ~3 �m. Given the rapid microstructural evolution of many types of irreversible transformation fronts--on the order of mm/s to m/s--nanosecond temporal resolutions are required to capture these processes. The dynamic transmission electron microscope (DTEM) at LLNL was developed to enable imaging of transient states during irreversible transformations with nanometer spatial and nanosecond temporal resolutions using a single-shot acquisition mode [1]. Recently, movie-mode DTEM was designed and implemented to provide multiple (up to 9) single-shot acquisitions in under ~1 �s, yielding frame rates that are on the order of 106 times higher than conventional in situ TEM frame rates [2]. Movie-mode DTEM employs an arbitrary waveform generator to deliver a user-defined (pulse duration and spacing) series of electron pulses, allowing complex irreversible transformation events to be tracked across microsecond timescales. Movie-mode DTEM acquisitions result in higher data throughput with reduced uncertainty. Here, we will provide an overview of movie-mode DTEM instrumentation and operation, with examples of its application to various materials science problems, including rapid alloy solidification (Figure 1), where solid-liquid interface evolution is monitored, and amorphous-crystalline phase transformations (Figure 2) involving semiconductor and phase-change materials. References [1] G.H. Campbell, J.T. McKeown, and M.K. Santala, Appl. Phys. Rev. 1 (2014) 041101. [2] T. LaGrange, B.W. Reed, and D.J. Masiel, MRS Bull. 40 (2015) 22. [3] This work was performed under the auspices of the U.S. Department of Energy, Office of Basic Energy Sciences, Division of Materials Sciences and Engineering for Microsc. Microanal. 21 (Suppl 3), 2015 1854 FWP SCW0974 by Lawrence Livermore National Laboratory under Contract DEAC52-07NA27344. Figure 1: Movie-mode DTEM image series acquired during the rapid solidification of an Al-3at.%Si alloy following laser melting. The entire solidification process is complete in ~50 �s. Each row of 9 images is a separate experiment that begins at the delay time above the leftmost image. Prior to movie mode, this entire data set would have required 23 separate single-shot experiments as opposed to the 3 movie-mode experiments shown here. Each frame was acquired with a 50-ns electron pulse; the inter-frame spacing was 2.5 �s. Figure 2: Movie-mode DTEM image series of laser crystallization of amorphous Ge. The 95-ns inter-frame time allows the crystallization front, moving at 10 m/s, to be imaged. The region imaged in the 20-ns exposures (left) is circled in the conventional TEM image (right).");sQ1[926]=new Array("../7337/1855.pdf","Dynamic Transmission Electron Microscope Study on the Crystallization of Ion-Bombarded Amorphous Germanium Thin Films","","1855 doi:10.1017/S1431927615010053 Paper No. 0926 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Dynamic Transmission Electron Microscope Study on the Crystallization of Ion-Bombarded Amorphous Germanium Thin Films Tian T. Li1, Melissa K. Santala1, Leonardus Bimo Bayu Aji1, Sergei O. Kucheyev1, HuaiYu M. Cheng2 and Geoffrey H. Campbell1 1 2 Materials Science Division, Lawrence Livermore National Laboratory, Livermore, CA, USA IBM/Macronix PCRAM Joint Project, IBM T. J. Watson Research Center, Yorktown Heights, NY, USA The structure-properties relationships of amorphous group IV semiconductors such as germanium (a-Ge) have attracted wide research interests over the past decades. The materials have great technological significance thanks to a plethora of applications, and understandings of their structural properties provide key scientific insights for those with similar covalently bounded networks. Pulsed laser induced crystallization is a common method to evolve the nanostructure in a-Ge. In this study, we report the correlations between the differences in the nanostructure of a-Ge thin films created by ion bombardment, and the laser crystallization kinetics measured using dynamic transmission electron microscope (DTEM). Previously, amorphous silicon (a-Si) prepared by different methods has have been shown to exhibit different medium range order (MRO), which gives rise to different mechanical properties and tendencies to structural relaxation under thermal annealing [1]. Changes in MRO can be measured using a statistical microscopy technique known as fluctuation electron microscopy (FEM), as shown in Figure 1. We sputter deposit ~30 nm of a-Ge thin film, and subsequently alter the nanostructure by Ar+ ion bombardment. We then observe the differences in MRO between the as-deposited and the treated samples using FEM. Amorphous germanium is known to crystallize with rate as high as ~10s of m/s during laser annealing. The movie-mode DTEM at LLNL provides us the unique capability to capture the structural evolution in situ during the crystallization process [2]. From a series of (up to 9) single-shot acquisitions in under ~1 �s, we can track the movement of the amorphous-crystalline boundary (Figure 2) with temporal resolution unavailable in conventional TEM experiments. We collect the crystallization kinetics data of both asdeposited and Ar+-treated a-Ge samples, and map the differences to the changes in the MRO. References [1] B. Haberl et al, J. Appl. Phys. 110 (2011) 096104. [2] G.H. Campbell, J.T. McKeown, and M.K. Santala, Appl. Phys. Rev. 1 (2014) 041101. [3] This work was performed under the auspices of the U.S. Department of Energy, Office of Basic Energy Sciences, Division of Materials Sciences and Engineering for FWP SCW0974 by Lawrence Livermore National Laboratory under Contract DEAC52-07NA27344. Microsc. Microanal. 21 (Suppl 3), 2015 1856 Figure 1: FEM variances (V(k)) versus diffraction vector (k) of four different amorphous Si samples: as-sputtered (as-sputt.), sputtered then annealed (ann. sputt.), as-implanted (as-imp.), and implanted then annealed/relaxed (rel. imp.). The differences in V(k) indicate the changes in the medium range order when the samples are subject to different preparation conditions and thermal treatments. Taken from [1], used with permission. Figure 2: (Left) A series of images collected using Movie-mode DTEM during in situ laser crystallization of amorphous Ge. The images are collected with 20-ns exposure and 95 ns intra-frame time. The crystallization front moves at ~ 20 m/s. (Right) conventional TEM image of the crystallized area after laser annealing.");sQ1[927]=new Array("../7337/1857.pdf","Simultaneous High-Speed DualEELS and EDS acquisition at atomic level across the LaFeO3 / SrTiO3 interface","","1857 doi:10.1017/S1431927615010065 Paper No. 0927 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Simultaneous High-Speed DualEELS and EDS acquisition at atomic level across the LaFeO3 / SrTiO3 interface P. Longo1, T. Topuria2, P. Rice2, A Aitouchen1, P.J. Thomas1 and R.D. Twesten1 1 2 Gatan Inc. 5794 W Las Positas Blvd, Pleasanton, CA, 94588, USA IBM Research Division, Almaden Research Center, San Jose, CA, 95120, USA Lanthanum ferrite LaFeO3 (LFO) is an antiferromagnetic perovskite oxide and has been investigated with great interest due to its potential applications in computer hard drives as read heads. LFO films are typically grown on SrTiO3 (STO) substrates due to the well matched lattice constants and the effects of chemistry, morphology and strain at the LFO / STO interface on the magnetic and electrical properties of the system. Hence, the study of the chemistry and elemental distribution at the atomic scale is important in order to understand the properties of these films. EELS and EDS in a STEM microscope have proven to be valuable tools for characterizing such materials. Here we show an approach where the simultanous high-speed acquisition of EELS and EDS spectra with an atomic scale probe allows the possibility to carry out a full compositional and chemical characterization of the LFO / STO interface. The analysis was carried out using a relatively short exposure time to reduce the effects of beam damage and sample drift which can be more pronounced when the data acquisiton is carried out at atomic level. To achieve efficient and fast joint EELS / EDS data acquisition, we have linked the STEM scan system with the EELS and EDS detectors via a single clock mastered off the EELS detector to ensure all the systems are in exact synchrony. To ensure data fidelity, the native detector applications are used to acquire the data into buffers and the data is transferred to the final application (in this case DigitalMicrograph) in blocks while the CPU is idle. Data were acquired at IBM San Jose, CA, USA using a probe-corrected, C-FEG STEM operating at 200 kV. A GIF Quantum ER EELS system equipped with DualEELS capability was used for the acquisition of energy-loss data and a large area (~1 sr) EDS detector was used to acquire the EDS data. EELS data were acquired in DualEELS mode and the low-loss (0 eV to 500 eV) and core-loss (400 eV to 900 eV) EELS spectra were acquired nearly simultaneously with 10 s transition time between exposures. The low-loss spectrum provides an accurate energy reference allowing absolute chemical shift measurements. The spectrometer was set up for moderate energy resolution analysis with a dispersion of 0.25 eV / channel yielding a measured energy resolution of 0.75 eV with 500 eV energy range in the spectrum. By looking at the shape of the near edge structure in the EELS spectrum, it is possible to extract information on the oxidation state, coordination state and much more. In addition, optical information present in the low-loss region is also available since the low-loss EELS data were recorded with each core-loss spectrum. EELS data were used to analyze chemical state information and generate elemental maps of the O, Ti, Fe and La. In the EELS spectrum, the main signals for the Sr are the M4,5 at 130 eV and the Sr L2,3 at 1940 eV. The M-edges have a poor peak-to-background ratio due their low energy while the L-edges are out of the energy range for this experiment (0 eV to 900 eV). The EDS data were acquired simultaneously with the EELS data and can be used to generate a Sr elemental map that can be combined with those obtained by EELS. Since the EDS signal does not provide any chemical information, only composition can be studied for Sr at the experimental conditions used in this paper. Figure 1 shows a colorized elemental map of Ti, Fe and La obtained using EELS and Sr using EDS. The color map suggests intermixing between Fe and Ti. Such intermixing can be confirmed and further Microsc. Microanal. 21 (Suppl 3), 2015 1858 investigated using EELS by looking at the fine structure of the edges across each atomic column. Figures 2a-c show EELS spectra of Ti L, O K and Fe L edges extracted from the selected regions in Figure 1. Changes can be observed in every spectrum although they are notable in the case of the O and Fe in Figures 2b,c with the Fe showing a chemical shift towards higher energy moving into the LaFeO3. This is a clear indication of a change in the oxidation state. The interface was also analyzed using STEM diffraction where diffraction patterns are taken point by point across the interface. This gives a further insight into the structure across the interface and information about strain and crystal distortion can be revealed. 1 2 3 4 Figure 1: Color maps of Ti L at 456 eV in green, La M at 832 eV in orange and Fe L at 708 eV in light blue obtained using EELS and Sr L at 1.81keV in red obtained using EDS. The numbers 1 � 4 correspond to the selected region where the EELS spectra in Figures 2a-d were extracted. Figure 2: EELS spectra extracted from the selected region in Figure 1; a) Ti L2,3-edges at 456eV; b) O K-edge at 532 eV; Fe L2,3-edges at 708 eV; d) Low-Loss region.");sQ1[928]=new Array("../7337/1859.pdf","Parallel Ion Electron Spectrometry (PIES): A New Paradigm for High-Resolution High-Sensitivity Characterization based on integrated TEM-SIMS","","1859 doi:10.1017/S1431927615010077 Paper No. 0928 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Parallel Ion Electron Spectrometry (PIES): A New Paradigm for HighResolution High-Sensitivity Characterization based on integrated TEM-SIMS S. Eswara Moorthy, D. Dowsett, L.Yedra, T. Wirtz Advanced Instrumentation for Ion Nano-Analytics (AINA), Luxembourg Institute of Science and Technology, 41 rue du Brill, L-4422 Belvaux, Luxembourg The Transmission Electron Microscopy (TEM) offers superior spatial resolution, but, the traditional analytical capabilities associated with electron microscopy such as the Energy Dispersive Spectroscopy (EDS) or Electron Energy-Loss Spectroscopy (EELS) are unfortunately inadequate for characterizing samples containing trace elements (at best 0.1 at. %) or for mapping isotopic distributions [1,2]. Another limitation is that investigations of light elements (such as hydrogen and lithium) are particular difficult or even impossible using these analytical methods. On the other hand, Secondary Ion Mass Spectrometry (SIMS) provides extraordinary chemical sensitivity (down to ppm or even ppb) and high dynamic range, but, offers poor lateral resolution [3]. However, to tackle modern problems in physical and biological sciences, the capability to simultaneously obtain high spatial resolution and high chemical sensitivity is of paramount importance. An ex-situ combination of TEM and SIMS in an attempt to overcome the limitations of the techniques taken individually is prone to sample modifications and other artefacts [4]. To overcome the intrinsic instrumental limitations, we have made an in-situ combination to complement the high-sensitivity of SIMS with the exceptional spatial resolution offered by TEM, by developing the correlative TEMSIMS technique. To determine the feasibility and to demonstrate the applications of the TEM-SIMS technique, we have developed a prototype instrument for TEM-SIMS based correlative microscopy (Figure 1). The pole-pieces of a Tecnai F20 were specially modified to accommodate the SIMS technique. A commercial FEI Magnum Ga+ FIB was attached to the TEM column to act as the primary ion column. The secondary ion extraction optics (extraction efficiency 90%) and a compact highperformance mass spectrometer were designed and developed in-house and are being continuously improved for optimal performance. A special sample holder which can be biased to high-voltages (�4.5 kV) was also developed in-house to enhance the collection efficiency of the secondary-ion extraction optics. To enhance the low intrinsic yield of secondary ions for non-reactive primary ion beams such as Ga+ for the TEM-SIMS we use reactive gas flooding [5]. Specifically, the enhancement of negative secondary ion yields due to Cs flooding and of positive secondary ion yields with O2 flooding were found to be up to four orders-of-magnitude. This enhancement of secondary ion yields leads to detection limits varying from 10-3 to 10-6 for a lateral resolution between 10 nm and 100 nm respectively (Figure 2). Sensitivities in the ppm range are possible, but at the cost of spatial resolution due to the inherent physical limit of SIMS. Nevertheless, it is possible to recover the structural details that were thus lost by overlaying high-resolution TEM image over the highsensitivity (but, poorer resolution) SIMS image. In this presentation, the strategies employed to overcome the technical challenges of the TEM-SIMS combination of techniques will be discussed and the new possibilities enabled by this correlative microscopy method will be highlighted with a focus on applications in materials science and Microsc. Microanal. 21 (Suppl 3), 2015 1860 biology. Two distinctive analysis protocols were identified: (a) TEM is used first to identify interesting structures followed by SIMS to know the chemical information and (b) SIMS is used first to identify chemical hotspots and then TEM is used for structural characterization. Details of these protocols and other related methods like isotopic labelling will also be discussed. Special emphasis will also be placed on TEM-SIMS characterization of samples containing light elements (low Z), which are particularly challenging with traditional analytical methods like EDS or EELS. References: [1] L. Reimer, H. Kohl, Transmission Electron Microscopy: physics of image formation (Springer, New York, 2008). [2] R. F. Egerton, Electron Energy-Loss Spectroscopy in the Electron Microscope (3rd edition), (Springer, New York, 2011). [3] D. S. McPhail, J. Mater. Sci. 41, (2006) 873. [4] K. Q. Ngo et al, Surf. Sci. 606, (2012) 1244. [5] P. Philipp et al, Int. J. Mass. Spectrom. 253 (2006) 71 Figure 1. Schematic of the TEM-SIMS setup (left) and the photo of the TEM-SIMS prototype instrument (right). Experimental O flooding 100 10-1 10-2 Detection limit without flooding Detection limit with flooding 2 Extrapolated No flooding Cesium flooding 10-1 10-2 Detection limit Useful yield Ga+ beam Si- signal UY = 6x10-5 without flooding UY = 1.6x10-1 with flooding 1 10 100 10-3 10 10 -4 10-3 10-4 10-5 10-6 GaAs matrix 10-7 COSiCuB+ Zn+ Cu+ -5 10-6 10-7 10-8 Lateral resolution (nm) Figure 2. Left: Detection limits using a Ga+ FIB with and without Cs0 flooding vs. minimum feature size: example for the detection of Si-. Right: Enhancement of secondary ion yields using reactive gas flooding under Ga+ bombardment Element");sQ1[929]=new Array("../7337/1861.pdf","Improving Analytical Efficiency of EDS using a Newly-designed X-ray Detecting System for Aberration Corrected 300 kV Microscope","","1861 doi:10.1017/S1431927615010089 Paper No. 0929 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Improving Analytical Efficiency of EDS using a Newly-designed X-ray Detecting System for Aberration Corrected 300 kV Microscope Takeo Sasaki, Hidetaka Sawada, Eiji Okunishi, Yu Jimbo, Yorinobu Iwasawa, Koji Miyatake, Shuichi Yuasa, and Toshikatsu Kaneyama JEOL Ltd., EM Business Unit, 3-1-2 Musashino, Akishima, Tokyo 196-8558, Japan Performance of elemental analysis with energy-dispersive X-ray spectroscopy (EDS) has been improved by using large-sized silicon drift detectors (SDDs) [1]. The detectors are required to place closer to the specimen holder to obtain more counts of X-ray signals because the solid angle of the X-ray detectors increases. The positions, sizes and number of the detectors have not been satisfactorily optimized to attain a high analytical efficiency, since the space to place the detectors is limited. We have developed a new X-ray detecting system for JEM-ARM300F that equips a cold field emission electron source. The key point of this system is to renovate an objective lens pole-piece and a specimen holder in addition to the SDDs. A newly designed wide gap pole-piece (WGP) accommodates new SDDs to place closer to the specially designed analytical holder so that a larger amount of X-ray signal can be detected with a very high efficiency. Figure 1 compares peak intensities of Ni-K obtained from a NiOx thin film at 80 kV with SDD1, SDD2, and SDD1 + SDD2 (Dual). The intensity of the SDD2 is approximately 1.8 times larger than that of SDD1 due to the shorter distance between the detector (SDD2) and a specimen. The SDD2 is placed in the direction of the sample holder rod and is located at a closer position to the specimen than SDD1, because the special specimen holder has a shorter length in the holder rod direction. Table 1 shows the solid angles and take off angles of the SDD1 and SDD2. The solid angle of SDD2 (1.08 sr) is about two times larger than that of SDD1 (0.55 sr), which accords well with the intensity ratio of the NiK spectra. Figure 2 shows atomic resolution EDS maps of SrTiO3[100] taken at 80 kV. Sr and Ti columns are clearly seen. It is worth to note that Ti-O columns and O columns are visible at an acquisition time of 13.5 minutes (number of pixels = 128�128). This proves a high analytical efficiency of the present system. It is clear from the short acquisition time that this X-ray detecting system is a very powerful tool for elemental analysis of damage-susceptible materials such as carbon materials. A spherical aberration coefficient (Cs) and a chromatic aberration coefficient (Cc) of WGP pole-piece (Cs=1.6 mm, Cc=2.4 mm) are larger than those of FHP pole-piece (Cs=0.7 mm, Cc=1.3 mm). (FHP is an optional objective lens pole-piece for ultra-high resolution in JEM-ARM300F and enables to resolve the Si to Si dumbbell of 45 pm in Si[114] sample [2].) We have tested the imaging performance of WGP. Figure 3(a) shows a Cs-corrected high angle annular dark field scanning transmission electron microscopy (HAADF-STEM) image (averaged with four sequential raw images) of Si[112] taken at 300 kV with a dwell time of 38 �sec/pix. The Si to Si dumbbell of 78 pm is clearly resolved. The Fourier transform of the image exhibits spots of (73 pm)-1 as well as (78 pm)-1 as shown in Figure 3(b). Figure 3(c) shows Cs-corrected TEM image of -Si3N4[0001]. The Si to N dumbbell of 95 pm is clearly imaged. These results have revealed ultra-high resolution STEM and TEM imagings are possible even with the WGP pole-piece. The wide gap permits a higher specimen tilting angle, and gives enough room for special holders of heating, cooling, gas environment etc. We emphasize that the WGP can be used for observations of many purposes. Microsc. Microanal. 21 (Suppl 3), 2015 1862 References: [1] S Kawai et al, Microsc. Microanal. 20. Suppl. 3 (2014) p.1150-1151. [2] H Sawada et al, Microsc. Microanal. 20. Suppl. 3 (2014) p.124-125. Solid angle [sr] Take-off angle [deg] WGP (Dual-SDD) SDD1 SDD2 0.55 1.08 25 29 Table 1 Solid angles and take off angles of SDD1 and SDD2 with the use of WGP. Figure 1. EDS spectra taken from a NiOx thin film at 80 kV. Blue triangles, red squares, and light green circles correspond to SDD1, SDD2, and SDD1+SDD2 (Dual), respectively. The intensities of the spectra were normalized using the peak intensity of Ni-K obtained with the SDD1. Figure 2. EDS elemental maps (radial-difference-filtered) of SrTiO3[001] taken at 80 kV with a probe current of 150 pA, a convergence semi-angle of 24 mrad, a map size of 128�128 pixels and an acquisition time of 13.5 minutes. Each scale bar represents 1nm. Figure 3. (a) Cs-corrected HAADF-STEM image of Si[112] taken at 300 kV (WGP pole-piece) with a dwell time of 38 �sec/pix (512�512 pixels), averaged with four sequential raw images. (b) Fourier transform (in power) of the image shown in (a). (c) Cs-corrected TEM image (raw data) of -Si3N4[0001] taken at 300 kV (WGP pole-piece).");sQ1[930]=new Array("../7337/1863.pdf","STEM in 4 Dimensions: Using Multivariate Analysis of Ptychographic Data to Reveal Material Functionality","","1863 doi:10.1017/S1431927615010090 Paper No. 0930 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 STEM in 4 Dimensions: Using Multivariate Analysis of Ptychographic Data to Reveal Material Functionality Stephen Jesse1,2, Miaofang Chi1,3, Albina Y. Borisevich1,3, Alex Belianinov1,2, Raymond R. Unocic1,2, Christopher T. Symons1,5, Eirik Endeve1,4, Richard K. Archibald1,4, Sergei V. Kalinin1,2, Andrew R. Lupini1,3 1 2 Institute for Functional Imaging of Materials, Oak Ridge National Laboratory, Oak Ridge, TN 37831 Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, TN 37831 3 Materials Science & Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831 4 Computer Science and Mathematics Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831 5 Computational Sciences & Engineering Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831 The scanning transmission electron microscopy (STEM) is a powerful platform for studying materials and their many varied properties (including structural, electronic, magnetic, ferroic etc.) locally and at their fundamental length-scales. In a general sense, STEM imaging can be represented as an information channel connecting the interactions of a highly focused electron beam and sample to the observer. However, in the current standard implementation, this channel is severely restricted by the instrumental linkage in which the information rich electron diffraction pattern (Ronchigram) is reduced to a single value (e.g. through the integrated intensity over a relatively large detector) at each beam location thus resulting in loss and distortion of information about important but subtle aspects of material properties. Previous theoretical and experimental work [1, 2, 3,4] suggests that full acquisition of the Ronchigram at each spatial location during a scan can enable super resolution, phase-contrast imaging, imaging of internal fields, and 3D sample reconstruction. Data acquisition and storage has evolved to a point now that it is possible to capture high resolution 4D ptychography data sets rapidly, and the use of large-scale compute facilities enables the processing and mining of these multi-GB data sets to distil the most salient aspects of the data while separating the statistically significant variations (signal) from noise. In the current work we have utilized the DE-12 camera (Direct Electron, LP, San Diego, CA), equipped with a 4096 x 3072 pixels Direct Detection Device (DDD�) sensor [5] installed on an aberration corrected FEI Titan operating at 300 kV combined through a custom FPGA control system to synchronize frame capture and beam positioning to acquire 4D scanning-scattering data sets. Typical scans of 192X192 pixel images containing 384X384 Ronchigrams at every pixel were captured in less than 1 minute. A host of multivariate statistical methods can be brought to bear to extract meaningful information from these large data sets. Statistical methods can be advantageous in that they are parallelizable and can be efficiently implemented on large-scale computing platforms, are model-free and operate with no pre-imposed expectation or bias, and are capable of elucidating inter-pixel correlations unlikely to be noticed by human observation alone. As an example, shown in Figure 1a is a micrograph rendered by integrating over the total intensity of Ronchigrams at each spatial location and shows the positions of atomic columns in thin film polymorph bismuth ferrite at a tetragonal/rhombohedral phase boundary. A combination of principal component analysis (PCA) and k-means clustering applied to the 4D data set enabled compression and sorting of the 16384 Ronchigrams into 25 spatially resolved groupings of similar scattering behavior. As shown in figure 1b, the different clusters are color-coded and the locations of atomic columns are indicated by an overlaid contour map. The most prominent result of this analysis is the clear differentiation of the two Microsc. Microanal. 21 (Suppl 3), 2015 1864 crystalline phases. Note this differentiation was based entirely on extracting subtle but statistically significant variations in Ronchigrams. Additionally, regular tiling of cluster arrangements commensurate with unit cell spacing provides a means to reveal, in a systematic way, the effects of local fields on electron scattering behavior. Figure 2 shows in greater detail the particular Ronchigrams associated with selected positions near an atomic column (indicated in figure 1b). Additional examples, more in-depth analysis methods, and relationships of analysis results to material properties will be discussed. [6] References: [1] J. M. Rodenburg, B. C. McCallum and P. D. Nellist, Ultramicroscopy, 48, (1993), p. 304. [2] N. Shibata et al, Nature Physics, 8, (2012), p. 611. [3] T.J. Pennycook et al, Ultramicroscopy, in press, (available on-line Nov. 2014) [4] C. Ophus et al, Microsc. Microanal. 20, 2014 [5] The authors would like to acknowledge Liang Jin and Benjamin Bammes for their assistance in allowing us to use the DE-12 system. http://www.directelectron.com/. [6] Research supported by: Oak Ridge National Laboratory's Center for Nanophase Materials Sciences (CNMS), which is sponsored by the Scientific User Facilities Division, Office of Basic Energy Sciences, U.S. Department of Energy, by the Laboratory Directed Research and Development Program of Oak Ridge National Laboratory, managed by UT-Battelle, LLC, for the U. S. Department of Energy, and by Division of Materials Sciences and Engineering Division, Office of Basic Energy Sciences, U.S. DOE. This study used the resources of the Oak Ridge Leadership Computing Facility at the Oak Ridge National Laboratory, and analysis under support of applied mathematics program at the DOE. Figure 1. (a) Micrograph of tetragonal/rhombohedral phase boundary of thin film Bismuth Ferrite. (b) Ronchigram based cluster map clearly differentiating phases as well as local but regular variations in scattering behavior commensurate with unit cell tiling. Color bar indicates cluster labeling. Figure 2. A zoomed-in view of the region indicated in figure 1b illustrating the Ronchigrams associated with specific clusters in the vicinity of an atomic column.");sQ1[931]=new Array("../7337/1865.pdf","Resolution Assessment of an Aberration Corrected 1.2-MV Field Emission Transmission Electron Microscope","","1865 doi:10.1017/S1431927615010107 Paper No. 0931 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Resolution Assessment of an Aberration Corrected 1.2-MV Field Emission Transmission Electron Microscope Yoshio Takahashi1, Tetsuya Akashi1, Tomokazu Shimakura1, Toshiaki Tanigaki1, Takeshi Kawasaki1, Hiroyuki Shinada1, and Nobuyuki Osakabe1 1. Central Research Laboratory, Hitachi, Ltd., Hatoyama 350-0395, Japan Observations of the structure and electromagnetic field at atomic resolution are critical for developing advanced materials. A lot of efforts are being made for achieving atomic resolution and for constantly improving it in transmission electron microscopes (TEMs) not only at the conventional in-lens specimen position (HR position) but also at a field-free specimen position (Lorentz position). In the last decade, 1/0.05 nm-1 information transfer for 300-kV TEM has been demonstrated [1], and 0.5-nm resolution in Lorentz microscopy has been reported [2,3]. We developed an atomic-resolution holography electron microscope (1.2-MV TEM) equipped with a cold field-emission gun and a CEOS hexapole Cs-corrector. In this paper, we report on the resolution assessment of the developed 1.2-MV TEM. We found that the resolution reached 0.043 nm at the HR position and 0.24 nm at the Lorentz position. Our strategy to define the instrumental resolution of the 1.2-MV TEM was to measure the maximum spatial frequencies both of a linear information transfer and of a no-reversal contrast transfer. If the maximum spatial frequency with no contrast reversal attained by the Cs-corrector was substantially higher than the linear information limit, we could state the instrumental resolution reached the information limit. Imaging resolution can also be checked by visualizing images of separated atomic columns or structures, which are restricted not only by the instrumental resolution but also by the beamspecimen interactions and the signal-to-noise ratio due to the total electron dose. For the HR position, a linear information transfer was verified using a chromatic lattice image [4]. Diffraction spots of 000, and 633, 642, 651 of <111>-oriented W produced under axial illumination were selected using two holes fabricated on a Cu plate situated at an objective aperture and then were interfered each other at an image plane. If lattice fringes are visualized, a linear information transfer with corresponding fringe spacing can be verified. Figures 1(a) and (b) show the chromatic lattice image of W and its Fourier transform (FT). Since W{633} fringes and the corresponding reflection spots were clearly visible, a 0.043-nm linear information transfer was confirmed. A two-dimensional coherent phase contrast transfer function (PCTF) depicted by using the residual aberrations showed no contrast reversal up to a spatial frequency of 1/0.034 nm-1. Both measurements enable us to conclude that the 1.2-MV TEM has 0.043-nm instrumental resolution. Imaging of 0.044 nm-separated dumbbells of <411>-oriented GaN was demonstrated. Approximately 3 nm-thick areas near a cleaved edge of GaN were observed because the exit wave no longer has a separated structure at the thick area due to beam broadening in the sample. Film was used for recording. The total dose was approximately 2x106 e-/nm2. Figure 2 shows (a) observed Ga dumbbells with 0.044 nm separation and (b) a simulated GaN image with the signal-to-noise ratio (SNR) that stems from the finite electron dose and the detection efficiency. For a field-free observation, the 1.2-MV TEM implements a dedicated imaging lens (Lorentz lens) above a conventional objective lens and an additional specimen stage (Lorentz position). The linear Microsc. Microanal. 21 (Suppl 3), 2015 1866 information transfer for the Lorentz position was examined by Thon ring patterns appearing in the Fourier transform (FT) of the image of amorphous carbon (10 nm) with Au particles (Fig. 3(a)). Images were taken by a CCD camera with a Nyquist frequency of 20 nm-1. The radial line profile of the Thon ring showed a 1/0.24 nm-1 linear information transfer (Fig. 3(b)). 111-reflections (1/0.235 nm-1) are also visible. A two-dimensional CTF estimated by the residual aberrations for the Lorentz position showed no contrast reversal up to a spatial frequency of 1/0.19 nm-1. Hence, the instrumental resolution at the field-free specimen position was concluded to reach 0.24 nm. We were also found that images of the {111} lattice with 0.235 nm spacing in polycrystalline Au particles could be obtained with a total dose of about 1x106 e-/nm2 (Fig. 3(c)). References: [1] C. Kisielowski et al., Microsc. Microanal. 14 (2008) p. 469. [2] R. E. Dunin-Borkowski et al., Microsc. Microanal. 18 (2012) p.1708. [3] A. K. Petford-Long et al., Proceedings of EMC (2012), PS.1.1. [4] T. Akashi et al., Appl. Phys. Lett. 87 (2005) p. 174101. [5] This research was supported by a grant from the Japan Society for the Promotion of Science (JSPS) through the "Funding Program for World-Leading Innovative R&D on Science and Technology (FIRST Program)" initiated by the Council for Science and Technology Policy (CSTP). We thank Dr. M. Haider, Dr. H. M�llar, and CEOS GmbH for developing the Cs-corrector. (a) (a) 0.043 nm (b) (b) W{6 3 3} 0.2 nm Figure 2. (a): Observed 0.044-nm GaN dumbbells. (b): Simulated image with SNR on electron dose. Figure 1. (a): Chromatic lattice fringes of tungsten at HR position. (b): FT of (a). (a) (b) 10.0 Count (x104) 5.0 1/0.24 nm -1 (c) 2.0 1.0 Au {111} 0 1 2 3 4 5 Spatial f requency (nm-1) 6 3 nm Figure 3. (a):Diffractogram of amorphous carbon with Au particles taken at Lorentz position, (b): Radial line profile at a sector in (a), (c): Image of Au {111} lattices with 0.235 nm spacing.");sQ1[932]=new Array("../7337/1867.pdf","Analytical Transmission Scanning Electron Microscopy: Extending the Capabilities of a Conventional SEM Using an Off-the-Shelf Transmission Detector*","","1867 doi:10.1017/S1431927615010119 Paper No. 0932 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Analytical Transmission Scanning Electron Microscopy: Extending the Capabilities of a Conventional SEM Using an Off-the-Shelf Transmission Detector* Jason Holm1 and Robert Keller1. 1. National Institute of Standards and Technology, Materials Measurement Lab., Boulder, CO 80305 Transmission-mode scanning electron microscopy techniques are seeing increased use in both materials science and biological applications [1]. One reason is that recently available off-the-shelf transmission detectors that can be implemented into nearly any scanning electron microscope (SEM) are enabling imaging modes not normally associated with conventional SEMs. For example, some modular units include bright and dark field transmission detectors, and one modular unit includes a high angle annular dark field (HAADF) detector with which elemental contrast can be observed. Although modular transmission detectors have demonstrated much utility [1-3], their full capacity has not yet been realized. One drawback is the inability to easily control the detector collection angle, and therefore select which electrons will be used to form images. Unlike scanning transmission electron microscopes (STEMs), conventional SEMs do not have post-specimen projector lenses that can be used to modify the collection angle. This means either (1) the specimen-to-detector distance must be adjusted to change the collection angle, or (2) an appropriate mask/aperture must be used to restrict the scattered electrons to a particular region of the detector corresponding to the desired collection angle. Furthermore, SEM transmission detectors with built-in specimen holders do not allow the sample position or orientation to be changed with respect to the detector or incident electron beam, and commercially available detectors with separate holders allow very little specimen positional control. (For example, see Fig. 1 in [2].) Inserting masks/apertures between the specimen and detector in such cases is challenging since very little space is available, and aligning an aperture with the detector and the optic axis is cumbersome at best. This contribution demonstrates a method to extend the analytical capacity of existing off-the-shelf modular detectors by implementation of (1) a new sample holder, and (2) a simple, inexpensive aperture system that can be used to select electrons scattered to any angle intercepted by the detector. It will be shown for one particular transmission detector that bright field acceptance angles are not limited to those dictated by the intended bright field detector, and that acceptance angles are easily controllable and modifiable over a wide range, from 0 mrad to > 1350 mrad. Implementation of other transmission imaging modes including annular bright field (ABF) [4], low angle annular dark field (LAADF) [5], and thin annular detector (TAD) schemes [6] is also possible. In particular, this work shows how the two developments can enable HAADF transmission imaging in a 15 year old SEM. To demonstrate the utility of a cantilevered, clamp-based sample holder and aperture system, Figure 1 shows an interior view of the SEM chamber with a transmission detector and EDS detector inserted. An annular aperture (not directly visible in the chamber image) was placed between the transmission detector and the sample. With this aperture in place, and the sample positioned approximately 6 mm above the detector, the dark field inner acceptance angle was approximately 250 mrad. Figure 2 shows dark field images of several 30 nm diameter Au and TiO2 nanoparticles on a carbon substrate recorded under these conditions. The image intensity differences for different materials are easily discernible in Fig 2a. Figure 2b shows an X-ray map of the sample, demonstrating that the most intense regions can *Contribution of the U.S. Department of Commerce; not subject to copyright in the United States. Microsc. Microanal. 21 (Suppl 3), 2015 1868 be assigned to the Au particles. Figure 2c shows that Au particles hidden under an agglomeration of TiO2 particles are not visible when viewed with a conventional secondary electron detector, but are easily discernible when imaging with the transmission detector and annular aperture system. This work shows that a relatively low cost upgrade can significantly extend the analytical capabilities of practically any SEM [8]. References: [1] T. Klein, et al., in Advances in Imaging and Electron Physics, P.W. Hawkes (Ed.), 171 (2012), 297. [2] N. Hondow et al. Nanotoxicology 5 (2011) 215. [3] M. Kuwajima et al. PLOS ONE 8 (2013) e59573. [5] S. D. Findlay et al. Ultramicroscopy 110 (2010) 903. [6] G. Rubin et al. Ultramicroscopy 113 (2012) 131. [7] J. M. Cowley, J. Electron Micros. 50 (2001) 147. [8] J. Holm acknowledges funding from the National Research Council Postdoctoral Research Associateship Program. Sample positioned here Transmission detector with masking aperture placed directly above the imaging diodes. Figure 1. A view of the SEM chamber interior with the transmission detector, sample holder, and EDS detector inserted. This is the setup used to generate the images shown in Fig. 2. Because of the camera angle, the sample and aperture are not visible in this image. (a) (b) (c) Figure 2. HAADF transmission images showing 30 nm Au nanoparticles and 30 nm TiO2 nanoparticles on a carbon substrate. The detector gain has been adjusted to show the carbon substrate. (a) An image showing intensity differences between the Au, TiO2, and C. (b) An X-ray map superimposed on the HAADF image. (c) Split-screen image showing Au particles under a pile of TiO2 particles. (HAADF image is on the left, secondary electron image is on the right.)");sQ1[933]=new Array("../7337/1869.pdf","Nanoscale Chemical Imaging via AFM coupled IR Spectroscopy","","1869 doi:10.1017/S1431927615010120 Paper No. 0933 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Nanoscale Chemical Imaging via AFM coupled IR Spectroscopy C.B. Prater1 M. Lo1, Q. Hu1, H. Yang1, C. Marcott2, and K. Kjoller1 1 2 Anasys Instruments, Santa Barbara, California 93101 USA Light Light Solutions, Athens, Georgia 30608 USA Conventional infrared spectroscopy is one of the most widely used tools in science and industry to identify materials via vibrational resonances of chemical bonds, but optical diffraction limits its spatial resolution to the scale of many microns. Atomic force microscopy (AFM) enjoys excellent spatial resolution and can measure mechanical, electrical, magnetic and thermal properties of materials, but has historically lacked the ability to perform robust chemical analysis. Two techniques, (1) AFM-based infrared spectroscopy (AFM-IR) and (2) scattering scanning near field optical microscopy (s-SNOM) have been developed which couple AFM with an IR source allowing the chemical identification capabilities of IR spectroscopy to extend to the nanoscale. As complementary techniques, AFM-IR and s-SNOM together provide an unrivaled capability to perform nanoscale chemical analysis on a diverse range of organic, inorganic, photonic and electronic materials. The AFM-IR technique achieves nanoscale spatial resolution by using the AFM probe as a local sensor of the IR absorption [1]. A sample is irradiated with light from a pulsed, tunable infrared laser. When the IR laser is tuned to a wavelength where the sample has an absorption peak, a portion of the incident IR light is absorbed and converted into heat resulting in a rapid thermal expansion of the absorbing region. This generates an impulse force on the AFM tip, inducing resonant oscillations of the AFM cantilever. IR absorption spectra can then be obtained by measuring the cantilever oscillation amplitude as a function of laser wavenumber. Because the amplitude of induced cantilever oscillation is directly proportional to the IR absorption of the sample [2], absorption spectra obtained by AFM-IR compare very well to conventional FTIR without the use of modelling or reference samples. More recent extensions of this technique have allowed chemical measurements to be performed on samples as thin as individual monolayers [3]. Figure 1 shows an example application of the AFM-IR measurement, nanoscale chemical measurements of the environmental degradation of a biomedical material. One challenge to implantable devices is the possible degradation of the material due to exposure to the in vivo environment. Polyurethane was exposed to this simulated environment for sufficient time to allow the initiation of surface degradation. The chemical mechanism of this degradation was then analyzed and mapped using the AFM-IR technique. The AFM-IR spectra clearly show the chain scission which occurs in the ether bonds of the polyurethane by the reduction in the peak at 1108 cm-1. It also shows the formation of alcohol groups by the increase in the 1176 cm-1 absorption band and carboxylate groups (COO-) shown by the increase in the broad bands at 1652 and around 1400 cm-1. These absorption bands can be used to map the degradation and determine how the degradation initiates and extends into the sample, information that was previously not achievable at this length scale. The s-SNOM technique has been used in the SPM field for more than two decades [4]. It uses a metallized AFM tip to enhance and scatter radiation from a nanometer scale region of the sample. The scattered radiation is detected in the far field and carries information about the complex optical properties of the nanoscale region of the sample under the metallized tip. Both the optical amplitude Microsc. Microanal. 21 (Suppl 3), 2015 1870 and phase of the scattered light can be measured typically using reference samples (e.g. gold or silicon) to correct the measurements for unknown phase contributions from the light source and tip. With appropriate models, these measurements can be converted into measurements of the complex optical constants (n, k) of the material under the tip. In some cases, the optical phase versus wavelength provides an approximation to a conventional absorption spectrum. As well as providing chemical imaging, the s-SNOM technique can also image electrical properties as demonstrated by the imaging of surface plasmon polaritons (SPP) in graphene [5], shown in Fig. 2. The SPPs are excited by the s-SNOM tip and then scattered or reflected by grain boundaries, edges or defects in the graphene. This generates interference which can be visualized with high spatial resolution using s-SNOM and provides information on the electronic structure of the graphene. References: [1] Dazzi, A., et al., Local infrared microspectroscopy with subwavelength spatial resolution with an atomic force microscope tip used as a photothermal sensor. Opt. Lett., 2005. 30(18): p. 2388-2390. [2] Dazzi, A., Theory of infrared nanospectroscopy by photothermal induced resonance. J. Appl. Phys., 2010. 107(12): p. 124519. [3] Lu, F.; Belkin, M., Infrared absorption nano-spectroscopy using sample photoexpansion induced by tunable quantum cascade lasers. Opt. Exp. 2011, 19, 19946. [4] Zenhausern, F., et al, Apertureless Near-Field Optical Microscope. Applied Physics Letters 1994 65 (13), p. 1623-1625. [5] Gerber, J. A. et al., Phase-Resolved Surface Plasmon Interferometry of Graphene. Phys. Rev. Lett., 2014. 113, 055502. Figure 1. a) AFM-IR spectra a) b) and b) AFM topography image of a section of a degraded polyurethane sample showing the chemical variation caused by the exposure of the polyurethane to a simulated in vivo environment. Figure 2. a) AFM topography and b) near field amplitude, collected at 925 cm-1 of a graphene sample showing the defects and grain boundaries which result in the plasmon interference in the near field image (indicated by the arrow).");sQ1[934]=new Array("../7337/1871.pdf","Introducing Nano-FTIR � Imaging and Spectroscopy at 10nm Spatial Resolution","","1871 doi:10.1017/S1431927615010132 Paper No. 0934 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Introducing Nano-FTIR � Imaging and Spectroscopy at 10nm Spatial Resolution Tobias Gokus1, Florian Huth1 and Andreas Huber 1. 1 Neaspec GmbH, Bunsenstrasse 5, 82152 Planegg (Munich), Germany Scattering-type scanning near-field optical microscopy systems (s-SNOM) allows to overcome the diffraction limit of light enabling optical measurements, imaging as well as spectroscopy, at a spatial resolution of 10nm not only at visible frequencies but also in the infrared or terahertz spectral range. S-SNOM employs an externally-illuminated sharp metallic AFM tip to create a nanoscale hot-spot at its apex [1]. The optical tip-sample near-field interaction is determined by the local dielectric properties (complex refractive index) of the sample and detection of the elastically tip-scattered light yields nanoscale resolved near-field images simultaneous to topography. Here we demonstrate that, based on the sSNOM concept, it is possible to perform molecular vibrational spectroscopy (nano-FTIR) of complex organic and inorganic nanostructures with 10-20nm spatial resolution can be realized [2]. For nano-FTIR spectroscopy the AFM tip of the sSNOM system is illuminated by a mid-IR broadband laser source while the light which is elastically scattered back from the AFM tip is analyzed with the help of an asymmetric Michelson interferometer. Nano-FTIR spectroscopy yields simultaneously amplitude and phase spectra of the materials under study and ultimately allows for directly determining the corresponding absorption as well as reflection spectra (see Figure 1). Furthermore, we find that nano-FTIR absorption spectra correspond well with transmission far-field FTIR spectra and therefore can be directly used for chemical characterization of unknown materials. Other applications show characterization of embedded structural phases in biominerals [3] or organic semiconductors [4]. The universal nano-FTIR approach presented here also allows for tailored system configurations where the ultimate spectral coverage can be achieved by using synchrotron-based broadband IR light sources [5] or pump-probe measurements at 10nm spatial resolution [6]. Equipping s-SNOM systems with cw light sources we demonstrate that Mid-IR s-SNOM near-field imaging can be performed at time scales of 30-300s per image. Use of material-selective frequencies in the mid-IR spectral range can be exploited to fully characterize polymer blends [7] or phase change materials with nanometer-scale domains. Quantification of free-carrier concentration and carrier mobility in doped semiconductor nanowires [8] or visualization and analysis of localized and propagating plasmons across graphene nanostructures [9] is achieved by amplitude- and phase-resolved infrared near-field imaging. Especially, the latter application holds great promises for graphene quality analysis as is possible to highlight grain boundaries and electronic defects which are not observable in standard AFM topography images. We envision that correlative IR graphene plasmon s-SNOM imaging and near-field photocurrent imaging will be a powerful tool for studying the electronic properties of graphene devices. Extending the concept of broadband-sSNOM spectroscopy to the THz-spectral range, we demonstrate nanoscale material characterization at THz-frequencies by coupling the free space beam of a THz-TDS system to a s-SNOM microscope [10,11]. Material specific near-field spectra of metals and highly doped semiconductors in the spectral range between 0.5 - 2.5 THz as well as near-field images (at fixed delay) can be obtained. The presented results open the door to study nanoscale light-matter interaction in the THzspectral range for all materials systems that can be investigate by conventional AFM measurements. Microsc. Microanal. 21 (Suppl 3), 2015 1872 Further, due the reflective optics design typically employed in s-SNOM systems we envision that also timeresolved near-field pump-probe experiments in the THz spectral range can be realized in order to study e.g. the charge carrier dynamics in semiconductors but also in biological complexes with nanoscale resolution. References: [1] F. Keilmann, R. Hillenbrand, Phil. Trans. R. Soc. Lond. A 362, 787 (2004) [2] F. Huth et al, Nano Lett. 12, 3973 (2012) [3] S. Amarie, et al, Beilstein J. of Nanotech. 3, 312 (2012) [4] C. Westermeier et al, Nature Comm. 5, 4101 (2014) [5] P. Herrmann et al, Optics Expr. 21, 2913 (2013) [6] M. Eisele et al, Nature Phot. 8, 841 (2014) [7] T. Taubner et al, APL 85, 5064 (2004) [8] J. Stiegler et a., Nature Comm. 3, 1131 (2012) [9] J. Chen et al, Nature 487, 77 (2012) [10] K. Moon et al, APL, 101, 011109 (2012) [11] A. Huber et. al, Nano Lett., 8, 3766 (2008) Figure 1. Nano-FTIR spectrum of a Si-SiO nano-structure. (a) Near-field reflectivity and absorption spectrum of SiO revealing a pronounced material-specific spectral signature. Detailed lineshape analysis allows identifying up to 5 distinct resonances at ca. 840, 1060, 1120, 1210, and 1365 cm-1 determining the spectral signature. Inset: IR near-field image of Si-SiO nanostructure at 1600cm-1 where the SiO spectral signature was measured, exhibiting a pronounced contrast due to different refractive index of materials. White line marks position of line profile. (b) Near-field spectroscopic line profile across SiO and Si interface (raw data, i.e. no filtering applied). Spectroscopic measurement along 1�m distance (10 nm pixel size, 100 spectra, 30s/spectra) exhibits abrupt change of measured spectral signature of SiO to a homogeneous spectral response on Si at 20-30 nm spatial resolution.");sQ1[935]=new Array("../7337/1873.pdf","New Hybrid Peak-Force Tapping/Near-Field Microscope for Nano-Chemical and Nano-Mechanical Imaging of Graphene Plasmons, Polymers and Proteins","","1873 doi:10.1017/S1431927615010144 Paper No. 0935 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 New Hybrid Peak-Force Tapping/Near-Field Microscope for Nano-Chemical and Nano-Mechanical Imaging of Graphene Plasmons, Polymers and Proteins M. Wagner1, G. Andreev1, K. Carneiro2, S. Habelitz2, T. Mueller1 1. 2. Bruker Nano Surfaces, 112 Robin Hill Road, Santa Barbara, CA 93117 University of California, Preventive and Restorative Dental Sciences, San Francisco, CA 94143-0758 Heterogeneous material systems in biomedical and chemical sciences as well as ever decreasing feature sizes in nanotechnology require new nanoscale material characterization techniques. Infrared spectroscopy can give valuable information on chemical composition, but conventional, far-field techniques such as Fourier-Transform Infrared (FTIR) spectroscopy are limited in spatial resolution to approximately the employed wavelength, i.e. several micrometer. Scattering-type scanning near-field optical microscopy (s-SNOM) is a near-field technique that is based on an atomic force microscope (AFM) and detects tip-scattered light in the infrared spectral region [1]. Spatial resolution in s-SNOM is only determined by the AFM tip radius and can reach down to 10nm, making s-SNOM a widely accepted tool for nanoscale chemical characterization. Another valuable AFM technique for material characterization is Bruker's peak-force tapping that allows precise force control between tip and sample down to 10s of pN which is essential for imaging fragile material systems [2]. Nano-mechanical properties such as modulus, deformation or adhesion can be quantitatively extracted from the measured force-distance curves down to atomic defect resolution. This technique can be complemented by Kelvin-probe force microscopy (KPFM) to extract electrical information such as the work function. Here, we present the first combination of optical, mechanical and electrical characterization techniques with sub-20nm resolution. We employ a hybrid system based on s-SNOM and peak-force tapping which is implemented in the Inspire instrument, built on the MultiMode AFM platform. The MultiMode AFM with peak-force tapping enables ultrahigh resolution imaging as demonstrated for atomic defects or the DNA double helix [3]. The s-SNOM imaging tool allows direct measurement of nanoscale absorption and reflection properties with up to 10 Hz imaging speed. In our presentation we will discuss how the different measurement modes are combined and we will give examples of the benefits of such a hybrid system. We address several questions ranging from graphene plasmonics to material distributions in polymers. For instance, Fig. 1 shows representative, quantitative data obtained on a PS-LDPE polymer blend, revealing softer and less reflective LDPE domains in the PS matrix. Amongst other examples, we highlight a biological application where amelogenin protein samples were investigated. Amelogenin is a protein that is critical to dental enamel formation [4,5]. In the presence of calcium and phosphate ions it self-assembles into ordered, selfaligned nanoribbon bundles, which suggests that it could form a template for phosphate-based apatite crystals that comprise dental enamel with similar ordering. To help clarify that open question, mapping the distributions of phosphate and hydroxyapatite nanocrystals within the bundles is necessary. With a nanoribbon width of <30nm, s-SNOM is the ideal tool to simultaneously map topography and chemical content. Figure 2 shows that the amelogenin bundles absorb at 1097cm-1 close to the phosphate resonance, whereas no absorption is observed at 1020cm-1 away from the phosphate absorption line. While the presence of phosphate could be identified using s-SNOM, no apatite nanocrystals with higher Microsc. Microanal. 21 (Suppl 3), 2015 1874 modulus than the ribbons could be distinguished in peak-force tapping in our preliminary data, indicating that for these in vitro preparation conditions apatite crystals have not formed yet (Fig. 3). We will also discuss a phosphorylated peptide derived from amelogenin amino acid sequence that assembles into ribbons without phosphate ions in solution. Here, s-SNOM is used to locate phosphate within the nanoribbon assembly, providing insight on the precise organization of the peptide molecules. In conclusion, we present a novel tool for combined near-field optical, nano-mechanical and nanoelectrical studies of heterogeneous material systems along with examples covering graphene plasmons, polymers and proteins involved in dental enamel formation. References: [1] F Keilmann and R Hillenbrand, Phil. Trans. R. Soc. Lond. A 362 (2004), p. 787. [2] B Pittenger, N Erina, and C Su, Bruker Application Note 128 (2011). [3] A Pyne et al., Small 10 (2014), p. 3257. [4] O Martinez-Avila et al., Biomacromolecules 13 (2012), p. 3494. [5] B Sanii et al., J Dent Res 93 (2014), p. 918. Figure 1. Quantitative nano-mechanical and nano-optical imaging of a PS-LDPE polymer blend. LDPE domains show higher deformation, lower modulus and lower reflection. Figure 2. Amelogenin nanoribbons in topography, infrared reflection and absorption at 1097cm-1 close to the phosphate resonance. Phosphate is concentrated in the nanoribbons and partially on the Si substrate from preparation. Figure 3. Topography and modulus map of Amelogenin nanoribbons. Hard apatite crystals are not present in the modulus image that reveals softer ribbons (dark areas) and hard Si substrate (bright areas).");sQ1[936]=new Array("../7337/1875.pdf","Quantitative Electron-Excited X-ray Microanalysis at Low Beam Energy","","1875 1875 doi:10.1017/S1431927615010156 doi:10.1017/S1431927615010156 Paper No. 0936 Microsc. Microanal. 21 (Suppl 3), 2015 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 � Microscopy Society of America 2015 Quantitative Electron-Excited X-ray Microanalysis at Low Beam Energy Dale E. Newbury and Nicholas W. M. Ritchie National Institute of Standards and Technology, Gaithersburg, MD 20899-8370 Quantitative electron-excited x-ray microanalysis performed at low beam energy (E0 5 keV) offers significant advantages: (1) The linear dimensions of the electron interaction volume scale approximately as E01.67 so that both lateral and in-depth spatial resolution are improved compared to the "conventional" beam energy range (10 keV to 30 keV). (2) Minimized x-ray absorption: As a direct consequence of the reduced beam penetration, x-ray absorption is significantly reduced and its impact on the analytical error budget is minimized. (3) For particle analysis, the greatly decreased electron range means that smaller particles can effectively be treated as bulk targets for quantitation. Instrumental advances have greatly improved low beam energy x-ray measurements: (1) Modern thermal field emission gun electron optics can deliver an analytically useful current into a beam diameter of 25 nm or less, minimizing the beam contribution to the limiting spatial resolution. (2) Large solid angle SDD-EDS systems, often consisting of arrays of detectors that can capture x-rays emitted over a broad range of azimuthal angle, enable collection of statistically robust spectra in reasonable measurement time that maximize the analytical information per unit electron dose. (3) The throughput of SDD-EDS enables collection of high count spectra, and the outstanding stability of SDD-EDS resolution and calibration with count rate provides the invariant peak shapes that enable successful multiple linear least squares peak fitting. Strongly interfering peaks can be accurately measured, even when the mutually interfering peaks have large ratios of relative intensity. Accurate SDD-EDS microanalysis performed with E0 from 5 keV to 10 keV of materials containing the low atomic number elements B to F, which must be measured with photon energy peaks below 1 keV, has been demonstrated [1]. These advances suggest that it is now a propitious time to further explore the limits of quantitation at low beam energy. Low beam energy x-ray microanalysis is subject to several challenges: (1) While low beam energy SEM imaging can be performed with landing kinetic energies of a few hundred eV or even lower, the minimum useful beam energy for x-ray microanalysis depends on the elements to be measured. For adequate excitation above the continuum background, it is desirable that the overvoltage, U = E0/Ec, where Ec is the critical ionization energy for the atomic shell of interest, is at least 2 for the highest Ec to be measured. Values of U in the range 1 < U 2 are of course possible, but the peak-to-background decreases sharply below U = 2. E0 = 5 keV is the lowest beam energy at which there is a measureable characteristic peak for the entire Periodic Table, except H and He. (2) The range of x-ray production depends strongly on Ec: R (nm) = (27.6 A)/(Z0.89 ) [E01.67 � Ec1.67] It is useful to construct an x-ray sampling depth axis parallel to the photon energy axis of the spectrum, as shown in Figure 1. In this case for 60Au-40Ag, AuM5-N7 (AuM5 = 2.21 keV) xrays are created to a depth of 74 nm, while AgL3-M5 (AgL3 = 3.35 keV) are created to a depth of 48 nm, meaning that there is a factor of 1.5 in linear dimension and 3.7 in volume over which the material being analyzed must be homogeneous to satisfy the basic requirement of the bulk quantification model. (3) The shallow sampling depth under low beam energy analysis conditions must be carefully considered when interpreting results. Surface oxide or other reaction layers and contamination that may be unnoticed at E0 = 20 keV can dominate the Microsc. Microanal. 21 (Suppl 3), 2015 Microsc. Microanal. 21 (Suppl 3), 2015 1876 1864 analysis results at low beam energy. This sampling depth effect is illustrated in Table 1, where the analysis of NIST SRM 481 (AuAg alloys) with various surface conditions is presented (kratio protocol with Au and Ag standards and NIST DTSA-II [2]). At a beam energy of 20 keV, the analysis of a polished but aged preparation of the four alloys shows anomalously low analytical totals but modest relative errors. Lowering the beam energy to 5 keV reveals that a significant surface corrosion layer exists containing S and Cl. After re-polishing to remove this layer, the analyzed values are found to correspond closely to the values on the SRM certificate. [1] D. Newbury and N. Ritchie, Microsc. Microanal., 20 (Suppl 3) (2014) 702. [2] N. Ritchie, DTSA-II available free at: www.cstl.nist.gov/div837/837.02/epq/dtsa2/index.html Table 1 Sample 20Au80Ag_20keV Relative error 40Au60Ag_20keV Relative error 60Au40Ag_20keV Relative error 80Au20Ag_20keV Relative error 20Au80Ag_5keV Relative error 40Au60Ag_5keV Relative error 60Au40Ag_5keV Relative error 80Au20Ag_5keV Relative error 20Au80Ag_5keV_Repolish Relative error 40Au60Ag_5keV_Repolish Relative error 60Au40Ag_5keV_Repolish Relative error 80Au20Ag_5keV_Repolish Relative error Raw sum 0.9304 0.9610 0.9690 0.9722 0.9855 0.9959 0.9951 1.005 1.005 0.9983 0.9897 0.9998 S (norm.) Not detected Not detected Not detected Not detected 0.0934 0.0311 0.0094 0.0051 Not detected Not detected Not detected Not detected Cl (norm.) Not detected Not detected Not detected Not detected 0.1061 0.0965 0.0450 0.0365 Not detected Not detected Not detected Not detected Ag (norm.) 0.7747 -0.15% 0.5821 -2.9% 0.3787 -5.1% 0.1850 -7.3% 0.721 -7% 0.5284 -12% 0.3706 -7.2% 0.2244 12% 0.7602 -2% 0.5955 -0.64% 0.3916 -1.9% 0.1946 -2.5% Au (norm.) 0.2253 0.46% 0.4179 4.4% 0.6213 3.5% 0.8150 1.8% 0.0796 -64% 0.3440 -14% 0.5754 -4.2% 0.734 -8.3% 0.2398 6.9% 0.4045 1.1% 0.6084 1.3% 0.8055 0.62% SRM 481: 60Au-40Ag E0 = 5 keV 0 0.5 1.0 1.5 2.0 Photon Energy (keV) 2.5 3.0 3.5 4.0 4.5 5.0 99 97 92 86 78 68 57 45 31 16 0 X-ray Production Range (nm) Figure 1. SDD-EDS spectrum of SRM481_60Au-40Ag at E0 = 5 keV with depth of x-ray production.");sQ1[937]=new Array("../7337/1877.pdf","Comparing the Intensities and Spectral Resolution Achieved by Wavelength-Dispersive Spectrometers on Electron Microprobes and SEMs.","","1877 doi:10.1017/S1431927615010168 Paper No. 0937 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Comparing the Intensities and Spectral Resolution Achieved by WavelengthDispersive Spectrometers on Electron Microprobes and SEMs. Stephen M. Seddio1 and John H. Fournelle2 1. Thermo Fisher Scientific, Fitchburg, WI, USA. 2. Department of Geoscience, University of Wisconsin-Madison, Madison, WI, USA. EMPs (electron microprobes) use WD (wavelength-dispersive) spectrometers that rely on curved diffractors and Rowland circle geometry. Today, EMPs are outfitted with as many as five WD spectrometers. Individual WD spectrometers are also available for SEMs. These spectrometers rely on either Rowland circle geometry or parallel beam geometry (hereafter, PB-WDS). The latter was developed to improve X-ray count rates in order to counter the generally lower beam current in an SEM relative to an EMP. The improved count rate in PB-WDS is the result of a much larger solid angle as a collimating optic is inserted deep into the SEM sample chamber during sample analysis. This study compares the X-ray intensities and spectral resolution achievable by the Rowland circle geometer, as mounted on an EMP and the PB-WDS geometry as singly mounted on an SEM. EMP measurements were made using the Cameca SX51 at the University of Wisconsin-Madison, which has five 160 mm diameter Rowland circle wavelength-dispersive spectrometers, three of which contain low-pressure P10 flow-though detectors and two of which contain high-pressure P10 flow-though detectors. SEM measurements were made using a Thermo ScientificTM MagnaRayTM PB-WDS, which contains a sealed Xe detector, mounted on a JEOL 7001F FE-SEM. Measurements were made under similar or, when possible, identical conditions (e.g., accelerating voltage, beam current, count times, and diffractor) on the same metal standards. In order to compare intensities across a wide range of wavelengths, X-ray intensities were measured on B K, C K, Al K, Si K, Ti K, Fe K, and Cu K. The diffractors used by the EMP were PC3 (MoB4C) for B; PC2 (NiC95) for B and C; PC1 (WSi60) for C; TAP for Al and Si; PET for Si and Ti; LiF200 for Ti, Fe, and Cu; and LiF220 for Cu. When possible, measurements were made concurrently using both high- and low-pressure P10 detectors on multiple spectrometers. If measurements were made on multiple spectrometers using the same diffractors and detector type (i.e., high- or low-pressure Ar), the reported intensities are those of whichever spectrometer yielded higher intensities. The diffractors used by the parallel beam WDS were MoB4C for B; NiC80 for B and C; WSi60 for C; TAP for Al and Si; PET for Si and Ti; and LiF200 for Ti, Fe, and Cu. A MagnaRay can concurrently contain six diffractors. The LiF220 diffractor was not available in the MagnaRay used in this study. The reported LiF220 Cu intensity for MagnaRay is the average of the intensities obtained by the factory from all MagnaRays containing LiF220 diffractors. In order to maximize count rates, detectors for both the EMP and the SEM were set to integral mode. Additionally, peak searches were done to assure that the spectrometers were positioned correctly to maximize count rates for a given element. Results are summarized in Figure 1 and Table 1.The PB-WDS yields dramatically higher intensities relative to the EMP for low-energy X-rays (e.g., B, C). The EMP generated higher intensities than the PB-WDS for Al and Si when measured on TAP. For higher energy X-rays (e.g., Ti, Fe, Cu), PB-WDS typically yields higher intensities compared to the low-pressure detectors of the EMP but lower intensities compared to the high-pressure detectors of the EMP. Figure 2 shows LiF wavescans from the MagnaRay (red) and EMP (blue). Microsc. Microanal. 21 (Suppl 3), 2015 1878 Figure 1. Measured intensities for selected elements. Diffrac Diffractor information is included. For EMP measurements, "low" and "high" refer to whether the intensities were measured with a low- or high-pressure detector. "LiF*" refers to a LiF220 diffractor. Figure 2. LiF wavescans of the Ti K and V K spectral region of a Ti-6Al-4V alloy. Microprobe data is blue. MagnaRay data is red. The overlap of the spectra is presented as purple for clarity. Table 1. WDS intensities measured using the EMP and SEM SX51 Diff. MoB4C NiC95 NiC95 WSi60 TAP TAP PET Det. kV N P10, low-P 10 6 P10, low-P 10 5 Ave Ave. 3218.7 384.0 3177.9 34.3 3014.6 3112.3 413.2 1530.2 1903.9 218.9 291.4 795.2 1668.9 544.8 677.2 1731.6 331.3 27.8 11.1 34.4 1.58 47.9 3.98 1.18 4.20 4.26 0.84 1.68 2.15 1.85 2.71 2.35 7.32 1.60 Diff. MoB4C NiC80 NiC80 WSi60 TAP TAP PET PET LiF LiF MagnaRay Det. kV N Ave. 9124.1 Xe 10 9 1619.3 4022.9 Xe 10 24 440.4 Xe 20 25 1390.2 1835.9 Xe 20 14 1418.4 Xe 20 13 1930.6 Xe 20 13 612.8 216.2 43.4 68.6 8.66 15.3 8.61 17.7 13.9 9.49 B Ka C Ka Al Ka P10, low-P 10 7 Si Ka Ti Ka Fe Ka Cu Ka P10, low-P 20 8 P10, low-P 10 20 P10, high-P 5 P10, low-P PET P10, high-P 20 7 P10, low-P LiF P10, high-P P10, low-P LiF 30 8 P10, high-P LiF220 P10, low-P LiF 30 4 P10, high-P LiF220 Xe 30 9 1164.3 5.32 LiF Xe 30 8 848.6 7.58 "Diff." refers to "diffractor." "Det." refers to "detector." "kV" refers to "accelerating voltage". "N" refers to the number of measurements averaged. "Ave." refers to the average of the measured intensities. " is the standard deviation of the measured intensities. """);sQ1[938]=new Array("../7337/1879.pdf","Evaluation of Combined Quantification of Cr-Ni Steel using EDS and WDS","","1879 doi:10.1017/S143192761501017X Paper No. 0938 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Evaluation of Combined Quantification of Cr-Ni Steel using EDS and WDS Philippe T. Pinard1, Ralf Terborg2, Tobias Salge3 and Silvia Richter1 1 2 Central Facility for Electron Microscopy, RWTH Aachen University, Aachen, Germany. Bruker Nano GmbH, Berlin, Germany 3 Core Research Laboratories, Natural History Museum, London, UK Classically, wavelength dispersive spectrometers (WDS) are associated with electron microprobe whereas energy dispersive spectroscopy (EDS) is the prominent technique to detect X-rays in scanning electron microscopes (SEM). This distinction is however less clear nowadays with electron microprobes equipped with EDS silicon drift detector and SEMs with WDS. As each spectrometer has advantages and limitations (e.g. resolution, count rate performance, detection efficiency, speed of measurement), there is an obvious advantage to combine both techniques to analyze as quickly and accurately as possible a given sample. The question is therefore which spectrometer should be used to measure which elements and under which analytical conditions. A series of 15 steel samples from Acerinox, S.A. were selected for this study, as they contain both major (Cr, Fe and Ni) and minor elements (Si, Mn, Co, Cu and Mo) with various concentrations. Measurements were simultaneously performed on a JEOL JXA-8530F microprobe equipped with 5 WDS and a Bruker XFlash EDS (FWHM MnK <125 eV) operated using the Probe for EPMA software. As pointed out by Ritchie et al. [1] the optimal beam current and counting time for EDS and WDS differ. To accommodate both techniques, measurements were performed using an accelerating voltage of 15 kV and beam currents of 8 nA (EDS: 120 s, longest shaping time; WDS: 30 s peak, 15 s background) and 80 nA (EDS: 40 s, shortest shaping time; WDS: 10 s peak, 5 s background). Measurements were also performed at 6 kV and 20 nA, using the L lines for the transition element metals. In order to apply standard based quantification pure element standards were acquired at 15kV, 8nA. For the 15 kV, 80 nA samples the 8 nA standards were used. EDS and WDS had the same accuracy for Cr, Fe and Ni (> 10 wt%), regardless of the analytical conditions used. This was not the case for Mn where EDS results at 80 nA were significantly worse than at 8 nA, even for high Mn containing samples (> 5 wt.%). Both EDS and WDS experienced problems for samples with low Mn (< 1 wt.%) (Fig. 1). We suspect that the interference of the Cr K� and, inadequate background positions, in the case of WDS, could explain the larger errors. The quantification of Si and Mo with concentrations greater than 0.25 wt.% gave consistent results, although the WDS values were systematically lower (Fig. 2). Potential peak shift between unknowns and standards should be checked. At 80 nA, WDS could accurately detect Co and Cu concentrations above 0.1 wt.%. Note that correcting the interference of the Fe K� on the Co K line was required. EDS was unable to correctly measure Co (max. conc. 0.34 wt.%). Accurate concentrations of Cu were only obtained above 2 wt.% (Fig. 1). The influence of the background should be investigated. At low beam energy, 6 keV, the results were worse for both techniques. As reported in other works [2,3] inaccuracy of the L-line quantification lead to large errors. Similar errors were obtained for both techniques (similar k-ratios). EDS gave slightly better results for major elements, but WDS had the edge for minor elements. These preliminary results show the complementarity of EDS and WDS to quantify major, minor and trace elements in steel samples. Both techniques have advantages and limitations. Further analysis and improvements will be presented. Microsc. Microanal. 21 (Suppl 3), 2015 1880 [1] N.W.M. Ritchie et al, Microsc. Microanal. 18 (2012), pp. 892-904. [2] X. Llovet et al, IOP Conf. Series: Materials Science and Engineering 32 (2012), p. 012014. [3] Pinard et al, Micros. Microanal. 20 S3 (2014), pp. 700-701. Fig. 1: Relative errors on composition of Mn, Co and Cu as a function of the nominal concentration from measurements at 15 kV, 8 nA. Fig. 2: Relative errors on composition of Si and Mo as a function of the nominal concentration from measurements at 15 kV, 8 nA. Fig. 3: k-ratios of Cr, Fe and Ni from measurements at 6 kV, 20 nA.");sQ1[939]=new Array("../7337/1881.pdf","Comparison of WDS and EDS Rare Earth Element Analysis","","1881 doi:10.1017/S1431927615010181 Paper No. 0939 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Comparison of WDS and EDS Rare Earth Element Analysis H. A. Lowers1 and P. K. Carpenter2 1 Central Minerals and Environmental Resources Science Center, U.S. Geological Survey, Denver, CO 80225 USA 2 Department of Earth and Planetary Sciences, Washington University, Saint Louis, MO 63130 USA Demand for rare earth elements (REE) is on the rise as more uses for these materials are found in the technology and energy sectors. Electron microscopy coupled with wavelength-dispersive spectroscopy (WDS) and energy-dispersive spectroscopy (EDS) analysis allows in situ analysis of REE containing minerals from prospective mineral deposits to help determine its potential as an economic resource. The analysis of REE minerals is complicated due to on- and off-peak interferences, background selection, availability of standards, and the number of elements present to analyze. We set out to compare standards and standardless based EDS to WDS to determine the best option in terms of time of analysis, accuracy, and element detection, to test EDS detector performance for REE analysis, and to evaluate the standard materials used for REE quantitative analysis. Standards-based EDS analyses were acquired at 15 kV, 50 nA probe current, 20 micron defocused beam diameter, and 200-600 seconds live count time on a JEOL 8200 electron microprobe with a 10 mm2 e2v Gresham silicon drift detector at Washington University. WDS analyses with off-peak and mean atomic number (MAN) background corrections were also acquired on the same instrument utilizing the same operating conditions. Natural and synthetic mineral standards, Edinburgh REE glasses [1], Roeder REE glasses [2], Drake and Weill REE glasses [3], and Smithsonian REE orthophosphates [4] were used for calibration and accuracy checks for both EDS and WDS analysis. The Edinburgh REE glasses and Smithsonian orthophosphates contain a single rare earth element in a Ca-Al-Si-O matrix and PO4 matrix, respectively. The Drake and Weill REE glasses contain multiple REEs, usually four, which were chosen to reduce to the number of peak interferences on the WDS spectrometer. A comparison of the measured k-ratio (k-measured) on the unknown using a multielement standard to the theoretically calculated k-ratio (k-calculated) has been used as a nonbiased method for evaluating the accuracy of the analysis. K-calculated is obtained by division of the elemental k-ratio for each secondary standard by the elemental k-ratio for the primary standard using k-ratio values calculated using the Armstrong (z) of the CalcZAF program [5]. The ratio k measured/k calculated is plotted versus elemental weight percent in figure 1. A value of one on these plots represents perfect agreement between the measured and observed values regardless of element concentration. The EDS data presented in Figure 1 indicate excellent agreement with the Drake and Weill REE standards but an overestimation of the measurement for the Edinburgh glasses due to pulse-pileup peaks in the REE L-family range which are not corrected for on the EDS system used. A wide dispersion of the ratio for the ~1 and ~0.1 wt% Roeder REE glasses indicates decreased accuracy by EDS at low concentrations but also possible errors in the accepted values for the REE in these trace element glasses. The accurate measurement of REE by EDS thus requires accurate removal of pulse-pileup artifacts and long count times for precise measurement at low concentrations. The off-peak background-corrected WDS data presented in Figure 1 reveals reasonable agreement among the standards with the exception of Ce, Er, and Tm in the Edinburgh REE glasses and the low level Roeder glasses. These inconsistencies suggest an error in the accepted element concentrations of these materials. The error in measurement of Tb in the Drake and Weill glass is due to an off-peak Microsc. Microanal. 21 (Suppl 3), 2015 1882 interference (Eu) and is eliminated with the MAN background correction (fig. 1). This k-ratio evaluation method allows inspection of data, reveals background placement issues, and possible problems with standard compositions regardless of the matrix correction scheme used. References [1] http://www.geos.ed.ac.uk/facilities/electron/REEStandards/ [2] P L Roeder, Canadian Mineralogist 23 (1985), p. 263. [3] M J Drake and D F Weill, Chem Geolog, 10 (1972), p. 179. [4] E Jarosewich and L A Boatner, Geostandards Newsletter, 15 (1991), p. 397. [5] http://probesoftware.com/Technical.html Figure 1. K-ratio comparisons of REE elements calibrated relative to USNM REE orthophosphates using EDS (top), WDS with off-peak background correction (left) and WDS with MAN background correction (right).");sQ1[940]=new Array("../7337/1883.pdf","Real-Space Simulation of Electron Scattering in Imperfect Crystals and Reconstruction of the Electrostatic Potential.","","1883 doi:10.1017/S1431927615010193 Paper No. 0940 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Real-Space Simulation of Electron Scattering in Imperfect Crystals and Reconstruction of the Electrostatic Potential. Wouter Van den Broek1 and Christoph T. Koch1. 1. Institute for Experimental Physics, Ulm University, Ulm, Germany. The Bloch-wave approach [1] is a powerful and efficient tool for the simulation of diffraction patterns of ideal crystals. However, when deviations from perfect crystallinity are present, like strain fields, defects or interfaces for instance, only approximate solutions are possible anymore. The multislice algorithm [1] (MSA) is preferable in these cases because it does not require any additional approximations for imperfect structures since it is based in real space. In order to achieve diffraction data up to high angles while at the same time resolving fine details like HOLZ lines, a small sampling distance and a large extent must be combined in real space. Furthermore, to accommodate to arbitrary specimen tilts or to the non-periodicity brought about by strain fields, defects or interfaces, small slicing intervals of the order of 0.01 to 0.1 nm are necessary. As a result, the demands on computer memory and processing speed are high. The improvement of graphics processing units (GPUs) in the last years has allowed to combine these two demands. Here, our forward dynamical electron scattering (FDES) software [2] is used. FDES is written in the CUDA programming language, with the CUFFT, CUBLAS and CURAND libraries and can simulate any combination of specimen-tilt series, beam-tilt series or focal series of high resolution transmission electron microscopy (HRTEM) images, diffraction patterns or convergent beam electron diffraction (CBED) patterns. Thermal diffuse scattering is accounted for with the frozen phonon approach. In Figure 1 FDES's capabilities are demonstrated by computing CBED patterns of a PbSe-CdSe coreshell particle with 1963 atoms. The cubic particle's structure is based on CdSe rock-salt. The unit-cell parameter, a, measures 0.61 nm. Pb-atoms are replacing the Cd-atoms in the cube's interior octahedron and are displaced by [-a/4, a/4, a/4]. The acceleration voltage is 40 kV and the convergence semi-angle is 15 mrad. The specimen is 1000 by 1000 pixels of 0.01 nm wide, resulting in a 0.1 nm-1 sampling in reciprocal space. The slice thickness is 0.01 nm. An average is taken over 50 frozen phonon configurations; the computation time is 247 s on a NVIDIA Tesla K20 GPU, i.e. less than 5 s per configuration. The Debye-Waller factors for Se, Cd and Pb equal 0.0118, 0.0157 and 0.0217 nm2 respectively. Preliminary results have shown that samples of over a million atoms can be simulated as well. Furthermore, FDES is an ideal test ground for inverse dynamical electron scattering (IDES), introduced by the authors in [3,4], where the object's three-dimensional electrostatic potential distribution can be retrieved from a tilt series of TEM recordings, for example from HRTEM, CBED or scanning confocal electron microscopy. Since IDES is based on the reformulation of MSA as an artificial neural network, it takes the dynamical scattering of the electrons into account exactly. A HRTEM tilt series is simulated of the PbSe-CdSe core-shell particle described above. The alpha tilt is varied from -10� to +10� in 2� increments while the beta tilt is held at 0�. Then the beta tilt is varied Microsc. Microanal. 21 (Suppl 3), 2015 1884 from -10� to +10� in 2� increments while the alpha tilt is held at 0�. The acceleration voltage is 80 kV, the focus and spherical aberration are -20 nm and 64.2 m, respectively, resulting in a point resolution of 0.17 nm. The specimen is 600 by 600 pixels of 0.025 nm wide and 250 slices of 0.02 nm tall. The dose is 100 electrons per pixel. Partial spatial coherence and the CCD's modulation transfer function are included as well. In order to save computation time, the IDES reconstruction is carried out with coarser parameters than the forward simulation: a slice thickness of 0.1 nm and an approximation to the specimen tilt by a shifted Fresnel propagator instead of the exact crystal tilts in FDES. Despite these coarser settings, an atomic-resolution reconstruction was obtained. Furthermore, 1-norm regularization is applied to the object and sparseness is increased with a generalized potential. Mass contrast was observed in the reconstruction so that by applying a threshold to the reconstructed intensities, the Pb-core could be isolated; see Figure 2. MSA is preferable for the simulation of imperfect crystalline structures. To meet the computational demands which come with these real-space calculations we presented GPU-based MSA software that is capable of accurate and fast computation of finely sampled large specimens. Its capabilities were demonstrated with the simulation of a CBED pattern of a core-shell particle and the simulation of a HRTEM tilt-series that was subsequently used in the three-dimensional reconstruction routine IDES. [5] References: [1] EJ Kirkland in "Advanced Computing in Electron Microscopy", (Springer, New York) p. 115. [2] W Van den Broek and CT Koch, IMC Proceedings (2014), IT-16-P-2748. [3] W Van den Broek and CT Koch, Phys. Rev. Lett. 109 (2012), p. 245502. [4] W Van den Broek and CT Koch, Phys. Rev. B 87 (2013), p. 184108. [5] The authors acknowledge the Carl Zeiss Foundation and the German Research Foundation (DFG, Grant No. KO 2911/7-1). Figure 1. Simulation of a (001) PbSe-CdSe coreshell particle without (left) and with (right) thermal diffuse scattering included. Logarithmic gray scale. Figure 2. Left, the Pb-core of the PbSe-CdSe coreshell particle. Right, the reconstructed core, retrieved by thresholding and thus demonstrating mass contrast.");sQ1[941]=new Array("../7337/1885.pdf","Simulating Inelastic Scattering in Scanning Transmission Electron Microscopy using STEM","","1885 doi:10.1017/S143192761501020X Paper No. 0941 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Simulating Inelastic Scattering in Scanning Transmission Electron Microscopy using �STEM A.J. D'Alfonso1, S.D. Findlay2 and L.J. Allen1 1. 2. School of Physics, University of Melbourne, Parkville, Victoria 3010, Australia School of Physics and Astronomy, Monash University, Clayton, Victoria 3800, Australia The ability to image specimens of condensed matter at atomic resolution has vastly expanded our understanding of advanced materials and their properties. However, as electron scattering is a highly complicated process much attention has been directed toward developing our understanding of imaging formation and interpretation. The simultaneous measurement of high angle annular dark field and electron energy loss spectroscopy images has been one approach toward improved interpretation [1], especially if chemical specificity is desired [2]. However, even with multimode imaging, it is widely acknowledged that in order to extract the maximal information present in atomic resolution images, or possibly to reveal potential ambiguities in image appearance, accurate simulations of the imaging process are critical [3]. We will discuss the application of image simulation to materials design problems, emphasizing the importance of using sophisticated simulations and models, coupled with careful experiment, to place electron microscopy on an absolute scale. One microscopy simulation suite, which is freely available for download, capable of simulating images based on elastic and inelastic phonon scattering, inner-shell ionization and subsequent x-ray generation is �STEM [4]. The �STEM suite is based on the multislice algorithm implemented using fast Fourier transforms. The advantages of this approach are that it is computationally efficient and lends itself naturally to parallelization. �STEM is available for both GPU and CPU computing models which allows detailed simulations to be performed rapidly on a desktop computer. Figure 1, taken from Ref. [5] shows a selection of imaging modes that may be simulated using �STEM, including those based on inner-shell ionization. �STEM is not limited to simulating periodic specimens, like that indicated in Fig. 1. As an example, consider the situation depicted in Figure 2(a) of a cerium dioxide nanoparticle illuminated by a defocussed STEM probe. Figure 2(b) shows an experimentally recorded Ronchigram (or Gabor Hologram) while Figure 2(c) shows the corresponding simulation. The experimentally recorded Ronchigram was subsequently used to reconstruct the phase image of the complex specimen transmission function, Fig. 2(d). Confirmation that the reconstruction process gave a unique solution was verified through forward simulation, Fig. 2(e). In this case, combining simulation and experiment was critical to confirm the experimental reconstruction. [1]. D.A. Muller, et al, Nature, 399, (1999), 758. [2]. P. Wang, et al, Phys. Rev. Lett. 101, (2008), 236102. [3]. C.L. Jia, et al. Nature materials, 13, (2014), 1044-1049. [4]. http://tcmp.ph.unimelb.edu.au/mustem/muSTEM.html [5]. L.J. Allen, A.J. D'Alfonso and S.D. Findlay. Ultramicroscopy, in press, (2014). [6]. This research was supported under the Australian Research Councils Discovery Projects funding scheme (Projects DP110102228 and DP140102538) and its DECRA funding scheme (Project DE130100739). Microsc. Microanal. 21 (Suppl 3), 2015 1886 1874 Figure 1 Images calculated using STEM for 100 keV electrons on a 300 � thick specimen of [0001] Si3N4. (a) Exit surface intensity due to elastically scattered electrons and (b) exit surface intensity due to thermally scattered electrons for plane wave illumination. In (c) and (d) the intensities in parts (a) and (b) after imaging by an aberration free lens with an aperture of 25 mrad are shown (e) A position averaged convergent beam electron diffraction pattern calculated using an aberration-free, coherent probe formed using an aperture of 9.6 mrad with the average taken over the unitcell. (f) The annular bright-field image and (g) a high-angle annular dark-field image, using a probe formed using a 25 mrad aperture. Inner and outer angles are given in the text. An elemental map based on the energy-dispersive x-ray signal for the Si K edge is shown in (h) and for the N K edge in (i), using a probe forming aperture of 25mrad. Each image is displayed on its own contrast scale. Figure 2 (a) Probe sample configuration, and the corresponding (b) experimental and (c) simulated Ronchigrams. Reconstructed phase image of the specimen transmission function from (d) experiment and (e) simulation. The incident 300 keV electron probe was formed using an aperture of 24 mrad and a defocus of -80 nm.");sQ1[942]=new Array("../7337/1887.pdf","Prospects for Detecting Single Vacancies by Quantitative Scanning Transmission Electron Microscopy","","1887 doi:10.1017/S1431927615010211 Paper No. 0942 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Prospects for Detecting Single Vacancies by Quantitative Scanning Transmission Electron Microscopy Jie Feng1, Alexander V. Kvit2, Andrew B. Yankovich1,2, Chenyu Zhang1,2, Dane Morgan1,2, Paul M. Voyles1,2 1. 2. Materials Science Program, University of Wisconsin-Madison, Madison, WI, 53706, USA Department of Materials Science and Engineering, University of Wisconsin-Madison, Madison, WI, 53706, USA Transmission electron microscopy (TEM) and scanning TEM (STEM) have made significant advances in direct imaging of point defects, including substitutional impurities, interstitial impurities, and selfinterstitials. However, imaging defects that decrease, rather than increase local intensity, such as vacancies, remains a significant challenge. Here we show in simulations that recent advances in highly quantitative, picometer-precision STEM imaging [1] may make it possible to detect single cation vacancies in a complex oxide both through the reduction of the column intensity and the small shifts in the neighboring atomic column positions. We use frozen phonon simulations to explore if vacancies in LaMnO3 perovskites can be imaged in the STEM [2]. At least 128 phonon configurations are required to ensure sub-picometer special precision and lower than 1% intensity fluctuations in simulations of 20 nm thick [100] LaMnO3. Simulation parameters were set based on a probe-corrected FEI Titan STEM at 200 kV. The probe convergence semi-angle was 24.5 mrad and the detector inner angles were 84.4 mrad and 8.48 mrad for high-angle annular dark-field (HAADF) STEM and annular bright field (ABF) STEM, respectively. Figure 1(a) shows the 3D structure of LaMnO3 (Pnma space group) and the 2D projection along [100]. There are two symmetry distinct near-neighbor La column separations along the x direction and one along the y direction. Figure 1(c) shows the vacancy visibility, defined as the percent reduction in intensity of vacancy-containing column with respect to a perfect column, which is 3% ~ 9% for a single La vacancy in a 10 nm thick sample and 1% ~ 6% visibility in a 15 nm thick sample. Figure 1(d) and (e) show that all three inter-column separations are measurably changed by a single La vacancy in both HAADF image and ABF image, and that all separation changes depend on the vacancy depth. The depth dependence may make it possible to not only detect a single La vacancy, but also accurately localize its depth. A similar analysis predicts a 1% ~ 3% reduced intensity and pm-scale contraction/expansion caused by a single Mn vacancy under the same conditions. The depth dependencies of vacancy introduced visibility and column displacements are likely caused by electron channeling effects [3]. The predicted changes are within the detection limitations of HAADF STEM imaging in experiment, which can now achieve sub-picometer precision and 1% intensity sensitivity by non-rigid registration and averaging of an image series [1]. Overall, these results predict that single cation vacancy imaging is possible in LaMnO3. Experiments, performed on [100] LaMnO3 film grown on DyScO3 substrate, have identified candidate La vacancies in STEM images, as shown in Figure 2. Detailed comparison of experiment and simulations may in the future enable full, 3D reconstruction of the defect structure in LaMnO3. Microsc. Microanal. 21 (Suppl 3), 2015 1888 References: [1] A. B. Yankovich, et al., Nat. Commun. 5 (2014), p. 4155. [2] E.J. Kirkland in "Advanced Computing in Electron Microscopy", 1st Edition, Springer, 1998. [3] P. M. Voyles, et al., Ultramicroscopy. 96 (2003), p. 251. [4] This work was supported by the US Department of Energy, Basic Energy Sciences, Grant DE-FG0208ER46547. Figure 1. Simulations for LaMnO3 [100]: (a) unit cell in perspective (left) and 2D projection (right). (b) Simulated HAADF (top) and ABF (bottom) images. (c) Intensity visibility for VLa as a function of vacancy depth for two STEM sample thicknesses. (d) Atomic column displacements in a 10 nm thick sample from HAADF images. (e) Atomic column displacements in a 10 nm thick sample from ABF images. Data inside shaded regions are not detectable. Separations are defined in (a). Figure 2. Experimental HAADF images of [100] LaMnO3: (a) visibility map and (b) intensity map. The column marked in the red box is likely to contain a single La vacancy.");sQ1[943]=new Array("../7337/1889.pdf","Propagation of Bessel Beams along Atomic Columns in Crystal: a Bloch Wave and Multi-slice Analysis","","1889 doi:10.1017/S1431927615010223 Paper No. 0943 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Propagation of Bessel Beams along Atomic Columns in Crystal: a Bloch Wave and Multi-slice Analysis Vincenzo Grillo1,2, Enzo Rotunno2, Benjamin McMorran3, Stefano Frabboni1,4 1. 2. CNR-Istituto Nanoscienze, Centro S3, Via G. Campi 213/a, I-41125 Modena, Italy CNR-IMEM Parco Area delle Scienze 37/A, I-43124 Parma, Italy 3. Department of Physics, University of Oregon, Eugene, Oregon, USA 4. Dipartimento FIM, Universit� di Modena e Reggio Emilia, Via G. Campi 213/a, I-41125 Modena, Italy After the recent introduction of holographic methods to produce Bessel beams, it is becoming interesting to understand the theoretical and applicative potentialities of these beams [1]. One of the most interesting properties of the Bessel beams can be tight back to the fact that they are the Fourier transform of a very small annulus in the Fourier space and are therefore similar to hollow cone illumination [2]. However while hollow cone generation is restricted to 0th order Bessel beams, the holographic approach permits to induce a spiraling phase, as in the most common Vortex beam generated by pitch fork holograms. We report on a detailed analysis on the propagation of Bessel beams of different orders on a Ga atomic column in a [100] oriented GaN crystal. The beam average convergence was 15 mrad and the beam energy was 300KeV. The analyses are based on the comparison between Bloch wave and multislice simulations. Considering the simplicity of the Bessel beam momentum spectrum and the symmetry of the material it is possible to give a description of the propagation as a function of a few states. This analysis permits a deeper understanding of the channeling phenomena and of the probe intensity oscillation in the propagation direction. For example in fig 1a,b,c, we can observe the analysis for the 0th order Bessel beam. The probe at the interaction in fig 1a is represented before the interaction. In fig 1b the intensity of the central part of the probe are plotted against the depth. Fig 1c is an effective excitation distribution as a function of the Bloch eigenvalue (all excitations for the same energy/eigenvalue are summed coherently). The probe intensity oscillation in fig 1b is more prominent than what observed in common aperture limited probes. The persistence of oscillation can be explained by observing in fig 1c that the oscillation are the beating between a small number of states (typical of Bessel beam narrow lateral spectrum) and the 1s state. Fig 2 a,b,c and fig 3a,b,c are similar sequence of probe shape, intensity evolution with depth and Bloch spectrum for beams with topological charge q= 1,2. In this case the symmetry selection rules forbid the excitation of the 1s. In the case of q=1 the resulting spectrum shows many peaks and therefore practically no oscillation. In the case of q=2 the selection rules and the limited momentum spectrum allow for only few excited states. As a consequence oscillations are again very clear. Bessel beam can be therefore a useful way to engineer the depth profile of the probe for possible depth resolved experiments. Microsc. Microanal. 21 (Suppl 3), 2015 1890 1878 References: [1] V. Grillo, E. Karimi, G C Gazzadi, S Frabboni,M R. Dennis, and R W. Boyd Phys. Rev. X 4, 011013 (2014) [2] T. Kawasaki, T. Matsutani, T. Ikuta, M. Ichihashi, T. Tanji Ultramicroscopy 110 (2010) 1332�1337 [3] E. Rotunno, M. Albrecht, T. Markurt, T. Remmele and V.Grillo 146 (2014) 62 1s Figure 1. a) Image of the Bessel probe L=0 ( hue encodes the phase ). b) Multislice simulation of the probe intensity on the atomic column c) Excitation of the Bloch states vs eigenvalue . Figure 2. a) Image of the Bessel probe L=1 ( hue encodes the phase ). b) Multislice simulation of the probe intensity on the atomic column c) Excitation of the Bloch states vs eigenvalue Figure 3 a) Image of the Bessel probe L=2 ( hue encodes the phase ). b) Multislice simulation of the probe intensity on the atomic column c) Excitation of the Bloch states vs eigenvalue");sQ1[944]=new Array("../7337/1891.pdf","Analysis of Dislocation Densities using High Resolution Electron Backscatter Diffraction","","1891 doi:10.1017/S1431927615010235 Paper No. 0944 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Analysis of Dislocation Densities using High Resolution Electron Backscatter Diffraction Arantxa Vilalta-Clemente1, Jun Jiang2, Ben Britton2, David M Collins1 and Angus Wilkinson1 1. 2. Department of Materials. University of Oxford, Parks Road, Oxford, OX1 3PH Department of Materials, Imperial College London, Prince Consort Road, London, SW7 2AZ, UK Cross-correlation analysis of electron backscatter diffraction (EBSD) patterns allows measurement of elastic strain and lattice rotation variations at a sensitivity of ~�10-4 [1]. Cross-correlation is used to measure shifts between sub-regions of test and reference patterns and simple geometry allows the elastic strains and lattice rotations to be calculated from the measured dispersion of pattern shifts. In deformed metals the lattice rotations are often significantly larger than the elastic strains and in these situations a pattern `remapping' approach has proved necessary to avoid artefacts in the strain fields [2]. One area of application for EBSD has been the determination of dislocation densities. The most explored route for quantifying the dislocation density has been to use the relationship, described by Nye [3] between the geometrically necessary dislocation (GND) density and the lattice curvature. In Nye's analysis the excess density of dislocations in the crystal is directly related to the gradient of the lattice rotation field that induced. Unfortunately solution of the reverse problem of finding dislocation densities from the measured rotation gradients often does not have a unique solution. This situation is exacerbated by the fact that only 6 of the nine possible rotation gradient terms can be established from EBSD on a single section. Despite these issues, a lower bound estimate of the total dislocation density can be made (though the densities of particular dislocation types are ambiguous). Figure 1 shows an example map of GND density distribution within a Cu polycrystal deformed to 10% tensile strain. Of course the GND density is only a fraction of the total dislocation density because any dipoles or multipoles between the measurement points cause no measureable rotation gradient. This leads to differences in the GND density as the step size is varied. Figure 1 shows the GND density recovered in the same region but with the rotation gradients calculated using three different length scales. This is shown in a more quantitative way in figure 2 which shows how the average GND density recorded in this map reduces as the effective step size is made progressively larger [4]. To estimate the total dislocation density a new approach has been recently proposed [5]. This borrows from peak profile analysis used in X-ray and neutron diffraction assessment of dislocation density. From the EBSD data the variations of stress from the mean value in each grain can be calculated and used to form a plot showing the probability of obtaining a given stress level. In all dislocated crystal we have analyzed the probability has the form of a central Gaussian-like part but has tails showing higher probabilities at the high stress levels with the probability following () = /. The form of these tails is consistent with the high stresses being generated by the localized stresses close to isolated dislocation cores, and the magnitude of the proportionality constant A can be used to determine the dislocation density. As the tails correspond to low probability, where experimental data tends to be somewhat noisy, rather than fitting the data directly we follow Groma's method [5] developed for analysis of X-ray diffraction peak intensities of using the restricted second moment V2 of the probability For the large shear stress xy close to an isolated edge dislocations with Burgers vector magnitude b along the x-axis and line direction along the z-axis normal to the sample surface of an isotropic elastic medium of shear modulus G and Poisson ratio, , it can be shown that () = () d Microsc. Microanal. 21 (Suppl 3), 2015 1892 Thus at large stresses we expect a plot of V2 against ln() to tend toward a straight line, and the gradient of this is related to the total dislocation density [6]. Figure 3 shows some examples of such plots for a ferritic steel in the annealed condition and after cold rolling to various reductions, and the total dislocation density obtained from the gradients is shown in the figure. Further evidence that the high stress values obtained at local positions by EBSD mapping is due to the incident beam being very close to dislocation positions is found by comparing the electron channeling contrast imaging (ECCI) and EBSD results. Figure 4 makes this comparison for a GaN sample which contains threading dislocations revealed as black/white dots in ECCI (figure 4a). The corresponding EBSD generated elastic strain fields show local `spikes' in the vicinity of the dislocations (xy map in figure 4b). () = ) ln( ( ) Figure 1: GND density maps showing log(GND in m-2) in the same region of a Cu sample deformed to 2% mapped at different effective step sizes. Figure 2: GND density variation with step size for Cu samples deformed to 2% and 10%. Figure 3: Plots of restricted second moment of stress probability distribution against ln(stress) from which the total dislocation density can be determined. Data for ferritic steel before and after cold rolling. Figure 4: (a) ECCI and (b) EBSD maps showing localized elastic strain ( xy) `spikes' near threading dislocations in GaN. References: [1] A.J. Wilkinson, G. Meaden, and D.J. Dingley, Ultramicroscopy 106 (2006), p. 307. [2] T.B. Britton and A.J. Wilkinson, Ultramicroscopy 114 (2012), p. 82. [3] J.F. Nye, Acta Metallurgica 1 (1953), p. 153. [4] J. Jiang, T.B. Britton and A.J. Wilkinson, Ultramicroscopy, 125 (2013), p. 1. [5] I. Groma, Physical Review B 57 (1998), p. 7535. [6] A.J. Wilkinson et al, Applied Physics Letters 105 (2014), p. 181907. [7] We acknowledge ESPRC funding under grants EP/K034332/1, EP/J016098/1, & EP/I021043/2.");sQ1[945]=new Array("../7337/1893.pdf","Nye Tensor Dislocation Density Mapping From Precession Electron Diffraction: Effects of Filtering and Angular Resolution","","1893 doi:10.1017/S1431927615010247 Paper No. 0945 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Nye Tensor Dislocation Density Mapping From Precession Electron Diffraction: Effects of Filtering and Angular Resolution A.C. Leff1, C.R. Weinberger2 and M.L. Taheri1 1. 2. Department of Materials Science and Engineering, Drexel University, Philadelphia, PA Department of Mechanical Engineering and Mechanics, Drexel University, Philadelphia, PA The Nye Tensor describes the geometrically necessary dislocation (GND) density in terms of the number of dislocations required to accommodate a given contortion in the crystal lattice. It has been used extensively in recent years to quantify GND densities from data acquired using electron backscatter diffraction (EBSD) in a scanning electron microscope [1,2]. Thanks to the development of precession electron diffraction automated crystallographic orientation mapping (PED-ACOM) for transmission electron microscopes (TEM) [3], the same type of spatially resolved orientation data produced using EBSD can now be acquired at a much smaller length-scale. By utilizing nano-scale orientation data to calculate GND densities, the detailed dislocation structure of deformed metals can be observed and quantified. In [4] a methodology for applying Nye tensor analysis to map GND densities from PED-ACOM data is presented. In the present work, this technique is demonstrated with a focus on the effects of angular resolution and data filtering on the dislocation density values calculated. PED-ACOM patterns were acquired and indexed using a JEOL 2100 LaB6 TEM equipped with NanoMEGAS SPINNING STARTM precession electron diffraction and ASTARTM ACOM systems with approximately 0.5� angular resolution and a step size of 10.4 nm. Five out of nine Nye tensor components are calculated and used to estimate the total GND density at each point in the scan using a least-squares method to provide the most continuous curvature between the orientation of a given point and its eight nearest neighbors. Fig. 1 shows GND density maps for a nanocrystalline Fe thin film (A,C,E,G) and a jet-thinned Cu TEM specimen (B,D,F,H). The average density values produced using this method are on the order of 1015 m2, higher than those reported from EBSD [1,2]. Three different methods of enhancing the dislocation maps are evaluated in order to determine what role, if any, noise due to comparatively poor angular resolution has on the values produced by this method. A low-pass Gaussian Fourier filter is used to remove high frequency noise such as that produced by from the calculated GND density values. This reduces the noise in the data visibly, causing a decrease in average density on the order of 10 11 m-2, well below the range of density values present. A Kuwahara filter is applied to the orientation data prior to the calculation of the Nye tensor. This results in substantive, qualitative changes to the maps produced and is therefore not recommended. An interpolation method [5] is used to enhance the angular resolution of the indexing software to approximately 0.3�. Improving the angular resolution in this fashion causes some change in the average density calculated for a given scan area, as would be expected when reindexing orientation data using different parameters; it does not result in a systematic decrease in density values, however, as would be expected if noise due to lack of angular resolution were causing the GND density values to be significantly inflated. Microsc. Microanal. 21 (Suppl 3), 2015 1894 References: [1] A.J. Wilkinson, D. Randman, Philosophical Magazine. 90 (2010) p. 1159. [2] T. Hardin, B.L. Adams, D.T. Fullwood, R.H.Wagoner, Mater. Sci. Forum. 702�703 (2011) p. 492. [3] E.F. Rauch and M. Veron, Materwiss. Werksttech. 36 (2005), p. 552. [4] A.C. Leff, C.R. Weinberger, and M.L. Taheri, Ultramicroscopy 153 (2015), p. 9. [5] E.F. Rauch, M. V�ron, Mater. Charact. 98 (2014) p. 1. [6] A.C. Leff and M.L. Taheri are grateful for support from: US DOE Basic Energy Sciences Early Career program (DE-SC0008274); National Science Foundation Faculty Early Career Program (#1150807); and DOE Nuclear Energy University Program (NE0000315). Figure 1. (A&B) GND density maps produced from raw data (no filtering). (C&D) GND density maps for the same regions as (A&B) produced from data that has been filtered using discrete Fourier transform analysis. (E&F) GND density maps produced using Kuwahara filtering. (G&H) GND density maps produced using interpolation method to refine the angular resolution. Color map units of m-2.");sQ1[946]=new Array("../7337/1895.pdf","Certified Reference Material for Strain Measurement Using EBSD","","1895 doi:10.1017/S1431927615010259 Paper No. 0946 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Certified Reference Material for Strain Measurement Using EBSD M.D. Vaudin1, W.A. Osborn1, L.H. Friedman1 and K. Siebein2 1. Materials Measurement Science Division, National Institute of Standards and Technology (NIST), Gaithersburg, MD 20899, USA 2 Center for Nanoscale Science and Technology, National Institute of Standards and Technology, Gaithersburg, MD 20899, USA The certified reference material (CRM[1]) for investigating the accuracy of the strain measurement method using electron backscattered diffraction has passed the majority of the tests required before release, and many of these results have recently been described [2]. The method, known as high resolution electron backscattered diffraction (HR-EBSD), measures the strain between positions on a sample in a scanning electron microscope by obtaining EBSD patterns (EBSPs) in a scanning electron microscope (SEM) at those positions and cross correlating several equivalent sub-regions in the patterns; the method has been described by Wilkinson et al. [3], and implemented in both commercial (Crosscourt [3,4]) and non-commercial (OpenXY[5,6]) software. A reference artifact consists of a 25 mm square chip cut from an (001)-oriented Si wafer with a film of Si1-xGex of thickness (t) deposited on the wafer by a commercial vendor; the film is patterned and has perfect epitaxy between the film and substrate. The CRM will consist of two chips of Si-Si1-xGex-on-Si with nominal (t, x) values of (35 nm, 0.3) and (50 nm, 0.2), respectively. Since Si1-xGex (shortened where appropriate to SiGe) has a larger lattice parameter than Si, the epitaxial SiGe film is compressed in the film plane and expands normal to the film, producing a tetragonally strained film. NIST reference values for the two tetragonal film strains will be supplied, and values for t and x for both chips will be given as NIST information values. The films are patterned so that among other features at the surface there are stripes of SiGe adjacent to pure Si, with widths from 50 m down to 0.5 m. Scanning across these stripes allows EBSPs to be collected from both unstrained and strained material, and pattern analysis software can then be applied to measure the tetragonal strain in the SiGe film. We determine the ratio () of the d-spacing of the (001) planes in the SiGe film divided by the (100) and (010) SiGe plane spacings, and the tetragonal strain is defined as -1. The (100) and (010) SiGe plane spacings are assumed to be equal as a consequence of the perfect epitaxy between the SiGe and Si; this assumption has recently been verified. Another important feature of the film pattern is a 12 mm � 20 mm pad of SiGe which has been designed for x-ray diffraction (XRD). The tetragonal strain has been verified using two XRD systems and the degree of epitaxy has been determined by a combination of techniques including XRD and x-ray photoelectron spectroscopy (XPS). Measurements of plane spacings both parallel to the chip surface (symmetric scans) and inclined (asymmetric scans) have been carried out using x-ray reciprocal space mapping around the 004 and 404 diffraction peaks from the substrate and film. Results indicate that for both (t, x) samples described above, the ratio of the in-plane spacings of the film and substrate are 1.0000(1), indicating perfect filmsubstrate epitaxy. It has also been found the plane spacing ratio for SiGe and Si (004) agree within 10-4 Microsc. Microanal. 21 (Suppl 3), 2015 1896 with results obtained from symmetric XRD scans measured using a two-circle x-ray diffractometer, and with calculations based on measurements of composition using XPS; many of these results were detailed in [2]. Now that the CRM, consisting of two artifacts with certified tetragonal strain values, has been validated with XRD data as described above, it can function in a way that mirrors the use of standards in energy dispersive spectroscopy (EDS) for quantification of elemental composition in the SEM. In the same way that compositional standards allow the accuracy of EDS in a SEM to be investigated, it becomes possible to assess the accuracy of HR-EBSD, calibrate an EBSD system for accuracy in strain measurement, and develop measurement protocols that maximize strain measurement accuracy. To achieve these goals, EBSPs will be collected at maximum pixel resolution from the same CRM samples using a number of in-house EBSD systems. The data will be collected at different SEM accelerating voltages and working distances to assess the impact of these experimental parameters. Tetragonal strains will be determined using CrossCourt and compared across systems and across experimental and analysis parameters. Initial results have already been collected from two systems and they show that the measured strains from the same sample have a relative difference of up to 10% and that diffraction patterns from the same system give different results depending on accelerating voltage, working distance and data analysis parameters. It must be noted that the precision of the technique, as indicated by the noise in strain values while scanning across a region of constant strain, can in some cases be as low as 1�10-4 strain. In a qualitative sense, HR-EBSD when used as a mapping tool can give extremely useful information about strain variation on a sample surface at spatial resolutions at 50 nm or better, with precision and resolution unobtainable by any other method. Indeed, much of the recent work on improving the technique of HR-EBSD has concentrated on the precision attainable, e.g., [7a,b]. When the strain data are to be compared with other techniques or with HR-EBSD strain data from different systems, potentially from different laboratories, and if the data are to be used in modeling analyses, it is vital to calibrate the results against a known sample such as that provided by the CRM described here. References: [1] NIST Special Publication 260 � 136 [2] MD Vaudin, WA Osborn, LH Friedman, JM Gorham, V Vartanian, RF Cook, Ultramicroscopy 148 (2015), p.94. [3] AJ Wilkinson, G Meaden, and DJ Dingley, Mater. Sci. Technol. 22 (2006), p. 1271. [4] Certain commercial equipment, instruments, or materials are identified in this paper to foster understanding. Such identification does not imply recommendation or endorsement by the National Institute of Standards and Technology, nor does it imply that the materials or equipment identified are necessarily the best available for the purpose. [5] J Kacher, C Landon, BL Adams and D Fullwood, Ultramicroscopy 109(9) (2009), p. 1148. [6] Brigham Young University (2015). OpenXY V1.0, github.com. [7] (a) TB Britton, J Jiang, R Clough, E Tarleton, AI Kirkland, AJ Wilkinson, Ultramicroscopy 135 (2013) 126; (b) TB Britton, J Jiang, R Clough, E Tarleton, AI Kirkland, AJ Wilkinson, Ultramicroscopy 135 (2013) 136.");sQ1[947]=new Array("../7337/1897.pdf","Novel Applications of Electron Channeling Contrast Imaging","","1897 doi:10.1017/S1431927615010260 Paper No. 0947 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Novel Applications of Electron Channeling Contrast Imaging Julia I. Deitz,1 Santino D. Carnevale,2 David W. McComb,1,3, Steven A. Ringel,2,3, Tyler J. Grassman1,2 1 2 Dept. of Materials Science & Engineering, The Ohio State University, Columbus, OH, 43210, USA Dept. of Electrical & Computer Engineering, The Ohio State University, Columbus, OH, 43210, USA 3 Institute for Materials Research, The Ohio State University, Columbus, OH, 43210, USA Microstructural characterization plays a critical role in crystalline semiconductor materials research, and such work has traditionally been performed using transmission electron microscopy (TEM). However, sample preparation for TEM work can be time intensive, and TEM specimen preparation can often damage the sample and features of interest, compromising their integrity. Electron channeling contrast imaging (ECCI) can serve as an alternative to traditional TEM in many instances, and because it is performed in a scanning electron microscope (SEM), ECCI effectively avoids many of these sample preparation issues. ECCI has a broad range of applications still to be exposed, and in this contribution, two such applications are discussed: quantum dot (QD) visualization and phase separation detection. We have demonstrated ECCI to be an ideal characterization tool for visualizing of defects within heteroepitaxial semiconductor structures [1], providing unprecedented access to valuable information regarding the formation and evolution of defects. We present here preliminary results regarding the further expansion of ECCI as applied to other epitaxial (opto)electronic materials systems. This includes, to our knowledge, the first demonstration of the use of ECCI for the visualization of subsurface epitaxial III-V QDs (InAs/GaAs) and the detection and characterization of crystallographic defects and phase separation in a III-V alloy (InGaP). All ECCI in this work was performed, on as-grown samples, using a FEI Sirion field-emission SEM, fitted with a pole piece mounted annular back scatter electron (BSE) detector with an accelerating voltage of 30 kV, a spot size of 5 (2.4nA), and a working distance of 5 mm. Application to Complex III-V Alloys: The InxGa1-xP ternary alloy system is of major importance to nearly all III-V based PV technologies. Nonetheless, InGaP is also a material system fraught with complex issues, such as atomic ordering and significant phase instabilities, making it a prime candidate for rapid microstructural characterization methods. Figure 1 presents preliminary data from an InGaP epilayer calibration growth nominally lattice-matched to GaAs0.9P0.1. The X-ray diffraction (XRD) reciprocal space map in Fig. 1(a) indicates a missed compositional target, and the splitting of the InGaP peak suggests phase separation. Low magnification ECCI, Fig. 1(b), shows a strong mottled appearance with no apparent correlation to any surface structure. Given the indication of phase separation provided by XRD, and the high sensitivity of ECCI to strain, the most likely source of this contrast is indeed phase separation, where the two lattice-mismatched phases possess different levels of strain, both with respect to the underlying substrate and each other; two-beam TEM yields similar contrast appearance in the presence of phase separation, depending upon the degree and extent of the separation [2]. Work is in progress to verify the identification of phase separation via ECCI in this sample and others. Of additional note here is that in the high magnification micrograph in Fig. 1(c) a large number of crystal defects � threading dislocations and stacking faults � are visible. The ability to visualize these defects despite the apparent phase separation, which in addition to general alloy scattering contributes to the degradation of the channeling quality, provides demonstration that ECCI is still a useful characterization technique in crystalline materials that are highly non-uniform. Microsc. Microanal. 21 (Suppl 3), 2015 1898 Application to Epitaxial Quantum Dots: Epitaxial QDs within the III-V compound semiconductor materials system typically form via the Stranski-Krastanov mechanism. Upon encapsulation, the QDs exert a high degree of strain on the surrounding material, producing strong strain fields that should strongly scatter the channeling electrons. Therefore, ECCI should enable the imaging of embedded QDs in plan-view geometry. To this end, Figure 2 presents ECCI micrographs of a single layer of InAs QDs embedded within a GaAs host at multiple imaging conditions. Clearly visible in Figs. 2(b) and 2(c) are small, round features consistent with the appearance of QDs (or their strain fields), while the standard BSE image in Fig. 2(a) shows no significant surface features. With the combination of ECCI's capabilities for imaging crystallographic defects and this demonstration of QD imaging, ECCI could be a powerful tool in QD research, enabling rapid characterization of such aspects as embedded QD density, size/shape, and uniformity, as well as investigation of potential QD-induced defect formation. Conclusion: Here, we demonstrate the use of ECCI for nondestructive analysis of, to our knowledge, two new areas: phase separation in complex alloys and subsurface epitaxial QDs. In both cases the results are comparable to those achieved via TEM analysis, but with the advantage of no specimen preparation and the use of SEM instrumentation. These extended applications of ECCI prove it to be an invaluable technique for PV characterization. [1] S. Carnevale, J. Deitz, et al, Appl. Phys. Lett. 104 (2014), p. 232111. [2] N. J. Quitoriano and E. A. Fitzgerald, J. Appl. Phys. 102 (2007), p. 033511. a b c InGaP GaAs0.9P0.1 2 �m GaAs g = (220) 500 nm g Figure 1. (a) XRD (400) reciprocal space map indicating lattice mismatch and phase separation in the InGaP epilayer. (b) Low (6956x) and (c) high (27852x) magnification ECCI micrographs in g = (220) diffraction condition. Threading dislocations (white arrows) and stacking faults (yellow arrows) are identified in (c). Figure 2. Single-layer InAs QDs in GaAs host as imaged via (a) standard surface-normal BSE, (b) g = (220) ECCI, and (c) g = (400) ECCI. The same region is picture in (b) and (c), and the blue rectangle surrounding three QDs, marked with arrows, provides frame of reference.");sQ1[948]=new Array("../7337/1899.pdf","Advances in HREBSD for Elastic Strain Measurement and its Application to Mechanical Testing","","1899 doi:10.1017/S1431927615010272 Paper No. 0948 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Advances in HREBSD for Elastic Strain Measurement and its Application to Mechanical Testing D J Dingley� and G M Meaden� �Department of Physics, University of Bristol, Tyndall Ave. Bristol BS8 1TH, UK �BLG Productions Ltd, Centre Gate, Colston Avenue, Bristol, BS1 4TR, UK The use of High Angular Resolution EBSD (HREBSD) for the measurement and mapping of elastic strain and geometrically necessary dislocation (GND) densities is becoming more common, especially for the study of polycrystalline materials and for in-situ straining and heating experiments. In the method, now frequently referred to as the Wilkinson method [1], multiple regions of interest of an EBSD pattern taken from a test region, are compared to a reference pattern taken from a nominally strain free region using a cross correlation procedure. In this way the relative distortion of the test pattern is determined. From this the full strain and rotation tensors are determined directly and from the rotation gradient between adjacent points in a map the GND density can be found. The current tendency is for the mapping of strain in polycrystalline specimens multiple times with intermediate thermo/ mechanical treatment and to map them over large areas. The sensitivity of the technique is 1 part in 10000 for strain measurement and 0.006 degrees for rotational measurement; this requires a pattern shift measurement equal to 0.1 of a camera pixel for a camera with 1000 x 1000 pixels. However, in order to secure this sensitivity care has to be taken to ensure accurate subtraction from the measured differences between reference and test patterns, those differences that are due to a/ the beam movements needed to collect the map and b/ the zooming of the EBSD pattern arising from changes in specimen to screen distance, when scanning an area of a steeply tilted sample. It has been shown that one way to track accurately the electron beam movement is to attach to the front of the EBSD detector a circular aperture mounted parallel to the phosphor screen of the camera. The attachment is referred to as a nose cone [2]. The position of the shadow of the aperture and its diameter provide information that can be directly related to the position of the diffraction pattern centre, PC, and sample to screen distance Z* for all points in a scan. The current investigation has concentrated on mapping the potential errors that occur in shift measurement with and without the use of a nose cone. We also correlated data from a series of scans of a polycrystalline sample which was removed from the microscope between scans. This was done to determine the reliability of using a reference pattern recorded in the first scan for use as a reference pattern for subsequent scans. HREBSD maps were obtained from perfect single crystal silicon specimens over areas ranging from 100 �m square to 1.5mm square. To test the reliability of using a single reference pattern as the reference for a sequence of scans, a polycrystalline nickel sample was mapped over an area of 15 �m, removed from the microscope, remounted and mapped again. The strain maps should be identical if PC and Z* positions between the two mappings are properly correlated. As illustrated in figure 1, it was shown that the limiting distance over which the normally used shift correction procedure could be relied on, for a specimen tilted 70 degrees from the horizontal, was �150 �m from the reference pattern. The figure shows the residual x shift for each data point after beam shift and zoom correction have been applied. The colour range in the image, red to blue, is Microsc. Microanal. 21 (Suppl 3), 2015 1900 �0.1 pixels and missing points are those that exceed this shift range (image width 300 �m, reference point at centre). The shift errors outside of this range are due to small deviations of the SEM magnification from its initial calibration value and uncertainties in the sample geometry. The result of attaching a nose cone and mapping the aperture shadow movement is shown in figure 2. It is for the case of a scan covering 800 �m x 1600 �m on a GaAs sample tilted at 70 degrees with beam step size of 100 �m. The shadows show that the beam movements did not form a rectangular scan over the specimen surface as commanded by the input scan dimensions but were in the form of a trapezium. Without the nose cone attachment this would not have been detected. The true movement of the EBSD PC can be calculated from the shadow movement [2] and is shown in figure 3. The importance of accurate location of the PC and Z* values is that the zoom corrections are highly dependent on both of these parameters. They are also paramount when using a reference pattern from one scan as the reference pattern for a second scan. Figures 4a & b are two such scans taken of the same Ni sample. The shadow data in this case showed that the PC values for the data point used as reference pattern for each the two scans, differed respectively by x: 9.5�m, y:20.2�m & z:8.1�m but their respective positions in the map differed by x:3�m, y:4�m. This implied a different correction was needed for PC and zoom values than would have been the case using map position correction. In the illustrated case, as there was no additional straining between the two scans, the maps should be identical. The mean difference between corresponding map positions in, for example, the resulting 11 strain values was �0.00038 (the maps have been cropped to permit easier identification of corresponding points). 1 2 3 Figures 1-3, Residual x shifts after subtraction of beam shifts and zoom effects, image width 300m (2) aperutre shadow centre movemenets , (3)back calculted PC positons from measured aperture shadow positions. a b Figure 4 Sequential mapping of deformed polycrystalline nickel base alloy. (map step size 0.5�m. Ref position marked with crosses) References [1] A.J.Wilkinson, G. Meaden, and D.J. Dingley, Ultramicroscopy, 106 (2006). p. 307. [2] European Patent Application No. 12717453.0");sQ1[949]=new Array("../7337/1901.pdf","Valence-loss EELS Spectroscopy of Refractory Plasmonic Nanomaterials","","1901 doi:10.1017/S1431927615010284 Paper No. 0949 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Valence-loss EELS Spectroscopy of Refractory Plasmonic Nanomaterials Andrew A. Herzing1, Urcan Guler2, Xiuli Zhou3 and Theodore B. Norris3 Material Measurement Laboratory, National Institute of Standards and Technology, Gaithersburg, MD USA. 2. School of Electrical & Computer Engineering, Purdue University, West Lafayette, IN USA. 3. Center for Ultrafast Optical Science, University of Michigan, Ann Arbor, MI USA. The interaction of light with nanostructures via plasmonic resonance offers a promising route to controlling the properties and propagation of light at length scales which are far below the diffraction limit. These nanostructures typically consist of noble metals, and can exhibit very strong, well-defined resonances. However, the promise of this technology has yet to be fully realized, which is due in large part to inherent limitations in the chosen materials, which often result in heavy losses in the optical range. Similarly, they can suffer from poor chemical and/or thermal stability, which severely restricts their use in practical applications. More recently, refractory transition metal nitrides have been proposed as a promising class of alternative materials for plasmonic applications. In particular, TiN has been found to exhibit ideal properties for many applications [1-3]. This material exhibits similar optical properties to those of Au, but is much more thermally stable, less expensive, and is compatible with traditional semiconductor fabrication technology. Herein, we report measurements of the local plasmonic response of TiN thin-films and nanocubes via monochromated electron energy-loss spectroscopy (EELS) in the scanning transmission electron microscope (STEM). 150 nm thick films of TiN were deposited on an MgO substrate via DC magnetron sputtering of a titanium target in an argon-nitrogen environment. The base pressure was 8x10-8 Torr and the substrate temperature during deposition was 800 oC. Cross-sectional specimens of the TiN film were prepared using a focused ion-beam (FIB) lift-out procedure after first protecting the TiN film surface by sputter-coating it with 25 nm of Au-Pd. This layer protects the surface during the deposition of a further Pt overlayer in the FIB prior to cross-sectional milling. Samples were also prepared from a solution of TiN nanocubes suspended in deionized water by drop casting onto a 15 nm thick silicon nitride membrane. The inset of Figure 1 shows a STEM-HAADF image of the MgO-TiN interface acquired with the crystals oriented along the <001> direction. The overall quality of the TiN film is quite high and is seen to grow epitaxially from the MgO substrate. Electron energy-loss spectroscopy (EELS) was carried out using a monochromated 150 keV electron beam with a spectral energy-loss resolution of 0.15 eV as measured by the FWHM of the zero-loss peak. EELS spectra of the valence loss response were acquired from the specimen using a monochromated probe with a current of 20 pA. Spectra were collected along a line beginning in the MgO substrate and terminating in the protective Pt overlayer. The energy-axis of all spectra were then aligned, and 20 spectra were integrated from the MgO substrate, MgO/TiN interface, and the center of the TiN film (Fig. 1). The spectrum from the MgO substrate exhibits features typical of this material [4] with a bandgap onset indicated by the abrupt increase in inelastic scattering near 7.5 eV. At the junction of the MgO and TiN film, a relatively sharp peak was observed at (2.05 + 0.01) eV. Further from the interface, at the center of the TiN film, a weaker peak was observed at (2.81 + 0.3) eV. This peak was followed by a nearly linear increase in inelastic scattering intensity that was continuous up until a discrete change near 11 eV. Multiple acquisitions from random points in the film showed that the spectral response did not vary with position. Similar results from the TiN nanocubes are shown in Figure 2. In this case, the spectral response from the interior of the particle nicely matches that collected from the center of the TiN film in Figure 1, and 1. Microsc. Microanal. 21 (Suppl 3), 2015 1902 represents a combination of the bulk and surface excitation with the former being much stronger than the latter. Subtle differences are observed in the low-loss response collected from the particle corner versus that collected near the cube face. These differences can be understood by comparison to optical simulations, and were found to be due to particle shape and surface chemistry effects. References: [1] U. Guler, et al, Appl. Phys. B 107 (2012) 285. [2] U. Guler, et al, Nano Letters 13 (2013) 6078. [3] G. V. Naik, et al, Opt. Mater. Express 2 (2012) 478. [4] C. C. Ahn and O. L. Krivanek, "EELS Atlas", Gatan, 1983. Figure 1. Low-loss STEMEELS analysis of MgO supported TiN thin film cross-sectional specimen. EELS spectra were collected from various points in the region depicted by the HAADF image (inset). Figure 2. Low-loss STEMEELS analysis of a single TiN nanocube shown in the HAADF image (inset). EELS spectra were extracted from a 2D spectrum image at the points shown by the colored dots.");sQ1[950]=new Array("../7337/1903.pdf","Bandgaps and Surface Inter-Band States in Photocatalysts with High Energy Resolution EELS.","","1903 doi:10.1017/S1431927615010296 Paper No. 0950 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Bandgaps and Surface Inter-Band States in Photocatalysts with High Energy Resolution EELS. Qianlang Liu1, Liuxian Zhang1, Katia March2, Toshihiro Aoki3 and Peter A. Crozier1 1. School for the Engineering of Matter, Transport and Energy, Arizona State University, Tempe Arizona 85287-6106 2. Laboratoire de Physique des Solides, B� timent 510, Universit�Paris-Sud, 91405 Orsay Cedex, France 3. LeRoy Eyring Center for Solid State Science, Arizona State University, Tempe, Arizona 85287, USA Photocatalysts have potential applications for solar fuel generation through water splitting [1]. The bandgap and inter-band states of the semiconductors significantly affects the performance and efficiency of the catalysts. Recent advances in STEM EELS monochromation now allow for routine ultra-high energy resolution of 15 meV or better in the low-loss region [2]. The ability to now correlate atomic structure and electronic structure in the low-loss and bandgap region of the energy-loss spectrum represents a powerful tool for characterization of electronic and optical properties of nanomaterial such as the high surface area, particulate systems that are generally used as catalysts. The band structure can vary among different nanoparticles depending on particle sizes, facets and also at the surfaces/interfaces of the semiconductors where charge transfer and photocatalytic reactions take place. With the innovation of high energy resolution EELS, it is possible to tackle the issues mentioned above by investigating the bandgap and the fine electronic structures inside the gap at the nano-level. This study focuses on TiO2 and Ta2O5 which are both UV absorption photocatalysts due to their large bandgaps. TiO2 is relatively abundant with simple crystal structures while Ta2O5 is very efficient with more complicated structures. Four different samples were obtained: anatase TiO2 (HP), Ta2O5 (HP), Ni/TiO2 (LP) and NiO/Ta2O5 (LP). TiO2 (HP) and Ta2O5 (HP) represent high purity TiO2 and Ta2O5 nanoparticles which were prepared following hydrothermal/solvothermal methods. Ni/TiO2 (LP) was obtained first by wetness impregnation on low purity commercial anatase TiO2 which are hundreds of nm in size then reduced @500� in flowing 5% H2/Ar for 2hrs [3]. It is used as a model of metal coC catalyst loaded on semiconductor to study the band structures at the interfaces. NiO/Ta2O5 (LP) as a model of semiconductor-semiconductor photocatalyst was prepared also first by impregnation but then oxidized @500� in flowing O2 for 2hrs [4]. Low-loss EELS was carried out on a monochromated and C aberration-corrected NION microscope operated at 60 kV with dispersions of 5 meV or 2 meV per channel. The energy resolution was better than 25 meV. Fig.1 shows the bandgap edges of Ta2O5 (HP) and TiO2 (HP) acquired under aloof beam conditions in which the electron beam was positioned at about 5 nm away from the particle surfaces. The measured values of Ta2O5 (HP) and TiO2 (HP) bandgap from more than 6 different spectra are 4.50 eV � 0.02 eV and 3.60 eV � 0.01 eV respectively. The bandgap is measured by extrapolating the straight portion of the fitted curve to the x axis. For anatase, the EELS measurement of 3.60 eV is larger than the wellknown ~3.2 eV but close to the recent theoretical calculation and experimental data [5]. Fig.1b also shows the bandgap of anatase increases to 3.71 � 0.03 eV as the Ti in the particle was reduced from Ti+4 to Ti+3 by the electron beam (Figure 1b insertion) which is contradictory to conventional optical measurements showing a narrower bandgap for reduced TiO2. The reason is not clear at this stage. On Fig. 2, inter-band states were clearly observed when the beam was on the surfaces of bare Ta2O5 and TiO2 of the NiO/Ta2O5 (LP) and Ni/TiO2 (LP) samples. Distinctive interstates peaks were at 1.13 eV, Microsc. Microanal. 21 (Suppl 3), 2015 1904 2.14 eV and 2.66 eV for Ta2O5 and 2.08 eV, 2.37 eV and 2.85 eV for TiO2. The peaks are only observed on the surface of the particles and they vanish when the beam is in the bulk Ta2O5 or TiO2. Those states are believed to be related to the oxygen vacancies produced from the high temperature reduction and/or impurities segregated to surfaces in the commercial TiO2 or Ta2O5 after high temperature heat treatment. The energies of these inter-band states vary among different particles depending on their local morphology and impurity concentrations. Peaks associated with different oxygen vacancy concentrations were also observed. More experimental data with theoretical simulation to explore the complicated electronic band structures of these photocatalyst materials will be presented. References: [1] Fujishima, A.; Honda, K. Nature 238 (1972), 37. [2] O.L. Krivanek et al, Nature 514 (2014), 209. [3] L. Zhang et al, J. Phys. Chem. (under review) [4] Q. Liu et al, Appl. Catal. B: Environ.172 (2015), 58. [5] M.Landmann et al, J. Phys: Condens. Matter 24 (2012), 195503 [6] The support from US Department of Energy (DE-SC0004954) and the use of NION microscope at John M. Cowley Center for HR Microscopy at Arizona State University is gratefully acknowledged. Figure 1: a) Low-loss EELS with insertion of STEM image showing beam was near surface of Ta2O5 (HP) particles; b) Bandgap edge from anatase TiO2 (HP) with insertion of Ti L2,3 edges before and after reduction by electron beam. Figure 2: a) Inter-band states on surface of Ta2O5 (LP) and low-loss EELS of vacuum. The insertion shows where the surface spectra were taken; b) Inter-band states on surface of TiO2 from Ni/TiO2 (LP) sample.");sQ1[951]=new Array("../7337/1905.pdf","Observation of Inter-Bandgap States in Doped Ceria via Monochromated EELS","","1905 doi:10.1017/S1431927615010302 Paper No. 0951 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Observation of Inter-Bandgap States in Doped Ceria via Monochromated EELS W.J. Bowman1, K. March2, T. Aoki3, C.A. Hernandez1, P.A. Crozier1 1. School for the Engineering of Matter, Transport and Energy, Arizona State University, 501 E. Tyler Mall, Tempe Arizona 85287-6106 2. Laboratoire de Physique des Solides, B�timent 510, Universit� Paris-Sud, 91405 Orsay Cedex, France 3. LeRoy Eyring Center for Solid State Science, Arizona State University, Tempe, Arizona 85287-1704, USA Recent advances in STEM EELS monochromation now allow for routine ultra-high energy resolution of 15 meV or better in the low-loss region [1]. The ability to now correlate atomic structure and electronic structure in the low-loss and bandgap regions of the energy-loss spectrum represents a powerful tool for characterization of electronic and optical properties of material defects and device features such as grain boundaries and interfaces. Here, we demonstrate an application of monochromated low-loss EELS for characterizing the electronic transition from valence band to inter-bandgap states in Pr0.1Ce0.9O2- (PCO) nanoparticles, an ionic/electronic-conducting solid whose cations have readily accessible trivalent/tetravalent redox couples (i.e. Ce3+/4+ and Pr3+/4+). The specimen for this study was heat-treated PCO nanoparticles (~30 nm) synthesized in the authors' lab via spray drying (rapid solution evaporation), which resulted in partially oxygendeficient particles (fig. A). Subsequent heating for 3 hr at 900 �C in air maximized the number of tetravalent cations (by filling O vacancies), causing the powder to appear dark red in color (fig. B; see discussion below). Energy-loss spectra were acquired using a monochromated Nion UltraSTEM100 at 60 keV in the energy-loss range up to ~5 eV using dispersions including 3 meV and 5 meV. Under these conditions the zero-loss peak (ZLP) full width half maximum was typically 18 meV to 25 meV to provide reasonable signal-to-noise. Data were acquired using the standard transmission geometry as well as an aloof-beam configuration, whereby the STEM probe was positioned in the vacuum some distance from the specimen (fig. C) to avoid the radiation damage. Core-loss EELS and optical observations corroborated findings based on lowloss data. An inverse power law fit was used to subtract the ZLP, and preliminary data (e.g. fig. D) revealed that the conduction band onset energy, i.e. the band gap, was approximately 3.2 eV-- measured from the intercept of the linear extrapolation of the conduction band edge with the energy-loss axis. Spectra showed a feature within the bandgap with a rising onset and extended plateau. The rise spanned the energy range of 1.6 eV to 2.4 eV and the feature's plateau extended from 2.4 eV to the conduction band edge. In the core-loss region, the separation between the Ce and Pr M5 white lines was monitored to identify the cations' oxidation state. Separations ranged from 46.2 eV to 46.9 eV, and were typically greater than the separations measured in reference material containing Ce4+ and Pr3+ which were 46.3 eV (see fig. E), which verified that the sample contained a substantial fraction of Pr4+ [2]. The inter-bandgap feature was attributed to transitions of electrons from the O-2p valence band into unoccupied Pr-4f levels lying slightly below the Ce-4f conduction band as shown in fig B. The energy range of the spectral feature's rising onset was assumed to be proportional to the Microsc. Microanal. 21 (Suppl 3), 2015 1906 Pr-4f band width, and the midpoint energy of the sloping onset (2.0 eV) was taken to be the Pr-4f band center position. This inter-band state associated with Pr doping results in broad optical absorption above ~2 eV, and causes the oxide to appear deep red in color [3]. This position and the measured energy of the Ce-4f conduction band onset (i.e. the direct optical bandgap) agreed closely with reports of optical bandgaps of PCO (2.1 eV) [3] and CeO2 (3.6 eV) [4]. Interestingly, the inter-band state persisted throughout time-resolved acquisitions, likely indicating signal delocalization in transmission mode. References and acknowledgements 1. O.L. Krivanek et al. Nature 514, 209-212 (2014). 2. C. L�pez-Cartes et al. Chem. Commun. 5, 644-645 (2003). 3. J.J. Kim et al. Chem. of Mater. 26, 1374-1379 (2014). 4. E. Ruiz-Trejo. Phys. Chem. Solids. 74, 605-610 (2013). 5. W.J.B. would like to acknowledge the NSF's Graduate Research Fellowship (DGE-1211230) for continued financial support. We gratefully acknowledge support of NSF grant DMR-1308085 and ASU's John M. Cowley Center for High Resolution Electron Microscopy. Figs. A & B. PCO (A) as-synthesized and (B) after heating. Band diagrams adapted from [2]. Fig. C. Representative HAADF STEM image illustrating aloof-beam EELS configuration. Fig. D. Low-loss transmission EELS of Pr-4f inter-band state, and ceria conduction band onset. Fig. E. Core-loss EELS of Ce and Pr M4,5 white lines in Ce4+/Pr4+ condition.");sQ1[952]=new Array("../7337/1907.pdf","Introduction to Plasmon Energy Expansion Thermometry","","1907 doi:10.1017/S1431927615010314 Paper No. 0952 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Introduction to Plasmon Energy Expansion Thermometry B. C. Regan1, William A. Hubbard1, E. R. White 1, Rohan Dhall2, Stephen B. Cronin2, and Shaul Aloni3, Matthew Mecklenburg4 Department of Physics and Astronomy & California NanoSystems Institute, University of California, Los Angeles, CA, USA 2. Department of Electrical Engineering, University of Southern California, Los Angeles, CA, USA 3. Molecular Foundry, Lawrence Berkeley National Laboratory, Berkeley, CA, USA 4. Center for Electron Microscopy and Microanalysis, University of Southern California, Los Angeles, CA, USA Obtaining local temperature measurements in a transmission electron microscope (TEM) is challenging. In liquid cell, gas cell, and heating experiments the temperature is typically determined using a thermocouple, optical pyrometer, or calibrated resistor. When using such methods one usually assumes that the temperature in the sample region is spatially homogeneous, since these techniques cannot resolve temperature gradients on small length scales. Here we describe a thermometry technique, plasmon energy expansion thermometry (PEET), which can make local temperature measurements on sub 10-nanometer length scales [1]. The technique does not use an external temperature probe, but rather induces the material under observation to act as a collection of tiny, local thermometers. In PEET the local density of a material is ascertained from its plasmon energy, and the local temperature is deduced from this density. In a TEM a beam electron can lose energy by creating a charge oscillation (called a plasmon) as it passes through a material. The plasmon's energy is proportional to the square root of the material's valence electron density. Thermal expansion or contraction changes this density as the material is heated or cooled [2]. The temperature-induced plasmon energy shift is thus related to a material's thermal expansion properties. A plot of this relation is shown in Fig. 1A for aluminum [3]. Typical aluminum electron energy loss spectra (EELS) for two different temperatures are shown in Fig. 1B. The small energy shift, only ~ -0.5 meV/K, is detected by curve fitting both the zero loss peak and the plasmon peak. The plasmon energy is taken to be the energy difference between the peak centers. Fig. 2 shows a device and the PEET data extracted from it. An electron beam-deposited platinum strap, seen as a dark diagonal line connecting the lower and left contacts, was used to heat one side of the serpentine aluminum wire in the red box (Fig. 2A). Plasmon energy maps of the wire with 0 and 1500 uW of power delivered to the heater are shown in Figures 2B and 2C, respectively. These maps reveal dark, crack-like, structures in the aluminum, which are caused by ~10-20 meV energy shifts at grain boundaries. Combining the two maps according to (C-B)/B (where the letters refer to the panels of Fig. 2) and applying the inverse of the curve in Fig. 1A gives the temperature map Fig. 2D. The pixel, or spectra acquisition rate was ~76 Hz for the data of Fig. 2D. The standard deviation of the measured temperature in various small (and thus presumably isothermal) regions ranges from 20 to 30 K, which corresponds to a temperature sensitivity of ~ 3K / . A variety of spectra acquisition rates yielded this same figure of merit. 1. Microsc. Microanal. 21 (Suppl 3), 2015 1908 References: [1] M Mecklenburg et al, Science 347 (2015), p. 629-632. [2] G Meyer, Zeitschrift fur Physik 148 (1957), p. 61-71. [3] A J C Wilson, Proceedings of the Physical Society, 54 (1941), p. 235-244. [4] This work was supported by NSF DMR-1206849, and in part by FAME, one of six centers of STARnet, a Semiconductor Research Corporation program sponsored by MARCO and DARPA. Data presented were acquired at the Center for Electron Microscopy and Microanalysis at the University of Southern California. Figure 1. (A) The fractional plasmon energy change vs temperature, calculated from the data in [3]. (B) Two EELS spectra from aluminum, each with a dispersion of 25 meV/bin, at room temperature (black) and ~100K hotter (red). The plasmon peak maximum bin for each temperature is indicated with an arrow, and the curve fit centers are indicated with a vertical line. Figure 2. (A) A bright field TEM image of a device. A heater connects the lower and left-hand contacts. PEET was applied to the region indicated with a red rectangle. (B and C) Plasmon energy maps. Each map is 154x398 pixels, and each pixel is 4 nm on a side. (D) Temperature map made from images (B) and (C), 151x396 pixels, with the same pixel size as in the plasmon images. A combination intensity scale and pixel count histogram is shown below each image.");sQ1[953]=new Array("../7337/1909.pdf","Nanoscopic imaging of energy transfer from single plasmonic particles to semiconductor substrates via STEM/EELS","","1909 doi:10.1017/S1431927615010326 Paper No. 0953 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Nanoscopic imaging of energy transfer from single plasmonic particles to semiconductor substrates via STEM/EELS Guoliang Li1, Charles Cherqui2, Nicholas W. Bigelow2, Gerd Duscher3, Patrick J. Straney4, Jill E. Millstone4, David J. Masiello*2 and Jon P. Camden*1 1 2 Department of Chemistry & Biochemistry, University of Notre Dame, Notre Dame, IN 46556 Department of Chemistry, University of Washington, Seattle, WA 98195 3 Department of Materials Science & Engineering, University of Tennessee, Knoxville, TN 37996, 4 Department of Chemistry, University of Pittsburgh, Pittsburgh, PA 15260 Localized surface plasmon resonances (LSPRs), the collective oscillations of conduction electrons in metallic nanoparticles, can produce intense near-fields at the resonance wavelengths. Plasmonic nanoparticles have been incorporated in the design of photovoltaic (PV) and photocatalytic devices, where they have been shown to enhance solar energy harvesting efficiency. Research has shown that the addition of plasmonic nanoparticles improves the efficiency of solar light harvesting via one or more of the following mechanisms1: (1) LSPR excitation leads to an increase in path length for incoming light via scattering, thereby increasing light absorption by the semiconductors; (2) energy transfer from the decay of an LSPR directly creates an electron-hole pair in the semiconductor, a process known as plasmon-induced resonant energy transfer (PIRET). Its efficiency relies on the overlap between the LSPR emission and the band gap absorption of the semiconductor2; (3) direct electron transfer (DET) from the nanoparticle to a semiconductor, in which an LSPR decays, through Landau damping, into a "hot" electron that may then scatter into the semiconductor if it has sufficient energy to overcome the Schottky barrier formed at the interface3. Mechanism (1) is only effective for photon energies above the band gap, while mechanism (2) and (3) involve photons with energies below or above the band gap, therefore, are of particular interest and importance. However, despite its importance, little is known about how PIRET and DET operate at the nanoscale, particularly at the level of a single nanoparticle. In this paper, we present a nanoscale EELS study of PIRET and DET on several Ag nanocube@substrate systems, where the cubes and substrates serve as plasmonic energy donors and acceptors, respectively. The substrates are carefully chosen to turn off or isolate the PIRET and DET energy transfer channels. To simplify the discussion, we focus on only the cube@silicon dioxide (SiO2)/boron phosphide (BP)/amorphous silicon (a-Si) results. Figure 1(a) demonstrates the correlation diagram of substrateinduced LSPR hybridization without energy transfer. Figure 1(b) describes schematically the EELS experiments. By tilting the samples, we are able to probe how the energy transfer modifies the proximal (D) and distal (Q) corner LSPR modes4 without the interferences from the edge and face modes. As are shown in Figure 2, SiO2 does not damp either D or Q mode whereas BP and a-Si impair the D modes remarkably, leaving the Q mode unaffected. This is because SiO2 is a transparent large band-gap (9 eV) insulator which is closed to both PIRET and DET; BP has the PIRET off (optically transparent) but the DET on (small band gap); a-Si has both PIRET and DET on. The efficiencies of PIRET and DET cause the D modes damp at different levels, but do not affect the Q modes which are far from the substrates. The experiments are carried out in a monochromated Carl Zeiss LIBRA� 200MC (S)TEM operated at 200 kV (150 meV energy resolution). The edge lengths of the Ag nanocubes range from 70 nm to 77 nm, the corner radii range from 10.2 nm to 12.6 nm. The SiO2 substrate is a commercial product, the BP and a-Si substrates are fabricated via conventional TEM specimen preparation procedures. Microsc. Microanal. 21 (Suppl 3), 2015 1910 Figure 1. (a) Diagram of substrate-induced LSPR hybridization of a Ag nanocube. The evolution of the surface charge distributions of the D and Q eigenmodes of the nanocube is schematically displayed as function of increasing substrate dielectric constant. (b) Experimental EELS setup. The electron beam independently addresses the proximal and distal corners of the nanocube by tilting the composite system. The substrate is probed by a beam position far from the nanocube. Figure 2. (a) Raw EEL spectra acquired at the proximal corner (solid lines), distal corner (dotted lines), and substrate (dashed lines), as described in Fig. 1b. ED and EQ denote the resonant energies, while D and Q denote the damping coefficients of the D and Q modes. (b) Z-contrast images of the tilted cubes. The solid lines represent the cube edges that are visible when viewed into the page, the dashed lines are cube edges that are blocked in the viewing direction. (c, d) D and Q EELS maps generated by plotting the spectral intensities over the spectrum image at ED and EQ. The proximal and distal faces are shown in the maps. The D mode (proximal) is highly damped for nanocubes on BP and a-Si, indicating significant energy transfer to the substrate. References [1] Cushing, S.K. and Wu, N., Interface 22 (2013), 63-67. [2] Li, J. et al., Nat Commun 4 (2013), 2651. [3] Clavero, C., Nat Photon 8 (2014), 95-103. [4] Zhang, S. et al., Nano Letters 11 (2011), 1657-1663.");sQ1[954]=new Array("../7337/1911.pdf","Time-Resolved Imaging of Electrochemical Switching in Nanoscale Resistive Memory Elements","","1911 doi:10.1017/S1431927615010338 Paper No. 0954 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Time-Resolved Imaging of Electrochemical Switching in Nanoscale Resistive Memory Elements William A. Hubbard1, E. R. White1, Alexander Kerelsky1, G. Jasmin1, Jared J. Lodico1, Matthew Mecklenburg2, and B. C. Regan1 1. Department of Physics & Astronomy and California NanoSystems Institute, University of California, Los Angeles, CA 90095 USA 2. Center for Electron Microscopy and Microanalysis, University of Southern California, Los Angeles, CA 90089 USA FLASH memory is reaching its scaling limit, but resistive random access memory (ReRAM) is considered a promising successor [1]. In ReRAM, metal electrodes sandwiching an insulating electrolyte form a digital memory element, where the presence or absence of a conducting path through the insulator represents one bit of information. The conducting filament is thought to form, atom-by-atom, when subject to a SET voltage applied across the electrodes, and to disintegrate when subject to a RESET voltage. We use scanning transmission electron microscopy (STEM) to image nanoscale ReRAM devices switching in situ. Operating the devices with small current limits slows the rate of filament formation and reduces confounding thermal effects, allowing us to obtain time-resolved images of filament formation and regeneration. We fabricated Pt/Al2O3/Cu ReRAM devices with a slant-vertical geometry. The "slant" gives good imaging access, while keeping the electrodes completely encased to ensure that conducting filaments form though the Al2O3 layer (as in the standard vertical geometry) and not across exposed surfaces or layer interfaces [2]. Figure 1a shows a STEM image of a pristine device, and Fig. 1b shows the same device after a 4.4 V SET bias has been applied. Applying the SET caused the current to jump to the 50 nA compliance limit as the device switched from its ~10 G OFF state to a ~20 M ON state. The STEM images show that, coincident with the SET, the device developed a 5 nm wide Cu filament connecting the electrodes. After a -4V RESET, the device was cycled repeatedly between the ON and OFF states. The low current compliance regulated the rate of the switching processes so that the SET and RESET processes could be time-resolved: each was imaged over several frames. The electronic and imaging data acquired under these conditions corroborate the electrochemical model for switching [3]. Devices tested with higher current limits showed evidence of heating during SET. Figures 1c-e show STEM images of devices that were switched with SET bias voltage values comparable to those of the device in 1a-b, but with current limits ranging from 100 nA to 700 nA (i.e. 2 to 14 to times larger). The resulting filaments are much wider than the filaments formed during low current SETs, and the ON resistances are < 1 k. The electrodes exhibit reconfiguration and recrystallization consistent with heating to temperatures near their respective melting points. However, the programmed current limit values used are still small enough that such high temperatures are unexpected. RESETs following such large-current SETs could require more than 1 mA, which often resulted in device failure (Fig. 1f). Using the larger compliance currents decreases the ON-state resistance RON, which allows transient currents larger than the compliance current limit to flow. These currents are sourced by stray capacitance C in the circuit, and can Joule-heat the device by hundreds of kelvins. A simplified circuit model of the ReRAM biasing setup is shown in Fig. 2. The time constant for the system is ~ C RON, Microsc. Microanal. 21 (Suppl 3), 2015 1912 and the maximum current I Q/ CV/ V/ RON. Therefore, the maximum power through a device during an ON switch is IV = V2/ RON. For a device with a RON = 1 k, the 1 mW SET heats the device above a thermal threshold. In comparison a RON = 10 M device switches with 100 nW and remains near room temperature. Small compliance currents thus enable time-resolved imaging of ReRAM cycling in situ. References: [1] I. Valov et al, Nanotechnology 22 (2011), p. 289502 [2] W. Hubbard et al, Microscopy and Microanalysis 20 (2014), p. 1550-1551 [3] S. Menzel et al, Physical Chemistry Chemical Physics 20 (2013), p. 6945 [4] This work has been supported by FAME, one of six centers of STARnet, a Semiconductor Research Corporation program sponsored by MARCO and DARPA, and by National Science Foundation award DMR-1206849. The authors acknowledge the use of instruments at the Electron Imaging Center for NanoMachines supported by NIH 1S10RR23057 and the CNSI at UCLA. a Pt b 50 nA 4.4 V c e 140 nA 4.4V d f 700 nA 4V 100 nA 4V 1 mA -1V Cu Figure 1. STEM images of a slant-vertical ReRAM device before (a) and after (b) its first SET (voltage and current limit shown). Similar devices (c-e) that were switched with larger current limits produced burlier connections with poorer cycling endurance. For instance, f shows the e device after a RESET attempt failed to change the device state to OFF. The scale bar in all images is 50 nm. V Rcomp= V Icomp C RHRS RLRS Figure 2. Circuit model of ReRAM device biasing. The resistance Rcomp is determined by the programmed current compliance Icomp. The ON state device resistance RON, a function of Icomp, decreases rapidly once Icomp exceeds a thermal threshold value that in our devices is 100 nA.");sQ1[955]=new Array("../7337/1913.pdf","Dynamic Rate Mechanism of V2O5 Coated SnO2 Nanowires for Lithium Ion Batteries Studied by in situ TEM","","1913 doi:10.1017/S143192761501034X Paper No. 0955 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Dynamic Rate Mechanism of V2O5 Coated SnO2 Nanowires for Lithium Ion Batteries Studied by in situ TEM Lifen Wang1,2, Zhi Xu1, Xuedong Bai1, Jianguo Wen2, and Dean J. Miller2 Beijing National Laboratory for Condensed Matter Physics and Institute of Physics, Chinese Academy of Sciences, Beijing 100190,China 2. Electron Microscopy Center - Center for Nanoscale Materials, Argonne National Laboratory, Argonne, IL 60439 Coating strategies are commonly employed to avoid lithiation-induced fracture and improve electrochemical performance of lithium storage materials with higher capacity and longer cycle life [1, 2]. Correlation of structure with electrochemical performance of such materials is needed to guide further design. This work mainly focuses on realizing a working lithium battery inside the transmission electron microscope (TEM) to measure and understand electrochemical mechanisms of vanadium pentoxide loaded tin dioxide (V2O5/SnO2) nanowire. We achieve this by using a nanomanipulator to assemble an open electrochemical cell inside a TEM that consists of target materials, a Li-containing counter electrode (Li or Li cobalt oxide), and a solid-state electrolyte (Li2O) [3]. To directly capture the dynamic structural changes of electrode materials during the lithiation process, open lithium cells were characterized via real-time in situ high resolution transmission electron microscopy (HRTEM) imaging, electron diffraction (ED) and electron energy-loss spectroscopy (EELS). The cells were set up with a single V2O5/SnO2 nanowire as the cathode material and cycled in galvanostatic mode. TEM images of the lithiation process at a lower rate (300 pA) are shown in Figure 1. The bright-field images in Fig. a, b-d show the effect of lithiation on overall nanowire morphology, revealing lithium intercalation into the interlayer spaces of V2O5 with only a subtle volume expansion while in the core, lithium alloys with Sn, giving rise to ~200% volume change. As seen in Fig. 1, cracks in the lateral direction due to the volume expansion mismatch are observed, and these cracks appear to be the main cause for the capacity loss at low rate. A rather different lithiation process is observed at a higher rate (3 nA) as shown in Figure 2. At this higher rate, we observe Sn nanoparticles growing along the nanowire and the alloy reaction is retarded in the first lithiation process. Layered material V2O5 in the shell with inactive Li2O in the core forms a more stable structure and active Sn nanoparticles aggregate along the stable nanowire, giving rise to a new nanostructure and better cycle life. A comparison of the lithiation mechanisms revealed by this in situ TEM approach reveals new insight into the dependence of lithiation and delithiation on charging rate for these V2O5/SnO2 nanowires. At low charge rates, lithium intercalates into the interlayer space of V2O5 in the shell and alloys with Sn in the core. Mismatch in volumetric expansion between the lithium intercalation-V2O5 compound and the lithium alloy-SnO2 causes mechanical fracture in the core/shell nanowire. The volume expansion of crystalline SnO2 is confined in the lateral direction by the V2O5 surface coating. This results in crack formation along the coating surface, which leads to capacity fade. In contrast, at higher rates, the alloy reaction of Sn with Li to form Li-Sn is retarded, and Sn particles that are the active host for Li aggregate on the surface of the host nanowire, minimizing deleterious effects of volume expansion and providing much better cycle performance. 1. Microsc. Microanal. 21 (Suppl 3), 2015 1914 References: [1] Jian Yan , Afriyanti Sumboja , Eugene Khoo , Pooi See Lee, Advanced Materials 23, (2011), p. 746. [2] LF Cui, R Ruffo, CK Chan, H Peng, Y Cui, Nano Letters 9 (2008), p. 491. [3] JY Huang, L Zhong, CM Wang, JP Sullivan, W Xu, LQ Zhang, SX Mao, Science 330 (2010), p. 1515. [4] This work was supported by National 973 projects (Grant Nos. 2012CB933003 and 2013CB93200) from the Ministry of Science and Technology, China. Research in the Center for Nanoscale Materials, including resources in the Electron Microscopy Center, was supported by the U. S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357. Figure 1. (a-d) Snapshots of SnO2 nanowire during the first lithiation process at lithiation current of 300 pA. Cracks that form along the coating shell are visible in (d). (e, f) Bright field images of the same area before and after the first lithiation process as marked by the boxed area in (a) and (f) showing cracks forming in the shell of the nanowire. Figure 2. (a) Bright field image of a pristine V2O5/SnO2 nanowire. (b-d) Images of the nanowire after the first lithiation process under 3 nA lithiation current showing Sn nanoparticles aggregating along the surface of the nanowire. !");sQ1[956]=new Array("../7337/1915.pdf","Elastic Properties of GaN Nanowires: Revealing the influence of planar defects on Young's modulus at nanoscale","","1915 doi:10.1017/S1431927615010351 Paper No. 0956 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Elastic Properties of GaN Nanowires: Revealing the influence of planar defects on Young's modulus at nanoscale Sheng Dai1,2, Jiong Zhao1, Xiaoguang Wang3, Zhiwei Shan3, Jing Zhu1 Beijing National Center for Electron Microscopy, School of Materials Science and Engineering, Tsinghua University, Beijing, 100084, China 2. Department of Materials Science and Engineering, University of Michigan, Ann Arbor, Michigan 48109, USA. 3. Center for Advancing Materials Performance from the Nanoscale (CAMP-Nano) & Hysitron Applied Research Center in China (HARCC), Xi'an Jiaotong University, Xi'an, 710049, China. 1. Gallium nitride (GaN) nanowire, a technologically important semiconductor has ignited the research attention in recent years. This materials has also been investigated extensively and proved to process enhanced physical properties and nanodevice applications. Besides its well-known optoelectronic and photonic properties, the study on the mechanical properties of GaN nanowires is essential and imperative for its piezoelectric applications and the designing for GaN-nanowire-based devices during processing and working. In our experiments, GaN nanowires with different structures were synthesized via chemical vapor deposition method. Both single crystalline (SC) GaN nanowires and a new structure, obtuse-angle twin (OT) GaN nanowires were obtained. High resolution electron microscopy was utilized to characterize the structures and the planar defects such as twin boundaries and stacking faults. The results (Figure 1) showed that (001) stacking faults have different relative orientations and volume fractions in SC and OT GaN nanowires, respectively. This is an ideal system for the investigation to explore the influence of planar defects on Young's modulus at nanoscale. In-situ electric-field-induced resonance method was utilized to reveal that SC GaN nanowires, along [120] direction, had the similar Young's modulus as the bulk value at the diameter ranging 92-110 nm. Meantime, the elastic behavior of the obtuse-angle twin (OT) GaN nanowires was disclosed both by insitu SEM resonance technique and in-situ TEM tensile test for the first time (Figure 2). Our results showed that the average Young's modulus of these OT nanowires was greatly decreased to about 66 GPa and indicated no size dependence at the diameter ranging 98-171 nm. A quantitative model based on the rules of mixture is proposed and the calculation results are in great accordance with our experimental data both in two kinds of GaN NWs. Our work demonstrated that the elastic modulus of one dimensional nanomaterials is dependent on the relative orientations and the volume fractions of planar defects. The Voigt model and Reuss model in classical mechanics still remain effective in our tested GaN nanowires and may be extended to some other one-dimensional namomaterials. Moreover, the SFs' effect on the elastic properties also indicates that one may construct new type of functional devices through planar defects engineering. Microsc. Microanal. 21 (Suppl 3), 2015 1916 References: [1] S Dai, J Zhao, M He et al. Nano Letters 15 (2015), p.8. [2] S Dai, J Zhao, M He et al. J. Phys. Chem. C 117 (2013), p.12895. Figure 1. HRTEM characterizations of GaN nanowires showing the planar defects and structure model. (a, b, e) are the results of single crystalline GaN nanowires. (c, d, f) are the results of obtuse-angle twin GaN nanowires. Figure 2. In-situ experiments to measure the Young's modulus of as-synthesized GaN naonowires. (a-b) In-situ SEM electric-field-induced resonance. (c-d) In-situ TEM tensile tests. (e) Young's Modulus E with Diameter d of GaN nanowires from resonance experiments.");sQ1[957]=new Array("../7337/1917.pdf","Quantification of Electrochemical Nanoscale Processes in Lithium Batteries by Operando ec-(S)TEM","","1917 doi:10.1017/S1431927615010363 Paper No. 0957 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantification of Electrochemical Nanoscale Processes in Lithium Batteries by Operando ec-(S)TEM B. L. Mehdi1, J. Qian2, E. Nasybulin2, C. Park3, D. A. Welch4, R. Faller4, H. Mehta5, W. A. Henderson2, W. Xu2, C. M.Wang5, J. E. Evans5, J. Liu2, J. -G. Zhang2, K. T. Mueller5,6, N. D. Browning1 1 Joint Center for Energy Storage Research, Fundamental and Computational Science Directorate, PNNL, Richland, WA, USA 2 Joint Center for Energy Storage Research, Energy and Environmental Directorate, PNNL, Richland, WA, USA 3 Department of Industrial and Manufacturing Engineering, Florida State University, Tallahassee, FL, USA 4 Department of Chemical Engineering and Materials Science, University of California, Davis, CA, USA 5 Joint Center for Energy Storage Research, Environmental Molecular Sciences Laboratory, PNNL, Richland, WA, USA 6 Department of Chemistry, Penn State University, University Park, PA, USA Lithium (Li)-ion batteries are currently used for a wide variety of portable electronic devices, electric vehicles and renewable energy applications [1,2]. In addition, extensive worldwide research efforts are now being devoted to more advanced "beyond Li-ion" battery chemistries - such as lithium-sulfur (Li-S) [3] and lithium-air (Li-O2) [4] - in which the carbon anode is replaced with Li metal. However, the practical application of Li metal anode systems has been highly problematic. The main challenges involve controlling the formation of a solid-electrolyte interphase (SEI) layer and the suppression of Li dendrite growth during the charge/discharge process (achieving "dendrite-free" cycling). The SEI layer formation continuously consumes the electrolyte components creating highly resistive layer, which leads to the rapid decrease of cycling performance and degradation of the Li anode [5]. The growth of Li metal dendrites at the anode contributes to rapid capacity fading (the presence of "dead Li" created during the discharge leads to an increased overpotential on charging) and, in the case of continuous growth, leads to internal short circuits (the dendrites contact the cathode) and extreme safety issues [6]. Here we demonstrate the application of an operando electrochemical scanning transmission electron microscopy (ec-(S)TEM) cell to study the SEI layer formation and the initial stages of Li dendrite growth - the goal is to develop a mechanism for mitigating the degradation processes and increasing safety. Bright field (BF) STEM images in Figure 1 A-C show Li metal deposition and dissolution processes at the interface between the Pt working electrode and the lithium hexafluorophosphate (LiPF6) in propylene carbonate (PC) electrolyte during 3 charge/discharge cycles. A contrast reversal caused by Li metal being lighter/less dense than surrounding electrolyte (Li appears brighter than the background in BF STEM images) allows Li to be uniquely identified from the other components in the system - the only solid material that is less dense than the electrolyte is Li metal. Using these images, we can precisely quantify the total volume of Li deposition, the thickness of the SEI layer (observed as a ring of positive contrast around the electrode) and alloy formation due to Li+ ion insertion during each cycle. Furthermore, at the end of each discharge cycle we can quantify the presence of "dead Li" detached from the Pt electrode, thereby demonstrating the degree of irreversibility (and degradation of Pt electrode) associated with insertion/removal of Li+ during this process. Such analyses provide significant insights into Li metal dendrite growth, which is critical to understand the complex interfacial reactions needed to be controlled for future Li-based and next generation energy storage systems. [7] Microsc. Microanal. 21 (Suppl 3), 2015 1918 References: [1] J. M. Tarascon, M. Armand, Nature, 414, (2001), 414, 359-367 [2] J. B. Goodenough, Y. Kim, Chem. Mater.,22, (2010), 587-603 [3] X. L. Jie, L. F. Nazar, J. Mater. Chem., 20, (2010), 9821-9826 [4] P.G. Bruce, S. A. Freunberger, L. J. Hardwick, J. M. Tarascon, Nat. Mater., 11, (2012), 19-29 [5] P. Verma, P. Maire, P. Novak, Electrochim. Acta, 55, (2010), 6332-6341 [6] J. Wen, Y. Yu, C. Chen, Mater. Express, 2, (2012), 197-212 [7] This work was primarily supported by JCESR, an Energy Innovation Hub funded by DOE-BES. The development of the operando stage was supported by the Chemical Imaging LDRD Initiative at PNNL. PNNL is a multi-program national laboratory operated by Battelle for the U.S. DOE under Contract DEAC05-76RL01830. A portion of the research was performed at the EMSL user facility sponsored by DOE-BER and located at PNNL. The multi-target tracking algorithm is supported by NSF-1334012. Figure 1. Cyclic voltammetry and BF STEM images of Li deposition and dissolution at the interface between the Pt working electrode and the LiPF6/PC electrolyte during the (A) first, (B) second and, (C) third charge/discharge cycles of the operando cell.");sQ1[958]=new Array("../7337/1919.pdf","Resolving the Atomic Structure of Materials Containing Light Elements by Annular-Bright-Field Electron Microscopy","","1919 doi:10.1017/S1431927615010375 Paper No. 0958 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Resolving the Atomic Structure of Materials Containing Light Elements by Annular-Bright-Field Electron Microscopy Lin Gu1, Dongdong Xiao1, and Yuichi Ikuhara2, 3, 4 1. Institute of Physics, Chinese Academy of Sciences/Beijing National Laboratory for Condensed, Matter Physics, Beijing, 100190, P.R. China 2. Institute of Engineering Innovation, The University of Tokyo, Tokyo, 113-8654, Japan 3. Nanostructures Research Laboratory, Japan Fine Ceramic Centre, Nagoya, 456-8587, Japan 4. WPI Advanced Institute for Materials Research, Tohoku University, Sendai, 980-8577, Japan Unraveling the correlations between atomic structure and properties of materials requires resolving the locations and chemical types of all the atoms that make up the structure. Imaging using high-angle scattered electrons in aberration-corrected scanning transmission electron microscopy (STEM), also refers to high angle annular dark field imaging (HAADF), allows giving atomic resolution images that can be directly interpreted over a wide range of specimen thicknesses. However, the image contrast is strongly dependent on the atomic number Z, which makes it difficult to obtain information on the structural arrangement of the light elements in materials which contain the light and heavy elements at the same time [1, 2]. Therefore, one effective strategy to enhance contrast from light atoms is to develop new imaging method which can weaken Z-dependence of image contrast so that light and heavy elements are visible simultaneously. The recently established annular bright-field imaging (ABF) using an annular detector located within a bright-field momentum range in aberration-corrected STEM has made direct imaging of light element come true, providing an unprecedented opportunity to reveal the configuration of light atoms which play a non-trivial role in relevant materials [3, 4]. Figure 1 shows the geometry of annular-bright-field imaging in aberration-corrected STEM, wherein an electron probe with convergent semi-angle is focused to sub-angstrom dimension and scans across the specimen, and an ABF detector at the post column subtends a detecting angle range q. The lithium and oxygen ions in LiFePO4, an important cathode for lithium ion battery, were clearly resolved, demonstrating the robustness of ABF imaging method. In this talk, we will present our recent efforts toward direct imaging of the lights atoms (e.g. Li, O, and Na) in electrode materials for rechargeable batteries and probing their atomic-scale structure evolution under different electrochemical states [5-8]. Microsc. Microanal. 21 (Suppl 3), 2015 1920 References: [1] N. D. Browning et al, Nature 366 (1993), 143. [2] S. Hillyard et al, Ultramicroscopy 58 (1995), 6. [3] E. Okunishi et al, Microsc. Microanal. 15 (2009), 164. [4] S. D. Findlay et al, Appl. Phys. Lett. 95 (2009), 191913. [5] Yang Sun et al, Nat. Commun. 4 (2013), 1870. [6] Yong-Ning Zhou et al. Nat. Commun. 5 (2014), 5381. [7] Haijun Yu et al, Angew. Chem. Int. Ed. 53 (2014), 8963. [8] This work was supported by funding from the NSFC (11174334, 51325206) and the Chinese Academy of Sciences. Figure 1. Schematic illustration of annular-bright-field imaging geometry. An electron probe with convergent semi-angle is focused to sub-angstrom dimension and scans across the specimen. An ABF detector at the post column subtends a detecting angle range q. A demonstration of lithium and oxygen sites within a LiFePO4 crystal along the [010] orientation is also shown.");sQ1[959]=new Array("../7337/1921.pdf","Large-Collection-Angle BF STEM Imaging of Compound Semiconductors","","1921 doi:10.1017/S1431927615010387 Paper No. 0959 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Large-Collection-Angle BF STEM Imaging of Compound Semiconductors T. Aoki1, M.R. McCartney2, and David J. Smith2 1 LeRoy Eyring Center for Solid State Science, Arizona State University, Tempe, AZ 85287-1704, USA 2 Department of Physics, Arizona State University, Tempe, AZ 85287-1504, USA The aberration-corrected scanning transmission electron microscopy (AC-STEM) is a powerful and versatile instrument, which can be equipped with a multitude of detectors including high-angle annulardark-field (HAADF), medium-angle annular-dark-field (MAADF), bright-field (BF), annular-brightfield (ABF), secondary electron (SE), and back-scattered (BSE), as well as energy-dispersive x-ray spectrometer (EDS) and electron-energy-loss spectrometer (EELS). Because HAADF STEM is an incoherent imaging mode with high atomic-number sensitivity ("Z-contrast"), and because it can be easily combined with EELS, it is most often utilized. However, modern AC STEMs can acquire multiple images from different detectors simultaneously, and these other imaging modes can also provide valuable materials information. Because each mode can be considered as complementary, the acquisition of several images allows better understanding and interpretation of materials microstructure. Phase contrast dominates in BF imaging when a small collection angle is used, as implied by the reciprocity theorem. However, when the collection angle is increased to be as large as the convergence semi-angle, phase contrast is suppressed and incoherent contributions dominate image formation [1]. In recent studies of compound semiconductor heterostructures [2,3], we have discovered the usefulness of large-collection-angle bright-field (LABF) STEM imaging, especially when used in combination with HAADF imaging. Here we describe and compare images of GaSb, ZnTe and GaAs, which were acquired with all three materials in [110]-type orientations, using a JEOL ARM-200F operated at 200kV. The LABF and HAADF images were taken with a convergence semi-angle of 21mrad, the BF collection angle was 21mrad, and the HAADF collection angle was 90mrad~220mrad. Figures 1(a) - (d) show LABF and HAADF images of GaSb (where ZGa = 31 and ZSb = 51), as well as corresponding intensity profiles. Figures 1(e) - (h) show images and line profiles for ZnTe (where ZZn = 30 and ZTe = 52), and Figures 1(i) - (l) show GaAs (where ZGa = 31 and ZAs = 33). With the convergence semi-angle equal to the BF collection angle, phase contrast was suppressed, no contrast reversals were observed, and LABF and HAADF images were both in-focus at the same time. In LABF, the positions of the atomic columns appear with dark contrast and show improved visibility, where atomic columns occupied by heavier atoms exhibit darker contrast than columns occupied by lighter atoms. When the Z difference is large (i.e., GaSb and ZnTe), the atomic columns of lighter element become blurred out in the HAADF image, yet they were clearly observed in the corresponding LABF, as seen by comparing Figs. 1 (a) and (c), and Figs. 1 (e) and (g). It is significant that even small Z-differences can be detected in LABF images, as shown by the case of GaAs and its intensity profile in Figs. 1(i) and (j). Overall, it is concluded that LABF STEM is another powerful AC-STEM imaging mode, with improved Zsensitivity, that is highly suited for studying complicated heterostructures at the atomic scale [4]. References: 1. J. Liu and J. M. Cowley, Ultramicroscopy 52 (1993) 335-346. 2. D. J. Smith, et al., Microscopy 62 (2013) S65-S73. 3. D.J. Smith, et al., J. Phys. Conf. Ser. 471 (2013) 012005. 4. We acknowledge the use of facilities in the John M. Cowley Center for High Resolution Electron Microscopy at Arizona State University. Microsc. Microanal. 21 (Suppl 3), 2015 1922 (a) (b) Ga Sb Ga Sb (c) (d) Ga Sb Ga Sb (e) (f) Zn (g) (h) Zn Te Te Zn Te Te Zn (i) (j) Ga (k) (l) Ga As As Ga Ga As As Fig. 1. (a) GaSb LABF image; (b) intensity profile of GaSb LABF image; (c) GaSb HAADF; (d) intensity profile of GaSb HAADF image; (e) ZnTe LABF image; (f) intensity profile of ZnTe LABF image; (g) ZnTe HAADF image; (h) intensity profile of ZnTe HAADF image; (i) GaAs LABF image; (j) intensity profile of GaAs LABF image; (k) GaAs HAADF image; (l) intensity profile of GaAs HAADF image.");sQ1[960]=new Array("../7337/1923.pdf","Atomic Scale Structure and Bonding Configurations in Monocrystalline Al1-xBxPSi3 Alloys Grown Lattice Matched on Si(001) Platforms","","1923 doi:10.1017/S1431927615010399 Paper No. 0960 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic Scale Structure and Bonding Configurations in Monocrystalline Al1-xBxPSi3 Alloys Grown Lattice Matched on Si(001) Platforms P. Sims1, T. Aoki2, J. Men�ndez3, and J. Kouvetakis1 1. 2. Department of Chemistry & Biochemistry, Arizona State University, Tempe, AZ 85287 USA LeRoy Eyring Center for Solid State Science, Arizona State University, Tempe, AZ 85287 USA 3. Department of Physics & Astronomy, Arizona State University, Tempe, AZ 85287 USA Atomic-resolution imaging and element-selective mapping of composition and structure in crystalline solids using aberration corrected microscopy with probe sizes comparable to atomic spacing is a powerful technique which has to date been successfully used to study known systems with classical structures such as hexagonal BN [1] and perovskites [2]. In this study we have used this method to directly investigate atomic scale structure, and bonding configurations in newly synthesized Al1-x BxPSi3/Si(001) alloys comprised of earth abundant elements. These materials are grown on Si buffered Si(001) via reaction of Al(BH4)3 and P(SiH3)3 using low pressure gas source molecular beam epitaxy. The reactions deliver tetrahedral AlPSi3 and BPSi3 building blocks, which are expected to incorporate intact into the crystal, creating a Si-like framework containing isolated Al-P and B-P donor-acceptor pairs. The addition of B into AlPSi3 facilitates lattice matching of the material with the Si template yielding fully relaxed structures devoid of mismatch induced defects as required for application in photovoltaic devices. In this study the bonding patterns of these materials are directly resolved and the constituent atoms are explicitly identified thereby yielding the alloy composition/structure at the Angstrom level. The work shows that the new materials are likely constructed via directed molecular assembly of chemical building blocks sharing the same chemical composition and basic tetrahedral motifs as the solid phases. Another unique functionality is the enhanced light absorption that is significantly extended relative to Si making these Si-based lattice-matched systems likely candidates as top junction materials in tandem solar cells designs for next generation Si-based PV technologies with efficiencies exceeding that of Si. XTEM (JEOL-4000EX) and probe-corrected STEM/EELS experiments were performed using a Nion UltraSTEM 100 equipped with HERMESTM and Gatan EnfiniumTM EELS spectrometer. Figures 1 (a) and (b) are XTEM/STEM images of the entire epilayer and interface region, demonstrating high quality epitaxy of monocrystalline layers devoid of interfacial dislocations and threading defects. The diffraction pattern reveals a Si-like structure with a lattice parameter nearly equal to that of Si, depending on the amount of B uptake, which reaches up to 6 at.% relative to Al. EELS indicates that the films are single phase alloys with uniform elemental distributions and not a mixture of Si and zinc-blende Al1-xBxP. Atomic resolution mapping identified distinct randomly oriented motifs for the Al-P pairs as expected for single phase-alloy with a disordered structure. The Si component is uniformly arranged throughout individual crystal columns consistent with a diamond lattice (Figure 2(d)). Raman spectroscopy provides evidence that the B atoms are incorporated as isolated B-P units with dilated bond distances relative to zinc-blend BP (Figure 3(a)). The single phase character of the samples is corroborated by triple axis XRD analysis of 004, 135 and 224 reflections (Figure 3 (b)). Measurements of the dielectric function using spectroscopic ellipsometry revealed higher absorption relative to bulk Si in the visible range [3]. References: Microsc. Microanal. 21 (Suppl 3), 2015 1924 [1] K. Suenaga et al, J. Electron Microsc. 61 (2012) p. 285. [2] D. Muller et al, Science, 319 (2008) p. 1073. [3] This work was supported by NSF and NREL. The authors acknowledge the use of facilities in the John M. Cowley center for High Resolution Electron Microscopy at Arizona State University. [4] P. Sims et al, Chem. Mater., submitted Figure 1. TEM/STEM images of Al1-xBxPSi3/Si(001); (a) XTEM of full epilayer grown lattice matched on Si(001) wafer. The arrows denote the location of the interfaces. (b) STEM HAADF image showing full alignment of the (111) lattice fringes. The close chemical and structural relationship between the two materials allows near perfect crystal integration devoid of interfacial defects. Figure 2. MLS fitted STEM EELS map of Al1xBxPSi3/Si(100); (a) HAADF image; (b) Al K map; (c) P K map; (d) Si K map; (e) Color Overlay; (f) structural model for the AlPSi3 in the same projection. Si Figure 3. (a) Raman spectra of Al1-xBxPSi3 and Si. The peak at 631 cm-1 is assigned to vibrations form isolated BP pairs in the structure [4]. Figure 3. (b) XRD -2 plots of Al1xBxPSi3 (blue) and AlPSi3 (red) showing 004 reflections of the diamond lattice");sQ1[961]=new Array("../7337/1925.pdf","Local Chemical and Deformation Profiles in InAs/AlSb Multilayer Structures for Quantum Cascade Lasers","","1925 doi:10.1017/S1431927615010405 Paper No. 0961 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Local Chemical and Deformation Profiles in InAs/AlSb Multilayer Structures for Quantum Cascade Lasers B. Warot-Fonrose1,4, J. Nicola�1,4, M. Vallet1,4, C. Gatel1,4, R.Teissier2, A. N. Baranov2, C. Magen3,4 and A. Ponchet1,4 1 CEMES CNRS-UPR 8011, Universit� de Toulouse, 31055 Toulouse, France 2 IES CNRS-UMR 5214, 34095 Montpellier, France 3 Laboratorio de Microscop�as Avanzadas (LMA), Instituto de Nanociencia de Arag�n (INA) ARAID and Departamento de F�sica de la Materia Condensada, Universidad de Zaragoza, 50018 Zaragoza, Spain 4 Transpyrenean Associated Laboratory for Electron Microscopy (TALEM), CEMES-INA, CNRS-Universidad de Zaragoza Optoelectronic devices are widely studied to master the growth process of the semiconductors and achieve the best performances. Some of these systems have very large conduction band offsets (OBC) , which make them attractive for use in devices using the in-band electronic transitions . For example, the multilayer InAs / AlSb deposited on InAs has been used with success to achieve quantum cascade lasers (QCL) for emitting short wavelength below 3 m [1]. These Quantum Cascade Lasers are the result of a complex architecture with many semiconducting layers deposited as very thin films. Any defect or strain accumulation could lead to the loss of optical properties. Among various structural parameters, the determination of the deformations at the local level is crucial. Interfaces are especially of major interest as they concentrate the chemical / structural transitions. In InAs/AlSb samples, the absence of any common atom between the barrier and the well induces that new bonds need to be formed. Depending on the AlAs or InSb type, compression or tension will be experienced by the interface. Local chemistry and strain state are therefore intimately linked and this work proposes to combine these two measurements on this technologically interesting system. The deformations are determined on a Cs corrected Tecnai microscope in high resolution mode. GPA analysis allows for the determination of deformation profiles across the layers and the interfaces. Special care has been taken to select proper images and imaging conditions to apply the treatment and an extensive study of the influence of the mask size used has been made. Combined with the deformation measurements, STEM-HAADF images have been acquired on a probe Cs corrected Titan (INA Zaragoza) to quantify the intensity variation at the vicinity of the interfaces. The EELS signal has also been recorded and correlated with HAADF images intensity. Several samples have been studied with specific interface preparation. The sequence of the opening of the different cells in the MBE chamber has been tuned to favor either compressive or tensile deformation at the interfaces. Considering local strain and chemistry, we estimated the interface composition and discussed the mechanisms of interface formation for the different growth sequences. One example is shown on figure 1c, where the deformation profile extracted from the deformation map is plotted. The tensile stress at both interfaces is clearly visible. Combined with STEM-HAADF image intensity analysis (figure 2), we found therefore that formation of the AlAs-type interface is spontaneously favored, due to the very high thermal Microsc. Microanal. 21 (Suppl 3), 2015 1926 stability compared to the InSb-type interface. We also showed that the interface composition could be tuned using an appropriate growth sequence [2,3]. References [1] J. Devenson et al, Applied Physics Letters, 91, (2007), 141106 [2] J. Nicolai et al, Applied Physics Letters, 104, (2014), 031907 [3] Acknowledgements: The authors are grateful to C. Crestou for TEM samples thinning. This work is supported by the French national project NAIADE (ANR-11-BS10-017) and the European Union Seventh Framework Programme under Grant Agreement 312483 - ESTEEM2 (Integrated Infrastructure Initiative�I3). Figure 1. a) HREM image of a InAs/AlSb/InAs sample, the white arrow indicates the growth direction, b) mapping of the deformation of the planes parallel to the interface, c) deformation profile obtain by the GPA method average on the whole width of the image (30 nm) Figure 2. STEM-HAADF image of the same sample as the one presented on figure 1 (left), intensity profile along the growth direction");sQ1[962]=new Array("../7337/1927.pdf","STEM Optical Sectioning for Imaging Screw Displacements in Dislocation Core Structures","","1927 doi:10.1017/S1431927615010417 Paper No. 0962 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 STEM Optical Sectioning for Imaging Screw Displacements in Dislocation Core Structures Hao Yang1, Juan G Lozano1, Timothy J Pennycook2, Lewys Jones1, Peter B Hirsch1 and Peter D Nellist1,3 1. 2. University of Oxford, Department of Materials, Parks Rd, Oxford, UK Faculty of Physics, University of Vienna, Vienna, Austria 3. EPSRC SuperSTEM Facility, Daresbury Laboratory, Warrington, UK Aberration corrected (scanning) transmission electron microscopes (STEM) with subangstrom resolution have advanced our knowledge of the atomic structure of edge dislocations viewed end-on, with the tensile or compressive strain normal to the dislocation being clearly visible. Atomic displacements associated with screw dislocations however cannot be observed in end-on images because the helical screw displacements are parallel to the viewing direction. Such screw dislocations do however show small rotational displacements in thin foils due to surface stress relaxations known as the Eshelby twist. [1] We show that the helical displacements around a screw dislocation can be generally imaged with the dislocation lying transverse to the electron beam by "optical sectioning" in highangle annular dark-field (HAADF) imaging in STEM. In optical sectioning, the few nanometre depth of focus of aberration corrected microscopes is utilized to extract information along the beam direction by focusing the electron probe at specific depths within the sample. This novel technique is applied to the study of the screw component in the dissociation reaction of a mixed [a+c] dislocation in GaN. A focal series of experimental images were recorded using a Nion UltraSTEM100 aberration-corrected STEM operating at 100 kV. A 1m thick sample of GaN, grown by metalorganic vapour phase epitaxy on a sapphire substrate, was thinned to be viewed along the a-axis. A dislocation was found lying in the plane of the sample, and characterized using weak-beam imaging to be of a mixed [a+c] type along [0001]. As the STEM electron beam is focused closer to the dislocation, the shearing of the (0002) planes becomes more localized in the image, and a more detailed observation of the screw displacements shows that the shearing occurs equally along two distinct lines along [0001] (see Figure 1). It is therefore apparent that the screw component of the dislocation has dissociated according to the reaction c=[�c+�c] confirming the assumption made in previous end-on observations. [2] The dissociation of screw is also confirmed by comparing experiment with image simulations of both dissociated and undissociated structure models, and a good match is found between experiment and simulated images of the dissociated screw model. Quantitative analysis of the screw displacement field under a set of specimen thickness, defocus, and screw dissociation width conditions shows that the dissociation distance is around 1.65nm in this work, and this method provides sufficient sensitivity to detect a dissociated screw in GaN with a dissociation distance as small as 1.1nm (see Figure 2). [2] Details of the experiment results of STEM optical sectioning as well as image simulations will be presented. [3] Microsc. Microanal. 21 (Suppl 3), 2015 1928 [1] Lozano, J.G. et al. Physical Review Letters 113, 135503. (2014) [2] Yang, H. et al. Nature Communications (Under Review) [3] The authors acknowledge funding from the EPSRC (grant numbers EP/K032518/1 and EP/K040375/1) and the EU Seventh Framework Programme: ESTEEM2. Figure 1. Experimental and simulated STEM-ADF image (inset) of the �[a+c] + �[a+c] dissociated dislocation lying perpendicular to the electron beam. The screw displacements associated with each of the partial dislocations can be observed, as indicated by the overlaid solid and dashed lines following the closer-to-focus stronger intensity peaks and furtherfrom-focus weaker intensity peaks, respectively. A simulated image of an isotropic elastic model of a dissociated dislocation with a 1.65nm dissociation distance is overlaid. The simulation was done with the beam focused 5 nm below the top entrance surface of a 10-nm thick foil. Figure 2. Determining screw dissociation through quantitative analysis of screw displacements near the dislocation. (a) The images in the fault region consist of pairs of closely-spaced peaks whose relative positions correlate with the amount of screw displacement. A Radon transform is used to measure changes in the column-pair angles across the dislocation. (b) The column pair angles have been measured for both the experimental and simulated images of dissociated screw (1.65nm dissociation distance) and undissociated screw dislocations, and a good match is found between experiment and the dissociated screw column angles, confirming the experimental observation of a dissociated screw. (c) The rate of angle change across the dislocation depends on the dissociation distance and probe defocus. By fitting sigmoid functions to the angle plots, and comparing the fitted slope-controlling parameter in the sigmoid functions of the experimental image with those of simulated images of different possible dissociation distance and defocus values, the dissociation distance of 1.65nm clearly gives the best agreement.");sQ1[963]=new Array("../7337/1929.pdf","A Multiple-Technique Approach for Resolving the Surface Structure of Lithium and Manganese Rich Transition Metal Oxides.","","1929 doi:10.1017/S1431927615010429 Paper No. 0963 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A Multiple-Technique Approach for Resolving the Surface Structure of Lithium and Manganese Rich Transition Metal Oxides. Alpesh Khushalchand Shukla1, Quentin Ramasse2, Colin Ophus3, Hugues Duncan4 and Guoying Chen1 1. Energy Storage and Distributed Resources Division, Lawrence Berkeley National Laboratory, Berkeley, United States. 2. SuperSTEM, Daresbury, United Kingdom. 3. National Center of Electron Microscopy, Molecular Foundry, Lawrence Berkeley National Laboratory, Berkeley United States. 4. Kinestral Technologies, South San Francisco, United States High-capacity lithium transition-metal oxides (Li1+xM1-xO2, where M is usually a combination of transition metals such as Mn, Co and Ni) have been extensively studied recently due to their potential application in high-energy Li-ion batteries. Structural differences observed on the bulk and the surface, particularly the enrichment of Ni and/or Co and lower TM oxidation states on the surface, have been attributed to effects of cycling [1]. However, recent studies [2, 3] have shown that such transition metalrich surface layers are also observed in pristine materials. To ultimately achieve improved rate capability and stability of the batteries, it is necessary to fully understand the structure and composition of these surface layers. In this paper, we report the results from structural and spectroscopic analysis of these complex oxides using a variety of techniques that span a wide length-scale, including aberration corrected STEM, electron energy loss spectroscopy (EELS) and X-ray energy dispersive spectroscopy (XEDS). Li1.2(Ni0.13Mn0.54Co0.13)O2 crystals with a plate-shaped morphology were prepared using a molten salt synthesis method and were studied using HAADF STEM imaging and EELS spectrum imaging in a Nion UltraSTEM electron microscope operating at 100 keV. XEDS experiments were conducted in an FEI Titan microscope equipped with a quad-silicon drift detector operating at 120 keV. HAADF STEM imaging using multiple zone axes revealed that the bulk of the primary particles consisted of variants of monoclinic phase, while the surface layer, which was typically 2 nm in thickness, had a spinel structure. An orientation relationship between the bulk and the spinel surface was found to be as follows: L23 ratios obtained from EELS spectra across the surface revealed that the oxidation states of Mn and Co, which was 4+ and 3+ in the bulk, decreased at the surface and the fine structure of O and Mn edge indicated the presence of spinel at the surface. EELS spectrum imaging also revealed higher concentration of Co and Ni on the surface. Analysis of Li K-edge showed a decreased concentration of Li at the surface, indicating the presence of antisite defects, which was also confirmed from HAADF STEM imaging. XEDS experiments confirmed the presence of Co and Ni segregation at the expense of Mn depletion along with depletion of oxygen at the surface, which explains the charge balance in the spinel surface in spite of lower oxidation states of Co and Mn. This result is consistent with earlier studies on oxygen-deficient spinels that have shown that spinels can easily accommodate oxygen vacancies [4]. Thus using a multiple-technique approach, we deduced that the surface of Li-rich layered oxides is a Co Microsc. Microanal. 21 (Suppl 3), 2015 1930 and Ni-rich, oxygen-deficient spinel with several anti-site defects. References: [1] B Xu et al, Energy & Environmental Science 4 (2011) p. 2223. [2] M Gu, et al, Nano Letters 12 (2012), p. 5186. [3] H Dixit, et al, ACS Nano 8 (2014), p. 12710. [4] A. Bergstein and P. Kleinert, Journal of Physics and Chemistry of Solids 26 (1965), p. 1181. [5] The authors acknowledge support of the National Center of Electron Microscopy, Lawrence Berkeley Lab, which is supported by the U.S. Department of Energy under Contract # DEAC02-05CH11231. Figure 1. (a) Structural model, STEM simulation and experimental HAADF STEM image showing spinel structure on the surface and monoclinic structure in the bulk, in [111]S and [103]M zone axes, respectively. (b) XEDS maps showing enrichment of mostly Co and some Ni on the surface. Figure 2. (a) Series of EEL spectra showing the surface (top) and the bulk (bottom) corresponding to the HAADF image in Figure 1a. (b), (c) and (d) L23 ratio calculated from (a) showing decreased oxidation states of Mn and Co at the surface.");sQ1[964]=new Array("../7337/1931.pdf","Structure of Ru/Pt Nanocomposite Films Fabricated by Plasma-Enhanced Atomic Layer Depositions","","1931 doi:10.1017/S1431927615010430 Paper No. 0964 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Structure of Ru/Pt Nanocomposite Films Fabricated by Plasma-Enhanced Atomic Layer Depositions Masahiro Kawasaki1, Chien-Nan Hsiao2, Jer-Ren Yang3 and Makoto Shiojiri4* 1. 2. JEOL USA Inc., 11 Dearborn Road, Peabody, MA 01960, USA National Applied Research Laboratories, Instrument Technology Research Center, 20, R&D Rd. VI, Hsinchu Science Park, Hsinchu, 30076, Taiwan, ROC 3. Department of Materials Science and Engineering, National Taiwan University, Taipei 10617, ROC 4. Kyoto Institute of Technology, Kyoto 606-8585, Japan * Present address: 1-297 Wakiyama, Kyoto 618-0091, Japan A nanoparticle comprising a Ru core covered with a shell of Pt atoms (Ru@Pt core�shell nanoparticle) has attracted considerable attention as a very active catalyst to remove CO from hydrogen feeds through selective oxidation since significant quantities of CO in reforming hydrocarbons poison the current hydrogen fuel-cell devices [1-3]. The particles are chemically prepared by deposition of Pt atoms from a deoxygenated solution of H2PtCl6 on the Ru nanoparticles dispersed on XC-72 active carbon [1] or by Pt coating by adding PtCl2 to Ru/glycol colloid [2-5]. Here, we report on the structural analysis of a new Ru and Pt nanocomposite film (Ru/Pt film), which was prepared by successive plasma-enhanced atomic layer depositions (PE-ALDs) of Pt and Ru, by means of analytical electron microscopy. PE-ALD of Pt films was performed using MeCpPtMe3 [C5H4CH3Pt(CH3)3] and Ar/O2 plasma. The Pt films deposited on the Si wafer at 200oC had a low resistivity of 16.2 cm. Ru films were grown using Ru(EtCp)2 [Ru(C2H5C5H4)2] precursor. The resistivity of the Ru thin film deposited on a Si wafer at 300oC was as low as 11 cm. For transmission electron microscopy (TEM) and scanning transmission electron microscopy (STEM) observations, a Ru/Pt film was prepared on a very thin amorphous carbon film supported on the Cu grid for TEM. The MeCpPtMe3 PE-ALD of 30 cycles was made on the carbon films heated at 300oC, followed by PE-ALD of 100 cycles using EBCHDRu [C14H18Ru] as the precursor at the same temperature. The EBCHDRu provided a comparable growth rate of Ru film with the Ru(EtCp)2. Analytical electron microscopy was performed with an atomic scale high-resolution (HR) JEOL ARM200FC microscope equipped with a cold field-emission gun and aberration correctors as well as a 60 mm2 silicon drift detector for energy dispersive X-ray spectroscopy (EDS), and with a JEM-2800 microscope with an objective lens of Cs of 0.7 mm attached with a JEOL 100 mm2 SDD for EDS. Analytical electron microscopy shown in Figures 1 and 2 and HRTEM revealed that the 30-cycle PE-ALD of Pt forms Pt ribbons 2~3 nm wide as a result of island growth on the C substrate. The following 100-cycle PE-ALD of Ru forms pure Ru ribbons 2~3 nm wide between the Pt ribbons, and changes the Pt ribbons to PtRu (7:3) alloy ribbons with the fcc A1 structure. The pure Ru ribbons are composed of Ru crystallites with the hcp A3 structure but poor in quality exhibiting ill-defined lattice fringes in HR-BF and HAADF STEM images. Hence the Ru/Pt film is a nanocomposite film comprising pure Ru ribbons and PtRu alloy ribbons with 2~3 nm in their widths, which resembles the Ru@Pt core�shell nanoparticles with the fcc A1 structure of PtRu and poor crystalline hcp A3 structure of Ru [3]. The structure of Ru/Pt film or the sizes of Pt and Ru particles are controllable by changing the PE-ALD condition such as the number of deposition cycles, deposition temperature and post-deposition heat treatments. The PE-ALD would be one of potential techniques to produce a fuel cell catalyst such as Ru@Pt core�shell nanoparticles. [1] S.R. Brankovic et al, Electrochem. Solid State Lett. 4, A217 (2001). Microsc. Microanal. 21 (Suppl 3), 2015 1932 [2] S. Alayoglu et al, Nature Materials 7, 333 (2008). [3] S. Alayoglu et al, ACS Nano, 3, 3127 (2009). [4] P. Ochal et al, J. Electroanalytical Chem. 655, 140 (2011). [5] N. Muthuswamy et al, International J. Hydrogen Energy, 38, 16631 (2013). Fig. 1. EDS mapping and electron diffraction of the Ru/Pt nanocomposite film. (a) HAADF-STEM image. (b,c) EDS maps of Pt-M and Ru-L, which were arranged from the original EDS data to display the relative atomic % of Pt and Ru. (d) Composite EDS map corresponding to (b) and (c). (e) Electron diffraction pattern, where the standard powder X-ray diffraction intensities for Pt and Ru crystalline are inset on the right and left, respectively. Fig. 2. (a, b) EDS spectra from local areas indicated by squares A and B in Fig. 1(a), respectively. The insets in (a) and (b) show the enlarged images of the corresponding areas to A and B in Fig. 1(d). Assuming the same absorption effect for Pt-M and Ru-L and using Cliff-Lorimer factors of 1.568 for Pt-M and 1.713 for Ru-L, from these spectra we estimated roughly the relative atomic % of Pt to Ru to be 71/29 (=~7:3) for area A and 3/97 (=~0:1) for area B.");sQ1[965]=new Array("../7337/1933.pdf","Microstructural and Chemical Analysis of HgCdTe/CdTe/ZnTe/Si (211) for Infrared Detectors","","1933 doi:10.1017/S1431927615010442 Paper No. 0965 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microstructural and Chemical Analysis of HgCdTe/CdTe/ZnTe/Si (211) for Infrared Detectors M. Vaghayenegar1, B. L. VanMil2, S. Simingalam2, Y. P. Chen2 and D. J. Smith3 1. 2. School for Engineering of Matter, Transport and Energy, Arizona State University, Tempe, AZ 85287 Army Research Laboratory, 2800 Power Mill Rd, Adelphi, MD 20783 3. Department of Physics, Arizona State University, Tempe, AZ 85287 HgCdTe, (MCT), has been the predominant material for Infrared (IR) focal-plane-array (FPA) technology for the past 50 years [1]. However, growth of high quality MCT on scalable preferred substrates, such as Si, has been hindered due to the dislocation density induced by the large lattice mismatch (i.e. 19.5%) between Si and MCT [2]. In the present work, MCT layers were grown by molecular beam epitaxy on 2x2 cm2 composite (211) substrates consisting of thin (~15 nm) ZnTe and thick (~10 �m) CdTe . The MCT structures consisted of LWIR or SWIR (nominal x=0.22 and 0.45, respectively) with varying thickness HgTe buffer layers (0, 4.7 or 18.8 nm thick) at the MCT/CdTe interface. The SWIR epilayers were grown with a final CdTe capping layer. Etch-pit density (EPD) measurements were obtained using Benson etch to determine which buffer reduced the defect density [3]. EPD indicates a slight improvement with a 15s (~4.7 nm) HgTe buffer layer. These samples were investigated by TEM to determine the effectiveness of the HgTe buffer layer as a defect blocking layer. <110> XTEM samples were prepared using standard mechanical polishing and dimpling to thicknesses of about 10-12 �m, with subsequent argon-ion milling at liquid nitrogen temperature to minimize possible ion-milling-induced artifacts [4]. For final thinning and to minimize formation of amorphous layers, low-angle, low voltage (approximately 2keV) milling was used. Cross section samples were studied using bright-field and high-resolution TEM, using Philips CM200-FEG, and JEOL JEM4000EX, respectively. Figure 1(a) shows SWIR material with 60 sec HgTe initial layer, with defects at the MCT/CdTe interface and in the upper regions of MCT that would be likely to affect the IR detector performance. Density of defects is larger at the CdTe/Si interface, Fig. 1(b), but decreases as the CdTe thickness increases. Higher magnification images of MCT/CdTe interface, Figs. 1(c) and (d), shows a (~14 nm) HgTe layer between the MCT/CdTe layers. Figure 2(a) shows a highly defective ZnTe layer between CdTe and Si which was intended to relieve strain during growth of the thick CdTe layer. Higher magnification inset at the left corner shows the presence of {111} stacking faults at the ZnTe/Si interface. The SAD pattern shows a 3.7 degree rotation between CdTe and Si crystal lattices, as expected based on previous work [5]. FFT calculations and EDS line scan profiles, Fig. 2(b), also confirm that this layer is ZnTe. Overall, the TEM observations showed that growth without the HgTe buffer layer had significantly more defective growth [6]. References: [1] M. A. Kinch., J. Electron. Mater. 39 (2010), 1043. [2] S. Farrell et al., J. Electron. Mater. 39 (2010), 43. [3] D. Benson et al, J. Electron. Mater. 38 (2009), 1771. [4] C. Wang et al., J. Vac. Sci. Technol. A24 (2006), 995. [5] S. Y. Woo et al., Appl. Phys. Lett. 102 (2013), 132103. [6] This work was supported by Army Research Office Grant #63749-EL.We gratefully acknowledge the use of facilities within the John M. Cowley Center for HREM at Arizona State University. Microsc. Microanal. 21 (Suppl 3), 2015 1934 (a) HgCdTe (c) HgCdTe CdTe CdTe (b) (d) CdTe HgTe CdTe Si Figure 1. XTEM images of MCT/MT/CT/Si Heterostructure: (a) Low-mag image of MCT/CT, (b) Low-mag image of CT/Si, (c&d) High-mag images of HgCdTe/HgTe/CdTe interface region. HgCdTe (a) CdTe ZnTe (b) Si CdTe ZnTe Si Figure 2. (a) XTEM images of CdTe/ZnTe/Si interface region, at two different magnifications with SAD pattern as inset, (b) EDS line scan profile across CdTe/ZnTe/Si interface.");sQ1[966]=new Array("../7337/1935.pdf","Beam Effects During In Situ Potential Cycling and Imaging of Sulfuric Acid and Platinum Electrodes","","1935 doi:10.1017/S1431927615010454 Paper No. 0966 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Beam Effects During In Situ Potential Cycling and Imaging of Sulfuric Acid and Platinum Electrodes Todd Brintlinger,1 Corey Love,2 and Olga Baturina2 1 2 Materials Division, U.S. Naval Research Laboratory, Washington, DC 20375 Chemistry Division, U.S. Naval Research Laboratory, Washington, DC 20375 A new class of transmission-electron-microscope (TEM) compatible cells has been developed and now allow the in situ and operando observation of numerous electrochemical reactions on the nanoscale.[1] This then requires a quantitative characterization of an electron beam's effect on an actively electrochemically cycled system. Previous research has included beamprecipitated formation of metal nanocrystals directly formed from metal-salt solution,[2, 3] as well as for the imaging of highly reactive Li-based electrochemical systems using both wet and dry electrolytes.[4, 5] These studies span the expected electron dose rate range from highly intense beams (1000s e-/ �2�s) designed to force crystallization to the smallest possible beam strengths (fractions of e-/ �2�s) for the minimally invasive imaging of electrochemical systems. To understand beam effects, we use a model system of platinum electrodes and sulfuric acid to investigate how the presence of an intermediate-to-low intensity electron beam (~1 e-/ �2�s) affects cyclic voltammograms (CVs), and how the CVs change with time following beam exposure in a flowing electrolyte. Figure 1 shows the unchanging platinum electrode in a flowing-electrolyte style electrochemical cell. Here, we are using the largest field-of-view and smallest beam intensity for the `standard' viewing conditions (2000x, intermediate condenser aperture, which is visible in image) in our TEM (JEM2200FS). Figure 2 shows the effect of turning the beam on and off during potential cycling. Here, we note a small increase in current when the beam is applied, with a ~1nA difference between the different CVs at -0.33 V, which is to be expected if the irradiated electrode collects all current from the primary electron beam. More importantly, a static potential shift to one of the `beam on' cycles (applied ex post facto by simple subtraction) puts the beam back on a very similar curve to the `beam off' CV. Finally, we see that `beam off' CVs can be reproduced following application of the beam (not shown) after a brief period of leaving the beam off, which indicates the flow of liquid through the cell allows for replenished electrolyte to replace the irradiated portions such that the original, beam-off CV is recovered. We will present these and other results using electrochemical impedance spectroscopy on this canonical electrochemical system, and discuss their impact on in situ TEM. References 1. 2. 3. 4. 5. Zheng, H., Y.S. Meng, and Y. Zhu, MRS Bulletin, 2015. 40(01): p. 12-18. Chen, X., et al., Acta Materialia, 2012. 60(1): p. 192-198. Zheng, H., et al., Science, 2009. 324(5932): p. 1309-1312. Zeng, Z., et al., Faraday Discussions, 2014. Liu, X.H., et al., Advanced Energy Materials, 2012. 2(7): p. 722-741. Microsc. Microanal. 21 (Suppl 3), 2015 1936 Figure 1: Bright-field transmission electron micrograph (Mag = 2000x) of commercial electrochemical cell with platinum electrodes and flowing 0.2M sulfuric acid. during cycles shown in Fig. 2. Electrode is not seen to change during cyclic voltammagram, nor are bubbles formed. Figure 2: Cyclic voltammagrams of entire cell during imaging of electrode in Fig. 1. Sweep rate is 100mV/sec.");sQ1[967]=new Array("../7337/1937.pdf","A study in the formation of Li7La3Zr2O12 as a Garnet-Type Ionic Conductor Synthesized by Flame Combustion.","","1937 doi:10.1017/S1431927615010466 Paper No. 0967 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A study in the formation of Li7La3Zr2O12 as a Garnet-Type Ionic Conductor Synthesized by Flame Combustion. Justin Roller1, Yang Wang2 and Radenka Maric3 1. 2. FEI Company, NA Nanoport, Portland, USA. Department of Materials Science and Engineering at the University of Connecticut, Storrs, USA. 3. Department of Materials Science and Engineering and Department of Chemical and Biomolecular Engineering at the University of Connecticut, Storrs USA. Solid-state lithium batteries capable of stability against chemical reaction with Li up to voltages higher than 5.5 V are considered a promising alternative to liquid or gel electrolytes that dominate the Libattery market [1]. In addition, they can be deposited as thin film batteries (TFBs) and integrated directly onto microprocessor chips [2]. The transition to solid-state dry batteries will allow for miniaturization due to a decreased electrolyte thickness, easier handling during manufacture, increased safety due to the lack of a flammable electrolyte coupled with a wide electrochemical potential window, and by the reduced environmental impact of oxide based electrolytes. Li7La3Zr2O12 (LLZ) is a garnet-type ionic conductor with bulk conductivities of 3 x 10-4 S cm-1 at 25�C. In order to maintain such a high ionic conductivity the LLZ must crystalize in the cubic Ia-3d space group and not the preferred tetragonal I41/acd space group which has an ionic conductivity that is 2 orders of magnitude lower [3]. Reactive Spray Deposition Technology (RSDT) was the route used to manufacture LLZ in an open atmosphere one-step process as a thin film [4]. In order to understand the mechanisms of nanoparticle formation, ensure complete precursor conversion, study the effect of process parameters on crystallinity, and map the elemental distributions of lanthanum and zirconium, a STEM and TEM study was undertaken. TEM grids were used to sample forming nanoparticles under various processing conditions such as fuel flow rate and equivalence ratios. In addition, films were deposited on various substrates and then subjected to annealing temperatures to examine the resulting morphology and crystal structure. The images and maps displayed in Figures 1-3 were obtained from samples prepared by wiping-off a portion of thin film deposited onto an Aluchrome YHf alloy and then sonicating the wipe in isopropanol (IPA). The IPA was then pipetted onto a 400 mesh grid having an ultrathin carbon film supported on a lacey film. All images were acquired on an FEI 200 kV Metrios TEM equipped with ChemiSTEMTM technology (X-FEG source and Super-X EDS). Figure 1 shows a 2 �m field of view (FOV) HAADF image and highlights the Z-contrast of the La (Z=57) and Zr (Z=40) relative to the lacey and thin film carbon. The TEM image and accompanied diffraction pattern (DP) highlight the thickness and diffraction contrast of the nanoparticle grains. The randomly oriented particles contribute to the polycrystalline spots observed in the DP. Figure 2 is a Super-X EDS map of the elements present. The La and Zr are distributed as circular particles ranging in diameter from 20-130 nm. In addition, there are areas where the La and Zr that are not circular, but exist as smaller clustered particles. This indicates different mechanisms of precursor conversion in the flame. The O is co-located with the La/Zr but also exists in other areas. Figure 3 highlights an area where both the nanoparticles and the clusters are present. Microsc. Microanal. 21 (Suppl 3), 2015 1938 References: [1] R Murugan, V Thangadurai, and W. Weppner, Angew. Chem. Int. Ed. 46 (2007) p. 7778. [2] J Tan and A Tiwari, Journal of the Electrochemical Society 1(6) (2012), p. Q57. [3] Y Wang and W Lei, Journal of Power Sources 275 (2015), p.612. [4] J Roller, J Renner, H Yu et al., Journal of Power Sources 271 (2014), p. 366. Figure 1. a) HAADF image with large 2 m FOV of LLZ nanoparticles and clustered areas. b) Same area imaged in bright field TEM with the associated c) diffraction pattern. Figure 2. Super-X EDS maps of LLZ highlighting the elemental distribution of Zr and La in the nanoparticles and clusters. Figure 3. Simultaneous acquisition of a) bright field, b) annular bright field, c) dark field, and d) HAADF images of LLZ nanoparticles and cluster areas.");sQ1[968]=new Array("../7337/1939.pdf","Annular Bright Field Scanning Transmission Electron Microscopy � Direct and Robust Atomic-Resolution Imaging of Light Elements in Crystalline Materials","","1939 doi:10.1017/S1431927615010478 Paper No. 0968 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Annular Bright Field Scanning Transmission Electron Microscopy � Direct and Robust Atomic-Resolution Imaging of Light Elements in Crystalline Materials Scott D. Findlay1 1. School of Physics and Astronomy, Monash University, Melbourne, Australia Light elements are key constituents of many advanced materials and energy management technologies � oxygen in dielectrics and superconductors, lithium in battery materials, hydrogen in hydrogen storage materials � making their reliable imaging an important goal for atomic-resolution structure analysis. Annular dark field (ADF) imaging, the paragon of directly interpretable atomic resolution imaging in scanning transmission electron microscopy (STEM), is dominated by strongly scattering heavy elements, making weakly scattering light elements difficult to detect. However, it was demonstrated six years ago that both light and heavy elements are directly and robustly visible in at atomic resolution STEM images recorded from an annular detector in the outer area of the bright field region [1,2]. This tutorial will present an overview of this technique, dubbed annular bright field (ABF) imaging, describing its implementation, conceptual underpinnings, limitations, and successful applications. Two distinct image-forming mechanisms contribute to the form of ABF images: elastic scattering that tends to forward-focus the electron probe when placed on columns of light elements (see left side of figure 1) and thermal scattering that tends to scatter the electron probe to high angles when placed on columns of heavy elements [2,3]. Both mechanisms reduce intensity in the outer area of the bright field region, and hence both light and heavy atomic columns appear as dark spots in ABF images, as in the example in the upper right of figure 1. Moreover, this proves to be true over a wide range of thicknesses and a moderate range of defocus values � lower right in figure 1 � meaning ABF images are not only directly interpretable but also robust, a highly desirable property for exploring unknown specimens. ABF has been successfully applied to visualizing both lithium [4,5] and hydrogen [6,7] within crystalline environments. However, for these lightest of elements the imaging remains challenging: direct interpretation may apply only over a restricted range of thicknesses, the signal can be of similar magnitude to the noise level, and the specimens tend to damage under the beam. Further limitations of the technique, such as contrast inversion with defocus in very thin specimens [8,9] and incorrect apparent inter-column spacing in materials with very small inter-column spacing [3], will be discussed. Intriguingly, ABF appears tolerant to modest distortions of atomic columns, enabling light element columns to be imaged in defect structures like grain boundaries [10]. Drawing on experience from both experimental and theoretical explorations, practical aspects such as implementation and optimum experimental geometry will be discussed, as will related techniques. For instance, using a detector complementary within the bright field region to an ABF detector can image light elements with improved spatial resolution [11] and taking the difference between this and the ABF signal can enhance the signal-to-noise [12]. A brief selection of applications to which ABF has been applied will be presented [13]. Microsc. Microanal. 21 (Suppl 3), 2015 1940 References: [1] E. Okunishi et al, Microscopy and Microanalysis 15(S2) (2009), p. 164. [2] S.D. Findlay et al, Appl. Phys. Lett. 95 (2009), p. 191913. [3] S.D. Findlay et al, Ultramicroscopy 110 (2010), p. 903. [4] Y. Oshima et al, J. Electron Microsc. 59 (2010), p. 457. [5] R. Huang et al, Angew. Chem. Int. Ed. 50 (2011) p. 3053. [6] S.D. Findlay et al, Appl. Phys. Expr. 3 (2010) p. 116603. [7] R. Ishikawa et al, Nature Materials 10 (2011) p. 278. [8] S. Lee et al, Ultramicroscopy 125 (2013) p. 43. [9] P.J. Phillips and R.F. Klie, Applied Physics Letters 103 (2013) p. 033119. [10] S.D. Findlay et al, Ultramicroscopy 111 (2011) p. 285. [11] M. Ohtsuka et al, Ultramicroscopy 120 (2012) p. 48. [12] S.D. Findlay et al, Ultramicroscopy 136 (2014) p. 31. [13] The author thanks all his collaborators in developing and applying ABF imaging, particularly at the Crystal Interface Laboratory, University of Tokyo, and JEOL Ltd. This research was supported under the Discovery Projects funding scheme of the Australian Research Council (Project No. DP110101570). Figure 1. Left: schematic of a STEM probe scattering through a crystal and the ADF and ABF detector geometry. Right upper: Experimental ADF and ABF images of LaAlO3 [001] for 200 keV electrons and a 23 mrad probe-forming aperture semiangle (courtesy of A/Prof. N. Shibata). The dark-cornered overlay shows the repeat-unit-averaged experimental image. The white-cornered overlay shows the corresponding simulation. Right lower: Simulated tableaux showing that ABF images are directly interpretable over a range thickness and defocus values almost as wide as that of ADF.");sQ1[969]=new Array("../7337/1941.pdf","Another Phase Plate in the Zoo: Reducing Charging and Optimizing the Design of Electrostatic Phase Plates","","1941 doi:10.1017/S143192761501048X Paper No. 0969 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Another Phase Plate in the Zoo: Reducing Charging and Optimizing the Design of Electrostatic Phase Plates Andreas Walter1, Siegfried Steltenkamp2, Daniel Rhinow1 and Werner K�hlbrandt1 1. Max Planck Institute of Biophysics, Department of Structural Biology, Max-von-Laue-Str. 3, D-60438 Frankfurt, Germany 2. caesar Research Center, Ludwig-Erhard-Allee 2, D-53175 Bonn, Germany Phase plates (PPs) enable in-focus transmission electron microscopy (TEM) of frozen-hydrated biological specimens. However, charging of PPs is a problem that has prevented their routine use in TEM of weak-phase objects [1,2]. In theory, electrostatic PPs are superior to thin-film PPs since they do not attenuate the scattered electron beam and allow freely adjustable phase shifts [3]. They consist of multiple layers of conductive and insulating materials, and are thus more prone to charging than thinfilm PPs. Although the origins of PP charging are not fully understood, empirical approaches to reduce charging of PPs have been at least partially successful [1,4,5]. However, charging of electrostatic PPs has not been addressed in detail. In this work we address possible origins of charging of Boersch phase plates (BPPs) and suggest a new design that is likely to reduce charging and would be superior to a BPP. First, we simulated the impact of electrostatic charges of BPPs on phase contrast transfer by performing finite-element calculations with Gmsh/GetDP and integrating the resulting threedimensional potential distribution with MATLAB. We numerically reproduced Fourier patterns that are observed experimentally with charged BPPs and cannot be fitted by a conventional CTF. We showed that surface charges trapped in the vicinity of the BPP central electrode cause circular power spectra, whereas charges along the supporting rods cause a 6-fold distortion. In the simulations, very large phase offsets (>15 ) needed to be applied to reproduce the experimentally observed power spectra (Fig. 1a). We also demonstrated that BPP charging occurs both deeply within the BPP and on its surface. We separated the relative contributions by two main experiments: 1) Monte Carlo simulations with CASINO indicated that the penetration depth of 200 keV electrons exceeds the thickness of a 5 �m thick BPP, causing inelastic scattering in both the upper and lower insulating Si3N4 layer. To address the effect of trapped charges in the insulator we used a FIB instrument to mill ring-shaped cavities around the central electrode on the upper and lower surfaces of the BPP. Subsequently, the cavities were filled with platinum by electron beam-induced deposition to mitigate charges trapped in the insulator in the vicinity of the central electrode. The comparison of Fourier transforms of images of an amorphous carbon film acquired with a BPP before and after fabrication of the charge drains showed that the charge drains reduced the rapid oscillation of Thon ring patterns that are characteristic of charged BPPs. This indicates that electrostatic charges trapped deeply in the BPP are a major factor in the charging of BPPs. 2) In analogy to environmental scanning electron microscopy where positively charged gas molecules in the specimen chamber neutralize negative surface charges trapped by insulators during SEM imaging, we developed a PP goniometer equipped with a gas injection system consisting of metallic nozzles pointing towards the PP chip. To analyse the influence of gas injection on the charging behaviour of BPPs in situ we injected argon in the vacuum of the back-focal plane. The argon flux through the goniometer was kept at flux rates < 0.05 ml/min. The pressure in the specimen chamber increased from ~10-7 mbar to ~10-5 mbar. Argon injection affected the Fourier pattern of charged BPPs considerably and immediately, indicating modifications of the BPP surface. Although argon injection did not reduce image artefacts caused by BPPs, the experiments showed that physical and chemical surface properties of PPs are critically important for PP charging, in accordance with other work [4,5]. Microsc. Microanal. 21 (Suppl 3), 2015 1942 1930 To minimize these charging artifacts and optimize the performance of electrostatic PPs, it is desirable to fabricate electrostatic PPs which expose as little material to the intense unscattered beam as possible. Additionally, an optimal electrostatic PP should impart a homogeneous phase shift to the unscattered beam and have a low cut-on frequency. We propose a new type of electrostatic PP that meets the above requirements and combines the advantages of a single coaxial PP and a BPP. It consists of three free-standing coaxial rods converging in the centre of an aperture (3-fold coaxial PP). To determine the optimum dimensions of the 3-fold coaxial PP and to quantify the homogeneity of the phase shift, we simulated the three-dimensional potential distribution and the induced phase shift with the finite-element method as mentioned above. The simulations shown in Fig. 1b indicate that a 3-fold coaxial PP with a rod width r below 1.2 �m is suitable for in-focus imaging of ice-embedded biomolecules. This device combines a rapidly decaying, 3-fold electrostatic field with a nearly constant phase shift of the unscattered beam. This ensures a homogenous and non-directional phase shift of the unscattered beam, which would result in close-to-optimal contrast transfer to near-atomic resolution. The production of a first prototype demonstrates that the free-standing rods are mechanically stable even without the electrode ring and that the fabrication of such a 3-fold PP is feasible. [1] A Walter et al, Ultramicroscopy 116 (2012) p. 62-72. [2] R Danev et al, Ultramicroscopy 109 (2009) p. 312-325. [3] K Schultheiss et al, Microsc. Microanal. 16 (2010) p. 785-794. [4] RM Glaser et al, Ultramicroscopy 135 (2013) p. 6-15 [5] M Marko et al, Journal of Structural Biology 184 (2013) p. 237-244 Figure 1. (a) Simulation of phase shifts, power spectra and EM images of ice-embedded ribosomes with a charged BPP nominally in focus. Experimentally observed artifacts in the power spectra were reproduced with positive electrostatic charges homogeneously distributed over the central electrode (1st column) with a charge density = 6*10-5 C/m2 or over the central electrode and along the rods (2nd column) with = 2*10-5 C/m2. (b1) Simulation of phase shifts calculated for 3-fold coaxial PPs with different rod widths, corresponding in-focus CTFs in red (conventionally defocused CTFs in blue), and images of ribosomes. The electrode voltages to produce a 90� phase shift at the position of the zero beam were 1450 mV for r = 2.5 �m and 700 mV for r = 1.2 �m. (b2) SEM images of the first prototype.");sQ1[970]=new Array("../7337/1943.pdf","Development of Phase Contrast Scanning Transmission Electron Microscopy.","","1943 doi:10.1017/S1431927615010491 Paper No. 0970 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Development of Phase Contrast Scanning Transmission Electron Microscopy. H. Iijima1, H. Minoda2, T. Tamai2, Y. Kondo1, T. Fukuda and F. Hosokawa1. 1. 2. EM Business Unit, JEOL Ltd., 3-1-2 Musashino, Akishima, Tokyo 196-8558, Japan. Department of Applied Physics, Tokyo University of Agriculture and Technology, 2-24-16 Naka-cho, Koganei, Tokyo 184-8588, Japan. Phase contrast transmission electron microscopy (P-TEM) is a powerful tool to enhance the image contrast of transparent materials such as ice-embedded biological specimens or polymer materials. In PTEM, a phase plate is placed at the back-focal plane (BFP) of the objective lens (OL). It gives a phase shift for scattered electron waves, resulting in a change of phase contrast transfer function (PCTF) from sine to cosine type. Eventually, phase variation of specimens is converted into intensity variation. Among various types of phase plates, a carbon film phase plate with a small central hole is the most practical [1]. However, there is a serious issue that high-density electron beam (cross-over) on the phase plate causes the charging and/or the alteration of the phase plate, resulting in decreasing the life time of the phase plate. To overcome this issue, we are developing phase contrast scanning transmission electron microscopy (PSTEM). Figure 1 shows the schematics of P-TEM and P-STEM. According to the reciprocity theorem, image contrast of P-STEM shows the same contrast in the P-TEM, if a phase plate is placed at a frontfocal plane (FFP) of an OL in P-STEM. In P-STEM, the life time of the phase plate is expected to improve, since a cross-over is not formed on the phase plate. In our experiments, A field emission electron microscope (JEM-2100F) equipped with a Shottky electron source was used to obtain a coherent small probe on a specimen. Phase plate is placed on a condenser lens aperture plane that conjugates to the FFP of the OL. Figure 2 compares conventional STEM (C-STEM) and P-STEM images of amorphous carbon film. A 30-nm-thick amorphous carbon film with a 5m-diameter central hole was used as phase plate. The power spectrum of Fourier transform for the P-STEM image shows that the PCTF changed from sine type to cosine type. It indicates that contrast converted image can be achieved in P-STEM as same as PTEM. Figure 3 shows the long-life time of the phase plate in P-STEM. Figure 3 compares the Fourier transforms of P-STEM image with fresh phase plate (Fig. 3 (a)) and the plate after 100 h use (Fig. 3 (b)) using a sample of amorphous carbon film. It clearly shows that the phase plate remained stable after 100 hours. This result proves the life time of the phase plate is greatly improved by using P-STEM. This development was supported by the program for "Development of Systems and Technologies for Advanced Measurement and Analysis" under JST. References: [1] R. Danev and K. Nagayama, J. Phys. Sci. Jpn. 70 (2001) p. 696. Microsc. Microanal. 21 (Suppl 3), 2015 1944 Fig. 1 Schematics of P-TEM (left) and the P-STEM (right). The phase plate is placed at the BFP of the objective lens in P-TEM and the FFP of objective lens in P-STEM. Fig. 2 C-STEM and P-STEM images of amorphous carbon film. Both images are taken at close to infocus. (a) C-STEM image. (b) P-STEM image. (c) and (d) Fourier transform of (a) and (b). (e) Intensity profiles of (c) and (d). Fig. 3 Comparison of power spectrum of P-STEM images taken with fresh phase plate (a) and 100 hours used phase plate (b). (c) Intensity profiles of (a) and (b).");sQ1[971]=new Array("../7337/1945.pdf","Electron Differential Phase Microscopy with an A-B Effect Phase Plate.","","1945 doi:10.1017/S1431927615010508 Paper No. 0971 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electron Differential Phase Microscopy with an A-B Effect Phase Plate. T. Tanji1), H. Niimi2), J. Usukura3), Y. Yamamoto1) and S. Ohta4) 1. 2. EcoTopia Science Institute, Nagoya University, Nagoya, Japan Graduate School of Engineering, Nagoya University, Nagoya, Japan 3. Graduate School of Science, Nagoya University, Nagoya, Japan 4. JEOL Ltd., 3-1-2 Musashino, Akishima, Tokyo, Japan Observations of week phase objects, such as thin films of light elements, thin polymer films, biological sections etc., are available by electron phase microscopy[1]. Many of phase plates utilized are thin film types. Some electrostatic types have been developed[2, 3], but they are not so general, because the fabrication of the filter with fine structures is very difficult. The mainstream of todays phase plate is the thin film type. This type of the phase plate, however, has some disadvantages, i.e. control of the film thickness, charging up, contamination and so on. We adopted the phase plate with a magnetic thin filament which generates the vector potential around itself by an Aharonov-Bohm (A-B) effect. The filament type phase plate with the A-B effect was proposed and constructed firstly by Nagayama. This type of the phase plate generates the differential phase contrast in the image, and has a longer life time than the thin film type. Any clear differential effect, however, has scarcely reported so far. We will report that the effect of a phase plate consisting of a Wollaston platinum filament less than 1 �m in diameter covered with ferromagnetic material, Sm-Co of 5 nm thick, deposited by a Pules Laser Deposition after demagnetization at 800�C. The filament with a clean surface selected by SEM is mounted on a single hole Cu grid, and re-magnetization in the field of 2.4T. The phase difference in the both side spaces of the filament measured by electron holography shows 1.5 rad as shown in Fig.1. Being set on the aperture holder, the phase plate is inserted in the back focal plane of the objective. Figure 2 shows images of a mouse photoreceptor cell, which is fixed with Os and without staining. The differential image in Fg.2(b) observed using the phase plate shows inner structures including mitochondria clearer than in the image taken ordinary TEM in Fig.2(a). The direction of the differentiation is shown by the arrowhead. The effects of the source size and the thickness of the filament on the differentiation will be estimated theoretically, and experiments by a 1000kV high voltage electron microscope will be reported, too. Refernce [1] K. Nagayama: J. Electronmicrosc., 60 (2011) S43. [2] H. Boersch: Z. Naturforsh. 2a (11947) 615. [3] T. Matsumoto and A. Tonomura: Ultramicroscopy, 63 (1996) 5. [4] The authors acknowledge Prof. T. Kato of Nagoya University and Dr. K. Ozaki of Advanced Industrial Science and Technology for their kind helps for the magnetization of filaments, and financially supported by Grant-in-Aid for Scientific Research 25246001. Microsc. Microanal. 21 (Suppl 3), 2015 1946 Fig.1 (a) Electron phase map reconstructed by electron holography. (b) A line profile along the arrow head in (a) which is averaged along the long side of the rectangle. The phase difference is about 1.5 rad. between both sides of the filament. a b 0.5 �m Fig.2 Images of a mouse photoreceptor cell without the phase plate (a), and with the phase plate (b).");sQ1[972]=new Array("../7337/1947.pdf","Lorentz Transmission Electron Microscopy for Imaging Magnetic Fields from a Perpendicular Ferromagnetic Stripe Domain Thin Film","","1947 doi:10.1017/S143192761501051X Paper No. 0972 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Lorentz Transmission Electron Microscopy for Imaging Magnetic Fields from a Perpendicular Ferromagnetic Stripe Domain Thin Film Taeho Roy Kim1, Olav Hellwig2 and Robert Sinclair1 1. 2. Department of Materials Science and Engineering, Stanford University, Stanford, CA 94305-4034. San Jose Research Center, HGST, a Western Digital Company, 3403 Yerba Buena Road, San Jose, CA 95135. Perpendicular magnetic thin films, widely used high density storage media, need further improvement in order to meet the increasing demand for data storage. To study them, it is important to understand both the microstructure and magnetic properties. One useful technique to study such films is magnetic force microscopy (MFM) [1]. However, MFM can only detect magnetic information of a material with no structural information such as grain structure. Here we apply Lorentz Transmission Electron Microscopy (LTEM) to this problem. This has been used successfully to study magnetic films with in-plane magnetic domains [2], but here we show the capability of LTEM for perpendicular ferromagnetic stripe domain thin films. We chose the Fresnel mode among various techniques in LTEM [3] because a simple experimental procedure of Lorentz lens defocus provides qualitative information on magnetization. The Lorentz force from a magnetic material deflects the electron beam and this deflection gives rise to contrast in the image. The results suggest that plan view images are comparable to MFM images shown in a previous study by Hellwig et al [4]. Furthermore, cross-sectional images allow observation of the magnetic fields emerging from the thin film that cannot be observed by MFM. We studied two Co/Pd multilayer films grown on a Si substrate with perpendicular ferromagnetic stripe domains. The number of Co/Pd layers of the cross-sectional sample was about 2.5 times more than that of the plan view sample, but our purpose of imaging magnetic fields is not affected. The detail of the Co/Pd multilayer film structure, fabrication method and stripe domain formation process are described elsewhere [4]. Both plan view and cross-sectional TEM specimens were prepared using a conventional method of grinding, polishing and ion milling. Magnetic imaging was then carried out on an FEI Titan 80-300 TEM at the Stanford Nanocharacterization Laboratory. The TEM was operated at 300 kV in Lorentz mode with the objective lens off to provide an environment free of magnetic fields. Images were acquired at large defoci up to about 2 mm and then were digitally processed with histogram equalization to show better the magnetic contrast. Plan view images are shown in Figure 1. The in-focus image shows a low magnification view of the grain structure and the defocused image shows stripe patterns due to magnetic domain contrast. The magnetic field components perpendicular to the electron beam at domain boundaries cause bright or dark contrast. In Figure 2, cross-sectional images are shown. The in-focus image shows that portions of the film are unintentionally etched away during TEM specimen preparation. The defocussed image shows magnetic domain contrast outside the film, in free space, which is not present where the magnetic film has been removed. There are bright lines due to the magnetic domain structure. Interestingly, these features are visible in the vacuum extending up to eight times the film thickness and their contrast weakens as the film becomes thinner. Microsc. Microanal. 21 (Suppl 3), 2015 1948 The study demonstrates the effectiveness of the LTEM for studying perpendicular ferromagnetic stripe domain thin films. Both plan view and cross-sectional images reveal the presence of magnetic domains. However the correlation with the grain structure has still not been achieved. [1] M. Alberecht et al, Appl. Phys. Lett. 81 (2002), p. 2875-2877. [2] K Tang et al, IEEE Trans. Magn. 32 (1996), p. 4130-4132. [3] C A Ross et al in "Wiley Encyclopedia of Electrical and Electronics Engineering: Magnetic Media, Imaging", J. Webster, (John Wiley & Sons, Inc., Hoboken) (1998), p.1-6. [4] O Hellwig et al, J. Magn. Magn. Mater. 319 (2007), p. 13-55. [5] The authors thank Western Digital Inc. for financial support for this study. Figure 1. Plan view images of a Co/Pd multilayer film using LTEM. (A) Low magnification in-focus image showing the 20 nm level grain structure. (B) Under-focused (about 750 m) image of the same region with stripe patterns representing magnetic domains. Figure 2. Cross-sectional images. (A) Focused image showing the film on a Si substrate. Arrows indicate the full stack region of the film with a thin layer of epoxy from the TEM specimen preparation on top. The dashed circle indicates the region where most of the film is etched away during the TEM specimen preparation. (B) Large under-focused (about 2 mm) image showing bright regions in a dark background due to magnetic fields from the film. The longest bright region implying strongest magnetic strength is located where the magnetic thin film is still remaining (the arrows from (A) are indicating the same location). In the dashed circle, no magnetic contrast is observed as expected. Small arrows indicate the possible location of perpendicular magnetic domains and their expected field directions in the film.");sQ1[973]=new Array("../7337/1949.pdf","Fabrication of Self-Supporting Annular Apertures for Use in the Transmission Electron Microscope","","1949 doi:10.1017/S1431927615010521 Paper No. 0973 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Fabrication of Self-Supporting Annular Apertures for Use in the Transmission Electron Microscope Liya Yu1, Aaron C. Johnston-Peck2, Andrew A. Herzing2 and Vincent Luciani1 1. Center for Nanoscale Science and Technology, National Institute of Standards Technology, Gaithersburg, MD 20899 USA 2. Material Measurement Laboratory, National Institute of Standards and Technology, Gaithersburg, MD, 20899, USA Apertures provide a way to further tailor the action of the transmission electron microscope (TEM). Recently there have been many reports looking at custom designed apertures, this was spurred by the discovery that an electron beam with quantized orbital angular momentum could be generated [1]. While much of the recent aperture design and fabrication work has been geared towards this endeavor many other applications for custom apertures exist. For instance, annular apertures can act as filters in the objective lens back focal plane to generate dark field images [2]. A focus ion beam (FIB) is typically used for fabrication of annular apertures due to the simplicity of processing. However, this approach limits the complexity of designs and requires three support arms in addition to the central beam stop, as shown in Figure 1(a). These unwanted extensions of the diaphragm eliminate electron transmission and can create unwanted diffraction effects, a particularly important consideration if the aperture is to function as the probe-forming aperture in the scanning TEM. Therefore, it would be beneficial to eliminate the need for support arms from the aperture design. A self-supporting annular aperture has been proposed, as illustrated in Figure 1(b), and fabricated in different dimensions to validate simulation results (Figure 2). Rather than thick periodically spaced metal arm, the central beam stop is supported by a thin layer of amorphous SiN. The SiN provides the necessary mechanical support but interferes little with the transmitted electrons. Apart from the FIB approach, i.e., directly ion-milling the design into bulk metal, fabrication techniques which allow layer by layer building of structures, are introduced to produce this free standing center block without the inclusion of electron absorbing supports. LPCVD nitride was deposited, patterned and etched to form the membrane window acting as the supporting structure. Photolithography provides an effective and flexible means of patterning, which even practically enables the fabrication of diaphragms with multiple aperture designs. Gold was selected to be the electron absorbing material for its high atomic number and low stacking strain behavior. Instead of a plasma or ion-milling etch, wet etch chemistry was chosen as a low energy noble metal etch method in order to preserve the integrity of the thin SiN window. The annular apertures fabricated by FIB and lithography were installed in a transmission electron microscope and their performance was compared through a series of tests. Additionally, their performance was modeled. Figure 2(d) and (e) show the calculation of point spread functions for two different designs. Degree of interaction between SiN and electrons which has implications for performance will also be evaluated. References: [1] B.J. McMorran et al, Science 331 (2011), p. 192. [2] S. Bals, R. Kilass and C. Kisielowski, Ultramicroscopy 104 (2005), p. 281. Microsc. Microanal. 21 (Suppl 3), 2015 1950 [3] A.C.J.-P. acknowledges support of the National Research Council Postdoctoral Research Associateship Program. Figure 1 (a) SEM image of an annular aperture fabricated using FIB. (b) Newly proposed structure of self-support annular aperture. Figure 2 (a)-(b) New style annular aperture fabricated with different inner diameters. (c)-(d) Simulated point spread function as a function of inner convergence angle (ICA).");sQ1[974]=new Array("../7337/1951.pdf","Low Dose Electron Holography: First Steps.","","1951 doi:10.1017/S1431927615010533 Paper No. 0974 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Low Dose Electron Holography: First Steps. Edgar Voelkl1, Rodney Herring2, Benjamin Bammes3 and David Hoyle4 1. 2. Hitachi High Technologies America, Inc., Clarksburg, MD, USA. University of Victoria, Victoria, Canada. 3. Direct Electron LP, San Diego, CA, USA. 4. Hitachi High Technologies Canada, Inc., Toronto, Canada. One of the key features of electron holography is its capability for providing access to the phase information of the image wave and ultimately to the object wave. This fact can be indicated easily by the definition of the image intensity that is recorded with standard recording devices like CCD cameras: I(x,y) = |a(x,y) exp(i(x,y))|2 = a2(x,y), with I(x,y) the intensity to be recorded, a(x,y) the amplitude of the image wave and (x,y) the phase of the image wave. Obviously, the phase information (x,y) is lost but is recoverable with holography [1]. Acquiring electron holograms always has been a technological challenge. On film, both the nonlinearity and the MTF (modulation transfer function) impacted the quality of the hologram as higher order sidebands are generated and higher frequencies are dampened. Although digital cameras are very linear up to at least 70% to their saturation, the MTF of almost every camera on the market drops to around 10% or below at the Nyquist limit and for sampling rates (pixels per interference fringe) s < 10, most MTFs are already below 50% at the location of the sideband (the main carrier of the holographic information). As a way out, or more as a compromise, electron holograms are recorded highly oversampled and the images obtained from the holograms are often downsized as they contain a lot of empty information due to the oversampling. Alternatively, holograms can be recorded with a binning factor of 2, 4 or even 8. In that case, the large binning factor increases the MTF at the Nyquist limit and oversampling (and empty information) is avoided. Both solutions are far from ideal, as, to draw a simple picture: a 4k x 4k camera is effectively reduced to a 1k by 1k camera or less as to be useful for electron holography. In addition, the DQE (detection quantum efficiency) of CCD cameras in general is << 1, simply because vendors create a compromise between a reasonable MTF and a reasonable DQE. It might be reasonable to say there are no dedicated cameras for electron holography. To compare existing CCD cameras for their use for holography, we followed the approach taken in [2] and simulated electron holograms on an electron-by-electron basis. With the availability of cameras that are exposed directly to the electron beam, like the DE-12 from Direct Electron [3], the holography community is closer than ever to the ideal camera for electron holography. For testing the usability of the DE-12 for electron holography, it was installed temporarily on the Hitachi HF3300V (or `STEHM') at the University of Victoria [4]. For the data presented in Fig.1, the brightness was adjusted such that an average of 130 electrons per frame at 150fps (frames per second) was obtained for images of the size of 512x512 pixels. Frames were pre-processed using an electron counting algorithm to record the precise location of each incident primary electron in each frame before summing to form the final hologram. The resulting holograms were then evaluated for average counts, fringe contrast and sampling rate. These parameters were then used to simulate the data via HoloWorks [5] by building the hologram from single electron events for an ideal camera. Then both the experimental and simulated holograms were reconstructed and the phase images evaluated after phase-tilt correction. For 1.19e- per pixel, a sampling rate s = 22.9 and fringe contrast � = 64.4%, the Microsc. Microanal. 21 (Suppl 3), 2015 1952 standard deviation was computed for the central area of each phase image and found to be 2/71 for the experimental data and 2/80 for the simulated data. Experimental and simulated data agree within 15%. Thus we conclude that the DE-12 is very close to an ideal camera for low-dose holography. The lowdose hologram in Fig.2a, recorded at an average of 4.19e- per pixel, �=44% and s=3.5 shows the potential of low-dose electron holography: the reconstructed phase image Fig.2b allows the measurement of the film thickness Fig.2c, basically invisible in Fig.2a at an extremely low dose. References: [1] "Introduction to Electron Holography," eds. E Voelkl, LF Allard and DC Joy (Kluwer Academic / Plenum Publishers 1999). [2] E. Voelkl, Ultramicroscopy, Volume 110, Issue 3, February 2010, Pages 199�210 [3] DE-12, Camera System from Direct Electron, LP, http://www.directelectron.com [4] University of Victoria, Canada. [5] HoloWorks, by HoloWerk LLC, http://www HoloWerk.com a b Fig. 1 a: Experimental data composed of 2386 frames recorded at 150 fps and 512 by 512 pixels. b: Simulated data based on fringe contrast, sampling rate and average pixel value. The standard deviation values within reconstructed phase images agree < 15%. a b nm 60 40 20 0 -20 0 c 100 200 300 pixels 400 500 Figure 2. a: Hologram from 3692 frames resulting in 4.3 e-/pixel; s = 3.5, � = 44%; b: rec. phase image showing amorphous C-film. c: line plot indicating the thickness of the C-film.");sQ1[975]=new Array("../7337/1953.pdf","A Practically Simple and Easy Approach for Minimizing the Influence of Fresnel Fringes on Phase Sensitivity Measured from Electron Holography","","1953 doi:10.1017/S1431927615010545 Paper No. 0975 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A Practically Simple and Easy Approach for Minimizing the Influence of Fresnel Fringes on Phase Sensitivity Measured from Electron Holography Zhouguang Wang Micron Technology Inc., Boise, Idaho, USA Electron holography has been applied for practical semiconductor device analysis for years. To precisely map p-n junction distributions, especially those with low doping levels in the devices, a high phase sensitivity measured from reconstructed phase image is required. Besides the fringe contrast and number of electrons in recorded hologram, an important factor that affects the phase sensitivity is Fresnel fringes. Fig. 1(a) is a typical hologram taken with no sample and the profile on the right shows the intensity modification by Fresnel fringes. This can be seen more clearly in the reconstructed amplitude image (Fig. 1(b)). The influence of Fresnel fringes on the phase is shown in Fig. 1(c). The phase profile indicates that a phase fluctuation up to 0.26 rad is induced by Fresnel fringes. Generally, two kinds of approaches can be utilized to reduce the effects of the fluctuation on the phase sensitivity: one is experimentally obtaining a hologram without Fresnel fringes [1, 2] and the other is numerically correcting [3]. These methods, however, need either special equipments or lengthy and complicated calculation, hard to be adapted for the use in semiconductor industry where a quick turn-around time is required. Practically we have employed a much simple and easy method to minimize the influence of Fresnel fringes. As shown in Fig. 1, the modification of Fresnel fringes on the intensity of interference fringes attenuates from the edge to the center of the interference band. If only the center part is used, the influence of the Fresnel fringes will be minimized. Figs. 2(a), (b), (c) and (d) are a series of reference holograms taken by a GIF CCD camera at biprism voltages of 15 V, 25 V, 35 V and 45 V respectively. The corresponding reconstructed amplitude images are given in Figs. 2(e), (f), (g) and (h) to show the influence of the Fresnel fringes. It can be seen clearly that with increasing biprism voltage the interference band expands, more and more Fresnel fringes move out of the recorded hologram. When only center part is recorded as shown in Figs. 2(c) and (d), the influence of the Fresnel fringes almost disappears completely. A one-dimensional Si/Si p-n junction with very low doping level was used for the evaluation of the method. Two holograms, A and B including their reference holograms were collected using a 2K�2K GIF camera. Hologram A was imaged using the edge part of the interference band, in which some Fresnel fringes were included. Hologram B, on the other hand, was formed by using only the center part of the interference band. Both holograms are reconstructed by using same parameters, and the resulting phase images and their corresponding profiles drawn along depth direction are shown respectively in Figs. 3(a) and (b). Although a reference hologram has been used for the reconstruction, the phase fluctuation induced by Fresnel fringes can be seen obviously in Fig. 3(a) so that no p-n junction can be recognized. In Fig. 3(b), however, there is no influence of the Fresnel fringes, as the result, we can clearly see a phase shift between p- and n-doped regions even the shit is very small (< 0.1 rad). The phase measurements in Si area yield the standard deviations of 0.162 rad and Microsc. Microanal. 21 (Suppl 3), 2015 1954 0.065 rad, respectively for Figs. 3(a) and (b), which explains why the p-n junction can be mapped successfully in the latter case. References: [1] K. Yamamoto, T. Hirayama and T. Tanji, Ultramicroscopy 10 (2004), p265. [2] K. Harada, A. Tonomura, Y. Togawa, T. Akashi and T. Matsda, Appl. Phys. Lett. 84(2004), p3229. [3] K. Yamamoto, I. Kawajiri, T. Tanji, M. Hibino and T. Hirayama, J. Electr. Microsc. 49(2000), p31. Figure 1. A typical hologram with Fresnel fringes (a) and its reconstructed amplitude (b) and phase (c) images. Their profiles drawn from red box clearly show the influence of the Fresnel fringes on both amplitude and phase. Figure 2. A series of holograms taken at different biprism voltages and their corresponding reconstructed amplitude images showing the reduction of the influence of Fresnel fringes. Figure 3. Electron holography for a 1D Si/Si p-n junction with very low doping level. (a) the phase image reconstructed from a common hologram with Fresnel fringes, the profile from it showing the influence of the Fresnel fringes on the phase; (b) the phase image from a hologram using only center part of the interference band, no Fresnel fringe effect so its profile revealing p-n junction location.");sQ1[976]=new Array("../7337/1955.pdf","Optimising Electron Holography in the Presence of Partial Coherence and Instrument Instabilities with Conventional and Direct Detection Cameras","","1955 doi:10.1017/S1431927615010557 Paper No. 0976 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Optimising Electron Holography in the Presence of Partial Coherence and Instrument Instabilities with Conventional and Direct Detection Cameras Shery L. Y. Chang1, Christian Dwyer1, Juri Barthel1, Chris B. Boothroyd1 and Rafal E. DuninBorkowski1 1. Ernst Ruska-Centre for Microscopy and Spectroscopy with Electrons, and Peter Gruenberg Institute, Forschungszentrum Juelich, Juelich, Germany Off-axis electron holography provides a direct means of retrieving the phase of the wavefield in a transmission electron microscope, enabling the measurement of electric and magnetic fields at length scales from microns to nanometers. To maximise the precision of the technique, it is important to acquire holograms using experimental conditions that optimise the phase resolution for a given spatial resolution. These conditions are determined by a number of competing parameters, especially the spatial coherence and the instrument stability. Here, we describe a simple, yet accurate, model for predicting the dose rate and exposure time that give the best phase resolution in a single hologram. Experimental studies were undertaken to verify the models of spatial coherence and instrumental instabilities that are required for the optimisation. The model is applicable to holography in both standard mode and Lorentz mode, and it is relatively simple to apply. In light of the considerable body of literature describing the factors governing the phase resolution in electron holography [1-3], we emphasize that we kept the models of spatial coherence and instrument instabilities as simple as possible. In addition, we have evaluated the performance of a state-of-the-art direct detection camera (DDC) for electron holography, and compare its performance in various operational modes with a conventional integrating charge coupled device (CCD) camera. We found that the commonly-used Gaussian model [2] is not suitable to describe the spatial coherence, and instead a bivariate Cauchy distribution convoluted with a Gaussian distribution is better suited. The fringe movement due to instabilities is well-modelled by the Langevin theory of Brownian motion, which improves upon previous models [4-5] since it is applicable to the practical range of exposure times used in experiments. These models can be determined with two data sets with a fixed biprism voltage and magnification: an intensity series (Fig. 1) to determine the spatial coherence and a time series (Fig. 2) to determine the time-dependent visibility. The optimum dose rate and exposure time (not considering specimen instabilities) for other combinations of biprism voltages and magnifications can then be predicted. The comparison of the phase errors at the optimum conditions (at 300kV) between a conventional CCD camera (UltraScan, Gatan) and a DDC camera (K2, Gatan) for a given exposure time (1sec) shows a two-fold reduction in the phase error. This improvement is largely due to the improvement of DQE and MTF offered by the DDC. This suggests that the use of DDC can have significant advantages in the application of electron holography. Microsc. Microanal. 21 (Suppl 3), 2015 1956 References: [1] H Lichte et al, Optik 77 (1987) p. 135. [2] H Lichte, Ultramicroscopy 108 (2008) p. 256. [3] F. Roeder et al, Ultramicroscopy 144 (2014) p. 32. [4] E. Voelkl, Ultramicroscopy 110 (2010) p. 199 [5] R. A. McLeod et al, Ultramicroscopy 141 (2014) p. 38. Figure 1. The normalized phase error as a function of normalized signal, plotted for selected exposure times (indicated in each graph). Also shown are predictions of phase errors (dotted lines) based on the fitting of the theoretical models. Figure 2. (a) Fringe displacement as a function of time, measured using 1000 holograms over a timespan of 67 minutes. (b) The average (black squares) and rms (red triangles) displacement for time intervals up to 400 seconds. The theoretical models are overlaid (dotted lines).");sQ1[977]=new Array("../7337/1957.pdf","Coherent Electron Interference of Diffracted Beams from Amorphous Materials","","1957 doi:10.1017/S1431927615010569 Paper No. 0977 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Coherent Electron Interference of Diffracted Beams from Amorphous Materials Rodney A. Herring CAMTEC, MENG, University of Victoria, British Columbia V8W 2Y2 Canada Two interference methods producing high contrast fringes from the diffraction intensities of amorphous materials have been made possible [1, 2]. Recent methods of k-space interference by means of an electron biprism have revealed that all electron intensities on the diffraction plane have sufficient coherence to form fringes. The intensities from amorphous materials are of interest because they represent their atomic structure, which to date has not been solved because of their missing phase information [3]. Relative phase information of the amorphous structure is possible using a wavefront-splitting method [1], whereas, absolute phase information of the amorphous structure, the truly desired information necessary for determining its structure factor, is obtainable using an amplitude spitting method (Fig. 1) [2]. In this method an amorphous layer exists on the surface of a crystal. The crystal's Bragg diffracted beams are used to carry the amplitudes and phases of the amorphous thin film. Interferograms were produced from the diffuse, speckle and ring intensities of many amorphous materials including a-C, a-Si, a-Ge, SiO2 , a-W, a-GaAs and an amorphous metal. The fringes in the interferograms were produced from electrons scattering from low to high angles further than the third amorphous ring in some cases. The absolute phase can then be related to the amorphous intensity's radial distribution for determination of their structure factors. The spatial frequency of the fringes carrying the phase information of the amorphous specimen, which in many cases determines the phase resolution limit, depends on the angle of overlay of the interfering Bragg diffracted beams with the higher spatial frequency Bragg beams transferring more specimen information in the interferogram. If the two crystal beams have equal but opposite phases, i.e., + g hkl, they cancel when interfered, leaving only two times the absolute amorphous phase, = ghkl + amor + gh kl + amor = 2amor . A fly in the ointment for determining the absolute phase of the amorphous specimen is that for convergent beams, there are still geometric unknowns such as the cross-over or focus probe position within the specimen giving rise to uncertainties in the part of the specimen that the left and right sides of the beam pass (Fig. 2). This shortfall is corrected by using a planar beam. First the planar beam passes through the amorphous layer and then it passes through the thin crystal producing the two carrier beams that are interfered as schematically shown in Fig 3. This method of beam interference offers the promise of being able to determine the phase of short- and longrange structures of amorphous materials. 1. Herring R A, Saitoh K, N, Tanaka N, and Tanji T (2010) Coherent Electron Interference from Amorphous TEM Specimens. J. Electron Microscopy 59: 321. 2. Herring R A, Saitoh K, N, Tanji T, and Tanaka N (2012) Electron interference from an amorphous thin film on a crystal transmission electron microscopy specimen. J. Electron Microscopy 61: 17. 3. Howie A (2004) Progress in Determining the Structure of Amorphous and Disordered Materials. Microsc Microanal 10 (Suppl 2) 784. Grants from JSPS, UVic, NSERC, CFI and BCKDF are gratefully appreciated. Microsc. Microanal. 21 (Suppl 3), 2015 1958 Fig. 1 � DBI of amorphous intensities showing in a) the amorphous intensities being carried by the Bragg diffracted beams, b) the interference of the amorphous intensities using a biprism, c) interference of an amorphous layer on a GaAs crystal. Fringes in box 1 shown in insert. Fig. 2 � Amorphous structure information difference between convergent and planar beam configurations. Fig. 3 � Interference of two planar, symmetrically Bragg diffracted beams from a crystal carrying amorphous amplitude and phase information. Interference is by means of an electron biprism.");sQ1[978]=new Array("../7337/1959.pdf","Three dimensional magnetic field reconstruction of artificial Skyrmion heterostructures","","1959 doi:10.1017/S1431927615010570 Paper No. 0978 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Three dimensional magnetic field reconstruction of artificial Skyrmion heterostructures Sheng Zhang1 , Amanda Petford-Long1 , Charudatta Phatak1 1 Materials Science Division, Argonne National Laboratory, Lemont, IL 60439, USA The Skyrmion spin structure is a topologically stable state in which the spins point in all directions wrapping around a sphere [1]. The Skyrmion configuration is not only important from a fundamental aspect in condensed matter physics as it can produce unconventional spin-electronic phenomena [2] but also from a technological point of view as a spintronic material [3]. In bulk crystals and thin films, the Skyrmion spin texture arises from the helical spin structure and is stabilized by the DzyaloshinskiiMoriya (DM) interaction. Lorentz microscopy has been previously used to study the real-space configuration of Skyrmions in thin films of Fe0.5 Co0.5 Si, and FeGe. However, since the magnetization of a Skyrmion state rotates in a continuous manner from one out-of-plane direction to the opposite one, a careful analysis of this state requires a complete three-dimensional analysis of magnetization. In this work, we will present the results from vector field electron tomography combined with Lorentz transmission electron microscopy (LTEM) to study such three-dimensional magnetization. The phase change e +m of an electron wave traveling through a thin foil can be expressed in terms of tomographic quantities; the electrostatic phase shift e corresponds to the scalar x-ray transform of the electrostatic potential, whereas the magnetic phase shift m is described by the vector x-ray transform of the magnetic vector potential [4]. The magnetic phase shift from the sample can be reconstructed using either off-axis or in-line holography. In this work, the tilt series was performed along two orthogonal directions. The recovered phase shift was then used to reconstruct the three-dimensional components of magnetization using standard iterative methods such as SIRT. We will discuss the reconstruction results and improvements achieved by using the iterative reconstruction methods. In our previous work on magnetic heterostructures, we have successfully shown that by controlling the macroscopic energy terms of magnetic layers, it is possible to stabilize novel topological states at room temperatures such as a "meron" state [5]. We will report on "artificial" Skyrmion lattices consisting of Co and Co/Pt multilayers that can be stabilized at room temperature. As a result, instead of having chiral DM interactions, the spin texture is controlled by the dipole-dipole interactions (magnetostatic interactions) and interlayer exchange interactions, which can lead to spin canting to create artificial Skyrmion spin structure. We have fabricated heterostructures consisting of [Co0.3 /Pt1 ]8 multilayers that have perpendicular anisotropy and thin layer of Co (10 nm) on top that has in-plane anisotropy as shown in Fig. 1(a). They were then patterned into disc structures using focused ion beam milling to introduce shape anisotropy to form vortex structures in the Co layer. Simulations were performed to understand the image contrast arising from the various layers in the heterostructures. Fig. 1(b)-(d) show the magnetization maps used to simulate the LTEM images, and Fig. 2(a)-(c) show the simulated underfocus images of the disc at -40 , 0 , and +40 tilt respectively. Fig. 2(d)-(f) show the experimentally obtained underfocus images at various tilt angles as indicated. In both simulated and experimental set of images, the domain contrast arising from the the out-of-plane magnetization can only be seen in the tilted images. It can also be seen that the additional contrast due to out-of-plane component changes contrast from black to white and vice-versa when tilted from -40 to +40 as Microsc. Microanal. 21 (Suppl 3), 2015 1960 indicated by the arrows in Fig. 2. We will report on the 3D magnetic field obtained from these discs and discuss the implication towards creationg of Skyrmion states. References [1] T. H. R. Skyrme, Proc. R. Soc. A, 262, 237 (1961). [2] Y. Onose, N. Takeshita, C. Terakura, H. Takagi, and Y. Tokura, Phys. Rev. B, 72, 224431 (2005). [3] X. Z. Yu et. al., Nat. Commun., 3, 988 (2012). [4] C. Phatak, M. Beleggia, and M. De Graef, Ultramicroscopy, 108, 503 (2008). [5] C. Phatak, A. Petford-Long, O. Heinonen, Phys. Rev. Lett., 108, 067205 (2012). [6] This work was supported by U.S. Department of Energy (DOE), Office of Science, Materials Sciences and Engineering Division. (a) Ti (2 nm) Co (10 nm) Co/Pt (10.4 nm) Pt (10 nm) Ti (2 nm) (b) (c) (d) 500 nm mx (Co) my (Co) mz (Co/Pt) Figure 1: (a) Schematic showing the various layers in the magnetic heterostructure. (b)-(d) shows the grayscale representation of the magnetization maps for the Co and Co/Pt layer. (a) (b) (c) 500 nm (d) (e) (f) 500 nm Figure 2: (a)-(c) Simulated underfocus images showing the contrast from the vortex in the Co layer and additional domain contrast from the Co/Pt layers under tilted conditions, (d)-(f) Experimentally obtained underfocus images showing the domain contrast under varying tilting conditions.");sQ1[979]=new Array("../7337/1961.pdf","Spin-Multislice Applied to the Electron Spin Interaction with Materials","","1961 doi:10.1017/S1431927615010582 Paper No. 0979 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Spin-Multislice Applied to the Electron Spin Interaction with Materials Vincenzo Grillo1,2,3, Spencer Alexander3 , Jan Rusz4 , Alexander Edstr�m4, Axel Lubk5, B J McMorran3, Ebrahim Karimi6 1. 2. CNR-Istituto Nanoscienze, Centro S3, Via G. Campi 213/a, I-41125 Modena, Italy CNR-IMEM Parco Area delle Scienze 37/A, I-43124 Parma, Italy 3. Department of Physics, University of Oregon, Eugene, Oregon, USA 4. Department of Physics and Astronomy, Uppsala University, Uppsala, Sweden 5. Triebenberg Laboratory, TU Dresden, Dresden, Germany. 6. Department of Physics, University of Ottawa, 25 Templeton, Ottawa, Ontario, K1N 6N5 Canada The multislice method is one of the most efficient and flexible methods to solve the propagation problem of a wave in the paraxial approximation. It has lent itself to a number of extensions which incorporate back-scattering effects and thermal motion due to phonons. However, the inclusion of relativistic effects, and in particular of spin, is a recent achievement [1][2][3]. In extending the multislice method, both Pauli equation- and Dirac equation-based approaches have been pursued. The Pauli equation in particular is probably sufficient to deal with most of the measurable effects in electron microscopy, however particular care has to be taken in Lorentz boosting the field in the appropriate reference frame [2]. To date, the Pauli equation-based approach has been applied only to the interaction of electrons with the external field of a lens system. In particular, ref [2] proposed an extension of the Fourier Transform (FT) -based method that allows for the inclusion of the components of the magnetic potential field orthogonal to the propagation direction. While these terms are naturally included in real-space multislice simulations [3], their inclusion in FT multislice algorithm has required a strong change. The extension of the formalism to include the interaction of electrons with the field inside of a material is more complicated. The main reason for this is that magnetic fields are typically extended over distances much larger than the lattice periodicity. We will therefore discuss a few possible approximations that can be used to describe this interaction, and especially the validity of a short-range cutoff that permits the incorporation of magnetic field effects on the atomic scales, where the field tends to be most intense. Fig1 shows a few columns of Fe [001] which are magnetized in the horizontal direction, orthogonal to the propagation direction. This situation could only be experimentally produced in Lorentz mode, since the objective field would produce a very large field along the direction of propagation magnetizing the sample in that direction. It is nonetheless useful as a case study. Fig 1a is an image of the exit wavefunction after interaction with the atomic electrostatic potential. Fig 1b is an image of the phase effect due to the Lorentz force produced by atoms, which are modeled as dipoles. Fig 1c is the part of the wavefunction that has undergone a spin flip due to interaction with the atomic magnetic field. According to mechanisms highlighted in a previous work [4], the spin flip is associated with a change in the orbital angular momentum of the wavefunction. Unfortunately, and, consistently with scalar multislice simulations, we find that Orbital Angular Momentum (OAM) is not conserved in the elastic interactions due to the non-circularly symmetric crystalline potential. A noticeable exception is the case when vortex states are coupled with the Bloch waves of lower transverse energy which feel a locally radially symmetric potential [5]. Fig 2 shows the simulated propagation of a very narrow vortex beam after interaction with a sample. Fig 2a is an image of the exit wavefunction of a probe concentrated on a Fe column. Figures 2b and c are images of the Microsc. Microanal. 21 (Suppl 3), 2015 1962 small fraction of the wavefunction that has undergone a spin flip. The difference between the two images depends upon the opposite spin-OAM projection of the initial state. We can see that the OAM of the final state is drastically different in the two cases. References: [1] Axel Rother, Kurt Scheerschmidt Ultramicroscopy 109 (2009) 154�160 [2] Vincenzo Grillo, Lorenzo Marrucci, Ebrahim Karimi, Riccardo Zanella and Enrico Santamato New journal Phys 15 (2013) 093026 [3] Alexander Edstr�m, Jan Rusz, IT-16-P-3287, 18th International Microscopy Congress 2014 [4] Ebrahim Karimi, Lorenzo Marrucci, Vincenzo Grillo, and Enrico Santamato Phys Rev Lett 108, 044801 (2012) [5] Huolin L. Xin and Haimei Zheng Microsc. Microanal. 18 (2012), 711�719 a) b) c) Figure 1. a) Exit wave after the interaction with the electrostatic potential of a Fe [100] cell. b) Lorentz phase effect introduced by the Fe atoms, which are modeled as dipoles. c) Wavefunction of the electrons that have undergone a spin flip during the interaction with the magnetic field. In both fig a and c, the intensity is proportional to the color brightness, and the phase is proportional to the hue. a) b) c) Figure 2. a) Exit wave after the interaction of a convergent probe with the electrostatic, magnetic and Pauli potential of a Fe [100] cell. b,c) Fraction of the exit wave that has undergone a spin- flip during the interaction in the case of (i) S.L >0, and (ii) S.L <0. The intensity is proportional to the color brightness, and the phase is proportional to the hue");sQ1[980]=new Array("../7337/1963.pdf","Nanoscale Strain Mapping in Embedded SiGe Devices by Dual Lens Dark Field Electron Holography and Precession Electron Diffraction","","1963 doi:10.1017/S1431927615010594 Paper No. 0980 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Nanoscale Strain Mapping in Embedded SiGe Devices by Dual Lens Dark Field Electron Holography and Precession Electron Diffraction Y.Y. Wang1, D. Cooper2, J. Rouviere3 C.E. Murray4, N. Bernier2, J. Bruley4 1. 2. IBM Micro-electronics division, 2070 Route 52, Hopewell Junction, NY 12570, USA CEA, LETI, MINATEC Campus, 17 rue des Martyrs, 38054 Grenoble Cedex 9, France 3. CEA, INAC, MINETEC Campus, 17 rue des Martyrs, 38054 Grenoble Cedex 9, France 4. IBM T. J. Watson Research Center, 1101 Kitchawan Road, Route 134, Yorktown Heights, NY 10598, USA For the past decade, stressors have been incorporated into the source and drain regions of the silicon semiconductor device to change the lattice constant of the current-carrying region in the channel, thereby altering the band structure of the semiconductor to enhance device performance. In semiconductor industry, it is critical to measure strain distributions at the nanometer scale. In recent years, dual lens dark field electron holography and precession electron diffraction are developed to obtain strain distribution at ~1 nm spatial resolution [1-7]. We use these two techniques to measure strain distribution of box shaped embedded SiGe devices and we compare our result with Eshelby inclusion simulations [8]. Fig.1 is the strain map obtained by dual lens dark field electron holography. The spatial resolution is about 2 nm with 1 nm fringe spacing. Dark field STEM image shows the box shaped embedded SiGe. The <220> strain map shows compressive strain in the channel region with large lattice constant in the embedded SiGe region. The <004> strain map shows slightly tensile strain in the Si region, with large lattice constant in SiGe region. Fig.2 is the strain map obtained by precession electron diffraction (PED). The probe size is about 2 nm. Fig.2(a) is the strain map along <220> direction and Fig.2(b) is the strain along <004> direction. Fig.2(c) is the shear strain map and Fig.2(d) is the crystalline rotation map. The strain map by PED is very similar to the one obtained by dark field electron holography. The shear strain shows high value at the bottom corner of SiGe and SiGe/Si boundary near the surface. The rotation map shows maximum 0.6o crystal rotation at the top surface. Fig.3 is the result of Eshelby inclusion simulation. Fig.3(a) is the simulation for the strain along <220> and Fig.3(b) is the simulation for the strain along <004> direction. The simulation results match well with measurement from dual lens dark field electron holography and electron precession diffraction measurement. The precession electron diffraction provides better S/N ratio maps than the one by dual lens dark field electron holography. However, the acquisition time and storage space for PED is ~103 and ~104 of dark field electron holography, respectively. In conclusion, using dual lens dark field electron holography and precession electron diffraction, we provided strain maps at high spatial resolution and demonstrated that to be valuable methods for semiconductor research and development. Microsc. Microanal. 21 (Suppl 3), 2015 1964 References: [1] M. Hytch, F. Houdellier, F. Hue, E. Snoeck, Nature 453 (2008), p. 1086. [2] Y.Y. Wang et al., Ultramicroscopy 101 (2004), p. 63; US patent: US 7,015,469 B2 (2006). [3] Y.Y. Wang et al., Ultramicroscopy 124 (2013), p. 117. [4] Y.Y. Wang et al., Applied Physics Letters 103 (2013), p. 052104. [5] Y.Y. Wang, A. Domenicucci, and J. Bruley, Microscopy Today, May (2014), p. 2. [6] J. Rouviere et al., Applied Physics Letters 103 (2013), p. 241913. [7] Y.Y. Wang et al., Applied Physics Letters 106 (2015), p. 042104. [8] J.H. Davies, J. Appl. Mech. 70 (2003), p. 655. a b c 50 nm 0 1.5% -1.5% 1.5% -1.5% 0 Figure 1. (a) STEM image; (b) <220> strain map by dark field holography; (c) <004> strain map. (a) (b) (c) (d) -1.5% 0 1.5% -1.5% 0 1.5% -1% 0 1% -1 o 0 1 o Figure 2. Strain map by PED (a) <220> map, (b) <004> map, (c) shear strain, (d) rotation map. Figure 3. Eshelby inclusion model: (a) <220> strain map; (b) <004> strain map.");sQ1[981]=new Array("../7337/1965.pdf","Strain Measurement through the Thickness of Crystal using DBI","","1965 doi:10.1017/S1431927615010600 Paper No. 0981 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Strain Measurement through the Thickness of Crystal using DBI Mana Norouzpour, Rodney Herring CAMTEC, Mechanical Engineering/University of Victoria, Victoria, Canada. Accurate measurement of strain, which plays an important role in physical properties of materials, has been the subject of intense works for decades. Moreover, knowing the strain in 3D at the atomic scale is highly desired for crystal growers and microelectronic device manufacturers since it creates undesirable defects that destroy the beneficial properties of the crystal, however positively, strain can enhance a device's performance by increasing the electron and hole mobility. Here we report a technique of DBI (Diffracted Beam Interferometry) that enables recovering the phase information, which is lost during recording by detectors. The retrieved phase provides information to measure the strain distribution through the thickness of crystals that can be compared to simulated strain profiles providing a method to determine the three dimensional strain distribution. As well, it is known that Higher Order Laue Zone (HOLZ) intensities have sufficient coherence to form high contrast fringes when self-interfered [1, 2]. HOLZ lines are well known for measuring strain with a remarkable sensitivity of 10-4 . In combination with DBI, they complement the developed method of CBED (Convergent Beam Electron Diffraction) used to map the direction and relative magnitude of the displacement field by fitting the Hough transforms of experimental and kinematically simulated HOLZ lines [3]. DBI involves self- interference of split HOLZ lines by applying a controlled range of positive voltages from 0 to ~30 on the electron optically focused biprism inserted in between the specimen and diffraction plane (Fig. 1a). The split HOLZ line within the 000 disc located close to the [320] zone axis, is formed from a Si substrate near the 0.7 0.3 / interface and the interference fringes running parallel to the line direction of the split HOLZ line (Fig 1b-1c). The experimental details include a JEOL JEM-2010F operated at 200 kV, a probe of~ 5 , specimen thickness of~ 200 , energy filtered electrons using a GIF Tridiem and the zero loss electrons having a 5 window. The measured phase profile across the split HOLZ line was obtained using Fourier reconstruction steps by means of Holoworks, a DigitalMicrograph subroutine, indicating a broad peak between 2 lower intensity peaks on either side (Fig. 2a). Simulations of possible phase profiles having displacement fields, , about the / interface, include a V-shape type, a parabola type and a bell-shape type [3], were performed. Only the phase profile of the bell-shape type and the experimental phase profile (Fig. 2b) fit well. Subsequently, with respect to the lattice parameter of the substrate or epitaxial layer 0.7 0.3, the strain profile through the thickness of the crystal has been determined for the first time (Fig. 2c). Considering the relative direction and magnitude of the displacement field while moving away from the Si/SiGe interface into the substrate [3], allows determination of the three-dimensional (3D) strain within the crystal (Fig. 2d). This technique can be applied to the large number of split HOLZ lines present in the diffraction pattern that represent distinct sets of bent lattice planes enabling all of the six strain tensors to be measured. Thus, the self-interference of split HOLZ lines takes the current strain measurement capability from 2D to full 3D crystal strain determination. 1. R. A. Herring et al, M&M, (2011), p.1232-1233. 2. R. A. Herring, J. Electron Microscopy, 58(3) (2009), p. 213-221. 3. Koh Saitoh et al, J. Electron Microscopy, 59(5) (2010), p. 367-378. Microsc. Microanal. 21 (Suppl 3), 2015 1966 b a c Figure 1. a) Schematic of DBI technique, b) Experimental split HOLZ line in the dark field mode, c) the formed interference fringes by DBI parallel to HOLZ line. a b c d Figure 2. a) Recovered phase profile across the split HOlZ line, b) bell-shaped simulated phase profile fits perfectly the experimental phase profile, c) strain plot through the thickness of the crystal using (b), d) 3D strain distribution within the crystal away from the Si/0.7 0.3 interface.");sQ1[982]=new Array("../7337/1967.pdf","Elastic Relaxation of Strained Silicon on Insulator (sSOI) Fins: Nanobeam Diffraction (NBD) and Simulations","","1967 doi:10.1017/S1431927615010612 Paper No. 0982 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Elastic Relaxation of Strained Silicon on Insulator (sSOI) Fins: Nanobeam Diffraction (NBD) and Simulations J. Li1, Pierre Morin2, Q. Liu2, K. Cheng1, N. Loubet2, B. Doris1, J. Gaudiello1 1 2 IBM, Albany, NY 12203 USA STMicroelectronics, Albany, NY 12203 USA juntaoli@us.ibm.com Channel strain engineering has been used since the 90nm node [1]. However, as technology scales, the effectiveness of conventional strain elements (such as Source/Drain stressors, stress liners) is significantly reduced. Intrinsically strained channel materials such as tensile strained-silicon on insulator (sSOI) and compressive strained SiGe, have been applied on planar devices [2, 3] to boost transistors. However, it is critical to understand the impact of Fin geometry on elastic relaxation in advanced 3D FinFETs structures. In this paper, we investigate the elastic relaxation in state-of-the-art sSOI Fins through Nano-Beam Diffraction (NBD) and calibrated 3D simulations. Intrinsically strained Si on insulator was used: initial lattice matched to Si0.7Ge0.3. This sSOI is initially biaxially strained. After Fin patterning (i.e., etch and cut), we studied the evolution of the strain in the Fin structures. The reported results only account for elastic relaxation. Fig.1 shows the cross scetional TEM views of the final Si Fins. Mechanical simulations were performed on Fin structures using a finite element method to solve the balance force equations. We used the nano-beam electron diffraction technique to determine the lattice deformation along the strained Si Fins formed on insulator (Fig. 2). Longitudinal strain along the length of a 380 nm Fin, at mid HFin, was then inferred from the Si substrate. The resulting profiles were used to validate mechanical simulation. Both measured and simulated profiles are shown on Fig. 2. An excellent matching is obtained. Stress contour map of the half strain Si Fin was shown in Fig.3 (left) and the elastic relaxation is clearly visible at the edges of the Fin and is more pronounced at the top of the Fin. Little relaxation is seen at the Fin/BOX interface where the Fin material is anchored to the oxide. Fig. 3 (right) illustrates the vertical transversal stress gradients from Fin bottom to top. When the Fin is cut, the strain is not maintained along the edges. In order to evaluate the stress relaxations from sSOI Si Fins with various Fin lengths, Focused Ion Beam (FIB) was used to simulate the Fin Cut process to form Fins. LFin was varied between 100 and 1000nm. To be consistent with current Fin formation techniques, LFin was defined by a FIB cut ending on the BOX for SOI. Foused Ga ion beam with beam energy of 30keV and beam current of 50pA was used to cut inisde the macro with extreme long (600�m) parallal Fins. Fig.4a shows the low mag STEM view of the sample prepared using the method decribed above. Fig.4b shows a zoom in view of the 860nm Fin. Fig.5 (top) shows the in plane lattice deformation map taken from the 860 nm Fin (The refernce was taken from the center of sSOI Fin). Fig.5 (bottom) illustrates the in plane lattice deformation profile along long sSOI Fin, at mid HFin. The investigation of strain relaxation/distribution in shorter Fins is still ongoing. In addition, similar stress relaxation studies on SGOI (SiGe on Insulator) and bulk SiGe FinFET structures will also be discussed. More detailed NBD strain analysis and FEM mechanical simulations will be included in this work. The NBD strain measurements correlated with FEM simulation on sSOI has been demonstrated that it can provide the high spatial resolution and high strain sensitivity for strain analysis in semiconductor Microsc. Microanal. 21 (Suppl 3), 2015 1968 materials. This technique can be used to investigate the elastic relaxation in state of the art sSOI Fins, SGOI and bulk SiGe FinFET devices. Reference [1]T. Ghani et al., IEDM, p.1161, 2003. [2]A. Khakifirooz et al., VLSI, p.117, 2012. [3] K. Cheng et al., IEDM, p.419, 2012. Acknowledgement: This work was performed by the Research Alliance teams at various IBM Research and Development Facilities. Fig.2 NBD measurement (blue line) and Simulation (orange line) of the longitudinal strain along the Fin length for a 380nm long sSOI Fin. Fig.1 (a) TEM cross-sections of sSOI Fins; (b) dark field STEM image and (c) 380 nm long sSOI Fin. (a) (b) Fig.3 Stress contour map (left) and the plane stress profile from Fin bottom to top (right). Fig.4 Simulated Fin Cut by Focused Ion Beam (FIB). Fig.5 In plane lattice deforamtion map. The reference was taken from the center of sSOI Fin (top). In plane lattice deformation profile along long sSOI Fin at mid HFin (bottom)");sQ1[983]=new Array("../7337/1969.pdf","Electrostatic-Potential Analysis of Charged Particles by Split-Illumination Electron Holography","","1969 doi:10.1017/S1431927615010624 Paper No. 0983 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electrostatic-Potential Analysis of Charged Particles by Split-Illumination Electron Holography Toshiaki Tanigaki1,2, Zentaro Akase2,3, Shinji Aizawa2, Hyun Soon Park4, Yasukazu Murakami2,3, Daisuke Shindo2,3 and Hiromitsu Kawase5 1. 2. Central Research Laboratory, Hitachi, Ltd., Hatoyama 350-0395, Japan Center for Emergent Matter Science (CEMS), RIKEN, Wako 351-0198, Japan 3. Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, Sendai 980-8577, Japan 4. Department of Materials Science & Engineering, Dong-A University, Busan 604-714, Republic of Korea 5. System Research & Development Center, Ricoh Institute of Technology, RICOH Co., Ltd., Kanagawa 224-0035, Japan Electron holography is a powerful technique for high-resolution evaluation of electromagnetic field distributions in and around materials. Precise electrostatic potential evaluations, however, have been difficult to perform because of the following reasons: The range of Coulomb electrostatic field is much longer than the lateral coherence length of the illuminating electron waves used in electron holography [1]; The incident electrons for observation as well as secondary electrons emitted from the sample can cause changes in electrostatic distributions in and around the sample with low-conductivity [2]. Split-illumination electron holography (SIEH) [3] is a suitable technique for these analyses because it can overcome the problem of lateral coherence length by setting the reference wave away from the sample. In this paper, we report the first practical application of SIEH in the analysis of the electrostatic potential distributions in and around charged insulating materials, particularly model toner samples containing polystyrene particles on a carrier particle used in electrophotography. Figure 1 shows a typical scanning electron microscopic image of toner particles on a carrier. Note that the image was obtained after holographic observations were performed. The electrostatic potential distributions around the toner particle indicated by an arrow were evaluated using SIEH. Figure 2 shows the schematic of SIEH electron optics. To prevent charging of the insulating samples by electron irradiation during the electron holographic observation, a mask Cu plate was placed at illuminating system. In addition, adjusting a mask focusing lens placed between the mask and the sample, the focused shadow of the mask was formed exactly on the sample. The electrostatic potentials were analyzed by phase shift simulation using a model in which the electrostatic potentials had axial symmetry along the axis penetrating the center of the toner and the carrier particles. Figure 3 shows the results: the potential distributions around the toner particles on the carrier were nonuniform [4]. The results obtained by the developed method have deepened our understanding of the potential distributions, which should lead to further improvements in electrophotography. This method is a powerful technique for clarifying the charge distributions of and around various types of insulating materials used in applied physics and engineering. Microsc. Microanal. 21 (Suppl 3), 2015 1970 References: [1] G. Matteucci et al., J. Appl. Phys. 69 (1991) p.1835. [2] M. Shirai et al., Ultramicroscopy 146 (2014) p.125. [3] T. Tanigaki et al., Appl. Phys. Lett. 101 (2012) 043101. [4] T. Tanigaki et al., Appl. Phys. Lett. 104 (2014) 131601. [5] This research was partly supported by a grant from the Japan Society for the Promotion of Science (JSPS) through the "Funding Program for World-Leading Innovative R&D on Science and Technology (FIRST Program)" initiated by the Council for Science and Technology Policy (CSTP). This work was partly supported by the Cooperative Research Program of "Network Joint Research Center for Materials and Devices" from the Ministry of Education, Culture, Sports, Science and Technology (MEXT). Figure 1. Scanning electron microscopic image of positive toner particles (yellow colored) on a carrier particle (green colored). Electrostatic potential distributions around the toner particle including area near carrier particle were analyzed by electron holography and electron phase shift simulation. Figure 2. Schematic of SIEH electron optics for observing electrostatic potential distributions of charged insulating materials. Figure 3. Analyzed electrostatic potential distributions around positive toner particle in the plane perpendicular to the electron beam (cross-sectional view).");sQ1[984]=new Array("../7337/1971.pdf","Crystalline Phase Mapping Associated to the Magnetic Flux in Cobalt Nanowires","","1971 doi:10.1017/S1431927615010636 Paper No. 0984 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Crystalline Phase Mapping Associated to the Magnetic Flux in Cobalt Nanowires Arturo Ponce1, Bethanie J.H. Stadler2, Francisco Ruiz1, Fernando Mendoza-Santoyo1, Mazin M. Maqableh2, Israel Betancourt1, Jesus Cantu1 and John E. Sanchez1 Department of Physics and Astronomy, University of Texas at San Antonio, San Antonio, Texas 78249, United States 2 Electrical and Computer Engineering, University of Minnesota, 4-174 EE/CSci Bldg., 200 Union St. SE, Minneapolis, Minnesota 55455, USA Electron holography is a powerful technique for mapping of the electric, magnetic fields and the crystalline potential within a transmission electron microscope (TEM). Particularly, the possibility to extract the magnetic behavior of a sample at nanoscale, at the remanent and excited states, transforms the microscope in another lab for the measurement of physical properties of materials. Parameters such as the external magnetic field produced by the objective lens, temperature, reversal magnetization, play an important role in the understanding of the magnetic properties of individual nanostructures like nanoparticles and nanowires [1]. Recently, phase orientation mapping in a TEM is more popular to indexing electron diffraction patterns and the formation of virtual images (bright field, dark field and annular dark field) by a correlation between an electron diffraction pattern and the rest of patterns in the area of the scanned areas [2,3]. The combination of electron holography and crystalline orientation phase mapping can provide of important information not only about the magnetic behavior but the different orientation of the magnetization as function of the orientations of nano-grains in a polycrystalline structure and extract quantitative information of the easy and hard axes in the nano-domains. In this work, we will present the analysis of a polycrystalline cobalt nanowire using both techniques, electron holography and crystalline phase orientation, in the same area. Previously the magnetic field of the objective lens (in a JEOL ARM200F microscope) was characterized by using a Hall sensor in a modified sample holder and using an external Gauss-meter for data collection. In this way, the remanent state of the objective lens at zero volts was characterized. Reversal magnetization was also performed by tilting the sample to obtain the component of the magnetic field in two different directions. In the Figure 1a we present a curve which describe contribution of the magnetic field to the sample position, in ground state we measure a field of 45 Oe, when the OL is switched off to a saturation field at the lens maximum excitation voltage. However the measurement was reproducible after applying a relaxation in the lenses, finding an error in the range of 0.25-0.35%. The magnetic measurements during the specimen loading procedure were very stable and no variations were detected. The electron hologram of the nanowire was processed using Holowork script in DigitalMicrograph, in Figure 1b a hologram is recorded, another hologram is also recorded as the reference to be used in the off-axis phase reconstruction process, its corresponding unwrapped phase image is displayed in Figure 2c. For crystalline phase mapping we performed a scanning over the sample with a lateral resolution 2 nm using a probe size of 1.1 nm under nano-diffraction mode. The electron diffraction patterns were recorded by using a precession angle of 0.9�. For magnetic mapping we recover the phase 1 Microsc. Microanal. 21 (Suppl 3), 2015 1972 of the hologram and amplify three times the cosine of the magnetic phase image. The results that will be discussed in the presentation will show also results of simulations of the magnetization and magnetic flux in the wire considering hard and easy axes in the hcp grains distributed along the cobalt nanowire. The correlation between the crystalline phase orientation map and the magnetization is shown in Figures 2a and 4b, respectively. The effect of the crystalline structure is clearly observed at the tip of the nanowire where the random crystal orientation of the grains induce some fluctuations in the magnetic flux. The wavy nature of the magnetization is also influenced by the grain orientation and the grain boundaries of the crystallites. In figure 4a, the Co nanowire z-orientation map is overlaid with the index map. The color key code of the orientations is depicted on the right. The Co nanowire is composed of several hcp grains with different orientations. [4] Figure 1. (a) Magnetic field in the specimen position in function of the OL excitation voltage. Black curve is the forward bias, and red is the backwards showing no significant hysteresis. The inset, shows a zoom in the linear part of the curve showing that there is some hysteresis in the lens. (b) Object electron hologram, the inset indicates the FFT from where we took one sideband to reconstruct the phase image c) Unwrapped phase image, displaying the phase variations characteristic of magnetic samples. Figure 2. (a) Crystalline phase orientation map, in which an inset of the map colors of the planes is included. (b) Magnetic phase recovered from the off-axis electron hologram. References: [1] Cantu-Valle J, et al, J. Magn. Magn. Mater. 379 (2015), p.294 [2] Rauch, EF, Mat -wiss u Werkstofftech 36 (2005) p.552 [3] RuizZepeda, F, Microsc. Res. Tech. 77 (2014) P.980 [4] This project was supported by NSF PREM DMR #0934218 and the Department of Defense #64756-RT-REP and the NIH RCMI Nanotechnology and Human Health Core (G12MD007591).");sQ1[985]=new Array("../7337/1973.pdf","New Quantitative Phase Reconstruction Technique using Hollow-cone Probe and Annularly Arrayed Detectors in STEM","","1973 doi:10.1017/S1431927615010648 Paper No. 0985 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 New Quantitative Phase Reconstruction Technique using Hollow-cone Probe and Annularly Arrayed Detectors in STEM Takafumi Ishida1, Tadahiro Kawasaki2, 3, 4 , Takayoshi Tanji2, 4 and Takashi Ikuta5 1. Department of Electrical Engineering and Computer Science, Nagoya University, Nagoya 464-8603 Japan 2. EcoTopia Science Institute, Nagoya University, Nagoya 464-8603 Japan 3. Nanostructures Research Laboratory, Japan Fine Ceramics Center, Nagoya 456-8587 Japan 4. Global Research Center for Environment and Energy based on Nanomaterials Science, Nagoya, 4648603 Japan 5. Department of Electrical and Electronic Engineering, Osaka Electro-Communication University, Neyagawa 572-8530 Japan The quantitative evaluation of the specimen structure in an image requires a projected potential image. The established method for evaluating the electrostatic potential in a specimen is electron holography which can directly reconstruct the phase of electron waves. In scanning transmission electron microscopy (STEM), a phase reconstruction technique that employs a multi-channel detector has been proposed [1,2]. The resolution of a reconstructed phase image based on the use of a multi-channel detector can be twice that of a conventional bright-field image. This means that we may expect higher resolution than electron holography. The annular bright-field phase (ABFP) imaging technique proposed by the present authors [3], which is a phase reconstruction technique using a multi-channel detector, has been demonstrated to be capable of high-resolution phase imaging. The reconstructed phase image obtained by ABFP imaging enables that quantification of the projected potential [4]. Figure 1 shows a schematic of ABFP imaging. A hollow-cone probe is formed with an annular objective aperture, and multiple images are simultaneously recorded with annularly arrayed detectors (AADs). Each image contains different information, because the AADs capture waves diffracted at different azimuthal angles. The phase image is reconstructed by the following processes. The images obtained by AADs are Fourier transformed. Then, a ring-shaped band pass filter is applied to each Fourier spectrum to extract the linear imaging components. Finally, the sum of these filtered spectra is inversely Fourier transformed to obtain the real and imaginary parts of the electron wave. The reconstructed phase image is determined by the arctangent of the ratio of the imaginary and real parts of the reconstructed complex amplitude wave. In order to evaluate quantitatively ABFP imaging, we performed the multislice image simulation. Figure 2(a) is an example of the simulated phase image of a single atom (Cu) reconstructed using ABFP imaging. Images were calculated for single neutral atoms of atomic numbers from 1 to 103. Since the absolute value of the phase shift cannot be determined in our reconstruction method, the relative phase shift measured by the height from the negative bottom to the positive top of the peak was defined as the maximum phase shift, as shown in Fig. 2(b). Figure 2(c) shows the maximum phase shifts for single atoms as a function of atomic number Z in ABFP imaging. A least-squares fitting curve shows that the maximum phase shifts are proportional to approximately Z0.6. The result of Fig. 2(c) is in good agreement with the simulated results of phase shifts obtained by electron holography [5]. This indicates that the projected potential was determined using the maximum phase shift, though it is a relative value. The reconstruction method can also be applied to thin crystal samples, and has the capability of counting the number of atoms in columns. Microsc. Microanal. 21 (Suppl 3), 2015 1974 In conclusion, we have developed a new phase reconstruction technique for high-resolution in STEM. This technique was quantitatively evaluated using image simulations. The results showed that the phase shifts caused by single atoms were proportional to Z0.6. Comparisons of experimental and simulated results for thin crystal samples will be discussed [6]. References: [1] M Landauer et al., Optik 100 (1995), p. 37 [2] M. Taya et al., Rev. Sci. Instrum. 78 (2007), p. 083705 [3] T. Ishida et al., Microscopy (in press) doi: 10.1093/jmicro/dfu098 [4] T. Ishida et al., Microscopy (in press) doi: 10.1093/jmicro/dfu113 [5] H. Lichite et al., Annu. Rev. Mater. Res. 37 (2007) p. 539 [6] This work was supported by JSPS KAKENHI Grant Numbers 18GS0211 and 24360020, and in part by the program "Global Research Center for Environment and Energy based on Nanomaterials Science" of MEXT, Japan. Figure 1. Schematic diagram of an STEM instrument for ABFP imaging. The hollow-cone probe is formed by an objective lens using an annular aperture, and the different azimuthal angle electrons that pass through the specimen are simultaneously acquired by the annularly arrayed detectors. The annular objective aperture inner and outer semi-angles 1 and 2 are determined so as to be coincident with the inner and outer angles 1 and 2 of the annularly arrayed detectors in the detection plane. Figure 2. (a) ABFP image of a single atom (copper). An estimated probe size is 0.1 nm (Va = 200 kV). (b) Line profile of (a) (A-A'). (c) Maximum phase shift due to a single atom as a function of atomic number. The dashed line is a least-squares fitting curve. The simulated phase shift is proportional to Z0.6.");sQ1[986]=new Array("../7337/1975.pdf","Analysis of GaAs Compound Semiconductors and the Semiconductor Laser Diode using Off-Axis Electron Holography, Lorentz Microscopy, Electron Diffraction Microscopy and Differential Phase Contrast STEM","","1975 doi:10.1017/S143192761501065X Paper No. 0986 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Analysis of GaAs Compound Semiconductors and the Semiconductor Laser Diode using Off-Axis Electron Holography, Lorentz Microscopy, Electron Diffraction Microscopy and Differential Phase Contrast STEM Hirokazu Sasaki1, Shinya Otomo1, Ryuichiro Minato1, Kazuo Yamamoto2, Tsukasa Hirayama2 Jun Yamasaki3 and Naoya Shibata4 1. 2. Furukawa Electric Ltd., Yokohama and Japan. Nanostructure Research Laboratory, Japan Fine Ceramics Center, Nagoya and Japan. 3. Osaka University, Osaka and Japan. 4. Institute of Engineering Innovation, The University of Tokyo, Tokyo and Japan. In order to develop and manufacture semiconductor devices which are key components of the optical telecommunication products, such as the semiconductor laser diode, it is essential to confirm whether it is manufactured as designed. Electric potential distributions of the semiconductor devices are designed in nanoscale, so two dimensional methods to evaluate the electrical potential in the semiconductors with a high spatial resolution are necessary for product management. The observation of the gallium arsenide (GaAs) model specimen and the analysis of the semiconductor laser diode were carried out by using the electron holography, which is one of the methods of the transmission electron microscope, and Lorentz microscopy. In the observation using the electron holography, not only pn junction but also interfaces which are in different dopant concentration regions of the 1x1019 and 1x1018 cm-3 regions and the 1x1018 and 1x1017 cm-3 regions could be observed [1]. Then, the analysis example for the semiconductor laser diode was introduced and described that these methods have been used practically. For other semiconductor electric voltage evaluation methods by TEM, electron diffraction microscopy [2] which is one method of phase reconstruction method, differential phase contrast [3] (DPC) which is one method of STEM are also effective and possible to be utilized complementarily with the electron holography. We will discuss about these method applied for semiconductor in this presentation. References: [1] H Sasaki .et al, Microscopy, 632014, 235. [2] J Yamasaki .et al, Appl. Phys. Lett., 1012012, 234105. [3] N. H. Dekkers & H. de Lang, Optik 41, 452 (1974). Microsc. Microanal. 21 (Suppl 3), 2015 1976 Figure 1. Phase image of the GaAs specimen reconstructed by the phase-shifting method. Figure 2. aLine profile of the phase image in the p-type region. (The arrows indicate a p-n junction.) bZn SIMS profile. cLine profile of the phase image in the n-type region. dSi SIMS profile.");sQ1[987]=new Array("../7337/1977.pdf","Three Dimensional Visualization of Electromagnetic Fields from One Dimensional Nanostructures","","1977 doi:10.1017/S1431927615010661 Paper No. 0987 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Three Dimensional Visualization of Electromagnetic Fields from One Dimensional Nanostructures Charudatta Phatak1 , Aur� lien Masseboeuf2 , Ludvig de Knoop2 , Christophe Gatel2,3 , Martin J. H� tch2 e y 1 Materials Science Division, Argonne National Laboratory, Lemont, IL 60439, USA CEMES-CNRS, 29, rue Jeanne Marvig, FR - 31055 Toulouse, France 3 Universit� Paul Sabatier, F-31000 Toulouse, France. e 2 One-dimensional (1D) nanostructures have been regarded as the most promising building blocks for nanoelectronics and nanocomposite material systems as well as for alternative energy applications [1]. Magnetic nanowires with circular cross-section, are of utmost importance from theoretical and technological aspects [2]. 1D carbon-based nanostructures such as carbon nanotubes are amongst the best candidates for field emission displays and new high-brightness electron sources [3]. The confinement effects in 1D nanostructures can alter their properties and subsequently their behavior significantly. Hence it is necessary to understand the strong effect of their size on their three-dimensional (3D) properties such as the magnetic and electric fields associated with nanowires and nanotubes completely before they can be used in applications. There are currently very few methods, which have the capability to visualize the complete 3D fields associated with nanowires. In this work, we show that using a combination of symmetry arguments and electron-optical phase shift data obtained using TEM, it is possible to recover the entire 3D magnetic or electric field in and around nanowires and nanotubes from a single image. The phase change e +m of an electron wave traveling through a thin foil can be expressed in terms of tomographic quantities [4]. The phase shift can be recovered experimentally using various techniques such as transport-of-intensity based methods or off-axis electron holography. Nominally, determining the 3D magnetic field or 3D elestrostatic potential requires recording a series of phase shift images as the sample is tilted about its axis. However, in a 1D nanostructure such as a nanocylinder that is uniformly magnetized along its long axis, the magnetic field possesses cylindrical symmetry with respect to the long axis. Similarly, carbon nanotubes under applied bias exhibit a cylindrically symmetric potential and electric field. Exploiting these conditions of cylindrical symmetry, we can reconstruct the full 3D magnetic or electric field associated with a nanowire from a single phase image using the inverse Abel transform. Simulations were performed for a uniformly magnetized sphere, which also possesses cylindrical symmetry to analyze the fidelty of the reconstruction algorithm. Fig. 1(a) shows the magnetic phase shift of such a uniformly magnetized sphere, (b) shows the derivative with respective to the horizontal axis (m /x) showing the cylindrical symmetry, and (c) the 3D reconstructed magnetic induction using the single image method developed in this work. Experiments were performed on nickel nanowires that were uniformly magnetized to reconstruct the 3D magnetic induction. Similarly the 3D electric field was reconstructed from in-situ biased carbon cone nanotips under varying applied bias. The phase shift was recovered using off-axis electron holography in a FEI Tecnai F20 TEM operating at 200 kV using the first transfer lens of the Cs-corrector as a Lorentz lens. The magnetic phase shift was obtained from the Ni nanowire by recording the phase shift images with the sample as-is and rotating it by 180 about its axis and then subtracting the two Microsc. Microanal. 21 (Suppl 3), 2015 1978 phase shifts. Fig. 2 (a) shows the magnetic phase shift of the Ni nanowire and (b) shows the reconstructed 3D magnetic induction. We will further discuss the details of the application and limitation of this method to reconstruct the 3D electromagnetic fields. References [1] C. M. Lieber, Solid State Communications, 107, 607 (1998). [2] C. Chappert, A. Fert, F. N. Van Dau, Nature Materials, 6, 813 (2007). [3] N. de Jonge, Y. Lamy, K. Schoots, T. H. Oosterkamp, Nature, 420, 393 (2002). [4] C. Phatak, M. Beleggia, and M. De Graef, Ultramicroscopy, 108, 503 (2008). [5] This work was support by U.S. Department of Energy (DOE), Office of Science, Materials Sciences and Engineering Division. (a) (c) By 0.8 0.6 0.4 (b) 0.2 0.0 y z x -0.2 Figure 1: (a) shows the magnetic phase shift image of a uniformly magnetized spherical nanoparticle, indicated by the dashed line, and magnetization vector is indicated by the blue arrow, (b) the derivative of the magnetic phase shift with respect to x showing the symmetry about the vertical axis of the image, and (c) the 3D reconstructed magnetic induction. (a) (b) |B| (T) 0.5 0.4 0.3 0.2 50 nm z y x 0.1 0.0 Figure 2: (a) shows the experimentally recovered magnetic phase shift from a Ni nanowire. The dotted lines indicate the border of the nanowire. (b) the 3D reconstructed magnetic induction from the nanowire.");sQ1[988]=new Array("../7337/1979.pdf","Progression of Focused Helium Ion Beam Milling in Gold Substrates","","1979 doi:10.1017/S1431927615010673 Paper No. 0988 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Progression of Focused Helium Ion Beam Milling in Gold Substrates E.M. Mutunga1, S. Tan2, A.E. Vlad�r3, and K.L. Klein3, 4* 1. 2. University of Tennessee, Knoxville, TN 37996 Intel Corporation, Santa Clara, CA 95054 3. National Institute of Standards and Technology, Gaithersburg, MD 20899 4. University of the District of Columbia, Washington, DC 20008 * Corresponding author: kate.klein@udc.edu The focused helium ion beam-solid state material interaction volumes differ in thin membranes and in bulk substrates [1, 2]. While nanometer-scale structures with feature sizes of 5 nm or smaller have been machined in thin membranes [3-5], inconsistent results have been observed in bulk substrates where the irradiated region often swells rather than mills [6]. The depth and breadth of the interaction volume scales with beam energy, and the sputter yield also varies significantly. In this study, we seek to explore and understand these differences by observing the progression of focused helium ion beam milling. For this we used transmission electron microscope images of the beam-irradiated cross sections of 100-nm-thick gold membrane and 400-nm-thick (essentially bulk) gold substrates. We varied the ion beam landing energies and the dose as well. Recent work [2] compared the interaction profiles of a 35 keV helium ion beam of varied dose in 140nm-thick single crystalline silicon membrane and bulk samples using TEM cross-sectional imaging. It was found that forward sputtering is up to 5 times higher than backward sputter yields and is more significant in thin membranes where the thickness approaches the nuclear stopping range of the ions in the bulk substrate. Here we look at the machining properties of gold using similar techniques and further explore the effect of beam energy on this interaction profile. Line arrays (Figure 1) have been patterned using a 35 keV helium ion beam in 100-nm-thick gold foil (a) and gold bulk substrate (b) using comparable dose. The gold bulk substrate thickness exceeds the ion range of 97 nm indicated by SRIM [7] helium ion point trajectories for a 35 keV beam. This results in increased helium implantation in the material and causes more pronounced surface distortion compared to the 100 nm gold foil where approximately half of the ions are transmitted through the thin membrane. Nanomachining in bulk substrates is difficult due to this increased helium implantation in the material, which causes both surface and sub-surface damage. It is therefore desirable to find the optimal sputter yield conditions in addition to using the lowest dose feasible. Modeling results show higher sputter yields relate to lower beam energies. A graph of sputtering yields for various gold sample thicknesses at 1-50 keV beam energies is shown in Figure 2, where the theoretical sputter yield peaks between 5-7 keV. It was also experimentally observed that for 20 keV as compared to 35 keV, the mill rate in a 100 nm gold foil increases by approximately 20 %. [1] E. Mutunga et al, EIPBN Proceedings (2014). [2] S. Tan et al, J. Vac. Sci. Technol. B 32, 06FA01 (2014) [3] L. Scipioni et al, J. Vac. Sci. Technol. B 28, C6P18 (2010). [4] J. Yang et al, Nanotechnology 22, 285310, (2011). [5] L. Scipioni, Carl Zeiss, Adv. Mat. Char.Workshop, U. Illinois, (2012). [6] R. Livengood et al, J. Vac. Sci. Technol. B 27 (2009) 3244. [7] J.F. Zeigler, M.D. Zeigler and J.P. Biersack, SRIM-2012.03 modeling freeware. Microsc. Microanal. 21 (Suppl 3), 2015 1980 Figure 1. HIM micrographs of one-pixel-wide line arrays patterned at 30 nm pitch over a 1 �m field-ofview using a 35 keV helium ion beam on (a) 100-nm-thick gold foil and (b) 400-nm-thick gold bulk samples at a dose of 3.12 x 105 ions/nm. Surface distortion on the bulk sample is evident due to the accumulation of dislocations and damage caused by the trapped helium in the surrounding and underlying areas. Figure 2. Plot showing theoretical backward sputter yield as a function of beam energy. Note the decreased sputter yield with increased beam energy for various thicknesses of gold.");sQ1[989]=new Array("../7337/1981.pdf","Investigation of Ar Ion-Milling Rates for Transmission Electron Microscopy Specimens","","1981 doi:10.1017/S1431927615010685 Paper No. 0989 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Investigation of Ar Ion-Milling Rates for Transmission Electron Microscopy Specimens Min-Hee Lee, Kyou-Hyun Kim* Advanced Process and Materials R&BD Group, Incheon Regional Division, Korea Institute of Industrial Technology, Incheon 406-840, Republic of Korea Recently, focused ion beam (FIB) is widely used to prepare a TEM specimen because the region of interest becomes narrow and specific. FIB system can precisely mill the specific area with a spatial accuracy of within few tens of nanometers. FIB, however, has problems such as Ga-induced damages near a sample surface which affect the investigation of original sample structure [1]. In addition, the thickness of TEM specimens prepared by FIB is commonly thicker than Ar-ion milled TEM specimens which makes difficult to obtain high quality of structural and chemical information from electron diffraction (ED) pattern, high resolution (HR) image, atomic resolution scanning-TEM (STEM) image, electron energy loss spectroscopy (EELS), etc. This then gives rise to the motivation that FIB-prepared specimens need to be further treated to reduce the artifacts induced by the high energy of Ga-ion sauce. By contrast with FIB, the typical Ar-ion milling system uses much lower accelerating voltage and shows no implantation effect from Ar-ion. The recent Ar-ion milling systems provide rocking mode or selective sector speed thinning methods to mill TEM specimens uniformly down to few nanometers in thickness. Also, the damage from Ar-ion can be reduced by using the low accelerating voltage of <1 kV. Based on the above, we applied the Ar-ion milling system to the FIB-prepared specimens in order to remove the damaged area induced by the Ga-ion sauce. Nevertheless, the FIB-prepared specimen is thin so that the Ar-ion milling rate needs to be monitored to prevent losing the specimen after the further milling using the Arion milling system. Motivated by the above considerations, we study here the Ar ion-milling rates for a couple of materials. For this study, we selected the single crystals of Si, GaAs, Ni and Cu. The selected specimens are first pre-thinned up to ~300 nm in thickness using FIB. In order to measure the Ar-ion milling rates, the FIB-prepared specimens were further Ar-ion milled at 1.5/3 kV with a 6� incident angle for 60 and 30 seconds, respectively. All samples were mounted on the center of Cu-grid as shown in Fig. 1. The Ar-ion beam directions are indicated by the arrows. The Arion milled specimens were then observed in TEM using convergent beam electron diffraction (CBED) to measure the thickness with help of Bloch simulation. For the Bloch simulation, we used the atomic scattering factors of Doyle and Turner [2], and the absorption parameters of Bird and King [3]. Figure 2(a) shows a medium magnification bright field (BF) image of Si single crystal. The BF image was recorded along the zone axis of [110]. The experimental CBED patterns for the Si single crystal were recorded from nine points as indicated in Fig. 2. Figures 2(b) and (c) respectively show the representative experimental CBED pattern and the corresponding simulated CBED pattern for the thickness of 279 nm. The sample thickness can be then estimated as 279 nm from Figs. 2(b) and (c). This measurement is repeated to calculate the Ar-ion milling rates for Si, GaAs, Ni, and Cu. As shown in Figs. 2(b) and (c), the simulated CBED pattern well agrees with the experimental CBED pattern. We believe that this study provides a direct reference to estimate a milling time not only for the FIB-prepared specimens but also for the mechanically thinned specimens. Microsc. Microanal. 21 (Suppl 3), 2015 1982 References [1] J. Mayer et al, MRS Bulletin 32 (2007), 400-407 [2] P. A. Doyle and P. S. Turner, Acta Crystallogr. Sect. A24 (1968), 390-397 [3] D. M. Bird and Q.A.King, ActaCrystallogr. Sect. A46 (1990), 202-208 [4] This work was supported by the research fund from Korea Institute of Industrial Technology (KITECH). Figure 1. FIB-prepared TEM specimen mounted on a Cu-grid. Figure 2. (a) A medium magnification image of Si single crystal prepared by FIB. (b) The (c) 275 277 279 experiment and (c) simulated CBED pattern to measure the thickness of TEM specimen. (c)");sQ1[990]=new Array("../7337/1983.pdf","Focused Ion Beam Micromachining Enables Novel Optics for X-ray Microscopy","","1983 doi:10.1017/S1431927615010697 Paper No. 0990 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Focused Ion Beam Micromachining Enables Novel Optics for X-ray Microscopy Kahraman Keskinbora1, Umut T. Sanli1, Corinne Gr�vent1, Michael Hirscher1 and Gisela Sch�tz1 1. Max Planck Institute for Intelligent Systems, Dept. Modern Magnetic Systems, Stuttgart, Germany X-ray microscopy is a strong analytical tool with a plethora of applications in physics, materials science and life sciences [1]. In many high resolution X-ray microscopes, a focusing optic such as a Fresnel zone plate (FZP) is utilized as a lens to form an image. The FZP is a diffractive optic composed of a series of concentric rings (zones) of radially varying grating period, where the width, r, of the outermost zone determines the resolution. Synchrotron radiation facilities are required as bright and coherent X-ray sources as the FZPs are usually quite limited in diffraction efficiency. The diffraction efficiency depends on FZP properties such as geometry, material, structure thickness and on the outermost zone width for small r. Conventional FZP profiles are binary and can have up to 10-40 % diffraction efficiency for ideal absorption and phase reversal zone plates, respectively. In practice the efficiencies are lower due to fabrication difficulties and errors. Electron beam lithography (EBL), the usual fabrication method, allows fabrication of binary FZPs with ultra-high resolution via processes of ever increasing complexity. However, due to the scattering of electrons within the resist, it is very difficult to fabricate ultra-high resolution FZPs with high structural thicknesses to achieve high aspect ratios. This, so called, proximity effect limits the aspect ratio, restricting the utilization of high resolution FZPs to softer X-ray energies. One special type of refractive/diffractive X-ray optic is the kinoform lens, with a continuous 3D surface profile. Ideally, it has a theoretical diffraction efficiency of 100 %. While in a real lens absorption would hinder 100 % efficiency, diffraction efficiencies well above 40 % are already demonstrated [2] using these lenses. Nevertheless, in order to fabricate these kinoforms, researchers resorted to step approximations using several consecutive overlay EBL steps, which is complicated and lead to much lower efficiencies than theoretically expected due to the vulnerability of the optics to fabrication errors amplified by error accumulations. To solve the issues that are intrinsic limitations of the EBL-FZPs we introduced novel, precise and mostly direct methods relying on the FIBs. The powerful method potentially allows for higher resolutions than EBL [3] and precise 3D sculpting capability enables the realization of highly efficient X-ray optics with high fidelity [4]. We have used a standard multi-purpose FIB instrument (Nova 600 NanoLab, DualBeam, FEI) to fabricate binary FZPs, kinoform lenses and multilayer FZPs (Figure 1). Fabrication of binary FZPs using IBL is a simple process (Figure 1a) which delivers high resolution FZPs in a very short time frame, in the order of 10 min, (e.g. 13 min [5]). Using these binary FZPs made out of gold, we were able achieve 30 nm effective r and half-pitch X-ray image resolutions down to 21 nm. Further progress down to 25 nm r (50 period), is possible (Figure 2a) via optimization of processes and materials. Various kinoform lenses were fabricated using a gray-scale IBL approach by taking advantage of the 3D fabrication capability of the FIB instrument (Figure 1b) [4]. It was possible to use these lenses for high efficiency soft X-ray focusing and imaging. The diffraction efficiency up to about 14 % was achieved and was limited by the strong absorption, a characteristic of the soft X-rays. Nevertheless, an experimental diffraction efficiency up to 90 % of the theoretical efficiency was achieved, thanks to the high surface quality of the lens (Figure 2b) and the high precision fabrication process [4]. Microsc. Microanal. 21 (Suppl 3), 2015 1984 Furthermore, in another approach, micro-machining and micro-manipulating capabilities of FIBs can also be utilized for high precision slicing, transfer and polishing of the multilayer (ML) FZPs, for both the hard and the soft X-rays (Figure 1c). By using an FIB, slices of a multilayer deposit have been successfully cut out and transferred onto a TEM grid where the surfaces were further polished to a fine optical finish and finally a Pt beamstop was deposited via focused ion beam induced deposition (Figure 1c). The proper function of the ML-FZP (Figure 2c), was demonstrated at a synchrotron radiation facility by resolving ~20 nm (half-pitch) features, without any apparent astigmatism [6]. References: [1] D. T. Attwood, "Soft x-rays and extreme ultraviolet radiation: principles and applications", Cambridge Univ Press, New York (2000). [2] E. D. Fabrizio, F. Romanato, M. Gentili et al., Nature, 401 (1999), 895-898. [3] W.-D. Li, W. Wu, and R. S. Williams, J. Vac. Sci. Technol., B, 30, (2012), 06F304-4. [4] K. Keskinbora, C. Gr�vent, M. Hirscher et al., Advanced Optical Materials, (2014), doi: 10.1002/adom.201400411. [5] K. Keskinbora, C. Gr�vent, U. Eigenthaler et al., ACS Nano, 7 (2013), 9788-9797. [6] K. Keskinbora, A.-L. Robisch, M. Mayer et al., Optics Express, 22 (2014), 18440-18453. Figure 1. Schematic summary of uses of FIB for X-ray optics manufacturing. a) Single step writing of a binary FZP. b) Single step writing of a kinoform lens using gray-scale IBL. c) Fabrication of the MLFZPs; slicing, transfer, polishing and beamstop deposition are made by using FIB. Figure 2. a) Binary zones with 50 nm period written with a FIB. b) A kinoform lens fabricated by using a gray scale FIB, written in a nano-crystalline Pd0.8Si0.2 alloy [4]. c) A multilayer FZP transferred onto a TEM grid and polished by FIB, ready for beamstop deposition.");sQ1[991]=new Array("../7337/1985.pdf","In Situ Probing Biological Structures by Combining Focused Ion Beam and Atomic Force Microscopy","","1985 doi:10.1017/S1431927615010703 Paper No. 0991 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In Situ Probing Biological Structures by Combining Focused Ion Beam and Atomic Force Microscopy Boyin Liu1, VahidReza Adineh1 and Jing Fu1 1 Department of Mechanical and Aerospace Engineering, Monash University, Clayton, VIC 3800 Understanding the heterogeneity of biological structures at the micro/nano scale can offer insights valuable for multidisciplinary research in cell biology and biomimicry designs. Here we propose to combine nanocharacterization tools, particularly Focused Ion Beam (FIB) milling and Atomic Force Microscopy (AFM) for probing the mechanical modulus and chemical signatures of biological structures. This proposed strategy overcomes the physical limit of AFM sampling depth, and allows probing the interior of biological samples and acquiring data previously inaccessible. To demonstrate the proposed approach, bacterial sample K. pneumoniae ATCC 13883 were treated with 2 mg/L polymyxin B for 24 hrs after cultivation. The bacterial cells were sliced at grazing angle ion beam at 9.7 pA to expose the cytoplasm and nucleoid regions for AFM probing (Figure 1) [1]. Figure 2a shows the Young's moduli of the bacterial cell interior measured in rehydrated condition. A large soft region was identified in the central region, which was surrounded by stiffer materials. A higher resolution map of stiffness (Figure 2b) was obtained on the surface of the sectioned cell with ~20 nm scanning intervals. The boundary between the cell and biofilm was distinct, and a number of soft regions were present and surrounded by stiffer features which presumably to be multi-protein complexes. Also, another study was conducted on an important sensory system: rat whisker. For an individual whisker sample, two cuts were milled by FIB first (Figure 3a). The handles of a microgripper in FIB/SEM were utilized to enter both sides of the "disk" followed by milling off the remaining attached part for free transfer (Figure 3b). The flat side of the disk sample representing the cross section of rat whisker was positioned on the substrate, and FIB platinum deposition was used to secure the sample. Figure 4a-c present the acquired AFM images of adhesion for the cortex-medulla, cortex and cuticle regions of the rat whisker interior cross section, while distinct nano/microscale morphologies could be observed. Representative distributions of modulus acquired from different regions of rat whisker interior are presented in Figure 4d, suggesting cortex medulla and cortex are softer than cuticle regions. Based on the acquired information, a complete 3D elastic modulus model could be constructed for simulating and studying the performance of rat whiskers [2]. Microsc. Microanal. 21 (Suppl 3), 2015 1986 References: [1] B Liu, MH Uddin, TW Ng, DL Paterson, T Velkov, J Li and J Fu Nanotechnology, 25(41) (2014), 415101. [2] VR Adineh , B Liu, R Rajan, W Yan and J Fu, Multidimensional characterization of biomechanical structures by combining atomic force microscopy and focused ion beam: a study of the rat's whiskers. (under review) [3] This work was performed in part at the Melbourne Centre for Nanofabrication (MCN) in the Victorian Node of the Australian National Fabrication Facility (ANFF), and Monash Centre for Electron Microscopy (MCEM). Figure 1. FIB milling of bacterial cells. (a) Tilt sample stage to grazing angle. (b) Slice the cell with FIB from the illustrated yellow line. (c) The cell after slicing with (d) top view. Scale bar: 1 �m. Figure 2. (a) Modulus map of bacterial cell interior and (b) with higher resolution. Figure 3. Retrieving and positioning rat whisker cross sections on the substrate using microgripper. The disk sample was cut off (a) and pinched (b), followed by being positioned on substrate with preferred orientation (c) and secured by platinum deposition (d). Scale bars: a) 100 m; b)-d) 50 m. Figure 4. Adhesion maps of rat whisker cross section after FIB milling including (a) cortex-medulla, (b) cortex and (c) cuticle regions. (d) Histogram s of the modulus distributions. (scale bar: 200 nm)");sQ1[992]=new Array("../7337/1987.pdf","Multilayer Fresnel Zone Plates for X-ray Microscopy","","1987 doi:10.1017/S1431927615010715 Paper No. 0992 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Multilayer Fresnel Zone Plates for X-ray Microscopy Umut T. Sanli1, Kahraman Keskinbora1, Corinne Gr�vent1, Adriana Szeghalmi2, Mato Knez3 and Gisela Sch�tz1 1. 2. Max-Planck-Institute for Intelligent Systems, Stuttgart, Germany Friedrich-Schiller-Universit�t Jena Institut f�r Angewandte Physik, Jena, Germany 3. CIC nanoGUNE, San Sebastian and IKERBASQUE Basque Foundation for Science, Spain Scanning transmission X-ray microscopy (STXM) is a powerful technique that allows chemical, structural and elemental specific characterization of materials with resolutions down to about 10 nm where the quality of the focusing optic is the limiting factor. The setup of a STXM is shown schematically in Figure 1. Among all optics the Fresnel zone plates (FZP) are one of the most popular optics in X-ray microscopy. A Fresnel zone plate, in its simplest form, is a set of concentric rings of alternating transparent and opaque nature. These rings constitute the zones of the FZP whose radii follow the relationship = + 2 2 /4 where rn is the radius of the nth zone, is the wavelength and f the focal distance. The resolution = R depends on the outermost zone width, rn. With the improvement of the FZPs and new X-ray sources, laboratory size X-ray microscopes, which do not need synchrotron radiation, recently became available. However to be able to compete with synchrotron sources, laboratory size X-ray microscopes need better focusing optics with higher efficiencies and resolutions at high energy radiation. The standard e-beam lithography based techniques for the FZP fabrication fail to deliver the needed high aspect ratios for high efficiencies and high resolutions at harder X-rays. Our research is focused on a new fabrication method that can deliver the required aspect ratio. In this method a nanometer smooth bare glass fiber is coated with alternating layers of materials to serve as opaque and transparent zones, by using the atomic layer deposition technique (ALD). The ALD is a very precise thin film deposition technique that can handle the extremely fine layer positioning accuracy required for this application. Virtually unlimited number of Multilayer FZPs (ML-FZPs) can be sliced out of the fiber by using a focused ion beam. This is depicted in Figure 2a and 2b. The ML-FZP thicknesses can be varied and optimized to the desired X-ray energy. We recently resolved 21 nm structures of a test sample as well as the innermost 30 nm structures of a Siemens Star test structure (see Fig. 3a) with an Al2O3-Ta2O5 FZP of outermost zone width, r = 35 nm and diameter, d = 38 �m at 1.2 keV [3]. This is, to date, the highest imaging resolution achieved by a ML-FZP to the best of our knowledge. In the hard X-ray range at 7.9 keV a sub-30 nm FWHM (fullpitch) resolution, was deduced from an autocorrelation analysis [3]. Besides improving the resolution by fabricating thinner zones, our research also covers new material couples for higher efficiencies that would compensate the efficiency losses due to volume effects for r < 25 nm. Furthermore, higher efficiency would also allow higher signal to noise ratio and faster imaging. In this work we also discuss the performance of such a new material couple, Al2O3-HfO2. The calculated efficiency of a lens made out of Al2O3-HfO2 as a function of X-ray energy and ML-FZP thickness is shown in Figure 3b. Microsc. Microanal. 21 (Suppl 3), 2015 1988 References: [1] D.T.Attwood, "Soft x-rays and extreme ultraviolet radiation: principles and applications", Cambridge Univ Press, New York (2000). [2] M. Mayer et al., Ultramicroscopy 111 (2011) 1706-1711 [3] K. Keskinbora et al., Optics Express, 22 (2014), 18440-18453. Figure 1. Schematic of STXM imaging setup. The order selecting aperture OSA eliminates unwanted diffraction orders. The sample is raster scanned and the transmitted light is collected by an APD. Figure 2. a) 6 step FIB slicing and polishing of ML-FZPs and b) SEM image showing excellent surface and zone quality. Figure 3. a) STXM image of a Siemens Star test pattern. The 30 nm features of the inner circle are clearly resolved. b) Calculated efficiency of an HfO2-Al2O3 FZP.");sQ1[993]=new Array("../7337/1989.pdf","FIB-assisted TEM Sample Preparation Refinement Using TRIM Simulations","","1989 doi:10.1017/S1431927615010727 Paper No. 0993 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 FIB-assisted TEM Sample Preparation Refinement Using TRIM Simulations Ashley Sutor1 and Bryan Gauntt1 1. Intel Corporation, CQN, Hillsboro, USA Focused ion beam (FIB) assisted TEM sample preparation uses an ion beam to precisely remove material in order to create electron transparent specimen. This technique enables the preparation of lamellae containing small targeted features, and can conveniently be used for specimen preparation from an expansive list of bulk materials [1]. The drawback for using high energy ion bombardment for TEM specimen creation is that it inevitably results in damage and other artifacts to the prepared TEM sample [1] [2], which can diminish the quality of the TEM image [2] thereby making image interpretation and analysis challenging. For the semiconductor industry, FIB-assisted TEM sample preparation poses increasing challenges as transistor size continues to decrease according to Moore's law. With the technology node dwindling from tens of microns in the 1960's to tens of nanometers by 2010 and beyond [3], there is an increasing emphasis on preparing thinner specimen, faster, and with minimal damage and artifacts. Hence, the optimal set of FIB parameters (i.e. energy and angle of incidence) are needed to prepare high quality ultrathin TEM samples. FIB conditions which allow for the creation of high quality sub 25nm lamellae were identified using the Transport of Ions in Matter (TRIM) computer program [4], which is based on a Monte Carlo code that simulates ion-solid interactions [5]. Ion damage depth and curtaining effects from dissimilar milling rates were evaluated using TRIM outputs: mean projected range, Rp, and sputter yield, SY, respectively. Computer simulations were verified with cross-sectional TEM samples prepared with various FIB conditions. References: [1] Mayer, Joachim, Lucille A. Giannuzzi, Takeo Kamino, and Joseph Michael. "TEM Sample Preparation and FIB-Induced Damage." MRS Bulletin 32.05 (2007), p.400-07. [2] Kato, N. I. "Reducing Focused Ion Beam Damage to Transmission Electron Microscopy Samples." Journal of Electron Microscopy 53.5 (2004), p.451-58. [3] Xiao, Hong in "Introduction to Semiconductor Manufacturing Technology". 2nd ed. 2012 (SPIE, Bellingham) p.5-6. [4] Ziegler, J. F., The Stopping Range for Ions in Matter (SRIM) Software package, 2013. <http://www.srim.org/>. [5] Ziegler, J. F., J. P. Biersack, and Matthias D. Ziegler in "SRIM: The Stopping and Range of Ions in Matter". 2008. (SRIM Co, Chester). [6] Ghandhi, Sorab Khushro. "VLSI Fabrication Principles: Silicon and Gallium Arsenide". 1983 (Whiley, New York), p.372. Microsc. Microanal. 21 (Suppl 3), 2015 1990 Figure 1. Damage Depth Evaluation. TRIM simulaiton for a Ga ion beam bombarding a silicon (Si) target at 10keV with an ion angle of incidence of 60 degrees with respect to the target surface normal. The plots shown are: (1) the Ion Collision Plot (left) which depicts the individual ion paths projected onto the x-y plane of the target, and (2) the Ion Distribution Plot (right) which plots ion concentration information as a functon of depth into the solid. The statistical distribution of ions in an amorphous solid is shown to be gaussian in nature with the peak concentration occuring at the mean projected range, Rp. In crystalline materials however, the presence of channeling can result in deviations from gaussian profile; channeling results in deeper penetration of ions into the solid than predicted for amorphous solids [6]. (A) Sputter Yield, SY (atoms/ion) 20 15 10 5 0 0 10 SY vs. (10keV) (B) Sputter Yield, SY (atoms/ion) 6 5 4 3 2 1 0 0 10 SY vs. (1keV) 20 30 40 50 60 70 80 90 20 30 40 50 60 70 80 90 Angle of Incidence, (degrees) Silicon (10keV) Tungsten (10keV) Angle of Incidence, (degrees) Silicon (1keV) Tungsten (1keV) Figure 2. Curtaining Effect Evaluation. Sputter Yield, SY, versus Ion Angle of Incidence, (Ga ion beam) with respect to the surface normal of selected target materials at (A) 10keV and (B) 1keV. Numerical data plotted was obtained from TRIM.");sQ1[994]=new Array("../7337/1991.pdf","Utilization of FIB Technique in TEM Specimen Preparation of GaN-based Devices for Dislocation Investigation","","1991 doi:10.1017/S1431927615010739 Paper No. 0994 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Utilization of FIB Technique in TEM Specimen Preparation of GaN-based Devices for Dislocation Investigation Jian-Guo Zheng1, Zhenguang Shao2 and Dunjun Chen2 1 2 Irvine Materials Research Institute, University of California, Irvine, CA 92697-2800, USA Jiangsu Provincial Key Laboratory of Advanced Photonic and Electronic Materials, and School of Electronic Science and Engineering, Nanjing University, Nanjing 210093, P.R. China Dislocations can be frequently observed in GaN-based devices which have various applications in optoelectronics as well as high-power and high-frequency equipment. The properties of GaN-based devices can be affected by dislocations [1-4]. For example, thread dislocations may serve as a nonradiative recombination center and affect electron mobility. Different doped GaN layers in the devices are normally fabricated on a substrate, and dislocations (type and density) change in these layers. Therefore, it is necessary to study dislocations in GaN-based devices, especially in the active doped GaN layers, in order to understand device behavior and improve device performance. Transmission electron microscopy (TEM) is almost an exclusive technique to observe high density dislocations inside the devices. TEM specimen preparation is the first crucial step toward a successful TEM experiment. To correlate device performance and dislocation information, a TEM specimen has to come from a specifically selected device (Fig. 1) and to be cut in some specific directions. Obviously, focus ion beam (FIB) technique available in a dual-beam system has a great advantage over traditional TEM specimen preparation techniques. There still could be a lot of challenges to prepare a TEM specimen from a selected device by using FIB. Besides specific location, the TEM specimen needs to have a large thin area including all layers of devices ranging from the substrate to the top layer. For dislocation study, the TEM specimen also needs to be free of specimen bending in the large area of interest and be thinned down to electron transparency with similar thickness in the area. We have successfully applied FIB technique to prepare TEM specimens for dislocation study of GaN-based devices. Fig. 2 shows such an example, where the arrow indicates the interface between the GaN layers and sapphire substrate. The image intensity above the interface is quite uniform, indicating the GaN layers have similar thickness in the electron beam direction. Fig. 3 is a corresponding weak-beam TEM image showing clearly how threading dislocations vary from the interface to the top of the device. This presentation will summarize a few methods to prepare good TEM specimens for dislocation study by using FIB and to prevent some possible problems such as specimen bending (Fig. 4). [1] S. Nakamura, Science 281 (1998), p. 956-961 [2] S. J. Rosner et al, Applied Physics Letters 70 (1997), 420 [3] N. G. Weimann and L. F. Eastman, J. Appl. Phys. 83(1998), 3656 [4] D. P. Han et al, Jpn. J. Appl. Phys. 54 (2015), 02BA01 [5] TEM work was performed at the Irvine Materials Research Institute (IMRI) at UC Irvine, using FEI Quanta 3D dual-beam system for TEM specimen preparation which was funded in part by the National Science Foundation Center for Chemistry at the Space-Time Limit under grant no. CHE0802913 Microsc. Microanal. 21 (Suppl 3), 2015 1992 Doped GaN layers substrate 1 Fig. 1 An array of 7x3 GaN-based devices imaged by SEM. 2 Fig. 2 SEM image of a cross-sectional TEM specimen of one GaN-based device prepared by using FIB. The arrow indicates the interface between sapphire substrate and doped GaN layers. 3 Fig. 3 Cross-sectional weak-beam TEM image of a GaN-based device shown in Fig. 2. Threading dislocations may pass through different doped GaN layers or terminate at the layer boundaries. 4 Fig. 4 Cross-sectional weak-beam TEM image of a GaN-based device, where the bend contour is visible as a broad bright band, indicating the specimen is bent.");sQ1[995]=new Array("../7337/1993.pdf","Nano and Microscale Patterning on Soft Matters with Ion Beam Irradiation","","1993 doi:10.1017/S1431927615010740 Paper No. 0995 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Nano and Microscale Patterning on Soft Matters with Ion Beam Irradiation Yeonuk Kim1, Jiachun Huang1, Ala Yasin Abuelfilat1, Jing Fu1 1. Department of Mechanical and Aerospace Engineering, Monash University, Clayton, VIC 3800, Australia With high precision capability of Focused Ion Beam (FIB), we fine-tuned the surface topology and physical properties of thin film soft materials. The first example is hydrogel which is known for its biocompatibility, and both physical and chemical properties of hydrogels enable an environment preferred for cell growth and proliferation [1, 2]. Surface properties at micro/nano scale also have a significant role at the interface of polymer/substrate to regulate cellular behaviors [3], and developing novel and effective surface modification approaches will be crucial for future tissue engineering applications. In this study, thin film hydrogel was coated on a silicon wafer followed by ion beam (keV Ga+) irradiation [4]. Sputtering yield of hydrogel as well as roughness and moduli were measured prior and after irradiation using Secondary Electron Microscope (SEM) and Atomic Force Microscopy (AFM). Our results showed that nano and micro scale features can be patterned by FIB, with controllable feature size and physical attributes at the surface of hydrogel. The Young's modulus of irradiated hydrogel determined using AFM force spectroscopy was revealed to be dependent on ion fluence. Compared to the original Young's modulus value of 20 MPa, irradiation elevated the value to 250 MPa and 350 MPa at 1 pC �m-2 and 100 pC �m-2 respectively. Cell culture studies confirmed that the irradiated hydrogel samples were biocompatible, with stable nanoscale patterns under physiological conditions [4]. This patterning approach can also be extended to other soft samples such as tumour tissues (Figure 2a). Monte Carlo simulation was performed first to investigate the ion-tissue interactions, based on the chemical compositions from database and EDX mapping (Figure 2b), followed by deploying FIB as a digitally controlled slicing tool. Site specific probing of thin section tumour tissue could be achieved with AFM in various imaging modes (Figure 2c), with minimal effects from the microtome sectioning. References: [1] A. Gaston, et al, "Nanopatterned UV curable hydrogels for biomedical applications" Microelectronic engineering, vol. 87 (2010), pp. 1057-1061. [2] A. Al-Abboodi, et al, "Three-Dimensional nanocharacterization of porous hydrogel with ion and electron beams" Biotechnology and bioengineering, vol. 110, no. 1 (2013), pp.318-326. [3] F. Boccafoschi, et al, "Study of cellular adhesion by means of micropillar surface topologies" Advanced materials research, vol. 409 (2012) pp. 105-110. [4] Y. Kim, et al, "Tuning the surface properties of hydrogel at the nanoscale with focused ion irradiation" Soft Matter, vol. 10 (2014), pp. 8848-8456. [5] Funding for this research was partly provided through Australia Research Council Discovery Project Grant (DP120102570 and DP120100583) and Interdisciplinary Research Seed Fund of Monash University. This work was performed in part at the Melbourne Centre for Nanofabrication (MCN) in the Victorian Node of the Australian National Fabrication Facility (ANFF). The author would like to thank the staff from MCN and Anatomical Pathology, Alfred Hospital for various supports. Microsc. Microanal. 21 (Suppl 3), 2015 1994 Figure 1. AFM images (top view) of surface morphology after ion milling (Ga+, 30 kV) with ion fluence (a) 0.1 pC/�m2 and (b) 100 pC/�m2. (c) The corresponding numerical values of surface roughness of pristine hydrogel and after ion irradiation with increasing ion fluence. (Scale bar: 5 �m) (d) Modulus of hydrogel before and after the ion beam irradiation. Figure 2. (a) Optical image of the multilayer structures of tumour tissue, and (b) Monte Carlo simulation of ion-tissue interactions with target properties based on database and EDX readings. (c) Young's modulus obtained in different layers of the tumour tissue sample by AFM force spectroscopy.");sQ1[996]=new Array("../7337/1995.pdf","Xe Plasma FIB-SEM with Improved Resolution of Both Ion and Electron Columns","","1995 doi:10.1017/S1431927615010752 Paper No. 0996 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Xe Plasma FIB-SEM with Improved Resolution of Both Ion and Electron Columns J. Jiruse1, M. Havelka1, J. Polster1 and T. Hrnc�1 1 TESCAN Brno, s.r.o., Brno, Czech Republic Combined plasma Xe ion source FIB (Focused Ion Beam) with SEM (Scanning Electron Microscope) was introduced three years ago [1]. It proved to be an important tool for ultra-fast milling, especially for the semiconductor industry. Besides 50-times higher milling rate compared to traditional Ga ion FIB, it eliminates the conductive contamination of integrated circuits and it is useful for processing of compounds such as SiGe, (In)GaAs and (In, Al)GaN. We demonstrated its utilization also for other fields, for example preparation of large TEM lamella [2] or the first large-scale FIB tomography [3]. Here we present the latest advances of Xe plasma FIB leading to improvement of lateral resolution more than two times. This allows expanding Xe ion beam applications into the area of traditional Ga FIB applications. A special piezo manipulator called the rocking stage is designed to minimize curtaining effects. The SEM column is improved as well by replacing conventional magnetic optics with recently developed immersion magnetic optics yielding ultra-high resolution imaging at low beam energies. The detection system comprises 8 electron detectors both in the chamber and in the column. Detectors of backscattered electrons have increased detection limits and efficiency, thus able to detect in the complete range of electron energies 0.2-30 keV. This is especially useful for imaging semiconductor samples and integrated circuits just after preparation of their large cross-sections by Xe plasma FIB, see Figures 1 and 2. A wide range of integrated analytical instruments is kept including EDX, EBSD, TOF-SIMS, EBIC and even AFM reported previously [4]. Actually the TOF-SIMS analyser benefits from Xe primary ions by better detection limit compared to Ga primary ions. Raman analysis in FIB-SEM [5] is also enabled here thus providing ultra-high resolution electron beam, ultra-fast milling Xe ion beam and finally photon beam in one instrument. References: [1] T Hrnc� et al, Proceedings of 38th Int. Symp. for Testing and Failure Analysis ISTFA (2012), 26. [2] A Delobbe et al, Microsc. and Microanal. 20 (2014), 298. [3] T Hrnc� et al, Microsc. and Microanal. 19 Suppl 2 (2013), 860. [4] J Jiruse et al, Microsc. Microanal. 18 Suppl 2 (2012), 638. [5] J Jiruse et al, Journal of Vac. Sci. Technol. B 32 (2014), 06FC03. [6] The research leading to these results has received funding from the European Union 7th Framework Program [FP7/2007-2013] under grant agreement n�280566, project UnivSEM. Microsc. Microanal. 21 (Suppl 3), 2015 1996 Figure 1. SEM image taken by low-energy In-Beam BE detector with acceleration voltage 2 kV. Left: Detail on top edge of Al metal via, silicon layers and polysilicon filling. Right: Large-area cross section through mold compound and Al bonding wires. Figure 2. SEM image of the solder bump, cross-sectioned by plasma FIB, taken by low-energy InBeam BE detector with acceleration voltage 1 kV (detail) and 5 kV (overview). Left: Detail of the bump showing some cracks in intermetallic compound layer and delamination in repassivation layer. Sn - Pb island boundary is also clearly visible. Right: Large-area cross section of the solder bump. The electrical and mechanical connection between the die and the substrate is critical for the flip chip package functionality. Inside Sn bump material there is visible formation of Pb dendritic structure points (white islands) caused by e.g. too rapid cooling. Intermetallic compound (IMC), repassivation layer (polyimide), under bump metallurgy (UBM) and integrated circuit layers are clearly visible due to the high material contrast of In-Beam BE imaging.");sQ1[997]=new Array("../7337/1997.pdf","Superconducting Nano Wire Circuits Fabricated using a Focused Helium Beam","","1997 doi:10.1017/S1431927615010764 Paper No. 0997 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Superconducting Nano Wire Circuits Fabricated using a Focused Helium Beam E. Y. Cho1, Meng K. Ma1, Chuong Huynh2, R. C. Dynes1 and Shane A. Cybart1 1. Oxide Nano Electronics Laboratory, Department of Physics, University of California, San Diego, La Jolla, California USA. 2. Carl Zeiss Microscopy, LLC., Peabody, Massachusetts USA Electrical circuits fabricated from high-transition temperature superconductors (HTS) are very difficult to pattern due to the lack of a reliable etching process. Chemical etching can be used for large features, but undercutting limits the feature size to tens of microns. Dry etching is required for smaller features, but there is no anisotropic reactive ion etch for these materials. Therefore dry etching must be done with isotropic argon ion milling. The ion milling process generates excess heat and unfortunately oxide superconductors are sensitive to high temperature. Overheating the material by ion milling causes it to deoxygenate which turns the superconductor into an insulator. Therefore to prevent overheating the critical dimension for argon ion milling is typically limited to a few microns. In this work, we demonstrate an alternative method to pattern HTS by direct write ion lithography using a focused helium ion beam. In this method we demonstrate the ability to pattern nano wires as small as 250 nm within the plane of an HTS film. The key to this approach is that HTS materials are very sensitive to disorder, the electrical transport properties transition from superconductor to insulator with increasing disorder [1]. Irradiation of the superconductor material generates point defects creating a highly insulating region where the beam is scanned. The high energy ions shoot through the thin film and are implanted into the substrate without removing any material, this process is much faster than etching for dense oxides. Furthermore, because helium is inert it does not react chemically with the YBCO like gallium ions in conventional focused ion beams. In addition, the insulating barrier has a much smoother edge than a wire edge patterned by ion milling. For our experiment, test samples were prepared by patterning 4 m wires with standard photolithography and broad beam ion etching from 30-nm thick YBCO films grown on sapphire. We chose this thickness because Monte Carlo simulations using the Stopping and Range of Ions in Matter (SRIM) software show that 30 keV helium ions will completely penetrate the film and implant into the substrate [2]. This ensures a uniform disordered region throughout the superconducting film thickness. Nano wires were made by irradiating insulating barriers to narrow down the 4 �m wires as shown in figure 1. In order to precisely determine the wire width we added a Josephson junction into the center of the nano wire [3]. Measurement of the Josephson junction parameters, maximum super current (IC) and voltage state resistance (RN), allow us to accurately determine the wire width. To pattern the sample we first used a dose of 6�1016 He+/cm2 to write a Josephson junction in the circuit, and then the dose was increased to 2�1017 He+/cm2 to write the insulating barriers that define the nano wire. Two test samples were made with wire widths of 250 nm and 500 nm, and a third control sample without narrowing the wire. Current-voltage characteristics of the samples were measured in a vacuum cryostat inside of a liquid helium dewar at 4.6 K. Figure 2 shows the results for 250 nm, 500 nm and 4 �m wide wires. All of the junctions have an ICRN product of about 400 V as expected because the ICRN product should be a constant of the material. This implies that material properties in the wire remained the same and that Microsc. Microanal. 21 (Suppl 3), 2015 1998 proportionally with the width ( ) as it should. These results provide strong evidence that the current only flows through the nano filament as intended, and that we were successful in direct writing of nano wires. there was no thermal damage. Furthermore, RN are 70, 38 and 5.6 , which scale inversely 1 proportionally with the width . IC for the junctions are 5.6, 10.3 and 70 �A, which scale This new technology provides an improvement in patterning HTS and enables fabrication of high density nano scale superconducting circuits and interconnects. [1] Valles Jr, J. M., et al. "Ion-beam-induced metal-insulator transition in Y Ba 2 Cu 3 O 7-: A mobility edge." Physical Review B 39.16 (1989): 11599. [2] SRIM, the stopping and range of ions in matter (2008) by J. F. Ziegler, J. P. Biersack, Matthias D. Ziegler. [3] Cybart, Shane A., et al. "Nano Josephson Superconducting Tunnel Junctions in Y-Ba-Cu-O DirectPatterned with a Focused Helium Ion Beam." arXiv preprint arXiv:1409.4876 (2014). Figure 1. Schematic representation of a direct write nano wire. A 4 �m YBCO wire was narrowed down to 250 nm with insulating barriers (red lines). A Josephson junction was inserted in the middle of the wire with a lower dose (orange line). (a) (b) (c) Figure 2. Current-voltage characteristics of YBCO Josephson junctions with wire widths of (a) 250 nm, (b) 500 nm and (c) 4 m. The red lines are the measured data and the black dashed lines indicate extracted RN.");sQ1[998]=new Array("../7337/1999.pdf","Manipulations of Submicro-fibers of Culex Pipiens with the Help of Nano-tweezers with Shape Memory Effect into Vacuum hamber of FIB.","","1999 doi:10.1017/S1431927615010776 Paper No. 0998 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Manipulations of Submicro-fibers of Culex Pipiens with the Help of Nano-tweezers with Shape Memory Effect into Vacuum hamber of FIB. Alexander Kamantsev1, Alexey Mashirov1, Pavel Mazaev1, Victor Koledov1, Vladimir Shavrov1, Vladimir Dikan2, Artemy Irzhak2, Alexander Shelyakov3 1. 2. Kotelnikov Institute of Radio-engineering and Electronics of RAS, Moscow, Russia. National University of Science and Technology "MISIS", Moscow, Russia. 3. National Research Nuclear University "MEPhI", Moscow, Russia. In recent years, a record small mechanical tools based on composites with shape memory effect (SME) were created [1-4]. Application of the technology of selective ion etching allowed for the creation of two-layer composite actuators and tools based on rapidly quenched nonmagnetic alloys with SME, such as Ti2NiCu [1]. These composite actuators can change their shape reversely and produce mechanical work using only "one-way" SME of the alloy [2]. Currently in the field of manipulation and manufacturing at the nanoscale, there is an urgent need to develop new functional materials in order to fill the gap between the dimensions of modern MEMS and real size of nano-objects to be manipulated. Recently the operation of nano-tweezers using layered composites based on alloy with SME driven by thermal actuation has been demonstrated [1-3]. The operation of the Ti2NiCu/Pt composite actuator driven by heating was demonstrated, with an overall volume of the actuator of less than 1 �m3 and thickness of active layer of the T2NiCu alloy being down to 70 nm [1]. These achievements of nanotechnology can be used certainly in various branches of biology. Developed technique allow to hold, move and manipulate of animate or inanimate objects of submicron and nanometer sizes, for example, carbon nanotubes, graphene sheets, bacteria, viruses, biological particles of different nature. The purpose of this paper is description of experiments on preparation of insect's fibers of submicron sizes into a vacuum chamber of FIB. The object of the study served as Culex pipiens. This species is distributed universally and has a large epidemic importance. The experiment consisted in the selection of a part of fiber, its separation from the insect's body (Fig. 1), displacement and attaching it on a copper grid (Fig. 2) for further TEM studies. The experiment was done into vacuum chamber of FEI Strata 201 FIB device with the help of composite nano-tweezers with SME. For positioning in space the nano-tweezers was attached to needle of Omniprobe micromanipulator (by CVD process). The whole process of manipulations and the characteristic scales are presented on Fig. 1 and Fig. 2. We are demonstrating on example of manipulations of insect's fiber that such microinstruments might help to resolve some unresolved problems of medicine and biotechnology, particularly in single cell manipulation, microsurgery, drug delivery systems, etc. [1] D. Zakharov, G. Lebedev, A. Irzhak, et al., Smart Matr. Struct. 21, (2012), p. 052001. [2] A. Irzhak, V. Kalashnikov, V. Koledov, et al., Techn. Phys. Lett. 36, (2010), p. 329. [3] A.V. Shelyakov, N.N. Sitnikov, A.P. Menushenkov et al., Thin. Sol. Films 519, (2011), p. 5314. [4] E. Kalimullina, A. Kamantsev, V. Koledov, et al., Phys. Stat. Solidi (C) 11, 5-6, (2014), p. 1023. [5] The authors acknowledge funding from the Russian Sciences Foundation, Grant 14-19-01644. Microsc. Microanal. 21 (Suppl 3), 2015 2000 Figure 1. Feeding of nano-tweezers and selection of a part of fiber on body of Culex pipiens. Figure 2. Separation of part of fiber from Culex pipiens body, displacement and attaching it on copper grid.");sQ1[999]=new Array("../7337/2001.pdf","Advantages of using Plasma FIB Over a Gallium LMIS Source","","2001 doi:10.1017/S1431927615010788 Paper No. 0999 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Advantages of using Plasma FIB Over a Gallium LMIS Source B.W. Arey1, D.E. Perera1, L. Kovarik1, J. Liu1, O. Qafoku2, A. R. Felmy2, R. Kelley3, T. Landin3, R. Alvis3 1 2 3 Pacific Northwest National Laboratory, Environmental Molecular Science Laboratory, 3335 Innovation Boulevard, Richland, WA 99354, USA Pacific Northwest National Laboratory, 902 Battelle Boulevard, Richland, WA 99354, USA FEI Co., 5350 NE Dawson Creek Drive, Hillsboro, Or 97124, USA For over 20 years, gallium liquid metal ion source (LMIS) technology has been the workhorse charged particle beam for the focused ion beam (FIB) industry. In recent years the gas field ion source (GFIS) and plasma field ion source (PFIB) are gaining in popularity as an alternative to LMIS. One of the driving forces is the gallium implantation into the material and the artifacts one must deal with. In our experiments we are exploring the interaction with the gallium source and xenon plasma source has on the CO2 reaction on olivine minerals. Fayalite, the iron-rich end member of the olivine series (Mg,Fe)2SiO4, occurs in abundance in the Earth's upper mantle and it can be a significant constituent of basaltic rocks. The isolated tetrahedral structure of fayalite can breakdown during reaction with aqueous solutions and it is known that the dissolution of fayalite is accelerated if high temperatures, acidic solutions and far from equilibrium conditions are present. During fayalite dissolution its constituents (Fe2+ and silica) will be released from the structure and can participate in precipitation reactions with nearby anions to form secondary minerals: (hematite, goethite, sulfites, carbonates, amorphous silica). In this study we will present the new methods for determining the spatial distribution of the light elements (Na, Li,and Ca) within the mineral fayalite. These elements may play a role in carbonation reactions of fayalite under super critical CO2 conditions. High Resolution Scanning Transmission Electron Microscopy (STEM), a highly-focused electron probe is raster-scanned across the material, and various types of scattering are collected as a function of position. The transmitted electrons at high scattering angle can be collected to form high-resolution, chemically sensitive, atomic number (Z-) contrast images. The x-rays generated can be collected using an energy-dispersive X-ray spectroscopy (EDS) detector and used to form high spatial resolution compositional maps. Electron energy losses can be detected using a Gatan image filter (GIF) to map the compositional and electronic properties of materials. Atom Probe Spectroscopy (APT), is a state-of-the-art analytical method that allows for three-dimensional elemental mapping with near atomic resolution. It is a powerful tool that can provide unique compositional information on a variety of materials systems such as complex alloys, semiconductor materials and devices, and ceramic material. Microsc. Microanal. 21 (Suppl 3), 2015 2002 This work was supported by the U. S. Department of Energy (DOE), Office of Basic Energy Sciences through a Single Investigator Small Group Research (SISGR) grant at Pacific Northwest National Laboratory (PNNL). This research was performed using EMSL, a national scientific user facility sponsored by the Department of Energy's Office of Biological and Environmental Research located at Pacific Northwest National Laboratory. PNNL is operated for DOE by Battelle Memorial Institute under Contract# DE-AC0576RL0-1830. Figure 1: Ga APT after reaction with CO2: Figure 2: Xe APT after reaction with CO2");sQ1[1000]=new Array("../7337/2003.pdf","Large volume 3D characterization by plasma FIB DualBeam microscopy","","2003 doi:10.1017/S143192761501079X Paper No. 1000 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Large volume 3D characterization by plasma FIB DualBeam microscopy T L Burnett1, 2, R Kelley3, B Winiarski1, 2, M Daly1, K Mani3, P J Withers1 1. 2. The University of Manchester, Materials Department, Manchester, M13 9PL, UK FEI Company, Achtseweg Noord 5, Bldg 5651 GG, Eindhoven, The Netherlands 3. FEI Company, 5350 Northeast Dawson Creek Drive, Hillsboro, OR 97124, USA There is a critical need to analyze many material systems in three dimensions (3D), for example to understand connectivity of phases, porous networks and complex shapes. There are several tools available for 3D characterization e.g. X-ray computed Tomography (CT) [1], Serial Sectioning SEM Tomography (SST) [2, 3], transmission electron microscopy [4] and Atom Probe [5]. Figure 1 demonstrates their complementarity in terms of the lengthscales they can access. The emergence of dual beam Focused ion beam (FIB-SEM) using Gallium ions has provided a means of accessing regions of interest of 50 m dimensions both for site specific TEM and SST in the SEM with slice thicknesses down to 10 nm. Although destructive it has enables 3D imaging of structure, grain structure (via EBSD) and chemistry (by EDS). However the limited volumes accessible through the Gallium Ion FIB present a severe limitation. The size of the volume that can be examined via SST and the depth from which the regions of interest identified by X-ray CT is also key to opening up possibilities through a correlative tomography framework [1]. There are many scenarios in materials science that require analysis across length-scales up to many 100's of microns that lie beyond what a Ga ion FIB can typically access, relating to grain microtextures, grain neighborhoods, grain boundaries, inclusions and cracks. In this paper we present the first examination of the capabilities of a new type of FIB-SEM that utilizes a Xe Plasma beam FIB (PFIB) for machining alongside an electron imaging column. We show that the FEI Helios PFIB can achieve a throughput 20-50x that of a Ga ion FIB opening up regions 100s of microns beneath the surface and allowing serial section tomography over volumes of hundreds of micron dimensions. The performance of the PFIB is demonstrated through the study of an austenitic and a ferritic steel both used within the energy sector. The PFIB was operated at 30kV and 59 nA beam current to collect typically 100's of slices ~100 � 100 �m2 cross section size and a slice thickness of 80 nm. In contrast to typical SST within a FIB a stage rocking mode (+/- 5) is sued to minimize curtaining that may be created during unidirectional FIB milling. Images using the Through lens detector (TLD) were captures at high resolution with the pixel size < 40 nm. These SEM-PFIB settings allowed us to collect 10-20 slices per hour (Fig. 2). The study here focuses on a ferritic steel used in pressure vessels with the region analyzed e just below a ductile fracture surface (Fig. 3). The distribution of carbides, as well as the voids associated with the ductile fracture, were imaged and analyzed. Our results indicate that curtain-free surfaces with a high detail level of microstructural features, e.g. carbides at the sized at ~100 nm. Results shows that the PFIB SST technique produces low levels of surface damage as evidenced by good channeling contrast. We will also present results showing Kikuchi patterns from the machined surface. Other ongoing experimental studies across a wide range of materials from paint to ceramics suggest the PFIB can routinely and quickly provide 3D serial section tomographs over dimensions of many hundreds of microns helping bridge the gap between conventional FIB and X-ray tomography. All results were aligned and analyzed using FEI-Avizo Visualization software. Microsc. Microanal. 21 (Suppl 3), 2015 2004 1. E. Maire and P. J. Withers: `Quantitative X-ray tomography', Int. Mater. Rev., 2014, 59, 1, 1�43. 2. M. D. Uchic, L. Holzer, B. J. Inkson, E. L. Principe, P. Munroe 'Three-Dimensional Microstructural Characterization Using Focused Ion Beam Tomography' MRS Bulletin, Volume 32, Issue 05, May 2007, pp 408-416 3. Borgh, I., et al., On the three-dimensional structure of WC grains in cemented carbides. Acta Mat., 2013. 61: 4726-4733. 4. C. K�bel, A. Voigt, R. Schoenmakers, M. Otten, D. Su, T-C. Lee, A. Carlsson, J. Bradley `Recent Advances in Electron Tomography: TEM and HAADF-STEM Tomography for Materials Science and Semiconductor Applications' Microscopy and Microanalysis, 11, Issue 05, 2005, pp 378-400 5. Blavettea, D., Duguay, S., Atom probe tomography in nanoelectronics. European Physical Journal - Applied Physics, 2014. 68(1): p. 10101-p1-12. 6. Burnett, T.L., McDonald, S.A., Gholinia, A., Geurts, R., Janus, M., Slater, T., Haigh, S.J., Ornek, C., Almuaili, F., Engelberg, D.L., Thompson, G.E., Withers, P.J., Correlative Tomography. Scientific Reports, 2014. 4: p. 4711 Figure 1. Comparison of Plasma focused ion beam (PFIB) tomography with other 3D microscopy methods. Figure 2. Avizo visualisation of 440 slices of an austenitic stainless steel sectioned with FEI Helios plasma FIB DualBeam at 59 nA Plasma current and 30kV electric field. The slice thickness is 100 nm and the sections face is 100 � 60 �m 2 with the pixel size of 38 nm. Combined PFIB milling and SEM imaging of a single slice is about 6 minutes. FEGSEM imaging at 2kV, 0.68nA, ETD detector, field-free mode. The inset shows a Kikuich pattern from the point. 4 �m 20 �m Figure 3. Ferritic steel from a nuclear pressure vessel sectioned with FEI Helios plasma FIB DualBeam at 59 nA Plasma current and 30kV. The slice thickness is 100 nm and the sections face is 100 �96 �m2 with the pixel size of 29 nm. Combined PFIB milling and SEM imaging of a single slice is about 8 minutes. FEGSEM imaging at 2kV, 0.80nA, TLD detector and field-free mode.");sQ1[1001]=new Array("../7337/2005.pdf","Bulge Testing for Strength Metrics of Detector X-Ray Windows","","2005 doi:10.1017/S1431927615010806 Paper No. 1001 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Bulge Testing for Strength Metrics of Detector X-Ray Windows Joseph Rowley1, Kendall Berry1, Robert Davis1, Richard R. Vanfleet1, Sterling Cornaby2, Mallorie Harker2, Richard Creighton2 1. 2. Brigham Young University, Dept. of Physics and Astronomy, Provo, USA Moxtek Inc., Orem, USA Bulge testing is a technique to measure thin film mechanical properties. First used by Beams[1] to test the strength of silver and gold foils, it has since been improved upon to more accurately model films by comparison with Finite Element Analysis(FEA)[2], and utilize different geometries to measure additional properties[3]. Elastic behavior can be studied, as well as plastic deformations[4]. The basic idea is applying pressure to one side of a film, then measuring the pressure and displacement. The resulting pressure vs. displacement data is then fitted to a cubic function, the coefficients of which will determine a materials Young's modulus and Poisson's ratio. X-Ray windows are thin solid films designed to withstand a pressure differential while remaining leak tight. Bulge testing is an excellent way to directly test windows mechanical properties. These results are of great benefit to x-ray window as well as x-ray detector design. For example, knowing the maximum deflection of a film under different pressures will let a detector be placed as close to a film as possible while maintaining the integrity of the detector. There are two basic x-ray windows designs. The first is a single suspended beryllium metal film, sometimes with an additional corrosion resistant protective coating. The other is a stack of thin films, incorporating a polymer film, on top of a support grid. The first type of film is well described by the bulge equation, while the other is not. However, testing can be done directly on the stack of thin films to determine strength of the film without the support structure. The information generated can aid in the design and assist a FEA model's accuracy. Fig. 1 is a schematic of the setup used for bulge testing. A Keyence LK-G5001P controller and LK-H052 laser displacement sensor are used to measure the displacement of the bulging film. A Newport 426 stage and Newport Newstep NSA12 are used to position and scan the sensor over the film. An electronic pressure regulator along with a pressure transducer control and measure the pressure applied to the film. An MFC measures the flow of air to the film. The laser displacement sensor is based on diffuse reflections. If a sample is too transparent, or has a mirror surface, the sensor does not operate. In these cases colloidal silver is applied to the sample, allowing diffuse reflections without altering the film and keeping the film displacement relatively unchanged. The ability to measure transparent and reflective films is very important for the success of this work. Fig. 2 is a graph of the bulge across the diameter of a circular film. This candidate window was 3.5 mm in diameter. From this graph it can be seen that the displaced film has a circular cross section. From several scans of the radius and different pressures, the strength of the film can be obtained Fig. 3 is a table displaying the displacement data at the center of the Moxtek's AP 5 small window at 1 atm. Microsc. Microanal. 21 (Suppl 3), 2015 2006 References: [1] J. W. Beams, "Mechanical Properties of Thin Films of Gold and Silver" (1959), p. 183 [2] Martha K. Small and W.D. Nix, Journal of Materials Research, 7 (1992), pp. 1553-1563. [3] J.J. Vlassak and W.D. Nix Journal of Materials Research, 7 (1992), pp 3242-3249. [4] Y. Xiang, X. Chen and J.J. Vlassak, Journal of Materials Research 20 (2005), pp 2360-2370. Figure 1. Schematic of Bulge Test Setup 70 60 Bulge Height(�m) 50 40 30 20 10 0 6000 6500 7000 7500 8000 8500 Horizontal Position(�m) 9000 9500 Data Circle Fit Figure 2. Bulge of X-ray window. Height of the bulge along the diameter of the film. AP5 small cycle Displacement at 0 atm Displacement at 1 atm number in micrometers in micrometers Cycle 1 0 71.885 Cycle 2 .187 71.914 Cycle 3 -.059 71.884 Figure 3. Table of Moxtek's AP5 small displacements at 1 atm");sQ1[1002]=new Array("../7337/2007.pdf","Elemental and Phase Analysis of the Stomatopod Dactyl Club by X-Ray Mapping","","2007 doi:10.1017/S1431927615010818 Paper No. 1002 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Elemental and Phase Analysis of the Stomatopod Dactyl Club by X-Ray Mapping Nicholas A. Yaraghi1, Lessa Grunenfelder2,3, Nobphadon Suksangpanya4, Nicolas Guarin4, Steven Herrera1, Garrett Milliron5, Pablo Zavattieri4, Leigh Sheppard6, Richard Wuhrer6, David Kisailus1,3 Materials Science & Engineering, University of California, Riverside, Riverside, California, USA Chemical Engineering & Materials Science, University of Southern California, Los Angeles, California, USA 3. Chemical and Environmental Engineering, University of California, Riverside, Riverside, California, USA 4. School of Civil Engineering, Purdue University, West Lafayette, Indiana, USA 5. Max Planck Institute for Colloids and Interfaces, Potsdam, Germany 6. Advanced Materials Characterization Facility, University of Western Sydney, Australia 2. 1. Nature creates multifunctional composite materials by assembling organic and inorganic constituents into complex hierarchical structures [1]. The ability to precisely control the phase, morphology, and local distribution of components has resulted in a number of unique biological materials that exhibit a wide range of mechanical properties. One such structure is the raptorial appendage (dactyl club) of Odontodactylus scyllarus, a species of stomatopod (marine crustacean). The dactyl club is a multiphase bio-composite material that exhibits exceptional damage tolerance from high energy loading events. Here we examine and quantify the local distribution of elements within the dactyl club by means of XRM and quantitative EDS. We identify distinct phases, which are then correlated with ultrastructural and mechanical analyses to derive structure-function relationships as well as provide insight into the formation of an impact-resistant natural material. Fresh dactyl clubs were obtained from both live and recently deceased specimens of O. scyllarus, which were maintained in an in-house sea water system. The clubs were first dissected and rinsed in DI water to remove residual salt. For XRM and EDS analyses, samples were embedded in epoxy (System 2000, Fibreglast, USA), sectioned using a low speed saw with diamond blade, and polished with progressively finer silicon carbide and diamond abrasive down to 50 nm grit. Polished sections were then mounted on carbon tape-coated aluminum stubs, carbon coated, and analyzed using a JEOL840 SEM operating at 20 kV. X-ray maps were post-processed using the "Chemical Imaging" software package within the Moran Scientific Microanalysis System [2]. Figure 1 provides an overview of the dactyl club structure as well as optical and electron micrographs of a polished transverse section. Backscatter electron microscopy (BSE) in Figure 1D highlights the heavily mineralized exocuticle region, which as identified by Weaver et al., is composed of oriented hydroxyapatite [3]. Figure 1E shows the results of an x-ray map collected at 20x magnification. The false-colored map reveals the local distribution of calcium, phosphorus, and magnesium. The map shows that the exocuticle region is highly concentrated in calcium and phosphorus, which corresponds to the crystalline calcium phosphate mineral phase present. Subsequently, the endocuticle region shows high concentrations of magnesium, which plays a role in stabilizing the amorphous calcium phosphate and calcium carbonate phases present [3]. Interestingly, there is a higher concentration of calcium located along the lateral sides of the endocuticle, indicating a more heavily mineralized sub-domain. This correlates well with the contrast observed in Figure 1D. The x-ray map also reveals a narrow phosphorus-rich region located at the interface between the endo- and exocuticle regions. This veneer Microsc. Microanal. 21 (Suppl 3), 2015 2008 may be important for controlling the transition from amorphous to crystalline calcium phosphate at the outer surface of the club. A B C D E Figure 1. A) Anterior of O. scyllarus highlighting the dactyl club, which is boxed in white; B) Optical micrograph showing the raptorial appendage and dactyl segment; C) Optical micrograph of a polished transverse section of the club denoted by the dashed line in (B); D) Backscattered electron micrograph showing heavily mineralized outer surface region; E) False-colored x-ray map showing distributions of magnesium, calcium, and phosphorus throughout the club. References: [1] P. Fratzl, R. Weinkamer, Progress in Materials Science 52, 1263-1334 (2007) [2] K. Moran, R. Wuhrer, Microchimica Acta 155, 209-217 (2006) [3] Weaver et al., Science 336, 1275-1280 (2012)");sQ1[1003]=new Array("../7337/2009.pdf","EPMA Analysis of Corroded Hot-Dip Galvanized Carbon Steel using the "Phase Map Maker" Software","","2009 doi:10.1017/S143192761501082X Paper No. 1003 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 EPMA Analysis of Corroded Hot-Dip Galvanized Carbon Steel using the "Phase Map Maker" Software Ryota Kamiyama1, Peter McSwiggen3, Naoki Kato1, Shigeru Honda1, Norihisa Mori1 and Charles Nielsen2 1. 2. JEOL Ltd., 1-2 Musashino, 3 Akishima, Tokyo 196-8558 Japan JEOL USA, Inc., 11 Dearborn Road Peabody MA 01960 USA 3. McSwiggen & Associates, 2855 Anthony Lane, St. Anthony, MN 55418 USA Scatter diagrams are a very useful technique for structure analysis of materials using EPMA map data. We developed the "Phase Map Maker", an automatic scatter diagram analysis program for the JEOL JXA-8230 / 8530F series EPMA. The "Phase Map Maker" was used in the analysis of hot-dip galvanized carbon steel. Hot-dip galvanizing is a widely used anti-corrosion technique. There are two beneficial anti-corrosion characteristics of this galvanization process. One is that of a protective coating, in which the zinc layer provides corrosion resistance. Another is that of sacrificial protection, in which the zinc provides a galvanic protection of the iron when iron substrate is exposed locally. A test piece of hot-dip galvanized steel was treated with a salt spray test. This treatment, according to JIS Z 2371, is a neutral 5% aqueous sodium chloride solution sprayed 1008 hours at 35�C. The test pieces from before and after the salt spray test were cut and made into polished cross sections. These samples were analysed and confirmed to have a layer (Zn), a layer (FeZn13) and a 1 layer (FeZn7), which is the general structure of hot-dip galvanized steel [1]. A scatter diagram analysis was applied to element maps from the two test pieces. Figure 1 shows the analysed result of a sample before salt splay. The results show that the plating layer is divided into two sub-layers, a zinc phase ( layer) and intermetallic compound phase. Quantitative analyses indicate that the intermetallic compound phase is the layer. Furthermore, the 1 layer was confirmed from line analyses focusing on the interface of the layer and substrate. These results show that most of the intermetallic compound is the layer, while the 1 layer occurs at the interface with the substrate. Figure 2 shows the result for the test piece exposed to salt splay. The plating layer has separated into two phases similar to the galvanized test piece, except the top layer of the salt-sprayed sample is an oxide layer (Fig. 2a and 2b). A chemical bonding state analysis was done on this top layer. The Zn-L spectrum of the top layer is consistent with that of the ZnO standard (Fig. 2c). In addition, quantitative analyses of the layer produced atomic concentrations of Zn and O of nearly 1: 1. These results are both consistent with a surface phase of ZnO. There was no clear difference between the intermetallic compound layer of the salt sprayed, treated sample and that of the untreated sample. On the other hand, it is worth to note that the layer found in the untreated sample disappeared in the treated sample, or the layer has oxidized in the treated sample. The scatter diagram analysis using "Phase Map Maker" has been found to be useful in the analysis of Microsc. Microanal. 21 (Suppl 3), 2015 2010 intermetallic compounds in hot-dip galvanized steel. References: [1] Y Wakamatsu, H Masumoto, M Yamane, M Onishi and T Shimozaki, Tetsu-to-Hagane Vol. 82 (1996) No. 1 P 75 a) b) Figure 1. Element maps (a) and phase map obtained by scatter diagram analysis (b) of an untreated galvanized steel sample. a) b) c) Figure 2. Element maps (a), phase map obtained by scatter diagram analysis (b) and chemical bonding state analysis result (c) of galvanized steel after being treated with a salt spray.");sQ1[1004]=new Array("../7337/2011.pdf","Utilizing Intensity Ratios To Characterize Glass Fragments Via Micro-X-ray Fluorescence","","2011 doi:10.1017/S1431927615010831 Paper No. 1004 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Utilizing Intensity Ratios To Characterize Glass Fragments Via Micro-X-ray Fluorescence Andrew H. Lee1 1. EDAX Inc., Mahwah, NJ USA Micro-X-ray fluorescence (micro-XRF) is a non-destructive elemental technique which utilizes a focused X-ray source to generate characteristic radiation from the sample. In "bulk" samples, when the X-ray source cannot penetrate through the sample, the count-rates (CPS) for each element is saturated. Weight fractions can then be calculated by employing a Fundamental Parameters calibration model. However, if the sample is not thick enough to absorb the primary X-rays (i.e. "thin" sample), then the count-rates will be influenced by the sample thickness. This effects the calculation of weight fractions, as the Fundamental Parameters model assumes samples are infinitely thick. Figure 1 shows nominal Xray probe depths at various energies and samples matrices. These probe depths are crucial in determining if a sample can be considered "bulk" or "thin" relative to the X-ray energy. In many microXRF applications, comparing and characterizing small fragments (i.e. determining whether they are the same or different) is required. However, the samples may have no consistent thickness or size, and generally must be analyzed "as is". Therefore, a method is necessary to be able to compare samples while compensating for the influence of sample thickness. This study will look at an alternative intensity-based methodology to characterize samples, using measurements from soda-lime glass fragments to illustrate the improved stability. In this method, weight fractions are ignored and instead, intensity ratios are used. Net intensities are calculated for each element, and then ratioed to the intensity of another element within the same sample. This is referred to as internal elemental ratios. It is based on the principle that while thickness will effect individual elemental intensities, the elements within that sample retain their relative ratios to each other (assuming consistent acquisition conditions). Figure 2 shows two graphs of Ca(K) intensities measured in various soda-lime glass standards with varying thicknesses and varying levels of calcium. Normally for bulk samples, the correlation between Ca(K) intensities and their weight fractions would have been relatively linear. However, since these samples were inconsistently thin relative to the incoming X-rays, the correlation between net intensities against given concentrations is relatively unstable. The data points do not create a well-fit regression line (R2=0.045) due to the influence of sample thickness. However, once Ca(K) intensities are normalized against the Si(K) intensity from the same sample, the correlation between the intensity ratios and their concentrations is significantly improved (R2=0.8877). This is critical in applications where glass characterization needs to be performed at a high degree of accuracy on thin samples, where unabsorbed X-rays can still penetrate through. By compensating for the influence of sample thickness, internal intensity ratios can provide improved correlation with elemental concentrations, while still allowing the sample to be analyzed non-destructively, with minimal preparation. Microsc. Microanal. 21 (Suppl 3), 2015 2012 Figure 1. Nominal probe depths of X-rays in various sampling matrices. If the sample thickness does not exceed the analysis depth at a given energy, it is considered transparent to X-rays (i.e. thin sample). Calcium (Ca) "raw absolute-data" plot 1500 1000 Calcium (Ca) "raw ratio-data" plot 1000�(analyte/Si) [intensity ratio] measured analyte intensity (c/s) 900 800 700 600 R2 = 0.8877 500 400 1250 1000 750 R = 0.0445 500 6.00 2 7.00 8.00 9.00 10.00 11.00 12.00 8 10 12 14 16 given % oxide 100�(analyte/Si) [oxide concentration ratio] Figure 2. Ca(K) intensities compared, the first graph (L) shows raw net intensities compared to their reported weight fractions (%). The influence of varying sample thickness creates a poorly fit regression line. Once the Ca(K) intensities were ratioed against Si(K) and re-plotted, their correlation to their reported weight fraction was substantially improved.");sQ1[1005]=new Array("../7337/2013.pdf","EDX Analysis of Low Concentration Dopant using HD-2700 Aberration Corrected STEM Equipped with Dual SDD","","2013 doi:10.1017/S1431927615010843 Paper No. 1005 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 EDX Analysis of Low Concentration Dopant using HD-2700 Aberration Corrected STEM Equipped with Dual SDD Takahiro Sato1, Yuya Suzuki1, Hiroaki Matsumoto1, Takahito Hashimoto1, Kuniyasu Nakamura1, Taiga Miura2, and Hisaharu Yoshida2 1. Science Systems Design Div., Hitachi High-Technologies Corporation, 1040 Ichige, Hitachinaka, Ibaraki 312-0033, Japan 2. Vacuum Device Inc., 1285-5 Iijima-cho, Mito, Ibaraki 311-4155, Japan A STEM instrument model HD-2700 equipped with two detectors (X-MaxN100TLE Silicon Drift Detector (SDD)) was useful for high special analysis of objects [1], in which each SDD has a 100 mm2 sensor and windowless configuration. High sensitivity and high through-put analysis with atomic resolution can be expected from this system. In the field of material sciences, especially next generation semiconductor, the analysis demand of lower concentration dopant elements such as arsenic, phosphorus and boron in silicon device has been increased. Generally, the detection of boron by EDX was not so successful because of the lower detection efficiency of characteristic X-ray for light element. The signal of electron energy loss spectroscopy (EELS) for light element is better than heavy element, so the EELS has been used to detect the boron [2]. In this study, we examined the sample preparation and analysis conditions using large solid angle dual SDD. A boron doped silicon wafer was used as an experimental specimen. Herein, the density of the irradiation boron is 3.89 x 1014 atom/cm2. Specimens were prepared by an NB5000 FIB-SEM with the Hitachi in-situ micro-sampling technique [3]. After fabrication at 40 kV, final thinning was performed at 5 kV, and a 100 nm thick sample was prepared. Subsequently, the sample was onto a double tilt analytical holder. The conditions for obtaining the EDX map were as follows: acceleration voltage is 200 kV, field of view is about 30 m x 25 m, pixel size is 128 x 96, acquisition time is about 30 minutes. Figure 1 shows a result of the EDX analysis. (a) is a BF-STEM image of the analysis area, Areas 1 and 2, in which amorphous layer in Area1 was observed on the top surface of silicon [110] single crystal in Area 2. (b) is superimposed EDX spectra of each area, in which boron peak were recognized in Area 1 spectrum (blue) and in Area 2 boron was not detected about 15 nm depth area (red). Figure 2 shows the elemental map (a) and its ADF-STEM image (b). Herein, two dimensional distribution of boron was successfully visualized. Figure 3 shows the line profile of B, O and Si from surface to 15 nm area extracted from the mapping results using spectrum imaging. Boron was mainly exists in range from surface to about 5 nm depth area. These results indicate that the EDX analysis using high sensitive dual SDD system is effective to the detection of low concentration dopant element. And the cold filed emission gun was also useful for the detection of lower concentrated materials due to the reduction of scattering of incidental electron beams. References [1] T. Hashimoto et al., Microsc. Microanal. 20, 2014, p.604-605. [2] K. Asayama et al., Applied Physics Express 1 (2008) 074001 [3]T. Ohnishi et al., Proc. 25th International Symposium for Testing and Failure Analysis, 1999, pp. 449453 Microsc. Microanal. 21 (Suppl 3), 2015 2014 (a) Area1 (b) Area2 Figure 1. Result of EDX analysis (a) BF-STEM image of the analysis area, (b) EDX spectra of each area V.acc., : 200 kV, Magnification : 4,000,000 (a) B-K (b) BF-STEM Born doped area 5 nm 5 nm Figure 2. EDX mapping result (a) B-K map, (b) BF-STEM image V.acc., : 200 kV, Magnification : 4,000,000 B-K 5 10 15 Figure 3. The line profile of B, O, and Si extracted from the mapping results using spectrum imaging.");sQ1[1006]=new Array("../7337/2015.pdf","Quantitative ED(X)S: The Zeta-Factor Method","","2015 2015 doi:10.1017/S1431927615010855 doi:10.1017/S1431927615010855 Paper No. 1006 Microsc. Microanal. 21 (Suppl 3), 2015 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 � Microscopy Society of America 2015 Quantitative ED(X)S: The Zeta-Factor Method Meiken Falke1, Andi K�ppel1, Igor Nemeth1 and Ralf Terborg1 1 Bruker Nano GmbH, Am Studio 2D, 12489 Berlin, Germany Energy dispersive X-ray spectroscopy is by now well-established for qualitative and quantitative composition analysis of electron transparent samples in SEM, TEM and STEM and is reaching from the mm to the atomic scale. Single mobile atomic impurities in non-ideal real life samples can be identified within seconds using high-end STEM instrumentation [1-2], which is a remarkable step towards analyzing the role of single atoms in various materials science problems, e.g. impurities on nanoparticle surfaces in catalysis. Although single atom spectroscopy is feasible now, the correct quantitative composition analysis of larger element mixtures, particularly mixtures of heavy and light elements still is a problem worth consideration. We report on the implementation of the -factor method [2-4] as an absolute EDS quantification method for electron transparent samples and opposed to the widely used relative CliffLorimer method. The latter can provide quantitative data on the accuracy level of a few at% and if using large solid and take-off angles even ppm already. The quantitative results from the Cliff-Lorimer method are either based on theoretical Cliff-Lorimer factors, calculated from cross-section and fluorescence yield data for particular elements at specific electron beam energies, or are only valid relative to a standard. An alternative absolute quantification procedure, the -factor method, has been developed by M. Watanabe. The method relies on having more information than in case of the relative Cliff-Lorimer approach. The electron dose must be known for all measurements and additionally, for the standard measurement, the standard thickness and density must be known. The approach then allows the quantification of sample compositions while accounting for absorption and fluorescence effects and determining e.g. the thickness for an unknown sample of interest. A further advantage is that from the factors obtained with one standard measurement the -factors for all the other elements can be obtained from the, afore mentioned, theoretically calculated Cliff-Lorimer factors. The -factor method has been implemented into the Bruker ESPRIT 2.0 software and tested using various standards such as Al- and Ti oxides, GaP [5] and Si3N4. For an initial test procedure a 30 nm Si3N4 foil (commercially available from Agar) was used as a standard and the same material with 60nm thickness was used as a test specimen. For this the foil was damaged by the electron beam to produce sample areas with folds of known thickness and composition. The spectrum from one of these areas (Fig 1) was processed to determine the net count number for the individual X-ray lines and then to compute the respective -factors for Si-K and N-K. Those -factors were then tested on a Si3N4 sample region of a different well known thickness and vice versa. The experimentally determined -values can be used to calculate a proportionality factor to the respective Cliff-Lorimer factors, theoretically obtainable from available atomic data for any beam energy. Based on the experimentally specified /Cliff-Lorimer factor ratio the Zeta-factors for all element K-lines can then be calculated (Fig.2) and used to quantify other sample compositions using the Microsc. Microanal. 21 (Suppl 3), 2015 Microsc. Microanal. 21 (Suppl 3), 2015 2016 2004 same instrumentation. Our tests suggest that for the method to be successful, the beam current and the thickness and composition of the standard must be known precisely [6]. [1] T. C. Lovejoy et al., Proceedings of M&M (2015), this volume. [2] R. M. Stroud et al., Proceedings of M&M (2015), this volume. [3] M. Watanabe, Z. Horita, and M. Nemoto, Ultramicroscopy 65 (1996) 187�198. [4] M. Watanabe & D.B.Williams, J. of Micr. Vol. 221. (2006) 89-109. [5] G. Kothleitner, et al., Microsc. Microanal. 20, (2014), 678-686. [6] We gratefully acknowledge helpful discussions with W. Grogger, G. Kothleitner and S. Fladischer from the ZELMI in Graz, Austria and sample courtesy of K. Voltz, University of Marburg, Germany Figure 1. Two areas of the Si3N4 foil and the respective spectra used for testing the Zeta-factor method. Figure 2. -factors for K lines calculated for 200kV based on the theoretically determined CliffLorimer-factors and the N- and Si--factors obtained experimentally using Si3N4 as standard.");sQ1[1007]=new Array("../7337/2017.pdf","Standardless Quantification Of Actinides By EPMA","","2017 doi:10.1017/S1431927615010867 Paper No. 1007 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Standardless Quantification Of Actinides By EPMA A. Moy1,2, C. Merlet1, O. Dugne3. 1. 2. GM, CNRS, Universit� de Montpellier II, Place E. Bataillon, 34095 Montpellier, France. CEA, DEN, DTEC, SGCS, LMAC, 30207 Bagnols-sur-C�ze, France. 3. CEA, DEN, DEC, SA3C, 13108 St Paul lez Durance, France. Electron probe microanalysis (EPMA) is an analysis technique used to quantify with a high accuracy the amount of elements present on a sample of unknown composition. However, for some elements, such as the actinides, quantitative EPMA is not possible to achieve due to the lack of suitable reference standard samples. To overcome this difficulty, standardless methods can be employed with the use of virtual standards. Standardless method requires the accurate knowledge of absolute characteristic x-ray intensities. These x-ray intensities can be obtained by analytical calculations or by Monte Carlo (MC) simulations with mean of physical parameters. However, these parameters are not always well known and can lead to inaccurate results when used in standardless quantification methods. Validation by experimental measurements of the calculated x-ray intensities is needed. Absolute M and M x-ray intensities were measured by electron impact on thick samples of Pt, Au, Pb, U and Th for energies ranging from the ionization threshold up to 38 keV. X-ray intensities were recorded with an electron microprobe using several high resolution wavelength-dispersive spectrometers and converted into absolute x-ray yields by evaluation of the detector efficiency. The experimental absolute M and M x-ray yields were compared to the absolute x-ray intensities simulated by the multi-purpose Monte Carlo code PENELOPE [1]. For the studied elements, good agreement was obtained between the experimental and the simulated data, allowing future use of the simulated x-ray intensities in standardless quantification methods (Fig. 1). For practical applications, the x-ray intensities of the most intense L and M x-ray lines were calculated for elements with atomic number 89 Z 99 and for accelerating voltage ranging from the ionization threshold up to 40 kV. For a convenient use, the data were stored in a database in function of the accelerating voltage. To convert the virtual intensities into intensities which can be used in practical applications, the characteristics of the spectrometer used must be determined experimentally. Indeed, the simulated x-ray intensity is related to the recorded signal by the detection efficiency of the spectrometer and by the spectral broadening of the studied x-ray line. To determine the spectrometer characteristics, an accurate method was employed [2-4]: - The spectral broadening induced by the spectrometer and by the natural line-width is determined by non-linear fitting of the x-ray line of interest by a set of Pseudo-Voigt functions. - The detection efficiency of the spectrometer, which depends of the measured photon energy, is obtained by comparison between the bremsstrahlung intensity emitted by a bulk standard of carbon or nickel and the bremsstrahlung calculated by the MC code PENELOPE. The virtual standard intensity of the studied x-ray line is finally given by the ratio of the calculated x-ray intensity and the normalized area of the x-ray line, timed by the spectrometer efficiency. To facilitate the implementation of the method, a software program was developed. The interface allows the user to manage a database of pre-calculated x-ray intensities, assists during the determination of the spectrometer characteristics and calculate virtual standard x-ray intensities for the studied element. Microsc. Microanal. 21 (Suppl 3), 2015 2018 The virtual value can then be used as a standard value in the microprobe software to perform classical quantitative analysis. Standardless quantifications of heavy elements, such as lead, thorium, uranium and plutonium, were tested. The tests were firstly performed on standard samples of PbS, PbTe, PbCl 2, vanadinite, and UO2 (Fig. 2) to confirm the method. Standardless quantification results were shown in agreement with classical quantification results. The measurements were secondly performed on actinide samples of U, ThO 2, ThF4 and (U-Pu-Am)O2, and with results obtained during previous measurement projects [5]. Actinide quantifications obtained with the virtual standards were found in good agreements with the expected results and show deviation from the expected value below 15%. This confirms the reliability of the x-ray intensities simulated by the MC code PENELOPE and the reliability of the developed method used to determine the spectrometer characteristics. Using this method, standardless quantification of actinide by EPMA can be obtained with a good level of accuracy without actinide material standards. References: [1] F Salvat et al, "PENELOPE-2012: A Code System for Monte Carlo Simulation of Electron and Photon Transport", (OECD/NEA Data Bank, Issy-les-Moulineaux, France). [2] C Merlet and X Llovet, Microchim. Acta 155 (2006) 199. [3] C Merlet et al, Phys. Rev. A 73 (2006) 062719. [4] A Moy et al, J. Phys. B: At. Mol. Opt. Phys. 47 (2014) 055202. [5] C Merlet et al, Microsc. Microanal. 14 (Suppl 2) (2008) p.1094. Figure 1. Experimental measurements and theoretical Figure 2. Comparison between standard and predictions of the absolute M and M x-ray yields of standardless quantification of Pb in a PbTe U. standard sample.");sQ1[1008]=new Array("../7337/2019.pdf","Focused Interest Group on Microanalytical Standards (FIGMAS): Assessing the Quality, Availability and Need for Standards in the Microanalytical Community","","2019 doi:10.1017/S1431927615010879 Paper No. 1008 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Focused Interest Group on Microanalytical Standards (FIGMAS): Assessing the Quality, Availability and Need for Standards in the Microanalytical Community Julien M. Allaz1, Owen K. Neill2 and Anette von der Handt3 1. 2. University of Colorado Boulder, Department of Geological Sciences, Boulder (CO), USA Washington State University, Peter Hooper GeoAnalytical Laboratory, School of the Environment, Pullman (WA), USA 3. University of Minnesota Twin Cities, Department of Earth Sciences, Minneapolis (MN), USA It has been recognized over the past years that different electron microprobe and scanning electron microprobe laboratories use different sets of standards or reference materials for quantitative analysis. Unfortunately, some of these standards either have become unavailable (e.g., some natural minerals from the Smithsonian Institution collection) or are only available to a restricted group of people (e.g., internal reference materials). Other synthetic materials are also available commercially or provided by other institutions and research centers. However, they sometimes lack either broad availability or acceptable characterization (e.g., NIST glasses, Corning glasses, Drake & Weill REE-glasses... [1,2]). Another important problem for the community is a clear assessment of standard quality (the "Good", "Bad" and "Ugly" of Carpenter [3]): "good" homogeneous standards with accurate compositional information and without impurities or inclusions are rare, whereas "bad" standards, which lack good characterization, are more common. Individual lab managers do commonly examine their own standard collections to re-evaluate compositional homogeneity and test the accuracy of published compositions. Standards are also frequently re-analyzed at individual labs using various techniques, and therefore multiple accepted compositions for individual standards may exist. This information about which standards are available, how to obtain new standards, and which existing standards are "Good", "Bad", or "Ugly" (to use Carpenter's [3] terminology) is incredibly valuable to the microprobe community, but such information is only rarely disseminated in any systematic way. In fact, the most complete published review of common microprobe standards is now 30 years old [4]. More recently, other researchers have reviewed limited sets of natural and synthetic Smithsonian microbeam standards to check for their homogeneity and the presence of mineral inclusions or impurities. This includes studies on the composition, quality, homogeneity and presence of impurity in many of these standards [5,6,7,8], a study on the presence of Pb in the synthetic REE-phosphate [9], an evaluation of the micro- to nano-scale impurities in Kakanui Hornblende [10], and a review of the quality of several pyroxene standards [11]. The community would benefit greatly from an organized and widely-available repository of information about standards and reference materials (composition, standard quality, availability, source, etc.). To address these issues, the Focused Interest Group for MicroAnalytical Standards (FIGMAS*) aims to promote and facilitate the creation of a community-wide standard materials collection that supports consistency and inter-laboratory comparison. Part of this effort will be creating an online database of the currently available standard and reference materials used in electron microprobe laboratories. This database will be used to assess the need of additional reference materials for quantitative electron microprobe analyses by energy- and wavelength-dispersive X-ray spectroscopy. In addition, the FIGMAS will assess the needs of the microprobe community regarding collection and production of new standard or reference materials. The FIGMAS will also collect known procedures for synthesizing Microsc. Microanal. 21 (Suppl 3), 2015 2020 or sintering a homogeneous standard material (glass or mineral), and, where possible, make these procedures available to the greater microbeam analysis community. If necessary, the FIGMAS will also serve to develop new such procedures. In the future, this database could be extended and crossreferenced to other reference material database already available for other analytical techniques (e.g., GEOREM for laser ablation ICP-MS, SHRIMP or SIMS [12]). The preparation of some round robin test of commonly used standard is a possibility. The FIGMAS seeks to determine the availability and quality of each standard and reference material used and to define them as Standard Reference Material (SRM), Reference Material (RM) or Certified Reference Material (CRM) following the definition from the National Institute of Standards and Technology (NIST). The group will particularly focus on criteria such as (a) the quantity available and average particle size, (b) the compositional homogeneity, (c) pervasiveness of impurities or inclusions, (d) the resistance to beam damage and to high vacuum, (e) the non-soluble aspect of the material in most eluent used for polishing and cleaning, (f) the long term stability (potential for oxidation, hydration, or change in crystal structure, etc.), and (g) the availability and quality of quantitative analyses for each of the material's constituent elements, including the analytical method used and the availability of a certificate from a trusted source (e.g., NIST or a similar recognized institution). The results of the FIGMAS will be made available on a website that the microbeam community can use as reference. This database will be accompanied by publications in peer-reviewed scientific journals. The analytical community will be given access to the database to facilitate the addition of new reference material data. FIGMAS committee members will be responsible for reviewing any entry made by a member of the community to validate its provenance, delete inappropriate or spurious entries, consolidate duplicates, and ensure completeness of the record. With the help of the wider microbeam community, we hope that this database will constantly evolve to widely distribute this valuable information, enhance inter-laboratory comparability, and permit some poor "bad" standards to, one day, join the "good" standards realm. * The request for the creation of this FIG is pending. Researchers and laboratory managers interested to join this group are encouraged to contact one of the authors (julien.allaz@colorado.edu, owen.neill@wsu.edu, or avdhandt@umn.edu). References: [1] P.K. Carpenter et al., J. Res. Natl. Inst. Stand. Technol. 107 (2002), 703�718. [2] M.J. Drake and D.F. Weill, Chemical Geology 10 (1972), 179-181. [3] P.K. Carpenter, Microscopy & Microanalysis meeting 14 (2008), 530-531. [4] J.S. Huebner and M.E. Woodruff, US Geological Survey Open File Report 85-718 (1985), 237 pp. [5] E. Jarosewich, J.A. Nelen and J.A. Norberg, Geostandard newsletter 4 (1980), 43-47. [6] E. Jarosewich, J. Res. Natl. Inst. Stand. Technol. 107 (2002), 681�685. [7] T.R. Rose, Microscopy & Microanalysis meeting 14 (2008), 528-529. [8] http://mineralsciences.si.edu/facilities/standards.htm [9] J.J. Donovan et al., The Canadian Mineralogist 41 (2003), 221-232. [10] E.P. Vicenzi and T. Rose, Microscopy & Microanalysis meeting 14 (2008), 522-523. [11] J. Fournelle, AGU Fall meeting (2012), V23C-2827. [12] http://georem.mpch-mainz.gwdg.de");sQ1[1009]=new Array("../7337/2021.pdf","Improved Background Correction for the Quantification of Actinide M-lines in EPMA","","2021 doi:10.1017/S1431927615010880 Paper No. 1009 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Improved Background Correction for the Quantification of Actinide M-lines in EPMA Xenia Ritter1,2, Philipp P�ml1, St�phane Br�mier1, Jasper Berndt2 1. European Commission, Joint Research Centre, Institute for Transuranium Elements, P.O. Box 2340, 76125 Karlsruhe, Germany 2. Institut f�r Mineralogie, Westf. Wilhelms-Universit�t, Corrensstr. 24, 48149 M�nster, Germany Electron probe micro-analysis (EPMA) is an important technique for a broad range of applications in nuclear sciences. One main target is to improve the safety of the nuclear fuel cycle, by studying the chemical and physical properties of spent nuclear fuel and its fission products, either solids, volatiles, or gases, after the irradiation [1]. Of particular interest for the nuclear scientist is the distribution and quantity of actinides in the fuel, either before or after irradiation. Actinides in nuclear fuel can be added during fuel fabrication (usually containing the fissile material), or are being produced during irradiation by neutron capture and/or alpha-decay of existing actinides. The fuel types most commonly studied by EPMA are uranium oxide and Mixed Oxide Fuel (MOX) containing a mixture of uranium and plutonium oxide. There are other concepts of nuclear fuel, e.g., thorium based compounds, or special target fuels for the burning of unwanted minor actinides (Np, Am, Cm), a process known as transmutation. Minor actinides provide a major contribution to the radiotoxicity of spent nuclear fuel. To reduce the spent nuclear fuel's radioactive inventory, the fuel can be reprocessed and minor actinides and fission products separated. The minor actinides can then be mixed with the source materials for fresh nuclear fuel and subsequently be irradiated in a fast neutron reactor. The fast neutrons will transmute part of the minor actinides into nuclides with a short half-life that will then decay to less dangerous or even stable nuclides. For the latter case, EPMA is of particular interest as an analytical tool, because it provides information about the behaviour and level of transmutation of the added minor actinides. In EPMA, the M-lines (~ 4.1 � 3.3 �) are used for the quantification of the actinides. They are commonly measured using PET crystals (pentaerythritol, reflecting plane 002, 2d = 8.742 �). In nuclear sciences, however, spectrometers are often equipped with -quartz crystals (reflecting plane 1011, 2d = 6.686 �). Quartz crystals have the advantage of a better energy resolution of the actinide peaks at the cost of intensity. Considering the complexity of the actinide M-line region, however, the gain in resolution is much preferred over the loss in intensity. The analysis of actinides is complex. Their electron shells produce a variety of X-ray M-lines, all in the same wavelength region. Consequently the lines are overlapping and the analysis of a material containing 2 or more actinides becomes troublesome. Additionally, also the background correction becomes very difficult. Fig. 1 shows as an example a PuO2 sample containing a small amount of Am. The number and density of M-lines results in an increased background, that is not following the continuum trend anymore. It is hence not possible to use the general background correction approach of just measuring the background left and right of the peak and linearly interpolating the intensity. On top of that, the Ar K absorption edge (3.871 �) falls exactly into this region, causing a change of both peak and background intensities (see Fig. 1). X-ray counters on wavelength dispersive (WD) spectrometers are generally gas flow proportional counters that are filled with 90 % argon � 10 % methane (P10) gas. Microsc. Microanal. 21 (Suppl 3), 2015 2022 For accurate analysis the background correction problem has to be solved. One approach would be to use the Mean Atomic Number (MAN) approach as described in [2]. However, to calibrate the continuum it is necessary to find standards not containing the element of interest that have both higher and lower mean atomic numbers compared to the unknown. For materials like UO2 or PuO2 this seems impossible to accomplish, because materials like BkO2 are not available. We hence decided to follow another approach, the multi-point background acquisition developed by Probe Software, Inc. [3] and described in more detail in [4]. This approach allows the analyst to measure multiple background points in the region of interest followed by automatic evaluation and regression. An example is shown in Fig. 1. In this case, a WD spectrum was acquired for the Pu and Am M-line range. From the spectrum, two background points were selected on the far left and two on the far right of the M-peak range, avoiding the increased intensity region and the sharp drop of intensity due to the Ar K absorption edge in the central part. Fitting these points (in this case exponentially, red curve in Fig. 1) much improved background values could be interpolated. The benefit of this approach is that it costs only a little extra time, while improving background correction significantly. In our contribution we will show WD scans for the actinides Th, U, Np, Pu, and Am. We will show spectra of samples containing these as major elements and discuss how to improve background correction considering a high gain in accuracy and little loss of time. In addition, our approach will be tested on a real irradiated fuel sample and the results discussed. References: [1] C Walker, Journal of Analytical Atomic Spectrometry 14 (1999) p.447. [2] JJ Donovan and TN Tingle, Journal of Microscopy and Microanalysis 2 (1996) p.1. [3] Probe Software, Inc., 885 Crest Dr., US-97405 Eugene, Oregon, U.S.A. [4] JM Jercinovic et al, IOP Conf. Series: Materials Science and Engineering 32 (2012) p.012012. Figure 1. M-line region of a PuO2 sample containing minor amounts of Am. Red points mark the four selected background positions, the red dashed curve is the regression curve. Note the significant drop in intensity in the centre of the spectrum, due to the Ar K absorption edge.");sQ1[1010]=new Array("../7337/2023.pdf","Phase Analysis of Large EDS Datasets with Matlab","","2023 doi:10.1017/S1431927615010892 Paper No. 1010 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Phase Analysis of Large EDS Datasets with Matlab Richard C. Hugo1, Steven Bernsen 2, Kathy Breen3, and Alex Ruzicka1 1. 2. Portland State University, Department of Geology, Portland, OR, USA New Mexico Inst. of Mining and Technology, Dept of Earth and Env. Sci., Socorro, NM, USA 3. US Geological Survey, Oregon Water Resources Center, Portland, OR, USA Today's SEM experimentalist can acquire prodigious amounts of data in short amounts of time. High-stability FEG-SEMs equipped with high-throughput SDD detectors are widely available. Commercial microanalysis systems control motorized specimen stages, acquiring hundreds of spatially aligned fields of view (FOV) in an overnight or weekend SEM session [1]. The resulting mosaic of EDS datacubes can easily comprise hundreds of gigabytes. Commercial phase analysis software is optimized for rapid, automated data exploration. In real time, powerful, proprietary algorithms reduce spectral dimensions via principal components analysis (PCA) and/or automated element identification, and identify and quantify phases via spectral and/or cluster analysis [2-4]. However, dataset size is limited by computer RAM so that algorithms can be completed in real time. Further, because PCA and cluster results are unique to a given field of view, phase analysis results cannot be applied to other specimens. We have developed a set of software tools in Matlab that extend datacube-processing capabilities to datasets much larger than available RAM. Our process maintains a database of individual datacubes stored on disk. Subsets of this database are temporarily loaded into RAM, processing templates are applied, and the result from each FOV is added to an output database. The output database takes the form of a tiled RGB TIFF image, which can itself be retrieved and re-processed in RAM-sized chunks by Matlab, or viewed with various image processing packages. A distinct advantage of our approach is that templates created for a single dataset can be applied to subsequent datasets for direct comparison of similar specimens. The workflow is described in Figure 1. Input data include a matched suite of BSE images, RAW datacubes, and metadata files which describe the location and dimensions of each datacube. Our data were collected with Oxford INCA software and exported to RAW form with INCABatch. Data flow between hard drive and RAM is managed by Matlab's blockproc function. Dimension-reducing options currently include PCA, spectral color mapping, and element ROI coloring. Cluster algorithms include k-means and supervised nearest-neighbor. Future capabilities may incorporate 4-channel output, hierarchical clustering, and other algorithms. After code review we will release all source code to the Matlab File Exchange website for others to use and improve [5]. References: [1] S Burgess et al, Microsc. Microanal. 20-Suppl 3 (2014), p. 640. [2] PG Kotula et al, Microsc. Microanal. 9(1) (2003), p. 1. [3] T Nylese and R Anderhalt, Microsc. Today 22(2) (2014), p. 18. [4] DS Bright and DE Newbury, J. Microscopy 216 (2004), p. 186. [5] The authors acknowledge PSU Faculty Enhancement funding to Hugo. The specimen was analyzed at the PSU Center for Electron Microsc. and Nanofab. with funding from NASA grant NX10AH33G. Microsc. Microanal. 21 (Suppl 3), 2015 2024 (B) (C) (A) (D) (E) (F) (G) Figure 1. Workflow for a complete phase analysis of a 10k x 12k EDS mosaic. The subject is a 1.6cm meteorite igneous inclusion. (A) BSE image mosaic is stitched with a cross-correlation algorithm. A representative FOV is then chosen from the BSE mosaic and dimension reducing algorithms are compared. The resulting datacube has three energy channels and is visualized as an RGB image. Energy reducing options include (B) PCA, (C) spectral color mapping, and (D) traditional energy-ROI coloring. After applying the desired dimension reducing template to the full dataset, a new representative FOV (E) is chosen for cluster analysis and a partial phase image is generated (F). Finally, the clustering template is applied to the full RGB dataset and the full phase image (G) is written in tiled TIFF format.");sQ1[1011]=new Array("../7337/2025.pdf","Investigation of Mg-Li-Ca alloys using a Wavelength Dispersive Soft X-ray Emission Spectrometer and EPMA","","2025 doi:10.1017/S1431927615010909 Paper No. 1011 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Investigation of Mg-Li-Ca alloys using a Wavelength Dispersive Soft X-ray Emission Spectrometer and EPMA H. Takahashi1, M. Takakura2, T. Murano2, M. Terauchi3, M. Yamasaki4, Y. Kawamura4, P. McSwiggen5 Global business promotion division, JEOL Ltd., 13F, Otemachi Nomura Bld. 2-1-1 Otemachi, Chiyoda-ku, Tokyo, 100-0004 Japan 2. SM business unit JEOL Ltd., 1-2 Musashino, 3-chome, Akishima, Tokyo 196-8558, Japan. 3. Institute for Multidisciplinary Research for Advanced Materials, Tohoku University, Sendai 980-8577, Japan. 4. Magnesium Research Center, Kumamoto University, 39-1, 2-chome, Kurokami, Chou-ku, Kumamoto, 860-8555, Japan. 5. JEOL USA, 11 Dearborn Rd. Peabody, MA 01960, USA. A new group of ultra-lightweight magnesium-lithium-calcium alloys have recently been developed. These innovative alloys have some superior characteristic including high specific strength. These unique, and long sought after features, will soon lead to the use of these alloys in personal computers, aircraft, trains, automobiles, and many other manufactured products. However, the microanalysis of these alloys is significantly more challenging due to the presence of lithium. The presence of lithium limits the available analytical techniques to either that of using ultra, low-energy X-rays, or a surface analysis technique. A new method for analyzing the ultra low-energy X-rays was applied to these challenging alloys. We chose to analyze the lithium and magnesium using the Li-K and Mg-L X-ray emission using a newly developed wavelength dispersive, soft X-ray emission spectrometer (WD-SXES) attached to an electron probe microanalyzer (JEOL EPMA JXA-8230)[ 1-3]. The sample investigated is in the ternary Mg-Li-Ca system. Sample preparation involved using an ion cross-section polisher (CP) to both polish and flatten the oxidized surface. The WD-SXES was tuned the lower limit to allow for the detection of the Mg-L emission at between 50 to 45eV. Spectrum element mapping was acquired using the lower energy Mg L, Li-K and Ca-L emissions. These elemental maps and the backscattered electron image are shown in Figure 1, and spectra from the precipitates and matrix are compared in Figure 2. The element mapping shows that the Mg and Li are in a higher abundance in the darker phase (BEI) than in the lighter phase. The spectra data provide additional insights into their structure. The sharp leading edge of the Mg-L emission line at 50 eV corresponds to the Fermi-edge, and its unique lower energy profile corresponds to the electron density of state due to the brillouin zone [4]. Therefore different peak shapes must correspond to structural states of the two phases. The Li-K emission is also clearly detected and can be seen to have a higher concentration in the precipitate than in the matrix. Using the WD-SXES to investigate these alloys clearly shows that this method of spectrum map is useful in distinguishing both the chemical bonding states and the distribution of magnesium and lithium between the phases. Reference [1] M. Terauchi, et al., J. Electron Microscopy, 61, 1 (2012). [2] T. Imazono, et al., Appl. Opt. 51, 2351 (2012). [3] H. Takahashi, et al., Microscopy and Microanalysis, 19 (supple. 2), 1258 (2013). [4] D. J. Fabian "Soft X-ray band and electric structure" 62, (1968) Acade. Press. 1. Microsc. Microanal. 21 (Suppl 3), 2015 2026 Figure 1. Elemental distribution of Mg, Li, and Ca and backscattered electron image using WD-SXESEPMA JXA-8230. Figure 2. Comparison of Mg-L, Li-K and Ca-L(n=3) spectra from the two phases in the Mg-Li-Ca alloy. Red solid line - Matrix; blue solid line - Precipitate.");sQ1[1012]=new Array("../7337/2027.pdf","Use of a Fast WDS Instrument for Identification of Minor EDS Peaks.","","2027 doi:10.1017/S1431927615010910 Paper No. 1012 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Use of a Fast WDS Instrument for Identification of Minor EDS Peaks. John Konopka Thermo Fisher Scientific The Electron Probe Micro Analyzer (EPMA) has been well established as providing the best possible resolution in x-ray microanalysis for over fifty years. These instruments and most of the Wavelength Dispersive Spectrometers (WDS) built for installation on SEMs employ a focusing geometry1. A different WDS instrument is now available that employs an x-ray optic used to focus x-rays from a sample onto a flat resolving crystal. Mechanically this device can be adapted to nearly any SEM port that can accommodate an EDS detector. EDS is excellent for identifying major peaks in a spectrum. The relatively poor energy resolution of EDS makes reliable peak identification difficult for minor peaks or for small peaks overlapped by larger peaks. Using WDS as a complement to an EDS analyzer combines the strengths of both technologies. EDS can acquire data from all elements in a sample in one acquisition. When needed, WDS can be applied just to the narrow range of energies where EDS data has not resolved a peak well enough for identification. Employing direct drive motors and lacking gears and belts this device can scan through x-ray energies very quickly. This high speed and the sensitivity of the x-ray optic combine to yield fast energy scans even in cases of low signal intensity, as is often encountered in the SEM. Figure 1. shows a challenging peak identification problem. This could be Mg Ka, As La or a combination. Figure 2. shows this resolved by a WDS energy scan. Even with relatively little data the high resolution of WDS unambiguously reveals the presence of both Mg and As. Figure 3. shows the identification of a trace of Cu La under a dominant Ni La peak. In semiconductors it is important that these elements not be present together. Because of the small feature size it is necessary to use low kV for the analysis which requires the use of these L lines. Not only must the overlap be identified but the absence of Cu must be confirmed. Figure 4. shows the identification of Hf Ma line. The sample was a few nanometers of Hf deposited on GaAs. Figure 5. shows the identification of a nanoparticle of Ir on a silicon oxide glass bead. The small size of the particle necessitated the use of a FESEM and low beam current. In spite of the low signal intensity the analysis had to be done quickly or the beam would drift off of the sample. Even so, the identification of Ir is unambiguous. All WDS data were acquired with a Thermo Scientific Magnaray WDS and NS7 EDS analyzer. The SEMs used are identified with each displayed spectrum. The fast, non-focussing geometry WDS instrument brings the high resolving power of WDS to the SEM/EDS analyst. Simplicity of operation, sensitivity and high speed remove barriers to adoption making it possible for most operators to have access to a technology which till now was restricted to more complicated, dedicated instruments. References: [1] KFJ Heinrich, "Electron Beam X-ray Microanalysis", (Van Nostrand, New York) p.104. Microsc. Microanal. 21 (Suppl 3), 2015 2028 Figure 1. EDS spectrum of guanacoite bearing rock. Note the unknown peak at about 1.28 keV. This is resolved in Figure 2. The WDS data were scaled by 3.6x. Figure 2. Identification of Mg and As. 1100s acquisition time. JEOL JSM-7100F FESEM. Figure 3. Identification of trace Cu in Ni. 450s acquisition time. Hitachi S-6600 FESEM. The WDS data were scaled by 12x Figure 4. Identification of Hf film on GaAs. 363s acquisition time. JEOL JSM-7001F FESEM. Figure 5. Identification of Ir nanoparticle. 22s acquisition time. Zeiss Merlin FESEM");sQ1[1013]=new Array("../7337/2029.pdf","The Reliability of EBSD-based Microstructure Analysis","","2029 doi:10.1017/S1431927615010922 Paper No. 1013 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Reliability of EBSD-based Microstructure Analysis Kemal Davut1,2, Stefan Zaefferer3 1. 2. Metal Forming Center of Excellence, Atilim University, Ankara, Turkey Department of Metallurgical and Materials Engineering, Atilim University, Ankara, Turkey 3. Department of Microstructure Physics and Alloy Design, Max-Planck-Institute for Iron Research, Duesseldorf, Germany The electron back-scatter diffraction (EBSD technique is an ideal tool to describe the multiphase nature of materials in a quantitative manner provided that the diffraction patterns of different phases can be distinguished clearly. EBSD has been widely used in past years and it still increases its popularity which is reflected as an progressive increase in the number of publications per year. The statistical reliability of EBSD based results, despite its popularity and widely usage, remains an open question since typical observation volumes for EBSD are usually small compared to other techniques such as XRD. In particular, the tendency to use smaller step sizes for materials with a finer microstructure limits the investigation to relatively small areas in order to complete the measurements within the available (or a reasonable) time. Therefore, generalized conclusions are frequently drawn based on information obtained from a relatively small volume of the material. Particularly for heterogenous materials, scans of smaller areas quite possibly give rise to inaccuracies. The aim of this study is to investigate the EBSD measurement parameters (i.e. step size, size of the probed and scanned areas) on the validity of phase volume fraction determination and of texture analysis. For that purpose the low-alloyed TRIP steel (described in the previous section), that contains about 10% metastable austenite in a matrix of ferrite and bainite, is used. The austenite is very heterogeneously distributed in the form of bands, mainly due to Mn segregation. This particular arrangement of austenite deteriorates the optimum mechanical properties and also makes the characterization of phase composition particularly important and difficult. In addition, the accompanying martensitic transformation in TRIP steels is highly crystallographic and makes the texture analysis critical. Therefore, the present TRIP steel is an ideal material to investigate the representativeness and reliability of EBSD datasets. EBSD scans of various area and step sizes were systematically analyzed in order to determine the optimum sampling conditions. Moreover, Cochran's formula was used to determine a theoretical optimum sample size for predefined levels of precision and confidence. A significant deviation of theoretical optimum sample size from the experimentally determined one was found. This deviation, for the most part, is a result of the heterogeneous distribution of the austenite phase. Errors that may arise from insufficient sampling and also from false indexing of EBSD patterns at phase boundaries were also discussed. In addition, a new mapping technique, that helps to cover larger areas and probe more grains while keeping the particular advantages of a small step size, is presented. This new technique can keep the measurement time short and datasets small with minor modifications in the commercial dataacquisition software. This technique can be fully automated and it keeps the electron beam in focus even on extremely large surfaces, which is crucial for correct indexing and hence for improving the reliability of texture analysis of hardly indexed austenite in the present TRIP steel. Lastly, the comparison of results obtained by EBSD and XRD revealed that the EBSD technique is reliable and representative Microsc. Microanal. 21 (Suppl 3), 2015 2030 provided that sampling and sample preparation are adequately done. References: [1] The authors acknowledge funding from the German Research Foundation (DFG - Deutsche Forschungsgemeinschaft) reference numbers BL 402/19-1 and ZA 287/5-1. Figure 1. The austenite volume fraction of the TRIP steel obtained by EBSD and XRD techniques. The EBSD side error bars indicate the error due to possible mis-indexing. The XRD side error bars are estimated to be 1%, coming from selection of fitting parameters for Rietfeld refinement as well as texture effects. Figure 2. The pattern quality maps obtained using the new mapping technique. Note that the small maps were taken 200 m apart from each other. For ease of post-processing and visualization these smaller high resolution maps are combined into a single larger one.");sQ1[1014]=new Array("../7337/2031.pdf","Dark-Field Imaging based on Post-Processing of Electron Backscatter Diffraction Patterns in a Scanning Electron Microscope","","2031 doi:10.1017/S1431927615010934 Paper No. 1014 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Dark-Field Imaging based on Post-Processing of Electron Backscatter Diffraction Patterns in a Scanning Electron Microscope Nicolas Brodusch, Hendrix Demers, and Raynald Gauvin Department of Mining and Materials Engineering, McGill University, Montreal, Quebec, Canada. Dark-field (DF) imaging can be performed by selecting a specific diffracted beam in the selected area diffraction pattern in conventional transmission electron microscope (CTEM) or in the convergent beam electron diffraction pattern in scanning transmission electron microscopy (STEM) mode [1]. The resultant micrograph provides high intensity of the objects in the probed volume that diffract in this particular direction. In contrast, dark-field micrographs can be obtained in STEM mode by capturing the signal from a specific range of scattering angles, with the most representative example being the high-angle annular dark-field imaging (HAADF) [2]. This leads to a contrast mostly based on atomic number differences between the different objects analysed [3]. These techniques were developed originally for CTEM and STEM. Because DF based on scattering angles is technically easy to obtain in a scanning electron microscope (SEM) by collecting the transmitted/diffracted signals with an electron detector below the thin specimen, it has been implemented in SEMs seriously since several years. This permitted taking advantage of the high contrast and low beam damage obtained at low accelerating voltages STEM in the SEM is now routinely achieved with a spatial resolution close to 1 nm in field-emission SEMs [4]. Despite these new possibilities, DF imaging only based on diffracted beams has not been achieved yet in a SEM. The mostly used diffraction technique in the SEM has been, since the discovery of Venables [5], electron backscatter diffraction (EBSD) which has a spatial resolution of roughly 20-30 nm and which needs a limited bulk surface preparation compared to CTEM or STEM. EBSD is assumed to be related to the electron channeling pattern (ECP) diffraction technique by the reciprocity theorem [6], although its angular resolution is, at this time, limited by the pixel resolution of the acquisition equipment. Figure 1 is a comparison between an ECP and an EBSP acquired at 20 kV from a [001] (001) silicon wafer. In this work, pseudo-Kikuchi patterns (EBSP) recorded via EBSD were stored and reprocessed by reporting pixels or clusters of pixels intensities from a specific location in a reference EBSP to reconstruct the final image (EBSD map). A resulting micrograph (called EBSD-DF image) was produced with a direct link to the diffracted beams in the EBSP and hence, to the crystallography of the sample, i.e., a DF image. The origin of the contrast is then similar to that of electron channeling contrast image (ECCI) as shown in Figure 2, in which EBSD-DF micrographs of an indented compressed iron specimen with different reflections are displayed. However, the post-acquisition processing is an invaluable advantage over ECCI because it allows generating multiple micrographs at the same time with only one set of EBSPs recorded in a beam raster fashion. This opens new ways of extracting and using the information contained in each EBSP and the main applications, at this point, are understanding deformation behaviors and interpretation [7] of channeling contrast [8]. Microsc. Microanal. 21 (Suppl 3), 2015 2032 References: [1] D.B. Williams and C.B. Carter, Transmission electron microscopy: a textbook for materials science. 2009: Springer. [2] S. Pennycook, Ultramicroscopy, 30 (1989), pp. 58-69. [3] O.L. Krivanek et al, Nature, 464 (2010), pp. 571-574. [4] P.G.Merli et al, Microscopy and Microanalysis, 9 (2003), pp. 142-143. [5] J. Venables and C. Harland, Philosophical Magazine, 27 (1973), pp. 1193-1200. [6] O.C. Wells, Scanning, 21 (1999), pp. 368-371. [7] N. Brodusch, H. Demers, and R. Gauvin, Ultramicroscopy, 148 (2015), pp. 123-131. [8] S. Kaboli et al, Journal of Applied Crystallography, (2015), Submitted. Figure 1. Comparison between an electron backscatter diffraction pattern (EBSP) and an electron channelling pattern (ECP) of a [001] (001) silicon wafer. (A) Raw and (B) digitally filtered EBSP and (C) ECP. Both were recorded with an accelerating voltage of 20 kV. The working distance was 10 mm for the ECP and the detector distance was 80 mm for the EBSP. Tilt angles were 0 and 80� for the ECP and the EBSP, respectively. (D) Line profiles extracted from (A-C) show the higher resolution obtained with the ECP compared to the EBSP. Figure 2. High contrast EBSD-DF images of a micro-hardness indent on compressed iron obtained using long EBSD detector distance for high angular resolution EBSPs with an accelerating voltage of 30 kV and a detector distance of 50 mm as a function of the virtual beam position on the high angular resolution reference EBSP. (A) Reference EBSP, (B) FSD image, (C) band contrast (BC) map and (D-H) EBSD-DF images with specific reflections marked by arrows in (A). The EBSPs image resolution was 1344 � 1024 pixels.");sQ1[1015]=new Array("../7337/2033.pdf","Real-Time Discrimination of Phases with Similar Kikuchi patterns but Different Chemistry through Simultaneous EBSD and EDS","","2033 doi:10.1017/S1431927615010946 Paper No. 1015 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Real-Time Discrimination of Phases with Similar Kikuchi patterns but Different Chemistry through Simultaneous EBSD and EDS J. Goulden1, S. Sitzman2, K. Larsen1, and R. Jones1 1 2 Oxford Instruments NanoAnalysis, Halifax Road, High Wycombe, HP12 3SE, UK Oxford Instruments America, Concord, MA, USA EBSD is a widely accepted, powerful analytical technique capable of a wide variety of microstructural analyses, including preferred crystallographic orientation, grain size, grain boundary characterization, phase distribution and inter-phase orientation relationship determination. The technique is adept at discriminating phases crystallographically, so crystallography-based phase identification and discrimination during mapping are powerful applications of EBSD. However, limitations exist, and there are cases where indexing by conventional inter-band angle matching alone fails to discriminate phases yielding very similar Kikuchi patterns, for example, in the case of samples containing more than one face centered cubic metal phase. A number of different solutions have been applied to solve this, both in real-time and in post-acquisition. In one real-time method, Kikuchi band width matching is used to exploit inter-planar spacing differences between phases. With high accuracy band edge detection methods, reliable comparisons between measured and calculated bandwidths for the phases involved are used to determine a best match. This method is useful for band width differences of about 10% or greater. Another real-time alternative is phase differentiation on the basis of chemistry, in which no minimum difference in crystal structure or unit cell parameter is necessary for discrimination. This technique uses EDS data acquired simultaneously with EBSD data for each point analyzed. When more than one viable solution results from indexing alone, EDS information for the point is used to weigh the results in favor of the solution whose EDS data best match reference information for the phases involved. This is illustrated in the example below. A hole in a copper gasket has been filled with solder, and the heat applied to melt the solder has caused a reaction between it and the surrounding copper. The associated phase changes were investigated by EBSD. The sample contains copper, lead, tin and copper-tin intermetallic phases. Copper and lead possess the same face-centered cubic crystal structure (space group 225), thus show the same Kikuchi bands and inter-band angles. Conventional EBSD indexing cannot tell these phases apart. Adding automatic band width matching gives a reasonable improvement in copper-lead discrimination (Fig. 1a), even when only moderate pattern quality settings are used for reasonable acquisition speeds under normal conditions. Simultaneously collected EDS maps (Fig.s 1b, 1c and 1d) indicate that the phases present show very different chemistries. AZtec TruPhase exploits these chemical differences to yield almost perfect differentiation between the copper and lead phases, as seen in Fig. 1e, and does not require band width matching. This poster will discuss this technique and its application to other samples. Microsc. Microanal. 21 (Suppl 3), 2015 2034 Figure 1a. Phase map from indexing including band width matching, showing reasonable phase discrimination with some remaining errors between the red Cu and yellow Pb phases. Figure 1b. Cu X-ray map. Note the lack of Cu in the Pb + Sn bands on the left side of the map. Figure 1c. Pb X-ray map. Note the lack of Pb in the copper region on the right. Figure 1d. Sn X-ray map. Figure 1e. EBSD phase map, collected with Aztec Truphase activated during indexing, shows almost perfect discrimination of the lead and copper phases.");sQ1[1016]=new Array("../7337/2035.pdf","Optimization of 3D EBSD in a FIB-SEM System Using a Static Sample Setup","","2035 doi:10.1017/S1431927615010958 Paper No. 1016 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Optimization of 3D EBSD in a FIB-SEM System Using a Static Sample Setup Julien Guyon1,2, Nathalie Gey1,2, Daniel Goran3, Smail Chalal4, and Fabi�n P�rez-Willard4. Laboratoire d'Etude des Microstructures et de M�canique des Mat�riaux, LEM3, CNRS ISGMP, Universit� de Lorraine, F-57045 Metz Cedex 01, France 2. Laboratory of Excellence on Design of Alloy Metals for low-mAss Structures ('LabEx DAMAS'), Universit� de Lorraine, France 3. Bruker Nano GmbH, Am Studio 2D, 12489 Berlin, Germany 4. Carl Zeiss Microscopy GmbH, Carl-Zeiss Str. 56, 73447 Oberkochen, Germany Recently, the need for a comprehensive characterization of grains and grain boundaries has led to the extension of electron backscatter diffraction (EBSD) into three dimensions (3D). 3D EBSD can be performed by serial sectioning the sample with robot assisted grinding and polishing [1], laser ablation [2], or focused ion beam (FIB) milling [3] in automated processes. In this note we report about an optimized experimental setup for 3D EBSD data acquisition in a conventional FIB-SEM instrument. In contrast to other 3D EBSD experiments in a FIB-SEM, in our setup the sample remains static during the entire experiment. The advantages in terms of data quality and throughput are discussed and illustrated with measurements on a coarse-grained INCONEL 718 nickel-based superalloy as a model material. Experimental Procedures: The static 3D EBSD setup was realized on a customized ZEISS Auriga 40 FIB-SEM instrument equipped with a high resolution Bruker e-Flash EBSD detector. The original chamber of the ZEISS Auriga 40 was redesigned to accommodate the EBSD camera at an azimuthal angle = -15.3� (see Fig. 1(a) ). The exact value of results from geometrical considerations, namely: a) FIB and SEM beam are at an angle of 54� for the Auriga 40; b) the FIB beam is parallel to the FIB prepared surface, which is c) tilted 70� from the horizontal [4]. The EBSD camera can be positioned as close as 15 mm to the FIBSEM coincidence point, which is at a working distance of 5 mm. Using the Application Programming Interfaces (API) of FIB-SEM and EBSD systems the 3D EBSD acquisition workflow was fully automated. An INCONEL 718 sample, roughly (10x10x5) mm� in size, was used to characterize the experimental setup and evaluate the resulting 3D-data quality. The nickel-based superalloy is well suited for this purpose, because it yields high-quality EBSD patterns on 30 kV FIB polished surfaces even at a high EBSD acquisition speed of 140 Hz. Moreover the recrystallized grains contain numerous coherent twins. Their well known {111} boundary planes are used as a reference to quantify raw data alignment. Main Results: Usually, in a FIB-SEM the sample is moved iteratively between milling and EBSD positions during the 3D EBSD experiment. Such stage movements � and corresponding stage settling times � are eliminated in the static setup. Additionally, no time is spent on image registration steps after each stage movement, which are needed to reposition accurately FIB and SEM beams relative to the volume of interest (VOI) 1. Microsc. Microanal. 21 (Suppl 3), 2015 2036 if the sample is moved. Thus, 3D EBSD acquisition is simpler and the analyzed volume rate increased when performed under static conditions. Figure 1(b) shows reconstructed 3D EBSD data using [5] for the INCONEL 718 sample. In the center of the VOI a twinned grain is observed. The VOI has been sectioned to show a reconstructed plane perpendicular to the original EBSD acquisition plane. This plane runs through the twinned grain. At this point, it is important to stress that neither drift correction operations were performed during the 3D EBSD run nor data post-processing after the 3D run to ensure alignment of the different sections to each other. Nonetheless the twin boundaries in the rendered surface appear almost perfectly flat, which demonstrates the impressive alignment quality of the raw 3D data. They correspond to a {111} plane. In the static setup it is conceivable to increase throughput further by performing the EBSD measurement while milling. Prerequisite is to be able to index the EBSD patterns despite the noise introduced by FIB. To explore this possibility, EBSD patterns of a selected grain were measured while milling simultaneously the sample with different FIB probe currents. Even for FIB currents up to 30 nA, the patterns were good enough to be indexed with high confidence. References: [1] M. Uchic et al, 1st International Conference on 3D Materials Science (2012) p. 195-202 [2] M. P. Echlin et al, Acta Mater. 64 (2014), p. 307-315 [3] S. Zaefferer and S. I. Wright, in "Electron Backscatter Diffraction in Materials Science", ed. A. J. Schwartz et al (Springer 2009), p. 109-122 [4] D. D�nitz and Ch. Wagner, US Patent US8901510 B2 (2014) [5] M. A. Groeber and M. A. Jackson, Integr. Mater. Manuf. Innov. (2014) 3:5 Figure 1. (a) Top view CAD drawing of the customized FIB-SEM instrument. The EBSD camera is positioned at an azimuthal angle = -15.3� to realize the optimal static 3D EBSD setup. (b) Reconstructed 3D EBSD volume of an INCONEL 718 sample. The volume has been sectioned along a plane perpendicular to the EBSD acquisition plane to show the alignment quality of the raw data.");sQ1[1017]=new Array("../7337/2037.pdf","Addressing Pseudo-Symmetric Misindexing in EBSD Analysis of -TiAl with High Acuracy Band Detection","","2037 doi:10.1017/S143192761501096X Paper No. 1017 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Addressing Pseudo-Symmetric Misindexing in EBSD Analysis of -TiAl with High Accuracy Band Detection Scott Sitzman1, Niels-Henrik Schmidt2, Alberto Palomares-Garc�a3, Roc�o Mu�oz-Moreno4, and Jenny Goulden2 1. 2. Oxford Instruments America, Inc., Concord, MA, USA Oxford Instruments NanoAnalysis, High Wycombe, Buckinghamshire, UK 3. IMDEA Materials Institute, Madrid, Spain 4. Rolls-Royce UTC, University of Cambridge, Cambridge, UK Recent technological developments in EBSD has enabled great improvements in indexing reliability and accuracy [1], including for complex, multi-phase samples and phases that traditionally present difficult indexing challenges, such as -TiAl. High speed cameras deliver this capability at relatively high speeds since higher quality settings, e.g. lower CCD binning, do not necessarily mean slow data collection rates. However, some individual phases continue to pose considerable indexing challenges, especially those that present extremely similar Kikuchi patterns to the EBSD camera for different crystallographic orientations, in which case the indexing engine may not clearly discern the correct orientation solution. This phenomenon is called "pseudosymmetry", as it commonly involves relatively high intensity bands in certain patterns with an apparent higher symmetry than the crystal structure actually possesses. In many cases, only very slight differences in inter-band angle separate candidate solutions, and only robust and accurate band detection may identify the correct one among them. Conventional Hough-based band detection methods used in commercial EBSD systems since the early 1990s are sufficiently accurate for the great majority of indexing requirements. High accuracy Hough transform settings improves band detection accuracy, and is useful in mitigating pseudosymmetric misindexing [2]. However, these settings result in greater image transform time and reduced data acquisition speeds, and even modified Hough-based band detection alone may not completely eliminate pseudosymmetry problems in the most chronic cases. New band detection refinement methods promise to improve EBSD indexing performance for some of the most chronic and important cases, and allow reasonable data acquisition speeds. These methods deliver higher accuracy band detection by iteratively comparing the positions of simulated bands with bands in the actual Kikuchi pattern image, using expected versus actual band widths and, importantly, accounting for the hyperbolic shape of bands on the phosphor screen. In addition to delivering high precision crystallographic orientation data and helping discriminate different phases with similar Kikuchi patterns, this method is sufficiently sensitive to resolve fine differences in inter-band angle to nearly eliminate many cases of pseudosymmetric misindexing. For especially difficult cases, known single-phase pseudo-symmetric orientation relationships may be specifically examined by the system to help further resolve these indexing issues. One important application is for -TiAl alloys, which are very promising jet engine turbine materials, combining low density with good oxidation and creep resistance. The high temperature deformation behaviour of these alloys needs to be better understood before they can widely replace the higher density Ni-base superalloys currently used; For example, an improved knowledge of the fundamentals of Microsc. Microanal. 21 (Suppl 3), 2015 2038 crystallographic slip and its interaction with the / lamellar variants could be critical. Microstructural characterization is key in this effort, and EBSD has an important role to play. Pseudosymmetry, however, is a major issue here (Fig. 1a), arising from -TiAl's close tetragonal c:a unit cell parameter ratio of 1.018, giving the Kikuchi patterns it generates a pseudo-cubic configuration, resulting in indexing inaccuracies with common 90� orientation solution mistakes about the primary prismatic axes. These mistakes show the same misorientations as boundaries between real -TiAl lamellae, causing further problems in revealing the true microstructure. Phase discrimination between coexisting (TiAl) and 2(Ti3Al) phases can also be difficult (Fig. 2a). Application of a new, automated band detection tool and system knowledge of the confronting pseudo-symmetry almost completely eliminate these issues (Fig.'s 1b and 2b) and are used in real-time during data collection, at normal acquisition speeds. These tools may be applied for other important pseudosymmetry problems in EBSD. For example, in some solar cell materials, chalcopyrite-type phases generate Kikuchi patterns that deviate only very slightly from higher symmetry patterns, creating orientation determination problems which can undermine grain size and orientation studies if not properly addressed. References: [1] K. Thomsen et al., Royal Microscopy Society EBSD 2014 conference proceedings (2014) [2] C. Zambaldi et al., J. Appl. Cryst. 42 (2009), p. 1092-1101 b. a. Figure 1. Comparison of results using conventional and special band detection methods for (TiAl). (a) Orientation map from a conventional band detection dataset, exhibiting multiple orientation errors seen as speckled pixels. (b) Orientation map from a dataset using special high accuracy band detection with system knowledge of the encountered pseudosymmetry, resulting in almost no indexing errors. No postacquisition cleaning has been performed on either map. Scale bars are 20 �m. a. b. Figure 2. Comparison of band detection methods using phase maps, with red = (TiAl), blue = 2(Ti3Al), black = no solution. (a) Conventional band detection method results in some phase discrimination mistakes, seen as individual, speckled blue and red pixels. (b) New band detection method results in nearly no indexing errors, as well as fewer non-indexed points. Maps are as-acquired, with no post-acquisition data cleaning. Scale bars are 5 �m. = 5 �m; gamma+alfa2; Step=0,2 �m; Grid123x49 = 5 �m; alfa2+gamma; Step=0,2 �m; Grid123x49");sQ1[1018]=new Array("../7337/2039.pdf","Can EBSD Patterns Be Used for Determination of Grain Boundary Inclination?","","2039 doi:10.1017/S1431927615010971 Paper No. 1018 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Can EBSD Patterns Be Used for Determination of Grain Boundary Inclination? Michael Chapman1 , Saransh Singh1 , Marc De Graef1 1 Dept. of Materials Science and Engineering, Carnegie Mellon Univ., Pittsburgh PA 15213, USA The functional properties of many polycrystalline materials are strongly correlated with the grain boundary character. While electron or optical microscopes can be used to study grain boundaries, they only reveal the 2D nature by means of cross sectional images. To obtain an accurate five-parameter description of grain boundaries, a full 3D characterization is required. While there exist several theoretical and experimental techniques to study the grain structures in 3D, e.g., stereology, serial sectioning using a Focused Ion Beam (FIB), and high energy x-ray diffraction at a synchrotron facility, these methods are either statistical, destructive or not easily accessible. In this contribution, we describe preliminary results of a novel technique, combining Monte Carlo electron trajectory simulations and dynamical Electron Backscatter Diffraction Pattern (EBSP) simulations to extract the grain boundary inclination from EBSD observations. The Monte Carlo simulations closely follow the model described in [1]. In this model, the material is assumed to be isotropic. It is also assumed that only elastic scattering events change the direction of the electrons, while the inelastic events are responsible for the energy loss. Apart from a few special energy loss mechanisms, e.g., Bremsstrahlung radiation, the electron loses energy in discrete collision events. However, the averaging of all such events along the trajectory of the electron leads to significant simplifications and it is assumed that the electron loses energy at a fixed rate per unit distance along its trajectory. This approximation is known as the Continuous Slowing Down Approximation (CSDA) and is valid for electrons having energies in the multiple keV range. The Rutherford scattering cross section is used to model the elastic scattering process. The dynamical simulation of the EBSPs follows the model described in [2]. The calculated patterns are tuned to the crystallography of the sample (crystal orientation, symmetry) and the microscope parameters (acceleration voltage, sample tilt, detector geometry etc.). The model calculates the depth and energy distribution of the electrons using Monte Carlo simulations and integrates the backscatter probability over the corresponding depth range to determine the intensity at a detector pixel as a function of energy and the pixel location. The backscatter probability is then given as the squared modulus of the electron wavefunction, which is typically computed using either the Bloch wave approach or the scattering matrix approach. The geometry used for the pattern simulations is shown in Fig. 1(a). The sample is composed of two grains with different crystal orientations. Far from the grain boundary, the observed pattern will be that of a single orientation, i.e., grain A or B. However, as the electron beam is rastered across the grain boundary, the observed pattern will become a superposition (mixture) of pattern contributions from both grains A and B. As a first order approximation, the mixing of the two patterns can be assumed to be linear, with relative weights depending on the number of backscattered electrons originating inside each of the respective grains, normalized by the total number of backscattered electrons. The number of electrons coming from the two grains as a function of the distance of the electron beam to the grain boundary and the grain boundary inclination angle with respect to the surface is shown in Fig. 1(b). Shallow grain boundaries show the largest changes in the number of BSEs. Fig. 2(a) and (b) show simulated EBSPs for two random crystal orientations and Fig. 2(c) shows the expected pattern when the number of BSEs is evenly split between the two grains. We will present preliminary results on Microsc. Microanal. 21 (Suppl 3), 2015 2040 the feasibility of extracting the grain boundary inclination angle from the mixture EBSD pattern using a dot-product analysis between pure A and B patterns and the mixture pattern; we will show that, at least for shallow inclination angles, the grain boundary inclination can indeed be extracted from the mixture EBSD pattern. References [1] D.C. Joy in "Monte Carlo Modeling for Electron Microscopy and Microanalysis", (1995, Oxford University Press, New York) p. 25 [2] P.G. Callahan and M. De Graef, Microscopy and Microanalysis, 19, 1255�1265 (2013). [3] Research supported by the Air Force Office of Scientific Research, MURI contract # FA9550-12-1-0458. Figure 1. a) Sample geometry with a grain boundary inclination angle and distance X0 between electron beam incidence and boundary; b) number of electrons from Grain A vs. inclination angle for six different X0 values. Figure 2. a) EBSP for Grain A, b) EBSP for Grain B, c) EBSP for the grain boundary between Grain A and Grain B where 50 percent of the electrons come from each grain.");sQ1[1019]=new Array("../7337/2041.pdf","Automated Dictionary-based Indexing of Electron Channeling Patterns","","2041 doi:10.1017/S1431927615010983 Paper No. 1019 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Automated Dictionary-based Indexing of Electron Channeling Patterns Saransh Singh1 , Marc De Graef1 1 Dept. of Materials Science and Engineering, Carnegie Mellon Univ., Pittsburgh PA 15213, USA Traditionally, crystal defects (dislocations, stacking faults etc.) have been studied using TEM image modalities, such as bright field-dark field, weak beam, and STEM diffraction contrast imaging. Although these techniques offer high spatial resolution, they suffer from a number of drawbacks: tedious sample preparation; limited available thin area; and lack of a guarantee that the defect remains identical to that found in bulk crystals. Electron Channeling Contrast Imaging (ECCI) has recently gained visibility as a technique for the study of near surface defects in crystals [1]. ECCI is an SEM image modality (which hence enjoys less demanding sample preparation requirements) and relies on the variation of the backscattered electron yield near crystal defects. It has been shown that the burgers vector of a surface penetrating dislocation can be identified using the same visibility criteria as used in the TEM. The ECCI technique, however, requires accurate knowledge of the diffraction conditions, which, in turn, requires determination of the crystal orientation. While Electron Back Scatter Diffraction (EBSD) can be used to determine the sample/grain orientation, it would be more convenient to have the ability to determine the orientation from an Electron Channeling Patterns (ECPs), since such patterns are acquired already as part of an ECCI observation. While there are several commercial packages available for the indexing of EBSD patterns, no such counterpart exists for ECPs. In this contribution, we introduce a dictionary based approach to index ECPs, based on previous work with EBSD pattern indexing [2]. The dictionary approach uses a physics-based forward model, described in detail in [3], to generate a set of simulated ECPs, utilizing the microscope geometrical parameters and symmetry information of the crystal being studied. The forward model estimates the depth distribution of the BSE1-type electrons using a Monte Carlo approach, and then integrates the backscatter probability over the corresponding depth range for electron exit directions sampled on a sphere. Details of the simulations, which can be performed using either Bloch waves or the scattering matrix formalism, can be found in [3]. This forward model is used to compute a so-called "master" channeling pattern, from which any ECP can be interpolated using bi-linear interpolation. The simulated master pattern for Nickel for a 30 kV acceleration voltage is shown in Fig. 1(a) using a modified equal-area Lambert projection from the hemisphere to a square. Fig. 1(b) shows the [001] zone axis pattern interpolated from the master pattern. Several representative dictionary patterns interpolated from the master pattern, along with the corresponding Euler angles, are shown in Fig. 2(a)-(e). In the dictionary approach, each ECP is reformatted as a normalized column vector (with or without average background subtracted), and the dot products between each experimental ECP and all dictionary pattern vectors are computed and ranked in decreasing order. The top k dot products (typically, k = 40 or so) and their associated Euler angles are kept for further analysis; this set of k patterns represents the k-nearest-neighbor (kNN) neighborhood of the experimental pattern in the dictionary. Cluster analysis of this neighborhood and the kNN neighborhoods of neighboring pixels provides information on whether the experimental pattern likely stems from a grain interior region or from a region close to a grain boundary. The Euler angle estimation of the lattice orientation at a particular pixel is performed using a directional statistics analysis [4] of the top k inner product matches. Since orientations can be mapped onto unit quaternions on the unit sphere S 3 in R4 , one must use 4D directional distribution functions; both the von Mises-Fisher (vMF) distribution and the axial Watson distribution have been Microsc. Microanal. 21 (Suppl 3), 2015 2042 found to be useful in this respect; they are defined by fvMF (x; �, ) = cvMF ()e��x , fWatson (x; �, ) = cWatson ()e(��x) ; � is the quaternion representing the mean direction, x is a unit quaternion, and is the concentration parameter (a measure for the spread of the distribution); the normalization factors are given by cvMF () = /(4 2 I1 ()) and cWatson () = exp(-/2)/(I0 (/2) - I1 (/2)), with Ii (x) a modified Bessel function of order i. For a given kNN, the corresponding unit quaternions are used in a Maximum Likelihood (ML) estimation framework to determine the most likely mean direction � and concentration parameter for each experimental ECP. The ML approach uses mixtures of vMF or Watson distributions, and explicitly takes crystallographic symmetry into account. The approach has been implemented on a combination of multi-core CPUs and a GPU to maximize performance. We will present ECP indexing results as well as an analysis of the robustness of the dictionary approach. References [1] D.R. Clarke, Philosophical Magazine 24 (1971) p. 973. [2] Y.H. Chen et al, Microscopy and Microanalysis, under review. [3] P.G. Callahan and M. De Graef, Microscopy and Microanalysis 19 (2013) p. 1255. [4] K.V. Marida and P.E. Jupp, "Directional Statistics," Wiley & Sons (1999). [5] Research supported by the Air Force Office of Scientific Research, MURI contract # FA9550-12-1-0458. 2 (a) (b) Figure 1: (a) Modified equal-area Lambert projection of the ECP master pattern for Ni at 30 kV. The top right quadrant is deliberately made lighter to show some prominent zone axis. (b) shows the [001] zone axis pattern for an angular range of 8o interpolated from the master pattern. (a) (63.08,47.39, 333.08) (b) (62.33,48.47,332.33) (c) (61.56,49.51,331.56) (d) (60.75,50.51,330.75) (e) (59.92,51.46,329.92) Figure 2: (a)-(e) Simulated ECPs for different Bunge Euler angles; Nickel, 30 keV.");sQ1[1020]=new Array("../7337/2043.pdf","Quantitative Analysis of Correlated 3D Strontium Titanate Datasets Collected by TriBeam and Diffraction Contrast Tomography","","2043 doi:10.1017/S1431927615010995 Paper No. 1020 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantitative Analysis of Correlated 3D Strontium Titanate Datasets Collected by TriBeam and Diffraction Contrast Tomography W.C. Lenthe1, M.P. Echlin1, M. Syha2,3, A. Trenkle2, P. Gumbsch2,4, T.M. Pollock1 1. 2. Materials Department, University California at Santa Barbara, Santa Barbara, USA Institute for Applied Materials IAM, Karlsruhe Institute of Technology, Karlsruhe, Germany 3. European Synchrotron Radiation Facility, Grenoble, France. 4. Fraunhofer IWM, Freiburg, Germany Much attention has been paid to the idea of the correlation of analytical techniques such as tomography. Recently, techniques for the acquisition of 3D tomographic and 4D time resolved datasets have emerged allowing for the analysis of mm3 volumes of material with nm-scale resolution. The TriBeam technique permits the acquisition of 3D EBSD datasets using a femtosecond laser to section material at unprecedented speed [1-2] with low damage [3-4] and high resolution [5]. Diffraction contrast tomography (DCT) [6], a synchrotron based X-ray technique acquires datasets non-destructively, permitting the repeated imaging of samples to collect 4D microstructural evolution [7] in crystalline materials. However, the vast majority of materials tomography datasets have been combined in a purely qualitative sense to date. In this work, a methodology for the precise alignment of tomographic datasets, including the alignment of the sample and orientation reference frames, and simultaneous identification and linking of grain structure between tomography datasets has been developed. The application of these algorithms to a pair of datasets collected from a single strontium titanate (STO) sample using both TriBeam tomography and synchrotron X-ray diffraction contrast tomography (DCT) will be presented. The resulting merged datasets have been quantitatively analyzed on the voxel scale and at the grain scale for the direct comparison of these two tomography techniques. Tomograms from a single sample were collected by DCT with time evolution and subsequently using the TriBeam, then merged onto one sampling grid for simultaneous quantitative analysis at both the grain and voxel scale. The two tomographic techniques employed collect diffraction information in fundamentally different ways. DCT illuminates the entire sample and simultaneously collects projections on a detector from all grains with orientations satisfying the Bragg condition. In contrast, EBSD patterns are collected serially for each voxel in the TriBeam experiment. Therefore, orientation reference frames and sample reference frames must be aligned, as shown in Figure 1, before quantitative analysis may be performed. Analysis of the merged datasets shows the resolution limits of the DCT technique and locations where the accuracy of the technique may be degraded. Errors in EBSD segmentation are also identified. Visualizations showing the difference in spatial locations of grain boundaries and centroids will be presented as well as quantitative comparisons of metrics for the quality of the dataset alignment algorithms. References: [1] M. P. Echlin et al, Review of Scientific Instruments 83(2) (2012), p. 023701. [2] M.P. Echlin et al, Advanced Materials 23 (2011), p. 2339. [3] S. Ma et al, Met. Mat. Trans. A 38 (2007), p. 2349. Microsc. Microanal. 21 (Suppl 3), 2015 2044 [4] Q. Feng et al, Scripta Mat. 53 (2005), p. 511. [5] M.P. Echlin et al, Materials Characterization: Tutorial Review 100 (2015), p. 1. [6] H.F. Poulsen et al, Journal of Applied Crystallography 34 (2001), p. 751. [7] D. Gonzalez et al, Acta Materialia 61 (2013), p. 7521. Figure 1: The reference frames are not aligned between the TriBeam tomogram (a) and the DCT tomogram (b), which must be aligned before quantitative analysis of the correlated microstructure can be performed.");sQ1[1021]=new Array("../7337/2045.pdf","Pattern Overlap and High Resolution Electron Backscatter Diffraction","","2045 doi:10.1017/S1431927615011009 Paper No. 1021 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Pattern Overlap and High Resolution Electron Backscatter Diffraction Vivian Tong1, Jun Jiang1, Angus Wilkinson2 and Ben Britton1 1. 2. Department of Materials, Imperial College London, Prince Consort Road, London, SW7 2AZ, UK Department of Materials. University of Oxford, Parks Road, Oxford, OX1 3PH, UK High resolution electron backscatter diffraction (HR-EBSD) is a technique that can routinely measure residual elastic strain and lattice rotations with very high precision (~10-4 in strain and rotation) [1-2] across large areas in a scanning electron microscope. The technique involves cross correlation of two or more diffraction patterns to measure subtle (sub pixel) image shifts and relate these to the relative strain and lattice rotation difference between the two patterns. Often interfaces are of significant interest in understanding the limiting performance in real component design and understanding residual elastic strain and lattice rotation can aid the generation of new materials and microstructures. However understanding the behavior of interfaces can be difficult, as in these regions the diffraction pattern generated in the SEM may contain information from the grains either side of the interface. This may limit the precision and spatial resolution of HR-EBSD. To explore limitations on the sensitivity of the HR-EBSD technique in the vicinity of grain boundaries and interfaces, we have performed a series of experiments using real and simulated pattern mixing, as well as some statistical analysis of the frequency of these regions that are likely to contain suspect data. EBSD patterns were captured from well-polished annealed Zircaloy-4 using a Zeiss Auriga CrossBeam instrument in high current mode with a large aperture (120�m) combined with a Bruker eFlashHR camera. The interaction volume of the electron beam with the sample was measured using a cross correlation method based upon the work of Chen et al [3]. The peak height measurements were compared with a simple simulation of the interaction volume, where the volume was assumed to be a half sphere crossing a vertical grain boundary of radius ~60nm (Figure 1). With knowledge of the interaction volume, simulations of pattern mixing were performed. Intensities from the two separate grains were mixed artificially (Figure 2) to explore the influence of a less dominant pattern on the cross correlation results. Pattern mixing fractions were related to distance from the grain boundary for our measured interaction volume. It was found that the accuracy, measured with an imposed pattern shift is only significantly (>2x10-4) affected when the interaction volume was ~18nm from the boundary. The likelihood of suspect data was evaluated through consideration of the EBSD sampling strategy and the interaction volume. In typical experiments, EBSD maps are generated using a discrete sampling of the microstructure in a grid and grain boundaries are typically curved and their exact position is rarely known. Therefore the frequency of sampling close to a grain boundary was explored through sampling of a simulated Voronoi microstructure (Figure 3). The probability of sampling near a grain boundary was found to be linearly dependent on the threshold distance and inversely proportional to the step size. Therefore the probability of points in the neighborhood of a grain boundary (i.e. 1 step either side of the boundary) can be estimated. With a step size of 0.2�m, 30% of points in this region likely to show some overlap (i.e. the pattern contains diffraction information from both grains) and only 9% points in this neighborhood are likely to show significant strain error (>2x10-4). Microsc. Microanal. 21 (Suppl 3), 2015 2046 Figure 1: Measurement of the electron beam-sample interaction volume through cross correlation of diffraction patterns. (a) schematic of the half-sphere model; (b) experimetal measurement of the volume; (c) verification using a half sphere model simulation. Figure 2: Experimentally obtained patterns from grains of Zircaloy-4 and the result of pattern mixing, experimentally found with overlap of the grain boundary with the interaction volume. Figure 3: (a) Simulation of an example microstructure and (b) measurement of the likelihood of a point near a grain boundary, i.e. within one step size, being within a specific distance of the exact grain boundary position. References: [1] A.J. Wilkinson, G. Meaden, and D.J. Dingley, Ultramicroscopy 22 (2006) p. 1271 [2] T.B. Britton, J. Jiang, P.S. Karamched and A.J. Wilkinson, JOM 65 (2013) p. 1245 [3] D. Chen, J-C. Kuo and W-T. Wu, Ultramicroscopy 111 (2011) p. 1488 [4] Acknowledgements: The authors would like to thank Drs Stuart Wright, Eleanor Clarke, Phani Karamched and Philip Littlewood for helpful discussions over the years concerning pattern overlap at grain boundaries and its effects on HR-EBSD measurements. Funding contributions are acknowledged from ESPRC through the HexMat grant (EP/K034332/1), Rolls-Royce plc and AVIC-BIAM.");sQ1[1022]=new Array("../7337/2047.pdf","Electron and X-ray Diffraction Measurements of Elastic Stress and Plastic Strain from Ultrasonic Impact Treatment of Aluminum-Magnesium Alloys","","2047 doi:10.1017/S1431927615011010 Paper No. 1022 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electron and X-ray Diffraction Measurements of Elastic Stress and Plastic Strain from Ultrasonic Impact Treatment of Aluminum-Magnesium Alloys L.N. Brewer1,2, E.F. Fakhouri2, M.E. Haggett2, , and K.N. Tran3 1. Metallurgical and Materials Engineering, University of Alabama, Tuscaloosa, AL, USA 2. Mechanical and Aerospace Engineering, Naval Postgraduate School, Monterey, CA, USA 3. Naval Surface Warfare Center Carderock Division, West Bethesda, MD, USA Ultrasonic impact treatment (UIT) is used to prevent fatigue and stress corrosion cracking by placing compressive stresses at the surface of a metallic component. The compressive stresses are generated by severe plastic deformation near the surface and can considerably evolve the microstructure. Mitigation of stress corrosion cracking is particularly important for aluminum-magnesium (Al-Mg) alloys in ship structures as they can sensitize under relatively mild thermal conditions and then experience severe intergranular corrosion and cracking in a marine environment.[1, 2] The connections between UIT processing parameters, plastic strain levels and distributions, and the resultant elastic compressive stresses are not well known, particularly for Al-Mg alloys used in ship structures. In this research, electron backscatter diffraction (EBSD), optical microscopy, and x-ray diffraction were used to determine the connections between impact amplitude and pin diameter on the distribution of plastic and elastic strains in the Al-Mg alloy AA5456. Four AA5456 plates were gas metal arc (GMA) welded and then ultrasonically treated using a series of impact displacement levels and pin diameters. Two of the plates were comprised of sensitized material from aging ship structures, while the other two plates were from unsensitized, control material. X-ray diffraction-based, residual stress measurements were performed as a function of distance from the weld for all conditions using cobalt k-alpha x-rays and the Proto Manufacturing iXRD instrument. Crosssectional samples for optical microscopy and EBSD were prepared by cutting the plate perpendicular to the weld and using standard metallographic techniques down to a finish of 0.05um colloidal silica. For EBSD, the samples were further electropolished using a solution of 90 ml ethanol and 10 ml perchloric acid. EBSD analyses were completed using a Zeiss Neon 40 scanning electron microscope at 20 keV beam voltage, 60 �m objective aperture, and a probe current of 1 nA. The EBSD signal was collected and analyzed using the EDAX OIM 6.0 software with a Hikari high-speed camera. The x-ray residual stress profiles show large compressive stresses across the weld, the heat affected zone, and the base metal after UIT (Figure 1). As the pin diameter decreased, the level of compressive residual stress increased, likely because of higher contact pressure for the smaller pins. The displacement amplitude did not affect the level of compressive residual stress. Sensitization level of the alloy also did not have an obvious impact on the level of compressive residual stress. The distribution of plasticity below the surface varied significantly from one portion of the weld to another (Figure 2). In the fusion zone, the large, annealed grains generated by weld pool solidification were heavily deformed by the UIT process and actually showed a distinct grain refinement at the surface of the UIT crater. This grain refinement after UIT has been observed previously for aluminummagnesium alloys and can actually result in nanocrystalline structures at the very surface of the UIT crater.[3] In the base metal, UIT increased the extent of plastic deformation somewhat over that present from the plate manufacturing, but not to the extent observed in the fusion zone. For the smallest pin diameters and largest displacement level, substantial sub-surface cracking was observed for the Microsc. Microanal. 21 (Suppl 3), 2015 2048 sensitized plates (Figure 3). This sub-surface cracking was not present in the unsensitized material. From these measurements, it is apparent that all levels of UIT produce substantial compressive stresses, but that an overly large contact stress may result in sub-surface cracking. Figure 1. X-ray residual stress profiles for GMA welded aluminum plate after UIT with three different pin diameters. The blue diamonds represent the residual stress profile of the welded sample without UIT. Figure 2. Grain orientation spread maps (GOS) using EBSD for the weld (A) and base metal (B) at the surface (top of map) and subsurface in a pin crater. The color scheme is in degrees. Red denotes an area with a large amount of intragranular misorientation. Blue denotes a region with very little intragranular misorientation. Figure 3. SEM backscatter image of sub-surface cracking in the heat affected zone of ultrasonically treated, sensitized AA5456 plate (1 mm pin diameter) References: [1] J. Springer et al., Fleet Maintenance and Modernization 2014, American Society of Naval Engineers (2014) p. 1-12. [2] R. Schwarting et al., Fleet Maintenance and Modernization 2011, American Society of Naval Engineers (2011) p.1-17. [3] K.N. Tran and L. Salamanca-Riba, Advanced Engineering Materials, 15 (2013) 1105.");sQ1[1023]=new Array("../7337/2049.pdf","Pushing the Limits of Cathodoluminescence Signal Detection: Analyzing 2D Materials","","2049 doi:10.1017/S1431927615011022 Paper No. 1023 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Pushing the Limits of Cathodoluminescence Signal Detection: Analyzing 2D Materials Anas Mouti1,2, Ritesh Sachan1, Andrew R. Lupini1, Matthew F. Chisholm1, Stephen Pennycook3 1. Oak Ridge National Laboratory, Materials Science and Technology Division, Oak Ridge, TN University of Kentucky, Chemistry department, Lexington, KY 3. National University of Singapore, Department of Materials Science and Engineering, Singapore 2. We report in this communication the analysis of hexa-BN atomic layers by cathodoluminescence in a scanning transmission electron microscope (STEM-CL). Both polychromatic imaging and light spectroscopy were performed varying thicknesses of BN, down to a few monolayers. Previous studies have reported spatially resolved structural and optical properties of BN flakes [1] ; our ultimate goal in this study is to collect useful data from a single monolayer, which would be the thinnest possible sample. We are conducting the research on a custom CL system that we have built and integrated onto a VG 601 STEM, equipped with an aberration corrector to increase probe current at higher magnifications. System characteristics include 2 str collection angle, 360-1000nm wavelength spectroscopy range, and 180900nm photomultiplier detection range. We report success in collecting spectra and polychromatic images from samples as thin as 5-6 monolayers. In Fig.1, (a) is the ADF picture of the edge of a BN flake, and A it's thinnest region, (b) is the corresponding polychromatic CL image, showing that signal is clearly collected from A. Fig. 1(c) is a high resolution image of A taken with a Nion UltraSTEM 100, from which it can be deducted that the thickness is about 5 monolayers. Finally, (c) is a CL spectrum acquired from the 5 monolayers. We discuss practical challenges such as beam damage with dose and acceleration voltage, and ways to improve the signal-to-noise ratio. We finally study the evolution of emission energy and intensity with different thicknesses. Our current research includes studying monolayers, and their interactions with dopants and plasmonic particles. References: [1] Bourrelier et al. ACS Photonics 2014, 1, 857-862 [2] This work was sponsored by US Department of Energy, Office of Science, Basic Energy Sciences, Materials Sciences and Engineering Division, as well as the Saudi National Science Fund. Microsc. Microanal. 21 (Suppl 3), 2015 2050 A 20 nm 20 (a) ADF (b) CL 400 450 500 550 600 Wavelength (nm) (c) High resolution of A (d) Spectrum from region A Figure 1: (a) Annular Dark Field (ADF) image of the edge of a BN flake, and (b) it's corresponding CL image, signal is detected from the thinnest area A of the flake. From the high resolution image of A in (c) it can be determined that A is about 5 monolayers thick. Finally, (d) is a spectrum acquired from A.");sQ1[1024]=new Array("../7337/2051.pdf","SPARC: a Cathodoluminescence Platform for Nanoscale Plasmonics in a Scanning Electron Microscope","","2051 doi:10.1017/S1431927615011034 Paper No. 1024 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 SPARC: a Cathodoluminescence Platform for Nanoscale Plasmonics in a Scanning Electron Microscope Toon Coenen1,2 and Jacob Jan de Boer1 1. 2. Delmic B. V., Delft, The Netherlands FOM Insitute AMOLF, Center for Nanophotonics, Amsterdam, The Netherlands In the field of plasmonics metallic nanostructures are used to confine light to deep-subwavelength volumes. Owing to the nanoscale size of such structures it is hard to resolve their optical properties with conventional optical microscopy. Recently, electron-beam spectroscopy techniques have emerged as powerful probes in nanoscience due to their ability to generate, probe, and control light at length scales far below the diffraction limit of light. Taking advantage of the extremely high spatial resolution, novel techniques have appeared that combine electron beam excitation with optical spectroscopy. Spatially-resolved cathodoluminescence (CL) spectroscopy, in which the electron-beam-induced radiation is collected inside an electron microscope, is one of these techniques that holds great potential for nanoscience. For a long time CL spectroscopy was mainly used in geology to analyze and identify minerals, but in the past two decades its scope has expanded significantly. Recently it has been used to study fundamental optical properties of a myriad of metallic, semiconductor, and dielectric (nano)materials in the fields of materials science and nanophotonics, including plasmonics and metamaterials. We have developed a special version of CL spectroscopy in which we can both effectively measure the emitted spectrum as well as the angular emission distribution (SPARC) [1-2]. The SPARC system is integrated with a standard commercially available scanning electron microscope (SEM). SEMs are relatively easy to operate and do not require electron-transparent samples. Additionally, the vacuum chamber is more spacious providing more flexibility. As a result SEM-CL is widely applicable and easy to use. Innovations in the SPARC CL-system include an improved light collection with a piezo-controlled parabolic mirror and the development of angle-resolved CL, which have further expanded the possibilities. In Figure 1(a) we show a photograph of the SPARC CL collection system. The piezoelectric positioning of the paraboloid mirror enables efficient light collection which is critical in plasmonic studies where CL signals are typically very low (10-4 photons/electron). Furthermore, by measuring the CL beam profile from the paraboloid with a 2D camera we are able to measure the angular profile, as every transverse point in the beam corresponds to a unique emission angle. In Figure 1(b) an illustration is shown of how the angle-resolved measurements are performed. The canonical structure in plasmonics is a single nanoparticle. For small nanoparticles (< 50 nm) the resonant features are relatively easy to determine as they are predominantly dipolar. However, for larger particles this can be more challenging due to retardation effects which lead to the presence of higher order multipoles (magnetic dipoles, electric quadrupoles etc.). In Figure 2(a) we show the CL spectrum of a gold nanoparticle (180 nm diameter, 80 nm high) on a silicon substrate as measured with a spectrometer. The spectrum is averaged over all scanning pixels that fell within the particle (5 nm pixel). As inset we show the spatial profile at the peak wavelength which a distinct doughnut pattern. Figure 2(b) shows the angular profiles collected at different excitation positions on the particles. Clearly the Microsc. Microanal. 21 (Suppl 3), 2015 2052 2040 particle acts as an efficient antenna and beams light away from the excitation position in a non-dipolar pattern. The CL spectrum, the nanoscale excitation distribution as well as the angular profiles provide valuable insights into the resonant behaviour of such particles [3]. Although this is just one example we have applied the technique to many other plasmonic systems as well and shown that the SPARC is a very powerful platform for studying plasmonics at the nanoscale. References: [1] T. Coenen, E. J. R. Vesseur, and A. Polman, Appl. Phys. Lett. 99, 143103 (2011). [2] T. Coenen, E. J. R. Vesseur, A. Polman, and A. F. Koenderink, Nano Lett. 11, 3779 (2011). [3] T. Coenen, F. Bernal Arango, A. F. Koenderink, and A. Polman, Nat. Commun. 5, 3250 (2014). Figure 1. (a) Piezo-controlled mirror manipulator system for efficient light collection (b) Graphical representation of angle-resolved detection of CL where light coming from a parabolic mirror is projected onto a 2D silicon camera. Figure 2. (a) Average CL spectrum for a 180 nm diameter gold particle on a silicon substrate. The inset shows the spatial profile at the peak wavelength (550 nm) where the white dashed line indicates the geometrical edge of the particle. (b) Angular emission profiles measured at 600 nm showing CL intensity as function of azimuthal and zenithal emission angles. for the same particle for central excitation (1) and four edge excitation positions (2-5) [3].");sQ1[1025]=new Array("../7337/2053.pdf","Measuring Gas Adsorption on Individual Facets of a Nanoparticle by a Surface Plasmon Nanoprobe","","2053 doi:10.1017/S1431927615011046 Paper No. 1025 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Measuring Gas Adsorption on Individual Facets of a Nanoparticle by a Surface Plasmon Nanoprobe Pin Ann Lin,1,2 John M. Kohoutek,1,2 Jonathan Winterstein,1 Henri Lezec,1 and Renu Sharma1 1 Center for Nanoscale Science and Technology, National Institute of Standards and Technology, Gaithersburg, MD 20899-6203, USA 2 Maryland Nanocenter, University of Maryland, College Park, MD 20742, USA Gas adsorption on metal nanoparticles is a fundamental step that controls a number gas-solid reaction processes, important in applications ranging from catalysis to gas sensing. Gas absorption depends on the nature of the gas molecules and atomic arrangement of the surface facets. These factors control the gas binding energies, which have not, so far, been directly measured. Here we show that electron-density changes, induced by gas adsorption on the active metal nanoparticle facet, cause shifts in surface plasmon (SP) energies that can be measured by electron energy loss spectroscopy (EELS). We employ a 1 nm-diameter electron beam from a monochromated 80kV electron source to locally (< 2 nm) excite SPs and measure their energies with 100 meV energy resolution. This high spatial and energy resolution allows us to resolve SP energy shifts in the range of a few meV, localized to individual nanoparticle facets. Using this technique in an environmental scanning transmission electron microscope (ESTEM), we are able to map in situ SP responses on different facets (i.e. corners and sides) of individual Au nanoparticles in vacuum, CO, and H2 at various gas pressures. Our results were further confirmed by finite-difference-time-domain (FDTD) simulations for the spatial localization of the electron beam excited SP nanoprobe. Figure 1a shows the STEM dark-field image of a triangular Au nanoparticle on a TiO2 support and the local SP excitation at the corner location A and side location B. Energy loss maps in the Au nanoparticle, obtained by FDTD simulations, reveal the highly-localized excitation volumes near the edge of the particle in both locations (Fig. 1b and c). SP-EELS spectra at both locations were collected in vacuum, and over a range of gas pressures. In the CO environment, the SP energy was observed to shift to higher energies and the magnitude of the energy shift was larger when probing location B than location A. (Fig. 2a). In contrast, in the H2 environment, the SP energy shifted to lower energies and the magnitude of the energy shift was larger when probing location A than location B. To confirm that energy shifts measured with such localized excitation are sensitive only to the electron density variations of the probed location, two models of electron density variations in a 1 nm skin depth at the corner and the side, respectively, were applied in our FDTD simulations. When the electron density varies at the corner, the SP energy varies only for probing location A (Fig.3a) but not when probing location B (Fig.3b). Similarly a variations in the electron density at the side yielded energy shifts in location B (Fig.3c), whereas no changes in location A (Fig. 3d). Therefore, we can conclude that the difference in magnitudes of gas-induced SP energy shifts reports a local phenomenon, and indicates crystallographically-specific absorption of the different gases on the Au nanoparticle facets. A detailed discussion of the SP energies measured for different facet-gas interactions and quantitative measurements of binding energies and number density of molecules adsorbed for different gasses on various facets of Au nanoparticles will be presented. Microsc. Microanal. 21 (Suppl 3), 2015 2054 Figure 1. (a) STEM image, scale bar is 20 nm. Simulated energy-loss intensity maps, normalized at the maximum adsorbed intensity in log scale represents the SP excitation intensity throughout a nanoparticle via a 1 nm beam (mark as a black dot) excitation at (b) the corner and (c) the side. Figure 2. Measured SP energy shifts as a function of gas pressure of (a) CO and (b) H2 when probing at corner location A and side location B. Figure 3. Nanoparticle models with surface charge density change in either (a,b) at the corner or (c,d) the side regions. Rainbow plots below each models are corresponding simulated spectra as a function of fractional surface charge density changes, where the most intense dark red represents the energy loss peak position of the sampled area of the nanoparticle.");sQ1[1026]=new Array("../7337/2055.pdf","Evaluating Adhesion Layers for Plasmonic Nanostructures with Monochromated STEM-EELS and Surface Enhanced Raman Spectroscopy","","2055 2055 doi:10.1017/S1431927615011058 doi:10.1017/S1431927615011058 Paper No. 1026 Microsc. Microanal. 21 (Suppl 3), 2015 Microsc. Microanal. 21 (Suppl 3), 2015 �� Microscopy Society of America 2015 Microscopy Society of America 2015 Evaluating Adhesion Layers for Plasmonic Nanostructures with Monochromated STEM-EELS and Surface Enhanced Raman Spectroscopy Steven J Madsen1, Ai Leen Koh2 and Robert Sinclair1 1 2 Stanford University, Department of Materials Science and Engineering, Stanford, CA 94305-4034 USA Stanford Nano Shared Facilities, Stanford University, Stanford, California 94305-4045, USA Nano-plasmonics is a rapidly growing field of study with applications in energy, healthcare and security. Plasmonic structures frequently consist of noble metals deposited onto a substrate in a `top-down' synthesis. This method requires an adhesion layer, often Cr or Ti, to bind the optically active metal to the substrate. Recent publications have demonstrated the presence of these adhesion layers cause damping of plasmon resonances, observed by optical scattering measurements [1]. Because Raman surface enhancement relies on plasmon resonances near the laser excitation wavelength, these layers can cause reduction in surface enhanced Raman spectroscopy (SERS) intensity [2]. The present study uses monochromated electron energy-loss spectroscopy in a scanning transmission electron microscope (STEM-EELS) to compare plasmon resonances of structures with and without commonly used adhesion layers at the nanometer scale. Raman spectra were collected for the same structures, and variation in SERS enhancement factors correlate with the changes observed in plasmon intensity. Nanopatterning was achieved by electron beam lithography of polymethyl methacrylate resist on either Si (for Raman) or thin SiN windows (for TEM). Samples were created with no adhesion layer, 2nm of Cr or Ti, or vapor deposited mercaptopropyltrimethoxysilane (MPTMS). 30nm of gold, the plasmonic and Raman surface enhancing material, is evaporated on top of the patterned resist and adhesion layer (if present). To avoid film delamination, the resist was not lifted off, leaving a structure shown schematically in cross section in Fig. 1. Fig. 2 shows a plan view STEM dark field image of a typical sample. STEM-EELS data, collected in a monochromated Titan at 300kV, show the presence of several plasmon resonances in the energy range 1.3-2.4eV, the expected range for gold nanostructures [3]. Fig. 3 shows EEL spectra collected from the marked regions in the inset. The samples with no adhesion layer and with MPTMS are similar, while both Ti and Cr result in dramatically reduced peak intensities. Lower energy localized surface plasmon resonance (LSPR) peaks are observed in the 1.5-1.7eV range at the marked location (Fig. 4). Here, gold with no adhesion layer shows the strongest peak, followed by MPTMS with the two metal underlayers Cr and Ti showing the weakest peaks. Samples were coated with a self-assembled monolayer of 4-mercaptopyridine dye and tested using Raman spectroscopy at 785nm and 532nm. Although there are large peaks observed in the 2.0-2.4eV range (Fig. 3), the excitation location has local rotation symmetry, which has been associated with `dark' plasmon modes not accessible with photons [4]. The Raman spectra using a 532nm (2.33eV) laser confirm that there is not strong SERS enhancement for this energy (Fig. 5). By exciting with a 785nm (1.58eV) source near the LSPR energy observed in Fig. 4, Raman signal strengths orders of magnitude greater than those from the 532nm laser can be observed. The SERS enhancement factors for a 1.58eV laser have the same trend as the EELS LSPR peak intensity at the same energy (Figs. 6,4) � the noadhesion-layer sample is greatest, then MPTMS, with Cr and Ti far lower. In conclusion, both Cr and Ti cause significant plasmon damping which was detected using STEM- Microsc. Microanal. 21 (Suppl 3), 2015 Microsc. Microanal. 21 (Suppl 3), 2015 2056 2044 EELS. These layers are also shown to reduce SERS enhancement factor, which is expected to be a direct result of reduced plasmon resonance. MPTMS causes less damping and is thus a more desirable option. [1] TG Habteyes et al, ACS Nano 6 (2012), p. 5702-5709. [2] T Siegfried et al, ACS Nano 7 (2013), p. 2751-2757. [3] FP Schmidt et al, Nano Letters 12 (2012), p. 5780-5783. [4] AL Koh, et al, Nano Letters 11 (2011), p. 1323-1330. [5] This research is supported by the Center for Cancer Nanotechnology Excellence and Translation (CCNE-T) grant funded by NCI-NIH to Stanford University U54CA151459. Figure 1: Cross section schematic of fabricated structures. Samples were generated with no adhesion layer, 2nm Ti, 2nm Cr, or vapor deposited MPTMS. Figure 2 (left): Dark field STEM micrograph of a representtative gold nanostructure. The circular regions are gold nanoparticles depressed relative to the surrounding film. Scale bar 200nm. Figure 3: Low-loss EEL spectra comparing plasmon resonances observed with different adhesion layers. Inset: location from which EELS data was collected marked in red. Figure 4: EEL spectra from the location indicated in inset. The sample without adhesion shows the strongest peak, and MPTMS is the least damping of the adhesion layers. Figure 5: Raman data collected from samples coated with 4-MP dye using a 533nm laser. SERS signal from the dye is not observed at this energy. Vertical axis offset for clarity. Figure 6: Raman data collected using a 785 nm laser. The trend in signal enhancement matches the trend in plasmon resonance observed at the excitation energy via STEM-EELS. Vertical axis offset for clarity.");sQ1[1027]=new Array("../7337/2057.pdf","Unconventional Surface Plasmon Excitations in Bi2Se3","","2057 doi:10.1017/S143192761501106X Paper No. 1027 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Unconventional Surface Plasmon Excitations in Bi2Se3 Cigdem Ozsoy-Keskinbora1, Nahid Talebi1, Hadj M. Benia2, Christoph T. Koch3, Peter A. van Aken1 1 2 3 Stuttgart Center for Electron Microscopy (StEM), MPI for Intelligent Systems, Stuttgart, Germany. Nanoscale Science, MPI for Solid States Research, Stuttgart, Germany. Institute for Experimental Physics, Ulm University, Ulm, Germany. The investigation of surface plasmon excitations has recently become a field of research due to their high potential to be applicable in sensors1, information technologies2, cancer research3 etc. These coherent delocalized electron oscillations are common at metal-dielectric interfaces. However, they also exist in highly doped semiconductors, conducting oxide system or graphene, i.e. in many other systems with high carrier mobility. This poses the question whether such resonances can also be observed at the insulator interfaces. Bismuth Selenide is an insulator material which has been investigated due to its thermoelectric behavior for a long time. In 2009, it became popular after being identified as a topological insulator, which means that it behaves as an insulator in the bulk and metallic at the surface. This also makes Bi2Se3 a potential candidate for surface plasmons resonances. Dirac plasmons in Bi2Se3 were first observed in 2013 with energy in the range of 0.5 � 1 eV4. The Dirac state is not the only reason for the existence of a plasmon resonance in Bi2Se3. The highly anisotropic tetradymites crystal structure has also highly anisotropic dielectric properties5 allowing plasmon excitation. In this study we would like to show energy filtered transmission electron microscopy (EFTEM) and a finite-difference frequency-domain (FDTD) study for investigating Bi2Se3 nanoplates and try to find an explanation for plasmon excitations. The EFTEM study was carried out using the 200 kV FEG-TEM Sub-Electron-Volt-Sub-�ngstromMicroscope (Zeiss SESAM) equipped with an electrostatic monochromator and the in-column MANDOLINE filter with a 0.2 eV slit width. As shown in Fig. 1 localized excitations exist at the surface of the Bi2Se3 crystal and featuring collective modes at different energies. These results supported by FDTD simulations to be able to explain the interplay between edge plasmons and surface plasmons which together make up the total electron energy loss signal shown in Fig.2. References: 1. Wang, Z et al, Analytical Chemistry, 86 (2013), 1430-1436. 2. Kosmeier, S. et al, Scientific Reports. 3 (2013), 1808. 3. Cai, W.; Gao, T.; Hong, H.; Sun, J., Science and Applications 2008, 17-32. 4. Di Pietro et al, Nat Nano 8 (2013),556-560. 5. Esslinger, M. et al, ACS Photonics 1 (2014), 1285-1289. Microsc. Microanal. 21 (Suppl 3), 2015 2058 Figure 1. EFTEM images which shows of localized plasmons excitation in a Bi2Se3 nanoplates. Figure 2. Propagation constant of plasmon polaritons propagating in a channel waveguide made of Bi2Se3.");sQ1[1028]=new Array("../7337/2059.pdf","Imaging Mass Spectrometry Using Ultra-high Mass Resolution Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometer, SpiralTOF","","2059 doi:10.1017/S1431927615011071 Paper No. 1028 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Imaging Mass Spectrometry Using Ultra-high Mass Resolution Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometer, SpiralTOF Takaya Satoh1, Ayumi Kubo1, Masaaki Ubukata2, Naoki Moriguchi3, Hisanao Hazama3, Kunio Awazu3, Michisato Toyoda4 1 JEOL Ltd., Akishima, Japan JEOL USA Inc., Peabody, MA, USA 3 Graduate School of Engineering, Osaka University, Suita, Japan 4 Graduate School of Science, Osaka University, Toyonaka, Japan 2 Imaging mass spectrometry (IMS) has been used for biological applications, to assess the distribution of proteins, peptides, lipids, drugs, and their metabolites in a tissue specimen. IMS has expanded during the last decade using matrix-assisted laser desorption/ionization (MALDI) time-of-flight (TOF) mass spectrometer, which adopted a linear and a reflectron ion optical systems. A reflectron MALDI-TOF mass spectrometer, using a delayed extraction technique, has higher mass resolution than linear MALDI-TOF mass spectrometer. However, its high mass resolution is available only within limited mass range, which isn't sufficient for analysis in low-molecular compounds such as lipids, drugs and drug metabolites. It is necessary to extend flight path length to improve mass resolution and mass accuracy in wide mass range. However, the flight path length of a reflectron TOF mass spectrometer is limited by its instrument size, and is difficult to be extended beyond certain length restricted by the instrument dimension. We developed a MALDI-TOF mass spectrometer with a spiral ion trajectory, SpiralTOF[1], to solve the issue. It has 17 m flight path length within a cubic vacuum housing of approximately 0.6m x 0.6m x 0.7m. The schematic of SpiralTOF, which consists of four toroidal electrostatic sectors, is shown in Figure 1. Each has eight stories made by nine Matsuda plates piled up inside a cylindrical electrostatic sector. The ions pass the four toroidal electrostatic sectors sequentially and revolve along a figure-eight-shaped orbit on a certain projection plane. During multiple revolutions, the ion trajectory shifts perpendicular to the projection plane every revolution cycle, thus generating a spiral trajectory. The flight path length of one revolution is 2.1 m. The total flight path of SpiralTOF was 17 m, which is 5-10 times longer than a reflectron TOF mass spectrometer. SpiralTOF achieved ultra-high mass resolution that could separate isobaric compounds, which differed only 0.1 u each other. The advantage of isobaric separation in IMS will be shown to take the IMS for lipids distribution on mouse brain tissue section as an example. The isobaric mass separation at m/z 820� 825 is shown in Figure 2. Three types of lipid peaks were well separated in mass spectrum and could show the different localization respectively. The high selectivity for drawing mass image is important for understanding clear localization of compounds, especially in low mass region. Further IMS measurements for drugs distribution on mouse brain tissue section will be reported in the presentation. References: [1] T. Satoh, T Sato, A. Kubo, J. Tamura, J. Am. Soc. Mass Spectrom. 22 (2011), p. 797. Microsc. Microanal. 21 (Suppl 3), 2015 2060 Figure 1. Schematic of time-of-flight mass spectrometer with spiral ion trajectory (SpiralTOF). The outer electrode of the left-top electrode is not included to show the ion trajectory (red line). Figure 2. Ultra-high resolution mass spectrum at m/z 820�825 in imaging mass spectrometry for lipids in a mouse brain tissue section. Three types of lipids were well separated in mass spectrum and could draw different mass images from them.");sQ1[1029]=new Array("../7337/2061.pdf","MALDI and LDI Imaging of Forensic Samples by Using A Spiral-Trajectory Ion Optics Time-of-Flight Mass Spectrometer","","2061 doi:10.1017/S1431927615011083 Paper No. 1029 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 MALDI and LDI Imaging of Forensic Samples by Using A Spiral-Trajectory Ion Optics Time-of-Flight Mass Spectrometer Masaaki Ubukata1, A. John Dane1, Robert B. Cody1, Donna Guarrera1, David C. Edwards1, Natasha Erdman1, Masateru Shibata1, Takaya Satoh2 1. 2. JEOL USA, INC., Peabody, MA, USA JEOL Ltd., Akishima, Japan Recently, matrix-assisted laser desorption/ionization (MALDI) imaging techniques have been developed for biological sciences to evaluate and understand the distribution of various chemicals on biological surfaces. In particular, this technique provides useful visual information about the locations of specific chemicals on surfaces. In this work, we explored the use of MALDI imaging for forensically relevant samples that included gunshot residue (GSR), fingerprints and ballpoint ink. These measurements were done using a spiraltrajectory ion optics time-of-flight mass spectrometer (SpiralTOF-MS [1]). This TOF system has a unique 17m flight path that provides ultrahigh-resolution mass spectra over a wide m/z range, even if the sample is not perfectly flat. Additionally, the m/z axis is very stable over the long measurement times required for MALDI imaging. GSR samples were obtained on an electrically conductive adhesive that was pressed against the back of a shooter's hand after a handgun was discharged. The fingerprint samples were collected on electrically conductive ITO glass slides from both a smoker and a non-smoker. The smoker's fingerprint was collected immediately after smoking a cigarette. 2,5-dihydroxybenzoic acid (DHB) was used as the MALDI matrix. This matrix was dissolved in methanol at a fixed concentration of 30 mg/mL. 1-2 mL of DHB solution was sprayed on each sample surface by using a commercial airbrush. Samples were then analyzed by using a high-resolution MALDI-TOFMS instrument. A polypropylene glycol (PPG) mixture of average MW 425 and 1000 were used as an external exact-mass reference standard. The GSR sample surface was analyzed for inorganic element distributions. As a starting point, we evaluated the mass accuracy of the SpiralTOF-MS with external calibration using two different DHB ions, C7H5O3+ and C7H6O4Na+, in the measured spectrum. Because their monoisotopic ions were saturated, the [M+1] isotope ions were used, and the calculated mass error for each was +6.4 mDa and 5.1 mDa, respectively. As a result, it was not necessary to use internal calibration for qualitative MALDI imaging measurements on the SpiralTOF-MS. Afterwards, the inorganic particle distributions for barium, barium oxide, lead, bismuth, calcium oxide, etc. were selected for imaging on the surface. As it turns out, the 138Ba m/z 137.9016 has a similar mass to the DHB [M+1] isotopic ion m/z 138.02671 (C7H5O3). Although the difference in their mass values is only 0.125 Da, the SpiralTOF-MS provided full separation for both ion peaks which in turn produced completely different images for each analyte on the surface. The distribution of the 138Ba m/z 137.9016 was randomly distributed across the surface while the DHB m/z 138.02671 was homogeneously present across the matrix surface. Next, we examined the nicotine distribution on both a smoker's and non-smoker's fingerprints (Figure 1). We focused on determining the chemical composition of the observed m/z 163.1224 on the fingerprint by using the accurate mass values with internal calibration method. DHB matrix ions were used as the internal calibrant for this accurate mass measurement. The mass error of the m/z 163.1224 was just -0.6 mDa when compared to the protonated molecule of a nicotine standard (m/z 163.1230, C10H15N2) using internal calibration. There was a strong nicotine distribution on the smoker's Microsc. Microanal. 21 (Suppl 3), 2015 2062 fingerprint that was not detected on the non-smoker's fingerprint. This is a very reasonable result considering the fact that the smoker is exposed to nicotine when smoking. The ballpoint ink samples (Figure 2) were subjected to a gold vapor deposition in order to improve the conductivity of the paper. No matrix was added to this sample surface (LDI imaging). The ink consisted of crystal violet as the main component and was easily detected directly on the surface of the handwriting sample with the gold vapor deposition. [1] T. Satoh, T Sato, A. Kubo, J. Tamura, J. Am. Soc. Mass Spectrom. 22(2011) 797-803 MALDI imaging of m/z 163.1 (Nicotine, [M+H]+) Smoker's fingerprint Non-smoker's fingerprint Figure 1. MALDI imaging of protonated nicotine on both smoker's and non-smoker's fingerprints. Figure 2. Imaging of hidden organic dyes for ballpoint pen ink obliterated with permanent marker.");sQ1[1030]=new Array("../7337/2063.pdf","Synthesis of Monolayer Molybdenum Disulfide and ToF-SIMS Characterization","","2063 doi:10.1017/S1431927615011095 Paper No. 1030 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Synthesis of Monolayer Molybdenum Disulfide and ToF-SIMS Characterization Xinqi Chen and Michael Ambrogio NUANCE Center, Northwestern University, Evanston, IL 60208, USA Two dimensional (2D) materials have attracted much attention in the past decade. This family of materials includes metallic graphene and semiconducting transition metal dichalcogenides such as MoS2. Due to the excellent mechanical flexibility, electron transportation property, and transparency, 2D materials have great potential applications in electronics, sensor, and supercapacitor. Mechanical exfoliation and chemical vapor deposition (CVD) are two common methods to prepare the monolayer and a few atomic layer films. Because of the high yield and good quality of the monolayer film, the CVD method becomes the main stream to prepare 2D materials. We prepared the MoS2 film with a similar method to reference 1 except for lower heating temperature and simpler cleaning procedure of the substrate. A 1 cm by 1 cm piece of silicon wafer with 500 nm SiO2 was put in acetone and ultrasonicated for 5 min. Then the substrate was cleaned with oxygen plasma for 20 min. 125 mg of sulfur and 30 mg of MoO3 were placed into separate ceramic boats. The sulfur was placed at the upstream and the MoO3 was placed at the center of the tube furnace. The polished-side of the wafer was put down on top of boat containing MoO3, using two Si wafers as spacers to provide excess ventilation for vapor to escape. Ar gas was flowed through CVD tube (~20 psig) for 15 min at room temperature. Then furnace was set to heat up to 650�C at a rate of 15�C per min and maintained for 15 min. Then it was turn off and the sample was allowed to cool down to room temperature. The sample was observed with SEM and the image (Fig. 1) shows the MoS2 grain size varies from a few micrometers to a hundred micrometers. The central area was covered with a continued film and the edge was decorated with triangular monolayer MoS2 crystalline. The Raman and photoluminescence (PL) spectra were shown in Fig. 2 and Fig. 3 separately. The Raman peak frequency difference between E2g and A1g mode is 20.4 cm-1. The PL peak is located at 668 nm and shows strong intensity. The results show that the film has atomic layer and high quality of crystalline. It indicates that this procedure is highly efficient to prepare high quality MoS2 atomic film. To further characterize the film, time of flight secondary ion mass spectroscopy (ToF-SIMS) was conducted on this sample. The ions of Mo+, S-, and Si+ have been mapped and the ion mass images are illustrated in Fig. 4. The results show that molybdenum and sulfur have been uniformly distributed in the film and there is a clear sharp boundary between the MoS2 crystalline and SiO2 substrate. It further indicates the high quality of the atomic layer film. In summary, a large scale of atomic layer MoS2 film has been prepared with a simple CVD method. The Raman and PL results show the high quality of the crystalline. At the first time, we showed the ion distribution with ToF-SIMS. Depending on the applications, the doping of 2D materials is often conducted to alter the electron transport property. Due to the ppm level of sensitivity and high lateral resolution, ToF-SIMS will be proven to be an informative characterization tool for 2D materials, especially for doped film. Microsc. Microanal. 21 (Suppl 3), 2015 2064 R Reference: [ A van de Zande et al, Nature Ma [1] er a aterials 12 (2 2013), P554 . F Figure 1. SEM image of CVD grown MoS2 film n Figu 2. Raman spectrum w excitation of ure n with 514. nm laser .5 Figure 3. Photolumin nescence spe ectrum with excitation o 514.5 nm laser of Figure 4. ToF-SIMS images of ion Mo+, S-, and Si+ (sca bar = 10 um) 4 S ale");sQ1[1031]=new Array("../7337/2065.pdf","An Investigation of the Pancreatic Islet Tumor Microenvironment with Time-of-Flight Secondary Ion Mass Spectrometry","","2065 doi:10.1017/S1431927615011101 Paper No. 1031 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 An Investigation of the Pancreatic Islet Tumor Microenvironment with Time-ofFlight Secondary Ion Mass Spectrometry Blake M. Bluestein1, Fionnuala Morrish2, David Hockenbery2, Lara J. Gamble1 1. 2. Department of Bioengineering, NESAC/BIO, University of Washington, Seattle, WA, USA Fred Hutchinson Cancer Research Center, Seattle, WA, USA Imaging time-of-flight secondary ion mass spectrometry (ToF-SIMS) provides molecular (chemical) information with subcellular spatial resolution. Here, imaging ToF-SIMS is used to analyze the tumor microenvironment in biopsies from a mouse model (Myc/p53-/-) with Myc-dependent inducible and regressible pancreatic -cell neoplasia. The oncogene Myc is overexpressed in a wide variety of human cancers and strongly effects cellular metabolism, including lipid metabolism [1,2] It is hypothesized that chemical mapping of neoplastic tissues will reveal uptake and synthesis of fatty acids within the islet tumor induced by Myc. Lipid analysis via imaging ToF-SIMS will provide further insight into metabolism in the tumor microenvironment. While imaging ToF-SIMS analysis of tumor tissue will provide a new perspective by visualizing tumor progression/regression, the system itself can also act as a model system for investigating stroma-tumor interactions in cancerous tissues. Pancreatic tissues were harvested and frozen in optimal cutting temperature (OCT) at 6 days post Myc induction. 4 �m cryosections were cut, with one used for H&E staining, one for ToF-SIMS analysis, and another for immunohistochemistry. High mass and high spatial resolution data was acquired with the pulsed 25 keV Bi3+ ion beam rastered over a 1 mm x 1 mm area (1280 x 1280 pixels). ROIs of the tumor and stromal tissue were then investigated further with imaging principal components analysis (PCA) to identify peaks that correspond to species of interest. Regions identified by analysis and PCA were then crossreferenced against immunohistochemical and H&E images to differentiate the tumor area from the surrounding tissue. ToF-SIMS data suggests a preferential uptake of fatty acids 18:3, 18:2 and possibly phosphoinositol from lower mass molecular fragments in the negative polarity within the tumor. The 6 day Myc-induced islet tumor exhibits a signal of myristic acid (C14:0), possibly a product of de novo fatty acid synthesis within the tumor. The tumor also exhibits an increased localization of sphingomyelin fragments and vitamin E compared to the surrounding tissue. Interestingly, the data shows an absence of Mg+ within the islet tumor (Figure 1C) and small, higher signal regions on the periphery of the tumor (Figures 1A and 2). These peripheral tumor regions exhibit an increased, localized signal of CN-, CNO-, C7H10O+, and Fe+. An ion image of CN- is shown in Figure 1A. Image PCA was applied to the entire image, illustrated in Figure 2 and reveals different chemistries within the tumor and surrounding acinar tissue. PCA was also applied to the selected tumor region images to spatially and chemically analyze within the tumor to compare different chemistries between different tumor sizes, which could indicate different tumor stage development, within the same tissue. Image PCA separated the high signal regions at the periphery of tumors from the bulk of the tumors, but further histologic correlations are needed to discern if these structures are inflammatory zones, mitochondrial dense regions, or related to vasculature. Continuing work includes cell staining to identify these structures (work by our collaborators at the FHCRC). Once these localized areas have been better defined, a comparison of the regions to ToF-SIMS identified chemistry may aid in a valuable understanding of the Myc oncogene and its effect on pancreatic -cell neoplasia. Microsc. Microanal. 21 (Suppl 3), 2015 2066 References: [1]Nilsson, J.A. and J.L. Cleveland, Oncogene, 22, p. 9007. [2]Morrish F. et al, J Biol Chem 285 (2010), p. 36267. Figure 1. Ion images of two separate tumors. The ion image CN- in A indicates protein rich regions within the tumor defined area, B shows the tumor regions rich with a fully saturated fatty acid, C14:0 and the ion image C displays the absence of Mg+ within the tumors. All images were binned by a factor of 4 pixels to increase contrast for this figure. Figure 2. Image PCA distinctly separates the tumors from the surrounding tissue. The positive scores (top, left image) corresponds to the positive loadings (graph at right, above y=0), where Fe+, protein rich regions, and lipid fragments separate the tumor from the outer tissue. Negative scores (bottom, left image) correspond to negative loadings (graph at right, below y=0) where Mg+ and hydrocarbon fragmented regions are observed in the tissue surrounding the tumor.");sQ1[1032]=new Array("../7337/2067.pdf","Manganese Segregation at Antiphase Boundaries Connecting ZrO2 Pillars in ZrO2-La2/3Sr1/3MnO3 Pillar-Matrix Structures","","2067 doi:10.1017/S1431927615011113 Paper No. 1032 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Manganese Segregation at Antiphase Boundaries Connecting ZrO2 Pillars in ZrO2-La2/3Sr1/3MnO3 Pillar-Matrix Structures Dan Zhou1, Wilfried Sigle1, Marion Kelsch1, Hanns-Ulrich Habermeier2, and Peter A. van Aken1 1. Max Planck Institute for Intelligent Systems, Stuttgart Center for Electron Microscopy, Stuttgart, Germany 2. Max Planck Institute for Solid State Research, Stuttgart, Germany Self-assembled vertically aligned nanocomposite thin films with two immiscible components heteroepitaxially grown on single crystal substrates1-4 have attracted tremendous research interest due to the advantages of utilizing both component functions and tuning material properties with high interface-tovolume ratio, hetero-epitaxial strain, or modifying the cation valence state. Anomalous magnetic anisotropy and modifications to the electric transport properties of La2/3Sr1/3MnO3 (LSMO) has been reported to be achieved by introducing non-magnetic ZrO2 pillars5-7. Whereas up to now, only macroscopic properties of ZrO2-LSMO pillar-matrix systems (charge transport and magnetism) have been studied, microscopic properties at the atomic level were not studied at all. Here we use high-angle annular dark-field (HAADF) imaging, annular bright-field (ABF) imaging and electron energy-loss spectroscopy (EELS) in aberration-corrected scanning transmission electron microscopy (STEM) to reveal the structure, composition and valence state at atomic resolution for the pillar-matrix interface region and antiphase boundaries (APB) connecting adjacent ZrO2 pillars in ZrO2-LSMO pillar-matrix structures with emphasis on the antiphase boundaries. Details about the pillar-matrix interface region can be found in our previous report.8 Atomic resolution EEL spectrum imaging (SI) reveals the substantial interdiffusion at the ZrO2-LSMO interface with Mn replacing Zr in ZrO2 (thus stabilizing the cubic or tetragonal phase) and Zr replacing Mn atoms in LSMO. Charge balance requires the combination of Mn valence state change and oxygen vacancy formation which are observed to segregate at the interface. As one way to relax the strain generated from the misfit of two components, three types of Mn segregated APBs were found connecting adjacent pillars. The crystal lattices on either side of the APB wall are displaced by an antiphase shift as can be seen in Figure 1, which includes two types of APBs. The Mn valence state in the channel was found to be the same as in the matrix. The APB wall planes are either {110} or {310}. The arrangement of heavy and light elements is revealed by simultaneously acquiring HAADF and ABF images, as shown in Figure 1c,d, and HAADF and EELS spectrum imaging. The role of the pillars and APBs regarding elastic strain and local electric fields will be discussed. The spin, charge, and orbital ordering in LSMO are extremely sensitive to local structural and elemental variations. Thus, our results provide a basis for understanding the origin of these properties.9 References: 1. Chen, A. P.; Bi, Z. X.; Hazariwala, H.; Zhang, X. H.; Su, Q.; Chen, L.; Jia, Q. X.; MacManus-Driscoll, J. L.; Wang, H. Y. Nanotechnology 2011, 22, (31). 2. Imai, A.; Cheng, X.; Xin, H. L. L.; Eliseev, E. A.; Morozovska, A. N.; Kalinin, S. V.; Takahashi, R.; Lippmaa, M.; Matsumoto, Y.; Nagarajan, V. Acs Nano 2013, 7, (12), 11079-11086. Microsc. Microanal. 21 (Suppl 3), 2015 2068 3. Zheng, H.; Zhan, Q.; Zavaliche, F.; Sherburne, M.; Straub, F.; Cruz, M. P.; Chen, L. Q.; Dahmen, U.; Ramesh, R. Nano Lett 2006, 6, (7), 1401-1407. 4. Liu, H. J.; Chen, L. Y.; He, Q.; Liang, C. W.; Chen, Y. Z.; Chien, Y. S.; Hsieh, Y. H.; Lin, S. J.; Arenholz, E.; Luo, C. W.; Chueh, Y. L.; Chen, Y. C.; Chu, Y. H. Acs Nano 2012, 6, (8), 6952-6959. 5. Gao, Y. Z.; Zhang, J. C.; Fu, X. W.; Cao, G. X.; Habermeier, H. U. Prog Nat Sci-Mater 2013, 23, (2), 127132. 6. Jin, Y.; Yao, X. C.; Jia, R. R.; Cao, S. X.; Zhang, J. C. J Supercond Nov Magn 2013, 26, (5), 1621-1624. 7. Gao, Y. Z.; Zhang, J. C.; Cao, G. X.; Mi, X. F.; Habermeier, H. U. Solid State Commun 2013, 154, 46-50. 8. Dan Zhou, W. S., Eiji Okunishi, Yi Wang, Marion Kelsch, Hanns-Ulrich Habermeier and Peter A. van Aken. APL Materials 2014, 2, (12), 10. 9. The research leading to these results has received funding from the European Union Seventh Framework Program [FP/2007-2013] under grant agreement no 312483 (ESTEEM2). Figure 1 (a) HAADF image of a plan-view 80 mol% LSMO-20 mol% ZrO2 sample showing two ZrO2 pillars connected by boundaries. (b) Magnified image of the boundary region shown in (a). Simultaneously acquired HAADF image (c) and ABF image (d) of the orange marked region in (b) including APBs with {110} and {310} boundary planes, respectively.");sQ1[1033]=new Array("../7337/2069.pdf","Atomic-Resolution EELS Study of Titanium Dopant Effects of Ca3Co4O9 Thin Film","","2069 doi:10.1017/S1431927615011125 Paper No. 1033 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic-Resolution EELS Study of Titanium Dopant Effects of Ca3Co4O9 Thin Film Xuan Hu, Patrick Phillips, Serdar Ogut, Robert Klie Department of Physics, University of Illinois at Chicago, Chicago, IL 60607 Thermoelectric materials have attracted significant attention over the last few decades. As one of the outstanding thermoelectric oxide materials, the incommensurately layered Ca3Co4O9 (CCO) exhibits a high in-plane Seebeck coefficient and a high thermoelectric figure of merit at high temperature. Many studies have reported that substitutional doping will increase the Seebeck coefficient [1], but understanding of dopant effects on local structure and electronic properties is still lacking. Using electron energy-loss spectroscopy (EELS) and first-principle calculations, the atomic and electronic structure of Ti dopants can be measured and the dopant effects on the thermoelectric properties can be analysed. Ti-doped CCO thin films (Ca3Co3.8Ti0.2O9) were deposited by Pulsed Laser Deposition (PLD). All the images and EEL spectra were acquired by using an aberration-corrected JEOL ARM200CF with a 200 kV cold field emission gun and post-column Gatan Enfina EELS spectrometer. Figure 1a-c) shows the EELS image of Co, Ti, and Ca signals, respectively. The integrated signal intensity of Ti L-edge has been used to determine the position of Ti dopants. We find that the Ti dopants mainly replace Co atoms in the Ca2CoO3 subsystem. We have analysed the near edge fine-structure of the Ti L-edge and compared it to the shape of reference spectra for Ti4+ and Ti3+ [2]. We determine the valence state of Ti to be 4+. Figure 1c) shows the Co L3/L2 ratio as a function of concentration ratio of Ti/Ca. As the Ti concentration increases, the intensity ratio of Co L3/L2 remains mostly unchanged. Since the Co white lines ratio is directly linked to the Co valence state [3], we determine the Co valence state of CoO2 subsystem to be (3.4�0.2) and the Co valence in the Ca2CoO3 subsystem to be (2.8�0.2). The results are close to the values of the Co valence of pristine CCO bulk [4]. This demonstrates that the Ti doping does not influence the Co valence, especially the mixed-valence state of Co in CoO2 layer, which means that hole concentration in the p-type CoO2 layers remains unchanged and the effect of Ti doping on the Seebeck coefficient should be negligible.[5] The experimental results will next be tested using first principles DFT modelling. Figure 2a) shows the unit cell of pristine 5/3 CCO with 66 atoms, which is used in our first-principle calculations. Based on this structure, the Ti dopants are studied by substituting Ti atom for one of the Ca atoms and Co atoms. All the 12 Ca sites and 16 Co sites have been considered. The calculated total energies show that Ti dopants are preferred on the Co sites of Ca2CoO3 subsystem compared to the Co sites of the CoO2 layer, which is in agreement with our experimental observation. Figure 2b) shows the partial density of states (PDOS) of angular �momentum resolved d-orbitals of the Ti atom, substituting the Co atom of the Ca2CoO3 subsystem, where all the Ti 3d orbitals are empty. In addition, the PDOS of the Ti p-orbitals shows that the Ti 3p orbitals are completely occupied. We can conclude that the valence of Ti dopant is 4+. However, the effects of Ti dopants and their locations on local electronic structure is still being examined. In this presentation, we will discuss out latest EELS result combined with first-principle calculations to analyse how Ti dopants influence the thermoelectric performance of CCO thin film [6]. Microsc. Microanal. 21 (Suppl 3), 2015 2070 2058 References: [1] C. J. Liu, L. C. Huang, and J. S. Wang, Appl. Phys. Lett. 89 (2006), p. 204102. [2] A. Ohtomo et al, Nature 419 (2002), p. 378-380. [3] Z. L.Wang et al, Micron 31 (2000), p. 571-580. [4] G. Yang et al, Physical Review B 78 (2008), p. 153109 [5] W. Koshibae et al, Physical Review B 62 (2000), p. 6869-6872. [6] This work was supported by a grant from the National Science Foundation (DMR-1408427). Figure 1: a) b) c) d) Figure 1. Ti doped Ca3Co4O9 thin film along [110]: a-c) EELS images for Co, Ti, and Ca signals, respectively. The red rectangle is the area of CoO column of Ca2CoO3 subsystem. d) Intensity ratio of Co L3/L2 as a function of concentration ratio of Ti/Ca, for Co in CoO2 layer (blue stars) and Co in CoO column of RS subsystem (red spots). Figure 2: a) b) Figure 2. First-principle simulation of Ti doped Ca3Co4O9: a) the unit cell of pristine 5/3 CCO with 66 atoms; b) the partial density of states (PDOS) of m-resolved d orbitals of Ti atom, which substitutes the Co atom of RS subsystem.");sQ1[1034]=new Array("../7337/2071.pdf","Atomic-Scale Quantitative and Analytical STEM Investigation of Sr--Doped La2CuO4 Multilayers","","2071 doi:10.1017/S1431927615011137 Paper No. 1034 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic-Scale Quantitative and Analytical STEM Investigation of Sr--Doped La2CuO4 Multilayers Y. Wang1, W. Sigle1, D. Zhou1, F. Baiutti2, G. Logvenov2, G. Gregori2, G.Cristiani2, J.Maier2, P.A. van Aken1 1. Max Planck Institute for Intelligent Systems, Stuttgart Center for Electron Microscopy, Stuttgart, Germany 2. Max Planck Institute for Solid State Research, Stuttgart, Germany Superconductivity in copper oxides arises when a parent insulator compound is doped beyond some critical concentration [1]. In the case of La2CuO4 (LCO), high-Tc superconductivity is obtained either by substituting La3+ with Sr2+ or by inserting interstitial O2- [2]. At internal interfaces, the enhancement of the superconducting critical temperature is influenced by the interfacial structure [3]. Recently, by using atomic layer-by-layer oxide molecular beam epitaxy (MBE), we have fabricated Sr--doped LCO multilayered structures on LaSrAlO4 (LSAO) substrate, in which some atomic planes of LaO were intentionally substituted by SrO. By varying the spacing between the LCO and SrO layers high-Tc superconductivity (~ 40 K) was obtained [4]. Here we lay emphasis on the detailed and quantitative STEM analysis. For the present contribution, we combine atomic-resolved quantitative STEM imaging with analytical STEM-EELS/EDX analysis to enhance understanding of high-Tc superconductivity at Sr--doped LCO interfaces with respect to the local lattice and oxygen octahedral distortion, as well as cation and electron hole redistribution. STEM investigations were performed using a JEOL ARM 200CF scanning transmission electron microscope equipped with a cold field emission electron source, a D-COR probe corrector, a 100mm2 Centurio EDX detector, and a Gatan GIF Quantum ERS spectrometer. Figure 1 (a) shows a cross-sectional HAADF STEM image of Sr--doped LCO multilayers, revealing that LCO and the LSAO substrate exhibit perfect epitaxy and show no local structural defects at the -doped interfaces. Due to the difference in atomic number (ZSr=38, ZLa=57), the atomic columns dominated either by La or Sr give rise to different contrast in the HAADF image. In the Sr--doped region the atomic column intensity is significantly lower than in pure LCO. The average image intensity profile in growth direction shows that at the Sr--doped region the image intensity has a relatively sharp drop of intensity followed by a slowly increasing intensity pointing to an asymmetric Sr distribution. Atomic-resolved HAADF and ABF images, which were simultaneously acquired at the Sr--doped region, are presented in Fig. 1 (b) and (c). The local lattice and copper-apical-oxygen distortions were quantitatively evaluated by image analysis. A detailed study on the redistribution of Sr and of electron holes at the interface was performed by a combination of atomic-resolved STEM-EELS/EDX. The Sr-L EDX and Sr-L2,3 EELS (Fig.2 a) line-scan profiles show that Sr is redistributed a few layers in LCO and has an asymmetric concentration profile. The electron holes across the Sr--doped interfaces were characterized by analysis of the O-K near-edge fine structure, as presented in Fig. 2 (b) and (c). These findings, suggesting a rather complex charge rearrangement mechanism, will be discussed. [5] Microsc. Microanal. 21 (Suppl 3), 2015 2072 References: [1] P.A.Lee et al., Rev.Mod.Phys. 78 (2006), p.17. [2] B.O.Wells et al., Science 277 (1997), p.1067. [3] A.Gozar et al., Nature 455 (2008), p.782. [4] F.Baiutti et al., submitted (2015). [5] The research leading to these results has received funding from the European Union Seventh Framework Program under Grant Agreement 312483-ESTEEM2 (Integrated Infrastructure Initiative I3). U. Salzberger is particularly acknowledged for TEM specimen preparation. Figure 1. (a) HAADF STEM image of Sr--doped LCO multilayers epitaxially grown on a LSAO substrate. Simultaneously acquired (b) HAADF and (c) ABF images of one Sr--doped area, on which the atomic columns have been located for quantification of the lattice and copper-apical-oxygen distortions. Figure 2. (a) Integrated Sr-L2,3 EELS line profiles across 4 Sr--doped regions. (b) O-K edge from a Sr-doped region and from LCO. (c) Integrated O-K pre-edge intensity profile across 4 Sr--doped regions.");sQ1[1035]=new Array("../7337/2073.pdf","Electronic Structure of New Line Defect in Strained NdTiO3 on SrTiO3","","2073 doi:10.1017/S1431927615011149 Paper No. 1035 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electronic Structure of New Line Defect in Strained NdTiO3 on SrTiO3 Jong Seok Jeong1, Mehmet Topsakal1, Peng Xu1, Renata M. Wentzcovitch1, Bharat Jalan1 and K. Andre Mkhoyan1 1. Department of Chemical Engineering and Materials Science, University of Minnesota, Minneapolis, MN 55455, U.S.A. NdTiO3 is a Mott insulator with an orthorhombic perovskite structure [1] and NdTiO3/SrTiO3 heterostructure is of interest for a new platform material to study two-dimensional electron gases which can be produced by charge rearrangement and/or atomic reconstruction at the interface [2,3]. Our recent study has shown that high-quality stoichiometry-controlled NdTiO3 film can be grown on SrTiO3 by a hybrid molecular beam epitaxy approach and can generate high carrier density in room temperature [4]. Bulk orthorhombic NdTiO3 (Pbnm) has lattice parameters a=5.525 �, b=5.659 �, and c=7.791 �. It is useful to consider tetragonal lattice parameters at=bt=d110=3.953 � and ct=d002=3.896 �--where the subscript "t" represents tetragonal lattice--since the tetragonal unit in the NdTiO3 lattice is analogous to the SrTiO3 cubic lattice (a=3.905 �). When it is grown on a SrTiO3, it yields strain to accommodate the lattice mismatch with the substrate. According to our scanning transmission electron microscopy (STEM) analysis, the NdTiO3 film grows on the SrTiO3 (001) with a specific crystallographic orientation--atct plane as in-plane of the heterostructure and bt as out-of-plane direction. Interestingly it is found that the strain in the NdTiO3 film interplayed with stoichiometry supplies driving force to form a new type of line defect to adapt the strain (Figure 1). Here, we present the analysis of the atomic and electronic structure of the line defect using analytical STEM with the assistant of density functional theory (DFT) calculation. High-angle annular dark-field (HAADF) STEM imaging, energy dispersive X-ray spectroscopy (EDX), and electron energy-loss spectroscopy (EELS) were performed using an aberration-corrected monochromatic FEI Titan G2 60-300 STEM equipped with a CEOS DOCR probe corrector, Super-X EDX spectrometer, and Gatan Enfinium ER spectrometer. First-principles calculations were performed in the framework of DFT as implemented in Vienna ab initio Simulation Package (VASP) code [5]. The defect has been detected in HAADF-STEM images viewed only along the ct-axis, indicating that the line defect is parallel to the ct-axis (Figure 1), which is in agreement with energy calculation results. The determined structure of the defect is described in Figure 1b and is known to have a relatively relaxed Ti-deficient structure (from EDX, not shown here). This defect structure is energetically more favorable than a strained structure without the defect. The calculated oxygen partial-density of states (DOS) and experimental O K core-loss EELS spectra show good agreement (Figure 2) and further confirm the validity of the determined defect structure. EELS and DFT calculation results indicate that Ti atoms in the core of the line defect have Ti4+ character in contrast to Ti3+ in the bulk structure, meaning alterations of the DOS and narrowing of NdTIO3 band gap occurred with defect formation. We also expect that other perovskite materials can possibly have this type of defect when they have proper amount of strain and certain crystallographic growth direction on substrates [6]. References: [1] AS Sefat et al, Physical Review B 74 (2006), p. 104419 [2] A Ohtomo et al, Nature 419 (2002), p. 378 Microsc. Microanal. 21 (Suppl 3), 2015 2074 [3] N Nakagawa et al, Nature Materials 5 (2006), p. 204 [4] P Xu et al, Applied Physics Letters 104 (2014), p. 082109 [5] G Kresse and J Hafner, Physical Review B 47 (1993), p. 558 [6] This work was supported in part by C-SPIN, one of the six centers of STARnet, a Semiconductor Research Corporation program, sponsored by MARCO and DARPA. STEM analysis was carried out in the Characterization Facility of the University of Minnesota, which receives partial support from NSF through the MRSEC program. Figure 1. (a) HAADF-STEM image of a line defect in a NTO film on a STO substrate, where the NTO film is viewed along the ct axis. (b) Perspective atomic model for the defect and its at, bt, and ct axis projections. Yellow spheres are Nd and green octahedrons represent Ti coordinated by six O. Figure 2. (a) ct axis-projected defect supercell (supercell boundary is presented by the pink box) for DFT calculation, where each position of O of interest are indicated. (b) Experimental O K core-loss EELS fine structures and calculated O partial-DOS; "Exp-On" and "Exp-Off" represent experimental EELS on and off the defect and the calculated partial-DOS for the O positions in (a) are accordingly labeled. The calculated O partial-DOS for pure NdTiO3 is also presented.");sQ1[1036]=new Array("../7337/2075.pdf","Characterization of Ultra Low-K Dielectric Materials by STEM EELS Elemental Mapping","","2075 doi:10.1017/S1431927615011150 Paper No. 1036 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of Ultra Low-K Dielectric Materials by STEM EELS Elemental Mapping Wayne W. Zhao, Michael Gribelyuk, and Jeremy D. Russell Engineering Analysis - Physical, Technology Development & Yield Engineering, GLOBALFOUNDRIES, Malta, New York, USA. Various materials including species, which are sensitive to the electron beam, have been introduced into the Si-based wafer process and integration. This requires the development of the novel analytical techniques, which would be capable to evaluate these structures without causing the beam damage. For example, the ultra-low-k (ULK, k is the dielectric constant) SiOC(H) is one of such materials, which is used extensively in the semiconductor industry to enhance properties of the copper-based advanced back-end-of-line (BEOL) interconnects[1]. Control and further development of these films requires information about their elemental composition with high spatial resolution [2~6]. In the electron microscope the extent of the electron beam damage depends on the accelerating voltage, the electron beam density and the total electron dose. The electron beam density was 9.987 nA / angstrom2, (calibrated with a Faraday cup). The electron probe size was 5 angstrom. The work has focused on the determination of the maximum dose, which would cause minimum electron beam damage of the SiCO(H) during analysis. The electron energy loss spectroscopy (EELS) technique with the Gatan Quantum GIF was used for the 2-D mapping of the elemental composition of SiCO(H), as EELS is sensitive to the light atomic weight elements and provides the minimum readout time [7~8]. A series of analysis was performed whereby the acquisition time at each pixel was varied. The size of the map and all other parameters were kept constant. We used neighboring identical structures of the SiCO(H) in the same TEM sample in each analysis. Demonstrated here is one example of the analysis of the SiCO(H) material, which was used to build a SRAM device. Figure 1 shows a STEM image and the individual EELS elemental maps acquired by Digiscan Spectrum Imaging [9], with a pixel time of 10 millisecond. The individual elemental line profiles of Si, C, O were extracted from each analysis and compared. They were used to determine the maximum dose, which does not cause the electron beam damage. In addition, the same maps were acquired by the TIA acquisition system. Since the readout time in TIA is longer compared to that when the Gatan Digiscan system is used, we reduced the acquisition time so that the dose is kept the same. The maps acquired by the Digiscan and TIA using the same dose were compared to evaluate the effect of the reduced acquisition time on the noise in the extracted elemental profiles. (Thanks go to Cynthia Martin for her excellent supports in preparing TEM samples nicely.) References: [1] Maex K. et al.: J. Appl. Phys. Vol. 93, (2003), pp. 8793�8800. [2] W. Zhao, et al., Microscopy & Microanalysis, Vol. 20 (Supplement 3), (2014), pp.362~363. [3] W. Zhao, M. Gribelyuk, et al., Proc. 38th International Symposium for Testing and Failure Analysis, (2012), pp. 347~355. Microsc. Microanal. 21 (Suppl 3), 2015 2076 [4] W. Zhao, et al., Microscopy & Microanalysis, Vol. 20 (Supplement 3), (2014), pp.1000~1001. [5] W. Zhao, et al., Microscopy & Microanalysis, Vol. 19 (Supplement 2), (2013), pp.902~903. [6] W. Zhao, Symp. Proc. the Material Research Society, 2002 Fall Meeting, (2002), Vol. 738, pp. G7.15.1~6. [7] R. Leapman and J. Hunt, Microscopy, Microanalysis, Microstructure, Vol. 2, (1991) pp. 231-244. [8] H. Harrach, et al., Microscopy & Microanalysis, Vol. 16 (Supplement 2), (2010), pp.1312~1313. [9] http://www.gatan.com/products/sem-imaging-spectroscopy/digiscan-ii-system. a b c d e f Figure 1. STEM EELS elemental mapping by Digiscan Spectrum Imaging, (a) HAADF-STEM image, and STEM-EELS elemental mapping by Digiscan spectrum imaging; (b) C (red); (c) N (cyan); (d) Si (brown); (e) O (green); and (f) Cu (yellow).");sQ1[1037]=new Array("../7337/2077.pdf","Investigation of Co-Doped BaTiO3 by Atomic-Resolution EELS","","2077 doi:10.1017/S1431927615011162 Paper No. 1037 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Investigation of Co-Doped BaTiO3 by Atomic-Resolution EELS Desai. Zhang1, T. Aoki2, P. Ponath3, A. B. Posadas3, A. A. Demkov3 and D. J. Smith4 1. 2. School of Engineering for Matter, Transport and Energy, Arizona State University, Tempe, AZ 85287 LeRoy Eyring Center for Solid State Science, Arizona State University, Tempe, AZ 85287 3. University of Texas at Austin, Department of Physics, Austin, Texas 78712 4. Department of Physics, Arizona State University, Tempe, AZ, 85287 BaTiO3 (BTO) is a prototypical ferroelectric perovskite that is an attractive candidate for applications such as negative-capacitance field effect devices [1]. Using SrTiO3 (STO) grown by molecular beam epitaxy as a buffer layer, high quality BTO films can be grown on Si (001) by atomic layer deposition (ALD) [2]. In addition, when BTO is doped with cobalt (Co), its room-temperature dielectric permittivity increases with doping [3]. In this work, Co-doped BTO (BCTO) films were grown on STO buffer layers on Ge (001) substrates. For microscopy characterization, cross-section TEM samples were prepared using mechanical polishing, dimpling and argon-ion-milling. The film growth quality was checked by high-resolution TEM images, such as Fig. 1, recorded using a JEM-4000EX operated at 400 keV with a structural resolution of ~1.7�. In addition to the high quality epitaxial layer, a flat STO top surface and an abrupt transition from STO to BCTO are visible. To determine the Co dopant distribution, the sample was examined using Nion UltraSTEM 100 operated at 60 keV to avoid possible beam damage. This microscope is also equipped with a special monochromator that offers energy resolution of <20 meV for EELS investigations [4]. As shown in Fig. 2, EELS compositional mapping of Co establishes a uniform elemental distribution. Because Co substitutes for Ti in the BTO lattice, the effect of Co doping on the Ti valence state is an important factor. Knowing that Ti3+ and Ti4+ cause different Ti-L2,3 EELS edge fine structure [5], the Ti oxidation state of BCTO was studied and the results are shown in Fig. 3. Figure 3 (a) shows EELS line scans taken from an area with thickness of ~30nm, which suggests that Ti3+ is present in BCTO but not in STO. However, EELS line scans taken on thicker areas show the existence of only Ti4+ in the film, as visible in Fig. 3 (b). Moreover, it was found that bypassing the monochromator could result in different morphology of the Ti-L2,3 edge. Two distinct peaks sometimes appeared to be visible when using the monochromator unlike spectra recorded with the monochromator bypassed. This difference is not yet understood and is under further investigation. In summary, this work has established that the Co dopant is distributed uniformly in the BCTO layer, but raises concerns that sample thickness may cause conflicting results that need to be understood when analyzing the Ti oxidation state in BCTO films. Further observations are in progress [6]. [1] S. Salahuddin and S. Datta, Nano Lett. 8, 405�410 (2008). [2] M. D. McDaniel, et al, J. Appl. Phys. 115, 224108 (2014). [3] P. Barik, et al, J. Amer. Ceram. Soc. 94, 2119�2125 (2011). [4] O. L. Krivanek, et al., Microscopy 62, 3-21 (2013). [5] Y. Shao, et al, Ultramicroscopy 110, 1014�1019 (2010). Microsc. Microanal. 21 (Suppl 3), 2015 2078 [6] This work was supported by AFOSR Grant FA9550-12-10494. We gratefully acknowledge the use of facilities within the John M. Cowley Center for High Resolution Electron Microscopy at Arizona State University. Figure 1. Cross-section high-resolution BCTO/STO/Ge(001) sample. TEM image of Figure 2. EELS composition map of Co, showing uniform Co distribution. Figure 3. EELS line scans from: (a) thin; and (b) thick, areas of BCTO/STO/Ge(001) sample with monochromator slit inserted.");sQ1[1038]=new Array("../7337/2079.pdf","Study on the Atomic and Electronic Structure in CrN (VN, TiN) Films using CS-Corrected TEM","","2079 doi:10.1017/S1431927615011174 Paper No. 1038 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Study on the Atomic and Electronic Structure in CrN (VN, TiN) Films using CS-Corrected TEM Zaoli Zhang1, Gerhard Dehm2 . Erich Schmid Institute, Austrian Academy of Sciences, Leoben, Austria . Max-Planck-Institut f�r Eisenforschung, D�sseldorf, Germany Email: zaoli.zhang@oeaw.ac.at 2 1 Transition metal nitrides have found wide-spread applications in the cutting- and machiningtool industry due to their extreme hardness, thermal stability and resistance to corrosion. The increasing demand of these nitrides requires an in-depth understanding of their structures at the atomic level. This has led to some experimental and theoretical researches [1-6]. The films used in this study were deposited by reactive direct current magnetron sputtering of a Cr/V/Ti metal target in an Ar+N2 atmosphere at a constant total pressure of 1 Pa, a target power of 6 kW, and a temperature of 350�C. A TEM/STEM JEOL 2100F operated at 200 kV and equipped with an image-side CS-corrector and a Gatan imaging filter (Tridiem) was utilized for characterizing the film structure. In this paper, we will present some recent results on the atomic and electronic structures of metal nitride thin films (CrN, VN and TiN) on MgO and Al2O3 substrate (Fig.1 and Fig. 2) using advanced TEM techniques, such as CS-corrected HRTEM/STEM, EELS/EDXS, quantitative atomic measurement and electron diffraction analysis as well as theoretical calculations. Interfacial detailed atomic and electronic structures are revealed and compared. Interface induced phenomena between nitride films and substrates are unveiled [2,3]. Particular study on the effect of N defects in the metal nitride (CrN) film has led to some interesting conclusions. Ordered nitrogen (N) vacancies were often found to well distribute at the {111} planes. Combining independent image analysis, such as atomic displacement/strain measurement using geometrical phase analysis, and spectrum analysis by examining the low loss and core loss, fine structure analysis, some generalized conclusions are drawn, which are: (i) a relationship between the lattice constant and N vacancy concentration in CrN is established [5], (ii) the change of ionicity in CrN crystal with the N vacancy concentration is shown; (iii) Particularly, a direct relationship between electronic structure change (L3/L2 ratio) and elastic deformation (lattice constants) in CrN films has been experimentally derived, revealing that the elastic deformation may lead to a noticeable change in the fine structure of Cr-L2,3 edge, i.e. L3/L2 ratio [1]. Such experiments point out an indirect approach to acquire electronic structure changes during the elastic deformation. The effect of randomly distributed defects in the films has been explored in a quantitative way using quantitative electron diffraction, combined with HRTEM and EELS analysis. Some quantitative relations are established. References [1]. R. Daniel et al, Acta Materialia 58 (2010), p. 2621. [2]. Z. L. Zhang, et al Physical. Review B 82(R) (2010) p060103-4 [3]. Z. L. Zhang, et al Journal of Applied Physics, 110 (2011) p043524-4 [4]. Z.L. Zhang et al Physical Review B 87 (2013) p014104. Microsc. Microanal. 21 (Suppl 3), 2015 2080 [5]. T.P. Harzer, et al Thin Solid Films 545 (2013) p154�160 [6]. A.S. Botana et al Physical Review B 85 (2012), p. 235118 [7]. Acknowledgement: Gabriele Moser and Herwig Felber are gratefully acknowledged for their help with sample preparation, thanks are given to Dr.Hong Li for ab-initio calculations. Thank Dr. Rostislav Daniel and Christian Mitterer in Montanuniversit�t Leoben, Leoben, Austria for delivering the materials. Figure 1. Left: HRTEM image of the CrN/Cr interface, a defective layer between Cr and CrN originated from the ordered N vacancy. Right: the anisotropic distribution of strain in the defective layer (exx). Figure 2. The change of white � line ratio with the lattice constant variations.");sQ1[1039]=new Array("../7337/2081.pdf","Quantitative Determination of Chemical Composition of Multinary III/V Semiconductors With Sublattice Resolution Using Aberration Corrected HAADF-STEM","","2081 doi:10.1017/S1431927615011186 Paper No. 1039 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantitative Determination of Chemical Composition of Multinary III/V Semiconductors With Sublattice Resolution Using Aberration Corrected HAADF-STEM Andreas Beyer1, Nikolai Knaub1 and Kerstin Volz1 1. Philipps-Universit�t Marburg, Faculty of Physics and Materials Science Center, 35032 Marburg, Germany Multinary III/V materials such as Ga(NAsP) or (GaIn)(NAs) have a great potential for applications like lasers or solar cells [1]. This is mainly due to the independent tunability of the bandgap and the lattice constant by intentionally changing the composition of the group III and the group V sublattice, respectively. Methods to determine the actual chemical composition of a multinary material often require rather crude assumptions or the combination of several techniques e.g. high angle annular dark field (HAADF) imaging and strain state analysis in scanning transmission electron microscopy (STEM) [2]. Aberration corrected STEM under HAADF conditions facilitates atomic resolution imaging and therefore allows to visualize the individual sublattices. Here we choose GaP, GaAs and their ternary alloy Ga(PAs) as a model system to investigate the influence of the chemical composition on the HAADF intensity for each sublattice independently. Epitaxial GaP, Ga(PAs) and GaAs layers were grown via metal organic vapor phase epitaxy on GaP and GaAs substrates, respectively. Electron transparent samples were prepared by conventional mechanical grinding followed by argon ion milling in a Gatan PIPS. For the TEM investigation, a <010> zone axis was chosen as it exhibits the maximum spatial separation of the individual sublattices. The characterization was performed using an aberration corrected JEOL JEM 2200 FS operating at 200 kV. Complementary frozen phonon simulations were carried out using the STEMSIM code [3]. Figure 1 shows HAADF measurements of GaP (a), Ga(P0.95As0.05) (b) and GaAs (c) samples with comparable thicknesses. A thickness in the range of 30 nm was determined by comparison of the measured intensity to simulated one. Figure 2 depicts the statistical evaluation of these samples, it was derived by procedure described in the following. In each image the individual column positions were determined with subpixel accuracy using the peak pairs code [4]. Utilizing an in house written Matlab code, the two sublattices were separated from each other and the intensity around a peak position was integrated. An integration radius 1/3 of the average next neighbor distance was chosen as this value shows the best tradeoff between a reduction of experimental noise and a loose of spatial information. Finally the measured intensities of the columns are drawn versus their relative frequency for the group V (Fig. 2 (a)) and the group III sublattice (Fig. 2 (b)), respectively (GaP in red, Ga(PAs) in green and GaAs in blue). As expected from the Z-dependence of the HAADF imaging, a shift of the group V values towards higher intensities is visible with increasing fraction of As. Moreover, the group III columns show a dependence on the As content as well, although they are occupied with Ga only. We attribute this behavior to a "crosstalk" of the individual sublattices caused by the finite size of the electron probe. Nevertheless, the dependence of the intensity on the composition is much more prominent on the partially occupied group V lattice. The behavior on both lattices is in agreement with simulations, if the microscope parameters, especially the actual probe size, are taken into account. Utilizing this statistical approach the influence of chemical composition Microsc. Microanal. 21 (Suppl 3), 2015 2082 and the crosstalk of one column to another can be quantified separately. This facilitates the quantitative evaluation of quarternary systems with atomic resolution as well. This contribution will show a method how aberration corrected STEM can be used to determine the composition of the sublattices of multinary III/V systems [5]. References: [1] B. Kunert, K. Volz, J. Koch, and W. Stolz, Appl. Phys. Lett., 88 (2006), , p. 182108. [2] T. Grieb et al., Microsc. Microanal, 20, (2014), pp. 1740�52. [3] A. Rosenauer and M. Schowalter, Microsc. Semicond. Mater. 2007, 120, (2008), pp. 170�172. [4] P. L. Galindo et al., Ultramicroscopy, 107, (2007), pp. 1186�93. [5] The authors acknowledge funding from the DFG in the framework of GRK 1782 (functionalisation of semiconductors). Figure 1. HAADF images of GaP (a), GaPAs (b) and GaAs (c) with comparable sample thickness of 30 nm. Figure 2. Statistical evaluation of the HAADF intensities on the group V (a) and group III lattices (b).");sQ1[1040]=new Array("../7337/2083.pdf","Determination of Local Chemistry Composition of Low-Dimensional Semiconductor Nanostructures Through the use of High-Resolution HAADF images","","2083 doi:10.1017/S1431927615011198 Paper No. 1040 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Determination of Local Chemistry Composition of Low-Dimensional Semiconductor Nanostructures Through the use of High-Resolution HAADF images D. Hern�ndez-Maldonado1,2,3, M. Herrera1, A.R. Lupini4, S. I. Molina1 Departamento de Ciencia de los Materiales e I.M. y Q.I., Facultad de Ciencias, Universidad de C�diz, Campus R�o San Pedro, s/n, 11510 Puerto Real, C�diz, Spain 2. SuperSTEM Laboratory, STFC Daresbury Campus, Daresbury WA4 4AD, UK 3. Department of Materials, University of Oxford, Parks Road, Oxford OX1 3PH, UK 4. Materials Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA The introduction of aberration-correction into Scanning Transmission Electron Microscopy (STEM) has vastly improved the ability to carry out chemical analysis of low-dimensional semiconductor nanostructures, which is a topic of high relevance in fields such as electronics and optoelectronics [1,2]. Among different techniques in STEM, Energy Electron Loss Spectroscopy (EELS) and Energy-dispersive X-ray Spectroscopy (EDX) are currently the most used to obtain a quantitative map of the structures' chemical composition. However, in practice, extracting information from an area of the size of a Quantum Dot (QD) with these techniques in STEM mode is a complicated task, due to the time necessary to record the signal. During this time drift effects or sample damage could occur. Thus a method that uses only the High Angle Annular Dark Field (HAADF) signal to analyse the composition could avoid these problems, because in a few seconds it is possible to image large areas. HAADF images of QDs have been widely used for a qualitative analysis of their chemical distribution and for morphological characterization [3]. However, due to the high dependence of the intensity with the atomic number (Z) of the HAADF images, the extraction of quantitative information from this sort of image has become a topic of great interest. Nowadays there are multiple options to interpret HAADF images in quantitative terms. It is possible to study the intensities associated with atomic columns through statistical analysis [4], phenomenological procedures [5] or to compare with simulations [6]. But in the case of the analysis of nanostructures like QDs all of these methodologies are affected by the dependence of the HAADF intensities on the strain fields originating from the nanostructure that alter the intensities [7], making the precise quantification of HAADF images a complicated task. This paper presents an alternative method to quantify the chemical composition of low dimensional nanostructures based on the analysis of the atomic column positions present in High-Resolution HAADF images. These atomic positions depend on the strain field of the nanostructure, which is closely related to the local composition of the structure. By a comparison between the strain maps measured from experimental HR-HAADF images and the strain maps obtained by Finite Element Method (FEM) applied to a model that describes the nanostructure, it is possible to determine the local chemical composition. We have applied this methodology to a nanostructure formed by two vertically-stacked InGaAs QDs. Figure 1a) is a HAADF image of the nanostructure used as an example. Bright areas are associated with the presence of In atoms. Images like this have been employed to construct a three-dimensional (3D) model where FEM calculations have been applied. Figure 1b) is the 3D model proposed for the structure under study, which is composed of different parts. During the process of comparison, the chemical concentration of each part of the model is 1. Microsc. Microanal. 21 (Suppl 3), 2015 2084 varied in a repetitive procedure until a good fit between the experimental and the calculated strain map is obtained. Figure 2a) is the experimental strain map of the nanostructure parallel to the growth direction and 2b) is the corresponding map calculated by FEM obtained after the fitting process. As a result of the process we obtain a composition profile along the nanostructure studied. Some considerations in the interpretation of the results, due to the fact that the depth of field of an Aberration-Corrected STEM is less than the thickness of the sample under study will be discussed. References: [1] G. Biasiol, et al. Physics Reports 500 (2011) 117-173. [2] S. Kadkhodazadeh. Micron 44 (2013) 75-92. [3] P. Wang, et al. Applied Physics Letters 89 (2006)072111. [4] S. Van Aert, et al. Physical Review B 87 (2013) 064107. [5] S. I. Molina, et atl. Ultramicroscopy 109 (2009) 172-176. [6] P.L. Galindo. Journal of Physics 522 (2014) 012013. [7] T. Grieb. Ultramicroscopy 129 (2013) 1-9. [8] This research was supported by the Spanish MINECO (project TEC2011-29120-C05-03 and CONSOLIDER INGENIO CSD2009-00013) and the Junta de Andaluc�a (PAI research group TEP-946 INNANOMAT). Co-funding from UE is also acknowledged. Work at Oak Ridge National Laboratory was sponsored by the U.S. Department of Energy, Division of Materials Sciences and Engineering. We thank L. Gonzalez and Y. Gonzalez from IMM for supplying the sample. Figure 1a) HAADF image of the nanostructure studied, which is formed by 2 QDs vertically stacked. b) Three-dimensional model of the previous structure created for the FEM calculations. Figure 2 a) Strain map obtained from the Geometrical Phase Analysis of a HR-HAADF image of the nanostructure presented in figure 1a). b) Calculated strain map of the 3D model proposed for the nanostructure.");sQ1[1041]=new Array("../7337/2085.pdf","TEM Characterization of Ball Milled CIS and CIGS Nanoparticles.","","2085 doi:10.1017/S1431927615011204 Paper No. 1041 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 TEM Characterization of Ball Milled CIS and CIGS Nanoparticles. I. I. Santana-Garc�a1, C. Kisielowski2 and H. A. Calderon1 1. 2. Departamento de F�sica, ESFM-IPN, M�xico D.F. 07338, M�xico. Molecular Foundry and JCAP, LBNL, One Cyclotron Road, Berkeley, CA 94720, USA. CIGS (CuInxGa(1-x)Se2) is a semiconducting photovoltaic material with the chalcopyrite crystal structure and a band gap varying with x from approximately 1.0 eV (CIS) to around 1.7 eV (CGS). These nanoparticles can be used to transform solar energy into electricity and are attractive for solar cells. In this investigation reactive mechanical milling is used for synthesis. This is a simple procedure where a chemical reduction is induced during milling at room temperature. It gives rise to a fine nanoparticle distribution if coagulation is restrained e.g. by adding a dispersing media. The nanoparticles in the present report have all been produced by reactive milling (x = 0.3), starting with chlorides and chemically reducing them with Na in an excess of NaCl to promote dispersion. Figure 1 shows the chemical distribution of CIS particles via EDS mappings. After 10 h of reactive milling, nanoparticles acquire a homogeneous distribution of the involved chemical elements. Similar results are determined for CIGS nanoparticles. Shorter milling times have shown clear chemical inhomogeneities in the nanoparticles compositions especially in the case of Ga containing materials. Figure 1 shows agglomerated nanoparticles at a relatively low magnification. The average particle size as determined by diffraction techniques varies from 5 to 7 nm but earlier TEM observation also shows that many smaller particles can be found. These semiconducting nanoparticles are rather beam sensitive. Figure 2 shows phase images of CIS nanoparticles and the changes that induce the beam interaction as a function of electron dose rate. In all cases, 40 experimental images are taken for the exit wave reconstruction procedure i.e., the particle is exposed for approximately 2 min [1]. As shown only rather low dose rates (lower than 60 e-/� s, can be used to preserve the as-synthesized condition of the nanoparticles. Higher dose rates produce unexpected changes in the agglomerates. Figure 3 shows phase images of CIS nanoparticles at a dose rate lower than the above determined limit. Different sizes are imaged, independent nanoparticles can be as small as 2 nm. Larger particles normally agglomerate and appear as crystalline domains in the phase images. The chalcopyrite structure can be readily identified in the atomic column distribution in Fig. 3 together with distinctively brighter columns that most likely represent heavier atoms. Figure 4 shows phase images of CIGS nanoparticles. In all cases, the electron dose rate is lower than 30 e-/� s since the presence of Ga increases dramatically the sample beam sensitivity and the corresponding bean induced changes. Particles have sizes that range between 2 and 5 nm and often their shape is irregular. When observed in detail and along a low index zone axis, the chalcopyrite structure can be readily recognized (see Fig. 4c). This investigation is complemented by simulation and modeling. References: [1] H.A. Calderon, C. Kisielowski, P. Specht, B. Barton, F. Godinez-Salomon, O. Solorza-Feria. Micron 68 (2014), p. 164. [2] Electron Microscopy was performed at the Molecular Foundry, which is supported by the Office of Science, Office of Basic Energy Sciences of the U.S. Department of Energy under Contract No.DEAC02-05CH11231. The research is partially supported by CONACYT (Proyecto FOINST. 75/2012, 148304 and 129207), IPN (COFAA, SIP). Microsc. Microanal. 21 (Suppl 3), 2015 2086 a b c d Figure 1. EDS Mappings of CIS nanoparticles (a) HAADF image, (b) Cu, (c) In, (d) Se (e) Na, (f) Cu-In, (g) Se-In, (h) Se-Cu, (i) and Cu-In-Se. Figure 2. Phase images of CIS nanoparticles at different dose rate for approximately 2 min. (a) 25, (b) 58, (c) 400, (d) 1400 e-/� s. a b c Figure 3. Low dose phase images of CIS nanoparticles after a EWR procedure. The expected chalcopyrite crystalline structure coincides with the observed atomic column distribution. a b c Figure 4. Low dose phase images of CIGS nanoparticles after an EWR procedure. The expected chalcopyrite crystalline structure coincides with the observed atomic column distribution.");sQ1[1042]=new Array("../7337/2087.pdf","Atomic Scale Study of Lomer�Cottrell and Hirth Lock Dislocations in CdTe","","2087 doi:10.1017/S1431927615011216 Paper No. 1042 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic Scale Study of Lomer�Cottrell and Hirth Lock Dislocations in CdTe Tadas Paulauskas1, Chris Buurma1, Brian Stafford1, Cyrus Sun2, Maria Chan3, Sivananthan Sivalingham1, Moon Kim2 and Robert F. Klie1 1. 2. University of Illinois at Chicago, Department of Physics, Chicago, USA University of Texas at Dallas, Department of Materials Science and Engineering, Dallas, USA 3. Argonne National Laboratory, Center for Nanoscale Materials, Argonne, USA Many useful and also detrimental properties of solids can be traced back to the underlying structure of dislocations and their behavior. Their presence has been studied extensively studied in the context of crystal growth on lattice mismatched substrates and dislocations may significantly alter a material's mechanical, thermal and opto-electronic properties [1]. Experimental knowledge detailing atomically resolved chemical structure of dislocation cores is highly desirable to advance fundamental understanding of these ubiquitous structures. In this study we present atomic scale analysis of two wellknown dislocations junctions in CdTe, the Lomer-Cottrell (L-C) and the Hirth lock. Despite being textbook examples of the lowest elastic energy stair-rod dislocations their atomic structure, in particularly in zinc-blende CdTe, were unknown. Stair-rod dislocations, usually identified in terms of their lack of mobility, may have important effects on the mechanical properties which raise concerns given the trend of continuous miniaturization of semiconductor devices. Furthermore, being pure-edge dislocations, L-C and Hirth locks are likely to be found in low-angle tilt grain boundaries. The dislocations under investigation here were studied as occurring naturally in poly-crystalline CdTe thin films incorporated into traditional CdS/CdTe solar cells. The post-deposition CdCl2 annealing process, commonly used in commercial poly-CdTe-based solar cells, was not performed on these samples in order to investigate the dislocations without any extrinsic dopants. Characterization is carried out in the aberration-corrected cold-field emission JEOL JEM-ARM200CF scanning transmission electron microscope (STEM) using high-angle annular dark field (HAADF). Chemical composition of the dislocation cores is demonstrated via atomic-column-resolved X-ray energy dispersive spectroscopy (XEDS) using windowless silicon drift detector, the Oxford Instruments X-Max 100TLE. Local strain fields associated with dislocations are measured from the images via the Geometrical Phase Analysis (GPA). In order to confirm identity of the dislocations we constructed Burgers circuits specially tailored for dislocation junctions since faulted material cannot be crossed twice in a circuit. To explain reaction steps and core configurations of the resulting dislocation junctions a double Thompson's tetrahedron is used which can take into account intrinsic/extrinsic stacking faults and polar variants of dislocations in CdTe [2]. Figure 1 a) shows X-ray spectrum map of L-C dislocation polar core variant which is seen to be composed of three Cd atomic columns and one Te at the center. In b) the Burgers circuit construction is used to identify nature of the dislocation as indicated. An unfaulted gap in the material (dashed box) is used to pass the circuit through. Partial dislocation introduced in this way is taken into account by the second small circuit. Figure 2 a) shows the Hirth lock dislocation. The core is more compact and is composed of two atomic columns (Te and Cd). Figures 2 b) and c) indicate the dislocation reaction steps in terms of the partial dislocations on two different {111} planes, which result in formation of the core. Microsc. Microanal. 21 (Suppl 3), 2015 2088 References: [1] Sutton, A. P. & Balluffi, R. W., "Interfaces in Crystalline Materials", New York: Oxford Science Publications, (1995). [2] Paulauskas, T. et.al.," Atomic scale study of polar Lomer�Cottrell and Hirth lock dislocation cores in CdTe", Acta Cryst. � A, (2014). [3] This research is supported by a grant from the Department of Energy Sunshot Program (DOE DEEE0005956). a) b) Figure 1. a) X-ray spectrum image of the L-C dislocation core (circled) which is associated with two intrinsic stacking faults (dashed SF). b) Two Burgers circuits construction is shown on an HAADF image for determination of Burgers vector of the stair-rod dislocation. Relevant crystallographic directions are shown as well. Inset shows magnified view of the core. a) b) c) Figure 2. a) HAADF image and Burgers circuits for the Hirth lock. b) Double Thompson's tetrahedron for the Hirth lock, showing dissociated perfect dislocations involved in the reaction on {111} planes. c) Dislocation reaction steps in terms of partial dislocations Burgers vectors, as per figure b).");sQ1[1043]=new Array("../7337/2089.pdf","Atomic resolution studies of the (K, Na)NbO3/SrTiO3 interface using aberration corrected STEM","","2089 2089 doi:10.1017/S1431927615011228 doi:10.1017/S1431927615011228 Paper No. 1043 Microsc. Microanal. 21 (Suppl 3), Microsc. Microanal. 21 (Suppl 3), 2015 2015 � Microscopy Society of America � Microscopy Society of America 2015 2015 Atomic resolution studies of the (K, Na)NbO3/SrTiO3 interface using aberration corrected STEM Chao Li1, Guang Yang1, Lingyan Wang1, Zhao Wang2 Electronic Materials Research Laboratory, Key Laboratory of The Ministry of Education& International Center for Dielectric Research, Xi'an Jiaotong University, Xi'an, China 2. Frontier Institute of Science and Technology, State Key Laboratory for Mechanical Behavior of Materials, Xi'an Jiaotong University, Xi'an, China Atomically-engineered interfaces between transition metal oxides offer novel electronic and coupled functionalities in thin films that are otherwise unapproachable through bulk synthetic chemistry due to the element composition modulation, lattice strain effect and various dislocations at the interface. Consequently, it is crucial to study the interface at atomic resolution for thin film growth, deposition quality and piezoelectric properties, etc. (K, Na)NbO3 (KNN) as an environment- friendly lead-free piezoelectric material has been extensively studied since the earlier research work achieved by Saito et al [1]. To date, efforts have been concentrated on KNN bulk materials and the KNN-based thin film research is limited. In this paper, a lead-free KNN thin film was epitaxially grown on SrTiO3 (STO) single-crystalline substrates by sol-gel method [2]. The microstructure of the KNN/STO epitaxial interface was investigated by scanning transmission electron microscopy (STEM) with electron energy loss spectroscopy (EELS) using JEOL ARM 200F electron microscope with probe spherical aberration corrector. The strain distribution of the KNN/STO epitaxial interface was mapped at the atomic scale from HAADF images by using the geometric phase analysis (GPA) [3]. Fig. 1a shows the HAADF-STEM images of the epitaxial KNN/STO interface region. The clear contrast difference indicates the sharp KNN-STO interface. It is noted that the brightness of one layer of atomic columns (marked by red rectangle) at the interface shows apparent differences in comparison with adjacent atoms. The HAADF-STEM image simulation (WinHREMTM) is conducted to quantify the atomic mixture along the electron beam direction in the experimental micrographs. Fig. 1b presents the schematic super-cell model of the KNN/STO interface and the corresponding simulated HAADF image (50 at% Ti was substituted by Nb atoms at the interface layer in the simulated image) is shown in Fig. 1c. The similarity of the intensity line profiles from both the simulated images and the experimental ones (Fig. 1d and 1e) indicates that the single layer at the KNN/STO interface contains 50 at% Ti and 50 at% Nb. In order to verify the intermixing layers along the interface, atomic resolution EELS mapping was applied. Fig. 2 (b-e) shows the EELS spectra image results from the interface region marked in Fig. 1a. It is clear that the Ti and Nb cations coexist at the 'B-site' (A: alkaline or rare-earth ions; B: transition metals in perovskite structure ABO3) which confirms the composition result obtained by the HAADF image simulation. The piezoelectric properties of KNN thin film will largely depend on the KNN/STO interface, in other words, the strain relaxation is important for its application. Fig 3a shows the HAADF image of the dislocation free interface and Fig. 3b-d are the corresponding GPA strain maps of Exx, Eyy and Exy, respectively. Fig 3b shows the relative a-lattice strain at the interface which indicates a gradual increase perpendicular to the thin film growth direction. However, there are little relative strain in perpendicular direction and out-of-plane rotation in both the film and substrate lattices. More 1. Microsc. Microanal. 21 (Suppl 3), 2015 2015 Microsc. Microanal. 21 (Suppl 3), 2078 2090 detailed analysis including dislocation at the interface and DFT calculation will also be discussed. References: [1] Y. Saito et al., Nature, 432 (2004) 84 [2] Q. Yu et al., Appl. Phys. Lett. 104 (2014) 102902 [3] M.J. H� et al ., Ultramicroscopy 74 (1998) 13 tch [4] The authors acknowledge the funding from National Natural Science Foundation of China (51202180), the Fundamental Research Funds for the Central Universities in China and the 111 Project of China (B14040). Figure 1. (a) Experimental HAADF image of the KNN/STO interface; (b) schematic atomic model of the interface; (c) simulated HAADF image of the interface, (d, e) intensity profile of the exprimental image and simulated image across the interface (along the dashed line), respectively. Scale bar, 1nm. Figure 2. EELS spectra imaging results of the KNN/STO interface. (a) HAADF-STEM image; (b) zoomed mapping area; (c, d) False-coloured Nb and Ti elemental maps, respectively; (e) combined elemental maps with Ti in green, Nb in red. Scale bar, 1nm. Figure 3. (a) A atomic-resolution HAADF imge of the stress released KNN/STO interface; (b)-(d) the calculated GPA result of xx, yy, xy, respectively(a). Scale bar, 2nm.");sQ1[1044]=new Array("../7337/2091.pdf","Solving the Controversy of Earth's Oldest Fossils using Electron Microscopy","","2091 doi:10.1017/S143192761501123X Paper No. 1044 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Solving the Controversy of Earth's Oldest Fossils using Electron Microscopy David Wacey1,2, Martin Saunders2, Charlie Kong3, and Martin Brasier4* 1. 2. University of Bristol, Bristol, UK. The University of Western Australia, Perth, Australia. 3. The University of New South Wales, Sydney, Australia. 4. University of Oxford, Oxford, UK; *deceased. Filamentous microstructures in 3.46 billion-year-old rocks (Apex chert) from Western Australia have been claimed to represent the oldest morphological evidence of life on Earth [1]. However, the biological nature of these filaments has been questioned on numerous occasions [e.g., 2] and this has led to one of the longest running and most controversial debates in palaeontology. Here we use a combination of focused ion beam (FIB) milling, TEM and SEM to decode the detailed morphology and chemistry of the Apex filaments and determine their formation mechanism. Critical to the claims of a biological origin for the Apex filaments is a suggested presence of cells, having three-dimensional wall compartments made of carbon [3]. In contrast, our TEM analyses of ultrathin wafers through four representative filaments reveal filament morphologies that are characteristic of a mineralic origin, with complex nano-scale intergrowths of mineral phases (Fig. 1a). Each filament is made up of multiple plate- or sheet-like grains of phyllosilicate (Fig. 1b, green), sitting within a matrix of microcrystalline quartz. Occasionally quartz is also seen inter-grown with the phyllosilicate within a filament. ChemiSTEM mapping shows that the phyllosilicate mineral(s) contain the elements K, Al, Si, O, plus variable amounts of Ba and minor Mg. Electron diffraction patterns of this mineral obtained in the TEM possess d-spacings consistent with a 2:1 layered phyllosilicate crystal lattice structure. This structure is found both in micas such as muscovite and some clay minerals. The nano-morphology of the phyllosilicate, appearing as a worm-like stack of crystals, closely resembles vermiculite, a common alteration product of mica. However, the chemical composition is spatially heterogenous on the nano- to micro-scale. This, together with the presence of barium, suggests that the phyllosilicate is likely a complex hydrothermal association of mica alteration products that are best termed vermiculite-like. Further ChemiSTEM mapping shows that carbon (Fig. 1b, yellow) and iron (Fig. 1b red) are closely associated with the phyllosilicate filaments. Both carbon and iron are seen interleaved between sheets of phyllosilicates within the body of the filaments, and also coat the outer margins of some parts of the filaments. In addition, carbon occurs away from the filaments within the quartz matrix where it forms a boundary phase between quartz grains (Fig. 1a). These data indicate significant redistribution of carbon both within and around the Apex filaments, in marked contrast to patterns found by us in bona fide fossil microbes from younger rocks [4]. Carbon interleaved between phyllosilicates within the filaments may resemble `cellular compartment walls' when investigated with lower spatial resolution (for example in optical work). Our higher spatial resolution analysis of supposed `cellular compartments' instead reveals very inconsistent compartment lengths (<50 nm up to c. 1 �m), with length/width ratios that match crystal growth patterns and are unlike any known microbial cells. 3D FIB-SEM data reveal further complexities to the filaments and additional insights into carbon distribution in their vicinity. These data demonstrate how the morphology of the filaments changes quite Microsc. Microanal. 21 (Suppl 3), 2015 2092 significantly over spatial scales of only a few micrometres along the length of a filament. In some FIB slices, their filamentous nature is clear and books of phyllosilicate crystals appear neatly stacked, while in other slices the filaments are seen to branch, suddenly thicken or be joined by additional microstructures. Furthermore, SEM highlights a number of nano-cracks within the chert matrix; these often feed right into the filaments and are filled with carbon. Like the TEM data, the SEM data are incompatible with these filaments being fossils of primitive filamentous organisms. Carbon distribution in Apex `microfossils' is, therefore, not comparable with true cellular morphology. Our non-biological formation model is: 1, Hydration of mica flakes (abundant in the country rock) during widespread hydrothermal activity resulting in vermiculite-like phyllosilicate formation. 2, Continued heating plus expulsion of water from phyllosilicate crystal lattices, causing exfoliation (i.e., accordion-like expansion at right angles to the cleavage plane), and creating the initial worm-like filamentous morphological expression of microfossil-like artefacts. 3, Adsorption of later hydrocarbons (and locally additional iron) onto the phyllosilicate, mimicking cell walls. We note that exfoliated vermiculite has high adsorption capacity for hydrocarbons resulting from the strong capillary action of slit-like pores between plate-like grains, encouraging its use for cleaning up oil spills [5]. [1] JW Schopf, Science 260 (1993), p. 640. [2] MD Brasier et al., Nature 416 (2002), p. 76. [3] JW Schopf and AB Kudryavtsev, Gondwana Research 22 (2012), p. 761. [4] D Wacey et al., Precambrian Research 220-221 (2012), p. 234. [5] The authors acknowledge funding from the European Commission and the Australian Research Council, and also acknowledge the Australian Microscopy and Microanalysis Research Facility. Figure 1. a) HAADF-STEM image of a filament from the 3.46 billion-year-old Apex chert. b) False colour ChemiSTEM three-element overlay map of area boxed in (a). The filament comprises stacks of sheet-like phyllosilicate grains (green) with carbon (yellow) and iron (red) interleaved between some of the sheets and around some of the filament margin. This distribution of phases is incompatible with a biological origin for the filament.");sQ1[1045]=new Array("../7337/2093.pdf","Standards-Based Quantitative EDS Mapping.","","2093 doi:10.1017/S1431927615011241 Paper No. 1045 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Standards-Based Quantitative EDS Mapping. Stephen M. Seddio1. 1. Thermo Fisher Scientific, Fitchburg, WI, USA. In many microanalytical software packages, standardless EDS (energy-dispersive spectroscopy) "semiquantitative" analysis is typically done with a single mouse click and yields results that appear to be accurate to 0.001 wt% and always with analytical totals of exactly 100%. Although semi-quantitative analysis has proven to be getting progressively more accurate [e.g., 1], it is still unable to account for many factors that can greatly affect the accuracy of an analysis such as surface roughness, detector window contamination, and secondary fluorescence of a neighboring phase [2]. Recent discussion [e.g., 1,2,3] has highlighted the importance of standards-based EDS quantitative analysis. However, such discussion has not carried over to EDS mapping. EDS spectral imaging, in which a full EDS spectrum is acquired and stored at each pixel, allows for the peak deconvolution algorithms and matrix corrections involved with EDS quantitative analysis to be applied to EDS mapping. Here, I investigate the importance of the use of standards in EDS quantitative mapping. EDS data were collected using a Thermo ScientificTM UltraDryTM SDD EDS detector mounted on a JEOL 7001F FE-SEM. EDS spectral images were processed using the Thermo Scientific NORAN System 7 X-ray microanalysis system. Quantitative mapping was done on an SPI Cu metal standard and lunar meteorite NWA 2727. Standards for EDS quantitative analysis were SPI metal and natural and synthetic mineral standards. Beam current was measured at the beginning of each mapping acquisition and was compared to an additional measurement after the mapping acquisition to ensure beam stability. For mapping the Cu metal standard, eleven beam-rastered, 64�48 pixel EDS spectral images were acquired at 20 kV until the average X-ray counts/pixel (i.e., counts per spectrum) were 500; 750; 1,250; 2,500; 5,000; 10,000; 15,000; 20,000; 30,000; 40,000; and 50,000. Although mapping a single element standard produces uninteresting map images, it provides an excellent opportunity to investigate the reliance of EDS quantitative analysis on a sufficient number of counts without confusion added by topographic effects. By extracting so-called "quant maps" using a Cu standard, 3072 quantitative analyses were calculated from each of the eleven spectral images (Fig. 1). Because the spectral images were all acquired on a Cu standard, all quantitative analyses should yield 100 wt% Cu. However, accurate results were not achieved until ~50,000 counts/pixel were acquired. Using semi-quantitative analysis, every pixel of each of the eleven spectral images would have yielded exactly 100 wt% Cu, which is the correct answer in this example, but would yield incorrect results in samples with more than one element in spectral images with insufficient counts. For mapping an area of NWA 2727, a 50,000 counts/pixel, 64�48 pixel (41.9�31.4 �m) EDS spectral image was acquired at 15 kV. Elemental quantitative maps were extracted. Additionally, analytical totals, pyroxene stoichiometry, and olivine stoichiometry maps were extracted (Fig. 2). The analytical total map is perhaps the most beneficial result of standard-based quantitative mapping in that it allows the user to quickly see which pixels yield poor analyses (i.e., high or low analytical totals). For example, in Fig. 1, there is a crack. One would expect that fewer X-rays reach the detector from a pixel representing the crack yielding a low total. However, with standardless analysis, this result is normalized to 100%, giving no indication of a pixel with a poor analysis. References: [1] NWM Ritchie and DE Newbury, Microsc. Microanal. 20 (2014) p. 696. [2] JH Fournelle, S Kim, and JH Perepezko, Surf. Interface Anal. 37 (2005), p. 1012. [3] DE Newbury and NWM Ritchie, Microsc. Microanal. 18 (2012) p. 1004. Microsc. Microanal. 21 (Suppl 3), 2015 2094 Figure 1. Average wt% Cu and standard deviation thereof as a function of X-ray counts in the Cu K K region of interest. Sample was 100 wt% Cu. Figure 2. a. Backscattered electron image of the mapped area. b. Map of analytical totals; low totals are fractures. c and d. Maps of how well the analysis at each pixel matches pyroxene (c; [Ca,Mg,Fe]2Si2O6) and olivine (d; [Mg,Fe]2SiO4) stoichiometries as calculated from standards-based quantitative analysis.");sQ1[1046]=new Array("../7337/2095.pdf","Survey for Fe-Si in Apollo 16 Regolith Sample 61501,22","","2095 doi:10.1017/S1431927615011253 Paper No. 1046 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Survey for Fe-Si in Apollo 16 Regolith Sample 61501,22 P. Gopon, J. Fournelle, M. Spicuzza, J.W. Valley Dept. of Geoscience, University of Wisconsin, Madison, Wisconsin 53706 USA Metallic iron (micron to submicron spherules) is relatively abundant in lunar regolith and is widely believed to have formed by reduction of Fe during space weathering (micrometeorite/ cosmic ray bombardment; [1]). Iron-silicides (Fe-Si) and native silicon require more reducing conditions than Fe0, and though predicted to be present in the lunar regolith have only been reported from one lunar meteorite and one lunar regolith sample [2,3,4]. Fe-Si and native silicon are only stable at extremely reducing conditions and have been proposed to form by a mechanism similar to that of metallic iron [3]. The stability of native silicon requires ten orders of magnitude lower oxygen fugacity than metallic iron at temperatures above 1600 K [5]. After the discovery of Si0 and Fe-Si in lunar regolith 61501,22 [2] and given the paucity of these phases in lunar samples reported in the literature, we systematically searched 61501,22 to locate additional FeSi specimens to determine their relative abundance and advance the understanding of their formation. Regolith grains for inspection were first selected under UV light (Apollo 16 regolith sample 61501,22) to re-create the conditions that led to the discovery of Fe-Si and Sio by Spicuzza et al. [2]. Further study of lunar Fe-Si from Apollo samples, however has been limited to non-destructive methods because until now there was only one 60-�m plagioclase grain containing Fe-Si and Sio from the original study [2]. During the initial analysis of Fe-Si [6], it was suggested that carbon (~1-2 wt. %, determined by EPMA) might be present in the Fe-Si, but the possibility of analytical artifacts has not been resolved. The presence of carbon was hypothesized to explain the ultra-reduced conditions. In order to determine with certainty the presence of carbon in these phases, a destructive technique (SIMS or atom probe) is necessary. Therefore we sought additional examples of Fe-Si/Sio from 61501,22 to test this hypothesis, We prepared 3 grain mounts of UV fluorescent grains (from 61501,22), in three size fractions (Mount 1: UV, 0.8-1mm, Mount 2: UV, 0.5-0.8mm, Mount 3: UV, < 0.5mm) as well as one mount with randomly selected UV and non-UV fluorescent grains (Mount 4: Random, < 0.5mm) to see if Fe-Si is limited to UV fluorescent grains. To date, two of the four mounts have been examined by SEM. In mount 3 (< 0.5 mm UV fluorescent, figure 1, a-c), we found examples of Fe-Si (a few nm to 2 �m in size, ~Fe3Si) in 10 grains (out of 441 grains). In mount 4 (Random < 0.5mm) we found a further 10 grains (out of 561 grains) containing Fe-Si. While we have not yet confirmed additional samples of Si0, these results increase by an order of magnitude the number of reported grains containing Fe-Si. Moreover, all found instances of Fe-Si were present encapsulated in a glassy matrix of anorthitic composition, referred to as glassy-anorthite (presumed glassy based on textural similarity to that present in [2]). Thus Fe-Si in this lunar regolith sample is relatively common, and may be more common in the lunar regolith in general than previously reported. Microsc. Microanal. 21 (Suppl 3), 2015 2096 Figure 1: BSE images of Fe-Si occurrences in Apollo 16 sample 61501,22. Grain number is in the format of mount- quadrant-row-column. a) Glassy An. spherule containing Fe-Si. b) Two glassy An. grains in an agglutinate, the top grain contains Fe-Si blebs while the bottom only Fe-metal. c) glassy An. grain in agglutinate containing Fe-Si, the right half of the grain is crystalline anorthite. d) Glassy An. grain containing Fe-Si. e) Glassy An. grain containing a filled crack consisting of glassy-An. and Fe-Si blebs. f) Glassy An. grain containing Fe-S and Fe-mtl, Fe-Si is restricted to the highlighted vein feature. References: [1] Hapke, B., 2001, J. Geo. Research, v. 106, no. 10, 39-73. [2] Spicuzza, M.J., et al., 2011, 42nd Lunar and Planetary Science Conference ( 2011 ), 16-17. [3] Anand, M., et al., 2004, Proc. of the Nat. Acad. of Sci. USA, v. 101, no. 18, 6847-51. [4] Nazarov et al., 2012, Petrology, v. 20, no. 20, 506-519 [5] Essene, E.J., and Fisher, D.C., 1986, Science (New York, N.Y.), v. 234, no. 4773, 189-93 [6] Gopon, P., et al., 2013, Microscopy and Microanalysis, 19(6), 1698�1708.");sQ1[1047]=new Array("../7337/2097.pdf","Sintering of Ice Spheres under Different Thermal Conditions","","2097 doi:10.1017/S1431927615011265 Paper No. 1047 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Sintering of Ice Spheres under Different Thermal Conditions Xuan Wang and Ian Baker Thayer School of Engineering, Dartmouth College, Hanover, New Hampshire USA Arrays of ice spheres, as geometrically-simplified specimens for complex snow structures, can be used to observe the microstructural evolution and investigate the mass transfer process during snow metamorphism. In this paper, 1-D arrays of polycrystalline ice spheres were produced using the method of Chen et al. [1] and placed in a precisely-controlled temperature gradient set-up. Different vapor transfer directions were applied to investigate the formation of faceted crystals and depth hoar structures. We also examined the effects of an imposed alternating temperature gradient (TG) along the ice spheres, which simulated the sintering condition occurring in the daily cycles of radiative heating and cooling in the topmost 20 cm of the snow layer [2]. Spherical ice spheres were produced by freezing 0oC deionized, degassed water droplets in liquid nitrogen. An array of ice spheres of ~3 mm diameter was arranged vertically inside an acrylic tube sealed with two copper caps. The acrylic tube was placed into a home-built temperature gradient setup. The evolutions of vertical arrays of ice spheres were investigated under four conditions for 48 h, with two groups of alternating TG metamorphism (ATG-1, ATG-2) and two groups of unidirectional TG metamorphism (TG-1, TG-2), see Table 1. A Skyscan 1172 micro-CT was used to examine the ice sphere specimens periodically using pixel size of 7.8 �m. The specimens were scanned every 12 h, and for the ATG conditions the TG direction was reversed every 12 h. The mass transfer mechanisms under the four thermal conditions can be quantified by the evolution of the neck area and the neck position (Figure 1). For ATG-1 and ATG-2, the neck area increased while the neck positions fluctuated as the TG direction was cycled and returned to the original position. The overall phenomenon is similar to that for isothermal sintering: the vapor diffuses from the spheres to the neck, but the neck position does not change. For TG-1 and TG-2, the neck area increased over time and the neck positions changed monotonously depending on the vapor transfer direction. Figure 2 shows the final neck morphologies for ATG-1, TG-1, and TG-2. Figure 2a shows a porous neck with rounded protrusions, while Figure 2(b, c) shows faceted and depth hoar structures around the neck. When the top was colder than the bottom, the cup-like depth hoar structures grew downwards, which means the inner surface faced downwards. For the opposite TG condition, the depth hoar structure grew upwards. In order to model the interaction between the depth hoar structure formation and temperature distribution, we increased our scan frequency under the TG-2 thermal condition and eight 2D vertical cross section images of ice spheres in different times were imported into COMSOLTM to model the temperature and vapor flux distribution. The problem was solved by considering only the thermal conduction in the ice and air domains. Figure 3 shows the modeled temperature and vapor flux distribution over time. The modeled vapor flux predicted the morphology in the next scan. The newly developed ice crystals around the neck evolved from the thin plates into faceted crystals and depth hoar structures, which directly simulated the formation of depth hoar structure under natural snow TG metamorphism. References: [1] Chen, S., I. Baker, and H.J. Frost, Surface instability and mass transfer during the bonding of ice spheres. Philosophical Magazine, 2013. 93(23): p. 3177-3193. Microsc. Microanal. 21 (Suppl 3), 2015 2098 [2] Pinzer, B. and M. Schneebeli, Snow metamorphism under alternating temperature gradients: Morphology and recrystallization in surface snow. Geophysical Research Letters, 2009. 36(23). [3] The authors acknowledge funding from Army Research Office contract 51065-EV. Table 1. Boundary conditions for ATG-1, ATG-2 and TG-1, and TG-1 conditions. Figure 1. The neck area and neck position evolutions for ATG-1, ATG-2, TG-1, TG-2 conditions. Figure 2. The final morphologies of the neck under (a) ATG-1, (b) TG-1, (c) TG-2 conditions. Figure 3. Interaction between depth hoar structure formation and the temperature distribution calculated using COMSOLTM.");sQ1[1048]=new Array("../7337/2099.pdf","Characterization of Individual Particles in Air Quality Program with Sem-Eds","","2099 doi:10.1017/S1431927615011277 Paper No. 1048 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of Individual Particles in Air Quality Program with Sem-Eds R. Ramirez-Leal, M. Valle-Martinez, and M. Cruz Campas Sonora State University, Ley Federal del Trabajo Final s/n, Col. Apolo Hermosillo, Sonora, Mexico c.p. 83100 ramzl036@yahoo.com.mx In major cities worldwide air pollution is a constant that is a big problem to solve. Of all the pollutants commonly found in the atmosphere, the suspended particles are one of the objects of study that attract our attention because of the complexity and variety of the same, especially those known as PM2.5, also known as respirable fraction. [2] In recent years, research efforts have focused primarily on the PM2.5 fraction, and found that a typical sample of this fraction may contain, in addition to the elemental and organic carbon associated with vehicle emissions from the city's traffic, other miscellaneous types of compounds, such as sulfates, mainly calcium and ammonium; chlorides, especially sodium; nitrates, mainly ammonium; biological materials and various other organic compounds. Particulate matter of aerodynamic diameter less than 2.5 �m (PM2.5), has been found to be associated with urban health problems. Many epidemiological studies show that atmospheric aerosols may produce adverse health effects, with recent studies revealing that coarser atmospheric particles are more related to respiratory diseases, whereas the finest particles seem to affect the cardio-vascular system. [3] Characterization of aerosol samples at the level of individual particles, using microanalytical techniques, generally permits to obtain more unambiguous and detailed information than bulk analysis, and so it simplifies recognition of the sources of pollution and their processes. [1] The City of Hermosillo is located between latitude 20 � 01' 00 "and 20 � 08' 30" north latitude and between the meridian 110 � 54' 30 "and 111 � 01' 00" west longitude at an altitude of 200 m above sea level; is the state�s capital, located to the Center-West of the coastal plain, in the Northweast region of Mexico This work is based on samplings collected of the monitoring station of the Air Quality Improvement Municipal Program in the downtown area for the City of Hermosillo with the sampler Thermo Scientific FH 62 C-14 The samples were analyzed using a scanning electron microscope JEOL JSM 5800-LV model coupled to a system of energy dispersive x-ray (EDS) EDAX DX prime brand with a lower detection limit per element of less than 0.1%. A section of about 1 cm� of the tape sample was cut into a circular shape and was placed in an aluminum sample holder by a strip of double-sided copper; were coated with a thin platinum film with thickness of approximately 100 � and placed into the SEM to determine the chemical composition and morphology of particles at the individual level. Based on EDS results, these particulate matter was primarily composed of Fe (w% 21.74), Si (w% 10.57), Ba (w%3.86), and Al (w%2.66). (Fig. 1). The structure of particles can be diverse and they usually present a two dimensional aspect. It is often found that such particles have a irregular morphology, spheroidal, outlying edges and fracture lines. Microsc. Microanal. 21 (Suppl 3), 2015 2100 Related to the chemical composition and morphology, the analyzed particles were classified into the most abundant groups such as soot, Si-rich particles, sulfates, metalrich particles. Table 1 The most of silica particles (probably Si oxides) and aluminosilicates (containing Al, Si, K, Fe and Ca) have irregular forms and come from soil. References [1] Ramirez-Leal, R., et al. (2009) Elemental Chemical Composition, Size and and M orphological Characterization of Individual Atmospheric Particles within an Air Quality Program. Microscopy and Microanalysis, 15, 1300-1301. [2] Ramirez-Leal,R., Esparza-Ponce,H and Duarte-Moller,A. 2007. Characterizatioin of inorganic atmospheric particles in air quality program with sem, tem and xas. Revista Mexicana de Fisica 53(3): 102-107. [3] Ramirez-Leal, R., Valle-Martinez, M. and Cruz-Campas, M. (2014) PhysicoChemical Characterization of Total Suspended Particles (TSP) Analysis by SEMEDS. International Journal of Advanced Research, 2, 815-817. Particle Type a) Litophilic source b) Industrial c)Resuspended Groups Oxides of silicealuminium Metal Rich Oxides of silicealuminium Sub-Groups Al, Si, Mn, K Bi, Fe and Pb Al, Si, K, Na, Pb and Ca Morphology Irregular Spheroidal Irregular and fracture lines Table 1.- Clasification main Sem-Eds of PM2.5 a b c Fig. 1 Particle Type: a) Litophilic source, b) Industrial, c) Resuspended");sQ1[1049]=new Array("../7337/2101.pdf","Towards Automated Segmentation Methods for 3D Tomography Studies of the Morphology of Carbon Nanoglobules in Chondritic Meteorites","","2101 doi:10.1017/S1431927615011289 Paper No. 1049 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Towards Automated Segmentation Methods for 3D Tomography Studies of the Morphology of Carbon Nanoglobules in Chondritic Meteorites K. J. Yakal-Kremski1,2, N. D. Bassim2, K. Scott3, and R. M. Stroud2 1. NRC-NRL 2. Naval Cooperative Fellow, 4555 Overlook Ave. SW, Washington, DC 20375, USA. Research Laboratory, Code 6366, 4555 Overlook Ave. SW, Washington, DC 20375, USA. 3. National Institute of Standards and Technology, Gaithersburg, MD 20899, USA. The presence of nano-scale, roughly spherical carbonaceous features has been shown in several samples of chondritic meteorites [1-4]. There have been a few proposed mechanisms by which these carbon nanoglobules may form [2, 5]. Quantification of nanoglobule prevalence, morphology, and relationship to other meteorite components can provide important evidence to elucidate how the nanoglobules form, and how the carbon contained therein may be altered by interactions with fluids and nearby minerals present on the asteroid from which the meteorite originated. It has recently been shown that 3D FIB-SEM tomography can be utilized to collect reconstruction volumes of chondritic meteorite grains, as was done on a grain from the Tagish Lake 5b sample [6]. A representative FIB-SEM cross-section is shown in Figure 1, showing the complex nature of the sample, as well as highlighting the selected nanoglobule regions of interest. Segmentation of the nanoglobules for the data set in [6] was achieved manually, which poses a significant challenge to further study of this and other samples, as it is very time consuming and subject to variation in the resultant output based on what individual is carrying out the segmentation. Several challenges arise when trying to formulate an efficient and effective automated method for segmentation of the nanoglobules and other constituent phases present in a chondritic meteorite particles. The porous nature and irregular geometries of the particles make segmentation of the entire volume challenging, as back-contrast from the far wall of pores makes automated identification difficult, and regions of which with similar contrast may be misidentified as the nanoglobules. Carbon epoxy infiltration, a common method for filling pores to even contrast, is not a viable method, as the nanoglobules are also carbonaceous, and would not be easily distinguishable from the background epoxy. Additionally, there are other carbon phases, large carbonaceous veins called clasts, which have similar contrast but need to be segmented as a separate phase. While not related to the segmentation of the nanoglobules per se, the sheer number of present phases is also daunting, which is reflected in the complex image histogram given in Figure 2. Identification of nearby phases may be important in determining how the nanoglobules may interact with the parent body [1, 7]. Despite the complex nature of the problem, an automated, or pseudo-automated, method for segmentation is desired, and will be attempted for the Tagish Lake sample. This method should also be portable for use with other meteorite samples or otherwise. An automated segmentation scheme would allow for much higher throughput, and a more regimented and less subjective definition for each of the phases desired for segmentation. This too would allow for further analysis of the Tagish Lake sample reported earlier, as only ~20% of the total reconstruction volume was analyzed due to the slow manual segmentation. Microsc. Microanal. 21 (Suppl 3), 2015 2102 References: [1] C. M. O'D. Alexander, et al., Geochimica et Cosmochimica Acta 71 (2007), p. 4380-4403 [2] L. A. J. Garvie and P. R. Buseck, Earth and Planetary Science Letters 224 (2004), p. 431-439 [3] K. Nakamura, et al., International Journal of Astrobiology 1 (2002), p. 179-189 [4] B. T. De Gregorio, et al., Meteoritics & Planetary Science 48 (2013), p. 904-928 [5] M. Saito and Y. Kimura, The Astrophysical Journal 703 (2009), p. L147-L151 [6] R. M. Stroud, et al., Proceedings of the 44th Lunar and Planetary Science Conference (2013) [7] C. D. Herd, et al., Science 332 (2011), p. 1304-1307 Figure 1. Cross-section of a FIB-SEM collected data set on a grain from a Tagish Lake meteorite sample (left). Image width is 3.1m. Many phases can be observed, including a large carbonaceous clast (C), a spongy magnetite particle (M), unfilled porosity, and many other phases. Insets on right are two nanoglobule regions of interest for identification and segmentation. Inset widths are each 140nm. Note that an internal void is present in the top nanoglobule, whereas the bottom nanoglobule appears dense. Figure 2. Histogram of a selected interior region of the microstructure cross-section shown in Figure 1. While individual peaks are distinguishable, there is a high extent of peak overlap. This makes simpler segmentation techniques difficult to use, due to high uncertainty of pixel allocation.");sQ1[1050]=new Array("../7337/2103.pdf","Combined Major and Trace Element Characterization of Tourmaline: Using EPMA to Address Elemental Fractionation by Laser Ablation","","2103 doi:10.1017/S1431927615011290 Paper No. 1050 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Combined Major and Trace Element Characterization of Tourmaline: Using EPMA to Address Elemental Fractionation by Laser Ablation Jared Wesley Singer1 and Marian Lupulescu2 1. 2. Earth and Environmental Science, Rennselaer Polytechnic Institute, Troy, NY, USA. Research and Collections, New York State Museum, Albany, NY, USA. Electron Probe Microanalysis (EPMA) and Laser Ablation Inductively Coupled Plasma Mass Spectrometry (LA-ICPMS) synergistically reinforce the accuracy and precision of the other. EPMA is a classic technique of X-ray microanalysis, but is subject to limitations in regards to light elements (H, Li, and Be, B) and trace element sensitivity. On the other hand LA-ICPMS is sensitive for trace element analysis and it is possible to measure Li and B. However, accuracy of LA-ICPMS hinges on internal standard (IS) compositions from EPMA and quantifiable fractionation behaviors. In this paper we present a method for total compositions using a fully-integrated EPMA/LA-ICPMS analytical strategy. EPMA provides an essential, independent analysis of laser ablation accuracy. Major element determinations were made by EPMA, using CAMECA model SX-100. Column settings included a 15 kV accelerating voltage, 20nA regulated beam current, and 5 micron spot size. Major element compositions were corrected in an iterative manner for Li, B, H, and O by stoichiometric formula recalculation of Selway and Xiong and the ZAF-type algorithm of Pouchou and Pichoir. Convergence is obtained in 3 or less iterations. Ideal totals within 1% were obtained for diverse species including elbaite, schorl, dravite, uvite, foitite and fluoro- tourmalines. Trace elements were measured by LA-ICPMS using a PhotonMachines Analyte.193, 4ns pulse, eximer laser coupled to a Varian 820 quadrupole ICP-MS. Spot analyses were conducted at 48% laser power, repetition rate 6Hz, 60s duration, spot size 80um square, and laser fluence of 4.95J/cm2. Time-dependent laser fractionation experiments were conducted under similar conditions for 300s durations. Helium carrier gas flowing at 0.70 L/min transports the ablated aerosol to ICPMS under dry, hot plasma conditions. IS 23 27 29 concentrations from EPMA are specified for Na, Al, and Si and concentrations are determined against NIST612 glass. Instantaneous fractionations (Ri) of major elements by laser ablation are calculated through direct comparison of LA-ICPMS spot analysis to results of EPMA for multiple internal standards (Ri = [C]LA(IS) / [C]EPMA ; table 1). Time-dependent fractionation (Rt) is calculated by the ratio of the initial to the final half of a 300s ablation (Rt =[C]i / [C]f ; Figure 1) . Major element data obtained by both EPMA and LA-ICPMS demonstrates large systematic errors (ranging >50%) that result from laser ablation processes. Instantaneous fractionations (depletions) are observed in the order F>>B>Fe>Mn~Na>Li>Si>Al>Ca (Table 1). Detailed analysis is given here for one elbaite specimen, demonstrating that time-dependent fractionation is not a suitable proxy for instantaneous fractionation (compare Table 1 and Figure 1); furthermore various tourmaline species show a diversity of time-dependent fractionations. For trace element determinations the magnitude of systematic error strongly depends on the choice of internal standard, however the analyst cannot know a priori which combination of analyte/IS is most appropriate. Fractionation behaviours of other tourmaline species and reference glasses will remain an area of continued research toward a multiple-IS correction scheme for accurate trace element data. Ongoing investigation includes composition of ablation ejecta (figure 2), dependence on laser fluence, crystal orientation dependence, and mechanisms of elemental fractionation during laser ablation. Microsc. Microanal. 21 (Suppl 3), 2015 2104 Instantaneous Fractionation Factor ( Ri = [C]LA / [C]EPMA ) Element Li B F Na Al Si Ca Mn Fe H O Total EPMA 0.92 3.47 0.93 1.39 22.77 16.02 0.28 0.293 0.05 0.38 stoichiometric 100.10 R (Na IS) 1.23 0.80 0.023 1.00 1.44 1.29 1.49 1.00 0.86 �0.06 �0.03 �0.15 �0.02 �0.21 � �0.03 �0.03 �0.44 R (Al IS) 0.88 0.56 0.016 0.70 1.00 0.89 1.01 0.70 0.63 �0.05 �0.16 �0.04 �0.22 � �0.05 �0.05 �0.43 �0.06 R (Si IS) 1.04 0.69 0.020 0.84 1.24 1.00 1.27 0.83 0.75 �0.15 �0.02 �0.21 � �0.02 �0.02 �0.42 �0.03 �0.05 Table 1. Major and minor elements of elbaite as determined by EPMA and instantaneous fractionation factors for LA-ICPMS are calculated from 60s spots for various IS (23Na, 27Al, and 29Si). Note that by definition, EPMA and LA-ICPMS are in agreement when R=1.0 Figure 1. Left: Time-dependent fractionation factors of elbaite (circle) and NIST612 (bar) show marked differences (Al IS; Al=1.0). Except for F and Hg, time dependent fractionations are smaller than instantaneous fractionations. Right: X-ray maps of the elbaite ablation pit (300s) shows Na enrichment in ejecta relative to Al and Si.");sQ1[1051]=new Array("../7337/2105.pdf","The First Solar System Solids as Revealed Through Slice-and-View Imaging","","2105 doi:10.1017/S1431927615011307 Paper No. 1051 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The First Solar System Solids as Revealed Through Slice-and-View Imaging Thomas J. Zega1 Lunar and Planetary Laboratory and Dept. of Materials Science and Engineering, University of Arizona, Tucson, AZ Introduction: Calcium-aluminum-rich inclusions (CAIs) are among the most significant materials within chondritic meteorites. In general, they are the largest physical objects in these primitive rocks, with some reaching centimeter sizes and being visible with the naked eye in hand specimens. They are composed of mineral phases that form at very high temperatures and which are predicted by thermodynamic models to be among the first solids to have condensed within a cooling gas of solar composition [e.g., 1]. Further, they have radiometric age dates that exceed those of all other solar materials [2]. CAIs therefore represent the first solids to have formed within our solar system, and analysis of them can provide a glimpse into the some of the earliest chemical and physical processes to have transpired during its formation [3-5]. The optical microscope and electron and ion microprobes have been used extensively to tease out information on the compositional properties of CAIs and have provided a wealth of information on them. Though steadily increasing, there are comparatively few studies that have investigated the microstructural aspects of refractory inclusions. With the advent of the focused-ion-beam scanningelectron-microscope (FIB-SEM) as a viable mineralogic tool [6] site-specific detailed analysis of CAI microstructure is possible. Here we apply FIB-SEM and slice-and-view imaging to reveal CAI microstructure in three dimensions. Samples and Methods: We examined a fluffy type-A CAI previously identified in a petrographic thin section of the Allende CV3 chondrite. This sample was chosen in order to investigate the microstructure of an inclusion believed to have formed via condensation in the early solar nebula. We used the FEI Helios 660 FIB-SEM located at FEI headquarters to slice and image a local region of the CAI measuring 30 �m wide. Each slice measured approximately 10 nm thick and a total of 410 slices were made. The image resolution is 3072 ! 2048 pixels with pixel dimensions of 10 nm ! 13 nm ! 10 nm for a total voxel size of 1300 nm3. Backscattered electron (BSE) images were acquired using an acceleration voltage of 2 keV and probe current of 800 pA. Results and Discussion: The local area that was measured is shown in Figure 1. This area is composed of the accretionary rim (AR) and the multi-layered shell around the CAI known as the Wark-Lovering Rim (WLR). The surface region of interest selected for slice-and-view measurements is composed of several mineral phases including pyroxene [(Mg,Fe)SiO3], anorthite (CaAl2Si2O8), melilite [(Ca,Na)2(MgAl)(Si,Al)2O7)], spinel (MgAl2O4), and hibonite (CaAl12O19). The pyroxene is the thickest surface layer ("13 �m) in this area of the WLR sequence followed by anorthite ("5 �m to 7 �m) and the spinel layer that is separated by smaller (<5 �m) subhedral melilite grains. A series of BSE images acquired from the slice-and-view data set are shown in Figure 2. The surface of the thin section occurs in the bottom of the image and reveals an undulating topology. Pyroxene is the thickest layer ("10 to 13 �m) in all slices of the data set, mirroring the trend observed at the surface and suggesting that its volume is relatively constant around the inclusion. Anorthite and melilite grains occur 1. Microsc. Microanal. 21 (Suppl 3), 2015 2106 on the right-most edge of the pyroxene layer and vary in size, from several �m to not appearing, through the image stack. Small (# 2 �m) grains of melilite and anorthite also occur within the pyroxene layer throughout the image stack. Several formation pathways have been proposed for the origins of WLRs around CAIs. These include condensation, metasomatic exchange, and flash heating, e.g., [3-5]. Pyroxene and melilite are considered primary phases formed via condensation in the early solar nebula, whereas some anorthite has been interpreted to form via secondary metosomatic alteration [3]. Previous analysis of this inclusion suggests that the pyroxene layer formed via condensation [7]. That we find anorthite and melilite inclusions within the pyroxene throughout the image stack points toward these phases having formed first, most likely via condensation, followed by nucleation of pyroxene around them. Examination of additional CAIs will reveal whether the observations made here are unique to this sample or are common to other type-A inclusions and thus indicative of broader scale processes within the solar nebula. References: [1] Ebel D.S. (2006) in "Meteorites and the Early Solar System II" ed. D.S. Lauretta and H.Y. McSween Jr. (University of Arizona Press, Tucson) p. 253. [2] Amelin Y. et al., Science 297 (2002), p. 1678. [3] Krot A.N. et al., Meteoritics 30 (1995), p. 748. [4] MacPherson G.J. (2004) in "Treatise on Geochemistry Volume 1 Meteorites, Comets, and Planets" ed. A. Davis (Elsevier, New York) p. 201. [5] Wark D. and Boynton, W.V., Meteoritics & Planetary Science 36 (2001), p. 1135-1166. [6] Zega T.J. et al., Meteoritics & Planetary Science 42 (2007), p. 1373. [7] Bolser D. et al., Meteoritics & Planetary Science (2013) p. #5358 (Abstract). [8] We acknowledge NASA for funding this work and Cliffe Bugge at FEI Company for help in acquiring the data set. Figure 1. BSE image of the WLR and CAI as viewed Figure 2. BSE images of several slices of the area shown in the white from the surface. Area measured with slice-and-view rectangle in Fig. 1. imaging outlined by the white rectangle. Spinel (Sp), melilite (Ml), anorthite (An), pyroxene (Px), and hibonite (Hb) occur in the WLR adjacent to the accretionary rim (AR).");sQ1[1052]=new Array("../7337/2107.pdf","Comparing Field Emission Electron Microprobe to Traditional EPMA for Analysis of Metallurgical Specimens","","2107 doi:10.1017/S1431927615011319 Paper No. 1052 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Comparing Field Emission Electron Microprobe to Traditional EPMA for Analysis of Metallurgical Specimens D.F. Susan, R.P. Grant, J.M. Rodelas, J.R. Michael, and M.C. Maguire *Sandia National Laboratories, Albuquerque, New Mexico The use of a thermal field emission electron source and other design changes have significantly increased the resolution of electron probe microanalysis (EPMA), especially in the 5-8kV range of accelerating voltage.[1,2] The analysis of particles/phases as small as 200 nm in diameter has been documented for geological and meteoritical specimens and a similar analytical volume was shown in a Sn-Ag solder alloy.[1,2] The following work highlights other examples of field emission EPMA for the analysis of fine-scale metallurgical microstructures. Features too small for traditional EPMA, such as fine-scale lamellar transformations and microsegregation in solidification microstructures, can now be analyzed successfully with FE-EPMA. The high resolution is obtained while simultaneously covering a long linescan length or a relatively large area for mapping. Figure 1 displays a two-phase lamellar structure formed in a multi-component precious metal alloy. The individual layers range from sub-micron to about a micron in width. A traditional EPMA line scan of six elements is shown in the left plot. A typical 15kV accelerating voltage was employed for this trace with one-micron spacing of analysis points. With conventional microprobe, the lamellar phase compositions can just barely be discerned; silver-rich and copper-rich phases are alternating in the microstructure. In some places only one analysis point is contained within a phase and there is likely significant overlap of analyzed volume from the surrounding phases. The Pd concentration appears relatively flat and the minor constituents of Au, Pt, and Zn show slight fluctuations in each phase but it is difficult to quantify the compositional changes. In the right-hand plot, an FE-EPMA line scan of similar length is exhibited, obtained on a JEOL JXA-8530F at 7kV with 0.1 micron spacing. The lamellar phase compositions are now easily determined and the segregation of the minor elements between the two phases is clearly evident as well. The data can be used for phase diagram analysis and to characterize the lamellar discontinuous precipitation mechanism.[3] In Fig. 2, FE-EPMA analysis of an austenitic stainless steel weld structure is presented. As shown in the elemental map of Ni, the FE-EPMA is able to discern fine scale intercellular segregation of Ni in the weld structure. In the linescan plot, the FE-EPMA results are compared to a traditional EPMA trace. Although the traditional EPMA trace is able to show the general segregation pattern, it does not accurately determine the peaks in Ni concentration. Again, only a few data points are collected in the peaks and valleys of composition. Note the two traces were taken in slightly different locations so the interdendritic patterns do not exactly align spatially. These examples show that FEEPMA can be a powerful analytical tool for studies of solidification/casting and welding as well as many other metallurgical processes producing fine scale compositional fluctuations. [1] P. McSwiggan et al., Micros. Microanal. 17 (Suppl 2), 2011, 624-625. [2] J.T. Armstrong et al., Micros. and Analysis, 27(7), Nov. 2013, 17-20. [3] D.F. Susan et al., Met. Mat. Trans. A, 45(9), Aug. 2014, 3755-3766. * Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary Lockheed Martin Corporation, for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000. Microsc. Microanal. 21 (Suppl 3), 2015 2108 5 mm 50 50 40 40 30 30 At % At % Au Pt Ag Pd Cu Zn 20 10 Au Pt Ag Pd Cu Zn 20 10 0 0 2 4 6 8 10 0 Distance (microns) 0 2 4 6 8 10 Distance (microns) Fig. 1. (top) FE-EPMA elemental maps showing concentration of Ag obtained from a multicomponent precious metal alloy. (bottom) EPMA linescan through the lamellar structure obtained with a traditional EPMA instrument and an FE-EPMA linescan at 7kV on a JEOL JXA-8530F. 14 12 10 Wt. % Ni 8 6 FE-EPMA Traditional EPMA 4 2 0 0 2 4 6 8 10 12 14 16 18 20 Distance (microns) Fig. 2. (left) BSE photo and Ni elemental map from a stainless steel weld microstructure. (right) Traditional EPMA and FE-EPMA linescans for Ni.");sQ1[1053]=new Array("../7337/2109.pdf","Influence of thermo-mechanical treatments on the microstructure and hardness of the Al-2024 alloy","","2109 doi:10.1017/S1431927615011320 Paper No. 1053 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Influence of thermo-mechanical treatments on the microstructure and hardness of the Al-2024 alloy C.G. Garay-Reyes, L. Gonz�lez-Rodelas, E. Cuadros-Lugo, I. Estrada-Guel and R. Mart�nez-S�nchez Centro de Investigaci�n en Materiales Avanzados (CIMAV) Miguel de Cervantes No. 120, 31109, Chihuahua, Chih., M�xico. The Al-2024 alloy as a heat-treatable material, exhibits certain excellent properties, such as high tensile strength, good damage tolerance and creep resistance. With the excellent thermal stability, these alloys are considered to be candidate materials for future applications in the aerospace industry [1]. Actually, recent demand for weight reduction in structural component calls for further enhancement of strength of commercial structural alloys. Increasing recognition has been given to the thermos-mechanical treatments recently as important techniques for improving the properties of metallic materials [2]. The plastic deformation process may be useful in the effect of aging process and helpful to improve the mechanical properties of Al-2024 alloy. Some studies have shown that the plastic deformation process may be useful in the effect of aging process, which is helpful to improve the mechanical properties of aluminum alloy and refine the precipitated phase of alloys notably [3]. However, there are few studies available concerning the effect of the pre-deformation degree on the microstructure and precipitation kinetics. The Al-2024 alloy was melted in a LINDBERG BLUE electric furnace at 740�C, degassed for 5 minutes with Argon gas (20 psi), using a graphite propeller at 490 rpm and finally 0.33 wt% of Al-5Ti-1B as grain refiner was added. The alloys were cast into steel molds preheated at 260�C, where specimens of approximately 101.39 mm long x 12.64mm wide x 9.57 mm height were obtained. Later these specimens were machined to obtain samples of approximately 97 mm long x 10 mm wide x 8mm high. Subsequently, hot rolling (pre-deformation) at 460 �C was carried out to reduce the thickness of the sample 50% and erase the as-cast microstructure. Subsequently, solution treatments at 495�C for 3, 5 and 7h were done in a LINDBERG-BLUE electric furnace followed by a quenching in water at 60�C. Thereafter, a cold rolling was carried out to reduce the thickness of the sample 5 and 15%. Finally, samples were cut approximately 19.46 mm long to be treated by aging in a FELISA furnace at 195�C for different periods of time. The microstructural evolution was studied by optical microscopy (OM), scanning electron microscopy (SEM) and transmission electron microscopy (TEM). The effect of homogenization-solubilization treatment time (3, 5 and 7h) on microstructure in Al-2024 alloy is shown by SEM micrographs in Fig. 1. It is observed from the results that in all homogenizationsolubilization treatment times, segregation decrease compared with as-cast conditions, but samples treated for 5 and 7h shown lower segregation. The Fig. 2 shows HRB and HV harness results as a function of aging time in the 2024 alloy, after 5 (a) and 15% (b) cold-working, additionally, the reference sample value is included. It is observed as the hardness in cold-working samples is greater than the reference samples. It is observed a direct effect of cold-working in hardness, high deformation - high hardness. Additionally, the cold-working affects the precipitation kinetics; it is observed that the time required to reach peak hardness in deformed samples is shorter than that observed in not deformed samples, it is expected a noticeable effect on the morphology, Microsc. Microanal. 21 (Suppl 3), 2015 2110 2098 size and distribution of precipitates. In addition, the precipitation kinetics is faster in the samples 15% cold-working compared to 5% cold-working and reference sample, which is in agree with reported in other studies [5]. References [1]S.K. Ghosh, J Mater Sci Technol 27 (2011), p. 193-198. [2] AL Ning, ZY Liu, SM Zeng, Trans Nonferr Met Soc China 16 (2006), p. 1341�1347. [3] Hengcheng Liao et al, JMEPEG 20 (2011), p. 1364�1369. [4] Zhanli Guo, Wei Sha, Mater Trans 43 (2002), p. 1273-1282. [5] Xiaofeng Xu et al, J Alloy Compd 610 (2014), p. 506�510. As-Cast Solubilized 3h Solubilized 5h Solubilized 7h Figure. 1Images obtained by SEM in Al-2024 alloy as function of homogenization-solution treatment time. Figure. 2 Hardness HRB and HV curves as a function of aging in Al-2024 alloy solubilized for 5h . Hardness of 5% and 15% cold-working and reference samples are shown.");sQ1[1054]=new Array("../7337/2111.pdf","Development of Quantitative Methods for Estimation of Aluminum Alloys Structure by Means of Image Analysis","","2111 doi:10.1017/S1431927615011332 Paper No. 1054 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Development of Quantitative Methods for Estimation of Aluminum Alloys Structure by Means of Image Analysis Alexander A. Kazakov 1, Alexander Kur1, Elena Kazakova1 1. St.Petersburg State Polytechnical University/Metallurgical Technologies Department, St.Petersburg, Russian Federation Quantitative methods for evaluation of the microstructure of hypoeutectic aluminum-silicon alloys were developed and implemented as a plug-in for the Thixomet PROTM image analyzer. Al-6Si-2Cu and Al10Si-2Cu-1Ni alloys in unmodified state and after modification were investigated. That allowed obtaining various degree of fineness of Al-Si eutectic that covered all classes of modification in accordance with the standard bar charts of the American Foundry Society (AFS). Coefficient of eutectic fineness of these alloys was developed: at the binary image of silicon particles the average distance between all pixels located within the particles to their boundaries was calculated. The reciprocal of this average distance is called a coefficient of eutectic fineness and describes all classes of modification in accordance with the standard bar charts of AFS [Fig. 1a]. Regression equations were calculated: (1) UTS(Al6Si2Cu) = 175.56-12.28*Vi-2.13*Vp+ 63.81*Kd UTS(Al10Si2CuNi) = 231.54-12.28*Vi-2.13*Vp+75.33*Kd (2) (3) A5(Al6Si2Cu) = 2.993-1.102*Vi-0.00964*Vp+2.059*Kd A5(Al10Si2CuNi) = 0.404-0.081*Vi-0.00964*Vp+0.714*Kd (4) UTS � ultimate tensile strength, MPa; A5 � relative elongation, %; Vi � volume fraction of coarse intermetallic inclusions (greater than 5 micron); Vp � volume fraction of pores; Kd � coefficient of eutectic fineness. They adequately describe the mechanical properties of the investigated cast alloys based on the developed coefficient of eutectic fineness and also the volume fractions of porosity and coarse (greater than 5�m) intermetallic inclusions evaluated by using ASTM E 1245 (Fig. 1 b, c). To evaluate fineness of eutectic, the volume fraction of coarse intermetallic inclusions and porosity, panoramic images of microstructure have been obtained at a magnification of �1000 with square 0.5 mm2. Intermetallic phases have been revealed by etching in an aqueous 30% solution of caustic soda for 5-10 s. To interpret the nature of the phases which have been found by microprobe analysis, thermodynamic modeling was applied, using FactSage software powered by SGTE databases [1]. Nomograms which illustrate the effect of each of the studied parameters of the structure on the mechanical properties of Al-6Si-2Cu and Al-10Si-2Cu-1Ni alloys were plotted. Developed methods for forecasting hypoeutectic aluminumsilicon alloy properties by their structure can be used in their production for quality control and for acceptance tests. Wrought aluminum alloy AA6063 has been examined in as-cast condition. Evaluation of grain structure and completeness of Fe-rich phase transformations has been carried out. To reveal the structure, anodizing with Barker's reagent for 2 min at 25 V dc was employed. A special algorithm to calculate histograms of the grain size distribution and to determine average grain size has been developed for digitalizing of obtained colored images (Fig. 2a). Microsc. Microanal. 21 (Suppl 3), 2015 2112 Wrought aluminum alloys usually contain essential amounts of Fe-rich intermetallics located on boundaries of dendritic cells [2]. These intermetallics are characterized by different morphology determined by their temperature-time nature and crystal structure: large plate-like -phase and compact spherical or skeletal -phase. -phase deteriorates surface quality of extruded product and decreases ductile properties of the alloy [3]. To eliminate this negative effect, a homogenization process is used, which results in transformation of -phase into -phase. Completeness of this process is measured by the ratio of -phase to -phase. To separate Fe-containing phases, chemical selective etching by 10% water solution of H2SO4 heated to 70 �C has been used; this process is accompanied by coloring of -phase particles in brown-black color, while -phase is outlined, which allows separation of these phases by gray level during image analysis. Another method to differentiate -phase from -phase particles using elongation ratio has also been applied: elongation ratio should exceed 5 for particle to be -phase (Fig. 2b). It has been found, that separation of Fe-containing intermetallics according to elongation ratio allowed distinguishing of Fe-rich phases more accurately in comparison with analysis based on gray level. References: [1] Bale C.W., Chartrand P., Degterov S.A., Calphad, 26, No. 2, (2002), p. 189. [2] Mondolfo L.F., Structure and properties of aluminum alloys (Metallurgiya, Moscow), p. 640. [3] Ji S., et al, Materials Science & Engineering, 564 (2013), p. 130. Figure 1. Evaluation of fineness coefficient according to the AFS grade of Al-Si eutectics (a), correlations between calculated and experimental values for ultimate tensile strength (b) and relative elongation (c) of Al-6Si-2Cu (red) and Al-10Si-2Cu-1Ni (blue) alloys. Figure 2. Recognition of grain boundaries in as-cast AA6063 alloy using new algorithm for colored images (a) and elongation ratio analysis of Fe-rich intermetallics (b).");sQ1[1055]=new Array("../7337/2113.pdf","Microscale torsion test to investigate initial yielding phenomenon of a singe cystal copper","","2113 doi:10.1017/S1431927615011344 Paper No. 1055 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microscale torsion test to investigate initial yielding phenomenon of a single crystal copper T. Yokoyama,*1 N. Shishido,*1,* 2 K. Koiwa*1,* 2, S. Kamiya,*1,* 2 H. Sato,*1,* 2 M. Omiya,*2, *3 M. Nishida,*1,* 2 T. Suzuki,*4 T. Nakamura,*4 T. Nokuo,*2, *5 and T. Suzuki*2, *5 *1 Nagoya Institute of Technology, Gokiso-cho, Showa-ku, Nagoya, Japan *2 Japan Science and Technology Agency, Chiyoda-ku, Tokyo, Japan *3 Keio University, 3-14-1 Hiyoshi, Kohoku-ku, Yokohama, Japan *4 Fujitsu Laboratories Ltd., 10-1 Morinosato-Wakamiya Atugi, Japan *5 JEOL Ltd, 3-1-2 Musashino, Akishima, Japan Mechanical property of copper is very important for the reliable design of semiconductor devices, because the interface between a copper line and the dielectric barrier layer is the week point in the device. In general, the line width of copper line is in the range from tens nanometer to micrometer, and the plastic property of a crystalline metal with small dimensions is highly sensitive to its dimension [1]. In addition, it appears that the loading condition significantly affects the plastic behavior of a micron scale metal. C Motz et al found the follow stress of almost 1GPa (i.e. nearly equal to theoretical strength of copper crystal) at the micro-bending test [2]. Considering these reports, in order to comprehend deformation behavior of the copper embedded in semiconductor device, we need to investigate mechanical response of the copper with small dimensions under heterogeneous stress distribution. In this paper, we twisted micron copper specimens with different dimensions, and investigated the size effect of its plasticity, especially focused on the initial yielding. Fig. 1(a) shows the specimen subjected to the torsion test with the small dimensions from submicron to microns. A single crystal plate of copper (purity 99.9999% and {110} surface orientation) was prepared by cutting out of the ingot grown by Czochralski method. The copper plate was milled by focused Ga+ ion beam to shape the specimen. The arc-shape on the {110} plane was formed by FIB milling from the top, and the bottom of the arc was removed from the lateral face. Finally, FIB stripped off the surface layer of 200nm thickness from the specimen with 10pA and 30kV of Ga+ beam in order to reduce the damage originated by Ga+ bombardment. The dimension of the fabricated specimen was represented by the specimen side length a depicted on Fig. 1(b). Three different types of specimens were successfully fabricated with a=0.3�m, 1.0�m, and 1.5�m. The stress distribution was shown in Fig. 1(c). Microscale torsion test was conducted inside the scanning electron microscope (JIB4600F, JEOL Ltd., Japan). Fig.2 shows the secondary electron image of the torsion test observed with the acceleration voltage of 15kV. The specimen was pushed on the end of the arc shape by the nanoindenter (PI87, Hysitron Inc., USA), which can measure load and displacement. The applied displacement of the nanoindenter was controlled with a constant rate of 10nm/s. Load-displacement curve of the specimen with a=1.0�m was shown in Fig. 3. The linear part of the obtained curve corresponds to the fully elastic response of the specimen. The proportional limit of the linear part can be defined as the initial yielding point, circled in Fig. 3. Maximum resolved shear stress on the slip system of the copper crystal at the initial yield point was calculated as the initial yield stress of the corresponding specimens. The obtained yield stress was plotted in Fig. 4. The cross section of the specimen twisted by the maximum torque is (001) plane and the slip plane with maximum resolved shear stress is (-111) as shown in Fig. 4 in yellow and in red, respectively. The yield stress is considerably higher than the bulk [3], and we confirmed the size effect on it. And, the obtained result from torsion test also showed the apparently-similar tendency to that of compression test [4]. However, compression is deformed uniformly, and torsion is deformed starting from the surface of specimen. This difference of morphology is expected to affect the size effect of initial yielding behavior. Microsc. Microanal. 21 (Suppl 3), 2015 2114 References [1] M D Uchic, et al., Science 305 (2004), p. 986-989. [2] C Motz, T Schoberl, R Pippan., Acta Maerialia 53 (2005), p. 4269-4279. [3] M F Horstemeyer, et al., Journal of Engineering Material and Technology 124 (2002), p. 322-328. [4] D Kiener, A M Minor, Acta Materialia 59 (2011), p. 1328-1337 Figure 1. The specimen subjected to the torsion test; (a) Specimen fabrication by FIB milling, out of a single crystal copper. (b) The specimen side length a represents the specimen dimensions. (c) The stress distribution of cross section has maximum torque. Figure 2. SEM image of 1�m specimen during torsion test; (a) Before twisting (b) After twisting. Figure 3. The load-displacement curve of the specimen with a=1.0�m Figure 4. Initial yield stress among different specimen side length");sQ1[1056]=new Array("../7337/2115.pdf","RIMAPS Studies on Structured Copper Foils","","2115 doi:10.1017/S1431927615011356 Paper No. 1056 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 RIMAPS Studies on Structured Copper Foils M Flores Zavala1, P L�thman2 and EA Favret3,4 1 2 Instituto Sabato, UNSAM � CNEA, Buenos Aires, Argentina KIST Europe Forschungsgesellschaft GmbH, Saarbr�cken, Germany 3 Instituto de Suelos, CIRN, INTA. Hurlingham. Argentina 4 CONICET. Buenos Aires. Argentina It is known that RIMAPS (Rotated Image with Maximum Average Power Spectrum) technique allows finding the main directions of any surface topography and describing it by simple geometrical figures [1]. The maxima of the RIMAPS spectrum indicate main directions and the arrangement of the elements that constitute the surface. RIMAPS has previously been used in several fields, such as materials science and biology [2] [3]. In this opportunity, the objective was to apply RIMAPS to analyze the micro-nanostructured surface of copper foils which were manufactured via an electrolytic plating route. A high purity copper base foil was electroplated using a specific electrolyte bath. A defined ultraflat surface structure in combination with given mechanical properties was obtained. Electroplating an anti-tarnishing layer (sample A, see figure 1) or a roughening treatment with anti-tarnishing layer (sample B, see figure 2) on the smooth base foil give a specific micro-nanotopography that could be characterized and differentiated by RIMAPS. The samples were observed using a scanning electron microscope (FEI Quanta 250). The surface elements of both samples differ in size. The RIMAPS spectra of samples A and B show many maxima, indicating several directions of the surface topography (see figure 3). The two spectra are quite similar. This result is not strange because the basic manufactured process is the same. For a better determination of the main directions (main maxima of the RIMAPS spectrum), a smoothing of the RIMAPS spectra based on Fast Fourier Transform was used (see figure 4). From these spectra it is possible to say about the topography of the samples that sample A has approximately 11 directions and sample B between 12 and 13 directions. If sample A is observed with a higher magnification surface elements like stars can be detected (see figure 5). Therefore we propose a model based on the form of a "star" for describing the main directions and arrangement of the surface elements of these copper foils (see figure 6). The number of lines that forms the star is the number of maxima of the smoothed RIMAPS spectrum in a rotation of 360�. The star representing the topography has 22 lines for sample A and 24 or 26 lines for sample B. Nevertheless other extra studies need to be done to validate this model. Nowadays RIMAPS spectra of star drawings with different number of lines are being calculated. Other parameters of the spectrum are also being considered, such as the integral of the curve which increases as the number of lines of the star increases. In conclusion, a priori RIMAPS analysis seems to be a tool for developing models of geometrical figures that describes, in a simple manner, the micro-nanotopography of the surface. In the present case, a model based on stars could be appropriated [4]. References [1] NO Fuentes and EA Favret, Journal of Microscopy 206 (2002), p. 72-83. [2] SM Romero et al, Microscopy and Microanalysis 20 (Suppl 3) (2014), p. 1340-1341. [3] EA Favret and P. L�thman, Microscopy and Analysis 21 (1) (2007), p. 7-9. [4] The authors acknowledge the technicians of the microscopy laboratory of CICVyA, INTA as well as Laurence Vast at Circuit Foil Luxemburg for providing copper sheets. Microsc. Microanal. 21 (Suppl 3), 2015 2116 Figure 1. SEM micrograph of sample A Figure 2. SEM micrograph of sample B 1.0 1.0 0.8 MAPS (a.u.) MAPS (a.u.) 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0.0 0 20 40 60 80 100 120 140 160 180 RI (degrees) 0.0 0 20 40 60 80 100 120 140 160 180 RI (degrees) Figure 3. RIMAPS spectra of figure 1 (red line) and figure 2 (black line) Figure 4. FFT smoothing (5 points) of RIMAPS spectra of figure 3 Figure 5. SEM micrograph of sample A with a higher magnification Figure 6. Sketch of the star model");sQ1[1057]=new Array("../7337/2117.pdf","Microstructure Evolution of Discontinuous Precipitation and Coarsening of Co Rod-like Precipitates in Supersaturated Cu-Co alloys","","2117 doi:10.1017/S1431927615011368 Paper No. 1057 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microstructure Evolution of Discontinuous Precipitation and Coarsening of Co Rod-like Precipitates in Supersaturated Cu-Co alloys 1 N. M. Suguihiro, 2I. G. Sol�rzano, 1E. B. Saitvitch Centro Brasileiro de Pesquisas F�sicas, Rio de Janeiro, Brazil Department of Chemical and Materials Engineering - PUC-Rio de Janeiro, Brazil 1 2 Discontinuous precipitation (DP) is a solid state reaction in which precipitation takes place in a moving interface, generally a moving, incoherent, grain boundary. The transformation product is arrays of precipitates growing cooperatively and perpendicular to the moving interface. Among hundreds of alloy systems in which DP takes place, Cu-Co alloy is of great importance due to ferromagnetic character of Co precipitates. In this alloy system, anisotropic arrangement of precipitates generates magnetic anisotropy and consequently drastic changes in its magnetic properties. It is known that in diluted Cu-Co alloys, DP takes place at relatively low temperatures, ranging from 450 to 700�C with Co rod-like precipitates coherent with the matrix [1]. However, interesting magnetic properties have been reported in supersaturated Cu-Co alloys in condition in which DP is the dominant mode of precipitation after isothermal aging treatments [2, 3]. For this reason, it is of fundamental importance to understand the development of DP and coarsening in CuCo alloys. In this paper we investigated the growth of DP and coarsening in supersaturated Cu-10at. %Co alloys upon isothermal aging at temperatures ranging from 450 to 650�C, for periods of 5, 10, 30 and 60min. Due to nanometric size of precipitates (about 2nm in diameter), a TEM/STEM JEOL JEM 2100F operating at 200kV under conventional, scanning and analytical transmission electron microscopy was used to analyze the microstructure in detail. Nano beam diffraction (NBD) mapping NANOMEGAS technique has been used to evaluate orientation relationships between precipitate colonies and parent matrix as well as preferential directions of discontinuous reactions upon grain boundary migration. Results show that DP was the dominant mode of precipitation. Discontinuous precipitates are rodlike Co-rich, fully coherent with matrix with fcc structure, as confirmed by diffraction patterns. We verified a good stability of these precipitates at 450�C up to 60min of aging, with DP development in all high angle grain boundaries. At higher temperatures, coarsening takes place rapidly consuming discontinuous precipitates. This coarsening is also grain boundary driven, so-called discontinuous coarsening (DC): the grain boundary moves consuming DP colonies replacing them by coarsen rodlike precipitates, with fcc structure and still coherent with matrix at early stages. However, under aging for longer periods of time DC precipitates lose coherency and change orientation in a similar fashion as demonstrated by Takeda for coherency loss of homogeneous Co precipitates [4]. Fig. 1a and b show well-defined DP and DC colonies respectively. Fig.2a shows the coherency loss of a rodlike DC precipitate by emission of dislocation loops, seen as the ring-like contrast. EDS Co mapping in Fig. 2b confirm the presence of precipitate phase . After longer aging times, DC precipitates become incoherent with the matrix and change orientation adopting the orientation relationship (100)//(110), as can be seen in the NBD mapping in Fig.3. References [1] A. Perovic, G. R. Purdy, Acta Metall. 29, (1980) 53. [2]E. F. Ferrari, W. C. Nunes, M. A. Novak, J Appl. Phys. 86 (1999) 3010. [3] N. M. Suguihiro et al. J. Mat. Sci. 49 (2014) 6167. [4] M. Takeda et al. Phys. Stat. Sol. 168 (1998) 27. Microsc. Microanal. 21 (Suppl 3), 2015 2118 [5] Authors acknowledge CNPq for financial support. Figure 1 � TEM bright field images. (a) DP colony in sample aged at 550�C for 60min. (b) Discontinuous coarsening after 30min at 550�C. Figure 2 � (a) TEM bright field of semi-coherent DC after 60min at 650�C. (b) Corresponding EDS Co mapping. Figure 3 � NBD in a sample aged at 650�C for 60min. (a) Reconstructed bright field image. Notice the DC colonies, with precipitates with dark contrast. (b) Crystalographic orientation map.");sQ1[1058]=new Array("../7337/2119.pdf","Effect on Microstructure and Hardness of A2024 Aluminum Alloy Doped Cerium Oxide Nanoparticle.","","2119 doi:10.1017/S143192761501137X Paper No. 1058 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Effect on Microstructure and Hardness of A2024 Aluminum Alloy Doped Cerium Oxide Nanoparticle. F.J. Baldenebro-Lopez1,2,3, J.A. Baldenebro-Lopez2, A. Santos-Beltr�n3, V. Gallegos-Orozco3, C.D. G�mez-Esparza1 and R. Mart�nez-S�nchez1 1. Centro de Investigaci�n en Materiales Avanzados (CIMAV), Laboratorio Nacional de Nanotecnolog�a, Miguel de Cervantes 120, 31109 Chihuahua, Chih., M�xico. 2. Facultad de Ingenier�a Mochis, Universidad Aut�noma de Sinaloa, Prol. �ngel Flores y Fuente de Poseid�n, S.N., 81223 Los Mochis, Sinaloa, M�xico. 3. Universidad Tecnol�gica de Chihuahua Sur, Carr. Chihuahua a Aldama Km. 3.5, 31313 Chihuahua, Chih., M�xico. Aluminum alloys possess interesting properties that attract much attention due to their light weight, high specific mechanical resistance, high thermal conductivity, and moderate casting temperature. Automotive and aircraft parts of aluminum alloys such as A2024 are reinforced with oxides for its superior properties such as refractoriness, high hardness, high compressive strength, wear resistance etc., and make them suitable for use as reinforcement material in metallic matrices. Incorporating ultra-fine particles of metal-oxides significantly improves mechanical properties of the metal matrix by reducing the inter-particle spacing and providing their inherent properties to the metal matrix since they get uniformly embedded into it. In aluminum alloys a hardening mechanism is promoted by the homogeneous dispersion of low solubility or total insolubility particles. Several kinds of ceramic materials, e.g. SiC, Al2O3, MgO and B4C, are extensively used to reinforce aluminum and its alloys [1]. On the other hand, rare-earth compounds have been used mainly as corrosion inhibitors for aluminum alloys. Several studies have been realized in terms of cerium oxide protective coatings to improve the corrosion resistance of the A2024 alloy. The contribution of cerium oxides in the particle-dispersion hardening has not been explored in the aluminum alloys; nevertheless the increase of hardness and wear resistance of Sn by the incorporation of CeO2 nanopowders was reported [2]. On the basis of this report and the knowledge about the capability of aluminum alloys to be hardened by particle-dispersion, the aim of this work is focused on the potential effect of cerium oxide nanoparticles (nanoceria) on the strengthening of the A2024 alloy synthesized by two routes: casting and mechanical milling followed by sintering. Microstructural and hardness analyses (in samples with and without cerium oxide additions) were conducted in order to determine the combined effect of nanoceria and processing route on hardness behavior of A2024 alloy. Nanoceria powders were dispersed into A2024 alloy by high energy mechanical milling during 1 h under argon atmosphere followed by cold-compacted. The bulk samples were obtained by two routes: sintering at 500�C during 2 h and melting at 750�C; both under argon atmosphere. The results show the presence of Ce-Cu-rich needles in the microstructure of the as-cast condition, while CeO2 nanoparticles were identified in the sintered condition and the fabrication process (sintering vs. casting) has a significant effect on the microstructure and mechanical properties of the final alloy. And one may conclude that by MA and sintering was possible to avoid the excessive growth of cerium needles. References: [1] Y Ansary, et al., J Alloys and Comp. 484 (2009), p.400-404. [2] A. Sharma, et al., J Alloys and Comp. 574 (2013), p.609-616. Microsc. Microanal. 21 (Suppl 3), 2015 2120 (a) (b) (a) Figure 1. TEM images of the nanoceria: (a) dark field, (b) bright field. Figure 2. TEM dark field images of the powders: (a) Al+CeO2 and (b) A2024+CeO2. (b) Figure 4. EDS-SEM mappings of the as-cast alloy matrix. Figure 3. SEM image of the A2024+CeO2 as-cast alloy. Figure 6. EDS-SEM mappings of the sintered alloy matrix. Table 1. Microhardness of A2024 aluminum alloy doped. Sample Hardness (HV) A2024 64.86 � 12.79 A2024 + 5%CeO2 81.36 � 5.64 Al + 5%CeO2 103.12� 15.20 Figure 5. SEM image of the sintered A2024+CeO2 alloy.");sQ1[1059]=new Array("../7337/2121.pdf","Characterization of Precipitate Phases in a NiCoAlFeCrTi High Entropy Alloy by Transmission Electron Microscopy.","","2121 doi:10.1017/S1431927615011381 Paper No. 1059 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Characterization of Precipitate Phases in a NiCoAlFeCrTi High Entropy Alloy by Transmission Electron Microscopy. C.D. G�mez-Esparza1,*, F.J. Baldenebro-Lopez1,2 , J.A. Baldenebro-Lopez2 , R. Corral-Higuera2, J. M. Herrera-Ram�rez1 and R. Mart�nez-S�nchez1 1. Centro de Investigaci�n en Materiales Avanzados (CIMAV), Laboratorio Nacional de Nanotecnolog�a, Miguel de Cervantes 120, 31109 Chihuahua, Chih., M�xico. 2. Facultad de Ingenier�a Mochis, Universidad Aut�noma de Sinaloa, Prol. �ngel Flores y Fuente de Poseid�n S.N., 81223 Los Mochis, Sinaloa, M�xico. High entropy alloys (HEA) have appeared as an important and new kind of metallic materials in the last one decade with great potential in both basic scientific knowledge and applications due to the their superior properties. It is known that chemical composition has a significant effect on the microstructure of these alloys, and it is important to investigate the influence of microstructure on the mechanical behavior. In previous investigations, results of the microstructural and mechanical characterization of a NiCoAlFeCrTi high entropy alloy produced by mechanical alloying and conventional sintering by scanning electron microscopy, X-ray diffraction and microhardness were reported [1]. The superior hardness of this alloy in comparison to other similar high entropy alloys was attributed to the formation of nanosized Ti-rich phase. However, it was interesting to note a wide range of microhardness values, hence nanoindentation test was performed to avoid the effect of porosity on the microhardness test. The results were not different and a wide variation of nanohardness values was obtained. It was suggested that these variations in microhardness were not only due to the typical porosity of sintered products but also to the significant variations of chemical composition of phases and the presence of different precipitate phases. Therefore, microstructural observations by transmission electron microscopy were performed to characterize the sub-micron and nanosized precipitate phases. Besides the formation of Ti-rich phase described in the previous work [1], the microstructure of the sintered NiCoAlFeCrTi high entropy alloy exhibits the formation of different precipitate phases. Fig. 1 shows TEM elemental mapping of equiaxial sub-micron dark precipitates. In Fig. 2, an overlay of elemental mapping gives evidence about the formation of Ti-rich and Al-rich precipitates. Fig. 3 presents a TEM close-up of bright needle-like precipitates and their chemical composition (Cr-Fe-rich) determined by EDS analysis. References: [1] C.D. G�mez-Esparza et al, Microsc. Microanal. 18, Suppl 2 (2012) p. 1920. Microsc. Microanal. 21 (Suppl 3), 2015 2122 Figure 1. Elemental mapping by STEM-TEM of the NiCoAlFeCrTi alloy. Figure 2. An overlay of the Ti (red), Al (green) and Cr (blue) elemental mapping on the TEM image showing the presence of two different equiaxial precipitates. Figure 3. TEM image and EDS results showing the presence and chemical composition of bright needle-like precipitates.");sQ1[1060]=new Array("../7337/2123.pdf","Direct Observation of the Strain Aging Effects Using the in-situ Heating and Straining Stage for TEM","","2123 doi:10.1017/S1431927615011393 Paper No. 1060 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Direct Observation of the Strain Aging Effects Using the in-situ Heating and Straining Stage for TEM Seung-Pyo Hong1, Seong-Il Kim1, Tae-Young Ahn1, Chang-Sun Lee2, and Young-Woon Kim1 1. Research Institute of Advanced Materials, Department of Materials Science and Engineering, Seoul National University, Seoul, Korea 2. Technical Research Laboratories Plate Research Group, POSCO, Pohang, Korea One of the new concepts for American Petroleum Institute (API) X100 grade line pipe steels was the strain-based design (SBD) approach [1]. In order to fulfill the increasing demands for the harsh environmental applications, such as the artic and seismic area, the SBD was considered as a key solution for the X100 line pipe steels. Even though the strength could be diminished during the processing or designing, large uniform elongation is the key requirement for those applications. Many researchers have been focused on the alloy design to fabricate line pipe steels meeting both the transport efficiency and the performance by combining with microstructure and mechanical property analysis [2]. Full size X100 steel plate and pipe with 32mm thickness were selected and investigated in this study. The pipe shaping was achieved through UOE (U-ing, O-ing, and Expansion) piping process. The plastic deformation history of the surface was different from that of the center during the process. Tensile stress was applied to bottom of the plate along the transverse direction (TD) which corresponds to outer side of the pipe. On the contrary, compressive stress was applied to top of the plate along the TD. Due to this uneven deformation history, from the surface to the center, it was expected that the dislocation density and structure were different through the thickness. Furthermore, the tensile stress along the TD led to compressive stress along the longitudinal direction (LD), resulting in the Bauschinger effect [3]. The UOE process is typically followed by the anti-corrosion coating process, which requires heating the pipe up to 200-250�C. In that temperature range, solute atoms, such as carbon and nitrogen, can be diffused into the dislocation cores [4]. The yield point phenomena were revealed during the following tensile test on LD. Therefore, it is believed that the deformation behavior was determined by the result of the competition between the strain aging and the Bauschinger effect. The average grain size was measured by using scanning electron microscope (SEM). The dislocation structures of the plate and pipe were observed and analyzed with selecting several layers through the thickness by using transmission electron microscope (TEM). To investigate both the strain and the thermal effect on the strain aging behavior of SBD X100 steels, in-situ heating and straining TEM stage was designed and applied to test the steels. Each step of process conditions, such as heating and applying stress, was simulated in the TEM while observing the microstructural change. The strain aging behavior and the Bauschinger effect were directly confirmed by observing dislocation motions. Microsc. Microanal. 21 (Suppl 3), 2015 2124 References: [1] H Motohashi and N Hagiwara, Journal of Offshore Mechanics and Arctic Engineering 129 (2007), p. 318. [2] D Seo et al, Proceedings of International Offshore and Polar Engineering Conference (2009) p.61. [3] A Abel and H Muir, Philosophical Magazine 26:2 (1972), p.489 [4] A Cottrell, Proceedings of the Physical Society 62 (1949) p.49. [5] This research was supported by the Nano�Material Technology Development Program through the National Research Foundation of Korea funded by the Ministry of Science, ICT & Future Planning (2011-0019984) and the Development program(No.10040025) of the Korea Evaluation Institute of Industrial Technology grant funded by the Korea government the Ministry of Trade, Industry and Energy. Figure 1. TEM images of dislocation structures of the steel after piping process: (a) near surface, (b) one quarter, and (c) center of the sample Figure 2. TEM images captured during the in-situ straining. Dislocations (marked as red lines) were gradually unpinned from the precipitate under continuous loading.");sQ1[1061]=new Array("../7337/2125.pdf","Controlled Radiolytic Synthesis in the Fluid Stage. Towards Understanding the Effect of the Electron Beam in Liquids","","2125 doi:10.1017/S143192761501140X Paper No. 1061 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Controlled Radiolytic Synthesis in the Fluid Stage. Towards Understanding the Effect of the Electron Beam in Liquids Patricia Abellan,123 Lucas R. Parent,1 Trevor H. Moser,4 Chiwoo Park,5 Naila Al Hasan,6 Prabhakaran Munusamy,7 Ivan T. Lucas,8 Ilke Arslan,1 Jay Grate,1 Ayman M. Karim,6 James E. Evans,7 and Nigel D. Browning.1 1. 2. Fundamental & Comput. Sci. Directorate, Pacific Northwest National Laboratory, Richland, USA. SuperSTEM Laboratory, SciTech Daresbury Campus, Daresbury, UK. 3 School of Chemical and Process Engineering, University of Leeds, Leeds, UK. 4. Dep. of Mechanical Eng. & Eng. Mechanics, Michigan Tech. University, Houghton, USA. 5. Dep. of Industrial & Manufacturing Eng., Florida State University, Tallahassee, USA. 6. Institute for Integrated Catalysis, Pacific Northwest National Laboratory, Richland, USA. 7. Env. Molecular Sci. Laboratory, Pacific Northwest National Laboratory, Richland, USA. 8. UPMC Univ Paris 06, Sorbonne Universities, Paris, France. Studying liquid samples in the (scanning) transmission electron microscope ((S)TEM) represents specific challenges as compared to the study of crystalline/amorphous solid specimens. Solutions decompose upon e--beam irradiation through radiolytic processes and chemical species are generated in the liquid phase. These species interact with the sample, most times diffusing away beyond the observation area. For experiments requiring the use of solvents with higher chemical complexity than water, such as organic solvents for battery research and for most synthetic methods, a larger variety of radicals will be produced � with the consequent increase of the number of chemical reactions involved.[1,2] The fundamental understanding of the degradation mechanisms associated with the interaction of the electron beam with different liquid systems is a prerequisite to the accurate description of any dynamic processes. The perfect control of the radiolytic processes using adequate solution mixtures and/or procedures for low dose imaging of liquids in the (S)TEM will expand the range of dynamic processes that can be accurately explored. In this presentation I will emphasize the importance of understanding and controlling the radiation chemistry of the solvent, which determines the nature and the yields of the reactions involved in the chemical processes induced by the incident electron beam. To illustrate that, I will present our most recent experiments where we achieve an unprecedented control of the amount of the reducing agent involved on the synthesis of well-controlled size (2.2 nm) palladium particles.[2] I will first detail why some solvents can generate a very low amount of free radicals and therefore enable the precise control of the effect of dose. This is especially important if we take into account the large dose rates typically involved in (S)TEM experiments (within 7-8 orders of magnitude above those used with typical radiation sources [3]). I will also describe particular solvents that can provide one-radical net reduction conditions with no need of added substances for scavenging the oxidizing free radicals. Under such control conditions, we demonstrate a match between the reaction kinetics and reaction products of the radiolytic and chemical syntheses of the size-stabilized Pd nanoparticles. We quantify the effect of electron dose in the STEM on the nucleation kinetics and compare these results to in situ small angle Xray scattering (SAXS) experiments investigating the effect of temperature during chemical synthesis. Finally, I will present a synthetic method for the preparation of cerium based nanoparticles, i.e. ceria (CeO2) and its precursor phase cerium hydroxide III (Ce(OH)3), using the effect of the electron beam on Microsc. Microanal. 21 (Suppl 3), 2015 2126 precursor solutions of Ce(III) and Ce(IV) species [4]. In this case we achieve uniform 2.9 nm Ce(OH)3 particles for their conversion into CeO2. Ceria nanoparticles are particularly interesting because of their redox properties - especially in the 3nm range - which make them great catalysts for a wide range of industrial applications and very promising materials for clinical purposes. In this work, we discuss and interpret the effect of the electron beam on the thermodynamics of cerium, using simplified Pourbaix diagram (E-pH diagram), and the importance of the kinetics of the reactions involved. Overall, I will introduce methodologies for the precise control of nanoparticle synthesis in the STEM (see images of the final particles in Fig. 1) and provide a mean to uncover the fundamental processes behind the size and shape stabilization of nanoparticles. These methodologies can also be applied to a broader range of experiments requiring the use of complex solvents. The interest of combining results from two different in situ techniques for liquids at the nanoscale, SAXS and STEM will also be emphasized. References: [1] P Abellan et al, Nano Letters 14 (2014) p. 1293 [2] P Abellan et al, "Controlled Radiolytic Synthesis of uniform Palladium nanoparticles", Submitted [3] N.M. Schneider et al, J. Phys. Chem. C 118 (2014) p. 22373 [4] P Abellan et al, "Radiolytic in situ synthesis of nanometric cerium compounds", Submitted [5] The work involving development of insitu stages and solutions preparation was supported by the Chemical Imaging Initiative; under the Laboratory Directed Research and Development Program at Pacific Northwest National Laboratory (PNNL). PNNL is a multi-program national laboratory operated by Battelle for the U.S. Department of Energy (DOE) under Contract DE-AC05-76RL01830. A portion of the research was performed using the Environmental Molecular Sciences Laboratory (EMSL), a national scientific user facility sponsored by the Department of Energy's Office of Biological and Environmental Research and located at Pacific Northwest National Laboratory. SuperSTEM is the U.K. National facility for Aberration-Corrected STEM, supported by the Engineering and Physical Research Council (EPSRC). Figure 1. Pd nanoparticles observed through a layer of liquid (a), and cerium hydroxide nanoparticles on a washed chip, (b). Both set of uniform size particles were grown in the fluid stage.");sQ1[1062]=new Array("../7337/2127.pdf","Development of Quantitative STEM for a Conventional S/TEM and Real-Time Current Calibration for Performing QSTEM with a Cold Field Emission Gun.","","2127 doi:10.1017/S1431927615011411 Paper No. 1062 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Development of Quantitative STEM for a Conventional S/TEM and Real-Time Current Calibration for Performing QSTEM with a Cold Field Emission Gun. Stephen D. House1, Long Li1,2, C.T. Schamp3, Russell Henry3, Dong Su4, Eric Stach4, and Judith C. Yang1 1. 2. Dept. of Chemical and Petroleum Engineering, University of Pittsburgh, Pittsburgh, PA (USA) (Current affiliation) RJ Lee Group, Monroeville, PA (USA) 3. Hitachi High Technologies America, Inc., Dallas, TX (USA) 4. Center for Functional Nanomaterials, Brookhaven National Laboratory, Upton, NY (USA) The details of the size and shape of nanoparticle (NP) catalysts significantly impact their catalytic activity and effectiveness. Thus, the ability to characterize these materials at the relevant scales is critical to the rational design of improved catalysts. High-angle annular dark-field scanning TEM (HAADF-STEM) is particularly suited for the study of heterogeneous NP catalysts as it provides directly interpretable contrast primarily dependent upon the atom type and material thickness. Important information about the 3D structure, however, can be lost due to the 2D projection nature of TEM. Quantitative STEM (QSTEM) can recover atomic-scale 3D structural information from a single HAADF-STEM micrograph by taking advantage of the fact that with digital detectors, we are essentially counting electrons. Through careful calibration and measurement of the image and microscope, the contrast (scattered electron intensity) can be explicitly related back to the number of atoms involved in the scattering. This presentation discusses our developments on two different QSTEM approaches, one based on a conventional TEM/STEM and another on an aberration-corrected dedicated STEM. The earliest QSTEM work performed demonstrated the feasibility of counting the number of atoms in ultra-small NP using sufficiently high collection angles (100 mrad) [1,2]. It was further shown that this method could also indirectly recover details of NP shape (e.g., spherical, hemispherical, or plate-like) [3]. Microscopes have advanced greatly in the intervening years, however, and the old VG STEMs used in these studies have all but disappeared. Here we present our adaptation of this method to enable its use on a conventional, non-aberration-corrected S/TEM, a JEOL JEM 2100F S/TEM with no special attachments or modifications to the microscope required. Two Au NP specimens, synthesized via UHV e-beam evaporation, were examined in this study: Au NP supported on an ultra-thin carbon (UC) film, and Au NP deposited onto a -Al2O3 scaffold. The NPs were <2 nm in size, typically ~1 nm. The necessary calibrations for QSTEM were accomplished by utilizing the free-lens control functionality of the JEM 2100F, allowing for the proper weighting and quantification of the intensity scattered from each NP. From these values, the scattering cross-section for each NP was calculated and the number of atoms determined, c.f., Figure 1. More recently, the introduction of aberration-correctors enabled the development of atomic-resolution QSTEM, wherein by normalizing the intensities of an atomically resolved HAADF-STEM micrograph into units of fractional incident beam current, the number of atoms in each atomic column can be calculated through comparison with image simulations [4]. In combination with energy minimization computations, estimations of the particle's 3D morphology and atomic coordination can be reconstructed [5]. In this presentation we will discuss our development of a QSTEM technique on the aberration-corrected Hitachi HD2700-C STEM at Brookhaven National Laboratory. The cold field emission gun (CFEG) electron source of the microscope produces superior spatial coherence and energy Microsc. Microanal. 21 (Suppl 3), 2015 2128 spread than the more common Schottky FEG sources. The cost of these gains is a reduced stability of the emission current in a CFEG, which decays continuously (and non-linearly) and can vary even within an image. This renders existing methods for calibrating currents in QSTEM, which rely on a constant beam current, unsuitable for use with a CFEG. Our approach to overcome the instabilities of the CFEG is to measure the incident probe current in real-time concurrently with image acquisition. To do this, hardware has been installed to measure the signal from electrons impinging upon the objective aperture. By correlating this signal with the incident probe current, measured using a Faraday cup TEM stage, c.f., Figure 2, the aperture can act as a beam monitor, providing a measure of the instantaneous probe current without impeding imaging. By acquiring the beam monitor signal simultaneously with HAADF images, each pixel of the image can be calibrated individually using the incident beam intensity at the moment that pixel was acquired. In addition to enabling QSTEM to take advantage of CFEG sources, this method is also less impacted by the non-uniformities present in all imaging detectors [6]. References: [1] MMJ Treacy and SB Rice, Journal of Microscopy 156 (1989), p. 211-234 [2] A Singhal, et al., Ultramicroscopy 67 (1997) p. 191-206 [3] JC Yang, et al., Materials Characterization 51 (2003), p. 101-107 [4] JM LeBeau, et al., Nano Letters 10 (2010), p. 4405-4408 [5] L Jones, et al., Nano Letters 14 (2014), p. 6336-6341 [6] This work was supported by DOE BES through grant DE FG02-03ER15476, and performed using the facilities at the Nanoscale Fabrication and Characterization Facility at the University of Pittsburgh and the Center for Functional Nanomaterials at Brookhaven National Laboratory, which is supported by DOE BES through contract DE-SC0012704. Figure 1. QSTEM measurements of Au NPs on (a,b) UC film and (c,d) -Al2O3. (a) and (c) show HAADF-STEM micrographs with selected NPs labeled with their sizes. (b) and (d) plot histograms of cluster size. Figure 2. Probe current plotted versus beam monitor signal, showing a linear relation between the two, increasing with time/extraction voltage up to what appears to be a stable region. Error bars are the standard deviations.");sQ1[1063]=new Array("../7337/2129.pdf","High-Speed Analysis of Pt Based Alloys at High Spatial Resolution using EELS","","2129 doi:10.1017/S1431927615011423 Paper No. 1063 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 High-Speed Analysis of Pt Based Alloys at High Spatial Resolution using EELS Paolo Longo1, R.D. Twesten1, M. Bugnet2, S. Prabhudev2 and G.A. Botton2 1 2 Gatan Inc. R&D, 5794 W Las Positas Blvd., Pleasanton CA, 94588, USA Department of Materials Science & Engineering, McMaster University, Ontario, Canada L8S 4M1 Platinum based nano-particles with a narrow size distribution are very important from a technological point of view due to their chemical and physical properties and applications in the field of catalysis, information storage, membrane exchange fuel cells etc. However, their composition and size distribution influence their chemical and electrical properties and also the surface activity. This is the reason why a lot of attention has been paid on the development of synthesis methods capable to control the size, the composition and elemental distribution across the nanoparticles. Hence, the study of the chemistry and the elemental distribution is important in order to understand the properties of the nano-particles systems. Analytical transmission electron microscopy (TEM) related techniques have proved to be valuable tools for characterizing such materials. Data was acquired using a probe and image corrected FEI Titan STEM operating at 300 kV (McMaster University). The microscope is also equipped with a fully upgraded GIF Quantum ERS as EELS spectrometer. Two different sets of samples were analyzed. The first one is Pt-Fe catalyst nanoparticles and the second one is TiOx supported Pt-Ru core-shell catalyst nanoparticles. Both sets of samples were deposited on a carbon film supported on a Cu mesh TEM grid. EELS data was acquired in DualEELS mode in order to have both the low- and core-loss regions in the case of the Pt-Fe nanoparticles and to have two different core-loss regions at high energy for the collection of the Pt M4,5-edges at 2120 eV and the Ru L2,3-edges at 2838 eV and at low energy for the acquisition of the Ti L2,3-edges at 456 eV and the Ru M4,5-edges at 279 eV. In the case of the Pt-Ru catalyst system, provided there is enough signal, it is clearly more advantageous the collection of the Ru L2,3-edges at 2838 eV given the high peak-to-background ratio and also the fact that it does not overlap with other edges. Figure 1 shows the color-coded maps obtained from a large Pt-Fe catalyst particle with a diameter of about 60 nm selected because of the large field of view and fine sampling were required to study the surface chemistry and distribution of elements simultaneously. EELS spectra were acquired using an exposure time of 5 ms, a step size of about 0.1 nm and the entire area was mapped within a few minutes. It is interesting to notice the presence of a nicely defined FeOx as a shell layer. In Figures 2a,b, EELS spectra extracted from the outer shell, inner shell and core regions in the particle show that the shell consists of FeOx. The compositional analysis obtained from EELS quantification across the entire particle shows Pt at 40% and Fe at 60%. The projected concentration of Fe seems to increase slightly when measurements are made towards the shell region. The same type of analysis was carried out across other particles of much smaller dimensions with a diameter of approximately 10 nm. Here the FeOx shell layer seems to be thinner and less pronounced than in the case of the 60 nm particle. The composition was measured using EELS and appears to be Pt at 60% and Fe at 40%. The composition seems to be size dependent. Fe seems to be more abundant in large particles and as result the FeOx shell layer becomes more pronounced Advances in electron sources, spectroscopic detectors and software control now make it possible to acquire EELS data from an analytical STEM with high efficiency for heavy elements and Microsc. Microanal. 21 (Suppl 3), 2015 2130 high-energy edges. Moreover, the acquisition of the data using the DualEELS capability allows the dynamic range to be increased and the possibility to correct for thickness and diffraction effects that can be significant for heavy materials. Figure 1: RGB composite image of Pt M4,5-edges at 2120 eV in red, O K-edge at 532 eV in blue and Fe L2,3-edges at 708 eV in green. The shell layer appears to be very well pronounced and show the presence of Fe and O only. The inset shows the region at the interface with the shell layer in more details. Figures 2: a) Fe L2,3-edges at 708 eV and b) Pt M4,5-edges at 2120 eV extracted from the core, inner and outer shell respectively. It is quite remarkable to see how the shape of the Fe L2,3-edges changes across the three regions. This could be an indication that the Fe is chemically changing across the particle");sQ1[1064]=new Array("../7337/2131.pdf","Verifying the Structure and Composition of Prepared Porous Catalytic Supports","","2131 doi:10.1017/S1431927615011435 Paper No. 1064 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Verifying the Structure and Composition of Prepared Porous Catalytic Supports Michael T.Y. Paul,1 Xin Zhang,1 Brenden Yee,1 Byron D. Gates1,* 1. Department of Chemistry and 4D LABS, Simon Fraser University, 8888 University Drive, Burnaby, BC, V5A 1S6, Canada (email: bgates@sfu.ca). The incorporation of catalytic nanoparticles (NPs) within nanostructured materials have been popular in recent years due to the excellent surface area to volume ratios that can be achieved for these materials and their ability to reduce the required loading of precious metal catalysts. The use of Pt and Pd NPs is especially attractive for power generating systems, such as proton exchange membrane and methanol oxidation fuel cells [1-3]. Further studies on these nanoparticle incorporated materials have revealed the underlying support materials can enhance catalytic properties and stabilities of the NPs [4, 5]. However, due to processing constraints, loading of functional NPs is often limited to 2-D supports [7, 8]. The incorporation of NPs into 3-D porous supports can be performed by sophisticated systems, such as physical vapor and atomic layer depositions [7-9]. In this study, we demonstrate a relatively simple and cost effective method, which was developed previously in our research group, for preparing various combinations of NPs loaded into 3-D structured materials [10-13]. The NP loading and spatial distribution were characterized by scanning electron microscopy (SEM), scanning transmission electron microscopy (STEM), and energy dispersive X-ray spectroscopy (EDS). Our technique could be used to prepare novel electrochemical and photochemical porous materials for increased catalytic efficiency and stability. The method for preparing NPs loaded into a 3-D porous support involved the preparation of NP coated spherical polystyrene templates, self-assembly of these templates, infiltration with a support matrix, and removal of the sacrificial templates. The NPs (Pd or Pt) were mixed with the polystyrene templates at room temperature for 20 min to facilitate the first step of this process [10, 12]. These decorated polymer spheres were purified to remove excess NPs, and self-assembled into thin films using either air-water interface self-assembly [10, 13], drop casting [12], spin coating [14], or vacuum filtration [15]. A support matrix was filled into the interstitial spaces between these sacrificial templates, and the polymer was then removed to produce nanostructured supports coated with NPs. This support matrix was deposited by either material specific electrochemical deposition [13] or sol-gel techniques [15]. Templates were removed either by high temperature air oxidation (>300�C for >8 h) or room temperature digestion in organic or alkaline solutions [10-15]. A typical SEM analysis of Pd NPs loaded onto the surfaces of Ni nano-bowls is shown in Figure 1A. Through the use of concentric backscattered (CBS) electron detection, the Pd NPs can be distinguished from the Ni support by the contrast according to differences in atomic (Z) number (Figure 1B). Loading of NPs is further confirmed with STEM based EDS mapping. We analyzed samples prepared with different loadings of Pt NPs supported by a TiO2 matrix. The EDS mapping indicated our ability to tune the loading of NPs on the sacrificial templates and to correlate this with loadings achieved within the porous support (Figure 2). Samples prepared with undiluted, 60% diluted, and 90% diluted concentrations of Pt NPs had corresponding Pt loadings of 1.6, 1.1, and 0.1 particles per 100 nm2, respectively. These results demonstrated that 3-D NP decorated materials can be prepared with a tunable loading of Microsc. Microanal. 21 (Suppl 3), 2015 2132 noble metal NPs on a support prepared by sol-gel or electrodeposition techniques. The combination of microscopy techniques used in this study was imperative to verify the 3-D composition and structures of the materials. The analysis by SEM with a CBS detector provided a direct method to image these materials. The combination of STEM imaging and EDS analyses provided further verification of the material composition and spatial distribution (and loading) of the NPs. Additional investigations to be discussed include the preparation of cross-sections using focused ion beam (FIB) and lift-out techniques to further study the fine structures of the porous materials and the interface between the NPs and support materials that are imperative for electrochemical and catalytic applications. References: [1] O. Antoine, Y. Bultel, R.J. Durand, J. Electroanal. Chem. 10 (2010) p. 638. [2] P. Waszczuk, J. Solla-Gullon, H.S. Kim, Y.Y. Tong et al, J. Catal. 203 (2001) p. 1. [3] Z.P. Sun, X.G. Zhang, Y.Y. Liang et al, Electrochem. Comm. 11 (2009) p. 557. [4] A.S. Eppler, J. Zhu, E.A. Anderson, G.A. Somorjai, Top. Catal. 13 (2000) p. 33. [5] N. Toshima, T. Yonezawa, New J. Chem. 22 (1998) p. 1179. [6] Z.Y. Zhang, L. Xin, K. Sun, W.Z. Li, Int. J. Hydrogen Energ. 36 (2011) p. 12686. [7] T. Teranishi, M. Hosoe, T. Tanaka et al, J. Phys. Chem. B 103 (1999) p. 3818. [8] M.H. Park, K. Kim, J. Kim et al, Adv. Mater. 22 (2010) p. 415. [9] B. Lu, X. Li, T. Wang et al, Nano Lett. 13 (2013) p. 4182. [10] B.K. Pilapil, M.C.P. Wang, M.T.Y. Paul et al, Nanotechnology 26 (2015) p. 055601. [11] J. van Drunen, B. Pilapil, Y. Makonnen et al, ACS Appl. Mater. Interfaces 6 (2014) p. 12046. [12] M.T.Y. Paul, B. Kinkead, B.D. Gates, J. Electrochem. Soc. 161 (2014) p. B3103. [13] B. Kinkead, J. van Drunen, M.T.Y. Paul et al, Electrocatalysis 4 (2013) p. 179. [14] P. Colson, R. Cloots, C. Henrist, Langmuir 27 (2011) p. 12800. [15] A. Blanco, E. Chomski, S. Grabtchak et al, Nature 405 (2000) p. 437. [16] This research was supported in part by the Natural Sciences and Engineering Research Council (NSERC) of Canada, the Canada Research Chairs Program (B.D. Gates), and CMC Microsystems through the MNT Financial Assistance program that facilitated access to materials characterization services. This work made use of 4D LABS shared facilities supported by the Canada Foundation for Innovation (CFI), British Columbia Knowledge Development Fund (BCKDF), Western Economic Diversification Canada, and Simon Fraser University. Figure 1. Scanning electron micrographs of Pd NPs decorated onto the surfaces of Ni nano-bowls as imaged under the field-free mode by: (A) an Everhart-Thornley electron detector; and (B) a concentric backscattered electron detector. Figure 2. (A) Transmission electron microscopy high angle annular dark field (HAADF) image and Pt, Ti, and O EDS maps of TiO2 supported Pt NPs. HAADF image and Pt EDS maps of samples prepared from (B) 60% and (C) 90% less Pt NPs.");sQ1[1065]=new Array("../7337/2133.pdf","A Correlative Study of HRTEM, HAADF-STEM, and STEM-EELS Spectrum Imaging for Biphasic Electrochemically Active TiO2","","2133 doi:10.1017/S1431927615011447 Paper No. 1065 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A Correlative Study of HRTEM, HAADF-STEM, and STEM-EELS Spectrum Imaging for Biphasic Electrochemically Active TiO2 Chang Wan Han1, Vinodkumar Etacheri2, Chulgi Nathan Hong2, Vilas G. Pol2, and Volkan Ortalan1 1 2 . School of Materials Engineering, Purdue University, West Lafayette, IN, USA . School of Chemical Engineering, Purdue University, West Lafayette, IN, USA Nanocrystalline TiO2 polymorphs have been widely studied as candidate materials for the energy harvesting and storage applications, including photovoltaic devices and rechargeable batteries due to its excellent performances as well as low cost [ 1�3] . Interestingly, it has been reported that a mixture of the polymorphs shows improved photocatalytic activity compared to that of a single TiO2 polymorph [ 4,5] . D. C. Hurum et al. found the improved activity of the biphasic TiO2, Degussa P25 - a mixture of anatase and rutile, is due to the synergetic charge transfer effect between them [6]. Recently, we also found that the synergetic effect in a biphasic TiO2 anode significantly improves the capacity and the rate performance of lithium ion batteries. Considering the fact that the biphasic form of TiO2 outperforms single polymorphic phases in the energy-related applications, various forms of biphasic materials would be further developed. Since the polymorphs in the biphasic materials have identical composition with each other, distribution of the constituent polymorph in a biphasic material is not readily obtainable at high spatial resolution. Therefore, most of reports just provide relative amount of constituent phases determined by powder X-ray diffraction and/or Raman spectroscopy and high-resolution transmission electron microscopy (HRTEM) images with low magnification in which the spatial distribution of the polymorphs are not clearly shown [5,7,8]. In this respect, a systematic approach to characterize the biphasic materials is essential. In this study, we performed a correlative approach, including HRTEM, high-angle annular dark field (HAADF) scanning TEM (STEM), and STEM electron energy loss spectroscopy (EELS) spectrum imaging to characterize a biphasic (bronze + anatase) TiO2 nanosheets showing excellent electrochemical performance in lithium ion batteries. HRTEM images collected at a high magnification (over 800kX) were used to characterize atomic structures of the polymorphs [Fig. 1a]. Although HRTEM has been a good approach for investigating atomic structures, it may not be suitable for morphological studies due to the complicated contrast mechanism. Moreover, a limited field of view in HRTEM at high magnification makes difficult to get the distribution of the comprising phases. Therefore, we performed HAADF-STEM imaging, providing morphological features of nanocrystalline TiO2 polymorphs [Fig. 1b]. Finally, EELS spectrum imaging was performed to reveal the relative distribution of each TiO2 polymorph in the biphasic TiO2 nanosheets. Unique high-resolution electron energy loss near edge structures of TiO2 polymorphs, collected by the JEOL ARM200CF, an aberration corrected STEM with a cold field emission gun allowed us to determine how nanocrystallites of each polymorph are distributed. Collected spectrum images were de-noised by principal component analysis (PCA) and the distribution was obtained by multiple linear least square (MLLS) fitting [Fig. 1c]. The results of a correlative analysis will be provided and the correlation between the structural features and the exceptional performance of the biphasic TiO2 nanosheets will be discussed. Microsc. Microanal. 21 (Suppl 3), 2015 2134 References [1] B. O'Regan et al, Nature. 353 (1991) p. 737�740. [2] A.G. Dylla et al, Acc. Chem. Res. 46 (2013) p. 1104�1112. [3] V. Etacheri et al, ACS Nano. 8 (2014) p. 1491�1499. [4] A.G. Agrios et al, Langmuir. 19 (2003) p. 1402�1409. [5] Y. Chimupala et al, RCS Adv. 4 (2014) p. 48507�48515. [6] D.C. Hurum et al, J. Phys. Chem. B. 107 (2003) p. 4545�4549. [7] G. Li et al, Dalt. Trans. (2009) p. 10078�10085. [8] M. Estruga et al, Nanotechnology. 20 (2009) p. 125604. [9] We thank the Purdue University and School of Chemical Engineering for their generous start-up funding. Figure 1. Results of correlative electron microscopy studies for the biphasic TiO2: HRTEM images showing the atomic structures of TiO2 nanocrystallites (Fig. 1a), HAADF-STEM image for morphologies (Fig. 1b), and STEM-EELS spectrum images revealing the distribution of the constituent phase (Fig. 1c)");sQ1[1066]=new Array("../7337/2135.pdf","Interface Sharpness in Amorphous Multilayer Heterostructures and their Effect on Quantum Confinement","","2135 doi:10.1017/S1431927615011459 Paper No. 1066 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Interface Sharpness in Amorphous Multilayer Heterostructures and their Effect on Quantum Confinement Andrew Thron, Adam Schwartzberg and Shaul Aloni 1 Molecular Foundry, Lawrence Berkeley National Laboratory, Berkeley, CA 94720 Atomic Layer Deposition (ALD) is advantageous over physical vapor depositions techniques due to its ability to control the deposition of materials one atomic layer at a time. Moreover, ALD's self-limiting growth mechanism and its chemical flexibility allow precise deposition of insulators, metals and semiconductors with unprecedented precision in thickness and composition. TiO2 has gained recent attention due to the wide variety of applications in solar energy conversion, such as water splitting, catalytic break down of organic pollutants, and solar cells. Quantum confinement was previously shown to occur in amorphous TiO2 films [1], and is suggested as means to extend TiO2 absorption from UV to the visible range of a spectrum. In order to increase the spectral range in which TiO2 can absorb light a multilayer structure can be created, where the thickness of alternating TiO2 layers is used to control its absorption. In this study the effect of interface sharpness on the optical absorption properties, in amorphous TiO2/SiO2 multilayer films, is investigated using scanning transmission electron microscopy (STEM), energy dispersive X-ray Spectroscopy (EDX), and Electron Energy Loss Spectroscopy (EELS). TiO2/SiO2 multilayer films, nominally 2nm/10nm thick, were created on Si (100) substrates using ALD. Optical absorptions measurements of the multilayered film showed a blue shift in the absorption edge of the TiO2, compared with that of a 40nm thick TiO2 film. Cross-Section samples were created using the standard lift-out technique in a focused ion beam microscope (FIB). Amorphous nature of the layers has been confirmed through TEM and selected area diffraction (Figure 1). EDS and EELS spectral images were collected at 80 and 200kV, at ambient and 77K temperatures to study the effects of beam damage, beam spreading and radiation induced diffusion. An annular dark field (ADF) image in Figure 2a shows nine TiO2 layers, which are brightest in contrast. From the EDX line scans, a significant Si signal is detected within the nominal TiO2 layers, which may suggest intermixing of the layers (Figure 2b). ADF images and EELS line scans where acquired at 200kV to negate any artifacts caused by beam spreading (Figure 2). EELS line scans confirm that a Si signal is detected in the TiO2 layers, and that the signal is most significant in the bottom half of the layer. A close examination of O concentrations profile in Figure 2d reveals a similar trend to that of the Si profiles, which suggests that the top of the SiO2 layers are rough. This gives insight into the morphology of the film, but it is difficult to characterize the chemical sharpness of the TiO2/SiO2 interface. Our TEM results suggest that SiO2 layers grow with a 2.3-3.0nm roughness under these conditions. Furthermore residual presence of Si in the TiO2 layers is associated either with long residency time of the SiO2 precursor or interdiffusion of Si into the TiO2 film. Microsc. Microanal. 21 (Suppl 3), 2015 2136 References: [1] King D.M. et. al., Nanotechnology, 19 (2008), 445401 Figure 1. (a) TEM cross sectional, overview image of the TiO2/SiO2 Multilayer film on the Si substrate. (b) Selected Area Diffraction Pattern acquired from the area highlighted by the red circle in (a). Diffraction spots originate from the Si substrate. The diffraction pattern at the center of pattern originate from the multiple layers in the film. Figure 2. (a) ADF image of the multilayer film where the EDX spectra were acquired. (b) EDX line scan of the multi layered film showing the change in atomic percent of Ti, Si, and O across the film. (c) ADF image acquired from the area of the multilayer film where the EELS line scan was acquired. (d) Changes in the average, relative composition of the multilayer film, calculated from the EEL edges.");sQ1[1067]=new Array("../7337/2137.pdf","Influence of Device Microstructure on The Optical Properties of Ge1-ySny (y = 0-0.11) LEDs Produced by Next Generation Deposition Methods.","","2137 doi:10.1017/S1431927615011460 Paper No. 1067 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Influence of Device Microstructure on The Optical Properties of Ge1-ySny (y = 00.11) LEDs Produced by Next Generation Deposition Methods. J. D. Gallagher1,T. Aoki2, P. Sims3, J. Menendez1, and J. Kouvetakis3 1 2 Department of Physics & Astronomy, Arizona State University, Tempe, AZ 85287 USA LeRoy Eyring Center for Solid State Science, Arizona State University, Tempe, AZ 85287 USA 3 Department of Chemistry & Biochemistry, Arizona State University, Tempe, AZ 85287 USA Ge1-ySny alloys represent a promising new class of IR semiconductors with tunable direct band gaps beyond that of Ge, allowing the optoelectronic capabilities of the material to be significantly extended into the mid IR. These alloys become direct gap semiconductors for y = 0.09 [1], making them an attractive alternative to Ge for application in purely group-IV interband lasers that are integrated onto Si-platforms. Significant progress has been made in recent years with the demonstration of high efficiency photodetectors and light emitting diodes containing up to 8% Sn [2,3]. However systematic optical studies of devices with concentrations near the indirect-to-direct transition are still lacking. A major problem with the synthesis of highly saturated materials with y > 0.08 is the strong dependence of the lattice parameter on the Sn content. This produces a significant lattice mismatch with Si and Ge platforms, making it difficult to integrate materials with sufficiently low defect densities as required for viable device performance. Furthermore, the associated compressive strains become an issue, rendering the material more indirect and adversely affecting light emission. For the fabrication of practical devices, relaxed films with large thickness are desirable but such films contain deleterious misfit dislocations that increase the non-radiative recombination rate. In this paper, we report the development of optimized growth protocols that enable fabrication of a new class of photodiodes featuring thick, bulk-like components with purposely designed microstructures. The device configurations comprise nGe/Ge1-ySny/p-Ge1-ySny stacks grown directly upon Si(100). The constituent layers feature defectengineered interfaces, allowing the fabrication of devices exhibiting strong direct-gap electroluminescence over a wide concentration range 0 y 0.11 for the first time. The intrinsic active regions are grown largely relaxed on Ge-buffered Si following low-temperature routes based on specialty Ge3H8 and SnD4 chemical sources. The devices contain only one defected interface between the intrinsic layer and the Ge buffer, while all other junctions are selected to be lattice-matched and thus devoid of dislocations. Figure 1(a) is a Z-contrast image taken with a JEOL-ARM 200F showing the entire layer sequence of a sample comprising n-Ge, i-Ge, i-Ge0.895Sn0.105 and p-Ge0.95Sn0.05 components. The i-Ge layer in this case serves as a spacer between the bottom n-type contact and the active region to block threading dislocations from penetrating through. It also mitigates possible diffusion of the P dopants across the junction. All layers are highly uniform, exhibiting sharp and well-defined interfaces as required for effective device performance. Figure 1(b) is a XTEM image acquired on a JEOL-4000EX, showing a flat free surface and a defect-free top interface. The bottom interface is defected, as evidenced by the presence of multiple edge dislocations and dislocation loops confined to the plane of growth. The microstructure of the interfaces was further characterized by high resolution STEM, and representative BF images are shown in Figure 2(b,c) for Ge0.915Sn0.085 and Ge0.895Sn0.105 devices. The main types of defects accommodating the lattice mismatch are found to be 60o dislocations and short stacking faults originating at the interface and penetrating down into the lower energy Ge buffer. Higher defect densities are observed in the more concentrated alloy samples, as expected, leading to undesirable non- Microsc. Microanal. 21 (Suppl 3), 2015 2138 radiative recombination, thus limiting the overall optical response. Figure 2(a) shows that in contrast to the bottom (n-i) interface, the top (i-p) counterpart is defect-free, thereby reducing non-radiative recombination in these devices. Figure 2(d) shows EL spectra vs composition for a representative set of samples across the entire 2-11% Sn composition range. Note that the emission intensities increase as a function of Sn, as expected due to the reduction of the direct-indirect edge separation. An anomaly to this trend is seen between the 2% Sn and 5.5% Sn plots in the Figure. This is attributed to an increase in non-radiative recombination as the mismatch strains between the device components in the 5.5% device partially relax via generation of misfit dislocations [4]. References: [1] J. D. Gallagher, et. al., Applied Physics Letters, 105 (2014), p. 142102. [2] H. H. Tseng, et. al., Applied Physics Letters, 102 (2013), p. 182106. [3] Wei Du, et. al., Applied Physics Letters, 104 (2014), p. 241110. [4] This work was supported by the NSF. We acknowledge the use of facilities at the John M. Cowley Center for High Resolution Electron Microscopy at Arizona State University. Figure 1. Panel (a) Z-contrast image of an entire device stack including the Si substrate. Panel (b) Diffraction contrast image showing the typical interface and bulk layer microstructure of the same device in low magnification. Figure 2 (a) STEM HAADF image shows no defects at the top i-p junction. (b,c) STEM BF micrographs of defective Ge/i-Ge0.915Sn0.085 and Ge/i-Ge0.895Sn0.105 interfaces. (d) EL spectra of Ge1ySny heterostructure diodes. The plots represent EMG fits of the direct gap emission.");sQ1[1068]=new Array("../7337/2139.pdf","Inter-Diffusion Characterization of SnOx/CuOx Grown on Cu Foil","","2139 doi:10.1017/S1431927615011472 Paper No. 1068 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Inter-Diffusion Characterization of SnOx/CuOx Grown on Cu Foil Gabriel Eng, Greg Baty, Matthew Hughes, and Zhiqiang Chen. Center for Electron Microscopy and Nanofabrication, Portland State University, Portland, OR, USA Hetero thin films junctions with multi band gaps have potential to improve the efficiency of solar energy absorption and photoelectrolysis [1, 2]. Considerable efforts have been made to fabricate stable thin film heterojunctions with different dopants. Elemental inter-diffusion on the hetero-junctions caused by thermal treatment is generally considered as undesirable for interface stabilities. However, elemental inter-diffusion may lead to a formation of gradient-alloyed thin films with band gap modification. This may result in various band gaps with wide range. Single layer CuO and SnO2 thin films have received intensive attention because of their potential in cost-effective solar cell technology. CuO and Cu2O, ptype semiconductors with band gaps of ~1.5 eV and ~2.0 eV respectively, have been materials of interest due to relatively high optical absorption and low cost [3]. SnO2, an n-type semiconductor with a band gap of ~3.7 eV, is of interest because of its high optical transparency and low electrical resistance [4]. However, increased power conversion efficiency depends on the interfacial engineering of the junction, which requires a detailed understanding of Cu and Sn interfacial diffusion behavior. In this research, CuO thin film was deposited onto a Cu foil via RF sputtering in a KJ Lesker thin film deposition system. SnO2 film was deposited above the CuO film. The as-synthesized films were annealed in a furnace at 300�C for one hour then in-furnace cooled overnight. Cross-section TEM specimens were prepared from the thin films with and without annealing via a dual beam FIB microscope. Imaging and composition analysis were performed at 200 kV on a FEI TECNAI F20 TEM with an Oxford SDD EDS detector and a Gatan Imaging Filter. XPS depth profile was carried out on a PHI VersaProbe II at a vacuum of ~1.2E-6 Pa, scanning 200 �m 15 kV 50 Watts X-ray beam, pass energy 23.5eV. Before annealing, two apparent layers can be seen in Figure 1 a): about 250nm-thick CuOx and 1250nmthick SnOx above the poly Cu foil. After annealing, three layers formed above the poly Cu foil. Beyond 290nm CuOx and 223~375nm SnOx layers, there is a 765~950nm inter-diffusion layer CuxSnyO1-x-y. Interestingly, this inter-diffusion layer is Sn-rich, as shown in Figure 1 b). An increase of about 50nm in CuOx film is attributed to oxidation of Cu substrate during annealing. However, the thickness of SnOx layer reduced from ~1250nm to ~223-375nm after annealing. It suggests that the inter-diffusion layer resulted from the diffusion of Cu into to SnOx layer. It is interesting to note that little Sn diffused into the CuOx layer. It indicates that Cu transportation in SnOx is much faster than that of Sn in CuOx. Further quantitative EDS line scanning profiles shown in Figure 1c shows that the alloyed layer is Snrich oxide film with Cu alloyed. XPS depth profiles (Fig. 2a. and 2b.) were taken with pre-sputtered, annealed and non-annealed samples. The non-annealed sample had distinct regions showing little diffusion. The annealed sample had a more gradual onset of Cu signal in the Sn layer consistent with Cu diffusion into the Sn layer. In both cases, the Sn signal ceased promptly after the interface indicating little Sn diffusion into the CuOx layer. The depth profile on the annealed sample was continued until the O signal was no longer present. This is consistent with the EDS analysis. Highly Cu alloyed SnOx-based alloy film is formed between SnOx and CuOx layer. In the both annealed and non-annealed samples, XPS depth profiles revealed peaks from the CuOx layer that are more consistent with Cu2O rather than the CuO. This may be attributed to deposition in an oxygen-deficient environment under vacuum. Microsc. Microanal. 21 (Suppl 3), 2015 2140 In conclusion, we found that after annealing a heterojunction of thin films of CuO and SnO2 at 300�C, Cu tends to diffuse into the SnOx layer to form alloyed SnOx but minimal Sn diffused into the CuOx layer. SnOx-based alloy thin film with Cu addition can be fabricated via simple thermal treatment. Future work would include variation of experimental parameters to tune the diffusion process and study its effects on the band structure of the heterojunctions. Continued efforts are being made to optimize power conversion efficiency in solar cell devices fabricated from CuOx and SnOx. a b c Figure 1. STEM HAADF micrographs and elemental maps by EDS of the a) non-annealed and b) annealed and an EDS line scan profile c) of annealed SnOx/CuOx. The annealed shows significant Cu diffusion into the SnOx layer. 100 90 80 70 p _ _ g p 100 p _ _ g p a Atomic Concentration (%) Atomic Concentration (%) 60 50 40 30 20 10 References: [1] Michael Woodhouse and B. A. Parkinsonm, "Combinatorial Sn Sn approaches for the identification and optimization of oxide semiconductors for efficient solar photoelectrolysis", Chem. Soc. Rev., Vol. 38, 197�210, 2009. [2] L. El Chaar, L.A. lamont, N. El Zein, "Review of photovoltaic technologies", Renewable and Sustainable Energy Reviews Vol. 15, 2165�2175, 2011. [3] H. Kidowaki, T. Oku, T. Akiyama, A. Suzuki, B. Jeyadevan, and J. Cuya, "Fabrication and Characterization of CuO-based Solar Cells," Journal of Materials Science Research, vol. 1, no. 1, 138-143, Dec. 2011. [4] S. Baco, A. Chik, and F. Md Yassin, "Study on optical properties of tin oxide thin film at different annealing temperature," Journal of Science and Technology, vol. 4, no. 1, 61-74, 2012. 50 40 30 20 10 0 O Cu O1s Cu2p3 Sn3d5 90 b 80 70 60 O Cu O1s Sn3d5 Cu2p3 Figure 2. XPS depth profiles of the a) non-annealed and b) annealed SnOx/CuOx. 0 0 0 200 400 600 500 1000 1500 800 1000 1200 Sputter Depth (nm) 1400 1600 1800 Sputter Depth (nm)");sQ1[1069]=new Array("../7337/2141.pdf","Zernike Phase Contrast Electron Microscopy: Observation of the Image Formation and Improvement of the Image Quality using Direct Detector","","2141 doi:10.1017/S1431927615011484 Paper No. 1069 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Zernike Phase Contrast Electron Microscopy: Observation of the Image Formation and Improvement of the Image Quality using Direct Detector Kazuyoshi Murata1, Naoyuki Miyazaki1 and Kuniaki Nagayama2 1. 2. National Institute for Physiological Sciences, Okazaki Japan. The Graduate University for Advanced Studies (SOKENDAI), Hayama, Japan. Zernike phase contrast transmission electron microscopy (ZPC-TEM) is a technique to enhance the image contrast of the object using a thin-film phase plate, which has been practically achieved by Nagayama's group recently [1]. In ZPC-TEM, carbon thin film having a central small hole is inserted into the back focal plane of the objective lens, which delays the phase of the scattered electrons at /2 and enhance the phase contrast by interfering between scattered and unscattered electrons. This method is very useful for observation of unstained biological samples using cryo-EM. However, several factors including an electron loss by phase plate, and drift and charging of the phase plate limit higher resolution information at present [2]. Here, we applied a direct detector CMOS camera [3] to improve the ZPC image. The divided sub-frames revealed the image formation detail of ZPC-EM. The result was applied for a selection of COMS image frames. The ZPC images summed selected frames showed an improvement of the image quality. For observation of the image formation of ZPC, amorphous thin carbon films were imaged with conventional-EM and ZPC-EM, respectively, with an underfocus condition using JEM-2200FS equipped with an electron source of 200kV FEG (JEOL Inc.). The images were recorded on a direct detector CMOS camera, DE12 (Direct Electron LP) for 1 sec exposure time divided with 25 serial frames of each 0.04 sec. The total electron dose was ~30 e-/�2. As Zernike phase plate, it was used an amorphous thin carbon film at the thickness of 50 nm having a 500 nm central hole, which corresponded to 1/50 nm cut-on frequency. The each frame was processed by FFT and the resulting CTF profiles were compared. For Cryo-EM observation of biological specimen, ice-embedded sapovirus VLPs [4] on EM grid were inserted into the electron microscope using Gatan 914 cryo-specimen holder (Gatan Inc.). The images were recorded with the same procedures mentioned above with the electron dose of ~20 e-/�2. The serial frames were aligned each other and summed after selecting the effective frames. The CTF profiles taken by an underfocus condition using conventional EM and ZPC-EM were calculated in each frame and compared along the time course (Fig. 1). While the CTF profiles were constant in conventional EM frames, they were not in ZPC-EM frames where the CTF appeared from the low frequency side after 0.24 sec exposure time (arrow in Fig. 1) and finally became a constant. These results suggest that it needs a pre-irradiation for several hundreds milliseconds to form and stabilize the ZPC images. It also means that the first several frames don't significantly contribute to the image formation. The frame-divided imaging by direct detector was applied for an unstained biological specimen. Fig. 2A and B represent the ZPC images of ice-embedded sapovirus VLP summed with all and selected frames, respectively. In the selected summed image, the first 6 frames to 0.240 sec were omitted. It looked similar to the all summed image, but the intensity of the power spectrum was slightly higher in the selected summed image than the all summed image. It suggests that the image quality (contrast) is improved by selecting the frames in ZPC-EM. Microsc. Microanal. 21 (Suppl 3), 2015 2142 Here we first present that the ZPC image is formed after several milliseconds of exposure. The contrast of ZPC cryoEM image enhanced by omitting the first several frames improves the image quality of the ice-embedded sapovirus VLP. This procedure would be believed to push the image resolution further in singe particle three-dimensional reconstruction of protein molecules using ZPC images in the future. References: [1] K Nagayama and R Danev Philos Trans R Soc Lond B Biol Sci. 363 (2008) 2153�2162. [2] K Murata et al. Structure 18 (2010) 903-912. [3] AF Brilot et al. J Struct Biol 177 (2012) 630�637. [4] GS Hansman et al. FEBS Let 580 (2006) 4047-4050. Figure 1. CTF profiles of amorphous carbon thin film in serial frames by conventional EM (A) and ZPC EM (B). Serial frames were recorded at every 0.04 sec up to 1 sec (Arrows). The CTF profile appears after 6 frames in ZPC image (arrow head) while it appears from the first frame in conventional EM. Dots shows zero nodes in CTF. CTF images of the first and last frames are inserted in the graphs. Figure 2. Ice-embedded sapovirus VLPs were recorded on the direct detector with serial frames of 0.04 sec each by ZPC-EM. The serial frames were aligned and summed with all (A) and selected (B: omitted the first 6 frames) frames, respectively. Scale 50 nm. C) Power spectrum of all and selected summed images.");sQ1[1070]=new Array("../7337/2143.pdf","Zernike Phase Plate Configuration at Intermediate Lens Position on JEM2200FS","","2143 doi:10.1017/S1431927615011496 Paper No. 1070 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Zernike Phase Plate Configuration at Intermediate Lens Position on JEM2200FS Wah Chiu, Caroline J.-Y. Fu, Htet Khant and Sohei Motoki* National Center for Macromolecular Imaging, Verna and Marrs McLean Department of Biochemistry and Molecular Biology, Baylor College of Medicine, Houston, TX 77030, USA *JEOL USA, Inc., 11 Dearborn Road, Peabody, MA 01960, USA Cryo-electron microscopy is an emerging tool for structural biology to study biological specimens in their native conformational states. High contrast images with high signal-to-noise ratio are always beneficial for extracting features out of subcellular components in a cell tomogram or visualizing small macromolecules in single particle cryoEM. There are recent advances in enhancing image contrast using direct electron detectors and Zernike phase plate. The application of direct electron detector has demonstrated its usefulness in enhancing the low and high resolution contrast to facilitate the near atomic resolution structure determination of single particles. Zernike phase plate offers impressive low resolution image contrast to reveal structural details of subcellular components in a crowded environment of cell tomograms. Zernike phase plate is made of a thin carbon film with a central hole, and is positioned at the focal plane of an objective lens. This configuration would change the modulation of the contrast transfer function from sine to cosine function. Therefore, the image contrast at low spatial frequencies is greatly enhanced. Implementation of Zernike phase plate has been done at the objective lens aperture position without making extensive modification to an existing microscope. We have used such configuration to collect data on a number of biological systems [1,2]. Alternatively, one can place the phase plate at a position conjugate to the back focal plane of an objective lens. For instance, in JEM-2200FS, one can place the phase plate at the focal plane of the objective mini lens, which is at the selected area (SA) aperture position (Figure 1). In this configuration, the effective focal length for the phase plate is doubled (f= 6.3 mm, which is about twice that of an objective lens). The extent of image contrast enhancement depends on the cut-on frequency of the Zernike phase plate, which is inversely proportional to the hole diameter and directly proportional to the effective focal length of the lens (Figure 2). Placing the phase plate at the SA position would enhance the contrast of the images when using the same hole size as most of the phase plates currently available as well as allow the objective aperture to be used simultaneously. In order to use such phase plate configuration in low dose operation, we have to implement an imaging protocol for search, focus and photo mode different from the conventional one. For example, projector deflector coils instead of image shift coils are used for off-axis focusing. A preliminary test with a number of frozen hydrated biological specimens shows promising results. A tomographic data collection and reconstruction of mammalian cells are being presented to illustrate its general applicability. Acknowledgement: This research has been supported by a NIH grant (P41GM103832). References Microsc. Microanal. 21 (Suppl 3), 2015 2144 [1] Danev, R & Nagayama, K (2008) Single particle analysis based on Zernike phase contrast transmission electron microscopy. J Struct Biol 161(2):211-218. [2] Dai, W, Fu, C, Raytcheva, D, Flanagan, J, Khant, HA, Liu, X, Rochat, RH, HaasePettingell, C, Piret, J, Ludtke, SJ, Nagayama, K, Schmid, MF, King, JA, & Chiu, W (2013) Visualizing virus assembly intermediates inside marine cyanobacteria. Nature 502(7473):707-710. Fig 1: Electron optical ray diagram for phase plates (PP) located at selected area aperture (SA). OL, OLA and OL Mini stand for objective lens, objective lens aperture and objective mini lens respectively. Fig 2: Relationship between cut-on periodicity, phase plate hole diameter and objective lens focal length.");sQ1[1071]=new Array("../7337/2145.pdf","Zernike Cryo-EM with a Direct Electron Camera Enables Tracking Protein Conformations in the Temporal Dimension","","2145 doi:10.1017/S1431927615011502 Paper No. 1071 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Zernike Cryo-EM with a Direct Electron Camera Enables Tracking Protein Conformations in the Temporal Dimension Yi-min Wu1, Jen-wei Chang2, Chun-hsiung Wang3, Kuniaki Nagayama4,*, Naoyuki Miyazaki5, Kazuyoshi Murata6 and Wei-hau Chang7 1. 2. Institute of Chemistry, Academia Sinica, Taipei Taiwan 115. Institute of Chemistry, Academia Sinica, Taipei Taiwan 115 3. Institute of Chemistry, Academia Sinica, Taipei Taiwan 115 4. National Institute for Physiological Sciences, Okazaki, Aichi, 444-8585, Japan 5. National Institute for Physiological Sciences, Okazaki, Aichi, 444-8585, Japan 6. National Institute for Physiological Sciences, Okazaki, Aichi, 444-8585, Japan 7. Institute of Chemistry, Institute of Physics and Genomic Research Center, Academia Sinica, Taipei Taiwan 115 * Current Address: Sokendai, Hayamacho, Kanagawa, 240-0193, Japan Direct electron camera built on CMOS chips with high quantum yield and video frame rate has begun to revolutionize biological cryo-electron microscopy (cryo-EM). By recording a molecular image as frames instead of an accumulated exposure, post-imaging computation are available, by which the chargeinduced or drift-caused image blurring are to be corrected in 2D by aligning the same particle images in the movie. As such, the number of near atomic and that of sub-nanometer structures are rapidly increasing, with notable examples including a mosaic virus [1] and a TRP tetramer channel [2] imaged by 300 kV field emission instruments. To test how far one can reach using an average 200 kV field emission instrument equipped with a direct electron camera with the above-mentioned motion-correction approach, we studied a DGNNV virus with icosahedral symmetry, ~ 35 nm in diameter, and reached a near atomic resolution for its capsid in the core region from ~15000 single particle images [3]. As we attempted to expand the same approach to asymmetric but smaller protein complex, we resorted to the usage of Zernike phase plate that provides superior contrast than conventional defocusing microscopy [4-5]. To this end, we used a 200 kV field emission instrument at Okazaki equipped with a direct electron camera. We chose RNA polymerase II complex (~12 nm) as a benchmark molecule as its known atomic structure would help data interpretation and the possibility of revealing it by using Zernike cryo-EM were examined both by simulations [6-7] and by experiment [8]. However, we encountered a ~20 � resolution barrier for which we summed all the motion-corrected frames as mentioned above to yield an RNA polymerase II structure on which the surface motifs were blurred, suggesting the existence of other type of motions during the imaging period cannot be corrected for in 2D, including orientation change of particles. To this end, we resorted to Scheres scheme of running averages of 6 movie frames [9], corresponding to 0.25 second, and reached a sub-nanometer resolution structure for RNA polymerase II using ~40000 single particle Zernike images. Interestingly, as we relaxed the target resolution for 3D reconstruction of RNA polymerase II to ~20 �, we found that the running averages of 3 movie frames would still give enough signals for accurately determining the orientation for particles as small as RNA polymerase II for the signals limited by small number of electrons could be compensated by the high contrast. Remarkably, such the first 3 movie-frame 20 � RNA polymerase II differs from the one-second accumulated RNA polymerase II in that many surface motifs become recognizable. Based on this Microsc. Microanal. 21 (Suppl 3), 2015 2146 finding, we went on to fraction the one second into various time points to ask whether or not the 3D reconstruction from the same set of particles would result in the same conformation at different time. To our surprise, significant rearrangement of the RNA polymerase II structure along the time trajectory was observed where most of the notable motions are consistent with those based on the X-ray crystal studies, including the motions of the clamp and the stalk. [1] Z Wang et al, Nat Commun. 5, p. 4808. [2] M Liao et al, Nature 504, p. 107�112. [3] CH Wang and WH Chang, in preparation. [4] R Danev and K Nagayama, Ultramicroscopy 88, p. 243-52. [5] W Dai et al, Nature 502, p. 707�710.. [6] WH Chang et al, Structure 18, p. 17�27. [7] RJ Hall et al, J Struct Biol. 174, p. 468-75. [8] YM Wu et al, J Physics D: Appl. Phys. 46, p. 494008 [9] A Brown et al, Science 346, p. 718-722 [10] The authors acknowledge funding from the Academia Sinica and NSF in Taiwan and NINS in Japan.");sQ1[1072]=new Array("../7337/2147.pdf","Off-Axis Electron Holography for the Quantitative Study of Magnetic Properties of Nanostructures: From the Single Nanomagnet to the Complex Device","","2147 doi:10.1017/S1431927615011514 Paper No. 1072 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Off-Axis Electron Holography for the Quantitative Study of Magnetic Properties of Nanostructures: From the Single Nanomagnet to the Complex Device C. Gatel1, A. Masseboeuf1, E. Snoeck1, F. Bonnilla2, T. Blon2, L.-M. Lacroix2, A. Meffre2, J.F. Einsle3, R.M. Bowman3, M.A. Bashir4 and M. Gubbins4 2. CEMES-CNRS / University of Toulouse, Toulouse, France Laboratoire de Physique et Chimie des Nano-objets (LPCNO), INSA, Toulouse, France 3. Centre for Nanostructured Media, School of Mathematics and Physics, Queen's University Belfast, UK 4. Seagate Technology, 1 Disc Drive, Springtown, UK 1. Electron holography (EH) is a powerful interferometric TEM method particularly efficient for the quantitative studies of local electrostatic and magnetic fields at the nanoscale over a field of view as large as few microns. We applied this method to study the magnetic properties of two very different systems: single 20 nm Fe nanocubes and a hard-disk drive writer in operando. Nanomagnets recently attracted considerable interest due to their possible application as building blocks for hard drive disks and permanent magnets or as nanobiological vectors for drug delivery and hyperthermia. Despite theoretical studies, the size-dependence of spin arrangements in single nanomagnets has not yet been evidenced experimentally due to sensitivity limitations of the investigation tools. The single domain limit, corresponding to the critical nanomagnet size separating vortex/single domain configurations, has never been observed although it will dictate the optimized size for applications. In such small nano-objects, micromagnetic simulations show that the magnetic internal structure changes from single domain (SD) to vortex states as the cube size increases (Fig. 1a). Some years ago, we reported symmetrical vortices, i.e. vortex of <001> axis, in isolated 30 nm single crystal Fe cubes with a 14 nm vortex core size [2]. Next, we showed that vortices can also be stabilized in the presence of dipolar interactions thanks to holes in the cubes inducing a pinning of the vortex core [3]. Here we will focus on smaller Fe cubes thanks to a new Hitachi TEM dedicated to EH with an unprecedented spatial resolution of 0.5 nm. Micromagnetic simulations were performed using the 3D version of the micromagnetic code OOMMF in order to determine the equilibrium magnetic configuration in the cubes with Fe bulk magnetic parameters. We measured single domain (SD) states (flower state) in 20 nm isolated cubes with a quantitative accordance of the inner magnetic structure between experience and simulations (Fig. 1b, c, d). It means that we experimentally demonstrate the transition between the single domain and the vortex magnetic states. It is in accordance with theoretical predictions of a transition around 23 nm for perfect Fe cubes [2,4]. Moreover, micromagnetic simulations show that, for cube sizes closed to the transition between SD and <001> vortex states, vortices of <001> and <111> axes are very close in energy (Fig 1a). We will show that some experimental holograms consists in <111> vortex states. In this presentation, we will present an overview of the different magnetic configurations that can be observed and predicted in Fe ferromagnetic nanocubes of variable sizes. The proliferation in society of mobile devices accessing data via the `cloud' is imposing a dramatic increase in the amount of information stored on hard disk drives (HDD) used by servers. To meet this request the HDD industry needs to achieve 2 Tb/in2 densities. This means significantly increased performance from the magnetic pole of the electromagnetic writer in the read/write head of the HDD. Current state-of-art writing implies complex magnetic pole of sub 100 nm dimensions, in an engineered Microsc. Microanal. 21 (Suppl 3), 2015 2148 magnetic shields environment, which needs to deliver strong directional magnetic field onto the smallest possible area of the recording media. Current understanding about the behavior, shape and strength of the magnetic field generated by this nanoscale electromagnet arises from indirect methods. Within a collaboration with Seagate Technology Company, a new set-up combining EH with in-situ electrical biasing was developed for the first time to quantitatively map the magnetic field generated by the writer as a function of applied electrical current up to +/-60 mA in real conditions [5]. A specific process of sample preparation was required to extract the active part of the device directly from the production line, without damaging the magnetic shields and the necessary elements for the current injection. The magnetic induction maps obtained by EH for various applied currents have been successfully simulated using micromagnetic simulation using the geometrical parameters of the writer. Our results allow separating the write pole and the shield contributions. The resulting quantified field maps demonstrate the key features requested in magnetic recording, namely a highly directional magnetic field with a low angular spread. References: [1] L.-M. Lacroix et al, J. Am. Chem. Soc. 131 (2009), p. 549 [2] E. Snoeck et al, Nano Lett 8 (2008), p. 4293 [3] L.-M. Lacroix et al, Nano Lett. 12 (2012), p. 3245 [4] W. Rave et al, J. Mag. Mag. Mat. 190 (1998), p. 332 [5] J.F. Einsle, et al, NanoResearch, in press (DOI 10.1007/s12274-014-0610-0) [6] The authors acknowledge the European Union under the Seventh Framework Programme under a contract for an Integrated Infrastructure Initiative Reference 312483-ESTEEM2 and the French National Research Agency under the "Investissement d'Avenir" program reference No. ANR-10-EQPX-38-01". This work is supported by the French national project EMMA (ANR12 BS10 013 01). Figure 1. (a) Energies of different magnetic configurations as a function of the Fe nanocube size (OOMMF calculations). (b) Experimental hologram of a 20 nm Fe cube. (c) Deduced induction map. (d) Calculated induction map (OOMMF calculations).");sQ1[1073]=new Array("../7337/2149.pdf","Three-Dimensional Magnetic Vortex Cores Visualized by Electron Holographic Vector Field Tomography","","2149 doi:10.1017/S1431927615011526 Paper No. 1073 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Three-Dimensional Magnetic Vortex Cores Visualized by Electron Holographic Vector Field Tomography Toshiaki Tanigaki1,2, Yoshio Takahashi1, Tomokazu Shimakura1, Tetsuya Akashi1, Ruriko Tsuneta1, Akira Sugawara1, and Daisuke Shindo2,3 1. 2. Central Research Laboratory, Hitachi, Ltd., Hatoyama 350-0395, Japan RIKEN Center for Emergent Matter Science (CEMS), Wako 351-0198, Japan 3. Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, Sendai 980-8577, Japan Spintronics, which uses electron spins, is expected to become a widely used device technology because it has advantages in terms of nonvolatility, data processing speed, electric power consumption, and integration densities compared with conventional semiconductor technologies. Since the functions of spintronics devices are controlled by changing the spin configuration, i.e., by changing the magnetic field vector distribution, their direct observation is important for understanding the mechanisms of spintronics devices. Vector field electron tomography (VFET) using a transmission electron microscope (TEM) is a powerful technique for visualizing three-dimensional (3D) magnetic vector distributions on the nanometer scale [1,2]. Here we report 3D magnetic vortices in stacked ferromagnetic discs in a nanoscale pillar using a 1 MV electron holography microscope and a dual-axis 360� rotation sample holder for the VFET [3]. The 1 MV holography electron microscope with high penetration power, which has been used to observe vortex behavior in high-Tc superconductors [4], is a promising instrument for observing practical spintronics devices because it does not require changing their inherent magnetic structures by sample thinning. The dual-axis 360� rotation sample holder can eliminate artifacts due to missing wedges while performing the VFET [5]. These electron microscopic technologies have been used to unveil 3D magnetic vortices in stacked ferromagnetic discs in a nanoscale pillar (Figure 1). Figure 2 shows 3D view of the reconstructed magnetic vortex cores. The obtained 3D magnetic vectors in the stacked magnetic discs clearly show counter clockwise (CCW) magnetic flux flows in both the upper and lower discs. The z direction of the magnetic vectors at the vortex core of the upper disc was up while that of the lower disc was down. These opposite z directions are indicated by coloring in blue and red. The tail-to-tail vortex cores were mutually repulsive and thus stabilized at a position offset from the structural center of the discs. Comparison of the observed 3D magnetic field vector distributions in the magnetic vortex cores with the results of micromagnetic simulations based on the Landau-LifshitzGilbert equation showed that tail-to-tail vortex configurations are one of the stable magnetization states. In conclusion, 3D magnetic field vector distributions in the stacked ferromagnetic discs were visualized using the electron holographic VFET performed using the 1 MV holography electron microscope with high penetration power and the dual-axis 360� rotation sample holder that prevented missing wedges. The obtained results demonstrate that the proposed electron holographic VFET is a promising technique for direct 3D visualization of the spin configurations in magnetic materials and spintronics devices. Microsc. Microanal. 21 (Suppl 3), 2015 2150 References: [1] G. Lai et al., J. Appl. Phys. 75 (1994) p.4593. [2] C. Phatak et al., Ultramicroscopy 108 (2008) p.503. [3] T. Tanigaki et al., Nano Lett. (2015) DOI: 10.1021/nl504473a. [4] A. Tonomura et al., Nature 412 (2001) p.620. [5] R. Tsuneta et al., Microscopy 63 (2014) p.469. [6] This research was partly supported by a grant from the Japan Society for the Promotion of Science (JSPS) through the "Funding Program for World-Leading Innovative R&D on Science and Technology (FIRST Program)" initiated by the Council for Science and Technology Policy (CSTP). Figure 1. Schematic diagrams of stacked ferromagnetic discs and axes used for 360� observations. Two Fe discs (light blue) with thicknesses of 20 nm (upper) and 30 nm (lower) were separated by a 10 nm Cr disk (green). The direction of the magnetic vortices in the disc is either clockwise (CW) or counter clockwise (CCW). At the magnetic vortex core, the direction of the magnetic vectors is either up or down. Figure 2. Three-dimensional (3D) view of the reconstructed magnetic vortex cores. The z-directional components are indicated by blue (+z) or red (-z). (a) The cross-sectional magnetic vectors. The magnetic vectors for the upper and lower vortex cores had opposite z directions. (b) 3D view of the tailto-tail vortex cores.");sQ1[1074]=new Array("../7337/2151.pdf","Towards Multiresolution Phase Retrieval using Electron Ptychography","","2151 doi:10.1017/S1431927615011538 Paper No. 1074 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Towards Multiresolution Phase Retrieval using Electron Ptychography Charudatta Phatak1 , Youssef Nashed2 , Tom Peterka2 1 2 Materials Science Division, Argonne National Laboratory, Lemont, IL 60439, USA Mathematics and Computer Science Division, Argonne National Laboratory, Lemont, IL 60439, USA In conventional transmission electron microscopy imaging, due to the limitations of electromagnetic lenses such as aberrations, and instabilities, the achievable spatial resolution is limited to significantly below the wavelenght of electrons. Furthermore, obtaining quantitative information at the highest spatial resolution is often times difficult. Using Ptychography, it has been shown that it is possible to eliminate the aberrations of electromagnetic lenses and obtain high resolution information beyond that limit of direct imaging [1,2]. The electron-optical phase shift of electrons passing through a thin foil sample in a TEM can be written as a combination of various components arising from the electrostatic potential (mean inner potential, local charge density variations), magnetic vector potential (local magnetization), crystalline lattice potential (atomic structure within the unit cell), and geometric phase (displacement field, strain). All these components of the phase shift typically consist of information at various spatial frequencies that need to be recovered. In this work, we use the extended ptychographical iterative engine (ePIE) to reconstruct the entire complex object wavefunction as well as the complex electron probe wavefunction [3]. The algorithm then iterates between reciprocal and real space and updates both the wavefunctions given the set of experimental constraints until convergence. By varying the mask used in reciprocal space during the reconstruction process, the effective resolution of the recovered phase shift can be controlled allowing a multiresolution recovery from the same data set. Simulations were performed to test the reconstruction procedure for electron ptychography data. The electron probe was simulated for a typical Cs corrected 200 kV JEOL 2100F FEG TEM. The diffraction images were then simulated using an input phase and amplitude of the object as a function of scanning the probe along a cartesian grid. Fig. 1(a) shows the simulated defocused electron probe, and (b) shows the corresponding diffraction pattern, and (c) shows the experimental defocused electron probe. The effects of lens aberrations such as astigmatism can be easily seen in the experimental image of the probe. Fig. 2(a) shows the input phase that was used for simulating the diffraction patterns, (b) shows a schematic of probe positions in a ptychographic scan overlaid on the simulated image, and (c) shows the preliminary result of the reconstructed phase shift. We will further discuss the effect of defocus of electron probe, overlap of probe positions during scan and unwrapping of recovered phase. Experimentally, we have implemented the ptychographic acquisition on a 200 kV JEOL 2100F equipped with a spherical aberration corrector and a dedicated Lorentz lens using a Digital Micrograph script. The spatial resolution of this microscope is limited to 0.5 nm in direct imaging mode. We will discuss the improvement of the spatial resolution particularly in the phase reconstruction performed using ptychography. References [1] M. J. Humphry, et. al., Nat. Commun., 3, 730 (2012). Microsc. Microanal. 21 (Suppl 3), 2015 2152 [2] A. J. D'Alfonso, et. al., Phys. Rev. B, 89, 064101 (2014). [3] Y. S. G. Nashed et. al., Optics Express, 22, 32082 (2014). [4] This work was supported by U.S. Department of Energy (DOE), Office of Science, Materials Sciences and Engineering Division. (a) (b) (c) 10 nm 10 nm Figure 1: (a) Simulated defocused electron probe for a defocus of f = 3 �m, (b) the corresponding simulated diffraction pattern, and (c) experimental defocused electron probe for a defocus of f = 2.3 �m. (a) (b) (c) j 50 nm i Figure 2: (a) Simulated input phase shift for 4 square magnetic islands showing the closure domain configuration, (b) schematic showing the probe positions during a ptychographic scan overlaid on the infocus image, and (c) the reconstructed phase shift (unwrapped) using ePIE algorithm.");sQ1[1075]=new Array("../7337/2153.pdf","Skyrmion Lattices Observed by High-Voltage Holography Electron Microscopes","","2153 doi:10.1017/S143192761501154X Paper No. 1075 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Skyrmion Lattices Observed by High-Voltage Holography Electron Microscopes Hyun Soon Park Department of Materials Science and Engineering, Dong-A University, Busan, Republic of Korea. Electron holography and Lorentz microscopy, using coherent electron waves emitted from cold fieldemission sources, provides opportunities for directly detecting and visualizing, in real space, the phase shifts of the electron waves due to the electromagnetic fields [1]. In 1992, vortices in a Nb superconducting thin film were observed by using a "coherent electron wave" Lorentz microscopy, simultaneously revealing their static lattice patterns and their dynamics in real time [2]. In 2009, neutron-scattering studies have shown the existence of skyrmion crystals in bulk helimagnet MnSi, by which 6-fold intensity patterns with a short lattice parameter (~17.5 nm) were detected [3]. Two of the most important areas of study in research on the skyrmion lattice are its topological magnetic configuration and nucleation/annihilation process during magnetic phase transition. Recent Lorentz microscopy studies (2010) have revealed the magnetic configuration of skyrimon lattices and the magnetic phase diagrams for Fe0.5Co0.5 Si thin samples in which the skyrmion lattice constant is 90 nm [4]. In 2012, the skyrmion lattices with a short lattice parameter (18 nm) in MnSi thin samples were observed by in situ observation using Lorentz microscopy [5]. Real-space imaging of skyrmion lattices has been performed by using the Fresnel (out-of-focus) method of Lorentz microscopy. Figure 1 shows the changes in the magnetic structure as a function of the applied magnetic field at 10 K in MnSi thin sample. Stripe magnetic domains, which were imaged as bright and/or dark lines, were clearly visible at zero applied magnetic field (a). Change in stripe domain contrast with increase in applied field was shown in b and c. Magnetic structures showing 6-fold symmetry appeared for magnetic field of 0.18 T (d) and gradual disappearance of skyrmion lattice with further increase in magnetic field was observed. Even though we reported the first direct observation of a short-period skyrmion lattice (18 nm) in a MnSi system [5], quantitative analysis of the magnetic flux flow inside and outside a skyrmion was quite difficult because of both the resolution limit (due to the defocused condition) and the unwanted artefacts (surface roughness or contamination) of thin samples. Precise phase measurement of weak phase objects such as skyrmions is very challenging because procedures are needed for averaging the phase images and separating the electric and magnetic vector potentials. Nevertheless, the advantage of electron holography compared to Lorentz microscopy, under just-focused condition, makes it possible to visualize a quantized magnetic flux with nanometer resolution, in addition to determining its density in the vicinity of skyrmions. Here we investigated the 2D magnetic flux distributions (Fig. 3a) of skyrmion lattices in helimagnet Fe0.5Co0.5Si thin samples with a stepped thickness as shown in Fig. 2 and estimated the 3D structures (Fig. 3b) of the skyrmion phases by using 300 kV and 1MV field-emission electron microscopes [6]. Our study demonstrates the potential of using a high-voltage holography electron microscope in various applications, such as the 3D visualization of magnetic fields in emergent matter systems and spintronics. The author thanks the late Dr. A. Tonomura for his valuable discussions and all the members of the FIRST Tonomura project. This research was supported by a grant from the JSPS through the `Funding Program for World-Leading Innovative R&D on Science and Technology', under the programs `Development and Application of an Atomic-resolution Holography Electron Microscope' and `Quantum Science on Strong Correlation'. References: Microsc. Microanal. 21 (Suppl 3), 2015 2154 [1] A. Tonomura, Rev. Mod. Phys. 59, 639 (1987). [3] S. Muhlbauer et al., Science 323, 915 (2009). [5] A. Tonomura et al., Nano Lett. 12, 1673 (2012). (2014). [2] K. Harada et al., Nature 360, 51 (1992). [4] X. Yu et al., Nature 465, 901 (2010) [6] H. S. Park et al., Nature Nanotech. 9, 337 Fig. 1. Magnetic-field dependence of magnetic structure in MnSi sample (Reprinted from Ref. 5) Fig. 2. Lorentz micrographs of Fe0.5Co0.5Si sample with a stepped thickness (Reprinted from Ref. 6) Fig. 3. 2D phase maps and 3D structure of skyrmion lattices in Fe0.5Co0.5Si thin sample. (a) Magnetic flux maps showing the phase shift due to magnetic vector potential. The difference between equi-phase lines corresponds to 0.1 rad. (b) Electron phase shifts of the helix and skyrmion as a function of sample thickness (carried out using 300 and 1,000 kV microscopes). Schematic illustration showing the 3D spin structure of a skyrmion and its phase shift profile for line A�B (in the x�z plane), integrated through specimen thickness t (Reprinted from Ref. 6).");sQ1[1076]=new Array("../7337/2155.pdf","Roles of self-assembly and beam damage in gas-assisted electron and ion beam induced processing","","2155 doi:10.1017/S1431927615011551 Paper No. 1076 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Roles of self-assembly and beam damage in gas-assisted electron and ion beam induced processing. Milos Toth1 1. School of Physics and Advanced Materials, University of Technology, Sydney, 15 Broadway, Ultimo, New South Wales 2007, Australia Gas-assisted focused ion beam (FIB) processing enables high resolution, direct-write deposition and milling of a wide range of materials. Here I will review recent advances in FIB processing with an emphasis on: � A new form of self-assembly in gas-assisted Ga+ FIB-induced processing [1], and � the use of oxygen ions for FIB milling of single crystal diamond [2]. Self-assembly can be used to overcome some of the limitations inherent to conventional, serial FIB methods in which a beam is scanned in a pattern that defines the nanostructure geometry. The limitations include throughput and the inability to fabricate complex 3D structures such as the Ga-filled, tapered GaF3 microcapillary shown in Figure 1. The structure was fabricated by irradiating GaN by a Ga+ FIB that was scanned repeatedly over the rectangle seen in the figure, in the presence of a XeF2 precursor gas. Pillar growth proceeds through a self-ordering mechanism that emerges from an interplay between a number of physical and chemical mechanisms that include the generation of excess, mobile Ga atoms at the substrate surface, dissociation of XeF2 adsorbates by secondary electrons that impact the pillar sidewalls, sputtering, and self-masking caused by the tapered pillar geometry. The process can be used to grow multiple ordered or disordered pillars in parallel, and is fundamentally different from convention FIB-induced deposition and milling mechanisms [1]. A second significant limitation of conventional Ga+ FIB processing techniques is damage and staining caused by the Ga+ beam. This problem can, in some cases, be alleviated by the use of ions other than Ga+. For example, milling of single crystal diamond by an oxygen FIB yields a damage layer and (implanted oxygen) impurities which can be removed by post-processing methods that are less intrusive than those used to process diamond milled by Ga+ ions [2]. The benefits and limitations of oxygen FIB milling of diamond will be discussed and compared to related electron beam processing techniques that enable minimally-invasive nanofabrication in the limit of negligible damage caused by momentum transfer and in the absence of staining caused by implantation. Specific examples of electron beam diamond processing techniques include damage-free etching using H2O as a precursor gas (Figure 2) [3], and chemical functionalization using NF3 as a precursor gas (Figure 3) [4]. References: [1] Botman, A., Bahm, A., Randolph, S., Straw, M. & Toth, M. Spontaneous Growth of Gallium-Filled Mi-crocapillaries on Ion-Bombarded GaN. Phys. Rev. Lett. 111, 135503 (2013). [2] A. A. Martin, S. Randolph, A. Botman, M. Toth & I. Aharonovich. Direct-write milling of diamond by a focused oxygen ion beam. Sci. Rep. (in press). [3] Martin, A. A., Toth, M. & Aharonovich, I. Subtractive 3D Printing of Optically Active Diamond Structures. Sci. Rep. 4, 5022 (2014). [4] Shanley, T., Martin, A. A., Aharonovich, I. & Toth, M. Localized chemical switching of the charge state of nitrogen-vacancy luminescence centers in diamond. Appl. Phys. Lett. 105, 063103 (2014). Microsc. Microanal. 21 (Suppl 3), 2015 2156 Figure 1. Schematic illustration (left) and a false-color SEM image (right) of a pillar grown by Ga+ ion beam bombardment of GaN in an XeF2 environment. The pillar is comprised of a solid, tapered, hollow GaF3 sheath that is filled with liquid Ga [1]. e Pillar PL intensity (a.u.) Virgin diamond 2"m" 620 640 660 680 700 Wavelength (nm) 720 740 Figure 2. The letters "NANO" etched into a microdiamond using H2O as the etch precursor gas (left) and photoluminescence (PL) spectra showing PL enhancement by a diamond waveguide fabricated by electron beam induced etching [3]. vacuum$ NF3$ vacuum$ H2terminated$ diamond$ fluorinate$ Figure 3. Schematic illustration of a direct-write process used to control the surface electronic structure and fluorescence properties of nanodiamonds by electron beam processing in a gaseous NF3 environment [4].");sQ1[1077]=new Array("../7337/2157.pdf","Site-Specific TEM Specimen Preparation of Samples with Sub-Surface Features","","2157 doi:10.1017/S1431927615011563 Paper No. 1077 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Site-Specific TEM Specimen Preparation of Samples with Sub-Surface Features Julia I. Deitz,1 Santino D. Carnevale,2 David W. McComb,1,3, Steven A. Ringel,2,3, Tyler J. Grassman1,2 1 2 Dept. of Materials Science & Engineering, The Ohio State University, Columbus, OH, 43210, USA Dept. of Electrical & Computer Engineering, The Ohio State University, Columbus, OH, 43210, USA 3 Institute for Materials Research, The Ohio State University, Columbus, OH, 43210, USA Characterization of crystalline defects has traditionally been performed in the TEM, as it has the required capabilities in terms of sensitivity and resolution to ensure sufficient accuracy. However, TEM analysis requires electron transparency, which requires preparation of thin specimens, typically achieved using focus ion beam (FIB)-based foil extraction and thinning. For specimens with features that cannot be seen on the surface, samples extracted via FIB are effectively selected "blindly." That is, the area from where the sample is extracted is, for the most part, randomly selected, and there is little or no prior knowledge as to what features will be included in the small, and extremely thin, region (approximately 10 �m � 5 �m � 100 nm). Such a small sample can provide a poor statistical representation of the material as a whole, and it does not allow for a desired feature to be obtained in an efficient manner. Here, we develop a method to locate sub-surface features using electron channeling contrast imaging (ECCI) in order to extract site-specific TEM specimens using the dual beam focused ion beam (FIB) instrument. ECCI, an alternative method for characterization to TEM, has begun to gain traction as a technique for the visualization of subsurface defects in crystalline materials, due in part to work at OSU on its application to epitaxial III-V compound semiconductor materials [1][2]. This powerful technique is performed using a field-emission scanning electron microscope (SEM) equipped with either a backscatter or forescatter electron (BSE, FSE) detector. Recently, we have demonstrated ECCI performed using the secondary electron (SE) detector as well. The ability to detect and image subsurface features with a standard SE detector means that is it now possible to alleviate the problem of unguided sample extraction with the FIB. It is standard for any FIB to have an SE detector, but not necessarily a BSE detector, meaning that ECCI can be performed on a sample in the FIB, and defects and features can be detected prior to extraction. Once a desired area is found, it can be marked with platinum and subsequently prepared for TEM characterization. This proposed method effectively eliminates the risk of "guess and check" sample preparation. Instead of merely hoping that the subject of interest will happen to have been captured via the "blind" FIB approach, where a TEM foil is taken from within the sample without any real guidance, we can pick the exact area to extract based upon the fact that we already know the subject of interest is actually there. This contribution serves to develop this ECCI-FIB technique for the extraction of site-specific TEM specimens with a FIB. All ECCI in this work was performed on as-grown samples, using either an FEI Helios or Nova NanoLab dual-beam instrument fitted with a secondary electron (SE) detector, using an accelerating voltage of 30 kV, a spot size of 5 (2.4nA), and a working distance of 15 mm. ECCI-FIB is first demonstrated on a GaP (50 nm layer) on Si sample occupied with a high density of misfit dislocations. Figure 1 shows three separate images taken with the SE detector in the FIB over the same area. Figure 1(a) is taken at standard surface-normal, surface-sensitive non-diffractive SE geometry, where sub-surface features are not expected to be seen. Figure 1(b) has a diffraction vector of g = 220, while 1(c) has the inverse diffraction condition of g = 220. It follows that 1(c) misfit Microsc. Microanal. 21 (Suppl 3), 2015 2158 dislocations should show inverse contrast to 1(b), and indeed this is confirmed, showing the SE diffraction follows the same mechanisms as BSE and TEM diffraction. Misfits that appear black in 1(b) appear white in 1(c) and vise versa. We will also present SE FIB work for more complex systems and different types of defects. When used in practice, after locating a feature of defect of interest with ECCI, a small amount of platinum can be deposited � to mark the location � with the e-beam, followed by standard ion beam platinum deposition. Subsequently, site specific FIB extraction can occur. In this presentation, we will discuss further details of this method to ensure accurate extraction with minimized beam damage. Conclusion: We demonstrate the use of ECCI for site specific extraction in the FIB for sub surface features. The project presented here serves to develop this method, ECCI-FIB, into an extensible tool for use in a wide range of research projects where sub-surface features need to be captured within the prepared specimens for TEM analysis. This application of ECCI removes risk of blind FIB work, giving guidance to the FIB user for accelerated research. References: [1] S. Carnevale, J. Deitz, T. Grassman, J. Carlin, Y. Picard, M. De Graef, S. Ringel, "Rapid misfit dislocation characterization in heteroepitaxial III-V/Si thin films by electron channeling contrast imaging," Applied Physics Letters 104(23), 232111 (2014). [2] S. Carnevale, J. Deitz, T. Grassman, J. Carlin, Y. Picard, M. De Graef, S. Ringel, "Applications of Electron Channeling Contrast Imaging for the Rapid Characterization of Extended Defects in III-V/Si Heterostructures," IEEE Journal of Photovoltaics, accepted (2014); Early Access: http://dx.doi.org/10.1109/JPHOTOV.2014.2379111. Figure 1: SE images of GaP/Si sample in the FIB, over the same area. 1(a), the SE is at standard, non diffracting geometry, while (b) has a diffraction condition of g = (220) and (c) with g = (220).");sQ1[1078]=new Array("../7337/2159.pdf","Micro-micromechanical Properties of Weld Metal","","2159 doi:10.1017/S1431927615011575 Paper No. 1078 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Micro-micromechanical Properties of Weld Metal Walter L. Costin1, A. K. Basak2 1 2 School of Mechanical Engineering, The University of Adelaide, South Australia, 5005 Adelaide Microscopy, The University of Adelaide, Adelaide, South Australia, 5005 Hydrogen Assisted Cold Cracking (HACC), also referred to as delayed cracking or cold cracking, is a weld defect which may form in the heat affected zone (HAZ) of the parent metal or the weld metal (WM). Failure usually occurs after the deposited weld has cooled down to temperatures below 200 �C and can be initiated within minutes to even days after welding [1] and attributed to micro-mechanical properties of the representing microstructures [2,3] in weld zone. Micro-fracture experiments, to determine the properties in specific microstructural regions, were studied in this study. Microscopic cantilever shaped test specimens were fabricated with Focused Ion Beam (FIB) techniques in localized region consisting of predominantly acicular ferrite microstructure followed by loading with nanoindentation for fracture tests. Three micro-cantilever shaped test specimens with pentagonal cross sections were fabricated in selected region as shown in Fig. 1. The methodology for the preparation was based on the technique developed by Davidson et al [4]. A narrow notch was then milled about 7 �m from the support of the beam to create an artificial stress concentrator with defined geometries as shown details in Fig. 2. The Scanning Electron Microscope (SEM) image (Fig. 3) of the tested specimen shows that the failure process was associated with the development of a plastic hinge with the rotation centre shifted to the beam support. This shift of the rotation center from the notch tip location towards the support may indicate a significant effect of the support on fracture mechanism due to insufficient depth of the notch. Relatively large plastic zone compared to the small sample size implies that linear elastic fracture mechanics (LEFM) is not applicable. However, recent investigations have used methods based on elasto - plastic fracture mechanics (EPFM) to obtain valid fracture toughness values for semi brittle materials [5]. The crack tip opening displacement (CTOD) was used as a parameter to describe the static fracture resistance of the tested material. The measurement of the CTOD was based on an analytical hinge model, which has been adopted for the notched micro-cantilever geometries. The results showed good correlation with the properties measured in macroscopic samples. Although the material tested in this study was not polycrystalline and less ductile, similar methods may be applicable for fracture toughness measurements in selected ferrite morphologies. References: [1] Barbaro F. J., 1999, "Types of Hydrogen Cracking in Pipeline Girth Welds," WTIA/APIA/CRC-WS International Conference Weld Metal Hydrogen Cracking in Pipeline Girth Welds Wollongong, NSW, Austalia. [2] Kuzmikova L., Barbaro F. J., Norrish J. & Li H., 2012, "Weld Metal Hydrogen Assisted Cold Cracking in High Strength Low Alloy Steels," SEAISI Conference & Exhibition, Bali, Indonesia. [3] Wongpanya P., Boellinghaus Th., Lothongkum G. and Hoffmeister H., 2009, "Numerical modelling of cold cracking initiation and propagation in S 1100 QL steel root welds," Welding in the World, 53(3). [4] Davidson J. L., and Olson D. L. R., 1997, Hydrogen management in steel weldments: joint seminar, Melbourne, Australia, 23rd October 1996: proceedings of the seminar, Published by Microsc. Microanal. 21 (Suppl 3), 2015 2160 the Organising Committee of the Joint Seminar on behalf of Defence Science and Technology Association and Welding Technology Institute of Australia. [5] Wurster S., Motz C. and Pippan R., 2012, "Characterization of the fracture toughness of micro-sized tungsten single crystal notched specimens," Philosophical Magazine, 92(14), pp. 1803-1825. Figure 1. Optical micrograph of selected crack (left) and FIB milled specimens for micro fracture experiments in acicular ferrite region (middle and right). Figure 2. Final specimen dimensions of FIB milled micro-cantilever. Figure 3. SEM of the test specimen before and after testing.");sQ1[1079]=new Array("../7337/2161.pdf","Three Dimensional Microstructural Characterization of Cathode Degradation in SOFCs Using FIB/SEM and TEM","","2161 doi:10.1017/S1431927615011587 Paper No. 1079 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Three Dimensional Microstructural Characterization of Cathode Degradation in SOFCs Using FIB/SEM and TEM Joshua Taillon1, Christopher Pellegrinelli1, Yilin Huang1, Eric Wachsman1, and Lourdes Salamanca-Riba1 1. University of Maryland, Materials Science and Engineering, College Park, MD, USA Solid oxide fuel cells (SOFC) present an efficient, clean, and flexible means of energy conversion, but the limited durability of the cells in practical applications has impeded their commercial adoption. Degradation occurs within the cathode upon long-term operation and exposure to various environmental contaminants, including H2O. The contaminants cause significant microstructural and compositional changes within the cathode that adversely affect activation, polarization mechanisms, and ionic and electronic conductivities. Previous works have demonstrated that a number of quantifiable microstructural characteristics can be directly related to SOFC performance, the most important of these being triple phase boundary length (LTPB) and pore surface area [1-2]. These parameters have not been examined during cell degradation, and further analysis under these conditions provides insight into specific cell degradation mechanisms, informing future fabrication and operation criteria. Three-dimensional quantifications of porous SOFC cathodes have been obtained after aging in both clean and contaminated atmospheres, under cathodic, anodic, and no polarization. The cathodes consisted of 50 wt% La1-xSrxMnO3 (LSM) and 50 wt% (Y2O3)0.08-(ZrO2)0.92 (8-YSZ), screen printed onto a YSZ electrolyte to form LSM-YSZ /YSZ/LSM-YSZ symmetric cells. The structure of these cathodes was quantitatively analyzed using a serial nanotomography technique within the FIB/SEM (Figure 1). The reconstructed volumes were quantified by a number of parameters, including LTPB, particle/pore size and distribution, surface area coverage, porosity, volume fraction, and tortuosity. Information relating to phase connectivity was compared for each cell through a skeletonization process. These parameters were analyzed in relation to electrochemical cell performance results, and correlated with composition changes observed via TEM/EELS. Figure 1 shows four reconstructed LSM/YSZ cathodes for different aging/polarization conditions. The average reconstructed volume was approximately 4000 m3. After reconstruction, the ratios of segmented LSM and YSZ matched those expected from the starting source material, indicating no significant redistribution of phases occurred during aging. Analysis of the phase surface area and volume revealed that the LSM particles were significantly larger and had much lower specific surface areas (per volume) than the other phases. There did not appear to be a significant effect on these parameters from aging in either environment. Likewise, the phase distribution was anisotropic with respect to the electrolyte/cathode interface, meaning there was no significant deviation from the average, regardless of location within the cathode. The connectivity of the phases was analyzed by computing a skeletonized representation of the segmented image data. In all LSM/YSZ cathodes, the YSZ and pore networks were found to be completely percolated, while the LSM phase was greatly fragmented, indicating that electronic conduction out of the cathode is limited in these cells. The tortuosity () of each phase was also calculated in each direction by tracking the phases' two-dimensional center of mass (slice-by-slice). The ratio of phase fraction (Vp) to tortuosity is related to the effective diffusivity (Deff) in a material (and thus Microsc. Microanal. 21 (Suppl 3), 2015 2162 r related to the conductivi [3]. For the YSZ ph e ity) r hase, this fr action was f found to be smallest in the cells a aged under H2O contam mination, sug ggesting that the presenc of water decreases Deff in these cells. No t ce s significant di ifference wa observed for LSM and the pore ph as f d hases. T final microstructural property an The nalyzed was a calculatio of the trip phase bou on ple work. The undary netw p points where three phas join repr e ses resent possib sites for the oxygen reduction reaction (OR ble r n RR), and p provide a pa athway for th reaction products to be conducted The conne he p b d. ectivity of th TPB netw he work was a analyzed to predict whe ether a porti of the network is ex ion xpected to b active by determinin if it is be y ng c connected across the volume of th cathode that has be sampled Isolated T he een d. TPB segme ents were p predicted to be inactive. The results of this analysis are p plotted in Fig gure 2a-b; t results s the show that w while both samples age in dry air had simila active TP fractions, the H2O-C s ed r ar PB , Cathodic sam mple was s significantly lower, pote entially imp pacting perfo ormance. Fi inally, detailed interfaci analysis of these ial c cathodes was performed with TEM-E s EELS, with which signi ificant cation (Mn) segre n egation was observed ( (Figure 2c-d) These vari ). ious parameters were the related to observed el en o lectrochemic performa cal ance. [4] F Figure 1: Vi isualizations of the four reconstructe LSM/YSZ composite cathodes. T s ed Z The a aging/polariz zation condit tions for eac are (a) Air (b) Air-Ca ch r, athodic, (c) H2O-Anodic, and (d) H2O OC Cathodic. Ph hases are dist tinguished by color: LSM in red (da b M ark), YSZ in yellow (ligh and pore phase ht), e e excluded for clarity. A portion of the bulk YSZ electrolyte is visible in th rear of ea reconstru r e e s he ach uction. F Figure 2: (a) Total, activ and inact TPB den ) ve, tive nsities for ea reconstru ach ucted LSM/Y sample (b) YSZ e. S Same data as (a), compar s ring active vs. inactive fractions of t TPB netw v f the work. (c) HA AADF-STEM signal M a EELS in and ntensity maps of the back s kground (d) and Mn-L2,3 (e) edge bet tween YSZ grains after NMF 3 d decompositio on. [ D Gostov et al, Jou [1] vic urnal of the American Ce A eramic Socie 94 (2011) p. 620. ety [ JR Smith et al, Solid State Ionics 180 (2009) p. 90. [2] h s [ CJ Gomm et al, AI [3] mes IChE Journa 55 (2009) p. 2000. al p [ The autho gratefully acknowled funding from the U. S. DOE, SECA contract DEF SEE0009084. [4] ors dge t J JAT addition nally acknow wledges fund ding through the NSF GR h RFP, grant D DGE 132210 06.");sQ1[1080]=new Array("../7337/2163.pdf","Ultrafast Infrared Nanoscopy with Sub-Cycle Temporal Resolution","","2163 doi:10.1017/S1431927615011599 Paper No. 1080 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Ultrafast Infrared Nanoscopy with Sub-Cycle Temporal Resolution Tyler L. Cocker1, Max Eisele1, Markus A. Huber1, Markus Plankl1, Leonardo Viti2, Daniele Ercolani2, Lucia Sorba2, Miriam S. Vitiello2 and Rupert Huber1 1. 2. Department of Physics, University of Regensburg, 93040 Regensburg, Germany NEST, CNR � Istituto Nanoscienze and Scuola Normale Superiore, 56127 Pisa, Italy Ultrafast microscopy of surfaces with simultaneous nanometer spatial resolution and femtosecond temporal resolution can be achieved by coupling ultrafast optical pulses to sharp metal tips [1-10]. In the case of scattering-type near-field scanning optical microscopy (s-NSOM), evanescent fields scattered from the tip apex provide a window into a drastically subwavelength world. Moreover, since these scattered fields can be measured using established far-field technologies, ultrafast infrared pulses emerging from a nanoscale volume can be detected with the best possible time resolution: faster than a single optical oscillation cycle [1]. Sub-cycle time resolution in the far- and mid-infrared can be achieved by ultrabroadband field-resolved terahertz spectroscopy [11,12]. This spectral range contains a host of important collective excitations in condensed matter systems, including plasmons, phonons and magnons. Both the dynamics and character of these low-energy excitations can change significantly as the dimensions of a system are reduced to the nanoscale. Their properties can also vary between different nanoparticles in an ensemble, as nanoparticle size, structure, and orientation is often heterogeneous. Direct, ultrafast measurements within single nanoparticles are thus of great interest. Here, we demonstrate a unique combination of ultrafast multi-terahertz (mid-infrared) spectroscopy with s-NSOM. Phase-stable multi-terahertz pulses are coupled to the tip of an s-NSOM and the scattered radiation is detected by electro-optic sampling. We record the oscillating electric near field with a time resolution given by the duration of the electro-optic gate pulse (10 fs, sub-cycle). Meanwhile, the radius of curvature of the s-NSOM tip apex defines our nano-spectroscopy spatial resolution (10 nm). We have applied our novel microscope to the study of ultrafast local carrier dynamics in a single, isolated indium arsenide nanowire. Ultrafast imaging of the scattered near-field intensity (Fig. 1(a) and 1(b)) provides a map of the photoinduced carrier density, while field-resolved nano-spectroscopy (Fig. 1(c)) reveals the local evolution of the plasma frequency with sub-cycle time resolution. The plasma frequency manifests itself as a resonance in our probe spectrum, and can be used to calculate the instantaneous local carrier density at the surface of the nanowire. The carrier population dynamics feature a decay on the picosecond scale, which agrees well with typical carrier trapping times. Surprisingly, though, we also observe a very fast initial decay (<50 fs). To reveal its origin we have employed a novel technique we call femtosecond tomography. This approach allows us to trace the electron dynamics at different probing depths inside the material. The ultrafast decay is found to only occur close to the surface of the wire and can be explained by the ultrafast build-up of a surface depletion layer in the nanowire [13]. Microsc. Microanal. 21 (Suppl 3), 2015 2164 References: [1] M Eisele et al, Nature Photon. 8 (2014) 841. [2] S Berweger et al, Nano Lett. 11 (2011) 4309. [3] J M Atkin et al, Adv. Phys. 61 (2012) 745. [4] M Wagner et al, Nano Lett. 14 (2014) 894. [5] M Wagner et al, Nano Lett. 14 (2014) 4529. [6] Y Terada et al, Nature Photon. 4 (2010) 869. [7] S W Wu and W Ho, Phys. Rev. B 82 (2010) 085444. [8] A Dolocan et al, J. Phys. Chem. C 115 (2011) 10033. [9] T L Cocker et al, Nature Photon. 7 (2013) 620. [10] S Yoshida et al, Nature Nanotech. 9 (2014) 588. [11] R Ulbricht et al, Rev. Mod. Phys. 83 (2011) 543. [12] P U Jepsen et al, Laser Photon. Rev. 5 (2011) 124. [13] The authors thank M. Furthmeier for technical assistance. This work was supported by the European Research Council through ERC grant 305003 (QUANTUMsubCYCLE), the Deutsche Forschungsgemeinschaft through Graduate Research College GRK 1570, and the Italian Ministry of Education, University, and Research (MIUR) through the Futuro in Ricerca 2010 grant RBFR10LULP (Fundamental Research on Terahertz Photonic Devices). T.L.C. acknowledges the support of the Alexander von Humboldt Foundation. Figure 1. Pump-probe multi-THz spectroscopy and microscopy of a single InAs nanowire. (a) Topography of the InAs nanowire recorded by AFM. (b) Scattered intensity of the multi-THz probe pulse as a function of near-IR pump/multi-THz probe delay time and AFM tip position. (c) Scattered near-field waveforms recorded at the spectroscopy position denoted in (a). The oscillating electric near field from a (10 nm)3 volume is traced with 10 fs temporal resolution");sQ1[1081]=new Array("../7337/2165.pdf","Lensless Imaging of Nano- and Meso-Scale Dynamics with X-rays","","2165 doi:10.1017/S1431927615011605 Paper No. 1081 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Lensless Imaging of Nano- and Meso-Scale Dynamics with X-rays Jesse N. Clark1,2, Mariano Trigo3,Tom Henighan1, Ross Harder4, Brian Abbey5, Tetsuo Katayama6 , Mike Kozina1, Eric Dufresne4, Haidan Wen4, Donald Walko4, Yuelin Li4, Xiaojing Huang7 ,Ian Robinson8 and David Reis1 1. 2. Stanford PULSE Institute, SLAC National Accelerator Laboratory, Menlo Park, USA Center for Free-Electron Laser Science, Deutsches Elektronensynchrotron, Hamburg, Germany 3. SLAC, SLAC National Accelerator Laboratory, Menlo Park, USA 4 Advanced Photon Source, Argonne National Laboratory, Argonne, USA 5 Department of Physics, La Trobe University, Bundoora, Australia 6 JASRI, Hyogo, Japan 7 National Synchrotron Light Source II, Brookhaven National Laboratory, Upton, USA 8 London Centre for Nanotechnology, University College London, London, United Kingdom There is a fundamental interest in studying pico-second dynamics at the nano- and meso-scale in particles and structures as it can provide insight into their mechanical and thermal properties out of equilibrium and during phase transitions. Imaging over such short time-scales has been very challenging due to stringent temporal requirements. Here we report two types of lensless imaging experiments which exploit the coherence of modern x-ray sources to image dynamics at the nano- and meso-scale. Lensless imaging relies on iterative phase retrieval to replace the role of a lens in a traditional imaging system [1]. Some of the advantages of this include greater sensitivity and removing some challenges with manufacturing high quality lenses. Using a spatially and temporally coherent source (such as xrays or electrons), diffraction patterns are collected from finite sized or extended objects. Iterative phase retrieval is then used to obtain a real-space image of the object by iteratively enforcing consistency with the measured data and real-space constraints. Using the short pulses from an x-ray free-electron laser, we demonstrate three-dimensional imaging of the generation and subsequent evolution of coherent acoustic phonons on the picosecond time scale within single nanocrystals [2] using Bragg coherent diffraction imaging [3]. The method demonstrated here does not rely on any a priori shape information and does not involve any ensemble averaging. The implications of this are broad, as it allows determination of the vibrational, mechanical and elastic properties of individual nanocrystals which may be irregularly shaped or contain defects such as stacking faults or dislocations. Additional examples of imaging time-dependent strain in larger crystals using a synchrotron source will also be shown. In a separate experiment, we demonstrate that in the absence of any temporal resolution, a dynamic system can still be imaged [4]. We will show that dynamics an order of magnitude faster than the intrinsic temporal resolution of the experiment can be imaged. This is achieved by using recent algorithm advances [5] within the lensless imaging technique known as ptychography [6,7]. Potential applications of this are many and include the possibility of imaging fluid flow, magnetic systems or multi-state samples at the nanoscale. Microsc. Microanal. 21 (Suppl 3), 2015 2166 [1] J. Miao, P. Charalambous, J. Kirz, D. Sayre, Nature 400, 342�344 (1999). [2] J. N. Clark, et al, Science 341, 56 (2013). [3] M. A. Pfeifer, et al, Nature 442, 63�66 (2006). [4] J. N. Clark, et al, Phys. Rev. Lett. 112, 113901 (2014). [5] P. Thibault & A. Menzel, Nature 494, 68-71 (2013). [6] J. M. Rodenburg, et al, Ultramicroscopy 107, 227 (2007). [7] P. Thibault, et al, Science 321, 379 (2008). [8] The authors acknowledge funding from Volkswagen Foundation. Figure 1. A) Three dimensional images of a nanocrystal showing the deformation due to an optical pump pulse for several different delay times. The experiment is compared to linear elasticity theory (Theory) and molecular dynamics simulation (Simulation). The scale bar is 100 nm. B) A test pattern is imaged when it is stationary (Static) and when it is oscillating (Dynamic) much faster than the temporal resolution of the experiment. The scale bar is 1000 nm.");sQ1[1082]=new Array("../7337/2167.pdf","High-Resolution Phase-Contrast Imaging at XFELs","","2167 doi:10.1017/S1431927615011617 Paper No. 1082 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 High-Resolution Phase-Contrast Imaging at XFELs A. Schropp1, R. Hoppe2, V. Meier2, J. Patommel2, F. Seiboth2, Hae Ja Lee3, B. Nagler3, E. C. Galtier3, B. Arnold3, J. B. Hastings3, and C. G. Schroer1 (1) Deutsches Elektronen-Synchrotron DESY, Hamburg, Germany (2) Technische Universitat Dresden, Germany � (3) SLAC National Accelerator Laboratory, USA Current x-ray sources of the fourth generation, also denoted by x-ray free-electron lasers (XFELs), produce intense and short x-ray pulses opening up completely new scientific possibilities for the investigation of fast dynamical processes [1, 2]. In this way snapshots of the state of matter can be recorded at specific moments in time. However, some experiments require as well high spatial resolution in order to visualize physical processes occurring on short length scales. While the temporal resolution is determined by the pulse length, the spatial resolution in x-ray imaging is mainly limited by the properties of the x-ray optics and diffraction effects. We built an x-ray microscope for the Matter in Extreme Conditions (MEC) endstation of the LCLS, which is especially adapted to the XFEL environment, and used the setup to carry out different experiments based on magnified x-ray phase-contrast imaging. The method allows one to image a sample with high spatial resolution and high sensitivity to small density changes within a sample. The xray microscope is based on a set of Beryllium compound refractive lenses (Be-CRLs) creating a secondary x-ray source closely in front of a sample. The sample is positioned in the divergent x-ray beam and a CCD detector records a magnified image of the illuminated area at a larger distance further downstream [cf. Fig. 1a)]. In a pump-probe scheme an optical laser hits the sample from the side, which initiates a shock wave propagating into the material, and the XFEL pulse probes the state of the material shortly after. In Figs. 1 b) � 1 e) a series of measured phase-contrast images recorded on Kapton at diff t delay times between the optical and XFEL pulse is shown. Larger features such as the position of the shock front and structures behind it can be observed directly in the image. However, due to diffraction effects it is necessary to evaluate the phase-contrast images numerically using phase retrieval methods in order to obtain high-resolution images of the sample. In addition, a quantitative analysis requires a precise knowledge of the illumination wave fi to disentangle features already present in the incoming x-ray beam from the actual object. Beam characterization was carried out using scanning coherent x-ray microscopy (ptychography). The method yields the wave fi distribution of the nano-focused XFEL beam with high spatial resolution (cf. Fig. 2) [3, 4]. In this contribution we report on current activities to implement high-resolution x-ray imaging at the Linac Coherent Light Source (LCLS) [5]. We demonstrate these new capabilities at the example of imaging of shock waves in different materials, outline the analysis procedure and discuss limitations related to fluctuations in beam intensity and position, bandwidth of the XFEL beam [6] and dispersive properties of the x-ray optics. Microsc. Microanal. 21 (Suppl 3), 2015 2168 a) d R l R0 b) drive laser magnified image 5 ns c) 10 ns d) f = 250 mm x focus sample L = 4143 mm shock front e) CRL detector 15 ns 20 ns 100 �m Figure 1: a) Experimental geometry for magnified phase-contrast imaging. b) � e) Phase contrast images measured on Kapton at different time delays between 5 ns and 20 ns. 2 mm Figure 2: Vertical intensity profile of a 150 nm-XFEL beam (FWHM). References [1] Chapman, H. N. et al. Femtosecond x-ray protein nanocrystallography. Nature 470, 73�U81 (2011). [2] Berrah, N. et al. Double-core-hole spectroscopy for chemical analysis with an intense X-ray femtosecond laser. PNAS 108, 16912�16915 (2011). [3] Schropp, A. et al. Full spatial characterization of a nanofocused x-ray free-electron laser beam by ptychographic imaging. Scientific Reports 3, 1633 (2013). [4] Schropp, A. et al. Scanning coherent x-ray microscopy as a tool for xfel nanobeam characterization. Proc. SPIE 8849, 88490R (2013). [5] Schropp, A. et al. Developing a platform for high-resolution phase contrast imaging of high pressure shock waves in matter. In Moeller, S. P., Yabashi, M. & Hau-Riege, S. P. (eds.) Proc. of SPIE, vol. 8504, 85040F (SPIE, San Diego, 2012). [6] Seiboth, F. et al. Focusing XFEL SASE pulses by rotationally parabolic refractive x-ray lenses. In J. Phys. Conf. Ser., vol. 499, 012004 (2014). 4 m");sQ1[1083]=new Array("../7337/2169.pdf","Fast Solid State Electron Detectors Based on the Principle of Silicon Drift Detectors for Efficient Soft and Hard Matter Analysis","","2169 doi:10.1017/S1431927615011617 Paper No. 1083 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Fast Solid State Electron Detectors Based on the Principle of Silicon Drift Detectors for Efficient Soft and Hard Matter Analysis A.Liebel1, G.Lutz2, U.Weber1, A.Niculae1, H.Soltau1 1 2 PNDetector GmbH, Otto-Hahn-Ring 6, D-81739 M�nchen, Germany PNSensor GmbH, Otto-Hahn-Ring 6, D-81739 M�nchen, Germany Fast imaging with electrons is a field of growing interest. Major reasons for high scan rates are the investigation of sensitive soft matter specimen (to minimize sample damage) or the growing demand on high throughput in industrial applications. Modern microscopes will need advanced detectors which are fast even under difficult imaging conditions. For Backscattered Electron (BSE) imaging or Scanning Transmission Electron Microscopy (STEM) most instruments use solid state (silicon) or scintillator detectors. The two major benefits of solid state detectors are the slim shape and the freedom in designing any preferred geometry. However, they have to be large to capture enough signal, but since the size of the detector also determines the signal capacitance the speed of the detector system at high signal amplification is limited. Hence, new readout concepts have to be implemented in order to accomplish solid state electron detectors which combine large active areas and high detection speed. Last year we introduced a new detector concept based on the principle of Silicon Drift Detectors (SDD) with integrated Field Effect Transistor (FET) [1]. These sensors are well known from Energy Dispersive X-ray (EDX) spectroscopy. They make use of the principle of sideward depletion to collect the signal charge from a large detector volume to a small anode with very low signal capacitance. Instead of detecting single X-ray photons, SDDs can in principle also be used to measure dynamic electron currents. However, in order to manage the wide variety of electron current intensities an additional feature has to be added in order to control the current amplification. In our new detector concept this is done by a second integrated FET which is connected to the anode. It can be used to regulate the mechanism of discharging the anode and therefore to adjust the gain to the signal intensity. That way, a new electron current detector has been created that features an extremely small signal capacitance (and therefore low detector noise) and a very fast first amplification stage integrated into the chip. Test samples of the new detector type have been fabricated with active areas up to 8 mm� and measurements with these structures were performed inside the SEM. Figure 1 shows an 8 mm� test detector installed inside the microscope in a configuration perpendicular to the beam axis. Figure 2a presents the measurement result of this structure where the maximum detector speed was determined by fast scanning over a knife edge. The achieved minimum signal rise time at 50 kOhm gain was well below 50 nsec which enables scanning with pixel dwell times of 50 nsec or less without the occurance of smearing effects. Increasing the gain reduces the detection speed (due to the small parasitic capacitance of the first amplification stage the signal rise time) but is still below 500 nsec for very high gain values of 10 MOhm (see Figure 2b). This is about 10 times faster than conventional BSE detectors under similar conditions. This allows for high scan rates even at very weak electron signals which is very beneficial, especially for soft matter analysis. Fast measurements with low primary currents and energies can reduce measurement time, sample damage and therefore preparation time. Microsc. Microanal. 21 (Suppl 3), 2015 2170 Finally, we fabricated the first larger annular BSE detector chip based on SDD technology. It has a total active area of 40 mm� and will be soon integrated and tested in an SEM. We will present measurements results and BSE images with the new detector that demonstrate its potential for fast and non-destructive imaging of soft and hard matter samples. [1] A.Liebel et al., Microscopy & Microanalysis, vol. 20, S3 (2014), pp. 28-29 Figure 1. Test detector fabricated last year with 8 mm� active area inside the SEM. Measurement results of a larger new detector in annular configuration will be shown. a) b) Figure 2. a) By fast scanning over a knive edge signal rise times of less than 50 nsec could be achieved with an 8 mm� test detector at 50 KOhm first stage gain. b) The measured rise time is still < 500 nsec at a very high gain of 10 MOhm as needed for weak signals which are typical for soft matter analysis. This is about 10 times faster than a conventional solid state BSE detector at similar gain.");sQ1[1084]=new Array("../7337/2171.pdf","Analytical Electron Tomography: Methods and Applications","","2171 doi:10.1017/S1431927615011630 Paper No. 1084 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Analytical Electron Tomography: Methods and Applications Georg Haberfehlner1, Angelina Orthacker1, Franz P. Schmidt2, Anton H�rl3, Daniel Knez2, Andreas Tr�gler3, Ulrich Hohenester3 and Gerald Kothleitner1,2 1. 2. Graz Centre for Electron Microscopy and Nanoanalysis, Graz, Austria Institute for Electron Microscopy and Nanoanalysis, Graz University of Technology, Graz, Austria 3. Institute of Physics, University of Graz, Graz, Austria. Electron tomography is a powerful technique for 3D characterization at the nanoscale. Recent developments focus on extracting a wide range of information about a sample in 3D [1]. Of special interest is the combination of electron tomography with spectroscopic techniques - EFTEM, EELS and EDS - to recover the information present in spectroscopic signals in three dimensions. Analytical electron tomography allows mapping of chemical variations and gradients, approaching the goal of full 3D elemental quantification [2]. Additionally, EELS tomography can be used to extract information about materials properties or chemical bonding [3,4]. In this presentation we will discuss the steps necessary to successfully combine spectroscopy and tomography and show respective applications. For analytical electron tomography a tilt series of spectrum images is recorded, over the widest tilt range possible for the actual sample geometry. The key to a successful reconstruction then lies within data processing, in particular in the extraction of the desired information from the spectroscopic signal, and the alignment and reconstruction of the tomographic tilt series. We will discuss novel, reliable and automated procedures for tomographic alignment together with advanced reconstruction algorithms, which incorporate additional information in the reconstruction process available from the sample. Particular emphasis is laid on compressed sensing algorithms, including total-variation (TV) minimization [5]. To give an example for a simultaneous EELS and EDS tomography and 3D mapping experiment, we demonstrate the reconstruction of phases in Al-alloys via TV minimization algorithms [2,6]. In this case the technique was applicable, since sharp interfaces between different materials in the sample were present, even though the quality and number of projections was limited. In addition we recovered spectral EDS and EELS data, by reconstructing each spectral channel, providing four-dimensional datasets with spectra available separately for each voxel (see Fig. 1). We will also discuss the power of EELS tomography for the extraction of optical material properties in the form of 3D surface plasmon fields for metallic nanostructures (see Fig. 2). Correct reconstructions need meaningful physical models as basis for special algorithms needed in the 3D reconstruction of the low energy-loss signal [7]. The paper attempts to give insight into the current status of analytical electron tomography, its potentials and current limitations. [8] Microsc. Microanal. 21 (Suppl 3), 2015 2172 [1] PA Midgley and JM Thomas, Angew. Chem. Int. Ed. 53 (2014), p. 8614. [2] G Haberfehlner et al, Nanoscale 6 (2014), p. 14563. [3] O Nicoletti et al, Nature 502 (2013), p. 80. [4] B Goris et al, ACS Nano 8 (2014), p. 10878. [5] R Leary et al, Ultramicroscopy 131 (2013), p. 70. [6] G Kothleitner et al, Microsc. Microanal. 112 (2014), p. 678. [7] A H�rl, A Tr�gler and U Hohenester, Phys Rev Lett 111 (2013), p. 076801. [8] This research has received funding from the European Union within the 7 th Framework Program [FP7/2007�2013] under grant agreement no. 312483 (ESTEEM2). We thank Jiehua Li and Peter Schumacher from University of Leoben and Harald Ditlbacher and Joachim Krenn from University of Graz for samples and discussions. Figure 1. 3D elemental maps of an AlSi-alloy and single voxel EELS and EDS spectra [2]. Figure 2. Dipole modes in a triangular metal-insulator-metal structure: (a) HAADF STEM image (b) energy-loss spectrum, (c) & (d) EELS maps of coupled dipolar modes visible in (b).");sQ1[1085]=new Array("../7337/2173.pdf","Multi-Dimensional Machine Learning Aided Analysis of a Nickel-Based Superalloy","","2173 doi:10.1017/S1431927615011642 Paper No. 1085 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Multi-Dimensional Machine Learning Aided Analysis of a Nickel-Based Superalloy David Rossouw1, Robert Krakow1, Zineb Saghi1, Catriona S M Yeoh1, Pierre Burdet1, Rowan K Leary1, Paul A Midgley1. 1. Department of Materials Science and Metallurgy, University of Cambridge, 27 Charles Babbage Road, Cambridge, CB3 0FS, UK. In the wake of improved X-ray detector efficiencies [1] and advanced data processing techniques [2-3] emerges the practicality of multi-dimensional electron microscopy, an analytical approach to materials characterization that combines spatial and spectral information [4]. In this work we combine electron tomography with energy dispersive X-ray (EDX) spectroscopy, a technique some term 4D microscopy, to obtain three-dimensional (3D) chemical information at the nano-scale. Such information is essential to fully understand structure-property relationships in materials. The approach is used to analyze a nextgeneration nickel-based superalloy, a critically important high strength material used in the aerospace and power industries. Mapping the 3D distribution of the alloying elements, both common and exotic, is essential to better understand the structural origin of its exceptional strength and ultimately aid the design of new superalloys capable of withstanding increased service temperatures. Here, a 1800 EDX spectrum image tomographic series is acquired from a focused ion beam-prepared superalloy needle specimen. The large 4D dataset is subsequently analyzed using a machine learning algorithm implemented in HyperSpy [5], called independent component analysis (ICA), to swiftly identify the major spectral components in the tilt series belonging to the individual phases present in the alloy. These components are found to be spatially and chemically representative of the matrix (gamma) and precipitate (gamma') phases present in the superalloy, enabling their size, distribution and composition to be conveniently and efficiently reconstructed in 3D (Fig. 1). The calculated composition of gamma and gamma' component spectra are in excellent agreement with atom probe data [6], providing the necessary validation of the machine learning-based analysis approach. The use of ICA on a tomographic tilt series dataset promises to be a powerful and efficient methodology to perform full 3D chemical analysis on a variety of nano-scale systems [7]. References: [1] H. S. Von Harrach et al. J. Phys. Conf. Ser. 241 (2010), p. 012015. [2] F. de la Pe�a et al. Ultramicroscopy 111 (2011), p. 169. [3] P. Burdet et al. Acta Mater. 61 (2013), p. 3090. [4] P. A Midgley and J. M. Thomas, Angew. Chem. Int. Ed. Engl. 53 (2014), p. 2. [5] www.hyperspy.org [6] L. Viskari, K. Stiller, Ultramicroscopy 111 (2011) p. 652. [7] Special thanks to Giorgio Divitini and Lech Staniewicz for preparation of the FIB prepared needle. D.R. acknowledges support from the Royal Society's Newton International Fellowship scheme. Microsc. Microanal. 21 (Suppl 3), 2015 2174 Figure 1. (a) The EDX detector geometry in relation to the on-axis specimen holder. (b) The acquisition of a 1800 EDX spectrum image tilt series of a needle specimen. (c) The selected region of interest close to the needle tip. (d) The ICA component 3D reconstructions. (e) Segmentation of the gamma' strengthening phase.");sQ1[1086]=new Array("../7337/2175.pdf","Off-axis chromatic scanning confocal electron microscopy for inelastic imaging with atomic resolution","","2175 doi:10.1017/S1431927615011654 Paper No. 1086 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Off-axis chromatic scanning confocal electron microscopy for inelastic imaging with atomic resolution Changlin Zheng1, Ye Zhu1, Sorin Lazar2, and Joanne Etheridge1, 3 1. 2. Monash Centre for Electron Microscopy, Monash University, Victoria 3800, Australia FEI Electron Optics, 5600 KA Eindhoven, The Netherlands 3. Department of Materials Engineering, Monash University, Victoria 3800, Australia The development of electron-optical techniques for geometric and chromatic aberration correction have significantly improved the spatial resolution of transmission electron microscopy (TEM) and enhanced its capability for chemical mapping [1]. In the present work, we develop an off-axis confocal mode in a scanning transmission electron microscope (STEM) which uses the intrinsic chromatic aberration of the imaging lens system to enable fast imaging with inelastically scattered electrons without using a spectrometer [2]. Furthermore, the off-axis confocal configuration opens the possibility of obtaining depth-sensitive inelastic images for 3D imaging [3, 4]. The principle of off-axis scanning confocal electron microscopy (SCEM) is shown schematically in figure 1a. It utilizes a double-aberration corrected S/TEM to enable a large confocal angle to be employed. In the off-axis confocal mode, the electron beam is incident on the sample at an angle, , to the optic axis. The tilted beam introduces a lateral shift of the scattered electron wave function (elastic and inelastic) at the back focal plane of the imaging lens. Furthermore, the scattered electrons are chromatically dispersed both parallel and perpendicular to the optic axis, effectively separating electrons with different energies. In this mode, electrons with a chosen energy loss can be focused at the confocal point using the image lens (fig 1b) and detected selectively using an integrating detector to form an inelastic STEM image (fig 2c). The method is illustrated experimentally in figure 2. A spherical-aberration corrected [5] Titan3 80-300 S/TEM fitted with a standard Schottky field emission gun is used to examine DyScO3. First, the effectiveness of the energy filtering in this mode is tested by placing the chromatic confocal point at the entrance aperture of a post-column spectrometer. Two separate spectra corresponding to two different imaging lens focal conditions are shown overlaid in Fig 2a. In one condition, energy-loss electrons corresponding to the Scandium L2,3 edge (~400-410 eV) are focused at the chromatic confocal point, in the other the Sc L2,3 pre-edge is in focus. The results prove only certain inelastically scattered electrons with selected energy loss (equal to the confocal energy) are focused at the confocal point (fig 2a). By collecting the chromatic confocal signal with a finite size BF detector, SCEM images are recorded (fig 2b and c). The image formed with the Scandium L2,3 on-edge signal clearly shows atomically resolved contrast (fig 2c) while the pre-edge signal only shows faint contrast (fig 2b) . The SCEM mapping is acquired with a pixel dwell time of 0.38 ms. This is much faster than conventional EELS spectrum imaging which is limited by the read out time of the CCD. More importantly, this off-axis chromatic confocal configuration offers the potential for fast three-dimensional chemical mapping when coupled with the improved depth and lateral resolution of the incoherent confocal mode [4, 5]. References: Microsc. Microanal. 21 (Suppl 3), 2015 2176 [1] K.W. Urban, J. Mayer, J.R. Jinschek, M.J. Neish, N.R. Lugg, and L.J. Allen, Phys. Rev. Lett. 110, (2013) p. 185507. [2] C.L. Zheng, Y. Zhu, S. Lazar, J. Etheridge, Phys. Rev. Lett. 112 (2014) p. 166101. [3] T. Wilson and C. Sheppard, Theory and practice of scanning optical microscopy (Academic Press, London ; Orlando, 1984). [4] C.L. Zheng et al, in preparation [5] CESCOR and CETCOR by CEOS GmbH. [6] Funding is acknowledged from the Australian Research Council Grants DP110104734 and LE0454166. Figure 1. (a) Schematic showing the Off-axis Chromatic SCEM mode. (b) The intensity distribution in the image plane after scattering from amorphous carbon in the off-axis chromatic SCEM mode. Figure 2. (a) EEL spectra (measured using a post-column spectrometer) arising from electrons around the confocal point in off-axis SCEM. Two separate spectra corresponding to two different focal conditions are shown overlaid � one focus corresponds to the Sc L2,3 pre-edge and the other to the Sc L2,3 on-edge. (b) (c) Off-axis SCEM image using (b) pre-edge and (c) on-edge core loss electrons of Sc L2,3 in DyScO3 (no post-column spectrometer is used) .");sQ1[1087]=new Array("../7337/2177.pdf","Depth Sensitivity of Atomic Resolution Aberration-Corrected Through-Focus STEM Imaging of Bi Dopants on Cu Grain Boundaries","","2177 doi:10.1017/S1431927615011666 Paper No. 1087 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Depth Sensitivity of Atomic Resolution Aberration-Corrected Through-Focus STEM Imaging of Bi Dopants on Cu Grain Boundaries C. A. Wade, M. Watanabe 1. Dept. of Materials Science and Engineering, Lehigh University, Bethlehem, PA A complete 3-dimensional view of materials at an atomic level has long been a goal of modern electron microscopy. Difficulties in obtaining 3-dimensinal images of materials at this high magnification level without the loss or distortion of information has been a challenge yet to be overcome [1]. The advent of an aberration-correction scanning transmission electron microscope (STEM) equipped with a cold fieldemission electron source has allowed much higher beam convergence angles to be utilized without parasitic aberration degrading image quality [2]. The benefit of using a higher beam convergence angle is two-fold: first, the higher convergence angle offers an improved lateral resolution as lateral resolution is inversely proportional to the beam convergence semi-angle, and second, the vertical resolution (along the optic axis) is improved as the vertical resolution is inversely proportional to the square of the beam convergence semi-angle. With these benefits large convergence angles are increasingly being used for STEM imaging in aberration-corrected systems. In this study the vertical resolution was measured by imaging single Bi atoms dispersed along a Cu bicrystal boundary as a function of the depth from the electron beam entrance (top surface) surface of the specimen. Of particular interest was how quickly vertical resolution would become degraded from various scattering events through the specimen and if through-focus STEM imaging was a practical approach to generating 3-dimensional images of these heavy dopant atoms along a grain boundary (GB). In the system studied Bi segregates almost exclusively to the Cu GB despite that the Bi atom location and the GB structure are not completely understood as it may change drastically depending on the geometric factors of the GB misorientation and the particular GB planes at GB of any given misorientaiton. By intentionally inclining the GB with respect to the incident electron beam, a specimen orientation in a near plane-view orientation is established, i.e. pseudo plan-view projection (PPP) imaging. From this orientation Bi atoms are laterally displaced from one another in the image projection that is formed allowing the image intensity of each Bi atom to be observed independently from any neighboring Bi atoms. The best vertical resolution of a single Bi atom observed in this study was 4.57 nm. Figure 1a shows an x plane atomic number contrast (Z-contrast) image formed from a series of images obtained in a through-focus image series acquisition. The bright areas of this x plane Z-contrast image correspond to the high intensity generated from electron scattering off of the high atomic number Bi atoms. Figure 1b shows the intensity profile taken over one Bi atom, which represents the vertical resolution measured as the full width at half maximum (FWHM) value of intensity. Similar FWHM values may be measured from Bi atoms at several depths through the Cu specimen allowing the impact that the distance from the top surface of the specimen has on vertical resolution to be seen. Figure 2 plots the measured FWHM value of vertical Bi intensity as a function of the depth from the top surface. From this limited sampling it can be seen that the vertical resolution deteriorates rather quickly with increasing specimen thickness. These experimental results for vertical resolution were compared with simulated images of Bi atoms at various thicknesses in a Cu matrix. Microsc. Microanal. 21 (Suppl 3), 2015 2178 References: [1] H. L. Xin and D. A. Muller, J Electron Microsc 58 (2009), p. 157-165. [2] A. R. Lupini et al, Microsc Microanal 15 (2009), p. 441-453. [3] The authors acknowledge support NSF through grants DMR-0804528 and DMR-1040229 0 Imaging Depth (nm) -10 FWHM 4.57 nm To Specimen Bottom Surface -20 -2 0 2 4 6 Intensity (10 counts) 6 (a) (b) Figure 1. A Z-contrast intensity image (a) showing the vertical distribution of electron high-angle scattering from 2 Bi atoms on the Cu grain boundary in a PPP imaging orientation with (b) showing the vertical resolution, measured as 4.57 nm, given as the FWHM of intensity from one Bi atom in the Zcontrast intensity image. FWHM of Imaged Bi Atom (nm) 5.2 5 4.8 4.6 8 10 12 14 16 Position from Top Surface of Bi Atom Being Imaged (nm) Figure 2. A graph showing the decay of vertical resolution as measured Bi atoms were imaged further from the top surface of the Cu specimen.");sQ1[1088]=new Array("../7337/2179.pdf","Super-X XEDS STEM Tomography of Precipitates in the LSHR Nickel Superalloy","","2179 doi:10.1017/S1431927615011678 Paper No. 1088 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Super-X XEDS STEM Tomography of Precipitates in the LSHR Nickel Superalloy S.J. Kuhr1, J.M. Sosa1, D.E. Huber, H.L. Fraser1 1. Center for the Accelerated Maturation of Materials, The Ohio State University, Columbus Electron tomography has been proposed to be advantageous for reconstructing sub-micron precipitates in nickel superalloys. However, not every type of electron image provides the necessary contrast for high fidelity tomographic reconstruction. The precipitates in Low Solvus High Refractory (LSHR) form coherently with the matrix and exhibit limited atomic number contrast relative to the matrix when imaged with traditional transmission (TEM) and scanning transmission electron microscopy (STEM) bright field/dark field imaging. Previous research has shown that energy filtered TEM (EFTEM) has displayed successfully morphologies based off the Cr-L3,2 edge[1,2]. However, EFTEM is hindered by poor signal to noise during image collection requiring prohibitive acquisition times. X-ray energy dispersive spectroscopy (XEDS) in STEM offers a robust method to collect a wide range of compositional information and enables spectral image (SI) collection that permits post-processing analysis of multiple elemental species. The primary objective of the work to be presented was to use Super-XTM XEDS in STEM to acquire compositional maps of LSHR to produce tomographic reconstructions that accurately depict precipitates within a dual-microstructure heat treatment (DMHT) gradient [3] as shown in Figure 1(a). Chromium is a significant component of the LSHR alloy at 12.3 wt% and segregates to the matrix, therefore chromium XEDS SI can provide strong contrast between the precipitates and the chromium-rich matrix. The XEDS SI shown in Figure 1(b) was used for the full precipitate reconstruction as displayed in Figure 1(c). The DMHT gradient contains a range of intricate morphologies that have proven difficult to characterize with two-dimensional methods. Three-dimensional characterization not only provides the morphology of the , but will also permit more accurate microstructural metrics for inputs for integrated computational materials (ICME) models. Images for Super-X XEDS tomographic reconstructions were acquired using a FEI TitanTM G2 60-300 S/TEM equipped with Super-XTM technology. The Super-XTM collection system is composed of four silicon drift detector (SDD) covering a 0.7 srad collection angle providing for fast and efficient XEDS SI collection with nanometer scale spatial resolution. A tomographic tilt series was acquired from needle-shaped LSHR specimens fabricated using a HeliosTM dual-beam FIB. Using a needle shape geometry and tilting along the longitudinal axis avoids projected thickness variations that usually occur with conventional thin foils. SIs were collected with acquisition times of 300s/image and 20 �s/pixel dwell times. The tilt series was collected from -62� to +62� at 2� increments. Due to the limited volume contained in the extracted needles, additional characterization techniques were applied for a more complete reconstruction. To this end, FIB serial sectioning was performed using images formed from secondary electron (SE) contrast [4]. Successive layers of the samples were removed via ion milling and subsequently SE imaged. Figure 2(a) displays the reconstructed volume of precipitates after image processing and feature extraction using MIPARTM [5]. Figure 2(b) shows the non-edge touching precipitates of the reconstructed volume. Both types of acquired data required rigorous image processing to account for subtle changes in contrast and image translation during data collection. Microsc. Microanal. 21 (Suppl 3), 2015 2180 The reconstructions produced by FIB revealed an apparent interconnected nature of the 200nm precipitates and also provided an ability to isolate a single precipitate. As this reconstruction was based on SE contrast, it provided no information about compositional differences between the precipitates and the matrix of the material. Conversely, the Super-X XEDS reconstruction offered compositional information as well as the ability to reconstruct both size scales of precipitation in the LSHR nickel superalloy. XEDS/STEM acquisitions were able to capture particles approximately 10 nm in size. It was determined that FIB serial sectioning provided a larger number of precipitates to analyze and the Super-XTM-XEDS reconstruction offered composition and precipitate morphology information with nanometer scale spatial resolution. From the observations made in this study, it was apparent that both reconstruction techniques provided valuable metrics that could be useful with ICME models. References: [1] [2] [3] [4] [5] G. B. Viswanathan, et al., Acta Materialia 53 (2005), p. 3041-3057. P.M. Sarosi, et al., Ultramicroscopy 103 (2005), p. 83-93. J. Gayda, et al., Superalloys (2004), p. 323-329. M.D. Uchic, et al., Ultramicroscopy 109 (2009), p. 1229-1235. J.M. Sosa, et al. Integrating Materials and Manufacturing Innovation 3 (2014), p. 18. (a) (b) (c) Figure 1. (a) High resolution scanning electron microscope image of LSHR after a etch (b) TitanTM chromium EDS map in gray scale (c) Full rendering of TitanTM data (a) (b) Figure 2. (a) volume element collected using serial sectioning performed in the DualBeamTM FIB (b) Reconstruction of non-edge precipitates");sQ1[1089]=new Array("../7337/2181.pdf","Nanoscale Mechanical Scattering Experiments towards the Local Analysis of Thermal Active Modes and Dispersion Interactions at Interfaces","","2181 doi:10.1017/S143192761501168X Paper No. 1089 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Nanoscale Mechanical Scattering Experiments towards the Local Analysis of Thermal Active Modes and Dispersion Interactions at Interfaces Ren� M. Overney Department of Chemical Engineering, University of Washington, Benson Hall, Seattle, WA 98195 Thermal active modes, i.e., molecular and submolecular mobilities, inherent to the relaxation behavior of solid condensed organic matter, such as polymers and molecular self-assemblies, and, interfacial and surface free energies, are of great importance for many practical applications. For instance, in organic second-order nonlinear optical materials (NLO), pursued for photonic device applications, the molecular mobility plays a pivotal role in producing stable NLO materials of high poling efficiency [1]. Other applications such as frictional sliding systems or nanocomposites, both involving nanoconstrained materials, depend on the surface/interfacial energies. Common to thermal modes and interfacial forces are that they are difficult to quantify, in particular, in systems that either possess an amorphous complex entropic structure, or are of small size. From scanning force microscopy (SFM) an approach has been devised over the years, providing direct and local insight into thermal modes and interfacial forces via an energetic analysis that is based on the time-temperature superposition principle [2]. Dubbed "intrinsic friction analysis" (IFA), it utilizes the mechanical scattering process between a sliding SFM tip in contact with the thermal active modes of the scanned sample. In the past, in particular, rotational and translational modes have been investigated involving complex organic systems, such as polymers and organic molecular glasses [2-5]. Recently, also molecular binding interactions [1] and surface dispersion interactions [6] could be energetically analyzed with IFA. In this paper, we will focus on both aspects, namely the molecular mobility in self-assembled systems depending on the molecular interaction strength, and the quantum electro-dynamic binding fluctuations between Van der Waals interacting surfaces. Based on a prior study involving dendritic organic non-linear optical (NLO) self-assembly molecular glasses, Fig. 1(a), an intermediate thermal phase regime was discovered that was found pivotal for both the phase stability and the degree of non-linear polarization. In the first part of this presentation, we focus on the energetic analysis of this intermediate phase regime in contrast to the two bordering "low" and "high" temperature phase regimes, Fig. 1(b), as function of the dendritic interactions, involving arene-perfluoroarene moieties, and the strength of the intrinsic molecular dipole fields. The presence of dendritic groups was found to fundamentally alter transition temperatures and the molecular relaxation behavior. Based on enthalpic and entropic energetic IFA analyses, thermally active modes in the low temperature regime were found to be intimately connected to the dendron structure, while in the intermediate phase regime a substantial amount of the total energy was found to be of cooperative entropic nature, Fig. 1(c). The multiple interactions (from dipole-dipole interactions to local non-covalent dendritic interactions) are discussed and summarized in a model that describes the thermal transitions and phases. Microsc. Microanal. 21 (Suppl 3), 2015 2182 While we focus in the first part in this presentation on the energetics in molecular mobility in self-assembled systems, in the second part, we will direct the discussion to IFA's applicability to spontaneous quantum electrodynamic fluctuations, i.e., Van der Waals interactions, between inorganic substances, Fig. 2, involving materials, such as layered materials (e.g., graphene, or molybdenum disulfide) and ionic materials (calcium fluoride). This study reveals IFA's ability to directly determine the Hamaker constant, and thus, gaining direct access locally to the interfacial energy, and consequently to the surface energy. Findings addressed here entail among others, good agreement of IFA data with the numerical analysis of EELS and optical reflectance data regarding Hamaker constant determination, and, a surprisingly strong reduction of the surface free energy from the macroscopic graphite system to single layer graphene. References [1] D. B. Knorr, S. J. Benight, B. Krajina, C. Zhang, L. R. Dalton, R.M. Overney, J. Phys. Chem. B, 116, (2012) 13793. [2] D.B. Knorr, R.M. Overney, J. Chem. Phys., 129, (2008) 074504. [3] S.E. Sills, T. Gray, R.M. Overney, J. Chem. Phys., 123, (2005) 134902. [4] T. Gray et al., NanoLetters, 8, (2008) 754. [5] D.B. Knorr, Jr., L. S. Kocherlakota, J. P. Killgore, and R. M. Overney, J. Membrane Sci., 346, ( 2010) 302. [6] B.A. Krajina, L.S. Kocherlakota, R.M. Overney, J. Chem. Phys., 141, (2014) 164707 Figure 1. (a) Schematic self-assembly NLO material. (b) Energetics in low/intermediate and high temperature regime. (c) Cooperative entropy (FF) in the three temperature regimes. Figure 2. Spontaneous quantum electrodynamic fluctuations coupled to friction F at velocity v at a fixed temperature.");sQ1[1090]=new Array("../7337/2183.pdf","High-Resolution Mapping of Quantitative Elastic Modulus of Polymers","","2183 doi:10.1017/S1431927615011691 Paper No. 1090 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 High-Resolution Mapping of Quantitative Elastic Modulus of Polymers Sergei Magonov1, Sergey Belikov1, Marko Surtchev1, Stas Leesment2, and Ivan Malovichko2 1 2 NT-MDT Development Inc, Tempe AZ, 85284, USA, NT-MDT Co. Building 100, Zelenograd, Moscow 124482, Russia Mechanical properties in Atomic Force Microscopy (AFM) are evaluated with the help of force curves, which are recorded at single or multiple locations of surface area. Here we report a progress in quantitative studies of elastic modulus of polymer materials with the Hybrid mode [1]. Although the mechanical properties can be derived from the deflection-versus-distance (D-vZ) and amplitude-versus-distance curves obtained in the contact and amplitude modulation modes, the use of the deflection-versus-time (D-v-t) curves recorded in the Hybrid mode is most efficient for on-line mapping of elastic modulus. In this mode a sample vertically oscillates at 1-2 kHz that is well below the resonances of the probe and scanner. In each cycle the probe deflects in response to the tip-sample forces to a set-point deflection level and retracts back. The related D-v-t curve is characterized by the baseline, approach and retracting parts. The wells and slopes of the approach/retract traces reflect the sample adhesion and deformation. When the microscope's optical sensitivity and spring constant of the probe are known, a force-versusdeformation (F-v-h) dependence can be retrieved from the D-v-t curve or related D-v-Z curve. The F-v-h curve can be analyzed with solid state deformation models (Hertz, DMT, JKR) to get the elastic modulus and work of adhesion. Due to 1-2 kHz rate of the D-v-t curves, they can be collected as the maps of up to 1024�1024 points during scanning with 1 Hz rate. A fast AFM controller enables on-line analysis of the force curves and the derived modulus and adhesion maps can be viewed simultaneously with the height image. A correlation between the elastic modulus values of neat polymers obtained in the Hybrid mode and macroscopic studies, and the use the modulus maps for compositional imaging of heterogeneous polymers are the main questions. In addition, we will address the spatial resolution of the mapping. The samples of polycarbonate (PC) and high-density polyethylene (HDPE) were prepared as blocks (thickness > 1 mm) by hot pressing. A film (thickness > 100 nm) of blend of PS with lowdensity polyethylene (LDPE) was made by spin-casting on a piece of Si wafer. The Hybrid mode studies were carried out with a scanning probe microscope Next (NT-MDT), which was placed in the electronically-controlled temperature, acoustic and vibrational enclosure. The low thermal drift (<0.2 nm/min) facilitates high-quality studies with scanning rate of 0.4 - 0.8 Hz, particularly, at the sub-micron scale. Si probes with spring constant near 30 N/m were applied. The probe spring constant probe and inverse optical sensitivity of the microscope were measured with thermal tune method. On-line analysis of D-v-t curves was made by fitting a retract part using DMT (Deruigin-Muller-Toporov) model. This elastic deformation model was used because a difference of the approach and retract parts of the cycle for studied polymers was small. The experimental results are summarized as follows. During scanning of 1 micron area of PC block we changed the force level from 20 nN to 30 nN and to 60 nN. The height image of PC, which shows a sample flat topography, is not shown here. The deformation and modulus maps, which were recorded simultaneously with the height image, are given in Figure 1. As a sample deformation increased stepwise, the elastic modulus stays in the 3.0-3.5 GPa range. The data are in line with the macroscopic PC modulus [2]. The same consistency was observed in our studies of other polymers with elastic modulus from tens of MPa to 7GPa. Study of polymer blends is illustrated by the height image and modulus map with the profile of PS/LDPE blend (Figure 2). Microsc. Microanal. 21 (Suppl 3), 2015 2184 The matrix and inclusions exhibit different modulus with quantitative values seen in the modulus profile. The modulus of the inclusions is below 0.5GPa - close to the macroscopic modulus of LDPE [2]. The matrix is characterized by the modulus of ~ 2.5GPa that matches the modulus of PS [2]. This assignment justifies the modulus mapping as the tool for compositional imaging of polymer blends and block copolymers. The spatial resolution of the modulus mapping was estimated in study of HDPE, which has a high lamellar content. In this case, the height image and modulus map with a profile (Figure 3) show the modulus changes on the 20 nm scale. This example of the mapping demonstrates its high resolution, which is comparable with the tip size. Similar resolution of the modulus mapping of block copolymers will be reported later. Figure 1. Deformation and modulus maps with the profiles along the white dashed lines for PC. During scanning the tip force has changed step-wise as shown in the maps. Figure 2. Height image and modulus map with a profile along the dashed white line for PS/LDPE. Fig. 3. Height image and modulus map with a profile along the dashed white line for HDPE. References [1] S. Belikov et al, MRS Proceedings (2013) p. 1527. mrsf12-1527-uu02-042. [2] J. Wen in "Physical Properties of Polymers. Handbook", ed. J. Mark, (Springer, 2007) p. 487.");sQ1[1091]=new Array("../7337/2185.pdf","Environmental AFM as a Probe of the Morphology, Viscoelasticity and Lubricity of Crosslinked Hydrophilic Biomedical Coatings","","2185 doi:10.1017/S1431927615011708 Paper No. 1091 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Environmental AFM as a Probe of the Morphology, Viscoelasticity and Lubricity of Crosslinked Hydrophilic Biomedical Coatings G. Haugstad1, C. Colling1, A. McCormick1, K. Wormuth1, M. Zeng2 1. 2. Characterization Facility, Univ. of Minnesota, 12 Shepherd Labs, Minneapolis, MN 55455 USA Boston Scientific Corp. , 1 SciMed Place, Maple Grove, MN 55311 USA We explore the morphology and tribo-mechanical properties of polyacrylate catheter coatings variably UV-crosslinked (1,4- butanediol diacrylate), as prepared and following variable tribological history under aqueous immersion (to mimick stresses during catheter deployment) to assess the lubricity and durability of coatings. A rich surface morphology is revealed over micro- to nanoscales, strongly dependent on both the extent of crosslinking and subsequent macrotribological processes. Submicron defect structures resulting from the sponge-coating deposition process include shallow (nm's) circular depressions hundreds of nanometers across ("cheetah spots") and much deeper/narrower "pinholes"; from the UV curing we find deep but larger in diameter "craters" plus "fissures" running between the deep holes and craters. The deepest defects exhibit modified behavior, as sensed in multiple AFM modes sensitive to dissipative properties (friction, adhesion, phase), consistent with these defects extending to the substrate, the Pebax catheter tube. Contact-mode "rippling", a form of wear due to shear forces during raster scanning by the tip, is a strong function of UV curing up to moderate curing times (10 seconds) but only weakly from moderate to long curing times (10-50 seconds). Examples of this phenomenological behavior are contrasted in Figure 1 for the cases of 5 and 10 second curing times. Slowly mapped AFM nanoindentation measurements are used to interrogate changes in both elastic and dissipative properties (i.e., viscoelasticity) of the coatings as a function of the above UV curing as well as subsequent aqueous macro-tribological history. Such nanomechanical responses are further explored under variable humidity and temperature (heated sample). A highly reproducible solvent-induced glass-rubber transition is identified over a narrow range of relative humidity (Figure 2), and found to be shifted to higher humidity on coatings that are "over cured" (50 seconds compared to 5- and 10second curings). These 50 second UV coatings are also found to be nonlubricious in the macrotribological tests. In the process of exploring humidity cycles, dramatic irreversible transformations in coating morphology are fortuitously discovered, whereby many of the defects present from sponge coating and UV curing are largely "healed". Fast force-curve mapping mode is use for (i) zero-wear, high-resolution imaging (including imaging before/after ripple-inducing scanning as shown in Figure 1), wherein the smallest dissipative domains in tip-sample adhesion images become even smaller for higher UV curing times, thus suggesting sensitivity to the size of cooperatively rearranging regions in the glassy state; and (ii) quantifying glass-rubber transition, sensed in stiffness and adhesion images as a function of sample temperature. Broadly, this study demonstrates (1) technologically relevant and scientifically rich phenomenology over scales ranging from tens of microns to tens of nanometers; (2) connections of nanoscale dissipative behavior to macroscale properties; and (3) the complementarity that multiple AFM methodologies [1] afford in analyzing lubricious hydrophilic coatings as are important to biomedical device technologies. This follows similar work on drug-eluting biomedical device coatings [2]. Microsc. Microanal. 21 (Suppl 3), 2015 2186 5 sec 10 sec Rippling, light load Same load, much less rippling 300 FIG. 1. Height images acquired in fast force-curve mapping (Bruker PeakForce QNM) comparing nanoscale rippling on 5-and 10-second UV-cured coatings from preceding contact mode scanning. FIG. 2. Humidity-dependent, raw force versus Z cycle plots (same scales) of coatings following different times of UV curing. Glass-rubber transition occurs at higher humidity for longer curing time. References [1] G. Haugstad, AFM: Understanding Basic Modes and Advanced Applications (Wiley, 2012). [2] G. Haugstad and K. Wormuth, "Nanomechanical Characterization of Biomaterial Surfaces: Polymer Coatings that Elute Drugs", in Industrial Applications of Scanning Probe Microscopy, ed. D. Yablon, (Wiley, 2013).");sQ1[1092]=new Array("../7337/2187.pdf","A New Variable Temperature Solution-Solid Interface Scanning Tunneling Microscope","","2187 doi:10.1017/S143192761501171X Paper No. 1092 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A New Variable Temperature Solution-Solid Interface Scanning Tunneling Microscope Abdolreza Jahanbekam1, Ursula Mazur1 and K. W. Hipps1 1 . Department of Chemistry and Materials Science and Engineering Program, Washington State University, Pullman, Washington, 99164-4630, USA Scanning tunneling microscopy (STM) has been widely used to investigate surface structures and electronic properties of adsorbed species on surfaces and also for observing chemical reactions on surfaces. The technique can be used in various environments such as vacuum, air and solution. Among these environments, ultrahigh vacuum (UHV)-solid and solution-solid (SS) interfaces get the most attention. If we compare sample preparation procedures for these two environments, it is much easier to prepare samples and take measurements at the SS interface, although STM images in UHV generally show higher resolution. If the sample cannot be evaporated without decomposition, SS interface studies may be the only choice. Generally, studies in UHV require that the molecules of interest be vapor deposited � which is impossible for many compounds. No matter how the sample is prepared, by the nature of the experiment, UHV studies are not compatible with chemical equilibrium involving material transport to and from the surface. SS interface studies, on the other hand, are easily adaptable to equilibrium studies. In SS studies, one can change the solvent in order to tune molecule-solvent and substrate-solvent interactions at the SS interface, thereby changing the ordering and structure of the adsorbed species. Gyarfas et al. demonstrated that the length of the alkane chain in different alkanoic solvents had a role in determining the surface structure of coronene on gold [1]. The repair of defects in self assembled layers at the SS interface can be promoted if there is a dynamic exchange between molecules on the surface and in the solution phase [2]. Different areas of technology and science profit from temperature dependent SS interface studies. In the area of technology, these studies can yield vital insights into the critical problems in catalysis, spincasting, friction, crystallization and organic electronics. Temperature dependent studies can yield a great wealth of information including diffusion rates, reaction rates, activation energies and thermodynamic quantities such as entropy and enthalpy of adsorption and/or surface reaction. Because different surface species reach equilibrium at different temperatures, and because some surface reactions are kinetically controlled, a study of a given solution-surface pair as a function of temperature can lead to the discovery of new materials and phases. While STM imaging studies of the SS interface have been in existence for more than two decades, temperature dependent studies have been rare. But one may ask why this fruitful branch of science is seldom seen in the scientific literature. The answer can be traced to the imposed limitation by the design of the commercial STM instruments. In the conventional "in air" design of systems for temperature dependent study, only the sample is heated. This is not the most desirable design for SS interface work. In fact, there are critical problems that are associated with this design. To solve these problems we have designed and built a new system in which all mechanical components of the STM system and the sample are heated in a controlled environment at the saturation vapor pressure of the solution under study (Figure1). This enables the user to image at the SS interface with some nonconducting volatile solvents for long periods of time (> 24 hours) which opens up another exciting door to this area of science. In order to fabricate the instrument we married an improved version of a commercially available STM head with a novel environmental chamber and an external preamplifier system of our own design. We have successfully tested our system by performing a temperature dependence study of cobalt(II) Microsc. Microanal. 21 (Suppl 3), 2015 2188 octaethylporphyrin (CoOEP) at the toluene/Au(111) interface. It is demonstrated that the lattice parameters remain constant within experimental error from 24 �C to 75 �C. We report the unit cell of CoOEP at the toluene/Au(111) interface (based on two molecules per unit cell) to be A = (1.36 � 0.04) nm, B = (2.51 � 0.04) nm and = 97� � 2�. We are also using this STM to study the competitive adsorption of a two component solute system at the solution-solid interface. In general, upon introduction of a multi-component solution to a substrate several scenarios are possible. Phase segregation, formation of randomly mixed monolayers, and coadsorption of components in an ordered fashion are among possible situations. As a model system, competitive adsorption of CoOEP and coronene molecules at the phenyloctane/Au(111) interface is being studied. We find that at room temperature CoOEP cannot be displaced from the gold surface by exposure to a near saturation solution of coronene. On the other hand, coronene is readily completely replaced by CoOEP near saturation solutions. We will report on the concentration and temperature dependence of the competitive adsorption/desorption and how it may lead to two or more phases present on the gold surface. References: [1] B. J. Gyarfas et al, Langmuir 21, 919 (2005). [2] M. Hibino et al, Thin solid films 273, 272 (1996). Figure 1. (a) Front view of controlled-temperature and controlled-atmosphere chamber. (b) Inside view of the chamber showing the STM assembly and a solution reservoir. (c) Backside view of the chamber with all of the connections. The parts are -- 1: Signal access module, 2: Pre-amplifier, 3: Chamber, 4: Port for evacuating the chamber, 5: Port for introducing gas into the chamber, 6: Outlet for fluid cooling, 7: Interface cable between STM and controller, 8: Digital voltmeter for reading thermocouple temperature, 9: Inlet for fluid cooling, 10: Thermocouple connection ports.");sQ1[1093]=new Array("../7337/2189.pdf","The Power of confocal Raman-AFM and Raman-SEM (RISE) Imaging in Polymer Research","","2189 2189 doi:10.1017/S1431927615011721 doi:10.1017/S1431927615011721 Paper No. 1093 Microsc. Microanal. 21 (Suppl 3), 2015 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 � Microscopy Society of America 2015 The Power of confocal Raman-AFM and Raman-SEM (RISE) Imaging in Polymer Research Ute Schmidt1, Wei Liu2, Jianyong Yang2, T. Dieing1, and Klaus Weishaupt1 1. 2. WITec GmbH, Ulm, Germany WITec Instruments Corp., Knoxville, TN, USA Polymers play an essential role in modern materials science. Due to the wide variety of mechanical and chemical properties of polymers, they are used in almost every field of application and are still a dynamic area in the development of new materials. For many of these developments knowledge about the morphology and chemical composition of heterogeneous polymeric materials on a sub-micrometer scale is crucial. In the past two decades, AFM (atomic force microscopy) was one of the main techniques used to characterize the morphology and phase separations in thin polymeric films. This measuring technique however relies only on mechanical contrast, leaving research on similar polymers out of reach. The combination of a confocal Raman microscope with an AFM one decade ago provided the ability to unequivocally determine the chemical composition of a material. By acquiring Raman spectra at every image pixel, the high spatial and topographic resolution obtained with an AFM can be directly linked to the chemical information provided by confocal Raman spectroscopy [1,2]. In polymer science, Raman spectra provide quantitative information about various features such as: chemical nature (structural units, type and degree of branching, end groups, additives), conformational order (physical arrangement of the polymer chain), state of the order (crystalline, mesomorphous, and amorphous phases), and orientation (type and degree of polymer chain and side group alignment in anisotropic materials). A milestone in microscopy, achieved last year, is the combination of SEM (scanning electron microscopy) with confocal Raman imaging. For polymer research an SEM provides the morphology and the crystalline structure of polymer composites together with the atomic composition mainly carbon and hydrogen. The combination with confocal Raman imaging allows the allocation of the unique chemical composition of the high resolution structures with diffraction limited resolution. The aims of this presentation are on the one hand to present results obtained with a confocal RamanAFM and a confocal Raman-SEM on the other hand, and highlight the benefits of such combined measuring capabilities. With the confocal Raman-AFM a three component polymer blend consisting of PS:EHA:SBR in the blending ratio 1:1:1 was analyzed. Figure 1 shows the high resolution topographic (a) and viscoelastic properties (b). The structures visible however are not present in the ratio of the blending. The Raman image (c) and the corresponding spectra (d) reveal an overlay of the EHA and SBR phase. Further information regarding substrate wetting and phase formation which could be derived by combining the results of both techniques will be presented. With the confocal Raman-SEM a polymer blend of PS:PMMA (blending ration 2:1) spin coated on a glass substrate was analyzed. Fig. 2 shows an overlay of a color coded Raman image consisting of PS:PMMA with the SEM image acquired from the same sample area. The SEM image reveals the fine structure of the two polymeric phases, whereas the chemical identification of the polymeric phases is provided by the confocal Raman image. References: [1] U. Schmidt, S. Hild, W. Ibach, O. Hollricher, Macromol. Symp. (2005), 230, p. 133 [2 ] U. Schmidt, J. M�ller, J. Koenen, in "Confocal Raman Microscopy", ed. T. Dieing, O. Hollricher and J. Toporski, (Springer Series in Optical Sciences 158, Berlin-Heidelberg 2010) p.237. Microsc. Microanal. 21 (Suppl 3), 2015 Microsc. Microanal. 21 (Suppl 3), 2015 2190 2178 Figure 1. Confocal Raman-AFM images of a three component polymer blend PS:EHA:SBR (1:1:1): AFM topography (a), AFM phase (b), color coded Raman image (c) and corresponding spectra (d). Figure 2. Confocal Raman-SEM imaging of a two component polymer blend PS:PMMA: SEM image (a), color coded Raman image overlaid on the SEM image (b), and corresponding spectra (c).");sQ1[1094]=new Array("../7337/2191.pdf","Dedicated X-Ray Mapping System with Single and Multiple SDD Detectors for Quantitative X-Ray Mapping and Data Processing","","2191 doi:10.1017/S1431927615011733 Paper No. 1094 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Dedicated X-Ray Mapping System with Single and Multiple SDD Detectors for Quantitative X-Ray Mapping and Data Processing Richard Wuhrer1 and Ken Moran2 1 2 University of Western Sydney, Advanced Materials Characterisation Facility (AMCF), Australia Moran Scientific Pty Ltd, 4850 Oallen Ford Road, Bungonia, NSW, 2580, Australia A JEOL 840 SEM/EDS system has been converted into a dedicated X-Ray Mapping (XRM) instrument, due to the needs of many users from varying disciplines including materials science and engineering, biology, geology and environmental science. The system performs 24 hours-7 days per week XRM electron microscopy (XRMEM). The system has just been upgraded with an Amptek 123 FAST123 silicon drift detector (SDD) with an Ultra Thin C1 window to allow for faster mapping. The XRMEM is setup to maximize its ability to operate with a larger working distance so that the instrument can perform x-ray maps at very low magnifications and high pixel resolution (1024x1024). This is now requested by a large number of users. With the use of the high count rate detectors, such as the Amptek SSD, our research team is collecting XRM's from over 3 hours and up to several days; this is especially true when trace elemental analysis and mapping is required. Quantitative X-ray mapping (QXRM) using multi-detectors and combined detectors, such as WDS/EDS, not only reduces mapping time, but also improves the ability to map minor and trace elements very accurately. One interesting outcome of this is that we now need to improve the collection efficiency of our WDS detectors to enable them to perform at these lower beam currents [1-3]. To take advantage of this increased precision and sensitivity of QXRM, careful attention to all those parameters necessary for high quality analysis is required. Important characteristics such as resolution, gain zero calibration, pulse pileup and other artifacts need to be well characterized for each detector used. Images are then created from the fully background corrected, overlap corrected and inter-element corrected calculation at each pixel in the image. As there is a very large dynamic concentration range to be considered (5 orders of magnitude) it is important to understand that the precision at each level will depend on how well this characterization is carried out [1]. To achieve this, time is no longer an issue but a means to accomplish a more precise result. Thus, if we can introduce more efficient detection then we can achieve this result in a shorter time. This may be achieved by counting more x-rays with one detector, as long as the critical quantitative characteristics of the detector is not compromised, or this can easily be achieved by adding more detectors. For QXRM to work correctly for multi-detector systems, the characteristics of each detector must be accurately determined so that the final quantification of the individual detectors can be summed. To accomplish this more effectively, the full spectrum at each pixel for each SDD detector should be saved. As a final check for consistency between detectors, a technique was developed called "Colouring Verification Technique (CVT)" that involves assigning a different RGB colour for each detector for the same element (Fig. 1). When combining the three maps of the same element, a grey scale map should be obtained (Fig. 1), not a colour map. This indicates a total correlation between the three detectors at the most critical final stage of quantification. Thus, the use of multiple detectors is a good way to determine which errors may be inherent in the specimen, the mathematical processing, and the detector systems. The use of this colouring technique for multiple detectors has also been found to be very valuable when analysing rough samples. We have further developed this colouring technique to show differences between raw intensity maps and quantification maps (Quantification Performance Test-QPT) [3]. In Fig. 2, the chromium Region Microsc. Microanal. 21 (Suppl 3), 2015 2192 of Interest (ROI) has been made red, the stripped intensity map is green and the quantitative map is blue. We can clearly see colouring around the boundaries of the tungsten carbide particles with the matrix, indicating that the results from the three processes are different. This is a classic example of why we need to quantify the results, for the difference between the intensity profile and the quantification profile is caused by a large atomic number correction on the chromium by tungsten (Z correction). We have confirmed this by plotting an intensity profile and the quantitative profile across this region (Fig. 2). Fig. 1: Quantitative chromium X-ray maps taken with detector 1 (red), detector 2 (green), detector 3 (blue), RGB images detector 1, 2 and 3 combined and pseudo colour map from the summed quantitative images of the three detectors, showing nickel as blue [1]. HWOF=100�m. Fig. 2: A different RGB colour is assigned to the combined sum of all detectors for an individual element. The RGB image shows a grey scale map, slightly coloured, indicating miss- correlation between the three elements. Also shown is the pseudo image for the three elements present (Ni, Cu, Cr) [3-4]. HWOF=110�m. Our research aims at developing post processing techniques to improve the quantitation of X-ray map data and to develop further post processing techniques for improved characterisation as an aid in assessing the practical properties of complex materials [4]. This also includes developing techniques of handling X-ray mapping data collected from multiple X-ray detectors around the column. References [1] K. Moran and R. Wuhrer, "Quantitative Bulk and Trace Element X-Ray Mapping Using Multiple Detectors", Mikrochimica Acta, Vol. 155, pp. 59-66 (2006). [2] K. Moran and R. Wuhrer, "X-ray Mapping and Interpretation of Scatter Diagrams", Mikrochimica Vol. 155, pp. 209-217 (2006). [3] R. Wuhrer, K. Moran and M. R. Phillips, "Multi-Detector X-Ray Mapping and Generation of Correction Factor Images for Problem Solving", Microscopy and Microanalysis, 14(suppl 2), 1108CD-1109CD (2008). [4] R. Wuhrer, K. Moran and M. R. Phillips,"X-Ray Mapping and Post Processing", Microscopy and Microanalysis, 12(suppl 2), 1404CD-1405CD (2006).");sQ1[1095]=new Array("../7337/2193.pdf","Quantitative Elemental Mapping with Electron Microprobe and Automated Data Analysis","","2193 doi:10.1017/S1431927615011745 Paper No. 1095 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantitative Elemental Mapping with Electron Microprobe and Automated Data Analysis Julie Chouinard1 and John Donovan1 1. University of Oregon, CAMCOR, Eugene, OR, USA While the utility of quantitative chemical mapping with electron microprobe (EPMA) can be overshadowed by the instrument time required to obtain such data, advancements in data acquisition and background corrections have made it a more practicable technique. The use of mean atomic number (MAN) background corrections can drastically shorten total run time by allowing the user to omit offpeak elemental maps by modeling the background for those elements instead [1]. Typically MAN corrections are limited to major elements, however, for simple matrices, a blank correction can be used in combination with MAN background to map trace elements, attaining similar accuracy to off-peak measurements while improving precision [2]. The advantages of mapping samples over doing single spot analyses are obvious: small scale chemical variations can be elucidated and differentiated from large scale features, for instance those seen in mineral grains with complex growth patterns; chemical information for phases proximate to those of interest can be later employed for determining the paragenesis of a rock sample or explaining away an apparent diffusion curve really attributable to secondary fluorescence; average compositions can be ascertained for materials that are heterogeneous at the micro- to millimeter scale; and countless other ways. The use of customizable and automated scripts can greatly aid in quickly extracting useful and often supplemental information from of these quantitative maps; these can be produced with programming languages and software packages such as Golden Software's Surfer� (modified Visual Basic), R, and MathWorks' MATLAB�. Several scripts have been made using the Scriptor application in Surfer� to automate and increase the speed of the analysis of mapped chemical data. To determine the composition over a diffusion boundary with a heterogeneous matrix, a script was crafted to average horizontal or vertical strips across a map [3]. Another was created to calculate the average composition of any delineated area, which can be used for defining the chemistry of a particular phase or reaction rim (Fig. 1). Scripts have also been generated to plot out the chemical changes along any traverse chosen on the map, such as [4]. Beyond chemical maps, programmable scripts have been developed to more efficiently process and examine spot analyses as well and several of these will be presented as well. The addition of acquisition and analysis features already available for single spot analyses will further improve EPMA mapping capabilities, decrease run time, and ultimately increase its usage as it becomes more cost effective. The ability to acquire quantitative, background and matrix corrected, energy dispersive x-day (EDX) maps for major elements simultaneous with wavelength dispersive x-ray (WDX) maps for trace elements is most notably absent. While interferences are still an issue for some EDX analyses, for many major elements it can provide reliable data with satisfactory precision and accuracy; combining EDX and WDX is already relatively commonplace for single point analyses as it often does not add any time to the run because the data can be compiled concurrently. Another acquisition option that would enhance mapping capabilities is the introduction of the time dependent intensity element corrections, without it elements like sodium (Na) that volatize and migrate in some matrices cannot be reliably mapped (Fig. 2). Microsc. Microanal. 21 (Suppl 3), 2015 2194 References: [1] J Donovan and T Tingle, Journal of Microscopy and Microanalysis 2 (1996) p. 1. [2] J Donovan and J Armstrong, Microscopy and Microanalysis 20 S3 (2014) p. 724. [3] J Barkman et al, Microscopy and Microanalysis 19 S2 (2013) p. 848. [4] J Chouinard et al, Microscopy and Microanalysis 20 S3 (2014) p. 750. Figure 1. Quantitative map of manganese (Mn) in a garnet; the average weight percent of Mn in rim of garnet (outer edge outlined in white) is 5.19%, compared to 4.22% in the core (bottom right corner outlined in white). Figure 2. Map of the Na concentration in a melt inclusion hosted by olivine [4]. The pattern is believed to be the result of Na migration in the glass and not a real chemical gradient (it was not seen in any other element analyzed); the integration of TDI into mapping may help resolve this issue.");sQ1[1096]=new Array("../7337/2195.pdf","Homogeneity Testing of Microanalytical Reference Materials by Electron Probe Microanalysis","","2195 doi:10.1017/S1431927615011757 Paper No. 1096 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Homogeneity Testing of Microanalytical Reference Materials by Electron Probe Microanalysis Dennis Harries1 1. Department of Analytical Mineralogy, Institute of Geosciences, Friedrich Schiller University, Jena, Germany. Well-characterized and suitable reference materials are key to accurate and reproducible chemical and isotopic microanalysis. The testing of homogeneity of potential microanalytical reference materials requires discriminating between the variations due to the instrumental precision and the variations due to true compositional differences. The theoretically easiest approach is to determine the instrumental precision by repeatedly analyzing the same spot on the sample and then compare different sites on the sample by an analysis of variance (ANOVA). However, in practical microanalysis this approach is expected to fail often, because repeated analysis of a single spot will often result in progressive degradation (or even loss) of the analyzed volume. In electron probe microanalysis (EPMA) this degradation is due to the build-up of surface contamination, induced diffusion, and structural damage. In consequence, the apparent instrumental variance is not applicable to sound statistical analysis. In principle, perfect homogeneity in materials does not exist. It has to be defined what scale and relative contribution of heterogeneity can be accepted for a given purpose. Testing for homogeneity then tries to answer the question whether a material is `fit for the purpose'. This question can be answered positively when significant heterogeneity cannot be detected with the method and its chosen conditions of operation. These are specifically the spatial and the `compositional' resolution, the latter determined by the instrumental precision and the resulting uncertainty. In the case that only one measurement per analysis spot is feasible, significance criteria for detectable heterogeneity can be derived in a statistically sound fashion if Poisson statistics of X-ray counts are the only significant contribution to the instrumental precision (and this precision is reasonably high). A traditional approach to this is the recently refined Homogeneity Index (H) [1, 2]. The Homogeneity Index compares the variance expected from counting statistics to the variance observed. If the latter is larger than the expected variance (H > 1), compositional heterogeneity may be significant. This can be tested based on F or chi-squared statistics. The answer is valid only for the specific set of measurement conditions that were used for the test. The total uncertainty budget of a reference value is, in the simplest (accurate) case, composed of uncertainty due to instrumental precision and compositional heterogeneity. The Homogeneity Index can be used to derive a relative contribution of compositional heterogeneity (sh,rel) to this total uncertainty budget. However, the approach is limited by the non-linear relationship between H and sh,rel (Fig. 1), resulting in a bias that yields apparently large contributions of compositional heterogeneity even if materials are almost perfectly homogeneous. Because the uncertainty of the Homogeneity Index is primarily determined by the number of measurements N, this bias and the danger of statistical type II error can be reduced by increasing N (Fig. 1). In consequence, the resolving power of a homogeneity test in terms of detectable heterogeneity is strongly related to the number of measurements. For example, in order to state that the contribution of compositional heterogeneity to the total uncertainty budget is less than 30%, a homogeneity test with N = 577 measurements has to be passed (H < 1.048). Microsc. Microanal. 21 (Suppl 3), 2015 2196 For 20% this number increases to more than 3000 measurements. What upper levels of heterogeneity contribution and analytical expenses are acceptable has to be decided by the user. The criteria discussed are intended for testing for homogeneity in terms of compositional resolution. These criteria are not suitable to assess homogeneity in terms spatial resolution. For example, it is not possible to address how the perception of homogeneity changes when the spot size or the average spacing between spots changes. This important question needs to be addresses by other means. References: [1] F.R. Boyd et al., Carnegie Institution of Washington Year Book 67 (1967), p. 210. [2] D. Harries, Chemie der Erde 74 (2014), p. 375 (doi: 10.1016/j.chemer.2014.01.001). Figure 1. The important role of the number of measurements N in homogeneity testing. Shown are histograms of 10,000 simulated homogeneity tests with N = 10 (a, b) and N = 100 (c, d) measurements per test. The (simulated and therefore known) expected relative contribution of heterogeneity E(sh,rel) to total uncertain is 0 (a, c; case of perfect homogeneity) and 80% (b, d; case of detectable heterogeneity).");sQ1[1097]=new Array("../7337/2197.pdf","Towards Statistically Representative Atomic Resolution 3D Nano-metrology for Materials Modelling and Catalyst Design.","","2197 doi:10.1017/S1431927615011769 Paper No. 1097 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Towards Statistically Representative Atomic Resolution 3D Nano-metrology for Materials Modelling and Catalyst Design. Lewys Jones1, Katherine E. MacArthur1, Jolyon Aarons2, Chris-Kriton Skylaris2, Misbah Sarwar3, Dogan Ozkaya3 and Peter D. Nellist1 1. 2. Department of Materials, University of Oxford, Oxford UK. School of Chemistry, University of Southampton, Southampton, UK. 3. Johnson Matthey, Reading, UK. Metallic nanoparticles are used widely in a variety of catalyst applications and further improvement in their performance requires materials characterisation at the nano-scale. The aberration corrected scanning transmission electron microscope (STEM) now allows for the routine imaging of metallic nanoparticles at atomic resolution. Especially useful is the annular dark-field (ADF) imaging mode where image contrast is related to sample mass or atomic number. It has been shown that, for pure metals, this approach allows for the number of atoms in atomic-columns to be counted [1] and for the 3D structure of nanoparticles to be revealed [2]. Recently we presented a method for the highthroughput automated analysis of ADF STEM data to produce 3D models of nanoparticles at atomic resolution [3]. By utilising prior knowledge about the structures this allows information such as facet types and surface-atom coordination numbers to be determined on a particle by particle basis (Figure 1) without the need for the multiple frame imaging of tomography. However, in many electron microscopy studies, few unique nanostructures are presented and this may lead to unintended observational bias. The next logical step then is to survey a larger number of particles and to move towards more general measurements of `average' particle properties. Here we present our progress towards this goal; the as received commercial catalyst samples were imaged at intermediate magnifications for particle-size analysis (~1-5Mx), Figure 1. Using this histogram, individual on-axis particles were chosen for high-magnification imaging (~10-20Mx), atomcounting, and model construction, Figure 2. As this approach yields the total number of atoms per particle these can be directly related back to the sizing histogram. Now we have a collection of 3D atomic models which we believe better represents the overall sample. None of these were so-called `perfect number' polyhedral and all contain many varied surface atom configurations with a variety of surface nearest-neighbour coordinations and facet types [3]. Next to explore the range of properties of these particles computer modelling is used. First our preliminary atomic model is relaxed at room temperature using Sutton-Chen potential. This relaxation is justified as it accommodates for any mis-measurement in the experimental atom-column positions and any beam-induced heating. During this step we observe little or no structural reconfiguration but often relaxation of bond lengths of a few percent. With the relaxed coordinates, the electronic properties of the particle can be calculated using density-functional theory (DFT). The ONETEP code for large-scale DFT calculations was used to obtain the electronic density and electrostatic potential of the particles (Figure 2). Using experimentally determined 3D models as inputs for materials modelling allows the inhomogeneous surface electronic properties of catalytic nanoparticles to be explored. Relating the nano-scale surface texture, electronic properties and chemical reactivity of metallic nanoparticle Microsc. Microanal. 21 (Suppl 3), 2015 2198 catalysts will become key information in designing new and improved systems [4]. References: [1] J. M. Lebeau and S. Stemmer, Ultramicroscopy 108 (2008), p. 1653�8. [2] S. Van Aert et at., Nature 470 (2011), p. 374�7. [3] L. Jones et al., Nano Lett. 14 (2014), p. 6336�41. [4] The authors acknowledge funding from EU FP7 grant 312483 (ESTEEM2), EPSRC grant EP/K040375/1 and Johnson Matthey, and S. VanAert and A. DeBacker for their useful discussions. Figure 1. Size-distribution histogram of a commercial batch of Pt nanoparticles as measured by intermediate magnification ADF STEM (450 particles total). High-resolution imaging from the same sample was used to rebuild atomic 3D models shown above the histogram. Numbered models indicate the number of atoms in each particle. Figure 2. Workflow of the combined experimental-computational nanoparticle analysis; a) the ADF STEM image recorded with a calibrated detector and used for the column-wise atom-counting, b) the rebuilt structural model from those atom counts (total atoms = 943, atom colour indicates coordination number [3]), c) the electronic density iso-surface plotted at 0.1e-/�3, and d) the surface in c) but coloured using with the values of the electrostatic potential. View d) reveals points of high potential where strong chemical affinity may be expected.");sQ1[1098]=new Array("../7337/2199.pdf","Probing Complex Nanostructures by Combining Atomic-Scale Theory and Scanning Transmission Electron Microscopy","","2199 doi:10.1017/S1431927615011770 Paper No. 1098 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Probing Complex Nanostructures by Combining Atomic-Scale Theory and Scanning Transmission Electron Microscopy Sokrates T. Pantelides1,2 and Stephen J. Pennycook3 1. 2. Department of Physics and Astronomy, Vanderbilt University, Nashville, TN 37235 USA Materials Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831 USA 3. Department of Materials Science and Engineering, National University of Singapore, Singapore 117576 Aberration-corrected scanning transmission electron microscopy has reached extraordinary levels of spatial and energy resolution, in both imaging and electron-energy-loss spectroscopy (EELS). In parallel, calculations based on density functional theory (DFT) using high-performance computers have made enormous strides in describing the atomic-scale properties of complex materials. A combination of theory and microscopy is a powerful approach to resolve long-standing issues and unveil new phenomena. This talk focuses on select recent applications of the combined approach. The first example is the elucidation of the origin of white light emission by ultrasmall CdSe nanoparticles [1]. Proposed explanations invoked surface and defect states, but could not account for the observed continuous spectrum. Atomic-resolution Z-contrast images found that nanoparticles larger than 2 nm in diameter have a solid crystalline core, but sub-2-nm particles are fluxional under the electron beam. Noting that, under uv excitation, energy is transferred to vibrations before a band-gap photon is emitted, quantum dynamical simulations at the corresponding temperature found that a sub-2-nm particle is similarly fluxional. These fluctuations cause the band gaps to vary continuously across the visual range on a femtosecond time scale (Fig. 1c), leading to white light (Fig. 1b). The second example is the resolution of the long-standing puzzle of the origin of ferromagnetism in insulating LaCoO3 thin films grown epitaxially under strain [2]. The insulating nature of the films led to the assumption that the films are stoichiometric (oxygen vacancies would make the material n-type). Several attempts using first-principles calculations could not reproduce the experimental data. Columnresolved EELS (Fig. 2) established that observed dark stripes in Z-contrast images and EELS maps are caused by ordered vacancies. DFT calculations found that such structures are indeed ferromagnetic. Finally, it was found that just insertion of the oxygen vacancies without relaxation introduces only Peierls-like minigaps within the metallic band structure. Relaxation of the structure, however, leads to a rupture of the energy bands and the appearance of a large energy gap, as observed (Fig. 2). The third example is the discovery of a new crystalline order in CuInS2 nanoparticles, called interlaced crystals [3]. In CuInS2, Cu and In atoms share one of two sublattices, but XRD cannot resolve whether the cations are ordered. Valence-balancing electron counting suggests that Cu and In cations should be ordered. DFT calculations indeed found that several ordering possibilities induce no strain at all and have essentially identical energies. Z-contrast images found that the different ordering possibilities coexist on a totally undisturbed perfect Bravais lattice (Fig. 3). Calculations found that domain and phase boundaries, which are not uniquely assigned, cost no energy. These are the features of what has been labeled as interlaced crystalline order. Such materials should be ideal for thermoelectric applications as the boundaries have negligible effect on electronic conductivity but reduced thermal conductivity. Microsc. Microanal. 21 (Suppl 3), 2015 2200 Finally, a new scheme that combines DFT with diffraction theory to simulate probe-position-dependent EELS will be described briefly [4]. Figure 1. (a-b) Experimental and simulated white light spectrum of sub-2-nm CdSe nanoparticles shown in the insets (c) Fluctuating band gap in the density of states produces the continuous spectrum. From Ref. [1]. Figure 2. Left: La M5 EELS map and O K EELS (shaded in green), showing O depletion in the dark stripes. Right: LCO spin-up bands after ordered vacancies are inserted, and after relaxation. Ref. [2] References: [1] T. J. Pennycook, J. R. McBride, S. J. Rosenthal, S. J. Pennycook, and S. T. Pantelides, "Dynamic fluctuations in ultrasmall nanocrystals induce white light emissions", Nano Lett. 12, 3038-3042 (2012). [2] N. Biskup, J. Salafranca, V. Mehta, M. P. Oxley, Y. Suzuki, S. J. Pennycook, S. T. Pantelides, and M. Figure 3. Z-contrast image of CuInS2 Varela, "Insulating Ferromagnetic LaCoO3- films: A nanoparticle and the same image using the phase induced by ordering of oxygen vacancies", same color for both Cu and In atoms, Phys. Rev. Lett. 112, 087202 (2014). revealing a perfect Bravais lattice. Ref. [3]. [3] X. Shen, E. A. Hern�ndez-Pagan, W. Zhou, Y. S. Puzyrev, J.-C. Idrobo, J. E. Macdonald, S. J. Pennycook, and S. T. Pantelides, "Interlaced crystals having a perfect Bravais lattice and complex chemical order revealed by real-space crystallography", Nature Commun. 5, 5431 (2014). [4] Collaborators in this work are identified in the cited references. Research at Vanderbilt was supported by Department of Energy grant DE-FG02-09ER46554. Microscopy was carried out at Oak Ridge National Laboratory, supported by the Department of Energy, Office of Science, Basic Energy Sciences, Materials Science and Engineering Division. Computations were carried out at the National Energy Research Scientific Computing Center, also supported by the Department of Energy.");sQ1[1099]=new Array("../7337/2201.pdf","Integrated Computational and Experimental Structure Determination for Nanoparticles","","2201 doi:10.1017/S1431927615011782 Paper No. 1099 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Integrated Computational Nanoparticles and Experimental Structure Determination for Min Yu1, Andrew B. Yankovich1, Amy Kaczmarowski1, Dane Morgan1 and Paul M. Voyles1 1. Department of Materials Science and Engineering, University of Wisconsin-Madison, Madison, Wisconsin 53706, USA. Atomic details of nanostructures are important to materials performance for catalysis, solar energy, optoelectronics, sensing and many other fields. However, solving the three-dimensional (3D) structure of nano-scale materials at the atomic level is challenging, especially for predicting metastable, out-ofequilibrium systems. Scanning transmission electron microscopy (STEM) provides structural images of materials at atomic resolution, but a single image provides only a two-dimensional (2D) projection of the structure, and three-dimensional tomographic imaging at atomic resolution with single-atom sensitivity remains challenging. Experimentally driven structural refinement approaches typically rely on minimizing the error between forward simulation from atomic models and the experiment data. Such optimizations are difficult with limited data and rely on knowing good initial guesses for the structure. They also typically make no direct use of information about the energy of the potential structures. Purely computational techniques, such as genetic algorithms (GAs) [1], have proven to be extremely effective at predicting the ground state structures of a wide range of complex structures, including clusters, crystals, and grain boundaries. However metastable configurations are often neglected by methods designed to find the global minimum of the system energy. We have developed an integrated GA optimization tool [2] that can reverse engineer the 3D structure of a nanoparticle by matching forward modeling to experimental STEM data [3,4] and simultaneously minimizing the system energy. This tool integrates the power of GAs for complex optimization and can find metastable structures guided by experimental data. We validated the algorithm by reproducing the structure and orientation of stable and metastable 309-atom Au nanoclusters using simulated STEM images with better than 0.5 pm fidelity for every atom. Then we determined the 3D structure of a ~6000 atom Au nanoparticle on amorphous carbon substrate based on a high precision STEM image [5]. STEM experiments were conducted on a FEI Titan microscope equipped with a CEOS probe aberration corrector. The experimental high angle annular dark field STEM image in Figure 1a shows a [110] oriented, ~8nm diameter Au nanoparticle having 2 distinct grains separated by a (112) twin boundary. In order to achieve better image signal to noise ratio and remove image distortions caused by instabilities in the probe and sample during image acquisition, we acquire a series of STEM images and utilize the non-rigid (NR) registration technique [4]. Inside the optimization, we simulate the STEM image using a computationally efficient convolution method with a non-linear scaling factor to counteract the over-estimation of intensities in thick sample, instead of frozen-phonon multislice simulations [3]. Based on the STEM image, we started GA optimization from a family of single crystals along Au (110). After 4000 generations, we find a structure of 4998 atoms shown in Figure 1b, which shows the same ellipsoid shape, twin boundary, and surface facets as experiment. The atomic column positions in the experimental and simulated STEM images differ by 0.18 � on average (Figure 1c). The fitness function decays monotonically through the GA evolution, while the energy term fluctuates (Figure 2). Overall, the integrated GA tool successfully optimizes the metastable Au nanoparticle atomic Microsc. Microanal. 21 (Suppl 3), 2015 2202 positions to match experimental STEM data while simultaneously achieving a locally stable structure. [6] References: [1] R. L. Johnston, Dalton Transactions (2003), p. 4193. [2] A. Kaczmarowski et al., Comput. Mater. Sci. 98 (2015), p. 234. [3] E. J. Kirkland, "Advanced Computing in Electron Microscopy" (Plenum Press, New York) (1998). [4] A. B. Yankovich et al., Nat. Commun. 5 (2014), p. 4155. [5] A. B. Yankovich, B. Berkels, W. Dahmen, P. Binev, P. M. Voyles (to be published) [6] The authors acknowledge funding from the National Science Foundation, Grant Number DMR1332851. Figure 1. Experimentally measured STEM image of Au nanoparticle (left); Simulated STEM image of the GA optimized structure (middle); Comparison of atomic column displacements between experiment and simulation (right). Figure 2. The energy, discrepancy between simulations and experiments, and total cost function over the course of the GA optimization.");sQ1[1100]=new Array("../7337/2203.pdf","Local Crystallography: Phases, Symmetries, and Defects from Bottom Up","","2203 doi:10.1017/S1431927615011794 Paper No. 1100 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Local Crystallography: Phases, Symmetries, and Defects from Bottom Up Alex Belianinov,1,2, Qian He,3 Mikhail Kravchenko,1,2 Stephen Jesse,1,2 Albina Borisevich,1,3Sergei V. Kalinin1,2 1 Institute for Functional Imaging of Materials, Oak Ridge National Laboratory, Oak Ridge, TN 37831. 2 The Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, TN 37831. 3 Materials Sciences and Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831. Advances in high resolution imaging and scanning transmission electron microscopies as well as probe-based microscopies, extend beyond simple visualization of atomic structure of matter by allowing high (10 pm or better) precision measurements of atomic positions. Such a high level of fidelity is sufficient to correlate the length and hence energy of individual bonds, as well as bond angles to functional properties of materials. We introduce an algorithm for local analysis of material structure based on the statistical analysis of individualized, local atomic neighborhoods. Our clustering analysis explores the connectivity of lattice, allowing for efficient chemical description and identification of phases and structural defects in multiphase samples. This analysis lays the framework for building image genomes and structure-property libraries, based on conjoining structural and spectral realms through local atomic behavior. Algorithms like k-means clustering on multi-phase regions allow identification of minute structural distortions, providing framework for description of local physical functionalities. As a model system, we have chosen mixed oxide Mo-V-M-O (M = Nb, Ta, Te and/or Sb), which is currently the most promising catalyst for many industrially important reactions, such as propane (amm)oxidation.[1] Our algorithm yields absolute positions of each atom in an image, as well as local descriptors such as column intensity and peak width determined at the refinement stage. Significant improvements in atom finding based on target defined search are made over the traditional peak fitting methods. [2, 3] Atom tilts as well as correction approaches based on the recovered image data are explored. [4].The number of nearest neighbors, or the search radius, can be defined separately and are chosen depending on the type analysis. In the simplest case neighbors are chosen based on dominate symmetry, e.g. 6 for hexagonal lattice, or 4/8 for cubic lattice. Once the local neighborhood is established, the clustering algorithm is utilized with an appropriate selection metric of interest to sort the local neighborhoods. Shown in Figure 1(a) is a raw image of Mo-V-M-O mixed MA and M2 phases. We have specified 6 neighbors as well as 6 clusters; the results of a distance to neighbor based metric are shown in Figure 1(b) and a 4 cluster angle to neighbor based metric in Figure 1(c). Rotation is accounted for by always placing the first neighbor atom in the same location relative to the center and filling the rest in a clockwise fashion. The analysis clearly distinguishes different areas of the image based on the similarity of chemical neighborhoods of their constituent atoms, whose role and effects are discussed in detail. Figure 1(d) is a FFT of the overall image shown in Fig. 1(a), panels (e) and (f) show average atomic neighborhoods for 6 and 50 neighbors respectively. Figure 1(g) is a clustering dendrogram for the distance to neighbor case used to define the final number of clusters. We believe this approach paves the way for full information recovery in high resolution imaging as well as allows classification for automatic identification of materials.[5] Microsc. Microanal. 21 (Suppl 3), 2015 2204 Figure 1. Two phase Mo-V-M-O (M = Nb, Ta, Te and/or Sb). (a) M1 and M2 mixed phase STEM image. (b) k-means clustering results for 6 neighbor, sorted by distance metric. (c) kmeans clustering results for 6 neighbor, sorted by angle metric. (d) FFT of image in (a). (e) 50 member neighborhood of the image in (a). (f) 6 member neighborhood of the image in (a). (g) Dendrogram for the 6 neighbor, sorted by distance metric. References [1] Shiju, N. R. & Guliants, V. V. Recent developments in catalysis using nanostructured materials. Applied Catalysis A: General 356, 1-17, doi:10.1016/j.apcata.2008.11.034 (2009). [2] Xiahan Sang, Adedapo A. Oni, and James M. LeBeau, Atom Column Indexing: Atomic Resolution Image Analysis Through a Matrix Representation, Microsc. Microanal., doi:10.1017/S1431927614013506 [3] Sarahan M.C., Chi M., Masiel D.J. and Browning N.D; Point defect characterization in HAADF-STEM images using multivariate statistical analysis, Ultramicroscopy, 111 (3) , pp. 251-257; 2011 [4] Jones L., Nellist P.D., Identifying and correcting scan noise and drift in the scanning transmission electron microscope (2013) Microscopy and Microanalysis, 19 (4) , pp. 1050-1060. [5] Research for all authors was supported by the US Department of Energy, Basic Energy Sciences, Materials Sciences and Engineering Division. This research was conducted at the Center for Nanophase Materials Sciences, which is sponsored at Oak Ridge National Laboratory by the Scientific User Facilities Division, Office of Basic Energy Sciences, U. S. Department of Energy.");sQ1[1101]=new Array("../7337/2205.pdf","HAADF/MAADF Observations and Image Simulations of Dislocation Core Structures in a High Entropy Alloy","","2205 doi:10.1017/S1431927615011800 Paper No. 1101 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 HAADF/MAADF Observations and Image Simulations of Dislocation Core Structures in a High Entropy Alloy T.M. Smith1, B.D. Esser1, E.P. George2, F. Otto2, M. Ghazisaeidi1, D.W. McComb1, M.J. Mills1 1. 2. Center for Electron Microscopy and Analysis, The Ohio State University, Columbus, OH Institute for Materials, Ruhr University, Bochum, Germany High entropy alloys (HEAs) are a new class of multi-component alloys in which the individual elements have similar concentrations. A single-phase solid solution HEA containing 5 elements (Co, Cr, Fe, Mn, and Ni) with equiatomic composition was first discovered by Cantor [1]. Among the surprising characteristics of this fcc HEA are: strong temperature dependence of the yield strength at temperatures around and below room temperature, relatively weak strain-rate dependence over the same temperature range [3]; very large hardening rates [2,3]; and large fracture toughness at room temperature [4]. These features are linked to deformation twinning and dislocation-mediated plasticity, yet presently there is insufficient knowledge of dislocation dissociation, stacking fault energy, or core structures in this alloy. The highly planar deformation involves dislocation arrays on active slip systems (Figure 1a and 1b). This characteristic could imply the presence of short range order, low fault energy, or supplementary displacements in the wake of glide dislocations. In the present study, an HEA sample, with the same nominal composition as above, was deformed to 5% plastic strain at room temperature. Post-mortem electro-polished 3mm disks were extracted and analyzed using a probe-corrected Titan G3 80-300kV along a [110] Zone axis. Figure 2a shows a high angle annular dark field (HAADF) image of a �<110> 60� mixed dislocation obtained at a short camera length (93 mm). A notable observation is a ~3 nm wide stacking fault, suggesting a modest stacking fault energy (SFE) in the alloy. Subsequent observations revealed a range in stacking fault lengths (28nm), possibly due to local chemical fluctuations affecting the SFE throughout the solid solution. Conversely, the edge dislocation shown in Figure 2c is a �[1 -10] Lomer dislocation, exhibiting an extremely compact structure, with no noticeable dissociation into a Lomer-Cottrell configuration. This dislocation is inconsistent with the 60� dislocation with respect to apparent stacking fault energy. The medium angle annular dark field (MAADF) images in Figure 2b and 2d, taken at longer camera lengths (145-230mm), contain stronger diffraction effects [5]. Zones of enhanced intensity ("plumes") can be observed in the MAADF, but not the HAADF images, implying the intensity is not a consequence of elemental segregation around the dislocation cores. Significant variations in the shape, size, and location of the "plumes" were observed around a dislocation core. Understanding the cause of these "plumes" can give insights into the structure of the dislocation cores through the thickness of the sample, and stimulate new ideas about how dislocations influence the surrounding lattice as they glide. The fact that the "plumes" are not observed in HAADF but appear and become brighter at increasingly longer camera lengths reveals that the contrast is either a natural consequence of: (a) the deviation from a perfect lattice near the dislocation core, (b) a slight misalignment either due to overall rotation of the dislocation in the thin foil or variation in the local separation distance between partials, or (c) a variation in the dissociation through the depth in the sample [6]. Atomistic simulations on simple systems were used to test the above possibilities systematically. Cells containing the dislocation cores corresponding to (a)-(c) were created. HAADF and MAADF images of these cells were then simulated using the �STEM multislice code [7] along with zone axis STEM simulations using CTEM soft[8]. The findings Microsc. Microanal. 21 (Suppl 3), 2015 2206 from these simulations combined with through-focus acquisitions to observe imaging effects through thickness will be discussed in detail. References: [1] B. Cantor et al. Mat Sci Eng a-Struct, 375 (2004) 213-218. [2] F. Otto et al. Acta Materialia, 61 (2013) 5743-5755 [3] A. Gali, E.P. George. Intermetallics, 39 (2013) 74-78. [4] B. Gludovatz et al. Science, 345 (2014) 1153-1158. [5] P.J. Phillips et al. Ultramicroscopy, 116 (2012) 47-55. [6] J.G. Lozano et al. Physical Review Letters, 113 (2014). 135103 [7] B.D. Forbes et al. Physical Review B, 82 (2010) 104103. [8] P.J. Phillips, M.J. Mills, M. De Graef. Philosophical Magazine, 91 (2011) 2081-2101. [9] Support for this work was provided by the US Department of Energy, Office of Basic Energy Sciences under Grant #DE-SC0001258 and the Center for Emergent Materials: an NSF MRSEC under award number DMR-1420451. This work was also supported in part by an allocation of computing time from the Ohio Supercomputer Center. Figure 1: (a) <112> Zone Axis BF (S)TEM image of dislocation arrays in a HEA (b) <111> Zone Axis BF (S)TEM image of dislocation arrays in a HEA. Figure 2: STEM images of dislocations in the HEA at two different camera lengths (HAADF vs. MAADF conditions). (a) and (b) are the same �<110>{111} 60� dislocation; (c) and (d) are the same �<110>{001} Lomer dislocation. Origin of the "plumes" in the MAADF images is presently unknown.");sQ1[1102]=new Array("../7337/2207.pdf","Strain-Mediated Asymmetric Growth of Plasmonic Nanocrystals: A Monometallic Au Nanorod-Au Nanoparticle Heterodimer.","","2207 doi:10.1017/S1431927615011812 Paper No. 1102 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Strain-Mediated Asymmetric Growth of Plasmonic Nanocrystals: A Monometallic Au Nanorod-Au Nanoparticle Heterodimer. Yihan Zhu1, Jianfeng Huang1, Yu Han1 1. Advanced Membranes and Porous Materials Center, Physical Sciences and Engineering Division, King Abdullah University of Science and Technology (KAUST), Thuwal 23955-6900, Kingdom of Saudi Arabia Colloidal noble metal nanocrystals (NCs) usually exhibit highly symmetric particle morphologies, as dictated by their intrinsic crystallographic symmetries. Breaking this morphological symmetry would bring vast variations to the plasmonic and optical properties of NCs, greatly enriching their applications in plasmonics, nanophotonics, sensing, and surface-enhanced Raman scattering (SERS). However, the synthesis of monometallic dimers via seed-mediated growth is difficult, because one of the main driving forces for dimerization is the lattice mismatch between two materials. When the growth material is the same as the seed, epitaxial growth on the entire surface of the seed particle is favored over the formation of a dimer.1 On the other hand, most reported dimers are constituted of two particles, whereas dimers with alternate configurations, for instance, a nanoparticle (NP) grown on a single-crystalline nanorod (NR) with a regular morphology, are yet to be explored. Given the high degree of symmetry of metallic structures, it is conceivable that growing a single NP on a highly regular NR would be difficult because it would require one site on the NR at which the NP is grown to be differentiated from many other symmetryequivalent sites. Here, we report that by employing specific strong ligands (thiols), an asymmetric AuNR-AuNP dimer can be synthesized from single crystalline AuNR seeds. Using electron microscopy and tomography, we reconstructed the three-dimensional morphology of the dimer crystal and identified that the newly grown AuNP is a multiple-twinned crystal preferentially residing at the "neck" region of the AuNR. The welldefined morphology of single crystalline AuNRs (with known surface facet indices) allows us to easily identify the formation of a dimer interface and to track the subsequent evolution of the AuNPs. The results provide important insights into the growth mechanism of this dimeric nanostructure, suggesting that the NP first nucleates at one (111) bridging facet of the NR, which further grows as a consequence of random twinning, and finally recrystallizes into a single multiple-twinned crystal. We found the key factors for successful synthesis include inhomogeneous surface strains on the AuNR and appropriate reduction kinetics, which are both associated with the use of ligands. A mechanism was proposed that describes the growth of AuNP in the dimer structure as follows. The chemisorption of ligands induces inhomogeneous strains on the surface of AuNR. Among different surface facets of the AuNR, the (111) bridging facets at the rod's "neck" are the most strained and are thus preferred locations for Au deposition to form a twinning structure via stacking faults that can largely relieve the surface strain.2 Once the first twin structure is formed on a (111) facet, this facet is differentiated from other symmetry-equivalent facets, becoming an "active" site for the further growth of Au through successive twinning, as driven by the same strain-relieving mechanism. New twins can form randomly in different <111> directions, resulting in various agglomerates of small grains. Worm-like structures are formed when most new grains grow laterally along the rod, while radiative twinning leads to cauliflower-like structures. At the final stage of synthesis, the agglomerates undergo a recrystallization process with small grains fused into large ones, accompanied by the disappearance of grain boundaries. As a consequence, Microsc. Microanal. 21 (Suppl 3), 2015 2208 different-shaped agglomerates gradually develop into single AuNPs with similar shapes. If the reaction is prolonged, the NPs have fewer grains and smoother surfaces. The key hypothesis of our proposed mechanism is that thiol ligands induce significant and inhomogeneous surface strain on the AuNR. In order to verify this hypothesis, we used HRTEM to characterize the assynthesized AuNR (with weaker ligand CTAB) and the AuNR incubated with thiol, and then we visualized their strain distributions using geometric phase analysis (GPA). Atomic-resolution HRTEM images were taken for both types of AuNR along the [110] axes, as the displacement of the densely packed (111) layers could be identified in this direction (Figures 1a, c). Interestingly, the GPA analysis showed that unlike the as-synthesized AuNR that exhibited little strain fluctuation over the entire crystal (Figure 1b), the thiol-incubated AuNR exhibited large shear deformation on the rod surface that was particularly localized on the (111) bridging facets in the "neck" regions (Figure 1d). Aligning the y-axis of the GPA map with the growth direction (i.e., the [001] axis) of the AuNR, the shear strain fields (xy) in left and right bridging facets have opposite signs, indicating the same type of shear deformation caused by gliding of the {111} planes (Figure 1d). The shear strain was determined to be from the <112>{111} slip. In the Bragg-filtered HRTEM image of the thiol-incubated AuNR, an array of rhombuses are delineated by connecting neighboring atomic columns (Figures 1e, f). The magnitudes of the strains in different areas can be intuitively identified from the obtuse angles () of the rhombus. Specifically, the angles measured at the side surface of the rod body (: 108.1�� 109.9� are very close to the value for a perfectly ABC-stacked ) bulk structure ( = 109.6�) (Figure 1f), whereas apparently smaller angles (: 102.5�� 104.3� are ) observed at the bridging facets (Figure 1e). These results clearly demonstrate that thiols can selectively induce large surface strain in the "neck" region of the AuNR, where the surface (111) planes undergo shear deformation along the [112] direction. References: [1] Sohn, K.; Kim, F.; Pradel, K. C.; Wu, J.; Peng, Y.; Zhou, F.; Huang, J. ACS Nano 3 (2009), 2191. [2] Paine, D. C.; Howard, D. J.; Stoffel, N. G. J. Electron. Mater. 20 (1991), 735. Figure 1. (a, c) Atomic-resolution HRTEM images of (a) the as-synthesized (CTAB capped) AuNR and (c) thiol-incubated AuNR taken along the [110] axes. (b, d) corresponding strain distributions of the shear component (xy, the magnitude cut-off is � 8%) determined by geometric phase analysis. (e, f) Enlarged Bragg-filtered HRTEM images of (f) region I and (e) region II, as marked in (c).");sQ1[1103]=new Array("../7337/2209.pdf","Using Electron Diffraction Techniques, CBED and N-PED to measure Strain with High Precision and High Spatial Resolution","","2209 doi:10.1017/S1431927615011824 Paper No. 1103 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Using Electron Diffraction Techniques, CBED and N-PED to measure Strain with High Precision and High Spatial Resolution J.L. Rouvi�re1-2, Y. Martin1-2, N. Bernier3, M. Vigouroux3, D. Cooper3, J.M. Zuo4 1. 2. Univ. Grenoble Alpes, INAC-SP2M, F-38000 Grenoble, France CEA, INAC-SP2M, F-38000 Grenoble, France 3. CEA, LETI, Minatec campus, 17 rue des Martyrs, 38054 Grenoble Cedex 9, France 4. Univ Illinois, Dept Mat Sci & Engn, 1304 W Green St, Urbana, IL 61801 USA Changes in the lattice parameters, i.e., introduction of strain, can modify material properties greatly. For instance, the band structure is modified with strain and this leads to changes in the transport or optical properties. In microelectronic devices, strain has been used to improve the mobility of charge carriers since 2003, with strain as low as 0.7% improving mobility by 50% [1]. Transmission Electron Microscopy (TEM) is presently the only technique that can measure the strain in individual nano-objects with high spatial resolution (about 1 nm) and high precision (about 10-4). Here we focus on recent developments we have made in two electron diffraction techniques: Off-axis Convergent Beam Electron Diffraction (CBED) and Nanobeam Precession Electron Diffraction (N-PED). Off-axis CBED can give 3D maps of the complete 3D strain tensor but it is computationally and experimentally demanding. In contrast, N-PED is a straightforward and precise technique, but it is limited to the projected 2D strain. In off-axis CBED the originality of our approach is to use both the deficient HOLZ lines of the transmitted beam and the excess HOLZ lines of the diffracted beams to measure the strain [2]. Using Bloch wave calculated CBED patterns as tests, we could retrieve 7 out of the 9 components of the deformation gradient tensor F. In particular, the volume of the cells can be determined (Fig. 1b). By using two different electron beam directions engendering an angle of 22�, we show that it is possible to determine the whole tensor F. In addition, the method can also be extended to the analysis of split HOLZ lines that allow measuring the variations of the strain tensor along the electron beam. Depending on the convergence angle (), the diffraction spots look like more dots or disks and the diameter of the incident beam varies (Fig. 1). When is about 2 mrad, the diffracted disks do not have a uniform intensity (Fig. 2f), which can complicate accurate determination of the position of the disks. It is why in Nanobeam Electron diffraction (NBED) is reduced in order to have spots with more uniform intensities (Fig. 2c) [3]. As these intensity variations are very sensitive to thickness, chemical composition and orientation, it turns out that the positioning of the disk centers is not very robust. This is why we introduced the precession of the beam [4]: the incident beam is tilted by an angle p and rotated around the original incident direction and a "descan" of the beam is applied after the sample in order to bring back the diffracted spot to their original positions (Fig. 1g-1h). The advantages of introducing precession are manifold: (i) uniformity in the disk intensity, (ii) presence of more diffracted spots, (iii) possibility to work with a relatively large convergence angle in order to reduce the beam diameter (Fig. 1j-1k); all this leads to an improved precision with a smaller probe [4]. Classically, PED is used in crystallography problems with p greater than 1�, in order to have diffracted intensities that approaches kinematical calculations. But in this condition, the probe diameter is increased significantly. As we want to avoid this probe diameter increase p below 0.5� is used in the technique N-PED. Best results were obtained on a FEI TITAN microscope equipped with a probe Cs-corrector and a 2kx2k CCD camera. Very smooth maps of the four components of the projected deformation tensor can be obtained with NPED (Fig.2). As in all diffraction techniques, the main drawbacks of N-PED for acquiring 2D maps are the slow speed and the large amount of data. A major advantage of diffraction-based techniques is to be Microsc. Microanal. 21 (Suppl 3), 2015 2210 able to analyze samples of non-uniform thickness and non-uniform composition along the electron beam. This will be illustrated by studying core shell nanowires (NWs) (Fig. 1l and 1m). [1] T Ghani et al, IEDM Techn. Dig. (2003) p. 11.6.1. [2] Y Martin et al, accepted in Ultramicroscopy. [3] A. B�ch� et al, Ultramicroscopy 131 (2013) 10. [4] J.L. Rouvi�re et al, Appl. Phys. Lett. 103 (2013) 241913. Figure 1. (a) Experimental CBED pattern observed in Si along [651, 441,31]. = 10 mrad. Both the transmitted and diffracted beams are used in the fitting of strain. (b) 18 different retrievals of the unit cell volume using a unique CBED pattern that was fitted 18 times with different starting values. The volume of the deformed crystal is retrieved only with the use of the excess lines. (c-g-f-k) Diffraction patterns obtained in different conditions, without (c and f) or with precession (g,h and k), with a slightly parallel beam (c and g,h) or a more convergent beam (f and k). (d-e-i-j) The associated images of the electron probe when going through Si. (h) Diffraction pattern similar to (g) but with descan off. (l) STEM image of tri-gate device composed of a Si NW surrounded by a TiN gate. (m) Core shell NW, composed of a Ge core surrounded with an amorphous SiN shell. Ge is dilated radially and compressed along its growth axis. Figure 2. Strain maps obtained in a device containing SiGe stressors and a Si3N4 layer above the SiO2 layer (the whited boxes) above the Si channel. (a-c-e-g) Experimental maps obtained with N-PED. (b-d-f-h) Maps obtained by Finite Element simulations. Note how the strain maps are smooth and are well reproduced by simulations where a -1.9 GPa stress is introduced in the Si3N4 layer.");sQ1[1104]=new Array("../7337/2211.pdf","4D-STEM Imaging With the pnCCD (S)TEM-Camera","","2211 doi:10.1017/S1431927615011836 Paper No. 1104 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 4D-STEM Imaging With the pnCCD (S)TEM-Camera M. Simson1, H. Ryll2, H. Banba3, R. Hartmann2, M. Huth2, S. Ihle2, L. Jones4, Y. Kondo3, K. M�ller5, P.D. Nellist4, R. Sagawa3, J. Schmidt2, H. Soltau1, L. Str�der2 and H. Yang4 1. 2. PNDetector GmbH, Sckellstra�e 3, 81667 M�nchen, Germany PNSensor GmbH, Otto-Hahn-Ring 6, 81739 M�nchen, Germany 3. JEOL Ltd.,3-1-2 Musashino Akishima Tokyo 196-8558, Japan 4. Department of Materials, University of Oxford, 13 Parks Road, Oxford OX13PH, UK 5. Universit�t Bremen, Otto-Hahn-Allee 1, 28359 Bremen, Germany 4D-STEM imaging describes a powerful imaging technique where a two-dimensional image is recorded for each probe position of a two-dimensional STEM image. For typical STEM images of 256x256 probe positions, a total of 65,536 2D images needs to be recorded. This amount of data can be recorded with the pnCCD (S)TEM camera in a practical timeframe. This camera uses a direct detecting, radiation hard pnCCD with a minimum readout speed of 1000 full frames per second (fps) [1]. With the pnCCD (S)TEM camera a 4D data cube consisting of 256x256 probe positions with a 264x264 pixel detector image for each probe position can be recorded in less than 70 s. Several measurements have been performed to prove the capability of the camera for 4D-STEM imaging, among them strain analysis, magnetic domain mapping and most recently electron ptychography. Electron ptychography is a 4D-STEM technique that has been described theoretically already in 1993 [2] but so far was limited experimentally by the low readout speed of existing cameras. In this technique, the intensity distribution in the bright field disk is recorded in 2D for each STEM probe position (Figure 1a). In an electron wave-optical approach the phase and amplitude information is extracted from the recorded intensity images. The reconstructed phase image shows enhanced image contrast compared to the conventional annular dark field image (Figure 2). Measurements with the pnCCD (S)TEM camera were carried out using a JEOL ARM200-CF in order to investigate different samples with the ptychographic phase reconstruction technique. Simulations show that the phase reconstruction technique does not require the full camera resolution of 264x264 pixels [3]. The pnCCD camera can be used in windowing and binning modes to further increase the measurement speed. Different combinations were tested at up to 4 000 fps (4-fold binning in one direction, i.e. 66x264 pixels, see Figure 1c). A 512x512 STEM image can then be recorded in just over a minute. Other 4D-STEM applications like strain analysis or magnetic domain mapping use the cameras's ability to precisely determine the exact position of the bright field disk on the detector. When traversing the sample the electron beam can be deflected due to different effects in the sample causing a movement of the BF disk's position. The advantage of using a 2D detector over conventional quadrant DPC detectors is the possibility to separate movements of the disk from intensity fluctuations inside it. Strain in the sample causes such movements and can be measured by 4D-STEM [4]. Imaging of magnetic fields inside the sample is possible using a Lorentz-like setup of the STEM, a 2D detector and analysis of the movements of the BF disk. Measurements demonstrating the technique with a nickel sample were done. Also dark field measurements profit from using 4D-STEM. By adapting the camera length, the 2D diffraction pattern can be recorded. In an offline analysis different STEM images can be reconstructed Microsc. Microanal. 21 (Suppl 3), 2015 2212 by using different parts of the detector images corresponding to different parts of the scattered beam. In conclusion we have shown that with the pnCCD camera new techniques in STEM are promoted and various fields of application benefit from recording the two-dimensional detector image. With its direct detection, high readout speed and radiation hardness the pnCCD (S)TEM camera makes recording of large 4D data cubes in short times possible and enables new science. [1] H. Ryll et al., Microscopy and Microanalysis 19 (2013), p.1160-1161. [2] J.M. Rodenburg, B.C. McCallum, P.D. Nellist, Ultramicroscopy. 48 (1993) 304�314. [3] H. Yang et al., Ultramicroscopy. doi:10.1016/j.ultramic.2014.10.013 [4] K. M�ller et al., Appl. Phys. Lett. 101 (2012), p. 2121101-2121104. a) b) c) Without binning Binning factor: 2 Binning factor: 4 Figure 1. Examples of recorded single raw images of the bright field disk for individual probe positions in STEM. Images were recorded without binning (a), 2-fold binning (b) and 4-fold binning (c). Pixel dwell times were 1000 �s, 500 �s and 250 �s, respectively. The intensity distributions in these images are analyzed to extract the phase information. Figure 2. Comparison of the conventional annular dark field (ADF) image and the reconstructed phase of a GaN sample using a ptychography method. Pairs of Ga-N columns show in the reconstructed phase, but the N columns are nearly invisible in the simultaneously recorded ADF image. Images contain 128x128 probe positions recorded at a dwell time of 250 �s.");sQ1[1105]=new Array("../7337/2213.pdf","GPU-Based Defect Image Simulations using the Scattering Matrix Formalism","","2213 doi:10.1017/S1431927615011848 Paper No. 1105 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 GPU-Based Defect Image Simulations using the Scattering Matrix Formalism Saransh Singh1 , Marc De Graef1 1 Dept. of Materials Science and Engineering, Carnegie Mellon Univ., Pittsburgh PA 15213, USA Systems of first order differential equations frequently occur in disparate fields spanning all of the sciences. Examples include the population evolution of interacting species in an ecosystem, the concentrations of molecules in a chemical reaction, the production of goods, the price of supplies in economic processes and so on. These equations also describe the dynamical scattering of electrons in a periodic lattice potential, in which case they are known as the Darwin-Howie-Whelan (DHW) equations. The DHWs describe how the probability amplitude of an electron wave changes as a function of depth in a crystal. This change depends on the difraction geometry and on the mutual interaction strength of different diffracted beams [1]. The equations are often written as: dg eig-g - 2isg g = i , dz qg-g g g where g is the amplitude of the beam diffracted by the lattice planes corresponding to the reciprocal lattice vector g, sg represents how well the Bragg condition is satisfied for the planes g, and qg-g describes the strength of interaction between the beams g and g . The equations are commonly written in the form of a matrix equation, given by: g=g 2sg dS = iAS, Agg = g=g dz qg-g where S is the column vector of unknown diffracted amplitudes and A is the structure matrix; the diagonal entries of A contain the diffraction geometry via the excitation errors sg , and the off-diagonal elements represent the beam interactions. The equations have a well-known solution of the form S(z0 ) = eiAz0 S(0) = S(z0 )S(0), relating the amplitude at depth z0 to the initial amplitude, which involves computation of the exponential of the structure matrix, resulting in the scattering matrix, S(z0 ). The presence of a lattice defect usually implies a broken translational crystal symmetry, resulting in atoms being displaced from their regular positions. Thus, a defect is usually quantified by a displacement field, R(r), that may vary continuously (e.g., misfitting inclusions and dislocations) or discontinuously (e.g., stacking faults) from one position to another. While there are other approaches to compute the amplitude changes of electrons traversing a periodic lattice (e.g., Bloch waves), the scattering matrix approach is particularly well suited for defect simulations as the defect contributions appear as phase factors in the off-diagonal terms only, while the overall structure of the equation remains intact [1]. The modified DHW equations, in the presence of one or more defects, take on the form: dg eig-g -ig-g - 2isg g = i e g , dz qg-g g where g 2g � R(r), with R(r) the total displacement at position r due to the superposition of all individual defect displacement fields. It should be noted that the defect phase shifts can be written in matrix form as Pgg (r) = e-ig-g , where the diagonal of the matrix contains 1s. Hence, the structure matrix in the presence of a defect is modified as Ad (r) = Agg Pgg (r), where the matrix product gg is performed element by element instead of as a regular matrix product. This decomposition into a perfect crystal factor A and a defect factor P has important consequences for the efficient numerical implementation of defect image simulations. Microsc. Microanal. 21 (Suppl 3), 2015 2214 Recently, Graphics Processing Units (GPUs) have become very popular for high performance computing applications, allowing for heterogeneous combinations of GPUs and CPUs to carry out computationally intensive tasks in a massively parallel way. The GPU has a very different memory structure and programming model from traditional serial programs; while the performance gain may be significant, the development effort and time involved in GPU programming is usually rather substantial. In this contribution, we present a new GPU-based algorithm for defect image simulations using the scattering matrix approach. In a typical CPU implementation, the intensity of each image pixel would be calculated in a sequential fashion; incorporation of multi-core techniques, for instance through the use of OpenMP constructs, can accelerate the computation, but even in the best case scenario, the speedup factor is usually less than the number of available cores. The GPU-based algorithm subdivides the image into smaller blocks and performs the computation of all pixels in a sub-block simultaneously. The perfect crystal structure matrix A is stored in global GPU memory, whereas the series of defect phase factor matrices Pj for each of the slices j along the integration column is stored in the local work item memory. The exponential of the defect structure matrix Ad is then calculated for each slice using scaling and squaring combined with the optimized Taylor expansion, considering the first nine terms in the expansion [2]. Multiplication of the first slice scattering matrix S1 with the initial probability amplitudes at the entrance surface, followed by repeated matrix-vector multiplications then completes the integration along the column. Such an approach minimizes the amount of data transfer between the CPU and the GPU, which would otherwise dominate the overall computation time. The scattering matrix formalism, as implemented on a GPU platform, provides a unified approach for defect image simulations for both the Transmission Electron Microscope (TEM) forescatter geometry and the Scanning Electron Microscope (SEM) backscatter geometry. The intensity of an image pixel in a TEM image is essentially given by the squared modulus of the relevant electron wave function component at the sample exit plane, i.e. |g (z0 )|2 , where g = 0 for bright field images, and g = 0 for dark field mode. STEM-mode diffraction contrast simulations can also be implemented by properly integrating the intensities over those diffracted beams that reach the detector [3]. For SEM image modes, the detected signal is typically given by a depth-integrated intensity. The observed backscattered intensity in an SEM can be described as [4]: Zi2 DWi z0 P(k) = (z)|(ri )|2 dz z0 0 iS where, Zi and DWi are the atomic number and Debye-Waller factor of the atom at the ith lattice site, (z) is a weight factor to account for the variation in the number of backscattered electrons as a function of depth and z0 is the maximum depth from which the electrons are backscattered; both parameters are computed via Monte Carlo trajectory simulations. The GPU-based scattering matrix approach provides a fast and efficient algorithm for defect image computations for both TEM and SEM mode; examples of simulation results will be provided, as well as an analysis of the speed-up achieved by moving from CPU to GPU-based computations. References [1] Marc De Graef in "Introduction to Conventional Transmission Electron Microscopy", (2003, Cambridge University Press, New York) p. 310, p. 462. [2] R.S. Pennington, F. Wang and C.T. Koch, Ultramicroscopy, 141, 32�37 (2014). [3] P.J. Phillips, M.J. Mills, and M. De Graef, Ultramicroscopy, 111, 1483�1487 (2011). [4] P.G. Callahan and M. De Graef, Microscopy and Microanalysis, 19, 1255�1265 (2013). [5] Research supported by the Air Force Office of Scientific Research, MURI contract # FA9550-12-1-0458.");sQ1[1106]=new Array("../7337/2215.pdf","EBSD Surface Topography Determination in a Martensitic Au-Cu-Zn Alloy","","2215 doi:10.1017/S143192761501185X Paper No. 1106 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 EBSD Surface Topography Determination in a Martensitic Au-Cu-Zn Alloy M. Chapman1 , P. Callahan2 , M. De Graef1 1 2 Dept. of Materials Science and Engineering, Carnegie Mellon Univ., Pittsburgh PA 15213, USA Dept. of Materials Science, University of California at Santa Barbara, Santa Barbara CA 93106, USA Au30 Cu25 Zn45 is a shape memory alloy with a composition that has been optimized to achieve a low hysteresis martensitic transformation, by increasing the degree of compatibility between the cubic austenite and the monoclinic martensite lattices [1]. This degree of lattice match has profound consequences for the multi-variant microstructure in that there is now no elastic energy penalty associated with the presence of variant-variant boundaries; this, in turn, can give rise to interesting curved martensite-austenite boundaries as well as the fact that the microstructure becomes completely irreproducible from one thermal cycle to the next. In our work, we attempt to verify a mathematical model [1] for the variant-variant geometries that occur during the transformation in Au30 Cu25 Zn45 ; in particular, we are interested in determining the complete transformation strain for each variant, including the out-of-plane component at the sample surface. Such a topographic measurement is made difficult because (a) the transformation temperature of the material is 50 C and (b) as stated above, every time the sample is transformed the martensite variant arrangement is completely different. Since crystallographic data needs to be collected along with the topography data to determine strain states and thus confirm the model, the topographic data needs to be collected simultaenously with the EBSD data. A novel technique that uses the background intensity of the EBSD patterns was used to achieve this result. The Backscatter Electron Surface Topography (BEST) method uses the EBSD detector and is based on the fact that, on average, backscattered electrons exhibit nearly specular reflection with respect to the surface of the sample; this can be modeled and confirmed by Monte Carlo (MC) simulations. For truly specular reflection, and assuming a spherical sample surface, the incident electron beam, the local surface normal, and the direction of maximum outgoing EBSD background intensity will all lie in a single plane. MC simulations can predict the deviation from the case of specular "reflection," which depends on the sample average atomic number, the microscope accelerating voltage, and the electron beam inclination angle, and can be expressed as a simple power series in the incidence angle. MC simulations were carried out for Au30 Cu25 Zn45 to determine the magnitude of specular deviation. An EBSD scan of martensitic Au30 Cu25 Zn45 is shown in Fig. 1a (Each pixel represents one EBSD pattern). Each 320 240 pixel EBSD pattern is fitted with a 2-D Gaussian distribution to determine the location of the maximum peak of the background intensity. This location is used to determine the relative shift of the peak from the center of the detector. The vertical component of the shift is depicted in Fig. 1b; green indicates a peak position 50 pixels above and black 13 pixels below the horizontal center line. Fig. 1c shows the horizontal component of the peak shift; red indicates a peak position 9 pixels left and black 6 pixels right of the vertical center line. The locations of the background peaks for the entire scan were binned and six predominant bins were identified. These six bins were randomly assigned colors and the resulting color map is shown in Fig. 1d. The background peak locations, along with the sample and experimental geometries allow for the calculation of the local surface normal. ^ The negative gradient of the surface height function, n(x, y) = rz(x, y), represents the local surface normal with components nx = @x z and ny = @y z. The surface height function, z(x, y), can then be retrieved by means of Fourier transforms to invert the gradient operator. The resulting surface height function, exaggerated in the vertical direction, is shown Fig. 2. Microsc. Microanal. 21 (Suppl 3), 2015 2216 References [1] Y Song, X Chen, and RD James, Nature 502 (2013), p. 85. [2] PG Callahan and M DeGraef, Microscopy and MicroAnalysis 19 2013, p.1255-1265. [3] Research supported by the Air Force Office of Scientific Research, MURI contract # FA9550-12-1-0458. Figure 1: a) Secondary electron micrograph of martensitic Au30 Cu25 Zn45 ; b) vertical position of the back- ground intensity peak on the EBSD pattern (see text for details); c) horizontal position of the background intensity peak; d) colorized map of the six clusters in the peak locations, indicating six different groups of sample surface inclinations that correspond to the individual martensite variants. Figure 2: Rendering of the reconstructed three dimensional surface of the martensitic Au30 Cu25 Zn45 alloy using BEST; the vertical dimension is exaggerated to highlight the topography. The surface is rotated 90 counterclockwise with respect to the images in Fig. 1.");sQ1[1107]=new Array("../7337/2217.pdf","High-Resolution Electron Backscatter Diffraction in III-Nitride Semiconductors.","","2217 doi:10.1017/S1431927615011861 Paper No. 1107 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 High-Resolution Electron Backscatter Diffraction in III-Nitride Semiconductors. Arantxa Vilalta-Clemente1, G. Naresh-Kumar2, M. Nouf-Allehiani2, Peter J. Parbrook3, Emmanuel D. Le Boulbar4, Duncan Allsopp4, Philip A. Shields4, Carol Trager-Cowan2 and Angus J. Wilkinson1. 1. 2. Department of Materials, University of Oxford, Parks Road, Oxford OX1 3PH, United Kingdom. Department of Physics, SUPA, University of Strathclyde, Glasgow G4 0NG, United Kingdom. 3. Tyndall National Institute, University College Cork, "Lee Maltings", Cork, Ireland. 4. Department of Electronic and Electrical Engineering, University of Bath, Bath BA2 7AY, United Kingdom. The large and increasing interest in III-nitrides semiconductors lies in the wide range of useful applications that can be achieved, from high electron mobility transistors (HEMTs) to light emitting LEDs and lasers. However, the III-nitride materials are usually epitaxially grown on foreign substrates, which lead to the formation of a large number of dislocations and significant strain variations in the epitaxial layers that seriously affect the performance of devices based upon them. The acquisition of high resolution electron backscatter diffraction (EBSD) patterns in the scanning electron microscope (SEM) is a very powerful method for the microstructural characterization of crystalline materials. EBSD is well established as a technique capable of measuring elastic strains, lattice rotations, and defect density in metallic materials, but until recently, it has not had much uptake for characterization of semiconducting materials [1]. This is largely as a result of the angular resolution limit of ~0.5� (~10-2 rads) of the conventional Hough-transformed analysis. The introduction of crosscorrelation based analysis of EBSD patterns has seen a step change in the angular resolution to ~10-4 rads which is sufficient to enable analysis of the much smaller misorientations and even local elastic strain fields that are more typical in semiconducting materials [2]. For EBSD measurements samples are tilted by 70 towards the EBSD detector (in our case Bruker eFlash HR with forescatter diodes FSDs). When the electron beam strikes the specimen, the electrons backscattered from the sample produce a diffraction pattern on the phosphor screen (Figure 1b). Changes in elastic strain and lattice rotations cause small shifts in the positions of zones axes and other features in the EBSD patterns. Cross-correlation is used to measure the shifts between each test pattern and a reference pattern on at least 35 sub-regions distributed across the pattern. The dispersion of shifts across the pattern is used to determine the change in lattice strain and rotation relative to the reference point. We have used the method on different III-Nitride semiconductor specimens. The EBSD measurements were made from linescan and 2D maps on the plan view of samples. We have measured the tilt, twist and elastic strain on different specimens. Histograms were constructed of the rotations about the surface normal (12 twist mosaic) and two orthogonal axes in the surface plane (13 and 23 tilt mosaics). Example histograms from a 900 nm thick GaN layer on sapphire are shown in figure 2.The width of the twist mosaic was found to be larger than the tilt mosaic. The twists within the layer are due to the edge components of threading dislocations (TDs), while tilts result from screw components. The EBSD results indicate that the edge type TDs are present in greater density than the screw type. Microsc. Microanal. 21 (Suppl 3), 2015 2218 Finally, we have also combined EBSD strain mapping with electron channeling contrast imaging (ECCI) observations of TDs [3]. Figure 3a shows an ECCI image from the top surface of 15 �m thick layer grown from a complex nano-dash template. The red circles indicate the positions of TDs found automatically by image analysis procedures. The TDs are highly clustered at the coalescence junctions between growth regions from different nucleation sites within the nano-dash template. Figure 3b is the 11 elastic strain variation measured by EBSD within the same field. As expected there is clear correlation between the positions of dislocation clusters identified by ECCI and locations at which the strain field changes most significantly. References: [1] A J Wilkinson, T. B. Britton, Materials Today 15 (2012), p. 366. [2] A J Wilkinson et al, Ultramicroscopy 106 (2006), p. 307. [3] G Naresh-Kumar et al, Phys. Rev. Lett. 108 (2012), p. 135503. [4] The authors acknowledge funding from EPSRC grant No: EP/J015792/1 & EP/J016098/1. Figure 1. (a) Schematic EBSD geometry and (b) example EBSD pattern obtained from GaN on sapphire. Figure 2. Histograms showing tilt and twist distributions in 900 nm GaN on sapphire. (a) DOI: 10.1 046/j .136 52818 .200 2.00 996. x (b) 10 �m 11 Figure 3. (a) ECCI showing TDs (marked with red circles) and atomic terraces and (b) one component (11) of the corresponding elastic strain field obtained by EBSD.");sQ1[1108]=new Array("../7337/2219.pdf","Nanoplasmonics in the TEM","","2219 doi:10.1017/S1431927615011873 Paper No. 1108 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Nanoplasmonics in the TEM Michel Bosman Institute of Materials Research and Engineering, A*STAR (Agency for Science, Technology and Research), 3 Research Link, Singapore 117602 This paper reviews recent developments in scanning TEM (STEM)-based characterization of localized surface plasmons. It will be shown that localized surface plasmon resonances can be mapped with STEM cathodoluminescence as well as with monochromated electron energy-loss spectroscopy (EELS) [1]. The spectral and spatial resolution of the techniques will be discussed, and examples will be given that demonstrate the unique capability of STEM-based plasmon analysis in comparison with other experimental techniques. Plasmon resonances on metal surfaces can be excited by light, transforming it from a free-space wave to a surface-bound oscillation. The modulation of light through surface plasmon resonances promises a range of novel technological applications, including nonlinear optics and fast circuitry. Fabrication and synthesis techniques are showing remarkable progress in obtaining nanometer precision in size and morphology of metal structures and particles, but characterization techniques typically lack this accuracy. An exception is STEM-based characterization, which has a long history in plasmon characterization [2-5], but has only recently become more widely used [6-10] thanks to instrumental developments [11, 12]. The first advantage of using STEM for surface plasmon characterization is the high spatial accuracy that the technique provides. We will show that in addition, EELS analysis will also give a more complete picture of the nature of the plasmon resonances. After all, far field optical characterization will mostly measure bright plasmon modes. With STEM on the other hand, highly localized fields surround the STEM probe, making this a clear near-field technique. As a result, bright as well as dark plasmon resonant modes can be excited, allowing a full-modal characterization, as shown for example in Figure 1 below [12]. The second advantage of this technique, is the possibility of measuring both atomic-resolution (S)TEM images and local spectra at the same location. A systematic series of measurements can now be obtained by doing monochromated EELS at several sample locations where the particle morphology and interparticle distances are measured. Even when the sample fabrication or particle synthesis does not yield a 100% success rate, measurements at selected locations will still give a full picture of the plasmon response, something that would not be possible with far-field techniques that measure many structures simultaneously over large areas. By combining imaging and local plasmon spectroscopy, we will demonstrate the occurrence of plasmon-enhanced electron tunneling [14, 15], and we will present in detail a method to control this novel phenomenon with the help of molecular monolayers [15]. The latter results are important not just as proof for the occurrence of plasmon-enhanced electron tunneling, but also as a demonstration how monochromated STEM-EELS can be used to measure the THz electronic properties of molecules. Microsc. Microanal. 21 (Suppl 3), 2015 2220 References: [1] A Losquin et al., Nano Lett. DOI: 10.1021/nl5043775 (2015). [2] C Powell and JB Swan, Phys. Rev. 115 (1959),p. 869. [3] PE Batson, Phys. Rev. Lett. 49 (1982), p. 936. [4] ZL Wang, JM Cowley, Ultramicrosopy 21 (1987), p. 347 [5] F Ouyang, PE Batson and M Isaacson, Phys. Rev. B 46 (1992), p. 15421. [6] J Nelayah et al., Nature Phys. 3 (2007), p. 348. [7] M Bosman et al., Nanotechnology 18 (2007), p. 165505. [8] B Schaffer et al., Phys. Rev. B (2009), p. 041401 [9] H Duan et al. Nano Lett. 12 (2012), p. 683. [10] D Rossouw and GA Botton, Phys. Rev. Lett. 110 (2013), p. 066801. [11] PC Tiemeijer, Ultramicroscopy 78 (1999), p. 53. [12] OL Krivanek et al. Nature 514 (2014), p. 209. [13] A Teulle et al. Nature Materials 14 87-94 (2015). [14] DC Marinica et al. Nano Lett. 12 (2012), p. 1333. [15] JA Scholl et al. Nano Lett. 13 (2013), p. 564. [16] SF Tan et al. Science 343 (2014), p. 1496. Figure 1. TEM image (left) showing a fused network of gold nanoparticles, each about 12 nm in diameter. The monochromated EELS maps (panels 2-6) show various infrared plasmon modes that are resonant in this network. The energy values at the bottom of each map indicate the resonant energy at which the plasmon mode is mapped for each panel. It can be seen that plasmon modes can resonate over relatively large distances of several hundreds of nanometers; see also [13].");sQ1[1109]=new Array("../7337/2221.pdf","Real-space Imaging of Plasmonic Modes of Gold Tapers by EFTEM and EELS","","2221 doi:10.1017/S1431927615011885 Paper No. 1109 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Real-space Imaging of Plasmonic Modes of Gold Tapers by EFTEM and EELS Wilfried Sigle1, Nahid Talebi1, Surong Guo1, Martin Esmann2, Simon F. Becker2, Ralf Vogelgesang2, Christoph T. Lienau2, Peter A. van Aken1 1. 2. Max Planck Institute for Intelligent Systems, Stuttgart, Germany Carl von Ossietzky Universit�t Oldenburg, Oldenburg, Germany Plasmonic nanoparticles have been extensively studied in the literature due to their ability of supporting localized surface plasmon (LSP) modes. Such structures can localize optical energy on the nanometer scale which opens up the field of optical nanoantennas. Application of nanoantennas in ultrafast optics requires large bandwidth. Unfortunately, the bandwidths of presently realized nanoantennas are small, despite their large radiative and Ohmic losses. Moreover, the coupling efficiency of far-field optical radiation to single nanoantennas is quite low, which is because of the extremely small volume of interaction. It was shown theoretically that tapered metallic nanostructures are able to couple propagating surface plasmon polaritons along their shaft adiabatically to nanolocalized plasmons at their apex [1]. This allows such tapered metallic waveguides to being applied in the field of sub-diffraction-limit nanofocusing, ultrafast photoemission, and near-field optical microscopy [2]. Here, we investigated both energy and spatial distribution of plasmonic modes being excited at the tip of gold tapers using electron energy-loss spectroscopy (EELS) and energy-filtering transmission electron microscopy (EFTEM). The EELS signal near the apex shows a broadband local density of optical states, as required for nanoantennas (Fig.1) in contrast to the narrow peaks from plasmon excitations along the shaft. This is also visible from EFTEM images (Fig.2) revealing strong intensity at the apex in the wide energy-loss range from 1.2 to 2.0 eV. We discuss the coupling of localized plasmons at the apex to the propagating plasmons at the shaft by considering the coupling efficiency and adiabatic behavior of the taper [4]. References: [1] M. Stockman, Phys. Rev. Lett. 93 (2004), p.137404. [2] M. Esmann et al, Beilstein J Nanotech. 4 (2013), p.603. [3] N. Talebi et al, Appl. Phys. a-Mater. 116 (2014), p.947. [4] The research leading to these results has received funding from the European Union Seventh Framework Program [FP/2007-2013] under grant agreement no 312483 (ESTEEM2). Microsc. Microanal. 21 (Suppl 3), 2015 2222 Figure 1. EELS spectra of a gold taper with an opening angle of = 45� for electron impact at the apex (purple) and at distances of 275 nm (black), 504 nm (green), 733 nm (red), and 962 nm (blue) from the apex. The spectra are shifted vertically for clarity. The zero-loss-peak contribution was subtracted from the individual spectra by using a power-law fit. Figure 2. Bright-field image and EFTEM images of conical Au tapers with an opening angle of (a) 35� in the energy-loss interval from 1.3 to 2.3 eV and of (b) 15� in the energy-loss interval from 1.2 to 2.2 eV. The color bar on the right symbolizes the energy-loss probability which is a measure of the z-component of the excited electric field [3].");sQ1[1110]=new Array("../7337/2223.pdf","Local Variations in the Surface Plasmon at Al Grain Boundaries and its Effect on the Optical Properties of Al Nanostructures.","","2223 doi:10.1017/S1431927615011897 Paper No. 1110 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Local Variations in the Surface Plasmon at Al Grain Boundaries and its Effect on the Optical Properties of Al Nanostructures. Andrew Thron1, C. G. Bischak2 , Scott Dhuey1 and Shaul Aloni1 1Molecular Foundry, Lawrence Berkeley National Laboratory, Berkeley, CA 94720 of Chemistry, University of California, Berkeley, Ca 94720 2Deparment Localized Surface Plasmon Resonances (LSPRs) are collective oscillations of free charge carriers which are driven by an external electric field. Light can be confined to nanoparticles and lithographically defined structures smaller than it's wavelength by coupling the electric field to LSPRs. The dispersion behavior of the LSPR is bound by the asymptotic limit, the surface plasmon frequency: = , (1) where sp is the surface plasmon frequency, p is the bulk plasmon frequency, and 2 is the dielectric function of the layer above the metal surface. With an inert metal such as Au, the surface plasmon frequency will stay constant in vacuum. Therefore, inherent variations to the LSPRs, in Au nanostructures, are largely associated with changes in shape, microstructure, and surface roughness. However, Knight et. al. observed a different behavior in Al disks, where the dipolar LSPR energy varied depending on the thickness of Al oxide layer [1]. Shifts in the LSPR were attributed to changes in the dielectric function of the oxide layer, which caused the surface plasmon to shift in energy. Decoupling the roll of shape, microstructure, and composition is necessary for successful design of Al plasmonic devices. Monochromated EELS in STEM has shown to be an effective technique in studying how microstructural changes influence the behavior of LSPR due to the high spatial resolution [2]. In this study Monochromated EELS and STEM shows that the surface plasmon varies locally across lithographically defined Al triangles, especially at the triple point between the grain boundary and the surface oxide layer. Lithographically defined Al triangles were created on 30nm thick Si3N4 membranes (Figure 1a). Using EELS, a decrease in the bulk plasmon peak energy was observed to occur at the grain boundaries (Figure 1b). This is attributed to a decrease in the charge carrier density at the grain boundaries due to a decrease in atomic density. A concurrent increase in the surface plasmon peak energy is also observed at the grain boundaries (Figure 1c). Using the relationship in equation 1, together with the measured surface and bulk plasmon peak energies, the dielectric function (2) is observed to vary across the Al triangle (Figure 1d). Specifically, 2 is observed to decrease at the triple points, where the grain boundaries meet the oxide layer. Changes to the LSPR supported by the Al triangle were not observed to due to their delocalized behavior (Figures 1e-i). However, local variations to the surface plasmon should contribute to scattering of the LSPRs. This study attempts to correlate the chemical and structural affects to the behavior of LSPRs. Further characterization of the oxide layer is needed in order to understand how variations in the oxide layer over the triple point and over the grain interior change 2. Moreover, in-Situ and ex- Microsc. Microanal. 21 (Suppl 3), 2015 2224 situ annealing experiments will be used to study the effect grain boundary density and structure have on the surface plasmon, and ultimately correlating how the surface plasmon contributes to LSRP scattering. References: [1] Knight M.W. et. al., ACS Nano, 8 (2014), 834-840 [2] Bosman M. et. al., Sci. Rep., 4 (2014), 5537 Figure 1. (a) Annular Dark Field image of lithographically defined Al Triangle. (b) Center of the Gaussian function fitted to the volume plasmon peak (c) Center of the Lorentzian function fitted to the surface plasmon peak. The overlaid black outlines are the positions of the grain boundaries observed in (b). (d) The change in the dielectric function 2, calculated from equation (1). The white outline highlight the positions of the grain boundaries. (e-i) Local Surface Plasmon resonant modes supported by the Al triangle.");sQ1[1111]=new Array("../7337/2225.pdf","Three-dimensional Surface Charge Reconstructions of Surface Plasmon Modes of Silver Right Bipyramids","","2225 doi:10.1017/S1431927615011903 Paper No. 1111 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Three-dimensional Surface Charge Reconstructions of Surface Plasmon Modes of Silver Right Bipyramids Sean M. Collins1, Martial Duchamp2, Emilie Ringe3, Zineb Saghi1, and Paul A. Midgley1. 1. 2. Department of Materials Science and Metallurgy, University of Cambridge, Cambridge, UK. Ernst Ruska-Centre, Forschungszentrum J�lich, J�lich, Germany. 3. Department of Materials Science and Nanoengineering, Rice University, Houston, USA. Plasmonic noble metal nanoparticles enable nanoscale confinement of electromagnetic fields for the development of chemical and biological sensing, photonic devices, and light trapping technologies. Scanning transmission electron microscopy and electron energy loss spectroscopy (STEM-EELS) has emerged as a key tool in experimentally observing the near field plasmonic modes of a wide variety of nanoparticle shapes and sizes [1]. When nanoparticle are at least approximately thin and invariant along the direction of the electron beam, the EELS signal can be related to a photonic localized density of states (LDOS) [2]. This LDOS, however, does not describe the three-dimensional fields or potentials surrounding the plasmonic nanoparticle [3], particularly in prismatic nanocrystals common in noble metal nanoparticle synthesis. Qualitative measurement of the three-dimensional EELS response has been demonstrated for a silver nanocube [4]. An alternative approach to reconstruct the surface charges corresponding to an eigenmode decomposition of the EELS response has been proposed theoretically [5], but has not been demonstrated experimentally to date. Here, we present surface charge reconstructions of the principal modes of silver right bipyramids. Experimental tilt-series STEM-EELS was acquired on the PICO microscope, a modified FEI Titan operated at 300 kV and equipped with a monochromator and a Quantum GIF for fast spectroscopy acquisition. Chemically synthesized silver right bipyramids with side length of approximately 50 nm were drop-cast onto molybdenum trioxide (MoO3) crystals dispersed on lacey carbon TEM grids. A straight-forward reconstruction of surface charge uses non-penetrating trajectories only [5], and so a partial MoO3 crystal substrate with a bipyramid resting against a crystal side in cross-section was selected for tilt-series STEM-EELS measurements. An annular dark field (ADF) tilt series was acquired concurrently with STEM-EELS for structural tomography of the particle morphology. ADF images were acquired in 3� increments from 68� to -67�. STEM-EELS maps were acquired at 0� and every 9� from 65� to 38�, limited by carbon contamination. The STEM-EELS data was processed by nonnegative matrix factorization (NMF) to separate modal contributions. Compressed sensing electron tomography (CS-ET) [6] was performed on the ADF data, the bipyramid morphology was extracted from the tomogram by threshold-based segmentation, and a surface mesh was generated for use in boundary element method (BEM) [7] simulations and surface charge reconstructions. The proposed reconstruction technique [5] uses conjugate gradient methods to minimize a cost function of the least-squares norm of the experimental data and a forward-calculated EELS response from a surface charge model. Given the prior knowledge of the particle morphology determined from CS-ET, the initial input surface charge distribution for cost function minimization was determined from BEM eigenmode calculations. To further constrain the cost function optimization algorithm to handle experimental noise and artifacts, an additional soft thresholding regularization penalty was used to identify solutions to the surface charge optimization that exhibited a small number of surface elements with high intensity, consistent with the experimental data and the highly localized surface charge Microsc. Microanal. 21 (Suppl 3), 2015 2226 densities typical of corner modes. Figure 1 presents surface charge reconstructions for two of the four principal corner modes observed experimentally. In the tilt series, two overlapping modes at 2.7 eV were not separated by NMF. However, two-dimensional maps of additional silver bipyramids revealed two separate modes in this spectral window. As determined from BEM simulations, the corner modes in an isolated right bipyramid comprise three dipole ((x,y) and z) and one quadrupole (z2) excitation, showing good agreement with the experimentally recovered modes. Symmetry-breaking by the substrate separates the degenerate (x,y) modes in the isolated bipyramid into modes at 1.3 eV and 1.8 eV. Significant surface charges on the substrate are observed in part due to artifacts in the experimental EELS maps exhibiting signal along the substrate-vacuum boundary. The method may be extended to higher energy modes localized along the edges and faces of the bipyramid, though these modes present additional challenges due to a high degree of multipolar mode overlap. The surface charge reconstructions presented here represent an excitationindependent response and can be used to predict the response to general illuminating fields with near or far field sources, offering a direct and quantitative assessment of the optical response of plasmonic particles from EELS [8]. References: [1] FJ Garc�a de Abajo, Reviews of Modern Physics 82 (2010), p. 209. [2] FJ Garc�a de Abajo and M Kociak, Physical Review Letters 100 (2008), p. 106804. [3] U Hohenester, H Ditlbacher and JR Krenn, Physical Review Letters 103 (2009), p.106801. [4] O Nicoletti et al, Nature 502 (2013), p. 80. [5] A H�rl, A Tr�gler and U Hohenester, Physical Review Letters 111 (2013), p. 076801. [6] R Leary et al, Ultramicroscopy 131 (2013), p. 70. [7] U Hohenester, Computer Physics Communications 185 (2014), p. 1177. [8] SM Collins acknowledges support of a Gates Cambridge Scholarship. The authors acknowledge funding from the European Research Council under the European Union's Seventh Framework Program (Grant No. FP7/2007-2013)/ERC Grant Agreement No. 291522-3DIMAGE and from the European Union's Seventh Framework Program under a contract for an Integrated Infrastructure Initiative (Reference No. 312483-ESTEEM2). Figure 1. Simulated and experimentally recovered surface charge distributions for (a) x and (b) z dipolar corner modes of a silver right bipyramid (each shown at two orientations). Surface charges are shown on a normalized color scale. Axes (top left) indicate the relative coordinates of each orientation.");sQ1[1112]=new Array("../7337/2227.pdf","High Resolution Characterization of Plasmonic Hybridization in Silver Nanostructures","","2227 doi:10.1017/S1431927615011915 Paper No. 1112 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 High Resolution Characterization of Plasmonic Hybridization in Silver Nanostructures E.P. Bellido1, I.C. Bicket1, J. McNeil1, G.A. Botton1 1 Department of Materials Science and Engineering, McMaster University, 1280 Main Street West, Hamilton, ON, L8S 4L7, Canada, gbotton@mcmaster.ca Surface plasmon resonances (SPR) in metallic nanostructures arise from the collective oscillation of conduction electrons, which create strong confined electric fields around the nanostructures. This confinement of electromagnetic (EM) energy at nanoscale dimensions holds potential towards the miniaturization of photonic devices [1]. Tremendous effort has been devoted towards optimization and design of nanostructures for several applications [2,3,4]. Most of these applications involve arrays of closely-spaced nanostructures: the plasmonic properties of the array differ from those of its isolated parts due to the interaction of evanescent fields. The study of SPR in these arrays, in particular the coupling of resonant modes, requires a characterization technique with both high spatial and energy resolution. Electron energy loss spectroscopy (EELS) meets these requirements, but has previously been limited to energies in the range of visible light or higher, mainly because of the relative intensities of the zero loss peak (ZLP) and low energy loss signal. In this work, we use electron beam lithography to fabricate circular arrays of rods (Fig. 1) and 30 nm high silver nano-square dimers (Inset Fig. 2) on 50 nm thick silicon nitride membranes. Using an ultrastable STEM-TEM (FEI Titan 80-300) system, operated at 80 keV and equipped with an electron monochromator and high-resolution electron energy loss spectrometer, we acquire spectral images of SPR modes with an energy resolution of 80 meV. To further increase the energy resolution and reduce the contribution of the ZLP, extending the range of detectable energies down to the mid-infrared region of the EM spectrum, we apply the iterative Richardson-Lucy deconvolution [5]. With this method, we obtain an effective energy resolution of 40 meV. The nano-square dimers are separated by 100 nm, as shown in the inset of Fig. 2. Due to the proximity of the two squares the structures couple through their evanescent fields, creating new modes that can be understood as the hybridization of the individual silver square SPR [6]. The result of the hybridization is the formation of multiple resonances that are energetically close to each other and can only be resolved by high energy resolution EELS. Figure 2 shows the deconvoluted spectra after 39 iterations at several positions on the dimer. Eleven plasmon peaks can be clearly identified, with the first peak at an energy of 0.28 eV, corresponding to the mid-infrared region on the EM spectrum. After deconvolution, we can identify peaks that are as close as 70 meV. As previously demonstrated, nanoantennas are able to efficiently transform optical radiation into SPR [7]. Far field radiation can excite short-range surface plasmon-polaritons (SR-SPP) in silver nanorods, which act as nanoantennas where the spatial distribution of the electromagnetic fields is correlated with the nanorod structure [2,3,7]. In this work, we also study the coupling and hybridization of SPR in nanoantennas. We extract high-resolution images and energy-filtered maps of hybrid SPR of 500 nm long nanorods in circular arrays with an inner gap radius of 50 nm; the number of rods in the array varies from two to eight. In these structures, we observe SP enhancement effects in the center for one or sometimes two of the low energy modes, despite minor asymmetries in the nanoantennas which may Microsc. Microanal. 21 (Suppl 3), 2015 2228 shift their individual resonant frequencies. Results suggest that these arrays do not require perfect resonance matching for enhancement effects to be observable. References [1] J.A. Dionne and H.A. Atwater, MRS Bulletin 37 (2012), p. 717. [2] D. Rossouw and G.A. Botton, Phys. Rev. Lett. 110 (2013) p. 066801. [3] D. Rossouw et al., Nano Lett. 11 (2011), p. 1499. [4] K.L. Kelly et al., J. Phys. Chem. B 107 (2003), p. 668. [5] E.P. Bellido, D. Rossouw and G.A. Botton, Microscopy and Microanalysis 20 (2014), p. 767. [6] E. Prodan et al., Science 302 (2003), p. 419. [7] L. Douillard, F. Charra, Z. Korczak, Nano Lett. 8 (2008), p. 935. Figure 1. Annular dark-field images of the nanorod arrays. Figure 2. Spectra showing the multiple hybrid plasmon resonances peaks of the two silver squares. The spectra were extracted from four positions color coded in the inset. Figure 3. Annular dark-field image of a circular array of four 500 nm rods (left) and its SPR showing the coupling and formation of hybrid modes from the multipolar resonances of the individual nanoantennas.");sQ1[1113]=new Array("../7337/2229.pdf","MALDI Imaging MS: Molecular Visualization of Tissues in Drug Discovery and Development","","2229 doi:10.1017/S1431927615011927 Paper No. 1113 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 MALDI Imaging MS: Molecular Visualization of Tissues in Drug Discovery and Development Stephen Castellino and M. Reid Groseclose GlaxoSmithKline, Drug Metabolism and Pharmacokinetics, Research Triangle Park, NC, USA Our current approach to risk assessment in the development of safe and efficacious drugs is primarily based on the parent drug plasma exposure (area under the concentration/time curve; AUC) in both animal models and in humans [1]. The successful implementation of this strategy is driven by a robust analytical methodology: LC-MS/MS. One of the key advantages of this "exposure-centric" approach is ease of translation from animal models to clinical. However, it has been recognized that measuring drug plasma exposure is not ideal given that most targets are not located within the plasma compartment and determining the tissue distribution of not only the parent drug but its metabolites would provide greater insight into our understanding of pharmacology and toxicology. What has been missing, until recently, is an analytical method capable of determining the tissue distribution of drugs and drug metabolites. Matrix assisted laser desorption/ionization (MALDI) imaging mass spectrometry (IMS) is uniquely suited for this task [2]. MALDI IMS is a highly sensitive technique (attomole), does not require isotopic labeling of analytes, and is capable of high spatial resolution (10�m pixels) and spectral resolution (FTICR; >104 resolving power (FWHM)). Furthermore, there is the opportunity to collect endogenous analyte distributions in the same experiment [3]. This allows one to co-register tissue histology images with the corresponding molecular images and visualize biology and chemistry in an integrated fashion. For MALDI IMS, the translation between animal models and clinical trials is though physiological based pharmacokinetic modeling and an enhanced understanding of biochemical mechanisms. In our experimental protocol, target tissues are removed at necropsy, flash frozen, and stored at low temperature until cryo-sectioning. Tissue sections (6-12 �m thickness) are thaw-mounted on glass slides prior to the application of the MALDI matrix. It is possible to perform whole rodent body MALDI imaging as well. The matrix is generally a small molecular weight highly conjugated molecule that absorbs the UV light from the laser and initiates desorption and the ionization process prior to detection by the mass analyzer. The matrix is usually applied as a fine aerosol containing a mixture of organic solvent and water. Tissue sections serial to those used for the MALDI IMS experiment are collected for histological staining (Figure 1). As is the case with other mass spectrometry methods, quantification of targeted analytes is possible if standards are available. Methods for quantitative MALDI IMS have been described in the literature [4-6]. Several examples illustrate the importance of understanding tissue distribution and the impact of MALDI IMS in drug development. When unexpected seizures were observed in a clinical Phase IIb study for an NNRTI inhibitor, MALDI IMS experiments with several animal models demonstrated that this adverse event was likely do to species differentiation of metabolite disposition and biotransformation. In this study, the exposure of the parent drug in plasma provided no clues to the mechanistic basis for the seizures in patients [7]. In a similar vein, monitoring the plasma levels of a late stage drug was not fully informing on the risk assessment of testicular toxicity in the rodent. In this case, MALDI IMS data demonstrated the testicular accumulation of long lived metabolites correlated Microsc. Microanal. 21 (Suppl 3), 2015 2230 with the time course of testicular injury. The risk assessment of testicular toxicity was linked to the biotransformation rather than the parent drug plasma level. In addition, several studies involving nepherotoxicity due to tubular deposits have been investigated using MALDI IMS. In one case, the deposits were determined to contain primarily calcium phosphate, which provided a basis for a mechanistic hypothesis. In the other case, the tubular deposits were shown to consist of several metabolites formed through oxidation pathways. In both instances, the parent drug was not directly linked to the toxicity. In summary, through the visualization of molecular tissue distributions combined with other imaging modalities, MALDI IMS is dramatically changing the way we discover and develop safe and efficacious drugs. While most of our efforts have focused on understanding drug related adverse events in the late stages of drug development, we have also found increasing demand and application in all stages of drug discovery and development including target engagement and PK/PD studies, (Figure 2). Figure 1. A) MALDI IMS pathway- tissue sectioning, matrix application, data collection, and ion images. B) Serial sectioning, histological staining and/or IHC, co-registration with ion images Figure 2. A) Optical Image of rat jejunum B) Ion image of drug from serial section C) H&E of rat liver with highlighted area of necrosis/inflammation D) Lipid ion image from serial section References: [1] P Muller and M Milton, Nature Reviews Drug Discovery (2012), p. 751. [2] R Caprioli, T Farmer, and J Gile, Anal. Chemistry, 69, (1997), p. 4751. [3] S Castellino, M Groseclose, and D Wagner, Bioanalysis, 3, (2011), p. 2427. [4] Reyzer, M, et al., J. Mass Spectrom. 38, (2003), 1081-1092. [5] Hamm G, et al, J. Proteomics, 16, (2012), p. 4952 [6] Groseclose M, and Castellino, S, Anal. Chemistry, 85, (2013), p. 10099 [7] Castellino, S, et al, Chem. Res. Toxicol., 26, (2013), p. 241");sQ1[1114]=new Array("../7337/2231.pdf","Development of Laser Ionization Imaging Mass Spectrometry for Multiple Drugs Administered to Cancer Cells","","2231 doi:10.1017/S1431927615011939 Paper No. 1114 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Development of Laser Ionization Imaging Mass Spectrometry for Multiple Drugs Administered to Cancer Cells Hisanao Hazama1, Hiroki Kannen1, Jun Aoki2, Michisato Toyoda2, Tatsuya Fujino3, Yasufumi Kaneda4 and Kunio Awazu1,5,6 1. 2. Graduate School of Engineering, Osaka University, Suita, Osaka, Japan Graduate School of Science, Osaka University, Toyonaka, Osaka, Japan 3. Graduate School of Science and Engineering, Tokyo Metropolitan University, Hachioji, Tokyo, Japan 4. Graduate School of Medicine, Osaka University, Suita, Osaka, Japan 5. Graduate School of Frontier Biosciences, Osaka University, Suita, Osaka, Japan 6. The Center for Advanced Medical Engineering and Informatics, Osaka University, Suita, Osaka, Japan Photodynamic therapy (PDT) is a less invasive treatment of cancer using a photosensitizing drug selectively accumulated in a tumor. By exciting the photosensitizing drug with a laser or other light sources, reactive oxygen species are produced and cancer cells are killed. PDT is expected for the treatment of cancer cells which gained resistance to anticancer drugs, while PDT is not suitable for treatment of a large tumor because the light cannot reach the deep region in the tumor. Thus, the combined therapy using chemotherapy and PDT has been proposed for treatment of a tumor including cancer cells which gained resistance to anticancer drugs. To evaluate and to optimize the efficacy of the combined therapy, it is important to measure and to control the spatial distributions of both the anticancer drug and the photosensitizing drug in the tumor tissue. In imaging mass spectrometry (IMS) using matrix-assisted laser desorption/ionization (MALDI), multiple drugs can be detected simultaneously. However, the detection sensitivity of IMS for drugs is lower than other drug imaging methods such as autoradiography, and some drugs could not effectively be ionized with MALDI. Therefore, a new ionization reagent, zeolite matrix, which is the mixture of zeolite and conventional MALDI matrix, has been developed to enhance the ionization efficiency of MALDI [1�3]. The detection sensitivity for an anticancer drug, docetaxel, was increased by 13-fold with the zeolite matrix [3]. In this research, simultaneous imaging of an anticancer drug and a photosensitizing drug administered to cancer cells was investigated using the zeolite matrix to confirm that the photosensitizing drug could accumulate in anticancer drug resistant cancer cells. An anticancer drug, docetaxel, and a photosensitizing drug, protoporphyrin IX (PpIX), were dissolved in a cell culture medium at the concentrations of 30 and 5 �mol/L, respectively. The human prostate cancer cell line PC-3 and its docetaxel resistant cell line PC-3-DR were cultured in the medium containing the drugs. After 1 hour, cells were washed with phosphate-buffered saline and centrifuged. After that, cell suspensions were prepared by removal of supernatant and addition of distilled water. Each cell suspension containing about 5000 cells with a volume of 1 �L was dropped onto a glass slide with a transparent and conductive coating of indium tin oxide (8237001, Bruker Daltonik GmbH, Germany) and dried. Zeolite matrix was made by mixing a zeolite, NaY5.6, with a MALDI matrix, 6-aza-2thiothymine, at 1:1 weight ratio and was dissolved with 50% acetone in water at a concentration of 10 mg/mL. 1 �L of the zeolite matrix solution was dropped onto the dried droplet of the cell suspension and dried. Mass spectra were measured using a MALDI time-of-flight mass spectrometer in the positive reflector mode (Voyager-DE PRO, Applied Biosystems, USA) and an external laser source at a wavelength of 355 nm, pulse width of 5 ns, and a pulse repetition rate of 20 Hz. Ion images of the dried droplets were obtained with the scanning mode IMS by automatically controlling the mass spectrometer Microsc. Microanal. 21 (Suppl 3), 2015 2232 using a software (MMSIT, Novartis, Switzerland). Figure 1 shows ion images of the sodium adduct ion [M + Na]+ of docetaxel at m/z 829.9 and protonated ion [M + H]+ of PpIX at m/z 563.3 simultaneously obtained from the PC-3 and PC-3-DR cells. Ion signal intensity of [M + Na]+ of docetaxel from PC-3 cells was significantly higher than that from PC-3DR cells. It is supposed that docetaxel is ejected from the PC-3-DR cells at a higher rate compared with that of PC-3 and cannot accumulate in the PC-3-DR cells, because the PC-3-DR cells have higher resistance to docetaxel compared with the PC-3 cells. On the other hand, [M + H]+ of PpIX was detected from both PC-3 and PC-3-DR cells with similar ion signal intensities. These results show that PpIX was accumulated in both PC-3 and PC-3-DR cells and that PDT would be effective for treatment of anticancer drug resistant cancer cells. The ion images in this research were measured in the scanning mode IMS at a spatial resolution of about 100 �m. However, cellular scale observation is preferable to distinguish the cancer cells with and without resistance to the anticancer drug in tumor tissues. Therefore, as the next step, cellular scale observation will be performed using a stigmatic mode imaging mass spectrometer with a spatial resolution of 1 �m [4�6]. References: [1] Y Komori et al., J. Phys. Chem. 114 (2010) p. 1593. [2] R Yamamoto and T Fujino, Chem. Phys. Lett. 543 (2012) p. 76. [3] H Kannen et al., Int. J. Mol. Sci. 15 (2014) p. 11234. [4] H Hazama et al., J. Biomed. Opt. 16 (2011) 046007, p. 1. [5] J Aoki and M Toyoda, J. Mass Spectrom. Soc. Jpn. 61 (2013) p. 23. [6] J Aoki et al., J. Phys. Soc. Jpn. 83 (2014) 023001, p. 1. Administered docetaxel and PpIX Control Administered docetaxel and PpIX Control 1 PC-3 cells PC-3-DR cells 1 mml 1 mml 0 (a) docetaxel (b) PpIX Figure 1. Ion images of (a) sodium adduct ion [M + Na]+ of docetaxel at m/z 829.9 and (b) protonated ion [M + H]+ of PpIX at m/z 563.3 obtained from the human prostate cancer cell line PC-3 and its docetaxel resistant cell line PC-3-DR. Control means the cells without administration of the drugs. Ion signal intensity [arb. units]");sQ1[1115]=new Array("../7337/2233.pdf","Mass Spectrometry Imaging of Microbial Interactions using Matrix-Enhanced Laser Desorption/Ionization Fourier-Transform Ion Cyclotron Resonance Mass Spectrometry","","2233 doi:10.1017/S1431927615011940 Paper No. 1115 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Mass Spectrometry Imaging of Microbial Interactions using Matrix-Enhanced Laser Desorption/Ionization Fourier-Transform Ion Cyclotron Resonance Mass Spectrometry Christopher R. Anderton,1 Rosalie K. Chu,1 Jared B. Shaw,1 Erika M. Zink,1 Carolyn A. Zeiner,2 Dennis Trede,3 Colleen M. Hansel, 4 and Ljiljana Pasa-Toli1 1 2 Pacific Northwest National Laboratory, Richland, WA 99354 School of Engineering and Applied Sciences, Harvard University, Cambridge, MA, USA 3 SCiLS GmbH, Bremen, Germany 4 Marine Chemistry and Geochemistry Department, Woods Hole Oceanographic Institution, Woods Hole, MA, USA Mass spectrometry imaging (MSI) methods offer the ability to gain both chemical and spatial information from biological samples of interest. In particular, matrix-enhanced laser desorption/ionization (MALDI) MSI facilitates detection of a variety of intact biomolecular species (e.g., lipids, peptides, and proteins) with spatial resolution regularly < 20 �m [1]. When analyzing complex systems, like biological samples, the need for high mass spectral resolution and high mass accuracy measurements to correctly identify and map molecules of interest is warranted. Fourier transform ion cyclotron resonance (FTICR) MS offers the highest mass resolving power and mass accuracy of any mass analyzer and has demonstrated greatly enhanced MSI performance [2]. Here, we will demonstrate the advancements our lab has made in MALDI MSI. In particular, we focus on characterizing the interactions of Eumycete fungi, key organisms in carbon degradation within the environment. We also explore new sample preparation methods and data analysis workflows. The advantages of using high performance mass spectrometers for studying biological material is illustrated in Fig. 1A, where many more spectral features can be resolved using FTICR verses more traditional MS (ToF) [2]. However, to attain unequivocal identification of detected molecules, tandem mass spectrometry must be used in conjunction with high performance mass spectrometers. We have equipped our high magnetic field (15T) FTICR-MALDI MS (SolariX, Bruker) with a UV laser (193 nm, ArF excimer, Coherent ExciStar XS) for UV photodissocation of selected ions within our ICR analyzer cell. Fig. 1B demonstrates the tandem MS capabilities of this method, where we observe a more comprehensive fragmentation of a lactosylceramide (LacCer) molecule than conventional tandem MS approaches. Sample preparation of agar-based microbial samples has been a significant limitation to MSI of microbial communities with MALDI, with only minor improvements being reported to the method the Dorrestein Research Group developed [3]. This basic method entails dry application of the matrix onto the hydrated agar sample by using a sieve, followed by agar dehydration to form a crystalline surface ready for MALDI analysis. We have explored a series of other agar conditions, sample adhesion, and matrix applications routes to advance the ability of MALDI MSI for imaging microbial communities inculcated on agar. In this we have also investigated the use of a robotic sprayer (TM-Sprayer, HTX Imaging) for matrix application in attempt to make highly homogenous, reproducible matrix films, while maximizing analyte extraction into the matrix crystal. Finally, the complexity and size of high mass resolution MSI data presents challenges in data Microsc. Microanal. 21 (Suppl 3), 2015 2234 processing. Opening and caching MSI files that are typically tens to hundreds of gigabytes in size, generating peak list with sufficiently tight tolerances of hundreds of peaks, and elucidating biological signatures of interest can be extremely challenging. We have been utilizing an advanced MSI processing software (SCiLS Lab), previously developed for ToF MSI workflows, for spatial characterization of detected molecules from microbial samples [4,5]. References: [1] B Spengler, Anal. Chem. 87 (2015), 64. [2] D F Smith, A Kiss, F R Leach III, E W Robinson, L Pasa-Toli, and R M A Heeren, Anal. Bioanal. Chem. 405 (2013), 6069. [3] J Y Yang, V V Phelan, R Simkovsky, J D Watrous, R M Trial, T C Fleming, R Wenter, B S Moore, S S Golden, K Pogliano, and P C Dorrestein, J. Bacteriol. 194 (2012), 6023. [4] T Alexandrov, M Becker, S-O Deininger, G Ernst, L Wehder, M Grasmair, F von Eggeling, H Thiele, and P Maass, J. Proteome Res. 9 (2010), 6535. [5] This work was performed at EMSL, a national scientific user facility sponsored by the Office of Biological and Environmental Research, U.S. Department of Energy (DOE). EMSL is located at PNNL, a multidisciplinary national laboratory operated by Battelle for the U.S. DOE. We would like to thank Professor Pieter Dorrestein (UCSD) for his many useful discussions, and Mr. Alain Creissen is thanked for his contributions and discussions in optimizing the TM-Sprayer for matrix applications. Figure 1. (A) Spectral overlay of ToF and FTICR MS data taken from the same rat brain section. This figure was modified from reference [2]. (B) MALDI-UVPD mass spectrum of the sodium adduct of LacCer (d18:1/8:0).");sQ1[1116]=new Array("../7337/2235.pdf","Multimodal molecular imaging: Insight into the complexity of biological surfaces through speed, resolution and identification.","","2235 doi:10.1017/S1431927615011952 Paper No. 1116 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Multimodal molecular imaging: Insight into the complexity of biological surfaces through speed, resolution and identification. Ron M.A. Heeren1,2, Anne L. Bruinen1, Nadine E. Mascini2, Gregory L. Fisher3, Tiffany Porta1,2 and Shane R. Ellis1 1 M4I, The Maastricht MultiModal Molecular Imaging Institute, University of Maastricht, Maastricht, The Netherlands. 2 FOM-AMOLF, Science Park 104, 1098 XG Amsterdam, The Netherlands 3 Physical Electronics, 18725 Lake Drive East, Chanhassen MN, USA The chemical complexity of biological surfaces is highly dynamic and subject to local changes in response to a changing environment. This chemical heterogeneity is a particular important parameter when considering treatment of diseases such as cancer. It is this inconceivably complex heterogeneity that makes tumors so difficult to treat as no single therapy targets all permutations of phenotypes and environment precisely. This implies that to make truly personalized tumor therapy reality a diagnostic method is needed that unravels this spatial and molecular complexity of tumor tissue. Here, we will argue that multimodal imaging mass spectrometry [1] is a diagnostic and prognostic tool that can elucidate tissue (and tumor) molecular heterogeneity and employ it to predict or design therapy response. A molecular view at different spatial and molecular levels is required to realize this ambitious goal. A combination of immunohistochemistry, Secondary Ion Mass Spectrometry (SIMS) and Matrix Assisted Laser Desorption Ionization (MALDI) imaging mass spectrometry is employed to classify and stratify different cellular phenotypes as well as to predict therapy response. The distribution of several hundreds of molecules on the surface of complex (biological) surfaces can be determined directly in a single imaging MS experiment. This enables molecular pathway analysis as well as the investigation of the role of the different molecular signals and their behavior under influence of a drastically changing physical, chemical or biological environment. Imaging MS is a label-free imaging technique that is capable of visualizing the molecular distribution on the surface of a biological sample. State-of-the-art molecular imaging mass spectrometry has evolved to bridge the gap between different disciplines such as MRI, PET, fluorescence imaging and histology. Among imaging MS modalities, SIMS is considered a high spatial resolution molecular imaging technique that addresses `clean' predominantly unprepared tissue surfaces. MALDI on the other hand is considered to provide a broader molecular scope on biological thin tissue sections that have been covered with a thin layer of minute matrix crystals that assist desorption and ionization. A particular advantage of MALDI is the detection in a higher mass range and the ability to characterize biological compounds with routine tandem MS. SIMS has, for a long time, be devoid of tandem MS based identification capabilities. Here, we present a novel high resolution tandem TOF-SIMS method to analyze and identify intact lipid ions from thin tissue sections. This new approach enables simultaneous surface screening with MS1 and targeted identification with MS2. The clinical acceptance of imaging MS greatly depends on the validation of the molecular images employed for patient and tissue classification. Modern pathology/hospital biobanks contain both chemically fixed (through formalin fixation and paraffin embedding) and cryopreserved material for Microsc. Microanal. 21 (Suppl 3), 2015 2236 large patient cohorts. Fundamental studies have, for a considerable amount of time, provided insight into signaling pathways of lipids, peptides and proteins directly from tissues of these biobanks. Their validation however was severely hampered by the lack of molecular imaging throughput. A number of innovative developments have emerged that significantly have increased the speed of analysis. We will demonstrate these advances based on a the rapid analysis of a number of different tumor models including breast cancer (BC) and cholangiocarcinoma (CCA). The (macro)molecular information obtained with high throughput imaging MS is combined with routine histology (H&E staining), allowing linking of molecular data with specific cellular structures. Molecular profiles of CCA will be correlated with (i) tumor (sub)class, i.e. intrahepatic vs. hilar/distal CCA, mucin-negative vs. mucin-positive intrahepatic CCA2, (ii) IDH mutation status and (iii) clinical outcomes including adjuvant chemotherapy response and disease-free survival. It is anticipated that this will yield new insights into the molecular pathology of cholangiocarcinoma and potentially novel therapeutic strategies. Moreover, molecular signatures predictive for clinical outcomes may form the base of personalized treatment. Chen et al [2] have shown that MALDI MS analysis of lipids distinguishes cancerous epithelium of cholangiocarcinoma from adjacent normal tissue, and between cholangiocarcinomas and colorectal cancers. Here we take this molecular based classification to another level through the direct application of imaging MS on a combination of larger tissue microarrays (figure 1), selected sections of patient tumors and xenografts References: [1] K. Chughtai and Ron M.A. Heeren, Chem. Reviews 110 (2010), p. 3237-3277. [2] Chen et al., Clin Chim Acta. 412 (2011) p. 1978-1982 [3] The authors acknowledge funding from the regional government of the province of Limburg, the Netherlands through the LINK program as well as the kind technical support from Waters Inc. and Bruker Daltonik GMBH. Dr. Tiffany Porta acknowledges support from the Swiss National Science Foundation within the "SNSF Early Postdoc.Mobility-Fellowships" program (grant P2GEP2_148527). Figure 1. Hematoxylin and eosin stained tissue microarray of patient-derived xenograft models of breast cancer with a corresponding characteristic tryptic peptide ion image at m/z 3187.4.");sQ1[1117]=new Array("../7337/2237.pdf","Standard Methods and Reference Materials for Performing a Complete EPMA WDS Instrument Diagnostic","","2237 doi:10.1017/S1431927615011964 Paper No. 1117 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Standard Methods and Reference Materials for Performing a Complete EPMA WDS Instrument Diagnostic D. C. Meier, E. S. Windsor, J. M. Davis, R. B. Marinenko, J. R. Anderson, F. Meisenkothen, and S. A. Wight Material Measurement Laboratory, National Institute of Standards and Technology, Gaithersburg, MD 20899-8371 An analytical laboratory quality assurance (QA) program should ideally confirm whether or not a given instrument is capable of making the best possible measurement per manufacturer specifications. There are two indispensable components of such a QA program. The first is a designed test analytical procedure for QA that can result in the best possible measurement given ideal conditions. The second is an ensemble of specimens and reference materials upon which the analytical procedure is performed that are suitable for testing the limits of the instrument's capabilities. The first component can be addressed by carefully standardized methods; the second, by carefully designed and analyzed reference materials. A comprehensive standardized QA method, in the context of electron probe microanalysis (EPMA) wavelength dispersive spectroscopy (WDS), is designed to assure that the function of each independent component of the instrument, including electron emission source, wavelength dispersion elements, photon counters, and spectrometer motion hardware, is tested. Furthermore, the experimental design should test each of these components to the limits of their capability in as efficient a manner possible, so as to maximize the analytical benefit while using minimal laboratory resources. Ongoing development efforts of such a standard, ISO CD 19463, will be discussed in the context of these design limits. Upon establishing the parameters of a standard QA method, the results of performing this experimental procedure will be demonstrated using NIST mineral glasses K-411 and K-412 from NIST SRM 470. The specific physical properties that make the SRM 470 glasses suitable specimens for QA, such as stability under electron beam and minimal compositional heterogeneity within and between specimens, will be discussed. The results of weekly QA compositional analyses of these glasses using suitable reference materials will be presented using a variety of statistical analysis, such as normality tests, and graphical representations, such as control charts, box plots, and bean plots (Figure 1), in order to demonstrate how actionable information can be extracted from periodic QA measurements. Finally, the primary limitation of K-411 and K-412 will be identified; specifically, these glasses are not composed of appropriate elements necessary to fully test the functionality of pentaerythritol (PET) or lithium fluoride (LiF) dispersion elements. For this reason, a glass engineered specifically to serve as the test specimen for EPMA WDS QA has been under development. The glass, designated ADM6XXX, is being integrated into a comprehensive QA program that effectively covers the analytical requirements of most of the periodic table, using a single test specimen and reference material specimen ensemble, capable of assessing the functionality of every spectrometer equipped with thallium acid phthalate (TAP), PET, or LiF dispersion elements on the instrument simultaneously. Under electron beam excitation, this engineered glass emits characteristic X-rays that include first order and second order diffraction lines for two characteristic K lines on each of these three crystals (Zn and Ge on LiF, Ca and Ti on PET, and Si and Al on TAP). The standardized QA procedure will be demonstrated analyzing the new materials in order to illustrate the probative value of a more comprehensive QA procedure for EPMA WDS. Microsc. Microanal. 21 (Suppl 3), 2015 2238 Figure 1. A control chart of bean plots of the Mg composition measurement over a period of months shows performance anomalies that were determined to stem from a fractured TAP crystal. Replacing the crystal (dashed vertical line) reduces the spread and improves the normality of the data. Figure 2. Mean composition of 18 specimens of ADM6XXX glass, confirming heterogeneity is within the compositional resolution of the instrument.");sQ1[1118]=new Array("../7337/2239.pdf","Testing of the DR Method for Image Sharpness Determination","","2239 doi:10.1017/S1431927615011976 Paper No. 1118 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Testing of the DR Method for Image Sharpness Determination M Matthews1 and J Shah2 1. 2. AWE, Aldermaston, Reading, UK Visiting Fellow, University of Bristol, Bristol, UK The current practice of specifying instrument resolution in SEM, of measuring the smallest visible gap between gold particles on a carbon substrate is, at best, not reliably repeatable and, at worst, dependant on the `eye of the beholder'. The International Standards Organisation Technical Committee on Microbeam Analysis, ISO/TC202, is developing a standard method for the quantitative determination of image sharpness in electron micrographs. Of the various procedures that have been appraised to date for their suitability as a standard the Derivative (DR) Method shows promise. The procedure, shown diagrammatically in Figure 1, is to first identify edges in the image area and draw profiles of predetermined length perpendicular to these. In the software version used for this study profiles with insufficient contrast are rejected (shown in red in Figure 1). The average of the slope parameters of an error function fitted through each of the identified profiles gives the sharpness value, s for the image. After initial testing of a set of eleven images distributed to the technical committee working group nine images were selected for detailed testing by the authors. From the shape of the two dimensional angular distribution plot of (where = S/2) it was possible to determine the degree of the noise distribution in the micrographs. The degree of astigmatism, if present was also detectable. Sharpness values were derived for each, varying individual input parameters, to determine the relative sensitivities of the method. The following key points were identified: � Although the procedure was able to extract edge profiles from a range of sample types it could not be used on every sample. For example two sample images failed to derive meaningful sharpness values. � All the images tested showed steady sharpness values for a given set of parameters when more than 100 � 200 edges were averaged. This was also the case for micrographs with considerable noise. � The computer program for calculating sharpness allows different input values of several parameters such as distance between the pixels processed, the profile (edge) length, number of edges (Figure 2) etc. The profile length parameter showed the greatest sensitivity, changing the derived sharpness value by up to �15% in the images tested. � The sharpness value reached a steady state with increasing profile length for the most of the images (Figure 2), although for one image this was not the case � It was not visually apparent which images would return convergent value of sharpness. Of the methods so far tested by the ISO technical committee the DR method shows the most promise. It can be applied to a range of image/sample types and generates a stable sharpness value with only 100 � 200 edges. However, checks need to be included in the procedure to test for the sensitivity to profile length. Microsc. Microanal. 21 (Suppl 3), 2015 2240 Figure 1 Image as acquired (upper left), edges identified and profiles drawn (upper right), error function fitted through each profile (lower left), average profile (lower right) derives sharpness value. 10 Default paramter values Distance between each pixel 1 (7) 8 Edge length 40 (20) Edge length 10 (20) Sigma 4 (2) 6.tif 10 6.tif 8 Sharpness 6 4 Sharpness 1 10 100 1000 10000 6 4 2 2 0 0 0 5 10 15 20 25 30 35 40 45 50 Edges Edge Length Figure 2 Sharpness results for a `well behaved' image.");sQ1[1119]=new Array("../7337/2241.pdf","The Use of Color in Elemental Maps and Electron-Microscope Images.","","2241 doi:10.1017/S1431927615011988 Paper No. 1119 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 The Use of Color in Elemental Maps and Electron-Microscope Images. R.F. Egerton 1. Physics Department, University of Alberta, Edmonton, Canada T6G 2E1. Color is used in elemental maps to provide a convenient way of displaying the results of electron-beam microanalysis. But such maps lack the intuitive appeal of gray-scale images in one important respect: there is no established consensus regarding the color assigned to each element. What are the prospects for standardization and how much information (including elemental concentration?) can be represented? Chemists have adopted a few color choices in connection with molecular modeling, such as the CPK scheme [1,2]; see Table 1. Some of these choices are mnemonic, for example white for colorless hydrogen, black for carbon, yellow for sulfur, green for chlorine. However, the CPK scheme is not universal and it covers only a small fraction of the common elements, which raises the question of how many elements could be represented if a standard color scheme were to be agreed upon. It has been claimed that the human eye can distinguish thousands or even millions of colors. Indeed, the standard 8-bit RGB scale allows (28)3 = 16,277,216 colors to be represented in a computer-stored image. However, the color differences involved (when the R, G or B value is changed by 1 bit) are subtle and often undiscernable. Ignoring 5% of the population (mainly males) who are color-blind, humans have trichromatic color vision. This three-input limit greatly restricts the number of distinguishable colors, compared to 1024-channel spectral analysis for example; there are many different spectra (metamers) that give rise to the same color sensation. Moreover, the number of colors that can be recognized (and assigned to a single element) is nodoubt a small fraction of the number than can be distinguished in a side-by-side comparison. In fact, it seems unlikely that all of the 92 elements would be individually recognizable. Although it is always possible to provide a key alongside the elemental map, there seems to be some consensus in the MSA community that the adoption of a few standard choices (as in Table 1) would be a step forward. A related problem is what to do when different elements overlap. Without precautionary measures, equal amounts of oxygen (red) and halogen (green) would create (through additive mixing on a display screen) yellow, which might be mistaken for sulfur. One solution (at the expense of spatial resolution) is to use a patterned texture whose cells are based on the elements present in each region. A different use of color is to represent the local intensity in an electron-microscope (TEM, SEM) image. Because the human eye can barely distinguish 16 grey levels [3], color allows more information to be represented in a single display. Although the choice of a false-color scheme is in principle arbitary, some choices are more logical, more intuitive, or more pleasing than others. One logical scheme represents increasing intensity by near-spectral colors from red to blue (as in Fig. 1a), corresponding to increasing photon energy or increasing temperature of an emitting object. However, many people associate red with heat, so an inverse-spectral scheme (blue to red; Fig. 1b) is probably more intuitive. The so-called hypsometric/bathymetric scheme is often used in cartography (deep blue for ocean depths, green for low-lying land, brown for higher, purple and white for mountains) and can be both pleasing and intuitive (Fig. 1c). Microsc. Microanal. 21 (Suppl 3), 2015 2242 Some of these color schemes allow a larger number of distinguishable levels than others. The eye appears to be more sensitive to small variations in the case of pale colors (center of the color triangle) and in the green region; see Fig. 2. In fact, pure spectral colors need to be avoided because they cannot be displayed via printing or on a RGB display such as a computer monitor [4]. References: [1] L Corey and L. Pauling, Rev. Sci. Instrum. 24 (1953), p. 621. [2] WL Koltun (1965), U. S. Patent 3170246. [3] AV Crewe, Scanning 3 (1980), p. 176. [4] The author thanks the Natural Sciences and Engineering Research Council of Canada for funding. element hydrogen carbon nitrogen oxygen phosphorus sulfur halogens metals atomic no. 1 6 7 8 15 16 9,17,35,53 26-29 Corey and Pauling (1953) white black blue red ----Koltun (1965) white black blue red purple yellow shades of green silver Table 1. CPK color assignments, based on Corey and Pauling [1] and on Koltun [2]. (a) (b) (c) Figure 1. Contour image showing three peaks and two valleys, in three color schemes: (a) with red to blue representing increasing height, (b) with blue to red representing increasing height, (c) with a hypsometric/bathymetric scheme, as used in cartography. The color key is to the right of each image. (a) (b) (c) Figure 2. Effect of (a) increasing the green content of pale gray by 10%, (b) increasing the intensity of green by 10%, and (c) increasing the intensity of blue by 10%.");sQ1[1120]=new Array("../7337/2243.pdf","Tomographic measurement of buried interface roughness","","2243 doi:10.1017/S143192761501199X Paper No. 1120 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Tomographic measurement of buried interface roughness Misa Hayashida1, Shinichi Ogawa2, Marek Malac1, 3 1 2 National Institute for Nanotechnology (NINT), Edmonton, Alberta, Canada. National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba, Ibaraki, Japan. 3 Department of Physics, University of Alberta, Edmonton, Alberta, Canada. Interface roughness of buried layers in nano-meter size devices often degrades performance of the devices because the roughness of the interfaces might affect expected growth of consequent layers during the device fabrication. To characterize the roughness of an exposed surface, atomic force microscope (AFM) has been widely used. On the other hand, when the interfaces are buried by additional device layers, it is impossible to measure the interface roughness by AFM. In this study, we present quantitative measurement of buried interface roughness by electron tomography using a STEM. The presented method allows to characterize interface roughness of the devices with a resolution of 1 nm. Figure 1 shows a multilayer rod-shaped sample of a stacked SiO2 - W - SiO2 on Si substrate structure used to characterize the roughness. An upper SiO2 - W interface exhibits larger roughness than a lower W- SiO2 interface. A cross sectional rod-shaped sample was fabricated by an FIB method. Nano-dot fiducial markers for accurate alignment of the tilt series were fabricated by SEM with a gas injection system [1]. The tilt series was acquired using annular dark field STEM mode of a Hitachi HF-3300 TEM / STEM with cold field emission source and MAESTRO computer control system [2]. The tilt step was 3� over the entire �90� tilt range eliminating the missing wedge problem. Ten images were collected at every tilt with 5 us dwell time to decrease effect of sample drift. All the images were summed after compensating each image for drift using standard cross correlation alignment. To evaluate effect of shot noise, 3, 5, 7 and 10 images were summed at each tilt resulting in four tilt series with different electron dose. Images were 512 X 1024 pixels with 0.47 nm pixel size. All four tilt series were reconstructed using Filtered Back Projection (FBP). To determine positions of the interfaces, median filter, low pass filter, Sobel filter and binary filter were applied to the slice images as shown in Figure 2. Then, the positions of interfaces between black (digital 0) and white (digital 1) were detected from the each Z slice. Approximately 7x10 4 points were detected in every tilt series. A plane was determined as it minimized the RMS distance from the interface for all points. Standard deviation of distances from this plane was taken as "interface roughness". The same processing was applied to all four tilt series. Figure 3 shows measured roughness decreases with increasing electron dose because the signal to noise ratio is increasing with electron dose and the image processing works more accurately. The measured roughness tends to an asymptotic value with increasing electron dose. The four series allow us to extrapolate the measured data to the asymptotic value, which in this case, is ~ 1.05 nm. To verify the performance of the data processing step, we also characterized roughness of computer generated test data and it showed a good agreement between the known input roughness and the measured roughness. [1] M. Hayashida, et. al., Ultramicroscopy, Vol 144, (2014), 50-57 Microsc. Microanal. 21 (Suppl 3), 2015 2244 [2] M. Bergen et. al. Microscopy and Microanalysis 19 (S2) (2013), 1394-1395. Financial support of Visiting Researcher grant from Alberta Innovates Technology Futures is gratefully acknowledged. This work was done when the first author worked at her earlier affiliation, AIST. (a) 50 nm (b) 50 nm Figure 1. ADF STEM image of test Figure 2. sample for evaluation of interface roughness. Nano-dot fiducial markers are (a) Z slice image of reconstructed volume (b) Filtered image of (a). The red line is the detected interface marked by two ellipses. between SiO2 and W layer. Figure 3. The measured roughness depends on signal to noise ratio (irradiation dose) and can be extrapolated to an asymptotic value of ~1.05 nm at a very high dose. The horizontal axis is the incident number electrons per pixel for the entire �90� series.");sQ1[1121]=new Array("../7337/2245.pdf","Highly Accurate Real Space Nanometrology Using Revolving Scanning Transmission Electron Microscopy","","2245 doi:10.1017/S1431927615012003 Paper No. 1121 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Highly Accurate Real Space Nanometrology Transmission Electron Microscopy Using Revolving Scanning J. H. Dycus1, J. S. Harris1, X. Sang1, C. M. Fancher1, S. D. Findlay2, A. A. Oni1, T. E. Chan1, C. C. Koch1, J. L. Jones1, L. J. Allen3, D. L. Irving1, J. M. LeBeau1 1. Department of Materials Science and Engineering, North Carolina State University, Raleigh NC 27695, USA 2. School of Physics and Astronomy, Monash University, Victoria 3800, Australia 3. School of Physics, University of Melbourne, Parkville, Victoria 3010, Australia Accurately determining crystallography at the nanoscale provides key understanding of materials behavior. X-ray and neutron based diffraction methods provide highly accurate and precise measurements, but are typically limited in their application for nanoscale materials by poor spatial sensitivity. On the other hand, scanning transmission electron microscopy (STEM) is capable of spatial resolutions below an angstrom, making atomic scale analysis routine. Moreover, high-angle annular dark-field STEM produces images that are directly interpretable with intensities scaling to the atomic number and total number of atoms in a column [1-2]. While, real-space distance measurements are possible with STEM, the effects of thermal drift and scan distortion hinder accurate metrology. In this talk, we will combine revolving STEM (RevSTEM) with a method for scan distortion correction to show accurate and precise real space length measurements for a nanostructured Bi2Te3-xSex alloy. We will show the effects of thermal drift can be corrected via measuring the drift parameters from multiple frames in an image series [3]. By using <100> silicon as a reference standard, we correct the effects from distortions introduced from the scan system, which can then be used for imaging samples of unknown crystallography. The atom columns in drift corrected image series are then indexed and assigned to a matrix representation, which yields information such as the lattice parameters on a unit cell-by-unit cell basis, shown in Figure 1a [4]. To validate the accuracy of the technique, samples of pure Bi2Te3 and Bi2Se3 are analyzed using XRD and the real-space STEM imaging technique. In each case, the error is below 0.1 %. A nanostructured sample of unknown composition, Bi2Te3-xSex, is then investigated with errors again below 0.1 % between XRD and STEM. Further, we will show that the sample composition can then be determined to within 1 at% by Vegard's Law. To relate structure and chemistry, atomic resolution EDS is then performed to determine the site-specific segregation of impurity atoms. The Bi2Te/Se3 crystal structure consists of quintuple layers in the sequence Te(1)-Bi-Te(2)-Bi-Te(1) with a van der Waals interaction between Te(1) columns, shown in Figure 2a. It is observed that Se impurities reside in the Te(2) position in the structure, as shown in Figure 2b. We will then discuss how bond distances and atomic positions can be measured. Finally, we will discuss how the experimentally observed variation in the bond lengths and an anomalously large van der Waals gap can be explained using density functional theory (DFT) [5]. Microsc. Microanal. 21 (Suppl 3), 2015 2246 References: [1] S. J. Pennycook and D. E. Jesson, Physical Review Letters 64 (1990), p 938. [2] J. H Dycus et. al., Applied Physics Letters 102 (2013), p. 081601. [3] X. Sang and J. M. LeBeau, Ultramicroscopy (2014), 28 - 35. [4] X. Sang et. al., Microscopy and Microanalysis (2014), 1764 - 1771. [5] JHD, WX, XS, and JML gratefully acknowledge support from the Air Force Office of Scientific Research (Grant No. FA9550-14-1-0182). SDF and LJA acknowledge support under the Australian Research Council's Discovery Projects funding scheme (DP140102538 and DP110102228). JHD acknowledges the National Science Foundation Graduate Research Fellowship (Grant DGE-1252376). The authors acknowledge the Analytical Instrumentation Facility (AIF) at North Carolina State University, supported by the State of North Carolina and the National Science Foundation. Figure 1. (a) RevSTEM image series of pure Bi2Te3 with boxes representing individual measured unit cells. (b) Histogram of the measured lattice parameters with the same color scale used for the box color in (a). Figure 2. (a) RevSTEM image series of the nanostructured Bi2Te2.7Se0.3 with a schematic of the structure overlaid. (b) Experimental and simulated atomic resolution EDS showing segregation of Se to the Te(2) position.");sQ1[1122]=new Array("../7337/2247.pdf","Atomic Resolution Imaging and Spectroscopy of Pt-alloy Electrocatalytic Nanoparticles","","2247 doi:10.1017/S1431927615012015 Paper No. 1122 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic Resolution Imaging and Spectroscopy of Pt-alloy Electrocatalytic Nanoparticles Sagar Prabhudev,1 Matthieu Bugnet,1 Cory N.Chiang,1 Michael Chatzidakis,1 Guo-zhen Zhu,2 and Gianluigi A. Botton1 1 2 Department of Materials Science and Engineering, McMaster University, 1280 Main St. W., Hamilton, Canada School of Materials Science and Engineering, Shanghai Jiao Tong University, Shanghai, 200240, P.R. China Platinum-alloy nanoparticles (Pt-alloy) are a system of great interest for the fuel cell electrocatalysis. The cathodic reaction (called the oxygen reduction reaction (ORR)) in a typical polymer electrolyte membrane (PEM) fuel cell is notoriously sluggish owing to its multi-electron pathway, requiring efficient catalysts to accelerate the kinetics. Traditionally, Pt was being used in this regard, but despite its exceptional catalytic activity it is now widely accepted to be an unrealistic choice due to depleting sources and exceedingly high costs. In an ongoing attempt to reduce the mass loading of Pt, there has been a tremendous research till date, constantly suggesting that a nanoscale alloying of Pt with 3d transition metals (TM) and the Pt group metals (PGM) is a more viable option. Fine� tuning the surface-structure during synthesis and phase transformations can enhance their catalytic activity and durability by manifolds. However, control and feedback on the synthesis demand characterization techniques that provide atomic-resolution chemical information. The recent developments in electron microscopy pertaining to its instrumentation (aberration correctors, ultrafast electron energy loss spectrometers, monochromators) and the innovation with respect to specimen holders (in situ liquid cell, in situ heating, tomography) have had significant benefits on the study of catalyst nanoparticles by providing chemical analysis and imaging on the atomic-scale [1]. This report describes an overview of recent examples and insightful novel findings related to the study of various catalyst nanoparticles, including Pt-Fe [2], Pt-Au and Pt-Au-Co [3] systems where aberration correction has played a major role in understanding the structures of these materials. Discussed below are two highlights central to our present work. Thermal annealing is one of the prominent modes of creating Pt-enriched surface structures (Figure 1) in the Pt-alloy systems and it is widely known that the catalytic activity is thereby enhanced due to a compressive strain induced on the Pt-shell [2]. However, the nanoscale phase transformation of an ensemble of alloy nanoparticles, polydispersed in both their size and composition, results in an assortment of materials with miscellaneous properties. Controlling their collective evolution and probing the interplay between compositional segregation and atomic-ordering is traditionally carried out via ex-situ approaches by deriving inference before and after annealing. Given the inherent dynamicity associated with nanoscale processes, an atomic-resolved and dynamic perspective has not been possible so far. Here we demonstrate an in situ atomic-resolution imaging and spectroscopy study carried out on the same Pt-alloy nanoparticles, over the course of thermal treatment [4]. While our STEM-HAADF results clearly demonstrate evolution of particle shape, size, ordering and sintering kinetics over the course of heat treatment, the EELS maps reveal new insights into the segregation (Figure 2) and phase separation (Figure 3) process. Structural degradation studies can screen various catalyst designs based on their durability and also, provide insights into the degradation mechanism itself. So far, these studies are solely limited to Microsc. Microanal. 21 (Suppl 3), 2015 2248 interpreting the cyclic voltammograms (during potential cycling). Here, we couple the traditional potential cycling (in a liquid electrolyte) with TEM bright-field imaging [5]. This report describes our recent findings on simultaneous investigation of both structural evolution and electrochemical responses of Pt-Fe nanoalloy catalyst particles, using an in-situ liquid cell inside a TEM. We illustrate how the coarsening mechanisms, including nucleation and growth, are not uniform, both in space and in time scale. For instance, the growth rate was found to be both site- and potentialdependent [6]. References [1] J Liu, ChemCatChem, 3 (2011), p. 934 1 3 2 [2] S Prabhudev et al., ACS Nano, 7 (2013), p. 6103. [3] X Tan et al., ACS Catalysis, 5 (2015), p. 1513. [4] S Prabhudev et al., (submitted). [5] G-z Zhu and S Prabhudev (equal contribution) et al., J. Phys. Chem. C, 118 (2014), p. 22111. [6] This work is supported by the CaRPE-FC network (Automotive Partnership Canada) and the microscopy was carried Pt Fe [110] at the Canadian Centre for Electron Microscopy, a facility supported by NSERC, the Canada Foundation for Innovation and McMaster University. 4 a a b 0 400 50 100 150 200 250 300 c! STEM-HAADF Sca STEM-HAADF intensity (counts) n. 1 [110] 300 200 1 4 Sca n. 2 100 Scan. 2 Scan. 1 0 50 100 5 nm 10 � 150 200 250 300 Pt vs Au 0 ! Pt ! Fe STEM probe position (a. u.) 2 Pt)rich#inner#shell# Fe#outer#shell# 3 d! Scan. 1 Scan. 2 e 600 0 a 50 100 150 200 250 f! b Intensity (arbitrary units) Fe vs Pt 8000 Scan. 2 Intensity (counts) Alloy core 200 Intensity (counts) Pt-rich inner shell Scan. 1 8 � 50 100 150 6000 4000 200 Intensity (arb. units) 5 !!!30000 !!! !!! 20000 10000 400 6 Fe L2,3Figure. 3: STEM-HAADF image (1) and EELS chemical imaging (2,3 and 4) of a thermally Pt M (2) annealed Pt-Au nanoparticle.4,5 and (3), 2D map of Pt and Au, respectively. (4), composite Pt versus Au map. Clearly, these maps evidence a phase separation between Pt and Au, corroborating with the immiscibility gap observed in the bulk phase 800diagram. 900 2000 2200 2400 Energy loss (eV) Pt Au 0 0 250 Distance (pixels) 2000 Figure. 1: a, Atomic resolution STEM-HAADF image of Pt-Fe ordered nanoparticle along [010] orientation. d, Multislice simulation of atomic model shown in c. b and e, Intensity profiles from line scans 1 and 2 corresponding to images in0a and d, respectively. f, 2-D surface relaxation mapping on a catalyst nanoparticle. The 0 600 700 color bar indicates calculated percentage relaxation with respect to bulk. 0.0 0.5 1.0 1.5 2.0 Fe outer a typical shell Distance (nm) b1 2 Fe#shell# 5 3 4 Fe Pt Fe vs Pt Figure. 2: (1), STEM-HAADF of a thermally annealed Pt-Fe particle. (2), Band-pass filtered image of a selected region from (1) Figure 1 red). (3) and (4), 2D EELS map of Fe and Pt, respectively. (5), the composite Fe versus Pt map (marked in | Atomic-resolution imaging and spectroscopy of Pt-Fe nanoparticles annealed at 800�C (1 h). a and b, STEM-HAADF and STEM-EELS compositional maps of two different particles from the heat-treated sample. a(1) and b(1), STEM-HAADF images. Parallel lines in a(1) indicate alternating bright (pink) and dark (blue) intensities. Particle-core in this region is close to [110] orientation. These bright and dark intensities correspond to atomic-columns of Pt and Fe, respectively. Thus, reveals ordering in a(1). a(2) and b(3), 2D EELS map of Fe. a(3) and b(4), 2D EELS map of Pt. a(4) and b(5), the composite Fe versus Pt maps. a(5), EELS line");sQ1[1123]=new Array("../7337/2249.pdf","Atomic-scale EDS Mapping for Chemical Imaging and Quantification of Interdiffusion in Self-assembled Vertically Aligned Nanocomposite Thin Films","","2249 doi:10.1017/S1431927615012027 Paper No. 1123 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic-scale EDS Mapping for Chemical Imaging and Quantification of Interdiffusion in Self-assembled Vertically Aligned Nanocomposite Thin Films Ping Lu1, Eric Romero1, Jon Ihlefeld1, Wei Pan1, Wenrui Zhang2, Haiyan Wang2, and Quanxi Jia3 1 Sandia National Laboratories, P O Box 5800, Albuquerque, NM 87185, USA Materials Science and Engineering Program, Texas A & M University, College Station, TX 77843-3128 USA. 3 Center for Integrated Nanotechnologies, Los Alamos National Laboratory, Los Alamos, NM 87545, USA 2 Recent technical advances in scanning transmission electron microscopy (STEM) and xray detector technology have made it possible to perform atomic-resolution chemical mapping using energy-dispersive x-ray spectroscopy (EDS) [1]. This ability allows for establishing a direct correlation between atomic-scale STEM images, such as high-angle annular dark-field (HAADF) images, and atomic-scale chemical EDS images, making chemical quantification of interfaces, defects and crystalline structure at the atomic-scale possible. Using these capabilities, we have lately quantified the chemical composition of an epitaxial (La0.7Sr0.3)MnO3 (LSMO)/BiFeO3 (BFO) quantum structure [2], cation occupancy in a Sm-doped SrTiO3 (STO) thin film and antiphase boundaries present within STO films [3], as well as determined structures of several intermetallic alloys [4]. In this study, we describe further use of these capabilities to quantify atomic-scale interdiffusion in self-assembled, vertically aligned nanocomposite (VAN) thin films. Self-assembled VAN thin films, consisting of two immiscible components heteroepitaxially grown on single crystals, have experienced an increase in research activity in recent years [5]. These structures offer the advantage of utilizing the functionalities of both components with the possibility of tuning the material's properties by tailoring the volume ratio of the two components, the interface-to-volume ratio and hetero-epitaxial strain. In addition, modification of interface properties such as structural and elemental distribution across the interface offers an additional dimension to generate new properties and functionalities. The atomic-scale characterization of the interface plays an essential role in understanding the structure-property relationships. Here we describe the characterization of NiO:LSMO and ZnO:LSMO VAN thin films. Fig. 1a shows a typical plan-view STEM HAADF image of NiO:LSMO film grown on STO substrate. The self-assembled NiO nanocolumns with an average diameter of 3 ~ 5 nm are evenly distributed in the LSMO matrix. Inset in Fig.1a shows a high-resolution image of a single NiO nanopillar within the LSMO matrix, indicating highly-epitaxial quality of these two phases. A white-contrast, ring-like shell is present around the NiO nanopillars in Fig.1a, signifying the presence of interdiffusion layers between the NiO and LSMO. Fig.1b shows an area used for atomic-scale EDS spectral-imaging. The EDS composite color-map from the area is shown in Fig.1c, along with the line-profile in Fig.1d. Fig.1e shows an excess La EDS map extracted from the spectral-imaging. A similar map is also obtained for Mn. These results indicate presence of both La and Mn-rich shell around the NiO pillars. This interlayer gives rise to the white contrast visible around the NiO pillars in the STEM images (Figs.1a, 1b). In addition, a significant amount of Ni is found in the LSMO matrix. Atomic EDS mapping in Fig.1f shows the Ni occupies the Mn lattice site in LSMO, with the overall composition of Microsc. Microanal. 21 (Suppl 3), 2015 2250 (La0.56Sr0.44)(Mn0.77Ni0.23)O3 in the matrix. A similar interdiffusion profile was found for the ZnO:LSMO VAN film grown on STO substrate, where the ZnO nanopillars were similarly surrounded by a thin La/Mn-rich layer. The details of these results will be presented in the future [6]. References 1. Von Harrach, H.S., Dona, P., Freitag, B., Soltau, H., Niculae, A. & Rohde, M. (2009). Microsc. Microanal. 15 (suppl.2) 208-209. 2. P. Lu, J. Xiong, M. Van Benthem . & Q.X. Jia App. Phys. Lett. 102, 173111 (2013). 3. P. Lu, E. Romero, S. Lee, J. L. MacManus-Driscoll & Q. X. Jia, Microsc. Microanal. 20, 1782-1792 (2014). 4. P. Lu, L. Zhou, M.J. Kramer & D. J. Smith, Sci. Rep. 4, 3945-3949 (2014) 5. Yu, P., Chu, Y.-H. & Ramesh, R. Oxide interfaces: pathways to novel phenomena. Mater. Today 15, 320-327 (2012). 6. Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the US Department of Energy's National Nuclear Security Administration under contract DE-AC0494AL85000. a b c La Sr Mn Ni 10 nm 4 nm d e f La L Sr L 0.39nm Mn K Ni K Fig.1. (a) STEM HAADF of NiO:LSMO VAN, showing the 3-5 nm NiO in LSMO matrix; (b) STEM image showing an area used for EDS spectral-imaging; (c) EDS composite color-map (La-red, Sr-green, Mn-blue, and Ni-cyan); (d) Line-profile across the NiO pillar along the dished line in (c); (e) Excess La map showing a La-rich layer around the NiO; and (f) EDS atomic-scale maps showing the La and Sr occupying the A-lattice site and Mn and Ni occupying the B-lattice site in the perovskite lattice.");sQ1[1124]=new Array("../7337/2251.pdf","Inversion of STEM EELS Data to Obtain Site Occupancy and Near Edge Structure.","","2251 doi:10.1017/S1431927615012039 Paper No. 1124 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Inversion of STEM EELS Data to Obtain Site Occupancy and Near Edge Structure. Mark P. Oxley1,2, Melissa J. Neish3, Leslie J. Allen3 and Matthew F. Chisholm2 1. 2. Department of Physics and Astronomy, Vanderbilt University, Nashville USA. Materials Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, USA. 3. School of Physics, University of Melbourne, Melbourne, Australia Aberration correction has led to routine scanning transmission electron microscope (STEM) images based on electron energy loss spectroscopy (EELS) with atomic resolution [1]. Such images are often obtained from very thin specimens, but not all specimens can be prepared in a way that leads to simple visual interpretation. Even for thin specimens, the interpretation of data is complicated by probe spreading due to both elastic and inelastic scattering processes [2]. Specimens prepared by ion milling are usually of the order of 40-60 nm thick and can be particularly hard to obtain clearly resolved spectrum images, even though the underlying structure is clearly seen in annular dark field (ADF) images. MnFePSi compounds exhibit a giant magnetocaloric effect, which makes them prime candidates for solid state refrigeration. These compounds are based on the hexagonal Fe2P structure where the Fe atoms occupy two distinct sites, with tetragonally coordinated 3f sites and pyramidally coordinated 3g sites, which we shall label Fe1 and Fe2 respectively. The addition of Mn and Si at varying levels can be used to improve the properties of these compounds. Studies using density functional theory and neutron scattering predict the Mn preferentially occupies the Fe2 site [3,4]. In this work we will examine Mn0.43Fe1.57P0.73Si0.27 using real space STEM EELS measurements to determine the position of the Mn atoms and the variation of the Fe L23 ratio between the Fe1 and Fe2 sites. To overcome the effects of both multiple elastic scattering and thermal diffuse scattering of the STEM probe we apply a recently developed inversion technique to obtain the underlying inelastic scattering potential from STEM EELS data obtained using an aberration corrected Nion UltraSTEM operating at 200 kV [5]. In Fig. 1 we show simultaneously acquired ADF (a) and EEL spectrum images based on the Fe L shell (b) and Mn L shell (c). The hexagonal structure is clearly seen in (a) and the structure is overlaid on an ADF simulation seen in (d). A probe forming aperture of 30 mrad. and an EELS collection semi-angle of 36 mrad. have been used. The specimen thickness was determined to be 65 nm from low-loss EELS data. While the positions of the Fe columns may be inferred from Fig. 1(b), the Mn L spectrum image in (c) does not provide any visual identification of the Mn sites. After the inversion procedure, the Fe atomic columns can be clearly seen in the recovered scattering potential shown in Fig. 1(e) with the Fe1 columns identified by the blue circles and the Fe2 columns by the red circles. Similarly the Mn L scattering potential shown in 1(f) clearly shows the preferential occupancy of the Fe2 sites by the Mn impurities. Not only are we able to determine the location of the Mn impurities, but the near edge structure of the different Fe sites can also distinguished. In Fig. 2(a) the averaged Fe L shell spectrum obtained directly from the experimental data is shown for both the Fe1 (blue lines) and Fe2 columns (red lines). There is a slight reduction in the Fe1 L3 peak compared to that of the Fe2 spectrum. This difference becomes far more distinct in the averaged spectra obtained from the Fe L-shell potentials obtained by inversion shown in Fig 2(b). Microsc. Microanal. 21 (Suppl 3), 2015 2252 This technique provides real space visualization of site occupancy and unlike diffraction based techniques, promises local site identification rather than averaged site occupancies. References: [1] M Bosman et al, Phys. Rev. Lett. 99 (2007) 086102. [2] B D Forbes et al., Phys. Rev. B 86 (2012), 024108. [3] N H Dung et al., Adv. Energy Mater. 1 (2011), p. 1215. [4] Z Q Ou et al., Magn, Magn. Mater. 340 (2013), p. 80. [5] N. Lugg et al., Appl. Phys. Lett. 101 (2012), 183112. [6] This research was supported under the Discovery Projects funding scheme of the Australian Research Council (Project No. DP110102228). This work was sponsored by US Department of Energy, Office of Science, Basic Energy Sciences, Materials Sciences and Engineering Division and supported in part by DOE Grant No. DE-FG02-09R46554. Figure 1. ADF image (a) and simultaneously acquired Fe L-shell and Mn L-shell spectrum images, (b) and (c) respectively, of Mn0.43Fe1.57P0.73Si0.27. (d) Simulated ADF image with structure inset. Recovered EELS scattering potentials for Fe L-shell and Mn L-shell ionization are shown in (e) and (f) respectively. Figure 2. Fe1 L-shell (blue lines) and Fe2 L-shell (red lines) obtained from the raw data (a) and the inversion process (b).");sQ1[1125]=new Array("../7337/2253.pdf","Periodic Artifact Reduction in Fourier Transforms of Full Field Atomic Resolution Images","","2253 doi:10.1017/S1431927615012040 Paper No. 1125 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Periodic Artifact Reduction in Fourier Transforms of Full Field Atomic Resolution Images Robert Hovden1, Yi Jiang2, Huolin L. Xin3 and Lena F. Kourkoutis1,4 1 2 School of Applied & Engineering Physics, Cornell University, Ithaca, NY, USA. Department of Physics, Cornell University, Ithaca, NY, USA. 3 Center for Function Nanomaterials, Brookhaven National Laboratory, Upton, NY, USA. 4 Kavli Institute at Cornell for Nanoscale Science, Ithaca, NY, USA. Modern electron microscopists have the luxury of studying the atomic structure of materials directly through real space imaging. With sub-Angstrom resolution that distinguishes atomic columns in a crystal, one can count the atoms across a grain boundary, identify the structure of interfaces, and locate individual defects and dopants. However, real-space analysis of crystals is not always sufficient and the discrete Fourier transform (DFT) of atomic resolution images is routinely used to mimic a diffraction pattern--with reciprocal lattice spots reflecting the symmetry and spacing of a specimen's crystal structure. However, when calculating a Fourier transform, periodic boundary conditions are imposed and sharp discontinuities between the edges of an image cause a cross-patterned artifact along the reciprocal space axes (seen in Fig. 1d). Traditionally, these DFT artifacts are reduced by applying a damping window to the real space image. However, windowing is not a universal approach--the window function can influence the resolution, shape, and intensity of spots in Fourier space [1]. More critically, by weighting the real space image the center of the window is overrepresented and important features toward the edges of the field of view can be missed. Figure 2 shows a LaVO3/SrTiO3 multilayer with one layer exhibiting a super-lattice structure. From the real-space image (Fig. 2a), this super-lattice is difficult to detect, but it clearly appears as additional peaks in the Fourier transform (Fig. 2d). In the windowed transform (Fig. 2e), however, this structure would be missed entirely--even when applying the relatively large Hann window (Fig. 2b). Here we demonstrate that the recently developed Periodic Plus Smooth Decomposition (P+S) technique provides a simple, efficient method for reliable removal of artifacts caused by edge discontinuities in atomic resolution S/TEM images (Fig. 1) [2]. In this method, edge discontinuities are reduced by subtracting a smooth background determined by solving Poisson's equation with boundary conditions set by the image edges [3]. The resulting Fourier transform reflects the original image's DFT, but without the cross pattern artifact (Fig. 1e). Unlike windowed Fourier transforms, P+S Decomposition maintains sharp reciprocal lattice peaks and uses all information from the image's field of view--shown in Fig. 2f, the super-lattice peaks are preserved. This approach provides an accurate estimation of the spectrum in regions where reciprocal lattice peaks occur. Since the Fourier transform is a linear operator, the DFT of the smooth component (Fig. 1f) shows the exact intensity removed from the unprocessed DFT. In a field where Fourier transforms are among the most common analysis tools, P+S Decomposition can provide substantial utility for aberration-corrected S/TEM. [4] [1] F.J. Harris, Proc IEEE 66, 51�83 (1978). [2] R. Hovden et. al, Microsc. Microanal., doi:10.1017/S1431927614014639 (2015). [3] L. Moisan, J. Math Imaging Visualization, 39 161-179 (2013). [4] This work used the CCMR EM facility supported by NSF MRSEC program (DMR-1120296). Microsc. Microanal. 21 (Suppl 3), 2015 2254 a. 2nm b. c. = + d. STEM Image e. Periodic Component f. Smooth Component = + DFT Moisan DFT DFT of Smooth Comp. Figure 1. P+S Decomposition of CdSe [111] STEM image. Discontinuity in the periodic boundaries manifest as cross pattern artifact in the DFT (d). The P+S Decomposition creates periodic boundary continuity (b) by removing a smooth background component (c) from the image. This removes the edge artifacts (e). The background component shows exactly what has been removed (f) to create the artifact free DFT (e). DFT's are log-absolute; scan direction is 26.5� clockwise from vertical. a. r Super-Lattice Laye b. r Super-Lattice Laye c. Super-Lattice Laye r 30 � d. Unprocessed superlattice spots e. Hann Window Periodic Decomp. spots missing f. superlattice spots Figure 2. LaVO3/SrTiO3 with one layer showing a super-lattice structure (a) that appears as additional peaks in the DFT (d). A windowed DFT heavily weights the center of the image (b) and consequently misses the super-lattice peaks (e). P+S Decomposition utilizes the entire field of view (c), preserving the atomic superlattice structure in the DFT (f). Sample grown by the group of H.Y. Hwang (Stanford University).");sQ1[1126]=new Array("../7337/2255.pdf","Aberration-corrected STEM of Four-atom Rhenium Nanowires Confined within Carbon Nanotubes","","2255 doi:10.1017/S1431927615012052 Paper No. 1126 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Aberration-corrected STEM of Four-atom Rhenium Nanowires Confined within Carbon Nanotubes Fan Zhang1, Pengju Ren1, Xiulian Pan1, Xinhe Bao1 and Jingyue Liu2 1. State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian 116023, China. 2. Department of Physics, Arizona State University, Tempe, Arizona 85287, USA. Due to the spatial confinement effect nanowires (NWs) of just a few atoms thick in diameter can possess novel quantum properties. These ultrathin NWs, however, tend to be both chemically and structurally unstable. Carbon nanotubes (CNTs), on the other hand, can be used as templates to synthesize nanostructures that remain stable [1-5]. For example, nanofilling of Eu can produce single-atom chains in double-wall CNTs (DW-CNTs) [3]. Self-assemble of graphene nanoribbons within single-wall CNTs and the helical twist and screw-like motion of the carbon nanoribbons have been observed [4]. Monocrystalline FeCo NWs inside CNTs with unique magnetic properties have been synthesized [5]. The strong metal-CNT interaction and the confinement effects of small diameter CNTs can induce the formation of novel metal phases that are generally not stable. We report here the discovery of selfassembled, ultra-long and atomically thin Re NWs with an unusual fcc-stacking pattern along the length of the CNTs. Re usually possesses a hcp structure which is extremely stable and no phase transition occurs under pressures to 216 GPa and temperatures up to its melting point [6]. Purified DW-CNTs with inner diameters < 1.5 nm were selected as templates to grow Re NWs. Volatile methyltrioxorhenium (Re(CH3)O3) was used as the precursor material to fill in the CNT channels. After being washed, dried, reduced in H2 for 2 h at 823 K, and annealed in He for 12 h at 473 K, the encapsulated organometallic Re species transformed into clusters dispersed within the CNT channels. Upon repeating the above annealing cycles for 3 times, the discrete Re nanoclusters self-assembled into continuous NWs within the CNT channels. Repeated annealing cycles produced longer Re NWs. Aberration-corrected STEM (AC-STEM), especially high-angle annular dark-field (AC-HAADF), technique was used to image the atomic arrangement of the ultrathin Re NWs within the CNT channels. Figure 1 shows a typical atomic resolution HAADF image of a Re nanowire, self-assembled within DWCNTs. The bright dots represent Re atoms. Some individual Re atoms (indicated by the white arrow in Fig. 1a), were also observed. By using the intensity of the individual Re atoms as an internal standard, one can deduce that the brighter dots in the center of the Re NW (Fig. 1a) contains two overlapping Re atoms and the dim dots on each side of the central brighter line of dots contain only one Re atom. Detailed analysis of many similar Re NWs (e.g., Fig 1b and 1c) suggests that the stacking of the Re atoms follows closely to that of the fcc Re projected along the [110] zone axis with the [1-10] direction along the length of the CNT. Figure 1c shows an interesting Re NW with different arrangement of the Re atoms along the CNT length. By comparing to the projections of a four-atom fcc Re NW (oriented with the [1-10] direction along the CNT length), it was identified that the different segments of the re NW shown in Fig. 1c correspond to the four-atom fcc Re NW projected along the [110] (indicated by the solid block arrow), the [111] (indicated by the hollow block arrow), and the [112] (indicated by the red arrow) zone axis, respectively. It is not clear why the Re NW existed in different configurations within the CNT. Stacking defects are, however, observable along the ultrathin Re NWs. DFT calculations revealed that the strong interaction between the Re atoms and the C atoms of the CNTs Microsc. Microanal. 21 (Suppl 3), 2015 2256 played a vital role in determining the formation of the fcc Re NW. Figure 2a shows the optimized model of Re4@(7, 7). The calculation of the lattice parameters suggests that the encapsulated Re fcc structure is distorted, i.e. compressed along [001] and stretched along [1-10], which implies an extremely strong interaction between the Re atoms and the C atoms of the CNT walls. Figure 2b and 2c shows, respectively, the electron density around the vicinity of the different Re atoms. The #1 and #3 Re atoms donate electrons to the C atoms and the other two Re atoms #2 and #4 do not show obvious electron transfer. This uneven interaction leads to a distorted Re fcc structure and causes deformation of the CNTs into a slight oval shape when viewed along its length. The inner spaces of CNTs can be used as nanoscale reactors to synthesize novel structures with unique properties. References: [1] J. Lee et al, Nature 415 (2002), p.1005. [2] L.-J. Li et al, Nature Materials 4 (2005), p. 481. [3] R Kitaura et al, Angew. Chem. Int. Ed. 48 (2009), p. 8298. [4] A. Chuvilin et al, Nature Materials 10 (2011), p. 687. [5] A. L. El�as et al, Nano Lett. 5 (2005), p. 467. [6] Y. K. Vohra et al, Phys Rev B 36 (1987), p. 9790. [7] The authors acknowledge the use of facilities in the John M. Cowley Center for High Resolution Electron Microscopy at Arizona State University. Figure 1. HAADF images of self-assembled, four-atom Re NWs within double-wall carbon nanotubes. Figure 2. DFT calculations of the interaction of the four-atom fcc Re NW with the carbon atoms of the double-wall CNT. The side scale bars indicate the degree of electron transfer from Re to C.");sQ1[1127]=new Array("../7337/2257.pdf","Collective Atomic Motion at a 90� <110> Tilt Grain Boundary in Gold","","2257 doi:10.1017/S1431927615012064 Paper No. 1127 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Collective Atomic Motion at a 90� <110> Tilt Grain Boundary in Gold M.L. Bowers1, A. Gautam1, C. Ophus1, F. Lan�on2, U. Dahmen1 1. 2. Molecular Foundry, National Center for Electron Microscopy, LBNL, Berkeley, CA 94720, USA Laboratoire de Simulation Atomistique (L_Sim), SP2M, INAC, CEA, 38054 Grenoble, France Grain boundaries in crystalline materials affect many macroscopic properties such as strength, electrical resistivity, and corrosion resistance. Processing techniques to enhance these properties are therefore often aimed at modifying the existing grain boundary content via processes such as recrystallization and grain growth. While much is known about the structural character of grain boundaries and the variables that affect their mobility, the atomic-scale mechanisms of migration are still poorly understood. Recent experimental work suggests that step nucleation and cooperative, string-like motion of many atoms near a boundary may contribute to its advancement [1,2]. Molecular dynamics simulations indicate that these atomic cascade events can be triggered by volume fluctuations at the grain boundary and may occur in the absence of an external driving force [3]. In this contribution, we investigate step propagation and collective atomic motion at an incommensurate 90� <110> tilt grain boundary in gold. The samples used in this study were prepared by physical vapor deposition of high purity Au onto <100> Ge substrates. This results in a {110} mazed bicrystal thin film where all grains are rotated 90� about a common <110> axis. Characteristic segments of these boundaries were imaged using the aberration-corrected TEAM microscopes at the National Center for Electron Microscopy in both HAADF-STEM and HRTEM modes at 300kV. Imaging parameters such as beam current and dwell time were systematically varied to optimize the detection of transient events, as well as to investigate the effect of the electron beam on event frequency. Large time series were acquired to allow for cumulative averaging of frames between events and enable precise atomic displacement measurements for comparison to simulations [4]. Custom MATLAB routines were employed to track and analyze structural fluctuations near the interface. Figure 1(a) shows an intensity-averaged image of an incommensurate {110}/{001} boundary segment that contains two steps in the boundary plane. The five-fold structural units that are characteristic of this boundary are outlined in yellow. Over the course of imaging, many structural fluctuations are observed at ambient temperature under the influence of the electron beam. Figure 1(b) shows a red/green/blue (RGB) overlay of frames before, during, and after one such event. The white atoms indicate sites that have not moved over the course of the experiment, while red and blue illustrate the trajectory of atoms during the cooperative motion. This clearly shows a string-like motion of atoms in the lower grain that serves to advance the boundary upward. This structural fluctuation persisted for several seconds and changed the local structure of the boundary segment as shown in Figure 2. The boundary steps have been defined in terms of the number of planes involved on each side of the step, n:m, where n is the number of {001} planes and m {110} planes. This structural analysis suggests that the observed collective motion of atoms consolidates a 1:1 and a 2:3 step into a single 3:4 step. The implications of the observed structural transitions for grain boundary migration will be presented. Molecular dynamics simulations are used to further clarify the experimental observations and to explore the z-component of the transition. The defect character of the atomic shuffles as well as consideration of intrinsic factors, such as stacking fault energy, are utilized in the description of this migration mechanism. Microsc. Microanal. 21 (Suppl 3), 2015 2258 References [1] Merkle KL, Thompson LJ. Mater Lett 2001:188. [2] Merkle KL, Thompson LJ, Phillipp F. Phys Rev Lett 2002;88:225501. [3] Zhang H, Srolovitz DJ. Acta Mater 2006;54:623. [4] Gautam A, Ophus C, Lan�on F, Denes P, Dahmen U. Ultramicroscopy 2015, in press. [5] Work at the Molecular Foundry was supported by the Office of Science, Office of Basic Energy Sciences, of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231 Figure 1: (a) Intensity average of 40 HAADF-STEM fast scan (0.5�s dwell time) images showing an incommensurate boundary segment containing two steps. Five-fold structural units are outlined in yellow. (b) RGB overlay of averaged intensities before, during, and after a structural event to highlight the atomic displacements. Figure 2: Grain boundary structure before (a) and after (b) collective rearrangement, colored by local coordination. Green and red correspond to the unique grain orientations, while yellow and orange show local distortions near the boundary. The boundary is highlighted for clarity.");sQ1[1128]=new Array("../7337/2259.pdf","Electron Holography Unveiled Formation Process of Extraterrestrial Magnetite","","2259 doi:10.1017/S1431927615012076 Paper No. 1128 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electron Holography Unveiled Formation Process of Extraterrestrial Magnetite Yuki Kimura1, Kazuo Yamamoto2 and Takeshi Sato3 1. 2. Institute of low temperature science, Hokkaido University, Sapporo, Japan. Nanostructures Research Laboratory, Japan Fine Ceramics Center, Nagoya, Japan. 3. Hitachi High-Technologies Corporation, Hitachinaka, Japan. Primitive minerals formed in the solar nebula coagulated into asteroids together with some water ice at the beginning of our Solar system. Interior temperature of an asteroid elevates above the melting point of water ice as a result of mainly the decay heat of radiogenic aluminium-26 [1] and possibly highly energetic impacts [2]. Then, primitive minerals were altered by aqueous processes. The degree of alteration depends on the size of asteroid and the time of asteroidal formation. Therefore, final products in a meteorite allow us estimation of the history of the early solar system. Magnetite particles founding in meteorites has been known as resultants of aqueous alteration in asteroids. Tagish Lake meteorite, which originated from a D-type asteroid located near the orbit of Jupiter [3], has a lots of magnetite particles with diameters of 110�680 nm in a significant form. Thousands to ten thousands of the magnetite particles have been assembled in three-dimensionally as colloidal crystals with spherical shape [4]. Whereas framboidal magnetite, which is aggregate of magnetite particles, has been found in several carbonaceous meteorites, Tagish Lake meteorite is only a case for discovery of colloidal crystal. Colloidal crystal is only able to form by an appropriate repulsive force between the particles in a solvent with limited space. If the magnetite particles were typical magnets, they would have readily coagulated as a result of the strong attractive magnetic interactions and they would not have produced a well-ordered structure [5]. Therefore, it is very surprising that how they arranged in three dimensionally and preserved the structure for 4.6 billion years. The residual magnetization of rocks and minerals is very sensitive to the chronology of their formation and to the temperatures that they have experienced [6]. By means of electron holography, we applied the paleomagnetic method at a nanometer scale (resolution ~6 nm) to magnetite particles isolated from colloidal crystals from the Tagish Lake meteorite and we inferred the occurrence of water in the parent body and formation process of magnetite particles. Electron holography permits the visualization of nanometer-scale magnetic properties and internal electric potentials of samples by recording electron-interference patterns (holograms). We used a transmission electron microscope (Hitachi HF3300-EH) in Japan Fine Ceramics Center. The TEM has been designed specifically for electron holography and has additional sample stage kept in a magnetic field-free environment. We selected colloidal crystals of magnetite together with two, three, four or six particles. The samples were not exposed to any artificial magnetic field at any stages during the experiments. The magnetite particles extracted from a colloidal crystal have a vortex structure with no external magnetic field, unlike common ferromagnetic materials. There is no report of any natural occurrence of such a vortex structure, although a theoretical prediction based on a three-dimensional micromagnetic simulation has been made of the existence of a uniformly magnetized single-domain vortex state at about 100 nm [7]. We summarized their circulating magnetic property permits the formation and Microsc. Microanal. 21 (Suppl 3), 2015 2260 preservation of the three-dimensional colloidal crystals of magnetite in a water droplet, which is required for arrangement by an appropriate repulsive force between the particles. The distribution of colloidal crystals in a small area and its shape suggest that many isolated liquid droplets were distributed in the asteroid and that large-scale assemblies of fluids were absent at least during the nucleation of the particles and the formation of the colloidal crystals. The uniformity of the size distribution and the similar morphology of the framboidal magnetite particles in the colloidal crystals suggest they were formed through homogeneous nucleation from a highly supersaturated solution in a single event. The solution can be formed by following scenario: Calcium and magnesium in a solution initially precipitate as a carbonate during changes in the aqueous environment of the asteroid. The resulting solution is then relatively enriched in iron ions, which was able to spatter by shaking of parent asteroid due to collision of other asteroid and flowed in a cavity of interior the parent asteroid. Recently, we started in-situ heating experiments to obtain more direct information about the formation history including the temperature and its period of the magnetite formation in a transmission electron microscope. In the presentation, we will show the primitive results. References: [1] M. Endress, et al., Nature 379 (1996), p. 701. [2] A. F. Rubin, Geochim. Cosmochim. Acta 90 (2012), p. 181. [3] T. Hiroi, et al., Science 293 (2001), p. 2234. [4] J. Nozawa et al., J. Am. Chem. Soc. 133 (2011), p. 8782. [5] A. P. Philipse, D. Maas, Langmuir 18 (2002) p. 9977. [6] B. P. Weiss et al., Science 290 (2000), p. 791. [7] M. Winklhofer et al., J. Geophys. Res. 102 (1997) p. 22695. Figure 1. TEM images of magnetite particles in a portion of a colloidal crystal recorded by Hitachi, HF3300-EH (300 kV acceleration voltage). (a) Bright-field image. The scale bar is 100 nm. (b) Phase image of (a) showing that the particles have a concentric circular magnetic field and a vortex structure. Arrows indicate the directions of the magnetization vectors in each magnetite particle. (Reproduced from Nature Communications with permission, doi: 10.1038/ncomms3649)");sQ1[1129]=new Array("../7337/2261.pdf","Focused Ion Beam Nanotomography of Chondritic Meteorites: Closing the Mesoscale Length Gap in Paleomagnetic Studies","","2261 doi:10.1017/S1431927615012088 Paper No. 1129 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Focused Ion Beam Nanotomography of Chondritic Meteorites: Closing the Mesoscale Length Gap in Paleomagnetic Studies Joshua F. Einsle1, Roger R. Fu2, Benjamin P. Weiss2, Takeshi Kasama3, Karl Fabian4, P�draig � Conbhu� 5, Wyn Williams5, Paul Midgley6, Richard J. Harrison1 1. 2. Department of Earth Sciences, University of Cambridge, Cambridge UK, Department of Earth, Atmospheric and Planetary Sciences, MIT, Cambridge, MA, USA, 3. Center for Electron Nanoscopy, Technical University of Denmark, Kongens Lyngby, Denmark, 4. Geological Survey of Norway, Trondheim, Norway, 5. School of GeoSciences, University of Edinburgh, Edinburgh, UK, 6 Department of Materials Science and Metallurgy, University of Cambridge, Cambridge UK. One of the greatest limits on the reliability of paleomagnetic information comes from the decreasing fidelity of signal with increasing age of the rock. The older the rock, the more complex is its geological history, and the more likely it is to have experienced conditions that altered, or even destroyed, the primary magnetic signal. This problem gets dramatically worse when it comes to extraterrestrial materials, such as meteorites, that predate the formation of the Earth itself. Nanoscale magnetic minerals sitting inside single silicate crystal hosts carry the most reliable paleomagnetic signals in meteorites. These small magnetic carriers adopt highly stable magnetisation states and are protected from alteration by their surrounding silicate. To interpret such paleomagnetic measurements, and assess their reliability, we need to relate the macroscopic remanence observations to the underlying magnetic properties of individual magnetic particles and to the collective magnetic behavior of the particle ensemble as a whole. Here we outline a combined experimental and computational methodology to quantify the 3D properties of magnetic particle ensembles at the mesoscopic length scale. Electron holography and tomography of individual particles are combined with FIB slice-and-view tomography to provide a complete characterization of the particle ensemble in terms of the domain state, volume, shape and interparticle spacing. This information is used as input to finite-element micromagnetic and kinetic Monte Carlo simulations to obtain a quantitative understanding of the system's response to changes in magnetic field, temperature and time. The Semarkona LL3.0 ordinary chondrite meteorite provides a case study to develop our new methodology. Single grains of the silicate mineral olivine (Mg2SiO4), containing thousands of submicron inclusions of metallic iron, have recently been used to determine the strength of the magnetic field in the early solar nebula [1]. These `dusty olivine' grains were first measured using a scanning SQUID microscope, enabling their magnetic remanence to be isolated from that of the surrounding meteorite matrix. Using holography and Fresnel-mode Lorentz imaging we confirmed the dominance of single vortex (SV) states in the majority of the remanence carriers (Fig 1b and 1c). In-field measurements demonstrated the high stability of this SV state, making them suitable carriers of paleomagnetic information. Focused Ion Beam nanotomography (FIBnT) was then used to reconstruct a 10 �m x 13 �m x 5.6 �m region from a single grain of dusty olivine (Fig 2a). By acquiring backscattered electron images for each 20 nm slice, we produced a quantitative reconstruction of the size, shape, orientation and distribution of magnetic Fe nanoparticles in the dusty olivine. This data was then used to produce highly detailed finite element micromagnetic models of individual particles, as well as a micromagnetic reconstruction of the entire ensemble.[2, 3]. This `nanopaleomagnetic' approach demonstrates for the first time how correlative microscopy, linking information obtained at the Microsc. Microanal. 21 (Suppl 3), 2015 2262 nano-micro-mm scale, can be used to obtain a physical understanding of the paleomagnetic signals carried by natural materials over 4.5 billion years old. References: [1] R. R. Fu, et al, Science, 346 (2014), p. 108. [2] T. P. Almeida, et al, Geophys. Res. Lett., 41 (2014), p. 1. [3] R. J. Harrison and I. Lascu, Geochemistry, Geophys. Geosystems, 15 (2014), p. 1. [4] The authors acknowledge funding from the ERC under ERC grant agreement 320750Nanopaleomagnetism and ERC grant agreement 291522 � 3DIMAGE. Figure 1. (a) SEM micrograph of the single grain of dusty olivine being studied. Rough dark gray material is mounting epoxy, while bright regions are the exposed olivine. (b) Magnetic induction map for a single 200 nm Fe particle in the vortex state. Figure 2. (a) 3D reconstruction of Fe nanoparticles for dusty olivine grain. (b) FEM modeling results for 400 nm Fe particle closest to the prolate population mean. Color scale shows helicity of magnetic field.");sQ1[1130]=new Array("../7337/2263.pdf","Correlating Mineralogy and Amino Acid Contents of Milligram-Scale Murchison Carbonaceous Chondrite Samples","","2263 doi:10.1017/S143192761501209X Paper No. 1130 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Correlating Mineralogy and Amino Acid Contents of Milligram-Scale Murchison Carbonaceous Chondrite Samples Aaron S. Burton1, Eve L. Berger2, Darren R. Locke3, Jamie E. Elsila4, Daniel P. Glavin4 and Jason P. Dworkin4 1. NASA Johnson Space Center, Astromaterials Research and Exploration Science Division, Houston, TX, USA 2. GeoControl Systems Inc � Jacobs JETS contract, NASA Johnson Space Center, Houston, TX, USA 3. HX5 � Jacobs JETS contract, NASA Johnson Space Center, Houston, TX, USA 4. NASA Goddard Space Flight Center, Solar System Exploration Division, Greenbelt, MD, USA Amino acids, the building blocks of proteins, have been found to be indigenous in most of the carbonaceous chondrite groups [1 and references therein]. The abundances of amino acids, as well as their structural, enantiomeric and isotopic compositions differ significantly among meteorites of different groups and petrologic types [e.g., 2, 3]. This suggests that there is a link between parent-body conditions, mineralogy and the synthesis and preservation of amino acids (and likely other organic molecules). However, elucidating specific causes for the observed differences in amino acid composition has proven extremely challenging because samples analyzed for amino acids are typically much larger (~100 mg powders) than the scale at which meteorite heterogeneity is observed (sub mmscale differences, ~1-mg or smaller samples). Thus, the effects of differences in mineralogy on amino acid abundances could not be easily discerned. Recent advances in the sensitivity of instrumentation have made possible the analysis of smaller samples for amino acids [4, 5], enabling a new approach to investigate the link between mineralogical con-text and amino acid compositions/abundances in meteorites. Through coordinated mineral separation, mineral characterization and highly sensitive amino acid analyses, we have performed preliminary investigations into the relationship between meteorite mineralogy and amino acid composition. By linking amino acid data to mineralogy, we have started to identify amino acid-bearing mineral phases in different carbonaceous meteorites. The methodology and results of analyses performed on the Murchison meteorite are presented here. All ceramic, glass and aluminum materials that were used for the study were heated in air at 500 �C for 18-24 hours prior to use to remove organic residue. Tweezers and forceps used for mineral picking were first subjected to sonication in 100% methanol, a 50/50 mixture of methanol and water, and 100% water. The cleaning and chromatography solvents used for this study were all high-performance liquid chromatography grade; ultrapure water (18.2 M, <4 ppb total organic carbon) was obtained from a Millipore Advantage A-10 water purification system. A ~20 mg sample of the Murchison meteorite was used for this study. The sample was gently crushed with a porcelain mortar and pestle, and portions transferred to a glass slide. Material on the slide was observed under a petrographic microscope, and grains were hand-picked by visual appearance (texture, crystal shape, opacity, etc.) to separate "matrix" and "non-matrix" components. We then collected the following samples: 1) bulk, unseparated material (2.8 mg); 2) a single grain containing matrix, non-matrix and sulfide materials (2.5 mg); 3) hand-picked grains containing predominantly matrix (1.2 mg); and 4) hand-picked non-matrix grains (1.0 mg). The samples were hot-water extracted for 24 hr at 100 �C in sealed glass ampoules, and the supernatant containing amino acids was removed and dried down under vacuum. Acid vapor hydrolysis and amino acid analysis were performed as described elsewhere [1, 3], except that liquid chromatography was performed on a Thermo-Dionex UHPLC 3000 Liquid Chromatograph, and amino acids were analyzed Microsc. Microanal. 21 (Suppl 3), 2015 2264 by UV-fluorescence detection (UHPLC-FD) and identified by comparing their retention times with those of known standards, and well previous amino acid analyses of the Murchison meteorite. The extracted meteorite sample residues were analyzed by scanning electron microscopy (SEM). Both back scattered electron images and energy dispersive Xray data were collected on a JEOL JSM-7600 field emission SEM. The matrix material of the Murchison meteorite was comprised primarily of fine-grained phyllosilicates, whereas our non-matrix fraction consisted of mostly mafic silicate minerals (Figure 1). Our expectation was that amino acids would be located in the matrix, and that the non-matrix fraction would be devoid of amino acids. This was only partially correct, however. Analysis of the matrix-containing samples (2.8 mg bulk, single grain, and matrix separates; Figure 2) revealed amino acid distributions that were fairly consistent with each other, with glycine as the most abundant amino acid, followed in descending abundance by amino-n-butyric acid, -alanine, -aminoisobutyric acid and -amino-n-butyric acid. Interestingly, the non-matrix material also contained amino acids but with a distinct composition from that of the matrix fractions; the non-matrix contained a large amount of glycine with much lower levels of the other amino acids. In comparison with a typical, relatively large-mass extraction of the Murchison meteorite (>100 mg), the overall amino acid abundances were comparable to the matrix fraction. However, the amino acid distributions were visibly different. In the large sample, -aminoisobutyric acid was the most abundant amino acid, followed in abundance by glycine, whereas in the small Murchison samples analyzed here, glycine and then -amino-n-butyric acid were the two most abundant amino acids of this set. Further studies are needed to understand the cause(s) of these observed differences in amino acid abundances and distributions. References: [1] Burton et al. Met. Plan. Sci. (in press) [2] Elsila et al. Met. Plan. Sci. 47 (2012) 1517-1536 [3] Glavin et al. Met. Plan. Sci. 45 (2010) 1948-1972 [4] Callahan et al. J. Chrom. A 1332 (2014) 30 - 34 [5] Glavin et al. Met. Plan. Sci 41 (2006) 889 - 902 [6] We acknowledge M. Zolensky (NASA JSC) for the Murchison sample. ASB acknowledges support from the NASA Early Career Fellowship and NASA Exobiology programs. Figure 1. The upper panels of this figure are back-scattered electron (BSE) images of the fractions post- hot water extraction (matrix on the right, non-matrix on the left). Below each BSE is a corresponding x-ray map. (Fe=red, Mg=green, Ca=blue, Al=cyan, Ti=magenta, S=yellow). The non-matrix fraction consists mostly of mafic silicate minerals with iron contents ranging between 2 and 40 wt.%. The matrix portion differs from the non-matrix in shape and chemistry. These post-extraction analyses serve as a quality check on the separation process, and confirm the identities of the amino acid hosts in each fraction. Figure 2. Comparison of selected amino acid abundances in samples from the Murchison meteorite. Amino acid abbreviations are: gly = glycine, bala = -alanine; gaba = -amino-nbutyric acid, aib = -aminoisobutyric acid, aaba = -amino-n-butyric acid.");sQ1[1131]=new Array("../7337/2265.pdf","Nanoscale Variation in Carbonaceous Matter from Primitive Meteorites Revealed by Aberration-Corrected STEM.","","2265 doi:10.1017/S1431927615012106 Paper No. 1131 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Nanoscale Variation in Carbonaceous Matter from Primitive Meteorites Revealed by Aberration-Corrected STEM. Bradley T. De Gregorio1, Rhonda M. Stroud2, Kate D. Burgess2,3 and Conel M. O'D. Alexander4 1. 2. Nova Research, Inc., Alexandria, VA USA. Materials Science and Technology Division, Naval Research Laboratory, Washington, DC USA. 3. ASEE NRL Postdoctoral Fellowship Program 4. Department of Terrestrial Magnetism, Carnegie Institution of Washington, Washington, DC USA. The majority of the organic matter (OM) at the surface of the early Earth was delivered by primitive meteorites and comets. By studying the characteristics of OM in meteorites and/or cometary dust, we can gain insight into the overall galactic pathway(s) of organic precursor molecules present in the early Solar Nebula (and in the collapsed molecular cloud from which it formed) that eventually provided components for prebiotic chemistry on Earth. In particular, we are interested in robustly distinguishing the signatures of OM that are representative of various asteroidal, nebular, and pre-nebular (e.g., molecular cloud) processes [1]. However, distinguishing these signatures is non-trivial and requires analysis of fine-scale heterogeneities to separate features with unique chemical signatures [2, 3]. To achieve the required spatial and spectral resolution, previous work has mostly relied on coordinated, multi-instrument analysis, such as XANES (characterization of C bonding at 0.1 eV energy resolution) + TEM (observation of sub-nm morphology) [3, 4]. TEM-based EELS is generally not suitable for these studies as typical operating conditions (e.g., 200 keV with a 200pA probe) can cause observable changes in the organic functional chemistry of meteorite OM. The feasibility of using an alternative "gentle" STEM approach (i.e., 60 keV in a UHV aberration-corrected microscope) was recently demonstrated [5]. In this study, we demonstrate the advantages of using the "gentle" STEM approach with both EELS and EDS for faster, more complete characterization of meteorite OM. Three insoluble organic matter (IOM) residues, created by acid dissolution of silicate minerals from the DOM08006, EET92042, and Murchison meteorites [1], were selected to represent different asteroidal processing regimes. Particles of IOM were embedded in S and then ultramictrotomed to obtain 30 nm sections, which were placed on lacey C TEM grids. Analyses were carried out at 60 keV with the Nion UltraSTEM200 at NRL, equipped with a Gatan Enfinium ER EEL spectrometer and a Bruker windowless, SDD x-ray spectrometer (~0.6 sr). Hyperspectral EELS datasets at the C K-edge were acquired from 250 nm � 250 nm regions at 8 nm/pixel, with a measured resolution (FWHM) of 0.30 eV. Subsequently, hyperspectral EDS maps were collected from the same region. The combined analysis time for both techniques was less than 45 min, with no detectable changes to the sample chemistry. With this EDS setup, we were able to quantify a 50% N enrichment in a 5 nm rind surrounding a spherical organic globular feature [Figure 1A, B]. The most primitive IOM sample, DOM08006, contains two main textural components, a "fluffy" material composed of connected carbonaceous nanoparticles, forming a porous network with high surface area, and a "compact" material of dense, non-porous carbonaceous matter, some of which forms large pieces extending several m2 in size [Figure 1C]. The EDS data reveal that the compact IOM has a high S abundance (6 at.%) [Figure 1D]. Furthermore, EELS data indicate the two textures contain distinct organic functional group chemistry [Figure 2]. Specifically, the fluffy IOM spectra show high intensity of the 285.0 eV * peak (due to aromatic C=C bonding) and of the 290.1 eV *1 peak (due to Microsc. Microanal. 21 (Suppl 3), 2015 2266 graphitic ordering), consistent with poorly graphitized C. This is consistent with the overall petrologic history of the meteorite, which has experienced heating on the parent asteroid at ~200 �C [6]. However, the compact, S-rich IOM lacks these heating spectral features, but rather has increased intensity at 286.7 eV and at 288.5 eV, which are indicative of carbonyl-bearing (C=O) modification of aromatic C [Figure 2]. This S-rich IOM appears to be absent in EET92042, and therefore may be a unique component in DOM08006 (and its related meteorites from the same parent asteroid). Whether this S-rich IOM is due to variation in precursor chemistry or is related to the particular processing history of the parent asteroid will be determined by its S bond distribution. Measurements of S-EELS and simultaneous C+S-EELS using the DualEELS are planned to address these questions [7]. References: [1] CMO'D Alexander et al, Geochimica et Cosmochimica Acta 71 (2007), p. 4380. [2] H Bussman et al, Science 312 (2006), p. 727. [3] BT De Gregorio et al, Meteoritics & Planetary Science 48 (2013), p. 904. [4] BT De Gregorio et al, Microscopy and Microanalysis 20(S3) (2014), p. 1694. [5] C Vollmer et al, Proceedings of the National Academy of Sciences 43 (2014), p. 15338. [6] GR Huss, AE Rubin and JN Grossman in "Meteorites in the Early Solar System II", ed. DS Lauretta and HY McSween Jr., (University of Arizona Press, Tucson, AZ) p. 567. [7] The authors acknowledge funding from NASA Origins Program, Grant NNH14AX71I. Dr. LR Nittler has also contributed essential discussions to this work. Figure 1. Meteorite IOM from DOM08006. (A) HAADF image and (B) EDS N abundance map of a 95 nm organic nanoglobule with a 5 nm N-rich rind. (C) HAADF image and (D) EDS hyperspectral map of "fluffy" and "compact" IOM textures. Figure 2. Hyperspectral EELS/EDS dataset from a hollow organic nanoglobule in DOM08006. A 7 nm nanodiamond is also present (red arrows and purple spectrum).");sQ1[1132]=new Array("../7337/2267.pdf","LOM Characterization of Heat-Resistant Cast Steels for the Lime/Cement Industry","","2267 doi:10.1017/S1431927615012118 Paper No. 1132 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 LOM Characterization of Heat-Resistant Cast Steels for the Lime/Cement Industry Juan Asensio1 and George F. Vander Voort2 1 Department of Materials Science and Metallurgical Engineering, The University of Oviedo, Oviedo, Asturias 33004, Spain. 2 Consultant - Struers Inc., 2887 N. Southern Hills Drive, Wadsworth, IL 60083-9293, USA. Heat-resistant cast steels are highly alloyed stainless steels designed to withstand high temperatures (600-1000�C), in corrosive environments. They are nickel-chromium-iron alloys usually with 0.20.75% carbon exhibiting a high strength at the elevated temperatures at which they operate [1]. The selection criteria for successful in-service operation include: corrosion resistance (oxidation, carburizing, nitriding and sulfidation); thermal fatigue and thermal shock resistance dependant on physical properties such as the lineal expansion coefficient, , and the thermal conductivity, ; metallurgical stability (mainly based in the ability to retard the formation of - phase); and lastly to exhibit good mechanical properties ordinarily tested at RT for simplicity by means of evaluation of harness and tensile testing [2]. In the present work three feeder links of a limestone continuous precalcining Lepol unit, operating between 300 and 750 �C (572-1382 �F) over one cycle, and with a thermal shock between 300-500 �C (572-932 �F), the last one owed to construction modifications derived from the adaptation of this roasting unit from cement to lime production. The advent of the present economical crisis has pushed manufacturers to the usage of cheaper fossil fuels, however, rich in sulphur (petroleum coke of 2% S) mixed with natural gas, and due to the low specification in sulphur of the steelmaking lime it is manufactured (S<0.040%-wt.), which in turn obligated setting the temperature at higher values to avoid sulphur pick-up near the gas exit section. The studied links were manufactured in accordance to DIN GX40 Cr Ni Si 22/9, the Cr-Ni-Si group designed for parts with a maximum operational temperature of 950�C, fair resistance to S2-, good thermal shock resistance, apt for parts subjected to high loads at oscillating temperatures. As-cast microstructure corresponds to refractory austenitic stainless steels, due to the gamma forming character of 9%Ni and 0.4%C which prevails over the alpha forming one given by 22% Cr and 1.5% Si. Carbon in the supply condition is usually distributed in equal parts, as interdendritic carbides on one side, and in solid solution of on the other. The links were set aside after a 5 year campaign and were identified as: sample A or "standardized" because it met all composition values in DIN standard; sample B with lower Ni than the value set in the specification, but with 0.5% Mo and 0.15% Nb added (not in the standard), both alpha and carbide forming elements. And lastly sample C, with a deficit in Cr and Ni with respect to the standard (Table 1). Samples for metallography were cut from a new link taken as a reference material (22% Cr-10% Ni), and of the three experimental compositions. The samples were mechanically ground and polished and etched with several etchants to reveal in light optical microscopy the relevant features to evaluate by quantitative metallography, namely the grain boundary carbides, grain carbides and phase (Fig. 1). The analysis revealed that damage by porosity was bigger in sample C (low Ni & low Cr) a fact which was coincident with high values of both the volume fraction of phase and gb carbides (M23C6) due to C precipitation and coarsening leading to gb decohesion. Sample B (Mo+Nb, low Ni), with the lowest grain size, displayed an intermediate fraction of porosity. Standardized sample A exhibited the lowest degree of intergranular porosity, coupled to low fractions of inter- and intragranular carbides (Figure 2). Microsc. Microanal. 21 (Suppl 3), 2015 2268 References: [1] "High-Alloy Cast Steels" in ASM Specialty Handbook: Heat-Resistant Materials, ed., J. R. Davis, ASM International, Materials Park OH, p. 200. [2] F.E. White and T.H. White, Traitement Thermique Revue 120-77 (1977), p. 43. [3] The authors acknowledge the technical staff of Cementos Tudela-Veguin, S.A. for allowing the access to their factory, survey of the experimental material and for valuable technical discussions. Element DIN Standard Sample A Sample B Sample C GX40 Cr Ni Si 22/9 "Standardized" "Mo+Nb added" "low-Ni & low-Cr" C 0.30 - 0.50 0.34 0.32 0.31 Si 1 � 2.5 1.353 1.580 1.533 Mn 0.593 0.580 0.752 1.5 Cr 21 � 23 21.099 21.938 20.243 Ni 9 - 11 10.43 8.25 7.81 0.216 Mo -------0.128 0.488 V -------0.0882 0.0791 0.0627 Nb -------0.0478 0.1346 0.0156 P 0.030 0.031 0.030 0.045 S 0.023 0.032 0.024 0.030 Table 1. Compositional range for DIN Standard and corresponding values for samples A- C. (%-wt.). A B C Figure 1. LOM of samples A (left), B (intermediate) and C (right). Etched with: 2.5g FeCl3, 5g picric acid, 90 ml ethanol, immerse for 15 s. Scale bar displayed at the bottom right: 100m. 2.0 1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 30 Vv (porosity)-% (a) 1.04 1.51 25 Vv (sigma)-% 20 15 10 5 0 22.14 (b) 19.84 0.30 1.12 STANDARDIZED Mo+Nb MICROALLOYED LOW Cr & LOW Ni STANDARDIZED Mo+Nb MICROALLOYED LOW Cr & LOW Ni ANALYZED COMPOSITIONS ANALYZED COMPOSITIONS Figure 2. Quantitative metallographic results for the three studied samples: (a) volume fraction of porosity or damage (left); and (b) volume fraction of sigma phase (right) after 5 years in operation.");sQ1[1133]=new Array("../7337/2269.pdf","Metallographic Sample Preparation and Characterisation of Oxide Scales on Hot-Rolled Steel Strips","","2269 doi:10.1017/S143192761501212X Paper No. 1133 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Metallographic Sample Preparation and Characterisation of Oxide Scales on Hot-Rolled Steel Strips Oktay Elkoca1, Kemal Davut2,3 1 Eregli Iron and Steel Works, Inc., Hot Rolled Products and Process R&D Dept., Eregli, Zonguldak, Turkey 2 Atilim University, Metal Forming Center of Excellence, Incek, Ankara, Turkey 3 Atilim University, Department of Metallurgical and Materials Engineering, Incek, Ankara, Turkey Oxide scales are formed on the surface of steel strips after hot rolling process. Since they affect the subsequent processes such as cold rolling, forming as well as the pickling process, the characterization of those oxide scales formed on the hot-rolled strips is of great importance. The pickling process, in particular, has been reported to be influenced by the scale's properties such as grain size, grain boundary character and the nature of the present phases in a rather unique way [1]. The oxide scales are usually composed of three different iron oxide phases, i.e. hematite, magnetite and wustite; which are usually located on the top, in the middle, and at the bottom of the scale layer, respectively. Sometimes the bottom layer of the oxide scale might have decomposition products of the wustite [2]. Total thickness of the oxide scale is increased at higher finishing and coiling temperatures of hot rolling process. During the cooling of the strip after coiling, the oxide scale continues to grow at the edge as well as the head and tail ends of the strip, where oxygen is still available. Also, the scale undergoes structural changes during cooling, because the wustite phase is unstable below 570�C. Previous examinations revealed that the thickness of the oxide layers at the edge region strongly depends on the coiling temperature, the shape of the hot rolled strip and the coil. The thick oxide layer could extend much further from the edges of the strips that have poor shapes, or that are being loosely wrapped. This causes insufficient pickling of the hot rolled strip during the pickling process (Fig. 1). The oxide scales are far more brittle than the steel substrate at room temperature [3]; moreover the mechanical properties of the scale layers are different from each other. This causes the oxide scales to be cracked or fragmented into pieces during the sample preparation. The cracks associated with the sample preparation could possibly and falsely be treated as genuine cracks in the scale [4-7]. Therefore, the metallographic sample preparation process is rather difficult and particularly important [8,9]. This work presents; firstly the metallographic specimen preparation procedures to preserve the true nature of oxide layers in a hot rolled strip; secondly the comparison of various methods to characterize the structures observed on the strip. Samples were cut from the regions across the width of a hot rolled strip of a commercial steel grade and they were characterized on the cross-sections perpendicular to the rolling direction. The metallographic sample preparation started with sectioning of samples cold-mounted in an epoxy resin by using an abrasive cutter. Grinding was performed using 320 grit and 600 grit SiC papers. Further grinding on finer abrasive papers (1000 grit, or 1200 grit) was found to be unnecessary. The most important step in the sample preparation was the coarse polishing step, which was preferably performed on a napless cotton cloth with a 6 micron diamond paste. The oxide scale layers were characterized by Nikon Eclipse MA200 optical light microscope Microsc. Microanal. 21 (Suppl 3), 2015 2270 (OLM), Jeol JSM 7001F Schottky field emission gun (FEG) scanning electron microscope (SEM), energy-dispersive X-ray spectroscopy (EDS) and Electron Back Scatter Diffraction (EBSD) technique. EBSD and EDS analysis were performed on the mentioned FEG-SEM equipped with Oxford INCA and CHANNEL 5 software packages. As a summary, this paper presents the comparison of OLM, SEM, EDS and EBSD techniques to characterize the oxide scales on hot-rolled steel strip. In addition, metallographic specimen preparation methods to reveal the true nature of oxides are also presented. The OLM and EDS techniques struggle to differentiate the different oxide layers due to the similarity of the local chemical composition among them. On the other hand, EBSD can identify these oxide layers much successfully as their crystal structures are rather dissimilar. Moreover, EBSD can give further information about the grain structure and local misorientations present within different scale layers. References: [1]. M. Zhang, G. Shao, Materials Science and Engineering A, 452�453 (2007) 189�193, [2]. R. Y. Chen and W. Y. D. Yuen, Oxid. Met., 59 (2003), 433. [3]. L. E. Samuels, Metallographic Polishing by Mechanical Methods, 3rd Ed., ASM, Ohio, (1982), 137. [4]. A. U. Malik and D. P. Whittle, Oxid. Met., 16 (1981), 339. [5]. H. T. Abuluwefa, R. I. L. Guthrie and F. Ajersch: Metall. Mater. Trans. A., 28A (1997), 1633. [6]. M. G. Fontana, Corrosion Engineering, 3rd Ed., McGraw-Hill, New York, (1987), 509. [7]. D. Caplan, G. I. Sproule, R. J. Hussey and M. J. Graham, Oxid. Met., 12 (1978), 67. [8]. L. B. Pfeil, J. Iron Steel Inst., 123 (1931), 237. [9]. L. Hachtel, Prakt. Metallogr., 32 (1995), 332. Figure 1. Insufficient pickling due to loose Figure 2. SEM micrograph of the sample wrapping, observed on the edge region of a taken from the edge region of the strip. hot rolled strip.");sQ1[1134]=new Array("../7337/2271.pdf","M�ssbauer Spectroscopy and Transmission Electron Microscopy Analysis of Transition Carbides in Quenched and Partitioned Steel","","2271 doi:10.1017/S1431927615012131 Paper No. 1134 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 M�ssbauer Spectroscopy and Transmission Electron Microscopy Analysis of Transition Carbides in Quenched and Partitioned Steel D. T. Pierce1, D. R. Coughlin2, D. L. Williamson3, K. D. Clarke2, A. J. Clarke2, J. G. Speer1, D. K. Matlock1, E. De Moor1 Advanced Steel Processing and Products Research Center, Colorado School of Mines, Golden, CO Los Alamos National Laboratory, Materials Science and Technology Division, Los Alamos, NM 3 Department of Physics, Colorado School of Mines, Golden, CO 2 1 Quenching and partitioning (Q&P) is a novel steel heat treatment that produces microstructures of martensite and retained austenite (Fig.1) [1]. Q&P consists of quenching to a temperature (QT) between the martensite start and finish temperatures, partitioning at a temperature the same or greater than the QT, followed by quenching to room temperature (RT). The goal of the heat treatment is to partition carbon (C) from martensite to austenite, thereby stabilizing the austenite prior to the final quench. Competing reactions such as transition carbide formation can reduce the extent of C partitioning, resulting in less retained austenite and mechanical property variations. The small volume fractions, carbide thicknesses below ~50 nm, and numerous overlapping peaks makes X-ray diffraction characterization of transition carbides challenging. In contrast, M�ssbauer spectroscopy (MS) with correlative transmission electron microscopy (TEM) is better suited for identifying and quantifying carbides. Most MS studies on transition carbides have focused on quenched and tempered microstructures in binary Fe-C steels with high C, extensive amounts of carbides, and MS spectra primarily comprised of resonance from a limited number of unique Fe sites [2]. Q&P steels with lower C and carbide fractions, alloying additions of Manganese (Mn), Silicon (Si), and other elements, and significant amounts of retained austenite in the microstructures have more complex MS spectra, requiring more precise analysis methods. A 0.38C-1.54Mn-1.48Si wt.% steel underwent austenitization, quenching to 225 C and partitioning at 400 C for 10 or 300 s. Microstructural characterization using a Philips CM200 TEM operated at 200 kV indicates the presence of -carbides (Fig.2). The (110) super lattice reflections in the selected area diffraction pattern in Fig.2b arise from the ordering of C atoms in the octahedral sites of the orthorhombic -carbide structure [2,3]. The MS spectra in Fig.3 were acquired at RT with a 57Co-Rh source and a spectrometer operating in the triangular constant acceleration mode. Resonance due to martensite, austenite and -carbide are present in the MS spectra. Five sextets with varying hyperfine fields and other sextet fitting parameters accounted for the total resonance of the complex distribution of Fe sites in the martensite [3]. Resonance due to Fe sites in austenite with zero C first or second nearest neighbors (A(0,0)) and with zero C first but with n (n=1, 2, 3 or 4) C second nearest neighbors (A(0,n)) were fit with two singlets [4]. Resonance due to Fe sites in austenite with one or two C first nearest neighbors (A(c)) was fit with a quadrupole split doublet [4]. The MS spectra also reveal magnetic six line resonances from two different Fe sites in -carbide: 1) Fe sites with three C nearest neighbors corresponding to stoichiometric -carbide, C(s), and 2) Fe sites with two nearest C neighbors corresponding to non-stoichiometric -carbide, C(ns), [3]. Fig.4 shows the MS spectrum of the 300 s sample on an expanded scale. The influence of both -carbide components on the spectrum is evident near the resolved 2 and 5 lines (labeled in Fig.3) of each resonance in Fig.4. -carbide fractions increase with increasing partitioning time. A correction [3] accounted for the saturation of the absorption spectrum due to sample thicknesses of ~30 m, which disproportionately attenuates the martensite and austenite resonance relative to -carbide. We calculated the recoilless fraction for -carbide, a fundamental parameter in quantitative MS analysis, to be 0.88. The recoilless fraction of -carbide is significantly greater than that for martensite and austenite (~0.81), making use of this parameter important for quantitative MS investigations [3]. The total fractions of -carbide, including s and ns components, were 1.4�0.3 and 2.4�0.3 at.% after partitioning for 10 and 300 s, respectively. Therefore, -carbides consume a significant portion (~24 and 41%) of the bulk C [5]. Microsc. Microanal. 21 (Suppl 3), 2015 2272 References: [1] J. G. Speer, D. V. Edmonds, F. C. Rizzo, D. K. Matlock, Curr Opin Solid St M 8 (2004), p. 219. [2] D. L. Williamson, K. Nakazawa, G. Krauss, Metall Trans A 10 (1979), p. 1351. [3] D. T. Pierce, D. R. Coughlin, et al., Acta Mater., http://dx.doi.org/10.1016/j.actamat.2015.01.024 [4] K. F. Laneri, J. Desimoni, G. J. Zarragoicoechea, et al., Phys Rev B 66;134201 (2002), p. 1. [5] This work was supported by the U.S. Department of Energy under Award Number DE-EE0005765. The support of the ASPPRC sponsors is gratefully acknowledged. DRC, KDC, and AJC acknowledge support from Los Alamos National Laboratory, operated by Los Alamos National Security, LLC, under Contract DE-AC52-06NA25396 for the U.S. Department of Energy. Fig.1 (a) Bright-field (BF) TEM image of two martensite variants with zone axes of [-111]1 and [001]2 and an austenite film with a zone axis of [011] after partitioning at 400C for 300 s. (b) Inset SAD pattern: A = (-110)2, B = (-11-1), C = (110)1, D = (11-1), (101)1, (110)2, and E = (200)2. Fig.2 (a) [-100] zone axis BF TEM micrograph showing -carbides within the martensite matrix in a sample partitioned for 10 s at 400 C, and (b) [-100] SAD pattern corresponding to the image in (a) where A=(0-20), B=(210), and C=(110) indicate superlattice reflections. Fig.3 MS spectra: relative counts as a function of source velocity for samples partitioned at 400 C for 10 (top) and 300 s (bottom). The stick diagrams represent the A(0,0), A(0,n), A(C), C(s) and C(ns) resonances and are not to scale. Martensite components have been removed for clarity. Fig.4 Expanded scale MS spectrum for the sample partitioned for 300 s from Fig.3. Relative counts, s-and ns-carbide subspectra, and fit including and neglecting -carbide for the sample partitioned at 400 C for 300 s. The quality of fit improves with the inclusion of the s-and ns-carbide subspectra.");sQ1[1135]=new Array("../7337/2273.pdf","Measurement of Grain Size Distributions","","2273 doi:10.1017/S1431927615012143 Paper No. 1135 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Measurement of Grain Size Distributions George Vander Voort1, Olga Kurochkina2 and Alexander Kazakov2 1 2 Consultant � Struers Inc., Wadsworth, Illinois St. Petersburg State Polytechnical University, St. Petersburg, Russia ASTM standard E1181 discusses the characterization of duplex grain size microstructures and establishes a number of category types. There are various ways to characterize grain structures, including both manual and automatic image analysis approaches based upon either lineal or areal analysis of the grains. The word "duplex" implies that the structure is divided into two parts and may not really be the best term in dealing with grain size distributions. E1181 does not define what a "normal" grain size distribution is, nor does any other ASTM grain size standard. Perhaps this should be the starting point before dealing with deviations from a normal or Gaussian distribution. The ASTM grain size number scale is not linear, but is based on a power series. So, for a specimen with a uniform distribution of grain areas, a plot of grain area versus the area % per class has a log normal distribution while a plot of the equivalent ASTM grain size class versus the area % of grains per class will exhibit a normal distribution. E1181 defines two basic types of duplex grain size distribution: random and topological in sections 8.3.1 and 8.3.2, respectively. There are three types of random duplex grain size distributions: the ALA (as large as), the wide range and the bimodal. An ALA type has a random distribution of individual coarse grains, differing in size by 3 or more ASTM grain size numbers from the average size and these individual coarse grains should comprise 5% or less of the area of the specimen. But, the "area of the specimen" is not defined. Does it mean the area of a 100X micrograph or the entire area of the polished surface? The wide-range duplex type is defined as an unusually wide range of grain sizes randomly distributed where the largest size differs from the smallest size grain by 5 or more ASTM grain size numbers. The bimodal type is defined as two distinct grain size populations, randomly distributed, such that the sizes differ by more than 4 grain size numbers, and that the two size populations together comprise 75% or more of the total surface area. What is the other 25% or less? The topological duplex conditions relate to a systematic variation in grain size across the section product form where the average grain sizes differ by 3 or more grain size numbers � called a "cross-section condition." A variation is the "necklace type" where individual coarse grains are surrounded by finer grains differing in size by 3 or more grain size numbers. The final type of topological duplex condition is called a "banded condition" where there are distinct bands with mean grain sizes differing by 3 or more grain size numbers. An analysis of the nature of the grain size distribution is easiest determined and evaluated by measuring the area of each grain in the field of view. Next, the individual grain areas are aligned in decreasing size and the sum of the grain areas per ASTM G number is determined. For example, the areas of all grains Microsc. Microanal. 21 (Suppl 3), 2015 2274 with an area between that for a G of 3.5 (11,405 �m2) and a G of 4.5 (area of 5703 �m2) would be added together and classed as grain size 4. This is repeated for all of the G classes present in the specimen. Next, the area % for each G class is calculated. A plot of G versus the area % per G class is constructed. The average grain size can be calculated from all of the measurements, particularly if the distribution is uni-modal. This work must start with evaluation of uni-modal grain size distribution specimens first, as shown in Fig. 1. Note that it covers 8 grain size classes and is, according to E1181, a duplex wide range condition. Figure 2 shows the microstructure of 330 austenitic stainless steel covering 9 G classes. It does exhibit a wide-range "duplex" condition. But, is this a "duplex" condition? The E1181 classifications of non-normal grain size distributions must be re-evaluated using this approach. Motor Lamination Steel Area % Per G Class 50 40 30 20 10 0 2 3 4 5 6 7 8 9 10 11 ASTM G a) Motor Lamination Steel (original 100X) b) Normal grain size distribution Figure 1. Grain size analysis of a steel with a uniform grain size distribution covering 8 grain size classes (no grains at G=3) with a skew of 1.43 and a kurtosis of 2.55 (891 grains measured). Austenitic 330 Stainless Steel Area % Per G Class 35 30 25 20 15 10 5 0 0 1 2 3 4 5 6 7 8 9 10 11 ASTM G a) RA 330 Austenitic Stainless Steel b) Wide range grain size distribution Figure 2. Grain size analysis of a steel with a wide-range grain size distribution covering 9 grain size classes (no grains at G-1), a skew of 7.58 and a kurtosis of 86.88 (416 grains measured).");sQ1[1136]=new Array("../7337/2275.pdf","Microstrcutural Stability of Nanostructured Ferritic Alloys (NFA)","","2275 doi:10.1017/S1431927615012155 Paper No. 1136 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Microstrcutural Stability of Nanostructured Ferritic Alloys (NFA) Reza Sharghi-Moshtaghin1, Shenyan Huang1, Richard DiDomizio 1, Laura Dial1 1.GE Global Research Center, 1 Research Circle, Niskayuna, NY 12309 Nanostructured ferritic alloys (NFA), are being developed for various high temperature applications where mechanical strength including creep properties and good resistance to irradiation damage is desired [1,2]. NFAs with a high chromium content (Cr >12%) present a fully ferritic matrix and are envisaged for applications in an extended range of temperature [3]. The high temperature strength is achieved by fine, homogeneously distributed nanoclusters in a ductile matrix that play an important role in enhancing mechanical strength [4]. The balance of chemical composition and most importantly Cr in these alloys is critical in order to maintain a ferritic structure and achieve desired corrosion and oxidation properties. However during thermal exposure at higher temperature and Cr levels due to phase stability or phase separation, high temperature -BCC may transform into iron rich (BCC) and '-Cr (BCC) phases. This phase transformation could change corrosion and/or mechanical properties of the alloy and affect performance of the material. In this study an NFA was prepared via a mechanical alloying process and subsequent hot consolidation, the detail of processing has been discussed elsewhere [5]. The chemical composition of the alloy used in this study is given in table 1. The billet was cut into pieces and aged at 427�C (800�F), 482�C (900�F) and 537�C (1000�F) up to 25,000 hrs. Each aged sample was removed from the furnace at a specific time interval and mechanical properties including Vickers hardness and tensile properties were measured. Microstructural evaluation on each sample was carried out using a 200KV Tecni-Osiris TEM equipped with 4 EDS detectors. TEM results showed that phase separation leads to the formation of the chromium-rich ' phase (Fig1,2). Also a tungsten rich laves phase was found at the grain boundaries after 2500hrs exposure at 482�C (Fig.2). Experimental results showed that tensile properties of the alloy are not significantly affected by annealing time (Fig. 3). This behavior is attributed to dislocation/NFA interaction and is under investigation. Reference: [1] G.R. Odette, JOM, Vol.66, No 12 (2014) 2427-2441 [2] M.L. Hamilton, et al., Fabrication Technique for ODS alloy MA957, Report PNL-13165, 2000 [3] A. Alamo et al., J. Nuc. Mater, Vol. 329-333, (2004) 338-341 [4] J.H. Ahh, S. Lee and J Jang, Advanced Materials Research Vols. 15-17 (2007) 696-701 [5] R. DiDomizio, S. Huang, L. Dial and M. Larsen, Met. Trans. A, Vol54A (2014) 5409-5418 Microsc. Microanal. 21 (Suppl 3), 2015 2276 Table 1. Chemical composition of the ODS alloy used in this study (wt%) Alloy NFA1 NFA2 Fe Bal. Bal. Cr 14 14 W 3 3 Ti 1.2 0.8 Y2O3 0.75 0.5 B 0.03 Figure 1. a- general microstructure shows the grain size of the alloy; b-STEM BF image shows 2-5 nm Nanoclusters formed in the microstructure in as-received condition, c- ' Cr particles formed in the microstructure after aging at 482�C (900�F) for 10000hrs. Figure 2. EDS elemental map show formation of ' Cr and Laves phsase in the microstructure after aging the sample at 482�C (900�F) for 10000hrs. Figure 3. Room temperature 0.2% yield stress and ultimate tensile stress and (b) plastic strain to failure for two NFAs as a function of anneal time at 900F.");sQ1[1137]=new Array("../7337/2277.pdf","ALCHEMI Studies of Spinel Oxides for SOFC Interconnect Alloy Coatings","","2277 doi:10.1017/S1431927615012167 Paper No. 1137 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 ALCHEMI Studies of Spinel Oxides for SOFC Interconnect Alloy Coatings Louis V. Gambino1, Neal J. Magdefrau2, and Mark Aindow1 1. Department of Materials Science and Engineering, and Institute of Materials Science, University of Connecticut, Storrs, CT 06269, USA 2. United Technologies Research Center, East Hartford, CT 06108, USA Manganese cobalt oxide (MCO) spinels are used as coatings on metallic interconnects for solid oxide fuel cells (SOFCs) because they inhibit outward diffusion of Cr while exhibiting polaronic conductivity that gives acceptable contact resistances [1]. We have recently shown that MCO-coated Crofer 22 APU develops a complex reaction layer (RL) at the alloy/coating interface, and we have proposed that the mechanism of formation is related to the site occupancies of Cr, Fe and/or Ni ions diffusing into the MCO from the alloy [2,3]. Here we report a study of these effects in which we firstly consider a series of model ceramic MCO spinels with Cr-, Fe- and Ni- substituted compositions. Powders of the appropriate compositions were produced by glycine nitrate combustion synthesis, and these powders were consolidated into ceramic pellets by pressing and pressure-less sintering. Conventional ceramic TEM samples produced by dimpling and Ar ion milling were analyzed using planar ALCHEMI on grains oriented close to the 400 systematic row. The data were plotted using the ordering tie line (OTL) approach introduced by Hou et al. [4]. In each case, the OTLs extrapolated to identify the sub-lattice compositions for the most highly ordered state consistent with the sense of order measured experimentally. These data were then compared with measurements from the RL in MCO-coated Crofer 22 APU coupons that had been oxidized in air at 800�C for 1000 h. FIB-cut cross sections through the coating/alloy interfaces were analyzed and ALCHEMI data were obtained at various points through the RL thickness corresponding to different levels of cation substitution into the MCO structure. Examples of the OTLs for ALCHEMI data obtained from Cr-substituted ceramic MCO samples are shown in Figure 1. For the samples with low and intermediate Cr contents the site occupancies show the expected trends with Co on the tetrahedral A sites and Mn and Cr on the octahedral B sites. For the high Cr sample, however, the Cr displaces some of the Mn onto the A sites so that at this composition the proportions of Mn are the same on both sites. The microstructure of the interface in the MCO-coated Crofer 22 APU after 1000 h oxidation as shown in Fig 2(a) contains: the alloy, a 1-2 �m thick fine-grained chromia layer, a RL 5�m in thickness, and the unaffected MCO coating. There is a compositional gradient within the RL and Fig. 2(b) shows the regions of low (10%), medium (28%) and high (44%) Cr content within the RL. The corresponding OTLs are shown in Fig. 3. For the regions of medium and high Cr content within the RL, the sense of the ordering is broadly consistent with that expected on the basis of the data from the Crsubstituted ceramic samples. For the low Cr sample, however, both the sense and extent of ordering are very different. This can be explained on the basis of the effects of other substitutional cations (mainly Fe) in the RL, and the fact that the MCO coatings are undergoing a structural transformation under the influence of a diffusive flux whereas the ceramic samples adopt an equilibrated cation distribution. References [1] Z. Yang, G. Xia, X. Li, J. Stevenson: Int. J. Hydrogen Energy, 32 (2007) p. 3648. [2] N.J. Magdefrau, L. Chen, E.Y. Sun, J. Yamanis, M. Aindow, J. Power Sources 227 (2013) p. 318. [3] L.V. Gambino, N.J. Magdefrau, M. Aindow. Mater. High Temp. 32 (2015) p.142. [4] D.H. Hou, I.P. Jones, H.L. Fraser. Philos. Mag. A 74 (1996), p. 741. Microsc. Microanal. 21 (Suppl 3), 2015 2278 Figure 1: OTL plots of ALCHEMI data obtained from ceramic Mn1.5-xCo1.5-xCr2xO4 spinels with x = 0.75 (High Cr), 0.5 (Med Cr) and 0.25 (Low Cr). The OTL endpoints mark to compositions for the tetrahedral (A) and octahedral (B) sites in the most highly ordered AB2O4 normal spinel structure. Figure 2: Bright field TEM images from a FIB-cut cross-section through an MCO-coated Crofer 22 APU sample that had been oxidized for 1000 h in air. Figure 3: OTL plots of ALCHEMI data obtained from grains in regions of the RL shown in Figure 2(b).");sQ1[1138]=new Array("../7337/2279.pdf","Linking Microstructure and Local Chemistry in Cu(In,Ga)Se2 Thin-Film Solar Cells","","2279 doi:10.1017/S1431927615012179 Paper No. 1138 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Linking Microstructure and Local Chemistry in Cu(In,Ga)Se2 Thin-Film Solar Cells E. Simsek1, Q.M. Ramasse2, D. Abou-Ras3, R. Mainz3, A. Weber3, H.-J. Kleebe4, P.A. van Aken1 1. 2. Max Planck Institute for Intelligent Systems, Stuttgart Center for Electron Microscopy, Stuttgart, Germany SuperSTEM Laboratory, STFC Daresbury Campus, Daresbury, UK 3. Helmholtz Zentrum Berlin, Berlin, Germany 4. Technische Universit�t Darmstadt, Institut f�r Angewandte Geowissenschaften, Germany Cu(In,Ga)Se2 (CIGSe) is one of the strongest candidates for thin-film photovoltaic applications with its high power conversion efficiencies up to 21.7% at laboratory scales [1]. In our study we focus on the microstructural evolution of structural defects, which may influence the electronic properties locally and limit the device efficiencies in polycrystalline CIGSe absorber layers. During the growth of the CIGSe absorber layers, using a three-stage co-evaporation process, samples pass through a Cu-rich phase at the end of the second stage [2]. It is known that this intermediate Curich stage is needed for defect annihilation [3,4]. A crystal structure transformation from a hexagonal to tetragonal chalcopyrite lattice also occurs in the second stage [3]. If these transformations are fast nonthermal-equilibrium transformations, planar faults form in the material. We prepared two Cu-poor CIGSe samples. The first sample (A) was produced by interrupting the growth process before it reaches the Cu-rich composition at the end of the second growth stage. The second sample (B) was grown by fully completing the three-stage co-evaporation process. The microstructure of the CIGSe thin film was analysed in depth and defect concentrations were determined by bright field (BF) and low angle annular dark field (LAADF) imaging using scanning transmission electron microscopy (STEM). The atomic structure and local chemistry of individual defects, like twin boundaries, stacking faults and dislocations, as well as of grain boundaries were studied in detail using high-resolution STEM in combination with electron energy loss spectroscopy (EELS). BF and LAADF STEM analyses showed large grains with low defect concentrations in sample B, whereas smaller grains with high defect concentrations were observed in sample A. The reason for these differences is the lack of the Cu-rich stage and recrystallization in the interrupted-process sample A. HR-STEM and EELS analysis showed striking chemical characteristics for a number of observed individual defects. The {112} twin planes, which often extend across the entire grain, exhibit homogeneous elemental distributions, which possibly do not affect the electrical properties [5]. However, at grain boundaries and complex defects we found Cu enrichment in correlation with In depletion. The local chemical change in these regions may alter the electronic properties. A Cu-segregated complex defect is shown in [110] projection in Figure 1. The segregation process apparently creates high local strain, which is subsequently relaxed via twin boundary formation in the chalcopyrite structure. Se-cation terminated twin boundaries are formed at {112} planes. This contribution on the CIGSe thin-film solar cells' growth process provides detailed information on the microstructural evolution, the defect formation and accompanied local compositional changes with the objective to further improve solar-cell performance via microstructure control. Microsc. Microanal. 21 (Suppl 3), 2015 2280 References: [1] ZSW press Release 12/2014 (2014) [2] A.M. Gabor et al., Appl. Phys. Lett. 65/2, (1994), p. 198-200 [3] H. Rodriguez-Alvarez et al., Acta Materialia 61 (2013), p. 4347-4353 [4] S. Nishiwaki et al., J. Mater. Res. 14/12, (1999), p. 4514-4520 [5] Y. Yan et al., Physica B, 401-402 (2007), p. 25-32 [6] The work was supported in part by Helmholtz Virtual Institute HVI-520 "Microstructure Control for Thin-Film Solar Cells". The research leading to these results has received funding from the European Union Seventh Framework Program [FP/2007-2013] under grant agreement no 312483 (ESTEEM2). SuperSTEM is the UK National Facility for Aberration-Corrected STEM, funded by EPSRC. Figure 1. (a) HAADF image of a complex defect twin boundary association. (b), (c), (d) elemental distribution maps extracted from the spectrum image acquired, showing Se, Cu and In maps, respectively. (e) Color-coded combination of the elemental maps.");sQ1[1139]=new Array("../7337/2281.pdf","Atomic-Resolution STEM-EDS Mapping of Grain Boundary Solute Segregation in Yttria-Stabilized Zirconia","","2281 doi:10.1017/S1431927615012180 Paper No. 1139 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic-Resolution STEM-EDS Mapping of Grain Boundary Solute Segregation in Yttria-Stabilized Zirconia Bin Feng1, Akihito Kumamoto1, Nathan R. Lugg1, Naoya Shibata1,2 and Yuichi Ikuhara1,3,4 1. 2. Institution of Engineering Innovation, The University of Tokyo, Tokyo, 113-8656 Japan PRESTO, Japan Science and Technology Agency, Saitama, 332-0012, Japan 3. Nanostructures Research Laboratory, Japan Fine Ceramic Center, Nagoya, 456-8587, Japan 4. WPI-AIMR Research Center, Tohoku University, Sendai, 980-8577, Japan Yttria-stabilized zirconia (YSZ) is one of the primary choices for the electrolyte material in solid oxide fuel cells (SOFC), due to its excellent ionic conductivity at high temperatures. Nevertheless, such performance is usually limited by the ionic conductivity at grain boundaries, which is at least two orders of magnitude lower than that of bulk [1]. As a result, many studies have been performed to investigate grain boundary properties in YSZ, focusing on atomic scale structures and chemical inhomogeneity [2, 3]. Despite the fact that the Y segregation in the grain boundaries has been demonstrated [2], the atomicscale mechanism of how solute segregation occurs is still unclear. Probing the grain boundary at atomic scale is therefore critical for a basic understanding of the chemistry at such interfaces, which in turn will provide new strategies for the design of high performance materials. In this study, the local Y distributions near YSZ grain boundaries were directly determined by atomic-resolution elemental mapping using STEM-EDS, equipped with a high-sensitivity silicon drift detector (SDD). Five types of YSZ (10 mol% Y2O3 doped ZrO2) bicrystals containing different coincident site lattice grain boundaries were first fabricated by diffusion bonding of two single crystals at 1600 �C for 15 h [2]. Elemental mapping was performed using a JEOL JEM-ARM200F cold FEG. The accelerating voltage was 200 kV, with the convergence semi-angle of 35 mrad. The probe size is about 1.0 � and the beam current is about 130 pA. The STEM-EDS system is equipped with double SDD-EDS detectors and the solid angle for the whole collection system is about 1.7 sr. Figure 1a shows a high angular annular dark-field (HAADF) STEM image of a pristine YSZ crystal from the [110] direction. Since the atomic column intensity in HAADF STEM image is approximately proportional to Z2 (where Z is the atomic number), the bright contrast represent the cation columns intermixed by Zr or Y. However, it is impossible to discriminate the Y atoms from the Zr atoms in HAADF STEM image because the atomic number of Y (Z=39) is very close to Zr (Z=40). Therefore, STEM-EDS mapping � which is able to isolate individual elements based on their X-ray emission lines was carried out to the pristine YSZ. The corresponding elemental maps are shown in Figures 1b (Zr) and c (Y). Elemental maps were extracted by selecting the following EDS energies for each element: Zr, using K of 15.776 keV and K of 17.668 keV and Y, using K of 14.958 keV and K of 16.738 keV. Such approaches enable us to clearly see the Y distribution at atomic resolution; the Y atoms are homogeneously distributed inside the perfect crystal. Figure 2 shows an example of HAADF STEM image of a 9 grain boundary indicated by the arrows. This HAADF-STEM image (in addition to others, not shown here) confirms that all the grain boundaries were bonded well at the atomic level without any amorphous or secondary phases. From elemental maps obtained by STEM EDS, we demonstrate that Y atoms indeed segregate to the grain boundaries. Microsc. Microanal. 21 (Suppl 3), 2015 2282 Moreover, there are some specific atomic sites which exhibit a stronger Y signal, suggesting that preferential segregation sites exist. Such segregation behavior can vary drastically depending on the grain boundary characters. In this presentation, we will combine these experimental results with theoretical calculations and discuss the atomic-scale Y segregation mechanisms in detail. References [1] X. Guo and R. Waser, Progress in Materials Science. 51 (2006) p. 151. [2] N. Shibata et al, Philosophical Magazine. 84 (2004), p. 2381. [3] M. Yoshiya and T. Oyama, Journal of Materials Science. 46 (2011), p. 4176. [4] B.F. and N.L. were supported as Japan Society for the Promotion of Science (JSPS) research fellow. Figure 1. (a) HAADF STEM image of the perfect crystal of YSZ observed from the [110] direction. (b) the corresponding EDS map of Zr K and (c) the EDS map of Y K. 1nm Figure 2. HAADF STEM image of a 9 model grain boundary.");sQ1[1140]=new Array("../7337/2283.pdf","Atomic-resolution STEM-EDS mapping of grain boundary solute segregation in yttria-stabilized zirconia","","2283 doi:10.1017/S1431927615012192 Paper No. 1140 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Atomic-resolution STEM-EDS mapping of grain boundary solute segregation in yttria-stabilized zirconia Bin Feng1, Akihito Kumamoto1, Nathan R. Lugg1, Naoya Shibata1,2 and Yuichi Ikuhara1,3,4 1. 2. Institution of Engineering Innovation, The University of Tokyo, Tokyo, 113-8656 Japan. PRESTO, Japan Science and Technology Agency, Saitama, 332-0012, Japan. 3. Nanostructures Research Laboratory, Japan Fine Ceramic Center, Nagoya, 456-8587, Japan. 4. WPI-AIMR Research Center, Tohoku University, Sendai, 980-8577, Japan. Yttria-stabilized zirconia (YSZ) is one of the primary choices for the electrolyte material in solid oxide fuel cells (SOFC), due to its excellent ionic conductivity at high temperatures. Nevertheless, such performance is usually limited by the ionic conductivity at grain boundaries, which is at least two orders of magnitude lower than that of bulk [1]. As a result, many studies have been performed to investigate grain boundary properties in YSZ, focusing on atomic scale structures and chemical inhomogeneity [2, 3]. Despite the fact that the Y segregation in the grain boundaries has been demonstrated [2], the atomicscale mechanism of how solute segregation occurs is still unclear. Probing the grain boundary at atomic scale is therefore critical for a basic understanding of the chemistry at such interfaces, which in turn will provide new strategies for the design of high performance materials. In this study, the local Y distributions near YSZ grain boundaries were directly determined by atomic-resolution elemental mapping using STEM-EDS, equipped with a high-sensitivity silicon drift detector (SDD). Five types of YSZ (10 mol% Y2O3 doped ZrO2) bicrystals containing different coincident site lattice grain boundaries were first fabricated by diffusion bonding of two single crystals at 1600 C for 15 h [2]. Elemental mapping was performed using a JEOL JEM-ARM200F cold FEG. The accelerating voltage was 200 kV, with the convergence semi-angle of 35 mrad. The probe size is about 1.0 � and the beam current is about 130 pA. The STEM-EDS system is equipped with double SDD-EDS detectors and the solid angle for the whole collection system is about 1.7 sr. Figure 1a shows a high angular annular dark-field (HAADF) STEM image of a pristine YSZ crystal from the [110] direction. Since the atomic column intensity in HAADF STEM image is approximately proportional to Z2 (where Z is the atomic number), the bright contrast represent the cation columns intermixed by Zr or Y. However, it is impossible to discriminate the Y atoms from the Zr atoms in HAADF STEM image because the atomic number of Y (Z=39) is very close to Zr (Z=40). Therefore, STEM-EDS mapping � which is able to isolate individual elements based on their X-ray emission lines was carried out to the pristine YSZ. The corresponding elemental maps are shown in Figures 1b (Zr) and c (Y). Elemental maps were extracted by selecting the following EDS energies for each element: Zr, using K of 15.776 keV and K of 17.668 keV and Y, using K of 14.958 keV and K of 16.738 keV. Such approaches enable us to clearly see the Y distribution at atomic resolution; the Y atoms are homogeneously distributed inside the perfect crystal. Figure 2 shows an example of HAADF STEM image of a 9 grain boundary indicated by the arrows. This HAADF-STEM image (in addition to others, not shown here) confirms that all the grain boundaries were bonded well at the atomic level without any amorphous or secondary phases. From elemental maps obtained by STEM EDS, we demonstrate that Y atoms indeed segregate to the grain boundaries. Microsc. Microanal. 21 (Suppl 3), 2015 2284 Moreover, there are some specific atomic sites which exhibit a stronger Y signal, suggesting that preferential segregation sites exist. Such segregation behavior can vary drastically depending on the grain boundary characters. In this presentation, we will combine these experimental results with theoretical calculations and discuss the atomic-scale Y segregation mechanisms in detail. References [1] X. Guo and R. Waser, Progress in Materials Science. 51 (2006) p. 151. [2] N. Shibata et al, Philosophical Magazine. 84 (2004), p. 2381. [3] M. Yoshiya and T. Oyama, Journal of Materials Science. 46 (2011), p. 4176. [4] B.F. and N.L. were supported as Japan Society for the Promotion of Science (JSPS) research fellow. Figure 1. (a) HAADF STEM image of the perfect crystal of YSZ observed from the [110] direction. (b) the corresponding EDS map of Zr K and (c) the EDS map of Y K. 1nm Figure 2. HAADF STEM image of a 9 model grain boundary.");sQ1[1141]=new Array("../7337/2285.pdf","Designer Nanoparticle-Liposome Hybrid Capsules for Drug Delivery","","2285 doi:10.1017/S1431927615012209 Paper No. 1141 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Designer Nanoparticle-Liposome Hybrid Capsules for Drug Delivery Joseph A Zasadzinski, Jeong Eun Shin, Natalie Forbes and Maria Ogunyankin. Chemical Engineering and Materials Science, University of Minnesota, Minneapolis, Minnesota 55455. Unilamellar liposomes can become the building blocks of increasingly complex nanostructures that cn optimize drug retention, targeting to specific sites, and controlled release rates. Tethering or encapsulating near infrared light adsorbing gold nanoparticles [1, 2] provides new targeting and release mechanisms with physiologically friendly near infrared light, while retaining the advantages of liposome biocompatibility and biodistribution. Multiple levels of encapsulation within multiple lipid bilayers [3, 4] can prolong drug retention by orders of magnitude over unilamellar liposomes. It is often difficult to maintain drug levels within the concentration range necessary to avoid toxicity while maintaining efficacy using systemic drug delivery. This therapeutic window can be expanded by altering drug biodistribution by using nanoscale delivery systems to maximize concentration at the disease site while decreasing toxicity by lowering drug concentrations elsewhere in the body. Nanoscale delivery vehicles can take advantage of the enhanced permeability and retention (EPR) effect, which is a passive targeting mechanism that exploits the rapid angiogenesis (formation of new blood vessels from existing ones) of tumor and inflammation sites. Whereas free drugs diffuse non-specifically, a nanocarrier can escape preferentially through the leaky vessels surrounding a tumor, which is the first step of the EPR effect. The ineffective lymphatic drainage in tumors helps retain liposomes that do extravasate and allows the liposomes to accumulate in the neighborhood of the tumor. The threshold size for extravasation into tumors is from 400 - 500 nm, but smaller nanoparticles with diameters less than 200 nm are believed to be more effective. The minimum size is about 5 nm, which is the kidney filtration cut off size. Encapsulating doxorubicin within PEGylated liposomes significantly reduced cardiotoxicity and increased the half-life in the systemic circulation. However, efficacy did not improve as expected from the increased liposomal doxorubicin accumulation at the tumor site [5] because of overly slow drug release. Liposomal cisplatin extended systemic half-life to 40 � 55 hrs [6] from the 15 � 20 min clearance of free cisplatin. However, cisplatin was not released at a therapeutic rate from the liposomes, even though 5- to 10-fold greater liposome encapsulated cisplatin accumulated in the tumors. This highlights the conflict between "retain" and "release" � factors that enhance drug retention often work against an optimal rate of drug release. Hybrid gold nanoparticle-liposome carriers provide a noninvasive method to initiate, control, and even stop drug release via membrane-localized temperature control. Plasmonic hollow gold nanoshells (HGN) are tethered (Fig. 1A) to or encapsulated within (Fig. 1B) liposomes via thiol-PEG-lipids to create a hybrid nanocarrier. The liposome-HGN are irradiated with picosecond pulses or continuous wave near infra-red (NIR) light. Picosecond pulses induce a rapid temperature rise in the HGN, leading to formation of vapor nanobubbles, similar to the cavitation bubbles caused by sonication, rupturing the liposomes and causing near instantaneous contents release. Continuous wave NIR irradiation leads to much smaller temperature increases near the liposome membrane that can induce permeability transitions in the liposome membrane, leading to contents release over minutes. The great advantage of using NIR light to induce release is that tissue, blood, etc. are relatively transparent to 650-950 nm wavelength light, allowing NIR transmission in soft tissues at depths up to several cm. Only complexes irradiated by the laser release drug, providing a targeting mechanism. We find that rapid doxorubicin release at the tumor site can reduce the necessary doxorubicin dosage by a factor of 10 or more. Microsc. Microanal. 21 (Suppl 3), 2015 2286 While liposomal doxorubicin has been the most successful of all nanoscale drug carriers, many drugs, such as the antibiotic ciprofloxacin and the chemotherapy agents vinorelbine, vinblastine and vincristine, leak out too rapidly from liposomes, which severely limits their efficacy. While coating the exterior of liposomes with polyethylene glycol (PEG) extends in vivo circulation times to 24 - 48 hours, PEG-lipids actually increase liposome permeability. To address premature release, we have made multicompartment structure based on liposomes within liposomes or "vesosomes" (Fig. 1C) [3]. Eukaryotes developed this nested bilayer structure as an alternative to optimizing the chemistry and physics of a single bilayer; each compartment has its own distinct bilayer membrane which separates different functions while protecting the internal contents from the environment [3, 4]. Having two or more bilayers between the environment and the drug contents can make a significant difference in drug retention in physiological environments. The additional bilayer barrier effectively eliminates drug release caused by exposure to phospholipase A2 (PLA2). Contents release from unilamellar liposomes after exposure to PLA2 was complete within 4- 8 hours. No release beyond background was induced by PLA2 from vesosomes over 24 hours. This indicates that the PLA2 cannot traverse the exterior bilayer of the vesosome to degrade the internal vesicles during the time of the experiment. The vesosome structure extended the half-life for release from vesosomes in serum to more than 50 hours (from about 2 hours for unilamellar liposomes), and was a delay in release for ~ 10 hours before any significant contents release occurred. Ciprofloxacin release could be extended from 10 minutes to 10 hours in serum. References [1] G. Wu, et al., J Am. Chem. Soc. 130,(2008), p. 8175. [2] N. Forbes, et al., Particles and Particle Systems Characterization 31,(2014), p. 1158. [3] C. Boyer and J. A. Zasadzinski, ACS Nano 1,(2007), p.176. [4] B. Wong, et al., Adv. Materials 23,(2011), p. 2320. [5] M. E. R. O'Brien, et al., Ann. Oncol. 15,(2004), p. 440. [6] W. C. Zamboni, et al., Cancer Chemother. Pharmacol. 53,(2004), p. 329. [7] This work was supported by National Institutes of Health (NIH) grant R01 EB012637. A B C 200 nm 100 nm 1 �m Figure 1. A) Cryo-TEM image of liposomes with tethered hollow gold nanoshells. The arrows point to encapsulated doxorubicin precipitates. B) HGN encapsulated within liposomes. Cryo-TEM is necessary to see within the liposome membranes. C) Freeze-fracture image of vesosomes. Arrow points to fractured exterior membrane. The interior vesicles are ~ 50 nm in diameter. This multiple membrane structure can slow contents release by two orders of magnitude.");sQ1[1142]=new Array("../7337/2287.pdf","Observations of in vivo Processing of Metal Oxide Nanoparticles by Analytical TEM/STEM","","2287 doi:10.1017/S1431927615012210 Paper No. 1142 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Observations of in vivo Processing of Metal Oxide Nanoparticles by Analytical TEM/STEM Uschi M. Graham1,4, Alan K. Dozier1, G�nter Oberd�rster2, Chen Wang1, Michael T. Tseng3, Joseph E. Fernback1, M. Eileen Birch1, and Burtron H. Davis4 1 Division of Applied Research and Technology, National Institute for Occupational Safety and Health, Cincinnati, Ohio, 45226 2 School of Medicine and Dentistry, University of Rochester, Rochester, NY 14642 3 Department of Anatomical Sciences and Neurobiology, University of Louisville, Louisville, KY, 40204 4 Center for Applied Energy Research, University of Kentucky, Lexington, KY 40511 High resolution transmission/scanning transmission electron microscopy (HRTEM/STEM) imaging coupled with advanced detectors allow one to probe nanomaterials in biological tissue sections in unprecedented detail. Here we report analytical TEM/STEM observations of different metal oxide nanoparticles (NPs) and compare their in vivo processing behavior: their breakdown, reactivity, and structural and chemical transformations. It was previously reported by Graham et al. that cerium oxide (CeO2) NPs, which have relatively low solubility even under extreme, controlled laboratory conditions (e.g., very low or high pH, varying ionic strength, high temperature), undergo in vivo processing in cellular structures.[1] This finding was evidenced through analytical STEM observations. In addition to CeO2, the occupational exposure risks of other widely used metal oxide NPs must be assessed. This research will lead to a better understanding of the role that in vivo processing plays with regards to toxicity or therapeutic effects of NPs.[2,3] Prepared tissue sections from inhalation studies of rats exposed to either silica (SiO2) or alumina (Al2O3) NPs were examined. SiO2 and Al2O3 NPs were selected based on relative solubility, much greater than and similar to CeO2, respectively, and for their industrial relevance and reported biological effects. After residence in rat lung tissue, metal oxide NPs underwent morphological changes and size variations, with some instances of significant reaction rims on their surfaces that are indicative of in vivo processing. Electron energy loss spectroscopy (EELS) and energy dispersive x-ray spectroscopy (EDS) in scanning (STEM) mode were utilized for elemental point analysis to provide profiling and mapping of compositional variations, phase changes, and oxidative states. Analyses were typically extended into areas adjacent to the NP (primary) deposits to detect any changes in elemental composition. The surface is thought to be main site of biological interaction, while the NP core is more isolated. The small probe (~0.2 nm) used allows compositional variations to be identified that span from the particle surface towards the NP core, while minimizing beam effects. Thus, reaction rims of the metal oxide NPs' surfaces were studied at nano and sub-nano scales. The surrounding tissue zones were also examined for any traces of in vivo processed NPs such as released ions and precipitates. The EDS/EELS spectrum images were collected on a 200 kV JEOL 2100F TEM/STEM. Figures 1 and 2 show precursor and in vivo processed lung SiO2 NPs respectively. EDS spot analysis of areas near NP deposits revealed migration of Si into the nearby tissue and intracellular environments, as indicated by the bright halo surrounding the SiO2 nanoparticles (Figure 3). References: [1] Graham U. M. et al., ChemPlusChem, 79, (2014), 1083. [2] Cassee, F. R. et al., Crit. Rev. Toxicol., 41, (2011), 213. [3] Zhao L. et al, Vaccine, 32 (2014), 327-337 Microsc. Microanal. 21 (Suppl 3), 2015 2288 Figure 1. Precursor silica. Figure 2. Silica transformation in lung. Figure 3. Dark Field STEM image showing bright cloud around lung silica particles. EDS analysis reveals cloud has a high silicon content.");sQ1[1143]=new Array("../7337/2289.pdf","Surprising high iron enrichment in hard dental tissues of rodents","","2289 doi:10.1017/S1431927615012222 Paper No. 1143 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Surprising high iron enrichment in hard dental tissues of rodents Vesna Srot1, Ute Salzberger1, Birgit Bussmann1, Bostjan Pokorny2,3, Ida Jelenko2 and Peter A. van Aken1 1. Max Planck Institute for Intelligent Systems, Stuttgart Center for Electron Microscopy, Stuttgart, Germany ERICo Velenje, Ecological Research and Industrial Cooperation, Velenje, Slovenia 3. Environmental Protection College, Velenje, Slovenia 2. Biominerals formed by living organisms display highly diverse composition and structure. Many of such biominerals are highly complex composite materials with excellent physical and mechanical properties [1, 2] that cannot be mimicked in the laboratory. These unique masterpiece architectures where organic matrix and crystalline or amorphous minerals are linked together are formed under conditions of moderate temperature, pressure and pH. Rodents possess long opposing pairs of through life continuously growing incisors, which are worn down by gnawing. The front surface of the incisors is enamel consisting of 96 wt% of inorganic material. The innermost part, softer dentine, forms the bulk of the teeth [3]. The surface of incisors of different rodent species shows characteristic orange-brown colour and is identified with the presence of iron [4]. Surface layer of rat�s enamel characterized by electron microprobe analysis [5] showed that the amount of iron (Fe) is 10-30 wt%. In this study, the microstructure and the chemical composition of incisors and molars of the feral coypu (Myocastor coypus Molina) were investigated in detail by using energy-dispersive X-ray spectroscopy (EDX) and electron energy-loss spectroscopy (EELS) combined with scanning transmission electron microscopy (STEM) imaging at high spatial as well as high energy resolution using VG HB501UX, Zeiss SESAM and Jeol ARM200F microscopes. The layer with a variable thickness was uncovered on the outer surface of the incisors which has not been observed before in rodent teeth. An annular dark field (ADF)-STEM image of a cross-sectional view showing the interface between the Fe-rich surface layer (Fe-SL) and Fe-rich enamel (Fe-E) is presented in Figure 1a. Based on our observations and analytical measurements the amount of iron in this surface layer is much higher compared to the concentration values reported in the literature. Within the iron-rich surface layer we surprisingly detected multiple iron-containing varieties. Studies of the electronic structure suggest that iron is present in predominantly 3+ valence state. O-K energy-loss nearedge structures (ELNES) acquired from different positions within the Fe-SL suggests the presence of different intermixing levels between Fe phosphate and Fe oxide/hydroxide (Figure 1b). Based on our energy-filtered TEM (EFTEM) investigations we proved that the interface between the Fe-E and above lying Fe-SL is relatively rough (Figure 2). In spite of the wide occurrence of iron in many living organisms, its function and incorporation is not unambiguously understood. Present discoveries will considerably enhance the understanding and function of Fe incorporation in hard dental tissues at the nanoscale level. Microsc. Microanal. 21 (Suppl 3), 2015 2290 References: [1] UGK Wegst and MF Ashby, Philos Mag 84 (2004), 2167. [2] AP Jackson and JFV Vincent, J Mater Sci 25 (1990), 3173. [3] BA Niemec in "Small animal dental, oral & maxillofacial disease" (2010), Manson Publishing Ltd, London. [4] EV Pindborg JJ Pindborg and CM Plum, Acta Pharmacol 2 (1946), 294. [5] A Halse, Archs Oral Biol 19 (1974), 7. Figure 1. ADF-STEM image (a) of the interface between Fe-rich surface layer (Fe-SL) and Fe-rich enamel (Fe-E). O-K ionization edges acquired from positions marked in (a) are shown in (b). In Fe-SL different intermixing levels between Fe oxide/Fe hydroxide (spectrum 1) and Fe phosphate (spectrum 2) were observed. O-K spectrum acquired from Fe-E is shown in spectrum 3. Figure 2. (a) Bright-field TEM image of the rough interface between Fe-SL and Fe-E. (b) Corresponding calcium (Ca) and iron (Fe) EFTEM maps.");sQ1[1144]=new Array("../7337/2291.pdf","Revealing the Internal Structure and Local Chemistry of Nanocrystals Grown in Hydrogel with Cryo-FIB Lift-Out and Cryo-STEM","","2291 doi:10.1017/S1431927615012234 Paper No. 1144 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Revealing the Internal Structure and Local Chemistry of Nanocrystals Grown in Hydrogel with Cryo-FIB Lift-Out and Cryo-STEM Michael J. Zachman1,2, Emily Asenath-Smith3, Lara A. Estroff2,3 and Lena F. Kourkoutis1,2 1. 2. School of Applied and Engineering Physics, Cornell University, Ithaca, NY 14853, USA. Kavli Institute for Nanoscale Science, Cornell University, Ithaca, NY 14853, USA. 3. Department of Materials Science and Engineering, Cornell University, Ithaca, NY 14853, USA. Hydrogels, three-dimensional polymeric networks with entrapped solvents, have gained increasing interest in a number of fields, including novel crystal synthesis. Compared to solution-based processes, crystal growth in hydrogels opens new routes to controlling morphology and function. Additionally, hydrogels have found applications in biomedical and biological research due to their biomimetic properties, which allow them to imitate the conditions surrounding cells. Understanding processes in hydrogels requires gaining access to their internal structures. Commonly, this requires removal of the liquid from the sample, as the hydrogel will dehydrate upon entering the vacuum of the microscope. Artifacts due to drying, however, can prevent imaging of the samples' native structure. Cryo-transmission electron microscopy (TEM) is a well-established technique in biology for imaging the near-native structure of thin hydrated specimens [1]. Thicker materials, however, require sample thinning in order to access the hidden structures and interfaces within. Over the last decade, cryofocused ion beam (FIB) milling approaches were developed for on-grid thinning of frozen samples up to microns thick [2]. For larger samples, an alternative approach, cryo-FIB lift-out, has recently been demonstrated [3-5]. In this method a thin slice, or lamella, is removed from the frozen bulk sample, attached to a TEM grid and thinned to electron transparency, while avoiding de-vitrification during the entire process. Here, we report progress in this technique, showing the ability to produce high-quality TEM lamellas from a hydrogel sample. Using cryo-scanning transmission electron microscopy (STEM) we image the structure of nanocrystals grown inside the hydrogel, as well as track local changes in chemistry and bonding with electron energy loss spectroscopy (EELS). For this work, iron chloride hexahydrate solution was added to the surface of a silica hydrogel and allowed to diffuse throughout its interior. Subsequent heating resulted in the growth of nanoparticles of an iron (oxy-, hydr-)oxide phase within the hydrogel. These structures were stabilized by snap-freezing in supercooled nitrogen, and then loaded into the cryo-FIB equipped with a Quorum PP3010T cryosystem. Figure 1 shows steps in the cryo-FIB lift-out process, from bulk sample to final lamella. To avoid instabilities due to sample charging, we first sputtered a thin layer of platinum on the surface of the hydrogel, which allowed preparation of a lamella with uniform thickness (2-5 �m thick, before liftout). Final thinning to 150-180 nm was performed with the lamella attached to the TEM grid (Fig. 1 c,d). The structure of the lamellas and the hydrogel-embedded nanoparticles were examined down to the nanoscale using annular dark field cryo-STEM. Convergent beam electron diffraction confirmed the particles' crystallinity. In addition, a comparison of the local bonding environments of the hydrogel and nanoparticles was obtained using EELS (Fig. 2). The potential impact of cryo-FIB lift-out in combination with cryo-STEM/EELS goes well beyond the field of crystal synthesis. In batteries and fuel cells for example, many critical liquid-solid interfaces are present, which have not been analyzed at high spatial resolution with the liquids intact. Microsc. Microanal. 21 (Suppl 3), 2015 2292 [1] J. Dubochet et al., Q. Rev. Biophys. 21 2 (1988), p. 129-228. [2] M. Marko et al., J. Microsc. 222 1 (2006), p. 42-47. [3] S. Rubino et al., J. Struct. Biol. 180 (2012), p. 572-576. [4] N. Antoniou, et al., 38th International Symposium for Testing and Failure Analysis (2012), p. 399. [5] C. Parmenter, et al., Microsc. Microanal. 20 (Suppl 3) (2014), p. 1224. [6] This work made use of the Cornell Center for Materials Research Shared Facilities which are supported through the NSF MRSEC program (DMR-1120296). Figure 1. Preparation of electron transparent lamellas of frozen-hydrated hydrogel specimens by cryoFIB lift-out: (a) Site-specific milling of a lamella from the bulk hydrogel sample. A sample temperature below the vitrification point is maintained using a gas-cooled Quorum cryo-stage. (b) Attachment of the frozen lamella to a TEM grid, using a cryogenically cooled OmniProbe needle and GIS water vapor. (c) Thinning to electron transparency (150-180 nm here, 2 �m scale bar). (d) A completed lamella imaged in the cryo-FIB and ready for transfer into the cryo-STEM. Figure 2. Annular dark field (ADF) cryo-STEM imaging of frozen-hydrated hydrogel lamellas prepared by cryo-FIB lift-out. Transfer from the cryo-FIB to the cryo-STEM can result in ice contamination (a), which hampers structural and spectroscopic analysis. Cryo-ADF-STEM of a contamination-free lamella reveals the structure of embedded iron (oxy-, hydr-)oxide phase nanoparticles (b). Crystallinity of these nanoparticles is confirmed by convergent beam electron diffraction (b, inset), and information about local bonding can be extracted using electron energy loss spectroscopy (c).");sQ1[1145]=new Array("../7337/2293.pdf","Chemical Imaging of Interfaces and in Interphases in Tooth Enamel","","2293 doi:10.1017/S1431927615012246 Paper No. 1145 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Chemical Imaging of Interfaces and in Interphases in Tooth Enamel Lyle M. Gordon1, Michael J. Cohen1, and Derk Joester1. 1 Materials Science and Engineering, Northwestern University, Evanston, IL, USA. Tooth enamel is the hardest tissue in vertebrates. Optimized to withstand the forces of mastication, it is composed of hydroxylapatite (OHAp) nanowires, thousands of which are bundled into rods that are organized in a three-dimensional weave. During tooth development, a preformed organic matrix is thought to be integral to the biological control over the precipitation of an amorphous precursor phase, its transformation into hydroxylapatite, and the growth of individual OHAp nanowires in enamel. This matrix is degraded during enamel maturation, but a small amount of organics remains in the final biocomposite, where its presence and that of water affect the mechanical properties. Once the tooth has erupted, enamel is affected by caries (tooth decay), a chronic infectious disease that affects nearly 100% of adults worldwide.[1] Caries commonly begins with the demineralization of enamel by acids produced in plaque biofilms. It has long been known that the susceptibility of enamel to dissolution is greatly dependent on the presence of magnesium, carbonate, and fluoride ions. However, mapping the distribution of organic and inorganic `trace' constituents is very challenging due to the complex 3D architecture, the importance of primarily low atomic number (Z) constituents, and the sensitivity of the sample to beam damage. Laser-pulsed atom probe tomography (APT), an imaging mass spectrometry technique of unrivalled spatial resolution (< 0.2 nm) and chemical sensitivity, allowed us to dramatically improve our understanding of the complex chemistry and structure of nano-scale organic/inorganic interfaces.[2] Others and we have recently expanded the use of APT to apatitic biominerals.[3-5] We report here on our recent discovery, by APT and correlative techniques, of a Mg-rich amorphous intergranular phase in regular enamel, and of iron-rich intergranular phases in pigmented rodent enamel (Figure 1), and the dramatic influence that these intergranular phases have on enamel mechanical properties and its resistance to acid corrosion.[6] We further discuss the localization of residual organic macromolecules, carbonate, and water in the intergranular phases and discuss the differentiation between organic and inorganic carbon (Figure 2).[7] References: [1] World Health Organization Media Centre, http://www.who.int/mediacentre/factsheets/fs318/en/. [2] L. M. Gordon, D. Joester, Nature 469, (2011). [3] L. M. Gordon, L. Tran, D. Joester, ACS nano 6, (2012). [4] J. Karlsson, G. Sundell, M. Thuvander, M. Andersson, Nano Letters 14, (2014). [5] E. A. Marquis, M. Bachhav, Y. Chen, Y. Dong, L. M. Gordon, D. Joester, A. McFarland, Current Opinion in Solid State and Materials Science 17, (2014). [6] L. M. Gordon, M. J. Cohen, K. W. MacRenaris, J. D. Pasteris, T. Seda, D. Joester, Science 347, (2015). [7] L. M. Gordon, D. Joester, Frontiers in Physiology 6, (2015). [8] The authors acknowledge funding from the National Science Foundation (NSF DMR-0805313, DMR-1106208, and DMR-1341391), the Northwestern University Materials Research Center (NSFMRSEC DMR-1121262), the International Institute for Nanotechnology, the Institute for Sustainability Microsc. Microanal. 21 (Suppl 3), 2015 2294 and Energy at Northwestern (ISEN), the Petroleum Research Fund of the ACS. LMG was supported in part by the Canadian National Sciences and Engineering Research Council. MJC was supported in part by NIH predoctoral Biotechnology Training Grant T32GM008449. Figure 1. Atom probe tomography reconstructions of A: regular mouse enamel; B: pigmented rat enamel; C: multiple grain boundary in mouse enamel; D: amorphous interphase in pigmented rat enamel. Figure 2. Atom probe tomography reconstruction of mouse inner enamel. A. Mg2+ segregated to simple and multiple grain boundaries. B. Nitrogen-containing ions that indicate the presence of biomacromolecules. C,D: CHO containing ions that can indicate both organic and inorganic carbon sources.");sQ1[1146]=new Array("../7337/2295.pdf","In situ Liquid S/TEM: Practical Aspects, Challenges, and Opportunities","","2295 doi:10.1017/S1431927615012258 Paper No. 1146 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In situ Liquid S/TEM: Practical Aspects, Challenges, and Opportunities Raymond R. Unocic Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, TN 37831 Recent advances in the development and implementation of platforms for in situ/operando liquid S/TEM experimentation have resulted in new research opportunities in the physical and life sciences, where it is now feasible to routinely image static and dynamic reaction processes of materials in their native liquid environments, at high spatial resolution, and under an external stimuli [1]. (Figure 1) The purpose of this tutorial is to provide a general framework for performing in situ liquid cell and electrochemical liquid cell experiments, and to understand the many physical and chemical interactions that can occur during electron beam irradiation. A comprehensive overview of factors such as accelerating voltage, electron dose, cumulative electron dose, and their influence on experimental measurements will be presented. It has been documented that the highly energetic electron beam, generally used for imaging and spectroscopy, can induce radiolysis within the liquid cell during analysis, which results in the formation of radiolytic species that are both reducing and oxidizing in nature [2]. This principal has been exploited for notable liquid cell nucleation and growth (N&G) studies from liquid phase solutions [3-5]. Calibration of electron dose is crucial such that the amount of radiation damage can be quantified to correlate the electron dose with dynamic observations of N&G processes. Figure 2 shows a typical example from an electron beam-induced N&G study of Pt nanaoparticle growth from a K2PtCl6 solution, where annular dark field (ADF) STEM images were acquired at two different electron dose rates. Additional experimental parameters, such as accelerating voltage, TEM vs. STEM imaging, and solution chemistry, have a synergistic influence on experimental results. The development of electrochemical liquid cells for in situ ec-S/TEM has enabled the dynamics of electrochemical processes from electrodeposition [6] to energy storage and conversion systems to be tracked and studied [7-10]. The ec-S/TEM platform incorporates microelectrodes that are directly patterned onto the microchip device such that electrochemical measurements can be performed to either induce an electrochemical reaction or to measure an electrochemical process; e.g., conventional electroanalytical techniques, such as chronoamperometry, cyclic voltommetry, and electrochemical impedance spectroscopy, can be applied during in situ ec-S/TEM yielding results that are consistent with the behavior of microelectrodes in microfluidic cells [11]. Figure 3 shows typical in situ quantitative electrochemical measurements acquired using an electrochemical microchip with a glassy carbon working electrode, Pt counter electrode, and Pt pseudo reference electrode using a [Fe(CN)6]3-/4- electrolyte [11]. The primary challenge for performing in situ electrochemical measurements continues to be the ability to separate the influence of the electron beam on the electrochemical measurement and selecting the optimal electron dose that is suitable for imaging but does not cause chemical reduction of the electrolyte. In situ liquid cell microscopy and in situ ec-S/TEM have proven to be viable characterization techniques used to investigate nanoscale chemical and electrochemical processes in liquid environments. With future developments in liquid cell microchip design, application of ultrafast imaging and spectroscopy methods, and the utilization of advanced data processing/data mining methods, new scientific breakthroughs are forthcoming [12]. Microsc. Microanal. 21 (Suppl 3), 2015 2296 References: [1] N de Jonge and FM Ross, Nature Nanotechnology 6 (2011) p. 695-704. [2] NM Schneider et al, Journal of Physical Chemistry C 118 (2014) p. 22373-22382. [3] H Zheng et al, Science 324 (2009) p. 1309-1312. [4] TJ Woehl et al, ACS Nano 6 (2012) p. 8599-8610. [5] KL Jungjohann et al, Nano Letters 13 (2013) p. 2964-2970. [6] MJ Williamson et al, Nature Materials 2 (2003) p. 532-536. [7] RL Sacci et al, Chemical Communications 50 (2013) p. 2104-2107. [8] Z Zeng et al, Nano Letters 14 (2014) p. 1745-1750. [9] RR Unocic et al, Microscopy and Microanalysis 20 (2014) p. 1029-1037. [10] ME Holtz et al, Nano Letters 14 (2014) p. 1453-1459. [11] RR Unocic et al, Microscopy and Microanalysis 20 (2014) p. 452-461. [12] Research supported by Oak Ridge National Laboratory's Center for Nanophase Materials Sciences (CNMS), which is a U.S. Department of Energy, Office of Science User Facility. Figure 1. Assembly view of a commercial in situ liquid cell and electrochemical liquid cell TEM holder [11]. Figure 2. Time-lapse series of ADF STEM images showing the influence of electron dose on the nucleation and growth behavior of nanostructured Pt during electron beam irradiation. (a-d) 50 e-/nm2s and (d-g) 415 e-/nm2s. Figure 3. Quantitative electrochemical measurements using electrochemical microchip devices for in situ ec-S/TEM and a 2mM potassium ferrocynide/ferricynide electrolyte. a) cyclic voltammetry, b) chronoamperometry, and c) electrochemical impedance spectroscopy [11].");sQ1[1147]=new Array("../7337/2297.pdf","Use of Ultrananocrystalline Diamond as a Phase-contrast Aperture Material","","2297 doi:10.1017/S143192761501226X Paper No. 1147 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Use of Ultrananocrystalline Diamond as a Phase-contrast Aperture Material Robert M Glaeser1, David Larson2, Benjamin Muddiman3, Simone Sassolini3, Stefano Cabrini3, and Jian Jin4 1. 2. Life Sciences Division, LBNL, University of California, Berkeley, Ca JCAP, LBNL, University of California, Berkeley, Ca 3. The Molecular Foundry, LBNL, University of California, Berkeley, Ca 4. Engineering Department, LBNL, University of California, Berkeley, Ca Boron-doped "ultrananocrystalline diamond" (UNCD), available as a 2 �m-thick layer on a "diamond on insulator" wafer (Advanced Diamond Technologies, Romeoville, IL), is investigated as a potential material for making devices that can be placed in the back focal plane of the objective lens of a TEM, the goal being to produce high contrast for in-focus images of weak phase objects. The resistance of diamond to knock-on damage when irradiated, even at 300 keV, is an attractive feature. In addition, diamond (indeed, carbon in general) is not expected to form a native oxide that is thicker than a monolayer. The ultrananocrystalline form of diamond is favored over a microcrystalline form because unwanted, "patchy" variations in surface potential (due to differences in the work-function for different crystal facets) are expected to decay more rapidly, i.e. on a length scale comparable to the domain size. We believe, however, that care must be taken to compensate for the contact potential (Galvani potential) between the UNCD device layer and the supporting (boron-doped) silicon wafer (handle). The tulip aperture [1] used here was made by first back-etching a window in the silicon with KOH followed by HF etch to remove the SiO2 "stopping layer", thus leaving a self-supported membrane of UNCD. A hard mask was then created on the top surface of the window by electron-beam lithography, after which unwanted UNCD was removed by oxygen plasma etch. Finally, the hard mask was removed with a second HF etch, leaving the bare, self-supporting tulip aperture shown in Figure 1. In preliminary results obtained so far, we found that the edge of a UNCD aperture exhibits significant charging when it is intentionally irradiated by a focused beam of 300 keV electrons at a temperature ~300 �C (Figure 2), but no such charging occurs when the same experiment is done at a temperature >600 �C (data not shown). In addition, after heating the aperture >600 �C, the edge no longer exhibited charging when it was again irradiated at ~300 �C (data not shown). This behavior suggests that (1) irradiation of UNCD at lower temperatures produces a Volta potential due to local changes in the chemical and structural composition of the surface, as has been suggested to happen for amorphous carbon [2], and (2) heating to >600 �C produces a global change in the surface, which makes the aperture immunne to further changes when irradiated. Still to be determined is whether applying a bias voltage to the UNCD "membrane", relative to the silicon frame (handle) will compensate the charging shown in Figure 3, currently thought to be due to the contact potential mentioned above [3]. References: [1] B Buijsse et al, Ultramicroscopy 111 (2011), p. 1688. [2] R Danev et al, PNAS 111 (2014), p. 15635. [3] This work was supported in part by NIH grant GM083039 and by User Project #3016 at The Molecular Foundry, LBNL. Microsc. Microanal. 21 (Suppl 3), 2015 2298 Figure 1. Optical image of our first UNCD "tulip" aperture. The dimensions of the rectangular, open aperture (with the tulip-like feature protruding in from the right) are ~0.5 mm by ~0.8 mm. The smooth, UNCD layer is closest to the viewer, supported on a silicon frame (handle), which is visible around the periphery. A 1 �m-thick SiO2 layer remains sandwiched between the UNCD layer and the silicon in the area outside the window that was etched into the silicon. Figure 2. Power spectra of images of a thin, amorphous carbon film, recorded with the tulip aperture at a temperature of ~300 �C. Intentionally touching an edge of the aperture with the focused, unscattered electron beam creates a local charging effect, manifested by a strong distortion (RIGHT) of the usual, focusdependent phase-contrast transfer function (LEFT). Figure 3. Power spectra of images of a thin, amorphous-carbon specimen. While the phase-contrast effect is well described by the usual defocus term when the electron diffraction pattern is in the open area of the aperture (LEFT), significant phase distortions occur close to the single-sideband (tulip-like) feature. Images were recorded at the previous defocus value (CENTER) and closer to focus (RIGHT).");sQ1[1148]=new Array("../7337/2299.pdf","Evolution of Magnetic Ring Designs for Phase Plates","","2299 doi:10.1017/S1431927615012271 Paper No. 1148 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Evolution of Magnetic Ring Designs for Phase Plates C J Edgcombe1 1 TFM Group, Dept of Physics, University of Cambridge, CB3 0HE, UK (email: cje1@cam.ac.uk) Ferromagnetic rings were proposed as phase plates both by the Tonomura group in 1983 [1] and independently in 2009 [2]. They offer the benefits of very low interception of scattered electrons and phase change independent of beam energy. For a phase change of /2, a ring of cobalt 20 nm thick need only be about 30 nm wide radially, within an aperture of diameter 50 micrometers. Characterisation of thin-film rings of cobalt has demonstrated their two stable states, one of which consists of the continuous loop or vortex of flux that is desirable for use as a phase plate [3]. A problem with the use of a magnetised ring in a TEM column is that the flux vortex has to remain continuous in the presence of the local axial (or out-of-plane) field. At the objective aperture position, this axial field can be 1T or more. First simulations suggested that a cobalt ring of suitable dimensions might be able to maintain a vortex in field of this size. However, a test showed that the loop was destroyed by a field between 0.3 and 0.5T. When the grain structure of the ring was included in the simulation, the maximum calculated field decreased to a value similar to that observed. There are other known disadvantages to placing a phase plate at the objective aperture [4]. To avoid these, methods have been developed (for example [5]) to place the plate at a conjugate position, further from the maximum field of the standard objective lens. Another method is to operate in Lorentz mode, turning off the main objective and using the selected-area position as the back focal plane of the Lorentz lens. Although the uncorrected aberrations in the Lorentz mode can be large, they can be corrected by existing methods [6]. Each of these techniques enables the plate to operate in a much lower axial field than that at the objective aperture. The large reduction in out-of-plane field that the plate must tolerate removes a major constraint on the design of a magnetic-ring plate. It is no longer necessary to use a material with the highest possible saturation magnetization (Ms). This has the benefit that the width-to-thickness ratio for the ring can be increased. The flux needed in the ring to provide a suitable phase change is approximately 10 15 Wb. This flux is the product of the cross-sectional area of the ring and the Ms of the ring material, so if Ms is decreased, the cross-section must increase. This increase can be obtained by extending the width of the ring in preference to its thickness. When cobalt is used, a minimum thickness of 20 nm is needed for the magnetisation to lie in-plane. This restricts the ring width for a phase change of 2 to about 30 nm. However, other materials of lower Ms may be able to remain magnetised in-plane at smaller thicknesses than is possible for cobalt, enabling the width to be extended further. An increase in the ratio of width to thickness provides two benefits. First, it raises the shape anisotropy, or energy penalty for flux to escape from the plane of the ring. This further constrains the flux to remain in-plane, thereby increasing the resistance of the vortex state to axial field. Secondly, it may be usable to reduce the effect of white haloes or stripes that appear around large objects when some Zernike phase plates are used to image weak phase objects [7]. Microsc. Microanal. 21 (Suppl 3), 2015 2300 In the simplest type of Zernike plate, the phase shift changes abruptly from zero to a fixed value, typically /2, at spatial frequency q = q0 = 2r2/f, where 2r2 is the diameter of the beam hole, is the electron wavelength, and f is the focal length of the lens. Calculations show that haloes and ringing occur around circular phase objects for which B >1, where B = q0b and 2b is the object diameter [8] (Fig. 1). However, one can imagine the /2 phase change to be distributed instead over a range of radii, outside which it remains constant. Fig. 2 shows the behaviour when the phase change increases linearly from zero at the axis to /2 at spatial frequency q0 and then remains at /2 for greater q.The effect of distributing the phase change in this way is that the ringing is eliminated and the object thickness is modelled more fully. If magnetic material is deposited as a wide ring with constant thickness, the phase change it induces will vary linearly with radius. Hence by suitable choice of dimensions and use of distributed magnetic material, it may be possible to replace the abrupt jump in phase of the simple Zernike plate by a smoother variation in phase, and so to increase the radii at which phase objects are accurately modelled. References [1] N Osakabe et al, patent application JP58-112718, June 1983. [2] C Edgcombe, EMAG 2009, J Physics Conference Series 241, p. 012005. doi: 10.1088/17426596/241/1/012005 . [3] C Edgcombe, A. Ionescu, J. C. Loudon, A. M. Blackburn, H. Kurebayashi and C.H.W. Barnes, Ultramicroscopy 120 (2012) p. 78; doi:10.1016/j.ultramic.2012.06.011 . [4] R Glaeser, Review of Scientific Instruments 84 (2013), p. 111101; doi: 10.1063/1.4830355 . [5] B Barton et al, Ultramicroscopy 111 (2011) p. 1696; doi:10.1016/j.ultramic.2011.09.007 . [6] B Freitag, M Bischoff, H Mueller, P Hartel and H S von Harrach, Microscopy and Microanalysis, 15 (Suppl. 2)(2009) p. 184. [7] R Danev, R M Glaeser, K Nagayama, Ultramicroscopy 109 (2009) p. 312. [8] C Edgcombe, Ultramicroscopy 136 (2014) p. 154, doi: 10.1016/j.ultramic.2013.09.004 . Fig. 1 Images simulated for /2 plate with step cut-on at frequency q0 and object sizes B = 1,2,4 and 8. Fig. 2 Images simulated for /2 plate with phase ramp from q = 0 to q0 and object sizes B = 1,2,4 and 8.");sQ1[1149]=new Array("../7337/2301.pdf","Development of Phase Contrast Scanning Transmission Electron Microscopy and its","","2301 doi:10.1017/S1431927615012283 Paper No. 1149 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Development of Phase Contrast Scanning Transmission Electron Microscopy and its application Hiroki Minoda1*, Takayuki Tamai1, Hirofumi Iijima2,Yukihito Kondo2 1 Department of Applied Physics, Tokyo University of Agriculture and Technology, 2-24-16 Naka-cho, Koganei, Tokyo 184-8588, Japan, 2 JEOL Ltd, 3-1-2 Musashino, Akishima, Tokyo 196-8558, Japan An interaction between electron waves and biological molecules composed of light elements is very weak. This makes it difficult to obtain their high contrast images in transmission electron microscopy (TEM). A phase plate (PP) for TEM was realized by Danev and Nagayama in 2001[1] and contrast of the biological molecules can be enhanced. They made Zernike-type phase plate which is a thin carbon film and has a small hole in its center. It is placed at a back focal plane (BFP) of the objective lens (OL). It gives a phase shift of a half to scattering waves by means of the mean inner potential of the thin film and the unscattered wave was passed through the hole without giving a phase shift. Thus, the phase difference was provided among the scattering waves and un-scattering waves. It acts as a filter in the Fourier space and modifies the phase contrast transfer function (PCTF) of the OL form the sine-type to the cosine-type. Accordingly, phase contrasts of the biomolecules would be visualized. There is a serious limitation in phase contrast TEM (P-TEM) because the PP system has to be installed in the BFP of the OL. The space round the BFP is limited and the PP can't be installed in commercial based high resolution TEM instruments because of their narrow gap of the OL pole piece. A special optical lens which can made a conjugate plane with the BFP of the OL is needed. To resolve this issue, we have been developing phase contrast transmission electron microscopy (P-STEM) by adjusting optical system including the PP system in the front focal plane (FFP) of the OL [2]. The PP for P-STEM is set in the condenser lens aperture plane (CLP) and the CLP was made to be the conjugate with the FFP by adjusting the optical illumination system. We could demonstrate the possibility to realize the optical system of P-STEM. Change in the PCTF from the sine-type to the cosine-type cold be demonstrated, but the contrast enhancement of the biomolecules had not clearly demonstrated due to the electron damage. To resolve this issue intensity of the incident electron beam should be reduced. The reduced intensity of the incident beam decreases a signal to noise ratio. Moreover, we have to use a longer camera length in P-STEM than that in conventional-STEM (CSTEM) to reduce a background intensity that was under a C-STEM imaging condition as shown in Figure 1. The longer camera length reduces the P-STEM signal intensity. To improve the low signal level we have developed a new bright field detector for STEM. Figure 2 shows one of the examples comparing images of a hole of a carbon film. Both images were taken under the same condition. It clearly shows that an image obtained with the new detector in (b) is much brighter than that with the old one in (a). Intensity of the image at the vacuum region (inside the hole) in (b) is 3.6 times larger than that in (a). Wwe have been observe biological molecules and some of the samples can provide the enhanced phase contrast. We will introduce some application in our presentation. [1] R. Danev and K. Nagayama, J. Phys. Sci. Jpn. 70 (2001), 696. [2] H. Minoda, T. Tamai, H. Iijima, F. Hosokawa and Y. Kondo, to be published in Microscopy. [3] This development was supported by SENTAN, JST. Microsc. Microanal. 21 (Suppl 3), 2015 2302 Figure 1 Projection image of the phase plate was formed on the detector plane and a camera length should be adjusted to fit the detector size to increase a signal to noise ratio. Figure 2 A comparison of the STEM bright field images obtained with an old (a) and a new detector. The image obtained by using the new detector is brighter than that obtained by the old detector. Signal intensity in (b) is about 3.6 times larger than that in (a).");sQ1[1150]=new Array("../7337/2303.pdf","High Efficiency Phase Contrast Imaging In STEM Using Fast Direct Electron Pixelated Detectors","","2303 doi:10.1017/S1431927615012295 Paper No. 1150 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 High Efficiency Phase Contrast Imaging In STEM Using Fast Direct Electron Pixelated Detectors Hao Yang1, Lewys Jones1, Henning Ryll2, Martin Simson3, Heike Soltau3, Yukihito Kondo4, Ryusuke Sagawa4, Hiroyuki Banba4, Timothy J. Pennycook5 and Peter D. Nellist1 1. 2. University of Oxford, Department of Materials, Parks Rd, Oxford, UK PNSensor GmbH, Otto-Hahn-Ring 6, 81739 M�nchen, Germany 3. PNDetector GmbH, Sckellstra�e 3, 81667 M�nchen, Germany 4. JEOL Ltd.,3-1-2 Musashino Akishima Tokyo 196-8558 Japan 5. Faculty of Physics, University of Vienna, Vienna, Austria Phase contrast imaging using electron elastic scattering has been shown to provide the highest efficiency for imaging weak phase objects for a given amount of radiation damage, compared to inelastic electron scattering as well as X-ray and neutron scattering [1]. Phase contrast imaging in scanning transmission electron microscopy (STEM) is attractive because of its flexibility in detector geometries without modifying the main optics of the electron column, and for the ability to simultaneously record analytical signals. Current bright-field STEM detector geometries include annular bright field (ABF) and differential phase contrast (DPC) imaging. As most phase information from weak scattering objects is contained within the bright field (BF) disc of the convergent beam electron diffraction (CBED) pattern, a detector geometry that makes full use of the variations in intensity of the BF disc is desirable to maximize the image contrast. We demonstrate high efficiency phase contrast imaging using a fast pixelated detector. By recording the CBED pattern at every probe position forming a 4-dimensional dataset, the phase can be reconstructed using a ptychographic phase reconstruction method based on one developed described by Rodenburg et al. [2,3] Taking the Fourier transform of the 4D dataset with respect to probe position allows interference between the BF discs and specific diffraction angles to be isolated. The interference inside the disc-overlapping region contains phase information, as shown in Figure 1. A phase image can be reconstructed by integrating such disc-overlapping regions for each spatial frequency contained in the final image. Through simulations, we find that for imaging weak phase objects, phase contrast imaging using ptychography with pixelated detectors gives the optimal phase contrast transfer function (PCTF), and generates the best phase contrast and low dose performances compared to existing imaging modes in STEM including BF, ABF and DPC. [4] Experiments on phase reconstruction using a fast pixelated detectors have been performed, and here we show results of an Au nanoparticle with five-fold twinning (Figure 2a-d), and gallium nitride GaN (Figure 2e-h). The experiments were performed using the pnCCD (S)TEM camera, viewed along a direct electron pixelated detector from PNDetector, mounted on the JEOL ARM200-CF aberration corrected microscope. The detector has a grid of 264x264 pixels and operates at a speed of 1000 framesper-second (fps). The detector can achieve a speed of up to 20,000 fps through binning/windowing. ADF images can be recorded simultaneously. Using the 4D dataset which records the diffraction patterns that contains the entire BF and part of DF regions, synthetic BF, ABF and DPC can be obtained along with the ptychographic phase reconstruction. The reconstructed phase of the Au nanoparticle (Figure 2c,d) shows a clear contrast on every atomic column. In comparison, ABF, as a nonlinear imaging mode, shows contrast that decreases towards the edge of the Au particle (Figure 2b). In the case of GaN, the N columns are hardly visible in the ABF image (Figure 2f), but are clearly resolved in the reconstructed phase (Figure 2g,h).[5] Microsc. Microanal. 21 (Suppl 3), 2015 2304 [1] Henderson, R. Quarterly Reviews of Biophysics 28, 171-193 (1995). [2] Rodenburg, J. M. et al. Ultramicroscopy 48, 304-314 (1993). [3] Pennycook, T. J. et al. Ultramicroscopy. doi:10.1016/j.ultramic.2014.09.013. [4] Yang, H. et al. Ultramicroscopy. doi:10.1016/j.ultramic.2014.10.013 [5] The authors acknowledge funding from the EPSRC (grant numbers EP/K032518/1 and EP/K040375/1) and the EU Seventh Framework Programme: ESTEEM2. Figure 1. The (a) amplitude and (b) phase of the interference between the central bright field beam (solid circle) and the diffracted beams (dashed circles) in the detector plane at a certain spatial frequency. This experimental result is obtained on an Au nanoparticle, by taking the Fourier transform of a four dimensional dataset which consists of the CBED patterns recorded at every probe position using pixelated detectors. The colour-bar in the amplitude and the phase is in arbitrary unit and in unit of radian, respectively. Figure 2. Simultaneous atomic resolution ADF, synthetic ABF image, and the reconstructed amplitude and phase of (a-d) an Au nanoparticle and (e-h) GaN bulk lattice viewed along the orientation. More clearly resolved image contrast near the edge of the Au nanoparticle is found in the reconstructed phase compared to the ABF image. For the case of GaN, the Ga-N column pairs are clearly resolved in the reconstructed phase, but the contrast of nitrogen is next to invisible in the ABF image. The probe forming aperture for both datasets is about 14.4 mrad, and the synthetic ABF has a collection angle of 7.2-14.4 mrad. Both the Au nanoparticle and GaN datasets contain 128x128 probe positions, and were recorded with a camera speed of 1000 and 4000 fps, respectively.");sQ1[1151]=new Array("../7337/2305.pdf","Electron-Beam Shaping","","2305 doi:10.1017/S1431927615012301 Paper No. 1151 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electron-Beam Shaping Roy Shiloh, Roei Remez, and Ady Arie Department of Physical Electronics, Fleischman Faculty of Engineering, Tel Aviv University, Tel Aviv 6997801, Israel In recent years, advances in nano-machining are thriving, allowing micro- and nano-scaled elements to be fabricated. Specifically, the authors have fabricated computer-generated holograms using nano-scale phase-masks [1], and demonstrated their operation in a standard TEM. Here, we continue to investigate applications for such phase-masks by imprinting the electron beam with familiar aberrations such as tilt, astigmatism, trefoil, and a spherical phase. One of the greatest struggles of electron microscopes, since their debut nearly a hundred years ago, was the reduction or amelioration of the effects of aberrations. Technological advances in fabrication methods allowed the microscopy community to take incremental steps towards the perfection of the electron microscope, and magnetic lens design in particular. A landmark in this direction of research was made by Scherzer (1936) [2] who showed that the existing, rotationally-symmetric, static and charge-free lenses cannot correct for spherical aberration - the dominant aberration at the time. Much later, this led to the successful demonstration of aberration correction using two electromagnetic hexapoles by Haider (1998) [3]. While somewhat more common today, spherical aberration-corrected microscopes are still very expensive. In lower magnifications, such as those used in Lorentz mode, it may be sensible to use a passive element which does not induce a magnetic field and does not require fundamental changes in the microscope's architecture. Such elements, in their early manifestations, have been previously introduced [4] but were not pursued. The rely on the mean inner potential of the material, in this case silicon nitride, to act as a refractive index for electrons through the expression, = 2 0 + 2 ( - 1) = 20 + where is the phase change imprinted on the beam by traversing thickness in the material, is the deBroglie wavelength associated with the electron's energy 0 and acceleration energy . is the mean inner potential, and the familiar form on the left, from light optics, associates the refractive index to the phase change. Thus, by selectively changing the thickness at pre-designed positions, a desired spatial phase-change is induced to the complete beam. We have fabricated a set of different phase plate for electrons that can potentially compensate different aberrations of the electron-optics system, as shown in Fig. 1. Two examples of generated electron beams that can compensate for astigmatism and trefoil are shown in Fig.2 below. Microsc. Microanal. 21 (Suppl 3), 2015 2306 (a) (b) (c) (d) 5um 2um 2um 2um Figure 1. TEM images of phase-masks imparting (a) spherical, (b) astigmatic, (c) trefoil, and (c) halfbeam tilt phase to the electron-beam. Contrast and brightness have been changed for visibility. (a) (b) Figure 2. Defocused TEM measurements of the beam, after traversing in some of the phase-masks in Fig.1, namely (a) astigmatism, and (b) trefoil. [1] R. Shiloh et al, Ultramicroscopy 144 (2014), 26. [2] O. Scherzer, Zeitschrift f�r Physik 101 (1936), 593. [3] M. Haider et al, Nature 392 (1998), 768. [4] Y. Ito et al, Nature 394 (1998), 49. [5] The work was supported by the Israel Science Foundation, grant no. 1310/13, and by DIP, the German-Israeli Project cooperation. [6] Presentation of this work in M&M2015 is made possible by a travel grant from the Israeli Ministry of Science and Technology.");sQ1[1152]=new Array("../7337/2307.pdf","A Phase Space Perspective on Electron Holography - Building Bridges Between Inline-, Off-axis Holography, Differential Phase Contrast and Diffractive Imaging.","","2307 doi:10.1017/S1431927615012313 Paper No. 1152 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A Phase Space Perspective on Electron Holography - Building Bridges Between Inline-, Off-axis Holography, Differential Phase Contrast and Diffractive Imaging. Axel Lubk1, Falk R�der1 and Hannes Lichte1 1. Triebenberg Laboratory, Institute of Structure Physics, TU Dresden, 01062 Dresden, Germany. Electron Holography (EH) was originally invented by D. Gabor in an attempt to correct the insurmountable spherical aberration of the objective lens in a Transmission Electron Microscope (TEM). The holographic principle proved so general that it found its way into a wide variety of techniques all aiming at reconstructing the phase information in a wave field of a particular radiation (not necessarily electrons) in one or the other way. As realized by D. Gabor, holography, in its classical appearance, is essentially nonlinear, that is, the reconstruction of a complex wave function from image intensities (almost) necessarily leads to some sort of root extraction. The associated ambiguities manifest as twin or shadow images, non-converging reconstruction algorithms or non-unique reconstructions, depending on the holographic technique. Further difficulties arise in the presence of partial coherence, eventually obscuring the reconstruction of a single wave function (or its phase). In an attempt to surmount these issues, a number of linear reconstruction schemes avoiding the issues of nonlinearity and partial coherence have been developed [1]. These so-called quantum state measurements seek a reconstruction of an object more general than the wave function, i.e., some representation of the density operator or elements of which. Here, we elaborate on this change of perspective, that is, we consider EH as a means to reconstruct the electron beam's quantum mechanical phase space distribution instead of its phase [2]. We thereby obtain a unified description of various holographic methods including related concepts such as superresolution. The unified framework allows for discussing the requirements (e.g., in terms of coherence), area of application (e.g., in terms of object properties) and generalizations (e.g., towards inelastic holography) of different types of EH, such as inline EH, off-axis EH or differential phase contrast (DPC). We particularly exploit the intuitive concept of phase space to visualize the various considerations as phase space diagrams. The central object of our considerations is the quantum mechanical density operator of the beam electrons in its various equivalent representations, namely the density matrix , , the Wigner function , and the ambiguity function , � all related by Fourier transforms along some axis (Fig. 1). Whereas the density matrix of a fully coherent wave function simply reads , = , = = , its generalization towards incoherent mixtures, |, allows for taking into account the effects of partial coherence introduced by the , | d d gun or inelastic scattering on the object. Most notably, the image intensity is = = , = = d , = , =0 - the main diagonal, momentum projection or Fourier transformed zero section in the corresponding representation. Moreover, we show that, both, coherent and incoherent aberrations, affect the different representations through particular linear operations [3], such as convolutions, which may be (partially) inverted (i.e., corrected) in case of a complete phase space reconstruction. Fig. 2 exhibits the generalized linear phase space reconstruction schemes of off-axis EH and inline (focal series) EH. Accordingly, off-axis EH corresponds to some sort of shift series [4], inline EH to a Microsc. Microanal. 21 (Suppl 3), 2015 2308 tilt series [5] and (generalized) DPC (not shown) to a Taylor expansion [6] in reciprocal phase space (ambiguity function). All of which can be used to assemble the complete phase space distribution provided that enough measurements (e.g., over a sufficient defocus range) have been performed. The above considerations form the basis for our experimental phase space reconstructions of (in)elastically scattered electrons in the framework of off-axis EH, inline EH and (generalized) DPC. As a consequence of persisting instrumental limitations, such limited astigmatic defocus ranges, the reconstructions are restricted, e.g., to objects homogeneous in one spatial direction (systems of thin films). Nevertheless, we are able to extract, for instance, a restricted phase space distribution of plasmon scattered electrons, thereby measuring object properties inaccessible by simple intensity measurements. In summary, we can therefore show that the phase space perspective provides both, a unified view on different holographic techniques and a novel route towards object characterization within the TEM. [1] M. Paris and J. Rehacek. Quantum State Estimation, Springer (2004). [2] M. Testorf and A.W. Lohmann. Appl. Opt. (2008), 47, A70-A77. [3] A. Lubk and F. R�der, Ultramicroscopy (2015). [4] F. R�der and A. Lubk, Ultramicroscopy (2014), 2014, 146, 103 � 116. [5] M. G. Raymer, M. Beck, D. McAlister, Phys. Rev. Lett. (1994), 72, 1137-1140. [6] A. Lubk and J. Zweck, Phys. Rev. A (2015), 91, 023805. [7] We acknowledge funding from the 17th round of the DIP Programme of the DFG. Figure 1. Different equivalent representations of the density operator of a Gaussian beam. The modifications (convolutions) of the Wigner function by spatial and temporal incoherence are indicated. Figure 2. (Reciprocal) phase space diagrams of inline and off-axis holography. Some important prerequisites of the corresponding setup, such as the band-limit of the quantum state in off-axis EH or the defocus interval from the near to the far field in inline holography are indicated. The holographic shift is the distance between object and reference region adjusted by the biprism voltage.");sQ1[1153]=new Array("../7337/2309.pdf","Lorentz-STEM imaging of Fields and Domains using a High-Speed, High-Dynamic Range Pixel Array Detector at Atomic Resolution","","2309 doi:10.1017/S1431927615012325 Paper No. 1153 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Lorentz-STEM imaging of Fields and Domains using a High-Speed, High-Dynamic Range Pixel Array Detector at Atomic Resolution Kayla X. Nguyen1, Robert Hovden2, Mark W. Tate3, Prafull Purohit3, John Heron3, 4, Celesta Chang3, Sol M. Gruner3, David A. Muller2,5 1. 2. Department of Chemistry and Chemical Biology, Cornell University, Ithaca, NY, USA School of Applied and Engineering Physics, Cornell University, Ithaca, NY, USA 3. Department of Physics, Cornell University, Ithaca, NY, USA 4. Department of Materials Science and Engineering, Cornell University, Ithaca, NY USA 5. Kavili Institute for Nanoscale Science, Cornell University USA Renewed interest in Lorentz microscopy of materials has accompanied the arrival of differential phase contrast imaging in scanning transmission electron microscopy (DPC-STEM). Utilizing an annular quadrant detector, small shifts related to the presence of electric or magnetic fields of the scattered beam are inferred from the difference in signal between the vertical or horizontal segments [1,2]. However, once the beam profile is not a top hat function [3], determination of spot shifts remains difficult quantitatively. DPC is also exquisitely sensitive to small changes in crystal tilt [4]. The challenge for quantitative Lorentz-STEM, like holography [5], is to accurately decouple the deflections resulting from crystallographic orientation, local atomic potentials, and the long-range electromagnetic fields. Here we demonstrate accurate Lorentz-STEM by collecting the entire diffracted beam throughout an image scan using a custom mixed-mode pixel array detector (MMPAD) [6] with exceptional highdynamic range--capable of 1 to >300,000 electrons per pixel. At 1ms readout and 10 pA probe current at 200 keV in a Tecnai F20, the high-angle scattering is detected with single electron sensitivity (Figure 1a,b) simultaneously with the unsaturated high-intensity central beam and crystallographic Bragg discs (Figure 1a). Thus, changes in the scattered beam's momentum are directly measured over all angles and a specimen's in-plane magnetic and electric fields can be extracted. Importantly, the influence of crystal tilt and absolute local sample thickness variations can also be accounted for. Figure 1c of the xcomponent of the magnetic field in Co imaged by Lorentz-STEM--a system commonly studied by traditional Lorentz-TEM and holography--shows the familiar magnetic ripple texture of the material. Fast MMPAD readout allows Lorentz-STEM imaging at atomic resolution. Since the entire diffracted beam is recorded, all of the standard STEM imaging modes--e.g. annular dark field (ADF) and bright field--can be specified post hoc [4]. Figure 1a shows an atomic resolution ADF image generated from the MMPAD alongside the mean beam deflections at a domain boundary in the ferroelectric, BiFeO3 showing both long and short range contributions (Fig. 2b,c). Over a larger field, the ferroelectric domains are clearly visible, with sharp boundaries (Figure 3a,b). In both, Fig 2 and 3, the vertical and horizontal components of the deflections the scattered beam are measured. Naively, these deflections could be converted directly to electric field. However, here the spontaneous polarization of the ferroelectric is accompanied and balanced by a lattice distortion and tilt. The gradient of the potential responsible for the beam deflection is comparably affected by crystal mistilt, thickness, atomic potential [7], and long-range electric fields. Our high-dynamic range PAD offers exciting opportunities for Lorentz-STEM--where crystal orientation, local atomic potentials, and electromagnetic fields can be determined from the full range of low to high-angle electron scattering. [8] [1] J. Chapman, J. Phys. D. 17, 623 (1984). Microsc. Microanal. 21 (Suppl 3), 2015 2310 [2] N. Shibata, et al., Nature Physics 8, 611-615 (2012). [3] Y., Yajima, et al., Bull. Col. Ibaraki Univ. (Nat. Sci) 58, 19-24 (2009) . [4] T.A. Caswell, et al., Ultramicroscopy 109, 304-311 (2009). [5] H. Lichte, et al., Ultramicroscopy 93, 199-212 (2002). [6] W. Vernon, et al., Proc. SPIE 67060U.1-11 (2007). [7] K. M�ller, et al., Nature Communications 5, 1-8 (2014). [8] Supported by the NSF MRSEC program (DMR 1120296) and the Kavli Institute at Cornell. Fig. 1. a) CBED pattern recorded in 1 ms extracted from a full PAD dataset on BiFeO3. b) Histogram of PAD intensities shows single electron quantization. c) Magnetic field from x-deflections of a Co film. Fig. 2. a) ADF-STEM signal obtained by integration from PAD dataset over 256x256 different beam positions at a BiFeO3 domain boundary. Simultaneous b) x and c) y components of the beam centroid deflection at atomic resolution. (65 sec acquisition for full PAD datatset). Fig. 3. a) x and b) y components of the beam centroid deflection ferroelectric BiFeO3. show domains and strain in");sQ1[1154]=new Array("../7337/2311.pdf","Hybrid Electron Holography","","2311 doi:10.1017/S1431927615012337 Paper No. 1154 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Hybrid Electron Holography Cigdem Ozsoy-Keskinbora1, Chris B. Boothroyd2, Rafal E. Dunin-Borkowski2, Peter A. van Aken1, Christoph T. Koch3 1. 2. Stuttgart Center for Electron Microscopy (StEM), MPI for Intelligent Systems, Stuttgart, Germany. Ernst Ruska-Centre for Microscopy and Spectroscopy with Electrons and Peter Gr�nberg Institute, Forschungszentrum J�lich, J�lich, Germany. 3. Institute for Experimental Physics, Ulm University, Ulm, Germany. When a plane wave, whether a photon, electron or neutron, passes through a transparent object, the exit wave provides information about the object's internal structure, refractive index, electric and magnetic fields within and near the specimen plus many other signals, depending on the type of radiation. The ability to detect the full exit wavefunction instead of only the intensity allows for numerical correction of the aberrations of the imaging system. Unfortunately, the phase information of a wave is lost when it hits a detector. Retrieving phase information with high reliability and sensitivity is a key challenge in many fields of optics. Gabor's 1948 holography experiments with electrons showed the potential for reconstructing the complete wavefunction for the first time1. Since then, many complementary approaches have been developed in different branches2 of optics, each with its own problems3, 4. In this study, we describe an approach for measuring the full exit wavefunction, which combines two different holography methods to overcome the deficiencies of each technique alone. We combine in-line electron holography (using a setup similar to Gabor's first experiment) to recover high spatial frequency information with an off-axis setup that allows low spatial frequency information to be recovered5. The synergistic combination of these two complimentary techniques is achieved by using the off-axis information as an initial guess for the in-line reconstruction algorithm, which then iteratively recovers the phase information. Figure 1 shows an outline of the procedure, which is followed during the reconstruction of the phase by the hybrid method. Initial experiments have been carried out successfully with electrons in a study of iron-carbon core-shell and gold nanoparticles. Figure 2 shows examples of the phase and amplitude that were obtained from several iron-carbon core-shell nanoparticles, even under suboptimal experimental conditions. References: 1. 2. 3. 4. 5. Gabor, D., Nature 161 (1948), 777-778. Cowley, J. M., Ultramicroscopy 41 (1992), 335-348. Koch, C. T.; Lubk, A., Ultramicroscopy 110 (2010), 460-471. Latychevskaia, T. et al, Ultramicroscopy 110 (2010), 472-482. Ozsoy-Keskinbora, C. et al, Sci. Rep. 4 (2014), 7020. Microsc. Microanal. 21 (Suppl 3), 2015 2312 Figure 1. Schematic outline of the wave reconstruction algorithm used in the present work. Figure 2. Full exit wavefunction obtained from a sample of iron-filled multi-walled carbon nano-onions using the hybrid method: a) amplitude; b) phase; c) amplitude and background-subtracted phase profiles.");sQ1[1155]=new Array("../7337/2313.pdf","Interface Magnetism Studied by Electron Holography with Multiple-biprisms","","2313 doi:10.1017/S1431927615012349 Paper No. 1155 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Interface Magnetism Studied by Electron Holography with Multiple-biprisms Yasukazu Murakami1 1. Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, Sendai, Japan. Revealing spin order in interface regions is important for both materials science and technologies. An intriguing phenomenon is the magnetic anomaly observed in an antiphase boundary (APB), which is a planar defect produced in ordered alloys. Atomic disordering within an APB has been believed to deteriorate ferromagnetism. Although this relationship induces useful functionalities such as pinning of magnetic domain walls, the magnetic degradation in APBs is deleterious for spintronic applications. Thus, a big challenge in materials science is to explore a distinct type of APB, in which the ferromagnetic spin order is not appreciably depressed; instead, the magnetization can be increased in APBs. Using electron holography with multiple biprisms, we have demonstrated that the magnetization can be amplified in APBs produced in Fe70Al30 [1]. In particular, we employed a technique of splitillumination [2] in order to acquire electron holograms from a portion away from the specimen edge. As illustrated in Fig. 1, an APB changes the geometrical phase of the B2-type superstructure in Fe70Al30. A thermally induced APB shows a finite thickness, in order to avoid energy penalty due to unfavorable atomic pairing [3]. We accordingly observe local atomic disorder in the APB region. For transmission electron microscopy studies, we focused on particular APB planes that were almost parallel to the incident electrons. A high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) image revealed a well-defined B2-type superstructure made of Fe columns and Fe/Al columns in the matrix regions: see Fig. 2. As shown in Fig. 2(b), difference in the peak intensity was obscured in the APB region, because of the atomic disordering (A2-type disordering) that occurred over this area. The width of the thermally produced APB appears to be 2-3 nm, although the result contains an uncertainty (approximately 1 nm) due to the ambiguity in determining the terminal positions. Figure 3 provides magnetic flux density maps (mapping of the phase gradient revealed by electron holography), observed as a function of temperature. For convenience, Fig. 3(a) shows locations of APBs in this view field. As shown in Fig. 3(b), the specimen appears to be magnetized approximately in one direction at 293 K; i.e., anomaly in APBs is not yet clear because of the significant magnetization in both matrix and APB regions. The magnetization in the matrix is significantly reduced by heating the specimen to 573 K. However, the magnetization in APBs remains pronounced at elevated temperatures, as shown in Figs. 3(c) and 3(d). Further heating makes the whole area of the specimen paramagnetic, as shown in Fig. 3(e). The results explicitly indicate that the ferromagnetic phase can be stabilized in the APB region. Our observations are unusual compared with those reported for other alloys in which APBs deteriorate ferromagnetic spin order. This study was supported by a grant from JSPS through the Funding Program for World-Leading Innovative R&D on Science and Technology (FIRST Tonomura Project). References: [1] Y Murakami et al, Nature Communications 5 (2014), p. 4133. [2] T Tanigaki et al, Appl. Phys. Lett. 101 (2012), p. 043101. [3] S M Allen and J W Cahn, Acta Metall. 27 (1979), p. 1085. Microsc. Microanal. 21 (Suppl 3), 2015 2314 Figure 1. Structure of APB produced in Fe70Al30 alloy showing B2-type structure. Reprinted from Ref. 1. Figure 2. Atomic disorder observed in the APB region in Fe70Al30. (a) HAADF-STEM image, acquired from an area with approximately edge-on APB. (b) Intensity profile measured in X1-Y1 line. Reprinted from Ref. 1. Figure 3. Magnetic anomaly in APBs observed in Fe70Al30. (a) TEM image showing the locations of APBs. (b)-(e) Mapping of the phase gradient, representing the in-plane magnetic flux density. Refer to the color wheel for the direction and magnitude of the magnetic flux. FM and PM represent ferromagnetic and paramagnetic, respectively. Reprinted from Ref. 1.");sQ1[1156]=new Array("../7337/2315.pdf","A Cryo-FIB Lift-Out Procedure for Cryo-TEM Sample Preparation at Soft-Hard Matter Interfaces","","2315 doi:10.1017/S1431927615012350 Paper No. 1156 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A Cryo-FIB Lift-Out Procedure for Cryo-TEM Sample Preparation at Soft-Hard Matter Interfaces K. Leifer Department of Engineering Sciences, Uppsala University, Box 534, Uppsala, Sweden TEM lamellae preparation using a focused ion beam (FIB) has become an important TEM sample preparation method in many research and development institutes. The TEM lamellae prepared with this method are typically some 10nm to some 100nm thick with a typical side length of some 10 microns. The preparation technique strongly relies on electron and ion beam induced deposits using Pt and W precursor gases. TEM lamellae for many very different materials ranging from soft to very hard materials could be successfully prepared. The FIB preparation is a unique tool for TEM lamella preparation when interfaces between hard and soft matter are to be analysed in the TEM. Using alternative methods such as conventional cross-sectional preparations or ultramicrotomy often fails for such materials. Whereas one of the drivers of the development of highly efficient FIB based TEM lamella preparation is semiconductor industry, micro-and nanostructured devices with hard matter interfaces are now about to be developed. Examples for such interfaces are implanted biomaterials, surfaces exposed to solutions and some battery structures. The knowledge about interaction between hard matter and living cells is crucial for the development of future medical and even pharmaceutical devices. We have developed a FIB based method to prepare TEM lamellae under cryo conditions. The particularity of this method consist in that a TEM lamella lift-out has been developed that allows to move the sample under cryogenic conditions to a TEM grid which can subsequently be transferred into a cryo-TEM holder as it is traditionally done at in the FIB lamella preparation at room-temperature. The sample can then be analysed at LN2 temperatures in the TEM. Some essential steps of the preparation method are shown in figure 1. In this presentation we will revise the current state of cryo-FIB lamella preparation. So far, FIB milling of frozen biological specimens with subsequent TEM observation has been carried out by transferring entire pencil like samples or samples that are milled on a C-Cu grid [1,2]. The pioneering work by Marko could, for the first time demonstrate that FIB preparation is a viable tool to locally prepare a biological sample. In this case, the sample must be grown or brought onto a TEM grid. Hayles et al. came up with a method where they successfully prepared a FIB lamella on a pencil like sample and demonstrated that amorphous ice remained stable after FIB preparation. In the ideal picture though, one would like to scan a sample surface for spots of interest with the FIBSEM under cryo conditions and subsequently, once this position is identified, the FIB lamella should be extracted from this spot on the sample. This was the idea that governed the development of the methodology that we have developed subsequently. We rebuilt and optimized parts of the cryo stage and a transfer chamber to bring the lamella into the TEM under cryo-conditions. Thus, from the moment, where we freeze our sample until the moment of observation in the TEM, the sample can be kept at low temperature. A critical point is the temperature stability during the cryo-lift-out procedure, which is determined by the temperature of the FIB manipulator. We measured the temperature close to the tip. These temperature measurements should enable to keep low temperatures in this critical step (figure 1). Microsc. Microanal. 21 (Suppl 3), 2015 2316 We demonstrate the cryo-FIB lift-out, cryo-transfer and cryo-TEM observation at the example of Aspergillus niger spores that are positioned on an Al support (figure 2). After cryo-FIB milling, both, Al support and the spores are sufficiently thin for HREM, energy filtered and BF/DF imaging [3,4]. [1] M Marko et al, Nat Methods 4 (2007), p. 215. [2] MF Hayles et al, J Struct Biol 172 (2010), p. 180. [3] S Rubino, S Akhtar, P Melin, A Searle, P Spellward, K Leifer, J Struct Biol 180, 2012, 572. [4] S Rubino, S Melin, P Spellward, K Leifer, (2014). J Visualized Exper: JoVE 89, 2014. [5] We acknowledge funding from the Knut and Alice Wallenberg Stiftelse. Tip_Temperature -155 Temperature (Deg C) -160 -165 -170 -175 -180 -185 -190 -195 0,5 1 1,5 2 2,5 Tip Number 3 3,5 4 4,5 Figure 1. Temperature of the cooled FIB manipulator tip for different tips. Figure 2. Some steps of cryo-FIB lamella preparation from a spore sample (A). After the cutting of the lamellae (B), the lamella is lifted out (C) and transferred to the TEM grid (D).");sQ1[1157]=new Array("../7337/2317.pdf","Temporospatial Relationship of Lipid Droplets and Mitochondria in Cardiac Muscle by Focused Ion Beam Scanning Electron Microscopy.","","2317 doi:10.1017/S1431927615012362 Paper No. 1157 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Temporospatial Relationship of Lipid Droplets and Mitochondria in Cardiac Muscle by Focused Ion Beam Scanning Electron Microscopy. Tsengming Chou1, Carole Sztalryd-Woodle 2, and Ru-ching Hsia3 1. 2. Laboratory for Multiscale Imaging, Stevens Institute of Technology, Hoboken, USA Department of Medicine, University of Maryland Baltimore, Baltimore, USA 3. Electron Microscopy Core Imaging Facility, University of Maryland Baltimore, Baltimore, USA Obesity is a rising health problem in the United States and many developed countries. The World Health Organization estimates that, in 2014, more than 1.9 billion adults were overweight worldwide, and more than 600 millions were obese [1]. Obesity is the result of an imbalance in lipid homeostasis and is evidenced by the excessive accumulation of cytoplasmic lipid droplets (CLDs) in cardiac muscles. The CLD is a dynamic organelle that interacts with various cell organelles such as the endoplasmic reticulum, endosomes and mitochondria. Our previous studies suggested that the LD specific protein, perilipin 5, is involved in promoting association of mitochondria to LDs [2]. Here we investigate the temporospatial relationship of LDs, mitochondria, endoplasmic reticulum (sarcoplasmic reticulum) and muscle fibrils in a 3D volume of mouse cardiac muscle using a focused ion beam scanning electron microscope (FB-SEM). Transgenic mice over-expressing perilipin 5 under the -myosin heavy chain promoter were fasted for 24 hr to increase the number of LDs before euthanasia. Heart muscles were excised and manually trimmed into thin slices, fixed, post stained with osmium tetroxide, dehydrated and embedded in spurs epoxy resin. Epoxy resin blocks containing heart muscle were trimmed and sectioned using an ultramicrotome to expose tissue surface. The block faces were glued onto a SEM-stub, painted with conductive paint, sputter-coated and loaded onto a Zeiss Auriga FIB/SEM. Back scattered SEM images were collected at a beam energy of 1.5kV, and ion beam millings were done at 1nA and 20nm per slice. A total muscle tissue volume of ~15 x 24 x 12 �m3 was analyzed in this study. Image alignment, segmentation and 3D analyses were done by using ImageJ and Avizo 9.0 software. LDs were present in abundance in cardiac muscles of transgenic mice. LDs were closely associated with mitochondria and the presence of LDs disrupted the normal distribution of mitochondria along the muscle fibrils. Vesicles of unknown function were observed in mitochondria in the heart muscle of transgenic animals, but not in wild type mice. Although it is unclear whether this is related to altered mitochondrial function in these mice, it is noteworthy that these vesicles appeared to originate between mitochondrial inner membrane stacks. A complete segmentation of each key organelle in the entire volume will be necessary to reveal the spatial relationship and interactions among the key organelles. Moreover, the membrane interactions predicted between LDs and between mitochondria, LDs and the sarcoplasmic reticulum may require higher magnification imaging. References: [1] World Health Organization Fact sheet No311. http://www.who.int/mediacentre/factsheets/fs311/en/ [2] H Wang et al, Journal of Lipid Research 52 (11), p. 2159. [3] This research effort used microscope resources partially funded by the National Science Foundation through NSF Grant DMR-0922522. Microsc. Microanal. 21 (Suppl 3), 2015 2318 Figure 1. The epoxy resin block face showing the exposed cardiac muscle tissue (A) and the back scattered SEM image (B) (LD, lipid droplet; Mt, mitochondrion; Ms, muscle fibril; SR, sarcoplasmic reticulum; Vs, vesicle of unknown function). Red arrow points to the excavated trench and the milled area. Yellow arrows point to the SR network. Figure 2. Single SEM slice demonstrating the close association of LDs with mitochondria (A). LDs and mitochondria were segmented using thresholding (B&C). B shows the 3D distribution of LDs (gold color); C shows a combined image of LDs and mitochondria (purple) in the 3D volume.");sQ1[1158]=new Array("../7337/2319.pdf","SIMS on FIB Instruments: a Powerful Tool for High-Resolution High-Sensitivity Nano-Analytics","","2319 doi:10.1017/S1431927615012374 Paper No. 1158 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 SIMS on FIB Instruments: a Powerful Tool for High-Resolution High-Sensitivity Nano-Analytics T. Wirtz, D. Dowsett, P. Philipp, J.-N. Audinot, S. Eswara Moorthy Advanced Instrumentation for Ion Nano-Analytics (AINA), Luxembourg Institute of Science and Technology, 41 rue du Brill, L-4422 Belvaux, Luxembourg FIB-based instruments play an ever more important role in materials science and also in life science. While such FIB-based instrumentation is an ideal tool for high resolution imaging (2D and 3D imaging) and nanofabrication (nanomaching, nanodeposition, specimen preparation), its analysis capability is currently limited. By contrast, Secondary Ion Mass Spectrometry (SIMS) is an extremely powerful technique for analyzing surfaces owing in particular to its excellent sensitivity, high dynamic range, very high mass resolution and ability to differentiate between isotopes. Adding SIMS capability to FIB instruments offers not just the prospect of obtaining SIMS information limited only by the size of the probe-sample interaction (~10nm) but also enables a direct correlation of such SIMS images with high resolution secondary electron images of the same zone taken at the same time. Past attempts of performing SIMS on FIB instruments were rather unsuccessful due to unattractive detection limits, which were due to (i) low ionization yields of sputtered particles, (ii) extraction optics with limited extraction and collection efficiency of secondary ions and (iii) mass spectrometers having low duty cycles and/or low transmission. In order to overcome these limitations, we have investigated the use of reactive gas flooding during FIB-SIMS and we have developed compact high-performance magnetic sector mass spectrometers with dedicated highefficiency extraction optics. In order to reach good detection limits when probing very small voxels in imaging applications, the ionization probability of the sputtered atoms and molecules needs to be maximized. When using typical ion species used in FIB-based instrumentation such as Ga or noble gases, the intrinsic yields are low compared to the ones found in conventional SIMS. However, the yields may be drastically increased by using reactive gas flooding during analysis, namely O2 flooding for positive secondary ions and Cs flooding for negative secondary ions [1-3]. Our results show that both negative and positive ion yields obtained with Ga+, He+ and Ne+ bombardment may be increased by up to 4 orders of magnitude when using such reactive gas flooding (Figure 1). This optimization of secondary ion yields leads to detection limits varying from 10-3 to 10-7 for a lateral resolution between 10 nm and 100 nm (Figure 2). The trade-off between detection limits and minimum detectable feature size in the SIMS mode shown in Figure 2 can be overcome in the correlative microscopy approach. The SIMS module can now be operated in a mode leading to excellent detection at the cost of poorer lateral resolution (e.g. 1 ppm @ 50 nm), but the sub 10 nm resolution is gained back by overlaying the secondary electron images obtained on the FIB, Dual-Beam or HIM instrument. The developed SIMS add-on system consists of three main components, namely the secondary ion extraction optics, the mass spectrometer and the integrated reactive gas flooding system. The emitted Microsc. Microanal. 21 (Suppl 3), 2015 2320 secondary ions are extracted by extraction optics with a maximized efficiency and without negatively impacting the focusing of the primary ion beam (broadening or distortion of the ion beam due to the electric fields). We have successfully designed such secondary ion extraction optics for several FIB-based instruments. The obtained extraction efficiency ranges from 40% (extraction voltages of a few 100V) up to 100% (extraction voltages of a few kV). These extraction optics are coupled to a specially designed compact high-performance magnetic sector double focusing mass spectrometer that we developed for the purpose of FIB-SIMS. The specifications of this mass spectrometer include highest transmission (100 %), high mass resolution (M/ M > 2000), full mass range (H-U) and parallel detection of several masses. The integrated gas flooding system can deliver O2 and Cs gas onto the analyzed area for the enhancement of positive and negative secondary ions, respectively. The results are very encouraging and the prospects of performing SIMS in combination with these high-resolution imaging and nanofabrication instruments are very interesting. In addition, the combination of high-resolution microscopy and high-sensitivity chemical mapping on a single instrument represents a new level of correlative microscopy. References [1] P. Philipp et al., Int. J. Mass. Spectrom. 253 (2006) 71 [2] T. Wirtz et al., Appl. Phys. Lett. 101 (2012) 041601 [3] L. Pillatsch et al., Appl. Surf. Sci. 282 (2013) 908 Experimental Extrapolated O2 flooding 100 10-1 10-1 10-2 Detection limit (C- in Si) No flooding Cesium flooding No flooding Cs flooding 10-2 10-3 10-4 10-5 10-6 10-7 10-8 Useful yield (no flooding): 7 x 10-6 30 keV Ga+ Useful yield (Cs flooding): 1 x 10-1 Useful yield 10-3 10-4 10-5 10-6 GaAs matrix 10 -7 10-9 CuB+ Zn+ Cu+ C- O- Si- 1 10 100 1000 Element Feature size (nm) Figure 1: Enhancement of secondary ion yields using reactive gas flooding under Ga+ bombardment 150 Intensity (counts) 2 Figure 2: Detection limit using a Ga+ FIB with and without Cs flooding vs. minimum feature size: example for the detection of C-. Figure 3: Co distribution in a tungsten carbide sample imaged by Ga+ FIB-SIMS (field of view: 50x50 m2): (a) without O2 flooding, (b) with O2 flooding");sQ1[1159]=new Array("../7337/2321.pdf","Application of Focused Helium Ion Beams for Direct-write Lithography of Superconducting Electronics.","","2321 doi:10.1017/S1431927615012386 Paper No. 1159 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Application of Focused Helium Ion Beams for Direct-write Lithography of Superconducting Electronics. Shane Cybart1, E. Y. Cho1, B. H. Wehlin1, Meng Ma1, T. J. Wong1, R. C. Dynes1, and Chuong Huynh2 1. 2. Oxide Nano Electronics Laboratory, Department of Physics, University of California San Diego, USA. Carl Zeiss Microscopy, LLC., One Corporation Way, Peabody, MA, USA. The 1986 discovery of high transition temperature (high-TC) superconductivity in copper-oxide materials set in motion an intense research effort to develop superconducting electronics operating at and above liquid nitrogen temperatures (77 K). Scientists and engineers worked with fervor to develop these new exciting materials but soon discovered that they were much more complicated than imagined. Anisotropic electrical properties and a very short superconducting coherence length eliminated the possibility of using classical superconducting electronic structures. These new materials demanded novel device architectures that proved very difficult to realize. Nearly three decades have passed and progress in high-TC superconducting devices has been very slow because process control at the sub ten nanometer scale is required to make high quality Josephson junctions: the basic building block of superconducting electronics. Recent advances in gas field focused helium ion beams [1] provide a new promising approach for direct-write lithography of high-TC superconducting materials [2] for the realization of a predictable and scalable high-TC superconducting electronics technology. In this work, we demonstrate a-b plane superconducting Josephson tunnel junctions for YBa2Cu3O7- (YBCO) by utilizing a 500-pm diameter focused helium ion beam to create a very narrow (~nm) tunnel barrier between two superconducting electrodes. The key to this method is that YBCO is very sensitive to point defects in the crystal lattice caused by ion irradiation [3]. Increasing irradiation levels has the effects of increasing resistivity and reducing the superconducting transition temperature. At very high irradiation levels YBCO becomes insulating and no longer conducts or superconducts [3]. To obtain smaller feature sizes required for high quality Josephson junctions we have employed a focused helium ion beam to directly disorder a thin 25 nm thick YBCO film. Initially large circuit features for electrical contacts, and 4-m wide strips of YBCO were patterned with conventional photolithography in a YBCO thin film that had an in situ deposited Au contact layer on top [2]. The starting YBCO film thickness was 150 nm, but the Au was removed and the YBCO was etched to a thickness of ~30 nm in the area intended for junctions. This thinning was to ensure that the helium ions fully penetrated the YBCO with very little lateral straggle. Samples were then loaded into a Zeiss Orion helium ion microscope and the 30 kV helium beam was scanned in a line across the 4-m wide superconducting bridges. Numerous test samples were written with ion fluence ranging between 1014 and 1018 He+/cm2. At the lower values very little reduced TC and Josephson current was observed. In contrast, at the higher doses the devices exhibited insulating behavior. In between these two extremes we were able to determine doses that could create very high-quality Josephson junctions with both metallic and insulating barriers. Figure 1 shows the ion damage profile for an ion implantation simulation for a dose of 6x1016 ions/cm2 into a 30 nm thick YBCO film. The Josephson barrier appears to be very uniform throughout the depth of the film and only 2 nm wide. After irradiations, test samples were characterized electrically by measuring the current-voltage (I-V) characteristics (Figure 2A). At low voltages the samples showed nearly ideal Josephson junction Microsc. Microanal. 21 (Suppl 3), 2015 2322 behavior with a zero voltage supercurrent that oscillated in magnetic field. At much higher voltages (IV) exhibited insulator behavior (Fig. 2B). Using ac techniques we measured the differential conductance (dI/dV) which revealed the YBCO superconducting energy gap near 33 mV. This feature is a result of quasi particle tunneling which provides strong evidence that we have created an insulating barrier less than 2nm wide. Figure 1. (left) Simulated 30 keV He ion irradiation of a YBCO film using Silvaco Athena. (right) A scale model of the focused helium ion beam creating a Josephson junction in a YBCO film. A B Figure 2. (A) Current voltage characteristics (I-V) for a YBCO Josephson junction written using a dose of 6x1016 ions/cm2. The inset shows the quantum diffraction of the critical current in magnetic field. (B) Current voltage characteristics (I-V) for the same junction at higher voltage bias. Here the non-linearity reveals that the barrier has insulating properties. Measurement of dI/dV reveals the YBCO superconducting energy gap. These results demonstrate the unique ability of focused helium ion beams for maskless direct write lithography of oxide tunnel barriers for electronic devices. This technique is not limited to superconductors and will work on any material that is sensitive to disorder. References: [1] B. W. Ward, J. A. Notte, N. P. Economou, J. Vac. Sci. Technol., B 24 (2006). [2] S. A. Cybart et al, arXiv: 1409.4876, http://arxiv.org/abs/1409.4876 (2014) [3] J. M. Valles et al, Phys. Rev. B. 39 (1989), p. 11599. [4] The authors gratefully acknowledge Garrett Schlenvogt for help with ion implantation simulations. This work was funded by the Air Force Office of Scientific Research and the UC Scholars Program.");sQ1[1160]=new Array("../7337/2323.pdf","Materials Processes Observed using Dynamical Environmental TEM at University of Illinois","","2323 doi:10.1017/S1431927615012398 Paper No. 1160 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Materials Processes Observed using Dynamical Environmental TEM at University of Illinois Aram Yoon1,2 , Wenpei Gao1, 2, Jianbo Wu1,2, Hyuk Park1, 2, J. Mabon2, W.L. Wilson2, Jian-Min Zuo1, 2* 1 Dept of Materials Science and Engineering, University of Illinois, Urbana-Champaign, IL 61801, USA 2 Frederick Seitz Materials Research Laboratory, University of Illinois, Urbana-Champaign, IL 61801, USA The progress in the development of dynamical environmental TEM (DETEM) at University of Illinois is reported here together with preliminary results that have been obtained in the study of nanostructure transformation and metal oxidation. The instrument development project was motivated by the need to improve time resolution in in-situ TEM study of various materials processes. TEM in general provides outstanding spatial resolution for atomic structure determination and small probes for electronic structure and chemical analysis. The use of imaging aberration corrector in an ETEM now enables atomic resolution visualization of structural transformations under variable temperatures and gas environments close to materials' real operational conditions (for a review, see ref [1]). However, the types of structural transformations that can be observed are limited by the sample stability and acquisition rate of electron images or diffraction patterns [2]. The development of MEMS based heating holders has improved the sample stability issue. For time resolution, the use of laser driven photocathodes provides superior time resolution by taking snap shots using ultrafast electron pulses [3]. While time resolution at 15 ns has been reported for recording a sequence of irreversible processes using pulsed electron beams [3], the time interval that can be recorded by this technique is currently very limited. A Hitachi H9500 80-300kV TEM with a LaB6 emitter is used as the development platform for DETEM. The major improvements already made include the design and installation of a gas handling and mixing system, a dual camera system with a Gatan Orius camera for video rate image recording and K2 IS camera for direct electron detection and fast recording at 400 fps for full frame transfer or 1600 fps at � frame transfer [4], and a high temperature sample heating holder up to 1500 �C [5]. The improvement of time resolution was demonstrated in the heating experiment of Au catalyst on silicon nanowires (NW). At 2.5 ms apart when the Au catalyst was heated to 400 �C, initial movements of Au catalyst were observed (Figure 1 shows an example), followed by a significant change in the wetting angle of Au nanoparticle. Afterwards Au was seen diffuse away along the NW surface as indicated by fast image contrast change. Further experiments on metal oxidation have been carried out thanks to the environmental capability of our TEM. Metal oxidation is one of the spontaneous chemical reactions. Because of its large impact on infrastructure and transportation, surface oxidation and its role on structure and property have attracted significant attentions, especially on the fundamental aspects of oxygen diffusion into the metal or metal ion transport. Figure 2 shows an in situ observation of oxidation of zircaloy-2, which is a nuclear cladding material. The experiment was carried using the special holder designed for the ETEM to provide gas injection and Joule heating directly on the sample [5]. We maintained the air pressure of 2x10-3 Pa, while heating the sample to 900�C during recording using the Orius SC200 camera (Gatan, Pleasanton, CA). The images capture the nucleation and decay of nanoparticles formed on the surface oxide of zirconium metal. Microsc. Microanal. 21 (Suppl 3), 2015 2324 References: [1] J. R. Jinschek, Chem. Commun.,2014, 50, 2696 [2] An example is the Model 622 Fiber optically coupled TV system for TEM made by Gatan, INC. [3] Kim, J. S., et al., Science 321, 1472 (2008) [4] http://www.gatan.com/files/PDF/products/K2_Datasheet_FL2web.pdf [5] T. Kamino, Journal of Physics: Conference Series 126, 012088 (2008) [6] Corresponding author Jian-Min Zuo, jianzuo@illinois.edu. The instrument development is supported by the NSF MRI Grant NSF DMR 12-29454 and University of Illinois. Aram Yoon and Jianbo Wu was partially supported by DOE BES under contract DEFG02-01ER45923. Wenepi Gao was supported by NSF DMR 0449790. We thank Prof. Brent Heuser for supplying zircaloy sample and discussions. Figure 1. Au catalyst movements on a silicon nanowire captured by a fast direct detection electron camera. The time stamp below shows frames captured into 6 seconds of observation with 2.5ms time interval. (a) (b) (c) (d) t = 0 sec (e) (f) t = 1 sec (g) t = 3 sec (h) t = 6 sec t = 7 sec t = 8 sec t = 9 sec t = 12 sec Figure 2. Oxidation of zircaloy observed at 900C in air with high resolution (magnification of 400K) recorded at video rate of 20 frames /second.");sQ1[1161]=new Array("../7337/2325.pdf","Strain Mapping during In-situ Deformation using a High-Speed Electron Detector","","2325 doi:10.1017/S1431927615012404 Paper No. 1161 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Strain Mapping during In-situ Deformation using a High-Speed Electron Detector C. Gammer1,2,3, J. Kacher1,2, J. Ciston1, C. Czarnik4, O.L. Warren5, A.M. Minor1,2 1. National Center for Electron Microscopy, Molecular Foundry, Lawrence Berkeley National Laboratory, Berkeley, CA 94720 USA 2. Department of Materials Science & Engineering, University of California, Berkeley, CA 94720 USA 3. Physics of Nanostructured Materials, Faculty of Physics, University of Vienna, Austria 4. Gatan, Inc., Pleasanton, CA 94588 USA 5. Hysitron, Inc., Minneapolis, MN 55344 USA Understanding the evolution of the local strain around individual defects during plastic deformation is of great importance for correlating defect structure with material properties. In-situ deformation in the TEM has provided great insight into the fundamental mechanisms occurring during deformation and many advances have been made in measuring the global plastic strain more accurately by using high precision transducers or image-correlation [1]. In addition, TEM has demonstrated the ability to measure strain fields around static dislocations [2], but measurements of the local elastic strain-field around individual defects during deformation are still lacking. In the present work we show for the first time that strain mapping can be carried out during continuous in-situ deformation in a TEM at the nanometer scale. We use scanning nanobeam electron diffraction (NBED) for the strain mapping. Figure 1 shows the experimental setup. The converged beam rasters over the sample and on top of the annular dark-field (ADF) image, a full diffraction pattern is recorded for every probe position. A strain map can then be calculated by measuring changes in the lattice constants from the diffraction patterns [3]. In addition, virtual dark-field images can be computed from the diffraction maps [4]. The main practical limitation for this method is the speed of the electron detector. To overcome this limitation we used a Gatan K2 IS direct detection camera operating at a frame rate of 400 f/s and recorded a series of 22 nanodiffraction maps (64x64pixel) to calculate a time dependent local strain-map from this data. A Hysitron PI-95 picoindenter is used to pull an AlMg sample in tension. Figure 2 shows images extracted from the time-resolved dataset. The ADF image (cf. Fig 2a) shows complex diffraction contrasts from defects and misorientations in the sample. In the colorcoded strain map (cf. Fig 2b) two moving dislocations are visible. The results facilitate the comparison of the local and transient strains occurring around moving dislocations with the global strains measured with the picoindenter. These results demonstrate how novel fast electron detectors enable real-time collection of diffraction maps at sufficient speeds to be combined with in-situ microscopy. The resulting datasets carry a large amount of information. Here we have shown how a time-resolved local strain-map can be extracted from a dataset acquired during in-situ deformation. [1] Q. Yu, M. Legros and A.M. Minor, MRS Bulletin 40 (2015) p. 62. [2] M. J. H�tch, E. Snoeck, and R. Kilaas, Ultramicroscopy 74 (1998) p. 131. [3] V.B. Ozdol, C. Gammer, M.C. Sarahan and A.M. Minor, Microscopy and Microanalysis 20 S3 (2014) p. 1046. [4] C. Gammer, V.B. Ozdol, C.H. Liebscher, and A.M. Minor, submitted to Ultramicroscopy. Microsc. Microanal. 21 (Suppl 3), 2015 2326 [5] The authors acknowledge support by the Austrian Science Fund (FWF):[J3397] and the Molecular Foundry, Lawrence Berkeley National Laboratory, which is supported by the U.S. Dept. of Energy under Contract # DE-AC02-05CH11231. Figure 1. Setup used for nanodiffraction strainmapping. During acquisition of an ADF image in STEM mode, a diffraction pattern is acquired for every probe position. A strain map is calculated from the nanodiffraction map by measuring changes in the lattice constants from the diffraction patterns. Figure 2. Frame extracted from the in-situ strain mapping video. (a) ADF image acquired during deformation showing diffraction contrast due to defects and misorientations in the sample. (b) The corresponding color-coded strain map shows two moving dislocations.");sQ1[1162]=new Array("../7337/2327.pdf","Fast Imaging of Carbon Nanotube Carpet Growth by Environmental TEM","","2327 doi:10.1017/S1431927615012416 Paper No. 1162 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Fast Imaging of Carbon Nanotube Carpet Growth by Environmental TEM Dmitri N. Zakharov1, Mostafa Bedewy2, Viswanath Balakrishnan2, Sebastian W. Pattinson2, Eric R. Meshot3, A. John Hart2, and Eric A. Stach1 1. 2. Center for Functional Nanomaterials, Brookhaven National Laboratory, Upton, NY 11973, USA The Laboratory for Manufacturing and Productivity, Massachusetts Institute of Technology, Cambridge, MA 02139 USA 3. Lawrence Livermore National Laboratory, Livermore, CA 94550 USA Hata, et al. [1] reported a major breakthrough in carbon nanotube (CNT) growth, which allows the creation of thick (>1 um) films of single-walled CNTs to be grown from a substrate (called carpets), broadly free of other carbon allotropes. This eliminates the additional step of CNT purification often required for further studies or applications. The simplicity of the method, in conjunction with the ability to control numerous growth conditions and parameters, has made it a model setup to study fundamental mechanisms in the growth and nucleation of CNTs. It has opened a pathway for investigations concerning growth effects on CNT diameter, number of walls, and chirality dependence as a function of an array of experimental growth conditions. For example, it was shown that catalyst particles undergo Ostwald ripening and subsurface diffusion during the growth process [2]. As a result there is active catalyst mass loss that triggers accumulating growth termination of individual CNTs, followed by abrupt termination of the CNT forest growth due to the loss of a self-supporting structure [3]. Here we investigate the initial stages of CNT carpet growth, the collective behavior of nucleated CNTs, as well as the nucleation and growth of individual tubes by environmental TEM (ETEM). The ETEM imaging is done with 2.5 millisecond time resolution inside an FEI imagecorrected Titan 80-300 equipped with a Gatan K2-IS fast rate camera. The camera utilizes direct electron detection CMOS chip that has a high resistivity to electron damage and a high sensitivity capable of delivering 400 frames per second at 2k x 2k. TEM samples are prepared in plan-view and cross-sectional configurations by electron beam assisted deposition of 10 nm Al2O3 and then 1 nm of Fe onto 30 nm thick SiO2 TEM windows. It is found that as deposited, the iron is present in the form of an oxide. Metallic iron catalyst particles form upon exposure to a reducing hydrogen environment (40 mTorr) at 750oC, however, after 30 minutes treatment only a partial reduction to metallic iron is observed (according to a series of selected area electron diffraction patterns). Further reduction is observed within the first several seconds following the introduction of acetylene (10 mTorr at 650oC) into the ETEM reaction chamber. Figure 1 shows frames from a movie showing the initial stages of CNT carpet growth in plan-view configuration. The cross-sectional configuration allows us to evaluate the growth rate of individual CNTs with time. It is observed that the growth rate directly correlates with changes in the CNT diameter and the generation of defects in the CNT wall (Figure 2), showing a tight correlation between CNT structure and growth modes. In general, the current ETEM setup allows us to study samples in both plan-view and cross-section configurations, to correlate CNT growth and nucleation with catalyst structure and morphology, as well as growth conditions (temperature, pressure of carbon containing precursors, reducing or buffer agents), and last but not least, correlate findings on a larger scale with ex-situ conventional CVD systems [4]. The power of this approach in understanding carpet growth will be highlighted in the presentation. Microsc. Microanal. 21 (Suppl 3), 2015 2328 References: [1] K. Hata, D.N. Futaba, K. Mizuno, T. Namai, M.Yumura, S. Iijima, "Water-Assisted Highly Efficient Synthesis of Impurity-Free Single-Walled Carbon Nanotubes", Science 306, pp. 1362 (2004) [2] S.M. Kim, C.L. Pint, P.B. Amama, D.N. Zakharov, R.H. Hauge, B. Maruyama, and E.A. Stach, "Evolution in Catalyst Morphology Leads to Carbon Nanotube Growth Termination", J. Phys. Chem. Lett. 1, pp. 918�922 (2010) [3] M. Bedewy, E.R. Meshot, H. Guo, E.A. Verploegen, W. Lu, A.J. Hart, "Collective Mechanism for the Evolution and Self-Termination of Vertically Aligned Carbon Nanotube Growth", J of Phys. Chem. C 113, pp. 20576-20582 (2009) [4] Research carried out in part at the Center for Functional Nanomaterials, Brookhaven National Laboratory, which is supported by the U.S. Department of Energy, Office of Basic Energy Sciences, under Contract No. DE-SC0012704. Figure 1. Plan-view configuration shows the initial stages of CNT carpet growth inside the ETEM. Individual frames (a) � (c) from a movie illustrate how both the catalyst particles and the CNTs evolve under growth conditions. Black arrows mark the merging of two catalyst particles that results in one tube terminating while the other tube continues to grow. White arrows point to a catalyst particle that has delayed tube nucleation. The white box highlight a change in the particle shape associated with the growth of a CNT and the formation of new catalyst particle. Figure 2. Individual frames (a) � (f) from a movie of individual three-walled carbon nanotube growth in cross-sectional configuration. (g) Growth rate of the nanotube versus growth time.");sQ1[1163]=new Array("../7337/2329.pdf","Compression Algorithm Analysis of In-Situ (S)TEM Video: Towards Automatic Event Detection and Characterization","","2329 doi:10.1017/S1431927615012428 Paper No. 1163 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Compression Algorithm Analysis of In-Situ (S)TEM Video: Towards Automatic Event Detection and Characterization Jeremy R. Teuton1, Richard L. Griswold2, B. Layla Mehdi1 and Nigel D.Browning1 1. Fundamental and Computational Sciences, Pacific Northwest National Laboratory, Richland, WA 99352, USA 2. National Security Division, Pacific Northwest National Laboratory, Richland, WA 99352, USA Precise analysis of both (S)TEM images and video are time and labor intensive processes. As an example, determining when crystal growth and shrinkage occurs during the dynamic process of Li dendrite deposition and stripping involves manually scanning through each frame in the video to extract a specific set of frames/images. For large numbers of images, this process can be very time consuming, so a fast and accurate automated method is desirable. Given this need, we developed software that uses analysis of video compression statistics for detecting and characterizing events in large data sets. This software works by converting the data into a series of images which it compresses into an MPEG-2 video using the open source "avconv" utility [1]. The software does not use the video itself, but rather analyzes the video statistics from the first pass of the video encoding that avconv records in the log file. This file contains statistics for each frame of the video including the frame quality, intra-texture and predicted texture bits, forward and backward motion vector resolution, among others. Avconv records 15 statistics for each frame. By combining different statistics, we have been able to detect events in various types of data. We have developed an interactive tool for exploring the data and the statistics that aids the analyst in selecting useful statistics for each analysis. Going forward, an algorithm for detecting and possibly describing events automatically can be written based on statistic(s) for each data type. Next, we applied this process to the series of (S)TEM images and videos from an in-situ liquid electrochemical stage for (scanning) transmission electron microscopes (in-situ liquid ec(S)TEM) demonstrating dynamic changes at the Pt working electrode surface during Li dendrite growth and dissolution process in liquid LiPF6 in PC electrolyte used in state-of-the-art Li-ion battery systems [2]. The electrochemical stability of electrolytes in the new generation of high energy densities Li-ion batteries is disturbed by both the oxidizing nature of the cathode and reducing nature of the anode, resulting often in the accumulation of insulating side products (thick insulating layer or crystal growth on the anode/cathode surface) which leads to capacity fading. After selecting the appropriate statistics, we were able to identify distinct changes such as the beginning of the crystals growth at the Pt surface, plateau of growth, and/or shrinking. Comparing our first pass algorithm to manually scoring the data, our algorithm typically found the events within �13 frames of the actual event. In one data set where the algorithm detected crystal growth much earlier than the human analyst (21 frames early), further investigation showed that the human analyst had missed the actual start of the crystal growth, due to the very subtle changes not readily apparent to the human eye. With further optimization on a larger data set, we hope to improve the identification accuracy of dynamic changes during electrochemical processes at the electrolyte/electrode interface during in-situ liquid ec-(S)TEM measurements, which would enhance time consuming video analysis and potentially replace traditional time and effort intensive techniques [3]. Microsc. Microanal. 21 (Suppl 3), 2015 2330 References: [1] Libav team, "Libav: Open source audio and video processing tools", libav.org [2] B.L. Mehdi, J. Qian, E. Nasybulin, C. Park, D. A. Welch, R. Faller, H. Mehta, W. A. Henderson, W. Xu, C. M.Wang, J. E. Evans, J. Liu, J. -G. Zhang, K. T. Mueller, N. D. Browning, (Submitted) [3] Research described in this paper is part of the Analysis in Motion Initiative at Pacific Northwest National Laboratory and was conducted under the Laboratory Directed Research and Development Program at PNNL. In-situ TEM work was primarily supported by Joint Center for Energy Storage Research (JCESR), an Energy Innovation Hub funded by the Department of Energy, Office of Science, Basic Energy Sciences. The development of the operando stage was supported by the Chemical Imaging Initiative, a Laboratory Directed Research and Development Program at Pacific Northwest National Laboratory (PNNL). PNNL is a multi-program national laboratory operated by Battelle for the U.S. Department of Energy (DOE) under Contract DE-AC05-76RL01830. A portion of the research was performed using EMSL, a national scientific user facility sponsored by the Department of Energy's Office of Biological and Environmental Research and located at PNNL. Figure 1. Interactive data explorer tool, showing selected statistics from two different data sets. Figure 2. Annotated plot of the Miscellaneous Bits video compression statistic");sQ1[1164]=new Array("../7337/2331.pdf","A New Discrete Tomographic Reconstruction Method for Electron Tomography","","2331 doi:10.1017/S143192761501243X Paper No. 1164 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A New Discrete Tomographic Reconstruction Method for Electron Tomography Ilke Arslan1, Toby Sanders1,2, Peter Binev2, Bruce C. Gates3, Alexander Katz4 1 2 Pacific Northwest National Laboratory, Richland, WA 99352, USA. University of South Carolina, Columbia, SC, 29208, USA 3 University of California-Davis, Davis, CA 95616, USA. 4 University of California�Berkeley, Berkeley, CA 94720, USA. We have developed a new and advanced reconstruction algorithm for electron tomography data using a combination of compressed sensing and discrete reconstruction methods. Our algorithm not only provides accurate reconstructions with limited projection images, but also with a limited tilt range. This algorithm offers a solution to the decades-old "missing wedge" artifact present in reconstructions with limited data. With the ability to provide high quality reconstructions with only ~30 images as opposed to the standard ~120, this algorithm opens the door to materials characterization of electron beam sensitive materials across all fields of science. This 3-D characterization method with 1nm to sub-nm resolution provides a new methodology for the advancement of nanotechnology. The reduced number of images necessary for a robust 3-D reconstruction also opens the door for 3-D chemical imaging, using EDX or EELS, where the number of tilts at which data are acquired is also limited. Standard methods of 3D reconstruction, such as weighted back projection (WBP) and simultaneous iterative reconstruction technique (SIRT), are not equipped to handle this lack of information, and result in significant blurring. Two of the recent successful algorithms are the discrete algebraic reconstruction technique (DART) and total variation (TV) minimization within compressed sensing (CS). The DART algorithm uses an algebraic reconstruction method (ARM) and pairs it with the prior knowledge that there are only a small number (two or three) of different materials in the sample, each corresponding to a different gray value in the reconstruction. An initial reconstruction is computed using the ARM and rounded to the chosen fixed gray values based on some threshold, and iteratively refined using the ARM. The method of TV minimization stems from the mathematical theory of compressed sensing and only recently became available due to new computational methods for solving the TV minimization problem. The method considers the characterization of real images and encourages the reconstruction to take larger jumps in gray values to create clear boundaries, hence creating a similar effect to that of DART. Figures 1-2 show a comparison of using DART with SIRT as an initial solution, and TVM as an initial solution. Using all of the projections for this data set (73 images), the two solutions are quite similar (Figure 1). However, using half of the data (37 images), it can be seen in Figure 2 that SIRT+DART does not reconstruct the small particles within the pores, while TVM+DART is able to reconstruct the particles. Further examples of these methods will be presented, covering different material geometries, such as layered zeolite materials, porous supports, and particles in supports. This research was funded in part by the laboratory directed research and development (LDRD) program at the Pacific Northwest National Laboratory, and in part by the DOE BES DE-SC0005822. PNNL is operated by Battelle under contract DE-AC05-76RL01830. Microsc. Microanal. 21 (Suppl 3), 2015 2332 73 Projections Figure 1. A comparison of DART reconstructions using SIRT and TV as initial solutions using all of the projections in a data set, in this case 73 images. The top images show the initial solutions, while the bottom images show slices through the final reconstruction. The graph displays the relative error at every iteration of both methods. 37 Projections Figure 2. A comparison of DART reconstructions using SIRT and TV as initial solutions using half of the projections in a data set, in this case 37 images. The top images show the initial solutions, while the bottom images show slices through the final reconstruction. The graph displays the relative error at every iteration of both methods.");sQ1[1165]=new Array("../7337/2333.pdf","Challenges and Opportunities in 3D Tri-gate Transistor Characterization","","2333 doi:10.1017/S1431927615012441 Paper No. 1165 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Challenges and Opportunities in 3D Tri-gate Transistor Characterization Jiong Zhang 1, Ling Pan1, Andrew A. Herzing2, Ray Twesten3, Paolo Longo3, Noel Franco1, Kevin Johnson1 and Zhiyong Ma1 1. Intel Corporation, Corporate Quality Network, Hillsboro, OR 97124 2. National Institute of Standards and Technology, Material Measurement Laboratory, Gaithersburg, MD 20899 USA 3. Gatan Inc., Pleasanton, CA 94588 Dimensional scaling has provided performance improvement and power reduction in the era of traditional MOSFET scaling. In the past decade, transistor performance has progressed through the introduction of innovations including strained silicon, high-k/metal gate, and novel architectures. Multigate devices have long held the promise of improved transistor electro-statics, offering improved performance at lower supply voltages and significantly reduced short channel effects [1]. Intel has deployed fundamentally different 3D tri-gate transistors manufactured at 22 nm [2], which is the first in the industry to exploit fin based tri-gate devices and combine the benefits of strained silicon and highk/metal-gate. With a smaller 3D tri-gate transistor, Intel can design even more powerful processors with incredible power efficiency. Critical to the success of the 3D tri-gate technology are the analytical techniques, supporting process development. Transmission electron microscopy (TEM) with its unique atomic resolution and analytical capabilities has become crucial and irreplaceable in technology research and development as well as product manufacturing in the semiconductor industry. The Intel 22nm transistor continued using epitaxial SiGe for PMOS strain engineering on the new 3D tri-gate architecture [2]. The new tri-gate structure raises more challenges for strain characterization. Optical-based methods can provide a route for macro-scale stress measurements at the die and wafer levels while TEM is the only tool providing nanoscale strain measurement. Traditional TEM strain techniques that work for planar transistors, such as convergent beam electron diffraction and dark-field electron holography, do not apply to tri-gate transistors due to the complicated 3D tri-gate geometry. The only e-beam based technique suitable for the tri-gate transistor is nano-beam electron diffraction (NBED). It has the advantage to well separate the device channel and the overlapping gate materials in diffraction space. Figure 1 shows the 2D strain mapping of a tri-gate transistor along the channel [110] direction and [001] direction. The 2D strain map elucidates the mechanical strain more intuitively and can be used to compare with simulation more accurately, guiding strain engineering development. The 3D tri-gate architecture also drives TEM characterization into three dimensional analysis. Tomography provides critical structure information which cannot be seen in typical 2D imaging due to the projection effect, especially when the architecture is more and more complicated and overlapped. Providing efficient and accurate 3D tomography results is industry priority in both tomography data acquisition and analysis. We have employed a subpixel registration method for the alignment based on minimization of the least squares difference of image intensities, as well as an iterative approach for the tilt axis adjustment, resulting in a significant reduction in reconstruction artifacts compared to traditional cross-correlation alignment and manual tilt adjustment [3] [4]. Figure 2 shows a scanning transmission electron microcopy (STEM) tomography result from a 22nm PMOS transistor. The missing wedge artifact is minimized with the full tilt needle sample, and the improved alignment as shown in Figure 2b. Figure 2c shows a preliminary Electron Energy Loss Spectroscopy (EELS) result at two different tilt angles from the same sample. Chemical tomography helps when the elements cannot be distinguished in Microsc. Microanal. 21 (Suppl 3), 2015 2334 STEM, and EELS tomography can provide additional bonding information in the fine structure, making it even more powerful. Transistor technology scaling has been pushing materials characterization into a new era in terms of resolution, sensitivity, multiple dimensions as well as automation for both structural and chemical analysis. Innovative metrologies are always needed to go beyond the requirement and speed of process development in order to maintain the continuity of Moore's Law. References: [1] D Hisamoto et al, IEDM (1989) p. 833 [2] C Auth et al, VLSI Technology (2012) p. 131 [3] P Thevenaz et al, IEEE Transaction on NBD Processing, volume 7 (1998) p. 27 Image Strain Map th [4] D Wolf et al, Proceedings of the 15 European Microscopy Congress [5] ORTEM lab engineers and artists are gratefully acknowledged for the TEM works to support the 22nm tri-gate technology [110] [001] Figure 1. NBED strain map of a tri-gate transistor in [110] channel direction and [001] direction. First NBD Strain Map Figure 2. STEM tomography of a needle PMOS tri-gate transistor. Slices in (b) are extracted at the locations denoted by arrows in (a). (c) STEM images and EELS maps at two different tilt angles for EELS tomography.");sQ1[1166]=new Array("../7337/2335.pdf","Robust Physical Alignment Models for Electron Tomography","","2335 doi:10.1017/S1431927615012453 Paper No. 1166 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Robust Physical Alignment Models for Electron Tomography Toby Sanders1, Micah Prange2, Peter Binev1, Cem Akatay3, Ilke Arslan2 1 2 University of South Carolina, Columbia, SC, 29208, USA Pacific Northwest National Laboratory, Richland, WA, 99352, USA 3 UOP LLC, a Honeywell Company, 50 E. Algonquin Rd., 60016, Des Plaines, IL. Tomography is a widely practiced technique for non-invasive 3-D imaging at all scales, with applications to many areas of science. In general, tomography involves three major phases: acquisition of 2-D projections of a sample, registration of the projections into a common coordinate system, and reconstruction of aligned data with a reconstruction algorithm. The acquisition methods and reconstruction algorithms have generated a great deal of interest due to the potential for improved resolution. On the other hand, the registration methods have not generated a great deal of research, and have typically used old hopeful models such as cross-correlation, or resorted to decorating the sample with gold particles for tracking. While one cannot improve the resolution capabilities with accurate alignment, inaccurate alignment or introduction of high density tracking particles can result in direct loss of resolution, making the alignment equally important. Most recently, others have begun to use the physical motion of the center-of-mass of the sample as means for a more mathematically justifiable and noninvasive alignment technique [1,2]. These methods have proven useful, even capable of atomic resolution in cases [1]. However, even theoretically these methods can be shown to fall short of sufficient robustness in some simple settings, such as when the projected sample is not a fixed volume. In our work, we develop more general alignment methods based on the center-of-mass, but allow for greater flexibility with our models by observing motions at many local scales, where the projected volume at these scales is in fact fixed. These observations allow us to continue to make use of the justifiable center-of-mass alignment methods and apply it in a much more general setting, making it a very robust approach. In figures 1 and 2, reconstructions of an alumina particle are shown, where the data was aligned with our method and cross-correlation. In figure 1, the plot of the position of a local center-of-mass in the sample is given as a function of the projection angle, showing that the path of this mass follows a mathematically feasible trajectory, where cross-correlation fails to do so. Also in figure 1, a reconstruction of this region from the two alignments shows better results with our alignment. In figure 2, the full 3-D reconstructions are visualized with a volume rendering, in a color map that varies smoothly from black into red, and then to white. With cross-correlation, the red glow around the reconstructed particle is a result of blurring due to misalignment. This research was funded by the Laboratory Directed Research and Development program at Pacific Northwest National Laboratory, under contract DE-AC05076RL01830. It was also funded in part by NSF grant DMS 1222390. Microsc. Microanal. 21 (Suppl 3), 2015 2336 [1] M.C. Scott, et al. in "Electron Tomography at 2.4 �angstrom resolution," Nature, 483, p.444- 447. (2012). [2] T. Sanders, et al in "Physically Motivated Global Alignment Method for Electron Tomography," Adv. Chem. And Struct. Im. (2015). Figure 1: Left column: Results from our alignment. Right column: Results from cross-correlation alignment. a,b, Location of a local center-of-mass as a function of the projection angle. In blue is the calculated center-of-mass and in red is the best-fit curve for a feasible path of the center-of-mass. c,d, The resulting reconstruction of this local region. Figure 2: a, 3-D reconstruction from our alignment. b, 3-D reconstruction from cross-correlation.");sQ1[1167]=new Array("../7337/2337.pdf","Reconstruction Strategies for Combined Tilt- and Focal Series Scanning Transmission Electron Microscopy","","2337 doi:10.1017/S1431927615012465 Paper No. 1167 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Reconstruction Strategies for Combined Tilt- and Focal Series Scanning Transmission Electron Microscopy Tim Dahmen1, Holger Kohr2, Niels de Jonge3 and Philipp Slusallek1 German Research Center for Artificial Intelligence GmbH (DFKI), 66123 Saarbr�cken, Germany KTH Royal Institute of Technology, Dep. of Mathematics, Lindstedtsv�gen 25, Stockholm, SE 100 44 3 INM - Leibniz Institute for New Materials, 66123 Saarbr�cken, Germany Email of the presenting author: Tim.Dahmen@dfki.de 2 1 Combined tilt- and focal series (TFS) scanning transmission electron microscopy (STEM) is a recently developed method to obtain nanoscale three-dimensional (3D) information of thin specimens [1], [2]. In this recording scheme, the specimen is tilted in relatively large angle increments, and a focal series is recorded for every tilt angle, which reduces the number of mechanical tilts. In addition, the overall tilt range is smaller than used for conventional tilt series. The computational problem of volume reconstruction from projections can be solved for this kind of data using iterative reconstruction algorithms. The first such method was the tilt-focal algebraic reconstruction technique (TF-ART), which is based on the idea of heuristic weighting and an unmatched projection/backprojection pair. But the reconstruction required high computational effort. To address this issue, the STEM transform has been formulated as a generalization of the well-known Ray-transform for parallel illumination, taking into account the reduced depth of field (higher beam converge angle) of aberration corrected STEM compared to TEM or standard STEM. It can be shown that the STEM transform is self-adjoint, i.e. *g = p * g = g, where is the STEM transform for projection direction , * the adjoint, g is the image, and p is the probe function [3]. With this result, it has become possible to develop a variation of the iterative reconstruction algorithm using the adjoint operator for the backprojection. The implementation of the new algorithm exploits the fact that the convolution required to evaluate the STEM transform can be precomputed per projection direction, resulting in an efficient implementation using linear interpolation. Another interesting result concerns the Fourier Transform (FT) of the STEM transform. In tomogaphic applications with parallel illumination, the Fourier slice theorem states that the FT of a projection contains exactly those 3D spatial frequencies of the volumetric function that lie on a rotated 2D plane perpendicular to the projection axis. For the STEM transform, a similar result can be derived and it can be shown that the FT of the STEM transform is non-zero in a set that corresponds to an inverse doublecone, i.e. all of Fourier space except a double cone of opening angle 1-, where is the opening angle of the electron beam. The STEM transform was thus formulated as a mathematical model applicable to STEM imaging with a convergent electron beam. It was shown that it is (1) a linear convolution, (2) a generalization of the Ray transform that contains the latter as the special case where the beam convergence semi-angle 0, and (3) self-adjoint, a result that facilitated a new iterative reconstruction algorithm for TFS based on a matched backprojection, which drastically improved the convergence rate, resulting in 60 times less iterations compared to previous methods. It also solved theoretical concerns about the convergence of the method, which was not guaranteed in the case of an unmatched projection/backprojection pair. This brings the combined tilt- and focal series one more step towards broad applicability by allowing the reconstruction of high resolution tomograms in feasible computation time. Microsc. Microanal. 21 (Suppl 3), 2015 2338 References: [1] T Dahmen, H Kohr, N deJonge and P Slusallek Microsc. Microanal. 20 (2014), pp. 548�60 [2] R Hovden et al Ultramicroscopy 140 (2014), pp. 26�31 [3] T Dahmen et al "Matched Backprojection Operator for Combined Scanning Transmission Electron Tilt- and Focal Series" under review Microsc. Microanal. (2015) Figure 1: Parametrization of the STEM Transform. The electron beam is modeled as a double-cone, with the vertex v at the focal point. The Beam opening angle is , the beam direction . By scanning the electron probe over the plane , an image with limited depth of field is created. Figure 2: Geometric interpretation of the Fourier slice theorem for the STEM transform in Fourier space. a) The frequencies covered by one projection correspond to the shape of a double wedge of opening semi-angle . b) If the tilt increment is chosen as =2, neighboring wedges overlap in a non-trivial shape (red). c) Considering a cross-section through the origin and perpendicular to the tilt axis, the wedges seamlessly cover the frequency space. d) A cross-section shifted along the tilt axis reveals a complex-shaped region in frequency space that contains information from more than one tilt direction. e) A cross-section even further along the tilt axis towards highest frequencies exposes that the region containing information from both tilt directions expands towards higher frequencies.");sQ1[1168]=new Array("../7337/2339.pdf","HAADF-STEM and Super-XTM XEDS Tomography of Complex Nano-scale Precipitates in a High Entropy Alloy, AlMo0.5NbTa0.5TiZr","","2339 doi:10.1017/S1431927615012477 Paper No. 1168 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 HAADF-STEM and Super-XTM XEDS Tomography of Complex Nano-scale Precipitates in a High Entropy Alloy, AlMo0.5NbTa0.5TiZr J.K. Jensen1, J.M. Sosa1, D.E. Huber1,2, G.B. Viswanathan1, R.E.A. Williams1,2, H.L. Fraser1 1. 2. Center for the Accelerated Maturation of Materials, The Ohio State University, Columbus, OH Center for Electron Microscopy and Analysis, The Ohio State University, Columbus, OH High entropy alloys (HEAs) are a relatively new class of materials garnering a great deal of attention due to their remarkable balance of properties, including high strength, toughness, ductility, and corrosion resistance. In contrast to conventional alloy systems, HEAs are based on 4 or more principal elements with near equimolar concentrations and tend to have simple microstructures due to the preferential formation of solid solution phases attributed to the high configurational entropy of mixing [1]. HEAs appear to offer new pathways to lightweighting in structural applications, new alloys for intermediate and elevated temperature components, and new magnetic materials. However, to realize this potential requires considerable alloy development that will rely on application of integrated computational materials (science and) engineering (ICME). The development of accurate computational models to predict alloy performance requires a detailed knowledge of the morphology and interconnectedness of microstructural features. Due to spinodal decomposition, these alloys often consist of a mixture of ordered and disordered phases on the nano-scale, and because of the compositional and structural complexity, it is desirable to characterize the morphology as well as the chemical nature of the precipitates for inclusion in ICME models. Electron tomography is a powerful characterization technique capable of resolving microstructural features at length scales as small as a few nanometers when coupled with the high spatial resolution available with modern scanning transmission electron microscopes (STEM). For STEM tomography specifically, an electron beam is transmitted through a sample at incremental degrees of rotation and images are recorded at each tilt. The images are then processed to reconstruct a three-dimensional representation of the area of interest. Tomographic reconstructions, in general, may be produced by any set of images collected though an angle of rotation with the quality of the reconstruction depending directly on the angular tilt range collected. A previous tomography study on the AlMo0.5NbTa0.5TiZr HEA was reconstructed from only high angle annular dark-field (HAADF) STEM images and successfully resolved the nano-scale precipitates in the microstructure [2]. While HAADF contrast was sufficient to resolve and segment the nano-scale precipitates in this HEA, the reconstruction provided only morphological information and no compositional data. In order to obtain a novel perspective of the 3D elemental segregation in the HEA, this paper will present the use of Super-XTM energy dispersive xray spectroscopy (XEDS) spectral images (SI) utilized in conjunction with HAADF-STEM images for an improved tomographic reconstruction of the AlMo0.5NbTa0.5TiZr alloy. Rather than using a conventional thin foil STEM sample geometry, a tapered needle-like sample was created using a FEI HeliosTM DualBeam FIB. This geometry was selected to eliminate thickness variations that occur with larger degrees of tilting with flat samples as well as to allow for 360� tilting using a Fischione 2050 on-axis rotation tomography holder. A tilt series of HAADF-STEM micrographs and Super-XTM XEDS SI were collected using a FEI TitanTM 60-300 STEM at 1� increments along the longitudinal axis of the sample. XEDS SI were collected for 300s with a dwell time/pixel of 20 s. All data was aligned, segmented and reconstructed using MIPARTM image processing software [3]. A Microsc. Microanal. 21 (Suppl 3), 2015 2340 weighted back projection was then applied using the TomoJ plugin in ImageJ. Image processing produced a three-dimensional reconstruction of the microstructure based off the HAADF images and the XEDS signals. Figure 1(a) shows the rendered 3D volume of the AlMo0.5NbTa0.5TiZr HEA, revealing a periodic microstructure composed of cuboidal and platelet precipitates. The reconstructions of the Zr and Ta EDX maps, shown in Figure 1(b) and 1(c) respectively, illustrate the segregation of Zr to the ordered B2 matrix and Ta to the disordered cuboidal and platelet precipitates. This 3D compositional variation was inherently difficult to observe through traditional 2D characterization, specifically the case of thin platelet in a stacked morphology where precipitate/matrix projection artifacts prevent local 2D compositional measurements. References: [1] J.W. Yeh et al, Materials Science Forum 560 (2007), p. 1. [2] J.M. Sosa et al, Materials Science and Technology (2015). Submitted [3] J.M. Sosa et al, Integrating Materials and Manufacturing Innovation 3 (2014), p. 18 [4] The authors acknowledge funding and support from the DAGSI program and the Air Force Research Lab (AFRL) in Dayton, OH. Thanks to Fischione for use of the 2050 on-axis tomography holder. Figure 1. (a) 3D HAADF reconstruction of AlMo0.5NbTa0.5TiZr tapered needle with segmented cuboidal and platelet precipitates superimposed with Super-XTM XEDS tomographic reconstructions of (b) Ta and (c) Zr elemental maps. (d) Composite reconstruction with Ta (red), Zr (green), and HAADF (white) 3D renderings.");sQ1[1169]=new Array("../7337/2341.pdf","Linear chemically sensitive electron tomography using DualEELS and compressed sensing","","2341 doi:10.1017/S1431927615012489 Paper No. 1169 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Linear chemically sensitive electron tomography using DualEELS and compressed sensing Alaa AlAfeef 1,2, W. Paul Cockshott1, Patrick Barges2, Ian Zuazo2, Joanna Bobynko3, Alan J. Craven3, and Ian Maclaren3 1 2 School of Computing Science, University of Glasgow, Glasgow, G12 8QQ, UK ArcelorMittal Maizi�res Research, Maizi�res-l�s-Metz, 57283, France 3 SUPA School of Physics and Astronomy, University of Glasgow, Glasgow, G12 8QQ, UK Electron tomography (ET) is now increasingly important for recovering the three-dimensional (3D) morphology of nanostructured materials in the physical and life sciences. ET typically involves the acquisition of a set of two-dimensional projection images at different tilts using (scanning) transmission electron microscopy ([S]TEM), followed by alignment and reconstruction using established algorithms to reconstruct a 3D volume that represent the physical morphology or 3D distribution of some other property of the specimen under investigation. In principle, the methodology is independent of the nature of the images and is applicable to any imaging technique that fulfils the projection requirement [1] such that the signal should change monotonically with the physical property of the sample. This condition is approximately fulfilled for mass-thickness contrast in bright field TEM of amorphous biological specimens, and high angle annular dark field (HAADF) STEM imaging of thin specimens. Consequently, both imaging techniques have been widely used in ET. Recently, ET has been performed using spectroscopic signals, including X-ray spectroscopy, energyfiltered TEM (EFTEM), and electron energy loss spectroscopy (EELS) in the STEM, to achieve a chemically sensitive 3D reconstruction. Whilst X-rays and EFTEM mainly allow the mapping of elemental contents, EELS offers additional possibilities for studying detailed chemistry. Early studies have already shown the feasibility of EELS-STEM tomography [2,3]. Nevertheless, EELS suffers significantly from multiple scattering, especially for thicker specimens, and this makes the backgroundsubtracted edge signal a non-linear function of thickness, which leads to reconstruction artifacts. Using DualEELS we can simultaneously record the low- and high-loss EELS signals for each pixel in a spectrum image, and this allows the subsequent deconvolution of the multiple scattering out of the highloss signal. This gives edges with integrated intensities that have a clear linear relationship to thickness. This was performed on VCx particles on a carbon extraction replica of a vanadium microalloyed highmanganese steel. DualEELS spectrum images (SIs) were recorded at approximately 10� intervals from about -50� to +50� using a JEOL ARM200F equipped with a Gatan GIF Quantum ER spectrometer. The data was processed by energy alignment of all SIs using the zero loss peak, noise reduction using principal component analysis, deconvolution to remove plural scattering, and production of vanadium maps with a 50 eV integration window using a recently developed procedure [4]. 3D reconstruction was performed using both the well-known SIRT algorithm, as well a new compressed-sensing based algorithm, DLET [5], which is a dictionary-based and very noise tolerant algorithm that can obtain highfidelity reconstructions from far fewer projections than are normally required by traditional ET methods. Figure 1 shows a comparison between SIRT and DLET for reconstruction using the V-L2,3 EELS signal. It can be noted that DLET gives a noticeable reduction in z-elongation due to missing wedge artifacts, which is a common feature of tomographic reconstructions (although this elongation is not totally Microsc. Microanal. 21 (Suppl 3), 2015 2342 eliminated), as well as reduced noise within and outwith the precipitate. Figure 2 shows line profiles through two orthoslices for both this EELS reconstruction of Figure 1, as well as a 3D reconstruction from the HAADF signal recorded simultaneously with the EELS spectrum image. These both show a very flat profile through the centre of the particle, and no apparent hollowing ( "cupping" artifact), which would be a signature of a non-linear relationship between signal and thickness. Thus, appropriate use of DualEELS enables linear chemically sensitive tomography on nanoparticles of an industrially important material. This combined with a new compressed sensing based algorithm, DLET, allows high-fidelity, low-noise reconstruction from relatively few projections and minimizes artifacts from the missing wedge in the tilt data. This could easily be extended to encompass changes in near-edge EELS fine structure in order to produce 3D views of chemical form or bonding in nanostructured materials. References: [1] P. W Hawkes, in "Electron tomography, 2nd edition", ed. J. Frank (Springer, New York). [2] K. Jarausch et al., Ultramicroscopy 109 (2009) p. 326. [3] L.Yedra, et al., Ultramicroscopy 122 (2012) p.12-18. [4] J. Bobynko, I. MacLaren, and A. J. Craven, Ultramicroscopy 149 (2015) p. 9. [5] A. AlAfeef et al., J. Phys.: Conf. Ser 522 (2014) P. 012021. [6] AAA acknowledges funding for a PhD studentship from the Lord Kelvin/Adam Smith Scholarship of the University of Glasgow. The work on high-Mn steels was funded by the EU Research Fund for Coal and Steel (Precipitation in High Manganese Steels, RFSR-CT-2010-00018). This work was made possible by generous provision of the MagTEM facility by SUPA and the University of Glasgow. Figure 1. A comparison of reconstructions of one VCx precipitate from the V-L2,3 signal using SIRT and the new DLET algorithm showing X-Y, and X-Z views, as well as a comparison to an idealized octahedron. Note that the distortion in the Z (missing wedge) direction is much reduced using DLET. Figure 2. Orthoslices through the centre of two reconstructions of a precipitate made using (Left) the V-L2,3 EELS signal, and (Right) the HAADF signal recorded simultaneously. Both were reconstructed using the DLET algorithm. In both cases, line profiles of density are shown corresponding to a horizontal line through the center of Orthoslices and neither show any cupping artifacts.");sQ1[1170]=new Array("../7337/2343.pdf","In-situ AFM and SEM Investigation of Slip Steps Evolving during Nanoindentation","","2343 doi:10.1017/S1431927615012490 Paper No. 1170 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In-situ AFM and SEM Investigation of Slip Steps Evolving during Nanoindentation Megan J. Cordill1, Josef Kreith2 and Georg E. Fantner3 1. 2. Erich Schmid Institute of Materials Science, Austrian Academy of Sciences, Leoben, Austria Department of Materials Physics, Montanuniversit�t Leoben, Leoben, Austria 3. Laboratoray for Bio- and Nano-Instrumentation, EPFL, Lausanne, Switzerland Since plasticity in metallic materials is usually attributed to the motion of dislocations, the knowledge of the evolution of the dislocation structure during deformation is the key to understanding the mechanical properties on all length scales. Especially for samples in the micrometer regime, the evolution of the slip-step pattern can be directly linked to the active dislocation sources. Depending on the number of operating dislocation sources on the different slip systems (slip planes) a characteristic slip-step pattern forms at the surfaces. The collective motion of dislocations and the formation of complex dislocation structures have been studied with nanoindentation using slip-step analysis. Scanning electron micrographs can give only information on slip-step orientation and slip-step distribution with no quantitative information on their heights. The heights are needed to estimate the number of dislocations that have reached the surface. Therefore, new correlated microscopy techniques are needed to further understand the origins of plasticity. The advent of integrated atomic force microscopes (AFMs) in stand-alone scanning electron microscopes (SEMs) has opened new possibilities due to quantitative 3D information on complex surface morphologies. This does not only reduce experimental times but also expands the information for both techniques in a comprehensive manner due to the quasi-simultaneous correlation within one setup. In material science, the investigation of slip steps structures and the correlation to the active dislocation sources during nanoindentation of metallic single crystals is necessary to examine size effects related to deformation. The collective density of dislocations and the formation of spatially complex dislocation structures have been studied with GETec's combined AFM and SEM (AFSEM TM). This system uses a highly flexible, in-situ tip scanning system with a wide range of scan-head modules. While the SEM acts as the high resolution navigation tool and can provide structural information via electron backscatter diffraction analyses, the integrated AFM solution provides laterally resolved, quantitative step height information with nm and even sub-nm resolution in XY and Z, respectively. This is an enormous advantage as it is very complicated to distinguish between real surface features on the lower nanoscale and coverage layers via SEM due to the electron beam penetration depth. By that, both techniques are complemented by each other in a straightforward manner providing comprehensive insights practically impossible or extremely complicated via individual technique or two separate instruments, respectively. Nanoindentation is a common measurement technique used to measure mechanical properties and plastic deformation events. However, difficulties arise when investigating size effects due to the size of the resulting indent imprints. With the combined imaging techniques nano- and micro-sized indents were located with the SEM while the AFM was employed to resolve the actual depth and resulting deformation with nanometer resolution (Figure 1). The AFM-SEM technique will be demonstrated through the evolution of slip steps emanating around nanoindentation imprints in a single crystal brass (Figure 2). It will be shown how the load-displacement curve corresponds to the resulting deformation Microsc. Microanal. 21 (Suppl 3), 2015 2344 [1-2], what slip planes are activated [3], and the number of dislocations will be estimated using the heights of slip steps and pop-in events [4]. Figure 1. SEM micrograph demonstrating how the AFM cantilever approaching a nanoindent. Figure 2. (a) Load-displacement curve, (b) AFM height image, (c) and height profile of a 500 �N indent. (d) Load-displacement curve, (e) AFM height image, and (f) height profile of slip steps that formed with a 10 mN indent. References: [1] M.J. Cordill, N.R. Moody, W.W. Gerberich, Materials Research Society Symposium Proceedings 976 (2006) p. 1. [2] M.J. Cordill, N.R. Moody, W.W. Gerberich, International Journal of Plasticity 25 (2009) p. 281. [3]W.F. Oele, J.W.J. Kerssemakers, J.Th.M. De Hosson, Review of Scientific Instruments 68 (1997) p. 4492. [4] D.F. Bahr, D.E. Kramer, W.W. Gerberich, Acta Materialia 46 (1998) p. 3605.");sQ1[1171]=new Array("../7337/2345.pdf","Deep Data Analysis of Atomic Level Structure-Property Relationship in an Iron Superconductor Fe1.05Te0.75Se0.25","","2345 doi:10.1017/S1431927615012507 Paper No. 1171 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Deep Data Analysis of Atomic Level Structure-Property Relationship in an Iron Superconductor Fe1.05Te0.75Se0.25 Anthony Giacomo Gianfrancesco,1,2,3 Alex Belianinov,1,2 Stephen Jesse,1,2 Sergei V. Kalinin1,2 1 Institute for Functional Imaging of Materials, Oak Ridge National Laboratory, Oak Ridge, TN 37831 The Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, TN 37831 3 UT/ORNL Bredesen Center, University of Tennessee, Knoxville, TN, 37916 2 Understanding the underlying physics of superconducting materials can be greatly facilitated by establishing a relationship between atomic structure and electronic properties at the nanoscale. We obtain the structural data through high-resolution scanning tunneling microscopy (STM), whereas scanning tunneling spectroscopy (STS) mapping of the same area provides information on the local electronic properties such as band structure and, in our case, the superconductive gap. We explore the electronic interactions via the multivariate statistical methods such as Principle Component Analysis (PCA) [1,2] and Bayesian Statistics [3,4] to analyze the data structure and reveal the correlations between structure and functionality. This approach is applied to chemically phase separated Fe1.05Te0.75Se0.25 superconductor, to identify effects of local phase separation, structural defects, and magnetic impurities on superconductive behavior. A low temperature STM was used to acquire atomically resolved images and perform STS on the same region. We then identify all the atoms in the 25x25 nm area shown in Fig. 1 and correlate topographic features with the spectral response by transforming two data sets to lie on identical imaging axis and then identify atoms with a multitude of structural and electronic variables through the local and spectral analysis. Atoms shown in Fig. 2(a) are clustered by their imaged height and spectra associated with those points are averaged together and plotted in Fig.2(b) using the same color code. We test our hypothesis of correlating height to band gap size Fig.2(c), where the band gap was found using similar fitting procedures as in [5]. The procedure outlined in the previous paragraph is also performed for the other types of structural data. The result is two separate arrays of structural and spectral variables on identical imaging axis. Figure 2(d) shows the correlation coefficient for each of the 25 spectral variables and 25 structural variables. For example, pixel (1,1) is the correlation coefficient of atom height to atom gap size and pixel (7,1) corresponds to the first principle component of the spectral response of the atom to the height of the identified atom. In this approach, the statistically significant atomic configurations are established and further used as an input into first principle modeling. Thus, the determined electronic structure is then compared to the local tunneling spectra. References [1] A. Belianinov, P. Ganesh, W. Lin, B. C. Sales, A. S. Sefat, S. Jesse, M. Pan and S. V. Kalinin, APL Mat. 2, 120701 (2014); http://dx.doi.org/10.1063/1.4902996 [2] M. Bosman, M. Watanabe, D. T. L. Alexander, and V. J. Keast, Ultramicroscopy 106(11-12), 1024 (2006); [3] E. Strelcov, A Belianinov, Y.H, Hsieh, S. Jesse, A. P. Baddorf, Y.H. Chu, and S. V. Kalinin, ACS Nano, 8 (2014), pp. 6449�6457. [4] N. Dobigeon, N. Brun, Ultramicroscopy, 120 (2012), pp. 25�34. [5] Lin W, Li Q, Sales B, Jesse S, Sefat A, Kalinin S, Pan M. Direct Probe of Interplay between Local Structure and Superconductivity in FeTe0.55Se0.45. ACS Nano 7 (3) pp. 2634-2641; (2013) Microsc. Microanal. 21 (Suppl 3), 2015 2346 Figure 1 Topographic scan of FeTeSe surface cleaved at Low Temperature. Scale bar = 5 nm Figure 2 a) Atomic Resolved Image with atoms located. b) Mean spectra associated with the color coded atoms. c) Gap Fit for those spectra. Colors correspond to spectra and atoms. d) Correlation coefficients of different structural and spectral information associated with each atom.");sQ1[1172]=new Array("../7337/2347.pdf","Nanoscale Photocurrent Microscopy for Thin Film Solar Cells Using Focused Electron Beam and Near-Field Optical Excitations","","2347 doi:10.1017/S1431927615012519 Paper No. 1172 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Nanoscale Photocurrent Microscopy for Thin Film Solar Cells Using Focused Electron Beam and Near-Field Optical Excitations Heayoung P. Yoon1,2, Yohan Yoon1,2, Paul M. Haney1, Sangmin An1,2, Joshua Schumacher1, Kerry Siebein1, Alline Myers1, and Nikolai B. Zhitenev1 1 2 National Institute of Standards and Technology, Gaithersburg, MD, 20889, USA, Maryland NanoCenter, University of Maryland, College Park, MD 20742, USA Characterization techniques based on scanning probe microscopy are increasingly used for investigating microstructure, compositions, and optoelectronic properties of photovoltaic devices. In photocurrent microscopy, excess carriers are generated by an injected beam of photons, which are collected by a Schottky contact or a p-n junction. The value of measured photocurrent corresponds to the local efficiency of carrier collection, which is determined by the local built-in potential and applied electric field, as well as the carrier recombination rate. This direct imaging technique has been frequently used for characterizing bulk properties of semiconductor devices (e.g., defects, diffusion lengths). Thin film solar cells (e.g., CdTe, Cu(InxGa1-x)Se2) are comprised of grains with sizes on the order of 1 m (Figure 1a). Previous work suggested that their inhomogeneous properties significantly impact the overall photovoltaic performance [1]. However, photocarrier dynamics in the microstructures (e.g., grain bulk, grain boundary, metallurgical junction) are currently not well understood. In this work, we measure local quantum efficiency of individual microstructures using near-field scanning optical microscopy (NSOM). Furthermore, we perform electron-beam induced current (EBIC) measurements that provide higher spatial resolution (< 20 nm). EBIC allows systematic controls of the carrier generation and of the size of interaction volume (tens of nm to a few um), well-suited for quantitative study. An Ohmic metal contact was formed to p-CdTe and n-CdS/SnO2 layers [2]. In order to measure the local response throughout the entire p-n junction, cross-sections of the device were prepared using focused ion beam (FIB) processes. For NSOM, we used a tapered optical fiber probe (200 nm diameter; Figure 1b) mounted on a tuning fork. Figures 1 (c) and (d) show the photocurrent maps at illumination wavelengths of 405 nm and 630 nm, respectively. The spatial resolution decreases at the longer wavelengths, as expected, due to the increase of the absorption depth. The bright contrast at the grain boundaries indicates high carrier collection efficiency, as proposed in previous works [2, 3]. We fit the line-scan collected near a center of a single grain under 635 nm illumination to estimate the minority carrier diffusion length (Ln) of 0.7 �m (Figure 1b) Figure 2 shows a series of cross-sectional EBIC images obtained at different acceleration voltages. The strong bright contrast at grain boundaries reflects the high carrier collections as seen in NSOM (Figure 1 c). Qualitatively, the 5 keV and 10 keV EBIC images are similar to that of 405 nm and 603 nm NSOM maps, respectively. This correspondence indicates that the calculated photon absorption and electron penetration depths are similar in magnitude. To extract material parameters, we performed least-squares fitting of the individual EBIC line scans using an analytical model that deconvolutes the collection efficiency and the generation profile [4]. Figure 2 (d) plots EBIC line scans overlaid with the model fit for a large, single grain. The extracted minority carrier diffusion lengths at different acceleration voltages were in the range of 0.6 �m to 0.8 �m, showing good agreement with the estimate from NSOM Microsc. Microanal. 21 (Suppl 3), 2015 2348 (Figure 2 b). Figure 2 also shows that the maximum collection efficiency is below 1. We attribute this reduced efficiency to strong surface recombination. Development of an analytical model that accounts for surface recombination in the depletion region and screening of built-in fields from carrier accumulation is in progress, and it will provide more information on the local properties (e.g., depletion width, carrier mobility and lifetime). In summary, we present local photocurrent microscopies of a cross-sectional CdTe solar cell using excitations by near-field optical illumination and by focused electron beam irradiation. The spatially and spectrally resolved efficiency maps confirm a higher carrier collection at grain boundaries. We estimate the minority carrier diffusion lengths of individual grains measured away from grain boundaries from NSOM and EBIC data and demonstrate an excellent agreement between the techniques. [1] S. Kumar and K. Rao, Energy Environ. Sci., 7 (2014), p.45 [2] H. P. Yoon et al., Solar Energy Mater. Solar Cells 117 (2013) [3] J. D. Poplawsky et al., Adv. Energy Mater., 1400454 (2014), p. 499. [4] P. M. Haney et al., arXiv:1410.4435v1, 2014. (a) Current (A) 10 -1 (b) 10 -2 200 nm 0.0 0.5 1.0 1.5 2.0 2.5 Distance from the p-n Junction (m) (c) (d) 1�m Figure 1. (a) Schematic of photocurrent microscopy. The inset shows a topography of a CdTe solar cell. (b) An NSOM line-scan profile at 635 nm illumination (white line in d). NSOM image at 405 nm (c) and 635 nm (d). The estimated photon absorption depth is 100 nm and 400 nm, respectively. (a) (c) 500 nm (b) Figure 2. Cross-sectional EBIC images at 3 keV (a) and 10 keV (b). The calculated penetration depth in CdTe is 40 nm and 310 nm respectively. (c) Measured EBIC linescans (solid lines) obtained for a large, single grain, overlaid with the model fit (dot lines)");sQ1[1173]=new Array("../7337/2349.pdf","Faster Time-Resolved Electrostatic Force Microscopy","","2349 doi:10.1017/S1431927615012520 Paper No. 1173 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Faster Time-Resolved Electrostatic Force Microscopy Durmus U. Karatay1, 2, Jeffrey S. Harrison2, Micah Glaz2, Phillip A. Cox2 and David S. Ginger1, 2 1. 2. Department of Physics, University of Washington, Seattle, WA 98105 Department of Chemistry, University of Washington, Seattle, WA 98105 Electrostatic force microscopy (EFM) is a widely used technique to study systems with nanoscale spatial resolution in various fields of science. Different EFM methods have been utilized to study phenomena such as mapping local electrical properties of graphene and measuring dielectric constants [1,2]. However, classic EFM methods usually do not capture time-domain information or, when they do, integrate over long times � on the order of minutes to hours. Our group has demonstrated that it is possible to perform time-resolved EFM (trEFM) to capture dynamic information, first on timescales as fast as 100 �s [3], and more recently as fast as 100 ns [4], and we have shown that these methods can provide important information about electronic processes such as photocurrent generation in devices with and without nanostructures. However, while offering impressive time-resolution, these methods are often slow in image acquisition. Here, we report the recent advances and findings regarding feedback-free trEFM. The motion of most AFM cantilevers can be modeled by damped driven harmonic oscillator equation: 2 0 () + 2 + 0 ()2 = cos() + , 2 where z is displacement of the tip, F0 is the driving force, m is the effective cantilever mass, 0 is the resonance frequency of the cantilever and Fe is an external force, which can be modeled as an exponential decay: ( 0) ()[1 - e-/ ]. Using signal processing methods we can extract the change in the instantaneous frequency of the cantilever which is related to the characteristic time of an exponentially decaying excitation. However, the phase of the cantilever at trigger time affects the cantilever response immensely. In order to quantitatively study the effects of phase on the data extraction, we have done experiments with a gold standard and voltage pulses ranging from 10 ns to 1000 ns. The resulting instantaneous frequencies due to triggering at different phases at = 100 ns can be seen in Figure 1.a. It is clear that some phases, namely 0� and 270�, cause non-ideal responses. Triggering at 180� gives the best response curves. In Figure 1.b, we show the time resolution of our equipment for a 300 kHz cantilever. We are able to differentiate signals with = 100 ns, which puts our time-resolution at 100 ns. We also have developed a real-time imaging system for feedback-free trEFM, where we can image the topography and the charging rate of the material at the same time. Currently, for a 128x128 image, it takes ~1 hour. In Figure 2, we show an image of an organic photovoltaic film composed of a polymer MDMO-PPV Microsc. Microanal. 21 (Suppl 3), 2015 2350 mixed with fullerene-derivative PCBM, taken using feedback-free trEFM. It can be clearly seen that the charging rate is dependent on local film morphology. Figure 1. (a) Deviation of cantilever's instantaneous frequency from the drive frequency for an exponential voltage pulse with = 100 ns, (b) Resolution of the FF-trEFM for different values of at trigger phase of 180�. Both results are for a 300 kHz cantilever. Figure 2. (a) Topography and (b) charging rate measurements for an MDMO-PPV:PCBM film. References: [1] T. Burnett et al., Nano Lett. 11 (2011) 2324. [2] E. Castellano-Hern�ndez et al., J. Phys. Condens. Matter 24 (2012) 155303. [3] D.C. Coffey, D.S. Ginger, Nat. Mater. 5 (2006) 735. [4] R. Giridharagopal et al., Nano Lett. 12 (2012) 893. [5]The authors acknowledge funding from NSF MRI DMR-1337173.");sQ1[1174]=new Array("../7337/2351.pdf","Correlative nanoscopy: super resolved fluorescence and atomic force microscopy towards nanoscale manipulation and multimodal investigations.","","2351 doi:10.1017/S1431927615012532 Paper No. 1174 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Correlative nanoscopy: super resolved fluorescence and atomic force microscopy towards nanoscale manipulation and multimodal investigations. Alberto Diaspro1,2, Jenu Chacko1,3 , Francesca Cella Zanacchi1,2, Reiner Oropesa1,2, Silvia Dante1, Claudio Canale1 1. 2. Istituto Italiano di Tecnologia, Nanoscopy and Nikon Imaging Center, Genova, Italy Department of Physics, University of Genoa, Genova, Italy. 3. Biomedical Engineering Department, University of California, Irvine, United States. Super resolved fluorescence microscopy combined with Atomic Force Microscopy allows to access different data sets and functionalities of investigated samples [1]. We named this hybrid approach coupling not specific force probing and fluorescence biochemical targeting - correlative nanoscopy. It opens opens new windows for approaching original questions to study behaviour of biological and materials science samples. Here we mainly report about the positive synergy given by merging single molecule localization and targeted read-out super resolution methods with scanning force microscopy in a unique blend that allows nanoscale manipulation and multimodal investigations. More specifically, STED AFM and STORM AFM have proven to be very useful in its unique way of identifying species from an AFM image. So far, the coupling between AFM and super resolved fluorescence optical microscopywas the first attempt to provide a chemical recognition to AFM by developing an integrated setup [2]. In the past, the main constraint of such an integration was given by low optical lateral resolution of the optical microscope, limited by diffraction at 250 nm. Hence the two instruments worked at completely different spatial scales and molecular specific targeting allowed by fluorescence tagging could not be exploited at the nanoscale. The advent of optical approaches with unlimited spatial resolution offered a unique challenging opportunity for a correlative nanoscopy approach [3]. Now, AFM is a widely used technique for the reconstruction of biomaterials topography with a nanometric resolution. However, despite its high resolution, the nonspecific nature of the imaging provided by AFM does not allow to answer the biological queries focused on specific molecular targets within cellular and/or cytoskeletal compartments. Nowadays, fluorescence based super resolution techniques allow achieving precise localization of selected molecular species with a resolution far below the diffraction limit. Within this scenario, one of the most advanced implementation of AFM is shown by the advantages of correlative approaches STED/AFM [2,4] and STORM/AFM [4,5], demonstrating a better comprehension of biological and biochemical questions and a strong integration of fluorescence imaging with the label free nature of the morphological information provided by AFM. We also show how the 3D imaging capabilities and multicolor STORM represents a powerful tool for a better overlay of the AFM and fluorescence information [4]. Moreover, the tip probing AFM approach opens a new window towards the possibility of mechanical nanomanipulation of cellular structures. Infact, the combined approach of AFM with the most advanced superresolution methods provides a golden opportunity for simultaneous nanomanipulation and imaging Microsc. Microanal. 21 (Suppl 3), 2015 2352 (with tens of nanometer resolution) of effects produced on specific molecule of interest in the cytoskeletal compartments of the cell [4, 6]. These techniques, based on correlative approaches, will smooth the way to a new generation of experiments able to couple topological information, local stiffness measurements and specific fluorescence imaging close to the molecular level. Figure 1. Schematic view of the STED-AFM coupling for correlative nanoscopy. Image credit J.Varghese Chacko. References: [1] C Smith, Nature 492 (2012) p. 293. [2] B. Harke, J. V. Chacko, H. Haschke, C. Canale, A.Diaspro, Optical Nanoscopy 1 (2012) p.3. [3] A. Diaspro, Il Nuovo Saggiatore 30 (2014) p.45. [4] J. Varghese Chacko, F.Cella Zanacchi, and A. Diaspro. Cytoskeleton (Hoboken) 70 (2013) p.729 [5] A. Monserrate, S. Casado, and C. Flors, ChemPhysChem. 15 (2014) p. 647. [6] J. Varghese Chacko, C. Canale, B. Harke, A. Diaspro, PLoS ONE 8 (2013) p.e66608.");sQ1[1175]=new Array("../7337/2353.pdf","Nanomineralogy of Meteorites by Advanced Electron Microscopy: Discovering New Minerals and New Materials from the Early Solar System","","2353 doi:10.1017/S1431927615012544 Paper No. 1175 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Nanomineralogy of Meteorites by Advanced Electron Microscopy: Discovering New Minerals and New Materials from the Early Solar System Chi Ma Division of Geological and Planetary Sciences, California Institute of Technology, Pasadena, CA 91125, USA; Email: chi@gps.caltech.edu Nanomineralogy is the study of Earth and planetary materials at nanoscales, focused on characterizing nanofeatures (like inclusions, exsolution, zonation, coatings, pores) in minerals and rocks, and revealing nanominerals and nanoparticles [1,2]. With advanced high-resolution analytical scanning electron microscope (FE-SEM with EDS and EBSD) and electron probe microanalyzer (EPMA), we are now capable to characterize solid materials down to nanoscales easier and faster. Nanofeatures have been discovered in many common minerals and rocks, providing insights into genesis and physical properties [3]. Nanominerals and nanoparticles have been revealed. New minerals and new materials with important geological significance have been discovered at micron to nanoscales. Nanomineralogy plays a unique role in Earth and planetary sciences research. During our ongoing nanomineralogy investigation of meteorites since 2007, more than 25 new minerals have been identified. Each of the new extraterrestrial minerals reveals distinctive forming environments with intensive variables (e.g., composition, temperature, pressure, fO2). The findings provide new insights into nebula, parent-body, or shock processes in the early solar system. 15 new minerals are from the Allende meteorite (Table 1), including eight refractory minerals: allendeite, hexamolybdenum, tistarite, davisite, grossmanite, panguite, kangite and paqueite. Refratory minerals are the first solid materials formed in the solar nebula. To date, ~50 refractory minerals plus about 15 presolar minerals mark the very beginning of the solar mineral evolution. There are now more than 4900 mineral species identified on Earth. Minerals in the solar system have evolved as a consequence of physical, chemical and biological processes over the past 4.568 billion years [4]. Presented here are a few new refractory, alteration, and high-pressure minerals formed in the early solar system, demonstrating how nanomineralogy works with an integrated SEM-EDSEBSD-EPMA approach. FE-SEM has been used for high-resolution electron imaging with EDS for fast elemental analysis and EBSD for crystal structure analysis. EBSD has been employed successfully for structure determination of new minerals since 2006 [5]. EPMA, pioneered by Castaing, is one of the most important analytical tools in Earth sciences [6], and it is responsible for discovery and/or quantitative elemental analysis of minerals since 1960s. Low-voltage EPMA has been carried out for quantitative analysis of our fine-grained new minerals at a submicrometer spatial resolution. Five new minerals (panguite, kangite, nuwaite, hutcheonite and adrianite) in Allende are also new materials. They are brand-new to science and can be exploited for finding engineering materials. Nature shows many advantages over technology on materials synthesis. References Microsc. Microanal. 21 (Suppl 3), 2015 2354 [1] C Ma, Eos Trans. AGU, 89 (2008), abs MR12A-01. [2] C Ma, American Mineralogist, 95 (2010) 188-191. [3] C Ma and GR Rossman, Microscopy and Microanalysis, 13 (Suppl. 2) (2007) 164-165. [4] RM Hazen et al, American Mineralogist, 93 (2008) 1693-1720. [5] C Ma and GR Rossman, American Mineralogist, 93 (2008) 154-157. [6] K Keil, Microscopy and Microanalysis, 5 (Suppl. 2) (1999) 572-573. [7] C Ma et al, American Mineralogist, 99 (2014) 654-666. [8] C Ma et al, American Mineralogist, 99 (2014) 198-205. [9] C Ma and GR Rossman, American Mineralogist, 94 (2009) 841-844. [10] C Ma and GR Rossman, American Mineralogist, 94 (2009) 845-848. [11] C Ma and GR Rossman, American Mineralogist, 94 (2009) 1491-1494. [12] C Ma et al, American Mineralogist, 97 (2012) 1219-1225. [13] C Ma et al, American Mineralogist, 98 (2013) 870-878. [14] C Ma, Mineralogical Magazine, 77 (2013) 10. [15] C Ma, Mineralogical Magazine, 77 (2013) 2704. [16] C Ma and AN Krot, American Mineralogist, 99 (2014) 667-670. [17] C Ma, Mineralogical Magazine, 77 (2013) 3002. [18] C Ma and AN Krot, Mineralogical Magazine, 78 (2014) 801. [19] SEM, EDS, EBSD, and EPMA analyses were carried out at the Caltech GPS Division Analytical Facility, which is supported, in part, by NSF grants EAR-0318518 and DMR-0080065. I would like to thank my collaborators O Tschauner (synchrotron), JR Beckett (meteorite), GR Rossman (Raman), AN Krot (meteorite), HC Connolly Jr (meteorite), AR Kampf (XRD) and TJ Zega (TEM), for joining me on this exciting journey of discoveries. Table 1. Fifteen new minerals discovered in the Allende meteorite. IMA No. IMA 2007-027 IMA 2007-029 IMA 2007-033 IMA 2008-016 IMA 2008-030 IMA 2008-042 IMA 2009-027 IMA 2010-057 IMA 2011-092 IMA 2012-079 IMA 2013-018 IMA 2013-029 IMA 2013-053 IMA 2013-054 IMA 2014-028 Mineral Name allendeite hexamolybdenum monipite tistarite davisite grossmanite hibonite-(Fe) panguite kangite majindeite nuwaite hutcheonite paqueite burnettite adrianite Formula Sc4Zr3O12 (Mo,Ru,Fe) MoNiP Ti2O3 CaScAlSiO6 CaTi3+AlSiO6 (Fe,Mg)Al12O19 (Ti4+,Al,Sc,Mg,Zr,Ca,)2O3 (Sc,Ti4+,Al,Zr,Mg,Ca,)2O3 Mg2Mo3O8 Ni6GeS2 Ca3Ti2(SiAl2)O12 Ca3TiSi2(Al2Ti)O14 CaVAlSiO6 Ca12(Mg5Si9)O32Cl6 Reference [7] [7] [8] [9] [10] [11] [2] [12] [13] [14] [15] [16] [17] [17] [18]");sQ1[1176]=new Array("../7337/2355.pdf","Soft X-ray Emission Spectroscopy on Chemical States of 3D-Transition Metal Elements with SEM","","2355 doi:10.1017/S1431927615012556 Paper No. 1176 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Soft X-ray Emission Spectroscopy on Chemical States of 3D-Transition Metal Elements with SEM M. Terauchi1, H. Takahashi2, T. Murano3, T. Imazono4, M. Koike4, T. Nagano5 and M. Koeda5 1. 2. IMRAM, Tohoku University, 2-1-1 Katahira, Aoba-ku, Sendai 980-8577, Japan Global business promotion division, JEOL Ltd., 2-1-1 Otemachi, Chiyoda-ku, Tokyo 100-0004, Japan 3. SA Technical Development Division, JEOL Ltd., 3-1-2 Musashino, Akishima, Tokyo 196-8558, Japan 4. Quantum Beam Science Center, Japan Atomic Energy Agency, 8-1-7 Umemidai, Kizu 619-0215, Japa. 5. Device Department, Shimadzu Corp., 1 Nishinokyo-Kuwabaracho, Nakagyo-ku, Kyoto 604-8511, Japan Soft X-ray emission spectrometer for 50-4000 eV had been attached to a SEM [1], and has been applied to 3dtransiton metals and compounds including battery cathode materials. X-rays originate from electronic transitions from valence bands (VB, bonding electron states) to inner-shell electron levels inform us of energy states of bonding electrons. Thus, L-emissions due to transitions of 3d valence electrons to 2p core-hole are very important to assign the chemical states of 3d-transiton metal elements. Furthermore, charge state is also important to discuss the physical property because of the amount of 3d electrons closely related to magnetism and conductivity of transition metal compounds. L, emissions attributed to 3d5/2,3/2 2p3/2 and 3d3/2 2p1/2 transitions, respectively, are suitable to probe valence states (bonding states). On the other hand, Ll, emissions are due to 3s1/2 2p3/2 and 3s1/2 2p1/2 transitions, respectively. The 3s and 2p electron levels of 3d-elements are core states. Thus, Ll, lines are appropriate to observe a change of binding energy of core electrons (chemical shift). TiO2 is an important photocatalytic material. Figure 1 shows L, emission spectra obtained from bulk TiO2 specimens having various morphology, i.e., rutile, anatase, and brookite, by use of the SXES-SEM instrument with a grating for 300-2200 eV developed by us [2]. A spectrum of metal-Ti is also shown for comparison. In a simple ionic model, TiO2 is considered as Ti4+O2-2. Ti4+ ion has no 3d electron. However, L, emissions are observed for each TiO2 material, reflecting covalent bonding between Ti and O atoms. Characteristic profile structures indicated by vertical lines vary for different TiO2 materials due to particular band structures, which are attributed to different crystal structures. When comparing the peak positions of L, spectra, those of TiO2 are located at the lower energy side than those of metal-Ti. They are opposite shifts for Ti4+ ion. This means that L, intensity profile influenced by the density of states of VB is not appropriate to detect chemical shift. Figure 2 shows Ll, emission spectra of bulk TiO2, rutile, anatase, and brookite along with metal-Ti. All data points of the respective spectra were obtained at the same time with the corresponding spectra shown in Figure 1, because the spectrometer with the present grating allows us a parallel detection for 300-2200 eV. As Ll, emissions are caused by transitions between inner-shell levels, intensity profiles of Ll show symmetric distribution concerning the peak top positons, except the presence of weak L intensity on the high energy side. It is clearly seen that Ll peaks of TiO2 materials are positioned at higher energy side than that of metal-Ti. These shifts of TiO2 materials are reasonable for cation chemical shift. Furthermore, shift amounts of rutiel and anatase are almost the same, but that of brookite is suggesting a larger positive valency than other two TiO2. Microsc. Microanal. 21 (Suppl 3), 2015 2356 As shown above, L-emission of 3d-transiton element is useful not only for probing valence electron state of bonding electrons by L, but also for charge state (valency) analysis by Ll,. SXES-SEM analysis of battery cathode materials which show particular characters depending on different charge states will be also presented. References: [1] M Terauchi, et al., Microscopy and Microanalysis 20 (2014), 629. [2] T Imazono, et al., Applied Optics 51 (2012), 2351. Figure 1. Ti-L, emission spectra obtained from bulk TiO2 specimens of rutile, and anatase, brookite as well as metal-Ti by using our SXES-SEM with a grating for 300-2200 eV [2]. Figure 2. Ti-Ll, emission spectra of bulk TiO2 of rutile, anatase, and brookite, as well as metal-Ti.");sQ1[1177]=new Array("../7337/2357.pdf","Direct and Indirect Observation of Lithium in a Scanning Electron Microscope; Not Only on Pure Li!","","2357 doi:10.1017/S1431927615012568 Paper No. 1177 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Direct and Indirect Observation of Lithium in a Scanning Electron Microscope; Not Only on Pure Li! P. Hovington1, M. Lagac�1, E. Principe2, S. Burgess3, A. Guerfi1, H. Demers4, R. Gauvin4, K. Zaghib1, 1. 2. Hydro-Quebec Reaserch Institute, Varennes, Quebec, Canada Tescan - Orsay Holding, USA 3. Oxford Instruments NanoAnalysis, High Wycombe UK. 4. Department of Mining and Materials Engineering, McGill University, Montreal, Quebec, Canada. Battery is one of the most used technologies in everyday life (cellular, electric vehicle or hybrid cars). Improvement in the specific capacity (energy by weight or volume) and charging rate has a potential to even significantly improve their used. Hence, the research in battery materials is very important, well founded and will have a direct economical and social impact. Most of the actual battery technology is based on the displacement of Lithium ion (Li+) from two active materials (i.e., graphite for the anode and LiFePO4 for the cathode). It is thus essential to determine the distribution and the amount of Li with a good spatial resolution (<< 1 �m) In terms of microstructural characterization Li is very difficult to analysed using conventional detector because it is a very light element and emits low energy x-rays (52 eV). Optical Emission spectrometry and XPS can easily detect Li but without any no good lateral resolution. In counterpart, Li has a relatively high sputtered yield and low backscattered coefficient. Mass spectra have been used in dedicated secondary ion mass microscope (SIMS), either static or Time of Flight (TOF) but with, again, with a limited lateral resolution. This work will present, for the first time, a scanning electron microscope coupled with a Ga ion beam that can easily detect Li using two different detectors; TOF-SIMS from collaborative work of Tescan and TOFWERK and Windowless energy dispersive spectrometer specially design for low noise and low energy works from Oxford Instrument NanoAnalysis. We present in Figure 1 an ion-mass spectrum together with ion induced secondary electron image and Li-7 chemical mapping of LiFePO4 particles. In the mass spectra, taken with less than 30 seconds, we clearly see a very high Li peaks together with a low Fe peaks. We note the very sharps peaks of Li at the mass 7 (charge =1) with no convolution which is ideal for chemical identification and localisation. The Li mapping images with a field of view of 1.5 �m clearly revealed the LiFePO4 particle with a spatial resolution much lower than 1 �m. One of the challenges with the TOF-SIMS will be to obtained quantitative results, especially on non planar surface. TOF-SIMS analysis on pre-flattened `thick lamella' is now underway to minimize this effect. Optimization of this instrument and new results will be presented. In addition to Li detection using the TOF-SIMS, we present, in Figure 2, Li K peaks not only from a Li standard, produced at Hydro-Quebec, but also on a LixS1-x polysulfide samples that are very important to understand the behavior of Li-S battery having a specific energy density 10x higher than standard Li-ion cathode material. To our knowledge it is the first time Li peaks is presented on a non pure Li film using EDS detector. For the Li films we clearly see a very high signal/noise Li peak and a good separation with the noise peak that is forming a small bump at the low end side of the Li peak. For the LixS1-x sample, the Li K peak can be clearly distinguished from the noise peak and from the S Ll peaks. Both peaks (Li K and S Ll) are found below 0.5 keV so analysis at 1 kV (or lower energy) at a very high spatial resolution can be made (< 30 nm) in the SEM. Li can also be indirectly observed using the variation of the BSE coefficient has a function of Li alloying in a Si anode. Figure 3 presents the variation of BSE as a function of composition for the Li-S complex. We can clearly see a decrease of 50% between Si and Li22Si5. This observation was clearly observed using in situ discharge of the Si anode particle (10x more energy than standard graphite anode) where some grains, or parts of grains, have greatly change their BSE yield with decreasing voltage. In conclusion, the used of two detectors capable of detecting Li and not only using pure Li film, is promising for Li detection with a good spatial resolution on technological bulk sample and, we think, will be able to accelerate battery development at Hydro-Quebec and also in similar research centers. Microsc. Microanal. 21 (Suppl 3), 2015 2358 Figure 1. Mass Spectrum taken from a FIB scanning on LiFePO4 particles. Also shown in insert ion induced secondary electron image and Li image (field of view 1.5 �m) a) b) Figure 2. Li spectrum taken on a pure Li laminated films (produce at Hydro-Quebec) and (b) from a polysulfide (LixSi1-x). Figure 3. Variation of the Backscattered coefficient (BSE) vs. Si in the SixLi1-x complex. Insert is showing a cross-section of the electrode at the beginning and at the end of the discharge (white phase=Si, gray phase = Li22Si5). Cycling is done in situ in the SEM.");sQ1[1178]=new Array("../7337/2359.pdf","High Spatial Resolution Quantification X-ray Microanalysis in a Field Emission Scanning Electron Microscope with an Annular Silicon Drift Detector","","2359 doi:10.1017/S143192761501257X Paper No. 1178 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 High Spatial Resolution Quantification X-ray Microanalysis in a Field Emission Scanning Electron Microscope with an Annular Silicon Drift Detector Hendrix Demers1, Nicolas Brodusch1, Patrick Woo2, and Raynald Gauvin1 1. 2. Department of Mining and Materials Engineering, McGill University, Montreal, Quebec, Canada. Hitachi High-Technologies Canada Inc., Toronto, Canada. The scanning electron microscope (SEM) was primary developed for imaging applications. With the introduction of the Si(Li) energy dispersive spectrometer (EDS), simultaneous imaging and x-ray microanalysis became possible. However, long working distance and high current were needed because the position and small solid angle of the EDS detector. SEM was initially and is still optimized for imaging applications, where the high spatial resolution is generally obtained at short working distance. This problem is still relevant today and unfortunately x-ray microanalysis is never performed in the best imaging conditions, i.e., not with the smallest probe size. With the introduction of an annular silicon drift detector (SDD) system, scanning electron microscopy is facing a revolution. This detector is inserted below the objective lens which gives a higher solid angle (up to 1.2 sr). In consequence, a lower working distance and probe current can be used. An improved spatial resolution becomes possible during x-ray microanalysis. For quantification microanalysis where the absorption of the x-ray is important, the value of the takeoff angle is important. Lower value increases the absorption in the sample. Also current correction models, suppose a fix value of takeoff angle. Because of the geometry of the annular SDD, the mean takeoff angle is 33 degree with a minimum of 24 and maximum of 50 degree at the optimum detector distance. However, preliminary results indicate that this large takeoff angle range does not affect the correction model when the absorption effect is small or moderate, i.e., using the mean takeoff angle gives accurate composition. Because of the position of the annular detector, Mylar windows are used to prevent the backscattered electrons (BSEs) to damage the SDD segments. The shape of the background is strongly affected by the window absorption at low x-ray energy. For accurate quantitative analysis, the calculation of peak net intensity depends on the background subtraction method used. However, for energy greater than 1 keV, the current background subtraction method seem enough with this annular SDD. An example of x-ray elemental maps of an Al-Mg diffusion couple acquired with annular silicon drift detector (SDD) at low accelerating voltage is shown in Figure 1. The accuracy of the quantification with an annular SDD was evaluated with the standardless [1] and f-ratio method [2,3]. The effect of this detector geometry and position on the correction model is currently studied. The annular SDD with is larger solid angle will clearly revolution the quantification microanalysis by moving from point analysis to quantitative micrograph with simultaneous electron imaging. Also, since the count rate can be as high as 1,500 kcps with our system, which lower significantly the detection limit of elements as well the minimum feature sizes of different phases that can be distinguished as shown in Figure 2. References: [1] Bruker ESPRIT software version 1.9. [2] P. Horny, E. Lifshin, H. Campbell and R. Gauvin, Microscopy and Microanalysis, 16 (2010), p. 821830. [3] R. Gauvin, Microscopy and Microanalysis, 18 (2012), p. 915-940. Microsc. Microanal. 21 (Suppl 3), 2015 2360 Figure 1. Quantitative x-ray maps of an Al-Mg diffusion couple sample at 5 kV. A and B shown the net intensity (with background subtraction) of Al and Mg, respectively. C Phase map obtained from the standardless quantification results. P3 is pure Mg, P4 Al12Mg17 phase, P1 Al3Mg2 phase, and P2 pure Al. The other identified components are from cracks filled with mounting resin. D Mg weight fraction pixel distribution calculated with standardless [1] and f-ratio [2,3] quantification methods. The phase diagram composition range of the two phases are also shown. Figure 2. High spatial resolution x-ray elemental map of an Al-Li 2099 alloy at 20 kV in SEM/STEM mode. The T1 (Al2CuLi) palettes observed in the bright-field micrograph (TE) are clearly visible in the Cu map.");sQ1[1179]=new Array("../7337/2361.pdf","Mapping Electronic Orbitals in Real Space","","2361 doi:10.1017/S1431927615012581 Paper No. 1179 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Mapping Electronic Orbitals in Real Space Stefan L�ffler1,2,3 1 2 Department of Materials Science and Engineering, McMaster University, Hamilton, Canada. University Service Centre for Transmission Electron Microscopy, Vienna University of Technology, Vienna, Austria. 3 Institute of Solid State Physics, Vienna University of Technology, Vienna, Austria. Electronic states are the key to most material properties including mechanical stability, chemical bonding, electrical conductivity, magnetism and optical properties. Because of their paramount importance across many fields, their direct mapping has attracted much attention in recent years. Using elastic electron scattering, total charge densities could be measured on the nanometer scale [1,2]. On the surface, molecular orbitals could be resolved using STM [3]. But the direct mapping of individual electronic states in real space in the bulk phase � the key to materials design � has been elusive so far. In this work, an overview is given over recent developments of mapping individual, specific electronic orbitals using transmission electron microscopy (TEM) and electron energy loss spectrometry (EELS). In TEM, it is possible to focus the probe beam to sub-�ngstr�m spatial resolution using a state-of-the-art aberration corrected instrument. By sending a focused electron probe through the sample, energy and momentum can be exchanged between the probe electrons and the sample electrons, giving rise to a unique energy and momentum distribution of the probe beam after the sample which can be measured. By selecting an appropriate energy range using a slit aperture, specific transitions to individual states in the sample can be selected [4], and by scanning the probe beam over the sample, information about the probability distribution of those states in real space can be deduced. Fig. 1 shows an example of an orbital map of the eg contributions to Rutile. To interpret the experimental data, a thorough understanding of the image formation mechanism and extensive computer simulations are indispensable. In the particular case of combining TEM and EELS, both elastic and inelastic scattering effects of the probe beam inside the specimen need to be considered [5]. The elastic scattering effects are usually taken into account using either the multislice approach or the Bloch wave approach. For modelling the inelastic scattering, a very versatile tool is the mixed dynamic form factor (MDFF) in the density matrix framework. Only recently, a novel approach to diagonalize the MDFF was developed [6]. It not only allows to combine realistic predictions of the electronic orbitals in the material (e.g., from density functional theory calculations) with the inelastic scattering calculation with reasonable computational effort, but also provides physical insight into the underlying scattering processes. One crucial result of this theoretical and numerical treatment is that the point-group symmetry of the atom under investigation plays a decisive role in the shape of the resulting maps. Only if the symmetry is low enough (as is the case in Rutile, but also near defects, interfaces, etc.), a bonding direction can be imaged. By combining simulations and experiments, it is thus possible to map orbital information in the bulk in real space using state-of-the-art TEMs. This will pave the way for completely new possibilities and materials design, in particular in the vicinity of defects and interfaces [7]. Microsc. Microanal. 21 (Suppl 3), 2015 2362 References: [1] JM Zuo et al, Nature 401 (1999), p. 49. [2] JC Meyer et al, Nature Materials 10 (2011), p. 209. [3] J Repp et al, Science 312 (2006), p. 1196. [4] W Hetaba et al, Micron 63 (2014), p. 15. [5] MP Prange et al, Physical Review Letters 109 (2012), p. 246101. [6] S L�ffler, V Motsch and P Schattschneider, Ultramicroscopy 131 (2013), p. 39. [7] The author acknowledges financial support by the Austrian Science Fund (FWF) under grant number I543-N20 and SFB F45 FOXSI. The experimental data in Fig. 1 was kindly provided by Matthieu Bugnet, Nicolas Gauquelin, Sorin Lazar, and Gianluigi Botton. The orbital visualization was kindly provided by Elias Assmann and Karsten Held. Fig. 1: Gaussian-filtered experimental (left) and simulated (center) map of the eg orbital contributions to the Ti L2,3 edge in 20 nm thick Rutile in [001] projection using an 80 keV probe beam. The overlays show the position of the Ti atoms (blue) and of the O atoms (orange). The ellipses (yellow) are guides for the eye only. Both images were replicated in a 3x3 rater for better visibility. The right panel shows the summed three-dimensional charge density of the eg Wannier functions around the central Ti atom.");sQ1[1180]=new Array("../7337/2363.pdf","Excitonic Calculations of ELNES: Low Energy and High Energy Spectra","","2363 doi:10.1017/S1431927615012593 Paper No. 1180 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Excitonic Calculations of ELNES: Low Energy and High Energy Spectra K. Tomita1 and T. Mizoguchi1 1 Institute of Industrial Science, University of Tokyo, Japan ELNES originates from electron transition from core orbital to conduction band, and the electron transition follows electron dipole transition. Thus, ELNES profile reflects the partial density of state of selected elements in the conduction band. By combining the aberration corrected STEM, electronic structure analysis with atomic resolution is possible. Thus, the EELS have been extensively observed for understanding the atomic and electronic structure of materials. On the other hand, theoretical calculation is indispensable to interpret the ELNES. Together with the improvements of spatial and energy resolutions of experimental ELNES, the speed and accuracy of the theoretical calculation for ELNES are also improved. Recently, ELNES calculations using one-particle, two-particle (excitonic), and multi-particle methods have been systematically performed[1], and it was demonstrated that multi-particle theory is mandatory to calculate L2,3 edge of transition metals and M4,5 edge of lanthanides[2], whereas two-particle (excitonic) theory is needed to reproduce ELNES of light elements, such as Li and Be. And, one particle calculation, which is conventional DFT-LDA/GGA method, can be applicable to ELNES except for those two cases [3]. However, importance of the excitonic effect on the high energy ELNES has not been discussed so far. In this study, we systematically performed the excitonic calculation, and the excitonic effects on both low and high energy ELNES were investigated. Furthermore, an application of the excitonic ELNES calculation to Li-ion battery was performed. The ELNES calculation was performed using first principles Full-potential Linearized Augmented Plane-Wave (FLAPW) method based on DFT-GGA theory. The excitonic effect was included by solving the equation of motion for the electron-hole two-particle Green function, which is Bethe Salpeter Equation (BSE) [4]. As reported before, BSE makes better reproduction of ELNES for Li-Kedge of LiCl[1] and Al-L23edge of AlN[2] comparing with GGA. In this study, the excitonic calculations were performed for the low energy ELNES (less than 100eV), such as Na-L23 edge, MgL23 edge and Si- L23 edge, and the high energy ELNES (higher than 500eV), such as N-K edge, O-K edge, and Mg-Kedge. Furthermore, an application of the excitonic calculation for Li-ion battery was also performed. Figure 1 shows the calculated and experimental Na-L2,3 edge of NaBr. All spectra were aligned using peak C. In the experimental spectrum, the sharp peaks A1 and A2 appear at the threshold and broad peaks B and C are followed. In the supercell calculation, although overall features can be reproduced, the position of the peak A is overestimated. On the other hand, the position of peak A becomes Microsc. Microanal. 21 (Suppl 3), 2015 2364 reproducible using the excitonic calculation. Similar trend has also reported in Li-K edge of Li halides. However, it is found that the peak split at the threshold cannot be reproduced even in the excitonic calculation. We then tried BSE calculation with spin-orbit coupling. Peak splitting is observed and also relative peak intensity and position are well reproduced using the spin-orbit excitonic calculation. On the other hand, we found that the excitonic calculation and supercell calculation give the almost identical spectrum to each other in the case of MgO O-K edge, as shown in Fig.2. We have systematically performed the excitonic calculation of low energy and high energy ELNES, and found that the excitonic effect is sometimes important even in the high energy ELNES. We are going to talk about those results in my presentation. Furthermore, some application of the excitonic calculation to the Li-ion battery will also be shown. References [1] T. Mizoguchi et al., Micron 41(2010) 695-709 [2] H. Ikeno et al., Phys. Rev. B, 83,155107 (2011) [3] W. Olovsson et al., Phys. Rev. B, 79,041102 (2009) [4] G. Onida et al., Rev. of Mod. Phys., Vol.74, Apr., 2002 [5] S. Nakai et al., Journal of the Physical Society of Japan, Vol.26, No.6, June, 1969 [6] T. Mizoguchi et al., Ultramicroscopy, 106(2006), 1120 1128");sQ1[1181]=new Array("../7337/2365.pdf","Probing Bonding Environments in Osmium-Based Double Perovskites Using Monochromated Dual Electron-Energy Loss Spectroscopy","","2365 doi:10.1017/S143192761501260X Paper No. 1181 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Probing Bonding Environments in Osmium-Based Double Perovskites Using Monochromated Dual Electron-Energy Loss Spectroscopy Bryan D. Esser1, Molly R. Ball1, Ryan C. Morrow2, Angela E. Goode3, Robert E. A. Williams1, Patrick M. Woodward2, Wolfgang E. Windl1, and David W. McComb1 1. 2. 3. Department of Materials Science and Engineering, The Ohio State University, 477 Watts Hall, 2041 College Road, Columbus, OH 43210, USA Department of Chemistry, The Ohio State University, Newman & Wolfrom Lab, 100 W 18th Ave, Columbus, OH 43210, USA Department of Materials, Imperial College London, South Kensington Campus, London SW7 2AZ, United Kingdom The ability to use the spin of the electron (��) for read/write capabilities in solid-state devices has the potential to double capacity and increase speed without altering weight. The key to "spintronics" is to find functional materials that deliver the required properties [1]. Double perovskite (DP) crystals have the generic formula, A2BB'O6, where A, B, and B' are transition metal atoms. DPs, especially halfmetallic ones, have been shown to exhibit high spin polarization and Curie temperatures above room temperature [2, 3]. To fully realize the potential of DPs as spin injectors, high quality, ordered and epitaxial thin films must be grown. This has proved challenging. The B and B' atoms occupy the same type of sites in the crystal and the properties are sensitive to the relative spatial pattern of these atoms. Our goal is to develop experimental methods that allow us to probe this arrangement with subnanometer spatial spectro-microscopy methods in order to directly probe on the atomic scale the relationship between structure, chemistry and properties in these complex oxides. We have investigated osmium-based DP (B'=Os) powders using monochromated dual electron-energy loss spectroscopy (EELS) in the scanning transmission electron microscope (STEM). The range of B site atoms studied were Cr, Fe, and Co. Powders with each B site atom were made with Sr or Ca, separately, on the A site. The B and B' atoms occupy octahedral sites in the crystal with six oxygen atoms at the apices of the octahedron. By varying the valance state of the A and B site atoms, the oxygen octahedra can tilt away from the orthogonal axes of the unit cell, which has an effect on the overlap of the bonding between the oxygen and the metal species. Ultimately, this can affect the magnetic properties of interest by altering the J-couplings within the material. STEM-EELS studies allow us to probe the unoccupied electron energy levels on the oxygen sites via the oxygen K-edge transition. By systematically varying the chemistry of the DPs, a reference set of oxygen K-edge spectra were obtained for comparison with thin film DPs grown for device applications, as seen in Figure 1. In order to fully explain the EELS, and where possible, X-ray absorption near-edge structure (XANES) data, density functional theory (DFT) was employed to calculate the density of states (DOS) around the Fermi energy. This, in turn, can help to distinguish different features within the experimental EELS data to better define the electronic configuration of each system. As an example, in Figure 2 it can be seen that the EELS and XANES data are in very good agreement for Ca2CoOsO6, with the calculated O-p states below the two spectra. The features in the calculated O-p states above the Fermi energy can be identified by looking at the hybridization between oxygen and the other elements in the full DOS calculation (not shown). In understanding the electronic states above the Fermi energy and the ways in which they manifest themselves in EELS spectra, one can track changes in those states around defects, interfaces, or in thin films. In studying the calculated electronic states along with the experimental data for a set of systematically altered materials, a much clearer understanding of the electronic and magnetic Microsc. Microanal. 21 (Suppl 3), 2015 2366 phenomena can be gained. This can help to guide future materials selection for materials of interest with applications in spintronics. [1] AJ Hauser, et al, Phys. Rev. B 85 (2012) p.161201(R) [2] KL Kobayashi, et al, Nature 395 (1998) p.677. [3] JM Serrate, De Teresa, and MR Ibarra, J. Phys: Condens. Matter 19 (2007) p.023201. This research was funded through the Center for Emergent Materials (CEM) at The Ohio State University (OSU) through a grant from the National Science Foundation (NSF). Calculations were made possible through resources made available by the Ohio Supercomputer Center (OSC). Figure 1. Sr- and Ca-based Osmate DP EELS spectra show changes in the density of states correlated with changes in valence state Figure 2. Comparison of EELS, XANES, and Calculated O-p states from DFT results for Ca2CoOsO6");sQ1[1182]=new Array("../7337/2367.pdf","Analytical Method for Thickness and Wrinkling Measurements of 2-D Zeolites","","2367 doi:10.1017/S1431927615012611 Paper No. 1182 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Analytical Method for Thickness and Wrinkling Measurements of 2-D Zeolites Prashant Kumar,1 Kumar Varoon Agrawal,1 Michael Tsapatsis,1 K. Andre Mkhoyan1 1 Department of Chemical Engineering & Materials Science, University of Minnesota, Minneapolis, MN 55455. Two-dimensional (2-D) zeolites are near single-unit-cell thick layers of silicon and oxygen atoms with precisely-sized pores of molecular dimensions [1]. It is critical to determine the thickness and structure of 2-D zeolite layers in order to understand their unique transport [2] and catalytic properties [3]. Thickness measurements of 2-D zeolites are often performed by X-ray reflectivity experiments [4] or by imaging cross-sectional samples in a transmission electron microscope (TEM) [5]. However, X-ray reflectivity measurements require fabrication of a periodic multilayer, which is not feasible for zeolite nanosheets with sub-micron lateral dimensions. On the other hand, conventional TEM (CTEM) images of cross-sections require in-depth analysis of thickness and image contrast. Due to these limitations and the electron-beam-sensitive nature of 2-D zeolites [6], a method for unambiguous determination of their thickness and structure has remained elusive. We developed an analytical method based on a set of TEM experiments, which can provide complete quantitative characterization of 2-D zeolites, including crystal structure, uniformity, thickness, and wrinkling. Since intensity modulation of a diffraction spot on tilting is a fingerprint of the thickness, and changes in the spot shape are a measure of the wrinkling, comprehensive three-dimensional (3-D) mapping of the reciprocal lattice was performed to monitor these changes. This was achieved by acquiring a tilt series of selected area electron diffraction patterns (SAED) of MFI-structure type zeolite nanosheets on a FEI Tecnai G2 F30 (S)TEM equipped with a TWIN pole piece at 300 kV accelerating voltage. A MFI-nanosheet suspension in octanol was prepared by following a method reported by Varoon et al [7]. The MFI crystal structure of as-synthesized nanosheets was confirmed by high resolution conventional TEM imaging (Figure 1a) and [010] zone axis SAED pattern (Figure 1b). Comparison of multislice simulated modulation of diffraction spot intensity with tilt to experimental data revealed that the nanosheets are 1.5 unit cells (uc) thick (Figure 1c). In order to investigate quantitatively the deviation from flatness (wrinkling) of the MFI nanosheet, we tracked changes in shape of diffraction spots upon tilting. The wrinkling of nanosheets in real space corresponds to precession of the rel-rods [8] into "cones" in reciprocal space. The diffraction spot shape changes are particularly pronounced at larger tilt angles. Quantification of these wrinkles was done by modeling two independent sine waves superposed in a- and c-directions of a flat nanosheet (Figure 1dg). Comparison with multislice simulated tilt series of SAED patterns for wrinkled nanosheet models revealed that the wrinkling is larger in the a-direction as compared to the c-direction (due to greater resistance of the MFI nanosheet to bending in c-direction). The average surface roughness of nanosheets was estimated to be 0.8 nm. References: [1] M. Choi et al, Nature 461 (2009), p. 246. [2] K. Varoon et al, AIChE. J. 59 (2013), p. 3458. Microsc. Microanal. 21 (Suppl 3), 2015 2368 [3] K. Na et al, J. Am. Chem. Soc. 132 (2010), p. 4169. [4] H. Jiang et al, J. Appl. Phys. 107 (2010), p. 103523. [5] W. Park et al, Chem. Mater. 23 (2011), p. 5131. [6] O. Ugurlu et al, Phys. Rev. B 83 (2011), p. 113408. [7] K. Varoon et al, Science 334 (2011), p. 72. [8] J. C. Meyer et al, Nature 446 (2007), p. 60. [9] This work was supported as part of the Catalysis Center for Energy Innovation, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences under Award DE-SC0001004. Figure 1. (a) Bragg-filtered high-resolution TEM image of MFI nanosheet along the [010] direction with the overlaid crystal structure. (b) SAED pattern taken along [010] zone axis. It is presented in reverse gray-scale color for better visibility. Tilt-axis is indicated with a double sided orange arrow. Diffraction spots used for thickness analysis are circled. (c) Variation in intensity with tilt angle (rel-rod map) of the (301) spot. Solid lines represent multislice simulated data and black dots represent experimental data. A good agreement with 1.5 uc thick nanosheet is observed. (d) Two sine waves in aand c-direction, the superposition of which provides the best description of the wrinkles in the MFI nanosheet. (e) Bragg filtered HR-TEM image shown in (a) overlaid with the estimated wrinkled nanosheet model. (f) a*b* and (g) b*c* projections of the (011) rel-rod constructed from experimental SAED patterns of a MFI nanosheet. Ewald sphere intersections with the rel-rod are plotted as solid blue and red lines. The cones determine the broadening of the rel-rods due to nanosheet wrinkling.");sQ1[1183]=new Array("../7337/2369.pdf","Determination of Surface Topography from Laser Ablation using EBSD","","2369 doi:10.1017/S1431927615012623 Paper No. 1183 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Determination of Surface Topography from Laser Ablation using EBSD P.G. Callahan1 , M.P. Echlin1 , T.M. Pollock1 , M. De Graef2 Materials Department, University of California at Santa Barbara, Santa Barbara, CA 93106, USA Department of Materials Science and Engineering, Carnegie Mellon University, Pittsburgh, PA 15213, USA In the last decade, serial sectioning with dual beam Focused Ion Beam-Scanning Electron Microscopes (FIB-SEM) has been used to successfully gather 3-D information from a variety of materials. The sizes of the 3-D volumes interrogated with this technique are limited by the milling rate of the Ga ion beams typically used. In order to collect larger volumes of data, a novel TriBeam system that combines ultrafast lasers with scanning electron microscopes has been developed [1]. The use of femtosecond laser ablation for machining allows for rapid machining of volumes of material approaching the mm3 scale. Machining with lasers results in a periodic surface roughness called the Laser-Induced Periodic Surface Structure (LIPSS); this surface roughness has a wavelength on the order of the wavelength of the machining laser [2]. A nickel sample was machined by laser ablation in the TriBeam system, and the resulting surface was scanned using electron backscatter diffraction (EBSD). The method used to determine the surface topography uses data collected with an EBSD detector and is based on the fact that the location as well as intensity of the backscattered electrons is recorded on the EBSD detector. It is the location of the background intensity that is useful in determining the surface topography. It has been shown previously that the background intensity on an EBSD detector can be modeled using Monte Carlo simulations [3]; the backscattered electrons contributing to the background show nearly specular reflection from the sample surface. The deviation from specular reflection is a function of atomic number and incidence angle. This deviation was determined for Ni using MC simulations. A plot of the intensity profile of the backscatter electrons on the EBSD detector versus the incidence angle of the electron beam on a Ni sample is shown in Fig. 1. These intensity profiles were determined via simulation and can be used to determine the deviation from specular behavior. The location of maximum intensity of the backscatter electrons on the detector is determined, and can be used to determine the exit direction of the backscattered electrons. This exit direction, along with the geometry of the experimental setup and deviation from specular behavior indicate the components of local surface normal on the sample, nx = -x z and ny = -y z. Next, the surface height function, or topography, is reconstructed from the local surface normals. The surface height function for the Ni sample is shown in Fig. 2. References [1] MP Echlin, M Straw, S Randolph, J Filevich and TM Pollock, Materials Characterization, 100, 1, 2015. [2] JE Sipe, JF Young, JS Preston and HM van Driel, Physical Review B, 27, 1141, 1983. [3] PG Callahan and M DeGraef, Microscopy and Microanalysis, 19, 1255, 2013. 2 1 Microsc. Microanal. 21 (Suppl 3), 2015 2370 Figure 1: The intensity profiles of backscattered electrons for different incidence angles for incident electron energy of 30 keV as determined from simulations. Increasing incidence angle corresponds to increased left skewness. Figure 2: The surface height profile of the Ni sample as determined from EBSD data is shown.");sQ1[1184]=new Array("../7337/2371.pdf","Comparison in 3D of Experiments on, and Simulations of Plastic Deformation of Polycrystals","","2371 doi:10.1017/S1431927615012635 Paper No. 1184 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Comparison in 3D of Experiments on, and Simulations of Plastic Deformation of Polycrystals A.D. (Tony) Rollett1, R. Pokharel2, R.A. Lebensohn2 & R.M. Suter3 1 2 Materials Science & Engineering Dept., Carnegie Mellon Univ., Pittsburgh, USA Materials Sci. & Technology, Los Alamos National Laboratory, Los Alamos USA 3 Materials Science & Engineering Dept., Carnegie Mellon Univ., Pittsburgh, USA This talk will address various aspects of image processing relevant to the plastic deformation of metals. The motivation behind the talk is to point out that the current state of comparisons between experiments and simulations of deformation of polycrystalline materials reveals some interesting challenges [1]. Addressing first the image processing issues, EBSD [2] relies heavily on image transformations of the diffraction patterns in which there is useful information about electron energy. HEDM [3] also relies on thresholding of the diffractograms for peak identification and more sophisticated methods are slowly being introduced [4]. Computer simulation of the experiments in which materials are plastically deformed has been accomplished almost entirely with the finite element method. In the past few years, however, an image-based approach5 that relies on the Fast Fourier Transform has started to be used because it offers a more efficient solution of the same equations (e.g. mechanical equilibrium) as the finite element approach. It is possible, for example, to import directly a measured 3D image from HEDM into the FFT simulation code and simulate with no need for the time-consuming step of creating a 3D mesh. Image-based approaches to simulation offer not only more efficient computation and data flow but they also prompt the development of more direct comparison at the level of the raw experimental data, which in this case is the set of diffraction images associated with each snapshot of the material. Turning to the background scientific issues, polycrystal plasticity is akin to deformation of a composite in which each grain has different properties by virtue of its anisotropic response to loading. It is important to demonstrate that we can validate crystal plasticity simulations in order to relate damage initiation such as cracks and voids to extreme values in stress, for example, as they relate to microstructural features such as triple lines. Specific examples will be given for tensile tests on an interstitial-free sheet steel, using electron back-scatter diffraction (EBSD [2]), and for a pure copper, using High Energy Diffraction Microscopy (HEDM [3]). An alternative to the standard finite element method is to model the mechanical response on the image itself using a spectral method based on Fast Fourier transforms (FFT [5]). In general, both experiments and simulations show that hot spots in stress or elastic energy density occur close to grain boundaries, triple lines and quadruple points [6]. Although correlations are found between hot spots and interfaces, special boundaries do not appear to play any role. Various measures of heterogeneity such as maps of grain average strain, Kernel Average Misorientation, Grain Orientation Spread, Intragranular Grain Misorientation and lattice reorientation can be used to make comparisons. As one example of the type of image processing being applied, we discuss a tensile test carried out on a sample of pure copper with a diameter of approximately 1 mm, for which HEDM maps were measured before and after a strain of 6 %. Calculations were performed using a standard viscoplastic model for crystal plasticity embedded in a code, vpFFT, that solves the equations of stress equilibrium and strain compatibility in a periodic unit cell built from a two- or three-dimensional image of the material [5]. A standard constitutive equation relates the strain rate to the local (tensor) stress resolved onto each slip Microsc. Microanal. 21 (Suppl 3), 2015 2372 system. For a given strain rate, the strain rate and stress fields and the corresponding effective stress are obtained iteratively. The code uses the Fast Fourier transform (FFT) to alternate between real and frequency spaces [3]. This permits convolution integrals to be replaced by point-wise tensor products (in frequency space), thus accelerating the computation. The initial state (annealed) of the material was used to instantiate the simulation and a cross-section near the center of the volume was used for comparisons. The change in lattice orientation, reduced to a magnitude, was computed for each point, Fig. 1(a), which reveals significant variations from grain to grain and within grains. The same pointwise lattice rotation was computed from the simulated result, Fig. 1(b). One advantage of using the lattice rotation is that it integrates changes over the strain interval, as compared, say, to computing orientation gradients. This comparison, as with others [1], shows only moderate agreement between experiment and simulation. The challenge to our understanding is therefore why crystal plasticity simulations give good results at the statistical level (e.g. texture development) but indifferent agreement at the local (grain) scale. The many contributions to this work are greatfully acknowledged [7]. References: [1] R Pokharel et al., Annual Reviews in Condensed Matter Physics 5 (2014) p. 317-346 . [2] AJ Schwartz, "Electron backscatter diffraction in materials science", Springer (2009). [3] SF Li and RM Suter, Journal of Applied Crystallography 46 (2013) p. 512�524. [4] J Lind et al., Acta materialia 74 (2014) p. 213-220. [5] RA Lebensohn, Acta materialia 49 (2001) 2723-2737. [6] AD Rollett et al., Modelling Simulation Materials Sci. Eng. 18 (2010) p. 074005 [7] Support from AFOSR, DOE/BES, NSF, NDSEG and LANL is gratefully acknowledged. Discussions with Francis Wagner, Marc De Graef, Chris Hefferan, Ricardo Lebensohn, Frankie Li, Ulrich Lienert, Jon Lind, Reeju Pokharel, Robert Suter, and many others are gratefully acknowledged. Fig. 1(a). Cross-section of specimen of pure copper, for which the magnitude of the lattice rotation (in degrees) has been computed pointwise between the initial state and after 6 % plastic strain in uniaxial tension. The diameter is approximately 1 mm. Fig. 1(b). The same cross-section as shown in 1(a) with the magnitude of lattice rotation between the initial state and after 6 % strain simulated with the vpFFT code.");sQ1[1185]=new Array("../7337/2373.pdf","Improved EBSD Map Fidelity through Re-indexing of Neighbor Averaged Patterns","","2373 doi:10.1017/S1431927615012647 Paper No. 1185 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Improved EBSD Map Fidelity through Re-indexing of Neighbor Averaged Patterns Stuart I. Wright1, Matthew M. Nowell1, Scott P. Lindeman1 and Patrick Camus2 1. 2. EDAX, Draper, Utah, USA EDAX, Mahwah, New Jersey, USA The Hough transform coupled with triplet indexing [1] has proven very robust in the automated determination of crystallographic orientation from EBSD patterns. However, when the incoming patterns are noisy the ability of these routines to correctly determine the orientation correctly can be compromised. This leads to noisy maps reconstructed from the orientation data, an extreme case is shown in Figure 1(a). A variety of clean-up routines have been developed to find isolated points and assign them orientations generally based on some correlation to the orientations of the surrounding pixels in the scan grid. As the original scan data becomes more and more noisy these clean-up routines [2] tend to produce more and more artifacts. A new approach to re-indexing of the stored patterns has been developed that recovers both non-indexed and mis-indexed points while reducing the number of artifacts. A confirmation of this capability can be observed in Figure 1(b). This approach is based on averaging the pattern at a scan point with the patterns of the point's neighbors. The details of this re-indexing approach and its limitations will be presented along with results obtained on four different samples: 1) A nickel alloy sample where increasing amounts of artificial noise have been added to the patterns. 2) A nickel alloy sample with varying EBSD camera settings producing patterns with different degrees of noise. 3) A heavily deformed magnesium sample. 4) Fine grained steel wire. Figure 1 shows the results obtained from the first sample with 20 iterations of applied Poisson noise. Figure 2 shows a summary of the results obtained on samples with varying degrees of camera noise. The results show the positive impact of the pattern averaging on re-indexing although not as dramatic as that achieved on the artificially noisy patterns. The averaging of the pattern at a point with the patterns of neighboring points is similar to increasing the size of the interaction volume. This effect is evident on the results obtained on the heavily deformed magnesium sample. On this sample the averaging led to a loss of fine structure in the resulting map. However, by judicious combining of the original results with the results obtained after averaging enables the advantages of the conventional indexing approach and pattern averaged approach to be realized. The results obtained on samples (1) and (2) were compared against a new indexing method based on matching of the experimental patterns with a dictionary of simulated patterns [3] as a baseline. This comparison confirmed the accuracy of the results as well as verified the indexing success metric based on the confidence index [4]. Microsc. Microanal. 21 (Suppl 3), 2015 2374 References: [1] SI Wright and BL Adams, Metall Trans A 23 (1992), pp. 759-767. [2] SI Wright, Mater Sci Technol 22 (2006), pp. 1287-1296. [3] YH Chen et al, Microsc Microanal (2015), under review. [4] DP Field, Ultramicroscopy 67 (1997), pp. 1-9. a b Figure 1. a) Orientation map reconstructed from EBSD patterns obtained on Nickel with 20 iterations of applied Poisson noise. b) Map reconstructed after applying the neighbor pattern averaging re-indexing algorithm. Figure 2. Indexing success rates for increasingly noisy camera conditions using the standard indexing routines and neighbor pattern averaging re-indexing.");sQ1[1186]=new Array("../7337/2375.pdf","A Novel Way for Determining Bravais Lattice Using a Single Electron Backscatter Diffraction Pattern","","2375 doi:10.1017/S1431927615012659 Paper No. 1186 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 A Novel Way for Determining Bravais Lattice Using a Single Electron Backscatter Diffraction Pattern Lili Li, Ming Han School of Materials Science and Engineering, Fujian University of Technology, Fuzhou, 350118, China An electron backscatter diffraction (EBSD) pattern always provides abundant crystallographic information but disappointingly low accuracy. A typical EBSD pattern usually contains dozens of visible Kikuchi bands that intersect each other to form over a hundred Kikuchi poles. The poles correspond to zone axes in real space or lattice planes in reciprocal space. The band widths are inversely proportional to the interplanar spacings of diffracting lattice planes, and the angle formed by the beam source and two band center-lines approximately corresponds to angle between two lattice planes. However, the band width measurement has a relative error of 5-20% [1] due to the complex profile. In addition, EBSD patterns always suffer from gnomic distortions [2]. Compensating the poor accuracy by the large amount of crystallographic information, a computer program EBSDL has been successfully developed to determine Bravais lattice of unknown crystals using a single EBSD pattern. The pattern center and the detector distance of the pattern as well as the accelerating voltage used are needed to execute EBSDL. Patterns from omphacite, diopside and ilmenite mineral crystals with non-cubic structure will be demonstrated as test cases to show the determining process and results. After manually detecting the bands from the patterns, EBSDL will automatically complete all calculations and give a solution with Bravais lattice type and parameters. Figure 1 presents the interface of EBSDL with an EBSD pattern that is determined to be omphacite with base-centered monoclinic. Also shown are a result dialog and a wizard dialog that orients users to the next step according to current status during the whole operation process. It should be pointed out that the lattice parameters determined from EBSD patterns appear to disregard the atomic type, suggesting that ordering of structure does not influence the EBSD patterns. Detailed analysis for the accuracy of the determination will be discussed. The program EBSDL is based on the Kikuchi bands detected from the pattern, independent from any a priori information about its crystal structure. The algorithm is also unrelated to any crystal symmetry information and is therefore applicable to all crystal systems. The application of the novel program is expected to endow scanning electron microscopy with the feasibility to directly determine Bravais lattice of unknown crystals in addition to the conventional imaging of the surface morphology. References: [1] D Dingley and S Wright, J. Appl. Cryst. 42 (2009), p. 234�241. [2] K Baba-kishi, Scanning 20 (1998), p. 117�127. Microsc. Microanal. 21 (Suppl 3), 2015 2376 Figure 1. A screen capture showing the graphic interface of the EBSDL with an opened test EBSD pattern that is determined to be omphacite. The dialog in the upper left corner is the wizard that floats on the interface during the whole process and is moveable by using mouse. The result dialog in the lower left corner presents the solution as well as the current, average and maximum errors of the bands and poles. The indices bands and poles that is the nearest from the cursor as well as corresponding error information can be instantly updated as the mouse moves on the pattern. The dialog in the upper right shows an as-measured reciprocal lattice plane and the one in the lower right presents the corresponding corrected counterpart associated with fitting grids. The vector that is closest to the cursor position is shown in red, and the error information and the ordinal number of the plane are shown in the panes of the status bar.");sQ1[1187]=new Array("../7337/2377.pdf","Tackling Characterization Challenges in High Deformation/Stress Steel Alloys Using Transmission Kikuchi Diffraction (TKD)","","2377 doi:10.1017/S1431927615012660 Paper No. 1187 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Tackling Characterization Challenges in High Deformation/Stress Steel Alloys Using Transmission Kikuchi Diffraction (TKD) Amir Avishai1, Kevin Abbasi1, Danqi Wang1, Nanthawan Avishai1, Dandan Wu2, Vikram Bedekar2, Scott Hyde2, Scott Sitzman3, and Arthur Heuer1,4 Swagelok Center for Surface Analysis of Materials, Case Western Reserve University, Cleveland, OH, USA. 2. Timken Technology Center, 4500 MT Pleasant Rd NW, N Canton OH, USA. 3. Oxford Instruments America, Inc., Concord, MA, USA. 4. Dept. of Mat. Sci. and Eng., Case Western Reserve University, Cleveland, OH, USA. Although steels have been extensively studied, the application of traditional characterization methods to investigate the microstructure still poses significant challenges. One example is White Etched Areas (WEA) that are microstructural alternations in bearings induced by dynamic loading conditions [1]. Another example is `white layers' at machined steel surfaces, which are generated by hard turning processes [2]. Both involve formation of nanostructured features at the surface that may lead to significant influence on surface-initiated damage, such as corrosion, fatigue and wear surface deformation. These highly deformed regions have grains that range in size from a few nanometers to 100nm and may consist of small pockets of retained austenite. In some cases preexisting carbides are no longer present in the deformed regions. In other processes such as low temperature carburization/nitridation, the challenges are not as much the structural refinement but primarily the very high level of lattice deformation and formation of nanometer size nitrides [3]. Here as well, the large stresses and very high level of interstitial alloying can result in local phase transformation that is not easily identified by scanning electron microscopy (SEM) or conventional transmission electron microscopy (TEM) without extensive effort. In these materials, sample preparation adds to the characterization challenge. Preserving the original microstructure without introducing any mechanical damage during preparation is critical. At the same time, producing adequate samples for investigating these nanometer scale features demands sample thickness and quality similar to high resolution TEM. All examples here were prepared by FIB lift-out, necessary in some cases due to the site specific nature of some of the features of interest, as in the WEA. In other cases due to the location of the features of interest at the sample surface, together with the need to avoid additional mechanical deformation and the presence of high residual stresses, FIB lift-out was required. Transmission Kikuchi Diffraction (TKD) in the SEM, a variation of the EBSD technique, was used to characterize these materials due to its sub10nm spatial resolution capability [4-6] and the EBSD/TKD system's ability to simultaneously collect EDS maps. In most cases, this spatial resolution enabled characterization of such nanostructured regions without the need for TEM, and even offered some advantages over conventional TEM, including the ability to collect quantitative phase and orientation mapping data and to analyze these data using welldeveloped EDS-EBSD post processing tools (Fig 1,2). Sample preparation was the biggest challenge in this work, balancing the need to obtain foil thicknesses on the order of one grain (necessary to minimize pattern overlap) while avoiding preparation-induced surface alteration. Relatively fast sample surface oxidation also meant that the experiments had to be carried out immediately after preparation or repolishing. Nevertheless, the ability to perform all stages of the work on a single FIB-SEM system demonstrates a major advantage of this technique. 1. Microsc. Microanal. 21 (Suppl 3), 2015 2378 References: [1] M.H. Evans, Material Science and Technology, Vol 28, No 1, 2012, p 3-22 [2] V. M. Bedekar, et. al., Microscopy and Microanalysis, 18, Supplement S2, July 2012, p 1778. [3] D. Wang, et. al., Metal Mater Trans A, 45, 2014, p 3578. [4] P.W. Trimby et. al., Acta Materialia, 62, 2014, p 69. [5] A. Avishai, et. al., Microscopy and Microanalysis, 20, Suppl. 3, Aug (2014), p 1476 [6] D. Wu, et. al., Microscopy and Microanalysis, 19, Supplement S2, Aug (2013), p 694. [7] The authors would like to acknowledge the support of Brandon Van Leer and Rick Passy of FEI. a. b. Figure 1. Dark field scanning TEM (STEM) image from a White Etch Area collected using forward scatter detectors in the "oriented dark-field" configuration. a). a lower magnification image of the whole foil, b) high magnification image showing the region of interest. a. b. c. Figure 2. Simultaneously collected data: a). EDS map of Fe, Cr and Al b). Phase maps, with ferrite in blue, austenite in red, and Fe3C in yellow c). TKD orientation map (inverse pole figure, z-direction coloring).");sQ1[1188]=new Array("../7337/2379.pdf","Rotation Axes Analysis of Deformed Magnesium Using Electron Backscatter Diffraction and Rotation Contour Contrast Reconstruction","","2379 doi:10.1017/S1431927615012672 Paper No. 1188 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Rotation Axes Analysis of Deformed Magnesium Using Electron Backscatter Diffraction and Rotation Contour Contrast Reconstruction Shirin Kaboli, Hendrix Demers, and Raynald Gauvin Department of Mining and Materials Engineering, McGill University, Montr�al, Canada. In a deformed polycrystalline microstructure, each grain has a non-uniform gray level in the backscattered electron (BSE) micrograph in a scanning electron microscope (SEM). A non-uniform gray level represents a local crystallographic contrast i.e., local misorientation inside a deformed grain. If deformation occurs in a progressive manner inside a grain, the variation of crystal orientation is not random. As a result, a regular crystallographic contrast in the form of parallel or concentric contours appears inside a deformed grain [1]. The rotation contour contrast (RCC) terminology was used to relate this contrast to the rotation of crystal about one or multiple rotation axes during deformation [2]. In this study, the RCCs in the form of cross-shaped and intersecting contours were observed in deformed Magnesium (Mg) grains in a BSE micrograph. This contrast was attributed to the rotation of the crystal about two rotation axes. The crystallographic directions of the two rotation axes were identified using electron backscatter diffraction (EBSD) and RCC reconstruction [3]. The Mg-0.2Al-0.3Ca (wt%) alloy was selected for rotation axes analysis in this study. The uniaxial hotcompression test was carried out at a temperature of 400 �C, a strain rate of 0.01s-1, and a strain of 0.6 using a 100 kN servo-hydraulic materials testing system. The BSE imaging was carried out at 10� specimen tilt, a 20 keV electron beam energy and a 15 mm working distance using a Hitachi SU-8000 cold-field emission SEM. The EBSD crystal orientation mapping was carried out at 80� specimen tilt (sample surface at 10� from the electron beam axis), a 20 keV electron beam energy and a 25 mm working distance. To index the electron backscatter diffraction patterns (EBSPs) recorded at 80� specimen tilt, a calibration routine was carried out with the standard single crystal silicon wafer specimen with a surface normal [001] and a [110] type reference direction. Using this calibration, the recorded EBSPs were indexed with the HKL Channel5 Flamenco software. A Matlab code was written to read the gray level of a selected pixel in the recorded EBSPs for RCC reconstruction. This pixel represents the position of a virtual electron beam in the EBSP [4]. Fig. 1a shows the BSE micrograph with RCC1 and RCC2 across a deformed Mg grain. Fig. 1b shows a reference EBSP (Pref) obtained from the center of the grain with the pattern center indicated with a red cross. Fig. 1c shows the magnified 1120 zone axis of the Pref for RCC reconstruction. For the first set of RCC reconstruction, five pixels were selected inside the 0002 Kikuchi band indicated with labels ae (set I in Fig. 1c). For the second set of RCC reconstruction, five pixels were selected close to the pattern center along a line parallel to the 1100 Kikuchi band indicated with labels f-j (set II in Fig. 1c). Fig. 1d and 1e show the reconstructed images for set I and set II, respectively. The reconstructed RCCs in image b and image g were very similar to the RCC1 and RCC2 in the BSE micrograph since pixel b and pixel g were selected near the pattern center (i.e., BSE micrograph and reconstructed images b and g have similar diffraction conditions). In set I RCC reconstruction, while RCC1 remained stationary inside the grain, RCC2 moved from the top to the bottom of the grain when the virtual electron beam position moved from pixel a to pixel e. The one-directional movement of the RCC2 from the top to the bottom of the grain in the images a-e indicated a one-axis rotation of the grain about the axis perpendicular to the Microsc. Microanal. 21 (Suppl 3), 2015 2380 0002 Kikuchi band. In set II RCC reconstruction, despite the variations in position, shape and width of the RCC1 with the movement of the virtual electron beam from pixel f to pixel j, the RCC2 remained stationary to a good approximation. However, the RCC1 moved in a lateral direction across the grain in images g to j. The one-directional movement of the RCC1 in a lateral direction across the grain in the images g-j indicated a one-axis rotation of the grain about the axis perpendicular to the 1100 Kikuchi band. As a result, the two rotation axes in this grain were identified as crystallographic directions perpendicular to 0002 and 1100 Kikuchi bands, respectively. References: [1] D C Joy et al, Proceedings of the 5th annual scanning electron microscope symposium part I and part II, workshop on biological specimen preparation for scanning electron microscopy, (1972), pp. 97104. [2] S Kaboli, H Demers, N Brodusch and R Gauvin, J. Appl. Crystallogr. (2014) Submitted. [3] S Kaboli, H Demers and R Gauvin, Ultramicroscopy. (2014) Submitted. [4] N Brodusch, H Demers and R Gauvin, Ultramicroscopy. 148(2015), pp. 123-131. Figure 1. Identification of the crystallographic directions of the two rotation axes in a deformed Magnesium grain. (a) The backscattered electron micrograph of a deformed grain with a rotation contour contrast (RCC) labeled RCC1 and RCC2 across the grain, (b) A reference electron backscatter diffraction pattern (Pref) obtained from the center of the grain in (a), with the pattern center indicated with a red cross, (c) The magnified 1120 zone axis for set I and set II RCC reconstructions, (d) The reconstructed images for pixels a-e in set I RCC reconstruction, (e) The reconstructed images for pixels f-j in set II RCC reconstruction. The two rotation axes in this grain were identified as crystallographic directions perpendicular to 0002 and 1100 Kikuchi bands, respectively.");sQ1[1189]=new Array("../7337/2381.pdf","Ferroplasmons: Novel Plasmons in Metal-Ferromagnetic Bimetallic Nanostructures","","2381 doi:10.1017/S1431927615012684 Paper No. 1189 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Ferroplasmons: Novel Plasmons in Metal-Ferromagnetic Bimetallic Nanostructures J.Ge1, A. Malasi2, N. Passarelli3, L. A. P�rez3, E. A. Coronado3, ,R. Sachan1,4, G. Duscher1,4 and R. Kalyanaraman1,2 1Department of Material Science and Engineering, University of Tennessee, Knoxville, Tennessee 37996, USA 2Department of Chemical and Biomolecular Engineering, University of Tennessee, Knoxville, Tennessee, 37996, USA 3INFIQC, Departamento de Fisicoqu�mica, Facultad de Ciencias Qu�micas, Departamento de Fisicoqu�mica, Universidad Nacional de C�rdoba, C�rdoba 5000, Argentina 4Material Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA New plasmonic materials are required to enable potential strong coupling of light and magnetism, which could lead to improved data storage and/or completely novel optical materials [1, 2]. Bimetallic materials are a promising new way to realize new and/or better plasmonic behaviors [3, 4]. In this paper we first use high-resolution transmission electron microscopy to present evidence for ferroplasmons, strong surface plasmons that appear in ferromagnetic metals when in the proximity of noble metal nanostructures. To better understand and explore the origin of these ferroplasmons we have developed a simple approach to synthesize electron transparent nanostructures so as to compare their plasmonic properties by electron energy loss spectroscopy and compare with calculations performed using the discrete dipole approximation (DDA) approach. Fig. 1(a) shows the high angle annular dark field (HAADF) image of a Co-Ag bimetallic nanoparticle formed on an amorphous carbon substrate by pulsed laser dewetting. A Zeiss Libra 200 TEM with a monochromator and energy filter was used to characterize the nanostructure by HAADF mode and electron energy-loss spectroscopy (EELS) at 200 kV. To understand the spatial nature of surface plasmons in these nanoparticles we generated the maps of plasmons from different energy windows, with the 0 to 3 eV data shown in Fig. 1(b-c) and 3 to 5 eV data shown in Fig. 1(d-e). For the 0 - 3 eV region, the strongest surface plasmon occurred at 2.6�0.3 eV and measurement of the scattering probability [Fig. 1(b)] and full width half maxima (FWHM) [Fig. 1(c)] showed that the Co region had a strong plasmon, called the ferroplasmon. In contrast, analysis of the 3 to 5 eV window showed that a surface plasmon occurred at energy of 3.5 � 0.3 eV and the scattering probability is shown in Fig. 1(d), while the FWHM is shown in Fig. 1(e). This peak was seen to be highly localized to the Ag region. Similar measurements done on pure Co nanoparticles did not show any evidence for these strong ferroplasmons seen when Co was in the vicinity of Ag. DDA modeling has suggested that ferroplasmons can be a result of electromagnetic coupling between the two metals [2]. To investigate this further, we developed a technique to rapidly prepare electron transparent nanostructures and perform EELS on at different energy windows and compare these experiments with DDA electrodynamics theory. The sample preparation involved a combination of nanosphere lithography and a carbon substrate float-off process. Fig. 1(f) shows the Z-contrast from HAADF imaging of pairs of non-touching bimetallic Co-Ag triangles Microsc. Microanal. 21 (Suppl 3), 2015 2382 synthesized by this technique on a-C. Fig. 1(g) shows the EELS spatial map of the scattering intensities in the window of 3.0 to 4.6 eV for Fig. 1(f). Fig. 1(h) is the theoretical prediction of scatter- ing intensity from DDA simulation at an energy of 4.26 eV. Good agreement between theory and experiment was evident, with large scattering coming primarily from Ag for this energy regime. The authors acknowledge support by the Army Research Office through grant W911NF-13- 10428. We also acknowledge CNMS2013-284 at the Center for Nanophase Materials Science, which is sponsored at ORNL by the Scientific User Facilities Division, Office of Basic Energy Sciences, U.S. Department of Energy for SEM characterization. Figure 1: (a) HAADF of a Ag-Co bimetallic nanoparticle. The individual metallic regions are marked in the figure. (b) Scattering probability of the 2.6�0.3 eV plasmon in ppm. (c) The FWHM of the 2.6�0.3 eV plasmon in eV. (d) Scattering probability of the 3.5�0.3 eV plasmon in ppm. (e) The FWHM of the 3.5�0.3 eV plasmon in eV. (f) HAADF of Ag-Co triangle pairs. The brighter triangles are Ag while the darker triangles are Co. (g) EELS map between 3 to 4.6 eV for the triangles in (f). (h) DDA simulation of the EELS from the triangles in (h) at 4.26 eV. References [1] A. N. Grigorenko, A. K. Geim, H. F. Gleeson, Y. Zhang, A. A. Firsov, I. Y. Khrushchev, and J. Petrovic, "Nanofabricated media with negative permeability at visible frequencies," Nature, vol. 438, pp. 335�338, 2005. [2] N. Passarelli, L. A. Perez, and E. A. Coronado, "Plasmonic interactions: From molecular plasmonics and fano resonances to ferroplasmons," ACS Nano, vol. 8, no. 10, pp. 9723�9728, 2014. http://dx.doi.org/10.1021/nn505145v. [3] R. Sachan, V. Ramos, A. Malasi, B. Bartley, G. Duscher, and R. Kalyanaraman, "Oxidation resistant ag nanoparticles for ultrastable plasmonic applications," Adv. Mater., vol. 25, pp. 2045�2050, 2013. [4] R. Sachan, A. Malasi, J. Ge, S. Yadavali, H. Krishna, A. Gangopadhyay, H. Garcia, G. Duscher, and R. Kalyanaraman, "Ferroplasmons: Intense localized surface plasmons in metal-ferromagnetic nanoparticles," ACS Nano, vol. 8, pp. 9790�9798, 2014.");sQ1[1190]=new Array("../7337/2383.pdf","Electron impact investigation of hybridization schemes in coupled split-ring resonators","","2383 doi:10.1017/S1431927615012696 Paper No. 1190 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Electron impact investigation of hybridization schemes in coupled split-ring resonators Qiuqun Liang1, Yuren Wen1, Thomas Reindl2, Nahid Talebi1, Peter A. van Aken1 1. 2. Max Planck Institute for Intelligent Systems, Stuttgart Center for Electron Microscopy, Stuttgart, Germany Max Planck Institute for Solid State Research, Stuttgart, Germany Split-ring resonators (SRRs), the meta-atoms of many magnetic metamaterials with negative permeability [1], have been widely applied to bio-sensing devices [2], antennas [3], perfect metamaterial absorbers [4] and filters [5]. The very wide applicability of SRRs in integrated photonic devices is based on the closedloop electric current that this element sustains, which in a dual representation effectively builds up a magnetic dipole moment. The concomitant induction of electric and magnetic moments in SRRs is utilized to realize negative refractive index metamaterials [1]. Besides its application in photonic circuitry, the electromagnetic interactions between the meta-atoms are also fundamentally interesting for investigating the coupling effects between induced moments. An intuitive understanding of the interaction between the plasmons of adjacent nanoparticles can be provided by the plasmon hybridization method [6], utilizing a hydrodynamic approximation for the conduction electrons. The result of such an approximation can be understood as the formation of bonding and anti-bonding final states. This hybridization method has been investigated in the dipolar limit for several simple geometries. We present a comprehensive experimental and theoretical investigation of the plasmon hybridization of coupled Au split-ring resonators (SRRs) by means of electron energy-loss spectroscopy (EELS), energyfiltered transmission electron microscopy (EFTEM), discrete dipole approximation (DDA), and Comsol Multiphysics calculations. The samples were patterned in a 30 kV electron beam lithography system (Raith eLINE). A 3 nm thick Cr adhesion layer and a 30 nm thick Au metal layer were deposited on a 30 nm thick Si3N4 membrane as a support grid. EELS and EFTEM experiments were performed in the Zeiss SESAM microscope. EELS can directly measure the photonic local density of states projected along the electron trajectory and directly probe the energy splitting in coupled meta-atoms. Additionally, we can selectively excite particular plasmonic modes by focusing the electron beam onto hotspot regions, since the plasmonic modes sustain unique spatial distributions with hotspots localized along the circumference of the metallic nanostructures. Various trajectories of the electron beam have been used to selectively excite different plasmonic modes in SRRs. In addition, different relative orientations of adjacent split rings have been used to study the hybridization of the induced charges and magnetic moments of the coupled SRRs according to the orientation of the induced moments. An in-depth understanding of the fundamental and high-order multipole plasmon coupling effects of the coupled SRRs, which have been not investigated beforehand, will provide significant perception into the design and optimization of magnetic metamaterials with desirable properties as well as their resonant behavior. We provide a complete scheme of the multimodal distribution of the coupled and single-element splitring resonators, with a precise attention to the hybridization of those modes according to the induced moment. Our study suggests a clear dominance of electric and magnetic dipole moments for even higher-order modes in the far-field radiation spectrum. The calculated radiated spectra from each individual electric and magnetic moments demonstrate that the role of the dipole moments play a key role in the plasmonic hybridization [7]. Microsc. Microanal. 21 (Suppl 3), 2015 2384 References: [1] Veselago, Sov. Phys. Usp. 10 (1968) 509 [2] HJ Lee and J G Yook, Appl. Phys. Lett. 92 (2008) 254103 [3] KB Alici, AE Serebryannikov and E Ozbay, J. Electromagn. Wav. Appl. 24 (2010) 1183 [4] Y Cheng et al, Appl. Phys. A 102 (2011) 99 [5] YD Dong, T Yang and T Itoh, Microw. Theory Tech. 57 (2009) 2211 [6] E Prodan et al, Science 302 (2003) 419 [7] QL gratefully acknowledges financial support from the Doctoral Training Program between Max Planck Society and Chinese Academy of Sciences. NT acknowledges the Alexander von Humboldt Foundation for the research fellowship. The research leading to these results has received funding from the European Union Seventh Framework Program [FP/2007-2013] under grant agreement no 312483 (ESTEEM2). Figure 1. (a) Calculated EELS maps and (b) experimental EFTEM image of the isolated Au SRR structure and a coupled Au SRRs structure at resonance energies of the fundamental, second- and thirdorder plasmonic modes. The parameters of the Au SRRs in the simulations are: thickness H = 30 nm, length L = 120 nm, width W = 30 nm and gap distance between the left and right SRRs G = 20 nm. The scale bars of the EFTEM images are 100 nm and the color codes for the intensities in (a) and (b) are presented accordingly.");sQ1[1191]=new Array("../7337/2385.pdf","Visualizing Plasmonic Coupling in Metamaterials and Applying Angular Resolved EELS","","2385 doi:10.1017/S1431927615012702 Paper No. 1191 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Visualizing Plasmonic Coupling in Metamaterials and Applying Angular Resolved EELS Felix von Cube1,2,3, Jens Niegemann4, David C. Bell1 , Stephan Irsen3 and Stefan Linden2 1. 2. School of Engineering and Applied Sciences, Harvard University, Cambridge, MA, USA. Physikalisches Institut, Universit�t Bonn, Bonn, Germany. 3. Electron Microscopy and Analysis (EMA), Center of Advanced European Studies and Research (caesar), Bonn, Germany. 4. Laboratory for Electromagnetic Fields and Microwave Electronics, Swiss Federal Institute of Technology (ETH), Z�rich, Switzerland. The metamaterial concept has introduced numerous fascinating phenomena to optics and photonics, as for example a negative index of refraction. In most cases, the origin of such a phenomenon can be traced back to the excitation of plasmonic resonances in the building blocks of the metamaterial. Hence, the knowledge of the plasmonic near-fields is of upmost interest for an in-depth understanding of the properties of a given metamaterial. One of the paradigm metamaterial is the split-ring resonators (SRR), it is a subwavelength metallic ring with a slit (Fig. 1a). We perform electron energy-loss spectroscopy (EELS) on lithographically defined gold SRRs to map the electric near-field distribution of their plasmonic resonances. We use EELS in combination with scanning transmission electron microscopy (STEM) allowing for nanometer spatial and sub eV energetic resolution [1]. For a given electron loss energy, strong EELS signals are expected, if the excited plasmonic resonance has a large electric field component Ez along the trajectory (z-axis) of the electron beam. In case of an isolated SRR, the first plasmonic resonance shows two strong electric field components at the arms of the structure (Fig. 1c). Metamaterials usually consist not only of one isolated structure, but of thousands of structures, therefore we are interested in the coupling behavior of two or more SRRs. In case of a SRR dimer, the electromagnetic coupling leads to the formation of two new plasmonic resonances, one slightly red shifted and one slightly blue shifted (Fig. 1 d,e). The formation of these new resonances can be explained with a classical dipole model, in which the dipoles oscillate either in-phase or out-of-phase. Furthermore, we observe that coupling in small SRR ensembles results in the formation of discrete eigenmodes where the number of modes corresponds to the number of SRRs. However, the EELS maps of large arrays, containing thousands of SRRs, show the formation of a quasi-continuum of modes, which is in accordance with the plasmon hybridization model [2, 3]. We also investigate the plasmonic near-field with angular resolved EELS. By tilting the sample, different electric field components of the plasmonic modes can be probed with the electron beam. This allows us to detect the Ex and Ey components of the electric near-field, in addition to the Ez component, which is probed in conventional EELS. The electron energy-loss spectra recorded under oblique incidence feature plasmonic resonances that are not observable under normal incidence (Fig. 2). Our experimental findings are reproduced by full numerical calculations based on the discontinuous Galerkin time-domain method [4]. In conclusion, we have used STEM-EELS to investigate the electric near-field of plasmon resonances on SRRs. We observe coupling effects in ensembles of two and more SRRs which result in the formation of Microsc. Microanal. 21 (Suppl 3), 2015 2386 new plasmonic modes. Additionally we have performed angular resolved EELS to get a three dimensional insight on the electric near-field [5]. References: [1] J. Nelayah et al, Nat. Phys. 3 (2007), p. 348. [2] F. von Cube et al, Nano Letters 13 (2013), p. 703. [3] P. Nordlander et al, Nano Letters 4 (2004), p. 899. [4] F. von Cube et al, Physical Review B 89 (2014), p. 115434. [5] S.I. and S.L. acknowledge the financial support of the DFG-Project LI 1641/2-1. F.v.C. acknowledges financial support by the DAAD. This work was performed in part at the Center for Nanoscale Systems (CNS), a member of the National Nanotechnology Infrastructure Network (NNIN), which is supported by the National Science Foundation under NSF award no. ECS-0335765. CNS is part of Harvard University. Figure 1. Dark field images of (a) an isolated split-ring resonator (SRR) and (b) a SRR dimer. The corresponding electron-energy loss spectroscopy (EELS) maps are displayed in (c-e). For the isolated SRR we find one plasmonic resonance at 0.75 eV. For the SRR dimer, two new resonances arise: one is red-shifted (d) and one is blueshifted (e), with respect to the resonance energy of the isolated structure. According to their optical properties, the low energy mode is called an optical bright mode and the high energy mode an optical dark mode. The white lines indicate the boundaries of the SRRs. The scale bars are 200 nm. Figure 2. (a) shows a dark-field image of a split ring resonator, the blue spot indicates the position of the fixed electron beam; the scale bar is 200 nm. (b-d) scheme the different sample orientations: (b) conventional (orthogonal) electron incidence, (c) sample tilt around the y-axis, and (d) sample tilt around the x-axis. (e) EEL spectra recorded at the position shown in (a) for different tilt-angles. The vertical dashed lines indicate the resonance-energies of the first-order and second-order plasmonic mode. For a tilt around the y-axis we find a strong resonance, which cannot be found under orthogonal electron beam incidence.");sQ1[1192]=new Array("../7337/2387.pdf","Edge modes of surface plasmon nanodisk antennas visualized in 3-photon PEEM","","2387 doi:10.1017/S1431927615012714 Paper No. 1192 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Edge modes of surface plasmon nanodisk antennas visualized in 3-photon PEEM R. C. Word1 and R. K�nenkamp1 1 Department of Physics, Portland State University, Portland, OR 97201, USA Photoelectron emission microscopy (PEEM) provides the high spatial resolution of electron microscopy and, at the same time, the ability to explore the breadth and complexity of light-matter interactions. The emission of imaging photoelectrons is based on the photoelectric effect, which is a fundamental quantum mechanical phenomenon. As a result PEEM images provide information about the photoemission thresholds of a specimen. The rate of photoemission depends on the electric field at the surface. Additionally, ultrafast lasers permit PEEM to be used at visible and infrared energies. In this regime, the yield of multi-photon PEEM (nP-PEEM) is proportional to incident intensity to n power, where n is the number of photons per electron needed to overcome the photoemission threshold. PEEM thus provides a high resolution, probe-free method of studying surface electromagnetic fields. The origin of these fields can be photonic or plasmonic. PEEM can be used, for example, to study light propagating in a slab waveguide [1], the time resolved propagation of long-ranging surface plasmon polaritons (SPP) confined to a planar metal surface [2], or, as in the present case, localized surface plasmon field distributions of nanoscale antennas [3]. Such characterization of optical and plasmonic antennas is important in the quest to develop nanoscale optical communication devices [4, 5] where there is particular interest in the infrared and visible spectral region [6, 7]. Here we report the direct observation of plasmonic edge modes or whispering gallery-type modes of nanoscale omni-directional disk antennas using near infrared three-photon PEEM [8]. Our antennas are milled with a focused ion beam (FIB) from gold on copper composite films on indium tin oxide substrate (ITO) on glass. An ultrafast laser (0 = 800 nm) incident at 60� to the surface normal generates surface plasmons at the edges of the disks by diffraction (Fig 1). Imaged photoelectrons transmit information on the strength and extent of the evanescent plasmon fields generated along the antenna rim and across its surfaces. High contrast between the pristine gold and milled ITO regions occurs because the excitation laser light energy of 1.55 eV, is sufficient for 3-photon emission from ITO (work function 4.2�4.5 eV) and possibly copper (4.5-4.8 eV), whereas four photons are needed for photoemission from gold (5.0-5.1 eV). It is important to stress that in these experiments a femtosecond pulsed laser is used to optically excite plasmon modes. Photoelectron emission occurs as the plasmonic excitation decays and interferes with the incident laser light. This detection process is different from the approach in EELS and cathodoluminescence where SPP modes are excited by electron beams [7, 9]. In our work resonances and constructive interference effects occur when the circumference of the receiver antenna is equal to an integer number of surface plasmon wavelengths (SPP = 775 nm) [10] as shown for a quadrupole mode in Fig 1. We have recently shown that the quadrupole field can be rotated in-plane to any desired position by controlling the angle of the incident light polarization. This control of the field position can be used to fine-tune the detection of particles in plasmonic sensor arrangements. Our experimental results agree well with the electric field distribution calculated by finite element method (FEM) simulations. Additional experiments are presently underway to observe higher and lower order modes, which FEM simulations suggest should also be observable. Of particular interest is the role the gap width, mill depth, and ITO layer plays in the development of the modes. Microsc. Microanal. 21 (Suppl 3), 2015 2388 Experimental details The disk antennas were milled using a FEI-235 FIB from a composite gold/copper metal film, which was vacuum evaporated onto ITO coated glass coverslips (SPI Supplies). The gallium beam current of the FIB was 30 pA. In a home-built aberration-corrected PEEM the antennas were exposed to 800-nm, 900-mW, 80-fs laser light from a Ti:sapphire laser. The laser spot size was approximately 100 �m, was incident at 60�, and had polarization set by a tunable wave plate. Photoelectron images were recorded with a CCD optically coupled to a phosphor screen with an image exposure time of 60 seconds. The images below are composite averages made from images of six identically prepared antennas. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] JPS Fitzgerald, RC Word, R K�nenkamp, Phys. Rev. B 89 (2014), p. 195129. C Lemke et al., Nano Lett. 13 (2013), p. 1053. L Douillard, F Charra, J. Electron. Spectrosc. Relat. Phenom. 189, Supplement (2013), p. 24. A Al�, N Engheta, Phys. Rev. Lett. 104 (2010), p. 213902. T Coenen, EJR Vesseur, A Polman, ACS Nano 6 (2012), p. 1742. J Lee, J Song, GY Sung, JH Shin, Nano Lett. 14 (2014), p. 5533. EJR Vesseur, A Polman, Nano Lett. 11 (2011), p. 5524. RC Word, JPS Fitzgerald, R K�nenkamp, Appl. Phys. Lett. 105 (2014), p. 111114. F-P Schmidt et al., Nano Lett. 12 (2012), p. 5780. F-P Schmidt et al., Nat. Commun. 5 (2014), p. 3604. This research was supported by the US-DOE Basic Science Office under Contract No. DE-FG0213ER46406. Figure 1. (Left) Drawing of a gold/copper nanodisk antenna on ITO/glass substrate indicating the direction and angle of incident laser light. Photo-emitted electrons are accelerated into the PEEM's electron optics above the surface. (Right) False-color three-photon PEEM images of the quadrupole mode of an antenna (shown as cube-root of the yield) compared to FEM simulations (as electric field norm squared) for four different light polarizations.");sQ1[1193]=new Array("../7337/2389.pdf","Direct Observation of Plasmonic Enhancement of Emission in Ag-nanoparticle-decorated ZnO nanostructures","","2389 doi:10.1017/S1431927615012726 Paper No. 1193 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Direct Observation of Plasmonic Enhancement of Emission in Ag-nanoparticledecorated ZnO nanostructures Jordan Hachtel1,2, Daniel Mayo3,4, Claire E. Marvinney3, Anas Mouti2, Richard Mu4, Stephen J. Pennycook5, Andrew R. Lupini2, Matthew F. Chisholm2, Richard F. Haglund1,3, Sokrates T. Pantelides1,2,6 1. 2. Vanderbilt University, Department of Physics and Astronomy, Nashville, TN USA Oak Ridge National Laboratory, Materials Science and Technology Division, Oak Ridge, TN USA 3. Vanderbilt University, Interdisciplinary Materials Science Program, Nashville TN, USA 4. Fisk University, Department of Physics and Astronomy, Nashville, TN USA 5. National University of Singapore, Department of Materials Science and Engineering, Singapore 6. Vanderbilt University, Department of Electrical Engineering and Computer Science, Nashville, TN USA The interactions between surface plasmon resonances (SPRs) in metals and photon emission in semiconductors open a wide range of tuning and enhancement applications in nanostructured devices [1,2]. Optical techniques, such as photoluminescence (PL), have been used extensively to study these effects, but do not have the ability to map out their nanoscale behavior. Scanning transmission electron microscopy (STEM) combined with cathodoluminescence (CL) spectroscopy provides us with a tool to map and analyze plasmon and exciton behavior with nanoscale precision. ZnO has been a prominent material in nanoscale optoelectronics due to its stable ultraviolet band-edge emission and highly-tunable visible defect emission. We combine emitting ZnO nanowires with an insulating MgO shell, and decorate the surface of the nanowire with Ag nanoparticles, shown in Figure 1a and 1b. The result is a versatile optical workbench, ideally suited for studying the way in which SPRs from the Ag nanoparticles enhance and interact with the exciton emission in the ZnO nanowires. The structure is first studied macroscopically through PL, as a function of MgO spacer thickness, shown in Figure 1c. The overall PL enhancement is dominated by cavity resonance effects, corresponding to Fabry-Perot and the combination of Fabry-Perot and whispering gallery mode resonances at 20 and 60 nm thicknesses, respectively [3]. However, by comparing the PL enhancement factor of the nanowires with nanoparticles to the ones without, we can see that there is a significant enhancement due to the presence of the plasmons that cannot be explained by cavity resonance effects. We turn to STEM to shed light on the plasmonic enhancement of the ZnO/MgO nanowires. We can observe the direct enhancement of the ZnO emission by comparing and contrasting annular dark field (ADF) imaging and CL imaging. Figure 2a and 2b show ADF and CL images of the same bare nanowire (a nanowire with no MgO shell). We horizontally sum and normalize the intensities of the nanowire in both images, Figure 2c, and observe that the peaks corresponding to the nanoparticles in CL are much broader than the peaks corresponding to the nanoparticles in ADF. Peaks in ADF correspond to the actual size of the nanoparticles, while the CL peaks correspond to the emission due to the nanoparticle. Hence, a CL peak from a plasmonic nanoparticle that is broader than the corresponding ADF profile of the same nanoparticle indicates plasmon-enhanced emission. However, we also know that plasmons are excited non-locally by the electron beam, and indeed the Microsc. Microanal. 21 (Suppl 3), 2015 2390 width of the CL peak extends past the ADF profile normal to the nanowire. There is no ZnO here, so the detected emission can only be from surface plasmons in the silver nanoparticle excited non-locally. We must therefore distinguish between non-local plasmon excitation and plasmon-enhanced ZnO emission. To this end we compare the horizontal and vertical ADF and CL profiles of a large nanoparticle at the top of the wire. In Figure 2d we see the horizontal CL profile is 20 nm broader than the ADF profile, showing the maximum range of non-local plasmon excitation. Figure 2e, on the other hand, shows a vertical CL profile 50 nm broader than the ADF profile, demonstrating enhancement beyond the range of non-local plasmon excitation. Hence, the presence of the plasmonic nanoparticles directly enhances the emission from the ZnO nanowires, even well away from the nanoparticles themselves [4]. References: [1] P. K. Jain, S. Eustis, and M.A. El-Sayed, J. Phys. Chem. B 110 (2006) 18243 [2] A. Manjavacas, F. J. G. de Abajo, and P. Nordlander, Nano Lett. 11 (2011) 2318 [3] D. C. Mayo et al, Thin Solid Films 553 (2014) 132 [4] This work was funded by NSF-EPS-1004083, NSF-TN-SCORE, DOD-W911NF-11-1-0156, DODW911NF-13-1-0153, DE-FG02-09ER46554, DE-FG02-01ER45916, and the DOE Office of Science BES Materials Science and Engineering Division. Figure 1. Plasmonic enhancement in Ag decorated ZnO/MgO core/shell nanowires. (a) HAADF and (b) EDAX images of the nanostructures demonstrating their composition and size. (c) Measured PL enhancement due to Fabry-Perot cavity resonances for structures with and without nanoparticles. (d) The enhancement factor due to the presence of the plasmonic Ag nanoparticles. Figure 2. Direct observation of plasmonic enhancement. (a) ADF and (b) CL images of `bare' nanowire. (c) The normalized intensity of the ADF and CL signals showing broad CL peaks. (d) Horizontal and (e) vertical line profiles of ADF and CL image. Demonstrating that the enhanced emission does not originate from non-local excitation of plasmon, and comes from the ZnO core.");sQ1[1194]=new Array("../7337/2391.pdf","Nanoscale Imaging Mass Spectrometry using Cluster Ion Beams.","","2391 doi:10.1017/S1431927615012738 Paper No. 1194 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Nanoscale Imaging Mass Spectrometry using Cluster Ion Beams. Nicholas Winograd Department of Chemistry, Penn State University, University Park, PA 16875. Imaging mass spectrometry has emerged, in recent years, to allow chemical composition to be elucidated with impressive spatial resolution and sensitivity. Many different approaches have been developed, each with specific attributes. Techniques employing soft ionization schemes such as MALDI and desorption electrospray ionization (DESI) have been successful particularly for biological samples since the target molecule is generally detected without fragmentation. Spatial resolution is generally limited to ~100 �m, although better resolution has been reported in special cases. Here, we focus on molecular secondary ion mass spectrometry (SIMS) using focused cluster ion beams to stimulate the desorption of molecular ions [1]. Since desorption results from energetic collisions between the primary cluster and the target molecule, significant fragmentation of molecular ions is usually observed. Despite this problem, there are several unique operational modes of the SIMS approach that present a unique space for materials characterization. These modes include the ability to acquire surface-specific chemical information and to perform 3dimensional imaging with sub-micron spatial resolution and nanometer-scale depth resolution. This approach is distinct from nanoSIMS where the energetic ion beam intentionally fragments the molecules to small fragments which are distinguished using stable isotope labeling. The development of cluster ion beams for imaging SIMS has been a real success story. About 15 years ago, it was shown that metal cluster ions such as Au3 and Bi3 could be generated from liquid metal ions sources and focused to a spot size of less than 100 nm [2]. These clusters were shown to produce mass spectra with much higher ion yields and less fragmentation than the more conventional atomic ion projectiles. Ion sources employing bigger clusters such as C60 were also commercialized and incorporated into existing time-of-flight (TOF)-SIMS machines [3]. These probes could not be focused as well as the liquid metal ion sources, but could still yield spatial resolutions in the range of 300 nm. Most recently, gas cluster ion beams (GCIB) produced by supersonic expansion, have been perfected for SIMS imaging, yielding probes consisting of, for example, argon clusters ranging in size from several hundred to several thousand atoms [4]. A critically important ancillary benefit of the cluster ion sources is that there is minimal chemical damage accumulation during erosion of the sample. The information in the mass spectrum is retained as a function of depth into the sample. Many studies using C60, for example, on well-defined layered structures such as those containing Irganox polymer additives, and lipid bilayers, showed that 10-20 nm depth resolution could be achieved through a thickness of several microns. Moreover, the GCIB sources are now able to discern sub-surface layers with 5-10 nm resolution. Hence, it is possible to acquire images with submicron spatial resolution, and to stack these images during molecular depth profiling to achieve a 3dimensional chemically-resolved picture with nanometer depth resolution [5]. There are two major problems that need to be resolved before this approach can be routinely applied. First, and foremost, is the issue of sensitivity. As the spot size of the primary ion beam is reduced toward zero, the number of molecules available for detection also approaches zero. Moreover, ionization efficiency for any of these desorption methods is thought to be in the range of 1 molecule in 108 to one molecule in 104, a very rare event indeed. The ultimate spatial resolution is generally limited by sample Microsc. Microanal. 21 (Suppl 3), 2015 2392 size, rather than by how tightly the ion beam can be focused, especially consider there are only ~106 molecules per layer per square micron. There is a second conundrum associated with the GCIB sources. With smaller cluster projectiles, ions are produced in greater quantity, but more molecules are fragmented, reducing the ionization efficiency. With the larger clusters, molecular desorption is very efficient, but the "softness" of the ionization means that the ionization efficiency is poor. In this talk, various solutions to this conundrum will be explored, including the use of Argon clusters doped with H-containing species to enhance [M+H]+ formation. The unique properties associated with cluster SIMS have altered the instrumentation landscape. The low duty cycle associated with traditional TOF-SIMS instruments complicates depth profiling and image quality. There has been a recent move toward the use of continuous ion beams, rather than pulsed ion beams to circumvent these complications. Several options will be discussed in this talk, including the use of a shaped field buncher to time focus a section of the continuous beam, which is then accelerated into a harmonic reflectron [6]. Finally, we present several applications of 2 and 3-dimensional imaging using cluster beams to characterize a variety of biological samples. For these applications, sample preparation is a key part of the measurement. For example, we are able to show the lipid distribution at the single cell level for HeLa cells by combining the high spatial resolution of the C60 source with the improved molecular desorption properties of the GCIB. In addition, we show that to achieve the best spatial resolution using tissue slices, it is necessary to avoid using a matrix coating, and to perform the measurements at cryogenic temperatures to avoid molecule migration and evaporation. Other issues such as the presence of matrix ionization effects, differential sputtering and charging will be discussed in detail. In general, the lecture will attempt to provide an assessment of where the technique stands with respect to other imaging mass spectrometry methods, what the current challenges are for improvement, and to speculate about future developments in the coming years. References: [1] J. C. Vickerman and N. Winograd, International Journal of Mass Spectrometry, ASAP (2015). DOI: 10.1016/j.ijms.2014.06.021 [2] D. Touboul et al., J. Am. Soc. Mass Spectrom. 16 (2005), p. 1608-1618. [3] D. Weibel et al., Anal. Chem. 75 (2003), p. 1754-1764. [4] N. Toyoda et al., Appl. Surf. Sci. 203 (2003), p. 214-218. [5] N. Winograd, Surf. Interface. Anal. 45 (2013), p. 3-8. [6] R. Hill et al., Surf. Interface Anal. 43 (2011), p. 506-509. [7] Funding from the Department of Energy (grant no. DE-FG02-06ER15803), The National Science Foundation (grant no. CHE-0908226) and the National Institutes of Health (grant no 9R01 GM11374620A1) are gratefully acknowledged.");sQ1[1195]=new Array("../7337/2393.pdf","Multimodal Imaging of Environmental and Biological Liquid Surfaces and Interfaces Using Imaging Mass Spectrometry","","2393 doi:10.1017/S143192761501274X Paper No. 1195 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Multimodal Imaging of Environmental and Biological Liquid Surfaces and Interfaces Using Imaging Mass Spectrometry Xiao-Ying Yu1 1. Fundamental & Computational Sciences Directorate, Pacific Northwest National Laboratory, Richland, WA 99354, USA. Imaging mass spectrometry (IMS), including time-of-flight secondary ion mass spectrometry (ToFSIMS), matrix-assisted laser desorption/ionization, or desorption electrospray ionization, uses the analytical power of mass spectrometry to generate chemical images illustrating the distribution of molecules in a sample. IMS generally requires that the analyte be transferred from the condensed phase, ionized, separated in vacuum, and detected. Imaging liquid surfaces and interfaces is a known challenge due to the high volatility of liquids in vacuum. We developed a vacuum compatible microfluidic interface, System for Analysis at the Liquid Vacuum Interface (SALVI), to enable surface analysis of liquids and liquid-solid interactions using ToF-SIMS. SALVI's detection window is an aperture of 2-3 m in diameter open to vacuum, which allows direct detection of the liquid surface. Liquid is withheld by surface tension within the aperture. SALVI is composed of a silicon nitride (SiN) membrane and polydimethylsiloxane (PDMS) microchannel in a block [1, 2]. Its applications to enable liquid ToF-SIMS as a new analytical tool were evaluated and reported. Most recently, we showed that solvent structure can be determined. Moreover, we demonstrated in situ correlative imaging of single mammalian cells using ToF-SIMS guided by structure illumination microscopy (SIM) for the first time. In addition, we extended SALVI for biofilm growth and real-time characterization [3]. We provided the first correlative chemical imaging of live biofilms using confocal laser scanning microscopy, nuclear magnetic resonance imaging, and ToF-SIMS. A ToF-SIMS V spectrometer (IONTOF GmbH, M�nster, Germany) was used mainly in these experiments. A pulsed 25 keV Bin+ (beam size: ~250 nm) ion beam with an incident angle of 45 degree off the normal was used as the primary ion beam for all measurements. The SIMS measurements were performed at the beam current of ~1.0 pA with a beam width of 130 ns and a repeated frequency of 20 kHz. The main chamber vacuum pressure was 2-4�10-7 mbar with SALVI inside the chamber, and the pressure only slightly increased to 3-5�10-7 mbar during measurements. This indicates that no spraying or fast spreading of aqueous solutions occurs through the aperture. Figure 1 depicts the results of the ionic liquid solvent structure observation using SALVI and in situ liquid ToF-SIMS. For ionic liquids with different CO2 loadings, the m/z spectra show clear spatial chemical inhomogeneity as predicted by molecular dynamic simulations. Figure 2 depicts the first correlative imaging results of a mouse lung C10 cell using SIM and ToF-SIMS. These new results demonstrated advancements of multimodal mesoscale imaging of liquid surfaces and liquid-solid interfaces in situ in their natural liquid environment using a universal sample holder, SALVI [4]. References: [1] L Yang et al., J. Vac. Sci. Technol. A 29 (2011), art. no., 061101, doi: 10.1116/1.3654147. [2] L Yang et al., Lab Chip 11 (2011), 2481-4. doi: 10.1039/c01c00676a. Microsc. Microanal. 21 (Suppl 3), 2015 2394 [3] H Xin et al., Analyst (2014), 139(7), 1609-13. doi: 10.1039/C3AN02262E. [4] H Shi et al., Catal. Sci. Technol. (2015), advanced on-line. doi: 10.1039/C4CY01720J. [5] The authors acknowledge funding from the Pacific Northwest National Laboratory (PNNL) Technology Development Program, Materials Synthesis and Simulation across Scales (MS3) Initiative and the Chemical Imaging Initiative (CII) Laboratory Directed Research and Development fund. Figure 1. Negative m/z spectra of an ionic liquid of 100 % CO2 loading at different locations along the microchannel. Figure 2. (a) Fluorescence SIM image of a C10 cell in SALVI. (b) ToF-SIMS m/z spectra acquired at the corresponding locations highlighted in green circles in (a); and (c) 2D ToF-SIMS images showing the spatial distribution of representative cell molecular fragments from location 1 in (a).");sQ1[1196]=new Array("../7337/2395.pdf","Molecular Characterization of Lubricant Degradation Produced in a Tribological Wear Test Using TOF-SIMS Imaging MS","","2395 doi:10.1017/S1431927615012751 Paper No. 1196 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Molecular Characterization of Lubricant Degradation Produced in a Tribological Wear Test Using TOF-SIMS Imaging MS Gregory L. Fisher1, Thierry Le Monge2, Jean M. Martin2, and Cyrielle Forest2 1. 2. Physical Electronics, Chanhassen, MN 55317, USA Ecole Centrale de Lyon, Tribology & Dynamic Systems, 69134 Ecully Cedex, France We have studied the effects of friction, produced in the course of a tribological wear test, on unsaturated fatty acids and saturated methyl esters used as lubricants in diesel fuel. The fatty acid and methyl ester additives provide necessary lubricating properties, and there is evidence of a protective tribochemical film that forms on metal contact surfaces when using fatty acid enriched biodiesel fuels. Prior investigation [1] has revealed that the presence of methyl esters alone reduces the frictional coefficient by 30%, while the addition of fatty acids can further reduce the frictional coefficient by up to another 50%. X-ray photoelectron spectroscopy (XPS) analysis of wear test specimens has indicated the evolution of a slightly dielectric tribo film in low wear areas. The dielectric tribo film was marked by increased hydroxyl and carbonyl functionality as well as decreased ester functionality compared to the initial surface. The present study focuses on the application of time-of-flight SIMS (TOF-SIMS) to determine the molecular evolution of lubricants used in biodiesel fuel and the potential metal-organic reaction products leading to formation of a tribo film. The test specimens were produced on a reciprocating cylinder-onflat tribometer with conditions intended to simulate the piston-cylinder contact geometry and dynamics that are typical of an internal combustion engine. A conventional diesel fuel (i.e. no addition of fatty acid methyl ester) was used for the present testing so as to isolate the molecular evolution of the fatty acid lubricants under the test conditions. The lubricants consists of C18 fatty acids, with one or two unsaturations, at a concentration of 800 part-per-million (ppm). The TOF-SIMS analysis was performed using a PHI nanoTOF II instrument equipped with a Bi cluster liquid metal ion gun (LMIG). The TRIFT mass spectrometer of the nanoTOF II provided an advantage for this study in that the wear track topography is effectively decoupled from the molecular characterization and imaging. High resolution molecular identification and imaging was accomplished using the HR2 imaging mode of the nanoTOF II's newly designed LMIG which achieves a spatial resolution (l) of approximately 400 nm at a mass resolution (m/m) of approximately 10,000 using a high analytical beam current. The HR2 imaging mode was an indispensable resource for high resolution molecular identification and imaging. There were no intact fatty acid lubricants observed on the wear test specimens, even outside the wear track area. The larger molecular decomposition fragments of the fatty acid lubricants were observed in the low wear regions of the wear track. Rearrangement products of the fatty acid lubricants were observed at the greatest concentrations localized to the high wear regions of the wear track. The fatty acid lubricants appear to transform in the presence of heat and oxygen to a mellitate-like intermediate in the process of forming the protective tribo film. Other rearrangement products, ostensibly formed with traces of sulphur and phosphorous that were clearly observed in the metal, include benzyl, cresyl, and naphthyl sulfonates and phosphates. The presence of the mellitates along with the benzyl, cresyl, and Microsc. Microanal. 21 (Suppl 3), 2015 2396 naphthyl side groups indicates a high degree of reduction and cyclization that evolved from the initial fatty acid lubricant additives. References: [1] J.M. Martin et al, Friction 1 (2013) p. 252. Figure 1. (A) Stage raster mosaic map images of the tribological wear track area in the positive ion polarity (top; Fe+, 56 m/z) and the negative ion polarity (bottom; C3H3O2-, 71 m/z). The total map area is 6.0 x 2.4 mm2, and the image is binned down to 256 x 256 pixels. High wear and low wear regions are indicated, and the region to the right of the white line is outside the wear track. (B) Positive ion polarity mass spectrum in the range of m/z = 200 � 700 from the entire image area. Peaks indicated with blue labels are a series of neutral C2H4 losses (m = -28.0316) with the highest m/z peak arising at C36H76N+. Various mellitate species appear at higher mass-to-charge ratios. (C) Negative ion polarity mass spectrum in the range of m/z = 200 � 700 from the entire image area. The mellitate, sulfonate, and phosphate moieties are identified.");sQ1[1197]=new Array("../7337/2397.pdf","High-Resolution Imaging of the Distributions of Cholesterol, Sphingolipids, and Specific Proteins in the Plasma Membrane with Secondary Ion Mass Spectrometry","","2397 doi:10.1017/S1431927615012763 Paper No. 1197 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 High-Resolution Imaging of the Distributions of Cholesterol, Sphingolipids, and Specific Proteins in the Plasma Membrane with Secondary Ion Mass Spectrometry Mary L. Kraft1, Peter K. Weber,2 Jessica F. Frisz1 , Haley A. Klitzing,1 Robert Wilson,1 Ashley Yeager,1 and Joshua Zimmerberg3 1. School of Chemical Sciences, University of Illinois, 600 S. Mathews Avenue, Urbana, IL 61801, USA. 2. Glenn T. Seaborg Institute, Lawrence Livermore National Laboratory, 7000 East Ave., Livermore CA 94551, USA. 3 Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD 20892, USA. The plasma membrane is a selectively permeable lipid bilayer that separates cells from their surroundings. Numerous different lipid species, cholesterol, and a variety of different proteins form the plasma membranes of mammalian cells. One class of lipids, the sphingolipids, and their metabolites serve both as structural components in the plasma membranes of mammalian cells, and as bioactive signaling molecules that modulate fundamental cellular processes. Though segregation of the sphingolipids into distinct membrane domains is likely essential for cellular function, the sphingolipid distribution within the plasma membrane and the mechanisms that regulate it are poorly understood. To overcome these disadvantages, we pioneered the use of high-resolution SIMS, performed with a Cameca NanoSIMS 50, for directly imaging metabolically incorporated, stable isotope-labeled lipids in actual cell membranes. The NanoSIMS 50 is a state-of-the-art magnetic sector secondary ion mass spectrometer that can image the elemental and isotopic composition at the surface (top 5 nm) of a sample with high sensitivity and as high as 50-nm-lateral resolution. In order to visualize the cholesterol and sphingolipids in the plasma membrane, we used metabolic labeling with 15N-sphingolipid precursors (15N-sphingosine and 15N-sphinganine) and 18O-cholesterol to selectively incorporate distinct stable isotopes, 15N and 18O, into the sphingolipids and cholesterol, respectively, in living mouse fibroblast cells. Then we chemically fixed the cells with a method that does not alter membrane lipid distribution [1,2]. Well-preserved cells with normal morphologies are identified by imaging with low voltage secondary electron microscopy (SEM). To increase secondary ion yields, the cell samples are coated with a thin (3-nm) iridium layer that does not alter the lipid distribution on the cells. Then we used a Cameca NanoSIMS 50 instrument to map the lipid-specific isotope enrichments on their surfaces better than 100-nm-lateral resolution. Using this approach, we have previously shown that the 15N-sphingolipids are enriched within distinct domains in the plasma membranes of fibroblast cells [2]. Here we report how we have used this approach to probe the mechanisms responsible for this sphingolipid organization. We investigated whether the sphingolipid domains are dependent on cholesterol-sphingolipid interactions, or the diffusion barriers that are established by the cytoskeleton and its associated membrane proteins by using SIMS to image the effects of cholesterol depletion and actin depolymerisation on 15N-sphingolipid distribution in the plasma membrane [2,3]. We also assessed whether these 15N-sphingolipid domains are co-localized with hemagglutinin, a specific membrane protein that is thought to have an affinity for sphingolipid-enriched membrane domains. Our results indicate that the sphingolipid organizations in Microsc. Microanal. 21 (Suppl 3), 2015 2398 the plasma membranes of fibroblasts are dependent on the cytoskeleton, but not on favorable interactions with cholesterol or hemagglutinin. References [1] HA Klitzing., et al., Secondary Ion Mass Spectrometry Imaging of Biological Membranes at High Spatial Resolution. In Methods in Molecular Biology: Nanoimaging Methods and Protocols, Sousa, A. A., and Kruhlak, M. J., (eds.) Humana Press, Totowa, New Jersey, (2013), Vol. 950, pp 483. [2] JF Frisz, et al., Proc. Nat. Acad. Sci. U.S.A. (2013) 110 (8), E613. [3] JF Frisz, et al., J. Biol. Chem. (2013) 288 (23), 16855. [4] The authors acknowledge funding from the Burroughs Wellcome Fund, the Intramural Program of the NICHD, NIH, the NIH Training Program in the Chemistry-Biology Interface T32 GM070421, the NSF under CHE�1058809, and Lab Directed Research and Development funding to LLNL. Dr. Kaiyan Lou is thanked for the synthesis of the 15N-sphingolipid precursors and 18O-cholesterol. Work at LLNL was supported by Lab Directed Research and Development funding and performed under the auspices of the U.S. DOE.");sQ1[1198]=new Array("../7337/2399.pdf","Standardization on Particle Size Distribution Measurement of Nanomaterials","","2399 doi:10.1017/S1431927615012775 Paper No. 1198 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Standardization on Particle Size Distribution Measurement of Nanomaterials Kazuhiro Yamamoto National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba, Japan Industrial applications of nanomaterials have recently been reported in many fields. Regulation of nanomaterials is being discussed by the Organization for Economic Co-operation and Development (OECD). The European Union (EU) announced their definition of nanomaterial in 2012. According to the EU definition, nanomaterial means a natural, incidental or manufactured material containing particles, in an unbound state or as an aggregate or as agglomerate and where, for 50 % or more of the particles in the number size distribution, one or more external dimensions is in the size range 1 nm - 100 nm. They also announced that the particle size of the primary particles in agglomerates or aggregates should be those that are considered. For regulatory purposes, it is necessary to measure the size distribution of nanoparticles based on the particle number concentration. The European Food Safety Authority (EFSA) recommended using at least two different analytical methods to identify nanomaterials for the EU regulation, one of which should be electron microscopy [1]. Transmission electron microscope (TEM) is a most useful technique that can provide precise information on the shape and size of the primary nanoparticles. Standardization on particle size measurement is performed Technical Committee (TC) 229 of International Organization for Standardization (ISO). Scope of ISO/TC229 is standardization in the field of nanotechnologies. In this study, standardization of particle size distribution measurement of nanomaterials by using TEM in ISO/TC229 is introduced. Interlaboratory comparison (ILC) tests of particle size distribution measurement on mono-dispersed nanoparticles, nanorods, and aggregates were examined in ISO/TC229. Gold nanoparticle was selected as a model material of mono-dispersed case. Maximum and minimum Feret diameter, and equivalent circle diameter (ECD) calculated from area of TEM image were measured. Gold nanorod was model material of anisotropic shaped material. Shaped factor and ECD were measured. In case of the bimodal size distribution, mixture of two kinds of silica nanoparticles was examined. Most of industrial nanoparticles are aggregated. Carbon black (CB) and titanium dioxide (TiO2) were selected as model materials of aggregated cases. TEM sample preparation is very important because nanoparticles are easy to aggregate in preparation. In this ILC test, chairpeople of each ILC prepare TEM specimens and distribute to participants of ILC. These ILC tests are now in progress, and data gathering by each chairperson is performing. In this presentation, some results of ILC by authors are shown. In ILC test of gold nanorod, two kinds of specimen are examined. Figure 1(a) and (b) shows TEM images of two gold nanorod specimens, named #1 and #2. Two specimens look like similar, and the difference of two is not clear. Size analysis of these gold nanorod is examined statistically. TEM images are recorded by using slow scan CCD camera at the resolution of 0.154nm/pixel. Area of projection images of each nanorods, maximum, and minimum Feret diameter are measured by using Image-J software. Equivalent circular diameter (ECD) is calculated from projection area, and figures 2(a) and (b) show histogram and cumulativetive distribution of ECD, respectively. Number of counted particles is 753 for specimen #1, and 840 for specimen #2. ECD distributions are in ranging 20 nm and 43 nm. Median diameter (d50) is 36.5nm for specimen #1, and 37.5 nm for specimen #2. Shape factor is defined as the ratio of maximum and minimum Feret diameters, and figure 3 shows cumulative distribution of shape factor. Shape distribution of specimen #2 Microsc. Microanal. 21 (Suppl 3), 2015 2400 is wider than that of specimen #1. Optical functions of gold nanorod are strongly depends on their size and shape, therefore the evaluation and standardization to measure size and shape is important. References [1] E.A.J. Bleeker et al, RIVM Letter report 601358001(2012). Figure 1. Typical TEM images of gold nanorod. Specimen #1 (a), and specimen #2 (b). Figure 2. Equivalent circular diameter distribution of gold nanorod specimens (a), and cumulative distribution (b). Figure 3. Cumulative distribution of shape factor.");sQ1[1199]=new Array("../7337/2401.pdf","Shape Recognition of Nanoparticles by High-Resolution SEM and TSEM","","2401 doi:10.1017/S1431927615012787 Paper No. 1199 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Shape Recognition of Nanoparticles by High-Resolution SEM and TSEM E. Ortel1, L. Pelluti�2, F. Pellegrino2, V. Maurino2, J. Mielke1, B. Powierza1, I. H�usler1, W. �sterle1 and V.-D. Hodoroaba1 1 2 BAM Federal Institute for Materials Research and Testing, Berlin, Germany. Department of Chemistry, University of Turin, Turin, Italy. Engineered nanoparticles (NPs) are used as an active material in sensors, photovoltaics, photocatalysis, etc. Numerous publications have shown that particular facets of NPs dramatically influence their performance, e.g. in photocatalytic reactions with TiO2 NPs [1, 2]. Therefore, information about the NP morphology expressed as area ratios of particular facets is highly demanded for the development of advanced nanomaterials. The accurate determination of the particle size distribution for spherical NPs is a task resolved rather easily by various techniques. However, measuring the morphology of individual NPs having complex 3D geometries like cubes, prisms or (bi)pyramids is challenging. Often, only time-consuming TEM and TEM-tomography experiments can resolve the 3D structure and facets of particles in the nanometer range accurately. We present new approaches based on i) top-view high resolution SEM and ii) in-depth view, transmission SEM (TSEM) for the determination of the full shape of facet-controlled NPs. From top-view high resolution SEM (approach i) we could identify the 3D geometry and the individual facet boundaries of NPs. A computer-generated 3D skeleton was manually matched to these facet boundaries, thereby yielding the necessary parameters for a full 3D description of the NP shape. The evaluation of the NP shape from TSEM (approach ii) relies on an automatic image analysis: The 2D projection of the particles on the image plane depends strongly on their orientation with respect to the incident electron beam. Taking only those particles into account for which the 2D projection coincides with an assumed silhouette for standing or lying NPs on the substrate, the size of the corresponding NP facet can be determined. Due to automated image analysis, statistically relevant amounts of data could be generated quickly and lead to reliable estimates of the NPs facet size. These general procedures to determine the NP shape is demonstrated on truncated bipyramidal TiO2 anatase NPs. The TiO2 NPs were synthesized by a ligand-assisted synthesis route, which provides a scalable model system with tuneable TiO2 NP geometries [1, 2]. Figure 1 presents top-view SEM micrographs of TiO2 anatase NPs. The truncated bipyramidal shape is clearly recognizable in Figure 1A. An enlarged part of A shows a single NP where individual facet boundaries can be identified (Figure 1B). A computer-generated 3D skeleton was matched to the facet boundaries (Figure 1C) which results in a full 3D description of the NP shape (Figure 1D). Figure 2 consists of TSEM micrographs of the truncated bipyramidal TiO2 NPs whereas A) shows the original micrograph and B) the same micrograph with NPs considered for the evaluation of the NPs shape. The NPs marked in red and yellow were identified as standing upright and flat-lying NPs on the support, respectively (Figure 2B). Hence the corresponding NPs lengths and widths could be measured from the TSEM micrograph. In order to validate the new methods and estimate the associated measurement uncertainties, TSEM images of randomly oriented bipyramidal particles with known dimensions were simulated and evaluated in the same way as the measured images. A good agreement was obtained between the known parameters and the evaluation results. Furthermore, for selected NPs, the obtained dimensions were confirmed by HRTEM measurements.[3] Microsc. Microanal. 21 (Suppl 3), 2015 2402 References: [1] C. Deiana et al., Phys. Chem. Chem. Phys. 15 (2013), p. 307. [2] C. Li et al., J. Am. Chem. Soc. (2015), DOI: 10.1021/ja5111078. [3] The research leading to these results has received funding from the European Union's Seventh Framework Programme (FP7/2007-2013) under grant agreements n�604577 (SETNanoMetro) and n�604347 (NanoDefine). A) B) C D 7nm 57nm 35 nm 30 nm Figure 1. Top-view SEM micrographs of truncated bipyramidal TiO2 NPs, A) overview, B) zoom-in on a single NP where individual facet boundaries can be identified, C) computer-generated 3D skeleton matched to facet boundaries D) full 3D description of the NP. A B 100 nm 100 nm Figure 2. TSEM micrograph of truncated bipyramidal TiO2 NPs, A) original micrograph, B) NPs considered for the evaluation of NPs widths (red boundaries) and lengths (yellow boundaries).");sQ1[1200]=new Array("../7337/2403.pdf","Assessment of Different Electron Microscopy Techniques for Particle Size Quantification of Potential Nanomaterials","","2403 doi:10.1017/S1431927615012799 Paper No. 1200 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Assessment of Different Electron Microscopy Techniques for Particle Size Quantification of Potential Nanomaterials Philipp M�ller1, Johannes Mielke2, Vasile-Dan Hodoroaba2, Ralf K�gi3 and Martin Ryner4 1. 2. BASF SE, Department of Material Physics and Analytics, Ludwigshafen, Germany. BAM Federal Institute for Materials Research and Testing, Division 6.8 Surface Analysis and Interfacial Chemistry, Berlin, Germany. 3. EAWAG aquatic research, D�bendorf, Switzerland. 4. Vironova AB, Stockholm, Sweden. While nano-scaled intermediate and consumer products are omnipresent in many industries, one challenge consists in the development of methods that reliably identify, characterize and quantify nanomaterials both as a substance and in various matrices. For product registration purposes, the European Commission proposed a definition for nanomaterial [1] which requires a quantitative size determination of the primary particles in a sample down to sizes of 1 nm. According to [1] a material is defined as nano if 50% of the primary particles are observed to comprise a smallest dimension <100 nm. The NanoDefine project [7] was set up to develop and validate a robust, readily implementable and costeffective measurement approach to obtain a quantitative particle size distribution and to distinguish between nano and non-nano materials according to the definition [1]. Among the available particle sizing techniques, electron microscopy was found to be one option meeting most of the requirements of the regulation [2-4]. However, the use of electron microscopy for particle sizing is often limited by cost per sample, availability in industry, particle agglomeration/aggregation, extremely broad size distributions, 2D materials and operator bias in case of manual evaluation. In the present study a matrix of substances carefully chosen to cover a maximum of industrial applications, shapes, material systems and size ranges were analyzed by electron microscopy in an interlaboratory round robin exercise. Different electron microscopic techniques like conventional highvoltage Transmission Electron Microscopy (TEM), (Transmission) Scanning Electron Microscopy (SEM and TSEM) as well as tabletop SEM and a low-voltage tabletop TEM instrument (MiniTEM) [5] were evaluated. Sample preparation for electron microscopy was standardized according to dispersion protocols which were developed and tested separately for any sample within the labs of NanoDefine [7] by light scattering techniques and TEM images of undispersed material. Figure 1 shows a comparison of electron microscopy images of a kaolin clay sample acquired by a) TEM, b) SEM and c) MiniTEM. In the present study `real' world materials were included. Full dispersion was not always possible. Objective automated evaluation of the particle size distribution is therefore very difficult. For some of the samples a good dispersion of the primary particles was achieved. The size distribution of those samples was determined by image evaluation (e.g. iTEM software applying watershed algorithm [6] or custom code of Vironova [5] included in the MiniTEM operation system) and the results were compared between the authors labs and between the different electron microscopy image modalities. Figure 2 a) shows a MiniTEM image of an iron pigment consisting of agglomerated primary particles which challenges the possibility to unambiguously identify individual particles for size and shape quantification. Figure 2 b) gives an example of well dispersed polystyrene spheres which were evaluated by automated quantitative image analysis [6], classifying the particles into different size categories (different colors in Figure 2 b)). The present study gives a comprehensive overview on recent development in the field of particle sizing by electron microscopy. Industry can benefit from advances in automated imaging and image evaluation Microsc. Microanal. 21 (Suppl 3), 2015 2404 as well as from new types of tabletop SEM and TEM solutions reducing some of the limitations of electron microscopy like cost, availability and operator bias. [7] References: [1] European Commission, Commission Recommendation of 18 October 2011 on the definition of nanomaterial, Official Journal of the European Union. 2011/696/EU (2011) p.38. [2] TPJ Linsinger et al, JRC Reference Reports, 2012, DOI: 10.2787/63490 [3] PJ De Temmerman et al, J. Nanopart. Res. 16 (2014), p.2177. [4] PJ De Temmerman et al, Powder Technology 261 (2014), p.191. [5] http://www.vironova.com/minitem [6] iTEM 5.2 (Build 3554),Olympus (Tokyo, Japan) [7] The research leading to these results has received funding from the European Union's Seventh Framework Programme (FP7/2007-2013) under grant agreement n� 604347 � NanoDefine (www.nanodefine.eu). Figure 1. Electron microscopy images of kaolin clay. a) conventional TEM; b) conventional SEM; c) miniTEM Figure 2. Examples for automated image evaluation. a) MiniTEM image of iron pigment; primary particles were not resolved and b) TEM image of polystyrene spheres separated by watershed algorithm");sQ1[1201]=new Array("../7337/2405.pdf","Three Dimensional Accurate Morphology Measurements of Polystyrene Standard Particles on Silicon Substrate by Electron Tomography","","2405 doi:10.1017/S1431927615012805 Paper No. 1201 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Three Dimensional Accurate Morphology Measurements of Polystyrene Standard Particles on Silicon Substrate by Electron Tomography Misa Hayashida1, Kazuhiro Kumagai2, Marek Malac1, 3 1 2 National Institute for Nanotechnology (NINT), Edmonton, Alberta, Canada National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba, Ibaraki, Japan 3 Department of Physics, University of Alberta, Edmonton, Alberta, Canada Polystyrene latex (PSL) nanoprticles (NPs) spheres are frequently used as size calibration standard over the 30 nm to 250 nm range. Their shape is assumed to be spherical. However, atomic force microscopy (AFM) results indicate difference between lateral and vertical dimensions [1]. Lateral dimensions (in the plane of the supporting substrate) are typically larger than the vertical (perpendicular to the substrate plane) dimensions. It is therefore necessary to accurately measure the shape of the PSL nanoparticles. Electron tomography in a (scanning) transmission electron microscope (S)TEM allows to accurately measure three dimensional (3D) shape of nanoscale objects. A water suspension of PSL nanoparticles (SC-0100-D, JSR Life Science) was deposited onto a silicon substrate, and allowed to dry overnight. To enhance the contrast of NPs in TEM image, they were coated with ~20 nm-thick osmium film by plasma-enhanced chemical vapor deposition, followed by ~200 nm thick sputtered carbon film. Prior to cutting process by focused ion beam (FIB), the area around target NPs was covered with additional thick carbon layer (~1.5 �m) to prevent damage in the cutting process. A rod-shaped sample with a single PSL NP was fabricated in the FIB to allow data collection without a missing-wedge eliminating the distortions arising from the missing wedge. Nano dot fiducial markers were deposited for high accuracy alignment using electron beam induced deposition of tungsten [2]. A Hitachi HF-3300 TEM/STEM equipped with cold field emission gun was used to collect tomographic tilt series in annular dark field mode (ADF). For the data acquisition, a MatlabTM based computer control system (MAESTRO) designed for Hitachi electron microscopes was used to control the data acquisition process [3]. At each tilt 512�1024 pix images with 5 s /pixel dwell time and about about 7 pA probe current. The tilt step was 3� over the entire �90� tilt range resulting in 61 images. Four images were acquired at each tilt for a total of 20 �s per pixel dwell time to decrease effect of sample drift while keeping acceptable signal to noise ratio. The images at each tilt were summed after compensating for drift using cross correlation alignment. Filtered back projection was used to reconstruct the three dimensional volume representation of the nanoparticle. Figure 1a shows a projected STEM image of a nominal 100 nm diameter PSL NP embedded in the rodshaped specimen. The bright contrast of the osmium layer, which accurately traces the surface of PSL NP enables us to accurately evaluate the shape the nanoparticle. Figures 1b) to d) show cross-sectional images along three perpendicular planes extracted from the reconstructed volume. The NP cross sectional shape can be fitted to a circle in a plane perpendicular to the sample rod Y axis. Cross sectional shape in planes parallel to the sample rod Y axis can be fitted to an ellipse. The shape is conveniently obtained by evaluating diameter of a circle fitted to sections perpendicular to Y axis at known distances from the silicon substrate. The method yields a 93.9 nm diameter and particle reaches Microsc. Microanal. 21 (Suppl 3), 2015 2406 85.5 nm above the substrate in Y direction. Therefore the particle is squished by 9 % in the Y direction becoming an ellipsoid with the short axis in the Y direction. The measured squish ratio agrees with the AFM measurements of nanoparticle assemblies [1]. Unlike the AFM measurements where the particle shape is inferred from periodicity of nanoparticle assembly on a substrate, here we report direct quantitative measurement of individual nanoparticle shape. The sub 1 nm resolution obtained here was afforded by the high accuracy alignment [2], high contrast osmium coating and low irradiation dose used for the data collection. The entire process that leads to the nanoparticle dimension measurement can be quantified in terms of accuracy of the individual steps in a traceable manner. [1] I. Misumi, et. al., SPIE Proceedings, Vol. 8378 (2012), 83780J. [2] M. Hayashida, M. Malac, M. Bergen, P. Li, Ultramicroscopy, Vol. 144, (2014), 50-57. [3] M. Bergen et. al. Microscopy and Microanalysis 19 (S2) (2013), 1394-1395. Ms. M. Ito and Dr. H. Sakurai (AIST, Japan) performed nanoparticle sample preparation. Financial support of Visiting Researcher grant from Alberta Innovates Technology Futures is gratefully acknowledged. This work was done when the first author worked at her earlier affiliation, AIST. Figure 1. A three dimensional 3D observation of PSL NP with 100 nm diameter which is included in rod shaped specimen. (a) A selected projection ADF STEM image from the tilt series. The sample consists of (from top) amorphous carbon layer, osmium (bright) on PSL NP. The fiducial nano dot markers are at the lower portion of the left most panel in a). The bright horizontal line between the NP and the silicon substrate is the osmium layer. The coordinate axes are indicated by arrows. The siliconnanoparticle interface is in the y plane perpendicular to the Y axis, which is parallel to the long axis of the sample. (b), (c) and (d) are example X-plane, Z-plane and Y-plane slices through the reconstructed volume. The dashed curve in (b) to (d) indicates the nanoparticle-osmium boundary that was fitted by an ellipsoid to obtain the nanoparticle 3D shape. The solid line indicates the position of the silicon substrate.");sQ1[1202]=new Array("../7337/2407.pdf","Repeatable and Transferable Processing for Electron Tomography: An Open Platform for Visualization and Reconstruction of 3D Materials","","2407 doi:10.1017/S1431927615012817 Paper No. 1202 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Repeatable and Transferable Processing for Electron Tomography: An Open Platform for Visualization and Reconstruction of 3D Materials Robert Hovden1, Marcus D. Hanwell2, Utkarsh Ayachit2, Yi Jiang3, Robert Maynard2, and David A. Muller1 1 2 School of Applied & Engineering Physics, Cornell University, Ithaca NY, USA Kitware, Inc., Clifton Park, NY, USA 3 Department of Physics, Cornell University, Ithaca, NY, USA Three-dimensional (3D) characterization of materials at the nano- and meso-scale has become possible with transmission and scanning transmission electron tomography[1-2]. This process requires advanced software tools where the final 3D visualization is critically dependent on the reconstruction algorithm and the parameters used to render the 3D image. Exacerbating this variability, electron tomography lacks the standards and tools for repeatable, transferable scientific analysis--the field requires open data formats, reconstruction algorithms, 3D visualization, and most importantly a way to share all processing steps from start to finish. To address this problem, we have developed an open-source platform--tomviz--for the tomographic reconstruction, analysis, and 3D visualization of materials. With a modern graphical interface, tomviz dramatically reduces the barrier of entry to materials tomography in research labs and universities. It is a transparent solution that avoids licensing fees and restrictions on redistribution--allowing researchers at user facilities to processes data off-site. tomviz can utilize the large quantities of memory and processing resources required to reconstruct, render, manipulate, and analyze voluminous 3D tomograms. The platform provides a robust graphical interface where objects can be rendered as shaded contours or volumetric projections (Figure 1). Multiple datasets, colormaps, and other visualization settings can be used in combination and these objects can be rotated, sliced, animated, and saved as image or video files. 3D data can be further analyzed through histograms, Fourier transforms, and filters--to name a few. The platform is open source, meaning that novel tomographic algorithms or mathematical operations can be readily implemented through its Python (with Numpy) scripting interface or the core C++ application (Figure 2). With tomviz, the full pipeline of data processing steps from reconstruction to visualization and analysis of 3D data can be presented, saved, and restored. This enables fully reproducible results for interlaboratory comparison--critical for fields where researchers share data with colleagues, or raise queries with experts. These pipeline files aid the publication of results, where open scientific methods and peer review are applied to S/TEM materials characterization. tomviz promotes the open Electron Microscopy Dataset specification[3] for large 3D dataset storage. Reproducible and shareable tomography analysis is necessary for the openness of science and meeting future open-data mandates. tomviz will address a multitude of key issues outlined in the DOE 2013 Data Crosscutting Requirements Review and pursuant to the Executive Order of May 9, 2013, Making Open and Machine Readable the New Default for Government Information. tomviz is publically available for download at www.tomviz.org [1] De Rosier, D. and Klug, A. Nature 217, 130-134 (1968) [2] Midgley, P.A. et al., Chemical Communications, 10, 907-908 (2001) [3] An Open File Format for Microscopy Data, Based on HDF5, http://emdatasets.lbl.gov/spec/ Microsc. Microanal. 21 (Suppl 3), 2015 2408 Figure 1 | tomviz 3D visualization of a hyper-branched CoP nanocrystal. An isosurface contours show the particle surface morphology with 2D slices through the particle highlight its high density core region. Volume rendering also shows morphology and changes in particle density. Figure 2 | tomviz is a graphical platform for tomographic reconstruction and visualization. Reconstruction methods in tomviz are run with a single click; no coding is required. After execution, data is transformed using generated Python code. This code is accessible and can be modified by the user to expand functionality. Through the Python and NumPy interface, custom image processing methods can implemented. The entire analysis and viz pipeline can be saved, restored, and shared.");sQ1[1203]=new Array("../7337/2409.pdf","Revealing New Atomic-scale Information about Materials by Improving the Quality and Quantifiability of Aberration-corrected STEM Data","","2409 doi:10.1017/S1431927615012829 Paper No. 1203 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Revealing New Atomic-scale Information about Materials by Improving the Quality and Quantifiability of Aberration-corrected STEM Data Andrew B. Yankovich1, Jie Feng1, Alex Kvit1, Thomas Slater2, Sarah Haigh2, Dane Morgan1, and Paul M. Voyles1 1 2 Dept. of Materials Science and Engineering, University of Wisconsin, Madison, WI 53706, USA. School of Materials, University of Manchester, Manchester, M13 9PL, UK. Recent advances in aberration-corrected scanning transmission electron microscopy (AC-STEM) have shown the world previously unattainable views into the atomic structure and composition of materials. Thanks to improved optics, experimental and environmental factors often limit the quality of information accessed by AC-STEM. However, collecting and processing AC-STEM data using new techniques from data science can help overcome these limitations, opening the door to new atomic-scale materials information by improving our ability to determine atomic column positions, measure 3D structures, detect single point defects, and determine the atomic-scale composition of materials. Typically for identifying atomic column positions in STEM images, practical limits are encountered before the fundamental signal to noise ratio (SNR) limit. The most common practical limit is distortions caused by instabilities during image acquisition. Recently, we developed a non-rigid registration (NRR) technique that overcomes this practical limit by correcting all types of image distortions present in STEM images [1][2]. Other methods have also been developed to address this issue [3]. The NRR and averaging of high angle annular dark field (HAADF) STEM image series greatly increases the image SNR and allows for sub-pm precision of beam insensitive single crystal materials. NRR and averaging HAADF STEM image series of Pt nanocatalysts have allowed measurements of pm-scale surface atom bond length variation that may help explain their catalytic activity. The atomic structure of a Au nanoparticle was measured with 1-2 pm precision using NRR at 1% of the dose of the Pt results [4], showing the high-precision NRR technique can be used on more beam sensitive material systems. Extracting 3D information from 2D TEM/STEM images is a long-standing problem in electron microscopy. Standardless atom counting allows for 3D atomic structure information to be measured from HAADF STEM and HRTEM images by comparing experimental and simulated image intensities [5][6]. Previously, the uncertainty in this measurement was limited by image Poisson noise. NRR and averaging allows for standardless atom counting with the uncertainty no longer dominated by Poisson noise [2]. The unique determination of the number of atoms in atomic columns is now possible, although not yet demonstrated due to other experimental limitations, such as sample preparation artifacts. Point defects are critical to the properties of a wide range of materials, including semiconductors and solid oxides used in fuel cells. Imaging substitutional and interstitial dopant impurity atoms is possible using HAADF STEM [7], as well as determining the depth of the dopant in the TEM specimen [8]. However, experimentally imaging single atomic vacancies remains a challenge. We have conducted extensive frozen phonon multislice simulations that predict the visibility of La vacancies in LaMnO3, both by the reduced atomic column intensity and by atomic column position distortions around the vacancy. NRR and averaging HAADF STEM images of LaMnO3 improves the SNR and the image precision sufficiently to potentially experimentally detect single La vacancies. Figure 1(a) shows a NRR and averaged HAADF STEM image of a LaMnO3 film. Figure 1(b) shows the visibility of each atomic Microsc. Microanal. 21 (Suppl 3), 2015 2410 column relative to its nearest 8 atomic columns. The atom column in (b) marked by the green box has a ~7 % visibility, matching simulated predictions of single La vacancies, making it a candidate single vacancy. Atomic-scale composition can be determined using STEM EDS spectrum imaging (SI), however due to a low efficiency of producing and collecting x-ray signals, long total dwell times that introduce spatial distortions are required. Currently, to minimize the distortions, multiple SIs are acquired using online drift-correction software, then summed, while the individual SIs and simultaneous HAADF images are discarded. Figure 2 shows that the quality of EDS SIs is improved by saving the simultaneously acquired raw HAADF and EDS SI series of Nd2/3TiO3 [9] and applying NRR. Figure 2 shows Nd and Ti elemental maps extracted from a NRR SI, standard drift correction SI, and a single long-dwell time SI. NRR of the EDS SI produces elemental maps with less spatial distortions, more atomic localization of x-ray signals, and higher contrast. Quantitative comparison between these techniques and additional methods of denoising EDS SIs will be discussed. [10] References: [1] Berkels et al. Ultramicroscopy, 138, 46 (2013). [2] Yankovich et al. Nature Communications, 5, 4155 (2014). [3] Sang et al. Ultramicroscopy, 138, 28-35 (2014). [4] Yankovich et al. Advanced Structural and Chemical Imaging, accepted (2015). [5] LeBeau et al. Nano Letters, 10, 4405 (2010). [6] Jia et al. Nature Materials, 13, 1044 (2014). [7] Voyles et al. Nature, 416, 826 (2002). [8] Hwang et al. PRL, 111, no. 26, 266101 (2013). [9] Azough et al. Chemistry of Materials, 27, 2, 497 (2015). Figure 1: (a) NRR and averaged [10] Work was funded HAADF STEM image of [100] by the Department of LaMnO3 grown on a DyScO3. Scale Energy, Basic Energy bar is 0.5 nm. (b) Visibility map of Sciences (DE-FG02each atomic column in (a). 08ER46547). Figure 2: (a) NRR and averaged HAADF STEM image of Nd2/3TiO3. (b)-(g) EDS elemental maps using the Nd L peak (b)(d)(f) and the Ti K peak (c)(e)(g). (b)&(c) Maps from a sum of 98 SIs after NRR. (d)&(e) Maps from a single long exposure SI. (f)&(g) Maps from a sum of 100 drift-corrected SIs. The total counts in each of the 3 SI methods are approximately constant. No images have been smoothed. Scale bar is 2.5 nm.");sQ1[1204]=new Array("../7337/2411.pdf","Opportunities in Angularly Resolved Dark-field STEM using Pixelated Detectors.","","2411 doi:10.1017/S1431927615012830 Paper No. 1204 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Opportunities in Angularly Resolved Dark-field STEM using Pixelated Detectors. Lewys Jones1, Hao Yang1, Katherine E. MacArthur1, Henning Ryll2, Martin Simson3, Heike Soltau3, Yukihito Kondo4, Ryusuke Sagawa4, Hiroyuki Banba4, and Peter D. Nellist1. 1. 2. Department of Materials, University of Oxford, Oxford, UK PNSensor GmbH, Otto-Hahn-Ring 6, 81739 M�nchen, Germany 3. PNDetector GmbH, Sckellstra�e 3, 81667 M�nchen, Germany 4. JEOL Ltd.,3-1-2 Musashino Akishima Tokyo 196-8558 Japan Pixelated sensors in the detector plane of the scanning transmission electron microscope (STEM) offer many opportunities for extracting valuable information from the bright-field (BF) disk including synthesising annular bright-field images, differential phase-contrast images, and ptychographic phase reconstruction [1]. However, by reducing microscope camera-length, these pixelated detectors can also be used to record the scattered electron flux outside the BF disk. This scattering has previously been recorded using annular dark-field (ADF) detectors however there the operator must choose what annular range the DF signal is to be integrated over at the point of the imaging and this can limit operational flexibility. Some instruments offer two or more annular DF detectors to record more than one angular range but in this case the ratio of collection angles of these two detectors is fixed. Use of two detectors allows low-angle or medium-angle (LAADF and MAADF) images to be recorded which have been found to yield useful strain or structural information [2], [3]. Here we demonstrate the first lattice resolution ADF data recorded using a pixelated detector. We show how various types of DF images can be synthesised including LAADF, MAADF and high-angle (HAADF) images. Moreover, the angularly resolved data recording allows for a shift from a situation where a user must choose from one of a limited set of available detector collection angles, to a regime where the required detector conditions can be chosen after the experiment. Figure 1 shows the readout from a pnCCD (S)TEM camera, a direct electron pixelated detector from PNDetector (264x264 pixels in this case), mounted on a probe corrected JEOL ARM200-CF. Owing to the massive difference in intensity inside and outside of the BF disk a beam-blanker was used to block the BF disk. Figure 1 also shows an enlargement showing the detection of individual electron scattering events at high angles meaning this approach also offers a route to direct quantification of the ADF signal without the need to normalise for detector sensitivity [4]. An ADF image was first focussed using a conventional detector with a PtCo nanoparticle as a test object (Figure 2). After focussing, the physical ADF detector was retracted and a 256x256 real-space probe array dataset recorded at a rate of 1000 detector plane frames per second (total scan time 66s). After the experiment, the imaging mode was selected in post-processing. First to view the overall data-quality the whole ADF signal was integrated (Figure 2, top). This data confirms that the magnification and resolution were sufficient to achieve atomic resolution, and that the scanning stability was sufficient at the relatively long 1ms dwell-time. This image also shows the position of a twin-boundary on the left side of the particle, indicated by the pair of red arrows. Using the 4D data-set scattering ranges were selected for integration including LAADF (25.8-26.7 mrad) and MAADF (30.1-32.7 mrad), Figure 2 bottom. In these plots the area to the left of the twin-boundary appears first light, then dark respectively. Microsc. Microanal. 21 (Suppl 3), 2015 2412 In conclusion, we have shown that use of a pixelated detector in the diffraction plane of the STEM allows for recording of a full 4D scattering data-set. Post-processing of this data allows for various imaging modes to be synthesised after the experiment itself. Tuning of the annular ranges of the virtual detectors allows specific contrast to be extracted around crystallographic, composition or strain features that may be washed out by larger physical ADF detectors [5]. References: [1] T. J. Pennycook et al, Ultramicroscopy in-press (2014). [2] S. Hillyard and J. Silcox, Ultramicroscopy 58 (1995), p. 6-17. [3] G. Zhu et al., Appl. Phys. Lett. 105 (2014), p. 231607. [4] J. M. Lebeau and S. Stemmer, Ultramicroscopy 108 (2008), p. 1653�8. [5] This work was supported by the EU grant 312483 - ESTEEM2 and EPSRC grant EP/K040375/1. The authors thank Johnson Matthey for the PtCo sample and for financially supporting KEM. Figure 1. pnCCD (S)TEM camera readout for an example real-space probe position. A beamstopper was used to mask the intense BF-disk. Enlargement shows the sensitivity of the detector to individual electron scattering events. Figure 2. Example images synthesised from the 4D dataset. Top panel shows the integrated ADF signal. Arrows indicate the position of a twinboundary. Lower panels show synthetic low-angle and medium-angle ADF images where the particle edge region appears bright and dark respectively.");sQ1[1205]=new Array("../7337/2413.pdf","Increased Fluctuation of Interatomic Distances in Distorted Structure of Stoichiometric LaMnO3.","","2413 doi:10.1017/S1431927615012842 Paper No. 1205 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Increased Fluctuation of Interatomic Distances in Distorted Structure of Stoichiometric LaMnO3. Alexander V. Kvit1,3, Jie Feng2, Andrew B. Yankovich1,2, Dane Morgan1,2 and Paul M. Voyles1,2 1. Department of Materials Science and Engineering, University of Wisconsin-Madison, Madison, WI, 53706, USA 2. Materials Science Program, University of Wisconsin-Madison, Madison, WI, 53706, USA 3. Materials Science Center, University of Wisconsin-Madison, Madison, WI, 53706, USA Non-rigid registration of aberration- corrected scanning transmission electron microscope (STEM) images enables sub-picometer precision in locating atomic column positions under favorable circumstances [1]. Reproducible high precision requires careful alignment of the microscope and selection of the acquisition parameters. Sample mistilt causes systematic variations in interatomic distances, but can be minimized using position-averaged convergent beam electron diffraction (PACBED) to <1 mrad. Low sample drift of 1 �/min results in larger field of view after registration. On our microscope, a FEI Titan operated at 200 kV, dwell time of 12 �s/pixel and for 256�256 pixel images is optimal. We recently found that images of LaMnO3 (LMO) film grown on DyScO3 (DSO) substrates exhibit an unusual elastic stress relaxation microstructure. They also consistently show precision of 3- 6 pm, measured as the standard deviation of repeated interatomic distances within the image field, despite being acquired under the conditions described above. Here we present evidence that the large spread in interatomic distances is intrinsic to the sample, and that it arises from the nature of stress relaxation in the LMO film. LMO (Pnma space group) is a good lattice match to DSO, but these materials have slightly different thermal expansion coefficients, creating stress in the film due to the change in lattice mismatch with temperature. The stress does not result in interface misfit dislocations. Instead, the film contains domains rotated by 90� with respect to each other and the substrate, as shown in Figure 1a. The different domains have different internal strain. Figure 1b shows a 2.3� tilt between the epitaxial LMO film and DSO substrate; Figure 2a shows a NRR image of a tilted domain, in which the unit cell corner angles are 92� and 88� degrees, not 90�. Figure 2b shows a simulated image of undistorted [100] LMO. The distances between La- atom columns DS0 and DS1 are not equivalent due to presence of oxygen columns on opposite sides of the La columns (see Figure 2e). From simulations, DS0 = 2.7575 �, DS1 = 2.8246 �, and DS2 = 3.8751 � for 10 nm sample thickness, which is close to thickness estimation from PACBED in the area shown on Figure 2a and 2c. We can distinguish such differences from NRR HAADF image on Figure 2a. Figure 2c shows a NRR low-angle annular bright field (LAABF, 8.48 � 18.7 mrad detector angle) image from the same area where oxygen columns can be easily distinguished except those that overlap with La columns. Simulated positions of all columns are shown by circles on experimental LAABF on bottom part of Figure 2c. Figures 2a and d show obvious elongation of the Mn columns. The La columns do not show such elongation, so the distortion does not arise from probe aberrations or sample mistilt. We therefore suggest that the distortion is a result of changes in octahedral tilt related to the same elastic stress that Microsc. Microanal. 21 (Suppl 3), 2015 2414 changes the La positions. Local variations in these distortions, either randomly or as a function of distance away from the substrate explain the larger intrinsic variability of interatomic distances we observe [2]. References [1] A. B. Yankovich, et el., Nat. Commun. 5, 4155 (2014). [2] This work was supported by the Department of Energy, Basic Energy Sciences (DE-FG0208ER46547). The facilities and instrumentation for microscopy were supported by the University of Wisconsin Materials Research Science and Engineering Center (DMR-1121288). Figure 1. (a) LAABF STEM image shown domains with [100] and [001] orientations; (b) Consequence of lateral stress between DSO substrate and LMO film: distortion of LMO structure and tilt of 2.3�. The image is taken in [100] zone axis. Figure 2. (a) NRR HAADF STEM image of LMO domain in [100] zone axis; (b) Simulated HAADF STEM image; with labeled distances between La columns DS0, DS1, and DS2; (c) NRR LAABF STEM image; circles show good correlations with simulated positions of O, Mn and La columns; (d) Zoomed NRR HAADF STEM image: elongation of Mn columns; (e) Simulated position of all columns.");sQ1[1206]=new Array("../7337/2415.pdf","Maximising Information from Aberration-Corrected STEM images: Applications to Plasmonic, Semiconductor and Battery Materials.","","2415 doi:10.1017/S1431927615012854 Paper No. 1206 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Maximising Information from Aberration-Corrected STEM images: Applications to Plasmonic, Semiconductor and Battery Materials. Y Zhu1,2, H Katz-Boon2, CL Zheng1, M Walsh2, C Dwyer3, L Bourgeois1, R Withers4 and J Etheridge1,2 1. 2. Monash Centre for Electron Microscopy, Monash University, Victoria 3800, Australia Department of Materials Engineering, Monash University, Victoria 3800, Australia 3. Ernst Ruska-Centre, Peter Gruenberg Inst, Forschungszentrum Juelich, D-52425 Juelich, Germany 4. Research School of Chemistry, Australian National University, ACT 0200, Australia Aberration correction has enabled STEM images to be taken with unprecedented spatial resolution. However, to realise the full potential of these images and maximize the information we can extract about the specimen, we need to measure and quantify the experimental parameters (such as spatial and temporal coherence functions, detector response etc) and understand the electron scattering processes [13]. Armed with this information, we can then tailor the electron-optical conditions to optimize the sensitivity of the image to desired structural information, so that it can be extracted from the image in an efficient, quantitative and practical way. We illustrate this approach with applications to functional materials being developed for plasmonic, opto-electronic and battery applications, for example: The measurement of the relative stability of different nanofacets on gold nanorods versus their atomic structure and size (Fig 1). Understanding facet stability is vital for understanding growth mechanisms and catalytic behavior, however, there is no established method for measuring the stability of facets that are just a few atoms wide. Using a quantitative analysis of ADF-STEM images to count atoms at successive time intervals and in two different crystal orientations, the crystal facet orientation and stability can be determined. This approach reveals the co-existence of high and low index facets on the same nanorod and, furthermore, that these facets have comparable stability [4]. The measurement of oxygen octahedral tilt angle, unit cell by unit cell, together with the corresponding strain, in the promising Li-ion conducting perovskite LixNd1-xTiO3 (Fig 2). The nanodomain structure of this material has been the subject of intense debate: is it due to a compositional phase separation or an octahedral tilt modulation [5]? By tuning the detector geometry to optimize the sensitivity of the image to oxygen and Ti positions, the oxygen octahedra can be imaged and the tilt angle measured quantitatively, even in thick crystals <150nm [6]. This reveals directly the modulation of the octahedral tilt angle in a nanodomain formation and enables the measurement of the higher order harmonics associated with this modulation that are not otherwise accessible from diffraction methods. The measurement of Al composition in beam sensitive AlGaAs/GaAs quantum tube nanowires (Fig 3). By comparing the absolute intensity distribution in experimental ADF-STEM images with a look-up table of intensities versus Al content calculated using all the measured experimental parameters [1-3], a spatial map of composition can be determined [7,8]. This is a rapid, low dose method, compared with EDS and EELS, ensuring precise, accurate and high resolution measurements. References: [1] C Maunders, C Dwyer, P Tiemeijer, J Etheridge, Ultramicroscopy 111 (2011) 1437 [2] C Dwyer, C Maunders, CL Zheng, M Weyland, P Tiemeijer, J. Etheridge App Phys Lett 100 (2012) Microsc. Microanal. 21 (Suppl 3), 2015 2416 [3] CL Zheng, J Etheridge Ultramicroscopy 125 (2013) 49 [4] H Katz-Boon et al. Nano Letts just accepted (2015) DOI: 10.1021/acs.nanolett.5b00124 [5] R Erni, et al. Nat. Mater. 13 (2014) 216 ; B Guiton & P Davies Nat. Mater. 13 (2014) 217 [6] Y Zhu, R Withers, L Bourgeois, C Dwyer, J Etheridge, submitted (2015). [7] H Kauko, CL Zheng, Y Zhu et al. App Phys Letts 103 (2013) 232111 [8] N Jiang, Q Gao, P Parkinson, J Wong-Leung et al. Nano Lett 13 (2013) 5135 [9] Funding is acknowledged from the Australian Research Council Grants DP110104734, DP120101573, LE0454166. We thank A. Funston, P. Mulvaney and C. Jagadish for specimens. Figure 1. (Left) An ADF-STEM image of a Au nanorod crossection and corresponding "atom map" (top left), part of a time series used to determine the relative stability of crystal nanofacet orientations (right). Figure 2. (Right) Map of absolute [001]-tilt angle, unit cell by unit cell, measured from a STEM image (left) with detector geometry optimized to reveal O and Ti sublattices in Li0.38Nd0.54TiO3. Figure 3. Cross-section of GaAs/AlGaAs quantum well tube heterostructured nanowire. (Left) HAADF image and (right) corresponding Al composition map.");sQ1[1207]=new Array("../7337/2417.pdf","Direct Observation of Interfacial Au atoms Using STEM Depth Sectioning","","2417 doi:10.1017/S1431927615012866 Paper No. 1207 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Direct Observation of Interfacial Au atoms Using STEM Depth Sectioning Wenpei Gao1,2, Shankar Sivaramakrishnan1,2, Jianguo Wen2, Jian-Min Zuo1,2 1 Dept of Materials Science and Engineering, University of Illinois, Urbana-Champaign, IL 61801 2 Frederick Seitz Materials Research Laboratory, University of Illinois, Urbana-Champaign, IL 61801 Interfacial atoms located between metal nanoparticles and supports are proposed active sites in catalysis, because of their distinct physical and chemical properties [1-3]. However, the atomistic details are difficult to resolve in the interface; the lack of knowledge has been a major obstacle toward unraveling their roles in chemical transformations. Here we report the detection of interfacial Au atoms on the rutile (TiO2) (110) surfaces thanks to the improved spatial resolution and depth of focus brought by aberration corrected scanning transmission electron microscopy (STEM). Au on TiO2 is selected because it shows remarkable catalytic activity as the sizes of Au particles reduce to ~3 nm or below, for the oxidation of CO. The Au catalysts are typically prepared by Au precipitation on titania support, followed by calcination in air or reduction under H2 at elevated temperatures. Extensive study has been done concerning the mechanism of CO oxidation catalyzed by gold and the role of interfacial gold atoms. However, direct observation of interfacial Au atoms has not been reported before. A major obstacle is the TiO2 support surface, which is often highly complex, undetermined and varies at the nanometer scale. With atomic resolution images recorded at different focuses along TiO2 [001], we have reconstructed the 3D intensity profiles of interfacial atoms. The results lend to direct support to the presence of interfacial Au atoms, embedded in a single interfacial layer. The experiment started with forming epitaxial Au nanocrystals (NCs) on rutile (110) by e-beam evaporation deposition followed by annealing in air. The sizes of the Au NCs ranged from 3.5 to 12 nm in width depending upon the annealing conditions. The samples are observed by aberration corrected (AC) scanning transmission electron microscopy (STEM) using the JEOL 2200FS installed at the Center for Microscopy and Microanalysis, Frederick Seitz Materials Research Laboratory, at 200kV. The microscope is capable of resolving atoms separated by 1 �. Figure 1 shows a Z-contrast image recorded along TiO2 [001] direction. Au atoms are brighter then Ti atoms due to its larger Z. An atomic model of Au nanocrystal on TiO2 [110] surface is shown in the inset, with Ti atoms in grey, O atoms in red and Au in yellow. The arrow indicates a distinct interfacial layer between Au NCs and the TiO2 (110) surface, with Au atoms embedded inside the layer. To locate the interfacial Au atoms in 3D, a focal series of Z-contrast images were recorded first from the Au nanocrystal near the interfacial region shown in Figure 1. The images were aligned using the cross-correlation method. The intensity profiles in the interface layers were then used to form the depth-sectioning images shown in Figure 2, for the Au, TiO and interfacial layers respectively. The intensity bands were marked using the color boxes to identify the column as a Au-column, Ticolumn or an interfacial-column. Notably, at the interface, the depth-sectioning image shows intermediate intensities between these of Ti- and Au-columns. Furthermore, there are intensity variations along the focus direction in the interfacial layer. For example, the column marked as A in Figure 2 resembles the intensity of Au columns observed in the Au layer, while the columns marked as T have similar intensities as that of Ti columns in the Ti layer. Regions marked as I show distinct Microsc. Microanal. 21 (Suppl 3), 2015 2418 intensity peaks that are very different from either the TiO or Au layers. Detailed interactions between Au NCs and the TiO2 support are thus revealed from these results. References: [1] T. Akita, M. Kohyama, M. Haruta, Accounts of Chemical Research, 46(8), 1773-1782 (2013) [2] D. Widmann, R.J. Behm, Angewandte Chemie Internatinal Edition, 50(43), 10241 (2011) [3] Z. Zhou, et al., Advanced Functional Materials, 18(18), 2801-2807 (2008) [4] The work reported here is supported by the NSF Grant No. DMR 0449790 and AC-STEM was carried out in the Frederick Seitz Materials Research Laboratory Central Facilities, University of Illinois. Figure 1 Figure 2 Figure 1, a Z-contrast image recorded along TiO2 [001] direction. Au atoms are brighter then Ti atoms due to its larger Z. An atomic model of Au nanocrystal on TiO2 [110] surface is shown in the inset, with Ti atoms in grey, O atoms in red and Au in yellow. A distinct interfacial layer is observed and indicated by the arrow. Figure 2, the focal series of Z-contrast images recorded from the Au nanocrystal near the interfacial region shown in Figure 1. The images were aligned using the cross-correlation method. These intensity profiles were then used to form the depth-sectioning images shown in Figure 2, for the Au, TiO and interfacial layers respectively. The intensity bands were marked using the color boxes to identify the column as an Au-column, Ti-column or an interfacial-column.");sQ1[1208]=new Array("../7337/2419.pdf","3D Multiscale Characterization of Discontinuities in Underwater Wet Welds.","","2419 doi:10.1017/S1431927615012878 Paper No. 1208 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 3D Multiscale Characterization of Discontinuities in Underwater Wet Welds. S. Paciornik1, L. F. Silva1, V. R. dos Santos1, N. Chawla2, G. Schneider3 and T. Bernthaler3 1. 2. Chemical and Materials Engineering, PUC-Rio, Rio de Janeiro, Brazil. Materials Science and Engineering ASU, Tempe, USA. 3. Materials Research Institute, Aalen University, Aalen, Germany. Underwater wet welding (UWW) is a critical procedure for the repair of offshore structures, mainly related to oil production and transportation. The harsh environmental conditions in which the weld is performed has strong consequences to the structural reliability of the welded parts. High cooling rates due to direct contact with water and the presence of hydrogen derived from water dissociation leads to the formation of defects, such as pores and cracks in the weld metal (WM), which adversely affect mechanical properties. During cooling, weld beads contract both in transverse and longitudinal directions. It is well established that longitudinal contractions are responsible for higher residual stress after welding. Consequently, in wet welds, the low toughness associated with high hydrogen contents in the WM can lead to nucleation of cracks [1,2]. Inclusions also tend to form as the available oxygen reacts with various elements forming stable oxides that may be eliminated in the slag or get trapped within the weld metal as spherical shaped inclusions. Welding depth relates to water pressure and partial pressures of oxygen and hydrogen. Thus porosity is directly dependent on welding depth. Oxide inclusion content tends to increase with welding depth but saturates as the solubility limit of oxygen reaches its peak value. Cracks depend on the availability of diffusible hydrogen, what depends on the specific type of electrode employed. For instance, rutillic electrodes tend to produce weldments with higher diffusible hydrogen in comparison to oxidizing electrodes. However, oxidizing electrodes tend to result in higher inclusions content, which is detrimental to the mechanical properties of the weld metal [3]. UWW is an excellent test case for multiscale analysis techniques due to the wide variation in the microstructural characteristics of pores, cracks and inclusions [4,5]. Discontinuity sizes range from nm (inclusions) to hundreds of �m (pores). Shape varies from spherical inclusions to elongated wormhole shaped pores and thin cracks. Pore spatial and orientation distribution is also complex, affected by the scape of gases from the liquid weld as it rapidly solidifies. Cracks show a predominant orientation transverse to the weld axis while inclusions are randomly distributed in the WM. These characteristics also highlight the relevance of 3D techniques, as traditional 2D microscopy would not be able to reveal these complex shapes and relationships. Thus, X-ray microtomography (�CT) was employed to reveal pores and cracks while FIB/SEM was used to characterize inclusions. Extensive image processing (noise filtering, image alignment, manual and automatic thresholding and post-processing) and 3D measurements (volume, shape) were performed with FIJI/ImageJ. Figure 1 shows 3D models for pores and cracks obtained from �CT. The expected pore elongation and crack orientation are clearly visible. Figure 2 shows a 3D model for inclusions obtained with FIB/SEM, confirming the random distribution of mainly spherical objects. These results highlight the relevance of analyzing discontinuities in 3D, employing complementary techniques to span the vast scale range involved [6]. Microsc. Microanal. 21 (Suppl 3), 2015 2420 References: [1] T. G. Gooch, Metal Construction 15 (1983), p 164. [2] T. G. Gooch, Metal Construction 15 (1983), p 206. [3] V. R. Santos et al, Welding Journal 91 (2012), p. 319. [4] S PACIORNIK et al, Proceedings of the 1st International Conference on 3D Materials Science (2012), p. 177. [5] E PADILLA et al, Materials Characterization, 83 (2013), p. 139. 6 The authors acknowledge funding from CNPq, FAPERJ and CAPES, Brazilian funding agencies. 1000 �m 1000 �m Figure 1. 3D models obtained from �CT. Pores (left), cracks (right). 3 �m Figure 2. 3D model of inclusions, obtained from FIB/SEM.");sQ1[1209]=new Array("../7337/2421.pdf","EBSD Texture Analysis as a Measure of Local Flow in Friction Stir Welding","","2421 doi:10.1017/S143192761501288X Paper No. 1209 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 EBSD Texture Analysis as a Measure of Local Flow in Friction Stir Welding D.J. Rowenhorst1 , K.E. Knipling1 , and R.W. Fonda1 1 U.S. Naval Research Laboratory, Code 6350, Washington, DC 20375 Friction stir welding (FSW) is a relatively new joining process wherein a rotating non-consumable tool is plunged into the joint line where it produces sufficient frictional and adiabatic heat to enable rotational flow, and thus mixing of the material across the joint line. As the tool is traversed along the joint line, a consolidated weld is deposited in the wake of the tool. There has been considerable interest in the exact nature of the complex material flow around the tool as it would provide crucial information for optimizing process models and avoiding defects within the weld. There have been a number of studies that have used either dissimilar metals [1] or tracers within the material to attempt to back out the flow within the weld [2, 3] however, it can be very difficult to correlate the post-processing locations of the tracers to their original placements and there is some evidence that the tracer material itself can have a significant affect on the FSW process [4]. Recent work using Electron Backscatter Diffraction (EBSD) analysis of the wake of FSW in aluminum alloys show that the deformation textures are dominated by pure shear textures [5], which is as expected with the majority of the material being sheared around the tool face and deposited in the wake as it traverses though the material. Proper analysis of these deformation textures requires that the crystallographic orientations be rotated to align the EBSD reference frame with the geometrically determined deformation frame of the spinning tool [6]. In FSW, this deformation reference frame is a strong function of the location within the deposited weld, which is unlike other deformation mechanisms such as rolling or forging where the direction of deformation is constant within the entire sample. It was observed that even after aligning the textures to the local idealized shear directions determined from the tool geometry, additional small rotations were required of the data to achieve exact alignment with the expected pure shear reference textures. This work recognizes that these small deviations are indicative of the variations of material flow in the wake as the material is deposited, and thus can act as a forensic tool to determine the exact direction of the material flow as a function of position in the wake of the tool. Figure 1 shows a large area EBSD scan that analyzed seven positions across the wake of a FSW of aluminum alloy 2519-T87, 1-inch plate, welded at 8 inches/minute at 140RPM. The welding tool used was threaded with three flats. The EBSD data was collected on a JEOL JSM-7001F field emission SEM equipped with a EDAX EBSD system. The EBSD step size was 1 �m, and each column of EBSD data is composed of four individual scans that were then stitched together to form a single column of data, each covering an approximate area of 900�m 3500�m. The columns were spaced 2.7 mm, with seven columns evenly spaced across the weld. Using the tangental direction of the texture bands within each column of data, the crystallographic texture was measured in 50�m segments (approximately 60 segments per column) and the strongest shear deformation textures in each column were extracted, as indicated by the white boxes in Fig. 1. The texture of each extracted region was then rotated into the geometrically determined deformation orientation, and then compared to the idealized shear texture. The small rotations needed to bring the measured and ideal shear textures into coincidence are indicative of the deviations from the geometrically predetermined shear flow. Microsc. Microanal. 21 (Suppl 3), 2015 2422 Most notably Fig. 1 shows that that there is an asymmetry in the flow from the retreating to advancing side of the weld, with most of the weld showing a material flow pattern fairly consistent with the idealized shear flow, while the advancing side shows significant deviations. At all the locations in the wake of the weld, there is small yet consistent amount of downward flow (along the tool rotation axis) of the material, which is attributed to the downward flow of the threads on the tool. [7] [1] M.N. Avettand-Fenoel et al, Metall. Mater. Trans. A, vol. 45 (2014) pp. 563�578 [2] T.U. Seidel and A.P. Reynolds, Metall. Mater. Trans. A, vol. 32A (2001) pp. 2879?-84 [3] J.A. Schneider in "Friction Stir Welding and Processing" ed. R.S. Mishra and M.W. Mahoney, (ASM, Materials Park, OH) p. 37 [4] R.K. Everett, et al, ASM Proceedings of the International Conference: Trends in Welding Research, (2013) pp. 842�847 [5] Fonda R. et al, Metall. Mater. Trans. A, vol. 44 (2013) pp. 337�344 [6] Fonda R. et al, JOM, vol. 66, pp. 149�155 [7] This work was funded by the Naval Research Laboratory under the auspices of the Office of Naval Research and from the Structural Metallics program of ONR. Figure 1: Directional shear texture analysis in the wake of a Friction Stir Weld. Deviations from the idealized shear reference frame indicate the directions of material flow at the time of deposition in the weld.");sQ1[1210]=new Array("../7337/2423.pdf","EBSD Characterization of Microstructural Variations in Solid-State Welds as a Function of Distance from the Weld Interface in Ti-17","","2423 doi:10.1017/S1431927615012891 Paper No. 1210 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 EBSD Characterization of Microstructural Variations in Solid-State Welds as a Function of Distance from the Weld Interface in Ti-17 Jonathan Orsborn1,2, Thomas F. Broderick3, Hamish Fraser1 1. Center for the Accelerated Maturation of Materials, Department of Materials Science and Engineering, The Ohio State University, Columbus, OH 2. Center for Electron Microscopy and Analysis, The Ohio State University, Columbus, OH 3. GE Aviation, Materials and Process Engineering, One Neumann Way, Cincinnati, OH Recent advances in welding technology have enabled the aerospace industry to reduce further the weight of aircraft, by welding titanium alloys reliably. Solid-state welding, in which two metal pieces are welded without either one melting, has recently been proven to produce reliable and consistent welds, and is rapidly becoming the preferred method for joining titanium. Rapid heating rates, short hold-times, plastic deformation, thermal gradients, and fast cooling rates all contribute to the non-equilibrium nature of the process. All of these factors vary with respect to their proximity to the weld interface, resulting in various regions of the welded piece. Thus, the resulting microstructure varies with distance from the weld plane, and can typically be divided into four zones: the dynamically recrystallized zone (DRX), the thermo-mechanically affected zone (TMAZ), the heat-affected zone (HAZ) and the base material (BM) [1]. Commercial titanium alloys have two equilibrium phases: the low-temperature hexagonal closepacked (hcp) phase, , and the high-temperature body-centered cubic (bcc) phase, . Ti-17 (Ti-5Al-4Cr4Mo-2Sn-2Zr) is a metastable- alloy, meaning that it can retain essentially 100% phase, upon fast cooling [2]. Since the mechanical properties of titanium alloys are particularly dependent upon microstructure, it is important to understand the phase transformations occurring and the microstructures produced during the solid-state welding process. Once the various regions of the weld can be characterized, the data can help better inform integrated computational materials engineering (ICME) models. In this study, samples of Ti-17-to-Ti-17 solid-state welds were sectioned via electric-discharge machining (EDM), normal to the weld interface. The cut face was then polished according to metallographic polishing procedures for titanium alloys outlined in [3]. Once polished, the surface was indented with a Buhler microhardness indenter, using 500 g load force, and a 10 s dwell time. The first indent was placed at the weld interface and subsequent intents were placed every 100 �m, extending to 10 mm from the weld interface, on each side. To ensure accuracy of the values, each sample received multiple, parallel, identical rows of indentations. The Vicker's Microhardness values were recorded and plotted in Figure 1. These microhardness indents could then be used as fiducial markers for recording of EBSD data, enabling precise correlation between microhardness and microstructure. Inverse pole figure maps were used to determine the amount of recrystallization in the beta phase, as well as the degree of dissolution of the alpha phase and whether any re-precipitation occurred. The microhardness profiles displayed symmetry with respect to the weld interface. The lowest hardness levels occurred at the weld interface, in the DRX zone, and generally increased as distance increased. Both sides of the interface showed local peaks and local minima in hardness at distances of approximately 1100 �m, and 1600 �m from the interface, respectively. The hardness returned to the baseline level, at a distance of approximately 2500 �m from the weld interface, which is in agreement with microstructural observations. EBSD data were collected at ~100 �m intervals for the first 2 mm, Microsc. Microanal. 21 (Suppl 3), 2015 2424 and at various locations beyond that distance. The data collected near the weld interface displayed grains smaller significantly smaller than those in the base material, indicating a significant amount of recrystallization, as shown in Figure 2. The EBSD data collected at the interface produced a noisy signal that was attributed to the large amount of deformation present. Data collected ~800 �m from the interface revealed the primary -laths had been fully dissolved and then rapidly re-precipitated as multiple laths with different orientations. It also showed evidence of recrystallization of the phase, but not to the same degree as the data collected at the interface. Data collected ~1600 �m from the interface showed partially dissolved -laths, with no recrystallization of the . The EBSD data, relevant to the points of interest, will be presented and discussed. References: [1] Attallah, M. M., et al., Metallurgical and Materials Transactions A 43 (2012), p. 3149-3161. [2] G. L�tjering G. and J.C. Williams in "Titanium", (Springer, Berlin). [3] George F. Vander Voort in "Metallography: Principles and Practice", (McGraw-Hill, New York). Figure 1. Microhardness profile traversing the weld interface, plotted to � 5000 �m from the interface. Figure 2. Complimentary alpha and beta phase inverse pole figure maps (top = , bottom = ), collected every ~800 �m, going from the interface to the base material at ~3200 �m away from the interface.");sQ1[1211]=new Array("../7337/2425.pdf","Optimization and Characterization of a Niobium-Platinum Resistance Spot Weld Using Scanning Transmission Electron Microscopy and Micropillar Compression Testing.","","2425 doi:10.1017/S1431927615012908 Paper No. 1211 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Optimization and Characterization of a Niobium-Platinum Resistance Spot Weld Using Scanning Transmission Electron Microscopy and Micropillar Compression Testing. Dan Sorensen1,2, Elizabeth Rosario1, Jason Heffelfinger1, Jong Seok Jeong3, Jason Myers4, John Lippold2, Antonio Ramirez2 Medtronic Inc., Minneapolis, MN, USA Ohio State University Department of Materials Science and Engineering, Columbus, OH, USA 3 University of Minnesota Department of Chemical Engineering and Materials Science, Minneapolis MN, USA 4 University of Minnesota Characterization Facility, Minneapolis MN, USA 2 1 In the medical device industry, high reliability, manufacturability, and miniaturization are important for designing next generation devices. In addition to the above requirements, materials used for implant applications must be bio-compatible, limiting the number of materials that can be used. The limitations in material selection are challenging when creating micro welds of dissimilar materials, an often unavoidable manufacturing requirement. A micro-resistance spot weld connecting mechanical components to a power supply is required to be made from niobium and platinum wires. The Pt-Nb binary phase diagram shows the presence of numerous possible phases that can form [1], all of which have low symmetry and different crystal structures than the cubic base materials [1,2]. The object of this study was to alter the welding parameters (welding current, electrode force, electrode polarity, weld period, and niobium wire surface roughness) to optimize the microstructure and phases present in the weld fusion zone. The small dimensions of the Pt and Nb wires being welded (304 �m and 381 �m respectively) and process parameters result in fusion zone thickness values ranging from 50 - 1000 nm, as seen in Figure 1, making scanning transmission electron microscopy the best method for imaging and chemical analysis of the fusion zone as it evolves upon changes in the welding process. In this analysis, samples for electron microscopy were prepared using focused ion beam milling (FEI Versa 3D) and analyzed using conventional transmission electron microcopy (CTEM), selected area diffraction analysis (SAD), and analytical scanning transmission electron microcopy (STEM). Analytical electron microscopy was performed using energy dispersive x-ray spectroscopy (XEDS) (Figure 1(a)) and electron energy loss spectroscopy (EELS) to characterize the local chemical variations that occur as a function of weld parameter changes. The results of the chemical analysis showed chemical distributions across the fusion zone changed as a result of welding parameters. The final portion of this analysis used micropillar compression experiments to create site specific test samples for mechanical testing experiments. Three micro pillars were fabricated with the fusion zone oriented at approximately 45 degrees to the compression direction in an attempt to force a shear fracture and determine the cohesion strength of the fusion zone with the base materials using a Hysitron TI-950 Triboindenter. Figure 2 shows a representative FIB cross section of a compressed pillar with its corresponding load-displacement curve. The cross section Microsc. Microanal. 21 (Suppl 3), 2015 2426 and load displacement curve show no sign of decohesion or fracture of the complex intermetallic fusion zone, signifying a robust joint. References: [1] ASM Handbook Volume 3: Alloy Phase Diagrams, ASM International, 1992. [2] R. Ferro and A. Saccone, Intermetallic Chemistry, Elsevier, 2008 Figure 1 � (a) Representative overview HAADF-STEM image of a Pt-Nb resistance weld joined using nominal settings with XEDS line scan results. (b) High magnification HAADF-STEM image of intermetallic phase observed in fusion zone. Figure 2 � (a) FIB cross sectioned micropillar of a Pt-Nb micro weld following compression and (b) corresponding load-displacement curve.");sQ1[1212]=new Array("../7337/2427.pdf","Metallography of Robonaut 2 Battery Spot Welds","","2427 doi:10.1017/S143192761501291X Paper No. 1212 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Metallography of Robonaut 2 Battery Spot Welds James E. Martinez1, Lucie B. Johannes1, Sandeep Yayathi2, Joshua M. Figuered2, Zoran M. Bilc3 1. 2. Structural Engineering Division, NASA Johnson Space Center, Houston TX, USA Software, Robotics and Simulation Division, NASA Johnson Space Center, Houston TX, USA 3. Flight Operations Directorate, NASA Johnson Space Center, Houston TX, USA Li-ion cells provide an energy dense solution for systems that require rechargeable electrical power. However, these cells can undergo thermal runaway, the point at which the cell becomes thermally unstable and results in hot gas, flame, electrolyte leakage, and in some cases explosion. The heat and fire associated with this type of event is generally violent and can subsequently cause damage to the surrounding system or present a dangerous risk to the personnel nearby. The space flight environment is especially sensitive to risks particularly when it involves potential for fire within the habitable volume of the International Space Station (ISS). In larger battery packs such as Robonaut 2 (R2), numerous Li-ion cells are placed in parallel-series configurations to obtain the required stack voltage and desired run-time or to meet specific power requirements. This raises a second and less obvious concern for batteries that undergo certification for space flight use: the joining quality at the resistance spot weld of battery cells to component wires/leads and battery tabs, bus bars or other electronic components and assemblies. Resistance spot welds undergo materials evaluation, visual inspection, conductivity (resistivity) testing, destructive peel testing, and metallurgical examination in accordance with applicable NASA Process Specifications. Welded components are cross-sectioned to ensure they are free of cracks or voids open to any exterior surface. Pore and voids contained within the weld zone but not open to an exterior surface, and are not determined to have sharp notch like characteristics, shall be acceptable [1]. Depending on requirements, some battery cells are constructed of aluminum canisters while others are constructed of steel. Process specific weld schedules must be developed and certified for each possible joining combination. The aluminum canisters' positive terminals were particularly difficult to weld due to a bi-metal strip that comes ultrasonically pre-welded by the manufacturer. This was further complicated as the maximum electrode force was limited to low-electrode force to prevent deflection of the aluminum can during welding. Other Li-ion cells are comprised of smaller diameter cylindrical steel canisters which are inherently capable of handling greater force from the electrodes. Allowing higherelectrode forces aids greatly in insuring a consistent resistance network for the weld. Various strategies were used to include adding projections to the tabs, slotting the tabs, and developing special electrode shapes. Essential variables such as load, time, power, voltage, and current also needed to be optimized to produce robust and defect-free welds. Overall lessons learned: � Developing good jigs is critical to insure the parts and electrodes are planer to one another and the location of the weld sites remains accurate and repeatable. � Maintaining strict control over materials is critical. Materials must be of a specific hardness and chemical composition to insure that a weld schedule is repeatable. Microsc. Microanal. 21 (Suppl 3), 2015 2428 � � Accuracy of the die used to stamp the projections is critical and worth the investment. Proper seasoning of the electrodes is critical to producing consistent welds. Once the electrodes have been properly seasoned, cleaning/dressing should be avoided until it is absolutely necessary. Reference: [1] NASA PRC-0009 Resistance Spot Welding of Battery Assemblies. Figure 1. Robonaut 2 Aboard the ISS. Figure 2. Nickel-Steel Spot Weld with Cracks on Both Sides of Interface. sec Figure 3. Nickel-Steel Cathode Showing Overload Deflection");sQ1[1213]=new Array("../7337/2429.pdf","Phase Transformations and Surface/Interface Properties in Functional Perovskites with Aberration-Corrected STEM/EELS","","2429 doi:10.1017/S1431927615012921 Paper No. 1213 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Phase Transformations and Surface/Interface Properties in Functional Perovskites with Aberration-Corrected STEM/EELS Jae H. Jang,1 Rohan Mishra,2,1 Young-Min Kim,3 Qian He,1 Jaekwang Lee,4 Mina Yoon,4 S.T. Pantelides,2,1 A.Y. Borisevich1 1 2 Materials Sciences and Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831 Department of Physics and Astronomy, Vanderbilt University, Nashville, TN 3 Korea Basic Science Institute, 169-148 Gwahak-ro, Yuseong-gu, Daejeon, 305-806, Korea 4 The Center for Nanophase Materials Science, Oak Ridge National Laboratory, Oak Ridge, TN 37831 Perovskite oxides exhibit a wide variety of electronic, magnetic, and ionic properties that make them invaluable for multiple applications in energy generation and storage. Local and global changes in oxygen stoichiometry often play a crucial role in determining these properties. For materials such as solid oxide fuel cell (SOFC) anodes and photocatalysts oxygen dynamics in the bulk and on the surface, respectively, forms the basis of their functionality. To fully understand the underlying mechanisms it is therefore necessary to characterize oxygen distribution and motion in conjunction with local electronic and magnetic properties at atomic scale. Aberration corrected STEM and EELS provide unprecedented spatial resolution for structural and chemical studies of these systems. Comprehensive image analysis via local crystallography approach enables tracking atomic column positions with picometer scale precision. In system with linear chemical expansion, local crystallography can be used to monitor oxygen content in each atomic column from interatomic spacings in HAADF images [1]. The ability to obtain chemical information from images, which can be acquired much faster than spectra, enables dynamic studies of oxygen motion with atomic resolution. For LaCoO3- thin films, electron beam induces a series of phase transformations as the oxygen deficiency grows, mirroring bulk behavior [2]. However, in addition to observing the overall phase evolution, atomically resolved dynamic data allows us to track transformation pathways and even metastable intermediate states (Fig.1). Configuration analysis from large dynamic datasets poses substantial challenges, best solved via big data approaches [3]. The detected configurations are then used as an input for Nudged Elastic Band Density Functional Theory (DFT) calculations, revealing energy landscape of the system. Such data can be crucial for optimizing low-temperature performance of the SOFC devices. Surfaces and interfaces of oxides can manifest properties not observed in the bulk material, enabling new functionalities. In a model vacancy ordered SOFC material lanthanum strontium cobaltite (LSCO) tetrahedrally and octahedrally coordinated Co ions alternate in the bulk, although valence state modulations are not present due to compensation by itinerant electrons due to metallic conductivity of the material. However, in the proximity of an insulating substrate, such as lanthanum strontium aluminum tantalate (LSAT) or neodymium gallate (NGO), tetrahedral and octahedral layers have distinct electronic and magnetic behavior (Fig.2). For LSCO/LSAT films, the cobalt layer adjacent to the interface is always octahedral, but for LSCO/NGO films an abundance of antiphase boundaries creates a patchwork of Co ion coordinations and therefore electronic and magnetic properties, suggesting that interface ionic conductivity for these structures may show response to electric and magnetic fields. Oxide surfaces can be studied by STEM/EELS in cross-section, enabling characterization of the surface states responsible for photocatalysis. In a model (LaFeO3)8/(Sr FeO3)1 perovskite digital superlattice, EELS studies indicated that, unlike the octahedral bulk, Fe at the surface is tetrahedrally coordinated. DFT calculations confirm structural observations and also suggests that, while the bulk of the superlattice is insulating and prefers G-type antiferromagnetic spin arrangement, for the surface it is energetically favorable to adopt ferromagnetic order and metallic conductivity. The effect of these phenomena on surface reactivity will be discussed, as well as the prospects of tuning this reactivity with electric and magnetic fields.[4] Microsc. Microanal. 21 (Suppl 3), 2015 2430 [1] [2] [3] [4] Y.-M. Kim et al., Nature Mater. 11, 888-894 (2012) Hansteen et al., J. Mater. Chem. 8, 2081-88 (1998) O.S. Ovchinnikov et al., these proceedings. The research is sponsored in part by the Division of Materials Sciences and Engineering, Office of Basic Energy Sciences, U.S. Department of Energy. JL and MY were supported by the Center for Nanophase Materials Sciences which is sponsored at Oak Ridge National Laboratory by the Office of Science, Basic Energy Sciences, U.S. Department of Energy. a b c d e f Figure 1. (a), (b) HAADF images and the corresponding (c), (d) lattice spacing maps and (e), (f) Co shift maps showing different stages of phase transformation in LaCoO3 thin films. Red boxes on Fig. 1(f) indicate metastable configurations. a Intensity ratio of Co L3/L2 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 10 20 30 Co L3/L2 Co L signal b Intensity ratio of Co L3/L2 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 10 Co L3/L2 Co L signal ADF intensity (a.u.) ADF intensity (a.u.) 20 30 40 50 n n Figure 2. HAADF images and overlaid HAADF, Co L integrated signal, and CoL3/L2 ratio profiles for (a) LSCO/LSAT thin film and (b) LSCO/NGO thin film. The L3/L2 ratio anomaly in (b) is indicative of the tetrahedrally coordinated cobalt in the vicinity of the interface.");sQ1[1214]=new Array("../7337/2431.pdf","In-Situ Liquid Cell TEM Studies of Electrochemical reactions in Lithium-ion Batteries","","2431 doi:10.1017/S1431927615012933 Paper No. 1214 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In-Situ Liquid Cell TEM Studies of Electrochemical reactions in Lithium-ion Batteries Wei Zhang1, Daan Hein Alsem2, Norman Salmon2 and Feng Wang1 1 Sustainable Energy Technologies Department, Brookhaven National Lab, Upton, New York, United States 11973 2 Hummingbird Scientific, Lacey, WA 98516 Lithium-ion batteries (LIBs) are the preferred energy storage devices for portable electronics and electric vehicles due to the long cycle life and high energy density [1]. The development of new electrodes for next-generation batteries requires better understanding how electrodes function by realtime tracking the electrochemical reactions in individual particles and the collective behavior of particles assembled in an electrode. Various available characterization methods, such as in-situ X-ray techniques, are powerful for real time tracking charge/discharge dynamics of electrodes in liquid electrochemical system, but with limited spatial resolution. Recently the development of in-situ TEM techniques enables real time observation of structural and chemical evolutions of electrodes during battery operation at high spatial and temporal resolution [2]. However, due to the use of an open-cell configuration in these platforms, solid or ionic liquid electrolyte needs to be used for compatibility with high vacuum in the TEM column. In order to study the lithium reaction with electrodes in liquid electrolyte as generally used in LIBs, we developed a liquid electrochemical cell with thin membrane windows to seal the liquid electrolyte, allowing for both TEM and X-ray measurements with the sample in liquid environments. Here, we present our recent results from the development and application of the liquid electrochemical cells for in-situ TEM studies of lithium transport and reaction in individual particles upon lithiation and delithiation. For electrochemistry experiments, the electrochemical capabilities were developed inside a liquid cell TEM holder (designed by Hummingbird Scientific)[3], as shown in Figure 1a and b. This holder uses micro-fluidic techniques and two micro-fabricated chips each having an electron transparent SiN viewing windows (Figure 1c), and uses a sealing mechanism in which the chips can be secured on the tip of the holder and sealed from vacuum in the TEM column by means of a mechanical sealing mechanism using miniature O-rings and a simple yet reliable clamping solution. The micro-fabricated chips possess three electrodes with different metal options with electrical biasing capabilities. The active electrode material can be attached to metal on-chip leads � those are only exposed to the electrolyte in the TEM viewing area of the cell. By using the liquid cell we performed in-situ studies of electrochemical reactions in single particles of two model systems, LiFePO4 and FeF2. They are the most extensively studied cathodes due to their intriguing phase transition behaviors, which are still under intense debate [2, 4]. For the in-situ measurements, a standard liquid electrolyte (e.g. 1M LiPF6 salt in ethylene carbonate and dimethyl carbonate) is sealed within the thin membranes (50 nm or below) of the top and the bottom chips. Figure 2 shows the typical morphology of FeF2 nanoparticles, with a plate-like shape of about a few micrometers long but only several tens of nanometers thick. Good contrast of nanoplates in the liquid electrolyte was obtained by bright-field TEM imaging. The orientation of those nanoplates in the liquid electrolyte was also determined by electron diffraction. They are mostly oriented along {001} or {110} directions, allowing us to study the orientation-dependent chemical and structural evolution in single- Microsc. Microanal. 21 (Suppl 3), 2015 2432 crystalline FeF2 particles during lithiation and delithiation, in a combination of TEM imaging, diffraction and valence electron energy-loss spectroscopy [5]. Results from in-situ TEM studies of the lithium transport and electrochemical dynamics, as well as the complex reactions at the electrode-electrolyte interfaces in individual FeF2 and LiFePO4 nanoparticles, will be discussed [6]. Figure 1. (a) Liquid flow path in liquid TEM flow holder, (b) in-situ TEM liquid holder tip, (c) crosssection of the liquid-electrochemical cell. Figure 2. (a) Bright-field TEM image of a single FeF2 nanoplate in the presence of liquid electrolyte, (b) a diffraction pattern recorded from the region masked by blue circle of FeF2 nanoplate in (a), indicating the [110] zone axis. References: [1] Tarascon, J.-M.; Armand, M. Nature 414 (2001), 359-367. [2] Wang, F.; Yu, H.-C.; Chen, M.-H.; Wu, L.; Pereira, N.; Thornton, K.; Van der Ven, A.; Zhu, Y.; Amatucci, G. G.; Graetz, J. Nature communications 3 (2012), 1201. [3] Abellan, P.; Mehdi, B. L.; Parent, L. R.; Gu, M.; Park, C.; Xu, W.; Zhang, Y.; Arslan, I.; Zhang, J.G.; Wang, C.-M. Nano letters 14 (2014), 1293-1299. [4] Malik, R.; Abdellahi, A.; Ceder, G. Journal of The Electrochemical Society 160 (2013), A3179A3197. [5] Katherine, J.; Evans, J.E.; Aguiar, J.A.; Arslan, I; Browning, N.D. Microscopy and Microanalysis 18 (2012), 621. [6] This work is financially supported by the Laboratory Directed Research and Development (LDRD) program at Brookhaven National Laboratory. Research carried out at the Center for Functional Nanomaterials, Brookhaven National Laboratory, is supported by the U.S. Department of Energy, Office of Basic Energy Sciences, under Contract No. DE-AC02-98CH10886.");sQ1[1215]=new Array("../7337/2433.pdf","Controlled Electrochemical Li Cycling in a TEM to Observe Li Morphology Evolution","","2433 doi:10.1017/S1431927615012945 Paper No. 1215 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Controlled Electrochemical Li Cycling in a TEM to Observe Li Morphology Evolution Katherine L. Jungjohann1, Katharine L. Harrison2, Andrew J. Leenheer1, Nathan T. Hahn2, and Kevin R. Zavadil2 1. Center for Integrated Nanotechnologies, Sandia National Laboratories, Albuquerque, New Mexico 87185 2. Sandia National Laboratories, Albuquerque, New Mexico 87185 To meet the increased demand for high power energy storage for grid and transportation applications, a stable highly efficient Li metal battery electrode is being investigated. The system is dependent on the formation of a solid electrolyte interphase that is capable of suppressing Li dendrite formation, which is the limiting characteristic that prevents application of high capacity Li metal anodes in current lithium ion batteries (LIBs). The SEI layer that is formed between the electrolyte and the Li metal electrode is dependent on the breakdown of the Li containing electrolyte. We investigated lithium bis(fluorosulfonyl)imide (LiFSI) in dimethoxyethane (DME) for suppressed Li dendrite formation [1]. This electrolyte is targeted for stable stripping of lithium at current densities up to 10 mA/cm 2 and Coulombic efficiencies above 99.1%. The morphological evolution of the Li deposition and stripping on copper electrodes was monitored using quantitative in situ scanning transmission electron microscopy (STEM) in a custom fabricated electrochemical cell. Electrochemical control within a TEM was pioneered by F. Ross in 2003 using MEMS technology [2] to study electrochemical processes of volatile electrolytes at nanoscale resolution [3]. The Center for Integrated Nanotechnologies has successfully designed a microfabricated liquid cell that can operate with picoampere current control over 10 ultramicroelectrodes while imaging within a TEM [4]. The electrodes are patterned onto a 50 nm SiNx membrane window, with a fluid gap around 150 nm. This design allows for multiple experiments to be performed within the same closed cell, where we studied varied current densities and beam conditions under identical environmental conditions. The electrochemical platform was custom patterned using e-beam lithography to deposit Cu leads coated with a passivation layer of Al2O3 and SiO2, where ~1 m2 Cu electrodes and large >100 m2 Si electrodes were exposed, Figure 1. The patterned electrochemical platform was loaded within a He filled glove box with 4M LiFSI in DME, then epoxy coated to hermetically seal the cell. LixSi quasireference electrodes were prepared within the TEM by applying 100 pA between two Si electrodes, and lithiation of the cathodically polarized Si could be observed by decreased contrast in the Si electrode. The cell was then set up as a three electrode experiment with Cu working electrodes, a Si counter electrode, and a LixSi quasireference electrode. Three Cu electrodes were cycled through 10 lithiation deposition and stripping steps at 2.25 mA/cm2. Each electrode was exposed to different total beam exposures of 0, < 4 e-/�2, and < 21 e-/�2 of STEM imaging during over the 10 cycles. The morphology of the deposited Li exhibited dendrite formation at each of the tested beam conditions. At the highest beam exposure, shown in Figure 2, many Li structures formed including dendrites, porous deposits, hexagonal grains, and small dense deposits. At this scale, Li prefers the formation of many nucleation sites in contrast to one single grain. These results provide an understanding of the deposit morphology at each step in addition to the final deposit structure which is typically analysed after coin cell studies [5]. Microsc. Microanal. 21 (Suppl 3), 2015 2434 References: [1] R. Miao et al, J. Power Sources 271 (2014), p. 291. [2] M. J. Williamson et al, Nat. Mater. 2 (2003), p. 532. [3] F. Wu and N. Yao, Nano Energy 11 (2015), p. 196. [4] A. J. Leenheer et al, J. Microelectromech. S. 99 (2015), 10.1109/JMEMS.2014.2380771. [5] This work was performed at the Center for Integrated Nanotechnologies (CINT), a U.S. DOE Office of Basic Energy Sciences (BES) user facility. Sandia National Laboratories is a multiprogram laboratory managed and operated by Sandia Corporation, a wholly-owned subsidiary of Lockheed Martin Corporation, for the U.S. DOE's National Nuclear Security Administration under contract DEAC0494AL85000. Portions of this work were funded by the Nanostructures for Electrical Energy Storage, an Energy Frontier Research Center funded by the U.S. DOE, Office of Science. Other portions were funded by the Joint Center for Energy Storage Research, an Energy Innovation Hub funded by the U.S. DOE, Office of Science. Figure 1. Custom patterned electrochemical platform containing tungsten electrodes, passivated then patterned with four Cu working electrodes, and four Si electrodes to be used as LixSi quasireference and counter electrodes. Figure 2. Ten Li deposition/stripping cycles, displaying the measured voltage over time under galvanostatic control, and representative bright field STEM images at the final Li deposition step for each cycle where Li deposits appear as bright structures. Shaded regions in data represent open circuit steps between cycles. Scale bar represents 1 �m.");sQ1[1216]=new Array("../7337/2435.pdf","Estimation of Nanoscale Current Density Distributions during Electrodeposition","","2435 doi:10.1017/S1431927615012957 Paper No. 1216 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Estimation of Nanoscale Current Density Distributions during Electrodeposition Nicholas M. Schneider1, Jeung Hun Park2,3, Suneel Kodambaka2, Haim H. Bau1, and Frances M. Ross3 1. University of Pennsylvania, Philadelphia, PA 19104, USA. University of California Los Angeles, Los Angeles, CA 90095, USA. 3. IBM T. J. Watson Research Center, Yorktown Heights, NY 10598, USA. 2. The control of interfacial morphology in electrochemical processes is essential for various applications. Morphological instability, particularly dendrite formation, can cause potentially catastrophic failure in rechargeable batteries and can lower the quality of electroplated coatings, yet may also be useful in forming porous deposits. Thus, it is important to understand the temporal development of morphology and the nature of the forces that govern the geometry of the electrode-electrolyte interface. Liquid cell electron microscopy allows us to image, in real time and with nanoscale resolution, the evolution of the solid-liquid interface during electrochemical deposition as a function of process conditions [1- 3]. Our nanoscale resolution allows us to infer the current density along the interface. Here, we use liquid cell electron microscopy to quantify the transient growth of interfaces during galvanostatic deposition of copper from acidic electrolytes. We extract the explicit electrode-electrolyte interface as a function of time, which gives, among other measures, the normal growth velocity at each point along the interface, a measure of the point-wise current density on a time and length scale previously inaccessible. We find that initial stages of growth can be understood through Kinetic Roughening Theory [4, 5]. At longer times, we identify a transition in growth mode consistent with the onset of diffusion limited growth. The non-uniform distribution of current density along the interface ultimately leads to a rapid increase in the amplitude of surface asperities [6]. The experiments were carried out with our custom made liquid cell, the nanoaquarium [7], operating in a three terminal configuration with Pt electrodes controlled by a Gamry potentiostat. The interface morphology evolution was imaged at video rate (30 images per second) as a function of current density, current pulsing and solution composition, in a Hitachi H-9000 TEM at 300kV. Figure 1 presents images of the interface recorded under conditions leading to the onset of a diffusive instability. We apply our in house-developed, unsupervised, non-parametric image processing algorithm [8] to extract the interface location as a function of time. This data is used to compute the point-wise normal speed, which is akin to the current density distribution in space-time as shown in Figure 2a. The current density distribution evolves in conjunction with the amplitude of the surface perturbations. Figure 2b shows the current density normalized by the average current density as a function of a point's height deviation from the mean. This measure matches well with the solution for a single sinusoid under galvanostatic conditions (solid line overlay) [6]. It therefore appears that the growth regime is dictated by the macroscale diffusion physics, while the morphology is the direct consequence of the nanoscale current density distributions [9]. References: [1] J. M. Grogan, N. M. Schneider, F. M. Ross, and H. H. Bau, J. Indian Inst. Sci. 92 (2012), p. 295. [2] N. de Jonge and F. M. Ross, Nat. Nanotechnol. 6 (2011), p. 695. [3] M. J. Williamson, R. M. Tromp, P. M. Vereecken, R. Hull, and F. M. Ross, Nat. Mater. 2 (2003), p. 532. Microsc. Microanal. 21 (Suppl 3), 2015 2436 [4] [5] [6] [7] [8] [9] D. P. Barkey, Adv. Electrochem. Sci. Eng. 7 (2002), p. 151. W. Schwarzacher, J. Phys.: Condens. Matter 16 (2004), p. R859. P. Fedkiw, J. Electrochem. Soc. 127 (1980), p. 1304. J. M. Grogan and H. H. Bau, J. Microelectromech. Syst. 19 (2010), p. 885. N. M. Schneider, "NMSchneider/Quantitative-Image-Analysis," github.com. [Online]. Available: https://github.com/NMSchneider/Quantitative-Image-Analysis. [Accessed: 13-Feb-2015]. The authors acknowledge funding, in part, from the NSF, grants 1129722, 1066573, 1031208, 1310639, and 1402872. Microscopy was performed at the IBM T. J. Watson Research Center with the valuable experimental contributions of Dr. Mark C. Reuter and Mr. Arthur Ellis of IBM. Figure 1. Instantaneous micrographs of the growth interface for Cu deposition from acidified CuSO4 with an experimentally measured current density of 1500 A/m2 for galvanostatic conditions with 400nA total applied depositional current. Red outline represents analyzed region. (a) t=6s (b) 200 nm t=0s Figure 2. (a) "Heat Map" of normal growth speed (akin to point-wise current density) with the measured transition time overlayed in green. (b) Current density normalized by average current density as a function of deviation from the mean height. Data extracted from the growth of Figure 1 at t = 0.0 (black), 2.0 (blue), 3.0 (green), 3.8 (red) s. Solid line represents sinusoidal solution from [6].");sQ1[1217]=new Array("../7337/2437.pdf","In situ Electrochemical TEM for Quantitative Nanoscale Imaging Dynamics of Solid Electrolyte Interphase and Lithium Electrodeposition","doi:10.1017/S1431927615012969","2437 doi:10.1017/S1431927615012969 Paper No. 1217 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 In situ Electrochemical TEM for Quantitative Nanoscale Imaging Dynamics of Solid Electrolyte Interphase and Lithium Electrodeposition Robert L. Sacci1, Jennifer M. Black2, Nina Balke2, Nancy J. Dudney1, Karren L. More2, and Raymond R. Unocic2 1 2 Materials Science Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831 Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, TN 37831 Recently, in situ electrochemical scanning transmission electron microscopy (ec-STEM) has enabled the direct visualization of lithium battery chemistry [1�4]. In this study, we seek to understand the Li electrodeposition process, dendrite formation, and solid electrolyte interphase (SEI) growth mechanisms via ec-STEM. Annular dark field (ADF) STEM imaging is used to extract quantitative information on the density of the SEI and Li nano-electrodeposits. A limiting factor in the performance of many portable electrical devices is the need for higher power energy storage devices, such as Li-metal batteries. Li-metal batteries are prone to short-circuiting when dendrites form during Li-electrodeposition (battery charging), which leads to spontaneous discharge and thermal runaway, making them unsafe for general use [5]. It is speculated that dendrite growth can be mitigated by controlling or quickly repairing the SEI that develops at the electrode/electrolyte interface [5]. The formation of the SEI and Li-electrodeposition are studied using an ec-STEM sample holder (Protochips PoseidonTM) that allows for electrochemical experiments and simultaneous STEM imaging within a volatile liquid electrolyte environment. The electrochemical microchip, shown in Figure 1a�b, consists of a microfabricated glassy carbon (GC) working electrode with Pt pseudo-reference and counter electrodes. Negative potential sweeps show cathodic currents that stimulated SEI formation and the nucleation and growth of Li electrodeposits on the GC electrode in 1.2 M LiPF6 in EC:DMC (Figure 1c). Simultaneous quantitative ADF STEM imaging tracks the dynamic nature of SEI formation, which appears as a bright thin layer at the GC edge (Figure 1d) and shows that the thickness and brightness of the SEI increases (Figure 1e), which correlates with the SEI exhibiting a maximum density near the electrode surface of 2.3 g cm-3. This value is similar to Li-salts (e.g., LiF density is 2.6 g cm-3) and agrees with the present model of the SEI: a dense inorganic inner layer and porous organic outer layer [6]. At more negative potentials, dark contrast, irregularly shaped nano-deposits form at the GC/SEI interface (Figure 1f). These dark electrodeposits are determined to be Li metal from quantitative ADF STEM image contrast analysis and confirmed by electron energy loss spectroscopy (EELS). The image analysis and electrochemical measurements are complementary techniques and are used to estimate the thickness of the electrodeposited thin film. The direct observation of SEI formation and Li electrodeposition dynamics on GC electrodes show the utility of in operando TEM investigations. Moreover, quantitative ADF STEM imaging provides detailed chemical information that offers further insight into the design of safer and longer lasting batteries [7]. Microsc. Microanal. 21 (Suppl 3), 2015 2438 References: [1] Z Zeng et al. Nano Letters 14 (2014) p. 1745-1750. [2] RL Sacci, et al. Chemical Communications 50 (2014) p. 2104-2107. [3] ME Holtz et al. Nano Letters 14 (2014) p. 1453-1459. [4] RR Unocic et al. Microscopy and Microanalysis 20 (2014) p. 1029�1037. [5] W Xu et al. Energy and Environmental Science 7 (2014) p. 513-537. [6] PB Balbuena and Y Wang, eds. Lithium-Ion Batteries: Solid-Electrolyte Interphase. Imperial College Press, London, (2004). [7] This work was supported as part of the Fluid Interface Reactions, Structures and Transport (FIRST) Center, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences. Electron microscopy conducted as part of a user proposal at Oak Ridge National Laboratory's Center for Nanophase Materials Sciences (CNMS), a U.S. DOE Office of Science User Facility. a b c d e f SEI Li Figure 1. (a, b) SEM images of three-electrode electrochemical microchip. Scale bars are 200 and 50 �m, respectively; (c) Cyclic voltammogram (CV) of glassy carbon in 1.2 M LiPF6 in EC:DMC showing Li electrodeposition; (d) operando ADF STEM images showing progressive SEI formation during negative potential sweep; (e) transient cross-section of operando ADF-S/TEM image intensities from SEI growth region. Scale bar is 0.5 �m; (f) operando ADF STEM image showing SEI (bright edge) and Li electrodeposits (dark particles). Scale bar is 2 �m.");sQ1[1218]=new Array("../7337/2439.pdf","Understanding the Role of Additives in Improving the Performance of Bulk Heterojunction Organic Solar Cells","","2439 doi:10.1017/S1431927615012970 Paper No. 1218 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Understanding the Role of Additives in Improving the Performance of Bulk Heterojunction Organic Solar Cells Wei Chen1,2, Seth B. Darling2,3 1. 2. Materials Science Division, Argonne National Laboratory, Argonne, Illinois, 60439, USA Institute for Molecular Engineering, the University of Chicago, Chicago, Illinois, 60637, USA 3. Center for Nanoscale Materials, Nanoscience and Technology Division, Argonne National Laboratory, Argonne, Illinois, 60439, USA Organic solar cells based on the bulk heterojunction (BHJ) represent one of the most promising technologies for next-generation solar energy conversion due to their low-cost and scalability. In the last ten years, the highest power conversion efficiency (PCE) obtained from organic solar cells has risen from 2.5 to 12%.1 Controlling BHJ morphology is essential for achieving high PCE. Processing BHJ layers with solvent additives has proven to be an effective strategy toward achieving BHJ morphology correlated with high efficiency. As examples, the incorporation of a small percentage of solvent additives, such as 3v% and 0.25 v% of 1,8-diiodooctane (DIO) in PTB7:PC71BM2 and DTS(PTTh2)2:PC71BM3, respectively, results in a nearly doubling of device efficiency. However, the mechanism underlying their role in improving power conversion efficiency (PCE) has remained elusive. The general picture of a correlation between solvent constituents and the morphology within the deposited active layer film is clear, but the nature of this correlation is unknown--and it must be deciphered in order to adopt a more rational approach to improving device function. Using a combination of characterization techniques including x-ray and neutron scattering as well as advanced electron microscopy, we probe the influence of additives on active layer morphology in several state-ofthe-art BHJ blend systems. Through the integrated analysis of real- and reciprocal-space data, we uncover new insights into the role of additives that can be generalized across multiple materials. In this work, in order to explore general principles of the impact of additives on organic solar cells, several high-performance materials were selected for study. PTB72 and PBDTP-DPP4 represent topperformers among conjugated polymers in polymer:fullerene BHJ devices and DTS(PTTh2)23 is among the leading solution-processed small molecule materials when blended with fullerenes. Morphology in these high-performance systems varies quantitatively, but each system shares some common structural motifs. Key among these is a hierarchical morphology comprised of crystallites several nanometers in scale, aggregates of crystallites spanning tens of nanometers, and donor-rich/acceptor-rich regions nearly an order of magnitude larger (Figure 1). The superior performance of polymer:fullerene BHJ solar cells is attributed to hierarchical nanomorphologies with optimum crystallinity and nanoscale intermixing of copolymers with fullerenes.5 In addition, these morphologies are not equilibrium structures, meaning that the kinetic pathway by which they are created is central. The solvent additives induce the change in the formation mechanism of polymer:fullerene nanomorphologies in the process of film casting. Additives have direct impact on the interactions between the molecular constituents in a BHJ blend, and those interactions can be explored using techniques such as small angle neutron scattering (SANS). Using this approach, we shed light on how additives position the system in a different region of the phase diagram in solution, where important aspects of the morphology take form. Together, one can readily see a significant qualitative shift in morphology starting in the solution and solidifying in the film when additives are present. This shift is observed in each of the systems studied. This new insight into the role of additives provides a pathway for rational design of optimized Microsc. Microanal. 21 (Suppl 3), 2015 2440 morphology in OPV active layers. References: (Times New Roman, 12 pt, left justified) [1] M. A.Green, K. Emery, Y. Hishikawa, W. Warta, E. D. Dunlop, Prog. Photovolt: Res. Appl. 2015, 23, 1. [2] Y. Liang, Z. Xu, J. Xia, S.-T.Tsai, Y. Wu, Li, G.; Ray, C.; Yu, L. Adv. Mater. 2010, 22, E135. [3] Y. M. Sun, G. C. Welch, W. L. Leong, C. J. Takacs, G. C. Bazan, A. J. Heeger, Nature Mater. 2012, 11, 44. [4] L. Dou, J. You, J. Yang, C.-C. Chen, Y. He, S. Murase, T. Moriarty, K. Emery, G. Li, Y. Yang, Nature Photon. 2012, 6, 180. [5] W. Chen, T. Xu, F. He, W. Wang, C. Wang, J. Strzalka, Y. Liu, J. Wen, D. J. Miller, J. Chen, K. Hong, L. Yu, S. B. Darling, Nano Lett. 2011, 11, 3707. [6] This work was supported by the U.S. Department of Energy, Office of Science, Basic Energy Sciences, Materials Sciences and Engineering Division. Use of the Advanced Photon Source (APS), the Electron Microscopy Center (EMC) for Materials Research, and the Center for Nanoscale Materials (CNM) at Argonne National Laboratory was supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357. The ALS at Lawrence Berkeley National Laboratory is supported by the Director, Office of Science, Office of Basic Energy Sciences, of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231. A portion of this research was conducted at the Center for Nanophase Materials Sciences (CNMS), the High Flux Isotope Reactor (HFIR), and Spallation Neutron Source (SNS), which are sponsored at Oak Ridge National Laboratory by the Office of Basic Energy Sciences, U.S. Department of Energy. We also acknowledge the support of the National Institute of Standards and Technology (NIST), U.S. Department of Commerce, in providing the neutron research facilities used in this work. This work utilized NCNR facilities supported in part by the National Science Foundation under Agreement No. DMR-0454672. Figure 1. Correlating hierarchical nanomorphologies to OPV performance especially promotion of exciton dissociation, in polymer:fullerene thin film solar cells.");sQ1[1219]=new Array("../7337/2441.pdf","Low Dose Electron Microscopy of ZnS�Bipy based Hybrid Organic�Inorganic Materials for Energy Applications.","","2441 doi:10.1017/S1431927615012982 Paper No. 1219 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Low Dose Electron Microscopy of ZnS�Bipy based Hybrid Organic�Inorganic Materials for Energy Applications. S. Ram�rez-Rave1, A. Galano2, R. G�mez1, C. F. Kisielowski3, H. A. Calderon4. UAM-Iztapalapa, Depto. Qu�mica, �rea de Cat�lisis y �rea de Qu�mica Anal�tica, Grupo ECOCATAL, M�xico 09340, D.F. Mexico. 2. UAM-Iztapalapa, Depto. de Qu�mica, �rea de Qu�mica Anal�tica, M�xico 09340, D.F., Mexico. 3. Molecular Foundry and JCAP, LBNL, One Cyclotron Road, Berkeley, CA 94720, USA 4. ESFM-Instituto Polit�cnico Nacional, Ed. 9 UPALM, Mexico, D.F., Mexico. A strong interest in organic�inorganic hybrid materials has developed lately because of the variety of options they can offer in the modulation of electronic properties. ZnS is an interesting material for photocatalysis with a good resistance to photocorrosion and adequate conduction band position. There are a few examples in the literature regarding ZnS structural surface modification by using organic aliphatic bidentate amines [1]. Additionally a recently published investigation shows the complete synthesis procedure using solvothermal methods and characterization of ZnS and ZnS�Bipy (4,4bipyridine) hybrid based organic�inorganic materials [2]. These materials are tested as catalysts in the photocatalytic reduction of 4-nitrophenol to 4-aminophenol, under UV light irradiation and show a high performance in photocatalytic activity. As for this work, the corresponding characterization by using electron microscopy is reported. Transmission electron microscopy has been performed in the TEAM 05 microscope (NCEM-LBNL) in conditions of low dose rate in TEM mode together with a routine in MacTempas � in order to apply the exit wave reconstruction procedure with 40 experimental images. Organic substances are normally rather beam sensitive and in particular 4,4-bipyridine tends to evaporate as the sample is observed in high dose rate. Figure 1 shows a typical experimental image of the ZnS-Bipy (4,4-bipyridine) hybrid based organic� inorganic materials taken at approximately 30 e-/�2s. It is rather noisy but some features can still be distinguished as shown by the inserted intensity profile. On the other hand, Fig. 2 shows the corresponding phase image after processing 40 experimental images taken at different defoci. This phase image shows both the inorganic and the organic materials forming a layered composite and with atomic resolution (see phase line profiles). The different domains are all in nanometric size and show a fine distribution of the organic and inorganic components as well as section where the two main phases can be readily identified. Figure 3a shows the phase image that corresponds exactly to the area in Fig. 1. The regular hexagonal atom arrangement corresponds to ZnS. There is also a section with a rather irregular hexagonal pattern most likely representing a thin layer of bipiridine. Figure 3b shows a model interpretation of the atomic distribution of ZnS along a [110] zone axis in the phase image. This has been done by assuming a cubic structure of the ZnS as suggested by X-ray diffraction [2]. Nevertheless the interatomic distances measured in the projection (Fig. 3b) point out to rombohedrical distorsions of the lattice. Such deviations from the atomic positions are in average larger that those normally found for nanoparticles or other nanophases where crystalline planes elastically deform as a consequence of the size. Modelling and image simulation of the organic-inorganic arrangement complement this work. References: [1] X. Huang and J. Li, J. Am. Chem. Soc., 129 (2007), 3157. 1. Microsc. Microanal. 21 (Suppl 3), 2015 2442 [2] S. Ramirez-Rave, A. Hernandez-Gordillo, H. A. Calderon, A. Galano, C. Garcia-Mendoza, R. Gomez, New J. of Chemistry. DOI: 10.1039/c4nj01891e. [3] Electron Microscopy was performed at the Molecular Foundry, which is supported by the Office of Science, Office of Basic Energy Sciences of the U.S. Department of Energy under Contract No.DEAC02-05CH11231. The research is partially supported by CONACYT (Proyecto FOINST. 75/2012, 148304 and 129207), IPN (COFAA, SIP). 1 2 3 (1) (2) (3) Figure 1. Experimental TEM image of ZnS-Bipiridine hybrid material at a defocus of -20 nm. The intensity profile shows a periodicity of 3.1 �. Figure 2. Phase image after EWR procedure of ZnSBipiridine material. The intensity profiles show the phase recovery Vs position. Both organic and inorganic components are imaged. Figure 3. (a) Phase image and (b) corresponding model interpretation showing both types of atoms in the ZnS component. Zn and S atom positions are indicated. The insert corresponds to the [110] projection of the unit cubic cell of ZnS.");sQ1[1220]=new Array("../7337/2443.pdf","Quantification of the critical dose for radiation damage to perfluorosulfonic acid membranes using soft X-ray microscopy","","2443 doi:10.1017/S1431927615012994 Paper No. 1220 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Quantification of the critical dose for radiation damage to perfluorosulfonic acid membranes using soft X-ray microscopy Lis G. de A. Melo, G.A. Botton and Adam P. Hitchcock Department of Chemistry and Chemical Biology - McMaster University, Hamilton, ON. Canada The rate limiting reaction in polymer electrolyte membrane fuel cells (PEM-FC) occurs in the cathode catalyst layer at the triple-phase boundary between the catalyst and electron conductor (Pt or Pt-alloy on carbon nanoparticles), the proton conductor (perfluorosulfonic acid (PFSA) ionomer) and the gas (transport of O2 and H2O through the porous structure of the carbon nanoparticles). The ionomer distribution affects the performance of PEM-FC and its optimization can decrease use of the expensive Pt/Pt-alloy catalyst nanoparticles. Analytical microscopies such as TEM-EELS, TEM-EDX and Scanning Transmission X-Ray Microscopy (STXM) provide a means to analyse ionomer distributions [1-3]. However, radiation damage by ionizing radiation (electron beams or X-rays) limits visualization of the distribution of ionomer at the nanoscale. In order to develop measurement protocols which minimize errors due to damage it is helpful to know the critical dose for radiation damage, which, along with careful evaluation of doses associated with different methods, can lead to more reliable analytical imaging of ionomer in PEM-FC cathodes. The critical dose is defined for this work as the dose for which the intensity of a spectral feature characteristic of PFSA is reduced by 1/e. Here, we quantify the critical dose for damage by ionizing radiation using the focused soft X-ray beam of STXM. Previous studies [4] have shown that critical doses for high energy TEM electron beams and X-rays are similar if the absorbed doses are correctly evaluated. An activated Catalyst Coated Membrane sample with a 3M PFSA membrane was embedded with epoxy and 100 nm sections were cut with an ultra-microtome and placed on a formvar-coated Cu TEM grid. 9-pad patterns were exposed in uniform regions of the membrane with the focused (~32 nm) X-ray beam using different exposure times and incident photon energies (see Figure 1). Each pad is 600 nm x 600 nm in size. Post patterning images at 292.4 eV (C1s *C-F) or 692.6 eV (F1s *C-F), as well as full C 1s and F 1s image sequences were used to quantify the spectral changes in each pad and thus the relative damage (see Figure 2). Two different exposure energies were used: 280 eV (pre C 1s edge) and 310 eV (C 1s continuum), chosen to explore possible photon energy dependencies of the critical dose. The absorbed dose is the amount of energy absorbed by a material divided by the amount of material as explained in [5]. The absorbed dose rate will change if the material loses mass due to damage. Four different analytical methods were used to determine the critical dose. These methods include approaches which use an estimated Dinf (analytical signal at infinite dose), those which optimize Dinf , and methods which take into account the reduction of the absorbed dose as damage progresses due to mass loss, which is the dominant outcome of radiation damage to ionomer. The first method used an exponential fit to OD(t) = OD() + C exp (-a/ac), where the input values are OD(t) (optical density measured for each pad from the 292.4 eV image taken after damage) and a (absorbed dose), yielding the critical dose, ac[6]. The second and third methods were as outlined in [5], considering no significant mass loss and mass loss, respectively. The fourth method to calculate critical dose is valid for mass loss cases and was based on the F 1s optical density value at 692.6 eV for each damage pad. The sources of both random and systematic errors for each method differ. Microsc. Microanal. 21 (Suppl 3), 2015 2444 For all four methods and all conditions explored, the derived critical dose values were similar within non-systematic uncertainties, yielding an averaged critical dose of 5�3 MGy, independent of the energy used to create the damage, or the energy used to analyze the damage. This is by far the lowest critical dose reported for polymeric materials to date. The next most sensitive material is polymethylmethacrylate (PMMA) which has a measured critical dose of 67�10 MGy [5,6]. This quantitative measurement, the first for PFSA to our knowledge, emphasizes the challenges of making analytically meaningful measurements of ionomer spatial distributions, especially at the highest spatial resolution scale, where the dose used increases substantially. Planned extensions of this work to evaluate radiation damage to PFSA under cryo-conditions and with water present in the cathode and membrane will be described [7]. [1] [2] [3] [4] [5] [6] [7] L. Guetaz, S. Escribano, and O. Sicardy, J. Power Sources, vol. 212, pp. 169�178, Aug. 2012. S. Yakovlev and K. H. Downing, Phys. Chem. Chem. Phys., vol. 15, no. 4, pp. 1052�64, Jan. 2012. D. Susac, V. Berejnov, A. P. Hitchcock, and J. Stumper, in ECS Transactions, 2011, vol. 41, no. 1, pp. 629�635. J. Wang, G. A Botton, M. M. West, and A. P. Hitchcock, J. Phys. Chem. B, vol. 113, pp. 1869�1876, 2009. T. Coffey, S. G. Urquhart, and H. Ade, J. Electron Spectrosc., vol. 122, pp. 65�78, 2002. J. Wang, C. Morin, L. Li, A. P. Hitchcock, J. Electron Spectros. Relat. Phenomena, vol. 170, pp. 25�36, 2009. Measurements made at Advanced Light Source beamline 5.3.2.2 STXM. The ALS is supported by the Director, Office of Energy Research, Office of Basic Energy Sciences, Materials Sciences Division of the U.S. Department of Energy, under Contract No. DE-AC02-05CH11231. We thank AFCC for supplying the sample. Research supported by NSERC, CFI, Canada Research Chairs and the Catalyst Research for Polymer Electrolyte Fuel Cells (CaRPE-FC) network. Figure 1 (a) 9-pad pattern created by STXM in PFSA membrane at 310 eV (exposure times of 2.5, 5.0, 7.5, 12.5, 25, 38, 50, 75 and 100 ms) measured at 292.4 eV. b) exponential fit and derived critical dose (ac) using method 1 c) relative damage in OD as function of dose (assuming no mass loss) � method 2 d) linear fit of ln(Dinf � D(t)) as function of dose. Figure 2 (a) color coded map of undamaged and damaged PFSA derived from full stack fit C 1s stack, (b) C 1s spectra of the 3 pads indicated in (a), (c) F 1s edge spectra for same pads from another area. Mass loss occurs only for F 1s edge.");sQ1[1221]=new Array("../7337/2445.pdf","Observing the Self-assembly of Metal-Organic Frameworks by In-Situ Liquid Cell TEM","","2445 doi:10.1017/S1431927615013008 Paper No. 1221 Microsc. Microanal. 21 (Suppl 3), 2015 � Microscopy Society of America 2015 Observing the Self-assembly of Metal-Organic Frameworks by In-Situ Liquid Cell TEM Joseph P. Patterson,1 Patricia Abellan,2 Mike S. Denny, Jr.,1 Chiwoo Park,3 Nigel D. Browning,2 Seth M. Cohen,1 James E. Evans2,4 and Nathan C. Gianneschi1* 1. Department of Chemistry & Biochemistry, University of California, San Diego, La Jolla, California, USA, 92093, USA 2. Fundamental and Computational Sciences Directorate, Pacific Northwest National Laboratory, P.O. Box 999, Richland, WA 99352, USA 3. Department of Industrial and Manufacturing Engineering, Florida State University, Tallahassee, FL 32306, USA. 4. Environmental Molecular Science Laboratory, Pacific Northwest National Laboratory, 3335 Innovation Blvd., Richland, WA 99354, USA Here we report a significant advancement in materials science made possible by the use of liquid cell transmission electron microscopy. Namely, the ability to observe self-assembly of Metal-Organic Frameworks (MOFs) in liquids with nanometer resolution. MOFs were studied by LCTEM in order to understand and demonstrate control over the dynamics and electron beam effects of these complex macromolecular, beam sensitive materials. With this control, we were then able to observe the nucleation and self-assembly of zeolitic imidazole framework 8 (ZIF-8) giving us the ability to extract growth kinetics, particle size distribution, particle morphology and crystal structure from a single experiment. The ability to observe dynamic self-assembled materials in liquids at the nanoscale is vital to our understanding of synthetic and biological materials. Liquid cell electron microscopy promises to revolutionize our ability to analyze these materials in liquids due to its unmatched spatiotemporal resolution. A key hurdle for analyzing synthetic or biological self-assembled materials by this method has been the ability to control their behavior under the electron beam and in the nanoliter liquid cells. We report the first observation of nucleation and self-assembly of metal-organic-frameworks by LCTEM providing an unprecedented level of information on their formation by a single technique. This result has far reaching implication as the ability to control and observe these processes for complex macromolecular, beam sensitive materials demonstrates the applicability of this technique for a wide array of soft and biological materials. LCTEM will become essential for our understanding of nanoscale processes for all materials in liquids; however there is current debate in the literature how useful this tool can be for beam sensitive materials like MOFs. Here we demonstrate that with a careful understanding of how the pre-synthesized materials behave in the liquid cell and under electron beam exposure, suitable conditions can be obtained to observe their self-assembly process with minimal beam effects. Using electron diffraction we confirm the same thermodynamic crystal structure is obtained as under bulk synthetic conditions and show for the first time that individual MOF particles growth through a surface reaction limited mechanism. Furthermore, we demonstrate for the first time the ability to control nucleation density in the cell by materials agitation (flow) rather than all previous studies where nucleation of inorganic crystals was controlled by the electron beam. Microsc. Microanal. 21 (Suppl 3), 2015 2446 References [1] de N Jonge, FM Ross, Nature Nanotechnology6, 2011, p. 695. [2] OM Yaghi et. al., Chemical Reviews, 112, 2012, p. 673. [3] TJ Woehl et. al. ACS Nano, 10, 2012, 8599 . Figure 1. Liquid Cell TEM and STEM images of UiO66 in water. A) high magnification STEM images B) and C) low magnification STEM images tracking particle motion over the period of 6 seconds. D) and E) TEM images after significant beam damage at 300 keV and F) STEM images after significant damage at 80 keV. Figure 2. Growth of ZIF-8 MOF by LCTEM A) - C) snapshots of a movie showing particle selfassembly in methanol and D) average growth kinetics extracted directly from movie analysis.");




var sQb=0;var sQc=1;var sQd=2;var sQe=3;var sQf=4;var sQg=5;var sQh=6;var sQi=7;var sQj=8;var sQk;var sQdl;var sQm;var sQn; var sQ98 ; var sQB=true;var sQC=1;var sQE=2;var sQD=3;var sQF=4;var sQP=false;var sQQ=true;var sQ46=true;var sQ83=false;var sQ73=false;var sQ14=true;var sQ17=false;var sQ90=-1;var sQ45=0;var sQT;function SMPSetFocus() { document.formSearch.txtSearch.focus(); }
function sQp(sQq) {if ((sQq>="a" && sQq<="z") ||(sQq>="A" && sQq<="Z") || (sQq>="0" && sQq <="9")) { return true; } else { return false; } }function sQr( sQt, sQu, sQv ) {var sQy;var sQz, sQ7, sQ2; var sQ3=0; var sQ42; var sQ43=true; var sQ44;sQ1[sQt][sQg]=0;for( sQ42=0; sQ42<sQ41.length; sQ42++ ){ if (sQ41[sQ42].length>0) { sQy=sQ1[sQt][sQu];if ( !sQ17 ){ sQy=sQy.toUpperCase(); } sQz=sQy.indexOf( sQ41[sQ42] );sQ44=sQ1[sQt][sQg];while (sQz >= 0){ sQ3 = sQ3 + sQz + 1;if(( sQm== -2 ) || ( sQm== -5 )) { sQ7 = false; } else { if (sQz == 0) { sQ7=false; } else { sQ7=sQp(sQy.charAt(sQz-1)); }  } if(( sQm == -3 ) || ( sQm == -5 )) { sQ2 = false; } else { if ( sQy.length - sQz == sQ41[sQ42].length) { sQ2 = false; } else { sQ2 = sQp(sQy.charAt(sQz + sQ41[sQ42].length)); }  } if (!sQ7 && !sQ2) { sQ1[sQt][sQj+sQ1[sQt][sQg]] = sQ3 - 1;sQ1[sQt][sQg] ++; } sQy = sQy.substring(sQz+1,sQy.length);while( sQp(sQy.charAt(0)) && sQy.length > 0 ) { sQy = sQy.substring( 1, sQy.length ) ;sQ3 ++; } sQz = sQy.indexOf(sQ41[sQ42]); } if( sQ1[sQt][sQg] == sQ44) { sQ43=false; } } } if( (sQ45==2) && (sQ43==false) ) { sQ1[sQt][sQg]=0; } } function sQ4(sQt, sQv) { var sQu=-1, sQ5; var ak=63250308986572; sQ1[sQt][sQf]=-1;while( ++sQu <= sQe && sQ1[sQt][sQf]==-1 ) { sQ5 = sQu==0 ? sQc : (sQu==1 ? sQd : (sQu==2 ? sQb : sQe));if( ((sQ5 == sQc ) && ( sQ46 )) || ((sQ5 == sQd ) && ( sQ83 )) || ((sQ5 == sQb ) && ( sQ73 )) || ((sQ5 == sQe ) && ( sQ14 )) ) { sQr( sQt, sQ5, sQv ); } else { sQ1[sQt][sQg]=0 } if ( sQ1[sQt][sQg] > 0 ) { sQ1[sQt][sQf] = sQ5; } } s="searchmaker";t=eval(s.charAt(2)+s.charAt(8)); } function sQ6() { var sQt, sQu, sQ79, sQ8;for( sQt=1; sQt<sQ1.length; sQt++ ) { sQ1[sQt][sQi] = sQt; } if( sQB ) {for( sQt=1; sQt<sQ1.length; sQt++ ) { if( sQ1[sQt][sQf] == sQc ) { sQ1[sQt][sQh] = (4-sQC) * 15000; } else if( sQ1[sQt][sQf] == sQd ) { sQ1[sQt][sQh] = (4-sQE) * 15000; } else if( sQ1[sQt][sQf] == sQb ) { sQ1[sQt][sQh] = (4-sQD) * 15000; } else { sQ1[sQt][sQh] = (4-sQF) * 15000 + sQ1[sQt][sQg]; } } for( sQt=2; sQt<sQ1.length; sQt++ ) { sQ79 = sQ1[sQt][sQh]; sQ8 = sQ1[sQt][sQi];for( sQu=sQt; sQu>1 && sQ79 > sQ1[sQu-1][sQh]; sQu--) { sQ1[sQu][sQh] = sQ1[sQu-1][sQh];sQ1[sQu][sQi] = sQ1[sQu-1][sQi]; }sQ1[sQu][sQh] = sQ79;sQ1[sQu][sQi] = sQ8; } } } function sQ9(sQt) { var sQA = sQ1[sQt][sQg]==1 ? "match" : "matches";sQk += ""; sQk += "<font face=\"Arial\" size=\"2\" color=gray>"; if( sQ1[sQt][sQf]==sQc ) { sQk += " - matched title"; } else { if( sQ1[sQt][sQf]==sQd ) {sQk += " - matched description"; } else { if( sQ1[sQt][sQf]==sQb ) { sQk += " - matched URL"; } else { sQk += " - " + sQ1[sQt][sQg] + " " + sQA + "";  } } } sQk += "</font>"; sQk += ""; } function sQG(sQt, sQH) { var sQI;var sQJ;var sQK = false; var sQdL=sQ1[sQt][sQj + sQH - 1]; sQI = sQdL<35 ? sQI=0 : sQI=sQdL-35;sQJ = (sQdL+35 > sQ1[sQt][sQe].length) ? sQJ=sQ1[sQt][sQe].length : sQJ=sQdL+35;while ((sQI>=0) && !sQK) { if(sQp(sQ1[sQt][sQe].charAt(sQI))) { sQI--; } else { sQK=true; } } sQI++;sQK=false;while ((sQJ > sQdL) && !sQK) { if(sQp(sQ1[sQt][sQe].charAt(sQJ))) { sQJ--; } else { sQK=true; } } sQk += "<BR><span class=search2>\".. "+sQ1[sQt][sQe].substring(sQI,sQdL); sQk += "<B>" + sQ1[sQt][sQe].substring(sQdL , sQdL + sQdl.length ) +"</B>";sQk += sQ1[sQt][sQe].substring(sQdL + sQdl.length ,sQJ) + " ..\"</span>"; } function sQN( sQt ) { if( sQP==false ) { var sQu=1; while ( (sQu < 4) && (sQu<=sQ1[sQt][sQg])) { sQG( sQt, sQu ); sQu++; } } } function sQR( sQy, sQdL ) { sQk += sQy.substring( 0, sQdL );sQk += "<B>" + sQy.substring( sQdL , sQdL + sQdl.length ) +"</B>";sQk += sQy.substring( sQdL + sQdl.length ,sQy.length ); } function sQU( sQt, sQT ) { sQk += "<p><font face=\"Arial\" size=\"2\" color=#000000>" + sQT +".</font> ";sQk += "<a target=\"_blank\" href=\""+sQ1[sQt][sQb]+"\"><span class=search1>"+sQ1[sQt][sQc]+"</span></a>";if (( sQ1[sQt][sQf]==sQe ) && (sQ45==0)) { sQN(sQt); } else { sQk+=""; } if( (sQ1[sQt][sQf]==sQd ) && (sQ45==0)) { sQk += "<br><font face=\"Arial\" size=\"2\" color=gray>Description:</font><font face=\"Arial\" size=\"2\" color=#000000> "; sQk += "";sQR( sQ1[sQt][sQd], sQ1[sQt][sQj] ); sQk += "</font>"; } else { if( sQ1[sQt][sQd].length > 0 ) { sQk += "<br><font face=\"Arial\" size=\"2\" color=gray>Description:</font></font><font face=\"Arial\" size=\"2\" color=#000000> "; sQk += "" + sQ1[sQt][sQd]; } else {  } } sQk += "</font>"; if( sQQ ) { sQ9( sQt ); } sQk += "<br>"; }; function sQV() { sQT = 0;if (! (sQn )) { for( var sQt=1; sQt<sQ1.length; sQt++ ) { if(( sQ1[sQ1[sQt][sQi]][sQg] > 0 ) && (( sQT < sQ90 ) || ( sQ90 == -1 ))) { sQU(sQ1[sQt][sQi], ++sQT); } }  } else { if( sQm == -4 ) { sQk += "<BR><font face=\"Arial\" size=\"2\" color=#000000>ERROR: The wildcard character (*) must be at the beginning or end of the text.</font>"; } }  } function sQW() { sQk += "<!DOCTYPE html PUBLIC \"-//W3C//DTD XHTML 1.0 Transitional//EN\" \"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd\"> <html xmlns=\"http://www.w3.org/1999/xhtml\"> <head> <meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" /> <meta http-equiv=\"Page-Enter\" content=\"BlendTrans(Duration=.01)\" /> <meta http-equiv=\"Page-Exit\" content=\"BlendTrans(Duration=.01)\" /> <title>MICROSCOPY & MICROANALYSIS 2015 | Search Results</title> <!--[if lte IE 8]> <link href=\"../27778953/style8.css\" rel=\"stylesheet\" type=\"text/css\" /> <![endif]--> <!--[if gte IE 9]> <link href=\"../27778953/style.css\" rel=\"stylesheet\" type=\"text/css\" /> <![endif]--> <![if !IE]> <link href=\"../27778953/style.css\" rel=\"stylesheet\" type=\"text/css\" /> <![endif]> <script src=\"search.js\" type=\"text/javascript\"></scr"; sQk +="ipt> <link rel=\"SHORTCUT ICON\" href=\"../462437/M&M.ico\" /> </head> <body onLoad=SMPSetFocus()> <div id=\"content-holder\" align=\"center\"><div id=\"content\"> <div id=\"right\" class=\"search-results\"> "; sQk += "<font face=\"Arial\" size=\"2\" color=#000000>";sQk += "<p>You searched for <b>" + sQ98 +".</b>"; sQk += "</font>" }; function sQY() { if(sQT==0) { sQk += "<p><font face=\"Arial\" size=\"2\" color=#000000>No pages matched your search.&nbsp;&nbsp;</font>"; } else { var sQA = sQT==1 ? "page shown." : "pages shown."; sQk += "<p><font face=\"Arial\" size=\"2\" color=#000000>" + sQT + " " + sQA + "&nbsp;</font>"; } sQk += "</div> <div id=\"left\"> <h2>Browse Scientific Program By:</h2><br /><h3>Table of Contents by Symposium</h3><ul><p class=\"indent\"><a href=\"plenary.html\" target=\"_self\">Plenary</a></p><p class=\"indent\"><a href=\"aiss.html\" target=\"_self\">Analytical and Instrumentation Science &nbsp;Symposia</a></p><p class=\"indent\"><a href=\"bss.html\" target=\"_self\">Biological Science Symposia</a></p><p class=\"indent\"><a href=\"pss.html\" target=\"_self\">Physical Science Symposia</a></p><p class=\"indent\"><a href=\"tftos.html\" target=\"_self\">Technologist Forum, Tutorials, and Outreach Symposia</a></p></ul><h3><a href=\"TOC.html\" target=\"_self\">Entire Program</a></h3><br /><h3><a href=\"author.html\" target=\"_self\">Author Index</a></h3><br /><h3><a href=\"keyword.html\" target=\"_self\">Keyword Index</a></h3><br /><h2>Global Search Engine</h2><p>Use this search engine to find papers at MM2015 containing your custom search parameters, for example: Author, Paper Title, Custom Keywords. A linked list of papers matching your supplied parameters will be automatically generated for you.</p> <p class=\"spacer\"></p> <form name=formSearch action=\"javascript:SMPStartSearch() //\"> <input name=txtSearch> <input name=send type=submit value=Search> </form> <p class=\"search-criteria\">For best results enter just one or two words. Searches on more than one word will be treated \"as a phrase\".</p> <p class=\"search-criteria\">Use the asterisk (*) character for results that contain all or part of a word.</p> </div> </div></div> <div id=\"header-holder\"><div id=\"header\" align=\"center\"> <a href=\"../Index.html\" target=\"_self\"><div id=\"header-link\"></div></a> <div id=\"header-width\"><p class=\"years\">73<sup class=\"sup\">ND</sup> Annual Meeting of the Micoscopy Society of America<br /> 49<sup class=\"sup\">TH</sup> Annual Meeting of the Microanalysis Society<br /> 48<sup class=\"sup\">TH</sup> Annual Meeting of the International Metallographic Society<br /><sup class=\"sup\"></sup><br /> <sup class=\"sup\"></sup> </p></div> <div id=\"navigation\"><a href=\"TOC.html\" target=\"_self\" class=\"no-underline\"><div id=\"button\">TABLE OF CONTENTS</div></a> <a href=\"plenary.html\" target=\"_self\" class=\"no-underline\"><div id=\"button\">WELCOME LETTERS & SOCIETY PAGES</div></a> </div> </div> <div id=\"footer-holder\"><div id=\"footer\" align=\"center\"><div id=\"society\"><a href=\"http://iumas.org/\" target=\"_blank\"><div id=\"IUMAS\"></div></a><a href=\"http://www.msc-smc.org\" target=\"_blank\"><div id=\"MSC\"></div></a><a href=\"http://www.asminternational.org/web/ims/home\" target=\"_blank\"><div id=\"IMS\"></div></a><a href=\"http://microanalysissociety.org\" target=\"_blank\"><div id=\"MAS\"></div></a><a href=\"http://www.microscopy.org\" target=\"_blank\"><div id=\"MSA\"></div></a></div></div></div> </body> </html>"; } function sQ11() { var sQ23 = sQk;document.open(); document.write(sQ23); document.close(); } function sQ03( sQy ) { var sQqh="";for( var sQt=0; sQt<sQy.length; sQt++) { if( sQy.charAt(sQt)=="<" ) { sQqh += "&lt;"; } else if( sQy.charAt(sQt)==">" ) { sQqh += "&gt;"; } else if( sQy.charAt(sQt)=="\"" ) { sQqh += "&quot;"; } else { sQqh += sQy.charAt(sQt); } } return( sQqh ); } function sQ93() { sQm = sQdl.indexOf("*"); if( sQm == 0 ) { sQm = -2; sQn = false; } else if (sQm == sQdl.length -1) { sQm = -3; sQn = false; } else if (sQm > 0 ) { sQm = -4;sQn = true; } else { sQn = false; } if ( sQdl.indexOf("*") != sQdl.lastIndexOf("*") ) { if( sQm == -2 ) { if( sQdl.lastIndexOf("*") == sQdl.length - 1 ) { sQm = -5; } else { sQm = -4; sQn = true; } } }if( ( sQm == -2 ) || (sQm == -5 )) { sQdl = sQdl.substring( 1, sQdl.length ); } if( ( sQm == -3 ) || (sQm == -5 )) { sQdl = sQdl.substring( 0, sQdl.length - 1 ); } } function SMPStartSearch() { var sQv;sQdl=document.formSearch.txtSearch.value;if (( sQdl.length > 0 )&&( sQdl != "*" )) { sQk = "";sQdl = sQ03(sQdl);sQ98 = sQdl; sQ93();if( sQ17 ) { sQv = sQdl; } else { sQv = sQdl.toUpperCase(); }if (sQ45 != 0) { sQ41=sQv.split(' '); } else { sQ41=sQv.split(); }if ( !(sQn ) ) { for( var sQt=1; sQt<sQ1.length; sQt++ ) { sQ4( sQt, sQv ); } sQ6(); } sQW();sQV();sQY();sQ11(); } } 

//  End of Search Maker Pro script.