Nanowires may be used as building blocks for many future applications ranging from microelectronics and nano-electro-mechanical systems to anode materials in Li ion batteries. This is due to their unique physical properties which are a consequence of their high aspect and surface-to-volume ratio. Particularly, nanowires with typical diameters of 100 nm and below have shown very high strength, which can be of the order of the theoretical strength of the respective material. In this paper, we will introduce a nanomechanical setup that allows for testing nanowires with diameters as small as 30 nm inside a dual beam SEM/FIB. The force measurement is accomplished by a MEMS-based load cell and the strain measurement by digital image correlation of the SEM pictures. Manipulation, transfer and alignment of the samples are performed with the help of a piezoelectric manipulator. Obviously, quantitative measurements with high accuracy are not trivial at this length scale. Therefore, the main sources of experimental errors such as misalignment, determining the sample diameter, or contamination of the sample surface are critically discussed. With this system, Au, Cu, and Si nanowires were tested in tension. For all nanowires, the yield and fracture strengths were found at very high stresses close to the limit of theoretical strength. In particular for the Si nanowires, the fracture strengths do show only a weak size dependence for wire diameters below 200 nm but does exhibit a large scatter in the strength values. This indicates the statistical nature of the fracture process which is analyzed by applying a weakest-link approach. It is shown that the scatter as well as the size dependence can be consistently described by Weibull statistics.

Both theoretical and experimental studies indicate mechanical size effects are related to dislocation mechanisms often influenced by the increased surface to volume ratios at the nanoscale. Mechanical size effects can originate from source limitation (due to source exhaustion and dislocation starvation) or from size-related dislocation mechanisms such as self-multiplication in small BCC volumes and source truncation. To investigate these various mechanisms in small bcc volumes, we have performed in situ transmission electron microscope (TEM) tensile tests of Mo-alloy nanofibers using a “push-to-pull” (PTP) device. By augmenting the in situ testing method with Digital Image Correlation (DIC) we can determine the true stress and strain in local areas of deformation. We found that the nanofibers exhibited clear exhaustion hardening behavior, where the progressive exhaustion of dislocation sources and starvation of dislocations increases the stress required to drive plasticity. Our presentation will focus on recent improvements to our mechanical testing techniques and critically discuss the various size strengthening effects observed during our in situ TEM tensile tests.

For the first time, the mechanical properties of submicron-sized metallic glasses (MGs) were studied through quantitative in situ tensile test inside a transmission electron microscope, which employs high-resolution measurements of the loading forces and accurate strain measurement with deposited markers on the gauge length. The quantitative experiment establishes that the small-volume Cu₅₀Zr₅₀ MG has fracture strength and elongation to failure reaching record high values of about 4 GPa and 6%, respectively. At the same time, the yield stress and elastic stain limit are found to be about twice as large as the already-high elastic limit observed in macroscopic samples, in line with model predictions of the intrinsic limit in the absence of heterogeneous shear band nucleation facilitated by extrinsic factors.

We report a series of in situ and ex situ compression tests of Ti-Al alloys with 6-7 wt.% Al in order to study mechanical size effects in alpha-Ti across several orders of magnitude. In Ti-Al solid solutions (HCP alpha-Ti), high lattice friction and short range ordering produce a planar slip behavior that was presumed to be relatively insensitive to size effects. For the in situ TEM nanocompression tests, Ti-7 wt% Al nanopillars from 280-380 nm in diameter were prepared by FIB from a polycrystalline sample. We analyzed the orientation of the nanopillars in the TEM and selected nanopillars having the (0001) plane inclined 45 - 60 degrees to the long axis (compression axis). In situ TEM nanocompression revealed that the pillars sheared along the (0001) plane as anticipated and that the critical resolved shear stress (CRSS) for basal slip was 450 MPa. Previous results reported that the CRSS for basal slip in bulk Ti-6.6Al is about 230 MPa at room temperature, and microcompression tests of pillars from 1-20 microns in diameter show relatively little size strengthening. Taken together, the in situ and ex situ compression tests show a relatively small size effect in Ti-Al solid solutions due to the predominant effects of the dislocation behavior as compared to the effects of source limitation or source truncation in small volumes. Acknowledgments: This work was supported by the Center for Defect Physics, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences. ET was supported by a JSPS fellowship. The in situ TEM work was performed at the National Center for Electron Microscopy, Lawrence Berkeley National Laboratory, which is supported by the U.S. Department of Energy under Contract # DE-AC02-05CH11231.

In situ straining experiments in the transmission electron microscope are the unique way to observe the dislocations under stress and so, to determine the strength of the obstacles opposing to the dislocations propagation. This represents an essential step for the determination of what is controlling the plastic deformation. In this paper, two examples will be described. The first case concerns extrinsic obstacles which are present in aluminum and intermetallic TiAl alloys. The strength of this pinning will be determined through the analysis of the critical configuration for the obstacle breaking in the framework of the dislocation theory. From this strength evaluation and from the measurements of the obstacle density, the obstacle overcoming mechanism will be correlated to the macroscopic shear stress. It will be shown that for aluminum alloys the pinning is controlling the dislocation propagation whereas that is not the case for TiAl. The second example is about the interface crossing by dislocations in lamellar TiAl alloys. In these alloys, the rectilinear interfaces are characterized by the different phases and by the orientation relationships between the lamellae. Several situations related to the type of interfaces, the nature of the incident dislocations and/or the orientation of the applied stress will be investigated. The strength of these various interfaces is deduced from the characteristics and the dynamics of the crossing mechanisms. The final aim is to understand the role of these interfaces on the mechanical properties of lamellar TiAl alloys.

Size effects leading to a ‘smaller-is-stronger&’ in micron and sub-micron scale plasticity have been commonly reported in recent literature. Still, less is known regarding the mechanisms responsible for the remarkable strength and hardening as well as the failure behavior of such miniaturized samples. This calls for direct in situ observation of the governing deformation mechanisms. We applied focused ion beam machining to fabricate non-tapered nano-tensile samples with dimensions in the 100 nm - 200 nm regime from single crystal copper. This tensile approach minimizes experimental constraints existing in common pillar compression studies and allows to study the failure behavior. Quantitative in situ tensile tests [1] on single and multiple slip oriented Cu tensile samples were performed in a Jeol 3010 transmission electron microscope operated at 300 keV with a Hysitron Picoindenter. We observe the operation of truncated spiral sources at high stresses as the dominant mechanism carrying plasticity, and both crystal orientations fail by localized shear. While tensile failure occurs after few percent plastic strain and limited hardening in the single slip case, extended homogeneous deformation and necking due to the activation of multiple dislocation sources in conjunction with significant hardening is observed for the multiple slip samples. Moreover, we show that the strain rate sensitivity of such FIB prepared samples is an order of magnitude higher than that of bulk Cu[2]. [1] Kiener D, Minor AM. Nano Lett. 2011;11:3816. [2] Kiener D, Kaufmann P, Minor AM. Adv. Eng. Mater. 2012:online.

Polycrystalline metals demonstrate a clear trend of strengths that scale as a power-law of their size. It has been established long ago in coarse grained materials that the strength is inversely proportional to the square root of the grain size (Hall-Petch law). However about 10 years ago, a breakdown of the Hall-Petch law has been noticed in metals with grain size smaller than 100nm. Because both the mean free path and the probability of multiplication of dislocation are constrained at such a grain scale, a probable change in the plasticity mechanism associated with very high yield stresses is supposed to occur. Exploring these mechanisms in very small crystallites is however complex and the nature of alternate deformation mechanism is still highly debated. In-situ Transmission Electron Microscopy (TEM) experiments have proved in the recent past years to be a powerful tool to probe elementary plasticity mechanisms at the pertinent time and length scale. Down to small scale, these experiments triggered deformation mechanisms to plasticity, including specific dislocation mechanisms (partial dislocation nucleation instead of perfect ones, surface dislocation emissionhellip;) or grain boundary mediated plasticity. In the present paper, I would like to report several observations of both intragranular and intergranular plasticity mechanisms in small grained Al (grain size between 100nm and 1µm) elaborated either by electrodeposition or by severe plastic deformation. Grain boundary (GB) plasticity mechanisms have been observed in specific conditions where usual dislocations mechanisms are prevented. In dislocation free nanocrystalline grains, we have found that stress assisted grain growth can be an efficient process to accommodate deformation. The amount of strain produced during this process can be measured and tentatively modeled. GB sliding due to the motion of interfacial dislocations is another possible mechanism leading to the formation of cavities. In this condition, strain up to several tens of percents has been achieved locally in films deformed uniaxially in tension. Intragranular dislocation mechanisms have been observed systematically in grains containing initially dislocation sources (i.e. more likely in larger grains) or when GB sources can emit dislocations inside grains. The formation of dislocation pile-ups against GB is a direct consequence of intragranular plasticity. The reversal motion of dislocations in the pile-up can explain quantitatively the Bauschinger effect observed during micro-yielding experiments.

Lately, ‘smaller is stronger&’ was widely reported in single crystals with the sample dimensions reduced to sizes less than ~ 20 mu;m in diameter [1-4]. Several new concepts have been brought forward to explain this phenomenon, such as ‘dislocation-starvation&’ [2, 5-7], ‘mechanical annealing&’ [1] and ‘source limited deformation&’ [4], etc. It is agreed that the dislocation starvation model dominated in the face-centered cubic (FCC) micropillars through molecular dynamics and dislocation dynamics simulations. But a body-centered cubic (BCC) micropillar is unlikely to be in a “dislocation-starved” state. The slow moving screw oriented dislocation may be able to generate new dislocations moving in the opposite direction before it exits from the surface, namely self-multiplication mechanism [8-10]. However, “dislocation-starved” state or “mechanical annealing” and size effect were recently reported in BCC Molybdenum pillars when the diameters decrease to hundreds of nanometers [11].Our experiment further demonstrates that “dislocation starvation” does occur in other BCC metals. Also, although the observation of a single dislocation moving through the pillar in real time is difficult, we effectively captured it by changing the testing method from normal bright or dark field TEM image mode to scanning TEM mode (STEM). Here we report our in situ compression test of a BCC single crystal Fe-3% Si pillar in a TEM. We found that the deformation is characterized by series of pronounced pop-ins accompanying stresses increase at these strain bursts. The increasing in stresses was found to be caused by the change from the edge dislocation dominated process to the screw dislocation dominated process. The underlying mechanism might be the change of dislocation sources or dislocation starvation/dislocation exhaustion. References: [1] Z. W. Shan, R. K. Mishira, S. A. S. Asif, O. L. Warren, A. M. Minor, Nature Mater. 7 (2008) 115. [2] M. D. Uchic, D. M. Dimiduk, J. N. Florando, W. D. Nix, Science 305 (2004) 986. [3] D. Kiener, W. Grosinger, G. Dehm, R. Pippan, Acta Mater. 56 (2008) 580. [4] C. A. Volkert, E. T. Lilleodden, Phil. Mag. 86 (2006) 5567. [5] W. D. Nix, J. R. Greer, G. Feng, E. T. Lilleodden, Thin Solid Films 515 (2007) 3152. [6] J. R. Greer, W. C. Oliver, W. D. Nix, Acta Mater. 53 (2005) 1821. [7] J. R. Greer, W. D. Nix, Phys. Rev. B 73 (2006) 245410. [8] C. R. Weinberer, W. Cai, PNAS 105 (2008) 14304. [9] J. R. Greer, C. R. Weinberger, W. Cai, Mater. Sci. Eng. A 493 (2008) 21. [10] S. Brinckmann, J.-Y. Kim, J. R. Greer, Phys. Rev. Lett. 100 (2008) 155502. [11] L. Huang, Q.-J. Li, Z.-W. Shan, J. Li, J. Sun, E. Ma, Nature Communication 2 (2011) 547.

Grain boundary engineering (GBE) is a proven way to increase the resistance of FCC metals to intergranular cracking and corrosion. However, the mechanisms through which engineered microstructures evolve are not well understood. Previous studies have characterized microstructure and grain boundary character distributions before and after GBE processing, providing empirical data from which several mechanisms have been proposed but none verified. The current hypothesis is that Σ3 type grain boundary nuclei form first through annealing twinning and that these nuclei lead to a proliferation of Σ3n boundaries due to geometric constraints. In order to verify this, in situ heating and straining experiments were carried out in a transmission electron microscope (TEM) to replicate GBE thermomechanical processing. High purity copper was used as an analogue model FCC material. The effects of heat and strain on boundary mobility were assessed during cyclic processing. Variations in the mobility of boundaries with different grain boundary character were assessed by utilizing the Nanomegas DIGISTAR/ASTAR orientation image mapping (OIM) method in TEM to assess the character of individual boundaries formed in situ. By isolating the effects of strain and annealing on the formation and mobility of Σ boundaries a better predictive model can be developed for future application of this process in various industries.

The 5XXX series aluminum alloys containing 5% magnesium are commonly used in structural applications requiring good corrosion resistance and weldability. They are a non-heat treatable materials which gain strength through primarily cold work and solid solution hardening with magnesium. Despite the strong characteristics the 5XXX series alloys are susceptible to sensitization. Over extended periods of time at elevated temperatures magnesium segregates from the matrix and forms secondary Al3MG2 (β phase) precipitates along the grain boundaries. This anodic β phase can be very susceptible to corrosive environments and can lead to intergranular stress corrosion cracking. It is well known that β phase forms at medium/high temperatures (150-300°C) but there is still some uncertainty about that actual formation and effect of stress on the kinetics of β phase growth. In-situ heating and straining in Transmission Electron Microscopy (TEM)techniques give promise to understanding the kinetics and mechanisms of β phase growth. In-situ TEM will allow for determination of the time, temperature and strain that is necessary for sensitization to occur. Having a better understanding of these parameters will give a more predictive model for the formation of β phase and ultimately a chance to prevent its formation. The work included in the research will aim to show the earliest stages of β phase growth giving a much better idea of what causes the onset of formation. Finally, orientation imaging will also be coupled with the in-situ experiments to see the dependency of grain boundary type and which are more susceptible to corrosion. Combining these techniques brings us a step forward in understanding the β phase formation in aluminum magnesium alloys.

The use of duplex stainless steels (DSSs), which consist of austenite and ferrite, is rapidly increasing due to their combined advantages of outstanding mechanical and corrosion properties. Generally, it is known that the austenite phase has higher strength comparing with the ferrite, though the difference between mechanical strength depends on the chemical composition in both phases. Recently, however an in-situ tensile test with neutron diffraction reported that the plastic yielding occurs first in austenite phase. In this study, to understand precisely the mechanical behavior of each phase, a dual-scale mechanical behavior of DSS was investigated by a nanoindentation and an in-situ tensile test in a high-resolution electron backscatter diffraction (HR-EBSD). Whereas the nanohardness value of austenite was mostly higher than that of ferrite, the incipient plastic deformation, which is usually measured as pop-in on load-displacement curve, occurred at lower stress in austenite comparing with ferrite. The pop-in stresses for both phases could be understood with the event of dislocation nucleation underneath the indenter. The lower pop-in stress of austenite could be explained as a result of its lower dislocation nucleation energy, which was calculated with the length of Burger&’s vector of dislocation measured by XRD and CBED. With in-situ tensile test in EBSD, the macro-scale tensile behavior including the effect of microstructure was observed. Orientation spread and misorientation distribution in each grain of austenite and ferrite were traced as the deformation proceeded. Based on these results, the mechanical behavior of each phase could be analyzed and compared with the small-scale nanoindentation behavior. According to the results of nanoindentation, hardness of austenite is higher than ferrite but plastic transition occurs as pop-in at lower stress on austenite than ferrite. And maximum shear stress underneath the indenter when pop-in occurs was around theoretical crystal yield strength. Result of XRD and CBED analysis, it was shown that pop-in occurs at lower stress on austenite than ferrite because austenite has lower dislocation nucleation energy in this specimen. And since austenite has higher hardening factor then ferrite, austenite has higher hardness value than austenite. This tendency of deformation was analyzed and also confirmed by in-situ EBSD.

The ALILE process enables fabrication of thin polycrystalline Si films at relatively low temperature (< 450°C) making it highly promising for applications in thin film photovoltaics. While the driving forces for the metal-induced crystallization are rather well understood the details of the materials transport during the layer exchange are largely unknown. In this work the microstructure of stacks of a-Si(100nm)/Al(50nm)/Quartz, annealed at 450°C, has been investigated at different length scales by combining optical microscopy, analytical SEM and analytical TEM [1]. The results indicate that the ALILE and crystallization reaction proceeds by forming 20-50µm wide dendritic “cells” with Al deficient centers. Excessive upward transport of Al by epitaxial growth out of the existing Al grains into the a-Si was observed in a rim of about 10 µm width around the cells and to a smaller extent even a few tens of micrometers away from the reaction front. Using in-situ TEM the lateral and vertical transport of Al at the expanding crystallization front could be directly visualized for the first time. We propose that Coble-type diffusion of Al along the Al grain boundaries and/or the Al/a-Si interface, driven by the compressive stress in the Al layer, is responsible for the massive long range lateral and vertical transport of Al. Our findings shed new light on the redistribution of Al and Si during the ALILE process. [3] [1] B. Birajdar, T. Antesberger, M. Stutzmann, E. Spiecker, pss (RRL) 5, 172 (2011) [2] B. Birajdar, T. Antesberger, B. Butz, M. Stutzmann, E. Spiecker, Scripta Materialia 66, 550 (2012) [3] The authors gratefully acknowledge financial support by the DFG via the Cluster of Excellence Engineering of Advanced Materials.

In-situ mechanical tests of amorphous carbon (a-C)/boron nitride nanotube (BNNT) and a-C/carbon nanotube (CNT) have been conducted to understand the mechanical performance of a-C as a welding material for load transfer between structures in nano-based structural materials. The experiments took place inside the vacuum chamber of a transmission electron microscope with an integrated atomic force microscope system, allowing nanomanipulation simultaneous to real time observation of the hybrid structures. Electron beam induced deposition (EBID) has been used for the deposition of a-C and modification of the structures. The pristine and failed structures were successfully welded with a-C, and a series of tensile, compressive, and lap shear tests were performed on the hybrid structures. The current work presents a-C welding as a viable method for the formation of stable tube-to-tube connections in boron nitride and carbon nanotube bundles and serves as a starting point for the improvement of the mechanical performance of nanotube based materials for space applications, where light-weight/low-budget/high-performance materials are sought after.

Uniform magnetic nanocrystals are synthesized via a templated growth in the presence of the iron-binding recombinant proteins, Mms6 and Mms13, respectively. Use of low-temperature synthetic approach permits control over size, shape, and orientation of the magnetic nanostructured crystals. The protein-templating mechanism is probed via numerous analytical techniques, including Transmission Electron Microscopy with the continuous flow Liquid Cell TEM Holder Platform, and magnetization measurements. Uniform magnetic nanocrystals can be further functionalized to address specific applications. Effect of the protein on the morphology and shape of resultant template magnetic nanometer-sized crystal is investigated.

Noble metal nanoparticles supported on reducible oxides are the top candidates as heterogeneous catalysts in various reactions such as CO oxidation and water gas shift reaction. Such supported catalysts show enhanced activity and durability and lower activation energy due to the metal-support interaction. As the catalytic property of the noble metal depends on the particle size, it is important to design stable catalysts with optimum size which is stable under the operating conditions of the catalysis. The thermal stability of the metals nanoparticles on oxide supports is a basic requirement for the use of these materials as catalysts as the operating temperature for many of the catalysis is high. The stability depends on the nature of the metal, support as well as the interface. In-situ transmission electron microscopy facilitates real time observation of the microstructural changes of the material. In this work, thermal stability of noble metals including Pt and Au nanoparticles on CeO₂ supports is investigated by in-situ heating in transmission electron microscope. CeO₂ with exposed (111) and (100) facets are investigated as supports. The experimental observations lead to the understanding of the kinetics of particle growth of metal nanoparticles on oxide supports and the effect of different crystal planes in stabilizing the noble metal nanoparticles.

Calcium carbonate is one of the most abundant building materials in biomineralization despite being a notably brittle compound. Numerous organisms are capable of constructing sophisticated and functional hybrid structures by the interplay of the mineral with an organic phase - mainly proteins. In order to understand this and to transfer the know-how to biomimetic mineral synthesis, it is important to elucidate the mechanism of interaction between organic and inorganic that lead to these hybrid materials and, in particular, how organics can be incorporated into the mineral. In this work, the negatively charged polyelectrolyte polystyrene sulfonate (PSS) is used to mimic the behavior of the negatively charged protein residues that influence calcium carbonate mineralization in e.g. mollusk shells. In addition, the sulfonate group in PSS is structurally similar to sulfate groups on polysaccharides that have been found to constitute the crystal nucleation site in these shells, therefore making PSS a suitable organic model. Wang et al. showed that the calcite crystallization utilizing the ammonium carbonate diffusion method in the presence of PSS yielded a family of well-defined mesocrystals — i.e. regular but porous scaffolds composed of well-separated, but almost perfectly 3D aligned calcite nanocrystals — on a glass substrate. Although the CaCO3 mesocrystal formation mechanism was suggested to proceed via an amorphous precursor, it remains unclear if this is the true pathway or if a different transformation process is followed through e.g. oriented attachment of nanocrystals or secondary nucleation on primary nanoparticles. Here we use both in situ TEM and in situ AFM to follow the formation process in detail. For in situ TEM a custom designed fluid cell was utilized, where the fluid containing CaCl2 and PSS was inserted via an access port between two Si3N4/Si(100)/Si3N4 wafers with 50-by-100 µm electron transparent Si3N4 membranes (~100 nm thickness). These two wafers were then bonded to and separated by a Si3N4 spacer nominally 300-400nm in thickness. A second access port was filled with solid ammonium carbonate, to slowly increase supersaturation by the decomposition of CO2 into the solution thereby inducing crystallization of CaCO3. Preliminary results using in situ AFM showed immediate adsorption of ~60-100nm PSS complexes on the Si3N4 wafer of the TEM fluid cell. In situ TEM typically revealed a core-shell structure of what appeared to be initial liquid-like or amorphous particles which, upon beam exposure after an hour-long reaction, nucleated an identical particle in direct contact with the original one. These results suggest that the initial mesocrystal formation might occur through a secondary nucleation of nanoscopic particles that build up into the eventual micron-sized mesocrystal obtained ex-situ. We detail the structural evolution and draw inferences about the formation mechanism of the resulting CaCO3 mesocrystals.

Abnormal grain growth is characterized by the development of a bimodal distribution of grain sizes during growth. The underlying mechanisms of abnormal grain growth caused by thermal annealing have been contributed to solute segregation, inclusions, and precipitation of new phases in multi-species materials. Arguably, in single component materials, the underlying mechanisms that contribute to abnormal grain growth are less understood. For these cases, the grain boundary character and its associated energy is considered to dominate which grains grow the fastest. In the present work, a series of elemental Cu, Ni, Fe and W thin films have been sputter-deposited and in situ annealed up to 750 deg. C in the TEM. The onset of abnormal grain growth has been quantified using precession enhanced diffraction based orientation mapping. By measuring the grain boundary misorientation and texture change of the film at different incremental temperature steps, the evolution of specific grains are tracked to reveal favorable near-neighbor grain boundary alignments. The results will be discussed in terms of comparing the two FCC and two BCC structures with and between each other. In addition, the experimental findings are used to refine Monte Carlo predictions of grain boundary misorientation distributions leading to a better understanding of the mechanisms which contribute to deviations from normal grain growth.

An o-ring sealed transmission electron microscopy (TEM) wet-cell with silicon nitride (Si₃N₄) windows was used to enclose a chlorauric acid (HAuCl₄) solution in the vacuum for the in-situ study. Gold nano structures were in-situ observed to develop in the HAuCl₄ aqueous solution under electron beam (EB) irradiation through wet-cell TEM. The nucleation, growth and coalesce of nanoparticles were observed in the EB irradiated region. The diameter of the nanoparticles ranges from 14 to 72 nm. Bubble-like materials from 6 nm to around 160 nm in diameter were also observed. Nanoparticles developed outside the EB irradiated region were found being larger than that inside the irradiated region. The Si₃N₄ windows were later separated to check the deposited structures in more detail. Electron diffraction from the nano features can be indexed by Au crystal. Gold was also observed by energy dispersive X-ray spectrum. More interestingly, nano liquid droplets were observed to fluctuate inside the bubble-like feature on the Si₃N₄ window under EB irradiation. Acknowledgments: The help from Prof. S. Dillon is greatly acknowledged. This work was supported by Shanghai Nano Project (11nm0507000) , Shanghai Leading Academic Discipline Project (B502), and Shanghai Key Laboratory Project (08DZ2230500).

Understanding the mechanical properties of nanomaterials is a subject of major interest in different fields of nanosciences and nanotechnologies. Particularly in microelectronics, metallic thin films used in interconnections are subjected to high current density and consequently high thermal loads. Thermal loads generate stresses through the differences in coefficients of thermal expansion with silicon substrate and oxides. Hence, determining the elastic to plastic transitions and the deformation modes becomes a critical issue to warrant device integrity. With the advent of in situ transmission electron microscopy (TEM)-nanoindentation, the mechanical behavior of small-scale samples can be investigated at critical interfaces, as plastic behavior can be triggered locally and the associated physical processes followed dynamically in the microscope. Here, we investigate the plastic behavior of passivated Al thin films. A dedicated TEM sample holder (Nanofactory Instruments), equipped with an actuator and a diamond mounted on a force-sensing MEMS device, allows indenting cross-sectional specimens that has been milled in a H-bar configuration using a focused ion beam (FIB). To study the dislocation behavior at the Al/SiO2 interfaces, the upper layer of silica, which is designed to distribute the stress imposed by the indenter and also to function as a protective layer during the FIB-milling process, is pressed into the diamond indenter. The in situ nanoindentation experiment was recorded on video with correlated force values. Finite element modeling helped converting the applied external load, into an estimation of the stress reaching the Al layer. The stress in the Al film of the specimen has also been estimated from the radius of curvature of dislocations. Both ways of estimating the stress in the Al film gave comparable results and served to estimate the yield stress of the Al film could. Especially, it was shown that dislocations tend to be captured by the Al/SiO2 interface. Pushed into this interface in compression, the dislocations did not go backwards. The result is comparable with in situ thermal cycling of similar samples, where the stress originates from difference in thermal expansion coefficients [1], except that here, thermal diffusion is reduced. To support the discussions, finite element models are built to calculate the transmitted stresses into the Al film by the indentation process, as well as to estimate the thermal stresses in the heating experiments [2,3]. References [1] M Legros et al, Acta Materialia 50 (2002), p. 3435. [2] P Mullner and E Arzt, MRS Proceedings 505 (1998), p. 149. [3] This work was supported by the EU through the ESTEEM project (Grant No. IP3: 0260019), and by the French National Agency for Research (ANR PNANO/HD STRAIN Project No. ANR-08- NANO-0 32).

Ostwald ripening that large particles grow bigger by consuming the smaller species is a key coarsening mechanism for a variety of physical and chemical processes. It plays an important role in the synthesis of monodisperse nanoparticles, geological rock texture formation and a variety of industrial reactions. However, Ostwald ripening often leads to undesirable consequences in many applications of nanoparticles thus it is desirable to limit this process by controlling the kinetic parameters. We study the growth of Bi nanoparticles in an engineered precursor-scarce environment in a liquid cell at an elevated temperature (180 degree C) using transmission electron microscopy (TEM). Observation reveals dynamics of oscillatory growth of individual nanoparticles, pairwise Ostwald ripening and anti-Ostwald ripening and a global collective oscillation. We achieved quantitative tracking of the volume trajectories of more than 30 nanoparticles with subsecond time resolution. It allows us to quantify the growth and dissolution rates of each individual nanoparticle, and build distance-dependent correlations between each pair of particles. By analyzing the pair correlation, we identified that a mass-transport zone is present around each particle, which couples to the observed growth kinetics. This study shed light on a new route for system engineering to reverse particle coursing by Ostwald Ripening. HZ thanks the funding support from U.S. DOE Office of Science Early Career Research Program.

Nanostructured surfaces with special wetting properties have potential to transform number of industries, including power generation [1], water desalination [2], gas and oil production [3], and microelectronics thermal management. The special wetting properties of these surfaces stem from the interaction of solid, liquid, and gas phases in the volume linking the droplet/bubble to the underlying substrate [4]. However, experimental determination of the geometry these complex interfaces has been limited by lack of appropriate imaging techniques. Here we demonstrate that imaging of such interfaces can be achieved using sample plunge freezing in liquid nitrogen slush, cryogenic temperature selective Focused Ion Beam (FIB) milling and SEM imaging. Plunge freezing is an established technique for preserving geometry of hydrated biological [5] and geological [6] specimens as well as colloidal suspensions and emulsions [7] for electron microscopy. We show that microscale water droplets condensed on a variety of substrates preserve their morphology during the plunge freezing process. After transfer into the microscope chamber, the frozen droplets can be selectively sectioned using FIB milling and imaged using SEM. We show examples of composite interfaces involving solid, liquid, and gas phases below water droplets condensed on superhydrophobic surfaces consisting of various nanostructures [8] as well as hybrid liquid-solid substrates [9]. We also discuss imaging artifacts and opportunities for 3D “destructive tomography”, elemental, and phase imaging of the interfaces. References: 1.(a) Dietz, C., et al., Appl. Phys. Lett. 2010, 97,033104; (b) Chen, R., et al., Nano Lett. 2009, 9 548. 2.Humplik, T., et al., Nanotech. 2011, 22,292001. 3.Smith, J. D., et al., Phys. Chem. Chem. Phys. 2012, 14, 6013. 4.(a) Cassie, A. B. D., et al., Trans.Farad.Soc. 1944, 40, 546; (b) Wenzel, R. N., Ind. Eng. Chem. 1936, 28, 988; (c) Patankar, N. A., Soft Matter 2010, 6, 1613; (d) Israelachvili, J. N., Intermolecular and Surface Forces. 3rd ed.; Elsevier: San Diego, 2011. 5.Dobro, M. J., et al., Meth. Enzymol. 2010, 481,63. 6.(a) Mikula, R. J., et al., Coll. Surf. A-Physicochem. 2000, 174, 23; (b) Boassen, T., et al., In Int. Symp. SCA, Vol. SCA2006-43 pp 1-6. 7.Binks, B. P., et al., Phys. Chem. Chem. Phy. 2002, 4, 3727. 8.Rykaczewski, K., et al., Soft Matter 2012. 9.(a) Wong, T.-S., et al., Nature 2011, 477 443; (b) Lafuma, A., et al., Europhys. Lett. 2011, 96, 56001.

As the device dimensions are scaled down, the current density in the metal interconnects increases. Under a high current density of ~106 A/cm2, voids and hillocks may form in the interconnectors due to the mass transport induced by electromigration, which leads to device failure. It is one of great technological importance to understand electromigration failure in thin film interconnects. Electromigration induced mass transport, in principle, can take place along different paths, such as lattices, grain boundaries, surfaces, and interfaces, which all have distinct activation energies of electromigration. Although most of the studies have shown that the Cu/Si3N4 interface is the dominant electromigration path, some have reported dominant path is the grain boundary, where it is proposed that the electromigration on Cu interconnects happens by coupling between grain boundary and interface diffusion. In order to understand the electromigration mechanism, in-situ electromigration sequence was observed under transmission electron microscopy (TEM). Single-level Cu lines with/without the Si3N4 capping layer on Si3N4 membrane were used as electromigration test structures. Firstly, a 50 nm-thick Si3N4 membrane was deposited onto the SiO2/Si substrate by a low pressure chemical vapor deposition. And then KOH preferred etching of Si substrate was carried out to make self-supporting Si3N4 membrane, with dimension of 2400mu;m×1200mu;m. On the Si3N4 membrane, Cu film of 50nm thickness was deposited by a ultrahigh vacuum DC magnetron sputtering, and patterned into a line width of the 3mu;m wide by photolithography and wet etching. To understand the effects of capping layer on electromigration mechanism, Si3N4 films of thickness 20nm was deposited as a dielectric capping layer on 3mu;m wide patterned Cu lines. Using a specially designed in-situ TEM stage, electric current was applied to the both ends of Cu line. Electromigration induced microstructural changes in surface and grain boundary of Cu line were recorded by in-situ TEM in real time under high current density of 1×106 A/cm2. Dominant path of electromigration of a single-level Cu line with/without the Si3N4 capping layer will be presented. This research was supported by the Nano-Material Technology Development Program (the Green Nano Technology Development Program) through the National Research Foundation of Korea funded by the Ministry of Education, Science and Technology (2011-0019984).

Transmission electron microscope (TEM) is usually taken as a tool to characterize the microstructures of the materials before/after deformation test. By incorporating a miniaturized load/depth transducer into a TEM holder, Hysitron has recently developed a novel and unique nanomechanical apparatus which enables one to acquire quantitative mechanical data while simultaneously recording the microstructure evolution of the materials during deformation, developing a one-to-one relationship between imposed stress and individual deformation event. This presentation will report the current progress of this unique technology as well as some of its applications. It was found that prior to the compression tests, the nickel pillars fabricated through Focused Ion Beam (FIB) were observed to contain a high density of defects. However, quite unexpectedly, the dislocation density was observed to decrease dramatically during the deformation process and, in some cases, even resulted in a dislocation-free crystal. The phenomena, which we termed as “mechanical annealing”, is the first direct observation of the dislocation starvation mechanism and sheds new light on the unusual mechanical properties associated with submicron- and nano- scale structures. However, due to the dislocation core structure difference between the screw and edge dislocations in body centered cubic (BCC) metals, it was believed that BCC metals are incapable of mechanical annealing and the sample size strengthening behavior should be less that in FCC metals. By in situ compression of nanopillars inside a transmission electron microscope, we demonstrate that with the pillar diameter decreasing to hundreds of nanometers, significant mechanical annealing does occur in BCC Mo. In addition, there exists a critical size (DC ~ 200 nm for Mo at room-temperature) below which the strengthening exponent in Hall-Petch like regression increases dramatically to that similar to FCC metals. We attribute the observed phenomena to the diminishing importance of lattice friction at high stresses, when the size-enhanced flow stress exceeds a single screw dislocation&’s lattice friction.

Microscale bending experiments were conducted inside scanning electron microscope using a nanoindentation device (PI 85 PicoIndenter, Hysitron, Inc.) to understand fracture properties of diffusion aluminide bond coatings (PtNiAl). The compositional gradient as well as the microstructural variation along the thickness direction makes diffusion aluminide bond coats an ideal candidate material for microscale fracture toughness study. Two different microbeam geometries, double cantilever beam and doubly clamped beam, were manufactured from the bond coat region using a focused ion beam. The crack initiation and propagation characteristics with varying Pt content and changing Ni:Al ratio were observed and the corresponding load-displacement data were obtained. While the clamped microbeams are used to obtain initiation fracture toughness of each individual zone of interest, the double cantilever beam specimen has the potential to determine propagation toughness and R-curve behavior across the entire coating thickness in one single test. Also, the double cantilever specimen is known to be stable under displacement control, whereas the doubly clamped beam is shown to be stable even under load control beyond a critical crack length to width (a/W) ratio for the beam dimensions used. The coating exhibits an R-curve behavior even though the bulk (Pt,Ni)Al itself is inherently brittle.

Polymeric electrospun scaffolds have been used in tissue engineering due to their microstructures, which mimic the fibrous networks in natural biological materials such as cartilage and blood vessels. An understanding of the failure of fibrous networks can not only facilitate the production of electrospun scaffolds with improved toughness, but also provides insight for treatment of diseases and conditions that involve soft tissue failure. Despite the importance, an understanding of the toughening mechanisms in fibrous materials, particularly at small scales, is still incomplete. The failure of polycaprolactone (PCL) electrospun scaffolds under mode I loading was experimentally studied at both macroscopic and microscopic length scales and compared with that of other nonwoven polymeric materials. Uniaxial tensile tests and fracture tests were first performed on PCL electrospun scaffolds. The detailed toughening mechanisms at the notch front were then examined by performing in-situ fracture testing of PCL scaffold in the scanning electron microscope (SEM). In the fracture tests, the scaffolds stretched more than a hundred percent without significant crack propagation. Blunting and necking occurred in the vicinity of the notch root, forming a region of intense deformation ahead of the notch. The examination of this notch region in the SEM showed that the randomly oriented fibers rearranged and formed parallel fiber bundles. The formation of fiber bundles increased during necking. These fiber bundles, which aligned parallel to the loading direction, resisted crack propagation. The understanding of toughening mechanisms presented here is to not only for the particular case of PCL electrospun scaffolds, but also for other fibrous materials, including ductile polymers. This study offers guidelines for the production of fibrous materials with enhanced toughness.

Magnesium is a lightweight metal that would find widespread use for structural applications if its room temperature formability can be enhanced. Basal slip has much lower CRSS than prismatic and pyramidal slip in Mg, but the two independent basal slip systems cannot satisfy the arbitrary shape change during complex deformation encountered in the shaping of any component. Therefore, deformation twinning often becomes important deformation modes. Recently we have run a systematic series of in situ TEM mechanical tests on pure Mg oriented for basal slip and deformation twinning, respectively, where we have quantitatively measured and characterized the basal slip behavior and the deformation twinning behavior. Strong crystal size effects on both deformation mechanisms were found. More importantly, the surface nucleation mechanism became increasingly significant at extremely small scales. Both the dislocation and deformation twinning behavior changed during plastic deformation in small-scale samples, resulting in high strength and also high ductility. The mechanism of external dimension refinement was further compared with internal dimension refinement where grain boundaries might be engineered to reduce plastic anisotropy and enhance both the strength and ductility in Mg. [Nano Lett. 12 (2012) 887]

Deformation twinning is an important plastic deformation mechanism in Hexagonal Close Packed materials (HCP) due to the lack of sufficient number of slip systems to accommodate general plastic flow by slip alone. It is evident from the recent literature that twinning can enhance the ductility of HCP materials and also has significant influence on the texture evolution in polycrystalline Magnesium (Mg) and its alloys. Hence, it is important to understand the propagation and growth of twins during plastic deformation. Most of the studies conducted hitherto are ex-situ and focused on understanding the influence of twinning on the stress-strain behavior. However, a detailed characterization of twinning during deformation is still lacking. We conduct in-situ micro tensile experiments on a polycrystalline Mg to understand the deformation twinning. Dog-bone shaped, micro-tensile specimens are made from pre-twinned samples. Preliminary results from in-situ experiments indicate that a large amount of twins are generated at the grain boundaries. A significant amount of offset, in the thickness direction, is noticed at the grain boundaries whose surrounding grains deform by twinning. The local stress-strain behavior is estimated using digital image correlation technique and correlated to the microstructural changes that occur during deformation.

Material processes subject to extreme driving forces and conditions far from equilibrium inherently occur on very short time scales, ranging from the femtosecond-scale processes occurring from non-equilibrated electron states to microsecond transient events of phase transformations and deformation processes. Typically, we are confined in the laboratory to conduct experiments near equilibrium or observe the material&’s state post process due to the resolution limitations of conventional analytical techniques. Though insight about material&’s behavior is gained from these observations, much of coupled and convoluted events of complex processes on short time scales not well understood that require technique that both high spatial and temporal resolution to observe nanoscale microstructural features evolving on short timescales. In effort to meet the need for studying fast dynamics and transient states in material processes, we have constructed a nanosecond dynamic transmission electron microscope (DTEM) at Lawrence Livermore National Laboratory to improve the temporal resolution of in-situ TEM observations. Prior DTEM hardware only allowed single-pump/single-probe operation, building up a process's typical time history by repeating an experiment with varying time delays at different sample locations. Movie Mode DTEM upgrade now enables single-pump/multi-probe operation. These technical improvement provide the ability to track the creation, motion, and interaction of individual defects, phase fronts, and chemical reaction fronts, providing invaluable information of the chemical, microstructural and atomic level features that influence the dynamics and kinetics of rapid material processes. For example, the potency of a nucleation site is governed by many factors related to defects, local chemistry, etc. While a single pump-probe snapshot provides statistical data about these factors, a multi-frame movie of a unique event allows all of the factors to be identified and the progress of nucleation and growth processes can be explored in detail. It provides unprecedented insight into the physics of rapid material processes from their early stages (e.g. nucleation) to completion, giving direct, unambiguous information regarding the dynamics of complex processes. This presentation will discuss the technical aspects of the Movie Mode DTEM technology in the context of recent material science studies using the novel in situ TEM capability. Work preformed at LLNL under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344 and supported in part by the U.S. Department of Energy, Office of Basic Energy Sciences, Division of Materials Sciences and Engineering.

The melting and solidification of metals and alloys is a ubiquitous manufacturing process used to fabricate components for an enormous number of applications. The properties and performance of these components is dictated by the phases and final microstructure of the solidified material, which is determined in turn by the thermal and compositional conditions that exist during solidification as well as thermodynamic and kinetic constraints of the materials system. During rapid solidification, pronounced deviations from equilibrium are known to occur [1-3], leading to the formation of unique microstructures and metastable phases with potentially useful properties. In-situ experiments that directly image the evolving microstructure and advancing solid-liquid interface during the solidification process can provide insight into the mechanisms and kinetics governing rapid solidification, leading to eventual improved control over component properties. Using the dynamic transmission electron microscope (DTEM), Al-Cu thin films were pulsed-laser-melted and the solidification of the alloys was investigated with in-situ time-resolved imaging. The Al-Cu system has been widely investigated and its thermodynamic and physical properties as well as regular eutectic (Al-Al₂Cu) are well defined [4], providing a model system for alloy solidification studies. Here, we present results from both time-resolved DTEM experiments and post-solidification characterization of multiple compositions of Al-Cu thin films (hypo-, hyper-, and eutectic alloys), and the effects of Cu content on the resulting microstructures, phase formation, and kinetics will be discussed [5]. References [1] Zimmermann, M. et al. Acta Metall., 37 (1989) 3305. [2] Jones, H. Mater. Sci. Eng. A, 137 (1991) 77. [3] Kurz, W. and Gilgien, P. Mater. Sci. Eng. A, 178 (1994) 171. [4] Murray, J. L. Int. Metals Rev., 30 (1985) 211. [5] This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory and supported by the Office of Science, Office of Basic Energy Sciences, Division of Materials Science and Engineering of the U.S. Department of Energy under Contract No. DE-AC52-07NA27344. Work performed at the University of Pittsburgh was supported by the National Science Foundation, Division of Materials Research, Metals & Metallic Nanostructures program through Grant No. DMR 1105757.

Sintering is usually considered as a key-step in the preparation of ceramics as it drives both density and microstructure of the final materials. Since such parameters could strongly influence the physico-chemical properties of materials during their life-cycle, the sintering of CeO₂ and ThO₂ pellets, as model compounds for the next generation of nuclear fuels, was investigated through an original approach based on the combination of dilatometric measurements and in situ HT-ESEM observations. On the one hand, the use of environmental microscope allowed observing in situ the behaviour of the compact during heat treatments between 1000°C and 1400°C. The subsequent image analysis led to the determination of original local kinetics parameters (i.e. at the grain scale), including grain boundaries mobility and individual grain growth rates. Such observations confirmed the general behaviour expected from theoretical models and numeric simulations but also evidenced unusual phenomena such as closed porosity elimination through the surface. When considering a larger population of grains, global kinetics parameters (i.e. at the pellet scale) were assessed. Particularly, the average grain size was plotted versus the holding time for the different temperatures considered. In these conditions, the nature of the mechanism controlling the global sintering rate was pointed out and the energy of activation corresponding to grain growth processes was evaluated, in good agreement with the data previously reported in literature. On the other hand, analogous experiments were performed by dilatometry and allowed monitoring both linear shrinkage and density of the samples. First, the value of the energy of activation obtained by HT-ESEM was confirmed by the use of the Dorn's method. Also, in the case of ThO₂, the combination of the two sets of data led us to establish for the first time a sintering map of this compound. This latter clearly evidenced two different zones in the evolution of the pellet microstructure driven by densification (d/d_calc. le; 92%) then by grain growth (d/d_calc. ge; 92%). The combination of HT-ESEM observations and of dilatometric measurements then appeared as a very rapid and reliable way of investigation of ceramics sintering which can be used efficiently to monitor the final microstructure of dense materials.

With electron beams smaller than the bond length of hydrogen, aberration-corrected scanning transmission electron microscopes (STEM) can currently image materials with resolutions below the shortest bond length in nature. However, these atomic resolution images are only 2D projections of a specimen. In order to determine the full 3D structure, one must acquire a series of STEM images over a range of specimen tilts. Unfortunately, the resolution of a 3D STEM tomogram is diminished by missing information from the restricted specimen tilt range and finite tilt increments. This typically limits volumetric resolutions of electron tomography to roughly 1 nm—twenty times worse than the best resolution in 2D projections. In addition, high-resolution (< 1nm) tomography of extended objects, to date, has not been possible. The sub-Ångstrom resolution of aberration-corrected electron microscopes is accompanied by a strongly diminished depth of focus, causing regions of “large” specimens (> 5nm) to appear blurred or missing. High-resolution 3D reconstruction of extended objects requires collecting information beyond a traditional tilt series. Here we present through-focal tomography that combines depth sectioning and traditional tilt-tomography to reconstruct extended objects, with high-resolution in three-dimensions. This novel technique fills in the missing 3D information by acquiring a through-focal image series at each specimen tilt. A through-focal tomographic reconstruction of porous PtCu nanoparticles used in the proton exchange membrane fuel cells (PEMFCs) revealed a nano-pore network throughout many particles on an extended Vulcan carbon support (~390nm). Experimental and simulated comparisons of through-focal tomography with traditional tomography methods reveal improved contrast and resolution over large fields of view.

Graphene, which is a single atomic layer of carbon atoms bonded in a hexagonal lattice, is among the few materials that are stable in two dimensions. Controlling the structure of such a thin material is difficult during synthesis or exfoliation so a post treatment may be the most viable option. In our studies, we show that graphene can be modified using a high temperature anneal and electron beam irradiation. For edge modification, a graphene flake was annealed from 400 to 1200 degrees Celsius while being examined under an electron microscope. As the temperature was stepped up to higher values, contaminants disappeared from the graphene lattice. At the highest temperatures, the graphene edges were observed to reconstruct into atomically straight bi-layer edges. This only occurred in areas where there was a concurrent electron beam irradiation. Without the sputtering of the electron beam, the annealed graphene sees little to no reconstruction. Our studies also show that the same mechanisms can be used on nanopores within the lattice. By focusing the electron beam onto the graphene, nano-pores or -holes were drilled into its structure. Using temperatures between 400 and 1200 degrees Celsius, the nano-pores were expanded or shrunk. Small diameter nano-holes were able to shrink completely, closing off the nano-pore, when being both annealed and bombarded by electrons. This gives a mechanism for controlling the final pore diameter by removing or blocking the beam exposure at the desired size. These simple post-synthesis treatments pave the way for scientists to modify the structure of graphene which can lead to tailoring their device characteristics. This work was supported by SWAN (GRC-NRI), AOARD-AFOSR (FA2385-10-1-4066) and the World Class University Program (by MEST through NRF (R31-10026)).

Detecting nano-scale temperature and temperature distributions is important for studies of heat generation and transfer in a wide range of engineering systems, such as microelectronic, optoelectronic and micromechanical systems. While in the past few decades, much progress has been made in the area of nano-scale temperature mapping techniques, no current method adequately combines high spatial resolution, high temperature sensitivity, and an ability to work in non-contact mode (such that the local temperature distribution is not perturbed by heat diffusion between the contact probe and the sample surface). In this presentation, we introduce a new nano-scale resolution non-contact in-situ temperature measurement technique (which we call thermal scanning electron microscopy, ThSEM). This technique is based on temperature dependent thermal diffuse scattering in electron backscatter diffraction (EBSD) in a scanning electron microscope (SEM). Unlike scanning thermal microscopy (SThM), which uses a contacting probe, ThSEM is a non-contact method and in contrast to optical temperature mapping techniques, ThSEM doesn&’t have the spatial resolution limitation that arises from the optical wavelength. While in the current work, a spatial resolution of less than 100 nm and a temperature resolution better than 10 °C are attained using a thermionic source SEM, it is possible for ThSEM to reach much higher temperature resolution and spatial resolution (< 10 nm) using field emission SEMs operating at lower beam energy. ThSEM preserves many of the merits of SEM and can zoom over a broad range of fields of view. Also, the hardware setup is very similar to the EBSD system in an SEM, which makes the integration of temperature mapping into SEM relatively straightforward. Moreover, multiple signals or contrast mechanisms, such as temperature distributions, grain orientation maps, topographic images, and elemental maps could be obtained from the same sample area simultaneously depending on the specific SEM capability. This technique thus adds a new channel - the temperature signal - to the collection of existing SEM signals. In this presentation, the feasibility of ThSEM for nano-scale temperature measurement is examined, the temperature dependencies of Kikuchi line intensities for a Si sample are presented and explained quantitatively using the Debye-Waller factor (DWF), parameters affecting temperature sensitivity are analyzed and optimized, and the spatial resolution of this technique is examined.

The mechanical deformation of CNT arrays (also called forests) is a complex, multi-scale phenomena and may be directly coupled to the performance of many applications including strain sensing and highly efficient interfaces. Numerous nanoindentation and compression studies have provided quantitative characterization of the mechanical properties of CNT arrays such as modulus and yield strength. The value of these techniques, however, are restricted in scope as they provide limited insight into CNT interactions and deformation mechanisms. Recently, in situ SEM compression testing has facilitated the direct observation of CNT arrays deformation and has provided valuable insight into the initiation and propagation of coordinated buckling. Even with the significant advance in capability afforded by in situ testing, the knowledge gained has been largely phenomenological. As a result, the in situ method alone is incomplete as a tool for describing CNT array mechanics and for quantitatively validating mechanical models. We demonstrate the use of 2-D digital image correlation (DIC) to spatially resolve strain and displacement of patterned CNT array columns based on in situ SEM compression sequences. The technique utilizes the native grayscale pattern established by the individual CNTs and CNT bundles within an array as a traceable speckle pattern for DIC analysis. In such a way, spatially resolved strain maps of CNT columns are generated based upon the motion of constituent CNTs for column widths of up to 100 mu;m and lengths of up to 75 mu;m. Upon validation of the DIC technique, we evaluated the compression of CNT columns with square cross sections having aspect ratios (defined as column length / column width) ranging from approximately 0.25 to 6.0. The columns deform in one of three distinct modes, which vary as a function of column aspect ratio. These modes include crushing, bending, and bottom-up buckle accumulation. In spite of the significantly different appearance of these deformation modes, DIC analysis reveals a consistent CNT array yield criterion of 5% local compressive strain for each column. Local strain in excess of 5% initiates coordinated buckling. Strain maps show that strain is highly non-uniformly distributed throughout the compression sequence. Rather, lateral bands of enhanced compressive strain form at relatively low global strain and grow in magnitude and dimension with increased global strain. Buckle formation generally initiates from within these bands. Outside of the localized bands, strain is significantly lower than the global applied strain by as much as an order of magnitude. Further application of in situ SEM DIC to CNT arrays and similar material systems is expected to significantly expand understanding of the mechanics of nanoscale hierarchical materials and provide a quantitative foundation for comparison to mechanical models.

Investigation of the mechanics of natural materials, such as spider silk, abalone shells, and bone, provided great insight into the design of materials that can simultaneously achieve high specific strength and toughness. Research has shown that their emergent mechanical properties are owed in part to their specific self-organization in hierarchical molecular structures, from nanoscale to macroscale, as well as their mixing and bonding. Following this inspiration we have addressed the design of carbon nanotube (CNT) based fibers and yarns by applying lessons learned from mulitscale experiments and simulations across multiple lengths scales. Carbon nanotubes (CNTs) are envisioned to be ideal building blocks in hierarchical macroscopic composite fibers due to their extraordinary strength and stiffness. Macroscopic materials based on CNTs, however, have been limited by weak shear interfaces between adjacent CNT shells and matrix elements. Initial studies have demonstrated that double-walled CNTs (DWNT) are very attractive building blocks for macroscopic high CNT density fibers. Here we present experimental-computational studies of shear interactions within hierarchical DWNT yarns that have furthered the understanding of a number of key mechanical mechanisms which contribute to the ultimate behavior of yarns. At the individual bundle level we have studied the shear behavior within DWNT bundles through in-situ SEM/TEM mechanical testing coupled with Molecular Mechanics (MM) and Density Functional Theory (DFT) modeling. In-situ SEM pullout experiments conducted on DWNT bundles revealed the typical sword-in-sheath failure mechanism and allowed quantification of the force required to shear a small inner bundle of DWNTs out from an outer sheath of DWNTs. In this study a normalized pullout force of 1.7 +/- 1.0 nN/CNT-CNT interaction for sliding of a smaller inner bundle of DWNTs out of a larger outer shell of DWNTs. Through comparison with MM and DFT simulations of sliding between adjacent CNTs in bundles it was identified that factors contributing to the pullout force included the creation of new CNT surfaces, carbonyl functional groups terminating the free ends, corrugation of the CNT-CNT interaction, and polygonilization of the CNTs in the bundle. In addition a top down analysis of the experimental results revealed that greater than one half of the pullout force was due to dissipative forces. This finding of behavior at the CNT bundle level significantly differed from the behavior of pullout in individual MWNTs for which dissipation is found to be negligible.

Understanding the interfacial stress transfer between carbon nanotubes (CNTs) and polymer matrices is of great importance to the development of CNT-reinforced light-weight and high-strength polymer nanocomposites. In this talk, we present our recent experimental work on studying the interfacial strength between Poly(methyl methacrylate) (PMMA) and the embedded individual CNTs. The interfacial strength of the CNT-PMMA interface is characterized using in-situ nanomechanical single-tube pull-out techniques. By pulling out individual tubes from the polymer matrix using atomic force microscopic force sensors inside a high resolution scanning electron microscope, the pull-out force and the embedded tube length are measured with resolutions of sub-nN and nm, respectively. Our results reveal the dependence of the pull-out force on the embedded tube length and identify the critical embedded tube length that leads to the maximum pull-out force. Our results show a mean interfacial fracture energy of ~0.175 J/m2 and interfacial shear stress of ~ 45 MPa for the CNT-PMMA interface. Our data on the CNT-PMMA interfacial strength are in reasonably good agreement with both the theoretical predictions by molecular dynamics (MD) simulations and those experimental data reported on other similar types of non-bonded nanotube-polymer interfaces (e.g. CNT-epoxy). Our results clearly demonstrate the shear lag effect in pulling out polymer-embedded nanotubes. Our in-situ single-tube pull-out experimental methodology can be readily extended to study the interfacial strength of both non-bonded and bonded interfaces formed by individual nanotubes and a variety of polymer matrices.

Grain boundaries are well known to influence the mechanical behaviour of materials. For polycrystalline metals strengthening by the Hall-Petch mechanism is a well-established concept. However, the strength of individual grain boundaries against dislocation plasticity, i.e. their barrier strength, remains still obscure. Recent advances in micro- and nanomechanical testing permit now to select individual boundaries and to probe their strength with a high spatial resolution. Additionally, the electron microscopy observations provide insights into the deformation mechanisms. In the present talk recent developments for miniaturized deformation experiments at room temperature and up to 400°C will be reported and results addressing the mechanical properties and deformation structure of individual grain boundaries will be presented.

Understanding the deformation mechanisms in nanocrystalline metals and alloys is crucial for improving their performance and stability as needed for technical applications. During the last couple years our understanding of the deformation mechanisms of nanocrystalline metals has improved a lot mostly based on in-situ deformation experiments using XRD. However, it is difficult to see and understand the local processes based on these bulk measurements. The local processes are typically investigated using classical BF/DF-TEM. However, varying contrast contributions due to local orientation, bending and defects make an accurate interpretation for nanometer sized grains difficult. This becomes particularly challenging during in-situ straining experiments as the grains are deforming and reorienting. Here we show a novel approach, combining modern TEM techniques with in-situ straining to follow the full crystallographic orientation of each individual grain in the field of view with nanometer resolution through the steps of increasing strain or cycles of straining. Hysitron&’s TEM Picoindenter is used to strain TEM lamella or thin films with nanometer precision, providing the corresponding stress-strain curves. Together with the Nanomegas orientation mapping system [1] operating on a Tecnai F20 in µP-STEM mode, it opens the possibility to obtain orientation maps with nanometer resolution at selected states during the straining and at the same time, follow the deformation continuously using fast STEM imaging of the area of interest. Data processing using the Mtex Toolbox for quantitative texture, grain size and orientation analysis is the next step towards data interpretation [2]. It enables good identification of the crystallographic orientation of all grains and sub-grains and the quantification of twins and other special boundaries. First investigations were conducted on magnetron sputtered Au and Pd samples. To avoid straining during handling and preparation of the TEM lamellas, the metals were directly sputtered onto TEM holey carbon grids with a 2 nm carbon layer and transferred to a Hysitron push to pull device using a DualBeam FIB. Straining of this film inside the TEM allowed us to follow the deformation processes in ultra-fine grained Au up to 9.5% strain before rapture of the sample. The analysis reveals a complex mixture of processes taking place during straining including growth of individual large grains by slowly ‘eating&’ smaller ones, twining/detwining and grain rotation (Figure 1). The results will be discussed in terms of the deformation mechanism and how it is related to the local grain constellation. References: [1] Rauch, E.F. et al., Zeitschrift für Kristallographie, pp. 103 225 (2010). [2] Bachmann, F. et al., Solid State Phenomena, pp 63, 160 (2010). [3] Support by the Deutsche Forschungsgemeinschaft (FOR714) is gratefully acknowledged.

Nanoindentation at elevated temperature is an increasingly popular area of research [1-3] with a several custom systems at various institutions and variety of manufacturers offering standard options for elevated temperature testing. These encompass a range of technical solutions for performing tests: indenter/sample heating, water cooling, heat shields, inert gas shrouding, vacuum chambers, etc. These all attempt to circumvent the challenges of thermal displacement drift and indenter/sample oxidation. A brief summary of these technical solutions and how they address these challenges will be made, specifically with regards to the modifications made to an in situ nanoindentation system for testing at elevated temperatures up to 500°C within an SEM. Procedures for the calibration of the tip temperature via Raman spectroscopy and precision thermocouple measurements will be discussed, and the mechanisms of thermal drift will be discussed as revealed via careful contact thermometry. Thermal drift is shown to be a minimum when the indenter and sample are at an isotherm, which is validated by direct thermal measurements. A wide variety of materials display interesting mechanical behaviour at elevated temperatures. By using an in situ SEM elevated temperature indenter, a unique ability of coupling observation and measurement of mechanical deformation has been achieved. In addition to standard reference materials such as fused silica, pure aluminium, and tungsten, results from several classes of materials will be shown as case studies: silicon, GaN nano-prisms and lead-free solders. References [1] N.M. Everitt et al. Phil. Mag. 91, 1221-1244 (2011). [2] Trenkle, Packard, and Schuh. Rev. Sci. Instrum. 81, 073901 (2010). [3] Zhi Chao Duan, A. M. Hodge, JOM, 61, p32-36 (2009).

Thin film components of conventional and flexible solid-state devices experience mechanical strain during fabrication and operation. At the bulk scale, small values of strain do not affect thermal conductivity, but this may not true for grain sizes comparable with the electron and phonon mean free paths and for higher volume fraction of grain boundaries. To investigate this hypothesis, we developed a MEMS based experimental thermo-electro-mechanical characterization platform. Using this setup in-situ TEM characterization of thermal and electrical conductivity of nominally 100-nm-thick aluminum films (average grain size 40 nm) is carried out as a function of tensile strain, using a 3-omega; technique. The analysis shows that mechanical strain decreases the mean free path of the thermal conduction electrons, primarily through enhanced scattering at the moving grain boundaries. This conclusion is supported by similar effects of mechanical loading observed on the electrical conduction in the nanoscale aluminum specimens.

High strength low alloy steels (HSLAs) are used in many applications, such as ship and automotive construction, energy production and transmission and structural applications. These alloys utilize small additions of alloying elements to modify recrystallization kinetics and control the final microstructure. The effects of many of these elements have been studied singly, but their synergistic interactions within more complex alloys are not well understood. Niobium (Nb) containing microalloyed HSLAS alloyed with molybdenum (Mo) and vanadium (V) are both known to retain strength at high temperature better than other HSLAS, but Mo is significantly more potent in this role. In this work, in situ TEM tempering (annealing) of Nb-based HSLAs containing either Mo or V additions is presented.. These experiments provide a direct measurement of the effects of Mo and V on Nb alloyed HSLAs, allowing microstructural evolution to be connected to thermal processing schedules . This data allows for the quantification of the effect of Mo and V on precipitation, grain growth and phase transformation kinetics. The in situ experiments are compared to theoretical calculations of the kinetic processes during annealing. This combinatory approach of in situ TEM processing and theoretical calculation yields a deeper understanding of phase transformation phenomena critical to the microstructural control and final mechanical properties in new HSLAs alloys.

Although nanograined materials are known to exhibit many enhanced properties relative to their coarse-grained counterparts, a reliable prediction of their behavior under external influences such as thermal energy requires an understanding of their unique microstructure. For the case of pulsed laser deposited (PLD) nanograined Ni films, previous work using in-situ heating transmission electron microscope (TEM) experiments revealed abnormal grain growth, complex defect structures, and evidence of metastable hcp grains that compete with stable fcc phase grains. A detailed understanding of these changes proved difficult to address using traditional TEM techniques, thus a newly enabled combination of in-situ heating TEM and precession electron microscopy was pursued. This approach allowed for dynamic observation of local microstructural changes together with quantitative data acquisition, as well as characterization of changes in grain size, phase, and texture during annealing of PLD nanograined Ni films. In the current work, in-situ TEM heating experiments were performed on nominally 25 nm thick PLD nanograined Ni films, constrained by the 5 nm SiNx substrate of the heating stage, for a total period of 300 s at 700 °C. Video acquisition during heating captured the formation of an abnormal grain at 700 °C and tracked its evolution in real time. During these experiments, precession data was acquired at room temperature, and at 120 s and 300 s of annealing. The key observations from this set of experiments are as follows: (i) the as-deposited Ni film exhibited a mixed out-of-plane texture of <101>_fcc, <001>_fcc , <112>_fcc and <111>_fcc that proceeded towards an equilibrium <111>_fcc texture with the formation of abnormal grains, (ii) the abnormal grain possessed a texture, (iii) the average fcc phase grain size increased from 28 nm to 44 nm, and (iv) fcc phase grain growth was accompanied by a concomitant increase in the hcp phase grain size (8 nm to 12 nm) and phase fraction (1% to 8.5%). These results will be compared to in-situ TEM heating/precession experiments performed at 550 and 600 °C, and will be discussed in the context of the existing microstructure evolution theories. This research was funded by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, Division of Materials Sciences and Engineering. 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.

This talk presents some of our recent efforts to study the fracture behaviors of metallic nanowires. In the first example, in situ uni-axial tensile tests of single crystalline copper nanowires were performed using a micro mechanical device inside a SEM chamber. Interestingly, both ductile and brittle-like fracture modes were found in the same batch of fabricated nanowires and the fracture modes appeared to be dependent on diameters of tested nanowires. From the analysis of fracture surfaces, sample morphologies and corresponding stress-strain curves, the competition between deformation and fracture mechanisms controlled by initial defects density and by the probability of dislocation interactions was attributed to this intriguing size dependent fracture modes transition. In the second example, we showed that, under uni-axial tensile loading, single crystalline ultrathin gold nanowires might also fracture in two modes, displaying distinctively different fracture morphologies and ductility. In situ HRTEM study suggested that the unexpected brittle-like fracture was closely related to the observed twin structures, which is very different from surface dislocation nucleation/propagation mediated mechanism in ductile fracture mode. Molecular dynamics (MD) simulations further revealed the processes of shear-induced twin formation and damage initiation at the twin structure/free surface interface, confirming the experimentally observed differences in fracture morphology and ductility.

Oriented attachment (OA) of molecular clusters and nanoparticles in solution is now recognized as an important mechanism of crystal growth in many materials, yet the alignment process and attachment mechanisms are poorly understood. To achieve this understanding we are investigating crystal nucleation and oriented attachment in a number of systems through in situ and ex situ TEM. Here we report the results of in situ TEM at sub-nanometer resolution and video rates using a custom designed TEM stage and fluid cell to directly observe oriented attachment of iron oxide and oxyhydroxide nanoparticles and titanium dioxide branched nanowires. Results on iron oxyhydroxide show that primary particles interact with one another through translational and rotational diffusion until a near-perfect lattice match is obtained either with true crystallographic alignment or across a twin plane. OA then occurs through a sudden jump-to-contact. Analysis of the acceleration leading to attachment indicates that it is driven by electrostatic interactions. The results on iron oxide nanoparticles show a progression of nanorod aggregation into disordered aggregates followed by gradual ordering into the final co-aligned mesocrystal, while analysis of the titanium oxide system indicates a highly organized aggregation and branching process. We detail the progression of the structural transformations and use quantitative analysis of the dynamics to constrain the underlying mechanisms.

Sintering of nanometric powder particles can be considered as competition of densification and crystal growth. We have used in situ transmission electron microscopy (TEM) to study the sintering behavior of nanometric nickel particles through current-assisted densification. Small agglomerates of Ni powder particles suspended on holey amorphous carbon films were directly contacted with a scanning tunneling microscopy (STM) tip and exposed to constant bias stress. Morphological changes of the particles were video-taped through in situ TEM imaging while the resulting current across the particle agglomerate was recorded as a function of time. After an initial incubation time we observed the onsets of densification of the particle agglomerate on the carbon support film while particles that were in contact with the STM tip and extended into the vacuum coalesced and exhibited grain growth. As a consequence, a roughly 20nm wide nickel nanorod grew between the STM tip and the particle agglomerate. During the observed Czochralski-type growth, planar defects were introduced into the growing nanostructure in the direction perpendicular to the growth direction. The growth defects are correlated abrubpt changes in resulting currents, i.e. specific ohmic resistances. Preliminary characterizations of the growth conditions suggest that grain growth of nickel is kinetically favored over densification of the particle agglomerate. Initial systematic experimental studies furthermore indicate that growth of the nanostructure is likely facilitated by electromigration of nickel. The observed growth of metallic nanostructures may provide new avenues for printing of metallic nanostructures with unprecedented accuracy.

Oxide glasses are widely used in optical and electronic devices, and micro- and nanoelectromechanical systems, where they are being fabricated for instance as particles, wires and pillars [1]. Such materials are well known to be brittle at or near room temperature and fracture when exposed to any mechanical deformation. Thus, it is difficult to deform them plastically in order to change their shape after fabrication as components in functional materials/devices. The reason why silica glass does not exhibit ductility at room temperature is its glass-transition temperature Tg of 1373 K [2] and thus the room temperature is too low to achieve viscous flow in order to promote any considerable deformation rate. Recently it has been shown that with moderate exposure to a high-energy electron beam typically used in the transmission electron microscope, effective shape changes in compression and tension (superplastic elongations) can be induced for amorphous silica when being on nanoscale [1]. By means of in situ nanomechanical experiments in a TEM it was shown in the same work that the flow stress strongly depends on whether the mechanical deformation is performed with or without the electron beam. Thus, conventionally brittle nanoscaled amorphous silica can be made ductile by electron beam irradiation without considerably changing the temperature. The electron-beam-assisted generation of structural and bonding defects by facilitating bond-switching events is believed to be the mechanism beyond this material behavior, which accommodates viscous flow through the whole sample body. In the present work detailed in situ nanomechanical experiments are performed on amorphous silica nanostructures inside the transmission electron microscope under defined conditions of e-beam irradiation. Beside the electron-beam-assisted viscous flow and shape change under mechanical load, it is demonstrated that the effect of the electron beam on the nanomechanical behavior of amorphous silica is reversible and can thus be considered for re-shaping of nanoscaled silica not only once, but multiple times. Furthermore, the reversibility may be exploited for generating anisotropic glass topology by e-bam assisted mechanical quenching. This offers new opportunities for fundamental studies on structure-property relations of glass at the nanoscale. [2] [1] K. Zheng, C. Wang, Y.-Q. Cheng, Y. Yue, Y. Han, Z. Zhang, Z. Shan, S. X. Mao, M. Ye, Y. Yin, E. Ma. Electron-beam-assisted superplastic shaping of nanoscale amorphous silica. Nature Comm. (2010) 1:24 doi:10.1038/ncomms1021. [2] The authors gratefully acknowledge financial support by the DFG via the Cluster of Excellence Engineering of Advanced Materials.

AlGaN/GaN high electron mobility transistors (HEMTs) possess vast potential applications in fields where high-power and high-frequency devices are needed [1]. Unfortunately, use of AlGaN/GaN HEMTs in high power applications results in unpredictable and catastrophic device degradation [2]. Various techniques have been used to detect and study the degradation mechanism using different techniques such as cathodoluminescence spectroscopy [3], atomic force microscopy [4], and transmission electron microscopy (TEM) [5]. One major advantage of using TEM as compared to other techniques is to identify line defects, pits and cracks beneath the gate and correlate their size change to the biasing duration [6]. Although theoretical calculations [7] proposed the role of inverse piezoelectric effect in the formation of cracks, direct correlation strain in a real-time biased sample is missing. In order to study the degradation mechanisms of HEMTs under applied bias voltage, cross sections of different devices were prepared using a focus ion beam (FIB) technique. A conventional FIB lift out process was followed and electron transparent samples were prepared. These steps are in good agreement with reported results [8] on sample preparation of HEMTs device for TEM studies. Samples were then thinned to electron transparency using low Kv mode in the FIB, for TEM analysis. Geometric phase analysis (GPA) revealed that strain existed in the pristine (unbiased) devices, which is due to the lattice mismatch between GaN and that of AlGaN layer. Our measurements confirmed presence of compressive strain of 0.8% perpendicular to the AlGaN/GaN interface and tensile strain of 2.1% parallel to the interface. Similar measurements on in-situ biased devices indicated increasing in the amount of strain in the AlGaN layer, as well as formation of pits and defects. These studies revealed a quantitative correlation of strain and defect evolution in AlGaN/GaN HEMTs during electrical biasing which led to degradation of electrical properties and eventually major physical failure.

We report our systematic study of magnetic reversal of permalloy islands in an square spin-ice geometry. In spin-ice the magnetic moments within a lattice obey “ice” rules analogous to the ice rule for solid water. It is a model system to investigate frustration, correlation, and shot/long range order-disorder process and has opened a new field of physics. Using in-situ Lorentz phase microscopy and electron holography we reveal the magnetic reversal process including nucleation, propagation, annihilation, and avalanche of magnetic charges, or emergent monopoles and the presence of flux channels similar to Dirac strings between the magnetic charge monopoles during the reversal. Real space imaging and statistical analyses suggest that positively and negatively charged monopoles always move together, connected by a common flux channel. The interaction between monopoles and flux channels is determined by local frustration and can explain monopole populations and various states of ordering [1]. We also report our development of the Multimodal Optical Nanoprobe (MON), a TEM sample stage that enables simultaneous measurement of optical, optoelectronic, electric transport, mechanical and structural properties of samples inside any S/TEM without compromising the microscope&’s performance. It provides the only means currently available for the simultaneous in-situ correlation of optical, spectroscopic, electronic and structural properties of complex materials and devices at length scales ranging from hundreds of micrometers to angstroms. Of particular technological importance is the ability to investigate the site- or location-specific properties of engineered material interfaces such as the p-n junctions for photovoltaic applications. The development of the MON was recognized by the 2011 R&D 100 Award and the 2011 Innovation Award [2] and was recently awarded a US patent [3]. Applications using this system will be demonstrated [4]. References [1] S. D. Pollard, V. Volkov, and Y. Zhu, ‘‘Propagation of magnetic charge monopoles and Dirac flux strings in an artificial spin-ice lattice&’&’, Phys. Rev. B, 85, 180402(R), Editors&’ Suggestion, (2012). [2] Y. Zhu, M. Milas, J.D. Rameau and M. Sfeir, and A. Danilov, 2011R&D100 Award (R&D Magazine), Multimodal Optical Nanoprobe for advanced electron microscopy", Innovation Award, Microscopy Today, Microscopy Society of America, 2011. [3] M. Milas, Y. Zhu, and J.D. Rameau, “A Transmission Electron Microscope Sample Holder with Combined Piezo Controlled Electric Biasing and Optical Lighting Capabilities”, U.S. Patent No. 8143593. [4] Research supported by U.S. DOE/BES, under Contract Number DE-AC02-98CH10886.

The dynamic transmission electron microscope (DTEM) is an instrument capable of nanosecond-scale time-resolved electron imaging and diffraction of transient phenomena and is used for the study of rapidly driven phase transformations [1]. The DTEM has been used to study crystallization of phase change materials, such as Ge₂Sb₂Te₅ [2], and is now being used to study GeTe. Phase change materials are used in optical memory and non-volatile random access memories, applications in which crystallization of the amorphous phase must be achieved within nanoseconds. GeTe is a binary-alloy phase change material that is of interest for memory applications because of its high crystallization temperature [3,4] and rapid switching speed [4]. Amorphous GeTe films were laser crystallized in situ in the DTEM. Images with 15-ns time resolution were recorded at various times (from 100 ns - 2 mu;s) after initiation of crystallization with a laser. We will present results on the measurement of crystal growth rates, which were observed to reach up to 4 m/s. We will also present results on the characteristic length scales of morphological instabilities that develop under a microsecond during laser annealing of amorphous GeTe. The length scale of instabilities in thin films are a function of the film thickness as well as intrinsic materials properties (atomic mobility and interfacial energies). A model of spinodal dewetting is used to connect our measurements to the GeTe properties. While the occurrence of phase transformations, such as crystallization and melting, may be used to estimate temperatures attained during in situ electron microscopy experiments, the direct measurement of the rapidly changing temperature profiles during laser-driven in situ TEM experiments is experimentally inaccessible. The calculations and simulation methods used to model the spatial and temporal temperature profiles during laser heating will be described. This work performed under the auspices of the U.S. Department of Energy, Office of Basic Energy Sciences, Division of Materials Sciences and Engineering by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. References [1] T. LaGrange et al., Ultramicroscopy 108 (2008) 1441-9. [2] M. K. Santala et al., J. Appl. Phys. 111 (2012) 024309. [3] S. Raoux et al., Appl. Phys. Lett. 95 (2009) 143118. [4] G. Bruns et al., Appl. Phys. Lett. 95 (2009) 043108.

The resolution of lithography techniques needs to be extended beyond their current limits to continue the trend of miniaturization and enable new applications. But what is the ultimate spatial resolution to which (electron) optical lithography can be pushed? We present the molecule-by-molecule deposition of an organometallic precursor with focused electron beam induced deposition (FEBID). In FEBID adsorbed precursor molecules react with a focused beam of electrons. The adsorbed precursor molecules are fragmented in the bombardment with the electrons and the non-volatile fragments form the desired deposit. To observe the molecule-by-molecule deposition we implement a dedicated scan routine in a Titan environmental scanning transmission electron microscope. The precursor gas is W(CO)6, used at a pressure of 2x10-5 mbar. With a focused electron beam of about 0.3 nm in diameter, we alternate deposition (exposure by the stationary e-beam) and imaging (over an area of 1.8 by 1.8 nm, centered around the exposed region). We collect the annular dark field (ADF) signal during imaging (measured in V). Secondary electrons (SEs), generated in the support by the incident electrons, play a dominant role in the deposition process. SEs contribute to the undesired broadening of the deposits when we use amorphous carbon membranes as support. The broadening exceeds our imaging area (1.8x1.8 nm), even for membranes as thin as 1 nm. We use few-layer graphene as support to avoid this inherent limitation. The ADF signal increases in steps during the growth with a step height of 0.6 V. Based on a calibration of the ADF signal we expect a voltage of 0.52 V for a dissociated W(CO)6 molecule. We conclude that we observe the molecule-by-molecule deposition. These experiments present the resolution limit of (electron) optical lithography techniques.

There have been significant interests in the manipulation of nanoscale objects driven by the desire to develop novel nanotechnological tools or devices. Despite many previous efforts it is still a great challenge to direct the movements of a nanoparticle at will. We report electron beam manipulation and simultaneous transmission electron microscopy imaging of gold nanoparticle movements in a liquid cell. Gold nanoparticles are trapped inside the electron beam and move dynamically towards the spot with higher electron current density. Their global movements follow the movements of the beam. The trapping force is estimated to be a few piconewtons at gradient of ~103 e/(nm2*s)/nm. Analysis on the trajectories of multiple nanoparticles trapped inside the beam reveals an attractive force between the individual nanoparticles. By rapidly reducing density of the electron beam, we further can ‘collect&’ nanoparticles on the membrane surface and assemble them into a cluster. The ability to manipulate nanometer size objects opens many opportunities to probe the interaction forces between nanoparticles and create useful tools for physical and chemical sciences. HZ thanks the funding support from U.S. DOE Office of Science Early Career Research Program.

The application of plasma physics to the manufacturing and processing of materials attracts significant academic and commercial interest. Low-temperature plasma-based techniques for the growth of high quality nanostructures are amongst the most convenient and effective in present-day nanotechnology. However, limited understanding of the associated mechanisms exists due to the extreme difficultly associate with real-time observation and the non-linear and localized nature of the interactions. The present research developed a novel in-situ platform for characterizing plasma-materials interactions in the transmission electron microscope (TEM). Ar-plasma was generated using a DC voltage applied between two commercial SiNx windows coated with Au thin film electrodes in a custom designed TEM holder. Enclosed environmental cells with nanoscale windows are uniquely suited for in-situ nanoscale investigations of plasmas generated/confined within them. Plasma induced sputtering, deposition, and rapid surface diffusion of Au thin films and nanoparticles were directly observed in-situ in the TEM. The sputter and deposition rates are highly dependent on the applied power and correlate well with the theoretical calculations.

Environmental transmission electron microscopy (ETEM) is an important tool for in situ catalyst characterization because reactive gases may have a strong influence on the structure and composition of materials. The ability to study materials at elevated temperatures under reactive gases makes it possible to investigate the atomic scale changes taking place during materials synthesis and in materials under reactor conditions. For example, by correlating catalytic reactor data with suitably designed ETEM observations, it is possible to map out structure-property relations in catalysts [1]. However, a significant problem with this approach has been the lack of any direct measurement of gas-phase products which may be generated by the catalyst during the structural observation. This limits the ability to explore the links between catalyst structure and activity. We employ electron energy-loss spectroscopy (EELS) of gases [2] to detect and quantify catalytic products directly inside the microscope reaction cell while simultaneously determining the nanoscale structure of the catalyst. CO oxidation on a Ru catalyst was investigated with operando TEM. In-situ catalytic activity measurements were performed with EELS and confirmed that catalysis could be detected inside the TEM using EELS. Catalytic conversions of about 1% can be detected with this EELS for this reaction. A detailed discussion on the challenges associated with developing operando TEM will be presented. Motivated by a desire to improve solar energy conversions, we have also incorporated a variable wavelength high-brightness light source into our FEI Tecnai F20 ETEM. In our design, we employ an Energetiq laser driven light source and couple light into the reaction cell of the microscope using a quartz fiber. The fiber is kept some distance from the TEM sample holder so that heating in reactive gas environments can be performed without damage to the fiber. We investigated the amorphization taking place on the surface of anatase nanoparticles during in situ exposure to light, heat and water vapor. The observed surface structure change is believed to be related to hydroxyl groups formed at photogenerated oxygen vacancies at the surface. [1] S. Chenna, R. Banerjee, P.A. Crozier, (2011), ChemCatChem 3(6): 1051-1059. [2] P. A. Crozier and Santhosh Chenna, (2011), Ultramicroscopy, 111 177-185. [3] The support from the National Science Foundation (NSF-CBET 1134464), US Department of Energy (DE-SC0004954) and the use of TEMs at the John M. Cowley Center for High Resolution Microscopy at Arizona State University are gratefully acknowledged.

Electrochemical devices can leverage the alloying/dealloying and plating/stripping of bismuth and bismuth bearing alloys in applications from sensors to energy storage. Key to the optimization of these devices is understanding of the structure of the deposition as related to the deposition kinetics of the codeposit. In the current study we employ a mixture of in situ TEM with ex situ SEM and XRD. We examine the co-deposition of bismuth with zinc as this couple is useful for energy storage applications as well as a testbed for trace metal sensing in water. The nanoscale deposition kinetics and morphological evolution of bismuth are studied with a liquid cell TEM. The effect of pH and Zn/Bi ration is studied, as well as the effect of DC vs. AC/pulse plating conditions. In all cases the resulting zinc-bismuth alloy composition is analysed by the use of XRD. It is observed that zinc grows behind a front of zinc-rich bismuth metal. We hypothesize this growth mechanism arise from a balance between the solubility of zinc in bismuth, the lattice mismatch between zinc and bismuth, and the electronic conductivity of the deposited material.

Bimetallic nanoparticles, consisting two different metallic elements, have received paramount enterprise in heterogeneous catalysis due to their superior and emergent catalytic activities as compared to their parental monometallic counterparts. However, the added degree of freedom gives extra complexity in studying which element is at work, especially in reactive conditions; wherein, preferential reactions or adsorption of reactive/carrying gas molecules on bimetallic surfaces can induce chemical segregation, structural changes, and elemental-specific phase transformations. As the number of atoms on the surfaces becomes comparable to that in the bulk in shrinking nanoparticles, the induced bimetallic redistribution and phase transformations can be immediately dramatic. Yet because of the lack of atomic-resolution real-space imaging capabilities in reactive gas environments, few experiments have reported atomic-scale dynamics of bimetallic segregation in nanoparticles. Here, using aberration-corrected environmental electron microscopy we report direct visualization of the real-time atomic-scale restructuring of CoPt nanoparticles in O2 and H2 gas environments. We found that these alloy nanoparticles underwent nearly complete Co-Pt segregation in an oxidizing environment. The atomic-scale dynamics reveals a two-stage reaction pathway—the near complete segregation of metallic Co followed by oxidation upon an additional thermal activation. In reduction conditions, however, segregated Co in the shell can be completely reshuffled back to the Pt-core lattices forming an alloy particle with only surface segregation. The unprecedented dynamics uncover the flexibility in chemical restructuring of bimetallic particles in reactive gas conditions. It provides a practical constraint and a guideline for the design and use of bimetallic nanoparticles for heterogeneous catalysis. It also provides insights why dealloyed structures can occur in intersoluble bimetallic systems and how corrosion precedes in nanoparticles. The in situ environmental TEM experiments were 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. We performed ex situ TEM experiments at National Center for Electron Microscopy (NCEM) of the Lawrence Berkeley National Laboratory, which is supported by the DOE under Contract No. DE-AC02-05CH11231. HZ thanks the funding support from U.S. DOE Office of Science Early Career Research Program.

The structure of the most typical gold nanoparticle catalyst, Au/TiO₂, was studied under reaction conditions using environmental transmission electron microscopy (ETEM) [1]. While in situ studies by ETEM have made important contribution to elucidating catalytic activity of gold nanoparticles, the fragility of TiO₂ under electron irradiation has caused conflicting results of in situ studies of Au/TiO₂. We studied structural reorganizations caused by electron irradiation systematically and then deduced intrinsic catalytic structure of Au/TiO₂ without electron irradiation. The electron-irradiation-induced structural reorganizations of Au/TiO₂ catalysts were categorized into "decoration", "pillar", "encapsulation" and "damage". These reorganizations were observed depending on electron current density and electron dose and were summarized as structure evolution diagrams. The diagrams show observation conditions where the reorganizations by electron irradiation are negligible. Under such conditions, we observed Au/TiO₂ catalysts in situ and found out that gold nanoparticles display stable polygonal interfaces with TiO₂ supports and that the perimeters of interfaces are bounded by sharp edges parallel to the <110> directions. Additionally, we observed that gold nanoparticles in Au/TiO₂ catalysts became rounded shape in O₂ gas, while they exposed the major {111} and {100} facets in vacuum and CO/air gas. These shape changes can be correlated with the catalytic activity of gold nanoparticles, similarly to the previous study of Au/CeO₂ catalysts [2]. [1] Y. Kuwauchi, H. Yoshida, T. Akita, M. Haruta, and S. Takeda, to be published (2012) [2] T. Uchiyama, H. Yoshida, Y. Kuwauchi, S. Ichikawa, S. Shimada, M. Haruta, and S. Takeda, Angew. Chem. Int. Ed. 50 (2011) 10157

Many processes in materials chemistry, such as oxidation and reduction in metals, ceramics and catalytic systems are dependent on the local environmental conditions (temperature, stress, gas, liquid etc). Therefore, although recent years have seen a paradigm change in (scanning) transmission electron microscopy with unprecedented improvements in spatial, spectroscopic and temporal resolution being achieved, to fully utilize these new capabilities in the study of structures and processes, mechanisms to control the environment around the sample beyond the inherent vacuum of the microscope column must be implemented. In this presentation, the development and implementation of two environmental stages will be discussed: an in-situ gas stage that allows atmospheric pressure in a range of reactive gases to be maintained around the sample while atomic resolution images are obtained, and an in-situ liquid that allows atomic scale images and electron energy loss spectra to be obtained from samples in solution. For both stages, obtaining reproducible and quantifiable results requires a complete calibration of the effect of the electron beam on the process being studied. After calibration, detailed quantitative measurements of the nucleation and growth of individual nanoparticles or different phases within nanoparticles can be determined and compared with ex-situ results. As these stages have been designed to be incorporated into both high spatial resolution aberration corrected (S)TEM, as well as into the high temporal resolution Dynamic TEM (DTEM), the specifics of the measurements for both types of microscopes will be described and comparisons with experiments performed in full environmental microscopes will be discussed. Pacific Northwest National Laboratory is operated by Battelle Memorial Institute for the U.S. Department of Energy under Contract No. DE-AC05-76RL01830. Development of in-situ stages was supported by DOE NNSA-SSAA grant number DE-FG52-06NA26213 and NIH grant number RR025032-01.

In-situ scanning/transmission electron microscopy (S/TEM) investigations of processes that require wet environments, i.e. in liquids, colloids, solutions, have recently provided insights into the colloidal growth of nanocrystals, nanowires and other nanostructures that thus far have been achieved empirically. In this investigation, we are expanding to study a heterogeneous environment containing metal colloids and salt solutions to understand the mechanism of templated growth induced by electron beam irradiation in the liquid cell, and the experimental parameters that control this process. Here we present real-time observations of palladium growth on gold nanosized surfaces using liquid cell STEM for the first time. Pd is grown from a PdCl2 aqueous precursor solution at several concentrations, in which Au nanoparticles with various sizes (5 - 50 nm in diameter) are introduced. The Pd growth is investigated as a function of the incident electron dose, solution concentration, solution thickness and template surface structure. The real-time observations demonstrated that Pd deposition on small Au nanoparticle surfaces (up to 15 nm in diameter) begins as an epitaxial process but undergoes a transition to dendritic attachment, radiating preferentially from the corners and/or edges. Comparison of the growth on individual nanoparticles to grouped nanoparticles with a small separation gap provides evidence for the attachment of charged (ionic) palladium species from solution onto the substrate particles, which themselves are carrying a steady state charge induced by the electron beam. This system constitutes a model for studying Pd electrodeposition processes in solution. Further implementation of the electrochemical capabilities of the liquid cell, i.e. biasing, will allow for external control of reactions within the cell and real-time observations of more complex electrochemical processes.

Scanning electron microscopy (SEM) has been widely used not only for high-resolution imaging of nanostructures but also for characterizing various properties with application methods, such as energy dispersive X-ray spectrometry (EDX) and cathodeluminescence (CL) analysis. Recently, in situ characterizations of nanomaterials in controlled atmospheres as well as in vacuum are now becoming indispensable, because functional nanomaterials have been applied to sensors, catalysts, battery cells, and so on. Thus, it is necessary to develop a characterization technique for an object in controlled atmosphere using a conventional SEM and its applied methods. In this study, we have developed a novel method of a sealed environmental membrane cell to characterize an object in an atmospheric pressure gas with commodity type apparatus of SEM, EDX, and CL without any improvements, such as differential pumping techniques. A thin silicon nitride (SiN) film supported by a Si frame is employed as a membrane window for the environmental cell, in which the object and the atmosphere gas are sealed together. An electron beam (EB) is irradiated to the object through the SiN membrane window, and the object is characterized with SEM, EDX, and CL. As a preliminary experiment, europium(II) iodide (EuI₂) in argon (Ar) was measured, because blue-fluorescent EuI₂ is very sensitive to oxidation and hydrolysis, exhibiting fluorescence disappearance in air. In the experiment, the sizes of the Si frame and the SiN membrane window were 8 mm × 8 mm and 100 mu;m × 4 mm, respectively. The thickness of SiN was about 20 nm. A bit of EuI₂ powder was put into a cell from the backside, and then it was capped with conductive double-faced tape and a glass substrate coated with indium tin oxide. The peripheral edges of the Si frame and the glass substrate were sealed with an epoxy adhesive. All the sample preparation procedures were performed in an Ar atmosphere, where the oxygen concentration was maintained to less than 1.0 ppm. It was transferred in air to SEM, where the electron accelerating voltage was set to 10 keV. In SEM images, particles of EuI₂ in the cell were observed clearly through the membrane window. In the EDX spectra, peaks assignable to Ar, I, and Eu were observed. Furthermore, as well as CL images of the emitting particles, a CL spectrum peaked at 437 nm was obtained, and the emission wavelength was in agreement with the reported fluorescence spectrum of oxidation-free EuI₂ [1]. These results conclusively show that EB was actually irradiated to the particles of EuI₂ in the cell through the SiN membrane, and then secondary electrons, fluorescent X-ray, and luminescent photons were successfully detected through it. It is noted that the property of gas barrier is high enough to maintain fluorescent property intact even after air exposure for more than one day, and indeed Ar gas was also maintained in the cell. [1] L. Wang, et al., J. Alloys Compd., 225, 174 (2005).

Electrochemistry characterization techniques such as cyclic voltammetry, chronopoteniometry, chronoamperometry, and electrical impedance spectroscopy are often employed to determine the kinetics of interfacial reactions, rate-limiting steps associated with oxygen reduction and evolution reactions, and electron/charge/mass transfer. These techniques alone do not permit the observation of localized interfacial reactions at high spatial resolution. With the development of membrane type in situ electrochemical cells for transmission electron microscopy, quantitative electrochemistry experiments in liquid environments can be performed with a good degree of fidelity provided that the numerous experimental artifacts associated with the method is understood. This talk will be focused on the use of electrochemical fluid cell microscopy as applied towards electrical energy storage systems and will address factors associated with successfully performing controlled nanoscale electrochemistry experiments. This research was supported by the U.S. Department of Energy&’s Energy Efficiency and Renewable Energy Office of Vehicle Technology Program. Microscopy was performed at the ORNL Shared Research Equipment (ShaRE) User Facility, which is sponsored by the Scientific User Facilities Division, Office of Basic Energy Sciences, U.S. Department of Energy.

Lithium ion batteries are the most promising energy sources. Recently, Lithium ion batteries especially are used as power sources of mobile phones and laptops. However, capacity of commercial lithium ion batteries is not enough to be adopted for high power energy consumption devices such as electric vehicles, next generation mobile phones, and robots. To increasing the capacity, many researchers have been focus on developing candidate materials. However, these materials have some huddles such as capacity retention. To overcome this problem, investigation of the interface reaction mechanism during charging/discharging process is critical. So microstructure analysis should be performed. In this study, we observed the charging and discharging process of anode materials by a in-situ TEM liquid cell battery system using a newly designed liquid cell TEM stage. The liquid cell TEM stage was compatible with a JEM-2010F(JEOL) microscope. Liquid cell was fabricated as follows. SiNx thin film was deposited on a Si (100) substrate by low stress LPCVD method. followed by a metal layer of 50nm-thick on the SiNx membrane by E-beam evaporation. Then SiO2 spacer layer, which is required to control the cell thickness, was deposited on SiNx thin film by PECVD method. The processed wafers were wet etched to make electron transparent window. Anode materials and Li metal were mounted onto deposited metal layer. Then two membranes were bonded by sealant epoxy glue for preventing electrolyte evaporation into TEM column. Finally liquid phase electrolyte was injected into cell under argon atmosphere. The liquid phase electrolyte thickness was maintained to 200~300nm by deposited SiO2 spacer layer. Microstructural changes of anode material and interface reaction between liquid phase electrolyte and electrode were recorded in real time by applying electrical potential to the metal layer to charge and discharge the anode materials. Real-time reaction paths of the liquid-cell battery will be presented. This research was supported by the Nano-Material Technology Development Program (the Green Nano Technology Development Program) through the National Research Foundation of Korea funded by the Ministry of Education, Science and Technology (2011-0019984)

In-situ scanning transmission electron microscopy (STEM) imaging through liquids is a promising approach for exploring materials processes. However, options for in-situ chemical identification are limited: X-ray analysis is generally precluded because the liquid cell shadows the detector, and electron energy loss spectroscopy (EELS) is degraded by multiple scattering in thick liquid layers. Here we explore the limits of EELS for studying chemical reactions in their native environments in real time and on the nanometer scale and apply the technique to metal deposition and nanoparticle growth. In-situ studies were conducted using a Protochips flow cell with 50 nm thick silicon nitride viewing windows spaced 150-250 nm apart. The determination of the optical gap, local electron density and thickness of the liquid by EELS is demonstrated through liquids such as those used in battery applications (propylene carbonate), nanoparticle growth (H2O and CuSO4/H2O) and polymer precursors (ethylene glycol). The potential for using EELS to understand in-situ STEM reactions is demonstrated for beam-induced deposition of metal films and nanoparticles. For instance, as CuSO4/H2O is irradiated during imaging, copper clusters nucleate on the nitride membrane. EELS confirms what we see via imaging by developing low-loss peaks characteristic of metallic copper in regions where there has been heavy copper deposition, but not outside the Cu clusters. Additionally, the increased Cherenkov radiation below the optical gap indicates the increased dielectric constant of the CuSO4/H2O solution compared to water according to the Frank-Tamm result, enabling the detection of the low concentration of CuSO4. Additional studies of the beam-induced growth of nanoparticles have been performed, enabling further understanding of growth conditions leading to metal deposition versus nanoparticle growth. From these techniques, in-situ imaging and valence EELS offers insight into the local electronic structure of chemical reactions and nanoparticle growth.

Quantitative imaging of the nano-to-microscale dynamics of fluidic processes is of critical importance to emerging technologies such as nano-engineered coatings and membranes for power generation [1], water desalination [2], and microelectronics cooling [3] applications. Recent advances in microscope and sample holder design enable imaging of such processes using in situ electron microscopy with nanometer spatial resolution [4]. Phenomena occurring in fluids encapsulated between two flat electron-transparent membranes can be imaged in a high vacuum TEM. An Environmental SEM (ESEM) has lower spatial resolution than TEM but does not constrain the geometry of the experiment to a two dimensional fluid film. For example, dynamics of complex three dimensional processes such as condensation [5] and freezing on nanostructured surfaces [6], nanoparticle motion on curved water-gas interfaces [7], and water microjets [8] can be imaged. Historically, low temporal resolution has been the main drawback of using in situ electron microscopy. While several approaches for ultrafast TEM have been developed recently [4a, 9], no counterparts for ESEM have been reported. Here we show that appropriately controlling the electron beam movement during ESEM imaging allows for studying dynamic fluidic processes with image acquisition rates of tens to hundreds of frames per second. We show the utility of our approach with examples of fast imaging of microdroplet condensation on superhydrophobic surfaces, rapid nanoparticle movement and self-assembly on curved water-gas interfaces, water surface flows with tracer nanoparticles, and liquid ejection through a microscale orifice. References: 1.Dietz, C.; Rykaczewski, K., et al., Appl. Phys. Lett. 2010, 97 (3), 033104-1-033104-4. 2.Humplik, T.; Lee, J., et al., Nanotechnology 2011, 22 (29), 292001. 3.Narayanan, S.; Fedorov, A. G., et al., J. Micromech.Microengin. 2010, 20 (7), 075010. 4.(a) de Jonge, N.; Ross, F. M., Nat. Nano. 2011, 6 (11), 695-704; (b) Stokes, D. J., Principles and Practices of Variable Pressure and Environmental Scanning Electron Microscopy. Wiley: 2008; p 234. 5.Rykaczewski, K., Langmuir 2012, 28 (20), 7720-7729. 6.Varanasi, K. K.; Deng, T., et al., Appl. Phys. Lett. 2010, 97 (23), 234102. 7.Rykaczewski, K.; Chinn, J., et al., ACS Nano 2011, 5 (12), 9746-9754. 8.DePonte, D. P.; Doak, R. B., et al., Micron 2009, 40 (4), 507-509. 9.(a) Stach, E. A., Mater. Today 2008, 11, 50-58; (b) Zewail, A. H., Science 2010, 328 (5975), 187-193.

Liquid cell electron microscopy provides unique and exciting information about processes ranging from particle nucleation and growth kinetics to diffusion and aggregation dynamics and atomic level rearrangements during coalescence. The ability to include an electrical bias in such experiments broadens the range of opportunities, allowing liquid cell TEM to be applied to studies of electrochemical nucleation and growth, reactions in batteries, and phenomena in corrosion or electrophoresis. In this presentation we will discuss the experimental issues associated with biasing in the liquid cell, illustrating with examples of different electrochemical processes. We first discuss the geometries that have been demonstrated for liquid cell biasing, describing the effects of cell geometry on diffusion and growth kinetics. We then consider the influence of the electron beam on liquid cell reactions. Radiolysis of water, forming hydrogen and other species, can have dramatic effects, while beam-induced heating is usually negligible. Radiolysis is particularly significant in electrochemical experiments since the complex chemistries that are of interest can interact with the radiolysis products leading to different possible consequences. We can see some of these effects in electrodeposition of copper and formation of Au particles where additives such as chloride are involved. We finally discuss electrochemical formation of dendritic or ramified structures and compare the morphologies seen during deposition and corrosion of metals, which have key relevance to reactions in batteries. If the experimental issues are well controlled, we believe that electrochemical liquid cell electron microscopy has an exciting future with its exceptional view of liquid phase dynamics in the presence of electric fields.

It is generally believed that the mechanism of lithiation of a material is governed by the material&’s phase and crystallographic orientation. We show that this is not true when the size of the material goes down to the nanoscale. Specifically, the surface of nano-sized materials can have significant effect on the lithiation behavior. In this work, we demonstrated that the lithiation behavior of nanowires can be controlled by surface modifications, taking germanium (Ge) nanowires as an example. Ge is considered to be one of the most promising candidate anode materials because of its high volumetric capacity (7366 Ah/L), second only to silicon (8334 Ah/L), and high gravimetric capacity (1384 mAh/g). In addition, Ge has high intrinsic electronic conductivity and lithium ion diffusivity, which enables high rate capability. Furthermore, it has been shown that nanostructured anode and cathode materials, including nanowires and nanoparticles, often exhibit improved performance compared to their bulk counterparts due to the short electron and ion transport paths and better accommodation of the large stress generated during lithiation/delithiation. In this work, we investigated the lithiation behavior of Ge nanowires with different surface modifications using in-situ transmission electron microscopy (TEM). We discovered that the lithiation behavior of Ge nanowires can be strongly altered by a small degree of surface modification: the lithiation of pure Ge nanowires results in isotropic expansion, i.e. swelling along both axial and radial directions progressing from the surface inward and resulting in a core-shell lithiation behavior; however, Si doping on the surface of pure Ge nanowires - introduced via a post-growth method - changes the lithiation behavior sharply, resulting in anisotropic expansion evidenced by elongation only along the axial direction (axial lithiation) and negligible swelling along the radial direction. Surprisingly, this axial lithiation behavior can be turned into core-shell lithiation behavior upon further surface coating on the surface of the Si-doped Ge nanowires. The underlying mechanisms for the different lithiation behaviors are discussed. These results shed light on the lithium ion transport properties in Ge and aid the design of Ge anodes for lithium ion batteries. Acknowledgment: 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.

Catalytic efficiency of a material mostly depends on its surface activity and more so in case of nanocrystals. Cerium oxide is a particularly interesting and versatile catalytic material, and relationships of activity and facet-types have been established by theory. It would be highly desirable to have a “nanoprobe” available which can measure or at least relatively compare oxygen extraction activity from particle to particle or facet-to-facet, however all established catalysis tests operate on macroscopic quantities of powders and deliver average results. We propose here to use dynamic motion of cations induced by electron irradiation and oxygen ablation as a “probe” to locally assess surface activity in a catalytic material. In-situ aberration corrected HRTEM serves as an ideal tool as it allows to quantify the surface dynamics as a function of time and atomic occupancies of lattice and interstitial sites. In extension of our early work we present here progress in three directions: (i) a first study of aberration corrected TV-rate recording of atomic motion has been achieved using the Sheffield JEOL 3100 R005 double aberration corrected microscope and a Gatan ORIUS digital video-CCD camera at a rate of 30 fps (animations will be available for presentation). (ii) We extend our studies of octahedral particles (G. Moebus et al, Adv. Func. Mater. 21(11), p.1970 (2011)) to specially prepared morphologies of cuboid shapes with extended {100} and only small {111} facets. The stability of these particles and their predicted superior surface activity compared to standard commercial powders make them highly relevant for applications. (iii) Apart from dynamic hopping sequences, we identify temporary cationic surface reconstructions with atomic occupancies outside bulk lattice positions of fluorite-ceria. Quantitative analysis of all these activities in terms of surface plot profiles and changes in interatomic distances will be presented. This underpins a novel capability of aberration corrected in-situ TEM to study dynamic surface reconstructions otherwise only known from SPM, which is in-applicable to nanoparticles. To support the idea of identification of hopping activity as a measure of oxygen extraction activity, supporting molecular modelling simulations are provided, which map the ease of extraction with spatial resolution across particles and across facets (U. M. Bhatta et al, ACS Nano, 6(1), p.421 (2012)).

Silicon is a promising candidate anode material for next generation lithium battery due to its superior capacity with a gravimetric capacity of ~ 4200 mAh g-1 and a volumetric capacity of ~ 8500 mAh cm-3. Upon lithiation, silicon will expand by 300% with dramatic anisotropic change, showing obvious elongation along the [110] direction for crystalline silicon. Associated with such a large anisotropic volume expansion, the silicon anode materials are subject to pulverization, leading to the loss of electrical contact and rapid capacity fading of the battery. With the general guidance of continuum mechanics on the deformation and fracture behavior of silicon upon lithiation, numerous fundamental microstructure design concepts have been emerged for mitigating the failure of the silicon anode caused by the large volume changes. Two strategies have been used: (1) Tailoring of silicon to a low dimension with special topological features. Typical example for this strategy includes silicon nanowires, nanorod, nanotube, crystalline or amorphous silicon film, hollow-structure silicon ball, double-wall silicon nanotube. (2) Making silicon based composite materials. Carbon is a commonly used conductive additive in the lithium electrode materials and has a variety of structures, ranging from particles, tubes, fibers, and even a single layer as graphene. Therefore, it is a natural approach to rationally design a composite material based on silicon and carbon. One of the key questions for this type of composite material is the lithium ion transport behavior across the network structure of silicon and carbon as well as the accommodation of the volume change of the silicon upon lithiation. Here, we performed in-depth study on the lithiation behavior of Si-C nanocomposite using in-situ transmission electron microscopy (in-situ TEM), with emphasis placed on lithium ion transport, phase transformation and shape accommodation characteristics of the nanocomposite. The information acquired allows us to elucidate the failure mechanism and limiting factors in designing anode materials based on silicon and carbon composite.

Lithium ion batteries are promising candidates for future electrical energy storage. For many applications strong improvements in terms of energy density are still needed. Silicon is an anode material with a very high theoretical capacity of 4200mAh/g. Unfortunately its reaction with lithium is not fully reversible and currently limits its use. For example volume expansions of up to 300% occur during its reaction with lithium. In order to better understand Lithium insertion and extraction into/from silicon, in situelectron microscopic experiments were performed. All experiments are based on silicon nanowires that were produced by the vapor-liquid-solid (VLS) technique or the metal-assisted etching (MAE) technique. Mechanical tests were performed on these wires using a home built setup for in situ tensile testing. In the experiments, the mechanical strength and failure of wires could be tested for different lithium contents and a clear loss in strength was identified. In addition to the mechanical tests, in situ electrochemical tests were performed. For these experiments a special test cell was built inside the microscope consisting of metallic lithium as the counter electrode and an ionic liquid based electrolyte which provided the ionic contact to the wire. Using this experiment the morphology of the lithiation process could be monitored in detail. The kinetics of lithium insertion and extraction were characterized and the reversibility of the process was investigated. In addition to detailed morphological data, results on the reversibility of the lithium storage process and its dependence on the state of charge will be given in this presentation.

In-situ transmission electron microscopy (TEM) is now being used to investigate physical, chemical and biological phenomena in liquids. Such observations could greatly benefit from atomic scale resolution offered by aberration corrected TEMs. However the relatively thick SiN windows necessary to encapsulate the liquid under vacuum and even the liquid trapped between the windows degrades the TEM resolution and signal to noise ratio (SNR). We introduce a new type of liquid cell using graphene sheets to encase colloidal growth solution for in-situ HR-TEM imaging. The flexibility, mechanical strength and impermeability of graphene allow for the solution to remain liquid within the TEM vacuum while Pt nanoparticles grow and combine under the influence of the electron beam. TEAM I, a modified FEI Titan S/TEM with spherical and chromatic aberration correction, was operated at 80 keV to provide nominal 1 Å image resolution without damaging the graphene sheets for prolonged movie acquisitions during the reactions. The series of TEM images with 3.85 fps show atomic columns, facets and twins of individual 2-5 nm Pt nanoparticles rotating, growing and coalescing while floating in the precursor liquid. We are able to track the positions of particles to measure diffusivity and determine that the particles preferentially coalesce along <111> directions possibly due to less ligand coverage on these facets. Liquid encapsulation between graphene sheets provides an ideal environmental cell to study nanoparticle morphology and dynamics with atomic resolution. The technique can readily be applied to a diverse range of systems for in-situ observations using HR-TEM.

A multifunctional scanning probe microscope (SPM) will be described that transparently integrates with a DualBeam SEM/FIB System. This is done without perturbing any of the capabilities of the Dual Beam in terms of detectors, gas injectors, analyzers etc while allowing for a completely exposed probe tip to be imaged online even with immersion objectives at working distances as short as 4 mm. In addition, the completely free motion of the rotation axis of the stage is maintained with the probe tip at the eucentric point, this makes it possible to orient the sample in any direction on any structure. The X and Y scan range of the atomic force microscopic (AFM) imaging achieves 35 microns with rough motion over 10 millimeters. This permits the SPM to tilt into position perpendicular to the SEM or FIB or under an angle for rapid and accurate placement of the probe tip at or on structures such as biopolymeric materials that are nanometric in X, Y and Z extent. Thus, not only can a structure&’s nanometric height be accurately profiled but this can be accomplished with the on-line excellence of SEM for X, Y metrology. Furthermore, electron and ion beam sensitive samples can be imaged and characterized by AFM at high resolution. Of special importance for on-line focused ion beam (FIB) fabrication is the Z extent of the AFM which is an unprecedented 35 microns. The design also allows placing the probe tip with an orientation facing up for fabrication or down for imaging. An additional integral part of the design is electron/ion beam friendly probes with unique characteristics. The functional capabilities of these probes will be described. All such probes are slender, long (>60 microns) with an axial ratio that can be as high as 10:1. All such probes designed for this combination of SPM with SEM/FIB have highly exposed probe tips. This, combined with the Z extent of the AFM imaging provides deep trench capabilities especially important for FIB fabrication. The system allows for standard imaging modalities such as elasticity and data on the relative elasticity of copper on silicon will be presented. Within the class of probes that have been implemented with SEM are probes for nanometric optical near-field imaging (NSOM) and data on GaN wires excited by electron beams will be shown as part of this presentation. Other functionalities existing for such probes exist such as thermal, magnetic and electrical. Such a combination of SPM with the Dual Beam capabilities of SEM/FIB is in effect a new form of analytical instrument which portends to be a disruptive technology affecting the potential of electron, ion and even scanned probe applications in fundamental and applied science.

In-situ transmission electron microscopy (TEM) method is powerful in a way that it can directly correlate the atomic-scale structure with physical and chemical properties. We will report on the construction and applications of the homemade in-situ TEM electrical and optical holders. Electrical transport of nanotubes [1, 2] and optical-electro-mechanical coupling of semiconductor nanowires have been studied inside TEM. Oxygen vacancy electromigration and its induced resistance switching effect have been probed in Pr0.7Ca0.3MnO3 and CeO2 films [3-5]. And the optical properties of carbon nanotubes [6] will be also included in this talk. References [1] W. Y. Zhou et al, Adv. Mater. 21, 4565 (2009). [2] K. H. Liu et el, J. Am. Chem. Soc., 131, 62 (2009). [3] Z. L. Liao et al, Appl. Phys. Lett. 94, 253503 (2009). [4] P. Gao et al, J. Am. Chem. Soc. 132, 4197 (2010). [5] P. Gao et al, Micron 41, 301 (2010). [6] K. H. Liu et al, Nature Nanotech. 7, 325 (2012).

The study of materials in liquids using a transmission electron microscope (TEM) is a field that is gaining significant momentum especially with the advent of commercially available holders. So far, the imaging of nanoparticle motion, nanoparticle growth, electrochemical reactions and biological samples has been demonstrated. Another area where this technique can have significant impact on is the study of the corrosion of materials. However, the imaging resolution in liquid cell experiments is significantly degraded because the electron beam passes through a liquid layer which can be more than a micron thick. This can make it difficult to characterize the detailed microstructure of polycrystalline metals and alloys, which are pre-dominantly the materials of interest in corrosion. Here, we introduce a method where the liquid layer is removed from the windowed area, without taking apart the flow cell, and thereby allowing us to image much higher resolution. The liquid can also be re-introduced and by repeating the procedure, we can take time-lapse images of the sample area at regular intervals. We use this method to study the corrosion of 50 to 100 nm thick Al thin films deposited onto one half of the flow cell using electron beam evaporation. When the flow cell is filled with water, the detailed grain structure of the polycrystalline film cannot be discerned clearly. By changing the liquid in the flow cell to a mixture of ethanol and water, the liquid de-wets from the imaged area, after a few minutes of irradiation with the electron beam. Then, the sample area gradually re-fills with the mixture after the electron beam is turned off. This way, we are able to collect successive nanoscale resolution images of the localized corrosion process over extended periods without changing the experimental configuration. First, we look at the effect of the initial grain size by comparing as-deposited films (50 to 100 nm grains) with films annealed at 450 °C (200 to 300 nm grains). We observed that the as-deposited films corroded in a dendritic manner where material was removed via localized tunnels, while the annealed films corroded by pitting which lead to uniform dissolution of the films. Our data also indicate that the pits did not initiate preferentially at the grain boundaries. Lastly, we apply these techniques to the study of localized corrosion rates and mechanisms as a function of the molarity of NaCl to compare corrosion in water and in salt water, and in samples where the surface chemistry is locally modified using a mass-selecting focused ion beam.

In a number of Li intercalation compounds in which an ordered array of Li is usually maintained, the control of point defects including cation disordering is of major significance for application to electrodes in rechargeable cells. Furthermore, as the chemically different environment induced by point defects leads to breaking of the ordered arrangement of atoms in crystals with a complex structure, mass and charge transport behaviors are also considerably affected by the presence of the defects. A variety of investigations on Li vacancies and cation intermixing have been reported for layered oxides. Recently, several notable experimental details revealing the atomic-scale point defects in olivine-type lithium metal phosphates, LiMPO₄ (where M = Fe, Mn) are available in the literature (S.-Y. Chung et al., Phys. Rev. Lett., 100, 125502 (2008); Angew. Chem. Int. Ed., 48, 543 (2009); Phys. Rev. Lett., 108, 195501 (2012)), as these phosphates have attracted a great deal of attention as alternative cathode materials in Li-ion cells over the past decade. Proper control and direct identification of their distribution in the lattice on the basis of crystal chemistry will be crucial steps toward enhancement of effective Li mobility during the intercalation reaction in olivine phosphates. Through an unprecedented combination of in situ high-temperature high-resolution electron microscopy, crystallographic image processing, geometric phase analysis, and neutron powder diffraction, we directly demonstrate that while the initial crystallites after nucleation during crystallization have a very high degree of ordering, significant local cation disordering is induced by rapid crystal growth in LiMPO₄ nanocrystals (S.-Y. Chung et al., Nano Lett., 12, 3068 (2012)). The findings in this study show that control of subsequent crystal growth during coarsening is of great importance to attain a high degree of cation ordering, emphasizing the significance of atomic-level visualization in real time.

Porous nanostructures are of great technological interest due to its potential applications in the field of catalysis, sensing, Li-ion batteries and super capacitors. The high surface area and the availability of accessible pores are the unique properties which makes them the ideal candidates for such applications. The thermal stability of these nanostructures is an important criterion in many of these applications including catalysis and sensing as the loss of surface area and the closure of the pores results in degradation of properties. In this work, we present the thermal stability studies on porous nanospheres of metals and oxides by in-situ heating in transmission electron microscope. Real time observation of the microstructural evolution of spherical nanoporous aggregates of metal and oxide nanoparticles reveals the stages of formation of hollow nanostructures. The observed transformation of nanoporous aggregates to hollow nanostructures is supported by a phase-field model for microstructure evolution.

Ferroelectric materials are characterized by a spontaneous electric polarization that can be reoriented between different orientations by an applied electric field. The ability to form and manipulate domains with different polarization orientations at the nanometer scale is key to the utility of ferroelectric materials for devices such as nonvolatile memories. The ferroelectric switching occurs through the nucleation and growth of favorably oriented domains and is strongly mediated by defects and interfaces. Thus, it is critical to understand how the ferroelectric domain forms, grows, and interacts with defects. Here we show the nanoscale ferroelectric switching of a tetragonal PbZr0.2Ti0.8O3 thin film under an applied electric field using in situ transmission electron microscopy. We found that the intrinsic electric fields formed at ferroelectric/electrode interfaces determine the nucleation sites and growth rates of domains and the orientation and mobility of domain walls, while dislocations exert a weak pinning force on domain wall motion. We also show that localized 180° polarization switching initially form domain walls along unstable planes. After removal of the external field, they tend to relax to low energy orientations. In sufficiently small domains this process results in complete backswitching. Our results suggest that even thermodynamically favored domain orientations are still subject to retention loss, which must be mitigated by overcoming a critical domain size.

BiFeO3 (BFO) is a multiferroic material that exhibits coupling between its antiferromagnetic and ferroelectric ordering. The magneto-electric coupling allows an applied electric field to control magnetism in devices based on this material. BFO films also have high Néel and Curie temperatures with a ferroelectric polarization along (111) of 90 mu;C/cm2. Due to these unique properties, BFO shows promise for use in both ferroelectric and magnetoresistive memories and in magnetic sensor technologies. In order for BFO to be incorporated into device technology, the ferroelastic switching mechanisms must be understood. One of the major factors controlling domain wall kinetics is the presence of defects in the material due to epitaxial strain, which have unique interactions with the domain boundaries. These interactions are not well understood, and could be further elucidated by in situ examination in a transmission electron microscope (TEM). In this study, we use an in situ biasing holder to apply DC voltage to the BFO thin films in the TEM. Varying the bias magnitude, polarity and electrode geometry allow for control of the internal electric field and thus the ferroelastic domains. Defects have been shown to have a profound effect on the switching dynamics of BFO films with various defects functioning as nucleation or pinning sites. By limiting and controlling both the initial nucleation and propagation and the relaxation of ferroelastic domains, defects are an important factor in determining the overall domain kinetics in BFO. Both ex-situ aberration corrected TEM and in-situ TEM characterization of defect densities and interactions with domains under an applied electric field are used to quantify the effects of defects on domain motion. Additionally, phase field simulations are performed to compare kinetics in various defect environments measured experimentally. Using these cross-comparisons, we can determine the effects of point defects and line defects on the kinetics of domain switching in BFO films. The Electron Microscopy Center and the research at Argonne are supported by the U.S. DOE Office of Science under contract DE-AC02-06CH11357.DE-AC02-06CH11357. Research at Berkeley was supported by National Science Foundation and computational resources was provided by National Energy Research Scientific Computing Center.

The cold field emission gun (C-FEG) is the brightest electron source available, which also exhibits the smallest energy spread [1]. This technology has been greatly improved over the years concerning the electron optics and the vacuum, but the same cathode materials are still in use [2]. We have recently developed a new C-FEG source using a carbon cone nanotip (CCnT) mounted on a standard tungsten cathode using a focused ion beam (FIB) [3]. This source exhibits very good spatial coherence properties, which could be useful for electron interferometry applications [4]. Here, we have inserted a CCnT inside a biasing transmission electron microscope (TEM) sample holder incorporating a nanomanipulator (Nanofactory Instruments), in order to approach the CCnT towards an Au-anode plate. We then ramped up the voltage between the nanotip and the anode from 0 to 100 V until the electric field around the tip was strong enough to allow the electrons to tunnel through the barrier and a field emission current could be recorded. When compared with the macroscopic electric field, the local field around the tip is enhanced, which is described by the field enhancement factor. This factor has been calculated from the current as a function of the applied voltage using the theory described by the Fowler-Nordheim (F-N) equations [5]. In addition to measuring the field emission current, off-axis electron holography was performed to measure the electric field directly as a function of applied voltage. Finite element modeling was performed to obtain a 3-dimensional model of the electrostatic potential and of the electric field. This allowed us to integrate the phase change along the beam path of the electrons through the potential [6]. By comparing this with the phase change obtained from the electron holography, a quantitative value of the electric field around the tip was deduced. We were able to show that the field enhancement factor determined directly from holography compares very favourably with that obtained from the F-N equations. The field enhancement factor depends critically on the distance between the tip and the anode [7]. We will discuss these factors through the use of modeling and relate to the in situ TEM measurements. References [1] OL Krivanek et al in “Advances in Imaging and Electron Physics” 153 (2008), ed. PW Hawkes, (Elsevier Academic Press, San Diego), p. 121. [2] AV Crewe, Rev. of Sci. Inst. 39 (1968), p. 576. [3] F Houdellier et al, Carbon 50 (2012), p. 2037. [4] F Houdellier and M Monthioux, CNRS, International Patent Application PCT/FR2011/052135. [5] RH Fowler and L Nordheim, Royal Society London Proceedings Ser. A 119A (1928), p. 173. [6] J Cumings et al, Phys. Rev. Lett. 88 (2002), p. 056804. [7] JM Bonard, Diamond and Related Materials 11 (2002), p. 763.

Electric field-assisted sintering (EFAS) is an effective densification technique for metals and ceramics powders to form poly-crystalline microstructures. Nanometric metal powders typically possess thin surface oxide layers that can hinder the sintering process. Here, electric field-induced dielectric breakdown of surface oxides is considered as a possible mechanism contributing to the efficacy of EFAS processing of metallic powders. Dielectric breakdown of the surface oxides on nanometric nickel particles was investigated through the direct application of an electrical bias to particles during in situ TEM observations. In this study, a scanning tunneling microscopy (STM) tip was placed in contact with nickel particle agglomerates and subsequent current-voltage characteristics were obtained using a double-tilt Nanofactory STM-TEM sample holder. High angle annular dark field imaging combined with simultaneous electron energy loss spectroscopy (EELS) allowed the first direct observation of field-induced surface oxide removal between two particles prior to neck formation and growth. Post-processing of the EELS spectrum images using principal component analysis (PCA) of Multivariate Statistical Analysis (MSA) software improved the signal to noise statistics. Increased conductivity, and thus current flow, with decreasing O/Ni ratio indicated electrical bias driven dielectric breakdown of the surface oxides. Almost stoichiometric NiO surface layers were reduced and the onset of neck formation and neck growth was observed with an initial neck diameter of 3 nm and an O/Ni ratio of 19%, while a subsequent measurement revealed neck growth to a diameter of 6 nm with an O/Ni ratio of only 13%. The electrical behavior proceeded from forward biased during the initial testing to ohmic behavior once the metal necks were formed. The results of the study suggest the application of high field strengths to powder compacts as a pretreatment step to form clean metal interfaces which can lead to accelerated consolidation at lower currents and, therefore decreased power consumption.

Perovskite (ABO₃ ) heterostructures offer a possible route for a paradigm shift in the design of nanoelectronics. Many ABO₃ materials exhibit functional phase transitions that can be controlled by subtle changes in applied voltage, stress, or temperature. Charge-ordered materials exhibit a sharp change in conductivity at the charge-ordering transition temperature, T_CO. In these materials, the valence electrons of particular atoms in a unit cell can be ordered along a crystallographic direction; this ordering, which occurs over the length of several unit cells, can result in novel electronic properties in thin films or bulk materials. In the case of one perovskite system, La_xSr_1-xFeO₃, (LSFO) the charge on the Fe atoms is ordered along the <111> crystallographic direction below T_CO. At this temperature, the resistivity of the material increases abruptly; this property has the potential for diverse applications in device fabrication. Heterostructures of La_xSr_1-xMnO₃ (LSMO) / La_ySr_1-yFeO₃ / La_xSr_1-xMnO₃ are deposited by molecular beam epitaxy (MBE). The T_CO is sensitive to variables such as the stoichiometry of the compounds and strain state of the LSFO. The strain state is manipulated, and geometric phase analysis (GPA) is applied to high-resolution transmission electron microscopy (HR-TEM) images to quantify the relative amount of strain in the LSFO layers. Charge ordered materials display ordering of the valence electrons over several unit cells, thus TEM is ideally suited for studying thin films of charge ordered materials. A cryo-holder is used in the TEM to observe the charge-ordered and charge-disordered states using nanobeam methods at various temperatures. The sensitivity of charge-ordering to the temperature and to the strain state of LSFO is reported. This information is vital for understanding how to tune variables when constructing novel electronic devices on the nanoscale. This work is supported by the Office of Naval Research (N00014-11-1-0664).