Symposium CM02 : In Situ TEM Characterization of Dynamic Processes During Materials Synthesis and Processing

Nucleation is the seminal process in crystal formation. While the classical picture of nucleation envisions addition of monomeric species to a growing nucleus that exhibits the structure of the bulk crystal, recent observations have revealed a rich set of hierarchical pathways involving higher-order species ranging from multi-ion complexes to dense liquid droplets that often form an initial transient crystalline or amorphous phase, which subsequently transforms to the final ordered phase. Our understanding of these multi-stage pathways and the factors that lead to their selection has been limited by availability of tools to investigate them in situ. Here we use liquid phase TEM to observe nucleation and transformation in the calcium carbonate and iron oxyhydroxide (FeOx) systems, both of which exhibit multiple polymorphs and are known for forming transient amorphous or disordered phases. For the calcium carbonate system, we used a series of additives — Mg, citrate and poly-acrylate — to modulate the transformation from amorphous calcium carbonate (ACC) to crystalline polymorphs. For citrate, PAA and low Mg content (≤2.5 mM), we find dissolution/re-precipitation dominates, with each additive extending the ACC lifetime and/or dissolution timescale, but the final polymorphs always exhibit the expected crystal morphology. In contrast, for [Mg] ≥ 5 mM, transformation occurs via loss of structural water seen both via an increase in electron density and changes in the Raman spectrum in the absence of morphological changes, leading to spheroidal Mg-calcite. TGA shows Mg brings excess water into ACC and molecular dynamics simulations predict this promotes atomic rearrangement. We hypothesize that slow dehydration coupled with ease of reorganization enables the isomorphic conversion. For the FeOx system we focused on crystallization of spindle-shaped hematite (Fe₂O₃) from ferrihydrite (5 Fe₂O₃-9H2O, fh). Our results show the spindles consist of atomically aligned domains organized into a second-order, hierarchical, rod-like structures penetrated by nm-scale pores. This structure forms as follows: First, the fh-to-hematite transformation is highly localized, with the initial hematite nucleating on fh surfaces. Second, spindles grow by addition of new hematite particles that nucleate from the solution. Whether the new particles form directly on existing hematite surfaces or in their immediate vicinity — after which they attach across the small remaining gap — is unclear. This process can even lead to formation of half-spindles with the growth direction pointing away from the mass of fh particles and into the fh-free bulk solution. Consequently, spindles do not form through fh attachment to hematite, nor even via random, independent hematite nucleation events followed by diffusion and attachment, but rather via an autocatalytic process in which fh provides a solute source, but the hematite interface drives secondary nucleation and attachment.

An ultrahigh vacuum TEM has been developed for in situ observation of ice (JEOL JEM-2100VL). A column of the microscope is evacuated by five ion pumps, two Ti-sublimation pumps, and two turbomolecular pumps. The pressure of the specimen chamber after baking is 1x10^-7 Pa measured by an ionization gauge. Furthermore, since the specimen is surrounded by liquid nitrogen shroud, the pressure around the specimen is supposed to be lower than 1x10^-7 Pa. We use a single tilt liquid He cooling holder (Gatan ULTST) for specimen cooling. Three ports are directed to the specimen surface with an incident angle of 55° for in situ studies. These are used for gas-inlet, UV irradiation (30 W deuterium lamp), and a quadrupole mass spectrometer. We used a 5-nm-thick amorphous Si film (SiMPore Inc.) as a substrate for ice deposition. To avoid the electron-beam damage to the ice sample, low magnification observation was performed with very low electron beam density. We will introduce two examples of observation on amorphous ices.
We have developed a new method for the formation of high-density amorphous ice, matrix sublimation method [1]. CO:H₂O = 50:1 ice was deposited at 10 K, which was then allowed to warm. After the sublimation of CO at around 35 K, high-density amorphous ice was formed. This is similar to that formed under high-pressure condition.
We found that UV-irradiated amorphous H₂O ice at 10 K shows liquid-like behavior at 50-140 K [2]. First, islands of crystalline ice Ic was made at 145 K, then cooled to 10 K and irradiated UV-rays. Crystalline ice Ic was easily amorphized within one minute. After 30 min. UV-irradiation, the sample was heated. We observed at T > 50 K that the height of the islands decreased and their area increased and finally overlapped like squashed liquid droplets. The viscosity estimated from the spreading velocity of islands is an order of 10⁷ Pa s, 5 orders of magnitude smaller than that at the glass transition temperature (10¹² Pa s), indicating that the UV-irradiated amorphous water ice behaves like a viscous liquid.

References
[1] Kouchi A. et al. (2016) Chem. Phys. Lett., 658, 287.
[2] Tachibana S. et al. (2017) Sci. Adv., 3, eaao2538.

Alloyed contacts, formed by thermal annealing of metal-semiconductor nanowires (NWs), are prescribed for lithography-free self-aligned gate processes. Prior research on nanoscale metallization has revealed large differences with their bulk counterparts, evoking reevaluation of the thermodynamics, kinetics, and resultant phases in alloyed and compound nanoscale contacts. The in situ heating inside a transmission electron microscope (TEM) permitted imaging of the atomic scale dynamics in contact phase transformation in elemental Si and Ge NWs but not yet in III-V NW channels. Here, we carried out in-situ heating TEM experiments to study the dynamics of contact metallization between Ni and In_0.53Ga_0.47As NW channels fabricated by top-down method, and observed at atomic resolution the initial stages of ledge formation and movement behaviors of Ni₂In_0.53Ga_0.47As (nickelide) in both the NW cross-section and along the NW channel. In the square cross-section of a <110> oriented InGaAs NW that is composed of {100} and {110} side-wall facets, the nickelide reacted in a layer-by-layer manner with ledge formation and movement on {111} facets and along <112> directions. To limit the naturally occurring intermixing between Ni and InGaAs NW due to latent heat during metal deposition, we tailored the interfacial structure of the InGaAs NW and found significant changes in the contact formation dynamics, captured by a model that was specifically developed for the cross-sectional geometry of NW channels. The dynamics of the reaction along the NW channel were more revealing. We observed consistent nucleation of strained single-bilayers that rapidly merge to form a stable double-bilayer, which moves on the reaction interface of In_0.53Ga_0.47As (111) || Ni₂In_0.53Ga_0.47As (0001). The single-bilayer ledges transfer into double-bilayers by collective gliding of Shockley partial dislocations and by forming a misfit dislocation. Consequently, the double-bilayer height became the unit height of the nickelide ledges in this phase transformation. We also monitored the solid-phase-regrowth (SPR) to incorporate dopants at the surface of the InGaAs channel under in situ TEM, and demonstrated the improved contact resistance by fabricating functional InGaAs MISFETs. Our in-situ studies provided guidance for the phase selection of crystalline self-aligned contacts in nanoscale channels cross-sections.

Nanoparticle dissolution is a common process in synthetic procedures where primary particles are replaced by more stable phases, as well as in environmental settings where they both serve as electron sinks or sources for microbes and are responsible for release of nutrients or contaminants. Iron oxides and iron oxyhydroxides, which are widespread in technological applications and environmental settings, are a prime example. To decipher the pathways that underlie dissolution and quantify the kinetic parameters controlling rates, we investigated dissolution of β-ferric oxyhydroxide (β-FeOOH) nanorods via in situ liquid phase (LP)-TEM. Previous ex situ studies that investigated dissolution of iron (III) hydroxides by proton or photoreductive attack found that photoreductive dissolution of is faster than by nonreductive dissolution (e.g. proton-promoted or ligand promoted thermal dissolution). Other studies concluded that the rate-determining step during dissolution in HCl is protonation of the surface together with formation of a chloride-Fe surface complex. Generally, dissolution of β-FeOOH nanorods in strong acids occurs end-to-end along the [010] direction and has been attributed to proton penetration and destabilization of so-called “tunnel structures” by removal of Cl ions. In our work, we find dissolution of β-FeOOH nanorods occurs both end-to-end (along [010]) and side-to-side ( along [100]) directions under the electron beam, even at extremely low electron dose rates. Dissolution involves two reaction steps: (i) beam radiolysis of the water to release H⁺ that promotes end-to-end dissolution; (ii) beam induced reduction of Fe(III) at the nanorod surface with subsequent release of Fe(II) into the solution, which drives side-to-side dissolution. To further understand the role of protons on dissolution, we investigated the effect of pH. The results show low concentration pH buffers reduce and even stop end-to-end dissolution, but side-to-side dissolution was still observed, leading to formation of dumbbell-shaped nanopartcles that eventually dissolve completely. Nanorods dissolution was totally inhibited for pH buffer concentrations higher than 100 mM. In addition, we found that chloride ions can inhibit end-to-end dissolution of β–FeOOH nanorods while promoting side-to-side dissolution at low concentration pH. The results provide strong evidence that radiolysis contributes to the reduction of Fe(III) to release of Fe(II) even at neutral pH, thus promoting dissolution of iron oxyhydroxides. Because, photolysis is a naturally occurring process in the environment that is similar to radiolysis, the findings of this in situ study may help to understand dissolution and transformation of iron oxides and iron oxyhydroxides in Nature.

We use liquid phase transmission electron microscopy to directly image the self-assembly of colloidal nanoparticles in solution, one-by-one in real-time. Depending on solvent conditions, a single type of anisotropic nanoparticles can lead to a wide variety of final structures not previously predicted: linear and cyclic “polymeric” chains, hierarchical plastic crystals, and highly ordered solids. In-situ monitoring of the dynamic pathways together with computation reveals interesting and novel phenomena in these systems due to inherent many-body coupling and discreteness at the nanoscale. We expect our study to open new opportunities in understanding the conformation, phase behaviors and collective dynamics on the nanometer length scale that is not accessible using other means.

Hierarchically organized nanoparticles are good candidates for plasmonics [1-2] and catalysis [3] studies due to its distinctive properties than its isolated forms. Self-assembly serves as a robust ‘bottom-up’ method to organize nanoparticles into desired arrays. The assembly process is often characterized by the ‘quench-and-look’ approach or indirect dynamic spectroscopic methods [4]. As such, the intermediate steps and the how the nanoparticle morphology, surface molecules participate in controlling the final assembled structures still remains largely unresolved.

Here, we use of in situ liquid cell electron microscopy to directly probe the self-assembly of chemically-synthesized nanoparticles of different morphologies; Au nanocubes, nanorods, nanospheres and nanobypiramids with surfactant molecules; cetyltrimethylammonium bromide (CTAB) at high temporal and spatial resolution. Our real-time observations reveal that there are different attachment pathways for side-to-side assembly of Au nanocubes: pre-alignment attachment and post-attachment re-orientation. We further investigate how the shape of nanoparticle and chemical surfactants can affect the nanoparticle assemblies to attain the lowest total energy state by monitoring the nanoparticle dynamics as a function of time. Understanding the physical and chemical interactions that govern the self-assembly could potentially lay the foundation for rational design of desired assembled nanostructures for application in catalysis, opto-electronics and drug delivery.

Acknowledgements
We acknowledge the support from Ministry of Education of Singapore (award No. MOE2015-T2-1-007).
References
S. F. Tan et al., Quantum plasmon resonances controlled by molecular tunnel junctions. Science 343: 1496-1499, 2014.
K. J. Savage et al., Revealing the quantum regime in tunnelling plasmonics. Nature 491: 574-577, 2012.
G. Prieto et al., Towards stable catalysts by controlling collective properties of supported metal nanoparticles. Nature Mater. 12: 34-39, 2013.
M. Grzelczak et al., Shape control in gold nanoparticle synthesis. Chem. Soc. Rev. 37: 1783-1791, 2008.

Chalcogenide phase change materials (PCMs), such as Ge-Sb-Te alloys, have outstanding properties, which have led to their successful use for a long time in optical memories (e.g., DVD-RAM or CD-RW) and, more recently, in non-volatile resistive memories with phase change random access memories. The latter are the most promising candidates to replace the current FLASH memories at CMOS technology nodes under 28 nm [1]. The main feature of PCMs are fast and reversible phase transformations between crystalline and amorphous states with very different transport and optical properties. Controlling their crystallization, however, is a challenge. In the abundant literature devoted to interface and size effects on PCM crystallization, mostly on thin films (in the sub-100 nm thickness range) capped with various materials, no interface effect was reported for film thicknesses above 30 nm. Recently, we established a completely new picture on the crystallization of relatively thick (100 nm) films of prototypical Ge₂Sb₂Te₅ (GST) and GeTe alloys. We revealed the impact of interfaces on their crystallization mechanism by scanning transmission electron microscopy (STEM) images that were obtained at room temperature after various annealing treatments [2]. We demonstrated that the crystallization temperatures of GST and GeTe thin films can significantly vary as a function of their surface state. Whereas GST and GeTe PCM alloys have been, and are still, the subject of a huge number of publications during the last decades, their crystallization mechanisms in thin films are not yet fully understood. In that context, acquiring STEM images in situ during annealing was required to get a full understanding of the crystallization process.

We will present new results on the crystallization steps of GeTe thin films at a nanometer scale obtained by combining in situ annealing with STEM cross section imaging of the films. Thanks to a novel sample preparation technique, we were able for the first time to observe directly in the microscope the effect of interfaces on the crystallization process of amorphous GeTe thin films [3]. Surface-oxidation promotes heterogeneous nucleation at the oxidized upper surface of the film. By contrast, perfectly capped, and thus non-oxidized films, crystallize at a significantly higher temperature through volume nucleation. We will also discuss the origin of surface nucleation in oxidized GeTe films. The latest findings demonstrate that one can control the stability of the amorphous phase of PCMs by controlling their interfaces, opening the route to new 3D memory cell architecture with improved data retention.

[1] P. Noé et al., accepted manuscript, Topical review in Sem. Sc. And Tech. (2017). http://iopscience.iop.org/article/10.1088/1361-6641/aa7c25
[2] P. Noé et al., Acta Mater. 110, 142 (2016).
[3] R. Berthier et al., J. Appl. Phys. 122, 115304 (2017).

Two-dimensional (2D) nanomaterials have shown superior performance in catalysis, sensing, and many other surface-enhanced applications. Although a lot of 2D materials have been made, the formation mechanisms especially those of non-layered 2D structures in solution processes are still unclear. We study the formation of 2D cobalt oxide and cobalt nickel oxide nanosheets using liquid phase transmission electron microscopy (TEM). Our direct observation reveals the two-step growth: 1) 3D nanoparticles are formed from the molecular precursor solution; 2) transformation of 3D nanoparticle into 2D nanosheets. Ab initio calculations show that a small nanocrystal with dominant edge energy transforms into a 2D structure when the nanocrystal grows further and consequently surface energy becomes dominant. Revealing of such a 3D-to-2D growth pathway and the competition between the negative surface energy and positive edge energy provides opportunities to the controlled synthesis of novel materials in solution.

Understanding the details of chemical reactions, particularly in biologically significant materials systems, has driven considerable effort in developing liquid cell stages to utilize the spatial resolution of the transmission electron microscope (TEM). However, containment of the liquid in the vacuum environment and scattering of electrons within the liquid cell and encapsulation window limit the achievable resolution. Recently, considerable effort has been invested in using 2D materials for liquid cell windows, with some success, particularly using graphene. In this case, the window is composed of pockets of water between a few sheets of graphene, creating a thin blister of liquid environment, minimizing scattering from the window and the liquid volume, known as a graphene liquid cell (GLC). Furthermore, recent work on the interaction of the electron beam and water provides a mechanism to use the electron beam to drive pH sensitive reactions. We apply this knowledge to drive pH dependent reactions in graphene liquid cells on hydroxyapatite (HAP), a biomaterial for which dissolution and precipitation processes are important. Using the electron beam to change local pH, we can examine the electron-beam induced change in pH, and the resulting dissolution/reprecipitation of HAP in GLCs. The reactions are captured in real time and processed using computer vision techniques to track the change in individual particles. These results demonstrate the world of new capabilities allowed by 2D window thin liquid cells and quantification of electron-beam/liquid interactions, which can be applied to many questions involving chemical reactions in liquid environments at the nano scale.

We observe the crystallization of a colloidal Kossel supracrystal from anisotropic nanoparticles using liquid-phase transmission electron microscopy. By tracking the nanoparticle motions one at a time, we find that the development of supracrystal facets follows a layer-by-layer and directional growth mechanism, resembling that for atomic crystals. The conceptual similarity between our colloidal crystal and atomic Kossel crystal also allows a precise kinetics control over the crystal growth process, which eventually leads to distinct equilibrium structures. Our study provides detailed dynamic information over the supracrystal growth process, which can serve as a general guideline for materials design and engineering from the bottom-up.

Polymeric gel-based and hydrogel nanocomposite systems are promising materials in nanomedicine, pharmaceutical, and biosensing applications. For example, Pluronic-based Au-nanocomposites have shown to be advantageous theranostic agents for multimodal imaging of carcinoma cells. The rich phase diagram of aqueous solutions of Pluronic block copolymers offers a pathway for a targeted formation and placement of Au nanoparticles and for a controlled ordering of spatial arrangement within the gel matrices of these materials, while the interactions between the individual blocks of the polymer leads to enhanced nanoparticles stability and biocompatibility.
Due to the viscous nature of the Au-Pluronic nanocomposite gels, their characterization is normally limited to Small Angle X-Ray Scattering (SAXS) and Cryo-TEM imaging of cryo-ultramicrotomed thin slices of either a freeze-dried, or a vitrified specimen. While this approach allows viewing the individual nanoparticles and gauging their size distribution, it does not provide sufficient information about spatial arrangement of the nanoparticles within the gel

We present a novel approach to imaging Pluronic-based Au-nanocomposites with the liquid cell in situ. A cold-stage nanoprinting system (Nano eNabler molecular printing) was used to control the deposition of 1-5 um droplets on SiN windows by a microfabricated surface patterning tool (STP) cantilever. The printed SiN windows were assembled for in situ experiments in Hummingbird Scientific Liquid Cell holder and imaging by using a Tecnai G²-F20 Scanning Transmission Electron Microscope operating at 200 kV in HAADF-STEM mode. We demonstrate that in situ liquid cell can be used for direct imaging of nanoparticles in viscous media and at various stages of drying, offering important clues to understanding how the nanoparticles are arranged in polymeric matrices. We also discuss specimen deposition protocol and the effects it makes on the imaging.

A deep understanding of the formation mechanisms of low-dimensional nanostructures from bottom-up processes is of great importance in order to exploit the controllability of the nanostructures and their applications in photovoltaics, electronics, sensors, etc. on an industrial scale.
Environmental transmission electron microscopy allows for monitoring gas-solid reactions at the nanoscale in real time. Examples of such reactions includes growth of 2D materials (graphene)[1], pseudo 1D materials (carbon nanotubes)[2] and semiconductor oxides such as CuO[3].
The high spatial resolution combined with spectroscopic capabilities enables fundamental insights during the dynamical growth processes of the nanostructured materials directly linked to growth parameters such as atmosphere composition and temperature.
Growth and growth termination of single wall carbon nanotubes are very dependent on the local environment of the catalytic seed particle forming the tubes [4]. Topology of the seed particle support, the state of the catalytic particle, and carbon source play a crucial role in the growth dynamics. Observing the initial solid carbon formation on seed particles at the atomic scale reveals the conditions leading to encapsulation of the catalyst particles (no carbon nanotube growth) and a carbon layer lift-off resulting in nanotube growth, respectively.
Nanostructures can also be formed by simple oxidation of metals such as copper. CuO nanowires are formed by oxidation of metallic copper in situ in the electron microscope. The exact geometry of the resulting structures depends strongly on the partial pressure of oxygen during the heat treatment. Low pressure results in 2-dimensional copper oxidation in contrast to the 1-dimensional CuO nanowires forming at higher pressures. Post-growth treatment of the oxide nanowires at elevated temperature at reduced partial pressure of oxygen results in dissolution of the formed nanowires [5]. The direct observation of oxidation and disintegration reveals the nature of the facet dependent reactions.
Gas to Solid (growth) and Solid to Gas (disintegration) reactions directly visualized by means of controlled atmosphere electron microscopy is a key factor in the understanding of the formation mechanisms and thereby a stepping stone towards the controlled synthesis of tailored materials.
[1] J. Kling et al. Carbon 99, 261 (2016)
[2] M. He et al., Scientific Reports 3, 1460 (2013)
[3] S. Rackauskas et al., Nano Letters 14, 5810 (2014)
[4] L. Zhang et al. ACS Nano 11, 4483 (2017)
[5] S. Rackauskas et al., Scientific Reports 7, 12310 (2017)

The introduction and tremendous advances in recent years in the capabilities of dedicated holders for testing materials in situ under a variety of external stimuli has enabled a wide range of ground-breaking studies. Nevertheless, however advanced this new instrumentation may be, the fundamental differences in the electron microscopy operating conditions between in situ conditions and those typically applied to ultra-high resolution (spatial and energy) work mean that it is still difficult to reconcile both types of information in a single experiment. Here a multi-length-scales correlative microscopy approach is used to combine high resolution scanning transmission electron microscopy and electron energy loss spectroscopy (STEM-EELS) spectrum imaging and in situ testing on the same nano-objects. In particular, we demonstrate how the addition or removal of native semiconductor material at the edge of a nanocontact can be used to determine the electrical transport properties of metal−nanowire interfaces. While the transport properties of as-grown Au nanocatalyst contacts to semiconductor nanowires are well-studied, there are few techniques that have been explored to modify their electrical behaviour. The approach taken here used an iterative analytical process that directly correlates multiprobe in situ transport measurements with subsequent aberration-corrected scanning transmission electron microscopy on the same nano-objects to study the effects of chemical processes that create structural changes at the contact interface edge. A strong metal−support interaction that encapsulates the Au nanocontacts over time, adding ZnO material to the edge region, gives rise to ohmic transport behaviour due to the enhanced quantum mechanical tunnelling path. Removal of the extraneous material at the Au−nanowire interface eliminates the edge-tunnelling path, producing a range of transport behaviour that is dependent on the final interface quality. These results demonstrate chemically driven processes that can be factored into nanowire-device design to select the final properties. Another application of this approach combines atomic-resolution TEM with EELS to examine the surface structure and oxidation state of VOx, supported on anatase TiO₂ nanoparticles, in situ, under alternating oxidizing and reducing environments. A two-step approach where chemical mapping by means of STEM-EELS spectrum imaging was used to address the homogeneity of the VOx distribution, while in situ TEM imaging was used to monitor the structural response to changes in the surrounding gas environment, reveals reveal a reversible transformation of the vanadium oxide surface between an ordered and disordered state, concomitant with a reversible change in the vanadium oxidation state, when alternating between oxidizing and reducing conditions.

Strategies for size- and shape-controlled synthesis of nanocrystals have been explored to design functional nanomaterials for catalysis, energy storage, biomedical, optical and electronic applications. In situ liquid scanning transmission electron microscopy (STEM) has the unique ability to image evolving chemical processes within a liquid environment and at high spatial and temporal resolution. Herein, we report two strategies for the directed synthesis of metallic and bimetallic nanoscale architectures using in situ liquid STEM. In the first approach, we developed a direct write, template free method to fabricate self-supporting, hollow metallic nanostructures and we interpret the formation mechanisms based on direct observations of nucleation and growth. A liquid phase precursor solution containing reducible chemistries is encapsulated between electron transparent silicon nitride membranes. The electron beam that is generally used for imaging stimulates radiolysis; thereby, promoting the dissociation of water (H₂O) molecules and the formation of complex radical species such as aqueous electrons (e_aq^-) and other reducing and oxidizing species. The highly reducing radiolysis generated species assist in the chemical reduction of metal ions from the precursor solution resulting in the formation of a metallic nanocrystal seed, which then acts as a catalyst generated H₂ gas forming a metal encapsulated hollow nanobubble. In the second approach, we utilize a custom-built electron beam nanopositioning and scan generator system to precisely control the position and electron dose of focused electron beam in an aberration corrected STEM to fabricated metallic and bi-metallic nanostructured materials. These approaches enable fundamental electron beam interaction studies as well as open a new pathway for direct-write nanolithography from liquid phase solutions.


This research was supported by the Center for Nanophase Materials Sciences, which is a DOE Office of Science User Facility.

For many applications, oxygen transport at solid-solid, solid-adsorbate, or solid-gas interfaces underpin many interactions taking place during processes of relevance to energy conversion. In non-stoichiometric oxides, creation and annihilation of oxygen vacancies is key to technologically relevant functionalities impacting fields such as catalysis, oxygen ion conductors and photochemical reactions. Environmental transmission electron microscopy (ETEM) allows gas-solid and solid-solid interactions to be explored in oxides with atomic resolution under in situ and operando conditions [1-3]. The range of stimuli available in the ETEM continues to grow and at ASU we are able to perform experiments in gas, heat and light. The ability to extract structural and bonding information with advanced imaging and electron energy-loss spectroscopy allows the dynamic changes taking place inside and on reducible oxide nanoparticles to be explored. This provides a powerful approach for elucidating the atomic-level view of the reactions taking place on oxide surfaces.
The oxygen exchange reaction is a fundamental process taking place on oxide surfaces involving the creation and annihilation of oxygen vacancies. We have investigated the structural rearrangements taking place on different surfaces of CeO₂ and TiO₂ nanoparticles. With negative C_s aberration corrected electron microscopy, oxygen columns may be directly visualized under favorable conditions especially when direct exposure detectors are employed. The creation of oxygen vacancies can also result in local cation relaxations, which are often easier to detect during in situ experiments. The interplay between oxygen vacancy formation energy, cation displacements and the oxygen exchange reaction will be explored and discussed. The chemical reactions associated with oxygen vacancies is a strong function of the ambient gas environment. In the presence of water vapor, surface hydroxylation can take place leading to an order-to-disorder transformation. Vacancies may also play a major role in oxidation catalysis where oxygen transfers to gas adsorbates. This process may be critical in the Mars van Krevelen mechanism for reactions such as CO oxidation. The interplay between lattice oxygen, gas adsorbates and metal nanoparticles can have a major impact on catalyst selectivity. This is highlighted in hydrocarbon reforming where oxygen vacancy formation may play a decisive role in carbon deposition, leading to catalysts deactivation.
References
1. F. Tao, and P.A. Crozier, Chemical Reviews, 2016. 116(6): p. 3487-3539.
2. S. Chenna and P.A Crozier, ACS Catal. 2 (2012) 2395.
3. B.K Miller et al, Ultramicroscopy 2015. 156: p. 18-22.
4. The support from the National Science Foundation (NSF-CBET 1604971, DMR-1308085, CHE-1508667), 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.

CeO₂ (ceria) and ceria-based materials have many applications in heterogeneous catalysis, largely due to their oxygen exchange properties [1]. Under reducing conditions, oxygen vacancies form on ceria surfaces accompanied by a subsequent reduction of neighboring Ce⁴⁺ cations to Ce³⁺ [2]. Moreover, a lattice expansion (~0.1-0.2 Å) occurs once oxygen vacancies are created on ceria surfaces [3]. The relative ease with which oxygen vacancies are created/annihilated on CeO₂ surfaces is strongly surface and size dependent and is an important catalytic property of ceria nanoparticles [4]. In situ environmental transmission electron microscopy (ETEM) can be used to observe dynamical processes that occur on ceria surfaces in reducing and oxidizing environments. Thus, atomic-level studies of CeO₂ surfaces may enable the reaction pathways and active sites to be determined for oxygen exchange reactions, a fundamental outstanding problem in heterogeneous catalysis.

An aberration-corrected FEI Titan ETEM equipped with a Gatan K2 IS direct detection camera (with high detection quantum efficiency and capable of up to 1600 fps) was used to image CeO₂ nanocubes. The image corrector of the microscope was tuned to an optimum negative C_s condition to enhance contrast from weakly scattering oxygen atomic columns – enabling simultaneous imaging of Ce and O atomic column positions. Imaging of (111), (110), and (100) CeO₂ surfaces was done in reducing (pO₂ < 10^-7 Torr) and oxidizing (pO₂ = 0.5 Torr) environments. Images were acquired at 40 fps with 1 sec. exposures to capture dynamic motion of CeO₂ surfaces between individual frames. Under reducing conditions, large displacements of Ce atomic column positions were observed on (110) and (100) surfaces and at step edge sites of (111) surfaces; atomic displacements were likely caused by creation/annihilation of oxygen vacancies. Ce atomic motion was diminished on all surfaces during oxidizing conditions. Molecular dynamics (MD) and density functional theory (DFT) calculations will be used to relate relaxed surface structures and estimated oxygen vacancy concentrations to experimental data. Furthermore, Ce atomic displacements will be correlated to surface-dependent oxygen vacancy formation energies [5].


References:
[1] Gorte, R.J., AIChE Journal 56 (2010) p. 1126-1135.
[2] Skorodumova, N.V., et al, Physical Review Letters 89 (2002) p. 166601.
[3] Marrocchelli, D., et al, Advanced Functional Materials 22 (2012) p. 1958-1965.
[4] Trovarelli, A. and Llorca, J., ACS Catalysis 7 (2017) p. 4716-4735.
[5] We gratefully acknowledge support of NSF grant DMR-1308085, the use of ASU’s John M. Cowley Center for High Resolution Electron Microscopy and use of the K2 IS camera courtesy of Gatan.

Crystallization is a dynamic event that involves multiple phenomena and length scales, and many physical phenomena can deviate from classical theories as samples approach relevant length scales. In situ crystallization experiments inside a transmission electron microscope (TEM) using nanostructured metallic glasses (MGs) provide a unique platform to study directly crystallization kinetics and pathways at unprecedented time and spatial resolution [1]. Here, we use Pt-Ni-Cu-P MG nanorods and in situ TEM technique to explore unexpected crystallization phenomena at the nanoscale. We report the previously unknown asymmetry behavior between critical heating and cooling rates in MGs, which disappears when the sample size is below 35 nm [2]. Moreover, in contrast to bulk MG samples that form polymorphic phases, we observe a formation of an unusual single crystalline phase, which has previously never been reported [2]. The metastable phase can be attributed to the lack of nuclei as the samples size approaches the nucleation length scale, resulting in the starvation of nuclei. Our findings provide insights of cluster and nucleation density by controlling the MG nanorod diameter, which can also affect the resulting crystallization phases.
[1] S. W. Sohn, Y. Jung, Y. Xie, C. Osuji, J. Schroers, J. J. Cha, “Nanoscale size effects in crystallization of metallic glass nanorods,” Nature Communications 6:8157 (2015).
[2] S. W. Sohn, Y. Xie, Y. Jung, J. Schroers, J. J. Cha, “Tailoring crystallization phases in metallic glass nanorods via nucleus starvation,” Nature Communications (2017), accepted.

The supported metal nanocatalysts often show different catalytic activity depending on the supporting oxide materials. This implies that support-metal interaction can redirect the chemical reaction on the metal catalyst and that an additional phase (overlayer) can form on the surface or at the interface of the metal catalyst. Although little is known about the structure and the properties of the overlayers, promising results have been reported that some phases can selectively activate the adsorption of certain types of molecules. Here, we investigate the structure and the stability of the overlayer on the supported metal nanocatalyst during oxidation using environmental transmission electron microscope.
The results show that the overlayers form and encapsulate the Pd metal nanocatalysts supported on the reducible oxide, e.g. titania, of which cations migrated to the metal surface. The structure of the overlayer resembles the surface structure of the support oxide, but with the unique lattice parameters. The stability of the overlayer was studied by varying the temperature and the pressure and the results are addressed by the thermodynamic terms. The surface and the interfacial stress are also considered to explain the unique structures on the metal nanocatalyst.

Investigating the surface dynamic structure of metals in gaseous environments is important for catalysts and gas sensing. Palladium is a well–known material which is used for hydrogen storage, hydrogen sensing and exhaust catalysis. Environmental TEM (ETEM) has been used for revealing the phase transition of palladium in hydrogen. Though the interaction between palladium surface and gases has been extensively studied, much less is known on the surface process at the atomic scale.

Here, we investigated the surface structure of a wedge–shaped palladium specimen in gases including hydrogen and oxygen by means of in-situ atomic resolution ETEM. As is well known, the surface of palladium is oxidized in air and at room temperature by several nanometers. After introducing hydrogen (100Pa) in ETEM, the native oxide layer (PdO) was reduced to metallic fcc palladium even at room temperature. During ETEM observation in hydrogen, the stable high index facets such as {311} and {511} appeared on the surface. After the exposure to hydrogen, we exhausted hydrogen and then introduced oxygen (100Pa) in ETEM. We confirmed that the surface oxide layer was reproduced. However, after prolonged exposure to hydrogen over 90 min., we have found that the surface was not oxidized. We could not detect the transition from palladium to palladium-hydrogen by electron energy loss spectroscopy (EELS) even after the prolonged exposure to hydrogen.

The results indicate that, although the phase transition did not occur, the prolonged exposure to hydrogen makes the oxidation–prevention state on the palladium surface. We will present atomic-scale in-situ movies on the surface dynamics in palladium in various processing.

Transparent conductive oxides (TCOs) are key materials for many applications, including inorganic and organic solar cells and light emitting diodes. In some cases, the TCO layer is subject to a high temperature processing step, as in the fabrication of silicon/mesoporous perovskite tandem solar cells¹, which may deteriorate its electrical properties. To address this issue, the optoelectronic properties and microstructure of tin oxide (SnO₂)-based TCOs was investigated as a function of temperature. Amorphous films of either SnO₂ or SnO₂ with a few at% of Zn (ZTO) were deposited by sputtering² and annealed in an in situ X-ray diffractometer (XRD, in air or vacuum) before characterising their optoelectronic properties. Cross-sections of the annealed films were analysed by transmission electron microscopy (TEM) imaging and energy-dispersive X-ray spectroscopy (EDX). Separate in situ TEM experiments included the acquisition of images, electron energy-loss spectra, diffraction patterns (including fluctuation electron microscopy, FEM) as a function of temperature.

Crystallising SnO₂-based TCOs that are deposited amorphous results in a sharp decrease in their electrical properties. Barriers formed by grain boundaries and fine changes in the film chemistry through interactions with the annealing atmosphere impede carrier transport and reduce the number of free carriers, respectively. Crystallisation is postponed by local compressive strain in the amorphous network, which is induced by smaller cations or oxygen deficiencies as these are thought to dampen vibrations and increase the energy barrier to crystallisation.³ Temperature-dependant XRD data acquired in air indicated that the addition of 5 at% of Zn postponed the crystallisation temperature from 350°C to 590 °C, while annealing these ZTO films in vacuum prevents the passivation of oxygen vacancies, resulting in a further increase in the crystallization temperature by ~400 °C. Annealing at higher temperatures or longer times to promote grain growth was found to be detrimental. Indeed, EDX revealed that Zn evaporates > 750 °C in air and > 950 °C in vacuum, creating voids that further hinder carrier transport. XRD, TEM and FEM data indicate that the amorphous structure of ZTO remains largely unchanged before crystallization.⁴

Through this multimodal approach, the sensitivity of the material to the annealing atmosphere and temperature could be investigated, while typical artefacts of each technique could be identified and accounted for in the results analysis.

[1] Werner et al. (2016), Applied Physics Letters 109 23
[2] Morales-Masis et al. (2016) Adv. Funct. Mater. 26, 384
[3] Zhu et al. (2014) J. Appl. Phys. 115, 033512
[4] Rucavado et al. (2017), Physical Review B 95, 245204

The ability to controllably position single atoms inside a material is a prerequisite for the creation of next generation atomic-scale devices. Recently, there has been interest in using scanning transmission electron microscopes (STEMs) as tools for atomic-scale fabrication. Advances in aberration correction allow for precise control over the size and position of the electron probe. Recent work has demonstrated the ability to control “beam damage” to modify materials in a way that may prove beneficial to the overall material properties. This ability to directly manipulate materials at the atomic scale offers new opportunities for quantum engineering, such as qubits based on individual donors in a semiconductor. The current state of the art in manipulation on the atomic scale includes the positioning and imaging of single atoms using scanning tunneling microscopy (STM) or atomic force microscopy (AFM). However, the movement of matter by these techniques is limited to two-dimensional surfaces. In order to achieve quantum device fabrication, it is necessary to be able to controllably position individual atoms within a three-dimensional crystal, but previously there existed no method for targeted manipulation of a specific atom within a bulk structure.

Here we demonstrate for the first time the ability to controllably move and place subsurface bismuth dopants in a silicon crystal at room temperature using STEM. Bismuth has advantages over phosphorous for donor-based quantum computing, and its large atomic number relative to Si makes this system an ideal test case for single-atom manipulation in STEM. The ability to controllably move Bi dopants is indicated by two findings from our density functional theory calculations: (1) the strain induced by a substitutional Bi dopant allows for Si vacancies to be preferentially created adjacent to the Bi atoms, and (2) there is no significant energy barrier for the Bi atom to hop into these vacancies once they have formed. Using the electron beam, we exploit this vacancy-mediated motion to direct and place Bi dopants below the surface in specific columns within an oriented Si crystal, an important step towards creating a functional quantum device. This result represents a promising advance in atom-by-atom fabrication, enabling the tuning of material properties through controlled dopant positioning.

This work was sponsored in part by US Department of Energy, Office of Science, Basic Energy Sciences, Materials Sciences and Engineering Division (ARL) and by the Laboratory Directed Research and Development Program of Oak Ridge National Laboratory, managed by UT-Battelle, LLC, for the U. S. Department of Energy (BMH, JS, PCS). DFT calculations (HS, STP) were supported by Department of Energy grant DE-FG02-09ER46554 and were performed at the ERDC DSRC with support from subproject AFSNW32473012.

We employed an in-situ electrochemical cell in the transmission electron microscope (TEM) together with ex-situ time-of-flight, secondary-ion mass spectrometry (TOF-SIMS) depth profiling, and FIB - helium ion scanning microscope (HIM) imaging to detail the structural and compositional changes associated with Na/Na⁺ charging/discharging of 50 and 100 nm thin films of Sb. TOF-SIMS on a partially sodiated 100 nm Sb film gives a Na signal that progressively decreases towards the current collector, indicating that sodiation does not proceed uniformly. This heterogeneity will lead to local volumetric expansion gradients that would in turn serve as a major source of intrinsic stress in the microstructure. In-situ TEM shows time-dependent buckling and localized separation of the sodiated films from their TiN-Ge nanowire support, which is a mechanism of stress-relaxation. Localized horizontal fracture does not occur directly at the interface, but rather at a short distance away within the bulk of the Sb. HIM images of FIB cross-sections taken from sodiated half-cells, electrically disconnected and aged at room temperature, demonstrate non-uniform film swelling and the onset of analogous through-bulk separation. TOF-SIMS highlights time-dependent segregation of Na within the structure, both to the film-current collector interface and to the film surface where a solid electrolyte interphase (SEI) exists, agreeing with the electrochemical impedance results that show time-dependent increase of the films’ charge transfer resistance. We propose that Na segregation serves as a secondary source of stress relief, which occurs over somewhat longer time scales.

Significant developments in micro-electrical-mechanical systems (MEMS) - based devices for use in transmission electron microscopy (TEM) sample holders have recently led to the commercialization of windowed gas cells that now enable the atomic-resolution visualization of phenomena occurring during gas-solid interactions at atmospheric pressure. In situ TEM study under atmospheric pressures provides unique information that is beneficial to correlating the structure-properties relationship of catalytic nanomaterials, particularly under real gaseous environments. In this talk, we illustrate the capability of this device as applied to our study of three catalyst systems: (1) In situ kinetic growth of free standing Pt nanowires; (2) Facet-dependent oxidation of Pt₃Co nanoparticles; (3) In situ investigation of the stability of Ir-dimer catalyst.
Atomic-scale insights into the kinetic growth of catalytic nanomaterials are critical for optimization of the synthesis process. We observe the growth of free standing Pt nanowire-network in solid state under a gaseous environment, by using in situ TEM approach. The captured dynamics controlled by the gas adsorbate shows a three-stage process including nucleation, coalescence and oriented attachment, which represent a potential route for nanostructure synthesis.
Moreover, an understanding of how the surface composition and structure of catalytic nanomaterials may be controlled by external means is useful for their efficient production. By taking advantage of the high spatial resolution of TEM, we study the surface composition and the dynamics involved in facet-dependent oxidation of equilibrium-shaped Pt₃Co nanoparticles in an initially disordered state. In brief, using our advanced in situ gas cell technique, evolution of the surface of the Pt₃Co nanoparticles was monitored at the atomic scale during their exposure to an oxygen atmosphere at elevated temperature, and it was found that Co segregation and oxidation take place on {111} surfaces but not on {100} surfaces. Another example we show here is the investigation of the stability of Ir-dimer catalyst. Beyond single atom catalysts, emerging Ir dimer atom catalyst has been successfully fabricated on iron oxide support, showing an ultra high durability for the water-splitting. The structure of Ir-dimer catalyst has experimentally demonstrated by atomic-resolution STEM imaging coupled with EDX/EELS elemental analysis. It is found that the Ir dimer active sites were strongly bonded on the iron oxide support through in situ TEM observation.

Heterogeneous catalysts accelerate chemical reactions by reducing the required reaction activation energy. The specific locations on the catalyst at which the activation energy is lowest – the so-called active sites – are poorly understood because catalytically relevant atomic structures only emerge under reaction conditions. Even with in situ TEM, determining atomic-level structure-activity relationships is difficult still due to the large number of surface structures forming dynamically during catalysis.

Discerning catalytically relevant structures from spectators may be possible by studying supported metal systems in which the active sites are localized to the metal-support interface. The rate of CO oxidation over Pt/CeO₂ has been shown to depend strongly on the perimeter length of the metal-support interface [1]. The interface facilitates the reaction through a Mars van Krevelen mechanism, whereby the CeO₂ lattice provides oxygen to oxidize CO to CO₂. Interfacial structures that enhance oxygen transfer, for example, through strain, may improve catalytic activity. However, at present there is no experimental data on the atomic structures that comprise the Pt/CeO₂ interface during catalysis.

Here, we seek to establish atomic-level structure-activity relationships by identifying and characterizing the active sites for CO oxidation over nanostructured Pt/CeO₂. Nanostructured CeO₂ cubes will be synthesized and loaded with 2 wt. % Pt by a photodeposition technique [2]. A quartz tube microreactor coupled to a gas chromatograph will be used to determine turn-over frequencies and activation energies. The ex situ reactor data will inform the reaction space explored during in situ TEM experiments. Atomic structures forming at and in the vicinity of the Pt/CeO₂ interface during CO oxidation will be visualized with aberration-corrected environmental TEM (AC-ETEM). The observed structures will be correlated with the catalyst’s overall activity through in situ mass spectrometry. Detectable conversions will be achieved through a unique porous microfiber pellet TEM specimen technique [3]. Dynamic particle and interfacial restructuring will be quantified by evaluating changes in particle morphology and interfacial lattice coherency. The observed structures will inform theorists simulating catalyst behavior and experimentalists studying other supported-metal systems. Also, understanding structure-activity relationships in this system will facilitate the engineering of highly active catalysts for the energy and environmental remediation reactions where Pt/CeO₂ is indispensably used.

[1] Cargnello, M. et al; Science 341, 771-773 (2013)
[2] Vincent, J. L. et al; Microscopy and Microanalysis 23(S1), 966-967 (2017)
[3] Miller, B. K. et al; Ultramicroscopy 156, 18-22 (2015)
[4] We gratefully acknowledge the support of NSF grant CBET-1604971 and ASU’s John M. Cowley Center for High Resolution Electron Microscopy.

Ceria supported Pt nanocatalysts have demonstrated superior catalytic properties for water gas shift reaction [1]. By downsizing noble metal nanoparticles to small clusters or single atoms, the effective use of noble metals can be maximized and unique catalytic properties can be realized [2]. In order to effectively increase the surface area of the ceria support and to explore the unique catalytic properties of small ceria clusters, we designed a hierarchical system that consists of Pt single atoms trapped onto small ceria clusters which are supported on ZnO nanowires. Our goal is to use the ceria nanoclusters to stabilize the Pt single atoms while simultaneously utilize the Pt-ceria interaction to tune the catalytic properties of the Pt₁-CeO₂-ZnO nanoscale architected catalytic system. In order to implement such a strategy, we used aberration-corrected environmental TEM to investigate the formation processes of the ceria clusters on the ZnO NWs. We further studied how the Pt single atoms or small clusters interact with the ceria nanoclusters under gas environment or at elevated temperatures. Dynamic observations of the movement of ceria clusters and the sintering behavior of the Pt single atoms will be discussed [3].


[1] Q Fu et al., Science 301 (2003), p. 935.
[2] J Liu, ACS Catalysis 7 (2017), p. 34.
[3] This work was funded by NSF under CHE-1465057. We acknowledge the use of facilities in the John M. Cowley Center for High Resolution Electron Microscopy at Arizona State University.

Compared with conventional nanoparticle-sized catalytic systems, isolated atoms or atomic-layered clusters anchored on dedicatedly-designed nanostructures can not only dramatically increase the atom utilization rate of expensive noble metals but also enormously improve their catalytic activity and selectivity. In situ scanning transmission electron microscopy (STEM) with single-atom-analysis capabilities is recognized to be a powerful and straightforward approach for studying the underlying mechanisms of thermal or chemical stability and enhanced catalytic performance of heterogeneous catalysts. Using the aberration-corrected STEM equipped with in-situ gas-cell or heating holders (i.e. Protochips Atmosphere and Fusion systems), we recently observed different forms of atomic distribution of platinum atoms (e.g. single atoms and atomic-thin layered clusters) stabilized on a unique diamond-graphene core-shell nanostructure and explored their thermal behaviors under reaction-relevant chemical atmosphere, through which their robust sintering-resistance in contrast to platinum catalysts supported on commercial alumina powders was directly demonstrated. Therefore, this carbon-supported catalyst possesses promising industrial application potential for low-temperature direct dehydrogenation of n-butane. We also found that its catalytic performance is highly correlated to the tunable hybrid level (i.e. graphene-layer numbers) and surface defect status of such core-shell carbon supports, which offers a novel platform for in-situ TEM investigation into the microscopic mechanisms of the strong metal-support interactions.

Recently, materials science benefits largely from in-situ transmission electron microscopy (TEM) using specialized holders, which separate small reactor volumes of (the sample and surrounding) liquids or gases from the high vacuum of the microscopy column. With these novel approaches, the mechanism of dynamical process can be revealed, reaction kinetics can be quantified, and even electrochemical processes can be studied in-situ in the microscope. Here, we report on the electrochemical deposition of iron on amorphous carbon electrodes in a liquid cell. The experiments were performed on a JEM-2100 (JEOL Ltd., Tokyo, Japan) microscope operated at 200 kV. The microscope is equipped with an energy dispersive X-ray spectrometer (EDXS). A Poseidon 510 TEM (Protochips Inc., Raleigh, NC, USA) holder was used together with a Gamry Reference 600 potentiostat (Gamry Instruments, Warminster, PA, USA) for conducting the electrochemical experiments (chronoamperometry). The liquid cell is comprised of two chips with 50 nm thin amorphous SiN_x windows, where the smaller one has a spacer height of 50 nm thereby controlling the total thickness of the liquid layer between the two windows. The upper, larger chip contains the amorphous carbon working electrode, the Pt reference and the counter electrodes. An iron sulphate containing solution with pH = 2 was used as electrolyte. Upon applying a potential of -1.2 V vs. Pt, the electrodeposition of an amorphous layer of iron is stimulated, which subsequently crystallizes to Fe and FeO_x, respectively. The nucleation of the crystal growth occurs instantaneously upon reaching a critical potential. The resulting deposition is discontinuous. In order to minimize artifacts such as the beam-induced radiolysis of water or secondary radical chemistry, the electron beam was blocked during the electrodeposition. The growth mechanism and the composition of the amorphous and crystalline deposits as derived from supporting EELS measurements will be discussed.

Fundamental study on the nucleation and growth of nanoparticles found itself at the forefront with the widespread use of liquid Transmission Electron Microscopy (LTEM) to investigate dynamic growth and assembly processes [1]. The most noble study use LTEM to observe and quantify the nucleation and growth of single colloidal platinum (Pt) nanoparticles [2]. Since then, several theories on the nucleation and growth of nanoparticles have been validated. In other study, other information was even added to existing knowledge. One instance was the revelation of a new hybrid growth process of gold (Au) on Pt icosahedral nanoparticles to form core-shell structures [3]. The aforementioned studies have been carried out by focusing on a small number of particles. Here, the electron beam-induced growth of an ensemble of Au nanoparticles was performed in situ. It was observed that the growth does not follow any single growth mode as proposed by the Liftshitz-Slyosov-Wagner (LSW) theory but is more likely a combination of both LSW growth and growth via coalescence.

[1] Ngo, T.; Yang, H. J. Phys. Chem. Lett. 2015, 6, 5051-5061.
[2] Zheng, H.; Smith, R. K.; Jun, Y. W.; Kisielowski, C.; Dahmen, U.; Alivisators, A. P. Science 2009, 324 (5932), 1309-1312.
[3] Wu, J.; Gao, W.; Wen, J.; Miller, D. J.; Lu, P.; Zuo, J.-M.; Yang, H. Nano Lett. 2015, 15 (4), 2711-2715.

The use of in situ liquid cell transmission electron microscopy (LCTEM) provides a powerful tool to directly observe nanoparticles (NPs) suspended in solution, but care must be taken to understand and control how electron beam illumination affects not just the NPs, but also their local environment, which can then in turn affect the NPs. In this report, we present direct visualization of the dynamics of oleic-acid-capped PbTe NPs using LCTEM equipped with a high speed camera (Protochips Poseidon P500 wet cell holder in a JEOL JEM2200FS operating at 200kV and a Gatan OneView with a full resolution 4000 x 4000 pixel images acquired at 25 fps which is binnable to 50 and 100 fps). We observe electron dose rate dependent etching and dissolution of PbTe NPs in toluene; ranging from no perceivable effect on the particles with lower dose rates (30 e^-/Ų/sec) to full dissolution with higher dose rates (150 e^-/Ų/sec). In addition, the dissolution effects are also observed to depend on local NP concentration. Despite suspension in an organic solvent (toluene) which produces very few chemically-active species under electron (beta particle) radiation, the radiolysis of the water adsorbed on the LCTEM SiN_x membrane during PbTe NPs loading results in multiple radiolytically-produced species. These radicals include several oxidizing species which cause etching of NPs. This behavior is initiated only after a threshold electron dose rate is received. This report provides direct evidence for the dissolution of PbTe NPs in organic solvents due to the radiolysis of adsorbed water. To prevent this dissolution, rigorous protocols to preserve anhydrous conditions, different capping agents, and sub-threshold dose rate imaging can be used. The authors acknowledge funding from the Office of Naval Research (Naval Research Laboratory Basic Research Program). This research was performed while N.B. and D.L.W. held National Research Council Research Associate Awards at the U.S. Naval Research Laboratory.

CuO, a p-type metal oxide semiconductor has been accepted as a potential material for gas sensor, bio sensors, photodetectors and batteries because of their abundant source, reasonable cost, ease of fabrication and compact size. The size, morphology, dimensionality and surface properties of nanostructures are important parameters that determine the performances of nanomaterials. Moreover, properties of CuO can be tuned by doping it with various metal atoms in order to satisfy the specific needs and applications. In the present work a facile and wet chemical synthesis method has been employed for the preparation of pure and Zn-doped polycrystalline 1-D CuO nanochain (Cu_1-XZn_XO, x=0 (S0), 0.01 (S1), 0.03 (S3), 0.05 (S5)). The exploration of the structural, morphological and optical properties of the pure and Zn-doped CuO samples have been carried by X-ray diffraction (XRD), Field emission scanning electron microscope (FE-SEM), energy dispersive X-ray analysis (EDX), Transmission electron microscope (TEM) and fluorescence spectroscopy. The Rietveld refinements of XRD data of the samples have been done using Fullprof software and the results mark the formation of single phase monoclinic structure and also confirmed the successful doping of Zn ions into CuO lattice. Moreover, decrease in cell volume and particle size has been observed for Zn doped samples, thus, showing the influence of Zn substitution on the crystalline structure of CuO. FESEM images of the samples clearly present the bundles of nanochains constituting of nanoparticles whose diameter gradually reduced from ~31.7 nm to ~23.2 nm by increasing Zn doping from x=0 to 0.05 in the CuO sample. Furthermore, EDX data confirm the presence of Cu, O and Zn elements in the sample. TEM results show that the nanoparticles become organized and assembled to form a nanochain like structure. High resolution transmission electron microscope (HRTEM) images of S0 and S3 samples display the lattice fringes of spacing 0.232 nm that correspond to (111) crystalline plane of the monoclinic CuO. Selected area electron diffraction (SAED) pattern shows several concentric diffraction rings masked by spots, thus, indicating the polycrystalline nature of the nanochains. From the fluorescence spectra of the samples, it has been observed that doping of Zn in CuO creates oxygen vacancies which are manifested by an increase in intensity of visible emission peak. This study suggest that the Zn doped CuO 1-D nanochain could be promising materials for optical and optoelectronic applications at a low cost.

The ferroelectrics nanostructures have numerous applications such as the electronic, magnetoelectric, photovoltaic, actuator and the nonvolatile memory ones [1-7]. For them, the stress-induced multi-phase coexistence such as morphtropic phase boundary (MPB) and thermaltropic phae boundary (TPB) can dramatically enhance their piezoelectricity and flexibility [2,5], leading to attractive potential applications. To understand the mechanism of the stress-induced multi-phase coexistence, we performed high-pressure in-situ transmission electron microscopy (TEM) and Diamond Anvil Cell (DAC) Raman scattering studies, on the series of nanometer free-standing single crystal (i.e clamping-effect-free and dislocation-free) samples prepared by an improved FIB methods.

ACKNOWLEDGEMENTS
This work has been supported by the Natural Science Foundation of Jiangsu Province,China (Grant No.BK20151382) , and the National Center for Electron Microscopy of Molecular Foundry at Lawrence Berkeley National Laboratory, for the support under the DOE Grant for user facilities.

REFERENCES
1. G. Catalan, J. Seidel, R. Ramesh, J. F. Scott, Rev. Modern Phys. 84, 119 (2012)
2. T. T. A. Lummen, et.al. Nat. Comm. 5, 3172 (2014).
3. M. Dawber, K. M. Rabe, J. F. Scott, Rev. Mod. Phys., 77, 1083 (2005).
4. J. F. Scott, Science 315, 954 (2007).
5. H. Lu, C. W. Bark, D. Esque de los Ojos, J. Alcala, C. B. Eom, G. Catalan, A. Gruverman, Science, 336, 59 (2012).
6. Y. L. Tang,Y. L. Zhu,X. L. Ma,S. J. Pennycook, et. al. Science 348, 547 (2015 ).
7. C. T. Nelson, P. Gao, J. R. Jokisaari, C. Heikes, C. Adamo, A. Melville, S. Baek, C. M. Folkman, B. Winchester, Y. Gu, Y. Liu, K. Zhang, E. Wang, J. Li, L. Q. Chen, C. B. Eom, D. G. Schlom, X. Q. Pan, Science 334, 968 (2011).

We have applied aberration-corrected transmission electron microscopy (TEM) imaging and electron energy loss spectroscopy (EELS) to study the structure and chemistry of epitaxial ceria thin films, grown by pulsed laser deposition onto (001) yttria-stabilised zirconia (YSZ) substrates [1]. There are few observable defects apart from the expected mismatch interfacial dislocations and so the films would be expected to have good potential for applications. Particular attention is paid to the transition from fully to partially coherent interfacial structures, and this is correlated to x-ray diffraction measurements [2]. Under high electron beam dose rate (above about 6,000 e-/Ųs) domains of an ordered structure appear and these are interpreted as being created by oxygen vacancy ordering. The ordered structure does not appear at lower lose rates (ca. 2,600 e-/Ųs) and can be removed by imaging under 1 mbar oxygen gas in an environmental TEM. EELS confirms that there is both oxygen deficiency and the associated increase in Ce³⁺ versus Ce⁴⁺ cations in the ordered domains. In situ high resolution TEM recordings show the formation of the ordered domains, which can be analyzed by the classical Avrami approach, as well as atomic migration along the ceria thin film (001) surface. The influence of thin film strain, by using different substrates, will also be considered [3].

[1] R. Sinclair, S. C. Lee, Y. Shi, W. C. Chueh, Ultramicroscopy, 2017, 176, 200.
[2] Y. Shi, S. C. Lee, M. Monti, C. Wang, Z. A. Feng, W. D. Nix, M. F. Toney, R. Sinclair, W. C. Chueh, ACS Nano, 2016, 10, 9938
[3] C. B. Gopal, M. Garcia-Melchor, S. C. Lee, Y. Shi, A. Shavorsky, M. Monti, Z. Guan, R. Sinclair, H. Bluhm, A. Vojvodic, W. C. Chueh, Nature Comm., 2017, 8, 15360.

At any stage in research and development of (new) functional nanomaterials, studies of these nanomaterials’ structure, properties, and function are critical, including detailed atomic-scale insights. To our advantage, in-situ electron microscopy (EM) enables visualization of structural evolution and thereby has become a powerful tool for characterizing actual state and function of those nanostructures under (near) operational conditions [1,2].
Applying atomic-scale EM techniques in in situ studies is, however, still extremely demanding. A key challenge is to establish in situ conditions in the close vicinity of the specimen while maintaining the microscope’s overall performance and stability. Ongoing activities also concentrate on methodological aspects of atom sensitive imaging while controlling electron beam / structure interactions [3,4]. Recent advances in in-situ atomic-scale EM will be highlighted.
Optimized in situ stages - based on MEMS technology – enable more accurate realization of experimental in situ conditions, and the opportunity to detect the material’s function simultaneously as well (= operando). Sample preparation in conjunction with MEMS cartridges as sample supports have been adopted [5]. Fine temperature control enables quantitative studies at elevated temperatures [6]. Others have utilized EM methods to measure the actual temperature (gradient) of the specimen more precisely [7]. Moreover, the integration of a heater into a gas-flow MEMS nanoreactor enables operando EM combining structural characterization of e.g. catalytic materials with simultaneous measurement of its activity for gaseous reactions [8].
These advancements open up for unprecedented experiments of dynamic phenomena in materials science to understand the structure-property-function relationship on the (sub)nanometer length scale.


[1] J. R. Jinschek, Chemical Communications 50 (2014) 2696.
[2] J. R. Jinschek, Current Opinion in Solid State & Materials Science 21 (2017) 77.
[3] J. R. Jinschek et al., Micron 43 (2012) 1156.
[4] S. Helveg, et al., Micron 68 (2015) 176.
[5] S. Vijayan, et al., Microscopy and Microanalysis 23 (2017) 708.
[6] L. Mele, et al., Microscopy Research and Technique 79 (2016) 239.
[7] F. Niekiel, et al. Ultramicroscopy 176 (2017) 161.
[8] S. B. Vendelbo, et al., Nature Materials 13 (2014) 884.

In an effort to identify suitable materials for storing high level radioactive waste, recent research suggests that pure and doped ZrO₂ nanocrystals could be suitable candidates. It has been suggested that when placed in an intense high level radiation environment these nanocrystals exhibit enhanced resistance to structural disintegration compared to their bulk counter parts. The mechanisms through which radiation loses energy results in a very large flux of low energy electrons. To date, radiation damage in ZrO₂ nanocrystals by electrons with energies below 200 keV has not been reported. We have found direct evidence of extensive radiation damage in ZrO₂ nanocrystals due to intense bombardment (2 x 10² electrons/nm² sec) by electrons with beam energies between 60 keV and 120 keV. Intense electron irradiation in this energy range produces a significant loss of Zr and O from the nanocrystal resulting in TEM images of hollow or “ghost” like structures which retain the stoichiometry of the nanocrystal. We present an explanation based on the Knotek-Feibelman process.

Real-time processing of digitally acquired data dates back to at least the 1980s¹, but decades later, most data processing still takes place long after acquisition, on a separate PC. Implementing real-time processing is especially important for in-situ experiments, in which the next experimental step is often determined by the changes observed in the microscope.
While computer processing power has increased exponentially, the amount of data collected in a single microscope session has also increased dramatically. Using Gatan’s latest CMOS based in-situ cameras, more than a TB of video can easily be acquired in a single day, even when the user is careful to limit the amount of data captured. This increases the difficulty to process the data in real-time. Although live FFTs are commonly available for analyzing data in real-time, as cameras get faster, only a subset of the acquired frames are actually analyzed at the microscope. For true real-time analysis, all the data generated by the detector would be processed live.
Gatan is working toward flexible solutions for live processing on the full stream of data from high-speed cameras. In this work, a scalable parallel file system with hundreds of TB of storage from DDN was used to store all data as it was acquired. While data is still being written to the filesystem, a 2^nd PC running Gatan’s GMS (Digital Micrograph) software is used to process this data via scripting. In this case, the script converted a selected area electron diffraction pattern into a 1D profile, in which subtle changes could be more easily discerned by the TEM operator. This processed data was again saved in DM4 format on the DDN file system and the analyzed data was accessed by GMS on the computer at the TEM displaying the analyzed data at the TEM next to the live view window. All of this was done in a few seconds, allowing the user to modify the experimental conditions based on this feedback.
This proof-of-concept experiment can be extended so that any amount of processing is applied and the analysis performed on an HPC platform. The key is this experiment demonstrates that the computational demands of the data analysis do not conflict with the demands of the data capture. The near real-time feedback to the TEM operator enabled phase changes to be easily recognized. Future advances will speed the data processing, increasing the frame rate of the processed data window.
¹W. Krakow, Applications of real-time image processing for electron microscopy, Ultramicroscopy. 18 (1985) 197–210.

Advancement in Li-ion battery technology is dependent on the development of new materials/electrolytes that can enable high capacity and provide stable, enhanced charge transfer across the electrode-electrolyte interface. To achieve these metrics, we have explored Li-metal anodes to understand the current limitations in their commercial use due to heterogenous morphology (dendrites) and performance within a solvent-in-salt electrolyte. We implement nanoscale imaging to understand the environmental and electrochemical parameters that control the Li morphology and solid electrolyte interphase structure, which determine the overall performance of the electrode. In-situ scanning/transmission electron microscopy (S/TEM) has enabled the exploration of the initial stages of Li deposition which can provide the baseline structure upon which the electrode evolution occurs [1].
All results were achieved using a custom microfabricated liquid-cell, developed at Sandia National Laboratories and offered through the user program at the Center for Integrated Nanotechnologies, the Electrochemical TEM Discovery Platform [2]. This platform provides customization of 10 electrodes within a single experimental cell, for multiple experiments within the same environment. Integration of Li-containing counter and reference electrodes provides the ability to cycle anode materials as part of a full cell, to monitor electrochemical processes at relevant potentials.
Many in-situ TEM cells suffer from side reactions with contaminants within the cell, due to electrode processing conditions or non-standard electrode materials for battery technologies. In this presentation, the challenges that are faced for using in-situ liquid-cell S/TEM to study processes relevant to bulk-scale battery processes will be discussed. In addition to electron beam effects, as these experiments were performed with the whole of the working electrodes being imaged to directly correlate the electrochemical data to the structural changes on the electrode surface. The real-time observation of Li deposition and stripping dynamics at the nanoscale has been used to identify the dominant factor in non-uniform Li morphology and has highlighted the complication of self-discharge as a potential barrier to commercialization. Our comparison of microscale electrochemcial observations in the TEM as compared to macroscale coin cell electrochemcial data for Li-metal anodes will be discussed [3,4].
References:
[1] A. J. Leenheer et al, ACS Nano 9 (2015), p. 4379.
[2] A. J. Leenheer et al, J. Microelectromech. S. 99 (2015), p. 1061.
[3] K. L. Harrison et al, ‘Lithium Self-Discharge and its Prevention,’ submitted.
[4] Sandia National Laboratories is a multi-mission laboratory managed and operated by National Technology and Engineering Solutions of Sandia, LLC., a wholly owned subsidiary of Honeywell International, Inc., for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-NA-0003525.

In this presentation, we will describe a vision for a future research paradigm, wherein a tight coupling of in-situ and operando experimental methods, data analytics and automated data analysis are coupled with artificial intelligence to direct how we use electron microscopy to characterize the mechanisms by which processing/structure/property relationships are determined. The presentation will be forward looking, and will incorporate research results and ideas culled from a variety of sources and authors: it will not be a typical presentation reviewing research. First, we will describe the motivations for working towards this type of research paradigm. These include the desire to speed up the rate of scientific discovery and time to market, as well as a more pedestrian desire to maximize the utilization of expensive instrument time. Second, we will review examples of autonomous research methods, both through data mining of the literature [1], and through the use of real time feedback and artificial intelligence methods to direct experimental outcomes.[2] Specifically, we will discuss how these approaches may be utilized in electron microscopy research in the near future, and the developments needed to bring this to reality. Third, we will describe how this approach can be used to explicitly and efficiently test operative hypotheses, and to efficiently understand the relevant experimental parameter space. This yields insight as to where detailed experimentation is most valuable. In this portion of the talk, the need for operando methods and correlative experimentation will be emphasized.[3] Finally, we will discuss this evolving research paradigm as it exists both within and provides challenges to established theories of scientific discovery.[4,5]

1. Science Mapping: A Systematic Review of the Literature, Chaomei Chen, Journal of Data and Information Science, 2, 1-40, 2017
2. Autonomy in materials research: a case study in carbon nanotube growth; P. Nikolaev, D. Hooper, F. Webber, R. Rao, K. Decker, M. Krein, J. Poleski, R. Barto and B. Maruyama, npj Computational Materials, 2, 16031 (2016)
3. Complex structural dynamics of nanocatalysts revealed in Operando conditions by correlated imaging and spectroscopy probes, Y. Li, D. Zakharov, S. Zhao, R. Tappero, U. Jung, A. Elsen, Ph Baumann, Ralph G. Nuzzo, E. A. Stach, and A. I. Frenkel.." Nature Comm., 6, 7583
2015.
4. The Structure of Scientific Revolutions, T.S. Kuhn, Chicago: University of Chicago Press, 1962.
5. A Sociological Theory of Scientific Change, S. Fuchs, Social Forces, 71(4), 933-953, 1993.

Cathode microstructure plays a vital role in the electrochemical performance of Li ion batteries. Understanding the crystallization process during annealing of electrode thin films can provide potential new avenues to control electrode microstructure in thin film batteries. A novel in-situ transmission electron microscopy (TEM) experiment is being designed to correlate microstructural evolutions to degradation mechanisms in LiV₃O₈ and V₂O₅ thin film battery systems. This design uses thin films of amorphous solid electrolyte (LiPON), amorphous carbon current collectors and a lithiated amorphous silicon anode.
Our first focus is understanding in detail the crystallization of the cathode films, which are amorphous as deposited. For these studies, c.50-100nm Li-V-O and V-O films are RF sputter deposited from LiV₃O₈ (Ar:O₂=3:1) and V targets (Ar:O₂=1:1) onto Si TEM discs which contain 50 nm thick silicon nitride membrane windows. Thermal annealing experiments are carried out both in vacuum (i.e. in-situ to the TEM) and Ar atmospheres for the Li-V-O thin films, and vacuum and ambient atmospheres for V-O films. Crystallization of Li-V-O thin films, depending on the annealing atmosphere (vacuum vs Ar) and temperature, resulted in either completely delithiated phases (V₂O₃, VO₂ and V₂O₅) or uncatalogued Li-V-O phases but not the expected LiV₃O₈ phase highlighting the challenge in maintaining Li concentration. We overcome this challenge by using thicker films (~1 μm), where the Li concentration at the middle of these films is close to LiV₃O₈ phase and formed LiV₃O₈ electron transparent films (~100nm) by controlled FIB nanofabrication.
In-situ TEM annealing results of sputter deposited V-O films show the onset of V₂O₅ phase crystallization around 250^oC in the areas not exposed to the e-beam. However, annealing to higher temperatures- 400^oC and 500^oC(1hr) in TEM atmosphere (pressure:10^-6 Pa) results in partial phase transformation to oxygen deficient V-O phases (VO₂, V₄O₇ and V₆O₁₃) depending on temperature and electron beam irradiation. Annealing of these as-deposited thin films in ambient atmosphere up to 500^oC (1hr) results in desired V₂O₅ phase. Thus, annealing in ambient atmosphere is observed to form V₂O₅ thin films with different microstructures for in-situ TEM thin film battery testing.
These studies establish the ability to maintain suitable cathode stoichiometry and microstructure in ultra-thin film form, which is a crucial step for the in-situ TEM battery studies we are developing, and potentially for other applications.
Acknowledgements: - This work is supported as part of the Center for Mesoscale Transport Properties, an Energy Frontier Research Center supported by the U.S. Dept. of Energy, Office of Science, Basic Energy Sciences (award #DE-SC0012673). Work at RPI made extensive use of the cleanroom and characterization facilities in the Center for Materials, Devices and Integrated Systems (cMDIS).

With ever increasing interest to prospective nanostructured materials for electrical energy storage (EES), the fact that strong correlations exist between the volume plasmon energy (E_p) and other intrinsic solid state parameters (valence electron density, cohesive energy, elastic moduli, thermal and electrical conductivities, etc.), open opportunities to probe and tailor properties of candidate materials in order to provide the ultimate performance and fully utilize their unique technological advantages. We discuss here relationships between the gravimetric capacity (C_g), elastic moduli, cohesive energy density and E_p of some prospective EES materials composed of Group I-IV elements. Periodic trends and factors influencing physico-mechanical properties, C_g and energy density of EES materials also will be considered. We show that the origin of connections between the material properties and the E_p lies in the nature of electron-ion interactions and in the essentially exponential decay of electron density with interatomic distance. For EES materials with metallic and preferentially covalent bonding, this is established by the universal binding energy relation (UBER), which describes the shape of the binding energy curve in a wide range of situations varying from cohesion in bulk solids to bimetallic adhesion, chemisorption on metal surfaces and bonding. In the UBER, the bulk modulus, B_m, is related to E_coh and the equilibrium Wigner-Seitz (WS) atomic radius, r_wse, via the dimensionless anharmonicity parameter:
η = r_wse/l, as B_m = (1/12π)r_wse^-3E_cohη² = (4π²e₀m/9e²h²N_ve) E_cohη²(E_p²-E_g²) (1),
where l is the characteristic length describing the width of the binding energy curve or the range over which strong forces act (this sets the range of the Hooke’s-law region), N_ve is the number of valence electrons per atom, and E_g is the band gap. From Eqn. (1), B_m ∝ E_coh/V_wse ∝ E_p²–E_g², where V_wse is the volume of the Wigner-Seitz cell at equilibrium. B_m, and E_coh/V_wse and can be linearized in a log-log scale as a function of E_p²-E_g² and described by an universal equation of the type P_m = A(E_p²-E_g²)^B, where P_m = C_g, B_m, or E_coh/V_wse, and A and B are lsf structure-dependent parameters. For B_m, and E_coh/V_wse, scaling can be deduced from Eqn. (1) since the valence electron density governs variations of both B_m and E_coh/V_wse with E_p and contribution from η only slightly reduces the exponential factor. Universality and scaling in E_p-property relationships indicate that E_p is an invaluable parameter to investigate the properties of engineering EES materials. We will describe how to employ these relations in order to determine and map nanoscale physico-mechanical properties of EES materials in situ using spatially-resolved low-loss EELS and STEM spectroscopic imaging and illustrate some applications with selected examples, including nanophase silicon and Si-Li alloys and conductive carbons utilized in composite electrodes for high-performance rechargeable batteries.

To probe the effect of electron beam irradiation in commercial electrochemical liquid cell transmission electron microscope (LCTEM) holders,[1] we use a model system of platinum electrodes and sulfuric acid to investigate how intermediate-to-low intensity electron beams affect cyclic voltammograms (CVs), and how the CVs change with time and beam current during irradiation in a flowing electrolyte. Using the largest field-of-view and small beam currents for the ‘standard’ viewing conditions in our TEM (a JEOL JEM2200FS), we see a potential shift (~ 100mV) in the entire CV when turning the beam on and off during potential cycling. For the same electrode geometry, increased electron beam currents cause increased currents in the oxidative and reductive portions of the CV, but the static potential shift remains constant. However, a larger physical separation between the reference and working electrode does change the static potential shift. Following Jiang, [2] we understand the static potential shift to arise from positive charge on both the top and bottom silicon nitride membrane. Using the measured potential shift and electrode geometry, we estimate the presence of ~10-100 aC at the top and bottom membrane. Finally, we see that ‘beam-off’ CVs can be recovered following irradiation after a brief period without irradiation, which indicates the flow of liquid through the cell allows new electrolyte to both replace the irradiated portions of electrolyte and to dissipate static charges such that the original, beam-off CV is recovered, but there is a finite time associated with this recovery which may impact the maximum speed of LCTEM experiments. We will present these results on this canonical electrochemical system and discuss the impact on LCTEM. The authors acknowledge funding from the Office of Naval Research (Naval Research Laboratory Basic Research Program). This research was performed while N.B. held a National Research Council Research Associate Award at the U.S. Naval Research Laboratory.

[1] H. Zheng, Y. S. Meng, Y. Zhu (Guest Editors) MRS Bulletin Vol. 40, Issue 1 (Jan. 2015)
[2] Nan Jiang, Micron Vol. 83, 79-92 (April 2016)

Yttrium iron garnet (YIG) has a range of applications in the fields of spintronics and magneto-optics due to its low damping constant and Faraday rotation. Utilizing thin YIG films, these properties can be taken advantage of in nanoscale devices, but attempts to do so have necessitated the usage of lattice matched substrates, such as gallium gadolinium garnet, limiting practical applications. Recent advances towards incorporating YIG into silicon based devices have been promising, showing that YIG can be grown on silicon with reasonably high quality by sputtering and annealing. These studies have rarely examined crystallization mechanisms or studied any fundamental crystallization behavior. Here we report an in situ TEM study of the crystallization behavior of YIG on SiO₂, specifically looking at fundamental crystallization properties and mechanisms.
Thin amorphous YIG films (20-100 nm) were deposited on SiO₂ TEM windows by RF sputtering. These YIG films were then crystallized in situ by a laser in the TEM column. The crystallization phenomenon was observed in bright-field as well as diffraction mode TEM. Bright-field TEM allows for studying nucleation rate, crystallization front velocity as well as crystallization geometries. Diffraction TEM allows for studying phase formation, percent crystallinity, instantaneous temperature, as well as texturing. Results from time resolved diffraction patterns show Avrami crystallization behavior of a highly textured nanocrystalline phase which forms prior to the YIG phase. The YIG phase also displays an Avrami growth behavior. These phases formed with Avrami constants (n) of 2 and 1 respectively at temperatures commonly used in annealing (800-1000°C). By continuing to study this behavior, we can optimize the temperature profile for higher quality YIG films on non-garnet substrates.
Sung, S. Y.; Xiaoyuan, Q.; Stadler, B. J. Appl. Phys. Lett. 2005, 87.
Block, A. D.; Dulal, P.; Stadler, B. J.; Seaton, N. C. IEEE Photon. J. 2014, 6, 1-8.
Gage, T. E.; Dulal, P.; Solheid, P. A.; Flannigan, D. J.;Stadler, B. J. Materials Research Letters 2017, 1-7.
Balinskiy, M.; Ojha, S.; Chiang, H.; Ranjbar, M.; Ross, C. A.; Khitun, A. Journal of Applied Physics 2017, 122, 123904.

We demonstrate a new design of graphene liquid cell consisting of a thin lithographically patterned hexagonal boron nitride crystal encapsulated from both sides with graphene windows [Kelly et al Nano Letters, in press 2018]. The resulting ultrathin window liquid cells are robust to vacuum cycling and can be produced with precisely controlled volumes and thicknesses as required for a specific experiment. The high stability of such cells allows us to demonstrate an order of magnitude improvement in the element mapping capabilities compared to previous cells, with 1 nm spatial resolution elemental mapping achievable using energy dispersive X-ray spectroscopy (EDXS). We apply this ability to observe the beam induced core shell structure of FePt nanoparticles. The presence of water was confirmed using electron energy loss spectroscopy (EELS) via the detection of the oxygen K-edge and measuring the thickness of full and empty cells.

We further demonstrate the atomic resolution imaging capabilities of these liquid cells by tracking the dynamic motion and interactions of small metal nanoparticles with diameters of 0.5−5 nm. A statistical analysis of ∼5000 measured displacements for individual particles found diffusivities of D = 3.25 × 10⁻³ nm² s⁻¹ for larger particles (with a mean size of 2.84 nm² and standard deviation 0.44 nm²) and D = 6.18 × 10⁻³ nm² s⁻¹ for smaller particles (with a mean size 1.26 nm², and standard deviation 0.55 nm²). These values are consistent with a previous observation of particles within graphene liquid cells,[Yuk et al Science 2012] but 10−100 times lower than those usually observed for SiN windowed liquid cells and over 10⁶ times smaller than expected values for bulk water.

This technology enables new opportunities for the application of graphene liquid cells to a host of in situ transmission electron microscope studies, providing a reliable platform for high resolution TEM imaging and spectral mapping.

In-situ transmission electron microscopy (TEM) of lithium ion batteries could shed light on the structural and chemical evolution of the electrode materials during electrochemical reactions. However, most currently used electrochemical cells in In-situ TEM study usually contain a low mass-loading of active materials, and sometimes, a point-contact between the electrolyte and the electrode, which result in a high impedance and high overpotential that are quite different from a real macro-scale cell. In this work, we developed a micro-cell with a configuration similar to a real macro-cell and with a relatively high mass-loading of active materials. This design enabled us to perform on-chip tests of galvanostatic discharge, cyclic voltammetry, and electrochemical impedance spectroscopy. Using this technique, we studied the lithiation behavior of Mn₃O₄ nanorods under a galvanostatic discharge mode and a constant potential mode, respectively. While the end states of lithiation under both modes are the same, it is only under the galvanostatic discharge mode that we can observe the step by step phase changes accompanying the three lithiation processes. This method has a great application potential to quantitatively study the structure changes under different charge/discharge stages of many multi-valent electrode materials.

In situ TEM promises to supplement conventional static (S)TEM imaging to provide direct observation of dynamic processes at the nanometer or atomic scale. This technique has been bolstered by the introduction of fast CMOS-based detectors, which can acquire “movies” much faster than older CCD-based cameras. However, observation of very high-speed processes remains challenging due to insufficient camera framerate, deleterious rolling shutter artifacts, and inadequate signal-to-noise ratio (SNR) at the low electron exposure in each high-speed movie frame. To address these challenges, we have introduced a large-format direct detection TEM camera, capable of running at up to several thousand frames-per-second with ultra-low noise and the choice of either a rolling or global shutter, for higher speed or reduction of image artifacts, respectively. This camera enables robust in situ TEM observation of very fast specimen dynamics.