Grain boundaries and interfaces of crystals have peculiar electronic structures, caused by the disorder in periodicity, providing the functional properties, which cannot be observed in a perfect crystal. In the vicinity of the grain boundaries and interfaces, dopants or impurities are often segregated, and they play a crucial role in the material properties. We call these dopants “function providing elements”, which have the characteristics to change the macroscopic properties of the materials drastically. In this study, the results obtained by Cs (spherical aberration) corrected HAADF STEM are demonstrated for well-defined grain boundaries and interfaces in oxide materials. On the other hand, such materials are composed of light elements, and light elements in the GB structural units play a crucial role in the material properties. Recently we have reported that annular bright field (ABF) STEM imaging is very powerful technique to produce images showing both light and heavy element columns simultaneously. In this study, crystal structures and GB atomic structures including light elements in several oxides such as ZnO, CeO2, Al2O3 and lithium battery materials are directly observed by ABF STEM.

In alpha-alumina, perfect dislocations typically dissociate into partial dislocations with the stacking fault on the {11-20} or {1-100} planes, whereas dissociation with the (0001) stacking faults is not observed. This would be because the fault energy of (0001) stacking faults is high compared with stacking faults on the other low-indices planes. However, there is no experimental evidence to support this consideration. To investigate the (0001) stacking fault, we fabricated an alumina bicrystal with a (0001) low-angle tilt grain boundary by diffusion bonding and observed the resultant grain boundary by transmission electron microscopy (TEM). It was found that the grain boundary consist of partial-dislocation pairs with the (0001) stacking fault. Furthermore, annular bright-field scanning TEM (ABF-STEM) was employed to characterize the atomic structure of the (0001) stacking fault. This technique visualizes not only the atomic columns of aluminum but also of oxygen. It was clearly found that the staking disorder across the (0001) stacking fault occurs on both the cation and anion sublattices. This is interesting because only the stacking faults on the cation sublattice have been reported in alumina so far. On the basis of our TEM observations, we constructed an atomic model of the (0001) stacking fault and then estimated its fault energy by first-principles calculations. The result showed that the fault energy of the (0001) is estimated to be one order of magnitude higher than that of {11-20} or {1-100} stacking fault. In the presentation, we will discuss the atomic structure of the (0001) staking fault and its fault energy in detail.
Acknowledgments: This work was supported by a Grant-in-Aid for Scientific Research on Innovative Areas "Nano Informatics" (grant number 25106003) from Japan Society for the Promotion of Science.

Grain boundaries (GBs) are known to have far-reaching effects on the electrical and mechanical properties of materials. Understanding the atomic scale mechanisms behind these effects requires an accurate determination of the interplay between GB structure and composition. Analysis of a range of grain boundaries using aberration corrected scanning transmission electron microscopy (STEM) has revealed a number of interesting trends: GBs from materials with different compositions can exhibit quite similar GB atomic structures, and GBs of the same material type can exhibit distinct structural variations due to the presence of vacancies and impurities Here, we will discuss the analysis of the atomic structures of [001] tilt grain boundaries in dense packing cubic crystals including FCC metals, perovskites, fluorites etc. A general model for the structure of grain boundaries in such similarly structured materials systems has been developed that is based on the crystallographic symmetry of the parent structures. This general model for grain boundary structures can, in principle, provide a means to infer the structure-property relationships in broad classes of materials. Furthermore, structural variations away from these predicted grain boundary symmetries can be interpreted as showing the frustration of symmetry caused by the incorporation of point defects (vacancies and impurities). To understand these chemical induced variations, and further quantify exactly how atomic scale variations at the boundary plane extend to the practical mesoscale operating length of the system, statistical analysis has been applied to the aberration corrected STEM Z-contrast images acquired from a series of SrTiO3 GBs doped with impurities, as well as GBs post-annealed in reduced atmosphere to incorporate oxygen vacancies. This analysis of GB similarity and variation provides insights into the structure-composition relationship in GBs and also provides an ability to experimentally determine the energetics behind the formation of grain boundary structures predict GB formation in various materials.
“The research described in this paper is part of the Chemical Imaging Initiative at Pacific Northwest National Laboratory under Contract DE-AC05-76RL01830 operated for DOE by Battelle. This work is supported in part by the United States Department of Energy, Basic Energy Sciences Grant No. DE-FG02-03ER46057. A portion of the research was performed using EMSL, a national scientific user facility sponsored by the Department of Energy's Office of Biological and Environmental Research and located at Pacific Northwest National Laboratory.”

We present the state of the art in strain mapping at the nanoscale using aberration-corrected high-resolution electron microscopy (HRTEM) [1] and dark-field electron holography (DFEH) [2]. The new holographic technique allows strain to be measured to high precision, with nanometre spatial resolution and for micron fields of view. It is also applicable to focused-ion beam (FIB) prepared samples which makes it ideal for studying devices of interest to the microelectronics industry. As with any TEM technique the samples are necessarily thin which allows some of the strain to be relaxed: the well-known thin film effect. In addition, dynamical scattering effects the strain information [3]. These issues will be addressed using a combination of finite-element modeling (FEM) and dynamical-scattering simulations.
Strain mapping at the highest resolution and certain sample geometries is still best carried out by the analysis of high-resolution images, obtained either by aberration-corrected HRTEM or high-angle annular dark-field scanning TEM (HAADF-STEM). Examples are nanocrystals [4] and multiferroic materials where the local polarization and strain can be obtained by analyzing the displacements of individual atomic columns [5]. As with DFEH, the experimental results must be compared with FEM and image simualtions. In some cases, a combination of high-resolution and DFEH is necessary, for example when studying embedded quantum nanodots [6].
This overview will, in particular, show results from the recently installed I2TEM microscope (Hitachi), an instrument specifically designed for DFEH experiments and aberration-corrected HRTEM over wide fields of view.
1 M.J. Hyuml;tch, E. Snoeck and R. Kilaas, Ultramicroscopy 74 (1998) 131-146.
2 M.J. Hyuml;tch, F. Houdellier, F. Hüe, and E. Snoeck, Nature 453 (2008) 1086-1089.
3 A. Lubk, E. Javon, N. Cherkashin, S. Reboh, C. Gatel, and M.J. Hyuml;tch, Ultramicroscopy 136, 42-49 (2014).
4 C.L. Johnson, E. Snoeck, M. Ezcurdia, B. Rodríguez-González, I. Pastoriza-Santos, L.M. Liz-Marzán, and M.J. Hyuml;tch, Nature Materials 7 (2008) 120-124.
5 A. Lubk, M. D. Rossell, J. Seidel, Y. H. Chu, R. Ramesh, M. J. Hyuml;tch, and E. Snoeck, Nano Letters 13, 1410-1415 (2013).
6 N. Cherkashin, S. Reboh, M.J. Hyuml;tch, A. Claverie, V.V. Preobrazhenskii, M.A. Putyato, B.R. Semyagin, and V.V. Chaldyshev, Appl. Phys. Lett. 102, 173115 (2013).

Domain walls in ferroelectric crystals have the potential to be used in nanoscale devices due to their unique properties and controllability at nanoscale dimensions. Understanding these properties requires a knowledge of both the electronic structure and the local electrostatic potential. However, direct characterization of electrostatic potential structures and the associated electric fields in ferroelectric domain walls has been extremely challenging.
In recent years, we have developed a new area detector, which we refer to as the "Segmented Annular All Field (SAAF)" detector, that is capable of atomic-resolution STEM imaging. This area detector can obtain 16 simultaneous atomic-resolution STEM images which are sensitive to the spatial distribution of scattered electrons on the detector plane. The detector has in-plane rotation capability which enables arbitrary alignment of the detector geometry to the crystallographic orientation of the sample. By taking the difference between the images from diametrically-opposed detector segments, we can form what are known as differential phase contrast (DPC) images. DPC STEM images approximate the gradient of the object potential (= fields) taken in the direction of the diametrically-opposed detector segments.
Having applied DPC STEM to the characterization of domain walls in ferroelectric materials such as BaTiO3 and LiTaO3, we will show the local electric field maps of domains and domain walls obtained. In the presentation, we will discuss the electrostatic potential structures of the 90° and 180° domain walls deduced from the experimentally obtained electric field maps.

The deposition of a ferroelectric oxide directly on a high-mobility III-V channel would potentially result in a single-transistor memory element where the polarization state of the ferroelectric layer would determine the state of the transistor. Towards this goal we deposited 8 nm of BaTiO3 on 2 unit-cells of SrTiO3 on GaAs (001) using oxide molecular beam epitaxy (MBE), and investigate the oxide/semiconductor interface using atomic-resolution HAADF/ABF imaging, electron energy loss spectroscopy (EELS) and first principles density functional theory (DFT). High-quality BaTiO3 films were achieved using a thin SrTiO3 nucleation layer grown with a ½ ML Ti template layer that allows for commensurate BaTiO3 films to be grown. Atomic-resolution Z-contrast and annular bright field (ABF) images of BaTiO3[100]/SrTiO3[100]/GaAs[110] and BaTiO3[110]/SrTiO3[110]/GaAs[100] exhibit atomically sharp interfaces and show no sign of interfacial diffusion or extensive sensitivity to the electron beam. ABF images also show that the first SrO monolayer contacting the GaAs substrate is highly oxygen deficient, and the SrTiO3 buffer layer has an out of plane polarization due to the presence of oxygen vacancies, which can be directly observed by the displacement between Ti and O columns. The electronic structure at the interface was determined using atomic-resolution EELS. The Ti L2,3 and O K edge spectra from the SrTiO3/GaAs interfacial Ti columns indicate the presence of oxygen vacancies and a distortion of the TiO6 octahedra. DFT calculations show that the O vacancies form preferentially at the SrTiO3/GaAs interface, where they polarize the SrTiO3, and in turn the BaTiO3, thereby inhibiting ferroelectric switching. However, piezo-force microscopy has shown the BaTiO3 films to be switchable. In order to resolve this apparent contradiction, direct observation of in-situ polarization switching following the application of an electrical bias to the ferroelectric oxide will be discussed.

Atomic-resolution spectroscopic imaging in state-of-the-art electron microscopes is now capable of unraveling bonding details at buried interfaces and clusters, providing both physical and electronic structure information [1]. The thousand-fold increase in electron energy loss spectroscopy (EELS) mapping speeds over conventional microscopes allows us to collect data from millions of spectra, enabling 2D mapping of composition and bonding at interfaces at atomic resolution as well as recording of statistically significant samples of inhomogeneous populations of nanoparticles. Using these techniques electronic and structural reconstructions at interfaces, microscopic inhomogeneities and atomic-scale interdiffusion can now readily be characterized and correlated with the macroscopic properties of the structure. Here, we will discuss the current state of aberration-corrected electron microscopy with specific focus on complex oxides. In recent years, this class of materials has seen an explosion of interest with the potential to both stabilize novel magnetic and superconducting phases at the interface between non-magnetic and insulating oxides as well the ability to incorporate the rich array of phases available in oxides into next-generation applications including fuel cells, batteries and magnetic memories. Key to the realization of the promised applications is the ability to understand and control the interfaces where electronic and structural reconstructions can dominate.
[1] D. A. Muller, L. F. Kourkoutis, M. Murfitt, J. H. Song, H. Y. Hwang, J. Silcox, N. Dellby, O. L. Krivanek, Science 319, 1073 (2008).

While in most complex oxides systems defects are detrimental to their functionality and applicability, they may be beneficial for high temperature superconductors (HTS), where they are found to be necessary to immobilize quantized vortices in the presence of magnetic fields, thereby allowing high currents to be carried. Therefore, determining the atomic structure of defects as well as understanding of how they behave and interact is critical to unravel and control their effect on the physical properties of HTS. Here we have combined aberration-corrected scanning transmission electron microscopy (STEM), electron energy loss spectroscopy (EELS) and density functional theory (DFT) calculations to unveil complex point defects which have not been previously identified, and to determine how they affect the crystal host structure on a single unit-cell level. STEM-EELS images reveal O-decorated Cu vacancies and local lattice distortions, while DFT calculations confirm their structure and chemistry and establish under which conditions they can be stabilized.
*Research sponsored by MICINN (Consolider NANOSELECT, CSD2007-00041, MAT 2011-28874-C02-01), Generalitat de Catalunya (Pla de Recerca 2009-SGR-770 and XaRMAE) and EU (EUROTAPES, FP7/2007-2013); RyC (JG). Research at ORNL U.S. supported by the U.S. DOE-BES, Materials Sciences and Engineering Division and through a user project supported by ORNL&’s Center for Nanophase Materials Sciences (sponsored by DOE-BES). Research at UCM supported by the ERC Starting Investigator Award and Fundacioacute;n Caja Madrid.

Thin film based solid state Li+-ion batteries hold technological promises for applications such as microelectromechanical systems (MEMS), smart cards, microsensors, and biochips. Thin film electrodes, especially those grown as oriented single crystals, can be utilized as model systems to study in details electrochemical processes on an atomic scale. In this work we investigated structures of different cathode films deposited by pulse laser deposition (PLD) on different orientation Nb:SrTiO3 substrates from LiMO2 (M=Co, Ni, Mn) targets. Structural similarities of Li-M-O phases and textured growth of the films make analytical TEM/STEM the techniques of choice, thus the films were characterized by electron diffraction, electron energy loss spectroscopy, and high-resolution imaging backed by simulation. The studied films include single composition films and films with varying composition of transition metals. HRTEM and HAADF imaging demonstrated the presence of structural variations within the films related to the different distribution of Li and M atoms in the framework of a oxygen sublattice. The films grow as epitaxially oriented facetted islands, coalescence of which results in the formation of vertically aligned domain interfaces and rough surfaces. Electrical and electrochemical properties of the films and its relation to the structure will be also presented. Effect of a film/substrate interface on electrical transport and performance of the battery will be discussed.

Electron vortex beams, carrying orbital angular momentum, have recently been produced with electron microscopy by interfering an incident electron beam with a grid containing dislocations [1]. Electron vortex beams have also been produced by carefully tuning the aberration coefficients using aberration correctors in electron microscopes [2].
Here, we will present the conditions required to produce an atomic size electron vortex beam in a scanning trasmission electron microscope using numerical and imaging simulations, as well as different levels of analytical approximation solutions that can then be practically implemented in different experimental setups. The electron optical conditions required to utilize electron vortex beams for atomic scale magnetic dichrosim, optical dichrosim measurements and for probing the electron momentum-transfer polarization, i.e. valleytronics, will also be discussed.
[1] J. Verbeeck, H. Tian, and P. Schattschneider, Nature 467, 301 (2010).
[2] L. Clark, A. Béché, G. Guzzinati, A. Lubk, M. Mazilu, R. Van Boxem, and J. Verbeeck, Physical Review Letters 111, 064801 (2013).
Acknowledgments.
This research was supported by the Center for Nanophase Materials Sciences (CNMS), which is sponsored at Oak Ridge National Laboratory by the Scientific User Facilities Division, Office of Basic Energy Sciences, U.S. Department of Energy (JCI), and by the Office of Basic Energy Sciences, Materials Sciences and Engineering Division, U.S. Department of Energy (SJP).

The introduction of silicon drift detectors has enabled the acquisition of high-resolution spectrum imaging maps using energy-dispersive X-ray spectroscopy (EDS) in an aberration-corrected scanning transmission electron microscope. More specifically, the windowless Oxford X-Max100TLE EDS detector offers a unique sensor design with a 100 mm2 active area closer to the sample for an ultra-high solid angle. In a conventional EDS detectors, low energy X-ray lines are partly absorbed by the polymer window and blocked by the ribs of the support grid. Removing the window and grid therefore has a significant effect for low energy X-rays, and can add to the sensitivity now possible with large solid angle designs in TEM. Light elements, such as beryllium, nitrogen and oxygen are enhanced with detection limits significantly increased due to much better counting statistics. Extremely low energy lines (less than 100eV) have now been observed. Most importantly, the new detector can be used in the ultra-high resolution pole piece without any effect on the spatial resolution.
Here, we will show that the count rates measured with the new Oxford X-Max100TLS detector are approximately 5-7 times higher than with the previous X-Max80 detectors. Consequently, we are now able to acquire atomic-resolution maps of materials including CdTe, AlGaN and SrTiO3, using our aberration-corrected, cold-field emission JEOL ARM200CF at UIC. We will show that X-ray mapping can be achieved for Al L (79eV), Li K (54eV) and Mg L (49eV) peaks in metallic materials. Finally, the simultaneous acquisition of electron-energy-loss and EDS spectra will be discussed.

The correlation between a material&’s luminescence properties with morphology, microstructure and local chemistry offers great benefit in the understanding of many technologically important materials and devices. This has encouraged a growing interest in performing cathodoluminescence microscopy in a STEM microscope [1,2]. Here, we present the development of a novel chatodoluminescence detection system suitable for a wide range of (S)TEM instruments that displays significantly enhanced collection efficiencies compared to existing systems [3] and show exemplar results from semiconductor heterostructures, nanorods, colloidal noble metal nanoparticles and biological samples where many compounds show luminescence properties.
The collection of the cathodoluminescence signal is achieved through miniature diamond-turned mirrors integrated into the tip of a conventional cryogenic side-entry TEM holder. Mirrors above and below the specimen provide a solid angle of collection of up to 7.3 steredian. Light is coupled out of the holder through two low-loss optical fibres to a Czerny-Turner optical spectrometer fitted with PMT and CCD detectors, providing a spectral resolution of about 0.4meV which is almost an order of magnitude better than the EELS system with the highest energy resolution currently present in the market. The system can be used in conjunction with most standard imaging and analytical techniques including EELS and EDS and this allows a full investigation of the material from the optical to chemical and compositional properties.
Different materials from biological to semiconductors are being investigated using the simultaneous acquisition of EELS and cathodoluminescence signals. Variations in the chemistry and composition of the material that can be observed using EELS influence the optical response and as result the luminescence properties. The paper will discuss the detailed analysis of such materials and the extra information that can be obtained when EELS and cathodoluminescence data are acquired simultaneously.
[1] Kim S. K., Brewster M., Qian F., Li Y., Lieber C. M. and Gradecak S., Nanoletters 9, 3940, 2009
[2] Zagonel L. F., Mazzucco S., Tence&’ M., March K., Bernard R., Laslier B., Jacopin G., Tchernycheva M., Rigutti L., Julien F. H., Songmuang R. and Kociak M., Nanoletters 11, 2011, 568
[3] Strunk H. P., Albrecht M. and Scheel H., J. of Microscopy 224, 2006, 79

The design of more efficient structures for diverse applications -- such as light gathering for energy conversion, catalysis, signaling using guided optical modes, and biological fluorescent tagging -- requires information about electronic and optical excitations in nano-scale and atomic-sized structures. Today we have available aberration corrected instruments that are routinely capable of atomic resolution imaging of isolated, non-periodic structures, and defect structures imbedded within periodic parent lattices, surfaces or 2-d single layer materials. This capability, coupled with 200 meV resolution EELS spectroscopy is being used today to explore optical and plasmonic behavior in many types of structures. We describe here work aimed at extending current spectroscopic capabilities well into the infrared and thermal energy scales by equipping the Nion Co, Ultra-STEM, [1] with a brightness preserving monochromator , [2,3] which we believe will be ultimately capable of delivering 10 meV energy resolution in a quiet room environment. At Rutgers, we have built such a quiet area within an existing older building using many passive strategies to control costs. Already low ambient magnetic fields have been reduced to about 2 micro-Gauss RMS using a shielded room of welded steel construction. This design uses eddy current AC shielding to give about a factor of 30 dB reduction at 30 Hz. A separate outer room reduces ambient noise to an NC-25 standard. Taking advantage of the thermal mass of the shielded room, and the insulating properties of the acoustic room, we obtain temperature fluctuations in the microscope enclosure room of about 0.1-0.2 deg C over several hours without chilled wall panels. Finally, the particular building was chosen for already good floor vibration of better than 1 micron/sec velocity. The building has a concrete block construction, placed on a concrete slab on the ground, well away from highways or other periodic interference, to avoid the need for a separate vibration isolation slab. We have also installed a 120V balanced power system, with the local ground tied to the center tap of the supply voltage, to aid in the control of ground currents. As in other installations, the operator&’s area is in an adjacent room, to maintain the quiet environment in the microscope room, but in this case we are also providing teaching and collaborative space there. We think these very complex, but also very reproducible and stable instruments may be, not only better research tools, but also better tools to teach the science of imaging and analytical characterization of the behavior of nano-scale objects.
1. O.L. Krivanek et al., Ultramicroscopy, 108 (2008) 179.
2. O.L. Krivanek et al., Phil. Trans. Roy. Soc., A367 (2009) 3683.
3. O.L. Krivanek and N. Dellby, US patent application (2010).

With the combination of monochromated electron energy loss spectroscopy (EELS) and aberration correction, it is now possible to map the optical properties of materials down to the nanometer level. Recently the energy resolution available from monochromators has significantly improved and we can now reach 20-30 meV routinely on ASU&’s NION UltraSTEM 100 allowing optical properties to be determined into the near infra-red with EELS. This capability can be applied to a wide range of different materials problems related to energy and the environment. Semiconducting nanoparticles can be employed in photocatalytic water splitting for hydrogen generation. Titania is the most stable water splitting catalyst but its efficiency is limited because of the large bandgap resulting in photon absorption only in the ultraviolet. It is possible to increase the absorption in the visible by doping the titania with nitrogen or treating with hydrogen. The exact origin of the enhanced absorption in not well understood. Here we perform bandgap mapping in a variety of titanias and will correlate the bandgap with local nitrogen content to obtain a more detailed understanding of the nature of the enhanced absorption. Monochromated EELS can also permit the dielectric function of submicron particles to be determined. This is particularly important when applied to carbonaceous particles produced by energy conversion processes because of their relevance to climate change. When appropriate spectral acquisition procedures are employed, it is possible to use Kramers Kronig analysis to determine the dielectric function over the full range of solar radiation from 200-2500 nm in wavelength. This information can then be employed to determine the radiative forcing properties of a variety of different forms of carbons.

Graphene edges and their functionalization by hydrogen influence the electronic and magnetic properties of graphene nanoribbons. The functionalization and structural engineering of edges are reported to greatly influence the properties of mesoscopic graphene structures. A variety of elements are known to attach to the edge of graphene, such as N, O, H, Fe and Si. However, to date hydrogen has been the most elusive to study due to its small atomic mass. Theoretical calculations predict saturating graphene edges with hydrogen lowers its energy and forms a more stable structure. Despite the importance, experimental investigations of whether graphene edges are always hydrogen terminated, are limited.
In our recent research activities, we utilized sub-Angstrom imaging of graphene edges by monochromated aberration corrected transmission electron microscope (AC-TEM) at an accelerating voltage of 80kV, combined with Density functional theory (DFT) calculations to examine armchair edge structures that are predicted to have shortened C-C bond lengths when they are not functionalized by hydrogen or other atoms. The JEOL 2200MCO, result of a bespoke developmental TEM system designed in partnership between Oxford and JEOL, equipped with imaging optimized monochromation system, offers 80pm spatial resolution at 80Kv in HRTEM mode. It can therefore resolve the location of atoms in the lattice structure of graphene precisely, which in turn can be used to determine the distance between atoms and the C-C bond lengths in graphene. Graphene edges produced by sputtering in vacuum and direct measurements of the C-C bond lengths at the edge show ~14% contraction relative to the bulk. DFT reveals the contraction is attributed to the formation of double/triple bonds at the absence of hydrogen functionalization. We use this property to show that hydrogen is not attached to the outermost bond of the armchair edge, indicating the formation of a triple bond, rather than that saturation of free edge states by hydrogen passivation. This was further confirmed by real-time tracking of bond lengths, where elongation occurred from functionalization by an additional atom to form a triangular edge state (first to observe), followed by the contraction of the same C-C bond when the additional carbon was removed. These observations give strong evidence that the edges of graphene are not always hydrogenated, which is important information for the design of future graphene electronic devices.

In oxygen conducting ceramics for intermediate temperature (300 - 600 °C) solid oxide fuel cells (IT-SOFCs), O^2- diffusion occurs via hopping into oxygen vacancies whose concentration can be modulated through doping with aliovalent cations (e.g. Ce⁴⁺ substitution by Gd³⁺ in CeO₂). Sluggish ionic conductivity in these electrolytes has been attributed to various defects which increase the activation energy for anion migration. The association of mobile oxygen vacancies with dopant cations, and the presence of highly resistive grain boundaries in polycrystalline electrolytes are well-accepted mechanisms which degrade total ionic conductivity [1]. The predominant explanation for high grain boundary resistivity in ceramics of high purity is the space charge double layer (SCDL) which results in vacancy-depleted regions emanating from grain boundary cores [2]. Recently it was shown that addition of 0 - 2 at% transition metal (TM) ions such as Cr, Fe, Ni and Cu to high purity gadolinium-doped ceria (GDC) enhanced grain boundary electrical conductivity by as much as 15 times by reducing the SCDL potential barrier [3]. We employ high spatial resolution monochromated electron energy loss spectroscopy (EELS) in an aberration-corrected scanning transmission electron microscope to probe the distribution of dopant cations at high purity Ce_0.9Gd_0.1O_1.95 grain boundaries modified with 0 - 2 at% TM ions. We also map the concentration of oxygen in the grain boundary region to better understand the influence of grain boundary structure and chemistry on oxygen vacancy distribution in materials relevant to IT-SOFCs.
1. Fergus, Jeffrey, et al., eds. Solid oxide fuel cells: materials properties and performance. CRC press, 2008.
2. Guo, X., Maier, J. Grain Boundary Blocking Effect in Zirconia: A Schottky Barrier Analysis. Journal of The Electrochemical Society, 148 (3) E121-E126 (2001)
3. Zajac, W., Suescun, L., Swierchzek, K., Molenda, J. Structural and electrical properties of grain boundaries in Ce0.85Gd0.15O1.925 solid electrolyte modified by addition of transition metal ions. Journal of Power Sources 194 (2009) 2-9.

In the aftermath of the developed aberration-corrected electron microscopy, new concepts emerge that address imaging at ultra-high resolution where most materials become radiation sensitive because of strong beam-sample interactions that increase quadratically with magnification. For example, it is unrealistic to expect that pristine surface structures or nano-composites can be imaged directly at a resolution below 0.1 nm while simultaneously maintaining sample integrity and single atom sensitivity. This contribution describes new concepts and technology that dramatically reduce such uncontrolled object alterations by the imaging electron beam, thereby allowing to characterize pristine structures as well as reversible and irreversible system excitations in a weak excitation approach on a single atom level. Applications shown that even a direct imaging of system conformations at atomic resolution with single atom sensitivity is now feasible and bears the potential to drastically expand views on how materials function. Specifically, we aim at recording elemental system excitations that occur at the onset of functional behavior at turn-over frequencies of chemical reactions. In recent work, such elemental excitations were detected in graphene and in a rhodium catalyst showing that collective system excitations may provide more insight into system dynamics than single atom trajectories [1]. Moreover, our ongoing investigations already demonstrated that the method is applicable in environmental electron microscopy to study gas-surface interactions or that the promotion of chemical reactions by various elements can be investigated in this manner. [2]
[1] Christian Kisielowski, Lin-Wang Wang, Petra Specht, Hector A. Calderon, Bastian Barton, Bin Jiang, Joo H. Kang, Robert Cieslinski, Real-Time, Sub-Ångstrom Imaging of Reversible and Irreversible Conformations in Rhodium Catalysts and Graphene, Phys Rev. B 88, 024305, 2013
[2] Electron microscopy was performed at NCEM, which is supported by the Office of Science, Office of Basic Energy Sciences of the US Department of Energy under Contract No. DE-AC02-05CH11231. The Dow Chemical Company and Helios SERC supported the development and application of the underlying low dose rate microscopy.

Novel two-dimensional (2D) heterostructures are fast becoming a major research area in condensed matter and material science. In this talk, we will present two examples of experimental and theoretical studies at the atomic scale of the stacking of two graphene layers, and graphene on hexagonal boron nitride (h-BN), using a combination of aberration-corrected scanning transmission electron microscopy (STEM) and first-principles calculations.
We have measured the optical response of twisted bilayer graphene (TBG) using monochromated electron energy-loss spectroscopy (EELS) in a low-voltage aberration-corrected STEM. The EEL spectra acquired from TBG present optical absorption peaks that depend on the twist angle. In particular, we find that band gap tuning in the optical range from the infrared to the ultraviolet can be achieved in TBG by controlling the twist angle between layers. The experimental features are well described by tight-binding calculations. Moreover, the theory developed provides insight on the electronic origin of the observed absorption peaks.
In the second example, we will show the emergence of a novel interesting electron-optical phenomenon present on 2D heterostructures. Specifically, the absorption spectra of a graphene layer on a h-BN layer under illumination with a dichroic signal was calculated, and the results indicate that the rotation angle between graphene and h-BN layers can be used as a tuning variable to achieve valley polarization, that is, to localize electrons to specific momentum valleys. We will discuss how the emergent field of valleytronics, in 2D heterostructures, can be accessed at the atomic scale using a monochromated aberration-corrected STEM and novel vortex electron probes carrying orbital angular momentum.

Acknowledgments
This research was supported by the National Secretariat of Higher Education, Science, Technology and Innovation of Ecuador (SENESCYT) (LB), a Wigner Fellowship through the Laboratory Directed Research and Development Program of Oak Ridge National Laboratory, managed by UT-Battelle, LLC, for the U.S. Department of Energy (WZ); National Science Foundation through grant No. DMR-0938330 (WZ), ERC starting Investigator Award, grant #239739 STEMOX, and Juan de la Cierva program JCI-2011-09428 (MICINN-Spain) (JS), by the Office of Basic Energy Sciences, Materials Sciences and Engineering Division, U.S. Department of Energy (SJP), and the Center for Nanophase Materials Sciences (CNMS), which is sponsored at Oak Ridge National Laboratory by the Scientific User Facilities Division, Office of Basic Energy Sciences, U.S. Department of Energy (JCI).

Aberration-corrected scanning transmission electron microscopy (STEM) at low voltage can now provide real space imaging and spectroscopy measurements at the atomic scale with single atom sensitivity. This opens new opportunities to study the structure of defects in 2D materials and probe their behavior beyond the ground states. Such studies, especially when combined with first-principles calculations, serve as an important step to correlate the defect structure with local properties, and help to create new functionalities in these 2D materials via controlled defect engineering.
At low electron beam energies and dose rates, some defects remain stable during imaging and/or spectroscopy, which allows reliable analysis of the defect structure and local bonding. Using low-voltage low-dose STEM imaging, we present a systematic study of intrinsic structural defects in high-quality CVD grown monolayer MoS2, and dislocation core structures composed of 5|7, 4|4, 4|6, 4|8 and 6|8 fold rings were imaged for the first time [1-3]. Quantitative image analysis allows us to map out the dopant distribution in MoSnot;2 one layer at a time, providing a feasible way to quantify the local doping level and measure the local band gap in doped MoS2 [4]. The fine structure in electron energy loss spectra acquired under optimized dose levels provides the sensitivity to determine the nature of the chemical bonding of single atoms, and we show that three-dimensional and planar bonding configurations for individual Si atoms in graphene can be directly discriminated [5].
Through direct momentum transfer, the electron probe can also be used to generate and excite defects in 2D materials, and the dynamics can be directly monitored via sequential imaging. Examples will be shown including reversible dynamics of a Si6 clusters trapped in a graphene nanopore [6], the systematic migration of vacancies in transition metal dichalcogenides, and the controlled in-situ fabrication of metallic nanowires within the semiconducting dichalcogenide layers.
[1] W. Zhou et al., Nano Lett., 13, 2615 (2013).
[2] S. Najmaei et al., Nat. Mater., 12, 754 (2013).
[3] X. Zou et al., Nano Lett., 13, 253 (2013).
[4] Y. Gong et al., Nano Lett., in press (2013).
[5] W. Zhou et al., Phys. Rev. Lett., 109, 206803 (2012).
[6] J. Lee et al., Nat. Commun. 4, 1650 (2013).
This research was supported in part by a Wigner Fellowship through the Laboratory Directed Research and Development Program of ORNL, the DOE Basic Energy Sciences (BES), Materials Sciences and Engineering Division, ORNL&’s Center for Nanophase Materials Sciences User Program sponsored by DOE-BES, DOE grant DE-FG02- 09ER46554, the JST research acceleration program, the Welch Foundation grant C-1716, the NSF grant DMR-0928297, CNS-0821727, OCI-0959097, the Army Research Office MURI grant W911NF-11-1-0362, the Office of Naval Research MURI grant N000014-09-1-1066, and the Nanoelectronics Research Corporation contract S201006.

There has been intense research into developing and characterizing new electrochemically active materials consisting of nanosized catalysts and thin films on various substrates. The desirable characteristics of nanoparticle catalysts, such as low mass loading, high efficiency and chemical stability, directly depend on the atomic-scale nature of the interface between the catalyst particles and their substrates. However, the research community still lacks sufficient knowledge of the angstrom-scale interaction between catalytic nanoparticles and their substrates.
We synthesize and characterize one such model interface, that between high-quality graphene surfaces and Pt nanocrystals grown directly on the graphene. We observe an atomically sharp interface, a high degree of crystalline order, and propose an epitaxial-like growth. In particular, we analyze the chemical nature of the interface with probe-corrected STEM-EELS at 80 kV. We aim to understand whether there is a molecular linker or a functional group that remains at the interface and helps bond the Pt atoms to the carbon support. EELS data, in particular C K-edge, demonstrate that the topmost graphene layer, while retaining crystalline order, is chemically different from underlying graphene layers, and we study the possibility that a C-Pt bond forms at the interface.
Our studies demonstrate that it is possible to investigate, at the atomic scale, the nature of the interfaces between supports and nanoscale catalysts. Understanding the exact atomic arrangement of such interfaces will help elucidate critical details about catalytic activity at the triple-phase boundary between supports, catalysts and electrolytes.

Oxide materials offer promise of multiple functionalities that can be tailored via interface engineering or chemical substitution. Their magnetic, electrical, and structural properties are tied to the subtle distortions of the crystallographic lattice from the perfect prototype. Atomic-scale understanding of all aspects of materials behavior: strain, polarization, charge transfer is necessary to understand and predict novel properties. Aberration corrected scanning transmission electron microscopy (STEM) can provide direct structural and chemical information at the unit cell level. Using atomic position detection and column shape analysis, precise local crystallographic characterization becomes possible. For materials with several competing functionalities, the properties can be affected both by structural order parameters and chemical factors, such as cation segregation, compositional gradients, or oxygen vacancy formation. These effects are especially significant in the vicinity of structural defects and heterointerfaces.
Interpreting experimental data in the framework of Landau-Ginsburg-Devonshire and /or Density Functional theory enables us to build a complete picture of the material behavior. Tracking both physical and chemical changes, we can decouple different charge compensation mechanisms (band bending vs. vacancy accumulation) at BiFeO₃ interfaces and domain walls. Tracking different order parameters in the vicinity of the BiFeO₃surface, we can resolve discrepancies in the estimates of the ferroelectric “dead layer” thickness obtained by different techniques. We can also study the origins of an unusual antiferroelectric phase in Na_1/2N_1/2Mg_1/2W_1/2O₃ and trace it back to octahedral tilt behavior. Finally, we can examine how decoration of the surface of a heterogeneous catalyst with a dissimilar phase affects polarization of the underlying lattice and catalytic properties of the system.
This research is sponsored by the Materials Sciences and Engineering Division, Office of Basic Energy Sciences of the U.S. DOE and through a user project supported by ORNL&’s Center for Nanophase Materials Sciences (CNMS), which is sponsored by the Scientific User Facilities Division, Office of Basic Energy Sciences, U.S. DOE.

The role of aberration-corrected scanning transmission electron microscopy (STEM) in materials characterization is examined with respect to Li-based cathode materials for battery applications. STEM-based methods are quickly becoming the most promising characterization tools for these materials, owed largely to the wide-range of techniques available on advanced STEM instruments, including the direct imaging of both heavy and light elements, and both energy-dispersive X-ray (EDX) and electron energy loss (EEL) spectroscopies. The current talk will focus on structural and chemical characterization of a Li-based cathode material, both in a pristine and irradiated state. Focus will remain on the nucleation of structural transitions, while also characterizing relevant parameters such as the manganese valence and oxygen presence. Various imaging modes, including high/low angle annular dark field (H/LAADF) and annular bright field (ABF), in conjunction with EELS, will be used extensively for this analysis.

Materials properties such as electronic transport, magnetic and optoelectronic responses can be greatly enhanced by isolated single dopants. These physical properties are sensitive to the atomic site and location (surface or bulk) of the dopants. Therefore to fully understand and control such properties, it is of critical importance to determine the three-dimensional single-dopant defect structure at the atomic-scale. Recent advances in electron optics have enabled scanning transmission electron microscopy (STEM) to directly determine the atomic structure of materials with sub-Ångström spatial resolution in the lateral directions. However, in a current electron microscope, the width of the point spread function along the optical axis, the ‘depth of focus&’, is still larger than 3 nm, which is insufficient to achieve depth resolution at atomic-scale, (although in some cases, the precision might be better than this limit). In the present study, we show the direct determination of the depth location of an isolated single Ce dopant in photo-luminescent wurtzite-type AlN. By combining quantitative atomic-resolution Z-contrast STEM imaging with frozen phonon image simulations on an absolute intensity scale, we can determine the depth of the single dopant atom to an accuracy of one atomic-spacing.

Diffusion in solids is a fundamental phenomenon and has been a topic of research for over two centuries now. Yet, experimental information about this atomic-scale phenomenon is limited to macroscopic measurements, such as tracking the movement of tracer atoms or measuring the change in concentration over time. These conventional experiments give us the diffusivity of a given species in a system, which is then combined with theoretical calculations to come up with the predominant atomic-scale mechanism. Naturally, such an indirect measurement relies on the accuracy of the calculations and is further complicated by the presence of defects in solids, which are ubiquitous. In this work, we will present direct observation of the diffusion pathways of different dopant atoms, such as Ce, Eu and Mn in bulk wurtzite-type AlN using aberration-corrected scanning transmission electron microscopy (STEM). By combining the STEM images with frozen-phonon image simulations and density functional theory based calculations, we can track the position of individual dopant atoms accurately to give full picture of vacancy- and interstitial-mediated diffusion of atoms inside the bulk solid.
Acknowledgements: This research was supported by the Materials Sciences and Engineering Division, Office of Basic Energy Sciences (BES), U.S. Department of Energy (DOE) and DOE grant DE-FG02-09ER46554. R. I. acknowledges support from JSPS postdoctoral fellowship for research abroad. This research used resources of the National Energy Research Scientific Computing Center, which is supported by the Office of Science of the U.S. DOE under Contract No.DE-AC02-05CH11231.

Recently, we have developed a novel 3D characterization technique to determine the local atomic structure of nanomaterials using high resolution electron tomography. This technique, termed equally sloped tomography (EST), is able to resolve local features at atomic resolution via a Fourier-based tomographic reconstruction method. Previously, we have resolved lattice structure as well as individual atoms and defects in single atomic species nanoparticles [1,2]. Here we have applied this methodology to 8-10nm bimetallic FePt nanoparticles. These particles form an ordered L1₀ state, in which the Fe and Pt segregate into alternately stacked (100) planes; the large magnetic anisotropy observed in these particles arises from this ordered state. The FePt particles can also form a superlattice structure, creating a material suitable for use in high-density magnetic memory devices [3]. Using EST to determine the 3D atomic structure of these particles, we can study the L1₀ state in unprecedented detail, and have identified not only structural defects, but also ordering faults in the L1₀ phase which ultimately will affect the magnetic functionality of the this nanomaterial.
[1] M. C. Scott, et al. Nature, 483, 444-447 (2012).
[2] C.-C. Chen, et al. Nature, 496, 74-77 (2013).
[3] S. Sun et al. Science 287, 1989-1992 (2000);

The development of new persistent luminescent phosphors in solid state lighting research is a challenge for advanced flat-panel applications. In persistent phosphors, two kinds of active centers are involved: emitters and traps. Emitters are centers capable of emitting radiation after being excited. Traps usually do not emit radiation, but store excitation energy and release it gradually to the emitters owing to thermal or other physical stimulations. Whereas the emission wavelength of a persistent phosphor is mainly determined by the emitter, the persistence intensity and time are determined by the trapping states which are generally associated with lattice defects or dopants. The phosphor CaTiO₃:Pr³⁺ has shown to be an attractive compound for field emission display applications because its transition ¹D₂ - ³H₄ at ~ 615 nm is close to the “ideal red”. Moderate intrinsic conductivity and resistance to high density electron irradiation make it ideal for the use in flat panel displays. But, despite the number of studies dedicated to the enhancement of the afterglow efficiency in this perovskite, no clear explanation of the mechanisms involved in the afterglow process has been given yet. Recent report [1] established that the incorporation of CaO excess in the perovskite lattice increased the persistence luminescence. The well-known defect chemistry of CaTiO₃ suggests CaO excess in the lattice can be achieved by the formation of Ruddlesden-Popper planar faults. These defects are soluble in the CaTiO₃ up to 1480 °C. At higher temperature, the segregation of RP- phases, i.e. Ca₃Ti₂O₇ and Ca₄Ti₃O₁₀, may occur. Ca excess can also induce Ti and/or O vacancies. This last mechanism should have a stronger influence on the persistent luminescence since they can act as traps and increase the decay time. However, any correlation between local structure-chemical defects and luminescent properties at nanometric level has been investigated up to date.
CaTiO₃:Pr with 0.2% Pr and different excess of CaO were synthetized, post-thermal treated at 1400 °C and 1500°C, and the excitation-emission spectra as well as the persistent luminescence measured. High resolution scanning and transmission electron microscopy by aberration corrected microscopes have been performed to characterize different type of defects. Atomically-resolved electron energy loss spectroscopy mapping were used to get inside the chemistry of defects. By means of unique access to cathodoluminescence spectroscopy in-situ STEM the optical response were analyzed, probed at nanometric scale for further correlation between the nanostructure and optical properties. [1] Eugenio H. Otal et al. Optical Materials Express, Vol. 2, Issue 4, pp. 405-412 (2012).

The formation of embryos of a new phase occurs by segregation and rearrangement of homogeneously distributed solute atoms. Here aberration-corrected scanning transmission electron microscopy has been used to capture the random motion of a single solute atom into a non-matrix lattice site that triggers one such phase transition. The images reveal the identity of the atoms involved in the transition and the sites they occupy. Further study and analysis with density functional theory unravels how something that initially appears so random and unimportant develops to a new phase that transforms the material to one with much more desirable properties. The results demonstrate that direct atomic level analysis of a complex structural evolution starting with the segregation of a single atom inside bulk materials has now become possible.
Research supported by the U.S. Department of Energy, Basic Energy Sciences, Materials Sciences and Engineering Division (Z.Q.Y, J.L., S.T.P., S.J.P., M.F.C.), by ORNL&’s Center for Nanophase Materials Sciences User Program, which is also sponsored by DOE-BES, by DOE grant DE-FG02- 09ER46554 (J.L, S.T.P.) and by the McMinn Endowment (S.T.P.) at Vanderbilt University. Research done by Q.M.H., H.Q.Y. and Z.Q.Y. at IMR was supported by the NSFC (51171189, 51271181, 51371178), and Liaoning Province.

Atomic-resolution structural and spectroscopic characterization techniques (scanning transmission electron microscopy and electron energy loss spectroscopy) are combined with nanoscale electrical measurements (conductive atomic force microscopy) to study at the atomic scale the properties of graphene grown epitaxially through the controlled graphitisation of a hexagonal SiC(0001) substrate by high temperature annealing. A scanning transmission electron microscopy analysis, carried out at 60KeV of beam energy, below the knock-on threshold for carbon to ensure no damage is imparted to the film by the electron beam, demonstrates that the buffer layer present on the planar SiC(0001) face delaminates from it on the (11-2n) facets of SiC surface steps, In addition, electron energy loss spectroscopy reveals that the delaminated layer has a similar electronic configuration to purely sp2-hybridized graphene. These observations are used to explain the local increase of the graphene sheet resistance measured around the surface steps by conductive atomic force microscopy, which we suggest is due to significantly lower substrate-induced doping and a resonant scattering mechanism at the step regions.

Aberration correction made possible two dimensional atomic resolution spectrum imaging in the scanning transmission electron microscope (STEM). It also led to a significantly reduced depth of field which was utilised to perform optical sectioning with atomic number contrast annular dark field (ADF) imaging and determine the positions of individual dopant atoms in three dimensions. Here we combine these corollaries of aberration correction to demonstrate three dimensional elemental mapping with atomic resolution electron energy loss spectroscopy (EELS). Atomic lateral resolution is critical as the optical transfer function of the STEM has a large missing cone, leading to excessive depth elongation for laterally extended objects. For instance, without atomic resolution a nanoparticle with a diameter of several nanometers shows a depth resolution of hundreds of nanometers. By collecting optical slices composed of atomic resolution spectrum images we improve the EELS depth resolution to the nanometer scale. We apply this depth resolution to a heterostructure composed of yttria-stabilised zirconia (YSZ) and strontium titanate (STO), a system which has sparked intense debate due its observed colossal ionic conductivity [1]. The 3D elemental mapping reveals islands YSZ buried within the STO in regions which appear perovskite like from 2D imaging, emphasising the importance of considering the possibility of three dimensional inhomogeneity. The results also highlight the unambiguous nature of EELS elemental mapping, revealing 3D compositional changes which cannot be determined through ADF optical sectioning with complete certainty.
Research sponsored in part by the UK Engineering and Physical Sciences Research Council through the UK National Facility for Aberration-Corrected STEM (SuperSTEM). Research at Oak Ridge National Laboratory was sponsored by the U.S. Department of Energy, Office of Science, Materials Sciences and Engineering Division (SJP).
[1] J. Garcia-Barriocanal et al., Science 321, 676 (2008).

Transition Al2O3 represents an important group of materials with very attractive surface and structural properties, which makes them a material of choice for a range of applications such as catalysts, catalytic supports, absorbents, hard protective coatings, abrasives, and membranes. Several important structural polymorphs derived from thermal decomposition of AlOOH Boehmite have complex structures, and to a large extent remain very poorly understood. Here we report a detailed atomic level STEM analysis of δminus;Al2O3, and show that δminus;Al2O3 is a phase that should be understood as a structural intergrowth of several crystallographic variants. The two variants, which represent the main component of δminus;Al2O3, are identified as δ1minus;Al2O3 and δ2minus;Al2O3. As a part of this study, we present a crystallographic approach that enabled us to unambiguously derive Al coordination from a series of low-index atomic level STEM HAADF images. In combination with XRD refinement and DFT calculations, we then report full crystallographic information for both δ1 and δ2minus;Al2O3 variants. In addition, we also discuss the energy of formation for δ1 and δ2minus;Al2O3 and other relevant transition Al2O3 phases, and show how energetic degeneracy leads to structural disorder and complex intergrowths among transition Al2O3 phases. The current findings have relevance for understanding thermodynamic stability and surface properties, and they provide important new information in the overall effort to rationalize the transformation processes in transition Al2O3. This research is part of the Chemical Imaging Initiative at Pacific Northwest National Laboratory. The work was conducted in the William R. Wiley Environmental Molecular Sciences Laboratory (EMSL), a national scientific user facility sponsored by DOE&’s Office of Biological and Environmental Research and located at PNNL.

Grain boundary physics are studied by the combination of aberration-corrected scanning transmission electron microscopy (STEM) and first-principles density-functional theory (DFT). The atomic structure and chemistry of grain boundaries were determined using Cs-corrected STEM. Based on the determined structure and chemistry, the electronic properties of the extended defects are studied by DFT calculations. We have studied grain boundaries in three representative thin-film solar cell materials including CdTe, CuInSe2, and Cu2ZnSnS4. We find that grain boundaries in these materials can be passivated by impurities. In CuInSe2, and Cu2ZnSnS4, grain boundaries can be passivated by O segregation. In CdTe, grain boundaries can be passivated by Cl segregation. Sufficient Cl segregation can lead to type conversion at grain boundary regions leading to local p-n junctions. These local p-n junctions help charge separation and improve current collection. The mechanism of grain boundary passivation will be discussed.

For photovoltaic materials, defects like intra-grain dislocations and grain boundaries (GBs) are usually considered to be major reasons for low efficiency, as they are likely to be carrier recombination centers. To understand the electrical activity of individual defects in CdTe solar cells, direct correlation from atomic structure to electronic property has been achieved by a combination of scanning transmission electron microscopy (STEM), electron energy loss spectroscopy (EELS), electron-beam induced current (EBIC) imaging, and density-functional theory (DFT) calculations.
Surprisingly, we find that intra-grain partial dislocation pairs do not create gap states but cause energy band bending, which instead assist the separation of electron-hole carriers, therefore reducing recombination [1, 2]. Moreover, after a CdCl2 heat treatment which significantly improves cell efficiency, we observe that a high concentration of Cl substitutes for Te in the GBs. Cl doping removes the gap states and further dopes electrons into the conduction band, resulting in an n-type region. The depletion regions between the n-type GB core and the surrounding p-type grains help separate electron-hole pairs, further improving cell efficiency. EBIC observations directly image the high collection efficiency of the GBs.
References:
[1] C. Li, J. Poplawsky, Y. Wu, A. R. Lupini, A. Mouti, D. N. Leonard, N. Paudel, K. Jones, W. Yin, M. Al-Jassim, Y. Yan and S. J. Pennycook, Ultramicroscopy 134, 113 (2013)
[2] C. Li, Y. Wu, T. J. Pennycook, A. R. Lupini, D. N. Leonard, W. Yin, N. Paudel, M. Al-Jassim, Y. Yan and S. J. Pennycook, Phys. Rev. Lett. 111, 096403 (2013).
Acknowledgment:
This research was supported by the US DOE Office of Energy Efficiency and Renewable Energy, Foundational Program to Advance Cell Efficiency (F-PACE), (CL, YW, JP, NP, WY, MAJ, YY, SJP), the Office of BES, Materials Science and Engineering Division (ARL), and through a user project supported by ORNL&’s Center for Nanophase Materials Sciences, which is also sponsored by DOE-BES. This research used resources of the National Energy Research Scientific Computing Center, which is supported by the DOE Office of Science under Contract No.DE-AC02-05CH11231.

Thin film solar cells based on Cu(In,Ga)Se₂ (CIGS) absorbers have demonstrated high conversion efficiencies up to 20.4 % on flexible substrate and therefore manifest a promising potential for the development of low-cost, high-efficiency solar cells. The efficiency of CIGS solar cells is strongly affected by copper and gallium concentration variations, which directly influence the band gap energy in nanoscale. Therefore, the measurement of nanoscale band gap variations is of high interest to improve the understanding of local mechanisms related to compositional inhomogeneities and to push the efficiency further towards the theoretical limit. Valence electron energy loss spectroscopy (VEELS) allows for probing electro-optical properties on nanometer scale. However, VEEL spectra can be influenced by various artifacts which complicate the data interpretation. Therefore, the extraction of the band gap energy by VEELS requires careful consideration of the results.
In this work, we present a systematic study that evaluates the feasibility and reliability of local band gap measurement in CIGS by VEELS. We determine the variation of the band gap energy across the evaporated CIGS layer experimentally, using a monochromated scanning transmission electron microscope (STEM). Further, we discuss the precision and accuracy of the results under consideration of error estimations and the comparison of the results with simulations: The measured band gap variation corresponds well to the expectations based on compositional gradients measured by energy dispersive x-ray (EDX) spectroscopy. As found by the analysis of different error origins, the precision is mainly limited by the acquisition reproducibility, given that the signal-to-noise ratio is high enough. Further, we estimate the influence of the specimen thickness by simulations and propose a correction for thickness effects. According to our simulations of scattering diagrams, the impact of radiation losses is negligible for specimens thinner than 100 nm. Conclusively, based on the findings of our study, the precision of the measurement is high enough to detect relative band gap variations of CIGS thin film solar cells. Therefore, VEELS provides a promising tool to address band gap variations on nanometer length scale, e.g. to investigate the influence of chemical inhomogeneities and dopant accumulations at grain boundaries.

Poly-crystalline CdTe-based thin film photo-voltaic devices have shown a great potential in large scale energy conversion applications. Commercial success of CdTe-based devices derives from the high absorption coefficient and ideal direct-band gap of the material which very effectively couples to our Sun&’s light spectrum as well as industrially established efficient manufacturing process of the modules. Despite this success conversion efficiency of CdTe has achieved very minor improvements over the last 20 years with current state-of-the-art 19.6% , while many theoretical predictions point towards high 20&’s%.
To overcome this stagnation and further drive cost-per-watt of the modules, better atomic-scale understanding of effects of post-deposition CdCl2 treatment as well as native dislocation structures and grain boundaries is needed. In this study, we investigate poly-crystalline CdTe thin film solar cells with and without CdCl2 treatment, as well as ultra-high-vacuum bonded CdTe bi-crystals with pre-defined misorientation angles. Aberration-corrected cold-field emission scanning transmission electron microscopy (STEM) analysis is carried out in the JEOL JEM-ARM200CF, using high-angle annular dark field (HAADF) and annular bright field (ABF) imaging. The instruments allows 70pm spatial resolution and 0.35 eV energy resolution at primary electron energy of 200 kV. Chemical composition and electronic environment of dislocation cores and grain boundaries is examined via atomic column-resolved X-ray energy dispersive (XEDS) and electron energy-loss spectroscopies (EELS).
We show that samples without CdCl2 annealing exhibit fluctuations in Cd-Te stoichiometry, which is particularly pronounced near defect structures. We also investigate immediate chemical environment of the defects for diffusion of impurities and show atomic-column resolved XEDS and EELS identification of commonly occurring dislocation core configurations, which could potentially provide detrimental carrier recombination centers.

Hydrogen peroxide (H₂O₂) is an important commodity chemical used for bleaching, disinfection, and chemical manufacture. At present, manufacturer&’s use an indirect large scale process in which anthraquinones are sequentially hydrogenated and oxidized in such a manner that hydrogen and oxygen are never mixed. For over a century, the identification of a safe, clean smaller scale process for generating H₂O₂ directly from gaseous H₂ and O₂ has been sought. Until recently, catalyst design for this direct reaction was solely focused on using supported palladium nanoparticles which require the use of bromide and acid to arrest peroxide decomposition, since palladium is also a very active catalyst for H₂O₂ hydrogenation to water. In 2002, we discovered that gold nanoparticles alloyed with palladium showed significant synergy for the direct reaction even in the absence of bromide and acid additives.¹ Over the past decade we have embarked on re-designing these supported Au-Pd materials in an effort to optimize their catalytic performance and improve their economic viability. In this presentation, we will explain the critical role that aberration corrected analytical electron microscopy has played in this design and optimization process. In particular, HAADF and XEDS spectrum imaging data have been utilized to provide the catalyst chemists with crucial clues on how to create new preparation protocols that facilitate the reproducible and simple manufacture of supported Au-Pd nanoparticles with the best size, morphology, composition and stability characteristics for catalyzing this demanding reaction.²
[1] P. Landon, P.J. Collier, A.J. Papworth, C.J. Kiely and G.J. Hutchings, Chem.Commun. 2002, 18, 2058-2059.
[2] J.K. Edwards, S. Freakley, A.F. Carley, C.J. Kiely and G.J. Hutchings, Accts.Chem.Res., (2014), (DOI: 10.1021/ar400177c), in press.

M1 and M2 phases in Mo-V-M (M = Te, Ta and etc) oxides have attracted great attentions as promising catalysts for selective oxidation/ammoxidation of propane[1]. AC-STEM was proven to be a powerful characterization tool for these systems, revealing local cation distribution[2], structure and chemistry of the crystal defects[3], polar domain structure[4], surface[5] and other features that enable understanding of atomistic mechanisms and optimization of catalytic performance. However, research has so far focused on single phases of this material, leaving effects such as synergy (disproportionate improvement of performance in phase mixtures of M1 and M2)[1] unexplained. In this work, we use AC-STEM to study a series of M2 phase catalysts (e.g. Mo-V-Te-Ta oxide) to get insights into this question.
We find that, for the systems where synergetic effects have been reported, certain facets of the M2 phase particles are often decorated with pentagons (e.g. Mo6O21), which are building blocks for M1 phase. These pentagons also appear to contain heavy cations (e.g. Ta) in higher concentration than the bulk of the material. The pentagons preferentially decorate (110) and (100) facets, but not (010) and (310) facets, suggesting that the arrangement where the M1 phase pentagons share one edge with the M2 phase hexagons is energetically favorable. While the surface decoration is usually limited to one layer of pentagons, in situations such as high defect density (stacking faults) or considerable surface roughness (concave regions), it can become true intergrowth, with nm-scale of M1 phase attached to M2 particles, a hitherto unobserved behavior.
It is also found that intergrown areas of the M1 and M2 phases are under considerable strain, which may shed light on the nature of the M1/M2 phase synergy in a real catalyst that consists of intergrown M1 and M2 phases. The study of underlying mechanism is still ongoing; nevertheless our results provide key information for modeling the synergistic effect between M1 and M2 phases. They also suggest an intriguing possibility to improve the catalytic performance by tuning an M1/M2 intergrowth hetero-structure compared to the traditional co-synthesis or physical mixtures of M1 and M2 phase composite catalysts.
[1] N. R. Shiju, and V. V. Guliants, Appl. Catal. a-General 356 (2009).
[2] J. J. Yu et al., Catal Comm. 29 (2012).
[3] W. D. Pyrz et al., Chem. Mater. 22 (2010).
[4] Y. H. Zhu et al., Chem. Mater. 24 (2012).
[5] Y. Zhu et al., Angew. Chem. 51 (2012).
* Research supported by the Materials Science and Engineering Division, U.S. Department of Energy (DOE), through a user project supported by ORNL&’s Center for Nanophase Materials Sciences, sponsored by the Scientific User Facilities Division, Office of Basic Energy Sciences, U.S. DOE, and by Chemical Sciences, Geosciences and Biosciences Division, Office of Basic Energy Sciences, U.S. DOE, under Grant #DE-FG02-04ER15604.

LaMnO3 particles are found to be catalytic for the oxygen reduction reaction in alkaline fuel cells. The performance of such catalysts depends primarily on the character of their surface terminations. Evaluating the surfaces on the atomic scale is a challenging task; however, recent advancements of aberration corrected transmission electron microscopy (TEM) instruments have allowed spatial identification of all atomic species in a material with a precision of a few picometers by comparison to multislice simulations of the projected atomic structure. Here, negative spherical aberration imaging is applied, where a negative value of the spherical aberration coefficient is employed in TEM mode, in order to determine the atomic terminations of orthorhombic, single crystalline LaMnO3 particles, tilted in different orientations. The results are compared to Wulff constructions that predict the equilibrium crystal morphology based on surface formation energies calculated by hybrid-exchange density functional theory. The discussion focuses on the ability to evaluate active surface sites of catalytic oxide structures using advanced TEM methods and comparison with simulations.

Fine-tuning nanocatalysts to enhance their catalytic activity and durability is crucial to commercialize fuel cells. In an ongoing attempt to reduce the mass loading of Pt/C in PEMFCs, there has been a tremendous research till date; constantly suggesting that a nanoscale alloying of Pt with 3d transition metals is a more viable option. Pt-Fe nanoalloys, in particular, have gathered much attention in recent years not just as a better catalyst to Pt/C but also because of its magnetic properties that are deployable in ultra-high density information storage. Recently [1], we introduced a new class of catalytic NPs to Pt-Fe system with an ordered intermetallic core encapsulated in a bilayer Pt-rich shell. We demonstrated that the observed exceptional increase in their activity and catalytic durability over existing Pt-Fe designs is attributable to a strained lattice and persistent atomic-order. In the light of these recent observations, we believe that the atomic-ordering and a compositional segregation of Pt are the two integral factors that dictate the overall performance of a nanocatalyst design. Our current work [2] presents an atomic-insight into the phase transformation of Pt-Fe nanoalloys obtained via an in-situ annealing in an aberration corrected STEM.
The in-situ heat treatment ranged RT to 200-400-600-700 until 800 degC and then quenched back to RT. At each temperature, the same nanoparticle (5 nm) was tracked, and a combination of atomic-resolution STEM-HAADF imaging and STEM-EEL spectroscopy was performed. The composition is known to exhibit a size-dependency at the nanoscale and hence, by tracking the phase transformation from the same particle over different temperatures, we have successfully overcome the error due to statistical interpretation from a large set of particles. While our STEM-HAADF results clearly demonstrate changes in particle shape, size, ordering and aggregation kinetics over the course of heat treatment, the EELS data reveal new insights into the segregation process. We confirm that Pt is not the only participant in the segregation process but is to be contending with Fe. The atomic-ordering starts to take place around 400 degC at the core and based on its initial composition and size, the particle evolves into an ordered nanoalloy with segregated Pt and Fe shells. Further, we illustrate through a model as to how such a thermally induced ordering process can lead to form various alloy configurations including stable intermetallics, intermetallic core-shells and Pt-rich NPs. We believe that a dedicated attempt to understand the nanoscale phase transformation as this is central to fine-tune the catalytic properties of alloyed-Pt NPs in general, and hence could redefine a new methodology to synthesize next generation fuel cell nanocatalysts.
1. Prabhudev, S.; Bugnet, M.; Bock, C.; Botton, G. ACS Nano 7 6103-6110 (2013)
2. Prabhudev, S.; Bugnet, M.; Zhu, G-Z.; Bock, C.; Botton, G. (manuscript under preparation)

Bond-length variations at the pm scale and the associated changes in electronic structure can have substantial influence on the activity of nanocatalyst surfaces. Measuring these displacements with electron microscopy requires high precision in locating atom positions. We have developed a new non-rigid (NR) registration technique for aberration-corrected Z-contrast HAADF STEM image series that enables sub-pm precision measurement of atom positions on robust samples, and pm-precision for nanocatalysts, 5-7 times better than previous approaches using STEM data. NR registering and averaging images of Au and Pt nanoparticles show pm-scale contractions and expansions of surface atoms. In particular, a Pt nanocatalyst on silica support exhibits contraction of atoms at a (1-11)/(-1-11) corner towards the particle center and expansion of a (1-11) facet, with very little lateral displacement. Standardless atom counting on the same NR registered STEM images shows that the Au and Pt nanoparticles are between 1 and 15 atoms thick with <1 atom uncertainty in a majority of the atom columns, providing new insight in the 3D structure. High precision in both positions and thickness are enabled by the extremely high signal-to-noise ratio after NR registration and averaging. High precision STEM imaging will open up new applications of STEM in understanding catalytic active sites, the displacement fields of defects and interfaces, and various forms of ferroic behavior of materials.

Bismuth vanadate has recently emerged as a promising candidate for photoelectrochemical (PEC) water splitting because of its ability to efficiently split water using sunlight, and the abundance and non-toxicity of its constituent elements.
BiVO₄ exists in three main crystalline forms: tetragonal zircon, monoclinic scheelite and tetragonal scheelite structures. In particular, the monoclinic BiVO₄ (m-BVO) phase exhibits a much higher photocatalytical activity due to its favorable band gap (2.4-2.5 eV) in the visible region of the electromagnetic spectrum, a conduction band positioned close to the hydrogen evolution potential, and a valence band position suitable for driving water oxidation under illumination. However, charge transport has been found to be a key limiting factor for its PEC performance.
It is believed that oxygen vacancies, the main ionic defect in BVO at low temperatures, can considerably improve the PEC performance of m-BVO by substantially increasing the donor density and enhancing charge transfer at the surface of the BVO grains. However, little is known regarding the concentration and the extent of the oxygen vacancy density in m-BVO at the nanoscale.
Considering the importance of BVO for PEC applications, we have examined the structural properties and the electronic and chemical behavior of a commercial bismuth vanadate powder using a combination of scanning transmission electron microscopy (STEM) and electron energy-loss spectroscopy (EELS).
EELS is used to investigate the valence of the vanadium ions in m-BVO by probing the vanadium L_2,3 edge. Large changes of the O-K and the V-L_2,3 edge fine structures are observed in all scanned grains. A shape change and a shift of the V-L_2,3 edge of ~1.3 eV to lower energies is observed at their surfaces. These observations are indicative of the presence of V⁴⁺ and V³⁺ species, as reported in previous studies. Thus, the valency of the surface vanadium ions is found to change owing to oxygen deficiency (vacancies) at the surface, with the reduction shell extending over approximately 5-10 nm from the surface. The oxidation state of vanadium is identified by comparing the measured V-L_2,3 edge fine structure to known V³⁺ and V⁴⁺ references. In addition, by using high-resolution STEM imaging we show that these electronic reconstructions at the surface of the grains do not entail modifications in the structural features of m-BVO.