Alumina (Al₂O₃) is one of the most common materials used in various industrial sectors. Despite such importance, atomistic-level details of the dislocations, which are responsible for plastic deformation of the material, have been identified only recently. Shibata et al. have reported the High-Resolution TEM (HRTEM) measurements on atomic-level structure of dissociated dislocations and their associated stacking faults[1]. We have performed a large-scale classical Molecular-Dynamics (MD) and hybrid Density-Functional (DF)/classical MD analysis on the electronic structure and the stability of the basal edge dislocation in α-alumina[2]. We have obtained the qualitatively consistent results that support the experimental observation of stable core-core distance of dissociated partials in the dislocation. In addition, effects of segregation of doped atom, such as Si and Ti, into dislocation core have been also investigated via the hybrid DF/MD method. Localized mid-gap states emerge at the doped cores as signatures of the dopant&’s electronic structure. As compared with previous researches on the edge dislocations, screw dislocations in α-alumina have not been studied very extensively. This is mainly due to the difficulty in finding the screw dislocations generated intrinsically in experiment. Recently, Tochigi et al. have reported observations of basal screw dislocation networks formed in an alumina low-angle twist boundary[3]. On the contrary to the basal edge dislocations, no dissociated partials have been observed in the case of basal screw dislocations. To elucidate causes on this observation, we performed a large-scale molecular dynamics computer simulation on the system with the screw basal dislocations in α-alumina, involving four dislocation cores with 288,000 atoms. We introduce partial screw dislocations dissociated along [1-100] and [0001] directions from a perfect dislocation, accompanied by {0001} and {1-100} stacking faults, respectively. Comparisons are made on the energies of these partials with that of perfect dislocation. The result reveals that the partial dislocations dissociated along both directions are unstable at 0 K. This is consistent with the fact that only perfect dislocation has been observed experimentally. We also analyze stability of the perfect screw dislocation with {0001} stacking fault at finite temperatures. The stability trend below 1000K is similar to that at 0K, while there is little difference in energy between perfect and partial dislocations above 1000K. Detailed analyses on the finite-temperature behaviors and a charge-distribution analysis via a variable-charge interatomic model will also be discussed. [1] N. Shibata et al., Science 316 (2007) 82; E. Tochigi et al., Acta Matter. 58 (2010) 208. [2] K. Tsuruta et al., Mater. Trans. 50 (2009) 1015. [3] E. Tochigi et al., Acta Mater. 60 (2012) 1293.

Nanocrystalline metals promise unique and improved properties, such as high wear strength, corrosion resistance, and enhanced thermoelectric performance. These gains are unfortunately mitigated by the inherent instability associated with the high fraction of atoms located at grain boundaries. Solute segregation to the grain boundaries can lower this energy penalty, and this approach has been successfully applied in a handful of systems to achieve some measure of stabilization. To better understand the conditions required for complete stability of a nanocrystalline structure we use a thermodynamic model for a regular nanocrystalline solution to construct nanocrystalline stability maps for use in alloy design. A modified ratio of the enthalpies of mixing and segregation is found to be the descriptive criteria for nanocrystalline stability. We adapt the Miedema surface segregation model to the grain boundary environment to provide the segregation enthalpy necessary to predict nanocrystalline stability, and use available materials data to predict the relative stabilities of hundreds of binary alloys.

Dopant atoms in a crystal are known to segregate to dislocations and form so-called Cottrell atmospheres in order to accommodate strain fields around dislocations. In addition, dislocations act as rapid diffusion paths of dopant atoms, which is known as the pipe-diffusion phenomenon. It has been shown that these characteristic properties of dislocations can be used to form novel nano-scale structures inside oxide materials. For instance, conductive Ti-rich nanowires in insulating sapphire single crystals have been fabricated by dislocations [1]. However, fundamental mechanisms of dopant segregation at dislocations are not well understood yet, especially in oxide materials. One successful method for introducing dislocations into crystals is the fabrication of low-angle tilt grain boundaries by bicrystal method. With this method, several dislocation structures can be introduced in nanometer order by controlling crystallographic orientations of two adjacent crystals; therefore, these model dislocation structures are advantageous for studying the interaction between dislocations and dopant atoms systematically. In a {11-20}/<1-100> 2° low-angle tilt grain boundary of alumina (α-Al₂O₃), a periodic array of climb-dissociated basal edge dislocations is introduced [2]. The core structures of those dissociated basal edge dislocations have been observed by scanning transmission electron microscopy (STEM). It was found that the upper partial dislocations and the lower partial dislocations are terminated by -Al-Al columns and -Al-O columns respectively, and being locally nonstoichiometric [3]. In the present study, several kinds of metal ions (Ni²⁺, Sr²⁺, Er³⁺, Zr⁴⁺ and Ti⁴⁺) are doped into the low-angle tilt grain boundaries of alumina in order to understand the dopant segregation behaviors around the partial dislocation cores. All of the metal doped grain boundaries consist of periodic arrays of b=1/3<10-10> and b=1/3<01-10> partial dislocation pairs with stacking faults in between. Z-contrast STEM images revealed that the dopant atoms segregate in the vicinity of the dislocation cores but not to the stacking fault regions. The ratio of the segregation amount to the upper/lower partials is likely to depend on the valence states of dopant atoms, but not on the ionic radii. More explicitly, we found that the upper partials tend to attract divalent cations, while the lower partials tend to attract tetravalent cations. These results suggest that the partial dislocation cores may be locally electrically charged; and thus, electrical interactions between dislocation cores and dopant atoms must be the key to understanding the dislocation segregation mechanisms in alumina. Referrences [1] A. Nakamura et al., Nat. Mater., 2, 453-456 (2003). [2] A. Nakamura et al., Phil. Mag., 86, 4657-4666 (2006). [3] N. Shibata et al., Science, 316, 82-85 (2007).

Grain boundary sliding is an important deformation mode in polycrystals, and it has been extensively investigated, for example, there are many studies on influences of the atomic geometry in the grain boundary region. However, it is important to investigate grain boundary sliding from the electronic structure of grain boundary for deeper understandings of the sliding mechanisms. In the present work, we investigated the grain boundaries sliding in pure and segregated bicrystals with classical molecular dynamics (MD) simulations and first-principles calculations. It is accepted that the sliding rate is affected by the grain boundary energy. However, there was no correlation between the sliding rate and the grain boundary energy in either the pure or the segregated bicrystals. First-principles calculations revealed that the sliding rate calculated by the MD simulations increases with decreasing minimum charge density at the bond critical point in the grain boundary. This held in both the pure and segregated bicrystals. It seems that the sliding rate depends on atomic movement at the minimum charge density sites.

UO2 pellets, used to power light water reactors, contain micron-sized grains separated by a network of grain boundaries. As the pellets go through the fission process, the resulting fission products must be accommodated within the UO2 lattice and grain boundaries provide important space to accommodate them. Therefore quantifying the way in which fission products interact with grain boundaries is essential for understanding, predicting, and ultimately controlling fuel performance. Here, atomic-level simulations using empirical potentials are used to examine the segregation of various fission products to one specific grain boundary - Σ5 tilt - in UO2, which is chosen as a model representative boundary in this material. The calculated segregation energies of fission products are typically in the range of several eVs. We find that two factors - the ionic charge and ionic radius of a given fission product - are dominant determiners of the segregation behavior and at times compete to determine the final segregation tendencies. In particular, larger aliovalent ions segregate to a specific uranium site that is characterized by smaller oxygen coordination and greater free volume, compared to other sites at the grain boundary. This work is supported by the Nuclear Energy Advanced Modeling and Simulation (NEAMS) program.

The deformation behaviors at the grain boundary and the grain interior were investigated in magnesium and its alloys using nanoindentation technique. The grain interior had a low strain rate sensitivity exponent, because its matrix was too large to be influenced by the grain boundary. The deformation mechanism in the grain interior was the dislocation slip. On the other hands, the grain boundary showed a high strain rate sensitivity exponent and was dominated by the grain boundary sliding due to the high diffusion rate at the grain boundary. The grain boundary affected the deformation behavior into the grain interior at the distance of 2 mu;m. In addition, the occurrence of grain boundary sliding was closely related to the grain boundary energy; grain boundary with high energy showed high strain rate sensitivity. Furthermore, the addition of solute atoms into magnesium tended to prevent the grain boundary sliding due to the decrease in grain boundary energy.

Tungsten and tantalum are neighbors in the Periodic Table of the Elements and they have many properties in common such as high melting point, high cohesive energy, high mass density etc. Their mechanical properties such as ductility or ductile-to-brittle transition temperature (DBTT) can be affected by the presence of impurities. However, the extents of the impurity effect on the two metals are rather different. Due to the presence of impurities, the increase of DBTT is more than 400K for tungsten, but is only about 20K for tantalum. Despite the ample experimental evidence, the nature behind this preferentially embrittling effect of impurities on the ductility of tungsten still remains elusive. In this work, based on the understanding that ductility is the competition between grain boundary (GB) separation and dislocation activities, we used density functional theory to systematically calculate the pristine and contaminated GB separation energy, the GB and dislocation core segregation energy of various impurities for tungsten and tantalum. The results show that for each impurity species, the GB and core segregation energies in tungsten are always significantly higher than those in tantalum, indicating that impurities in tungsten are more likely to segregate to GB regions and the vicinity of dislocation core to influence them. The binding energy difference between GB and free surface site for each impurity species in tungsten is always higher than that in tantalum, indicating that the presence of impurities, if deemed undesirable, will cause a greater reduction in GB separation energy for tungsten. The analyses of the chemical and mechanical effects of impurities suggest that tungsten is more sensitive to impurities because of its low lattice constant and high bulk modulus despite other possible causes.

Atomistic simulation of the uniaxial straining of nanocrystalline palladium along different loading directions at room temperature demonstrated that both the elastic and plastic behavior displayed a pronounced anisotropy after some deformation. Surprisingly, even the Young&’s modulus can significantly change with the deformation history of the sample. Some of these changes are found to be reversible. They either relax or can be removed by applying the opposite deformation. A mechanism based on excess-free-volume migration in the grain boundaries is proposed to explain such behavior. To explore this further we have introduced significant amounts of porosity into nanocrystalline palladium structures and found that a significant part of the void volume fraction reduces during the sample preparation. The Young&’s modulus and the flow stress decrease considerably with increasing porosity. Similar general trends, decreasing elastic stiffness and decreasing flow stress with increasing free volume content (or decreasing density), is found in amorphous materials as for example diamond-like carbon. Mechanical deformation can modify the free volume content and thereby the properties significantly. It is thus possible to create grain boundary-like regions with a larger free volume between amorphous regions of higher density. Such structures will be investigated.

In the plane-wave density-functional theory (DFT) methods, physical quantities such as energy and stress are given as an integral or average throughout the supercell. If such quantities can be given in each local region in the supercell, nature of defective systems such as grain boundaries (GBs) and interfaces between dissimilar materials can be effectively analyzed. Historically, energy-density [1] and stress-density [2] methods to calculate local energy and local stress were proposed within the plane-wave pseudopotential methods. However, practical applications have not been performed enough, due to the difficulty in the gauge-dependent problem in these methods. Recently, we have clarified the formulation of the energy density and stress density in the projector augmented wave (PAW) method for the first time [3], and proposed that local energy and local stress can be given as unique physical quantities if the energy and stress densities are integrated within proper local regions where the gauge-dependent terms are integrated to be zero. This strategy has been successfully applied to atomic-layer resolved stresses on Al surfaces [3] using our package software QMAS (Quantum MAterials Simulator) for the PAW method [4]. In this paper, we present our recent applications of local-energy and local-stress analysis to GBs and interfaces in metals and ceramics. For defective systems, it is necessary to define local regions to satisfy the condition for the gauge-dependent problem effectively. We adopt atom-centered regions via fuzzy-Voronoi or Bader integration schemes. We deal with tilt and twist GBs in Al and Cu, tilt GBs in Fe and SiC, and metal-carbide interfaces. Results of metallic GBs are compared with those by the embedded-atom method (EAM) potentials. We show that the local-energy and local-stress calculations are quite effective to investigate the stability, mechanical behavior and impurity effects of various GBs and interfaces [5]. Acknowledgement: The present study was supported by the Grant-in-Aid for Scientific Research on Innovative Area, “Bulk Nanostructured Metals”, by JST Industry-Academia Collaborative R&D Programs, and by the Strategic Programs for Innovative Research, MEXT, and the Computational Materials Science Initiative, Japan. References: [1] N. Chetty and R.M. Martin, Phys. Rev. B 45, 6074 (1992), [2] A. Filippetti and V. Fiorentini, Phys. Rev. B 61, 8433 (2000), [3] Y. Shiihara, M. Kohyama and S. Ishibashi, Phys. Rev. B 81, 075441 (2010), [4] S. Ishibashi, T. Tamura, S. Tanaka, M. Kohyama and K. Terakura, http://qmas.jp, [5] S. Ogata, Y. Umeno and M. Kohyama, Modell. Simul. Mater. Sci. Eng. 17, 013001 (2009).

Models of crystal plasticity should cover many length scales and overlap each other with effective scale transitions. Using Molecular Dynamics (MD) techniques we study the impacts of an edge dislocation on a {(914),(19-4),0} grain boundary (GB) in Al under load. This GB minimizes the anisotropy effect caused by the application of a shear stress on a bicrystal with fcc structure and hence allows a comparison between 3-dimensional MD and 2-dimensional Dislocation Dynamics (DD) modeling. Several thousands of GBs having the same misorientation were constructed, different only in very small microtranslations of one of the two crystals parallel to the GB plane prior to final relaxation by MD. The lowest-energy GB of these was selected for further study. The resulting GB structure has such a complexity that there are 15 different impact points for an approaching dislocation. Applying iterative MD and energy/stress minimization techniques we force a dislocation to move towards the GB on each of the 15 glide planes. After absorption of the dislocation the 15 atomic displacement fields are analyzed in terms of dislocation spreading, yielding values that can be subsequently used in DD simulations.

The practice of adding impurities to control the evolution of crystalline interfacial microstructures is based on rough estimates on the drag force exerted by the resultant impurity cloud on the moving interfaces. However, in spite of several existing theoretical frameworks, there is no reliable estimate on the extent of the drag due to challenges are in extracting two key ingredients, i) intrinsic mobilities of grain boundaries and ii) the impurity-grain boundary interaction energies, thereby limiting efforts to control microstructural evolution during processing of commercial purity alloy systems. In this talk, we present results of atomic-scale simulations on extracting the variation in these two quantities for a series of high-angle grain boundaries in the Al-Mg system. We find dramatic variations in the impurity-grain boundary interactions with boundary type and the interaction strength renders the anisotropy in intrinsic mobilities inconsequential under typical processing conditions. Extraction of the effective (extrinsic) mobility using existing theoretical frameworks confirms the effect, and we find excellent agreement between our predictions and mobilities extracted via prior in-situ experiments. A conclusion, we discuss the implications of the results on microstructural evolution during annealing phenomena.

The evolution of grains in nanocrystalline materials is examined using the phase field crystal (PFC) method. The strength of the PFC method lies in its ability to follow the atomic scale motion of a process that occurs on diffusive timescales. Due to the atomic scale resolution of the PFC method, the evolution of the dislocation structure of a grain boundary and the local atomic displacements of atoms near the boundary during grain growth can be determined. For example, we find that grain boundary dislocation interactions give rise to a repeating faceting-defaceting transition during grain shrinkage that leads to a different relationship between the grain area and time than that seen in classical curvature-driven grain growth. The evolution of the dislocation structure during grain growth in nanocrystalline materials can also induce grain rotation and, in particular, translation. The results of the simulations will be presented and discussed.

Strontium titanate based materials have great interest for a wide range of applications in microelectronics namely in tunable microwave devices, due to the high electric-field dependence of the permittivity and more recently as a possible thermoelectric oxide material. A grain growth anomaly in strontium titanate (SrTiO3) was recently reported [1]. Between 1200 to 1600 degree C grain growth kinetics did not follow the classical Arrhenius-type temperature dependence. Changes in the faceting behavior of the grain boundaries at high temperatures were pointed as the possible explanation for the anomalous grain growth behavior observed in SrTiO3 [1, 2]. In this work the anomaly in the dependence of grain growth kinetics on the temperature of strontium titanate is revisited. New data is presented which lead to a reinterpretation of the phenomenon. Discontinuities on the evolution of grain growth with the sintering temperature were observed, defining four grain growth regimens with transitions at temperatures around 1500, 1550 and 1605 degree C. These transitions correspond to grain size decreases, despite the increasing sintering temperature. These discontinuities were also observed in the conductivity response of the system (grain and grain boundaries) determined by impedance spectroscopy. By using complementary experimental tools, we proved that this anomalous behavior is directly related to the presence of liquid phase, which amount varies with the temperature and showing trends consistent with a scenario of retrograde solubility, reported for metals and semiconductors, but not reported before for ceramics. These new results have great scientific and technological relevance in tailoring the microstructure and dielectric response of SrTiO3 based materials and using grain boundary behavior for materials design. References [1] M. Baurer, D. Weygand, P. Gumbsch, M.J. Hoffmann, Scripta Mater, 61 (2009) 584-587. [2] M. Baurer, H. Stormer, D. Gerthsen, M.J. Hoffmann, Adv Eng Mater, 12 (2010) 1230-1234.

A classical force-field model with partial ionic charges was applied to study the behaviour of oxygen vacancies in the perovskite oxide strontium titanate (SrTiO₃). The dynamical behaviour of these point defects was investigated as a function of temperature and defect concentration by means of molecular dynamics (MD) simulations. The interaction between oxygen vacancies and an extended defect, here a Σ3(111) grain boundary, was also examined by MD simulations. Analysis of the vacancy distribution revealed considerable accumulation of vacancies in the envelope of the grain boundary. Possible clustering of oxygen vacancies in bulk SrTiO₃ was studied by static lattice calculations within the Mott-Littleton approach. All binary vacancy-vacancy configurations were found to be energetically unfavourable.

It is well established that grain boundaries and interfaces act as sinks for radiation-induced defects, promoting the radiation tolerance of materials. What is less clear is exactly how boundaries separately modify the thermodynamics and kinetics of defects, especially larger defect clusters that might be produced during collision cascades, to provide this enhanced radiation tolerance. Here, we use accelerated molecular dynamics (AMD) and adaptive kinetic Monte Carlo (AKMC) techniques to examine the mobility of defect clusters in a variety of grain boundaries in Cu, with special emphasis on how that mobility depends on the character of the grain boundaries. In particular, we find that different boundaries interact very differently with defect clusters, with some greatly retarding the mobility of such clusters and others having relatively little effect, compared to bulk behavior. These results have implications not only for radiation damage evolution, but for other applications where boundary-mediated mass transport is crucial.

Ferroelectric lead zirconate titanate (PZT) is one of the most important electroceramic materials and applied in actuators, sensors, memory devices, etc. In all these cases, PZT has to be insulating and mass or charge transport along and across interfaces is usually undesirable. Still, high electric fields as well as high temperatures may cause substantial diffusion or migration of ions in bulk and along/across grain boundaries and this can drastically change the overall materials properties. However, information on the ion transport kinetics in PZT bulk and grain boundaries is scarce. In the first part of this contribution it is shown how combination of several experimental tools helps analyzing a number of interface-related ion transport phenomena in donor-doped PZT: i) Very fast 18O tracer diffusion along grain boundaries was observed at temperatures above 600°C. ii) Strongly accelerated oxide ion (tracer) diffusion in PZT was triggered by application of high electric fields; interestingly the fast diffusion was perpendicular to the field direction. iii) At temperatures of about 350-500°C, PZT multilayers with co-sintered Ag-Pd electrodes underwent strong resistance degradation under field load. The highly conducting region was located in a near-surface zone of several 10 micrometer thickness. Conductive atomic force microscopy (c-AFM) and chemical analysis revealed deposition of metallic silver paths in grain boundaries. Their formation started close to the anode and is not explainable by conventional electrochemical metal deposition. In many regards SrTiO3 is similar to PZT or BaTiO3, all being large band gap perovskite-type oxides, and it often acts as a kind of prototype material for such oxides. Detailed knowledge is available on its defect chemistry and also on structure and transport properties of grain boundaries in polycrystalline SrTiO3. However, defect chemistry and electrical properties of very thin SrTiO3 layers are not very well-understood yet; interfacial space charge layers may play an overwhelmingly important role there. In the second part of this contribution, we present the results of a combined 18O tracer diffusion and impedance study on acceptor-doped SrTiO3 layers of about 100 nm thickness, deposited on Al2O3, LaAlO3, undoped SrTiO3 and Nb-doped SrTiO3. Among others, pronounced near-surface space charge zones could be detected in tracer diffusion experiments and a strong dependence of the effective tracer diffusion coefficient on the substrate was found. Anisotropy of mass and charge transport properties are discussed and kinetic results are correlated with structural information from HRTEM and XRD studies.

Alumina (Al₂O₃) is a thermally grown oxide that efficiently enables the alumina-forming alloys to withstand high operating temperatures and oxidizing environments. The growth of alumina scale is closely related to the diffusion process of oxygen and aluminum in the material. To acquire knowledge of diffusion mechanisms and predict diffusion coefficients, we have calculated all the involved energetics about the O and Al diffusion through the grain boundaries of α-Al₂O₃ using first-principles density function theory method. In this study, we examined specifically the diffusions of O and Al via vacancy exchange mechanism along low-energy Σ3 (1 0 -1 0) prismatic grain boundaries (whose energies are in the range of 0.25 to 0.39 J/m²) and high-energy Σ3 (0001) basal grain boundaries (whose energies are in the range of 2.12 to 2.38 J/m²). It was found that both the vacancy formation energy and diffusion activation energy were lower on the grain boundaries than in bulk α-Al₂O₃. Moreover, our results show that the vacancy formation energies are significantly lower on the high energy grain boundaries than on the low energy grain boundaries of Al₂O₃. Furthermore, our results from first-principles climbed image nudged elastic band calculations indicated that the energy barriers were lower by more than 2 eV for atomic O diffusion and lower by about 0.8 eV for atomic Al diffusion on the high energy Σ3 (0001) basal grain boundaries than in the bulk. However, the calculated energy barriers for atomic diffusion on the low energy Σ3 (10-10) prismatic grain boundaries are just lower by a very small amount as compared to those in bulk Al₂O₃. Hence, the atomic diffusion in α-Al₂O₃ was predicted to be much faster along the high-energy grain boundaries than that along the low-energy grain boundaries. In addition, we examined the correlation between the energetics of diffusion and the geometric arrangement of atoms in the bulk and different grain boundaries of α-Al₂O₃.

We use atomistic modeling to investigate the structure and transport of isolated point defects at low-angle twist grain boundaries in MgO, chosen as a simple model oxide. We find that point defects delocalize at misfit dislocations at the grain boundary and the extent of delocalization depends on the length of the misfit dislocation segment. Delocalized point defects migrate from one misfit dislocation segment to another misfit dislocation segment first by localizing at the misfit dislocation intersection and then delocalizing at another of the misfit dislocation segments. Point defect dynamics in these model ceramic grain boundaries are compared with those in metal heterointerfaces and the implications of such point defect dynamics for defect annihilation under extreme conditions, such as under irradiation, are discussed. This material is based upon work supported as part of the Center for Materials at Irradiation and Mechanical Extremes, an Energy Frontier Research Center funded by the U. S. Department of Energy, Office of Science, Office of Basic Energy Sciences under Award Number 2008LANL1026

Multiferroic materials of current interest tend to have (anti)ferroelectric and (anti)ferromagnetic properties arising at separate instabilities which may be widely separated in temperature. Although they may be multiferroic, they do not have multiferroic phase transitions and their domain walls are likely to be (anti)ferroelectric or (anti)ferromagnetic but not both simultaneously. Multiferroic materials which involve magnetism and ferroelasticity, on the other hand may undergo genuinely multiferroic transitions at which both properties appear simultaneously. In these cases the domain walls can be either (anti)ferromagnetic (180°) walls or W and Wprime twin walls which are both ferroelastic and (anti)ferromagnetic (90° walls for, e.g. cubic → tetragonal, 60° and 120° walls for e.g. hexagonal → monoclinic). A key property is the strength of coupling of the magnetic order parameter with strain and the dynamic character of the different ferroelastic twin walls can be investigated through their tendency to give rise to acoustic attenuation in measurements of elastic properties. A number of materials have been investigated in this context using resonant ultrasound spectroscopy (RUS). In FexO, magnetic ordering is coupled with a pseudoproper ferroelastic transition and shear strains of up to a few % are observed. Twin walls in the rhombohedral phase are antiferromagnetic/ferroelastic, and give rise to strong attenuation at RUS frequencies. In the weakly ferromagnetic (canted antiferromagnet) phase of Fe2O3 the monoclinic symmetry-breaking strain is sufficiently small as to not yet have been detected with laboratory diffraction techniques but ferroelastic W and Wprime walls appear to be present. From a strain analysis of the orientations of these, an estimate of the symmetry-breaking shear strain at room temperature is ~5permil;. Single crystal samples give strong acoustic attenuation and, even for these low strains, the twin walls behave as if they are normal ferroelastic walls with respect to an applied stress. On the other hand, coupling between magnetic ordering and strain in KMnF3 perovskite appears to be minimal and there is no evidence of acoustic loss due to processes other than motion of twin walls associated with the octahedral tilting (improper ferroelastic) transitions. In the low temperature Pnma structure, which is weakly ferromagnetic (canted antiferromagnetic), the ferroelastic twin walls are immobile. Pinning of ferroelastic twin walls by defects is most likely to occur if the walls are thin, but ferroelastic twin walls which are also magnetic walls will tend to adopt the thicker configuration of the latter if there is significant coupling with strain. Thus they will tend to remain mobile over wider temperature intervals. For systems with multiple order parameters there are theoretical possibilities for controlling the magnetic properties of the twin walls themselves, again depending on the form and strength of strain coupling.

Ferroelectric ceramics with perovskite structures, like Pb(Zr,Ti)O3 (PZT), which is widely used, e.g., for electromechanical actuators and sensors, or (K,Na)NbO3 (KNN), which is a promising lead-free substitute for PZT, have dense polycrystalline microstructures. The crystallites have complex domain structures, and they contain atomic dopants in order to optimize the functional performance. The objective of this work is an investigation of the fundamental behaviour of the ferroelectric polarisation at domain walls, grain boundaries, or atomic dopants in PZT and KNN by means of first-principles supercell calculations using density functional theory (DFT) and classical atomistic simulations with empirical ionic potentials. The focus of this talk will be on isolated Cu substitutionals and defect complexes of these with charge-compensating oxygen vacancies in KNbO3. In this study we found a morphotropic phase transition in Cu-doped KNbO3 similar to one which was reported earlier for Li-doped (K,Na)NbO3. We estimated the energy needed for switching the ferroelectric polarization in a crystal region with a defect complex. This energy determines how defect complexes can pin domain walls and contribute to ferroelectric hardness and/or large piezoelectric strain.

Hexagonal manganites RMnO3 with R=Ho-Lu, Y and Sc exhibit have received much attention because of their intriguing physical properties such as magnetoelectric effect [1]. Hexagonal YMnO3 exhibits two structural phase transitions from a paraelectric P63/mmc structure to an intermediate paraelectric one at 1270 K, accompanying the tilting of the MnO53+ polyhedra and corrugation of the Y3+ layers. Subsequently, additional displacements of the MnO53+ polyhedral and the Y3+ ions at 920 K took place and a spontaneous polarization along the [001] direction appears. At lower temperature of 70 K, the antiferromagnetic transition occurs [2]. In addition, characteristic domain structures consisting of six FE and structural antiphase domains appears in the ferroelectric phase, which can be identified as the “cloverleaf” pattern [3-5]. Recently, the domain wall structures in hexagonal manganites (LuMnO3 and TmMnO3) were investigated at the atomic scale [6]. In this study, in order to elucidate the structural characteristics of the domain walls in the FE phase of hexagonal YMnO3, we have applied high-angle annular-dark-field (HAADF) imaging technique to real-space imaging of the FE and structural antiphase domain walls in hexagonal YMnO3 and mapped the atomic shifts of Y and Mn ions. Two types of 180°domain walls can be identified. One is the transverse and longitudinal domain walls with atomic displacements of a/3 and the other is the transverse domain walls with atomic displacements of 2a/3. In contrast, the displacements related to the MnO5 polyhedral remain intact across the domain walls. These features of the FE and structural antiphase domain walls should be important to understand unusual physical properties in hexagonal manganites. In addition, we have investigated changes of the FE and structural antiphase domain walls by partial substitution of Ti4+ ions for Mn3+ ones in YMnO3. It is revealed that the FE displacement along the [001] direction disappear and correlation length of the Mn3+ trimerization becomes short less than 10 nm. [1] T. Katsufuji et al., Phys. Rev. B 66, 104419 (2001). [2] T. Lonkai et al., Phys. Rev. B 69, 134108 (2001). [3] T. Choi et al., Nat. Matter 9, 253 (2010). [4] M. Lillenblum et al., J. Appl. Phys., 110, 052007 (2011). [5] K. Kobayashi et al., (in press). [6] Q. H. Zhang et al., Phys. Rev. B 85, 020102(R), (2012).

To explore charge-controlled physics of transition metal oxide thin film heterostructures and develop new classes of materials with competing functionalities, understanding the atomic mechanism of structural order parameter changes has remained a goal of active research during last two decades. Plentiful theoretical studies of interface mediated coupling have been reported to decode the atomic mechanisms responsible for coupling phenomenon between polarization and spin exchange. However, direct atomic level observations of polarization and electronic phenomena have not kept pace. An approach combining quantitative aberration corrected scanning transmission electron microscopy (STEM) and advanced EELS techniques, providing atomic scale insight into structural order parameter changes and chemical behaviors, enables us to directly reveal multiple aspects of interface behavior. Here, we report the application of this approach to studies of BiFeO3 (BFO) )/(La,Sr)MnO3 (LSMO) interfaces where a change in local valence state can be detected for different polarization states of BFO. In this study we examined a 50-nm thick BFO film grown on 5-nm of MnO2-terminated LSMO on STO substrate. STEM EELS analysis (Nion UltraSTEM operated at 100 kV and equipped with Gatan Enfina EEL spectrometer) was used to verify the interface termination and evaluate cation intermixing. For opposite polarization orientations in BFO (i.e. downward and upward polarizations with respect to the interface), we discovered different valence states for near-interface Mn cations. Simultaneously conducted unit-cell-by-unit-cell mapping of structural distortions indicated that a local lattice parameter increase at the interface is observed only for downward polarization orientation, implying polarization charge driven localized oxygen vacancy injection. Atomic-scale observations as used in this study are therefore critical for uncovering the interplay of electronic, ferroelectric, and ionic/chemical aspects of interface behavior * This research is sponsored by the Division of Materials Sciences and Engineering (YMK, SJP, SVK, AYB), Office of BES of the U.S. DOE, and by appointment (YMK) to the ORNL Postdoctoral Research Program administered jointly by ORNL and ORISE. Instrument access via ShaRE user facility, which is supported at ORNL by Office of BES of the U.S. DOE, is gratefully acknowledged.

Doping with Al and O atoms in beta-Si3N4 network gives rise to constitute a new type of solid solution material called beta-SiAlON, exhibiting superior outstanding chemical, thermal and mechanical properties. The scanning transmission electron microscopy (STEM) observations of interstitial rare-earth atom (e.g., Eu, Yb and Ce) incorporation into beta(LEDs). On the other hand, recent analytical TEM studies at nanoscopic scale clearly showed the incorporation of transition metal atoms, e.g., Fe, Cr and Ti into beta-SiAlON crystal structure [1-2]. However, a simple question where do these atoms sit in the beta-SiAlON lattice is still not answered. For the sake of the clarity at the atomic-scale, we report the use of various state-of-the-art aberration-corrected STEM/TEM instruments for experimental results and first-principles calculation routes for theoretical studies on the fully densified gas pressure sintered SiAlON samples doped with transition metal atom Interestingly, Z (atomic number)-contrast atomic-resolved STEM images reveal that titanium (Ti) atoms substitute with silicon (Si) and/or aluminum (Al) atomic columns in the basic tetrahedron units of beta-SiAlON lattice. In addition, first principles calculation results strongly suggest the existence of Al and O impurities in beta-Si3N4 network plays a vital role on the doping capabilities of Ti atoms within beta-SiAlON lattice. It is anticipated that these acquired atomic-scale experimental and theoretical results, for the first time, will pave the way to new approaches in the production of transition metal atom-doped beta-SiAlON ceramics for different applications. [1] H. Yurdakul and S. Turan, Cer. Internationals, 37, 1501-1505 [2] H. Yurdakul and S. Turan, http://dx.doi.org/10.1016/j.ceramint.2012.04.028. [3] This research was supported by TUBITAK 2214-coded Ph.D. Scholarship Program (H.Y.) and DOE Basic Energy Sciences, ORNL-Division of Materials Science and Engineering (USA).

The plastic deformation of hexagonal close packed (hcp) metals is normally dominated by single slip (basal or prismatic) and twinning rather than cross-slip which frequently occurs in face-centered cubic (fcc) metals. While most fcc metals shows good ductility, a large range of ductility is found in hcp metals. Here we perform molecular dynamics simulations of hcp Mg and Zr to reveal the key deformation mechanisms of nucleation and interaction of slip and twinning. Mg and Zr have very different c/a ratios (Mg: 1.624 close to the ideal value of 1.633 and Zr: 1.93) and dominant deformation modes (Mg: basal, Zr: prismatic); they thus may be taken to represent typical properties in various hcp metals. Grain-boundary nucleated processes and conventional slip and twinning are compared in textured and random polycrystalline structures. This work of DK and SRP were supported by DOE NERI contract DE-FC07-07ID14833. The work of ZL and MJN was supported by the by the Consortium for Advanced Simulation of Light Water Reactors (www.casl.gov), an Energy Innovation Hub (http://www.energy.gov/hubs) for Modeling and Simulation of Nuclear Reactors under U.S. Department of Energy Contract No. DE-AC05-00OR22725. The work of MVM was sponsored by NSF Award Number DMR-0845868.

This work presents the results of a comparative molecular dynamics study using interatomic potentials describing fcc metals with different stacking fault energies (SFE). Fully periodic polycrystalline samples with an average 40nm grain size and random tilt boundaries were deformed to a 9% strain level. Low SFE materials present highly non-planar grain boundary structures. Grain boundaries with these non-planar structures emit dislocations at lower stresses than their higher SFE counterparts. The implications for the mechanical response of these materials are discussed.

Imaging light atoms requires optimization of electron optics, detector configuration and electron acceleration voltages. Here we report progress in imaging oxygen atoms at the interface of gold Nano Crystals (Au NCs) grown on rutile TiO2, which have attracted considerable interest for their remarkable size-dependent catalytic activities, especially for catalyzing the oxidation of carbon monoxide. Several mechanisms have been proposed to explain such surprising chemical activities and almost all of them involve interactions at the Au NCs and support interface, which renders the study of interfacial structure tremendously important. Scanning transmission electron microscopy (STEM) with annual dark-field (ADF) detector and aberration corrector (AC) is a powerful tool to characterize interfacial structures at atomic resolution, Au and Ti atoms can be resolved at 0.1 nm or better using high angle ADF. Oxygen atoms, however, are difficult to detect using ADF-STEM. Recent reports have shown that Annular Bright Field (ABF) STEM can be used to observe light atoms [1-3]. Here we explore the application of ABF-STEM for imaging interfacial light atoms. To enable high resolution AC-STEM investigation of NC interfaces, we have developed a novel NC synthesis technique for preparing Au NCs on vicinal TiO2 surfaces. Au nanoparticles about 2~5 nm in diameters, were first deposited onto the single crystal TiO2 substrate, pre-thinned and annealed in air, by e-beam evaporation deposition at room temperature. Then the sample was annealed in air at high temperatures to form epitaxial Au NCs. To detect the oxygen atoms, a JEOL JEM 2200FS AC-STEM was modified for ABF-STEM imaging. Optimal imaging conditions were tested by experiment and simulations, including the benefits of using lower acceleration electrons to avoid radiation damage. Oxygen atoms are resolved experimentally and the result was confirmed by multi-slice STEM simulation. To enhance the resolution, we performed spatial averaging during the processing of raw images, which can automatically detect and average the similar areas with a template from the raw image, effectively reduce the S/N ratio and hence improve the contrast. With above techniques, we then explored the possibility to detect oxygen atoms at the interface, in order to measure the interfacial registry of metal-oxides. The results will be presented and discussed in the background of gaining better understanding about interfacial effect in real catalysts. This work is supported by NSF the NSF Grant No. DMR 0449790. Electron microscopy was carried out in the Frederick Seitz Materials Research Laboratory Central Facilities, University of Illinois. References: [1] Findlay, S. D. et al. Ultramicroscopy 110, 903-923 (2010) [2] Findlay, S. D. et al. Ultramicroscopy 111, 285-289 (2011) [3] Ishikawa, R. et al. Nature Mater. 10, 278-281 (2011)

One family of the complex perovskite Ba3BNb2O9 has shown desirable dielectric properties in the field of microwave and satellite communications [1], and some have photocatalytic properties [2]. The cation ordering of these compounds have effect on the dielectric properties, where it was reported that stacking faults and domains are the cause of disordered state [3]. In our work, the physical properties at low temperatures down to 2 K have been measured and shown the materials having multiferroic properties. The cation ordering and disordering and microstructures of these compounds are studied by advanced scanning transmission electron microscope (STEM) techniques at atomic resolution, such as high angle annular dark field imaging (z-contrast imaging), annular bright field imaging, STEM spectrum imaging and electronic structures by electron energy loss spectroscopy. Ref: [1] H. Kagata, et. al., J. Appl. Phys. 33 (1994) 5463. [2] J. Jin, et. al., J. Phys. Chem. B 107 (2003) 4936. [3]Y. Liu et. al., Physica B 385 (2006) 564.

Adhesion of soft-materials onto surfaces of inorganic solids is a key aspect of designing composite material systems with controlled thermal and mechanical properties. To elucidate the role of dispersive coupling between adherent and substrate on resultant bi-material properties, we performed a comprehensive atomistic simulation study of adhesive properties of interfaces between aromatic rigid-rode poly-[(4,4'diphenylene) pyromellitimide] nanometer-size layers and a dimer-reconstructed silicon surface. The effects of interactions across the interface and interface-induced ordering on adhesion are scrutinized in detail via classical molecular dynamics and compared with experimental results, obtained using a novel laser-induced stress-waves technique. Ab initio calculations are employed to investigate the effect of point-like and linear surface defects on adhesive properties. Furthermore, to account for non-planarity of substrate surfaces, we developed theories of adhesion onto rough substrates, thereby quantifying the role of non-planarity in shaping the adhesive behavior. The implications of defects and the non-planar nature of the substrate for energetics of adhesive coupling are discussed in detail, within the frameworks of methods employed.

Direct Atom-resolved Imaging of Ordered Defect Superstructures at Oxide Grain Boundaries Zhongchang Wang,1,* Mitsuhiro Saito,1 Susumu Tsukimoto,1 and Yuichi Ikuhara1,2 1WPI, Advanced Institute for Materials Research, Tohoku University, Sendai, Japan 2Institure of Engineering Innovation, The University of Tokyo, Tokyo, Japan The ability to spatially resolve and chemically identify atoms in complex defect regions would bring atomistic understanding of structure-property interplay closer to reality [1]. This is particular significant for the widely used polycrystalline materials, where their prevalent intercrystalline defects, i.e., grain boundaries (GBs), are universally implicated in affecting remarkably their properties and hence many of their applications. However, this resolving is still extremely challenging [2] because GBs have long been known to serve as effective sinks for atomic defects and impurities, which in turn may drive structural transformation of GBs, resulting in modification of material properties. Regardless of origin of the segregated defects, such intricate interplay associated with the embedded nature of GB complicates our efforts to pinpoint exact site and chemistry of the entities in the defective regions, thereby hindering the development of a deeper understanding of how specific defects mediate property shift. Here, we show that advanced transmission electron microscopy, spectroscopy, and first-principles calculations can provide three-dimensional images of complex multi-component GBs with both atomic resolution and chemical sensitivity. The unprecedented resolution of these techniques allows us to demonstrate that, even for the simple cubic oxide MgO, GBs can accommodate complex ordered defect superstructures that induce pronounced electron trapping states in band gap of MgO [3]. These results point to existence of new impacts associated with defect-GB interactions in ceramics, and demonstrate that atomic scale analysis of complex multi-component structures insides materials is now becoming possible. [1] P. D. Nellist et al., Science, 305, 1741 (2004). [2] C. L. Jian and K. Urban, Science, 303, 2001-2004 (2004). [3] Z.C. Wang, M. Saito, K. P. McKenna, L. Gu, S. Tsukimoto, A. L. Shluger, and Y. Ikuhara, Nature 479, 380 (2011).

In recently discovered intercalated iron-selenide superconductors AFe2-xSe2 (A= K, Li, Na, Rbhellip;), superconductivity with a Tc of 30 K appears to coexist with an exceptionally strong antiferromagnetism and large magnetic moment on the Fe sites. This finding has generated intense experimental efforts in order to understand the relationships between superconductivity, magnetism and structure in these materials. In spite of the large chemical and structural similarity between AFe2-xSe2 and the prototypical iron arsenide superconductors Ba(K)Fe(Co)2As2 (122 Fe-pnictides), it has not been possible to synthesize the former compounds as electronically homogeneous superconducting single crystals. Several experiments have suggested the existence of spatially separated superconducting and antiferromagnetic phases, which might be related to the presence and/or order of Fe vacancies. Based on the investigation of the closely related TlFe1.6Se2 compound, we provide evidence that nanoscale phase separation in ordered and disordered vacancy regions leads to a different magnetic ground state than the ground state of single-phase counterparts, suggesting that magnetoelastic coupling between different phases is important for the emergence of superconductivity in AFe2-xSe2. Here we show how real-space, aberration-corrected scanning transmission electron microscopy can be a critical tool for investigating nanoscale, crystallographically coherent phase separation into ordered and disordered regions. We will show distinctive signatures in the electron energy loss spectroscopy (EELS) fine structure of the Fe-L2,3 edges for these compounds and discuss the results in light of a comparison with 122 Fe -pnictides. Research sponsored by US DOE Office of Science, Materials Science and Engineering Division.

Most oxides, whether bulk or thin film, are generally prepared in polycrystalline form. Therefore, it is not surprising to find that grain boundaries often play a key role in controlling charge and mass transport through these materials. The exact role that they play depends on the nature of the grain boundary stoichiometry, the charge and size of the segregants and impurities, the degree of coverage by second phases and the grain boundary to volume ratio. In this presentation, we focus on semiconducting or mixed conducting (ionic + electronic) oxides in which space charge fields are present in the vicinity of the boundaries. In the case of ZnO, depletion of electrons near the grain boundaries results in space charge barriers and an overall reduction in conductance by as much as 10 orders of magnitude. Here we utilize high cation diffusion along the grain boundaries to transport barrier creating dopants to activate specific boundaries. In the case of nanostructured CeO₂, we attempt instead to create an accumulation layer at the boundaries to enhance ion transport along those boundaries. In the final example, we investigate the impact of segregants on charge transfer across a heterointerface between a mixed conductor and adsorbed gas molecules.

An existing barrier to the development of high performance solid oxide fuel cells is the poor ionic conductivity at room temperature of typical oxide-ion conductors, making high operating temperatures necessary for oxide ion mobility to be sufficiently fast for operational devices. One strategy for improved room temperature ionic conductivities is to exploit the "nanoionic" effects demonstrated by thin films and heterostructures, where enhanced ionic conductivities have been reported compared to corresponding bulk samples. The canonical example when considering heterostructures is that of CaF₂|BaF₂ [1]. As the thickness of the domains is reduced, ionic conductivity increases, attributed to the increasing contribution of space charge regions; with these predicted to form as a consequence of the fluoride ion chemical potential differing between the two component phases, driving a separation of Frenkel pairs across the CaF₂|BaF₂ interfaces [2]. CaF₂ and BaF₂ both adopt the fluorite structure, but have different lattice parameters, giving rise to epitaxial strain or mismatching at interfaces. This strain can be relieved by the formation of dislocations, which may facilitate enhanced conduction of ions [3]. Even in the absence of extended defects, strain can be expected to modify the energies of point defects and their diffusion barriers relative to the corresponding bulk materials, and it is not clear to what extent experimentally observed conductivity enhancements can be attributed to space charge formation or strain effects in this heterostructured material. We have performed atomistic simulations of a model PbF₂|PbF₂ heterostructure as a "strain-free" structural analogue to the CaF₂|BaF₂ system. PbF₂ undergoes a superionic transition at high temperature, and is fluorite structured. By including an adjustable difference in the chemical potential of the fluoride ions, Δmu;(F^-), between neighbouring domains, we have modelled the redistribution of mobile fluoride ions and corresponding change in ionic conductivity produced in response to an interfacial difference in chemical potential in an ideal system, and have assessed how this behaviour varies as a function of the magnitude of Δmu;(F^-). [1] Sata, N., Eberman, K., Eberl, K. & Maier, J. Nature 408, 946-949 (2000). [2] Guo, X. & Maier, J. Adv. Mater. 21, 2619-2631 (2009). [3] Jin-Phillipp, N. Y. & Phillipp, F. Philosoph. Mag. 84, 3509-3516 (2004).

Proton-conducting Y-doped BaZrO3 combines a high chemical stability with good bulk conductivity. Nevertheless, the small grain size (< 1 micrometer) and blocking character of the grain boundaries still impede a broader application, e.g. in intermediate temperature fuel cells. Dense Y- and Sc-doped BaZrO3 ceramics were obtained by "Spark-Plasma-Sintering" at 1600°C (5 min, 50 MPa). Additional long-time annealing (20 h at 1700°C) decreases the grain boundary resistance by 1-3 orders of magnitude, while grain size and bulk conductivity remain constant. TEM-EDXS analysis indicates a pronounced dopant accumulation in the grain boundary region for the annealed samples [1]. The comparable degree of segregation for Sc3+ (negligible size mismatch to Zr) and Y3+ (large mismatch) suggests a positive excess charge in the grain boundary core as main driving force for the dopant segregation. Within the space charge model, this dopant accumulation in the grain boundary core and/or the space charge zone will decrease proton depletion and thus increase total conductivity. Further evidence for the validity of the space charge model is obtained from impedance measurements under large DC bias [2]. An extremely large-grained sample with >100 micrometer grain size and Pt microcontacts allows to measure impedance spectra across individual grain boundaries. DC bias values of up to 1 V per boundary strongly decrease the grain boundary resistance as well as capacitance. Both findings support the model that a space charge depletion zone for protons (and oxygen vacancies) is the main reason for the blocking behavior of the grain boundaries. Grain boundaries in other acceptor-doped perovskites with large band gap (Fe-doped SrTiO3 [3], (La,Sr)GaO3 [4]) also exhibit a positve excess charge in the grain boundary core, which is caused by the segregation of oxygen vacancies to the core. The similar behavior of BaZrO3 grain boundaries suggests that a high excess vacancy concentration in the core, favored by geometric constraints in the distoretd core structure, is a quite common feature for perovskites. Possibilities for decreasing the positive core charge will be discussed. [1] M. Shirpour et al., J. Phys. Chem. C 116 (2012) 2453 [2] M. Shirpour et al., Phys. Chem. Chem. Phys. 14 (2012) 730 [3] R.A. De Souza, J. Fleig, J. Maier, O. Kienzle, Z.L. Zhang, W. Sigle, M. Rühle, J. Am. Ceram. Soc. 86 (2003) 922 [4] H.J. Park, S. Kim, J. Phys. Chem. C 111 (2007) 14903

Experimentally, it is known that small amounts (up to 200 wppm) of thorium added to iridium metal segregates to grain boundaries and forms a precipitate, Ir₅Th, which refines grains, increases inter-grain cohesion and thereby increases the high temperature ductility of resulting alloys. Cerium is similar in many ways: it is chemically similar (in the same column of the periodic table as thorium but one row lower), segregates to grain boundaries, and forms an isomorphic grain pinning precipitate, Ir₅Ce. Despite these similarities, cerium nonetheless fails to provide as much cohesion between grains as thorium. Traditional models of impurity effects at grain boundaries fail to satisfactorily explain these contrasting behaviors of thorium and cerium doped iridium alloys. We develop alternative models which focus on the interfaces between phases in the two phase system, Ir and the intermetallic precipitate. First principles calculations show that there is a structural similarity between the Ir and precipitate phases, which leads to a natural orientation relationship, a small interfacial energy, and a low lattice misfit. Ir₅Th has a strong formation energy, relative to the pure Th and Ir phases; in contrast, Ir₅Ce has a significantly weaker binding. Various models of thin precipitate phases are studied, demonstrating that particularly for Ir₅Th, nanolayers of the precipitate are likely to form, with little driving force to coarsen. In contrast, the Ir₅Ce phases are less likely to form, particularly as nanolayers, due to their energetics. Research supported by the U.S. Department of Energy (DOE), Basic Energy Sciences, Materials Sciences and Engineering Division.

Interfaces play a crucial role in materials behavior and their evolution is of particular interest to understand the kinetics processes taking place during irradiation in the case of structural nuclear materials, or during high temperature exposure in the case of Ni based superalloys or Zr alloys. Atomic scale characterization including transmission electron microscopy and atom probe tomography are used to probe structure and chemistry of specific interfaces in materials under dynamic conditions. Several examples will be discussed, which will include radiation can induce non-equilibrium segregation (RIS) of solutes in ferritic steels, the growth of oxide films on Ni base superalloys and the development of oxides on Zr alloys.

Grain boundaries have long been known as having pervasive influences on various bulk properties, and such effects tend to dominate in most practically important nanocrystalline devices. Grain boundary doping has been demonstrated as a powerful method to alter the atomic and electronic structures,[1-2] therefore enables us to have an effective control over various material properties. As an example in this research, we focus on the rear earth doped titanate, praseodymium (Pr) and Europium (Eu) doped SrTiO3, which have been shown to have luminescence[3-4] and high temperature ferroelectric[5] properties. In this research, a series of undoped, Pr doped and Eu doped SrTiO3 bicrystals containing grain boundaries were fabricated with diffusion bonding technique[6]. Atomic structures of these grain boundaries were characterized using aberration corrected scanning transmission electron microscope (STEM), and the optical and dielectric properties were analyzed from electron energy loss spectroscopy (EELS). Based on these analyses, direct comparison of the atomic structures and optical properties in different dopant systems were obtained. Variations in the grain boundary atomic structure were analyzed statistically using the Z-contrast STEM-HAADF images. In the undoped grain boundary, two distinct atomic arrangements were discovered, one being a symmetric structure and the other being rigid-body shift structure that contains a rigid-body translation along the boundary plane, and statistical analysis of the grain boundary population shows the rigid-body shift structure is the prevalent structure. In the case of Pr doped and Eu doped grain boundaries however, the symmetric structure is the prevalent structure, but with different segregation behaviors. Further discussions on EELS analysis and the doping mechanism from first principles calculations will be shown in the presentation. [7] References: [1] Buban, J.P., et al., Science. 311 (2006) 212. [2] Klie, R.F., et al., Nature. 435 (2005) 475. [3] Jiang, C.G., et al., Appl. Phys. Lett. 94 (2009) [4] Okamoto, S. and H. Yamamoto, J. Lumines. 102 (2003) 586. [5] Duran, A., et al., J. Appl. Phys. 97 (2005) [6] Yamamoto, T., et al., J. Am. Ceram. Soc. 83 (2000) 1527. [7] This work is supported by the United States Department of Energy 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, which is operated by Battelle for the U.S. Department of Energy under contract DE-AC05-76RL01830. Part of this work was also conducted in Research Hub for Advanced NanoCharacterization, The University of Tokyo, supported by the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan.

Cerium oxide (CeO2) has attracted much attention in recent years for its applications in catalysis as well as electrolyte materials for solid oxide fuel cells. It has been reported that grain boundaries (GBs) play an important role in the oxygen transport properties in CeO2. The effects of GBs have been widely investigated using polycrystalline materials, however, most of those studies have taken account of averaged effect of many different GBs included in the material. Precise understanding of structure-property relationship requires knowledge on the atomic-scale details. In this study, the atomic structure of model CeO2 GBs were studied by the combination of aberration-corrected transmission electron microscopy (STEM), electron energy loss spectroscopy (EELS) and theoretical calculations. Yttria-stabilized zirconia (YSZ) bicrystals containing Σ3 and Σ5 GBs were first fabricated by diffusion bonding. Then, CeO2 thin films were heteroepitaxially grown on the bicrystal substrates by pulsed laser deposition (substrate temperature: 900 °C, oxygen pressure: 3.0 × 10-3 Pa). STEM observations were performed using JEOL JEM-2100F equipped with CEOS Cs-corrector. EELS spectra were acquired in STEM mode by an Enfina spectrometer (Gatan Inc). For theoretical approach, static lattice calculations were first carried out to determine the most stable GB translation state. The structure with minimum energy was further studied by density functional theory (DFT) calculations within local-spin density approximation (LSDA) + U formalism. Our experimental and theoretical approaches have successfully revealed the atomic structures of the CeO2 GBs. In the case of the Σ5 GB, nonstoichiometric GB core structure with oxygen vacancies is found to be the most plausible [1]. On the other hand, the Σ3 GB has the stoichiometric GB core structure [2]. Further investigation reveals that the density of oxygen sites with coordination deficiencies is lower for Σ3 GB, indicating the structural distortions at the Σ3 GB are not as pronounced as those at the Σ5 stoichiometric GB. Therefore oxygen vacancies would be formed in the Σ5 GB, while the Σ3 GB is stable enough without introducing oxygen vacancies. Our results suggest that GB geometry has a significant effect on the oxygen stoichiometry at the CeO2 GBs and, therefore, different GBs may have different degrees of effects to the oxygen transportation properties. [1] H. Hojo, T. Mizoguchi, H. Ohta, S.D. Findlay, N. Shibata, T. Yamamoto and Y. Ikuhara, Nano Lett, 10, 4668-4672 (2010). [2] B. Feng, H. Hojo, T. Mizoguchi, H. Ohta, S.D. Findlay, Y. Sato, N. Shibata, T. Yamamoto and Y. Ikuhara, Appl Phys Lett, 100, 073109 (2012).

Solid-state dewetting occurs in thin continuous metal films when capillary instabilities drive the nucleation and growth of holes. The film/substrate interface energy plays a crucial role during this process as it dictates the tendency of a thin metal film to adhere to a substrate. However, the majority of past studies have not considered atomic-scale characterizations of interface reactions during dewetting [1]. Ni films were DC magnetron sputtered onto Si (100) substrates covered with or without SiO2 layers. Nominal film thicknesses of 5nm and 15nm were deposited to investigate the dependence of interface chemistry and structure on varying film thicknesses. Cross-sectional interface characterization by scanning transmission electron microscopy in tandem with electron energy loss spectroscopy revealed unexpected Ni diffusion through an ultrathin native SiO2 layer [2]. At room temperature we observed resulting silicide formation with gradual changes in chemical composition. Independent of as-deposited film thickness, an equilibrium metal film thickness of roughly 6.5nm remained on the top of the native SiO2 layer. In-situ annealing of as-deposited films demonstrated that the reaction of Ni with Si below the native SiO2 layer competes with agglomeration of the remaining Ni film above the oxide layer. In addition to the results mentioned above we will report preliminary data on the characterization Ni films deposited on thick thermally grown SiO2 layers before and after solid-state dewetting. [1] R. Clearfield et al., Appl. Phys. Lett. 97 (2010). [2] A. M. Thron et al., Acta Mater. 60, 2668 (2012).

An understanding of the formation mechanism of core/shell nanostructure is significant for the controllable synthesis of nanomaterial architectures with tailored properties. Our current understanding of the growth of core/shell nanostructures are mostly based on the conventional ex situ studies, and inevitably it lacks intermediate details related to the formation dynamics of the nanostructures. Herein, we investigated the structural evolution of Pb/Fe3O4 core/shell nanparticles during solution growth using liquid cell transmission electron microscopy. We fabricated a self-contained liquid cell which allows the study of chemical reactions in a thin liquid layer sandwiched between two silicon nitride membranes. About 30 picoliters growth solution of iron(II) acetylacetonate (100 mM) and lead(II) acetylacetonate (60 mM) dissolved in triethylene glycol was loaded into the liquid cell. Nanoparticles started to nucleate and grow after the electron beam was casted on the thin liquid layer for several seconds, and four distinct stages of the structure evolution were observed. Initially, nanoparticles composed of iron and lead were produced and grew into 5 nm; secondly, phase segregation took place which led to the formation of nanoparticles with a lead core and a thin iron shell; thirdly, the iron shell was oxidized into orthorhombic Fe2O3 and sizes of the nanoparticles continue to increase. At the later stage, the Fe2O3 shell transformed into the cubic Fe3O4 phase under electron beam irradiation. During this stage, bubbles nucleated at the interface between the Fe2O3 shell and the Pb core. Bubbles grow and migrate rapidly in the shell and collapse on the surface. The rich dynamics observed by liquid cell TEM enlightens us the design of core/shell nanostructures with unique functions. We performed TEM experiments at Materials Science Division and National Center for Electron Microscopy (NCEM) of the Lawrence Berkeley National Laboratory, which is supported by the U.S. Department of Energy (DOE) under Contract No. DE-AC02-05CH11231. HZ thanks the funding support from U.S. DOE Office of Science Early Career Research Program.

The microstructure and stability of Cu/Ta interface has attracted recent large-scale molecular -dynamics simulations using different interatomic potentials [1,2]. For technological applications, this system has received great attention due to its use in Cu interconnects, Ta being deposited as a diffusion barrier. In particular, the thermal stability of very thin layer of Cu grown on Ta have experimentally shown a large and unusual mass transport through solid state dewetting. Our case study concerns Cu on Ta[001] substrate: we evidenced by solid state dewetting 5 nm of Cu film in UHV environment on Ta[001], the unique formation of epitaxial Cu[001] crystallites. In this work, microstructure of the interface and within the deposits are computed using the Cu-Ta interatomic potentials proposed in the literature[1,2]. It is found that the first layer of Cu is pseudomorphic with the bcc Ta. The following epitaxial relationship are observed for the remaining structure Cu(001)//Ta(001) and Cu[110]//Ta[100] using data from [1] and in disagreement with [2]. The transition between the bcc and fcc structure is very sharp, over 1-2 planes of Cu. The interfacial properties and structure are computed between 0K and 1000 K. The simulations results are compared and discussed with experimental characterizations made by cross-section High Resolution TEM and with recent coherent X-Ray diffraction on individual crystallites of Cu [3] which allow us to measure the residual 3D strain field. [1] A.Hashibon, A.Lozovoi,Y.Mishin, C.Elsasser, P.Gumbsch, Phys.Rev.B 77, 094131 (2008) [2] I.Lazic, P.Klaver, B.Thijsse, Phys.Rev.B 81,045410 (2010) [3] G.Beutier , M.Verdier, G.Parry et al.,Thin Solid Films in press (2012)

With moderate electrical conductivity and dielectric properties, Fe3+[Ni2+Fe3+]O4 (NFO), with an inverse spinel structure and room temperature ferrimagnetic behavior is promising for high frequency microwave applications and spintronic devices. As a magnetic material, NFO has an energetically-favored crystallographic direction, the easy axis, which is the preferred direction of domain magnetization. In textured NFO ceramics, the easy-axis direction in all grains is approximately parallel. As a result, the textured ceramics show magnetic anisotropy and it is possible to maximize the magnetic anisotropy by engineering the crystallographic texture along the easy-axis direction. In this study, chemical solution processing is employed to grow highly (111)-textured, polycrystalline NiFe2O4 (NFO) on (111) platinized Si and c-plane sapphire substrate. In the former, the grown films show complete out-of-plane texture in the (111) direction with random in-plane orientation (uniaxial texture), while in the latter, (111) out-of-plane and (110) in-plane orientation is achieved. theta;-2theta; X-ray diffraction, X-ray pole figures and electron diffraction are used to study film texture. It is shown that texturing initiates by nucleation of (111) planes at the film/substrate interface and decreases with increasing film thickness. As the NFO magnetic easy-axis is <111>, the out-of-plane magnetization exhibits improved Mr/Ms and coercivity with respect to randomly oriented films on silicon substrates. References: 1. E. P. Wohlfarth and K. H. J. Buschow, Ferromagnetic materials : a handbook on the properties of magnetically ordered substances (North-Holland, Amsterdam ; Oxford, 1980). 2. U. Lüders, A. Barthélémy, M. Bibes, K. Bouzehouane, S. Fusil, E. Jacquet, J. P. Contour, J. F. Bobo, J. Fontcuberta, and A. Fert, Advanced Materials 18, 1733 (2006).

Liquid metal embrittlement (LME) is the phenomenon of reducing the tensile strength when a metallic material is stressed in contact with a liquid metal. In spite of decades of studies, by experiments, thermodynamical analysis and lately by ab-initio calculations, this phenomenon is still fundamentally not understood. In particular, its occurrence and its intensity in a given Solid Metal/Liquid Metal (SM/LM) are still not well understood. In most cases, the embrittlement switches a normally ductile fracture to an intergranular brittle fracture, suggesting the importance of the grain boundaries (GB), as observed for some SM/LM couples in which atoms of LM are segregated at GBs of the SM (a few atoms up to 2 mono layers in the Cu/Bi system). The presence of LM in SM GBs could be the microscopic cause of the LME. We have studied the LME phenomenon for the couple Cu/Hg from an experimental and computational point of view. We compared the LME behavior of standard high purity OFHC copper with GB-engineered copper (containing a high fraction of sum;3 GB in the specimen). Experimentally the special sum;3 GB in copper are less prone than general GB to LME by liquid mercury. We have investigated the difference in LME induced fracture between the symmetric sum;3(111) and the symmetric sum;5(210) GB by ab-initio calculations. While the special sum;3 is a low-energy GB, the sum;5(210) has an energy representative of a general GB. The key quantity of such modeling is to evaluate in a Griffith approach the energy difference γGB - 2 γSL . This informs us on the trend for a grain boundary to break into two SM/LM interfaces. The SM/LM interfaces are Cu(111)/Hg and the Cu(210)/Hg modeled using ab-initio molecular dynamics. In addition, γGB is determined for different values of grain boundary coverage Theta in accordance with experimental observation of LM segregation at GB. The results of these calculations from the sub-monolayer level to the full solid/liquid interface will be presented and compared with the experimental trend. Finally, ab-initio tensile tests on the sum;3(111) and the sum;5(210) GB with(out) LM atoms segregant were performed. The calculated applied fracture stress on GB will be compared with the results of mesoscopic modeling in a typical copper microstructure during the fracture process. The difference in magnitude between the two types of modeling will be discussed to evaluate if ab-initio tensile tests are relevant for grain boundary LME.

Recent interest in exfoliated single layers of layered transition metal dichalcogenides (LTMDs), often referred to as inorganic analogue of graphene, have opened up new prospects for the technological applications of 2D crystals. Polymorphism is one of the unique features of LTMDs that arises from the structural instability induced by electron doping. For instance, MoS2 and WS2 exhibits stable 2H polymorph as well as metastable 1T phase. Our recent studies indicate that two phases coexist in single layers of MoS2 and WS2 when exfoliated via lithium intercalation [1]. Here we report direct high resolution HAADF-STEM observation of chemically exfoliated MoS2 and WS2 single layers [2]. We demonstrate that the exfoliated sheets consist of electronically dissimilar polymorphs that are lattice matched such that they form coherent interfaces with each other. Our STEM images reveal the co-existence of metallic (1T) and semiconducting (2H) phases in nanometer-sized domains. We also report observation of 2a x a superlattice due to Jahn-Teller distortion and microscopic ripple structure which is responsible for the stability of 2D materials. [1] G. Eda, H. Yamaguchi, D. Voiry, T. Fujita, M.W. Chen, M. Chhowalla, Nano Lett. 11, 5111 (2011). [2] G. Eda, T. Fujita, H. Yamaguchi, D. Voiry, M.W. Chen, M. Chhowalla, Under review.

Advanced nuclear energy systems will require materials that perform for extended periods of time under conditions of elevated temperature, corrosion, and constant irradiation. Radiation induced segregation and the formation of chromium rich carbides along grain boundaries has been shown to be a key factor in irradiated assisted stress corrosion cracking in Fe- and Ni-based alloys. In order to develop new materials that are radiation tolerant, a systematic study is needed to determine the fundamental response of solute segregation under high temperature irradiation at specific grain boundary structures under these extreme environments. This research focuses on understanding the response of specific grain boundaries (misorientation and grain boundary plane) under high temperature and ion irradiation. 316L stainless steel and a model stainless steel alloy were thermomechanically processed to induce a high fraction of twin and twin variant grain boundaries. After thermomechanically processing to induce a range of grain boundary types, heavy ion irradiation was completed at high temperature. Following the irradiation, a grain boundary specific FIB-EBSD preparation method was employed to investigate the interaction of irradiation induced defects and solute segregation with specific grain boundaries by STEM-EDS, HR-TEM, and atom probe tomography. Our results indicate variation exists in radiation induced solute segregation with both grain boundary misorientation and grain boundary plane normal for a range of boundary types. In particular, irradiation solute segregation and non-equilibrium thermal segregation varied most dramatically for changes in the inclination angle for Sigma 3 grain boundaries. Overall, we report on the dependency of solute segregation under ion irradiation with grain boundary structure and indicate key variations that illustrate the need to understand the full grain boundary character distribution under extreme environments. This work was supported by the Division of MSE, Office of BES, U.S. DOE. Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin Corporation, for the U.S. DOE&’s NNSA under contract DE-AC04-94AL85000.

Metal oxide nanoparticles such as SiO₂, ZnO or TiO₂ are established as commodity materials while current research discovers continuously new applications in catalysis, sensors or food supplementation to name only a few. For example TiO₂ nanoparticles are a promising material for renewable energy applications in photovoltaics and photocatalytic processes. The performance of these particulate materials depends considerably on the primary particle size, aggregation and crystal structure. Gas-phase synthesis allows making these materials economically in large quantities with close control of the properties mentioned above considerably determined by particle-particle sintering. Theoretically derived sintering rates were found to be in agreement with experimental observations for big nanoparticles (d_p = 10 - 100 nm). However little is known for the sintering rate of small TiO₂ nanoparticles (d_p < 10 nm) although this would be the key size range where nanoparticles begin to exhibit their extraordinary performance and exciting new properties. Here1, sintering of two touching rutile TiO₂ nanoparticles (d_p < 5 nm) has been investigated by classic molecular dynamics simulations using a force field with constant charges to close the gap of knowledge between molecular clusters and bigger nanoparticles. The MD was accelerated with GPUs on a low cost consumer graphics card and a modified version of the MD code hoomd blue 0.8. This combination made it possible to reach required residence times until full coalescence within a feasible amount of time. Observations of the atom trajectories reveal that surface atoms exhibit a much higher mobility than bulk ones indicating that sintering by surface diffusion dominates at these particle sizes and temperatures. The simulations provide also a detailed insight into the atomistic structure and evolution of the contact point between the two particles and especially the grain boundaries that begin to develop there depending on the initial orientation of the crystal planes of the two sintering particles. The sintering rate is determined by calculating the surface area evolution representing the ions of the particles as overlapping spheres. For the smallest particle diameters, the MD-obtained sintering rates were smaller than that predicted by theory and experimental observations for larger particles (d_p > 10 nm) converging however to that theory around 5 nm. An expression for the sintering rate of rutile TiO₂ nanoparticles as function of primary particle size and temperature has been extracted from the MD data, bridging the gap of knowledge from a few molecules to several nanometers, the key size range for nanoparticle properties and performance. This sintering rate facilitates the design of large scale manufacture of such small nanoparticles based on phenomenological models or allows engineering estimations of the particle morphology. 1. Buesser B, Gröhn AJ, Pratsinis SE, J. Phys. Chem. C 2011, 115, 11030

Lead Titanate (PbTiO₃) is a representative ferroelectric material with large tetragonality and polarization. 90° domain configuration is the key issue for ferroelectric and piezoelectric properties of this material, and the configuration can be tuned] by the film thickness via the residual strain. We have elucidated the nucleation and growth mode of the 90° domains from the viewpoint of misfit dislocations [1]. However, crystal lattices around the 90° domain walls are heavily strained owing to the lattice correspondence between the adjacent domains. Particularly, the lattice constraint along the bottom boundary of the 90° domain (a-domain), i.e. c-axis of tetragonal lattice, is much larger than that of the a-axis in the SrTiO₃ (001) substrate. The purpose of the present study is to elucidate the atomic arrangement around the 90° domain walls in the PbTiO₃/SrTiO₃ thin films. The films were epitaxially grown on the SrTiO₃ (001) substrates by pulsed-MOCVD method. Atomic resolution domain wall analysis was conducted by an aberration corrected HRTEM (TITAN80-300, 300kV, FEI) and an aberration corrected STEM (JEM-ARM200F Cold FEG, 200kV, JEOL). TEM and STEM observations show that the films typically possess 90° domain configurations with misfit and threading dislocations. Some 90° domains penetrate the film towards the surface, while another domains are terminated inside the film and are short in length. The 90° domain wall has a finite width, in which the lattices of tetragonal symmetry lower the symmetry owing to lattice deformation. Atomic resolution HRTEM images show that the atomic displacement drastically rotated just on the centerline of the domain wall. This result also indicates the fact that the spontaneous polarization can rotate exactly at the center of the domain wall. On the other hand, the bottom of the 90° domain need to contact with the SrTiO₃ substrate, although there is a lattice mismatch as large as about 6% at room temperature. ABF/HAADF-STEM images clearly show that the bottom of the 90° domain contacts the substrate at the corner of the domain just on a pair of partial edge-dislocations with b=1/2[-202]. Furthermore, the bottom structure is composed of plates of stacking faults between the dislocations. Thus, it is found that the domain wall structure at the bottom end of the a-domain is complex in order to relax the large lattice mismatch. [1] T. Kiguchi, K. Aoyagi, Y. Ehara, H. Funakubo, T. Yamada, N. Usami, and T.J. Konno, Sci. Technol. Adv. Mater. 12, 034413 (2011).

A first principles calculation is used to study the electronic structure of AgTaO$_3$/SrTiO$_3$ (001) heterostructure. Formation of a two-dimensional hole gas at the (AgO)/(TiO2) p-type interface is presented. The Ag 4d states strongly hybridize with the O 2p states and contribute to the hole gas. It is demonstrated that the holes are confined to an ultra thin layer (~4:9 A) with a considerable carrier density of ~ 10^{14} cm^{-2} . We estimate a hole mobility of 18.6 cm^2 V^{-1} s^{-1}, which is high enough to enable device applications.

The electronic and magnetic properties of the cubic pyrite CoS2 /FeS2 interface are studied using the all-electron full-potential linearised augmented plane wave method. We find that this contact between a ferromagnetic metal and a non-magnetic semiconductor shows a metallic character. The CoS2 stays close to half-metallicity at the interface, while the FeS2 becomes metallic. The magnetic moment of the Co atoms at the interface slightly decreases as compared to the bulk value and a small moment is induced on the Fe atoms. Furthermore, at the interface ferromagnetic ordering is found to be energetically favourable as compared to anti-ferromagnetic.

Characterizing the interplay between grain size and mechanical strength of polycrystalline cubic-zinc sulfide (c-ZnS) is of vital importance for developing c-ZnS based infrared windows with superior rain and sand erosion resistance. Towards this end, using molecular dynamics (MD), we characterize the impact of grain size on the mechanical strength of polycrystalline c-ZnS, with a special emphasis on examining nanometric grain sizes, which based on the Hall-Petch relation, can be expected to result in a marked increase in the material's yield strength (and consequently hardness). In this work we employ the bond-order Tersoff potential to represent interatomic interactions. To characterize the Hall-Petch relations of c-ZnS, we utilize multimillion-atom MD simulations of samples with pre-existing nanocracks and grain sizes ranging from 3-22 nm which are subjected to uniform uniaxial strain-rates. We observe distinct Hall-Petch and inverse Hall-Petch regimes and the transition is a pronounced function of nanocrack dimensions and size. Further, the grain size impacts crack propagation leading to either intragranular (< 17 nm) or intergranular (17-22 nm) failure.

The oxidation of metals plays a critical role in many technologically important processes including high-temperature corrosion, heterogeneous catalysis, and fuel reactions. Oxygen surface chemisorption is believed to occur at the very initial stage of the oxidation reaction, which induces consecutive surface reconstructions as the exposure to the gaseous environment changes. Thermodynamic data of surface phase transitions has been firmly established via extensive theoretical modeling. However, the kinetic picture of how the phase transitions occur is still missing to a large extent, which makes the interpretation or prediction of surface oxidation solely using the thermodynamic data incomplete and sometimes invalid. For example, significant discrepancy between equilibrium phase diagram predictions and experimental realities can be frequently observed. Therefore, the knowledge of the dynamical evolution at the atomic scale is highly desirable. In this work, using first-principle molecular dynamics (FPMD) and density functional theory (DFT), we investigate the kinetics of the phase transition on the Cu(110) surface induced by oxygen chemisorption. Our results reveal that considerably large energy barriers exist for the (2×1) phase to c(6×2) phase transition, and the reaction path of this transition is also demonstrated. We expect that our results can be further utilized to dynamically simulate the early-stage process of metal oxidation. Furthermore, the insights and approach obtained from this study can be extended to understand the microscopic processes of oxygen chemisorption-induced surface restructuring during the oxidation of transient metals.

Zinc oxide has long been utilized as a suitable material for varistor devices, which protects electric circuits with the voltage-dependent resistance. It is thought that this kind of characteristic is generated by double Schottky barrier formed at the grain boundaries (GBs) due to the presence of dopant elements. Therefore, it is important to evaluate the atomic-scale structures in order to understand the structure-property relationship at GBs. As a simple approach to observe GB structures, ZnO single GBs have been fabricated by joining two single crystals. Here, we have investigated [0001](13-40) Σ13 ZnO symmetric tilt GBs doped with/without doping Pr (praseodymium), which is a typical dopant element used for ZnO varistor. The GB structure has been observed using conventional TEM and aberration-corrected STEM. HAADF (High-angle annular dark field) and ABF (Annular bright field) modes have been utilized for STEM imaging. Moreover, first-principle calculations were carried out to obtain stable GB atomic arrangements. It was found that the atomic arrangement of the undoped ZnO Σ13 GB can be described by the array of SUs (structure units). Here, two types of SUs that we call SU C and D are observed. On the other hand, we found that Pr doping induces structural transition from the undoped case. Transition of SUs C and D to SUs A for the Pr-doped case occurs. Our close inspection revealed that Pr atoms selectively segregate at Zn sites with largest space, and this would be due to the larger ion size of Pr than that of Zn. Our extensive study is currently on going, and we suggest that a series of ZnO GB structure may be predicted. Further discussion will be given in my presentation.

Spontaneous and piezoelectric polarizations in III-nitride semiconductor are crucial in optoelectronic and electronic devices. In the case of GaN-based high electron mobility transistors (GaN-HEMTs), the threshold voltage and on-current can be controlled by adjusting the amount of polarization differences between GaN and AlGaN layers. The accurate prediction of polarization related properties has been demonstrated from first-principles calculation using the polarization theory based on the Berry-phase. Although the polarization theory was successful for predicting polarization properties of bulk III-nitride semiconductors, it is not straightforward to obtain atom-resolved local polarization, which is essential to investigate the interface polarization and impurity effect. As solutions to this problem, we suggest a method to calculate local electric polarization from Wannier functions with no need for any high-symmetry reference system. Calculated electric polarizations of bulk AlN, GaN and InN with our local polarization method show good agreements with previous reports. We also reveal calculated electric polarization in the interface between AlN and GaN layers and the effect of atomic impurity on electronic polarization of neighboring atoms. Our atom-resolved local polarization method enables accurate predictions of 2-dimensional electron gas (2DEG) density induced at AlN/GaN hetero-junction interfaces and spontaneous polarization at nano-structured materials.

The surface morphologies of AlGaN/AlN/GaN HEMT structures were examined by using the atomic force microscopy (AFM).These HEMT structures have been grown by metalorganic chemical vapor deposition (MOCVD) onto the sapphire (0001) substrates where the thicknesses of AlN interlayers were varied from 0 to 5 nm. After the growth of GaN buffer layer, the AlN intermediate and AlGaN top layers were subsequently deposited at 1130 degree C in an trimethylaluminum(TMAl) and ammonia (NH3) atmosphere. The surface of AlGaN layers shows the thick- and fine-thread patterns. It was found that the root-mean-square (RMS) roughness of the samples with 0, 0.5, 2.5 and 5nm AlN interlayer thickness are 0.503, 0.534, 0.534 and 0.601nm, respectively. Although the cracks and rougher surfaces show that the qualities of AlGaN barrier layers have slightly degraded in the samples with thicker AlN layer. These phenomena could be attributed to the lattice mismatch and the growth temperature. In addition, the room-temperature 2DEG mobility and density analysis were performed on the AlGaN surfaces and the measured results were discussed in detail.

Perovskite oxides such as barium-strontium titanate, lead zirconate-titanate, or potassium-sodium niobate are technologically important and scientifically exciting materials. Many perovskites are ferroelectric and possess high dielectric constants, which make them attractive for modern microelectronic applications. A successful implementation however requires not only information about the intrinsic functional properties but also a thorough understanding of crystal defects that influence the functionality as well as the mechanical and structural stability of such components. In the present work we investigate the properties of dislocations in perovskite materials by means of atomistic simulations, using both accurate first-principles calculations (density functional theory) and computationally efficient atomistic simulations with classical interatomic potentials [1]. In paraelectric strontium titanate the structures of dislocation cores are analyzed and compared to high-resolution transmission electron microscopy observations. The calculated Peierls energies and stresses for these dislocations under different applied loads give insight into their mobilities and are related to the macroscopic mechanical behavior. In ferroelectric perovskites we study the motion of a domain wall under an applied electric field, and the pinning by a dislocation at the atomic scale. [1] P. Hirel, M. Mrovec, C. Elsaesser, Acta Mater. 60 (2012) 329

The local magnetic moments were found to be a linear function of 3d L3/L2 white line ratio in electron energy loss spectroscopy (EELS) [1]. We have applied this TEM/EELS technique to measure the local magnetic moments at the grain boundaries in iron and found that the local magnetic moments were increased up to 2.63 mu;B at grain boundaries according to grain boundary character defined by the orientation relation between two adjoining grains. However, the local magnetic moment obtained from the grain interior of Ni was 1.03mu;B, which is much higher than that of theoretical one. This is probably due to significant enhancement of the L3/L2 white line ratio for the materials with transition metal atoms having 3d counts exceeding nine like Ni [1]. Thus, one of motivations of this study is to find a new relation between the magnetic moment and the L3/L2 white line ratio for Ni using some alloys and compounds containing Ni atoms. In addition, the local magnetic moments of well-characterized grain boundaries in pure Ni will be evaluated using the new function obtained. We obtained a new liner function between the magnetic moments and the white line ratio for Ni. Here, the theoretical magnetic moments of materials (Ni, NiAl, Ni3Al) subjected to TEM/EELS measurement were obtained by the ab-initio calculation using the VASP code. The average magnetic moment measured in the grain interior of nickel was 0.58 mu;B. This is in good agreement with the saturate magnetic moment of nickel of 0.62 mu;B. Of particular importance is the find that the magnetic moments are higher at the grain boundaries than in the grain interior, and this enhancement at grain boundaries depended significantly on the misorientation angle of grain boundaries. The local magnetic moments increased up to approximately 1.0 mu;B as the misorientation angle increased, and showed a maximum around the 50°. The enhancement of the local magnetic moments at Σ5 (0.68 mu;B) in nickel was approximately 17%, in comparison with the grain interior. The ab-initio calculation revealed the local magnetic moment of atoms in the (210) Σ5 grain boundary plane being 0.68 mu;B, which is approximately 10% larger than the value of the grain interior. Furthermore, the average of local magnetic moments at the (111)Σ3 grain boundary, that is the coherent twin boundary, in nickel was found to be 0.58 mu;B, being almost the same as that of the grain interior. #268;ák et al.[2] showed by ab-initio calculations that such increase in the local magnetic moments at grain boundaries are closely related to the atomic free volume at grain boundaries. There is no volume change at (111) Σ3 grain boundary compared with the grain interior. This is a possible reason for non-enhancement in the local magnetic moments near (111) Σ3 grain boundary.[1] D. M. Pease, A. Fasihuddin, M. Daniel and J. I. Budnick: Ultramicroscopy, 88 (2001).[2] M. #268;ák, M. Scaron;ob and J. Hafner: Phys. Rev., B78 (2008), 054418.

The role of defects is a focus of the ongoing discussion about the electronic properties of the conducting interface between the two insulators LAO and STO. Besides the model of charge transfer due to the polar nature of LAO, it is generally accepted that defects in vicinity of the LAO/STO interface can have a large impact on the electronic properties. At the extreme, the interdiffusion of La-ions on Sr-sites, and the creation of oxygen vacancies within the STO lattice, have even been considered as sole origin of the interface conduction. In this study, the LAO/STO-interface will be discussed from a defect chemical point of view. It will be shown that cation vacancies in the STO lattice, i.e. Sr-vacancies and Ti-vacancies, have to be considered in order to draw a complete picture of the LAO/STO-interface. The conducting LAO/STO-interfaces are fabricated by pulsed laser deposition (PLD) of LAO on TiO2-terminated STO single crystal substrates, and by the growth of LAO and STO on LSAT single crystal substrates. The defect structure of the LAO/STO-interface is investigated by means of high temperature conductance measurements in equilibrium with the surrounding atmosphere (HTEC). In the investigated temperature range (950K-1100K), the decisive defect equilibria at the LAO/STO-interface are activated, and hence, strive for a well-defined thermodynamical equilibrium state which is related to the ambient oxygen pressure (pO2). The resulting conductance characteristics contain information about the chemical reactions, which take place at the LAO/STO-interface, and the corresponding defect concentrations. The results will be discussed in terms of the defect chemistry model of doped STO. Using this method, it can be excluded that mobile oxygen vacancies are the sole origin of the conducting interface [1]. In fact, the temperature- and oxygen partial pressure independent plateau region, which is found for reducing atmospheres, indicates the presence of immobile interfacial donor states. In addition, HTEC measurements on LAO/STO systems grown on LSAT substrates reveal a decrease of the interfacial electron density proportional to pO2^(-1/4) for oxidizing conditions. This indicates the formation of Sr-vacancies, which act as acceptor-type electron traps in the vicinity of the LAO/STO-interface [2]. [1] F. Gunkel et al., Appl. Phys. Lett. 97, 012103 (2010). [2] F. Gunkel et al., Appl. Phys. Lett. 100, 052103 (2012).

The percolation threshold in a ceramic composite depends on the processing conditions used to make them along with the size and shape of the filler. In this study, borosilicate glass microspheres were used as the matrix material and antimony tin oxide (ATO) was used as the filler. The microsphere/ATO composites were fabricated by hot pressing using a method that allowed the ATO to be confined to the spaces between the glass particles, forming percolating networks at low volume fractions of the filler. Prior to hot pressing, the electrical response of the powder mixtures was also evaluated using a custom set-up and die, as a function of the applied pressure. By comparing the results of these two methods, a correlation of the microstructures obtained and the ac conductivity determined from impedance spectroscopy allow for valuable insights in structure-property-processing relationships in these materials.

Metals-polymer interactions are important for coatings, catalysis, and other applications involving a heterogeneous interface between the two dissimilar materials. One method for producing polymer-metal interfacial structures is through vapor-phase deposition of one material, typically the metal, onto the other material. The properties of the final material depend on the form of the metal being deposited (e.g., atoms versus clusters), their incident energy, and the type and structure of polymer used as the substrate. In the present work, we examine the deposition of Cu atoms and clusters on amorphous and crystalline polymethylmethacrylate (PMMA) and polystyrene (PS) surfaces. The objective is to determine the way in which the interactions of the metal and the polymer are influenced by an oxygen pre-treatment of the polymer, the form of the Cu, and the incident energy of the Cu. In particular, energies of 5, 10 and 20 eV/atom are considered. The approach is molecular dynamics simulations using Charge Optimized Many Body (COMB) potentials and Density Functional Theory calculations. This work is supported by the National Science Foundation (grant CHE-1005779).

Phase change materials, mainly Te-based chalcogenide alloys have been extensively used to apply for non-volatile memory device. The weakness of materials based on the phase change phenomena is that the switching bit suffers from stress caused by density difference between the amorphous and the crystalline states. In this study, the microstructural evolution of the Ge2Sb2Te5 film sandwiched by SiO2 dielectric layer was investigated with annealing process of fast ramping rate (1140 oC/min) and near melting temperature using X-ray diffraction, Raman and TEM. In addition, the deformation of the FCC structure in Ge2Sb2Te5 film was estimated under triaxial tensile stress based on the density functional theory and Birch-Murnaghan equation of state(EOS). The FCC structure undergoes phase separation and irreversible deformation due to the decreasing viscosity and tensile stress, which effectively compensates the densification. However, the presence of trigonal structure in Ge2Sb2Te5 film dominantly induces void nucleation and growth because of the broken Van der Waals bonding.

The kinetics of a solid-liquid interface of iron with a triple point is investigated by performing a large-scale molecular dynamics simulation on a graphics processing unit (GPU). We have developed our own code for carrying out MD simulation on a GPU [1] and achieved about 100 times the speed of our previous code developed for a single CPU calculation [2,3]. The faster code enables the handling of a million atoms in MD simulations over a period of nanoseconds with ease. Using this code with a GPU, the evolution of the grain boundary groove is examined by considering the relaxation of a solid-liquid biphasic system with a triple point, where the grain boundary is in vertical contact with the solid-liquid interface at a temperature below the melting point (i.e., undercooling). It is confirmed that a grain boundary groove evolves at the triple point of a solid-liquid interface with a bccΣ3(111)<110> tilt grain boundary during solidification, whereas a solid-liquid interface with a twin boundary of small grain boundary energy (bccΣ3(112)<110> tilt grain boundary) is almost planar except in the vicinity of the triple point. This agrees with the experimental observation and its conventional interpretation that the balance between the grain boundary energy and solid-liquid interface determines the structure of the grain boundary groove [4]. [1] K. Oguchi, Y. Shibuta, T. Suzuki, J. Jpn. Inst. Metals, article in press. [2] S. Tateyama, Y. Shibuta, T. Suzuki, Scr. Mater. 59 (2008), 971. [3] Y. Shibuta, T. Suzuki, J. Chem. Phys, 129 (2008) 144102. [4] Y. Shibuta, K. Oguchi, T. Suzuki, ISIJ Int. submitted.

Piezoelectric AlN films are remarkable for their application as film bulk acoustic wave (BAW) resonators. The performances of BAW devices are enhanced where piezoelectric AlN films are (001) preferred orientation. Thus, various sputter deposition methods have been investigated and it is concluded that (001) textured AlN films without an incorporation of oxygen exhibit high piezoelectric response. In this study, highly (001) textured AlN film has been fabricated using a continuously deposited 3-nm thick Co(Pt) layer. Since the insertion of the ultra-thin Co(Pt) layer is conducted without the interruption of deposition, the interface of AlN/Co(Pt) is considered as oxygen-free. According to cross sectional high resolution transmission electron microscope images, there is no amorphous AlN at the AlN/Co(Pt) interface, suggesting the incorporation of oxygen is reduced using the continuous sputter deposition. This results in a direct growth of AlN crystals with c-axis preferred orientation on Co(Pt)(111). The effect of the Co(Pt) layer insertion is considerably strong and it enables us to fabricate (001) textured AlN film under the deposition condition where polycrystalline growth is preferred. The preferred orientation of AlN film with the Co(Pt) layer was further examined using X-ray diffractometer equipped with a 2-dimentional detector. The diffracted intensities are collected as a function of (2theta;, psi;) where 2theta; is a diffraction angle and psi; is an angle between substrate normal and diffraction plane normal. It is demonstrated that the c-axis preferred orientation of AlN film is considerably strong since hkls including non 00l such as 101, 102 and 103 are observed as spots. X-ray rocking curve full width at half maximum of 002 is 2.7° and such AlN film may exhibit high piezoelectric response. The residual stress in the (001) textured AlN film is estimated using the sin2psi; method as 1.6 GPa (tensile). It indicates that the film is fabricated at the sputter deposition condition which is not suit for (001) deposition. AlN film without the insertion of Co(Pt) layer was also analyzed and it was revealed that the film is almost polycrystalline, however, it has a weak texture. Note that such texture cannot be detected using a conventional x-ray diffraction profile.

Motivated by the promise of using grain boundary (GB) engineering to increase intergranular fracture toughness in polycrystals, we investigated fracture along several tilt GBs in nickel via atomistic simulations. By analysis of atomic debonding process at the crack tip, we found that GB may induce a change in fracture mode from brittle cleavage to crack propagation with copious dislocation emission, compared with that in the corresponding single crystal. We interpreted the GB-induced transition based on Rice and Thomson&’s model and found this transition may come from the change in local stress intensity at the crack tip caused by the GB misfit dislocations as well as the emitting dislocations. This work was funded by the BP Materials and Corrosion Center at MIT.

Cubic boron nitride (c-BN) is known for its outstanding mechanical properties, thermal properties, and wide band gap. Superior hardness makes c-BN the material of choice for bullet-proof vests, nuclear reactor coatings, and industrial cutting tools. Alternative methods are described to synthesize nanocrystalline c-BN, starting with an ammonia-borane precursor. In the first method, the precursor is heat treated in a controlled environment, forming a high-surface-area nanopowder. Subsequently, the nanopowder is consolidated by a novel high-pressure sintering method to yield nanocrystalline material. In the second method, the precursor is fed via a bubbler system into a high-enthalpy plasma, forming vaporized species that upon rapid quenching also yield a high-surface-area nanopowder. Similarities and differences between these two types of nanopowders are described, as well as their responses to pressure-assisted sintering, taking into consideration variations in processing parameters. It is shown that nanopowders with metastable structures can be generated, which facilitate densification during high-pressure sintering. Results indicate that mixed hexagonal- and cubic-BN nanopowder can be consolidated at pressures and temperatures below that normally required for processing c-BN. Major technical issues associated with sintering are addressed, such as elimination of residual porosity and exaggerated grain growth. Various analytical techniques are used to determine mechanisms and kinetics of phase transformations during nanopowder synthesis and consolidation. In addition, nano-indentation testing is used to determine the dependence of hardness on grain size in the nanoscale range - data that is currently lacking.

The nanoscale structure of metal surfaces is an important factor in determining catalytic behavior. Rhodium is a versatile metal catalyst with application in CO-oxidation, NO-reduction, hydrogenations, electro-oxidations, and hydroformylation reactions. Controlling the structure of metal nanocrystals comprising this precious metal remains a challenging undertaking, despite the success that has been achieved with the more noble metals Ag, Au, Pd, and Pt. The difficulty in control of Rh can be attributed to its less positive reduction potential, which typically requires the use of stronger reducing agents and/or higher temperatures. We report the controlled synthesis of bimetallic nanocrystals incorporating rhodium using halides and shape-controlled metal seeds to direct the rhodium overgrowth. The growth obtained is epitaxial regardless of the seed crystal's size, morphology, and identity. The halide ions were found to play a crucial role in the growth mechanism. The size of the overgrowth can be controlled by the amount of Rh added to the growth solution. The resulting Pd-Rh NCs show promising activity for electro-oxidation of formic acid and hydrogenations in the gas phase. The Rh epigrowth can also be used to tune the surface plasmon resonance of Au seed crystals. This represents a powerful new method for control of rhodium and allows for the synthesis of equally novel nanostructures for a variety of potential applications.

Efforts of many scientists for more than half a century have resulted in a substantial understanding of the response of various materials to irradiation. However, the accumulated knowledge accumulated and theoretical developments have not led to a proper understanding of damage accumulation processes to help develop damage resistant materials. The basic reason for this has been the assumption that the primary damage due to fast particle irradiation, regardless of their nature and energy, consists of single vacancies and self-interstitial atoms (SIAs), both diffusing three-dimensionally. This has led to development of numerous models. However, none of these models have been able to predict successfully the damage accumulation as a function of crystal structure and material composition. Moreover, the models predict damage accumulation to be quite similar in crystals of different symmetry which contradicts experimental observations. For example it is well established that BCC materials swell much less than FCC and only HCP crystals exhibit radiation growth. In this report we shaell try to rationalize the situation taking into account the more realistic nature of primary damage produced by different particles, properties of dislocations and diffusion mechanisms of defects produced via irradiation in crystals of different symmetry. It is shown that the primary damage is more complicated than originally assumed. The nature of the primary damage assumed within the framework of Production Bias Model leads e.g. to the same maximum swelling rate of ~1%/dpa in all metallic cubic crystals and radiation growth rate in HCP crystals. It will be further shown that the diffusion properties of defects and their clusters in crystals of different symmetries are responsible for the difference in damage accumulation due to irradiation. Finally, a unified framework for the theory of radiation damage in solids is discussed. This produces a sound basis for designing radiation damage resistant materials.

Grain boundaries in materials can serve as defect sinks for absorbing radiation-induced defects. However, different grain boundaries may interact with defects differently. Here we use atomistic simulation methods such as molecular dynamics and molecular statics to investigate the role of grain boundary structure on affecting the defect-boundary interaction during irradiation. We use copper and uranium dioxides as model systems for metals and oxides respectively. Several representative grain boundaries are investigated such as twin, twist, tilt, asymmetric tilt, and “general” grain boundaries. In copper, we investigated how defects interact with pristine boundaries, how boundaries modify defect production during cascade damage, and how defects interact with damaged boundaries. We found some similarities between different boundaries such as bias absorption of interstitials during cascade damage and subsequent enhanced interaction with vacancies. However, we also found some differences in defect-boundary interaction between different boundaries that are related to the boundary structures. In uranium oxides, we primarily investigate how defects interact with pristine boundaries and how boundaries modify defect production. Again, we found similarities and differences between different boundaries. The defect-boundary interaction behavior in metals and oxides is compared and discussed. This work is in collaboration with many researchers from the two Energy Frontier Research Centers (EFRCs) at Los Alamos National Laboratory and Idaho National Laboratory, both are funded by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences.

The interaction of radiation with materials controls the performance, reliability, and safety of many nuclear power system structural components. Improvements in radiation damage tolerance may be attainable if interfaces can be manipulated at the nanoscale to give optimal interface stability and point defect recombination capability. Ion irradiation experiments, a combination of physical measurements and atomistic modeling are being used to study the effect of misfit dislocation density on radiation tolerance of Cr_xV_1-x/MgO interfaces. Epitaxial films 100 nm thick of Cr, V, and their alloys are deposited on MgO(001) substrates by molecular beam epitaxy. By controlling the composition of the Cr_xV_1-x film, the lattice mismatch with MgO can be adjusted so that the misfit dislocation density varies over a wide range. A variety of damage states are produced by ion irradiation at room temperature using 1 MeV Au⁺ ions to doses ranging from 0.4 to 300 displacements/atom. High-resolution transmission microscopy and Rutherford backscattering spectroscopy (RBS) measurements are used to examine mixing, defect structures, defect distribution, and interface evolution following irradiation. The accumulation of damage in the film and the substrate has been investigated using RBS in channeling and random geometries. Results show that Cr_xV_1-x/MgO interfaces withstand high dose irradiation, and that the damage level decreases with increasing misfit dislocation density. Atomistic modeling is being done to simulate the interaction of point defects with the Cr_xV_1-x/MgO interface to better understand the early stages of defect generation and annihilation in and near the interface. This work is coupled with spatially dependent cluster dynamics methods to examine damage evolution on temporal scales more relevant to experimental observables. The integration of these methods and the results of the irradiation experiments are described.

Coherent x-ray Bragg projection ptychography (BPP) is a new lensless imaging technique in which the shape and distribution of local unit cell displacements in nanoscale domains (or crystals) can be mapped with ~7 nm in-plane resolution. When applied to thin film ferroelectrics, the lattice features imaged with BPP can be converted to a quantitative measure of local polarization within domains and across domain boundaries. This approach is complimentary to piezoresponse force microscopy (PFM), enabling the study of nanoscale ferroelectricity in terms of local unit cell structure and the corresponding piezoresponse. Here, we apply both techniques to investigate the stabilization of 180 degree stripe domains in a 27 nm thick epitaxial (001) PbTiO3 ferroelectric thin film grown on a 1 degree miscut SrTiO3 (001) substrate. Substrate miscut has been shown to influence domain behavior and morphology in ferroelectric thin films, and here we study the case in which the substrate terrace separation (~22 nm) is of order the equilibrium domain width. We find that the ferroelectric domain walls line up with substrate terraces edges and adopt their spacing, posing intriguing possibilities for control of ferroelectric domain boundaries and opening the door for in-operando study of domain wall dynamics and stabilization.

The multiple functional properties of transitional metal oxides are enabled by minute displacements of atoms from high symmetry positions, mesoscopically described by order parameter fields; examples include polarization in ferroelectrics, cooperative Jahn-Teller distortions, and octahedral tilt systems. The explosive growth in applications of functional oxide interfaces in the last decade has brought forth the challenge of understanding these behaviors and their coupling on a level of a single interface or domain wall. In this talk, I will illustrate the applications of local crystallographic mapping by scanning transmission electron microscopy for deciphering functional properties of oxide interfaces. For BFO-LSMO interfaces we observe exciting interplay between octahedral tilting distortions, dielectric properties, and polarization that in some cases can stabilize antiferroelectric form of BFO. This demonstrates a paradigm of octahedral tilt engineering of interface behavior, complementing the well-established strategies based on misfit strain and polarization. In the second part of the talk, I will discuss the role of oxygen vacancies in functional oxides. The intrinsic non-stoichiometry and high diffusional mobility of vacancies in oxides make them an active player in the physics of oxide interfaces. Using local crystallography approach, I will demonstrate that in some systems local vacancy concentrations can be determined quantitatively from lattice spacings, bringing the concept of chemical expansivity to the atomic level. Interestingly, vacancy dynamic and ordering can be described by the effective non-conserved order parameter fields, and the corresponding free energy parameters can be extracted from STEM data. In the final part of the talk, I will discuss prospects for decoupling physical and vacancy-mediated behaviors at interfaces via atomic-scale observations.

Interfaces with adjacent space charge layers play a major role in modifying the electrical properties of ionic solids. Defect transport properties across and along these interfaces can either be impeded or enhanced depending on the interface charges. The concentrations of point defects and thus the conductivity can be adjusted by several methods such as aliovalent doping, or introducing interfaces (grain boundaries, heterointerfaces). Line defects (with charged) cores represent an additional, intermediate possibility. Grain boundary cores of perovskite and fluorite structured oxides with large band gap such as SrTiO3, CeO2, ZrO2 (Y stabilized)[1, 2] are typically positively charged due to the presence of excess anion vacancies. Positive grain boundary cores impede the transport of positively charged defects like holes, cation interstitials and anion vacancies by formation of depletion layers. However, in case of TiO2 also negative core charges may be formed[3]. For this reason, we decided to investigate the effect of line defects on conductivity in TiO2 rutile. Line defects (dislocations) are created in [001] and [110] oriented single crystal TiO2 by uniaxial compression at elevated temperature. TEM revealed that the dislocations favorably lie on {110} slip planes irrespective of the compression axis. Conductivity was measured in [001] (along the slip planes) and in [110] (perpendicular to the slip planes) as a function of oxygen partial pressure (pO2) and temperature. For the [001] sample, the increased conductivity at high pO2 (10-5 - 1 bar) with decreased pO2-dependence is assigned to an enhanced ionic conductivity due to positive carriers. In contrary, no changes were observed in case of the [110] sample despite the presence of dislocations. At very low pO2 (i.e. in n-type regime), the conductivity remained same as for samples without dislocations. The enhancement in conductivity and variation in the point defect concentrations can be explained by negatively charged dislocation cores and adjacent space charge layers. The proposed space charge model also explains the unchanged conductivity in the n-type regime. Further, the nature of the positive defects contributing to the ionic conductivity will be discussed based on oxygen tracer exchange and SIMS analysis. The present observations on TiO2 suggest that the formation of line defects by mechanical effects can be an alternative method to tune the electrical properties of ionic solids. The effects strongly depend on the sign of the dislocation core charge. Interestingly, similar experiments on SrTiO3 show the opposite effect with positive dislocation core and space charge depletion layers. [1] X. Guo and R. Waser, Progress In Materials Science, 2006, 51, 151. [2] I. Denk, J. Claus and J. Maier, Journal of Electrochemical Society, 1997, 144, 3526. [3] Q. L. Wang, O. K. Varghese, C. A. Grimes and E. C. Dickey, Solid State Ionics, 2007, 178, 187.

Diamond-like carbon (DLC) has gained much attention as a super-low friction material for automotive engines, aerospace equipments, and so on. The detailed understanding of the tribochemical reactions of the DLC is strongly required for clarifying its super-low friction mechanism and designing more efficient super-low friction materials. Here, classical molecular dynamics simulation is frequently employed to investigate the friction behaviors on atomic scale. However, the classical molecular dynamics method cannot simulate the chemical reaction dynamics. Therefore, we developed our original first-principles molecular dynamics simulator “Violet” [1] and tight-binding quantum chemical molecular dynamics simulator “Colors” [2] for the elucidation of the tribochemical reaction dynamics. First, we applied our first-principles molecular dynamics simulator to the investigation on the tribochemical reaction dynamics of the H-terminated DLC and clarified that the hydrogen-hydrogen repulsion at the interface leads to the super-low friction of the DLC. Our tight-binding quantum chemical molecular dynamics simulations on the larger models also support the above conclusion. Moreover, our tight-binding quantum chemical molecular dynamics simulations suggest the formation of H2 molecule at the friction interface by the tribochemical reaction of hydrogen atoms which terminate carbon atoms [3]. The super-low friction coefficient of 0.07 is realized after the formation of H2 molecule. Then we suggest that the formation of vapor phase at the DLC-DLC interface is the reason for the super-low friction of the DLC. Furthermore, we investigated the tribochemical reaction dynamics of the OH-terminated DLC. The friction coefficient of the OH-terminated DLC is significantly lower than that of the H-terminated DLC under a severe condition of 7GPa load. It suggests that the OH-terminated DLC can solve the instability problem of the H-terminated DLC under a very severe condition [4]. We also simulated the tribochemical reaction dynamics of the H-terminated DLC under methanol condition [5]. The simulation results show that the methanol molecules decomposed at the friction interface and then the OH-terminations are observed. Therefore, we concluded that the methanol condition is very effective to achieve the super-low friction because the OH-termination is realized by the tribochemical reactions of methanol molecules. [1] T. Shimazaki and M. Kubo, Chem. Phys. Lett., 503 (2011) 316. [2] M. Koyama, M. Kubo, A. Miyamoto et al., J. Phys. Chem. B, 110 (2006) 17507. [3] K. Hayashi, M. Kubo et al., J. Phys. Chem. C, 115 (2011) 22981. [4] K. Hayashi, M. Kubo et al., in preparation. [5] K. Hayashi, M. Kubo et al., Faraday Discuss., in press.

We present the microscopic structure, physical properties and defects evolution in novel nanocarbon hybrids of single- and multi-walled carbon nanotubes (SW and MWCNTs) and detonation nano diamond forming truly tetragonal-trigonal nanocomposite ensemble that were subjected to 50, 100 and 103 kGy gamma ray doses. They were solution cast as thin films and characterized using various analytical tools to investigate hierarchical defects evolution much below threshold. This work is a summary of our recent work [Gupta et al. J. Appl. Phys. 107 (2010) and Gupta et al. J. Appl. Phys. 106 (2009)] where radiation-induced microscopic defects seem to be stabilized by UDD, especially in the case of SWCNTs as compared with MWCNTs when subjected to the maximum irradiation dose. Quantitative analyses of multi-wavelength Raman spectra assessed through variation in the prominent D, G and 2D Raman bands elucidated radiation-induced microscopic defects. A minimal change in the position of D, G and 2D bands and a marginal increase in intensity of the defect induced double resonant D band and Raman active 2D band are some of the implications suggesting combined radiation coupling and resilience. The in-plane correlation length (La) was also determined following TK relation from the ratio of D to G band (ID/IG) besides microscopic stress/strain. Moreover, we also suggest the following taking into account of intrinsic defects of the constituents: a) charge transfer arising at the interface due to the difference in electro-negativity of CNT C-sp2 and UDD core (C-sp3) leading to phonon and electron energy renormalization; b) mis-orientation of C-sp2 at the interface of MWCNT and UDD shell (C-sp2) resulting in structural disorder; c) softening or violation of the q ~ 0 selection rule leading to D band broadening and a minimal change in G band intensity; and d) normalized intensity of D and G bands with 2D band help to distinguish defect-induced double resonance phenomena. Furthermore, an attempt was made to identify the nature of defects (charged versus residual) and it seemed the electrical properties were more labile to irradiation than structural properties. This work was primarily conducted at the University of Missouri and MURR while the author (S.G.) was working there.

A rational control of the structure of Single Wall Carbon Nanotubes during their synthesis is highly desirable, but currently limited by our poor understanding of nucleation and growth mechanisms and the lack of direct evidence on the actual state of the catalyst particle / nanotube interface. To progress towards an atomic scale understanding, we use a carefully assessed tight binding model for nickel and carbon [1, 2] to numerically investigate different aspects of the CCVD synthesis process. Owing to significant technical improvements of our grand canonical Monte Carlo code [2], we can extend our previous calculations [3] of carbon adsorption isotherms to nanoparticles (NPs) up to 807 Ni atoms, in a broad temperature range. We thereby study the carbon solubility and physical state of the metal catalyst as a function of size, temperature and carbon chemical potential conditions corresponding to nucleation and growth of SWNTs. Combining experimental information from Transmission Electron Microscopy and atomistic computer simulation, we try and understand the relation between the diameters of the tube and the metallic NP from which it grows [4]. We then study the wetting of the NPs with respect to sp2 carbon walls, that strongly depends on carbon concentration, and emphasize its role in the growth of tubes. This enables us to identify conditions leading to experimentally observed situations: aborted growth by encapsulation of the metal NP with carbon, growth termination by detachment of the tube from the NP and continuous growth under mild carbon chemical potential, temperature and feeding rate conditions. [1] H. Amara, J. M. Roussel, C. Bichara, J.-P. Gaspard and F. Ducastelle, Phys. Rev. B 79, 014109 (2009). [2] J. H. Los, C. Bichara and R. Pellenq, Phys. Rev. B 84, 085455 (2011). [3] H. Amara, C. Bichara and F. Ducastelle, Phys. Rev. Lett., 100, 056105 (2008). [4] M.-F. C. Fiawoo, A.-M. Bonnot, H. Amara, C. Bichara, J. Thibault-Pénisson and A. Loiseau, Phys. Rev. Lett. 108, 195503 (2012).

We use a combination of theory (density functional calculations and simulations using classical potentials) and experiments (Z-contrast imaging & electron energy-loss spectroscopy using aberration-corrected scanning transmission electron microscopy) to probe the structure and electronic properties of grain boundaries in graphene and the interactions of grain boundaries with point defects. Grain boundaries exhibit an alternation of pentagonal and heptagonal rings where the local strain is compressive and extensive, respectively. Vacancies are attracted at pentagons and carbon interstitials are attracted at heptagons [1]. In both cases the point defects merge in the grain boundary and release strain. The grain boundaries also mediate vacancy-interstitial annihilation. Furthermore, grain boundaries attract O adatoms and bind them more strongly. Cooperative activity among O atoms can lead to preferential etching in grain boundaries. Z-contrast images obtained with an aberration-corrected scanning transmission electron microscope reveal twisted grain boundaries that undergo continuous evolution under the electron beam via conversion of pentagon-heptagon ring pairs into hexagon-hexagon ring pairs and vice versa. Molecular dynamics simulations using classical potentials probe the stability of straight boundaries and the activation energies for the observed evolution. Electron-energy-loss spectroscopy in the low-loss region (valence excitations) has been used to explore plasmon localization, a phenomenon that was recently observed to occur at impurities and grapheme edges [2]. The potential for grain boundaries acting as plasmonic wave guides will be discussed. An image of a highly symmetric grain boundary loop has been obtained and its structural, electronic, and magnetic properties have been explored by theory, including the possibility that single impurity atoms at the center of the loop may impart unique properties to this topological defect. This research was supported by NSF grant No. DMR-0938330 (WZ, J-CI), DOE grant DE- F002-09ER46554 (STP), by ORNL&’s Shared Research Equipment (ShaRE) User Facility, Oak Ridge National Laboratory, which is sponsored by the Scientific User Facilities Division, Office of Basic Energy Sciences (J-CI), by the Office of Basic Energy Sciences, Materials Sciences and Engineering Division, U.S. Department of Energy (SJP, JL,STP), and by the Defense Treat Reduction Agency grant HDTRA1-10-1-0016 (BW,YP,STP) [1] B. Wang, Y. Puzyrev and S. T. Pantelides, “Strain-enhanced defect reactivity at grain boundaries in polycrystalline graphene”, Carbon 49, 3983 (2011). [2] W. Zhou, J. Lee, J. Nanda, S. T. Pantelides, S. J. Pennycook, and J.-C. Idrobo, “Atomically localized plasmon enhancement in monolayer grapheme”, Nature Nanotechnol. 7, 161 (2012).

Catalytic chemical vapor deposition (CCVD) is the most widely used technique for growing single-walled carbon nanotubes (SWCNTs). It is also the most promising one in terms of upscaling and structural control. However, it is commonly observed that CCVD-grown SWCNTs contain defects. Controlling the defect density of SWCNTs and graphene-like materials in general is a key issue for the control of their properties. We will report on the influence of the CCVD parameters on the defect density of SWCNTs using in situ and ex situ Raman analyses and HR-TEM observations [1,2]. In situ Raman monitoring was notably used to determine experimental conditions allowing the preparation of SWCNT samples free of pyrolytic carbon and not altered by air exposure. Using the Raman D band to quantify the SWCNT defect density brought two unexpected results. First, the evolutions of the G/D ratio with the synthesis temperature display two distinct Arrhenius-type regimes. Second, these Arrhenius plots systematically display a convex shape, that is a lower apparent activation energy with increasing temperature. HRTEM and Raman analyses support that long and straight SWCNTs with few -curvature-inducing defects are grown at high temperature while much shorter and more tortuous SWCNTs and carbon nanostructures are grown at low temperature. Kinetic models were tested to reproduce the experimental evolution of the G/D ratio as a function of the growth parameters. We found that two different models can account for the convex shape of the Arrhenius plots: (i) a model considering a change of intermediate states as a function of the temperature (for instance d