First-principles calculations can make quantitative predictions of the thermodynamics and diffusion properties of hydrogen in metals. We will discuss examples of using these methods in tandem with large materials libraries to identify new materials for applications involving high temperature hydrogen purification and high temperature hydrogen capture. In the former application, we have screened large collections of disordered alloys, intermetallics, and amorphous metal alloys as metal membranes. In the latter application, we have used first-principles computational thermodynamics to identify novel complex metal hydrides with desirable properties.

One attractive aspect of organic optoelectronic materials is their significantly large and complex design space, providing great opportunity to achieve diverse physical and chemical properties. As such, there is a pressing need for the development of efficient and low-cost methods to explore the vast chemical space and advise experimental efforts. In this work, we examine the feasibility of virtual screening of organic light-emitting diode (OLED) and organic photovoltaic (OPV) materials based upon quantitative structure-property relationships (QSPR) for their charge transfer characteristics. Our QSPR study indicates Marcus reorganization energy of a molecule - which has an exponential impact on the charge transfer rate - can be well illustrated by a small group of topological descriptors from 2D molecular structures. The predictive trend allows building an effective in silico design scheme for a wide selection of organic semiconductor materials and screening a large group of virtually generated candidate compounds without computationally demanding electronic structure calculations. With the help of Schrodinger&’s new Materials Science Suite, we demonstrate the design workflow from building candidate libraries to optimizing QSPR models can be performed within a single intuitive graphical user interface environment. The work illustrates the prospect of simplifying the complex workflows of cheminformatics-based materials design, which can rapidly accelerate the discovery and optimization process for organic semiconductor materials.

The objective of this work is to develop and apply advanced virtual diffraction techniques to characterize atomistic simulations of polymorphic alumina surfaces and interfaces. Virtual diffraction techniques produce computational analogues to both selected area electron diffraction patterns and two-theta x-ray diffraction line profiles allowing direct experimental validation of atomistic simulations. Traditional characterization methods within atomistic simulations are often limited to cubic or other high-symmetry crystal structure and none are capable of uniquely identify alumina phases due to the subtle differences buried within their non-cubic crystal structures. In this work, virtual diffraction methods are advanced to compute diffraction intensities from atomic simulation data across a high resolution mesh of reciprocal space, which eliminates the need for a priori knowledge of the crystal structure being modeled or other assumptions concerning the diffraction conditions. Virtual selected area electron diffraction patterns are constructed by viewing the projection of a hemispherical slice of the reciprocal lattice mesh lying on the surface of the Ewald sphere along a specified zone axis. X-ray diffraction line profiles are created by binning the intensity of each reciprocal lattice point by its associated scattering angle, effectively mimicking powder diffraction conditions. Together, virtual x-ray diffraction line profiles and selected area electron diffraction patterns uniquely identify the different phases as well as characterize the nanoscale distortions associated with atomic relaxations near interfaces.

In this presentation, we introduce a comprehensive study of the binary systems of the platinum group metals with the transition metals, using high-throughput first-principles calculations. These computations predict stability of new compounds in 37 binary systems where no compounds have been reported in the literature experimentally, and a few dozen of as yet unreported compounds in additional systems. Our calculations also identify stable structures at compound compositions that have been previously reported without detailed structural data and indicate that some experimentally reported compounds may actually be unstable at low temperatures. With these results we construct enhanced structure maps for the binary alloys of platinum group metals. We also introduce the aflowlib.org API (application programming interface), an interface allowing computational scientists to download thermodynamic/electronic structure information with simple commands embedded in their own scripts/programs. We will show how to plot PGM phase diagrams in the users personal laptop. Research sponsored by DOD-ONR. [Hart, Curtarolo, Massalski, Levy, A comprehensive search for new phases and compounds in binary alloy systems based on platinum group metals, using a computational first principles approach, in press, Phys. Rev. X, (2013)]

Molecular dynamics (MD) computer simulations have been used to evaluate the structure of thin (~1-4nm) amorphous intergranular films (IGFs) in oxide and nitride ceramics. The simulations predict adsorption of modifier ions from the silicate IGFs onto specific bounding crystal surfaces that poison those surfaces, altering grain growth and subsequent morphology and mechanical properties. Subsequent HAADF-STEM studies corroborate the location of such modifier ions at the same surfaces as seen in the simulations. In the case of the IGF in silicon nitride ceramics, the HAADF-STEM results raise questions regarding the effect of La versus Lu adsorption on grain growth. The experimental data show the presence of each species on the prism-oriented nitride surface, but grain growth is significantly different in each, in which La inhibits growth whereas Lu does not. Finding the adsorbate at the same surface in HAADF-STEM is clearly not enough to signal poisoning. Our recent MD simulations reproduce the specific adsorption of Lu and La at the sites observed in HAADF-STEM, but also show that the presence of oxygen at specific N surface lattice sites plays a significant and required role on Lu adsorption, but has almost no effect on La adsorption. EELS studies have shown that oxygen is present at sites on the prism surface of silicon nitride in contact with the oxide IGF. By evaluating the energetics of Lu at specific surface adsorption sites and the effect of the melt state during liquid-phase sintering on growth behavior, the simulations identify the mechanism for the apparently opposite effect of similar adsorption of these rare-earths on the nitride surface coupled with the very different growth behavior.

Selective surfaces for solar to thermal energy conversion require low optical reflectance in the visible and near infrared region (high solar absorptance α) and a high reflectance in the infrared beyond 3 mu;m (low thermal emittance ε). This is achieved with the introduction of metallic inclusions into a dielectric matrix. These materials can be prepared by Ni electrochemical coloration of nanoporous alumina previously obtained by anodization of an Al foil. It results in surfaces with typical values for α and ε of 0.82 and 0.07, respectively. This work describes the use of a numerical simulation for improving this performance. The simulation predicts the optical reflectance of the surfaces from compositional depth profiles. A finite differential elements algorithm was used for solving the wave propagation in depth dependent dielectric constant. Effective medium theory was used for obtaining the optical properties of the composite from tabulated data. Several concentration depth profiles were simulated to obtain the corresponding optical reflectance spectra and the dependence on the α and ε parameters were studied. It is concluded that the selectivity is highly dependent on the interference effect originated in the different layers. Moreover, it was found that α can be improved (keeping ε at a relative low value) by increasing alumina thickness, metal inclusion content or increasing molecular weight of metallic inclusion.
To confirm these results several samples were experimentally characterized varying the main parameters of the preparation process: anodization duration, metallic inclusions and its electrodeposition potential and duration. Optical reflectance was measured for all samples by different setups for each spectral region. Selected samples were studied by XPS (X-ray Photoelectron Spectroscopy) to obtain the atomic concentration profile. Comparing the predicted optical properties from XPS measurements with the measured ones validated the numerical simulation. The correlation with the numerical simulation results allowed improving the α of the Ni containing samples to 0.85. Samples with Cu and Ag inclusions were also prepared, giving place to 0.90 and 0.05 for the best values obtained for α and ε, respectively. Although these values are still within usually reported range, the simulations provide a way for further selective surface improvements for solar to thermal energy conversion. Values between 0.95 and 0.97 may be obtained for samples with alumina thickness of 1.5 mu;m and metal inclusions with 40 % of peak volumetric filling fraction. Moreover, in this direction the simulation predicts that for Ag inclusion the step-like behavior can be further shifted beyond 3 mu;m.

Chlorinated organic compounds are the most primary contaminant of the groundwater, causing a significant environmental problem. Due to their persistency and toxicity, many efforts have been done to treat them for the last few decades. Especially, a variety of iron bearing soil minerals (e.g., magnetite, pyrite, green rust) has shown a remarkable reduction capacity for the chlorinated compounds such as tetrachloroethylene (PCE) and carbon tetrachloride (CT). Vivianite, monoclinic octa-hydrated ferrous phosphate, also showed an important role on the enhanced reductive dechlorination of CT. Because vivianite can be easily formed in the phosphorous-enriched natural reducing environments, it may significantly influence the contaminant fate and transport in subsurface. However, there is no definite dechlorination pathway that can explain how these chlorinated compounds are degenerated by vivianite, and the efficiency of vivianite in dechlorination pathway is also not clear. Therefore, by using the combination of experiments and density functional theory (DFT) calculations, we investigate the dechlorination reaction mechanism of various chlorinated compounds on the vivianite surface. We discuss which chlorinated compounds are favorably decomposed using vivianite, and further discuss the feasibility of groundwater treatment technology based on iron bearing soil minerals.

Quantum chemistry calculations of acrylamide spectra
Sheng D. Chao
Institute of Applied Mechanics, National Taiwan University, Taipei 106, Taiwan ROC
e-mail address: sdchao@spring.iam.ntu.edu.tw
Acrylamide (2-propenamide, C3H5NO) has been used for industrial polymerization and food chemistry [1-4]. We have carried out the MP2 and density functional (DF) calculations using basis sets up to aug-cc-pVTZ. The isolated acrylamide molecule was found to have two stable configurations. We then consider 12 conformers of the acrylamide dimer. The basis set superposition errors (BSSEs) were corrected by the counterpoise method of Boys and Bernardi. Subsequently the full potential curves were obtained. It is shown that correlation interactions contribute significantly to the stabilization. Also, we have found the M05 and X3L density functionals can yield results close to the MP2 calculations. Using these data we simulated the infra-red and Raman spectra. The overall agreement with the experiment is achieved while for the high frequency C=O and NH modes we found significant evidence for double proton transfer reactions.
[1] W. L. Claeys, K. De Vleeschouwer and M. E. Hendrickx, J. Agric. Food Chem., 2005, 53, 9999-10005
[2] FSA. Food Standard Agency Study of Acrylamide in Food Background Information & Research Findings; Press briefing 17.05.02; Food Standard Agency, United Kingdom, 2002
[3] SNFA. Analytical Methodology and Survey Results for Acrylamide in Foods; Swedish National Food Administration, 2002
[4] FDA. Acrylamide in Food, Food and Drug Administration, United States, 2004

The use of synchrotron X-ray spectroscopy for materials characterization is of growing importance, particularly as new in situ capabilities come on line for analysis of electrochemical processes, catalysis, and solar harvesting. X-ray absorption spectra (XAS) provide a direct probe of local element-specific electronic structure from which chemical and structural properties can be inferred. However, such XAS studies often require theoretical interpretation, particularly in the absence of a pre-existing spectral fingerprint. Recently, we have developed a first-principles approach to simulate XAS based on density functional theory. Through molecular dynamics sampling, we have found significant and non-trivial spectral contributions due to structural disorder and finite temperature dynamics. The successes of this method in predicting the spectra of a wide range of gas and condensed phase systems have led us to develop a web-based interface which permits users to launch such simulations on remote high-performance computing resources once they provide a structural model for the system. However, the process of extracting the origins of particular spectral line-shape variation is nontrivial and definitely does not always follow a simple Gaussian renormalization of a spectrum one might compute for a representative reference system. To explore this issue, we have developed a framework to extract the most significant degrees of structural variation that drive spectral variations based on statistical and principal component analysis. This defines a framework for machine-based analysis of spectra computed from a multitude of atoms sampled from different configurations that can extract useful and intuitive understanding implicitly visible in the ensemble-averaged X-ray absorption spectrum.
This work was performed at The Molecular Foundry, supported by the Office of Science, Office of Basic Energy Sciences, of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231.

Aqueous phase stability of materials is important for predicting corrosion, dissolution, and for the synthesis of materials in aqueous environments. A Pourbaix diagram, or an E-pH diagram, is a representation of aqueous phase electrochemical equilibria pioneered by Marcel Pourbaix [1]. Experimentally determining Pourbaix Diagrams is painstaking, as we need not only the free energy of aqueous ions, but also of all solid phases that a system can exist in.
Recently, we released a high-throughput app [2], which utilizes the Materials Project (www.materialsproject.org) database of DFT calculations on solids, and experimental data of ions in solution gathered from various tabulations to generate stability diagrams in aqueous solutions. The methodology is based on the ability to reference experimental data and DFT energies to compare them on a consistent framework [3]. In this talk, we wish to outline possible applications of the app in the prediction of corrosion, dissolution, etc. and hence demonstrate the versatility of our open-source Pourbaix diagram generation app.
[1] M. Pourbaix, Atlas of Electrochemical Equilibria in Aqueous Solutions, National Association of Corrosion Engineers, Houston, TX (1974).
[2] https://materialsproject.org/apps/pourbaix/
[3] K. Persson, B. Waldwick, P. Lazic, G. Ceder, Physical Review B, 85, 235438 (2012)

The availability of large amounts of data generated by high-throughput computing or experimenting has generated interested in the application of machine learning techniques to materials science [1]. Machine learning of materials behavior requires the use of feature vectors or descriptors that capture the essential compositional or structural information that is most likely to influence a property. We will present a new method for assessing the similarity of material compositions. A similarity measure is important for the classification and clustering of compositions. The similarity of the material compositions is calculated utilizing a data-mined ionic substitutional similarity based upon the probability with which two ions will substitute for each other within the same structure prototype. The method is validated via the prediction of crystal structure prototypes for oxides from the Inorganic Crystal Structure Database. It performs particularly well on the quaternary oxides, predicting the correct prototype within 5 guesses 90% of the time. We expect that this compositional similarity measure can be used to classify other properties as well.
[1] C. Fischer et al, Predicting Crystal Structure: Merging Data Mining with Quantum Mechanics, Nature Materials, 5 (8), pp. 641-6 (2006). G. Hautier et al, Data Mined Ionic Substitutions for the Discovery of New Compounds, Inorganic Chemistry, 50 (2), 656-663 (2011).

We present a demonstration of Bayesian inference applied to a representative problem in materials research. Our goal is to infer the length scale of a spatially varying field on a 2-D substrate from the behavior of a phase-separating film deposited on the substrate and modeled by the Cahn-Hilliard equation. Inference was performed based on a combination of parametric and non-parametric model approximations followed by the use of Bayes&’ rule to determine the probability distribution of the underlying length scale. The techniques demonstrated are general and may be extended to other problems involving the assessment of reduced order models in materials science.
This work was supported by the US Department of Energy (DOE), Office of Basic Energy Sciences under Award No. DE-SC0008926.

Scientists are often posed with the problem of finding a good choice of experimental parameters that optimize a particular quantity in the presence of ambiguity about the underlying system. This ambiguity arises from unknown physical constants, competing (and often poorly understood) mechanisms, and measurement or calibration error. Instead of an exhaustive and potentially expensive search over all such parameters, the method of Optimal Learning uses Bayesian statistics and heuristics to guide an experimenter through parameter space in an efficient manner, with the goal of learning about the underlying physical system through each experiment. We give an overview of the Optimal Learning method and the heuristics involved. We present how a meaningful baseline prior belief is elicited, using the domain knowledge of the experimenter. We then present a real-world application of the Optimal Learning technique to characterizing the stability of nano-emulsions stabilized by gold nanorods.

Concern for supply restrictions of rare-earth metals has spurred intense interest in the discovery of new compounds that do not contain critical elements yet still exhibit high saturation magnetization and intrinsic coercively. There have also been efforts in optimizing magnetic properties in older alloys. Both are daunting tasks given the high energy product of existing rare-earth based alloys. Criticality of Dy, in particular, is driving the need to developing new alloys for the higher operating temperature regime of traction motors and some generators. Discovery of new compounds, however, requires a more sophisticated approach than simple “trial and error”. Ames Laboratory, in collaboration with a number of universities and laboratories have been embarking on a comprehensive research program to combine a series of integrated computational and experimental efforts to both discover and design new compounds with promising magnetic properties. Experimental materials discovery will include both bulk and thin film combinatorial synthesis. Additionally, we are developing high throughput thermal analysis and in situ XRD capabilities to characterize the phase space of these multi-elemental libraries as a function of temperature. The computational efforts of this research include both density functional theory and adaptive genetic algorithms to identify new compounds. Specific examples of materials discovery and new insights into improvements of existing alloys will be presented.

Although for I-III-VI2 ternaries the chalcopyrite (CP) phase [1] is thermodynamically the most favorable, recent advances in growth techniques make it possible to synthesise other crystal phases, i.e. the wurtzite (WZ) phase. The WZ CuInS2 compound has been the subject of recent research efforts because of its potential as a light-absorbing material in printed and flexible PV devices, light-emitting diodes, and nonlinear optical devices. It has been observed experimentally that the band gap of WZ CuInS2 is near the red edge of the visible spectrum, and that this material has high optical absorption coefficients and substantial photo-stability. Once assembled in a solar cell device via vapour-phase deposition, it has been shown to achieve the power conversion efficiency of 7.5%. Several experimental groups have been able to synthesise this material recently in various nanocrystalline forms. [2-4]
We have examined theoretically the structure, electronic and optical properties of wurtzite CuInS2. We use hybrid exchange density-functional theory in the B3LYP approximation, to study the relative stability of possible competing WZ structures. The B3LYP approximation has been shown to yield highly accurate estimates of, not only band gaps, but also the whole electronic dispersion throughout the Brillouin zone, for semiconductor materials in the zinc-blend, wurtzite, and chalcopyrite structures. [5] For the most stable WZ CuInS2 polymorph, we also examine the optically induced creation of a bound exciton, and we estimate its binding energy using hybrid exchange time-dependent density-functional theory [6]. We have recently shown that this approach is capable of describing to high accuracy the optical response of weakly bound excitonic systems.
We have characterised 3 possible WZ CuInS2 lattices, only one of which has a non-vanishing band gap. In this system, the ideal WZ structure is perturbed by the presence of two inequivalent cation sites, which results in the removal of the degeneracy near the top of the valence band and contributes an additional crystal field splitting of 88 meV. Both CP CuInS2 and WZ are found to absorb strongly in the visible region, with Lambert- Beer attenuation coefficients more than 5 times higher than in conventional III-V semiconductors. In both materials, the optical response in the visible region is dominated by intense excitonic features, with binding energies of the order of 20-30 meV in CP-CuInS2 and 10 meV in WZ. The latter system also exhibits a sizeable (4.4%) anisotropy in the optical response in the xy plane, in contrast to a perfect tetragonal (CP-CuInS2) or a perfect WZ structure.
[1] S. Siebentritt at al, Progress in Photovoltaics 18, 390 (2010).
[2] C. Sun et al, Chem. Mat. 22, 2699 (2010).
[3] B. Koo et al, Chem. Mat. 21, 1962 (2009).
[4] Y. Qi at al, J. Phys. Chem. C 113, 3939 (2009).
[5] C. L. Bailey at al, Phys. Rev. B 81, 205214 (2010).
[6] L. Bernasconi at al, Phys. Rev. B 83, 195325 (2011).

We report a simple one-pot, low temperature polycondensation reaction between paraformaldehyde and bisanilines to form hemiaminal supramolecular networks (HSNs) that undergoes thermally-induced cyclization at high temperatures to form poly(hexahydrotriazine)s (PHTs). HSNs prepared from bisanilines are mechanically strong, and, at elevated temperatures undergo cyclization to form extraordinarily high modulus (14.0 GPa) PHTs. HSN and PHT properties are characterized through a suite of spectroscopic and mechanical techniques. A proposed mechanism for the HSN and PHT forming reactions is presented, and DFT calculations support spectroscopic characterization and model experiments for the formation of HSN as an intermediate to PHT. Although completely inert to organic solvents and alkaline aqueous solutions, PHT and HSN thermosets were found to be easily digested in strong acid and revert back to their starting monomers, thereby rendering them recyclable high-modulus materials.

In order to realize the order of magnitude acceleration of novel material commercialization envisioned by programs such as the Materials Genome Initiative one must be able not only to identify novel materials in silico but also to controllably synthesize and validate their predicted properties. This is a non-trivial task thus, novel strategies are needed that facilitate linking experimental synthesis inputs with theoretical outputs, such as enthalpy of formation. The primary objective of this work is to develop a multi-tiered physics-based method for the predictive reactive-sputtering deposition of phase-pure metastable materials. In particular, my talk will focus on our recent work in implementing a sputter modeling approach combining dimensionless angular atomic flux with sputtered particle thermalization, to arrive at the energy distribution of incident atoms and film compositions at any point on the substrate. I will discuss recent work by my group using this technique in the fields of high-temperature Fe and Ni-based oxidation resistant materials and self-protecting YSZ-based thermal barrier coatings. In both instances it is important to be able to monitor the phase of a base material (ie β-NiAl) undergoing a destabilizing thermochemical reaction with a gas or solid across the entire phase diagram. Here, the time dependent positions of phase boundaries as a function of synthesis and high temperature exposure to harsh oxidative environments are determined through a combination of optical spectroscopy and diffraction measurements. Data sets in such cases become large rapidly and correlating base-phase stability and the nucleation and growth of beneficial/deleterious phases with synthesis and processing conditions is an onerous task. Challenges in addressing these issues with current state-of-the-art materials data minimization/visualization techniques will be discussed, and a perspective for the integration of these results with DFT calculations will be set forward.

This lecture will describe some recent progress in the methods for rapid and precise characterization of samples composed of noble-metal nanocrystals protected by self-assembled monolayers (SAMs). As these materials are frequently prepared and size-purified in solution, a key step is the transition intact to the vapor phase and onto a support in vacuum. The electrospray method provides an efficient and 'soft' transfer from the dilute solution, which accounts already for the "electrospray revolution" in macro-molecular biology. Recently electrospray has been applied to large metal clusters and smaller metallic nanocrystals, in the range to ~ 3-nm (1000 metal atoms). The time-duration and material quantity required for such analysis, employing high resolution and ultra sensitive mass-spectrometric detection of the intact structure, has been reduced to ~ 30-seconds and the microgram-level. Further reductions to the few picomole (~ 100-ng) level are achieved by employing the nano-electrospray sources. Progress in high-speed electron diffraction analysis of single smaller metal nanocrystals and their protective SAMs, is also described. Finally, we outline the special challenge presented by coupling vacuum-deposition of nanocrystals from the electrospray source to the supports used for high resolution electron microscopy and rapid electron diffraction analysis.

A high-throughput (HTP) approach to rapid materials discovery will not only result in many new functional materials, but also in many unforeseen new structure-property relationships and has the potential of revolutionizing computational materials theories and our understanding of metastable materials. This approach is based on a three step iterative discovery cycle: HTP production of combinatorial materials libraries, HTP characterizations of these libraries, and unsupervised, automated and machine-learned approaches of extracting often hidden and unforeseen features and relationships from them. These new insights and theories will drive the next generations of combinatorial materials libraries.
This talk will focus on second step of the discovery cycle and narrate recent development at Stanford Synchrotron Radiation Lightsource on a facility that can screen up to 2000 samples, via x-ray diffraction and fluorescence, on combinatorial libraries printed on Silicon wafers and glass plates.

We show simultaneous molecular chemistry and bond orientation information over a large area (18 mm x 13 mm) obtained with high transmission using a novel parallel process full field secondary electron spectrometer. This unique spectrometer (LARIAT MKI) incorporates: a full field soft x-ray source (NSLS beamline U7A); electrostatic and magnetostatic electron optical elements to discriminate energy and depth of the secondary electron distribution and; a highly parallel electron detector. The rapid parallel process produces a series of two dimensional images as the incident soft X-ray energy is scanned above a K or L absorption edge. Synthesis of the of the image stack produce spatially resolved NEXAF spectra containing information about the chemistry (including bond concentration) and orientation of the surface-bound molecules with better than 100-micron lateral resolution and sub-monolayer molecular sensitivity. Rapid image-based full field capture of NEXAFS spectra over large areas permits new combinatorial approaches to data acquisition and analysis. The power of the combinatorial imaging NEXAFS method is described through a variety of examples including: i) simultaneously probing concentration and molecular orientation of single-strand DNA micro array sensors; ii) semifluorinated molecular gradients; iii) and organic electronic combinatorial device arrays. Other possible applications described include the surface orientation and chemistry of continuously graded polymer films and graded or patterned self-assembled monolayers. We also envision combinatorial imaging NEXAFS as an in-situ probe for catalyst discovery using micro arrays to directly image catalytic chemical activity of thousands of catalysts simultaneously under reaction conditions.
We will also briefly describe improvements of the second generation system (LARIAT MKII), currently in the testing phase on beam line U8 at NSLS. The MKII will be permanently installed on new Brookhaven National Laboratory synchrotron source, NSLS II to be completed in 2015 The key hardware differences in the LARIAT MKII configuration as compared to the MK I model is the use of superconducting magnet components and additional electrostatic elements in the electron optics assembly. The use of superconducting magnets is necessary to attain the significantly higher magnetic fields required to achieve ~600X increase in areal resolution. This overall resolution improvement derives from a factor of four increase in imaging area while simultaneously improving the lateral resolution by a factor of up to twenty (~5µm). There is also the possibility to operate this instrument in a nano scale resolution modality using zone plate optics to produce a type of scanning reflection x-ray microscopy.

The last several decades have seen an explosion of models that describe, from physics, all aspects of material behavior. In more recent times, ‘Big Data&’ has emerged in the form of automated data acquisition from instruments. A traditional materials modeling approach to fuse these two sources of information is to use the data to estimate calibration parameters for the physics-based model, thereby biasing the simulated answer towards one consistent with the known experimental results. A traditional signal processing approach reverses these roles and uses a physics model to bias analysis of data to produce results that are consistent with the known physics of structure formation. Using a Bayesian model originally developed for image analysis, it will be shown that this methodology is ideally suited for analysis of materials data because of the physical analogues of the ‘prior knowledge&’ used to perform this bias. Specifically, the Markov Random Field (MRF), which is identical to the Ising model of ferromagnetism, is used a prior that reproduces surface phenomena, as well as equilibrium phase constituent fractions. Examples will be given of Bayesian regularized inversions of microscope image data that can be adjusted in ‘physically correct&’ ways in order to compensate for signal degradation due to noise.

While high-throughput inorganic library synthesis is relatively well-established, high-throughput characterization of properties is a much greater and idiosyncratic challenge, and high-throughput structure determination is in its infancy. We use cosputtering to routinely prepare thin film composition spreads and have developed a suite of techniques to measure electrical and electrocatalytic properties. We have also developed experimental techniques for rapidly collecting x-ray diffraction data using high-energy synchrotron radiation, an approach in which a wide range of scattering vectors is intercepted by a two-dimensional detector. Manual XRD data analysis and interpretation is time-consuming and highly labor intensive. We have addressed the challenge of creating efficient computational methods for XRD analysis using tight integration of statistical machine learning methods, to deal with noise and uncertainty in the measurement data, with optimization and inference techniques that incorporate the rich set of constraints inherent in the underlying materials physics and chemistry. We have found that adding a so-called human computation component (along the lines of the FoldIt project) can dramatically improve the efficiency of the automated tasks. This approach is enabling us to gain insights and knowledge from the deluge of data, capturing and distilling structure/processing/property relations in multicomponent materials.

We are actively developing visualization and analysis techniques to rapidly extract trends in composition-structure-property relationships from large data sets from combinatorial libraries. Because data often take on spectral or higher dimensional formats, it is helpful to apply data reduction schemes involving cluster analysis. Examples of techniques we have applied to date include hierarchical clustering using metric multidimensional scaling as the metric, non-negative matrix factorization, and the mean shift theory. They have proven to be extremely useful in deciphering the distribution of structural phases across composition spread samples using diffraction data. We have also applied our techniques to Raman spectra and hysteresis curves. We will also discuss applications of new regression techniques including the relevant vector machines. This work is funded by ONR and DOE.

We present the Autonomous Rapid Experimentation System (ARES), a platform for synthesis, characterization and model development for carbon nanotubes (CNTs). We grow carbon nanotubes by heating a catalyst in the presence of feedstock gases, and use the same laser for both thermal activation of the growth process and Raman excitation. This enables us to perform in-situ analysis of growth kinetics and nanotube characteristics. Because the system is fully automated, we are able perform sequences of experiments one hundred times faster than standard techniques. The rapid serial nature of the method, as opposed to a parallel combinatorial approach, enables us to change the synthesis conditions for each individual experiment.
The ARES method has generated thousands of experiments, and thus now requires us to employ statistical and data mining techniques to analyze the results. We will present our analysis of the probability and kinetics of single wall nanotube growth events compared to multiwall growth, compared to no growth. Among the data mining and modeling tools used by ARES are linear regression, symbolic regression, neural networks and random forests. These tools enable ARES to feed model results into experiment design, and in the near future we will close the loop by feeding experimental results back into the models. The combination of the integrated model development, refinement, verification and validation, and faster experimentation allow ARES to produce high volumes of useful data while leveraging it intelligently and reducing the need for human involvement. The resulting understanding of the CNT growth kinetics and its relationship with morphology may lead to better control and understanding of CNT growth.

Lanthanide-doped nanocrystals can be utilized as low-background biological imaging probes due to their ability to convert infrared excitation into visible emission - a process known as upconversion. The brightness, emission wavelengths, and excitation wavelengths of these lanthanide-doped particles, however, have been limited by a poor understanding of energy transfer within the nanocrystal and an unavoidable trade-off between brightness and size. Here, we describe the use of a high-throughput computational package to simulate the spectrum and energy transfer pathways of arbitrary compositions of NaYF4 nanocrystals doped with multiple lanthanide ions. Using this in silico combinatorial method, we screened for nanocrystal compositions that exhibit high luminescence intensity and high spectral purity at unique excitation and emission wavelengths. These hits were then experimentally refined using a high-throughput nanocrystal synthesis robot and a laser-scanning plate reader. Finally, we used single-nanoparticle upconversion microscopy to investigate the effects of surfaces on the upconverted emission of a size series of nanocrystals. This combined theoretical and experimental approach allowed us to predict and experimentally verify new compositions of upconverting nanocrystals that are brighter and more spectrally pure than previous nanoparticles at unique excitation and emission wavelengths.

When it comes to solar cells, the function of photovoltaic conversion of light into electricity stems primarily from the absorbing material to sustain a large electron-hole quasi Fermi level (QFL) splitting in operation conditions. The value of the QFL that can be achieved in given conditions is determined by the amount and nature of defects and can be compared to that of a defectless material (so called radiative limit). This QFL splitting would be therefore the prime quantity to know about a material when investigating its suitability for photovoltaic conversion, but unfortunately it has not been possible up to now to measure it. In this presentation, we will present a technique based on photoluminescence Hyperspectral Imaging (PL-HI) that is able to measure directly the QFL with a spectral resolution of 2 nm and a spatial resolution near the diffraction limit [Delamarre et al., APL 2012]. We will show that the measured QFL is a good approximation of the cell voltage (within tens of meV) and will also show how this technique can be used to measure contactless the local conversion efficiency of an absorber material in thin film solar cells using exemples from Chalcogenide and III-V solar cells. Finally, we will show that this method enables fast data acquisition and high throughput.

The Materials Project (http://www.materialsproject.org) aims to leverage the information age for materials design through advanced scientific computing and innovative design methods. The methodology is scaled to all inorganic compounds and beyond, and our goal is to disseminate that information and design tools to the larger materials community. The Project was launched online in October 2011 and since then we have computed and imported > 30,000 inorganic compounds into the database, which can now be freely accessed and searched over through the web interface for structural properties, local environments and coordination, XRD, electronic structure, Li-ion electrode properties, reaction energies and more. Recently, the high-throughput materials design engine and workflow was extended to predictions of novel electrolytes for the next generation energy storage solutions. Examples of feedback between experiments and modeling delivering critical insights for a number of cases will be presented, covering inorganic solid materials, solid stability in aqueous media, as well as liquid organic electrolyte systems.

Recent advances in computing power and ab initio techniques present good opportunities for building a vast theoretical database of material properties. Fast and accurate high-throughput calculations can be achieved by aligning proper computational methods into a sophisticated automatic procedure. By minimizing human interventions, even though every procedural step should be carefully verified and tested in advance, massive data of material properties could be obtained within a reasonable computational resource. The obtained materials map may enable us to identify a material with unexpected property that can provide a breakthrough in various applications.
As an attempt to make such a materials map, we developed an automation code for computing band gap and static dielectric constant of various oxides by employing VASP code as the core engine for the ab initio calculations. The automation code generates input files from the structure data that are automatically extracted from ICSD. Human interventions are minimized in our codes and every automation step and reliability of the computed property was carefully examined in the first stage. We employ HSE06 and LDA functional for calculating band gap and dielectric constant respectively, as they are known to produce good results for each property. Various optimization procedures precede the core computations, for instance, filtering redundant or non-experimental data from ICSD and optimizing computational parameters to reduce the cost without compromising the accuracy. As results, we manage to calculate properties of more than 1000 oxides and identified some oxides with unexpected properties of large band gap and high dielectric constants. By analyzing the correlation between different physical properties, we identify the fundamental condition that enables such unique combination of material properties.

We explore the phase stability of paradigmatic perovskite ferroelectrics as a function of temperature, using first-principles density-functional perturbation theory. A high-throughput screening of all possible space groups compatible with macroscopic constraints allows us to identify microscopic configurations that are mechanically stable even in paraelectric cubic phases. Extensive first-principles molecular dynamics confirm the existence of these configurations at finite temperature, while density-functional perturbation theory allows to determine the free energy and the critical temperature for the solid-solid phase transitions. The open-source AiiDA framework we developed (www.aiida.net) allows to automatically manage the complex workflows needed for these high-throughput calculations, thanks to its ability to automatize the job execution, manage workflows, tune the job parameters, monitor the calculations, and organise all data in shared databases.

High throughput (combinatorial) materials science methodology is a relatively new research paradigm that has enabled rapid and efficient materials discovery, screening, and optimization. Combinatorial methods can address the extremely high cost and long development times of new materials, and their introduction into commerce, and are therefore consistent with the goals of the Materials Genome Initiative (MGI). Going forward, MGI will be an important driver of high throughput methodologies; not only will combinatorial methods serve as a rapid technique for validating properties predicted by modeling and simulation, but also they will be used to generate experimental data that can iteratively improve such models. This talk will review the progress of high throughput experimentation for materials in the electronic, magnetic, optical, and energy-related arenas. A major challenge for combinatorial methodology will be the ability to rapidly manage large amounts of data in a variety of formats. Thus, we will discuss the use of materials informatics to address the ever-increasing amount of data resulting from modern experimentation, as well as the speed at which it is generated. Such an approach is required to identify, from myriad choices, those materials that give rise to materials with a particular desired property. As an example, a probabilistic clustering analysis algorithm for mining large amounts of x-ray diffraction data for phase identification will be discussed. Further, we will describe the high-throughput characterization of thermochromic films for “smart,” energy efficient window coatings. The origin of the thermochromic behavior of our films is a phase transition from monoclinic VO2 to tetragonal VO2, accompanied by a change in near infrared reflectance. High throughput methodologies were used to screen binary and ternary oxides based on the VO2 system for compositions that exhibit the phase transformation in the desired (ambient) temperature range. A novel, high throughput measurement tool for spectroscopic reflectance has been constructed for this research, and data collection, handling, and analysis have been fully automated.

Conducting polymers consisting of conjugated heterocyclic chains are one of the most frequently studied classes of organic materials for next-generation organic field-effect transistors (OFETs), organic light-emitting diodes (OLEDs), and flexible photovoltaic materials. Despite their widespread importance, the conventional approach to developing new conducting polymers for these applications is still dominated by heuristic experimental guesswork by synthetic chemists. This experimental approach has had significant success in the past, but is ultimately time-consuming and costly due to the nearly limitless number of promising candidate materials. In this talk, I will present our application of high-throughput predictive computational approaches for rationally guiding experimental efforts for designing new semiconducting polymers [1-2]. In particular, we show that the use of first-principles calculations can dramatically reduce the number of promising synthetic targets for experimental design and gives a rational understanding of how different chemical functional groups modulate electronic properties to ultimately guide the organic synthesis. From our computational screening methodology, we observe that a unique pattern of aromatic/quinoidal effects naturally arises, which directly controls the band gap and mobility of these conjugated systems. Finally, based on our first-principles calculations, we draw attention to a vinylene-linked, benzannulated pyrrole polymer as a low-band-gap material with extremely interesting electronic properties.
[1] B.M. Wong and J.G. Cordaro, J. Phys. Chem. C, 115, 18333 (2011).
[2] J.D. Azoulay, Z. A. Koretz, B.M. Wong, and G.C. Bazan, Macromolecules, 46, 1337 (2013).

The need for computational materials and chemistry codes to utilize effectively high-end computing platforms is often taken as read; there is no doubt that this presents a major challenge for the discipline. In 2012, leading scientists from Europe across all major user disciplines updated the Scientific case for HPC in a White Paper entitled “The Scientific Case for High Performance Computing in Europe 2012-2020; from Petascale to Exascale”. This wide-ranging report captured the current and expected future needs of the scientific communities through the conclusions of five sector-based panels. We focus here on the needs in Materials Science; while Exaflop machines are essential and higher computational power will enable significantly increased accuracy, some important fields will be limited by throughput and data management.
Just as the drive to exascale platforms remains an elusive goal, so commodity-based clusters continue to provide a cost effective solution for the material scientist and computational chemist. Detailed cluster architecture - node and interconnect - determines the ability of such clusters to address the differing demands from the fields of capability and capacity computing. We consider the performance of a number of key codes - NWChem, GAMESS-UK, ONETEP, CASTEP, Quantum Espresso, DLPOLY and GROMACS - on a variety of HPC systems. We identify the challenges in scaling to large processor counts and the associated bottlenecks.
Finally, just as performance is perhaps the key metric for experienced practitioners, so to the novice user ease-of-use is arguably of greater importance, with the availability of web-based scientific gateways providing an essential tool. We consider the availability of such resources as part of the HPC Wales High Performance Computing (HPC Wales) service in the UK. Funded to provide computational services to industry and SMEs in particular, we outline the services provided to the Advanced Materials and Manufacturing communities.

In quantum dots (QD), the excitons are spatially confined and their energy spectrum, which controls many physical properties of interest, can be adjusted over a wide range by tuning composition, density, size, lattice strain and morphology. The formation of QDs joined by a thin flat wetting layer, known as the Stranski-Krastanow (SK) morphology, is a general growth mode observed in many epitaxially-strained thin solid films. These features make semiconductor QDs attractive for the design and fabrication of novel electronic, magnetic and photonic devices. The success of this endeavor has mainly been enabled by research leading to reliable means for estimating forces in small material systems and by establishing frameworks, in which the integrity and/or functionality of the systems is satisfied. The material failure continues to be a main technology-limiting barrier and thus, the subject of capillary-driven morphological evolution of surfaces and interfaces, especially under the action of applied force fields e.g., electrostatic and thermo-mechanical, is still a challenging materials problem. In such nano-scale systems, the magnitude of the surface roughness, diffusion anisotropy and texture orientation may have a significant influence on the thin film surface evolution. Here we demonstrate the effects of anisotropic material properties (the crystal texture, surface Gibbs free energy, diffusivity) and strain relaxation on morphological evolution of QDs and occurrence of wetting layer. Our study based on continuum level dynamical simulations will be presented for the spontaneous evolution of an isolated thin solid droplet on a rigid substrate under stress field by utilizing various combinations of the surface texture and the direction. The simulations showed that there is a threshold value for the stress level under which the formation of isolated islands observed; whereas at higher stress levels we observed the formation of SK-type islands connected with a very thin wetting layer. We also showed that with controlling the crystallographic orientation of the deposited film, thus the anisotropic surface properties, it is possible to obtain various QD morphologies ranging from semi spherical to pyramidal nano structures. Supported by TUBITAK grant no 111T343 and TUBA GEBIP.

We investigate, both by experiments and simulations, how grain boundary (GB) network configurations affect macroscopic properties of polycrystalline materials using gallium permeability through GBs in aluminum as a case study.
Our in situ experiments show that gallium permeability proceeds non-uniformly through the aluminum GB network and that the penetration velocity is a function of the GB crystallography.
The form of such a function is determined by simulating the diffusion kinetics for different combinations of GB crystallographic parameters until the result matches the experiments.
A similar approach may be useful in predicting properties of polycrystalline materials that are governed by mass transport through GBs, such as short-circuit ionic conductivity in solid oxide fuel cells, electromigration in interconnects and lithium transport in battery electrodes.
This work was supported by the US Department of Energy, Office of Basic Energy Sciences under award No. DE-SC0008926.

The need for affordable growth techniques in GaN that improve crystallinity and reduce native defects, leads to exploration of polymers suitable for ceramic conversion, with adequate chemistry and crystallography. Following a “materials by design” paradigm, we have produced polycrystalline GaN fibers by the polymer-derived-ceramic (PDC) technique. The first stepping-stone in this approach revolves around the exploration of cellulose acetate as carrier polymer of the metal-rich polymeric mixture suitable for fiber shaping, crosslinking, polymer-to-ceramic conversion, and crystallization.
Post-ceramidization TEM analysis revealed wurtzite-polymorphic fibers resulting from complex nucleation and grain growth mechanisms, being mostly unconstrained during initial polymer-to-ceramic conversion and reaching a competition stage later on. Determining parameters towards preferential nucleation sites and subsequent crystallographic configurations point at polymer architecture. Beyond fluid dynamics during electrospinning, this works highlights the importance of carrier polymers, as their templating role is likely, at first instance, to regulate nucleation centers and molecular dynamics involved during early aggregation. The role of pre-ceramic polymer synthesis is well established, and a good understanding exists, on the transformation dynamics of polymers to produce Si-based ceramics. Our work suggests similar knowledge is needed towards an optimized synthesis of GaN with special attention to molecular dynamics between carrier polymer and metallic molecules. Optical characterization of PDC-GaN revealed weak cathodoluminescence band-to-band recombination at room temperatrue and active luminescence centers mostly, from shallow donors, to which N vacancies could be contributing.
Albeit, non-incremental value of PDC-GaN will be fully unveiled upon a combinatorial ”synthesis by design” paradigm. This approach would correlate carrier polymer architectures and resulting microsctructures by understanding grain nucleation, growth, and point defect generation. In particular, this study suggests that grain size control, and eventually maturation to single crystal, as well as improved N adsorption could be key efforts.

As an imaging instrument, the helium ion microscope (HIM) offers a unique combination of resolution, contrast, and surface sensitive imaging. Using the high brightness helium ion source, the helium ion beam generates raster-type images with sub-nanometer resolution. Early adopters have used this instrument for the imaging of nanoscale features on a broad range of materials including electrical insulators such as glass, ceramics, polymers, and bio-materials. Examples will be presented.
Based upon these imaging results, there has been a growing desire to introduce nano-scale fabrication and nano-scale analysis capabilities on the same instrument. The newly released ORION NanoFab allows several new capabilities to allow researchers to fabricate, image and analyze all on one platform. The new features include heavy ions beams of neon and gallium for sputtering with high precision (< 2 nm probe size) and high rates. Also included on the same platform is a focused electron beam to allow for e-beam inspection, and the generation of analytical X-Rays for elemental identification. Other accessories include a pattern generation capability, a gas injection system, and an EDS detector. Also under development are a SIMS (secondary ion mass spectrometry) capability at the 10 nm length scale. Simulations, and early experimental results will be presented.
Together, this set of technologies on a single platform allow for in situ fabrication using the helium beam, neon beam and gallium beam. Fabrication can include sputtering, beam assisted deposition or etching, and lithography. Imaging of the result can be achieved with the surface sensitive helium beam, or the electron beam. Finally, elemental analysis will be made possible with the SEM EDS and Neon based SIMS.

Size exclusion chromatography (SEC) is the main method of determining mean molar masses and molecular mass distributions for a wide variety of complex polymers and macromolecules, and has remained in heavy use despite a lack of reliability and precision for new, advanced materials. These drawbacks are due to the size-based mechanism of separation intrinsic to the process. As a complimentary separation process, liquid adsorption chromatography at critical conditions (LACCC) is simultaneously promising and problematic because of the large parameter space presently available for customization of surface properties and the ternary optimization of solvent, solute, and surface necessary to find ideal separation conditions; namely, at critical conditions where the mechanism of separation becomes independent of molecular weight, depending instead on the type of repeat unit. By creating 2D model substrates using polymers grafted to the interface, we create an analogous system in which the adsorption process near critical conditions can be observed directly. In this way, we gain fundamental insight into the contribution of the interface structure and chemistry to the LACCC separation process, which will be useful in the rational design of stationary phases for this technique.
Thin polymer film swelling is in principle observable by a number of techniques, such as ellipsometry or quartz crystal microbalance with dissipation. However, we intend to observe the nanometer-scale shifts in polymer conformation at low grafting density. Therefore, we use neutron reflectivity to observe the process in the initial stages because of its relatively high contrast and sensitivity. We demonstrate the relevance of this technique by emulating published polystyrene LACCC conditions in cyclohexane and dimethylformamide. We then calculate a dimensionless interaction parameter directly related to the free energy of adsorption, and compare it to the adsorption partition coefficient derived from the literature LACCC experiment. Indeed, such direct measures of interaction parameters and partition coefficients are critical in descriptors in a wide variety of interfacial dynamic processes.

An important goal of synchrotron characterization techniques such as x-ray absorption (XAS), non-resonant inelastic scattering (NIXS) and pair distribution function (PDF) is to determine the configuration of atoms in the sample. These techniques are particularly important when x-ray diffraction cannot be applied due to poor or non-existent crystallinity. Efficient and effective approaches to determine atomistic configurations from these synchrotron characterization results will allow an acceleration of materials research and design, as well as maximize the utility of in-situ experiments. In this talk, we will discuss approaches that combine first principles computation and machine learning algorithms to allow efficient determination of atomistic configurations from synchrotron characterization data.

The large penetration depth of a hard X-ray probe is interesting in a broad range of applications in materials science research because of its bulk sensitivity and the compatibility with in-situ conditions. Most hard X-ray techniques, however, address the atomic structure while analysis of the electronic configuration is hampered by the large spectral broadening and only few tools for theoretical analysis are available. The presentation will discuss the information that can be obtained on the electronic structure using resonant hard X-ray emission spectroscopy. The various techniques including magnetic circular dichroism will be briefly introduced and recent results on Pt, CeO2 and Fe3O4 nanoparticles will be presented.
The full information content in inner-shell spectra can only be accessed by theoretical modeling of the data. We used density functional theory (FEFF, ORCA, Wien2k) codes and ligand field multiplet theory to simulate the experimental data. We find that many spectra can be modeled using a surprisingly simple approach. This allows for a detailed analysis of the electronic structure. The limitations of this approach are discussed.

This paper focused on size/composition/shape/structure and function relationship in C-H bond activation using noble, transition metal and doped oxides. Our experimental studies are based on 1) fabrication of technologically relevant supports, 2) physical size- and composition selected cluster deposition with atomic precision control, 3) ex situ and in situ microscopies and 4) in situ synchrotron X-ray characterization of cluster size/shape and oxidation state under realistic working conditions, combined with mass spectroscopy analysis of the reaction products. The experimental studies are complemented with DFT calculations. Using model size-selected clusters and cluster synthesized by wet chemical routes, this contribution will outline the applicability of the outlined approach on example of selective C-H bond activation in the oxidative and non-oxidative dehydrogenation of cyclohexane and cyclohexene on subnanometer cobalt oxide clusters, cluster-based assemblies, and nanometer size Co3O4 and (Au)Co3O4 particles under both oxidative and non-oxidative conditions.
References
1. “Oxidative Dehydrogenation of Cyclohexane on Cobalt Oxide (Co3O4) Nanoparticles: The Effect of Particle Size on Activity and Selectivity”, by E. C. Tyo, C. Yin, M. Di Vece, Q. Qian, S. Lee, B. Lee, S. Seifert, R. E. Winans, R. Si, B. Ricks, S. Goergen, M. Rutter, B. Zugic, M. Flytzani-Stephanopoulos, Z. Wang, R. E. Palmer, M. Neurock, and S. Vajda, ACS Catal. 2, p. 2409minus;2423 (2012)
2. “Support-Dependent Performance of Size-Selected Subnanometer Cobalt Cluster-Based Catalysts in the Dehydrogenation of Cyclohexene", S. Lee, M. Di Vece, B. Lee, S. Seifert, R. E. Winans and S. Vajda, Chem. Cat. Chem. 4, p. 1632-1637 (2012)
3. “Oxidative Dehydrogenation of Cyclohexene on Size Selected Subnanometer Cobalt Clusters: Improved Catalytic Performance via Evolution of Cluster-Assembled Nanostructures", S. Lee, M. Di Vece, B. Lee, S. Seifert, R. E. Winans and S. Vajda, Phys. Chem. Chem. Phys., 14, p. 9336 - 9342 (2012)
4. “Subnanometre Platinum Clusters as Highly Active and Selective Catalysts for the Oxidative Dehydrogenation of Propane”, S. Vajda, M. J. Pellin, J. P. Greeley, C. L. Marshall, L. A. Curtiss, G. A. Ballentine, J. W. Elam, S. Catillon-Mucherie, P. C. Redfern, F. Mehmood and P. Zapol, Nat. Mater. 8, 213-216 (2009)
5. “Improved Oxidative Dehydrogenation of Cyclohexane via Decoration of Cobalt Oxide (Co3O4) Nanoparticles with Gold Clusters”, E Tyo, M. DiVece, S. Lee, Z. Wang, R. Palmer, Q. Qian, M. Neurock, S, Vajda et al., in preparation

Information from traditional ex situ characterization techniques falls short in accurately describing complex materials systems, such as solid oxide fuel cells, which require specific conditions for optimal operation. Operando x-ray absorption spectroscopy (XAS) fills the gap by providing element-specific information on the chemical phenomena responsible for performance deficiencies. A La0.6Sr0.4Co0.2Fe0.8O3-δ (LSCF) thin film quasi-symmetrical cathode cell was tested in air contaminated with H2O and CO2 at 400 and 700°C and simultaneously examined with glancing angle operando XAS using a custom built testing platform, which is also described. Whereas impedance spectroscopy indicates increased polarization resistance in the cathode in both atmospheres at 750°C, XAS near-edge and extended fine structure data indicated that the oxidative behavior of H2O and CO2 on Fe and Co cations have diverging behaviors with increasing temperatures. In particular, the degree of oxidation according to absorption edge shift varies with temperature and cation and comparison with the impedance data showed some correlation between the degree of oxidation and amount of increase in the polarization resistance as a function of temperature. The local atomistic structure of the Co was also observed to be more severely affected by H2O and CO2 than that of the Fe, which showed stability in its local structure under all of the testing conditions.
Synchrotron-based in situ x-ray photoelectron spectroscopy was performed to complement the study with surface-specific information. Spectroscopy at high temperature and post-gas exposure discovered the removal and formation of a carbonate species, the irreversible formation of a new Co phase, and the related evolution of different oxygen moieties, such as surface, lattice, and hydroxide oxygen, all as a function of temperature and atmosphere. The combined information suggests that the segregation of the Co cation from the perovskite phase into an oxide phase via the formation and subsequent decomposition of a carbonate phase at operating temperature is strongly suspected as a mechanism by which H2O and CO2 cause degradation in the LSCF cathode.
Additionally, a comparison of the results from operando and in situ testing revealed that cathodic bias has some interaction with the CO2 gas. While cathodic bias caused a slight reduction of Fe and Co and CO2 gas caused an oxidation of Fe and Co, the combination of those parameters in the operando experiment caused more severe oxidation of both Fe and Co. In conclusion, the operando results cannot be predicted by their in situ components, strengthening the rationale for conducting operando experiments, despite their challenges.

Most of the electrochemistry processes occur within the thin layer of electrolyte at the electrolyte / electrode interfaces, commonly denoted as the electrical double layer (EDL). In spite of some classic EDL theories, very limited experimental information is available about these solvent or solute species within such EDLs. We have developed in-situ liquid cells to study such electrolyte / electrode interfaces by means of soft x-ray absorption spectroscopy. Because the fluorescence x-ray photon has much larger mean free path in condensed matters than the secondary electrons, by comparing the total fluorescence yield (TFY) and total electron yield (TEY) spectra, we can extract useful information about the compositional, structural or chemical difference between the bulk and the interfacial electrolyte. Under different bias, by using a mechanical chopper system to modulate the incident x-ray, the TEY current becomes alternating. This allows us to separate the tiny AC TEY current from the dominant faradaic current so that we can obtain surface-sensitive TEY signal under electrochemical conditions.
With this in-situ and operando XAS technique, we characterized the interfacial water at the gold electrode surfaces. It was found that the interfacial water layer has significantly different hydrogen-bonding network structure compared to the bulk water. Under different bias, the polar water molecules will respond to the external electrical field and reorient at the gold electrode surface, which significantly changes the amount of distorted or broken hydrogen bonds. Copper under-potential deposition (UPD) on gold electrode was also studied to demonstrate the surface-sensitivity of such XAS technique.

[1]. J. J. Velasco-Velez, C. H. Wu, M. B. Salmeron, in preparation.

The extraordinary mechanical, thermal, electronic and optical properties of carbon nanotubes (CNTs) have garnered considerable interest in a wide range of academic fields. CNTs are attractive candidates as fillers in composite materials, considerably improving the properties of the host. Recently, vast research has been undertaken to investigate the active properties of some polymer-CNT composites, with potential applications ranging from Micro-opto-electro-mechanical systems (MOEMS) to neuroscience. The development of this field is hindered, however, as the mechanics of actuation at the molecular level in these composites is still not well understood and a unified model explaining this behaviour is missing. Consequently, there is no method to predict whether specific materials exhibit this “smart behaviour”.
Near edge X-ray absorption fine structure (NEXAFS) spectroscopy is a valuable synchrotron technique for the study of nanocomposites, offering a wealth of information about local chemistry of the system, all with excellent energy resolution. Additionally, the use of a highly polarized beam allows for assessment of conformational arrangement of soft matter, which is of critical importance when the system is governed by non-covalent interactions.
In this study, films of thermally active EVA|MWCNT composites are studied through in-situ NEXAFS spectroscopy in an attempt to correlate spectral progression of the system with macroscopic observations of mechanical actuation. The effects of straining to promote CNT alignment and the role of dispersants are also addressed. Systematic variations in NEXAFS spectra with increasing temperature identify emissions from specific chemical groups as possible actors in the actuation mechanism, and suggest a high degree of conformational effects. An actuation model based on molecular orientation is later proposed which supports our findings. These are the initial steps towards forming a systematic database aiming at active behaviour prediction of polymer nanocomposites, and the production of materials by design.

X-ray Absorption Spectroscopy (XAS) is one of the core competencies of synchrotron science. With sensitivity both to chemical state and to the local configurational environment, the unique view of matter offered by XAS has become an indispensable tool for a broad range of scientific disciplines. With the advent of high-flux, third-generation synchrotron sources, new experiments probing the structure of inner shell electrons are now possible. In this talk, I will discuss recent developments in crystal-based spectrometry. Among the new capabilities are XAS with energy below the level of the core-hole lifetime, X-ray emission spectroscopy providing a complementary view of electromic and chemical state, X-ray Raman spectroscopy which allows access to low energy absorption edges and dipole forbidden transitions, and XAS measured at ultra-low absorber concentration. Special emphasis will be placed on instrumentation under development for beamlines at NSLS-II.

The perhaps most exciting characteristic of the scanning tunneling microscope (STM) is its exceptional sensitivity to surface states combined with unprecedented atomic resolution. The STM probes the filled and empty density of states a few Å above the surface in the vacuum. It is generally assumed that bulk states rapidly decay into the vacuum region, whereas the decay of the surface states is expected to be slower. Hence, surface states should dominate when measuring the electronic structure by an STM. Thus, this technique is considered to be an ideal tool to identify the energetic position of surface states, which in turn govern the physical properties of the surfaces. However, a controversial debate arising from previous STM experiments and first principles calculations on the GaN (10-10) surface raises the question to what extend STM is sensitive enough to surface states. In order to address the aforementioned discrepancy we combine scanning tunneling spectroscopy and first principles calculations and we demonstrate that STM may fail to probe the physically most relevant parts of the density of states of a surface, i.e. the energetic position of the surface state and the band edges, and thereby suggest wrong physical properties, such as surface band gaps and origins of Fermi level pinning [1].
Thick slabs consisting of 48 GaN layers were used in order to describe both the surfaces stares and the states at the bulk CBM as well as their dispersion and decay into the vacuum. Our calculations clearly show that (i) the empty surface state is in the fundamental band gap, (ii) is characterized by strong and flat dispersions around the Γ point and at the edges of the BZ, and (iii) at the onset has a very small density of states as compared to bulk states, such that conventional STM mapping modes fail to probe it. We further analyzed the decay of the partial charge densities of the states at the onset of the surface state and at the bulk CBM. We find that the latter has 2 orders of magnitude stronger signal than the surface state minimum. Using the ab initio computed decay lengths of the various states we designed a tip-sample distance ramp approach. The experiments are in excellent agreement with the calculated band structure and confirm the presence of an intrinsic surface state within the fundamental band gap as well as the dominant character of the surface state only at small tip-sample separations. The above results demonstrate that in order to probe the true and physically most relevant parts of the density of states by STM, it is compulsory to consider in detail the effects of the band dispersion and decay of the individual states into the vacuum and design appropriate measuring modes. This proposed approach is general and will be useful to identify surface and defect states whenever they are characterized by large dispersions and strong energy dependencies in the DOS.
L. Lymperakis et al., APL 103, 152101 (2013).

Observing reacting single atoms on solid catalyst surfaces under controlled reaction conditions of gas environment and elevated temperature is a key goal in understanding heterogeneous gas-solid catalytic reactions. In-situ real time aberration corrected environmental (scanning) transmission electron microscopy (E(S)TEM) provides new information from direct analysis of dynamic surface and sub-surface structures of reacting catalysts. In this paper in-situ AC ETEM and AC ESTEM studies of oxide catalysts and supported metal nanocatalysts important in the chemical industry are presented. They provide direct evidence on the atomic scale of dynamic processes at oxide catalyst surfaces and single atom dynamics in catalytic reactions. The ESTEM studies of single atoms show that nanoparticles act as reservoirs of ad-atoms. The results have important implications in catalysis and nanoparticle studies. They include the possibility that single atoms are themselves acting as unique catalytic species as well as reflecting atoms in transit between competitively growing nano-particles. With the new analytical technology it is in principle possible to establish the nearest neighbour environment and hence bonding of each atom present discreetly, in loosely ordered rafts and as part of more established crystal structures. These data can then be co-related with chemical reaction and catalysis performance results, and used to understand, control and improve them. Nanoparticle catalyst nanostructures are inherently unstable and over time both the selectivity and activity may change; usually for the worse. There are large commercial and even economic, as well as environmental, benefits from understanding, controlling and mitigating changes in structure and performance of existing solid state catalysts and to help to stimulate innovative new solutions on the basis of atomic scale studies informed by the new AC ESTEM development.

GaN is of great interest in optical and high power devices. It has a wide, direct band gap, shows strong chemical stability, and can withstand high electric breakdown fields. Innovative technologies aim at ever-smaller devices, with features in the nano-dimensions being achieved. However, devices at the nano-scale are strongly affected by fluctuations promoted by applied fields, trapped internal charges, and surface screening. It is therefore necessary to build a comprehensive model describing charging dynamics to fully understand opto-electronic behaviours typical of in operando GaN.
In this work, an in operando device is presented, where a grounded Au-GaN interface is bombarded by an electron beam. In-situ, time-resolved cathodoluminescence (t-CL) and specimen current (SC) have been simultaneously monitored during electron beam irradiation, which promoted distinctive secondary electron contrast from the irradiated regions (SE-IR). Under extreme irradiation conditions the size of SE-IR evolved non-linearly with magnification and increased with irradiation time. Under moderate irradiation conditions, all SE, SC, and t-CL were affected by electron beam parameters environmental exposure. The interplay of all three factors: continuous e-beam irradiation, size (and sign) of SE-IR, and environmental history have been modelled by a 2D transistor to explain SC evolution and SE emission, and to infer mechanisms responsible for near band edge-CL (NBE-CL) degradation upon e-beam bombardment. The proposed 2D transistor explains experimental data (SC and SE-IR) through modifications of Schottky barrier height and depletion width at the Au/GaN interface in the different scenarios. The model also predicts the qualitative behaviour of time constants in NBE-CL. The proposed 2D transistor model aims at providing parameters conducive to optimized operation in GaN devices, where the NBE electron-hole recombination rates are minimally affected by device charge transport and electric field dynamics. This approach will provide an optimized “device by design”, in terms of operation parameters and environmental exposure, crucial towards radiation-hard electronics and microsystems in harsh environments.

The demand for emerging materials in our society can no longer be fulfilled by simple build - test algorithms. Materials by design has emerged as a means to design materials using computational modeling that considers atomistic to continuum length scales. In situ mechanical measurements can be performed on these same length scales, acting in both predictive and substantiation roles. These experiments are aided by advanced instrumentation, where a single device can explore more than six orders of magnitude in force and displacement, and five in the time domain. This is then combined with imaging techniques before, during, and after testing. Structure property relationships are explored through targeted compression, tension, scratch, and indentation tests. These tests yield the basic material parameters; elastic modulus, strength, and fracture toughness, starting with a variety of ideal materials, simple metals, silicon, and HOPG, then working towards more complex systems of interest to the engineering community. Further experimentation leads to insight on deformation mechanisms and the ultimate mode of failure, acting as input parameters for further simulation.

Micromachined nanocalorimetry sensors have shown excellent performance for high-temperature and high-scanning rate calorimetry measurements. Scanning AC nanocalorimetry allows thin-film calorimetry measurements at rates from nearly isothermal to as high as 2,000 K/s, bridging the gap between traditional calorimetry (<1 K/s) and DC nanocalorimetry (>1e3 K/s). Here, we combine scanning AC nanocalorimetry with in-situ x-ray diffraction (XRD) to facilitate interpretation of calorimetry measurements. Time-resolved XRD during in-situ operation of nanocalorimetry sensors using intense, high-energy synchrotron radiation allows unprecedented characterization of thermal and structural material properties. We demonstrate this experiment with detailed characterization of the melting and solidification of elemental Bi, In, and Sn thin-film samples. We apply a similar technique to characterize the glass forming ability of ternary Au-Cu-Si alloys using a combinatorial approach and show that the critical cooling rate changes by an order of magnitude over a very small range of compositions. The measurements are performed using a device that contains an array of individually addressable nanocalorimetry sensors. Combined with XRD, this device creates a new platform for high-throughput mapping of the composition dependence of solid-state reactions and phase transformations.

Developing new functional materials is of prime importance for achieving efficient and large-scale energy storage. This task also requires the development of new in-situ experimental techniques to analyse performance and identify bottlenecks. Interpretation of in-situ data can benefit greatly from first principles calculations. In this talk, we outline our recent success in employing a novel form of density functional theory for the treatment of systems in contact with a liquid environment, joint density-functional theory (JDFT), to interpret results from in operando valence electron-energy-loss spectroscopy. Specifically, we study images of lithium ions cycled between a nanocrystalline lithium-iron-phosphate cathode and an aqueous electrolyte by developing a unique combination of nonlinear polarizable continuum models with a new excited-state Kohn-Sham framework. We find that this combination gives excellent agreement with experiment for electronic excitation energies of molecules in electrolyte environment.

Transparent conductive oxides (TCOs) is a technologically significant class of materials that are commercially used in optoelectronic devices such as solar cells, flat-panel displays, light-emitting diodes. The prototypical compounds are In2O3:Sn (ITO), ZnO:Al (AZO) and SnO2:F (FTO). Despite the heavy industrial use of these materials, the physical origins of simultaneous high optical transparency and large electrical conductivity remain controversial. For example, undoped indium oxide thin films have been recently demonstrated to have thickness-dependent high conductivity that results from surface donor-type defects [1], challenging the traditional oxygen vacancy donor defect model.
In this contribution we reveal the non-equilibrium origin of high electrical conduction in gallium zinc oxide ZnO:Ga (GZO) thin films using complementary contributions from high-throughput combinatorial experiments, first-principles modeling and in-situ electrical transport measurements as a function of temperature (T) and oxygen partial pressure (pO2) [2]. Specifically, gallium zinc oxide thin film sample libraries prepared using combinatorial pulsed laser deposition with temperature gradient across the glass substrate show that the maximum electrical conductivity occurs in low-pO2 and low-T regime. In contrast, constrained-equilibrium first-principles theoretical defect model suggests that that conductivity should increase with increasing temperature. The in-situ electrical measurements of high-quality ZnO:Ga layers on both amorphous and crystalline substrates reconcile these two contradicting observations by discovering a non-monotonic dependence of Ga:ZnO electrical conductivity on temperature. Starting from room temperature at ambient pressure, the conductivity first drops by 5 orders of magnitude up to 500C, but then the trend reverses and matches well the theoretically predicted dependence.
The results of these studies indicate that high electrical conductivity of commercial-grade Ga:ZnO TCO occurs by virtue of its non-equilibrium state. This discovery calls for development of new theoretical and experimental tools for predictive design of metastable materials. This work also exemplifies the importance of complementary iteration between combinatorial experiments, theoretical modeling and in-situ characterization for materials by design.
This research was supported by U.S. Department of Energy
[1] PRL 108 016802 (2012)
[2] APL, in press (2013)

Methods and software infrastructure for computational catalyst screening are presented. A special focus will be to illustrate how one can improve the reliability by including quantitative error estimation. A fitting methodology for empirical exchange-correlation functionals, which reproduces catalysis-relevant benchmark properties reasonably while allowing for quantitative error estimates, is presented. Methods for finding the morphology of the catalytic materials under reaction conditions together with models allowing the inclusion of adsorbate-adsorbate interactions in mean-field microkinetics are discussed.

Nanocatalysis, which combines the design of heterogeneous catalysts (monodispersity, shape control, presence of ligands) with the monitoring of their surface activity using new microscopic and spectroscopic methods, has recently emerged as a major new field for the rational design of improved catalysts. Provided that understanding at the atomic level reactive processes that occur on their surface allows the fine-tuning of nanocatalysts, first-principle calculations can guide the conception of nanocatalysts, both in terms of activity and selectivity.¹ Here we show relevant theoretical descriptors for adsorption strength based on the d-band center model of Hammer and Noslash;rskov.² This conceptual DFT approach, in line with the Sabatier principle and the Brönsted-Evans-Polanyi relationship, could be a useful guide to design efficient nanocatalysts with sites having a specific activity. According to the qualitative concept of Paul Sabatier, catalytic properties will be hindered if the reactants adsorb too strongly, whereas no reaction will occur if the interaction is too weak. The Sabatier principle applied to the NP case is illustrated by the monoelectronic descriptors introduced in this work, which will be depicted as color maps which give a straightforward point of view of possible reactivity spots.³ The main outcome of this approach is the definition of effective d atomic orbitals for each surface atom, which energy depends on the field generated by the other metal atoms and by the surface ligands. On the basis of our DFT calculations, we conjecture that a catalytic activity will be maximized by optimizing the energy of the effective d AOs of the active sites. This can be achieved by modulating the ligand-field exerted on the surface metal atoms. Some examples will be shown for cobalt and ruthenium nanoclusters.
[1] Gerber, I. C. & Poteau, R., Theoretical nanocatalysis: where are we now? in Nanomaterials in catalysis, Serp, P. and Philippot, K. (ed.) Wiley-VCH, 2013
[2] Hammer, B. and Noslash;rskov, J. K., Electronic factors determining the reactivity of metal surfaces, Surf. Sci., 1995, 343, 211-220.
[3] del Rosal, I.; Mercy, M.; Gerber, I. C. & Poteau, R. Ligand-Field Theory-Based Analysis of the Adsorption Properties of Ruthenium Nanoparticles, ACS Nano, 2013, 10.1021/nn403364p.

New selective catalysts have been conceived by using results from studies on model systems, with the help of both modern surface-science techniques and quantum mechanics calculations, and developed by using colloidal and other types of self-assembly chemistries. An example of this approach will be provided where catalysts consisting of dispersed platinum metal nanoparticles with narrow size distributions and well-defined shapes were prepared and tested for the selective promotion of carbon-carbon double-bond cis-trans isomerization reactions in olefins. It was shown that the selective formation of the cis isomer could be controlled by using tetrahedral particles with exposed (111) facets. A discussion of the surface science and DFT calculations used to predict and explain this behavior will be presented.

Recent advances of in situ experimental techniques allow researchers to probe the solid-liquid interface in electrochemical systems under operating conditions. These advances offer an unprecedented opportunity for theory to predict properties of electrode materials in aqueous environments and inform the design of energy conversion and storage devices. To compare with experiment, these theoretical studies must model microscopic details of both the liquid and the electrode surface. Joint Density-Functional Theory (JDFT), a computationally efficient alternative to molecular dynamics, couples a classical density-functional, which captures molecular structure of the liquid, to a quantum-mechanical functional for the electrode surface. JDFT can compute thermodynamic averages, such as solvation free energies in aqueous and non-aqueous electrochemical environments, with accuracy comparable to thermal fluctuations. Leveraging this accurate and microscopically detailed description of the physics of solvation, we present a JDFT exploration of SrTiO₃, which can catalyze solar-driven water splitting, in an electrochemical environment. We determine the geometry of an activated SrTiO₃ (001) surface and explore why this novel geometry is correlated with higher activity for water splitting. JDFT predictions of the specular X-ray crystal truncation rods for SrTiO₃ show excellent agreement with measurements performed during water splitting at the Cornell High Energy Synchrotron Source (CHESS).

Solar thermal fuels (STF) store the energy of sunlight, which can then be released later in the form of heat, offering an emission-free and renewable solution for both solar energy conversion and storage. However, this approach is currently limited by the lack of low-cost materials with high energy density and high stability in the charged state. Previously we have predicted a new class of functional materials that have the potential to address these challenges [Kolpak and Grossman, 2011]. Recently, we have developed an ab initio high-throughput screening approach to accelerate the design process and allow for searches over a broad class of possible candidate materials. The high-throughput screening algorithm we have developed can run through large numbers of molecules composed of earth-abundant elements, and identifies possible metastable structures of a given material. Corresponding isomerization enthalpies associated with the metastable structures are then computed. Using this high-throughput simulation approach, we have discovered molecular structures with high isomerization enthalpies that have the potential to be new candidates for high-energy density STF. We have also discovered physical design principles to guide further STF materials design through the correlation between isomerization enthalpy and structural properties from our result sets.

Boron subphthalocyanine chloride is a promising organic photovoltaic donor material, having the highest reported open circuit voltages among bilayer OPVs when coupled with C60. We designed analogues of this molecule where boron and chlorine were substituted with other trivalent cations and halogen elements, respectively. Structure and electronic properties were explored using Density Functional Theory (DFT) with added Van der Waals interactions. Relaxed molecular morphologies revealed and inverse relation with the degree of openness of the porphyrin ring umbrella and the diameter of the trivalent site. From the relaxed molecular morphologies, time-dependent DFT was used to compute the absorption spectra. Crystal structures were predicted for each designed molecule. The energies of electronic and optical excitations of the crystals were calculated using the GW/Bethe-Salpeter equation method. We found that exciton binding energies are strong in these materials (~0.5 eV for B-SubPc-Cl) and significantly affect the optical absorption spectra.

The bulk photovoltaic effect describes the ability of inversion symmetry breaking materials to produce intrinsic photocurrents and photovoltages. It was first observed over 50 years ago; unfortunately, early experiments revealed weak response, typically for light outside the visible spectrum. Recently, however, the multiferroic bismuth ferrite has been observed to produce strong photovoltaic response to visible light, suggesting that the effect has greater potential than previously thought and renewing interest. Additionally, we have previously demonstrated the ability to compute, from first principles, the bulk photocurrent using the theory of “shift current,” and have successfully reproduced experimental results in BTO and BFO. This ability has allowed for understanding of the structural and chemical properties generating large bulk photovoltaic response and the design of high-performing materials. In this talk we present three polar oxides with the LiNbO3 structure that we predict to have band gaps in the 1-2 eV range and very high bulk photovoltaic response: PbNiO3, Mg1/2Zn1/2PbO3, and LiBiO3. The first is has already been synthesized, and the others are very similar to known materials. All three have band gaps determined by cations with d10s0 electronic configurations, leading to conduction bands composed of cation s-orbitals and O p-orbitals. This both dramatically lowers the band gap and increases the bulk photovoltaic response by as much as an order of magnitude over previous materials, demonstrating the potential for high-performing bulk photovoltaics.

Next generation targets for inertial confinement fusion (ICF) experiments may require the presence of an ultra-low density (< 25 mg/cm^-3) carbon-based foam layer on the inside of the millimeter-sized spherical ablator shell. The foam layer has a thickness of 20-200 mu;m and serves as a scaffold to introduce high Z dopants for diagnostics or to define the shape of the deuterium-tritium fuel layer. Previously, we have demonstrated that foam-lined ICF targets can be made by using prefabricated ablator shells as a mold in which the foam film is casted by sol-gel chemistry while the shell is rotated.[1] This requires a suitable polymer-based sol-gel chemistry that can withstand the shear forces experienced during the coating process.[2, 3] To achieve uniform coatings on the inside of rotating shells, however, requires detailed understanding on how viscosity of the sol-gel precursor solution and the rotational velocity affect the uniformity of the coating.
To identify promising combinations of viscosity and rotational velocity for these coating experiments, we performed conventional computational fluid dynamics (CFD) simulations based on the Navier-Stokes equations (commercial code Star-CCM by CD-Adapco, Melville, NY, USA). The two-phase (air and liquid) system was simulated using the volume-of-fluid (VOF) method. Promising combinations of viscosity and rotational velocity were then tested using in-situ radiographic imaging at beamline 2BM at the Advanced Photon Source, Argonne National Laboratory, which provides the high X-ray flux and temporal resolution required for these experiments. Spatio-temporal image analysis provided crucial information on how the foam precursor liquid is distributed in the rotating shell. This information allowed us to optimize the coating process and to achieve the high coating uniformity required for this application.
Work at LLNL was performed under the auspices of the U.S. DOE by LLNL under Contract DE-AC52-07NA27344.
1. Biener, J., et al., A new approach to foam-lined indirect-drive NIF ignition targets. Nuclear Fusion, 2012. 52(6): p. 062001.
2. Dawedeit, C., et al., Tuning the rheological properties of sols for low-density aerogel coating applications. Soft Matter, 2012. 8(13): p. 3518-3521.
3. Kim, S.H., et al., Exploration of the versatility of ring opening metathesis polymerization: an approach for gaining access to low density polymeric aerogels. Rsc Advances, 2012. 2(23): p. 8672-8680.

CRUD (acronym for Chalk River Unidentified Deposit) is predominately a nickel-ferrite (NiFe2O4) deposit that accumulates on hot surfaces of nuclear fuel rods during reactor operation. The presence of CRUD modifies the heat transfer between the fuel rods and coolant and can induce localized corrosion of surface the fuel clad. Besides these unwanted effects boron, which is a strong neutron absorber, can accumulate within the CRUD, triggering shifts in the core neutron flux and fluctuations in the reactor power level. Therefore it is crucial to understand and predict the mechanisms by which B is trapped into the CRUD. As a first step, the incorporation of B as a point defect into the crystal structure of NiFe2O4 has been investigated using the Density Functional Theory (DFT) framework. To obtain the formation energies of various interstitial and substitutional B-defects, theoretical results have been combined with experimental thermo-chemical data. Assuming solid-solid equilibrium conditions, the main factors that limit the incorporation of B are (i) the relatively narrow stability domain in which the host NiFe2O4 is stable and (ii) the formation of ternary Fe-B-O and Ni-B-O compounds. The present study also investigates the incorporation of B assuming solid-liquid equilibrium between NiFe2O4 and the surrounding aqueous solution under conditions of pressure, temperature and pH characteristic to PWRs.

Understanding and controlling the behaviour of dislocations is crucial for a wide range of applications, from nano-electronics and solar cells to structural engineering alloys. Transmission electron microscopy (TEM) revolutionised materials science through its ability to directly image dislocations but is limited to applications involving thin, electron transparent samples. The diffraction of more penetrating X-rays provides a perfect complement. However, thus far the quantitative measurement of the strain fields due to individual dislocations, particularly in the bulk, using micro-diffraction has remained elusive. Here we report the first characterisation of a single dislocation in a freestanding 130 nm thick GaAs/In0.2Ga0.8As/GaAs membrane by synchrotron X-ray micro-beam Laue diffraction. A thin sample was chosen to allow a direct comparison with TEM images of dislocations within this sample. Our experimental X-ray measurements of lattice rotations and strains agree with textbook anisotropic elasticity solutions for single dislocations, providing one of few experimental validations of this fundamental theory. However, unlike TEM, our measurements are not limited to thin samples. Indeed, based on the experimental uncertainty in our measurements, we predict that 3D measurements of lattice strains and rotations due to individual dislocations in the material bulk are feasible provided a sufficiently small focal spotsize of the polychromatic X-ray probe beam can be achieved. These findings have important implications for the in-situ study of dislocation structure formation, self-organisation and evolution in the bulk.

Computational methods have become effective tools in development of new materials and processes, and the Calphad (Calculation of Phase Diagrams) method has been identified as an essential tool for Integrated Computational Materials Engineering (ICME). Calphad software is used within ICME either as a stand-alone tool or coupled with other software, such as software for diffusion, phase field or casting simulations. In addition to the use in ICME, results from Calphad calculations can be used to identify regimes for critical experiments and can be used to aid their interpretation.
The Calphad method was originally developed for the functional description of thermochemical properties and the calculation of phase equilibria in alloy systems. The strength of the Calphad method is that it is currently the only method for the calculation of multi-component, multi-phase systems with condensed phases exhibiting homogeneity ranges. Although the Calphad method is not truly predictive, descriptions of binary and ternary system can be combined for the prediction of multi-component systems since strictly quaternary phases are extremely rare in metallic systems. The approach used by the Calphad method is also well suited for the description of other phase-based properties, including diffusion mobilities, molar volumes and bulk moduli, and has the potential to treat electrical and thermal conductivity and even two-phase properties, such as interfacial energies.
Data are needed for development of model parameters used by the Calphad method, and also for validation of computational results. Classical collection of data from publications is time consuming and prone to introduction of errors. Appropriately designed data and file repositories can significantly accelerate the data collection process. In the fast developing world of information technology these repositories must be flexible enough to accommodate evolving data needs while being self-consistent in providing unique identifiers for materials.
Examples for the application of the Calphad method and prototype repositories and their data schemas developed at NIST for Calphad files and data used in Calphad assessments will be presented.

Incredible advances over the past decade in computational physics as well as high-throughput materials synthesis and characterization are enabling a fundamentally new approach to new materials discovery and development. This theory driven approach to the computational design of new materials and material properties aims to directly guide the experimental materials development rather than retroactively explaining the observed properties. This new paradigm for developing and discovering materials with specific desired functionalities and of understanding the property, structure, morphology, and composition relationships requires tight coupling of theory, experiment and characterization to accelerate the rate of materials evolution and innovation. Here, we present our recent and on-going work on the development of p-type transparent conducting oxides (p-TCOs) as a case study in the application of Inverse Design approaches to materials development.
A₂BO₄ spinel oxides can be classified into four Doping Types depending on the relative energy ordering and level within the gap of the cation anti-site defect acceptor and donor levels. Doping Type II (DT-2) spinels where the acceptor lies above the donor and, in addition, the donor lies in valance band, are natural host materials for p-type conduction. High-throughput theoretical screening finds Co₂ZnO₄ to be the prototype DT-2 material. In particular, the Co_Td site defect level is resonant in the valence band, making it electrically neutral thus allowing the electrically active Zn_Oh acceptor to yield p-type conductivity, independent of the concentration of Co_Td defects. Resonant elastic x-ray diffraction (REXRD) site occupancy measurements on bulk ceramic samples grown in air at 800 °C confirm this basic prediction. Further, intentional non-equilibrium growth to increase the Zn_Oh concentration due to either quenched-in cation-site-occupancy disorder or incorporation of excess Zn should be an effective doping strategy. Experiments using combinatorial co-sputtered Co_2-xZn_1+xO₄ thin film “libraries” with intentional composition gradients on 2”x2” glass substrates support this. More definitive REXRD site occupancy measurements on as-deposited and annealed Co₂ZnO₄ and Co₂NiO₄ films grown epitaxially on SrTiO₃ by pulsed laser deposition confirm this. Finally, 17 candidate extrinsic dopants were evaluated theoretically yielding Li, Mg and Ni as the most promising. Ni doping was tested via combinatorial co-sputtering and found to be effective. In fact, the conductivity of Co₂Zn_1-xNi_xO₄ increases monotonically with increasing Ni all the way to Co₂NiO₄ for which σ asymp; 100 S/cm.
This work is supported by the U.S. Department of Energy, Office of Science, BES, under Contract No. DE-AC36-08GO28308 to NREL as part of the DOE Energy Frontier Research Center "Center for Inverse Design".

Hierarchical, architected materials have the potential to be transformational for a variety of applications, providing components which simultaneously offer the best performance attributes of ceramics, metals, and plastics. Hierarchical materials are materials which concurrently realize functional features on multiple length scales (sub-micron up to a millimeter level). This allows large void space in a material structure to be filled with load bearing members, adding compliance to the material without significantly increasing the density. Researchers have pursued hierarchical material structures for almost 20 years, leveraging inspiration from both nature and architecture, but few have shown success in synthetically re-creating these structures at multiple length scales. Prior attempts to experimentally fabricate hierarchical structures using 3D printing have only produced structures at large (mm) scales while micro-fabrication techniques using expensive custom equipment to produce sub-micron structures do not translate well to high-volume, low-cost production. Palo Alto Research Center (PARC) has invented a novel approach for manufacturing large area hierarchical materials based on electrohydrodynamic film patterning (EHD-FP) that mitigates both of the aforementioned shortcomings.
EHD-FP enables the rapid fabrication of hierarchical, architected materials with features at multiple length scales while creating a process which can easily scale up to quickly create large area patterned films at low cost. This talk focuses on evaluating a series of two-dimensional (2D) EHD-FP kagome and triangular truss structures made with ultraviolet (UV) curable polymers. Using the EHD-FP process, films with hierarchical spatial features can be easily and rapidly created with low viscosity UV cross-linked polymers. The EHD-FP process also has the unique ability to infuse these hierarchical features with nanoparticles that are aligned during the patterning phase, providing nanoscale structure. Estimations of mechanical toughness and strength will be presented based on experimental tensile testing results. Experimental results will be compared with modeling expectations of improvements in material properties. A finite element model (FEM) in COMSOL®, a commercial simulation package, is used to further explore the impact of varying geometrical parameters and levels of hierarchy in the EHD-FP films in order to recommend an optimal subset of damage tolerant hierarchical material structures. Our results provide confirmation of some of the underlying mechanics for hierarchical materials, allowing for a path towards transformative, large area hierarchical materials with superior functionality over bulk constituents.

Advanced materials have been identified by the European Union as one of the Key Enabling Technologies (KETs) for the coming years. KETs are the cornerstone of the ‘Leadership in Enabling Technologies&’ priority of Horizon 2020 - the current multi-annual (2014 - 2020) framework programme for research and innovation of the European Union. The development of new, industrially viable advanced materials requires an efficient modelling approach to shorten the development process of materials-enabled products. This requires strong interaction between the different modelling communities and industry. Because of the complexity and long timescale of the code development and validation processes, the support of the European framework programmes such as Horizon 2020 makes an important contribution.
We will present Horizon 2020, with special attention to the place of materials research and innovation, as well as a brief overview of past and current European activities in the field of materials modelling and its use for the design of advanced materials with specific properties, leading to the development of materials-enabled products.

In the present work we show transformations of perovskite crystalline structures by changing the temperature in situ TEM using in a sample-holder which ranges from 90 to 390 Kelvin. The materials studied have been structures based on LaBaCo2O5.5 (LBCO) and BaTiO3 (BTO). The structural transformations have been evaluated by electron diffraction patterns taken in a JEOL ARM 200F microscope. As consequence of these structural changes the physical properties also are affected, in this way we have performed the off-axis electron holography technique to study the contribution of the electric and magnetic fields in the reconstructed phases obtained from the holograms. In addition, aberration-corrected scanning/transmission electron microscopy (STEM) and related analytical techniques were employed to understand the nature of these unusual giant magnetoresistance effect and anomalous magnetic properties the double perovskite LBCO structure. We have observed in LBCO a structural transformation a low temperature (90 K) which shows orderly oxygen vacancies and confirmed by line profile electron energy loss spectroscopy. Oxygen vacancy and nanodomain structures were studied by STEM. For BTO we observed evolution of its grain boundaries as well as structural changes by electron diffraction. The off-axis electron holography was used to obtain the reconstructed phase and to quantify the electric field contribution of individual BTO grains.

Electron tomography (ET) is a powerful tool for reconstruction of the 3D structure of materials at sub-nanometer resolution. Significant advances in ET instrumentation have been made in recent years, yet current reconstruction algorithms for inversion of the projection data do not properly model the image formation process, or incorporate prior knowledge about the material, and therefore yield poor results. Model Based Iterative Reconstruction (MBIR) provides a framework for tomographic reconstruction that incorporates a forward model which explicitly describes the imaging process, and a prior model for the object, to obtain reconstructions that are qualitatively superior to current methods such as Filtered Back Projection (FBP) and quantitatively accurate. Here we present a novel MBIR algorithm for multi-modal ET with forward models for both High Angle Annular Dark Field (HAADF) and Bright Field (BF) imaging modalities and a Generalized Gaussian Markov Random Field prior model. The algorithm also identifies and classifies anomalous measurements from Bragg scatter in the BF data, allowing for the use of BF-ET in the analysis of crystalline materials, which has not been possible to date. MBIR results on simulated as well as real data show that the method can dramatically improve reconstructions of HAADF and BF-ET data from crystalline and non-crystalline samples compared to FBP. Current and future work includes improving scalability of the algorithm in a high performance computing setting, and incorporation of more detailed physics-based models into tomographic data inversion process.