Critical to the development of improved solid oxide fuel cell (SOFC) technology are novel compounds with high oxygen reduction reaction (ORR) catalytic activity and robust stability under cathode operating conditions. We have screened approximately 2145 distinct perovskite compositions for potential use as high activity, stable SOFC cathodes, and have verified that our screening methodology qualitatively reproduces the experimental activity, stability, and conduction properties of well-studied cathode materials. We used the calculated oxygen p-band center as a first principles-based descriptor of the surface exchange coefficient (k*), which in turn correlates with cathode ORR activity. We used convex hull analysis under operating conditions in the presence of oxygen, hydrogen, and water vapor to determine thermodynamic stability. This search has yielded 52 potential cathode materials with good predicted stability in typical SOFC operating conditions and predicted k* on par with leading ORR perovskite catalysts. We also used our established trends in predicted k* and stability to suggest methods of improving the performance of known promising compounds. The materials design strategies and new materials discovered in our computational search will help enable the development of high activity, stable compounds for use in future solid oxide fuel cells and related applications.

We demonstrate automated generation of diffusion databases from high-throughput density functional theory (DFT) calculations. A total of more than 230 dilute solute diffusion systems in Mg, Al, Cu, Ni, Pd, and Pt host lattices have been determined using multi-frequency diffusion models. We apply a correction method for solute diffusion in alloys using experimental and simulated values of host self-diffusivity. We find good agreement with experimental solute diffusion data, obtaining a weighted activation barrier RMS error of 0.176 eV when excluding magnetic solutes in non-magnetic alloys. The compiled database is the largest collection of consistently calculated ab-initio solute diffusion data in the world.

Fast oxygen transport materials are necessary for a range of technologies, including efficient and cost-effective solid oxide fuel cells, gas separation membranes, oxygen sensors, chemical looping devices, and memristors. Strain is often proposed as a method to enhance the performance of oxygen transport materials, but the magnitude of its effect and its underlying mechanisms are not well-understood, particularly in the widely-used perovskite-structured oxygen conductors. This work reports on an ab initio prediction of strain effects on migration energetics for nine perovskite systems of the form LaBO3, where B = [Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Ga]. Biaxial strain, as might be easily produced in epitaxial systems, is predicted to lead to approximately linear changes in migration energy. We find that tensile biaxial strain reduces the oxygen vacancy migration barrier across the systems studied by an average of 66 meV per percent strain for a single selected hop, with a low of 36 and a high of 89 meV decrease in migration barrier per percent strain across all systems. The estimated range for the change in migration barrier within each system is ±25 meV per percent strain when considering all hops. These results suggest that strain can significantly impact transport in these materials, e.g., a 2% tensile strain can increase the diffusion coefficient by about three orders of magnitude at 300 K (one order of magnitude at 500 °C or 773 K) for one of the most strain-responsive materials calculated here (LaCrO3). We show that a simple elasticity model, which assumes only dilative or compressive strain in a cubic environment and a fixed migration volume, can qualitatively but not quantitatively model the strain dependence of the migration energy, suggesting that factors not captured by continuum elasticity play a significant role in the strain response.

Perovskites with fast oxygen ion conduction can enable technologies like solid oxide fuel cells. One component of fast oxygen ion conduction is low oxygen migration barrier. Here we apply ab initio methods on over 40 perovskites to produce a database of oxygen migration barriers ranging from 0.2 to 1.6 eV. Mining the database revealed that systems with low barriers also have low metal-oxygen bond strength, as measured by oxygen vacancy formation energy and oxygen p-band center energy. These correlations provide a powerful descriptor for the development of new oxygen ion conductors and may explain the poor stability of some of the best oxygen conducting perovskites under reducing conditions. Other commonly-cited measures of space, volume, or structure ideality showed only weak correlation with migration barrier. The lowest migration barriers (< 0.5 eV) belong to perovskites with non-transition-metal B-site cations, and may require vacancy-creation strategies that involve no dopants or low-association dopants for optimal performance.Comment: Final draft and SI in single PD

This work demonstrates how databases of diffusion-related properties can be developed from high-throughput ab initio calculations. The formation and migration energies for vacancies of all adequately stable pure elements in both the face-centered cubic (fcc) and hexagonal close packing (hcp) crystal structures were determined using ab initio calculations. For hcp migration, both the basal plane and z-direction nearest-neighbor vacancy hops were considered. Energy barriers were successfully calculated for 49 elements in the fcc structure and 44 elements in the hcp structure. These data were plotted against various elemental properties in order to discover significant correlations. The calculated data show smooth and continuous trends when plotted against Mendeleev numbers. The vacancy formation energies were plotted against cohesive energies to produce linear trends with regressed slopes of 0.317 and 0.323 for the fcc and hcp structures respectively. This result shows the expected increase in vacancy formation energy with stronger bonding. The slope of approximately 0.3, being well below that predicted by a simple fixed bond strength model, is consistent with a reduction in the vacancy formation energy due to many-body effects and relaxation. Vacancy migration barriers are found to increase nearly linearly New J. Phys. 16 (2014) 015018 T Angsten et al with increasing stiffness, consistent with the local expansion required to migrate an atom. A simple semi-empirical expression is created to predict the vacancy migration energy from the lattice constant and bulk modulus for fcc systems, yielding estimates with errors of approximately 30%.