Multiscale Modeling of Compositional Stresses in Nonstoichiometric Oxides

University Of Pennsylvania, Philadelphia PA

Investigators

Abstract

Oxide materials are of immense importance for energy conversion and energy storage technologies. Their working principle is based on the coupling of transport between ions and electrons. Relevant use includes solid oxide fuel cells, catalysts and electrolyzers, and these effects have also been employed in chemical sensors, electrochemical transducers, and in advanced electronic memories and computing devices. The goal of this work is to define novel computational methods to study how the oxygen vacancies interact with defects such as grain boundaries, dislocations and facets. This will be studied in polycrystalline ceria films used in electronic and energy systems and in nanocrystals used in catalysis, thus influencing functional properties. A large effort and many resources are now being invested to develop efficient methods to improve the catalytic properties and transport in oxide materials. Thus, the insights gained from increased knowledge of defect mechanics at interfaces and grain boundaries could have economic impact, mostly in the energy and electronics industry. Results and methods will be embedded into college and graduate level eductation. Interactive software will be produced to relate mechanics concepts to the broader public. Since the interactions between defects in oxide materials are determined both by atomic-scale phenomena and by the elastic and electrostatic interactions of defects over length scales of hundreds of nanometers, the project will adopt a multiscale approach. The structure of grain boundaries and defects in oxide materials (with ceria as a concrete example) will be predicted using a novel genetic algorithm technique and validated with high-resolution measurements. Based on the structure, this project will study how the formation energies and interaction of defects are influenced by stresses and space charges near the grain boundaries. The distribution of oxygen vacancies and dopants near surfaces and grain boundaries will be determined using a combination of state of the art quantum simulations and a new method to determine the strains of charged defects. Using the information from these atomic-scale simulations, the project will develop a fully-coupled continuum electro-chemo-mechanical model to predict the stresses in polycrystalline oxide films and nanocrystals and to model the influence of stresses on oxygen vacancy density and hence transport. For nanocrystals and polycrystalline films, the project study the effect of stresses on catalytic activity, which will be validated by coulombic calorimetry studies and HRTEM observations. Finally, the fully-coupled electro-chemo-mechanical model will be used to model and predict how strain patterning and defects such as dislocations influence metal-insulator transitions in oxides (with NiO as an example) and the predictions will be validated with experiments. The project will provide an opportunity for graduate and undergraduate students to both carry out experimental work in a leading industrial lab and to develop advanced computational skills. The progress made in the computational methods will be included in the course that the PI has created to promote hands-on simulation experience. Simulation modules enabling the communication of complex mechanics concepts to the broader public will be developed.