Distributions of local electrochemistry in heterogeneous microstructures of solid oxide fuel cells using high-performance computations

A high-throughput, high-performance computational approach is used to simulate local electrochemistry in three-dimensions of solid oxide fuel cell electrodes, with the aim of understanding distributions of performance values within microstructures. Simulations are carried out on 47 three-phase cathodes, whose lateral {vertical} dimensions are 22 {15} times their average particle size (0.46 μm). The 47 microstructures are spread across four distinct groups having different standard deviations in their particle size and/or local volume fraction distributions. The average performance simulated compares favorably to two accepted effective medium theory (EMT) models, but a significant discrepancy between the locally-resolved simulations and the EMT models arises from the Ohmic transport terms. This is borne out further by local electrochemical values, specifically distributions of local activation overpotentials and regions of extremely high current densities: values often connected with degradation. The impact of particle size and volume fraction distributions on the distribution of performance values–both within and between microstructural groups—is highlighted throughout. Results from this study indicate that high-performance simulations in a high-throughput, large-data workflow can elucidate performance characteristics that are not captured by continuum level models using EMT. Understanding of these detailed performance characteristics is expected to lead to more durable and reliable electrochemical cells. [Display omitted] •A finite element method (ERMINE) simulated morphology-resolved, local electrochemistry in SOFCs•High-throughput, high-performance computations (ERMINE) simulated performance of 47 cathodes•Distributions of performance parameters are analyzed within and between large volume cathodes•ERMINE distributions underscore the importance of local ionic currents in electrode performance•Simulations reveal how particle size and volume fraction distributions impact electrochemistry