A theoretical study of oxygen anion diffusion through La and Ca doped SrTiO3 lattice structures: molecular dynamics and ML approaches

Abstract

The diffusion of oxygen ions in doped perovskite materials is of crucial importance for applications such as solid oxide fuel cells and electrolyzers. This study employs a synergistic approach combining MD simulations and machine learning techniques to investigate the influence of Ca and La dopants on oxygen anion diffusion in A-site deficient SrTiO₃. MD simulations were performed on SrTiO₃ super-cells with varying concentrations (5%, 15%, and 25%) of Ca and La dopants, both individually and in combination. Bulk oxygen anion diffusivity values were calculated from the mean squared displacement data obtained from the MD trajectories. k-Means clustering analysis was employed to identify localized clusters of oxygen anions and analyze their displacement patterns, shedding light on the impact of dopants on the surrounding oxygen anion migration. The results reveal that Ca doping enhances oxygen anion diffusivity, while La doping exhibits a more localized effect, restricting oxygen anion mobility in its vicinity due to increased cation–anion interactions. The co-doping of Ca and La leads to an intermediate behavior, with enhanced diffusion compared to La-doped systems but lower than Ca-doped systems. The study provides insights into the mechanisms governing oxygen anion diffusion in doped perovskites and highlights the potential of integrating computational approaches for optimizing materials design for energy-related applications.