(427g) Predicting Diffusion Barriers in Alloy Nanocatalysts Using a Transition-State Cluster Expansion

Kinetic processes such as diffusion and dissolution play an important role in determining the atomic structures of metastable alloy nanocatalysts. To enable the simulation of atomic diffusion in nanostructured materials at experimentally relevant sizes, kinetic Monte Carlo (KMC) models parameterized by density functional theory (DFT) calculations have been widely used. However, accurately predicting the kinetic barriers for diffusion in substitutional alloys in realistic simulations still remains challenging due to the wide variety of local environments that may exist around the diffusing atom. We present a Pt-Ni-vacancy cluster expansion model that explicitly includes transition states as a sublattice, enabling us to accurately predict the activation energies of vacancy-mediated diffusion in Pt-Ni nanoparticles. We assess the performance of the cluster expansion compared with other simple models, such as the broken-bond model and a model inspired by Marcus theory, on the same dataset, and find that the cluster expansion has significantly lower prediction error. The present work provides an accurate and efficient way to predict diffusion barriers, which can be implemented in KMC simulations to understand the effects of atomic diffusion on the structure and stability of nanocatalysts.