Atomistic simulation of grain boundary migration induced by non-equilibrium solute distribution

Abstract

While atomistic simulation is now a widely used method for investigating grain boundary migration, it is usually based on the so-called artificial driving force. In this study, we report a new general atomistic method for simulating grain boundary migration based on a tunable physical driving force that is induced by unbalanced solute distribution. This is motivated by the previous experimental reports that non-equilibrium distribution of solute atoms, e.g., solute atoms are more enriched/depleted in one grain relative to the other, is commonly observed during diffusion induced grain boundary migration. We show that the solute-induced driving force caused by the unbalanced free energy in the two neighboring grains strongly depends on the type of the solute atoms, which can be used to tune both the type (e.g., attraction or repulsion) and magnitude of the driving force. Furthermore, we find that different solute types (with varying atomic size and cohesive energy) interact with the grain boundary in different ways resulting in solute segregation or anti-segregation. Moreover, the lattice strain resulting from atomic size mismatch, which has been proposed to be an important driving force for diffusion induced grain boundary migration, is found to be not always needed to induce the grain boundary migration. We hope that while this study showcases a new atomistic method of simulating grain boundary migration, the results of this investigation also aids the understanding of the phenomenon of diffusion induced grain boundary migration, for which clear physical mechanisms remain elusive after decades of research.