Reinforcement learning-guided long-timescale simulation of hydrogen transport in metals

TL;DR

This paper introduces a reinforcement learning-based method for long-timescale atomistic simulations of hydrogen diffusion in metals, effectively bridging the gap between simulation and experimental timescales.

Contribution

The authors develop a novel reinforcement learning approach that enables long-timescale diffusion simulations, outperforming traditional methods like Metropolis-Hastings in sampling efficiency.

Findings

Hydrogen diffusivity in metals matches experimental data

The method accelerates sampling of low-energy configurations

Effective simulation of diffusion in complex alloys

Abstract

Atomic diffusion in solids is an important process in various phenomena. However, atomistic simulations of diffusion processes are confronted with the timescale problem: the accessible simulation time is usually far shorter than that of experimental interests. In this work, we developed a long-timescale method using reinforcement learning that simulates diffusion processes. As a testbed, we simulate hydrogen diffusion in pure metals and a medium entropy alloy, CrCoNi, getting hydrogen diffusivity reasonably consistent with previous experiments. We also demonstrate that our method can accelerate the sampling of low-energy configurations compared to the Metropolis-Hastings algorithm using hydrogen migration to copper (111) surface sites as an example.