Fine-Tuning a Universal Machine-Learned Interatomic Potential for Oxygen Plasma Interactions with WS${}_2$

Abstract

Molecular dynamics simulation of plasma-surface interactions requires an interatomic potential that is simultaneously accurate, computationally efficient, and able to describe many elements and bonding types in reactive systems. In principle, a foundation model for machine-learned interatomic potential (MLIP) can meet these demands. We explore the use of the Universal Models for Atoms (UMA) model, developed by Meta FAIR, for the interactions of oxygen plasma species on a multilayer of WS${}_2$, a promising 2D material. Starting from the pretrained uma-s-1p1 model under the Open Catalyst 2020 (OC20) task, we apply an iterative fine-tuning loop. Even in the absence of fine-tuning, the pretrained model reproduces the production-scale observables of interest, namely, chemisorbed S and O coverage under 15 eV O${}^{+}$ and O${}_2^{+}$ bombardment. These results were obtained without spin polarization and Hubbard $U$ correction. Nonetheless, fine-tuning with spin polarization and a Hubbard $U$ correction reduces the energy and force mean absolute error (MAE) to $4.5\times 10^{-3}$ eV/atom and $0.076$ eV/angstrom, respectively.

29 pages, 9 figures in the main text, 8 figures in appendix section, and 2 supplementary videos