Machine Learning Interatomic Potentials for Reactive Hydrogen Dynamics at Metal Surfaces Based on Iterative Refinement of Reaction Probabilities.

Autor:	Stark WG; Department of Chemistry, University of Warwick, Gibbet Hill Road, Coventry CV4 7AL, U.K., Westermayr J; Department of Chemistry, University of Warwick, Gibbet Hill Road, Coventry CV4 7AL, U.K., Douglas-Gallardo OA; Department of Chemistry, University of Warwick, Gibbet Hill Road, Coventry CV4 7AL, U.K., Gardner J; Department of Chemistry, University of Warwick, Gibbet Hill Road, Coventry CV4 7AL, U.K., Habershon S; Department of Chemistry, University of Warwick, Gibbet Hill Road, Coventry CV4 7AL, U.K., Maurer RJ; Department of Chemistry, University of Warwick, Gibbet Hill Road, Coventry CV4 7AL, U.K.; Department of Physics, University of Warwick, Gibbet Hill Road, Coventry CV4 7AL, U.K.
Jazyk:	angličtina
Zdroj:	The journal of physical chemistry. C, Nanomaterials and interfaces [J Phys Chem C Nanomater Interfaces] 2023 Dec 04; Vol. 127 (50), pp. 24168-24182. Date of Electronic Publication: 2023 Dec 04 (Print Publication: 2023).
DOI:	10.1021/acs.jpcc.3c06648
Abstrakt:	The reactive chemistry of molecular hydrogen at surfaces, notably dissociative sticking and hydrogen evolution, plays a crucial role in energy storage and fuel cells. Theoretical studies can help to decipher underlying mechanisms and reaction design, but studying dynamics at surfaces is computationally challenging due to the complex electronic structure at interfaces and the high sensitivity of dynamics to reaction barriers. In addition, ab initio molecular dynamics, based on density functional theory, is too computationally demanding to accurately predict reactive sticking or desorption probabilities, as it requires averaging over tens of thousands of initial conditions. High-dimensional machine learning-based interatomic potentials are starting to be more commonly used in gas-surface dynamics, yet robust approaches to generate reliable training data and assess how model uncertainty affects the prediction of dynamic observables are not well established. Here, we employ ensemble learning to adaptively generate training data while assessing model performance with full uncertainty quantification (UQ) for reaction probabilities of hydrogen scattering on different copper facets. We use this approach to investigate the performance of two message-passing neural networks, SchNet and PaiNN. Ensemble-based UQ and iterative refinement allow us to expose the shortcomings of the invariant pairwise-distance-based feature representation in the SchNet model for gas-surface dynamics. Competing Interests: The authors declare no competing financial interest. (© 2023 The Authors. Published by American Chemical Society.)
Databáze:	MEDLINE
Externí odkaz:	Zobrazit plný text záznamu