A Generalizable Machine-learning Potential of Ag-Au Nanoalloys and its Application on Surface Reconstruction, Segregation and Diffusion

TL;DR

This paper introduces a deep learning-based interatomic potential for Ag-Au nanoalloys, achieving high accuracy in modeling surface phenomena and surpassing empirical force fields, with strong generalization capabilities.

Contribution

A novel machine-learning potential for Ag-Au nanoalloys trained on first-principles data, accurately modeling surface reconstruction, segregation, and diffusion without explicit surface configurations.

Findings

Accurately predicts Au (111) surface reconstruction

Successfully models Au segregation on Ag-Au surfaces

Outperforms empirical force fields in adsorption and diffusion simulations

Abstract

Owing to the excellent catalysis properties of Ag-Au binary nanoalloy, nanostructured Ag-Au, such as Ag-Au nanoparticles and nanopillars, have been under intense investigation. To achieve high accuracy in molecular simulations of the Ag-Au nanoalloys, the surface properties are required to be modeled with first-principles precision. In this work, we propose a generalizable machine-learning interatomic potential for the Ag-Au nanoalloys based on deep neural networks, trained from a database constructed with the first-principle calculations. This potential is highlighted by the accurate prediction of Au (111) surface reconstruction and the segregation of Au towards the Ag-Au nanoalloy surface, where the empirical force field failed in both cases. Moreover, regarding the adsorption and diffusion of adatoms on surfaces, the overall performance of our potential is better than the empirical force fields. We stress that the reported surface properties are blind to the potential modeling in the sense that none of the surface configurations is explicitly included in the training database, therefore, the reported potential is expected to have a strong ability of generalization to a wide range of properties and to play a key role in the investigation of nanostructured Ag-Au evolution, where the accurate descriptions of free surfaces are necessary.