Prediction of surface reconstructions using MAGUS

TL;DR

This paper introduces a new module within MAGUS for predicting surface reconstruction configurations, leveraging machine learning, graph theory, and diverse structure generation methods, validated on silicon and silicon carbide surfaces.

Contribution

The paper presents a novel module for surface reconstruction prediction integrated into MAGUS, combining advanced structure sampling and transfer ideas from cluster predictions.

Findings

Successfully predicted known surface reconstructions of Si and SiC.

Discovered a new SiC surface model in Si-rich conditions.

Validated the module's effectiveness on multiple surface types.

Abstract

In this paper, we present a new module to predict the potential surface reconstruction configurations of given surface structures in the framework of our machine learning and graph theory assisted universal structure searcher (MAGUS). In addition to random structures generated with specific lattice symmetry, we made full use of bulk materials to obtain a better distribution of population energy, namely, randomly appending atoms to a surface cleaved from bulk structures or moving/removing some of the atoms on the surface, which is inspired by natural surface reconstruction processes. In addition, we borrowed ideas from cluster predictions to spread structures better between different compositions, considering that surface models of different atom numbers usually have some building blocks in common. To validate this newly developed module, we tested it with studies on the surface reconstructions of Si (100), Si (111) and 4H-SiC(1-102)-c(2x2), respectively. We successfully gave the known ground states as well as a new SiC surface model in an extremely Si-rich environment.