Machine learning potentials for modeling alloys across compositions

TL;DR

This paper introduces a machine learning approach that combines information theory to optimize sampling and design potentials capable of accurately modeling alloy properties across diverse compositions and structures.

Contribution

The authors develop a novel method integrating information theory with machine learning to improve the accuracy of alloy modeling across all compositions and structures.

Findings

Successfully predicts alloy properties like stacking-fault energies and phase diagrams.

Demonstrates robustness through extensive comparison with experimental data.

Effectively captures complex chemical arrangements in alloys.

Abstract

Materials properties depend strongly on chemical composition, i.e., the relative amounts of each chemical element. Changes in composition lead to entirely different chemical arrangements, which vary in complexity from perfectly ordered (i.e., stoichiometric compounds) to completely disordered (i.e., solid solutions). Accurately capturing this range of chemical arrangements remains a major challenge, limiting the predictive accuracy of machine learning potentials (MLPs) in materials modeling. Here, we combine information theory and machine learning to optimize the sampling of chemical motifs and design MLPs that effectively capture the behavior of metallic alloys across their entire compositional and structural landscape. The effectiveness of this approach is demonstrated by predicting the compositional dependence of various material properties - including stacking-fault energies, short-range order, heat capacities, and phase diagrams - for the AuPt and CuAu binary alloys, the ternary CrCoNi, and the TiTaVW high-entropy alloy. Extensive comparison against experimental data demonstrates the robustness of this approach in enabling materials modeling with high physical fidelity.