Machine learning potentials for modeling alloys across compositions
- 1Department of Materials Science and Engineering, Massachusetts Institute of Technology, Cambridge, MA, USA
- 2School of Chemical, Materials and Biological Engineering, University of Sheffield, Sheffield, UK
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 materials 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.
Related
- Machine learning-driven inverse design of high-hardness multi-principal element alloys through symbolic regression
- From scrap to prime: Designing high-performance alloys from dirty feedstocks with machine learning
- Correcting the Liquid Densities of Universal Machine-Learning Interatomic Potentials Using Local-Number-Density Repulsion
- pdb2reaction: End-to-End Reaction-Path Elucidation from PDB Structures Using Machine-Learning Interatomic Potentials
- Applicability Domain of Universal Machine Learning Interatomic Potentials at Solid-State Battery Interfaces
BibTeX
@article{sheriff2026machine,
title = {{Machine learning potentials for modeling alloys across compositions}},
author = {Killian Sheriff and Daniel Xiao and Yifan Cao and L.R. Owen and Rodrigo Moura Freitas},
journal = {Science Advances},
year = {2026},
month = jun,
doi = {10.1126/sciadv.aea9951},
url = {https://doi.org/10.1126/sciadv.aea9951},
note = {CC BY-NC},
}
RIS
TY - JOUR
TI - Machine learning potentials for modeling alloys across compositions
AU - Killian Sheriff
AU - Daniel Xiao
AU - Yifan Cao
AU - L.R. Owen
AU - Rodrigo Moura Freitas
PY - 2026
DA - 2026/06/19
JO - Science Advances
DO - 10.1126/sciadv.aea9951
UR - https://doi.org/10.1126/sciadv.aea9951
AB - 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 materials 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.
SN - 2375-2548
ER -