General-purpose machine-learned potential for 16 elemental metals and their alloys

KK Song and R Zhao and JH Liu and YZ Wang and E Lindgren and Y Wang and SD Chen and K Xu and T Liang and PH Ying and N Xu and ZQ Zhao and JY Shi and JJ Wang and S Lyu and ZZ Zeng and SR Liang and HK Dong and LG Sun and Y Chen and ZH Zhang and WL Guo and P Qian and J Sun and P Erhart and T Ala-Nissila and YJ Su and ZY Fan, NATURE COMMUNICATIONS, 15, 10208 (2024).

DOI: 10.1038/s41467-024-54554-x

Machine-learned potentials (MLPs) have exhibited remarkable accuracy, yet the lack of general-purpose MLPs for a broad spectrum of elements and their alloys limits their applicability. Here, we present a promising approach for constructing a unified general-purpose MLP for numerous elements, demonstrated through a model (UNEP-v1) for 16 elemental metals and their alloys. To achieve a complete representation of the chemical space, we show, via principal component analysis and diverse test datasets, that employing one-component and two-component systems suffices. Our unified UNEP-v1 model exhibits superior performance across various physical properties compared to a widely used embedded-atom method potential, while maintaining remarkable efficiency. We demonstrate our approach's effectiveness through reproducing experimentally observed chemical order and stable phases, and large- scale simulations of plasticity and primary radiation damage in MoTaVW alloys. Machine-learned potentials are accurate but often lack broad applicability. Here, authors develop a general-purpose neuroevolution potential for 16 metals and their alloys, achieving efficient and accurate predictions of various physical properties.

Return to Publications page