General-purpose machine-learned potential for 16 elemental metals and their alloys
Abstract
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 feasible 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. This work represents a significant leap towards a unified general-purpose MLP encompassing the periodic table, with profound implications for materials science.
Keywords
Cite
@article{arxiv.2311.04732,
title = {General-purpose machine-learned potential for 16 elemental metals and their alloys},
author = {Keke Song and Rui Zhao and Jiahui Liu and Yanzhou Wang and Eric Lindgren and Yong Wang and Shunda Chen and Ke Xu and Ting Liang and Penghua Ying and Nan Xu and Zhiqiang Zhao and Jiuyang Shi and Junjie Wang and Shuang Lyu and Zezhu Zeng and Shirong Liang and Haikuan Dong and Ligang Sun and Yue Chen and Zhuhua Zhang and Wanlin Guo and Ping Qian and Jian Sun and Paul Erhart and Tapio Ala-Nissila and Yanjing Su and Zheyong Fan},
journal= {arXiv preprint arXiv:2311.04732},
year = {2024}
}
Comments
Main text with 17 pages and 8 figures; supplementary with 26 figures and 4 tables; source code and training/test data available