Universal materials model of deep-learning density functional theory Hamiltonian

Yuxiang Wang ¹Yang Li ¹Zechen Tang ¹He Li ²Zilong Yuan ¹Honggeng Tao ¹Nianlong Zou ¹Ting Bao ¹Xinghao Liang ¹Zezhou Chen ¹Shanghua Xu ¹Ce Bian ¹Zhiming Xu ¹Chong Wang ¹Chen Si ³Wenhui Duan ⁴Yong Xu⁵

扫码查看

作者信息

1. State Key Laboratory of Low Dimensional Quantum Physics and Department of Physics,Tsinghua University,Beijing 100084,China
2. State Key Laboratory of Low Dimensional Quantum Physics and Department of Physics,Tsinghua University,Beijing 100084,China;Institute for Advanced Study,Tsinghua University,Beijing 100084,China
3. School of Materials Science and Engineering,Beihang University,Beijing 100191,China
4. State Key Laboratory of Low Dimensional Quantum Physics and Department of Physics,Tsinghua University,Beijing 100084,China;Institute for Advanced Study,Tsinghua University,Beijing 100084,China;Frontier Science Center for Quantum Information,Beijing 100084,China
5. State Key Laboratory of Low Dimensional Quantum Physics and Department of Physics,Tsinghua University,Beijing 100084,China;Frontier Science Center for Quantum Information,Beijing 100084,China;RIKEN Center for Emergent Matter Science(CEMS),Wako 351-0198,Japan
折叠

Abstract

Realizing large materials models has emerged as a critical endeavor for materials research in the new era of artificial intelligence,but how to achieve this fantastic and challenging objective remains elusive.Here,we propose a feasible pathway to address this paramount pursuit by developing universal materials models of deep-learning density functional theory Hamiltonian(DeepH),enabling computational model-ing of the complicated structure-property relationship of materials in general.By constructing a large materials database and substantially improving the DeepH method,we obtain a universal materials model of DeepH capable of handling diverse elemental compositions and material structures,achieving remarkable accuracy in predicting material properties.We further showcase a promising application of fine-tuning universal materials models for enhancing specific materials models.This work not only demonstrates the concept of DeepH's universal materials model but also lays the groundwork for devel-oping large materials models,opening up significant opportunities for advancing artificial intelligence-driven materials discovery.

Key words

Large materials model/Universal materials model/Deep-learning density functional theory/Artificial intelligence-driven materials discovery

引用本文复制引用

基金项目

Basic Science Center Project of National Natural Science Foundation of China(52388201)

National Natural Science Foundation of China(12334003)

National Science Fund for Distinguished Young Scholars(12025405)

National Key Basic Research and Development Program of China(2023YFA1406400)

Beijing Advanced Innovation Center for Future Chip(ICFC)()

Beijing Advanced Innovation Center for Materials Genome Engineering()

Shuimu Tsinghua Scholar program()

出版年

2024

科学通报(英文版)

中国科学院

科学通报(英文版)

CSTPCD

ISSN：1001-6538

段落导航