报告题目：Toward large scale materials simulation by Kohn-Sham DFT

报 告 人：Guo Hong 　加拿大皇家科学院院士

报告时间：2023年6月20日 上午9:15~10:15

线下地点：粤海校区 致原楼1007

(线上)Zoom Meeting ID：811 8831 3474

Zoom会议链接： https://mcgill.zoom.us/j/81188313474

报告摘要：

In this seminar I shall present some recent progress and our on-going effort to develop a large-scale materials simulation capability based on the Kohn-Sham density functional theory (KS-DFT). For large length scales, the ground state is efficiently solved by our general-purpose KS-DFT solver RESCU. RESCU can compute ground state electronic structure for systems comprising many thousands of atoms on modest computer resources. I shall briefly discuss the computational mathematics behind the efficiency gain. For Born-Oppenheimer ab initio molecular dynamics (AIMD), we apply a sparse Gaussian Process (sGP) regression, on top of the RESCU-AIMD, to reach both large length scales and timescales, i.e. a few thousand atoms, hundreds of picoseconds to several nano-seconds, using a small computer cluster. Certain electron induced chemical reactions can be captured by the RESCU-AIMD/sGP approach.

主讲嘉宾简介：

郭鸿，麦吉尔大学讲座教授，加拿大皇家科学院院士，美国物理学会会士，深圳大学访问教授。1976年四川师范大学本科，1981年通过CUSPEA赴美留学，获匹兹堡大学博士学位，研究领域：纳米电子输运理论、材料物理和计算物理等。

欢迎各位老师和同学参加!