摘要：In this talk, I will introduce the methods developed in my group, especially the machine learning and graph theory aided crystal structure prediction method (MAGUS) [1]. In addition, I will show some applications of these methods combined with first-principles calculations, for instance, the predictions of new possible compounds (helium-water, helium-ammonia, helium-methane, helium-silica, silica-water, etc) in the interior of giant planets or exoplanets, and their exotic new states under planetary high-pressure and high-temperature conditions (superionic state, plastic state, and their coexistence) [2-6]. These new compounds and their states may have some important implications for giant planets, including demixing, magnetic field, erosion of the rocky core, etc.

参考文献：

1.    Kang Xia et al., “A novel superhard tungsten nitride predicted by machine-learning accelerated crystal structure search”, Sci. Bull. 63, 817 (2018).

2.    Cong Liu et al., “Mixed coordination silica at megabar pressure”, Phys. Rev. Lett. 126, 035701 (2021).

3.    Cong Liu et al., “Multiple superionic states in helium-water compounds”, Nature Physics 15, 1065 (2019).

4.    Cong Liu et al., “Plastic and Superionic Helium Ammonia Compounds under High Pressure and High Temperature”, Phys. Rev. X 10, 021007 (2020).

5.    Hao Gao et al., “Coexistence of plastic and partially diffusive phases in a helium-methane compound”, Natl. Sci. Rev. 7, 1540 (2020).

6.    Hao Gao et al., “Superionic Silica-Water and Silica-Hydrogen Compounds in the Deep Interiors of Uranus and Neptune”, Phys. Rev. Lett. 128, 035702 (2022)

简介：孙建，南京大学物理学院教授，国家杰出青年基金获得者（2021）。研究方向为：高压物理、计算凝聚态物理、材料设计与模拟、行星深部物质等。发展了机器学习和图论辅助的晶体结构预测新方法（MAGUS）；预言了多个新材料，若干被实验证实；预言了若干系统在高温高压下的超离子态、塑晶态和协同扩散态等新奇物态。已发表学术论文90余篇，重要期刊（Nat. Phys./PRL/PRX/PNAS/JACS）论文18篇。曾获2011年中国国家自然科学二等奖（第五完成人）、2013年“国家海外高层次青年人才”、2014年GRC Van Valkenburg奖。担任中国物理学会高压物理专业委员会委员、中国材料学会计算材料学分会委员、《物理学进展》副主编、《高压物理学报》编委。