[20p-C202-1~8] 15.7 Crystal characterization, impurities and crystal defects

[20p-C202-1] Development of Interatomic Potentials for Binary Metals
applying First-Principles Calculations and Machine Learning

〇Takeru Miyagawa¹, Yugo Sakai¹, Akio Yonezu¹, Kazuki Mori², Nobuhiko Kato², Keiji Ishibashi³ (1.Chuo Univ., 2.CTC, 3.COMET Inc.)

Keywords:Combinatorial thin film, Molecular dynamics, Neural network potential

Although molecular dynamics (MD) has been widely used in recent years to analyze materials at the atomic and molecular level, there is a problem that MD cannot be applicable for novel materials because interatomic potentials do not exist for these materials. In this study, we created interatomic potentials for a new binary metals applying first-principles molecular dynamics (first-principles MD) and neural networks (NN). MD simulation was then performed using the generated interatomic potentials, and physical properties were evaluated.