[19a-B01-1~10] 23.1 Joint Session N "Informatics"

[19a-B01-2] Development of Artificial Neural-Network Interatomic Potentials for Silicon Grain Boundaries with Carbon Segregation

〇Yusuke Noda¹, Tatsuya Yokoi¹, Atsutomo Nakamura¹, Katsuyuki Matsunaga^1,2 (1.Nagoya Univ., 2.JFCC)

Keywords:machine learning, neural network, silicon grain boundary

In this study, we have constructed artificial neural-network interatomic potentials for silicon-carbon binary systems, and evaluated energies and atomic forces in silicon grain boundaries with carbon segregation. We have succeeded in predicting energies and atomic forces obtained from DFT calculations using the artificial neural-network interatomic potentials, and confirmed the high prediction performance compared with conventional empirical interatomic potentials.