[18p-S2_204-1~10] 23.1 Joint Session N "Informatics"

2:30 PM - 2:45 PM JST(5:30 AM - 5:45 AM UTC)Presentation by Applicant for JSAP Young Scientists Presentation Award

[18p-S2_204-3] On-demand phase-field modeling: Three-dimensional Landau energy for HfO₂ through machine learning

〇(M1)Yusuke Tamura^1,2, Kairi Maduda², Yu Kumagai^2,3 (1.Tohoku Univ., 2.IMR, Tohoku Univ., 3.OAS, Tohoku Univ.)

Keywords:

phase-field method,machine learning,ferroelectrics

HfO₂ is an unusual material in which ferroelectricity is stabilized upon thin-film formation, and its emergence involves multiple atomic-displacement modes. To describe this behavior within a phase-field method, a multivariable Landau energy functional is required; however, constructing such a functional entails a prohibitively large computational cost. In this study, we employed machine learning to substantially reduce the computational cost while constructing a highly accurate Landau energy functional. Furthermore, using phase-field simulations based on the constructed functional, we verified the emergence of polarization in both bulk and thin-film HfO₂.