[22p-52A-1~17] 23.1 Joint Session N "Informatics"

2:45 PM - 3:00 PM JST(5:45 AM - 6:00 AM UTC)Presentation by Applicant for JSAP Young Scientists Presentation Award

[22p-52A-7] Materials design for dielectric materials using graph neural networks and Monte Carlo sampling

〇(M2)Yuho Shimano¹, Alex Kutana¹, Ryoji Asahi¹ (1.Nagoya Univ.)

Keywords:

Materials infomatics,Graph neural networks,Monte Carlo sampling method

Doped rutile TiO₂ has attracted much attention as a material that develops a colossal dielectric constant (CP material). Using first-principles calculations and machine learning, we found that there is a correlation between the local structural distortion and the increase in the dielectric constant. We also developed a method to obtain the expected value of dielectric constant of a large cell by using graph neural networks and Monte Carlo sampling method, and clarified the dependence of dielectric constant on temperature and doping concentration in co-doped materials.