[16a-PB04-1~1] 13.1 Fundamental properties, surface and interface, and simulations of Si related materials

[16a-PB04-1] Modeling to Predict Impurity Distribution from Time Evolution of Electron in Nanoscale Semiconductors

〇Masakazu Muraguchi¹, Kazuki Harada¹, Yuuki Kobayashi¹, Yoshitaka Itoh¹, Tota Suko² (1.Hokkaido Univ. of Sci., 2.Waseda Univ.)

Keywords:Electron dynamics, Machine Learning, Time development calculation

Machine learning has been used to model the relationship between the time evolution of electron density and the impurity distribution, and we have predicted the impurity distribution inversely from the electron conduction process.In this report, taking advantage of solving an inverse problem, the impurity distribution obtained as a result of the prediction was incorporated into the Hamiltonian and the physical simulation was performed again.The electronic conduction process and permeability obtained by the recalculation were used as the evaluation index of the prediction results. This enabled us to improve the accuracy of the prediction model.