[15p-P06-1~15] 23 Joint Session N "Informatics" (Poster)

[15p-P06-8] Photocurrent Prediction of Multi-Element-Doped Hematite Photoelectrodes by Two-Step LASSO Regression

〇Takuma Nishimura¹, Yoshitaka Kumabe^1,2, Yosuke Harashima^3,4, Mikiya Fujii^3,4, Takashi Tachikawa^1,2 (1.Grad. Sch. of Sci., Kobe Univ., 2.Mol. Photosci. Res. Center, Kobe Univ., 3.MS, Nara Inst. Sci. Tech., 4.DSC, Nara Inst. Sci. Tech.)

Keywords:

photocatalyst,machine learning,data analysis

Clean hydrogen production reaction by photocatalysts using sunlight is attracting attention as an environmentally friendly energy production method. Hematite, commonly known as a red rust, can process water splitting reaction photoelectrochemically, leading to hydrogen production on the counter electrode. In this study, we prepared multi-element-doped hematite photoelectrodes and acquired experimental data, including Raman spectra. Towards developing high performance hematite photoelectrodes, we worked on constructing a machine learning model to predict photocurrent as an indicator of hydrogen production capability in order to achieve high performance in future works.