[8a-N302-1~11] 18.1 Materials Informatics

11:30 AM - 11:45 AM JST(2:30 AM - 2:45 AM UTC)Presentation by Applicant for JSAP Young Scientists Presentation Award

[8a-N302-10] Application of Machine Learning Techniques to Multielement-doped Hematite Photocatalysts

〇Takuma Nishimura¹, Yoshitaka Kumabe^1,2, Yosuke Harashima^3,4, Mikya Fujii^3,4,5, Takashi Tachikawa^1,2 (1.Grad. Sch. of Sci., Kobe Univ., 2.CLiPI, Kobe Univ., 3.MS, NAIST, 4.DSC, NAIST, 5.CMP, NAIST)

Keywords:

photocatalyst,machine learning

Hematite (α-Fe₂O₃), commonly known as a red rust, can promise as a photoelectrode for clean hydrogen generation via photoelectrochemical water splitting reaction. However, rapid recombination of photogenerated carriers hinders practical applications. To facilitate the future discovery of high-performing samples, we collected diverse experimental data (compositions, X-ray diffraction patterns, Raman spectra, UV-vis transmittance spectra, scanning electron microscopy images) for multielement-doped hematite. In this study, we attempted to identify the critical factors for predicting photocatalytic activity using this dataset and machine learning techniques and provide insights from a chemical perspective.