Use of metamodels for rapid discovery of narrow bandgap oxide photocatalysts

Haoxin Mai; Tu C Le; Takashi Hisatomi; Dehong Chen; Kazunari Domen; David A Winkler; Rachel A Caruso

doi:10.1016/j.isci.2021.103068

Use of metamodels for rapid discovery of narrow bandgap oxide photocatalysts

iScience. 2021 Aug 30;24(9):103068. doi: 10.1016/j.isci.2021.103068. eCollection 2021 Sep 24.

Authors

Haoxin Mai¹, Tu C Le², Takashi Hisatomi³, Dehong Chen¹, Kazunari Domen^{3

4}, David A Winkler^{5

6

7}, Rachel A Caruso¹

Affiliations

¹ Applied Chemistry and Environmental Science, School of Science, STEM College, RMIT University, GPO Box 2476, Melbourne, VIC 3001, Australia.
² School of Engineering, STEM College, RMIT University, GPO Box 2476, Melbourne, VIC 3001, Australia.
³ Research Initiative for Supra-Materials (RISM), Shinshu University, 4-17-1 Wakasato, Nagano 380-8553, Japan.
⁴ Office of University Professors, the University of Tokyo, 2-11-16 Yayoi, Bunkyo-ku, Tokyo 113-8656, Japan.
⁵ Monash Institute of Pharmaceutical Sciences, Monash University, Parkville, VIC 3052, Australia.
⁶ School of Biochemistry and Genetics, La Trobe University, Kingsbury Drive, 3042 Bundoora, Australia.
⁷ School of Pharmacy, University of Nottingham, NG7 2RD Nottingham, UK.

Abstract

New photocatalysts are traditionally identified through trial-and-error methods. Machine learning has shown considerable promise for improving the efficiency of photocatalyst discovery from a large potential pool. Here, we describe a multi-step, target-driven consensus method using a stacking meta-learning algorithm that robustly predicts bandgaps and H₂ evolution activities of photocatalysts. Trained on small datasets, these models can rapidly screen a large space (>10 million materials) to identify promising, non-toxic compounds as candidate water splitting photocatalysts. Two effective compounds and two controls possessing optimal bandgap values (∼2 eV) but not photoactivity as predicted by the models were synthesized. Their experimentally measured bandgaps and H₂ evolution activities were consistent with the predictions. Conspicuously, the two compounds with strong photoactivities under UV and visible light are promising visible-light-driven water splitting photocatalysts. This study demonstrates the power of machine learning and the potential of big data to accelerate discovery of next-generation photocatalysts.

Keywords: catalysis; chemistry; computational chemistry.