Dataset Information

Use of metamodels for rapid discovery of narrow bandgap oxide photocatalysts.

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.

SUBMITTER: Mai H

PROVIDER: S-EPMC8455646 | biostudies-literature | 2021 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Use of metamodels for rapid discovery of narrow bandgap oxide photocatalysts.

Mai Haoxin H Le Tu C TC Hisatomi Takashi T Chen Dehong D Domen Kazunari K Winkler David A DA Caruso Rachel A RA

iScience 20210830 9

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 H<sub>2</sub> evolution activities of photocatalysts. Trained on small datasets, these models can rapidly screen a large space (>10 million materi ...[more]

PMID: 34585115

Similar Datasets

Project description:Background: The development of organic/inorganic metal halide perovskites has seen unprecedent growth since their first recognition for applications in optoelectronic devices. However, their thermodynamic stability and toxicity remains a challenge considering wide-scale deployment in the future. This spurred an interest in search of perovskite-inspired materials which are expected to retain the advantageous material characteristics of halide perovskites, but with high thermodynamic stability and composed of earth-abundant and low toxicity elements. ABX 3 chalcogenides (A, B=metals, X=Se, S) have been identified as potential class of materials meeting the aforementioned criteria. Methods: In this work, we focus on studying tin zirconium selenide (SnZrSe 3) relevant physical properties with an aim to evaluate its prospects for application in optoelectronics. SnZrSe 3 powder and monocrystals were synthesized via solid state reaction in 600 - 800 °C temperature range. Crystalline structure was determined using single crystal and powder X-ray diffraction methods. The bandgap was estimated from diffused reflectance measurements on powder samples and electrical properties of crystals were analysed from temperature dependent I-V measurements. Results: We found that SnZrSe 3 crystals have a needle-like structure (space group - Pnma) with following unit cell parameters: a=9.5862(4) Å, b=3.84427(10) Å, c=14.3959(5) Å. The origin of the low symmetry crystalline structure was associated with stereochemical active electron lone pair of Sn cation. Estimated bandgap was around 1.15 eV which was higher than measured previously and predicted theoretically. Additionally, it was found that resistivity and conductivity type depended on the compound chemical composition. Conclusions: Absorption edge in the infrared region and bipolar dopability makes SnZrSe 3 an interesting material candidate for application in earth-abundant and non-toxic single/multi-junction solar cells or other infrared based optoelectronic devices.

Dataset Information

Use of metamodels for rapid discovery of narrow bandgap oxide photocatalysts.

Publications

Use of metamodels for rapid discovery of narrow bandgap oxide photocatalysts.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets