Machine learning aided design of perovskite oxide materials for photocatalytic water splitting
| Autor: | Minjie Li, Wencong Lu, Tian Lu, Qiuling Tao, Ye Sheng, Long Li |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
Materials science
Band gap Energy Engineering and Power Technology 02 engineering and technology 010402 general chemistry Machine learning computer.software_genre 01 natural sciences Cross-validation symbols.namesake Electrochemistry Perovskite (structure) business.industry 021001 nanoscience & nanotechnology Pearson product-moment correlation coefficient 0104 chemical sciences Random forest Support vector machine Fuel Technology symbols Artificial intelligence Gradient boosting 0210 nano-technology business computer Photocatalytic water splitting Energy (miscellaneous) |
| Zdroj: | Journal of Energy Chemistry. 60:351-359 |
| ISSN: | 2095-4956 |
| DOI: | 10.1016/j.jechem.2021.01.035 |
| Popis: | Suffering from the inefficient traditional trial-and-error methods and the huge searching space filled by millions of candidates, discovering new perovskite visible photocatalysts with higher hydrogen production rate ( R H 2 ) still remains a challenge in the field of photocatalytic water splitting (PWS). Herein, we established structural-property models targeted to R H 2 and the proper bandgap ( E g ) via machine learning (ML) technology to accelerate the discovery of efficient perovskite photocatalysts for PWS. The Pearson correlation coefficients (R) of leave-one-out cross validation (LOOCV) were adopted to compare the performances of different algorithms including gradient boosting regression (GBR), support vector regression (SVR), backpropagation artificial neural network (BPANN), and random forest (RF). It was found that the BPANN model showed the highest R values from LOOCV and testing data of 0.9897 and 0.9740 for R H 2 , while the GBR model had the best values of 0.9290 and 0.9207 for E g . Furtherly, 14 potential PWS perovskite candidates were screened out from 30,000 ABO3-type perovskite structures under the criteria of structural stability, E g , conduction band energy, valence band energy and R H 2 . The average R H 2 of these 14 perovskites is 6.4% higher than the highest value in the training data set. Moreover, the online web servers were developed to share our prediction models, which could be accessible in http://materials-data-mining.com/ocpmdm/material_api/ahfga3d9puqlknig ( E g prediction) and http://materials-data-mining.com/ocpmdm/material_api/i0ucuyn3wsd14940 ( R H 2 prediction). |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect