Study on the electrocatalytic CO2 reduction performance of covalent organic framework materials based on machine learning
In order to accurately predict the catalytic performance of covalent organic framework materials (COFs) for electrocatalytic carbon dioxide and analyze the influencing factors affecting the catalytic effect, this study collected COFs structure data and experimental data from 44 literatures, and used...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
EDP Sciences
2025-01-01
|
| Series: | E3S Web of Conferences |
| Online Access: | https://www.e3s-conferences.org/articles/e3sconf/pdf/2025/30/e3sconf_epemr2025_01008.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | In order to accurately predict the catalytic performance of covalent organic framework materials (COFs) for electrocatalytic carbon dioxide and analyze the influencing factors affecting the catalytic effect, this study collected COFs structure data and experimental data from 44 literatures, and used machine learning methods. Six regression models were trained and evaluated with COFs structure data and experimental data as features and Faraday efficiency as output. By evaluating the fitting coefficient, mean absolute error and the fitting effect of the test set, the extreme gradient boosting (XGB) model has the best performance. Through the visual analysis of the partial dependence diagram and the individual expectation condition diagram of the XGB model, the coordination metal is Ni, the coordination metal content is greater than 10 %, the pore limit diameter is in the range of 2.5nm-12.5nm. The COFs with tetragonal crystal system have high Faraday efficiency. This research method can not only accurately predict the catalytic performance of covalent organic framework materials (COFs) for electrocatalytic carbon dioxide, but also provide a reference for the screening of catalysts according to the structural characteristics. |
|---|---|
| ISSN: | 2267-1242 |