跳到主要导航 跳到搜索 跳到主要内容

Accelerating the Prediction of g-C3N4-Supported Dual-Atom Catalysts for Photocatalytic CO2 Reduction to CO and HCOOH: A Machine Learning and DFT Combined Approach

  • Yingmei Bian
  • , Yanxin Wang
  • , Zijun Yang
  • , Zexiang Yin
  • , Heng Zhao
  • , Yuan Liu
  • , Hainan Shi
  • , Yaqiong Su
  • , Yida Deng
  • , Haozhi Wang
  • Hainan University
  • Tianjin University
  • Liaoning Normal University
  • Nankai University
  • Fudan University
  • Central South University

科研成果: 期刊稿件 › 文章 › 同行评审

42 引用 (Scopus)

摘要

Dual-atom catalysts (DACs) show promise for photocatalytic carbon dioxide (CO2) reduction due to their high atom utilization and synergistic effects. However, finding efficient combinations is challenging because of the large number of possibilities. In this study, a high-throughput screening of TM1TM2@g-C3N4 catalysts is conducted using density functional theory (DFT) and machine learning (ML) to identify promising candidates for CO2 photoreduction. The results reveal that the ML algorithm can successfully achieve the relationship between the descriptors of the DACs and the limiting potentials (UL) of CO and HCOOH products. Among four ML models, the feedforward neural network (FNN) achieves the highest accuracy. The ML model successfully predicts the limiting potentials (UL) and screens RuHf@g-C3N4 (UL = −0.72 V) and VV@g-C3N4 (UL = −0.79 V) as promising CO producing catalysts, and MoMo@g-C3N4 (UL = −0.28 V) and CrMo@g-C3N4 (UL = −0.32 V) as efficient HCOOH producing catalysts. DFT validation shows low average errors (−0.05 eV for CO, 0.02 eV for HCOOH). Compared to pure DFT, the FNN model reduces screening time by 56% (CO) and 42% (HCOOH). The ML framework not only successfully screens highly promising catalysts but also provides a solid theoretical basis for the subsequent experimental synthesis for energy conversion.

源语言英语
期刊论文编号e03855
期刊Advanced Energy Materials
卷15
期39
DOI
出版状态已出版 - 21 10月 2025

联合国可持续发展目标

此成果有助于实现下列可持续发展目标:

  1. 可持续发展目标 7 - 经济适用的清洁能源
    可持续发展目标 7 经济适用的清洁能源

学术指纹

探究 'Accelerating the Prediction of g-C3N4-Supported Dual-Atom Catalysts for Photocatalytic CO2 Reduction to CO and HCOOH: A Machine Learning and DFT Combined Approach' 的科研主题。它们共同构成独一无二的学术指纹。

引用此