Elucidating the impact of oxygen functional groups on the catalytic activity of M-N4-C catalysts for the oxygen reduction reaction: a density functional theory and machine learning approach

Liang Xie; Wei Zhou; Yuming Huang; Zhibin Qu; Longhao Li; Chaowei Yang; Yani Ding; Junfeng Li; Xiaoxiao Meng; Fei Sun; Jihui Gao; Guangbo Zhao; Yukun Qin

doi:10.1039/d3mh02115g

Elucidating the impact of oxygen functional groups on the catalytic activity of M-N₄-C catalysts for the oxygen reduction reaction: a density functional theory and machine learning approach

Mater Horiz. 2024 Apr 2;11(7):1719-1731. doi: 10.1039/d3mh02115g.

Authors

Liang Xie¹, Wei Zhou^{2

1}, Yuming Huang¹, Zhibin Qu¹, Longhao Li¹, Chaowei Yang¹, Yani Ding¹, Junfeng Li¹, Xiaoxiao Meng¹, Fei Sun¹, Jihui Gao¹, Guangbo Zhao¹, Yukun Qin¹

Affiliations

¹ School of Energy Science and Engineering, Harbin Institute of Technology, Harbin 150001, P. R. China. gaojh@hit.edu.cn.
² State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology, Harbin 150090, P. R. China. hitzhouw@hit.edu.cn.

PMID: 38277153
DOI: 10.1039/d3mh02115g

Abstract

Efforts to enhance the efficiency of electrocatalysts for the oxygen reduction reaction (ORR) in energy conversion and storage devices present formidable challenges. In this endeavor, M-N₄-C single-atom catalysts (MN₄) have emerged as promising candidates due to their precise atomic structure and adaptable electronic properties. However, MN₄ catalysts inherently introduce oxygen functional groups (OGs), intricately influencing the catalytic process and complicating the identification of active sites. This study employs advanced density functional theory (DFT) calculations to investigate the profound influence of OGs on ORR catalysis within MN₄ catalysts (referred to as OGs@MN₄, where M represents Fe or Co). We established the following activity order for the 2eORR: for OGs@CoN₄: OH@CoN₄ > CoN₄ > CHO@CoN₄ > C-O-C@CoN₄ > COC@CoN₄ > COOH@CoN₄ > CO@CoN₄; for OGs@FeN₄: COC@FeN₄ > CO@FeN₄ > OH@FeN₄ > FeN₄ > COOH@FeN₄ > CHO@FeN₄ > C-O-C@FeN₄. Multiple oxygen combinations were constructed and found to be the true origin of MN₄ activity (for instance, the overpotential of 2OH@CoN₄ as low as 0.07 V). Furthermore, we explored the performance of the OGs@MN₄ system through charge and d-band center analysis, revealing the limitations of previous electron-withdrawing/donating strategies. Machine learning analysis, including GBR, GPR, and LINER models, effectively guides the prediction of catalyst performance (with an R² value of 0.93 for predicting ΔG_{*OOH_vac} in the GBR model). The E_g descriptor was identified as the primary factor characterizing ΔG_{*OOH_vac} (accounting for 62.8%; OGs@CoN₄: R² = 0.9077, OGs@FeN₄: R² = 0.7781). This study unveils the significant impact of OGs on MN₄ catalysts and pioneers design and synthesis criteria rooted in E_g. These innovative findings provide valuable insights into understanding the origins of catalytic activity and guiding the design of carbon-based single-atom catalysts, appealing to a broad audience interested in energy conversion technologies and materials science.