Combining First-Principles Modeling and Symbolic Regression for Designing Efficient Single-Atom Catalysts in the Oxygen Evolution Reaction on MoCO MXenes.

In this study, we address the significant challenge of overcoming limitations in the catalytic efficiency for the oxygen evolution reaction (OER). The current linear scaling relationships hinder the optimization of the electrocatalytic performance. To tackle this issue, we investigate the potential of designing single-atom catalysts (SACs) on MoCO MXenes for electrochemical OER using first-principles modeling simulations. By employing the Electrochemical Step Symmetry Index (ESSI) method, we assess OER intermediates to fine-tune the activity and identify the optimal SAC for MoCO MXenes. Our findings reveal that both Ag and Cu exhibit effectiveness as single atoms for enhancing OER activity on MoCO MXenes. However, among the 21 chosen transition metals (TMs) in this study, Cu stands out as the best catalyst for tweaking the overpotential (η). This is due to Cu's lowest overpotential compared to other TMs, which makes it more favorable for the OER performance. On the other hand, Ag is closely aligned with ESSI = η, making the tuning of its overpotential more challenging. Furthermore, we employ symbolic regression analysis to identify the significant factors that exhibit a correlation with the OER overpotential. By utilizing this approach, we derive mathematical formulas for the overpotential and identify key descriptors that affect the catalytic efficiency in the electrochemical OER on MoCO MXenes. This comprehensive investigation not only sheds light on the potential of MXenes in advanced electrocatalytic processes but also highlights the prospect of improved activity and selectivity in OER applications.