Machine learning enabled rational design of atomic catalysts for electrochemical reactions

Atomic catalysts (ACs) with ultimate atomic utilization and unique catalytic properties have emerged as promising high-performance catalysts because of their great potential for enabling the efficient use of metal resources and achieving atomic economy. The rational design of ACs, however, remains cost-prohibitive for both experimental and computational studies, due to the vast design space required. There exists surging interest in using machine learning (ML) to accelerate the study of the catalytic properties of ACs and the screening of high-performance ACs. Herein, we highlight and summarize recent advances in the ML-enabled rational design of ACs with special emphasis on how to select high-performance ACs efficiently for various electrochemical reactions, including the oxygen evolution reaction (OER), oxygen reduction reaction (ORR), hydrogen evolution reaction (HER), N-2 reduction reaction (NRR) and CO2 reduction reaction (CO2RR). The current challenges and future prospects for the development of this emerging field are also summarized.

Automated summary

Atomic catalysts (ACs) have shown great potential as high-performance catalysts due to their ultimate atomic utilization and unique catalytic properties. However, their rational design is cost-prohibitive, and machine learning (ML) has emerged as a promising approach to accelerate the study of ACs and screen high-performance catalysts efficiently. This review highlights recent advances in the ML-enabled rational design of ACs, focusing on the efficient selection of high-performance ACs for various electrochemical reactions.