Artificial intelligence accelerated catalyst design for high-efficiency ammonia electrosynthesis

Ammonia electrosynthesis represents a pivotal sustainable technology for the energy transition towards carbon neutrality. Among others, single-atom catalysts (SACs) have drawn significant interest due to their advantages including maximum metal utilization, homogeneous and low-coordination metal active centers. However, the conventional design of SACs based on density functional theory is constrained by high computational costs and system complexity. Artificial intelligence (AI) emerged as a transformative paradigm, enabling the rational design of SACs and related catalytic architectures. This review systematically outlines the fundamental mechanisms of SACs towards ammonia electrosynthesis. We further elaborate on the data-driven strategies for building catalyst descriptors and databases, with emphasis on reviewing cutting-edge AI applications in high-throughput catalyst screening, structure-activity relationship modeling, reaction mechanism analysis, and multi-metal-atom catalyst design. Finally, we summarize the pivotal shift in the design workflows driven by AI, analyze the persistent challenges, and discuss future pathways toward industrial implementation.