@article{Xiao2026, 
author = {Yang Xiao and Bo Wang and Weizuo Li and Jianuo Liu and Xinru Ren and Cheng Qian and Xiugeng Ye and Jiangcheng Shi and Huang Zhou and Yafei Zhao and Yuen Wu},
title = {Atomic-precision oxygen evolution materials: From empirical trial-and-error to AI-driven intelligent manufacturing},
year = {2026},
journal = {Nano Research},
volume = {19},
number = {4},
pages = {94908378},
keywords = {artificial intelligence, oxygen evolution reaction, mechanisms, material manufacturing, trial-and-error},
url = {https://www.sciopen.com/article/10.26599/NR.2026.94908378},
doi = {10.26599/NR.2026.94908378},
abstract = {As a key bottleneck in proton exchange membrane water electrolysis (PEMWE) for hydrogen production, the acidic oxygen evolution reaction (AOER) poses significant challenges due to its harsh reaction environment and sluggish kinetics. Addressing this bottleneck relies heavily on advancing atomic-precision oxygen evolution materials (APOEM). Traditional empirical trial-and-error (ETAE) approaches, while laying foundational groundwork, are inefficient in navigating the vast compositional and structural space of APOEM. They often fail to precisely tailor active sites to achieve balanced activity, stability, and cost, which limits progress in overcoming the intrinsic limitations. APOEM, by contrast, enables atomic-level control over active site configuration, ligand environments, and electron structures. This precision is critical for optimizing the adsorption/desorption kinetics of reaction intermediates and mitigating catalyst dissolution under acidic conditions, thus addressing the shortcomings of ETAE-driven material development. This review systematically summarizes the synergistic effects of atomic-level engineering, mechanistic insights, and artificial intelligence (AI)-driven screening in enhancing APOEM performance. Subsequently, four atomic-scale engineering strategies, including single-atom/clusters, defects, interfaces, and strain, are systematically reviewed. Finally, this review explores how AI-driven intelligent manufacturing (ADIM) transforms APOEM development. ADIM integrates AI, machine learning, high-throughput computing, and automated synthesis. Unlike ETAE, which relies on manual experimentation and serendipity, ADIM accelerates the screening of APOEM candidates, predicts structure–property relationships, and optimizes atomic configurations at scale.}
}