Toward Practical Design of High‐Entropy Catalysts for Chlorine Evolution Reaction via Pareto‐Guided Multi‐Objective Bayesian Optimization Enabled by a Robotic AI‐Chemist

ABSTRACT The electrocatalytic chlorine evolution reaction (CER) is essential to modern chlor‐alkali industry, yet conventional RuO 2 catalysts suffer from parasitic oxygen evolution. High‐entropy ruthenium oxides (Ru‐HEO) are promising alternatives, but their practical design is hindered by complex composition‐structure‐performance relationship. Herein, we construct a Pareto‐guided multi‐objective Bayesian optimization framework to enable autonomous high‐throughput exploration of quinary Ru‐HEO system. Through this trade‐off strategy, we identify compositions that efficiently balance mass activity, Cl 2 selectivity and material cost. The leading Ru‐HEO catalyst with only 8.4 at% Ru achieves a remarkable activity of 5083 A g −1 Ru at 1.50 V versus RHE and maintains excellent 100‐h stability, outperforming commercial RuO 2 and the state‐of‐the‐art catalysts reported. Integrated into a photovoltaic‐electrochemical (PV‐EC) prototype device and tested under simulated diurnal illumination, it sustains >95% selectivity, a maximum solar‐to‐chemical (STC) efficiency of 14.6% and projected Cl 2 production costs as low as $0.177 per kg. Our work establishes a closed‐loop, AI‐accelerated research paradigm that integrates multi‐objective optimization with robotic experimentation, offering a generalizable and expedited pathway toward high‐performance electrocatalysts for sustainable chemicals manufacturing.