Closed-Loop Multi-Objective Optimization for Cu-Sb-S Photo-Electrocatalytic Materials' Discovery.

Area of Science:

• Materials Science
• Catalysis
• Renewable Energy

Background:

• Copper antimony sulfides are earth-abundant materials with broad light absorption, making them promising for photo-electrochemical water splitting.
• Their catalytic performance in hydrogen evolution is highly sensitive to crystalline structure and atomic composition.

Purpose of the Study:

• To optimize the photo-electrocatalytic hydrogen evolution of copper antimony sulfides (Cu-Sb-S) by exploring their compositional space.
• To develop and implement an integrated workflow combining high-throughput experimentation and machine learning for catalyst optimization.

Main Methods:

• A closed-loop workflow integrating a robotic platform, characterization techniques, and a machine learning (ML) optimization model was employed.
• Multi-objective optimization was used to efficiently explore the Cu-Sb-S compositional space.
• Photocurrent testing was performed at 0 V versus the reversible hydrogen electrode (RHE) to evaluate catalytic activity.

Main Results:

• The integrated experiment-ML loop identified optimal conditions in just nine cycles.
• An optimized Cu-Sb-S composition (9:45:46 ratio) on F-doped SnO2 (FTO) glass achieved a photocurrent of -186 µA cm⁻².
• This optimal material exhibited a bandgap of 1.85 eV and 63.2% Cu¹⁺/Cu species content.
• The targeted search yielded a non-obvious CuSbS composition with 2.3 times higher activity than random sampling.

Conclusions:

• An intelligent, closed-loop workflow can rapidly optimize complex material systems for catalysis.
• Optimized copper antimony sulfides demonstrate significantly enhanced photo-electrocatalytic activity for hydrogen evolution.
• This approach accelerates the discovery of efficient and earth-abundant catalysts for sustainable energy applications.