Data-Driven and Interpretable Machine-Learning for Performance-Determining Interactions Governing C-C Coupling and C₂⁺ Selectivity in Cu-Catalyzed CO₂ Electroreduction

Key Points

• The study aims to enhance the understanding of interactions affecting C-C coupling and C₂⁺ product selectivity in CO₂ electroreduction using machine learning.
• Developed an interpretable machine-learning framework to analyze the interactions between catalyst properties and reaction conditions.
• Evaluated the performance of the framework in predicting C-C coupling efficiency and selectivity for C₂⁺ products.
• Demonstrated significant improvements in predictability of C-C coupling interactions.
• Achieved enhanced selectivity for C₂⁺ products through optimized electrochemical parameters.