Active learning-guided catalyst design for selective acetate production in CO electroreduction.

Area of Science:

• Electrochemistry
• Materials Science
• Computational Chemistry

Background:

• Acetic acid is a vital industrial chemical with growing global demand.
• Electrochemical CO reduction reaction (CORR) over copper (Cu) presents a sustainable pathway for acetate production from waste carbon.
• Limited mechanistic understanding hinders the development of highly selective CORR catalysts.

Purpose of the Study:

• To develop an AI-driven multi-scale simulation framework for elucidating the CORR mechanism.
• To identify key descriptors governing CO-to-acetate selectivity.
• To guide the discovery of novel, highly selective CORR catalysts.

Main Methods:

• Integration of grand-canonical density functional theory (GC-DFT) and microkinetic modeling (MKM).
• Application of active learning for catalyst optimization.
• Experimental validation using zero-gap electrolyzers.

Main Results:

• The DFT-based MKM identified CO-CH coupling as the acetate formation pathway.
• CH* binding energy was determined as the critical descriptor for acetate selectivity.
• AI optimization predicted Cu/Pd (2:1) and Cu/Ag (3:1) as superior catalysts.
• Experimental results showed significantly enhanced Faradaic efficiencies for acetate production with Cu/Pd and Cu/Ag catalysts compared to pure Cu.

Conclusions:

• The study establishes an AI-driven framework for mechanistic understanding and catalyst design in CORR.
• The findings provide a data-driven strategy for advancing selective electrocatalysis.
• Optimized Cu-based alloys demonstrate enhanced performance for sustainable acetate production.