Universal and Interpretable Descriptor to Design Dual-atom Catalysts for Multi-Type C-C Coupling with Ultrahigh C₂₊...

Area of Science:

• Materials Science
• Catalysis
• Computational Chemistry

Background:

• Dual-atom catalysts (DACs) show promise for C-C coupling reactions, producing multi-carbon (C2+) products.
• Designing DACs with high activity and selectivity for CO2-driven C-C coupling is challenging due to complex inter-metal interactions.

Purpose of the Study:

• To propose a universal and interpretable descriptor for metal-nitrogen-carbon (M1-M2-N4-C) dual-atom catalysts in C-C coupling.
• To evaluate the descriptor's effectiveness in predicting catalyst activity and selectivity for various C-C coupling pathways.

Main Methods:

• Developed a descriptor based on inherent atomic properties (electronegativity, d electron number).
• Applied the descriptor for high-throughput screening of over 400 graphene-based DACs.
• Integrated machine learning (ML) with physical insights.

Main Results:

• The descriptor accurately evaluates activity and selectivity for various C-C coupling processes (CH*–CH*, CO*–CO*, CHO*–CO*, CH*–CH2*).
• FeZr@NC predicted as a superior catalyst for C-C coupling, achieving a C2+ yield of 22.6%, significantly outperforming existing catalysts.
• Descriptor predictions showed minimal discrepancies with computational results across different coupling reactions.

Conclusions:

• The proposed descriptor offers practical guidelines for designing advanced DACs based on atomic properties.
• This work facilitates the rational design of highly efficient catalysts for C-C coupling reactions.
• The descriptor's universality allows extension to various C-C coupling processes.