CO Oxidation Catalysis Decoded: An Electronic Structure Descriptor Perspective

Area of Science:

• Catalysis and Surface Science
• Computational Chemistry
• Materials Science

Background:

• Carbon monoxide (CO) catalytic oxidation is a key model reaction for understanding catalysis.
• Traditional descriptors like CO adsorption energy lack universality due to material specificity.
• Electronic structure descriptors offer a more intrinsic and potentially universal approach.

Purpose of the Study:

• To systematically review and categorize electronic structure descriptor systems for CO oxidation.
• To explore hierarchical relationships between different descriptor types.
• To investigate pathways for developing dynamic and multiscale descriptors.

Main Methods:

• Systematic categorization of four electronic structure descriptor systems: energy-level, d-band, charge/valence, and spin structures.
• Analysis of hierarchical relationships across electronic properties.
• Exploration of dynamic and multiscale descriptor construction.
• Application of physics-based descriptors and interpretable machine learning.

Main Results:

• Identified four key electronic structure descriptor systems and their interrelations.
• Demonstrated the decisive role of these descriptors in CO oxidation reaction pathways.
• Showcased potential for dynamic and multiscale descriptor development.
• Highlighted the utility of physics-based descriptors and ML for predictive modeling.

Conclusions:

• Electronic structure descriptors provide a unified framework for understanding CO oxidation catalysis.
• Dynamic and multiscale descriptors are crucial for capturing complex catalytic processes.
• Physics-based descriptors coupled with interpretable ML can establish causal models for catalytic performance.