Identifying Outstanding Transition-Metal-Alloy Heterogeneous Catalysts for the Oxygen Reduction and Evolution...

Area of Science:

• Materials Science
• Catalysis
• Artificial Intelligence

Background:

• Developing efficient heterogeneous catalysts is crucial for energy applications.
• First-principles calculations are accurate but computationally expensive for screening many materials.
• Global data-centric models may miss niche materials with optimal properties.

Purpose of the Study:

• To use a local artificial intelligence approach, subgroup discovery (SGD), to identify specific material descriptors for optimal oxygen adsorption.
• To find constraints on these descriptors that lead to desirable adsorption energies for oxygen reduction and evolution reactions.
• To guide the rational design of novel alloy catalysts.

Main Methods:

• Utilized a dataset of 95 oxygen adsorption energies from density-functional-theory calculations.
• Applied subgroup discovery (SGD) to identify rules correlating material properties with adsorption energies.
• Analyzed atomic, bulk, and surface properties as potential descriptors.

Main Results:

• Identified specific rules (SG rules) for transition-metal surfaces with outstanding oxygen adsorption properties.
• Found constraints that yield oxygen adsorption energies within the Sabatier-optimal range for oxygen reduction.
• Discovered descriptors that maximize deviations from linear-scaling relations for oxygen evolution.

Conclusions:

• SGD effectively identifies local structure-property relationships for catalyst design.
• The derived SG rules provide physicochemical insights into optimal adsorption.
• This approach facilitates the targeted design of advanced alloy catalysts for electrochemical reactions.