Screening of Silver-Based Single-Atom Alloy Catalysts for NO Electroreduction to NH₃ by DFT Calculations and Machine...

Area of Science:

• Materials Science
• Electrochemistry
• Computational Chemistry

Background:

• Electrocatalysts for nitrogen oxide reduction reaction (NORR) are crucial for environmental remediation and ammonia synthesis.
• Single-atom alloy (SAA) catalysts show promise but require efficient methods for performance prediction.
• Current exploration of SAAs is limited by the lack of rapid and reliable performance prediction tools.

Purpose of the Study:

• To screen transition-metal atom doped Ag-based SAAs for NORR to ammonia.
• To develop and apply machine learning (ML) and data-analytics approaches for predicting SAA catalytic performance.
• To elucidate the mechanism governing NORR activity and selectivity.

Main Methods:

• Comprehensive screening of transition-metal doped Ag-based SAAs using density-functional theory inputs.
• Application of regression machine learning algorithms and compressed-sensing data analytics.
• Utilizing a descriptor Φmax(η) to evaluate catalytic activity and selectivity.

Main Results:

• Cu/Ag and Zn/Ag SAAs demonstrated efficient NO activation and hydrogenation with suppressed hydrogen evolution reaction.
• NH3 selectivity was found to be dependent on the s orbitals of the doped single atom near the Fermi level.
• Catalytic activity correlated strongly with the local active site environment and intrinsic features, quantified by Φmax(η).

Conclusions:

• The study provides a mechanistic understanding of NORR to NH3 on SAAs.
• Identified Cu/Ag and Zn/Ag as promising electrocatalysts for NORR.
• Established a design principle for guiding the development of highly active and selective SAA catalysts.