Rational design of Pt-anchored single-atom alloy electrocatalysts for NO-to-NH3 conversion by density functional theory and machine learning

Electrochemical NO reduction reaction (NORR) toward NH₃ synthesis emerges as a promising approach to eliminate NO pollution and generate high-value-added products simultaneously. Therefore, exploring suitable NORR electrocatalysts is of great importance. Here, we present a design principle to evaluate the activity of single-atom alloy catalysts (SAACs), whose excellent catalytic performance and well-defined bonding environments make them suitable candidates for studying structure-activity relationships. The machine learning (ML) algorithm is chosen to unveil the underlying physics and chemistry. The results indicate that the catalytic activity of SAACs is highly correlated with the local environment of the active center, that is, the atomic and electronic features. The coeffect of these features is quantitatively verified by adopting a data-driven method. The combination of density functional theory (DFT) and ML investigations not only provides an understanding of the complex NORR mechanisms but also offers a strategy to design highly efficient SAACs with specific active centers rationally.