Coverage-Dependent Lateral Interactions Shape the Electrocatalytic Activity of High-Entropy Alloys.
While the rich diversity of surface sites on high-entropy alloys (HEAs) is essential for tuning electrocatalytic activity, the coverage-dependent lateral interactions that shape reactive interfaces are often neglected in theoretical studies. Here, we develop a machine learning interatomic potential (MLIP)-enabled framework to model the oxygen reduction reaction (ORR) within an Ag-Ir-Ru-Pd-Pt-Cu-Rh-Re alloy composition space. By tracking the binding strengths of O* and OH* intermediates during competitive coadsorption on crowded surfaces, this framework highlights the key role of lateral interactions, including attractive hydrogen-bond networks and electrostatic repulsion, in evaluating electrocatalytic activity. Incorporating these coverage-induced effects improves agreement with reported PtIr and AgPd composition-activity trends relative to an isolated-site baseline. We further show that increasing compositional complexity within the studied alloy space can amplify lateral repulsion under finite-coverage conditions, broadening binding strength distributions and reducing the population of optimal active sites. The competition between local electronic optimization and coverage-dependent lateral interactions gives rise to a volcano-shaped activity-entropy relationship, offering guidance for the rational design of HEA-based ORR electrocatalysts.