Defect-Engineered h-BN as a Platform for Single-Atom HER Catalysts: Descriptor Screening Refined by Electrochemical Stability Analysis

TL;DR

This study uses density functional theory and electrochemical analysis to identify stable, single-atom HER catalysts on defect-engineered h-BN, emphasizing the importance of stability filtering for practical catalyst discovery.

Contribution

It introduces a multi-step screening framework combining descriptor analysis and electrochemical stability to identify robust single-atom HER catalysts on h-BN.

Findings

Cu@VN and Pd@VB are near-thermoneutral for HER.

Electrochemical stability analysis refines candidate selection.

Pd@VB remains stable across a broad potential-pH range.

Abstract

Defect engineering enables hexagonal boron nitride (h-BN) to act as a platform for stabilizing isolated metal atoms, yet systematic identification of catalytically viable motifs remains limited. Here, density functional theory is used to screen transition and coinage metals anchored at B, N, and BN vacancies in h-BN for hydrogen evolution reaction (HER) activity. Cohesive-energy benchmarking reveals that B vacancies provide the strongest thermodynamic stabilization of single atoms, while electronic-structure analysis demonstrates vacancy-dependent modulation of conductivity and metal charge state. Hydrogen adsorption free energies identify Cu@VN and Pd@VB as near-thermoneutral candidates comparable to Pt(111). However, incorporation of electrochemical stability through Pourbaix analysis significantly refines this selection: Cu@VN is unstable at low pH and susceptible to OHads poisoning, whereas Pd@VB remains stable and catalytically accessible across a broad potential-pH range. These results show that descriptor-based HER screening can generate an expanded pool of candidates, but rigorous electrochemical filtering is essential to identify truly robust systems. The presented multi-step strategy provides a general framework for rational discovery of single-atom catalysts on defect-engineered 2D supports.