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

1. Problem Statement

The Hydrogen Evolution Reaction (HER) is critical for sustainable hydrogen production, yet current high-performance catalysts rely heavily on scarce noble metals like Platinum (Pt). Single-Atom Catalysts (SACs) offer a solution by maximizing atomic utilization, but identifying stable and active motifs remains challenging.

• The Support Challenge: Hexagonal boron nitride (h-BN) is an attractive 2D support due to its chemical inertness and large surface area, but its pristine form is too insulating and inert to anchor metal atoms effectively.
• The Screening Gap: While defect engineering (creating vacancies) can activate h-BN, systematic identification of which metal-vacancy combinations are truly viable is limited. Previous studies often rely solely on hydrogen adsorption energy ($\Delta G_{H^*}$) as a descriptor, neglecting thermodynamic stability against aggregation and, crucially, electrochemical stability under operating conditions (pH and potential).

2. Methodology

The authors employed a multi-step computational screening protocol using Density Functional Theory (DFT) to evaluate transition and coinage metals (Ni, Cu, Ru, Rh, Pd, Ag, Ir, Pt, Au) anchored on defect-engineered h-BN.

• Computational Framework:
  • Software/Method: VASP with PBE-GGA functionals, PAW potentials, and DFT-D3 dispersion corrections.
  • Models: 4×4 supercells of monolayer h-BN containing three defect types: Boron vacancy ($V_B$), Nitrogen vacancy ($V_N$), and Boron-Nitrogen divacancy ($V_{BN}$).
  • Key Calculations:
    1. Binding Energy ($E_b$): Compared against elemental cohesive energy ($E_{coh}$) to assess thermodynamic stability against clustering.
    2. Electronic Structure: Analysis of Band Gaps, Density of States (DOS), and Bader charge transfer to understand conductivity and metal oxidation states.
    3. HER Activity: Calculation of Gibbs free energy of hydrogen adsorption ($\Delta G_{H^*}$) using the Computational Hydrogen Electrode (CHE) model.
    4. Stability Filters:
       • Kinetic Stability: Ab initio Molecular Dynamics (AIMD) at 298 K.
       • Electrochemical Stability: Construction of Pourbaix diagrams to map stability domains (dissolution vs. surface poisoning by OH or H) across pH and electrode potential.

3. Key Contributions

• Multi-Level Screening Protocol: The paper introduces a rigorous filtering strategy that moves beyond simple $\Delta G_{H^*}$ descriptors. It integrates thermodynamic anchoring, electronic modulation, kinetic stability, and electrochemical stability (Pourbaix analysis) to identify truly robust catalysts.
• Defect-Dependent Electronic Modulation: The study elucidates how specific vacancy types ($V_B$, $V_N$, $V_{BN}$) fundamentally alter the electronic environment, charge state, and conductivity of the anchored metal, directly influencing catalytic activity.
• Refinement of Candidate Selection: The work demonstrates that descriptor-based screening alone can yield false positives. By applying electrochemical stability filters, the pool of viable candidates is significantly narrowed, revealing that a near-thermoneutral $\Delta G_{H^*}$ is insufficient without electrochemical resilience.

4. Key Results

A. Vacancy Characteristics and Metal Anchoring

• $V_B$ (Boron Vacancy): Provides the strongest thermodynamic stabilization for transition metals (Ru, Rh, Ni, Ir) due to strong coordination with surrounding N atoms. It often induces metallic behavior (band gap closure) and results in positively charged metal centers (electron transfer to N).
• $V_N$ (Nitrogen Vacancy): Generally offers weaker stabilization. Metals here tend to acquire negative charges (electron donation from B-rich environment).
• $V_{BN}$ (Divacancy): Shows intermediate behavior but is unique in stabilizing all tested metals below their cohesive energy limits.

B. HER Activity Screening ($\Delta G_{H^*}$)

• Initial screening based on $\Delta G_{H^*}$ identified two top candidates comparable to Pt(111):
  1. Pd@$V_B$: $\Delta G_{H^*} \approx 0.083$ eV (near-thermoneutral).
  2. Cu@$V_N$: $\Delta G_{H^*} \approx 0.048$ eV (near-thermoneutral).
• Other metals showed either too strong (e.g., Pt, Ru) or too weak (e.g., Ag, Au) hydrogen binding depending on the vacancy site.

C. Stability Analysis and Final Selection

The application of stability filters drastically altered the ranking:

• Cu@$V_N$:
  • Thermodynamics: Binding energy is close to Cu's cohesive energy, suggesting a risk of agglomeration.
  • Electrochemical: Pourbaix analysis reveals instability. In acidic conditions, Cu dissolves ($Cu^{2+}$). In alkaline/neutral conditions, the active site is poisoned by hydroxyl ($OH_{ads}$) coverage, blocking HER activity.
  • Verdict: Rejected due to lack of electrochemical robustness.
• Pd@$V_B$:
  • Thermodynamics: Strong binding ($E_b \ll E_{coh}$), ensuring resistance to clustering.
  • Kinetic: AIMD simulations confirm structural stability at 298 K.
  • Electrochemical: The Pourbaix diagram shows a wide stability window where the clean Pd site is stable, transitioning directly to $H_{ads}$ coverage at cathodic potentials without dissolution or competitive $OH_{ads}$ poisoning.
  • Verdict: Selected as the optimal candidate.

5. Significance

• Identification of a Robust Catalyst: The study identifies Pd anchored at a Boron vacancy in h-BN (Pd@$V_B$) as a highly promising, stable, and active SAC for HER, potentially offering a cost-effective alternative to bulk Pt.
• Methodological Advancement: The paper establishes that electrochemical stability analysis (Pourbaix diagrams) is an essential, non-negotiable step in computational catalyst discovery. Relying solely on $\Delta G_{H^*}$ can lead to the selection of catalysts that are thermodynamically active but electrochemically unstable or poisoned under operating conditions.
• General Framework: The proposed multi-step strategy (Thermodynamics $\rightarrow$ Electronic Structure $\rightarrow$ Kinetic Stability $\rightarrow$ Electrochemical Stability) provides a transferable framework for the rational design of single-atom catalysts on various 2D defect-engineered supports.