Enhanced Hydrogen Evolution Performance at the Lateral Interface between Two Layered Materials Predicted with Machine...

Area of Science:

• Materials Science
• Catalysis
• Computational Chemistry

Background:

• Sustainable hydrogen production requires economical and effective catalysts.
• Low-dimensional interfacial engineering enhances catalytic activity for the hydrogen evolution reaction (HER).
• Two-dimensional lateral heterostructures (LHSs) offer tunable properties for catalysis.

Purpose of the Study:

• To investigate hydrogen adsorption Gibbs free energy (ΔG_H) in various 2D LHSs using DFT.
• To develop universal descriptors for predicting HER catalytic activity in 2D materials.
• To employ machine learning (ML) to identify promising HER catalyst candidates.

Main Methods:

• Density Functional Theory (DFT) calculations to determine ΔG_H for hydrogen adsorption in MX₂/M'X'₂ and MX₂/M'X' LHSs.
• Development of three universal descriptors based on atomic information near adsorption sites.
• Training ML models (regression and classification) using DFT results and experimental data.

Main Results:

• LHS MX₂/M'X' interfaces exhibit enhanced HER reactivity due to metallic behavior and stronger hydrogen absorption.
• ML models achieved high accuracy (R²=0.951 for regression, F₁=0.749 for classification).
• MoS₂/ZnO identified as the most promising HER catalyst with ΔG_H = -0.02 eV and low overpotential (-171 mV).

Conclusions:

• Interfacial engineering of 2D LHSs is a viable strategy for developing efficient HER catalysts.
• The developed descriptors and ML models can accurately predict catalytic performance.
• MoS₂/ZnO presents a superior candidate for sustainable hydrogen production.