Machine-learning-accelerated screening of hydrogen evolution catalysts in MBenes materials

Abstract

Machine learning (ML) models combined with density functional theory (DFT) calculations are employed to screen and design hydrogen evolution reaction (HER) catalysts from various bare and single-atom doped MBenes materials. The values of Gibbs free energy of hydrogen adsorption (ΔG_H*) are accurately predicted via support vector algorithm only by using simply structural and elemental features. With the analysis of combined descriptors and the feature importance, the Bader charge transfer of surface metal is a key factor to influence HER activity of MBenes. Co/Ni₂B₂, Pt/Ni₂B₂, Co₂B₂, Os/Co₂B₂ and Mn/Co₂B₂ are screened from 271 MBenes and MXenes as active catalysts, with the near-zero ΔG_H* of 0.089, −0.082, −0.13, −0.087 and −0.044 eV, respectively. Finally, stable Co₂B₂ and Mn/Co₂B₂ are considered as the excellent HER catalysts due to |ΔG_H*| < 0.15 eV over a wide range of hydrogen coverages (θ from 1/9 to 5/9). The present work suggests that ML models are competitive tools in accelerating the screening of efficient HER catalysts.