High throughput screening for electrocatalysts for nitrogen reduction reaction using metal-doped bilayer borophene: A combined approach of DFT and machine learning

Abstract

NH3 plays an important role in industrial production, while it is still a challenge to develop a promising catalyst for NH3 synthesis. In particular, the metal borides have attracted much attention because of their excellent catalyst performances towards N2. In this paper, we have theoretically studied the N2 reduction reaction (NRR) by employing the transition metal (TM = Co, Cr, Cu, Fe, Mn, Mo, Ni, Sc, Ti, V, Zn, Nb, Pd, Rh, Ru or Zr) doped bilayer borophene (BB). After the screening, it is found that Sc-doped BB (Sc@BB) exhibits excellent NRR performance with the limiting potential of -0.43 V, and the hydrogen evolution reaction (HER) could be well suppressed. The limiting potential could be further improved into -0.40 V by applying 0.30 % uniaxial strains to the system. In addition, the machine learning method was employed to understand the catalytic mechanism, the results revealed that the NRR performance of TM@BB is strongly related to the inherent properties of TM atoms, such as the d-electron number, electronegativity, first ionization energy, atomic radius, and atomic number. Therefore, our paper not only proposed a promising electrocatalyst for NRR, but also further extend the application of transition metal-boron based two dimensional materials in NRR.