First principles and machine learning based superior catalytic activities and selectivities for N2 reduction in MBenes, defective 2D materials and 2D π-conjugated polymer-supported single atom catalysts†
Abstract
Production of ammonia through an electrochemical process suffers from two challenging issues, namely low catalytic activity and low Faraday efficiency. Here, we boost the N2 reduction to NH3, while inhibiting the hydrogen evolution reaction (HER) using novel two-dimensional (2D) transition metal borides (MBenes), defect-engineered 2D-materials, and 2D π-conjugated polymer (2DCP)-supported single-atom catalysts (SACs). Density functional theory (DFT) calculations show that nitrogen molecules can be captured in the hollow sites of MBenes, with significant increases in the adsorption strength and NN bond length. Also, defective 2D-materials formed by the vacancy sites of Te, Se and S expose N2 molecules to a specific environment adjacent to three transition metals, which drastically improves the catalytic activity and selectivity (by dramatic increase in the NN bond length up to 1.38 Å). We report a new mechanism for the nitrogen reduction reaction (NRR) as a combination of dissociative and associative mechanisms. A machine-learning (ML) based fast-screening strategy to predict efficient NRR electrocatalysts is described. Overall, TaB, NbTe2, NbB, HfTe2, MoB, MnB, HfSe2, TaSe2 and Nb@SAC exhibit impressive selectivities over HER with overpotentials of 0.44 V, 0.40 V, 0.24 V, 0.60 V, 0.17 V, 0.17 V, 0.64 V, 0.37 V and 0.58 V, respectively. This study opens a new doorway to overcome previous drawbacks of 2D-materials for NRR.