Theoretical design of two-dimensional catalysts for nitrogen reduction and oxygen evolution/reduction reactions

Abstract

Energy and environment are the two most crucial issues for the sustainable future of human society. Electrocatalysts have been playing an important role in the energy conversion/storage and control of environmental issues, thus gaining consistent attention from researchers. Among all the electrocatalytic reactions, nitrogen reduction reaction (NRR), oxygen evolution/reduction reaction (OER/ORR) can particularly convert small molecules (N2, H2O, and O2) into value-added chemicals, provide sustainable energy resources, and decrease the carbon emission by an environmentally friendly way at the same time. The central task of the current research into electrocatalysis is the rational design of catalysts with high reaction activity, high stability, and low cost. Two-dimensional (2D) catalysts, due to their highly exposed active sites induced by the low-dimensional feature and unique physical/chemical properties, have attracted researchers' attention since proposed. Up to now, a large number of 2D catalysts have been reported theoretically or experimentally for NRR, OER, and ORR. Nevertheless, the reaction mechanisms and activity origin of the 2D catalysts are not fully unravelled, and a systematic design principle has not been proposed yet. With the development of the computing power of supercomputers, the theoretical/computational design of the electrocatalysts by first-principles calculations has become an effective approach to understand the catalytic reaction at the atomic level and provide guidelines for experiments. In this thesis, we propose the theoretical design principles for 2D catalysts, and further combine several computational approaches including first-principles density functional theory (DFT) calculations, ab initio molecular dynamics (AIMD) simulations, and machine learning (ML) to design several groups of novel 2D catalysts for electrocatalytic NRR, and bifunctional OER/ORR. Furthermore, we investigate their activity and selectivity origins. Electrocatalytic NRR is a clean and sustainable alternative to the conventional Habor-Bosch process for ammonia production. We first designed a novel set of 2D electrocatalysts—pentagonal transition metal monophosphides (penta-MP, M=Ti, Zr, Hf) for NRR. Penta-MPs are predicted to be dynamically, thermally, and mechanically stable through DFT and AIMD simulations, and their quasi-planar structures and metallicity can enhance the N2 activation on the surface. We also plot the theoretical NRR Gibbs free energy diagrams, which suggest that NRR proceeds through the distal mechanism on penta-MP. In particular, penta-TiP exhibits a low NRR overpotential of 0.56 V and high selectivity of NRR over hydrogen evolution reaction (HER), which is beneficial for NRR. Compared with conventional catalysts, atomically dispersed single-atom catalysts (SACs) generally exhibit higher atomic utilization efficiency close to 100%, high activity and selectivity. We further extend the 2D NRR electrocatalysts to 2D SACs, namely transition metal-tetracyanoquinodimethane (TM-TCNQ). Through a two-step screening process by ab initio calculations, we systematically studied 17 TM­TCNQ systems, and among them, Sc-and Ti-TCNQ are excellent candidates with large N2 adsorption energy and low *N2→*N2H barriers. The N2 activation on them is effective due to the 'acceptance-donation' mechanism, and Gibbs free energy diagrams show that they exhibit low NRR overpotentials of 0.33 and 0.22 V, respectively, through an enzymatic-consecutive mixed pathway. The selectivity of NRR over HER and the thermodynamic/kinetic stability of the Sc-/Ti-TCNQ are also validated. The design of bifunctional catalysts for OER/ORR has emerged as an intriguing topic with applications in metal-air batteries and fuel cells. Considering the merits of SACs, we design a group of them supported on C2N monolayer as promising bifunctional 2D OER/ORR catalysts. In particular, Rh@C2N exhibits lower OER overpotential (0.37 V) than the IrO2(110) benchmark with good ORR activity, while Au@C2N and Pd@C2N are superior ORR catalysts (with the overpotential of 0.38 and 0.40 V) than Pt(111) and their OER performance is also outstanding. More importantly, we discover the origin of the bifunctional catalytic activity by DFT calculations and ML. By DFT, we find a volcano-shaped relationship between the catalytic activity and ΔGO, and finally link them to the normalized Fermi abundance, a parameter based on electronic structure analysis. We further unravel the origin of element-specific activity by ML modeling based on the random forest algorithm that takes outer electron number and oxide formation enthalpy as the two most important factors, and our model can give an accurate prediction of ΔGO with much reduced time cost. In conclusion, the computational design principles have been applied on multifarious 2D catalysts, including 2D materials with basal plane activity and SACs supported on 2D materials, to investigate their NRR/OER/ORR activities. These studies provide a theoretical guideline for the future applications of 2D catalysts for the efficient energy conversion and alleviation of environmental issues.