Electrocatalytic
reduction of N2 to ammonia (eNRR) provides
a sustainable alternative to an energy-intensive Haber–Bosch
process. However, the poor activity and selectivity of the eNRR catalysts
limit their large-scale applications. Recently, transition metals
(TMs) doped graphitic carbon nitride-based single-atom catalysts (SACs)
have emerged as a very promising class of catalysts. Inspired by their
activity and selectivity for a range of reactions, using density functional
theory we investigated TMs (3d, 4d, and 5d series) anchored on h-C4N3 as possible catalysts for eNRR. We employed
a search scheme for finding an efficient TM SAC for eNRR, based on
its optimum N2 adsorption, N2 protonation feasibility,
and selectivity against HER. The optimum bond strength of TM–N
bond is characterized by a change in the magnetic moment of the metal
upon N2 adsorption, and the charge transferred (Δq) from TM to N2 molecule. We also established
the number of valence electrons (group number) of the TM as a potential
descriptor to determine the feasibility of N2 protonation,
which ultimately decides the activity and selectivity of an eNRR catalyst.
SACs with TMs belonging to the groups 5–7 are found to have
the highest activity. Mo- and W-SACs emerge as the most suitable candidates
after the third level of screening, which is based on the selectivity
toward eNRR. The overpotentials for both Mo- and W-SACs are 0.02 and
0.33 V versus SHE, respectively. The thermodynamic analysis suggests
that Mo-based SAC is the most active catalyst for eNRR with an ultralow
overpotential and high faradaic efficiency for eNRR. The Mo-SAC mimics
the naturally occurring nitrogenase enzyme, which also has a Mo atom
as the active site. Our in-depth analysis provides a rational design
of a new class of highly efficient catalysts for the electrochemical
NRR under ambient conditions.