Despite recent intense interest in the development of catalysts for the electrochemical nitrogen reduction reaction (ENRR), mechanistic understanding and catalyst design principles remain lacking. In this work, we develop as trategy to determine the density of initial and steady-state active sites on ENRR catalysts that follow the Mars-van Krevelen mechanism via quantitative isotope-exchange experiments. This method allows the comparison of intrinsic activities of active sites and facilitates the identification and improvement of active-site structures for ENRR. Combined with detailed density functional theory calculations,w es how that the ratelimiting step in the ENRR is likely the initial N Nb ond activation via the addition of ap roton and an electron to the adsorbed N 2 on the Nv acancies to form N 2 H. The methodology developed and mechanistic insights gained in this work could guide the rational catalyst design in the ENRR.Ammonia synthesis via the Haber-Bosch process is ap illar of modern agriculture,which converts the abundant but inert dinitrogen in the atmosphere to the key precursor of nitrogenbased fertilizers (N-fertilizers). [1] Despite more than acentury of optimization, the Haber-Bosch process remains energy and carbon intensive. [2][3][4][5][6][7] Distributed and modular ammonia synthesis via the electrochemical nitrogen reduction reaction (ENRR) at or close to ambient conditions,p owered by renewable electricity is an attractive alternative because it allows as-needed and on-site production of ammonia, and in turn N-fertilizers,from N 2 and water. [3] Widespread adoption of the ENRR for ammonia production could drastically reduce the carbon footprint of agricultural activities.ENRR is also compatible with the intermittencyo fr enewable energy sources,e .g., solar and wind energy,a sa mmonia and Nfertilizers can be produced and stored when renewable electricity is abundant or even in surplus.Ar ecent flurry of studies on the ENRR led to the discovery of many catalysts with remarkable ammonia production rates and selectivities. [8][9][10][11][12][13][14][15][16][17][18] Mechanistic insights gained so far lean heavily on computational investigations, with relatively little experimental input and insufficient theory/experiment integration. Forexample,reported ammonia production rates are typically normalized by the geometric area or electrochemical surface area determined by capacitance measurements, [19] which make the comparison of the intrinsic activity among different catalysts challenging. Catalyst development based on the comparison of (electrochemical) surface area normalized activities risks missing highly active structures/phases if they exist in low densities. Our recent work highlights this risk, as VN 0.7 O 0.45 ,rather than the bulk VN phase,inthe vanadium nitride (referred to as the oxygen-modified VN catalyst, or VNO below) is identified as the active phase. [10] Thus,i dentification and quantification of active sites in the ENRR are key to developing catalyst design principles. [20]...