Nitrogenase catalyzes the reduction of dinitrogen (N-2) to two ammonia (NH3) at its active site FeMo-cofactor through a mechanism involving reductive elimination of two [Fe-H-Fe] bridging hydrides to make H-2. A competing reaction is the protonation cif the hydride [Fe-H-Fe] to make H-2. The overall nitrogenase rate-limiting step is associated with ATP-driven electron delivery from Fe protein, precluding isotope effect measurements on substrate reduction steps. Here, we use mediated bioelectrocatalysis to drive electron delivery to the MoFe protein allowing examination of the mechanism of H-2 formation by the metal-hydride protonation reaction. The ratio of catalytic current in mixtures of H2O and D2O, the proton inventory, was found to change linearly with the D2O/H2O ratio, revealing that a single HID is involved in the rate-limiting step of H-2 formation. Kinetic models, along with measurements that vary the electron/proton delivery rate and use different substrates, reveal that the rate-limiting step under these conditions is the H-2 formation reaction. Altering the chemical environment around the active site FeMo-cofactor in the MoFe protein, either by substituting nearby amino acids or transferring the isolated FeMo-cofactor into a different peptide matrix, changes the net isotope effect, but the proton inventory plot remains linear, consistent with an unchanging rate-limiting step. Density functional theory predicts a transition state for H-2 formation where the S-H+ bond breaks and H+ attacks the Fe-hydride, and explains the observed H/D isotope effect. This study not only reveals the nitrogenase mechanism of H-2 formation by hydride protonation, but also illustrates a strategy for mechanistic study that can be applied to other oxidoreductase enzymes and to biomimetic complexes.