The catalytic cycle for H-2 oxidation in [NiFe] D. gigas hydrogenase has been investigated through density functional theory (DFT) calculations on a wide variety of redox and protonated structures of the active site model, (CO)(CN)(2)Fe(mu-SMe)(2)Ni(SMe)(2). DFT calculations on a series of known LFe(CO)(CN)(L')(n-) (L = Cp or Cp*, L' = CN, CO, CNCH3; n = 0, 1, 2) complexes are used to calibrate the calculated CO bond distances with the measured IR stretching frequency. By combining this calibration curve with the energy and CO bond distance of the DFT calculations on the active site model and the experimental IR frequencies on the enzyme, the redox states and structures of active site species have been determined: Ni-B is a Ni(III)-Fe(II) species, Ni-SI(a) is a Ni(LI)-Fe(II) species, Ni-SI(b) has a protonated terminal sulfur (Ni bound), Ni-R is a Ni(II)-Fe(II) dihydrogen complex with H-2 bound at Fe, and Ni-C is a Ni(III)-Fe(II) species with an Fe-H-Ni bridge. The latter species returns to W-SI through a Ni(I)-Fe(II) intermediate, which is potentially observable. Protonation of the Ni bound terminal sulfur results in a folding of the Fe(mu-S)(2)Ni framework. Dihydrogen activation is more exothermic on the Ni(III) species than on the corresponding Ni(II) or Ni(I) species. Our final set of proposed structures are consistent with IR, EPRI ENDOR,and XAS measurements for these species, and the correlation coefficient between the measured CO frequency in the enzyme and the CO distance calculated for the model species is 0.905.