The first half-reaction of nitric oxide synthase (NOS) is investigated by means of quantum mechanical/molecular mechanical (QM/MM) calculations. An energetically feasible arginine hydroxylation path was found only when the iron-oxy complex accepted one proton from an external source. The so formed species has not been considered in heme chemistry; it is described as Por(+*)Fe(III)-OOH and is characterized by the same molecular constituency as the more known ferric-hydroperoxide species, compound 0, but has a cation-radical porphyrin moiety. The reaction itself is found to involve proton coupled electron transfer (PCET) and oxygen coupled electron transfer (OCET) steps en route to the formation of compound I and the ultimate monooxygenation of arginine. The cofactor H(4)B turns out to be a key player in the mechanism acting alternatively as an electron donor (when neutral) and an electron sink (when in its radical-cation state) and, thereby, providing the electron transfer component in the various coupled proton and oxygen transfer steps (see Scheme 4 ). The various pieces of this mechanism account for many of the experimental observations, such as the following: (a) the origins of the second proton supplied to the heme, (b) the elusiveness of compound I, (c) the inactivity of peroxide-shunt pathways in NOS first half-reaction, (d) the inhibition of the H(4)B analogue 4-amino-H(4)B due to protonation at the N3 position, (e) the roles of Trp188 (iNOS numbering) and the crystal water at the active site (W115), and so on. Alternative mechanistic hypotheses are tested and excluded, and a new mechanism for the NOS second half-reaction is proposed.