The salt [K(18-crown-6)]2[Ru(CN)2(CO)3] ([K(18-crown-6)]2[1]) was generated by the reaction of Ru(C2H4)(CO)4 with [K(18-crown-6)]CN. An initial thermal reaction gives [Ru(CN)(CO)4]-, which, upon ultraviolet (UV) irradiation, reacts with a second equiv of CN-. Protonation of [1]2- gave [HRu(CN)2(CO)3]- ([H1]-), which was isolated as a single isomer with mutually trans cyanide ligands. The complex cis,cis,cis-[Ru(pdt)(CN)2(CO)2]2- ([2]2-) was prepared by the UV-induced reaction of [1]2- with propanedithiol (pdtH2). The corresponding iron complex cis,cis,cis-[Fe(pdt)(CN)2(CO)2]2- ([3]2-) was prepared similarly. The pdt complexes [2]2- and [3]2- were treated with Fe(benzylideneacetone)(CO)3 to give, respectively, [RuFe (μ-pdt)(CN)2(CO)4]2- ([5]2-) and [Fe2(μ-pdt)(CN)2(CO)4]2- ([4]2-). The pathway from [3]2- to Fe2 complex [4]2- implicates intermetallic migration of CN-. In contrast, the formation of [5]2- leaves the Ru(CN)2(CO) center intact, as confirmed by X-ray crystallography. The structure of [5]2- features a "rotated" square-pyramidal Fe(CO)2(μ-CO) site. NMR measurements indicate that the octahedral Ru site is stereochemically rigid, whereas the Fe site dynamically undergoes turnstile rotation. 57Fe Mössbauer spectral parameters are very similar for rotated [5]2- and unrotated Fe2 complex [4]2-, indicating the insensitivity of that technique to both the geometry and the oxidation state of the Fe site. According to cyclic voltammetry, [5]2- oxidizes at E1/2 ∼ -0.8 V vs Fc+/0. Electron paramagnetic resonance (EPR) measurements show that 1e- oxidation of [5]2- gives an S = 1/2 rhombic species, consistent with the formulation Ru(II)Fe(I), related to the Hox state of the [FeFe] hydrogenases. Density functional theory (DFT) studies reproduce the structure, 1H NMR shifts, and infrared (IR) spectra observed for [5]2-. Related homometallic complexes with both cyanides on a single metal are predicted to not adopt rotated structures. These data suggest that [5]2- is best described as Ru(II)Fe(0). This conclusion raises the possibility that for some reduced states of the [FeFe]-hydrogenases, the [2Fe]H site may be better described as Fe(II)Fe(0) than Fe(I)Fe(I).