dinitrogen · non-innocent ligands · osmium · radicals · rutheniumWith a limited number of well-understood exceptions (such as the square-planar d 8 configuration and complexes with strongly p-donating ligands), complexes of transition metals that belong to the 4d and 5d series most typically follow the so-called EAN (effective atomic number) rule by adopting a closed-shell 18-electron configuration. Relative to the 3d metals, the metals of the 4d and 5d series have more diffuse and more polarizable d orbitals, leading to greater bond covalence and to lower pairing energies and consequently to lower stability for open-shell configurations, especially in the lower oxidation states and especially for the late transition metals.[1] Coordinative saturation of the metal center is achieved by coordination of a suitable number of one-and two-electron ligands and/or by forming metal-metal bonds in oligonuclear clusters. Nevertheless, enforcing mononuclear metalloradical configurations through the use of specially designed ligands can lead to peculiar properties and reactivity. A good example is the use of bulky porphyrins or geometrically constrained alkene-containing ligands to give stable porphyrin rhodium(II) complexes or alkene rhodium(II) and alkene iridium(II) complexes that have shown interesting applications in C À H activation and in catalysis. [2,3] For the heavier Group 8 elements ruthenium and osmium, complexes with the formal oxidation state I and a metalloradical character were previously produced, generally by electrochemical methods, only as unstable species or could not be isolated. Pilloni et al. pointed out as early as 1977 that the coordination geometry could be an important factor allowing the generation of stable d 7 mononuclear complexes and that the presumed lability of the chloride ligand in complex [Ru(dppp) 2 Cl] (dppp = 1,3-bisdiphenylphosphinopropane) opens decomposition pathways through disproportionation.[4] A longer-living but still not isolable ruthenium(I) complex, [Ru(PP 3 )Cl], [5] was obtained by using the "Sacconitype" tetradentate phosphine ligand PP 3 = P(CH 2 CH 2 PPh 2 ) 3 , which is known to stabilize M(L 4 ) moieties either in an octahedral (with two additional cis ligands) or in a trigonalbipyramidal (with one additional axial ligand) coordination geometry. A key to further stabilize the construct could be the incorporation of the anionic function into the tetradentate ligand to yield a M(XL 3 ) moiety. On the basis of this idea, Peters has recently developed tris(o-phosphinophenyl)silyl ligands (their relationship to the Sacconi system is illustrated in Scheme 1), and shown that they are capable of stabilizing dinitrogen complexes of Fe [6][7][8] More recently, the same ligand (with R = iPr) has successfully been used to generate the first stable mononuclear ruthenium(I) and osmium(I) metalloradicals (see Scheme 2).[9] The precursor complexes [(SiP The two dinitrogen complexes (1 and 2) were also investigated electrochemically, revealing a reduction and an oxidation. The red...