Unraveling the Major Differences between the Trinuclear Cyclopentadienylmetal Carbonyl Chemistry of Cobalt and That of NickelA Theoretical Study

The geometries and energetics of the trinuclear cyclopentadienylmetal carbonyls Cp3M3(CO)n (Cp = η5-C5H5); M = Co, Ni; n = 3, 2, 1, 0) have been investigated by density functional theory. The cobalt and nickel systems are found to be rather different owing to the different electronic configurations of the metal atoms. For cobalt, the small calculated energy separation of 5.0 kcal/mol between the two lowest-energy singlet Cp3Co3(μ3-CO)(μ-CO)2 and Cp3Co3(μ-CO)3 tricarbonyl structures accounts for the experimental results of both isomers as stable species that can be isolated and structurally characterized by X-ray crystallography. The corresponding Cp3Ni3(CO)3 species in the nickel system are predicted not to be viable owing to exothermic CO dissociation to give the experimentally observed very stable Cp3Ni3(μ-CO)2, which is found to be the lowest-energy isomer by a substantial margin of ∼25 kcal/mol. In all of the low-energy Cp3M3(CO)n (n = 2, 1) structures, including that of the experimentally known triplet spin state Cp3Co3(μ3-CO)2, all of the carbonyl groups are face-bridging or face-semi-bridging μ3-CO groups bonded to all three metal atoms of the M3 triangle. In the lowest-energy carbonyl-free Cp3M3 (M = Co, Ni) structures, agostic C–H–M interactions are found using hydrogens of the Cp rings. In addition, the lowest-energy Cp3Ni3 is the only structure among all of the low-energy Cp3M3(CO)n (M = Co, Ni; n = 3, 2, 1, 0) structures in which each Cp ring is a bridging rather than terminal ligand