The QTAIM approach to chemical bonding between transition metals and carbocyclic rings: a combined experimental and theoretical study of (eta(5)-C5H5)Mn(CO)3, (eta(6)-C6H6)Cr(CO)3, and (E)-{(eta(5)-C5H4)CF=CF(eta(5)-C5H4)}(eta(5)-C5H5)2Fe2.

Experimental charge densities for (C(5)H(5))Mn(CO)(3) (2), (eta(6)-C(6)H(6))Cr(CO)(3) (3), and (E)-{(eta(5)-C(5)H(4))CF=CF(eta(5)-C(5)H(4))}(eta(5)-C(5)H(5))(2)Fe(2) (4) have been obtained by multipole refinement of high-resolution X-ray diffraction data at 100 K. The resultant densities were analyzed using the quantum theory of atoms in molecules (QTAIM). The electronic structures of these and related pi-hydrocarbyl complexes have also been studied by ab initio density functional theory calculations, and a generally good agreement between theory and experiment with respect to the topological parameters was observed. The topological parameters indicate significant metal-ring covalency. A consistent area of disagreement concerns the topology of the metal-ring interactions. It is shown that because of the shared-shell bonding between the metal and the ring carbons, an annulus of very flat density rho and very small wedge rho is formed, which leads to topologically unstable structures close to catastrophe points. This in turn leads to unpredictable numbers of metal-C bond paths for ring sizes greater than four and fewer M-C bond paths than expected on the basis of the formal hapticity. This topological instability is a general feature of metal-pi-hydrocarbyl interactions and means that a localized approach based on individual M-C(ring) bond paths does not provide a definitive picture of the chemical bonding in these systems. However, other QTAIM indicators, such as the virial paths, the delocalization indices, and the source function, clearly demonstrate that for the n-hapto (eta(n)-C(n)H(n))M unit, there is generally a very similar level of chemical bonding for all M-C(ring) interactions, as expected on the basis of chemical experience.