From our combined experimental and computer modeling study we found a structurally and kinetically well-defined second coordination shell around chromium(III) ions in aqueous solution. Strong hydrogen binding due to polarization of first coordination sphere water molecules leads to a mean coordination number of 12.94 water molecules in the second shell and to short first shell hydrogen−second shell oxygen distances of about 1.4 Å. The experimentally measured exchange rate constant of = (7.8 ± 0.2) × 109 s-1 (ΔH = 21.3 ± 1.1 kJ mol-1, ΔS = +16.2 ± 3.7 J K-1 mol-1) corresponds to a lifetime of 128 ps for one water molecule in the second coordination shell and compares very well with a lifetime of 144 ps as observed from molecular dynamics simulation of a [Cr(H2O)6]3+ complex in aqueous solution. The geometry and the partial atomic charges of [Cr(H2O)6]3+ were determined by density functional theory (DFT) calculations. Water exchange from the second coordination shell to the bulk of the solution proceeds between a H2O sitting in the second shell and an adjacent one which just entered this shell from the bulk. By a small rotation of the first coordination shell water molecule, one of its two hydrogen bonds jumps to the entered water molecule and the one which lost its hydrogen bond leaves the second shell of the [Cr(H2O)6]3+. This associative reaction mode is a model for water exchange between water molecules which are bound by strong hydrogen bonds, as in the case for strongly polarizing 3+ ions such as Al3+ or Rh3+. Furthermore, the exchange phenomenon between second sphere and bulk water involving only two adjacent water molecules is strongly localized and independent of other water molecules of the second shell. In this respect it may be considered as a starting point for a study of water exchange on a protonated metal oxide surface.

Quantum chemical calculations based on density functional theory have been performed on Cr(CO)6, (eegr6-C6H6)Cr(CO)3 and (eegr6-C6H6)Cr(CO)2(CS) at the local and nonlocal level of theory using different functionals. Good agreement is obtained with experiment for both optimized geometries and metal-ligand binding energies. In particular, a comparison of metal-arene bond energies calculated for the (eegr6-C6H6)Cr(CO)3 and (eegr6-C6H6)Cr(CO)2(CS) complexes correlates well with kinetic data demonstrating that substitution of one CO group by CS leads to an important labilizing effect of this bond, which may be primarily attributed to a larger pgr-backbonding charge transfer to the CS ligand as compared with CO.

The parametrization of the EHMO-ASED method we have recently suggested for organometallics is shown to be also applicable, in principle without any modification, to derive the major structural parameters of second-row transition metal systems such as carbonyls or metallocenes. Furthermore, this model leads to satisfactory results when used to calculate the structure of compounds as large as (N-methylindole)tricarbonylchromium(0) or (phenylo.xazoline) tricarbonylchromlum(0) with full geometry optimization of the ligands.