Assignment of the electronic spectra of [Mo(CN)(8)](4)(-) and [W(CN)(8)](4)(-) by Ab initio calculations.

CASPT2 calculations are performed on the dodecahedral and square antiprismatic isomers of the [Mo(CN)(8)](4)(-) and [W(CN)(8)](4)(-) complexes. The high-energy experimental bands above 40000 cm(-)(1) are assigned to MLCT transitions. The experimental observed trend of the extinction coefficients for the molybdenum and tungsten complex is reproduced by our CASSCF oscillator strengths. All bands below 40000 cm(-)(1) can be ascribed to ligand-field transitions, although small contributions from forbidden MLCT transitions cannot be excluded. In order to account for all experimental bands in the electronic spectrum of these octacyanocomplexes, a dynamic equilibrium in solution between the two isomeric forms must be hypothesized. Spin-orbit coupling effects are found to be more important for the square antiprismatic isomers; in particular, large singlet-triplet mixings are calculated for this isomer of [W(CN)(8)](4)(-). Ligand-field and Racah parameters as well as spin-orbit coupling constants are determined on the basis of the calculated transition energies. The obtained values for these parameters support the recently proposed model for exchange interactions in magnetic clusters and networks containing pentavalent octocyanometalates of molybdenum and tungsten.

The electronic structure and spectrum of Mo(CN)8(3-).

State of the art CASSCF and CASPT2 calculations have been performed to elucidate the nature of the electronic transitions observed in the experimental spectrum of the octacyanomolybdate(V) cation. Assuming a triangular dodecahedral structure for this complex gives a convincing agreement between theory and experiment. All absorption bands are assigned to low-lying charge-transfer transitions involving excitations from ligand orbitals to 4dx2-y2. The calculated molecular orbitals reveal weak pi interactions between the metal and ligand orbitals, compared to much stronger sigma interactions. This calculated electronic structure substantiates the previous hypothesis concerning the giant spin ground states of magnetic clusters and networks containing Mo(CN)8(3-) as a constituent part.