CCL:G: Metal oxidation states and charge separation

["Serge Gorelsky" <gorelsky()gmail.com>]

 Sent to CCL by: "Serge Gorelsky" [gorelsky()gmail.com]
>  Sent to CCL by: "Richard Andrew Huhn" [huhnra#%#mit.edu]
>  I have a crystal structure with two metal-ligand complexes in the unit cell. It
>  is of the type L-M-L L'-M-L'. We are unsure of the oxidation states of the
>  metal: it may be, for example, L-M(0)-L L'-M(II)-L', or L-M(I)-L[+]
>  L'-M(I)-L'[-].
>  Are comparisons possible of DFT energies / geometries for these two hypothetical
>  species, and if so, how does one specify in G03 the charge separation in the
>  latter species? Literature references to similar calculations would be helpful!
>  Thank you for your help,
 this topic has been discussed in the CCL posts before. In a nutshell,
 you need to give Gaussian 03 the guess wave functions that are the sum
 of fragment wave functions such as L-M(0)-L L'-M(II)-L', or
 L-M(I)-L[+] L'-M(I)-L'[-]. This can be easily done by AOMix-CDA (see
 Appendix II in the AOMix manual at www.sg-chem.net) and GUESS=CARDS
 keyword in Gaussian 03.
 --
 Best regards,
  Serge Gorelsky