PC GAMESS-related discussion club

Alex Granovsky
gran@classic.chem.msu.su

at present, GVB code in PC GAMESS cannot be combined with DFT and TD methods.

If the complete active space for MCSCF calculations can be reasonably truncated still capturing all the important physics, you can try CASSCF. The direct SOSCF CASSCF code in PC GAMESS is very efficient (and was very recently considerably improved again) and allows to handle systems with up to, say, 2000 basis functions and reasonable active spaces routinely, especially if running in parallel. The semidirect fitting-based MCQDPT2 code in PC GAMESS is also very efficient and allows to handle problems of virtually any size.

However, if the active space of the reasonable size cannot be selected for your particular systems, you have no ways how to improve your results beyond TDDFT/TDHF.

Best regards,
Alex Granovsky



On Tue May 27 '08, shimshon wrote
---------------------------------
>I am using TDDFT and TDHF to calculate two photon properties.
>The size of my molecules is around 30-60 heavy atoms, and I commonly use DZV and a basis set.

>I calculate transition dipole moments and excitation energies for the first few excited states.

>A suitable scaling & benchmarking give satisfactory agreement with experiments.

>For some cases, higher theory is needed - probably due to charge transfer on the excited state that demands a use of more then a single reference computation.

>I would be grateful for any suggestion regarding any improved method that could be used for this size of molecules.

>(maybe some kind of VB will do? could it be combined with TD methods?)

>Thank you in advanced,
>Shimshon Kallush