Hi Yavar,

I think in the plane wave community people use (maximally localized) Wannier functions for localizing wave functions. I'm sorry I don't know if transformations to Gaussian-type orbitals have been described (I reckon they have).

I'm guessing that you want to transform to a Pople basis set, so you can compare your cluster results to a molecular DFT code and do a proper comparison of clusters and periodic systems?

CRYSTAL is a periodic DFT code that can use Pople basis sets naturally. It can also treat 0D systems.

There are other periodic DFT codes that use atomic orbitals (DMol3, FHI-AIMS, ADF/BAND) which can all treat clusters and periodic systems on equal footing with the same basis set.

Dear  Fedor ,

Thanks for your time and consideration,

I will try your comments soon and I Know about Crystal capabilities ,

but there is a small problem with solid state codes, Considering PCM models in quantum chemistry codes, may affect resulting wave-functions.

And I am trying to project the wavefunctions obtained from a periodic code (for example PWscf or SIESTA)  on Pople basis space.

I will write more for you,

Thanks again

Best

PCM for periodic systems is available in BAND which has COSMO for 1D and 2D periodic systems. You can use the same basis functions (Slaters, not Gaussians), integration etc. for 0D systems either in BAND or in the molecular DFT code ADF.

SM12 (Generalized Born solvation model from Minnesota) for ADF and BAND is currently under development.