CCL:G: post-Hartree claculations, advice needed

["Alex A. Granovsky" <gran[]classic.chem.msu.su>]

Dear Matthias,

 

I would recommend PC GAMESS for very fast and memory efficient MP4 calculations.

 

Best regards,

Alex Granovsky

 

 

----- Original Message -----

From: Matthias Gottschalk gottschalk~~gfz-potsdam.de

To: Granovsky, Alex, A.

Sent: Friday, June 08, 2007 12:35 PM

Subject: CCL:G: post-Hartree claculations, advice needed

Dear colleagues

I hope that somebody could give me some advice.

I want to calculate energies for dimers and trimers to extract accurate potentials using post Hartree methods (also for "large" distances > 5 Angstroms). I intend to do this for closed- and open-shell systems (UHF) consisting of simple atoms (in the moment).

Because I am not a computational chemistry specialist I did some reading and as a result I concluded that methods like MP4, CCSD and QCISD might be appropriate using a basis set like aug-cc-pVQZ might be appropriate.

I did success-full calculations for Ar dimers with gaussian g03 reproducing published results which made me confident. I failed for trimers using these evolved methods so long, however (might be due to that I have only a 32 bit compiled version and the required real memory exceeds 2.2 GB possible with 32 bit, I do have 8 GB, however). MP2 trimers with more simple basis sets work, but that might be not what I want?

Now my questions are:

1. are the methods described above adequate?

2. which of the many programs available could be recommended to use?

Thanks

Matthias

-- 

Prof. Dr. Matthias Gottschalk

GeoForschungsZentrum

Sektion 4.1

Telegrafenberg

14473 Potsdam

Germany

tel/fax  +49 (0) 331 288-1418/1402