PC GAMESS-related discussion club

Alex Granovsky
gran@classic.chem.msu.su

thanks for answering Pasha's question. I'm attaching the archive with your sample to this message and hope it will be useful for others.

I would like to mention the alternative way - you can just specify the negative nuclear charge to ignore that nucleus in calculations but put  proper basis functions on its place automatically.

Best regards,
Alex


On Thu Nov 22 '07, Pedro Silva wrote
------------------------------------
>On Thu Nov 22 '07, Solntsev Pasha wrote
>---------------------------------------
>Dear Pasha,

>I have placed a demonstration for He2 on my homepage. You can download it from:

> www2.ufp.pt/~pedros/science/bsse.zip

>Basically, you should prepare three files. The first one (let's call it "complete molecule") has all of your atoms and basis sets. The others have all basis sets but only a set of the atoms (let's call them "fragment A" and "fragment B"). These latter files should be prepared in ther following way:
>a) prepare your files as usual, using COORD=UNIQUE. Then

>b) The basis set should be included explicitly, with all exponents, etc. in $DATA, rather than in $BASIS. Basis sets can be conveniently downloaded from

http://www.emsl.pnl.gov/forms/basisform.html

c) Substitute the charge of the atoms which should be absent from the calculation but whose basis functions should be kept.

>Run your three files. The real interaction energy between the fragments A and B should be :

>FINAL ENERGY "complete molecule" - FINAL ENERGY "fragment A" -FINAL ENERGY "fragment B".

>If you compute the energies for separated fragments A and B WITHOUT the extra basis functions on the absent atoms, you get the wrong interaction energy. The difference between the two computed interaction energies is the BSSE.

>I hope I have been clear (and that there are no major errors in this post :-)

>Pedro Silva

http://www2.ufp.pt/~pedros/science/science.htm

This message contains the 17 kb attachment [ bsse.zip ] BSSE calculation sample