[molpro-user] Counterpoise correction for ion-molecular complexes

[Kirk Peterson]

Dear Shengfu,

your input file below seems to be mixed between Li and Na.  After your dummy,He directive presumably you want to
calculate Na+ and this is consistent with your wf command. Note however that the default frozen core for Na is 1s2s2p,
hence 10 electrons.  Therefore unless you redifine the electrons that are being correlated in the ccsd(t) step with a "core"
directive, there are no electrons to correlate and presumably this results in the "there is no orbital..." error (please make
sure you use a basis set designed for correlating these electrons, e.g., cc-pwCVTZ).  Note also that
Na+ (or Li+) is closed shell, so specifying rccsd(t) here will "cost" you about a factor of 2 greater cpu than just ccsd(t)  (the
same applies for your closed-shell complex He-Na+).  Also in the
next dummy statement you give "dummy,Li" where it seems like it should be "dummy,Na" in this calculation.  Of course if
this is just He atom, there are no triples since there are only 2 electrons and just "ccsd" would suffice (but specifying ccsd(t)
will not result in an error).

regards,

Kirk


On Feb 1, 2010, at 4:20 AM, Yang, Shengfu F. (Dr.) wrote:

> Dear All,
> 
> I have got a problem with the counterpoise correction for ion-molecular complex. For example, part of the input file is: (Na+He)
> 
> {rhf; wf, 12,1, 0);
> rccsd(t);
> E1 = energy;
> 
> dummy, He;
> {rhf;wf,10,1,0}
> rccsd(t);
> E2 = energy;
> 
> dummy, Li
> hf;
> ccsd(t);
> E3 = energy;
> 
> dummy;
> ....
> 
> The error is always something like "there is no obital ...." for the counterpoise correction of Li+. Can I ask what the solution is for this problem?
> 
> Thanks a lot for your help!
> 
> Sincerely Yours,
> 
> Dr. Shengfu YANG
> -------------------------
> Department of Chemistry
> University of Leicester
> Leicester LE1 7RH
> UK
> Tel: +44-116 252 2127
> Fax: +44-116 252 3789
> -------------------------
> _______________________________________________
> Molpro-user mailing list
> Molpro-user at molpro.net
> http://www.molpro.net/mailman/listinfo/molpro-user