[molpro-user] Spin-orbit coupling calculation problem in CI program - Number of Active orbitals

Tomáš Šolomek tom.solomek at gmail.com
Thu Oct 11 15:03:28 BST 2012


Dear Molpro users,

I've been trying to calculate the spin-orbit couplings for a series of
molecules, which span up to 100 electrons. When I run my calculations I can
find the following error printed once the CI program is entered.

 TOO MANY ACTIVE ORBITALS: 35 THIS VERSION ALLOWS 32

I somehow sorted the problem out and I verified my reasoning from couple of
posts to the molpro-user mailing list. The obvious solution would be
increasing the number of frozen-core orbitals which I apparantly did but
with no success. Is there any specific reason why this didn't work in this
particular case or did I miss something obvious?

This is my input together with the "core" directive.

 {  multi;
   occ, 47; closed,37;
   rotate,31.1,40.1,0;
   rotate,35.1,41.1,0;
   rotate,47.1,52.1,0;
   wf,84,1,0;state,1;weight,1;
   wf,84,1,2;state,1;weight,1;}

  {ci;wf,84,1,0;occ,47,closed,37,core,15;save,3010.1,state,1;noexc}
  {ci;wf,84,1,2;occ,47,closed,37,core,15;save,3040.1;state,1;noexc}

This particular molecule has 12 heavy atoms. I need to use (10,10) active
space for CASSCF wavefunction. I expected the "core" directive to cut
orbitals 3 more from the active orbitals in CI program to get to the
maximal number of allowed orbitals (32; 47-15). However, it seems that the
program still counts only the inner-shell orbitals of heavy atoms, i.e. 12
(47-12). When I increased the number of frozen-core orbitals to 18, nothing
has changed and the job crashed. What might be wrong?

Thank you for any help with this. Cheers


tomas
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.molpro.net/pipermail/molpro-user/attachments/20121011/df3759fb/attachment.html>


More information about the Molpro-user mailing list