CASPT2 BVAT energy contributions nonzero for 1-electron active space

[ Molcas user's WWWBoard ]

Posted by Jose Luis Pascual on October 02, 2008 at 09:59:52:

Hello all

I'm trying to make MS-CASPT2 calculations using CASSCF wavefunctions as starting point including just one electron in the active space. SS-CASPT2 for root 1 goes OK, but SS-CASPT2 for the second root does not converge.
The only thing I find strange is that BVAT correlation energy contributions are not zero, but I think they should be exactly zero as they are double excitations from the active space. As the active space comprises just one electron, they should be zero.

Am I wrong in this point?? Is this a bug??

Help will be very welcome. Thanks in advance

JLP

Follow Ups:


Post a Followup

Name:
E-Mail:

Subject:

if B is 1s22s22p1, what is Li?

Passfield:

Comments:

[ Follow Ups ] [ Post Followup ] [ Molcas user's WWWBoard ]