[molpro-user] Error in CASSCF optimization

[Werner Győrffy]

Dear Thomas,

This error can be avoided by using closed-shell HF orbitals as initial 
orbital guess. You should run HF before the CASSCF calculation.

Regards,

Werner.

On 06/25/2014 01:30 PM, Thomas Vent-Schmidt wrote:
> I want to do an open shell CASSCF optimization with frozen core, but
> I always get this error. Anyone an idea how to solve the problem?
>
> Thanks Thomas
>
> PROGRAM * MULTI (Direct Multiconfiguration SCF)       Authors: P.J.
> Knowles, H.-J. Werner (1984)     S.T. Elbert (1988)
>
>
> Number of frozen core orbitals:  13 (  7  3  2  1 ) Number of
> closed-shell orbitals: 15 (  6  5  3  1 ) Number of active  orbitals:
> 12 (  4  3  3  2 ) Number of external orbitals:    159 ( 56 45 34 24
> )
>
> State symmetry 1
>
> Number of electrons:     9    Spin symmetry=Sextet    Space
> symmetry=2 Number of states:        1 Number of CSFs:      11583
> (13068 determinants, 52272 intermediate states)
>
> MCSCF GRADIENTS ONLY POSSIBLE WITH CLOSED-SHELL SCF FROZEN CORE
> ORBITALS ? Error ? FROZEN CORE ORBITALS NOT FOUND ? The problem
> occurs in FIND-ORB
>
> ERROR EXIT CURRENT STACK:      MAIN
>
> _______________________________________________ Molpro-user mailing
> list Molpro-user at molpro.net
> http://www.molpro.net/mailman/listinfo/molpro-user
>