cipt2 calculation of a excited state

[Benoît BRAÏDA]

Le jeudi, 27 mar 2003, à 09:01 Europe/Paris, Kalju Kahn a écrit :

> Hello,
>
> I am fairly new to molpro and I am not sure if the problem I am having
> with cipt2 calculation of the exited state is a result of my limited
> understanding of the method or a program bug.  I am running a cipt2
> calculation on the third exited state of benzonitrile.  The molecule
> has C2v symmetry and the exited state belongs to A1 irreducible
> representation.  The A1 ground state calculation is successful and I
> can do RS2 calculations in the exited state after using level shifts
> around 0.15.  The relevant portion of my cipt2 input looks like this:
> --------------------------------
> basis,roos(3s2p1d/2s)
>
> hf
>
> multi
> occ,15,6,8,2
> core,6,0,2,0
> closed,15,0,8,0
> wf,54,1,0
>
> cipt2
> occ,15,6,8,2
> core,6,0,2,0
> closed,15,0,8,0
> wf,54,1,0
> maxiter,80,100
> state,1,3
> refstate,1,3
>

I suppose that what you've experienced was a variational collapse of 
the MRPT2 wf of your 3rd exited state during your PT2 calculation.

Try to put :

state,3

instead of :

state,1,3
refstate,1,3

(the three first MRPT2 roots should be calculated at the same time, 
taking the three first MCSCF roots as refs.)


Best regards,

Benoît


••••••••••••••••••••••••••••••
Dr Benoît BRAIDA
ICMAB-CSIC
Campus de la UAB
08193 Bellaterra
Barçelona
Espagne (Spain)
Phn : +34 / 93 580 18 53 ext : 319
Fax : +34 / 93 580 57 29
Mail : braida at icmab.es
••••••••••••••••••••••••••••••