[molpro-user] Calculating molecular oxygen excited state

[Lorenzo Lodi]

Jacky LIEVIN wrote:
> Dear Lorenzo,
> 
> Please find below a modified version of your input. The results are  
> looking fine:

[cut]

Thanks a lot Jacky, it did the trick and it's now working fine!
I spent some time studying your sample input (I'm new to state-averaged 
calculations) and there a few things I'm not too sure of.
For the excited-state state-averaged CASSCF your input reads:
{MCSCF;
   WF, NELEC=16, SYMMETRY=1, SPIN=0;state,2;lquant,2,0
   WF, NELEC=16, SYMMETRY=1, SPIN=0;state,1;lquant,2}

The second line specifies the two lowest-states with "1" (=Ag) symmetry, 
namely the <sigma g+> and <delta g xx-yy>.
If the third line is to specify the degenerate <delta g xy> state with 
symmetry "4" (=B1g), shouldn't we specify SYMMETRY=4 ?

Also, I have a slightly different question. Is it advisable/necessary to 
in the excited-state RHF step to have it to converge to the desired 
excited state too? Couldn't we use the ground state as RHF reference for 
the subsequent steps?

Thanks again,

Lorenzo

> Modifications are:
> 1) the use of state averaged MCSCF
> First for ensuring the degeneracy of the delta state (one component of  
> symmetry 1 and one of symmetry 4): group theory tells us that deltag  
> correlates to Ag + B1g in D2h symmetry
> Second for describing the second state (sigma + g) of symmetry 1:  
> sigma + g correlates to Ag
> Both electronic states must thus be calculated in a single MCSCF state  
> averaged calculation defined by 2 WF directives.
> Note the lquant directive given on each WF line and for each  
> calculated state
> Note the degeneracy CASSCF(1)=CASSCF(2)= energy of delta state