[molpro-user] pi state in c8v symetry

[liuxj at iccas.ac.cn]

I am sorry I didn't express my problem clearly in last letter
1. I want calculate a linear molecule at pi state. I gave c2v symmetry then
used lquant card. But I didn't know how to give the wf card(i.e. wf,8,?,0;
lquant,1 ).
2.  And I calculated a sigma state for a linear molecule.
when I give different occ card(i.e. a different active space), it converged
to different energy. how can I know which active space is better. multi
card is as follows:
multi;occ,5,1,1;wf,8,1,0;lquant,0;
multi;occ,6,,,1;wf,8,1,0;lquant,0; 

 ----------------------------------------------- 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 


 

E.P.Lee at soton.ac.uk 写: 

dear all
I wanna caculate pi state in c8v symmetry.
I give c2v symmetry then use lquant card, then which symmetry should I give
in wf card?
wf,8,?,0; lquant,1

	lquant is only used for a linear molecule. lquant,0 is for
a Sigma state; lquant,1 is for Pi state; etc. I don't think lquant can
be used for a molecule with a C8v symmetry.

And I puzzled with another result,which is for sigma state.
when I give different occ card, it converged to different energy. how can I
know which is the exact convergence. multi card is as follows:
multi;occ,5,1,1;wf,8,1,0;lquant,0;
multi;occ,6,,,1;wf,8,1,0;lquant,0;

	The absolute value of the computed total energy depends on the
size of the CI (casscf is full CI within the active space). It is
obvious that with different occ (i.e. a different active space), you
will get a different energy. It is just like using a different basis set
in a HF calculation, which will obviously give a different HF energy. 

	I am not sure, what you mean by exact convergence. If you mean
energy convergence w.r.t the active space used in a casscf, probably it
will never converged. ... . Really you need a very large active space
and the limit is a full CI. ... .