State assignation in state-average CASSCF

8 views
Skip to first unread message

Giorgio Visentin

unread,
Feb 9, 2026, 8:18:10 AM (3 days ago) Feb 9
to molpro-user
Dear Molpro users and developers,
I am computing the potential energy surface associated with the (1b2)^-2 ^2A1 state of the doubly charged ion of water. I used the CASSCF method to compute the surface and averaged the calculation over four different states of the same symmetry as the state I am willing to model, in order to avoid the optimisation procedure to flip within quasi-degenerate states. Thus, I get four surfaces, as shown in the attached test output file. 
My question is the following: how may I identify each state from the output file? Thus far, when using the state-averaged CASSCF level, I had no  trouble in the state identification, as the density of states was not so high as in this case. 

Thanks for your attention. 
Best wishes,
Giorgio 
H2O_MRCI_En_Exc_AS5.out

peterso...@gmail.com

unread,
Feb 9, 2026, 9:39:44 AM (3 days ago) Feb 9
to Giorgio Visentin, molpro-user
Dear Giorgio,

a couple of things:

  1. Molpro doesn’t support multidimensional arrays such as gs(i,j,k).  You need to set a linear array counter that runs over all i, j, k (initialize it outside of the loop over i, and increment it in the innermost loop).
  2. the only way to identify the states is to compare the character of the CI eigenvectors at each geometry. The states seem pretty clean and are well separated, so this doesn’t look like it should be too difficult.

regards,

-Kirk
--
You received this message because you are subscribed to the Google Groups "molpro-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email to molpro-user...@googlegroups.com.
To view this discussion, visit https://groups.google.com/d/msgid/molpro-user/ca87c7f1-729a-466a-a5dc-8b317634d78dn%40googlegroups.com.
Reply all
Reply to author
Forward
0 new messages