[ Molcas user's WWWBoard ]

Posted by Steven on August 16, 2014 at 20:37:27:

In Reply to: RASSI posted by ABID on July 27, 2014 at 17:33:30:

: Dear All
: I am performing some RASSI calculations, I have a very strange problem with my project output.
: When I use the keyword LUMOrb in my calculations it gives a strange error of convergence, However if key word is removed then it gives very strange ground state configuration having excitation(Root 1 contains excitation) I found that if we use LUMOrb and calculations goes to end then root 1 is perfect.

I don't see how you can be using the keyword LUMorb for rassi? Do you mean you use if for rasscf? LUMorb only has an effect on reading starting orbitals in rasscf.

Convergence in rasscf will certainly depend on the starting orbitals, and can be a difficult problem if you are not used to dealing with selection of active orbitals.

It seems you are using a rather large active space, did you check that the orbitals in the active space are correct? The error given by rassi points to a possible problem with wavefunctions having (too) different active orbitals.

: Last part of my output when we use LUMORb is pasted here.

Please don't dump large parts of code.

Follow Ups:

Post a Followup



if B is 1s22s22p1, what is Li?



[ Follow Ups ] [ Post Followup ] [ Molcas user's WWWBoard ]