Re: RASSCF CI convergence


[ Molcas user's WWWBoard ]

Posted by Wenli Zou on February 24, 2005 at 16:05:57:

In Reply to: RASSCF CI convergence posted by John Nash on June 21, 2004 at 22:04:39:

Hi John,
In order to get a normally converged RASSCF result, You can
increase the number in CIMX option (the maximum is 200);
increase or reduce the convergence thresholds in THRS option suitably;
change the states in multi-state RASSCF computation;
or,
change your start orbitals. The start orbitals can be previous converged HF orbitals using another occupations, RASSCF orbitals with different spin multiplicity, irreducible representation, or active space, or HF/RASSCF orbitals computed previously at a nearby bond distance.



Follow Ups:



Post a Followup

Name:
E-Mail:

Subject:

if B is 1s22s22p1, what is Li?

Passfield:

Comments:


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