EXPBAS desymmetryzation

Subject:

if B is 1s22s22p1, what is Li?

Passfield:

Comments:
: Hello, : I have troubles converging CASSCF without symmetry (oscillatory behavior with no help from level shifting or turning off the quasi newton), while the computation with symmetry on converges smoothly. : I would like to use the orbitals from symmetric computation as a start for the one with no symmetry. : How would I do that properly? : I tried this: : 1) run gateway/seward -> scf -> casscf with symmetry on : 2) copy rasorb to inporb and runfile to runfil1 : 3) execute expbas with the noexpand and dessymetrize keywords : 4) copy the desymorb to inporb : 5) run gateway/seward with symmetry off : 6) run casscf with the lumorb command : No matter how I modify the input, the energy of the second casscf differs from the first one, even though the active space size, spin and root are equivalent. : I would understand the second casscf converging to the different energy, but it differs from the first iteration, even if I use the cionly keyword. : Does it have something to do with ordering of atoms in gateway/seward or orbitals in casscf? : Thank you for any help. : Lukas