Molcas Forum

fullmetalchem15t

Member

Registered: 2018-05-14

Posts: 4

C-PCM average state RASSCF for symmetry 2

I'm trying to get RASSCF average state energies for a C-PCM solvated PdCl4^2- complex. The states are divided into two symmetries. According to the manual, in order to get excited state energies that correspond to electronic transitions, generate solvation charges that are in equalibrium with ground state, then do a non-equilibrium calculation for the excited states. My second symmetry is having difficulty optimizing the orbitals under the method described (the ground state is in the first symmetry). When it does converge, the degeneracy of my states is destroyed. The type of calculations I'm trying to do, preserving degeneracy is extremely important. The only way I get the orbitals to optimize is by generating the solvation charges in equilibrium to an excited state of symmetry 2 (with the RFROOT keyword) but I do not trust the energies I get out of this method. I tried using C1 symmetry which worked but I had the same issue when I tried to get triplet states.  I have none of these orbital optimization issues for this system without solvation. I feel like there is something critical I'm missing. My input for my SEWARD and the RASSCF for symmetry 1 and 2 are below.

&SEWARD  &END
HIGH cholesky
symmetry
xyz
Basis set
Pd.ANO-RCC-VTZP
Pd         0.000000000      0.000000000      0.000000000   angstrom
End of basis
Basis set
Cl.ANO-RCC-VTZP
Cl1        0.000000    2.341369    0.000000   angstrom
Cl2        2.341369    0.000000    0.000000   angstrom
End of basis
Angmom
0.0 0.0 0.0
AMFI
RF-Input
PCM-Model
solvent
water
CONDuctor version
End of RF input
End of input

&RASSCF &END
Spin
1
Symmetry
1
nActEl
32 2 0
TypeIndex
CIRoots
87 87 1
Iter
100 100
LUMORB
levshift
1.0
orbappear
compact
End of input

>> COPY $Project.RasOrb  $HomeDir/$Project-sym1.RasOrb
>> COPY $Project.JobIph  $HomeDir/$Project-sym1.JobIph
>> COPY $Project.RunFile $HomeDir/$Project-sym1.RunFile

>> COPY $HomeDir/$Project-sym1.RunFile RUNFILE
>> COPY $HomeDir/$Project-sym1.RasOrb INPORB

&RASSCF &END
Spin
1
Symmetry
2
nActEl
32 2 0
TypeIndex
CIRoots
66 66 1
Iter
100 100
LUMORB
levshift
1.0
orbappear
compact
MAXORb
2
NONEquilibrium
End of input

Ignacio

Administrator

From: Uppsala

Registered: 2015-11-03

Posts: 1,085

Re: C-PCM average state RASSCF for symmetry 2

Wild guess: could it be that solvation changes the order of states such that the CIRoot should be changed in order to preserve degeneracies?

You could also try the SupSym keyword to enforce additional symmetries in the orbitals.

fullmetalchem15t

Member

Registered: 2018-05-14

Posts: 4

Re: C-PCM average state RASSCF for symmetry 2

Thank you for your input. The issue can't be the ordering of roots because for symmetry 2, I'm requesting the maximum number of roots possible for this symmetry. My guess, which may be incorrect is that Molcas doesn't know how to handle the generated charges from the ground state for the states in symmetry 2 since the ground state is not in symmetry 2. The think this because if I use C1 symmetry, the orbitals converged but not for the triplet states. Note that I'm having the same issue with the triplet states regardless of whether I'm using C1 or Ci symmetry. The only way I can't get the orbitals to converge for symmetry 2 is to generate charges for a non-degenerate in symmetry 2.