Molcas Forum

XiaolinLiu

Member

Registered: 2022-11-15

Posts: 1

Calculating spin-orbit coupling matrix elements with solvent effects

Hello, I am working on calculating SOC matrix element between S0 and T1 states to estimate the relative order of decay rate of T1 states of several molecules in solution. The singlet and triplet RASSCF calculations are carried out using optimized T1 geometry and are as follows.
Singlet RASSCF:

 &GATEWAY
Title
CuAnQ
RICD
COORD
 105
Angstrom
 Cu                -2.82347600   -0.14317300   -0.02750200
...
Basis set
ANO-RCC-MB, Cu.ANO-RCC-VDZP
RF-input
PCM-model
solvent
acetonitrile
R-MIn
1.0
SPHEre radius
1   1.7475
...
CONDuctor version
end of rf-input
End of input

&SEWARD &END
>> COPY $Project.ScfOrb INPORB
&RASSCF
LUMORB
Spin
1
Inactive
230
Nactel
2 0 0
Ras2
7
CiRoot
28 28 1
RFRoot=1

The triplet RASSCF input is mostly the same, I'm only showing the difference here:

&RASSCF
LUMORB
Spin
3
Inactive
230
Nactel
2 0 0
Ras2
7
CiRoot
21 21 1
RFRoot=1

The next step is using RASSI. This is what I'm not sure about. Initially, I had:

>> COPY CuAnQ_MeCN_T1_pcm_scf_singlets_casscf_nroot_28.JobIph JOB001
>> COPY CuAnQ_MeCN_T1_pcm_scf_triplets_casscf_nroot_21.JobIph JOB002
&RASSI
Nr of JobIph
 2 28 21
 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Spin
MEES
MESO
SOCO
0.01
HEFF
EJOB
RFpert

then I realized it was unclear which state the reaction field calculated is based on. Does the input indicate the reaction field in JOB001 will be used for RASSI? If so,
I have to change the order. Please comment on the inputs and suggest best practices for doing such calculations since I haven't found one in the forum. Thanks!