Support and discussions for Molcas and OpenMolcas users and developers
You are not logged in.
Please note: The forum's URL has changed. The new URL is: https://molcasforum.univie.ac.at. Please update your bookmarks!
You can choose an avatar and change the default style by going to "Profile" → "Personality" or "Display".Hi,
I want to obtain the zero-field splitting parameters of a Mn(II) compound at a CASPT2 level.
I have run the RASSCF for the ground S=6 root and the 24 roots of the S=4 manifold and subsequently CASPT2 for S=4 is run for the lowest 17 roots only. Then I run RASSI pointing to the corresponding two IPH files - the energies for the S=4 roots are the CASPT2 ones, while that for the S=6 root is that of the CASSCF, which is not what I want. The input is
&RASSCF
FileOrb= 1.RasOrb
Spin= 6
Charge= 0
CiRoot= 1 1 1
ORBA= FULL
MAXORB= 1
Nactel= 5 0 0
RAS2= 5
>> COPY $Project.JobIph 1_IPH
>> COPY $Project.RasOrb 1.RasOrb
>> COPY $Project.rasscf.h5 1.rasscf.h5
&CASPT2 &END
Title
caspt2 for the ground state high-spin sextet
Frozen
50
End of input
>> COPY $Project.JobIph 1PT2_IPH
&RASSCF
FileOrb= 2.RasOrb
Spin= 4
Charge= 0
CiRoot= 24 24 1
ORBA= FULL
MAXORB= 24
Nactel= 5 0 0
RAS2= 5
>> COPY $Project.JobIph 2_IPH
>> COPY $Project.RasOrb 2.RasOrb
>> COPY $Project.rasscf.h5 2.rasscf.h5
&CASPT2 &END
Title
mscaspt2 for the 17 lowest-lying roots of the spin quartet multiplet
Frozen
50
MultiState
17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
End of input
>> COPY $Project.JobIph 2PT2_IPH
&RASSI
SPIN
MEES
HEFF
EPRG= 7.0D-1
Nr of JobIph= 2 1 17
1
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
IPHN
1PT2_IPH
2PT2_IPH
PROP
3
'ANGMOM' 1
'ANGMOM' 2
'ANGMOM' 3
&SINGLE_ANISO
MLTP= 1; 6
QUAX= 1
I also tried passing the CASPT2 energies via the HDIAG keyword, but it did not work either.
Does anyone know how to fix this?
Thanks
Offline
Try passing the JOBMIX files instead of JOBIPH, and add EJOB to RASSI
Offline
I did not save the JOBMIX files... I need to repeat one of the CASPT2 sections then. I guess there is no other option?
Offline
If you have the $Project.caspt2.h5 files, try with them (but without EJOB in that case).
Offline
No, I don't... The job should be done by tomorrow and I'll be able to check, thanks
Offline
Hi. It doesnt seem to have worked. This is a summary of the energies
S=6 root | RASSCF root 1 -2847.96595377 | CASPT2 Root 1 -2852.39956228 | RASSI State 1 0.00000000
S=4 roots | RASSCF root 1 -2847.83604544 | CASPT2 Root 1 -2852.28861695 | RASSI State 2 -2852.29240452
" | RASSCF root 2 -2847.83543096 | CASPT2 Root 2 -2852.28922489 | RASSI State 3 -2852.29233670
" | RASSCF root 3 -2847.83507952 | CASPT2 Root 3 -2852.29190332 | RASSI State 4 -2852.29089348
" | RASSCF root 4 -2847.83397319 | CASPT2 Root 4 -2852.29055854 | RASSI State 5 -2852.29003402
" | RASSCF root 5 -2847.83212023 | CASPT2 Root 5 -2852.29188780 | RASSI State 6 -2852.28934670
" | RASSCF root 6 -2847.82931146 | CASPT2 Root 6 -2852.28899352 | RASSI State 7 -2852.28928852
" | RASSCF root 7 -2847.82912887 | CASPT2 Root 7 -2852.28874381 | RASSI State 8 -2852.28838489
" | RASSCF root 8 -2847.82698481 | CASPT2 Root 8 -2852.28785550 | RASSI State 9 -2852.28798435
" | RASSCF root 9 -2847.82691971 | CASPT2 Root 9 -2852.28765373 | RASSI State 10 -2852.28760814
" | RASSCF root 10 -2847.81264203 | CASPT2 Root 10 -2852.27661701 | RASSI State 11 -2852.27640539
" | RASSCF root 11 -2847.81148670 | CASPT2 Root 11 -2852.27474792 | RASSI State 12 -2852.27438828
" | RASSCF root 12 -2847.80640317 | CASPT2 Root 12 -2852.27066116 | RASSI State 13 -2852.27061674
" | RASSCF root 13 -2847.80445826 | CASPT2 Root 13 -2852.26905564 | RASSI State 14 -2852.26862894
" | RASSCF root 14 -2847.80404207 | CASPT2 Root 14 -2852.26776097 | RASSI State 15 -2852.26719099
" | RASSCF root 15 -2847.79546817 | CASPT2 Root 15 -2852.26409016 | RASSI State 16 -2852.26390068
" | RASSCF root 16 -2847.78633866 | CASPT2 Root 16 -2852.25608261 | RASSI State 17 -2852.25550864
" | RASSCF root 17 -2847.78487349 | CASPT2 Root 17 -2852.25439518 | RASSI State 18 -2852.25392774
" | RASSCF root 18 -2847.73546293
" | RASSCF root 19 -2847.73511225
" | RASSCF root 20 -2847.73363316
" | RASSCF root 21 -2847.73288096
" | RASSCF root 22 -2847.73225356
" | RASSCF root 23 -2847.73074586
" | RASSCF root 24 -2847.73036880
and this is the input for RASSI I used
&RASSI
SPIN
MEES
* When reading JOBMIX files, use EJob. HEFF prevents recalculation of spin-free Hamiltonian
EJob
HEFF
EPRG= 7.0D-1
Nr of JobIph= 2 1 17
1
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
IPHN
1PT2_MIX
2PT2_MIX
PROP
3
'ANGMOM' 1
'ANGMOM' 2
'ANGMOM' 3
where 1PT2_MIX and 2PT2_MIX have been saved with >> COPY $Project.JobMix 1PT2_MIX and >> COPY $Project.JobMix 2PT2_MIX after the corresponding CASPT2 section.
Last edited by danielreta (2023-01-31 10:40:20)
Offline
I'd suggest you try a smaller/simpler test case and add the complexity later. This seems to work for me:
&GATEWAY
Coord = 1
Mn 0.0 0.0 0.0
Group = NoSymm
Basis = ANO-RCC-VDZP
RICD
&SEWARD
&RASSCF
Spin = 6
Charge = 0
NactEl = 5
RAS2 = 5
CIRoot = 1 1 1
> COPY $Project.JobIph 1_IPH
&CASPT2
> COPY $Project.JobMix 1PT2_MIX
&RASSCF
FileOrb = $Project.rasscf.h5
Spin = 4
Charge = 0
NactEl = 5
RAS2 = 5
CIRoot = 12 12 1
> COPY $Project.JobIph 2_IPH
&CASPT2
Multistate = 9
1 2 3 4 5 6 7 8 9
> COPY $Project.JobMix 2PT2_MIX
&RASSI
NrOfJob = 2 all
IphNames
1PT2_MIX
2PT2_MIX
EJob
SpinOrbit
Offline
Thanks. I have tried your example. Now, the RASSI energy for the S=6 state is the CASPT2 one, but those for the S=4 are neither the RASSCF nor the CASPT2 ones - if you take the energy differences between the CASPT2 roots and the RASSI ones there is variation
Offline
They are the MS-CASPT2 energies, and the corresponding H matrix is diagonal (because the CASSCF states have already been mixed). That should be equivalent to reading the full SS-CASPT2 H matrix.
Offline
oh! My bad! Ok, that makes sense then.
So in summary, if one reads from the JobMix and uses EJOB, it works. Thanks
Offline