Different results of wavefunction overlap

[Li Yuan]

Dear molpro experts,

I calculated the wavefunction overlap between states belong to two different  geometries by changing one bond length as 0.2 angstrom. For the first test, I used CASSCF wavefunction. I run the same calculations two times, but the results of wavefunction overlap are quite different. Could please anyone help me understand this ? I attach my output files and the overlap results for your review. Thank you very much!

Best regards,

Yuan

[Alexander MITRUSHCHENKOV]

Could be a bug, I’ll look at this.

Best

Alexander

------------------------------------------------------------

Dr. Alexander Mitrushchenkov, IGR
Laboratoire de Modélisation et Simulation Multi Echelle
UMR 8208 CNRS
Univ Gustave Eiffel
5 Bd Descartes
77454 Marne la Vallée, Cedex 2, France

Phone:    +33(0)160957316
Fax:      +33(0)160957320
e-mail:   Alexander.Mitrushchenkov AT univ-eiffel.fr

--
You received this message because you are subscribed to the Google Groups "molpro-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email to molpro-user...@googlegroups.com.
To view this discussion, visit https://groups.google.com/d/msgid/molpro-user/bad2cc24-7ff9-4275-81c3-6315afa9dbccn%40googlegroups.com.
<overlap.zip>

[Li Yuan]

Hi Alexander, 

Thanks for your response. I have many calculations relying on the wavefunction overlap. If you find a way to fix this, please let me know. I have the access to source code, thus I can modify the code from my side. Thank you!

Best regards,

Yuan

[Alexander MITRUSHCHENKOV]

You need to use BIORTH option to get transition properties with different orbitals, displaced geometries being a particular case. Please see

for more details. Without BIORTH options results are meaningless.

Best regards

Alexander

------------------------------------------------------------

Dr. Alexander Mitrushchenkov, IGR
Laboratoire de Modélisation et Simulation Multi Echelle
UMR 8208 CNRS
Univ Gustave Eiffel
5 Bd Descartes
77454 Marne la Vallée, Cedex 2, France

Phone:    +33(0)160957316
Fax:      +33(0)160957320
e-mail:   Alexander.Mitrushchenkov AT univ-eiffel.fr

Le 27 sept. 2025 à 02:38, Li Yuan <yli2...@gmail.com> a écrit :

[Li Yuan]

Hi Alexander, 

Thanks for your help.  I have followed your suggestion and put input like this:

{ci;trans, 5160.2, 5260.2, biorth,ov;        !Compute overlap and transition density <R|R+DR>
dm,8100.2}
{ci;trans, 5161.2, 5261.2, biorth,ov;        !Compute overlap and transition density <R|R+DR>
dm,8200.2}

Previously, I followed the manual of molpro, the input example for LiF:

https://www.molpro.net/manual/doku.php?id=non_adiabatic_coupling_matrix_elements

After your suggestions, the result of overlap is reasonable now. I tried bond length at 2.40 and and 4.00 Angstrom. The diagonal overlap matrix elements are very close to 1. But I found the CI coefficients of same state at different geometries changes. Would you please review my input file ? I think this is the typical input for my case, but not sure if I get it correctly. I appreciate your help.

Thank you,

Yuan

[Alexander MITRUSHCHENKOV]

Just one small note:

If you will use BIORTH option AND need not only overlaps but some other properties (dipole/velocity/H etc — has not much sense for displaced geometries but just in case), AND use full MRCI (i.e. not CAS-like) wavefunctions, you have to add

bort_debug(17)=1

just before calling ci;trans. This is also valid for the same geometry but different orbitals. So, if properties other than overlaps are required for full MRCI, please check that the line

'Version with bi-normalized virtual orbitals’ is present in the output

This has been fixed in curent development version.

Best wishes

Alexander

------------------------------------------------------------

Dr. Alexander Mitrushchenkov, IGR
Laboratoire de Modélisation et Simulation Multi Echelle
UMR 8208 CNRS
Univ Gustave Eiffel
5 Bd Descartes
77454 Marne la Vallée, Cedex 2, France

Phone:    +33(0)160957316
Fax:      +33(0)160957320
e-mail:   Alexander.Mitrushchenkov AT univ-eiffel.fr

To view this discussion, visit https://groups.google.com/d/msgid/molpro-user/09b7055d-fdb6-4724-babf-62defc72c884n%40googlegroups.com.

[Alexander MITRUSHCHENKOV]

Correction to my previous mail, it should be

bort_debug(16)=1

If you have access to the source code, you can find all bi-orth print/debug flags in src/mrci/citra.F90

Best

Alexander

------------------------------------------------------------

Dr. Alexander Mitrushchenkov, IGR
Laboratoire de Modélisation et Simulation Multi Echelle
UMR 8208 CNRS
Univ Gustave Eiffel
5 Bd Descartes
77454 Marne la Vallée, Cedex 2, France

Phone:    +33(0)160957316
Fax:      +33(0)160957320
e-mail:   Alexander.Mitrushchenkov AT univ-eiffel.fr

To view this discussion, visit https://groups.google.com/d/msgid/molpro-user/09b7055d-fdb6-4724-babf-62defc72c884n%40googlegroups.com.

[Alexander MITRUSHCHENKOV]

Hi,

First, you do not need to add ‘OV’ after BIORTH, overlaps are calculated anyway.

As to CI coefficients, these clearly depend on the orbitals. For CASSCF, you may freely rotate core, closed and active orbitals. The CI coefficients will follow. To have as much as similar orbitals, you can try DIAB orbitals option of CASSCF. Still, for such big geometry steps, I am not sure the orbitals and/or CI coefficients with resemble and/or if their difference can have any physical meaning.

Best

Alexander

------------------------------------------------------------

Dr. Alexander Mitrushchenkov, IGR
Laboratoire de Modélisation et Simulation Multi Echelle
UMR 8208 CNRS
Univ Gustave Eiffel
5 Bd Descartes
77454 Marne la Vallée, Cedex 2, France

Phone:    +33(0)160957316
Fax:      +33(0)160957320
e-mail:   Alexander.Mitrushchenkov AT univ-eiffel.fr

To view this discussion, visit https://groups.google.com/d/msgid/molpro-user/73c89c45-89a2-43f8-a276-3550c9f7a05cn%40googlegroups.com.
<mrci_ov.out>

[Li Yuan]

Hi Alexander,

Thank you very much for all these guidances. I need more time to digest these. I really appreciate.

Best regards,

Yuan