Will BOD/NAdOs be available for non-singlet multi-determinant WFs?

Aerael · 2021-09-01 12:57:28

Dear Prof. Tian Lu,

I have used the Bond Order Density (BOD) / Natural Adaptive Orbitals (NAdO) methods. [200,20]
For single determinant methods, everything is fine and well.
Unfortunately, your implementation does not seem to support multi determinant wave functions.
Will there be a way to construct meaningful BOD/NAdOs from e.g. CASSCF wave functions?

AOM creation seems to work well (with valid AOM results), even for non-singlet multiplicities, so I suppose this is not the issue?

Also, the respective error message given if you try to make NAdOs of non-singlet multideterminant WFs
"Error: Only restricted and unresticted single-determinant wavefunction is not supported!"
should probably read
"Error: Only restricted and unresticted single-determinant wavefunction are supported!"

Thanks so much in advance for your answer and thank you for all the good work!
Daniel

sobereva · 2021-09-01 14:01:01

Dear Daniel,

The message is indeed wrong, it should be "Error: Only restricted and unresticted single-determinant wavefunction are supported!", thank you for pointing out this problem, I will fix it.

BOD/NAdO can also be used for non-singlet multiplicities in Multiwfn, such as triplet.

Currently the formalism of BOD/NAdO for multiconfigurational wavefunctions is not availble, so BOD/NAdO doesn't work for CASSCF wavefunction currently.

Best regards,

Tian

Aerael · 2022-09-12 14:34:25

Dear Tian,

I wanted to kindly ask again if there might be an opportunity to expand the NAdO creation in MultiWFN to multi determinant wave functions.
NAdO author Pendàs published another paper this year which mentions cumulative densities and their relation to NAdOs but does not go into detail either:
https://doi.org/10.1002/jcc.23861

Thanks so much for your consideration!
Daniel

sobereva · 2022-09-12 14:38:35

Dear Daniel,

Thank your for bringing this paper into my attention. Currently I don't have a clear plan to include this into Multiwfn, because DFT wavefunction is adequate for most cases, and there are much more important functions to implement into Multiwfn.

Best regards,

Tian

Multiwfn forum

#1 2021-09-01 12:57:28

Will BOD/NAdOs be available for non-singlet multi-determinant WFs?

#2 2021-09-01 14:01:01

Re: Will BOD/NAdOs be available for non-singlet multi-determinant WFs?

#3 2022-09-12 14:34:25

Re: Will BOD/NAdOs be available for non-singlet multi-determinant WFs?

#4 2022-09-12 14:38:35

Re: Will BOD/NAdOs be available for non-singlet multi-determinant WFs?

Board footer