Author Topic: Create wfn file with natural orbitals for Multiwfn  (Read 3730 times)

_mtkt_

  • Newbie
  • *
  • Posts: 2
  • Karma: +0/-0
Create wfn file with natural orbitals for Multiwfn
« on: May 19, 2021, 10:21:48 AM »
Hello turbomole users,

I want to fit Tresp-charges with the program Multiwfn and therefor I need a wfn file. In the manual of Multiwfn (p.837) they explain how to do this with Gaussian (keywords PBE1PBE/6-31g(d) TD density=transition=2 out=wfn), which creates natural orbitals which are saved in a wfn file.
My problem is that I cannot create such an wfn with turbomole. I have already tried proper and $wfn but the wfn is not the same as in Multiwfn.
Is there any possibility to create this wfn file?

Best regards,
_mtkt_

uwe

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 558
  • Karma: +0/-0
Re: Create wfn file with natural orbitals for Multiwfn
« Reply #1 on: May 20, 2021, 02:46:40 PM »
Hi,

you need the excited states density (or the difference density between excited state and ground state density) exported to a wfn file?

Add $wfn to the control file and run egrad - egrad is calculating the excited state density (as it is needed for the TDDFT gradients) and egrad will use the excited state density when generating the wfn file as requested by the $wfn keyword.

Open the resulting wfn file. If the first line says:

SCF/DFT density

then the data is for the ground state density. If the line is:

TDSCF/TDDFT density

the data is for the TDDFT density of the excited state you have asked egrad to calculate the gradients for ($exopt).

That's at least what egrad in the current version (7.5) is doing, but it should also work with older versions (not sure how far this feature goes back, actually)...



_mtkt_

  • Newbie
  • *
  • Posts: 2
  • Karma: +0/-0
Re: Create wfn file with natural orbitals for Multiwfn
« Reply #2 on: May 21, 2021, 12:48:34 PM »
Thanks for your answer and your suggestion!

At first sorry, I forgot an important word: I need natural transition orbitals (NTOs) in the correct .wfn-format for Multiwfn.

I have tried egrad with $wfn $exopt 1 but it is still not the same .wfn file which I need.
Furthermore I also tried tm2molden after escf calculation, because Multiwfn normally can deal with these files and create NTOs in a .wfn-file but for this I need an additional .log- or .out-file ("Please input path of Gaussian/ORCA output file or plain text file, electron excitation information will be loaded from this file", from Multiwfn Terminal). I don't have any .log or .out file which works here.


Best regards,
_mtkt_

uwe

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 558
  • Karma: +0/-0
Re: Create wfn file with natural orbitals for Multiwfn
« Reply #3 on: May 26, 2021, 05:03:58 PM »
Hi,

did you add
$wfn
$exopt 1
in two different lines?

As you have the full transition density when running an egrad job, exporting wfn will transform it to natural transition orbitals which are then written to the wfn file.
 

marand

  • Full Member
  • ***
  • Posts: 25
  • Karma: +0/-0
Re: Create wfn file with natural orbitals for Multiwfn
« Reply #4 on: November 09, 2023, 01:22:13 PM »
I am sorry to enter the conversation without invite, but I would like to know if it is possible to generate using either turbomole or wfn the natural orbitals for the difference density, specifically if the latter comes from subtracting electronic densities obtained in two separate calculations (e.g. ground state KSDFT and DeltaDFT fot the lowest triplet state).
Thank you for any assistance on the matter.
Yours sincerely
Marcin Andrzejak

uwe

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 558
  • Karma: +0/-0
Re: Create wfn file with natural orbitals for Multiwfn
« Reply #5 on: November 28, 2023, 11:43:29 AM »
Hello,

if you have the difference density (as it is generated by egrad) I wonder why you need NTOs in addition? NTOs are a quick way to get an idea of how the difference density does look like (where the electrons come from and where they are going to).
If you have the difference density available, you can simply plot it (using +/- view) to see the change in density directly.
Or do I miss something important?

Best Regards