Charge displacement curve

shailja · 2022-12-17 10:51:27

I am trying to plot the charge displacement curve (CDC) (section 4.13.6 in the tutorial) for a non-linear system. When I plot the integral of the electron density difference qualitatively, I get a similar shape of the curve as provided in the tutorial with a high electron density value (~200 a.u.). The plot is attached herein. Now I have mainly two doubts as follows:

1) Is such kind of high electron density flow possible depending upon the system and strength of the electric field (in my case, it is field=Z+100 (0.001 a.u.) in gaussian)?

2) I saw an option in the multiwfn for multiplying or dividing the grid values with a factor. What is the purpose of these options? And how could one select the value of the factor and justify it?

sobereva · 2022-12-17 13:05:43

1 The data is evidently out of reasonable range. An electric field of 0.001 a.u. should not cause so large charge displacement. Please check Fig. 3 of my paper ChemPhysChem 2021, 22, 386 –395, in which I studied local integral curve of electron difference of cyclo[18]carbon under external electic field.

To help you, I need more information, including all commands you inputted in Multiwfn, and the wavefunction files used for plotting (you can upload to netdisk and paste downloading link here, or send me via E-mail)

2 Commonly that option is not needed, at least it is irrelevant to present problem.

shailja · 2022-12-17 16:22:57

Thank you for your kind reply. I have sent you all files via email (to sobereva@sina.com).

sobereva · 2022-12-18 05:21:19

I have checked your files. The problem comes from that the atomic coordinates in the two .wfn files are not identical. To obtain a meaningful electron density grid data, the atomic coordinates must be exactly the same between the wavefunction files. If you open the two .wfn files via text editor, you can easily find the atomic coordinates (starting from line 3) are quite different.

You should consider adding "nosymm" keyword during Gaussian calculation, which requests Gaussian not to automatically translate and rotate the system to standard orientation.

shailja · 2022-12-18 15:48:24

Thank you for your kind help.

I have regenerated the wfn file (without field) after reordering the atomic coordinates (kept the same as with field). I have sent you my new files and integral curve (now in the presence of 0.005 au), which seems qualitatively and quantitatively unreasonable. Can you please have a look to see if the WFN files still have problems or if there are other issues?

sobereva · 2022-12-20 06:25:43

The two .wfn files you sent to me still have different atomic coordinates.

zef50.wfn:

  P    1    (CENTRE  1)   0.00000000  0.00000000  0.00000000  CHARGE = 15.0
  B    2    (CENTRE  2)   0.00000000  0.00000000  7.71670419  CHARGE =  5.0
  N    3    (CENTRE  3)   0.86682304  0.00000000  4.86700286  CHARGE =  7.0
  N    4    (CENTRE  4)  -0.82848050 -0.37642400  3.29084088  CHARGE =  7.0
  C    5    (CENTRE  5)   0.07332515  3.08232283  8.23738421  CHARGE =  6.0
  C    6    (CENTRE  6)   1.98523477 -1.59222087  9.51242400  CHARGE =  6.0
...

without_filed.wfn:

  P    1    (CENTRE  1)   0.00000000  0.00000000  0.00000000  CHARGE = 15.0
  B    2    (CENTRE  2)   0.00000000  0.00000000  7.68806692  CHARGE =  5.0
  N    3    (CENTRE  3)   0.86016115  0.00000000  4.83288856  CHARGE =  7.0
  N    4    (CENTRE  4)  -0.83403499 -0.40621237  3.26314925  CHARGE =  7.0
  C    5    (CENTRE  5)   0.06557420  3.10240064  8.09199119  CHARGE =  6.0
  C    6    (CENTRE  6)   2.01880905 -1.54711334  9.45285460  CHARGE =  6.0
...

To 100% guarantee that the atomic coordinates in the two .wfn files are exactly the same, please performing single point calculation with and without external electric field using identical coordinates to obtain the their respective .wfn files, and meantime do not forgot to add "nosymm" keywords.

shailja · 2022-12-20 14:21:54

Thank you for your valuable suggestions. I understood to get the grid data, atomic coordinates should be precisely the same in both wfn files (with and without an electric field).

This worked well for CDC. However, I wonder about this new .wfn file of without field that I got after a single point from with field geometry Because this without-field geometry is not the actual one (it's like with field). So, this gives me around a 1.8 kcal/mol difference in energy between the actual optimized geometry without field (previously used for CDC) and this new one.

To get the difference in charge, I should use the actual geometry of without a field (but if I use that, atomic coordinates are not 100% the same).

Can you please let me know if I am missing something?

sobereva · 2022-12-21 07:32:54

You can use the geometry obtained either with or without electric field for plotting CDC, the conclusion should be basically the same.

Multiwfn forum

#1 2022-12-17 10:51:27

Charge displacement curve

#2 2022-12-17 13:05:43

Re: Charge displacement curve

#3 2022-12-17 16:22:57

Re: Charge displacement curve

#4 2022-12-18 05:21:19

Re: Charge displacement curve

#5 2022-12-18 15:48:24

Re: Charge displacement curve

#6 2022-12-20 06:25:43

Re: Charge displacement curve

#7 2022-12-20 14:21:54

Re: Charge displacement curve

#8 2022-12-21 07:32:54

Re: Charge displacement curve

Board footer