Multiwfn forum

Ossama

Member

Registered: 2019-11-12

Posts: 20

Electric dipole moment

Hi
Forgive me if this is a naïve question!

I find that the electric dipole moment of a molecule, calculated after the grid of electron density is calculated, is dependent on the origin coordinate. This makes the obtained result rather meaningless. I suppose if the nuclear charges were incorporated in the calculations this would make the obtained result origin-independent, at least for neutral molecules or a zero-net charge distribution.

1- Am I right?

2- And if so, is there a handy way to calculate the electric dipole moment without excluding the nuclear charges?

Thanks,,

sobereva

Tian Lu (Multiwfn developer)

From: Beijing

Registered: 2017-09-11

Posts: 1,833

Website

Re: Electric dipole moment

For neutral systems, electric dipole moment is independent of choice of origin. This can be easily proved, and you can easily find relevant knowledge via Google. Only charged systems have origin-dependent dipole moment.

I don't know how did you calculate electric dipole moment, perhaps there is something wrong.

Ossama

Member

Registered: 2019-11-12

Posts: 20

Re: Electric dipole moment

Dear Prof. Lu,

As mentioned in Multiwfn manual (3.6- Outputting and plotting specific property within a spatial region):

"If what you calculated is electron density, the molecular dipole moment evaluated based on grid data is also printed out."

I carry out the following steps:

              5 Output and plot specific property within a spatial region (calc. grid data)
              1 Electron density
              3 High quality grid  , covering whole system, about 1728000 points in total

I've an example here, Cl2 molecule, where two .wfn files were used. There is no difference between them other than the coordinates origin.

The first one (Cl2.wfn) gives:

Electric dipole moment estimated by integrating electron density
X component:        4.596854 a.u.   11.684035 Debye
Y component:       -1.037455 a.u.   -2.636946 Debye
Z component:       -0.035310 a.u.   -0.089750 Debye
Total magnitude:    4.712602 a.u.   11.978239 Debye

The second one (Cl2-reoriented.wfn) gives:

Electric dipole moment estimated by integrating electron density
X component:       -0.218321 a.u.   -0.554918 Debye
Y component:       -0.386888 a.u.   -0.983370 Debye
Z component:       -0.035310 a.u.   -0.089750 Debye
Total magnitude:    0.445638 a.u.    1.132699 Debye

Cl2.wfn

Cl2-reoriented.wfn

Last edited by Ossama (2022-08-07 12:25:08)

Ossama

Member

Registered: 2019-11-12

Posts: 20

Re: Electric dipole moment

I tried to modify the X, Y and Z components of the electric dipole moments by adding the effect of the two chlorines' nuclei (for both files) but I failed to get identical values of the total magnitude.

sobereva

Tian Lu (Multiwfn developer)

From: Beijing

Registered: 2017-09-11

Posts: 1,833

Website

Re: Electric dipole moment

Please do not calculate dipole moment in this way. The integral information (including dipole moment) shown after calculating grid data in main function 5 is evaluated simply based on uniformly distributed grid, the result is quite inaccurate if electron density around nuclei is large and varies sharply. This problem is especially prominent if your system contains relatively heavy atoms, because heavier atom has sharper variation of electron density around nucleus.

In Multiwfn the best way of calculating electric dipole moment is using subfunction 5 of main function 300, which is fastest and most accurate, because it is fully based on analytic integral between basis functions.

Ossama

Member

Registered: 2019-11-12

Posts: 20

Re: Electric dipole moment

Thanks, Prof. Lu

It worked.

I downloaded the latest version (3.8) since the function you mentioned didn't exist in the previous versions. Both files of Cl2 gave a dipole moment magnitude of zero, which, of course, is the expected value for a homonuclear diatomic molecule.

My regards,