[CP2K-user] [CP2K:18751] Calculating dipole moments for static dielectric constant calculation. Please help me decide whether and how to use Ewald summation.

Nathalie Smith nathaliesmith26 at gmail.com
Tue May 2 11:20:53 UTC 2023


Dear CP2K users, 

I hope it's okay to insert my own question into this old conservation. 
Sorry in advance if it's not. I tried to make my own post about my issue, 
but didn't get any answers so far.

I am also trying to calculate static dielectric constants of 3D periodic 
systems using the fluctuation formula, but I also get values that are 1-2 
orders of magnitude off when using PERIODIC TRUE and the Berry Phase dipole 
moments. 

Is there something fundamental about the output of the dipole moments that 
I don't understand? I thought that for every step, the components (x, y and 
z) of the total dipole moment of my unit cell are printed (in Debye and in 
a.u.). 

I gave my input file for the MD simulation and my python script for the 
calculation of the dielectric constant 
here: https://groups.google.com/g/cp2k/c/KU_JGMdWrro .
Please be aware that I now know that I need to omit the factor 4*pi, 
however I still get very wrong results.

I would be super thankful for any help with my problem! I am getting quite 
desperate with this problem...

Best wishes,

Nathalie Smith

hut... at chem.uzh.ch schrieb am Donnerstag, 4. Februar 2021 um 16:04:49 UTC+1:

> Hi
>
> that is wrong advice. You have to use the Berry phase dipole
> in periodic settings.
> Probably you were using the polarization/volume, check the units!
> You might have to multiply by the cell volume (again units!).
>
> regards
>
> Juerg Hutter
> --------------------------------------------------------------
> Juerg Hutter Phone : ++41 44 635 4491 <+41%2044%20635%2044%2091>
> Institut für Chemie C FAX : ++41 44 635 6838 <+41%2044%20635%2068%2038>
> Universität Zürich E-mail: hut... at chem.uzh.ch
> Winterthurerstrasse 190
> CH-8057 Zürich, Switzerland
> ---------------------------------------------------------------
>
> -----cp... at googlegroups.com wrote: -----
> To: "cp2k" <cp... at googlegroups.com>
> From: "Stns" 
> Sent by: cp... at googlegroups.com
> Date: 02/04/2021 02:45PM
> Subject: Re: [CP2K:14691] Calculating dipole moments for static dielectric 
> constant calculation. Please help me decide whether and how to use Ewald 
> summation.
>
> Dear Juerg:
>
> Thank you very much for confirming that. 
>
> Your reply makes sense, but I have been using "PERIODIC FALSE" because 
> someone advised me that that's the only way to get "traditional" dipole 
> moments that can be used with fluctuation formulas for calculating the 
> dielectric constant. When I previously didn't use it, I was getting Berry 
> Phase dipole moments that led to dielectric constant values that were many 
> orders of magnitude off. If there are any comments you can make about this, 
> I would be grateful. 
>
> It sounds like you are saying that I should use "PERIODIC TRUE." If you 
> have time and see any other changes I should make to be able to use 
> fluctuation-type formulas for calculating the dielectric constant and get 
> reasonable values, I would welcome any tips. Many thanks. 
> -- 
> You received this message because you are subscribed to the Google Groups 
> "cp2k" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to cp2k+uns... at googlegroups.com.
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/cp2k/8dfc787a-4480-4016-a0b5-6a88fe080280n%40googlegroups.com
> .
>
>

-- 
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+unsubscribe at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/a1cc549a-ded8-44f6-a1f1-69b6b8bf489cn%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20230502/053afd0c/attachment.htm>


More information about the CP2K-user mailing list