[Pw_forum] born effective charges

[Stefano Baroni]

In your case it would be much more effective to calculate the  
effective charges as the forces linearly induced by an applied  
homogeneous electric field (just one electric-field calculation),  
rather than as the polarization linearly induced by a nuclear  
displacement. The two definitions do of course coincide (they both  
amount to calculating the mixed second derivative of the crystal  
energy with respect to the electric field and a nuclear  
displacement), but the calculation of the force response to an  
electric field only requires just 1 calculation, rather that 100 (in  
your case)!

S.

On Oct 9, 2006, at 4:40 PM, Vivek Ranjan wrote:

> Hi !
>
> I am trying to calculate born effective charges for a system  
> consisting of
> more than hundred atoms. I am doing it using linear response, i.e.  
> I relax
> the structure, then I do a scf calculation and then I run a gamma  
> point
> phonon calculation. After the programme prints the dielectric  
> constants,
> it must calculate the born effective charges. With the given computer
> resources at my disposal, it cannot complete this part. So, I must  
> restart
> the calculation from the previous step. I use "recover.true." flag.  
> But it
> never seems to go beyond the dielectric constant calculation part. Is
> there a way I can do this calculation ?
>
> Thanks,
> Vivek
>
> PS : I know that I can calculate polarization using berry phase
> calculation. Then I can move atoms one-by-one and calculate effective
> charges by finite difference. But that will be doing many  
> calculations, so
> I want to avoid this method.
>
>
>
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum

---
Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  
Trieste
[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)

Please, if possible, don't  send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html



-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20061009/8a6f23b8/attachment.htm