[DFTB-Plus-User] Defect + magnetism

toufik esssakhri tousak at hotmail.fr
Tue Mar 10 11:18:19 CET 2020


Thank you very much,
In fact, I am comparing my results with DFT calculations, from DFT dos_up and dos_down
are different. As you said, to break spin symmetry I set RelaxTotalSpin to yes. Present atoms
are not magnetic. The origin of magnetism is the defect. Thus Iam not using SpinPerAtom
initialisation. Here is the spin polarisation block I am using:

  SpinPolarisation = Colinear {
    UnpairedElectrons = 1
    RelaxTotalSpin = Yes
  }
  SpinConstants = {
    ShellResolvedSpin = Yes
    C = { -0.0306  -0.0251  -0.0251  -0.0255 }
    F = { -0.0370  -0.0312  -0.0312  -0.0299 }
  }
I wonder if I am missing something

Best Regards


De : DFTB-Plus-User <dftb-plus-user-bounces at mailman.zfn.uni-bremen.de> de la part de Bálint Aradi via DFTB-Plus-User <dftb-plus-user at mailman.zfn.uni-bremen.de>
Envoyé : mardi 10 mars 2020 10:59
À : dftb-plus-user at mailman.zfn.uni-bremen.de <dftb-plus-user at mailman.zfn.uni-bremen.de>
Cc : Bálint Aradi <aradi at uni-bremen.de>
Objet : Re: [DFTB-Plus-User] Defect + magnetism

Dear Toufik,

On 3/9/20 9:22 PM, toufik esssakhri via DFTB-Plus-User wrote:
> Thank for your response,
> Attached the density of states, we can see the dos_up  = dos_down
> from the band.out file  we get the defect levels :

To me, the band.out snippet and the dos seems to me consistent. You have
two peaks in the DOS: around -8.3 and -8.6, the one at -8.6 is roughly
twice as big as the one at -8.3. You can decrease the broading as a
command line argument to dp_dos, but to me, it seems to be OK as it is.

Now, whether spin-up and spin down must differ is a different question.
It depends on your system. You probably want to set RelaxTotalSpin = Yes
so that you have one common Fermi-level for both spin channels. And it
may be necessary to start from a spin-polarized initial guess (see
InitialSpins) to break the spin symmtetry.

  Best regards,

  Bálint


--
Dr. Bálint Aradi
Bremen Center for Computational Materials Science, University of Bremen
http://www.bccms.uni-bremen.de/cms/people/b-aradi/


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.zfn.uni-bremen.de/pipermail/dftb-plus-user/attachments/20200310/da3e1810/attachment.htm>


More information about the DFTB-Plus-User mailing list