[gpaw-users] How can the vacuum potential be found out

Jens Jørgen Mortensen jensj at fysik.dtu.dk
Tue Feb 7 08:11:54 CET 2012


On 06-02-2012 17:30, Georg Kastlunger wrote:
>
> 	
>
> 	
>
> 	
>
> 	
>
> Thanks again for your help,
> but unfortunately we are primarily interested in the case of periodic
> boundary conditions.
>
> Is there maybe anybody else who might know how the zero energy is
> defined in periodic cells?

For that case, the average of the Coulomb potential is zero, where the 
Coulomb potential is the potential we get from solving the Poisson 
equation "nabla v_coulomb=4 pi rho", where rho is the pseudo charge 
density.  The total effective potential is the v_coulomb+v_XC.

Jens Jørgen

> Regards,
> Georg
>
>
> On 02/06/2012 05:22 PM, Ask Hjorth Larsen wrote:
> >  Hi
> >
> >  On Mon, 6 Feb 2012, Georg Kastlunger wrote:
> >
> >>  Which boundary do you mean?
> >>  Is it the value at one certain point in the cell or an average of an
> >>  interface between unit cells?
> >>
> >>  Best regards,
> >>  Georg
> >
> >  The potential is zero at all the non-periodic cell boundaries.  This
> >  means one cannot shift the potential arbitrarily - it is unique, and
> >  so are the eigenvalues.
> >
> >  Each eigenvalue equals the derivative of the total energy with respect
> >  to the occupation of the corresponding state (keeping wavefunctions
> >  and density constant).
> >
> >  (The same is not true if the cell is fully periodic; then I'm not sure
> >  what determines the potential and eigenvalues)
> >
> >  Regards
> >  Ask
> >
> >
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://listserv.fysik.dtu.dk/pipermail/gpaw-users/attachments/20120207/b9c0448f/attachment.html 


More information about the gpaw-users mailing list