[Wannier] Re_ Wannier functions in real space

[Oier Arcelus]

>Oier,

  >In reading this I?m wondering if you have missed something fundamental about wannier functions: The WF are not periodic in the crystallographic unit cell. In the limit of infinite k-point sampling they are isolated - and if >we sample the Brillouin Zone with a NxNxN mesh then the WF are periodic in a NxNxN supercell of the unit cell.
  >There is a practical illustration of this in example01 (GaAs) such has a very coarse sampling the BZ, and hence you see the periodic images of the WF appear in the plots, if the plotting supercell is large enough.

  >Jonathan


Oh, thats a big mistake from my part. So I guess It is sufficiente to consider a supercell sufficiently big so that the Wannier functions are small  enough by the time they arrive to the supercell border, and then just consider the exchange correlation potential in that supercell.


Oier.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/wannier/attachments/20170216/54672512/attachment.html>