Problem with WANNIERIZE

Dear EPW users,

I’m new to EPW and trying to learn it. As I was checking the examples I’m having some trouble with the "wannierize" part, as I’m not quite familiar with the "Wannier90" code. I’m consulting the Wannier90 docs, but it will be a great help if someone give me a brief description of the following input keywords (I’m using the input from EPW/examples/mgb2):

[code:1qwn0kmf]nbndsub = 5,
nbndskip = 0

dis_froz_max= 8.8
proj(1) = ‘B:pz’
proj(2) = ‘f=0.5,1.0,0.5:s’
proj(3) = ‘f=0.0,0.5,0.5:s’
proj(4) = ‘f=0.5,0.5,0.5:s[/code:1qwn0kmf]’

On what basis I have to assign the values for "nbndsub" and "dis_froz_max" for any other system?

Best regards
Subhodip

Dear Subhodip,

You can see it as follow:

  • nbnd is the number of band that you have in your nscf calculations, say 40

  • nbndskip is the number of bands you skip from the bottom for the Wannierization, say 10

  • nbndsub is the number of band you include in your Wannierization from the bottom, say 30.
    In this example, the Wannierization will be done only on band 11-30.

  • dis_froz_max is the energy window used for disentanglement. For an explanation of disentanglement, please refer to the Wannier doc.

Best,

Samuel

Thank you Samuel.

Dear Samuel,

I’m trying to perform a calculation with my system consulting MgB2 example, Wannier90 doc and what you have told. The job terminates without any error after writing the .epb files.

[code:2nzsv2rb] ===================================================================
irreducible q point # 30
===================================================================

 Symmetries of small group of q:  4
      in addition sym. q -> -q+G:

 Number of q in the star =    1
 List of q in the star:
      1  -0.562508414  -0.746685321  -1.316479934
 dvanqq2      :    315.42s CPU    340.52s WALL (      65 calls)

    q(   64 ) = (  -0.5625084  -0.7466853  -1.3164799 )

 Writing epmatq on .epb files

[/code:2nzsv2rb]
What could be the probable reason for this?

Best regards
Subhodip