[Wien] NiSe2

[Peter Blaha]

> >So either do not use irrep at all; or when you calculate the
> >bandstructure,
> >you need to make some tricks and put the k-vectors slightly away from
> >the surface (Take the X point at 0.499 instead of 0.5, and so on)
> 
> Thanks again but how can I make it? Modifying e.g. "xcrysden.klist"?

My suggestion is that you should NOT use irrep at all. You can still 
calculate the bandstructure. Use a "dense" mesh (many points) and plot 
eventually points only (not lines). For a small case like NiSe2 this 
should not be a problem.

Alternatively, you need to modify case.klist_band.  (xcrysden.klist is 
created by xcrysden and copied in w2web later on (when setting the kmesh)
to case.klist_band.    
x lapw1 -band    uses this klist_band file.

And yes, you have to edit this file by hand and change the coordinates as 
you desire. (the format is   running-index,  ix, iy, iz, idiv;   so
499 0 0 1000      means kx=499/1000; ky=kz=0

 
                                      P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-15671             FAX: +43-1-58801-15698
Email: blaha at theochem.tuwien.ac.at    WWW: http://info.tuwien.ac.at/theochem/
--------------------------------------------------------------------------