[Wien] FER: Fermi level position problem

[Peter Blaha]

>  I am calculating on different systems, but I have the same problem:
> plotting the band structure the Fermi Level is often 0.2 - 0.4 eV lower
> in energy (down into the Valence Band) than the expected position (after
> have introduced the value from .scf file into .insp file). The problem
> is for exemple in calculating graphite. I have done convergence in
> charge with 0.0001 of convergence parameter. Is it a problem of
> convergence? Or is it something else?

I would not expect it to be a convergence problem.

Either you are doing something wrong (changing XC potential, RKmax, Energy-
parameters, different structure or lattice parameters, wrong nr.of electrons
in case.in2,...)

or you use a (crude) "shifted" k-mesh in the scf cycle (if your valence-
band maximum (EF) is at Gamma, but Gamma is not in the scf k-mesh, then EF
will be "lower" and when plotting a bandstructure including Gamma it
looks as if EF is inside the valence bands.

or you use a large broadening scheme ? Use TETRA

                                      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/
--------------------------------------------------------------------------