[WIEN]: Bandgaps calculations

[Peter Blaha]

Of course, the Kohn-Sham eigenvalues are not supposed to be the experimental
measured excitation energies. Nevertheless, many people use them as such.

Experience (and some theoretical considerations as the "discontinuity problem")
has shown, that for insulators/semiconductors the gap comes out about 50 % too 
small (and for very small gap semiconductors it may vanish at all); while
for metals the bands (in particular around EF) are pretty good (though not
perfect) in many cases (fermi surfaces, angle resolved photoemission).

Thus I'd expect that for MgB2 the bands agree quite well with PES or FS
measurements.

I would NOT use the Engel-Vosko GGA (option 25) for metals !


> Dear WIEN users,
>    If the gaps are not correct with the three options in the program; then in
> the case of several bands the bandedges should not be very precise. Is that
> correct?
>    For example, in MgB2 there are four bands at Ef, but two of them have the
> upper edge at ~0.8eV is this value correct?
>    Is the GGA option 25 stil available and would it give the correct values
> for these cases?
>    Yours
> 
>          Pablo de la Mora
> -------------------------------------------
> Peter Blaha wrote:
> 
> > ====
> > Peter Blaha <pblaha at theochem.tuwien.ac.at>
> > submitted the following contribution:
> > ====
> > 
> >  
> > > I make calculations of AlN,GaN and InN in wurtzite structure using
> > > WIEN97.10
> > > 
> > > The bandgaps for AlN y GaN are 50% below of the experimental value, but
> > > the bandgap for InN is zero.
> > >    
> > 
> > This is what LDA or the common GGAs give.
> > 
> >  
> > > How I can make calculations more precise of bandgaps in WIEN97.10?
> > >    
> > 
> > You can use GGA option 25. This is due to Engel and Vosko. It gives much
> > better bandgaps (often even pretty good), but you have to pay a price:
> > Never use the total energy! Equillibrium volumes,... are way too large.
> > 
> 
> 


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