[Pw_forum] for bandgap information

[Nicola Marzari]

On 22/09/2014 13:23, Neeraj Goel wrote:
> i have attached my CuInS2 , non- self consistent file. what is the
> velue of badgap in this case. and how will i get this.
>
>                                         regards
>
>                                       Neeraj Goel
>
>                                  Indian School of Mines
>

The direct band gap is going to be equal or smaller than {minus the highest
occupied state plus the lowest unoccupied state} calculated at each key
point, while the indirect band gap is going to be equal or smaller
than {minus the highest occupied state at any key point} plus {the lowest
occupied state at any k-point}.

As your mesh fineness increases, these values will converge to the right 
result,
and if you'll understand also if the gap is direct or not.

*BTW, great if anyone could volunteer to include these calculations in 
the SCF and
NSCF code (since they are very easy to do), with the due caveats.*

Ideally, a plot of the band structure along high-symmetry directions 
will help -
do it first with germanium or silicon, and remember that the band 
structure in
DFT is in any case not particularly good, and not meant in any formal sense
to reproduce the correct quasiparticle excitations (even if the DFT band 
structure
can be quite ok, a part from a smaller gap than you should have, if you use
LDA or GGA, and quite good gaps if you use much more expensive HSE 
functionals or the
like).

				nicola

>
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://pwscf.org/mailman/listinfo/pw_forum
>

-- 

----------------------------------------------------------------------
Prof Nicola Marzari, Chair of Theory and Simulation of Materials, EPFL