Dear Cao,
I'm pretty sure that at the DFT level (without SO) the DOS of graphene has
to be linear close to the Dirac point (which corresponds to the Fermi energy
for pristine graphene).
You can easily understand this by looking at the band structure studied with
the primitive unit cell. The gap that you see is due to the discrete k-point
sampling and the peculiar band structure of graphene. I also think that the
effect is amplified by the fact that you use the tetrahedron method, which was
devised for three-dimensional systems (though it seems to work fine also for
graphene). Most probably, for your k-point mesh the two closest eigenvalues
across the Dirac point are spaced by about 0.2eV.
I don't think this is a real problem for your application.
HTH
GS
Il giorno 21/mag/2012, alle ore 14.13, Cao TF ha scritto:
> Dear QE users,
> I have calculated the density of states (DOS) of pure graphene. The supercell
> used in my work is a rectangular cell containing 60 carbon atoms. But
> something has puzzled me greatly. There is always a little energy gap (~0.2
> eV) in the DOS, which contracts with the fact that graphene is zero gap
> semiconductor. Although I have change many parameters. This problem has not
> been solved. Here I give the input file and I will be greatly appreciated if
> you can give me some suggestions.
> For nscf.in
> &CONTROL
> title = 'graphene layer' ,
> calculation = 'nscf' ,
> restart_mode = 'from_scratch' ,
> outdir = './tmp' ,
> pseudo_dir = '/lustre/ISSP2/tfcao/pseudo' ,
> prefix = 'graphene' ,
> /
> &SYSTEM
> ibrav = 8 ,
> a = 12.31 , b = 12.79312342 , c = 12.000 ,
> nat = 60 ,
> ntyp = 1 ,
> occupations = 'tetrahedra' ,
> tot_charge = 0.0,
> nspin = 2 ,
> starting_magnetization(1) = 0.5 ,
> ecutwfc = 30.0 ,
> ecutrho = 300.0 ,
> /
> &ELECTRONS
> conv_thr = 1.0d-8 ,
> mixing_mode = 'local-TF' ,
> mixing_beta = 0.7 ,
> diagonalization = 'david' ,
> /
? Gabriele Sclauzero, EPFL SB ITP CSEA
PH H2 462, Station 3, CH-1015 Lausanne
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://www.democritos.it/pipermail/pw_forum/attachments/20120522/b42e965d/attachment.htm
