Dear Elie, it looks like you have way too few k-points for your nscf calculation. For a well converged DOS it isnecessary to have sufficient dense k-grid. By increasing the number of k-points you'll see, that the DOS will be much smoother, because up to know the 'waves' in your DOS doesn't seem to be physical, I think they are artificial and a result of a low k-sampling.
If the grid is dense enough and your DOS doesn't change anymore you can also decrease the smearing and improve the behaviour of your spectra close to the gap. Hope that helps Lars 2011/8/2 Elie Moujaes <elie.moujaes at hotmail.co.uk> > Dear all, > > I am trying to reproduce calculations done on the electronic DOS of a 60 > carbon system. I have got that the DOS is around 0.2 at the Fermi level > whereas the calculations done previously got that it is exactly zero. In > addition their curve seems sharper than mine. I used degauss=0.01 in nscf > and scf calculations. the grid for scf was 4 4 1 whilst it was 9 9 1 for > nscf. You will find attached the ps file of the DOS. Can anyone advise me on > that? > > Thanks > > Elie Moujaes > > University of Nottingham > NG7 2RD > UK > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > -- Lars Matthes Institut f?r Festk?rpertheorie und -optik Friedrich-Schiller-Universit?t Jena Max-Wien-Platz 1 07743 Jena Germany Phone: +49.3641.947163 Mail: Lars.Matthes at uni-jena.de -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20110802/5a40bb60/attachment.htm
