Dear Peng:
To show the difference between DOSes, the best way is put the Fermi level at
same place. Normally, (E-Ef) is preferred by lots of researches as the variable
of x-axes. Manually shifting DOS for matching is not a good idea, since HOMO
should be the standard and adjusted to the same position (although Fermi level
given by PWscf is LUMO for semiconductor or insulator).
BTW, in PWscf, nscf calculation also provides Fermi energy, which usually is
slightly different with that from scf calculation and displayed by setting
verbosity='high' in &control namelist.
--
GAO Zhe
CMC Lab, Materials Science & Engineering Department,
Seoul National University, South Korea
At 2012-09-08 01:16:39,"Peng Chen" <pchen at ion.chem.utk.edu> wrote:
Dear All,
I calculated the total density of states (in the attachment) for both afm and
fm states of an insulator. I shifted the curves so that the band gap starts at
0 eV. In the top figure, the Fermi energy obtained from the scf calculation
falls between the gap. So if I plot Fermi surface, can I get reliable result?
In the bottom figure, I shifted the spin down DOS of fm state so that it can
match with the spin down DOS of afm state. I am not sure if that is the right
way to show the DOS difference of these two states.
Another question is about the accuracy of calculation. In the scf calculation,
I made the total energy converged within 0.001 Ry with related to the ecutwfc,
degauss, kpoints... Then I did band structure calculation. Can I say the
error of calculated band energy is within 0.001 Ry?
Best Regards.
Peng
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://www.democritos.it/pipermail/pw_forum/attachments/20120908/7e79df07/attachment.htm
