[Pw_forum] scf and nscf and smearing

[Stefano de Gironcoli]

If you want to compute total energy (scf calculation) of a  metal you 
need to specify some
strategy to deal with the Fermi energy. You have to specify either 
occupations="tetrahedra"
(that requires the use of an automatically generated k-ponit grid) or 
occupations="smearing"
(in this case also smearing and degauss variables must be set;  see 
INPUT_PW for their definitions).

When you do a band structure (nscf) calculation  the scf potential is 
read from file and the
(previously computed) scf  hamiltonian is diagonalized on the k-points 
specified in the nscf input..
The Fermi energy is not computed, nor it is the charge density or the 
energy.
Thus, you don't need to specify occupations, smearing or degauss but 
they don't harm.
If you leave them in the input file they will (should) be simply ignored.

best wishes,
Stefano de Gironcoli


W. YU wrote:

>Dear PW users,
>
>For metal system, in scf calculation, it is required
>to use occupation='smearing'. Do we still required to
>include  it and the same other parameters such as
>degause in the following nscf calculations as we do in
>the scf calculation?
>
>thanks
>
>
>
>
>		
>__________________________________ 
>Do you Yahoo!? 
>Yahoo! Mail - now with 250MB free storage. Learn more.
>http://info.mail.yahoo.com/mail_250
>_______________________________________________
>Pw_forum mailing list
>Pw_forum at pwscf.org
>http://www.democritos.it/mailman/listinfo/pw_forum
>  
>