Hi, I would suggest the following link (from Prof. Marzari's site) for a
better understanding:
http://theossrv1.epfl.ch/Main/ElectronicTemperature
bests,
sonu
*With kind regards,*
*S Kumar | Post Doc*
*ICReDD, JAPAN*
___
Quantum ESPRESSO is supporte
Before any scf calculations one does not know about the levels distribution,
so adding smearing may help in cases where the levels are too close near
HOMO. Otherwise, it wont harm anything else.
Mahmoud
From: Kazume NISHIDATE
To: xavier neto , Quantum ESPRESSO users Forum
Date: Tu
If the HOMO-LUMO gap is not enough small, then smearing with a small degauss
will not add any unoccupied levels to the calculations. However, if degauss
is large enough to do so, then the calculated results are far from the GS
properties.
Mahmoud
From: Kazume NISHIDATE
To: xavier net
Dear Xavier,
> So, I can use the Fixed occupation and, if the energy levels show a small
bandgap, I introduce the >Smearing. Is it right?
Yes, exactly.
>If yes, I just have to add the nbnd function in the input file?
Yes. It suffices to put nbnd equal to N/2 + 1 (N is number of electro
Dear users,
I tried to perform a vc-relax calculation for the tetragonal phase of
CaSiO3, to obtain equilibrium parameters and attain zero pressure. Attached
below is my input file and the end part of the output of vc-relax
calculation. Please have a look at the optimized cell-parameters matrix.
Th
> I've watched/read some tutorials about QE, and some of these showed
> that the occupation of semiconductors should be "smearing", other
> ones that it should be "fixed",
The occupancies of metals and semiconductors should be "smeared" to
obtain a good convergence of the calculation.
In the plan
Dear Mahmoud Payami,
Thank you for your answer. It helps me a lot.
So, I can use the Fixed occupation and, if the energy levels show a
small bandgap, I introduce the Smearing. Is it right?
About the second question. I've read that the result of HOMO - LUMO,
obtained after the SCF calculatio
Dear Dr. Tarek Hammad,
Is "nbnd = 278" enough to describe the bands above the Fermi level? Also,
maybe, "emax=15" has constrained your calculation, not sure.
Best,
Hari Paudyal
On Mon, Nov 1, 2021 at 2:12 PM Tarek Hammad
wrote:
> Dear QE team and users
>
> I ran qe to get the DOS for spin
Dear QE team and users
I ran qe to get the DOS for spin polarized case after writing the
required scf, nscf, and DOS files.
However, DOS plot looks strange such that the spin-up density of states
above EF were almost absent!!!
I attached with this message the following files:
scf, nscf
