gap's problem

[fatimazahra]

Hi everyone
I have a system with odd number of electrons, I have ran calculation with quantum espresso, I got that the system is a semiconductor with a band gap (1.20 eV), when I run with yambo, the system is metal, and I get this message in r-setup:

[02.05] Energies [ev] & Occupations
===================================

Fermi Level [ev]:-3.386976
Electronic Temp. [ev K]: 0.2585E-1 300.0
Bosonic Temp. [ev K]: 0.2585E-1 300.0
El. density [cm-3]: 0.812E+23
States summary : Full Metallic Empty
0001-0007 0008-0008 0009-0024

[WARNING]Metallic system
N of el / N of met el: 15.00000 0.98075
Average metallic occ.: 0.980745
X BZ K-points : 100

[Daniele Varsano]

Dear Fatima,
Most probably this is due to the electronic temperature (did you use any smearing in the qe run?).
You can anyway avoid this problem by redoing the setup by setting:

Code: Select all

ElecTemp=0

in your input file (remove first the ndb.* datatbase calculated previously).

In any case please provide your QE output and complete yambo report and we will have a look.
Next: the initialization subforum is more appropriate than the self-energy section.

Best,
Daniele

[fatimazahra]

Hi Daniele
yes, I need to use smearing in QE calculation,
I redid the setup with ElecTemp=0, but I didn't avoid my problem,

[Daniele Varsano]

Dear Fatima,
this is rather strange. Could you please post the complete yambo report and qe output?
Best,

Daniele

[Zahra Taghipour]

Dear Developers,

I intend to use GW correction on DFT bandgap of my sample. The material that I am working on is a semiconductor with band gap as small as 0.1 meV!
Due to convergency issue, I have to use Smearing. But, this led yambo to see my material as metaic!

In order to test different parameters, I switched to the well-known and simpler case of GaSb! The PP that I have been using are GA.PBE-D-MT_FHI.UPF, and SB.PBE-MT_FHI.UPF. However, even in the case of GaSb, and without using Smearing, I get BG=0! And again, Yambo sees it as metalic and after a long calculation of GW, I get the same corrections on both CB and VB, at gamma point (The gap at gamma point is supposed to be around 0.8 meV).

I already tried to set the ElecTemp equal to zero. But it did not help!
I have attached the result of QE and Yambo for GaSb.

I would appreciate if you could help me resolve this issue.

Thanks,
Zahra Taghipour,
PhD student,
University of New Mexico, USA

[Daniele Varsano]

Dear Zahra,
from what I can see, the QE ground state calculation gives you:

Code: Select all

 highest occupied, lowest unoccupied level (ev):     6.5346    6.5346

ie, a metallic beahvour or at least a gap smaller then 0.1meV.

For what concern the yambo setup (and GW calculations), the electronic temperature it is not set to zero:
r_setup:

Code: Select all

 .-Input file : (none)
 | ElecTemp=0.1000E-3     eV    # Electronic Temperature
 | BoseTemp=0.1000E-3     eV    # Bosonic Temperature

o-PP_181Ry-nbnd300.qp

Code: Select all

# | gw0                          # [R GW] GoWo Quasiparticle energy levels
# | ElecTemp=0.1000E-3     eV    # Electronic Temperature
# | BoseTemp=0.1000E-3     eV    # Bosonic Temperature
# | FFTGvecs=  5817        RL    # [FFT] Plane-waves

Please be sure you are running the right input file!!

Best,
Daniele

[Zahra Taghipour]

Dear Daniele,

I have tried to set the ElectTemp to zero as follow:

Running: yambo -i -V gen -F 00init.in
-> this already shows ElectTemp=0!!!

Running yambo -F 00init.in
-> This reads 00init.in as input and gives r_setup as output, however, this shows:
{ Electronic Temp. [ev K]: 0.1000E-3 1.160 }

Other than this, I do not know how I am supposed to change the ElectTemp to zero! Please help me to resolve this issue. It has been a while that I am struggling with it without any result!

Attached, are 00init.in and r-setup files, if needed!

Zahra,
PhD student,
UNM, USA

[Daniele Varsano]

Dear Zahra,
ok the problem here is that the QE ground state calculation outcome is a metallic system.
Even if you put a zero temperature, yambo cannot converge the Fermi level and need to add an electronic temperature to do that.
If you think that this is an artifact, you should try to get a non-metallic behavior from the qe calculations, for instance you can try to add unoccupied states in the scf calculation and work with "fixed" occupations, or you can ask suggestions to the pw_forum.

Best,
Daniele

[kerensanni]

Hi everyone
I have a system with odd number of electrons, I have ran calculation with quantum espresso, I got that the system is a semiconductor with a band gap (1.84 eV), when I run with yambo, the system is metal.
I changed the Electemp to 0. Is still the same issue. I had to use smearing in my quantum espresso calculation because of the Manganese in my system. Pls what should I do.

Thanks,

Nagura Jonah,
Undergraduate student,
Department of Physics,
University of Ibadan.

[Daniele Varsano]

Dear Jonah,
please post your setup report file together with quantum espresso input and output.

Best,
Daniele