Dear Giuseppe,
Thank you for very pertinent comment. Of course, I have meant
stoichiometric materials, not the defect structures. Indeed, in the
latter case band structure is not well defined notion, but probably DOS
for some supercell approximation of defect structure is more robust.
Sincerely yours,
M. V. Kondrin
On 01/12/2023 01:06 PM, Giuseppe Mattioli wrote:
Dear M. V. Kondrin
The "scissors operator" strategy can be appropriate in the case of
stoichiometric materials, but it may be not when you investigate
defects or dopants in a host matrix, where the band structure (and the
DOS) can depend on the localization of electronic charge on defects
that can be seen as "deep" or "shallow" by different GGA or EXX
functionals. The best method to use depend on Rameswar's purposes.
Best
Giuseppe
Quoting mkondrin <mkond...@hppi.troitsk.ru>:
I also performed calculations using HSE06 functional for obtaining
electron DOS (using projwfc.x). It is compatible with hybrid
functionals and usually provide correct values of electron bandgaps.
After that I just used "scissors" operation on electron band
structure obtained using standard PBE to match bandgap obtained
previously.
Approximately in the same words I described this operation in the
recent paper which is under review now.
Hope this helps.
Sincerely yours,
M. V. Kondrin
On 01/12/2023 11:03 AM, Giuseppe Mattioli wrote:
AFAIK, nscf calculations are not compatible with EXX functionals.
There are a few workarounds, like manually adding the band path as
"fake" k-points with weight=0 to the regular grid obtained by the
automatic distribution used in the scf calculation.
I generally perform EXX calculations of isolated or very large
systems @Gamma. More experienced users may add more accurate
information.
HTH
Giuseppe
Quoting Rameswar Bhattacharjee <rb1...@georgetown.edu>:
Hi Giuseppe,
Thank you very much for your detailed reply. I followed the
instructions
and was able to complete the SCF calculation. Also, I did an NSCF
calculation using the same combination you suggested as a pre-step to
perform the band calculation. My ultimate aim is to get the band
gap using
a hybrid functional such as PBE0.
However, I am not able to run the band calculation with this
combination
got an error like below:
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Error in routine exx_grid_init (1):
wrong EXX q grid
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
It seems the k-points grid I am using is not suitable for EXX
calculation
although I used the same grid for PBE and it was perfect. I am sure
someone
in the forum had performed band structure using hybrid functional
and would
be able to help me resolve this. Any help would be highly appreciated.
Here is my input for band calculation using PBE0
------------------------------------------
&control
restart_mode='from_scratch'
calculation='bands'
prefix='Ethyne-Anthracene',
pseudo_dir = '/home/rb1820/QE-PP/NC',
outdir='./scratch'
/
&system
ibrav = 0,
tot_charge=0,
nat=36,
ntyp=2,
ecutwfc = 90,
ecutfock= 150,
input_dft='pbe0'
vdw_corr='grimme-d2',
nbnd = 60
/
&electrons
electron_maxstep = 1000
mixing_beta=0.7
diagonalization='david'
diago_david_ndim = 2
conv_thr=1.0d-8
/
&ions
/
&CELL
cell_dynamics='bfgs',
press = 0.0
cell_dofree = 'x',
/
ATOMIC_SPECIES
H 1.008 H_ONCV_PBE-1.2.upf
C 12.0107 C_ONCV_PBE-1.2.upf
ATOMIC_POSITIONS {angstrom}
K_POINTS { crystal }
11
-0.5 0.0 0.0 1.0
-0.4 0.0 0.0 1.0
-0.3 0.0 0.0 1.0
-0.2 0.0 0.0 1.0
-0.1 0.0 0.0 1.0
0.0 0.0 0.0 1.0
0.1 0.0 0.0 1.0
0.2 0.0 0.0 1.0
0.3 0.0 0.0 1.0
0.4 0.0 0.0 1.0
0.5 0.0 0.0 1.0
CELL_PARAMETERS {angstrom}
6.933462797 0.000000000 0.000000000
0.000000000 20.000000000 0.000000000
0.000000000 0.000000000 20.000000000
---------------------------------------------------------------
Thanks
Rameswar Bhattacharjee
Georgetown University
On Wed, Jan 11, 2023 at 3:21 AM Giuseppe Mattioli <
giuseppe.matti...@ism.cnr.it> wrote:
Dear Rameswar Bhattacharjee
I don't know if this is the actual source of error, but I strongly
recommend to use norm-conserving pseudopotentials when performing EXX
calculations. You have generally no speed-up with US because ecutrho
is more or less often the same, and in your case you are using
tremendously high (and likely useless) values for ecutwfc and
ecutrho.
Moreover, using NC you can finely tune the density basis set used to
build Fock integrals with ecutfock, otherwise stuck to the ecutrho
value. This combination of values is probably responsible for a very
high memory request (the code prints an estimate at the beginning of
the output). For example, *using C and H ONCV pseudopotentials*, you
should be on the reasonably accurate side using
ecutwfc=90.0
ecutfock=135.0~180.0
ecutrho is automatically fixed at ecutwfc*4=360.0 in the case of NC
HTH
Giuseppe
Quoting Rameswar Bhattacharjee <rb1...@georgetown.edu>:
Hi Everyone,
I am trying to get a band structure using a hybrid functional. As
suggested
previously in this forum I choose PBE0 functional with pbe-ultrasoft
pseudopotential. The calculation starts fine and scf iteration
converges
smoothly. But the job is stuck and does not complete with the
last line
"ACE
projected onto 60 (nbndproj) and applied to 60 (nbnd)
bands". I am
not sure what mistake I am making here or if it is just a memory
issue as
no error message is printing. Any suggestion would be highly
appreciated. I
am providing last few lines of the output and my input str for
reference.
Thank you.
Last few lines of the scf output
-------------------------------
highest occupied, lowest unoccupied level (ev): -3.5627
-3.2294
! total energy = -312.15802972 Ry
estimated scf accuracy < 3.2E-09 Ry
convergence has been achieved in 12 iterations
Using ACE for calculation of exact exchange
EXX grid: 4188907 G-vectors FFT dimensions: ( 100, 288,
288)
ACE projected onto 60 (nbndproj) and applied to 60 (nbnd)
bands
----------------------------------
Input str
------------------------------
&control
restart_mode='from_scratch'
calculation='scf'
prefix='Test',
pseudo_dir = '/home/QE-PP',
outdir='./scratch'
/
&system
ibrav = 0,
tot_charge=0,
nat=36,
ntyp=2,
ecutwfc = 140,
ecutrho = 1400.0,
input_dft='pbe0'
vdw_corr='grimme-d2',
nbnd = 60
/
&electrons
electron_maxstep = 1000
mixing_beta=0.7
diagonalization='david'
diago_david_ndim = 2
conv_thr=1.0d-8
/
&ions
/
&CELL
cell_dynamics='bfgs',
press = 0.0
cell_dofree = 'x',
/
ATOMIC_SPECIES
H 1.008 H.pbe-rrkjus_psl.1.0.0.UPF
C 12.0107 C.pbe-n-rrkjus_psl.1.0.0.UPF
ATOMIC_POSITIONS {angstrom}
K_POINTS automatic
12 1 1 0 0 0
CELL_PARAMETERS {angstrom}
6.933462797 0.000000000 0.000000000
0.000000000 20.000000000 0.000000000
0.000000000 0.000000000 20.000000000
----------------
Rameswar Bhattacharjee
Georgetown University
GIUSEPPE MATTIOLI
CNR - ISTITUTO DI STRUTTURA DELLA MATERIA
Via Salaria Km 29,300 - C.P. 10
I-00015 - Monterotondo Scalo (RM)
Mob (*preferred*) +39 373 7305625
Tel + 39 06 90672342 - Fax +39 06 90672316
E-mail: <giuseppe.matti...@ism.cnr.it>
_______________________________________________
The Quantum ESPRESSO community stands by the Ukrainian
people and expresses its concerns about the devastating
effects that the Russian military offensive has on their
country and on the free and peaceful scientific, cultural,
and economic cooperation amongst peoples
_______________________________________________
Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users
GIUSEPPE MATTIOLI
CNR - ISTITUTO DI STRUTTURA DELLA MATERIA
Via Salaria Km 29,300 - C.P. 10
I-00015 - Monterotondo Scalo (RM)
Mob (*preferred*) +39 373 7305625
Tel + 39 06 90672342 - Fax +39 06 90672316
E-mail: <giuseppe.matti...@ism.cnr.it>
_______________________________________________
The Quantum ESPRESSO community stands by the Ukrainian
people and expresses its concerns about the devastating
effects that the Russian military offensive has on their
country and on the free and peaceful scientific, cultural,
and economic cooperation amongst peoples
_______________________________________________
Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users
_______________________________________________
The Quantum ESPRESSO community stands by the Ukrainian
people and expresses its concerns about the devastating
effects that the Russian military offensive has on their
country and on the free and peaceful scientific, cultural,
and economic cooperation amongst peoples
_______________________________________________
Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users
GIUSEPPE MATTIOLI
CNR - ISTITUTO DI STRUTTURA DELLA MATERIA
Via Salaria Km 29,300 - C.P. 10
I-00015 - Monterotondo Scalo (RM)
Mob (*preferred*) +39 373 7305625
Tel + 39 06 90672342 - Fax +39 06 90672316
E-mail: <giuseppe.matti...@ism.cnr.it>
_______________________________________________
The Quantum ESPRESSO community stands by the Ukrainian
people and expresses its concerns about the devastating
effects that the Russian military offensive has on their
country and on the free and peaceful scientific, cultural,
and economic cooperation amongst peoples
_______________________________________________
Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users
_______________________________________________
The Quantum ESPRESSO community stands by the Ukrainian
people and expresses its concerns about the devastating
effects that the Russian military offensive has on their
country and on the free and peaceful scientific, cultural,
and economic cooperation amongst peoples
_______________________________________________
Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users