[QE-users] Problem with PBE0 calculation

mkondrin mkondrin at hppi.troitsk.ru
Thu Jan 12 09:44:21 CET 2023 

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 <rb1820 at 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.mattioli at 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 <rb1820 at 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.mattioli at 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 at 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.mattioli at 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 at lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users

More information about the users mailing list