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 <[email protected]>:
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 < [email protected]> 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 <[email protected]>: > 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: <[email protected]> _______________________________________________ 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 [email protected] 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: <[email protected]> _______________________________________________ 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 [email protected] https://lists.quantum-espresso.org/mailman/listinfo/users
