[QE-users] pseudopotential for spin-orbit calculation with hybrid functional

JULIEN, CLAUDE, PIERRE BARBAUD julien_barbaud at sjtu.edu.cn
Tue May 26 13:18:24 CEST 2020



Dear users, 




I am currently trying to run a calculation taking spin-orbit coupling (SOC) into account, using a hybrid functional (PBE0) for a more accurate determination of the band gap of my material (FAPbI3). I am running QE 6.4 


Trying with SOC alone, no issue. However, when I try to combine it with PBE0, I realize that the combination of both doesn't work for a USPP potential as the scf run crashes before even starting with the following error message: 

Error in routine setup (1): 
Noncolinear hybrid XC for USPP not implemented 





I would use another kind of pseudopotential, but the only PPs proposed on the QE website for atoms such as I or Pb ar either *rrkjus* (which is of USPP type) or *kjpaw* (and I read in EXX_example that PAW PPs are not supported for hybrid calculations). Anyway, if tried to run with both kinds, and I actually get the same error message as above regardless (because they are actually both USPPs?) 




Knowing this, what (fully relativisitic) PP should I choose to run a calculation with SOC+PBE0? Does it mean that QE effectively does not support SOC+hybrid functional for atoms such as Pb or I, for which I can only find USPP/PAW pseudopotetials? 


If it is of any help, the input file is attached 





Thanks in advance 

Julien 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20200526/e280d815/attachment.html>
-------------- next part --------------
&CONTROL
  pseudo_dir="/THFS/home/shu_kwy/util/pseudo"
  prefix='cubic_6.36623'
  calculation = 'scf'
  outdir='./out_SOC'
  max_seconds=7200
/
 
&SYSTEM
  input_dft='pbe0', nqx1=2, nqx2=2, nqx3=2,
  exx_fraction=0.166,
  x_gamma_extrapolation=.true.,
  exxdiv_treatment='gygi-baldereschi',
  lspinorb=.true.,
  noncolin=.true.,
  starting_magnetization=0.0,
  nat= 12
  ntyp= 5
  ibrav= 1, a=6.36623,
  ecutwfc=50, ecutrho=400,
  nbnd=64
/
 
&ELECTRONS
  mixing_beta = 0.3
  conv_thr =  1.0d-9
  electron_maxstep=20
/
 
&IONS
  
/
 
&CELL
  press_conv_thr=0.05
/
 
ATOMIC_SPECIES
C   12.011  C.pbesol-n-rrkjus_psl.1.0.0.UPF
N   14.007  N.pbesol-n-rrkjus_psl.0.1.UPF
H    1.008  H.pbesol-rrkjus_psl.0.1.UPF
Pb  207.2   Pb.rel-pbesol-dn-rrkjus_psl.1.0.0.UPF
I   126.90  I.rel-pbesol-n-rrkjus_psl.1.0.0.UPF


ATOMIC_POSITIONS (angstrom)
H        3.9132685770     2.1949237260     4.5818352540
H        4.7891514320     4.3213346350     3.6717990180
H        2.4219539810     3.2264032400     2.1956159760
H        3.9398199850     4.9819812450     2.2685828100
H        2.5221427740     1.4946600340     3.7426779320
Pb       0.1962305740     0.0744428740     6.3436044810
C        3.1976425550     3.2384436710     2.9725789760
I        6.2697830120     3.2198407720     6.0523760340
I        3.3873580660     6.2895201700     0.1342846240
I        0.4355962900     0.0976052520     3.1606482660
N        4.0255299560     4.2511327860     2.9895373240
N        3.2296288180     2.2393576460     3.8168542150


K_POINTS automatic
4 4 4 0 0 0 


More information about the users mailing list