[Pw_forum] Does SOC conflict with HSE06?

lilonghua foursea at 163.com
Wed Aug 24 02:30:05 CEST 2011


Dear all,

Does SOC (spin-orbit coupling) conflict with HSE06 in pwscf?
SOC is ok! HSE06 is also OK! But when combining SOC and HSE06, the calculation stop at EXX calculation!
This is my input file:

 &control
  verbosity='default'           ! 'high'|'low'|'minimal'
  restart_mode='from_scratch'   ! 'from_scratch'|'restart'
  prefix='PbTe'
  pseudo_dir='./'
  outdir='./tmp'
  calculation = 'scf'
 /

 &system
  ibrav=0
  nat=    2
  ntyp=   2
  ecutwfc = 35.0, ecutrho = 140.0  ! ecutrho (4-12*ecutwfc)!
  occupations='fixed'  ! 'tetrahedra'|'smearing'|'fixed'|
  nbnd= 16                  ! number of bands, need special

!+++++ DFT Hybride functional +++++
 input_dft='hse', nqx1 = 1, nqx2 = 1, nqx3 = 1
 exxdiv_treatment='g-b' ! yukawa,erfc_simple,vcut_ws,vcut_spheric,none
 x_gamma_extrapolation = .true.  ! if cut_ws or vcut_spheric, = .false.
 !exx_fraction =0.25, screening_parameter=0.16  !see Modules/func.f90
 lspinorb=.true.
 noncolin=.true.

 /

 &electrons
  diagonalization='david'      ! 'cg'|'-ndiag N'
  electron_maxstep=300
  conv_thr = 1.0e-8
  mixing_mode='plain'          ! 'TF'|'local-TF'
  mixing_beta = 0.5
  mixing_ndim=8
 adaptive_thr=.true.          ! use for EXX
 conv_thr_init=1.d-3          ! use for EXX
 conv_thr_multi=1.d-1         ! use for EXX

 /

ATOMIC_SPECIES
Pb  207.200  Pb.rel-pbe-nc.UPF
Te  127.600  Te.rel-pbe-nc.UPF
ATOMIC_POSITIONS {crystal}
Pb    0.0000000000000   0.0000000000000   0.0000000000000
Te    0.5000000000000   0.5000000000000   0.0000000000000
K_POINTS {automatic}   !gamma,crystal_b
  4  4  4 0 0 0
CELL_PARAMETERS {cubic}
   6.104762563       6.104762563       0.000000000
   6.104762563      -6.104762563       0.000000000
   6.104762563       0.000000000       6.104762563

  Best regards,
Longhua Li
2011-08-23 				
----------------------------------------
foursea at 163.com





More information about the users mailing list