[QE-users] Issues with band convergence in a magnetic system

[Rishi Bhandia]

Hello,

I’ve been trying to conduct a spin polarized, SOC, band calculation for the fcc Heusler Co2TiGe and am then planning on using these results to compute the Berry curvature using wannier90. However, while I have been able to be successfully run the scf calculation just fine, I’ve been having issues with achieving convergence in the band calculations.

The error message I get is:
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
     Error in routine c_bands (1):
     too many bands are not converged
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

I’ve checked the structure that I use multiple times using various visualization software. The magnetization per unit cell I’m getting is close with what has been obtained in experimental measurements, however, I’m unclear as to what parameters I should adjust to achieve convergence in the non-sc part of the calculations.

Here is my input file. The cutoff parameters, magnetization and electronic parameters were chosen by use of the
https://materialscloud.org/work/tools/qeinputgenerator tool:

&CONTROL
  calculation = 'bands'
  outdir = './'
  prefix = 'Co2TiGe'
  pseudo_dir = './'
  verbosity = 'high'
/
&SYSTEM
  degauss =   1.4699723600d-02
  ecutrho =   1.0800000000d+03
  ecutwfc =   9.0000000000d+01
  ibrav = 2
  nat = 4
  nbnd = 10
  lspinorb = .true.
  noncolin = .true.
  nspin = 4
  ntyp = 3
  celldm(1) = 11.0171
  occupations = 'smearing'
  smearing = 'cold'
  starting_magnetization(1) =   2.9411764706d-01
  starting_magnetization(2) =   1.0000000000d-01
  starting_magnetization(3) =   4.1666666667d-01
/
&ELECTRONS
  conv_thr =   3.2000000000d-09
  electron_maxstep = 80
  mixing_beta =   4.0000000000d-01
/
ATOMIC_SPECIES
Co     58.9332 Co.rel-pbesol-spn-rrkjus_psl.0.3.1.UPF
Ge     72.61 Ge.rel-pbesol-dn-rrkjus_psl.1.0.0.UPF
Ti     47.88 Ti.rel-pbesol-spn-rrkjus_psl.1.0.0.UPF
ATOMIC_POSITIONS crystal
Ti           0.0000000000    0.00000000000.0000000000
Co           -0.2500000000     0.7500000000 -0.2500000000
Co           -0.7500000000     2.2500000000 -0.7500000000
Ge           0.5000000000 0.5000000000     -0.5000000000
K_POINTS {crystal_b}
6
  0.50  0.00  0.50  30 !X
  0.00  0.00  0.00  30 !G
  0.50  0.50  0.50  30 !L
  0.50  0.25  0.75  30 !W
  0.375 0.375 0.75  30 !K
  0.00  0.00  0.00  30 !G

Thanks,
Rishi Bhandia
PhD student
Johns Hopkins University

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20191217/c4f8207c/attachment.html>