[gpaw-users] k-point sampling

Thomas Olsen tolsen at fysik.dtu.dk
Thu Aug 30 19:49:10 CEST 2018


Hi

The GW calculation needs a uniform k-point grid, so you cannot just add a k-point. But you can shift the the grid such that it "hits" the K point.
Alternatively you can use a 9x9 grid or a Gamma-centered 12x12 grid.

/Thomas
________________________________________
Fra: gpaw-users-bounces at listserv.fysik.dtu.dk <gpaw-users-bounces at listserv.fysik.dtu.dk> på vegne af Kangli Wang via gpaw-users <gpaw-users at listserv.fysik.dtu.dk>
Sendt: 30. august 2018 16:40
Til: gpaw-users at listserv.fysik.dtu.dk
Emne: [gpaw-users] k-point sampling

Dear gpaw user,
  I have a question about adding high sysmmtry point to k-point sampling
fo r G0W0 calculation. We know for MoS2, 10*10*1 k-point sampling does
not include (1/3, 1/3,0) for G0W0 canculation. So I want to add this
point to 10*10*1 sampling. This is the script:
***
from gpaw.response.g0w0 import G0W0
from ase.build import mx2
from gpaw import GPAW, PW, FermiDirac
from ase.dft.kpoints import monkhorst_pack
import numpy as np

structure = mx2(formula='MoS2', kind='2H', a=3.184, thickness=3.127,
                size=(1, 1, 1), vacuum=3.5)
structure.pbc = (1, 1, 1)

Ecut = 400
kpts = monkhorst_pack([10 , 10, 1])
k = np.array([1./3, 1./3, 0])
kpts = np.vstack((kpts, k))

calc = GPAW(mode=PW(Ecut),
            xc='PBE',
            basis='dzp',
            kpts=kpts,
            occupations=FermiDirac(0.01),
            txt='MoS2_out_gs.txt')

structure.set_calculator(calc)
structure.get_potential_energy()
calc.write('MoS2_gs.gpw', 'all')

calc.diagonalize_full_hamiltonian()
calc.write('MoS2_fulldiag.gpw', 'all')

for ecut in [80]:
    gw = G0W0(calc='MoS2_fulldiag.gpw',
              bands=(8, 18),
              ecut=ecut,
              truncation='2D',
              nblocksmax=True,
              q0_correction=True,
              filename='MoS2_g0w0_{}'.format(ecut),
              savepckl=True)

    gw.calculate()
***


  But it always show wrong information:

***
GPAW CLEANUP (node 18): <type 'exceptions.AssertionError'> occurred.
Calling MPI_Abort!
rank=15 L11:   File
"/sw/chem/gpaw/1.4.0/mpp1/lib/python2.7/site-packages/gpaw/response/g0w0.py",
line 350, in __init__
rank=15 L12:     assert -1 not in kd.bz2bz_ks
rank=15 L13: AssertionError
***

  Could you give me some suggestions or help besides using a larger
k-points? Thank you in advance!
  All the best.
  Kangli

_______________________________________________
gpaw-users mailing list
gpaw-users at listserv.fysik.dtu.dk
https://listserv.fysik.dtu.dk/mailman/listinfo/gpaw-users



More information about the gpaw-users mailing list