9
$\begingroup$

I use VASP to do my DFT calculation. I have to do a self consistent calculation to generate my CHGCAR. I want to ask is there anyway to write the KPOINTS file to generate K-meshing grid which can exactly lying in $(\frac{\pi}{a} , \frac{\pi}{a}, \frac{\pi}{a})$ point in square Brillouin zone boundary? As I used Monkhost-Pack grid, which can only guarantee that your point grid can be located at Gamma point.

$\endgroup$
6
  • $\begingroup$ In principle it is possible to add the k-points that you want explicitly in VASP, is this what you are asking, or are you asking more generally about sampling k-point grids (irrespective of VASP)? $\endgroup$
    – ProfM
    May 9 at 7:37
  • $\begingroup$ @ProfM Yes. But when I use this method It will give an error as follow: "error in ibzkpt_hf: two k-point are equivalent". But I have check that no K-point are equivalent in my file except the TRIM in boundary which connected by reciprocal lattice vector and differ by a sign. $\endgroup$
    – JensenPang
    May 9 at 10:19
  • $\begingroup$ I would say that two points separated by a reciprocal lattice vector are equivalent; why do you want to include both? $\endgroup$
    – ProfM
    May 9 at 10:36
  • $\begingroup$ @ProfM Because I turned off the symmetry TAG in the INCAR file, and the KPOINT grid should be symmetry in the Brillouin zone. I try to fit run VASP2Wannier90 that's why I want to include the high symmetry point in my file. $\endgroup$
    – JensenPang
    May 9 at 10:43
  • 2
    $\begingroup$ If you turn off the symmetry with ISYM = -1, the point group symmetry found for your structure is not imposed on the Kohn–Sham orbitals. However, there is no way to turn off periodic boundary conditions in VASP! So $\vec{k}$ points separated by reciprocal lattice vectors are always equivalent. $\endgroup$
    – mt.huebsch
    May 22 at 1:28

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.