Page 1 of 1

Hybrid HSE06 KPOINTS SYMMETRY

Posted: Tue Nov 03, 2015 6:32 am
by 5-2274
Hello admin,

I am trying to do calculate the bandstructure of CdS. When I do not use any hybrid functional, vasp uses symmetry to reduce the number of kpoints and uses only the irreducible Briilioun zone points. However, with the same INCAR and KPOINTS file and just adding the hybrid flags causes VASP not to use any symmetry in kpoints at all. I am wondering if this is normal or is there something that I am doing wrong?
Thank for your help in advance.

INCAR:
mytitle
IBRION = -1
ENCUT = 275
ISMEAR = 0
SIGMA = 0.01
PREC = Accurate
NSW = 0
NELMIN = 1
ALGO = All
TIME = 0.4
LSORBIT = .TRUE
LHFCALC = .TRUE.
HFSCREEN = 0.2
PREFOCK = Fast

Re: Hybrid HSE06 KPOINTS SYMMETRY

Posted: Tue Nov 03, 2015 11:06 am
by admin
All correct.
In the calculation of the non-local Fock exchange energy (Eq. 6.59)
http://cms.mpi.univie.ac.at/vasp/vasp/I ... _Fock.html
the sum over k runs over all k-points in the BZ.
Therefore in your calculation you observe two sets of k-points.
The irreducible set produced by IBZKPT
and the full-kpoint grid produced by IBZKPT_HF.

Re: Hybrid HSE06 KPOINTS SYMMETRY

Posted: Tue Nov 10, 2015 9:48 pm
by 5-2274
Hello admin,

Thanks for the reply.
But, irreducible set produced by IBZKPT should be symmetry reduced?
In my case, no symmetry operations are used for both IBZKPT and IBZKPT_HF meshes.
For instance, for a wurtzwite structure with 8*8*6 kpoints grid generated using Gamma automatic mesh with 0 0 0 shift, both IBZKPT and IBZKPT_HF reports 384 kpoints?
Am I doing something wrong?