Dear Admin,
I have been having a strange problem recently. I compressed a initially Hexagonal lattice with PSTRESS. I see that I get the following warning
VERY BAD NEWS! internal error in subroutine IBZKPT:
Reciprocal lattice and k-lattice belong to different class of lattices. 168
I turned the symmetry off by selecting ISYM = 0
the lattice vectors in the POSCAR file are as follows
16.0429608991080400 0.0001624393306630 0.0000675493164954
8.0216215016549270 13.8946440331429200 0.0004603463756877
0.0000185063212862 0.0001409205690510 4.1841738261691970
my KPOINTS file looks like
Automatic mesh generation
0
Monkhorst
1 1 4
0 0 0
Should I overlook this warning? or is it that something is wrong with my calculations?
I shall be thankful for your response.
error in IBZKPT
Moderators: Global Moderator, Moderator
-
TAT
- Newbie
- Posts: 39
- Joined: Thu Aug 02, 2007 7:28 pm
- License Nr.: 691 5-1
error in IBZKPT
Last edited by TAT on Tue Mar 10, 2009 3:25 am, edited 1 time in total.
-
admin
- Administrator
- Posts: 2921
- Joined: Tue Aug 03, 2004 8:18 am
- License Nr.: 458
error in IBZKPT
the Bravais matrix of the lattice after compression is not hexagonal anymore but triclinic (at least if you just put one atom in 0 0 0, which would not lead to any further symmetry reduction), nevertheless such a simplified input does not give the symmetrization error with a 1x1x4 mesh.
Therefore proably the atomic positions reduce the symmetry even further ("168" indicates that for real and reciprocal space space a triclinic and a simple lattice (without any symmetry) are found by the symmetry analysis and symmetrization routines.
there are 2 ways to proceed:
please check if the results of the compression run really hold (have you increased ENCUT by at least 1.3* max(ENMAX), set PREC=Accurate and have a look if all steps are fully converged electronically?. If so and the structure really changes from hexagonal to triclinic, please increase SYMPREC to a higher precision than 1e-5 or reduce the number of digits in the Bravais martix (and probably the atomic positions) to a reasonable number. I am not sure whether it makes much sense to give 16 digits (10-16Ã… precision ) for these numbers, even in a high quality calculation. Please eg. check up to which one the results remained constant during the last steps.
If you want to keep the input unchanged, please ignore the symmetry warning.
Therefore proably the atomic positions reduce the symmetry even further ("168" indicates that for real and reciprocal space space a triclinic and a simple lattice (without any symmetry) are found by the symmetry analysis and symmetrization routines.
there are 2 ways to proceed:
please check if the results of the compression run really hold (have you increased ENCUT by at least 1.3* max(ENMAX), set PREC=Accurate and have a look if all steps are fully converged electronically?. If so and the structure really changes from hexagonal to triclinic, please increase SYMPREC to a higher precision than 1e-5 or reduce the number of digits in the Bravais martix (and probably the atomic positions) to a reasonable number. I am not sure whether it makes much sense to give 16 digits (10-16Ã… precision ) for these numbers, even in a high quality calculation. Please eg. check up to which one the results remained constant during the last steps.
If you want to keep the input unchanged, please ignore the symmetry warning.
Last edited by admin on Tue Mar 10, 2009 12:33 pm, edited 1 time in total.