The ABINIT Users Mailing List ( CLOSED )

[verstraete@pcpm.ucl.ac.be]

OK: the answer is written in the log file:
> tetrahedron: problem with a composite k-point grid.
>  Only simple lattices are supported. Action : use nshiftk=1.
> tetrahedron: skip subroutine.

Right now tetrahedron only accepts uniform kpoint grids. This means only
1 shiftk. Normally abinit automatically generates a non-orthogonal
kptrlatt with only 1 shiftk from any kptrlatt or ngkpt + shiftks you
input.

HOWEVER: this supposes you have given a set kptrlatt+shiftk which is
compatible with the system's symmetry. The grid you gave was a FCC grid,
and your system is rhombohedral. According to the automatic tests in
abinit, the optimal grids for rhombohedral are

kptrlatt n 0 0   0 n 0   0 0 n
nshiftk 1
shiftk 0.5 0.5 0.5

Since you added nshiftk 4, the grid cannot be reduced to a uniform grid...

I insist on this point because I imagine it will come up again with other
people. You can generalize the tetrahedron method, but it's not very
useful since we always use uniform MP grids.

Matthieu

-- 
===================================================================
Matthieu Verstraete               mailto:verstraete@pcpm.ucl.ac.be 
PCPM, Boltzmann, pl. Croix du Sud, 1            tel: 010/ 47 86 81
B-1348 Louvain-la-Neuve  Belgium                fax: 010/ 47 34 52