Dear Siesta users:
I now have a puzzle on the k point sampling.
Before, I usually use a cubic in real space,
for example:
%block LatticeVectors
a 0.0 0.0
0.0 b 0.0
0.0 0.0 c
%endblock LatticeVectors
So its corresponding reciprocal space is also a cubic:
a' 0.0 0.0
0.0 b' 0.0
0.0 0.0 c'
when I use the k sampling for this kind cell
%block kgrid_Monkhorst_Pack
4 0 0 0.0
0 4 0 0.0
0 0 1 0.0
%endblock kgrid_Monkhorst_Pack
I think it means 4, 4 and 1 k point will be used along a', b' and c' will be
used. (Am I right?)
But if I use a non-cubic cell,
for example:
%block LatticeVectors
a1 a2 0.0
b1 b2 0.0
0.0 0.0 c
%endblock LatticeVectors
so the corresponding reciprocal space is also a non-cubic:
maybe like the following:
a1' a2' a3' (say Vector1)
b1' b2' b3' (say Vector2)
c1' c2' c3' (say Vector3)
If I also use a K point sampling like:
%block kgrid_Monkhorst_Pack
4 0 0 0.0
0 4 0 0.0
0 0 1 0.0
%endblock kgrid_Monkhorst_Pack
This time, are 4, 4 and 1 the k points along the reciprocal Vector1, Vector2
and Vector3 respectively?
Maybe, my discription could make someone confused.
In other words, is the block kgrid represented in real space lattice vectors or
in reciprocal space vectors or just in cartesian coordinates?
Any reply will be appreciated.
Best
Guangping
2011-08-03
Guangping Zhang
