Dear Lily,
this depends not only on the number of atoms, but also on your system.
Do you want to calculate a slab, a bulk or a molecule and the elements
are also important.
Will it be a metallic system or semiconducting ?
A good idea is to vary your k-grid parameter and see if it converges.
Range for k-points per axis can be between 1 and 500 (at least that was
the highest I had so far for density of states calculations of
"pseudo-metallic" systems).
Also consider the accuracy you need/the question you want to answer with
your calculation.
Maybe you don't need a too good k-point sampling, as you are not
interested in bandstructure and density of states. Therefore an average
sampling might be sufficient.
Best wishes
Heiko
On 07/07/11 23:20, lily zheng wrote:
Thank you so much for your explanation,Roland!
Usually what is the thumb of rule to get Monkhorst matrix? for
instance, if I only calculate 2 atoms system? while if I calculate
1000 atoms system, what the monkhorst matrix should be like?
Many thanks
Lily
--
Heiko Dumlich
Freie Universität Berlin
Fachbereich Physik - AG Reich
Arnimallee 14
D-14195 Berlin
GERMANY
Phone: +49-30-838 56157
