The VASP Guide says in
http://cms.mpi.univie.ac.at/vasp/vasp/node205.html
the following:
Code: Select all
There are some rules of thumb, which you should check
whenever making a calculation: For bulk materials the
number of plane waves per atom should be between 50-100.
A smaller basis set might result in serious errors. A larger basis
set is rarely necessary, and is a hint for a badly optimized
pseudopotential. If a large vacuum is included the number of
plane waves will be larger (i.e. 50% of your supercell
vacuum -> number of plane waves increases by a factor of 2).80 percent (using RWIGS in POTCAR to calculate volume occupied
by the atoms). The default settings for ENCUT is 400, which gives
a number of plane waves of around 244 (I averaged over the k-points).
This means 244/11 is approximately 22 plane waves/atom.
According to the above 'rule of thumb' this is far from accurate.
Is that the right conclusion?
And what is the factor when vacuum is 80 percent?
Furthermore, I have calculated the total energy of this system as
function of ENCUT, assuming that a larger ENCUT will use a larger
number of plane waves, which then should result in a more accurate
total energy. However, I get the following data:
http://surfion.snu.ac.kr/~lahaye/vasp/encut.jpg
The upper panel shows how the total energy varies with ENCUT,
and the lower panel shows the average number of plane waves
as function of ENCUT.
I expected that the total energy would gradually converge as
ENCUT is larger. However, it has a maximum first around
ENCUT=500 before convergence starts. I find this very strange!!!
By the way, when not giving ENCUT input explicity in INCAR, the
default is 400 (from POTCAR), which is exactly the ENCUT that is
equal to the converged total energy (at ENCUT > 800) before going
to the maximum. Is this where the VASP pseudo potential has made
a magical and very clever choice?
I hope somebody can help me understand better what is happening
here.
Thank you,
Rob.