For 500 Ry I get zero forces on atoms for bulk
GaAs, whereas for 300 Ry I still have some not vanishing values.
That's a bit surprising for me. Usually forces should vanish around
200-250Ry. So I'd rather recheck whether you have a good k-grid.
Soft confinement of orbitals could also help a lot, especially since you
need very good convergence in forces.
Here is my optimized GaAs basis:
%block PAO.Basis
As 3 -0.195
n=4 0 2 E 73.28644 5.7
6.879 4.278
n=4 1 2 E 44.09132 6.25442
8.04645 5.19876
n=4 2 1 E 97.53993 1.03121
7.50695
Ga 3
n=4 0 2 E 80.12335 5.74628
7.16526 4.82650
n=4 1 2 E 87.59084 7.11605
7.61660 4.41252
n=4 2 1 E 93.09519 0.13086
6.26045
%endblock PAO.Basis
Or you could try to let SIESTA automatically soften the orbitals as
implemented in SIESTA 3 (PAO.SoftDefault).
I want to study a different configuration of one pair of Mn ions on the
surface of GaAs. I want to see the difference in the energy , electronic
structure,
This is more a conceptual problem. How the system is prepared
experimentally? Is it an MBE deposition of atoms, or surface prepared by
annealing in vacuum, where the Mn atoms come from?
Depending on the preparation procedure the final geometry can be very
different.
You need to know that you are modeling the right thing, otherwise you'll
spend ages of time on getting very accurate results on the system which
is wrong from the very beginning.
What is the surface orientation? Is it (100)?
What is the unit cell size that you use? Is it (2x4)?
What is the initial surface reconstruction? GaAs does not exist
unreconstructed. It would take a lot of time for CG procedure just to
relax the surface if you don't provide the right geometry.
