Page 1 of 1
110 kpoints
Posted: Thu Jun 27, 2013 8:27 pm
by ccccc
Dear all
I want to calculate band structure of a FCC (1x1)(110) surface. How should i decide the kpoints of this surface? I saw different things on internet so I am confused about kpoints. The 110 surface is a rectangular. So is it OK if i use these points?
0 0 0
0.5 0 0
0.5 0.5 0
0 0.5 0
Thank you
110 kpoints
Posted: Wed Jul 03, 2013 5:23 pm
by lumacido
First: Have you think in automatic generation?
File:
Automatic Mesh
0
Automatic
25
That use to be acceptable. However, if you really want to use it you should check if your choice is correct: increasing the number of them (and check the convergence in your desire property).
Second: Instead of setting coordinates, it is easier to set the number of them.
Third: You should use just one in the direction perpendicular to the surface and more points in the shorter direction.
You have set 2 2 1 for x, y and z directions, respectively. If the cell is rectangular, maybe will be better 3 2 1 or 4 2 1 (if x is smaller than y).
Sooo, I suggest you to use this as KPOINTS file:
K-Points
0
Gamma centered
3 2 1
0 0 0
Check energy (or another property you are interest in) and repeat with 4 2 1 or 3 3 1, 6 4 1 or 5 5 1, etc. Until you are satisfied with the solution.
(I hope you have optimized you sistem before, have you?).
<span class='smallblacktext'>[ Edited Wed Jul 03 2013, 05:33PM ]</span>
110 kpoints
Posted: Wed Jul 03, 2013 5:47 pm
by ccccc
Thank you for your answer dear lumacido. i think the kpoints you suggested can be used for relaxation of the surfaces. But as far as i know, i need to use specific kpoint coordinates to find a surface's band structure. For example 111 surface needs to set Gamma - M - K points and they have different coordinates.