Page 1 of 1

optimization of iron slab(110)

Posted: Wed Mar 17, 2021 3:41 pm
by rachita_panigrahi
i am optimizing iron slab(110) suface in vasp having 234 atoms.. i have doubt on kpoints.can you just help me

Re: optimization of iron slab(110)

Posted: Thu Mar 18, 2021 11:46 am
by merzuk.kaltak
A good choice for the KPOINTS file depends on the used POSCAR, specifically the length of the lattice vectors a,b,c defined in the POSCAR.
If N1, N2, N3 are the number of k-points in the first, second and third direction in the first Brillouin cell, like for instance in this KPOINTS file

Code: Select all

Gamma-centered KPOINTS
0
Gamma
N1 N2 N3
The I would set N1 and N2 and N3 such that N1 * a ~ N2 * b ~ N3 * c, where a, b, c is the length of the lattice vectors defined in the POSCAR.
If you have some tricky POSCAR it is helpful to visualize it in VESTA to find the lengths of the vectors.
For tight convergence you can multiply N1,N2,N3 with a common factor.
Here is an example of a POSCAR that is stretched

Code: Select all

Example cubic
1
4 0 0 
0 6 0
0 0 12
Since c is the largest vector set N3=1. Because a is 3 times smaller than c and b is in between a and b, I would use N1=3, N2=2.
If you use N3=2, then you should double N1 and N2 as well and use N1=6, N2=4 and so on.

Re: optimization of iron slab(110)

Posted: Thu Mar 18, 2021 5:22 pm
by rachita_panigrahi
here i am attaching the lattice vector

Code: Select all

 30.1232986450         0.0000000000         0.0000000000
        0.0000000000        30.1232986450         0.0000000000
        0.0000000000         0.0000000000        30.1232986450
and i am taking KPOINTS
1 1 1
is it correct?

Re: optimization of iron slab(110)

Posted: Thu Mar 25, 2021 10:38 am
by merzuk.kaltak
The choice 111 seems reasonable.