Compute KPOINTS and Kpath

Topics here may be considered for their own forums if appropriate.

Compute KPOINTS and Kpath

Postby sandhu407 » Sun Sep 24, 2017 1:07 am

Hello ,

I am a new user of aflow and in this field i have few basic questions for KPOINTS and its path

1) Compute KPOINTS with KPPRA= (>1)
What does KPPRA means and what should be its values if we want to find the KPOINTS mesh with resolution density of 2pix0.05 A.

2) Computer KPOINTS with dK=(<1.0)
what does mean by "dk" ? i have tried its different value but the answer is always as; Result ******************

3) Kpath in the reciprocal space for band structure calculations,

Can we use this kpath to calculate the band-structure of any given supercell poscar in the input. If there are many points , can we chose any close loop of the kpoint path?

These are very basic questions for an expert, looking for your reply.

Posts: 1
Joined: Sun Sep 24, 2017 12:49 am

Re: Compute KPOINTS and Kpath

Postby coreyoses » Tue Oct 10, 2017 5:26 pm

Hi there,

Apologies for the long delay. Some code maintenance was required.

To answer your questions:
1. KPPRA stands for k-points per reciprocal atom. We describe the routine in our Standards paper: doi=10.1016/j.commatsci.2015.07.019. Input KPPRA (6,000-10,000 recommended) is the lower bound of the product of the (total count of k-points in the 3-D grid) and (number of atoms in the cell). The smallest mesh to satisfy this requirement is calculated. Effectively, cells with more atoms will yield smaller grid meshes.
2. The function required some maintenance, but it is fixed and will be available in the new release (keep an eye out for updates under "AFLOW Version History"). Input dK is the desired smallest distance between grid points in reciprocal space. The smallest mesh to satisfy this requirement is calculated.
3. No, the kpath determined is for the input structure (and uniform supercells) alone. A non-uniform expansion of the input is considered a derivative structure and may have a lower symmetry, and thus a different kpath. Furthermore, the kpaths provided are constructed from the irreducible part of the first Brillouin zone, with additional consideration for other zero-slope points. A full description is provided here: doi=10.1016/j.commatsci.2010.05.010. A closed loop of high-symmetry k-points is not enough.
Corey Oses
AFLOW Developer
Posts: 49
Joined: Wed May 31, 2017 2:23 pm

Return to Miscellaneous topics

Who is online

Users browsing this forum: No registered users and 1 guest