Page 1 of 1

Kpoints in nscf calculation

Posted: Sun Nov 27, 2016 3:08 pm
by Wangwei Lan
Dear all,

I am very new in quantum espresso and EPW. Recently, I want to calculate electron-phonon coupling to determine the Tc. However, I have a very general question regarding running nscf calculation. I've searched both QE and EPW forum but can't get an answer. I hope you can help me.

In the example, GaN, I found in the nscf_epw.in file, the kpoints are like this:

K_POINTS crystal
8
0.0 0.0 0.0 0.125
0.5 0.0 0.0 0.125
0.0 0.5 0.0 0.125
0.0 0.0 0.5 0.125
0.5 0.5 0.0 0.125
0.5 0.0 0.5 0.125
0.0 0.5 0.5 0.125
0.5 0.5 0.5 0.125

My general question is that how do you get the kpoints here? In the example, the number is "8", I notice this is the same number as the number of directories in gan.save directory, which are K00001,...K00008, so is there any relations between these two?

Thanks very much.

Wangwei Lan
Florida State University

Re: Kpoints in nscf calculation

Posted: Mon Nov 28, 2016 3:00 pm
by sponce
Dear Wangwei Lan,

The nscf grid should be an homogeneous grid between 0 and 1 that corresponds to your coarse initial grid in EPW.

A 2x2x2 k-point grid will therefore indeed give
8
0.0 0.0 0.0 0.125
0.5 0.0 0.0 0.125
0.0 0.5 0.0 0.125
0.0 0.0 0.5 0.125
0.5 0.5 0.0 0.125
0.5 0.0 0.5 0.125
0.0 0.5 0.5 0.125
0.5 0.5 0.5 0.125

The k-points for the nscf needs to be generate by hand (you can make a small script to do this).

Best,

Samuel

Re: Kpoints in nscf calculation

Posted: Mon Dec 05, 2016 4:53 pm
by Nandan
You could also use:
wannier90-2.0.1/utility/kmesh.pl 2 2 2

nandan.