ABINIT Discussion Forums

Hi there

I'm a novice to Abinit, but I'm trying to learn how to do GW calculations on some novel silicon structures (Still FCC though). I'd like to calculate the GW correction at points other than the Gamma point, and I've gone through the tutorials but am a little confused. I'm getting caught up on specifying the k-grid needed to capture the point, and have been running into all sorts of errors. I even got a lovely one asking me to contact the Abinit group.

I believe the trouble comes down to that my grid does not contain the k-points that I want to test. I can capture them if I use a large enough mesh, but my primitive cell is about 34 atoms, so I really don't want to use an extravagantly large mesh. Is there a way to still use a small mesh but capture the k-points I'd like to measure? (L,W,K,Gamma, and X?)

Thanks!
Lewis Patterson

Hi Lewis,

As far as I know, there is no way to avoid large grids to get the desired critical points.
First, to calculate GW corrections on a k-point, you need to have it in your grid. Then, the reason you've been experiencing all kinds of bug, is that the k-point grid need to be regular, and to comply with the symmetries of your system.

Good luck with the rest!