hBN QP bandstructure

Submitted by kanak on Thu, 03/02/2017 - 08:12

Dear researchers

I have been running the hBN code in BGW. The code was posted in a workshop. There was no reference to check the results.
I have got the following bandstructure for monolayer hBN:
As you can see, the data has been extracted from the entire Brillouin zone. As the final interpolation step in inteqp, I used K point grid from entire Brillouin zone. The simulated valance band closely matches with the simulation reported in (please refer to fig 5(b)):
The conduction band energy levels, however, does not match. Is this because the k grid is too coarse for sigma simulation? Or this is because I have tried single shot GW i.e. G0W0 only?

Has anyone in this forum tried this before? Please let me know.

Submitted by babarker on Sat, 03/04/2017 - 13:27

Hello Kanak,

The choices for most of the paramters used in the workshop example were picked so that users could complete a large number of calculations in a short period of time. The number of empty states, the screened coulomb cutoff energy, and other parameters need to be carefully converged to achieve physical results. Most papers usually have discussion of their most important parameters either in the text or in the supplementary material.