Graphyne

[NW]

Hello,
I have recently calculated the bandstrucutre, DOS and transmission of a Graphyne by DFT-1/2 method. The output results show a 0.2eV band gap in DOS and transmission but in bandsturture  I got 0.77eV! Why the values of the band gap are not equal?  I attached the results.

[Petr Khomyakov]

It might be that the (in)direct band gap is at some other k-point(s), not the ones you used for plotting the band structure.

[NW]

Thanks Petr.
So how can I fix it? I used different k-points such as 1, 17, 15 or 1, 18, 16 or 1, 23, 21 but the output band structure did not change.

[Petr Khomyakov]

It is not about the number of k-points, but rather the k-point routes in the Brillouin zone that were used to compute the band structure. Could you post the script related to this calculation?

[NW]

Yes for sure.

[Petr Khomyakov]

You should have set the lattice type of your structure to Simple Monoclinic in the Builder: Bulk Tools->Lattice Parameter, and then use the Band Structure analysis object that would set a correct path corresponding to the lattice type of your 2D crystal. In your script, no specific lattice type was set.

The second thing is that you have not used  a sufficient number of k-points for DOS calculations; it should be increased significantly, see Appendix in Phys. Rev. B 87, 075414 (2013) for clarification of the issue. In fact, the band gap of your system is vanishing, i.e., close to zero, similarly to graphene.

Enclosed you can find the script with the corrections I have mentioned. 

[NW]

Thanks again Petr for useful advice. I have learned a lot of new things in this simple example.

Just  I want to know why did you choose Simple Monoclinic?

[Petr Khomyakov]

Just  I want to know why did you choose Simple Monoclinic?

This is because your structure corresponds to this Bravais lattice type, judging from the lattice vectors specified in your python script (which can also be seen in the Builder->Bulk Tools-> Lattice parameters), see the following paper describing all the Bravais lattice types (e.g., their lattice vectors) and related Brillouin zones (BZs) and special points in the BZ,   Computational Materials Science 49 (2010) 299–312.