Strange Band-structure of ZrSe2

[naomi]

Hello Guys,

I have tried to get bandstructure of ZrSe2. Something strange happened (Please look at image precisely, you will see VBM is higher than CBM).
could you help me to get right bandstructure with 0.498eV bandgap. 

I attached images and Python file.
I simulated two times with different value, however, both simulations did with GGA/PAW/PBE
One simulation run with 150 Hartree and 89 k Sampling. the other one with 75 Hartree 41 K sampling. If you check both images you will see similar results

If you need extra information, Please let me know

Thanks in advanced

[Petr Khomyakov]

Is there any particular reason that you have adopted a triclinic unit cell for ZrSe2? Have you optimized your structure?

Another thing is that using the GGA or LDA density functional might not allow you to calculate the band gap with the required accuracy, see J. Phys. Chem. C 119, 13169 (2015).

An additional note, I do not think you need that many k-points (71x71x1) to sample the Brillouin zone for self-consistent calculations. Also you may try using a more accurate pseudopotential/basis set such as SG15/Medium.