ABINIT Discussion Forums

Dear all,

I am doing a supercell calculation of MoS2 bulk and I have made a 2 x 2 x 1 supercell by translating. (input file attached)
As I am using a supercell my bandstructure has to be unfolded back into 1st BZ using fold2Bloch. I do as said in the tutorial of hydrogen atom on the abinit webpage. I am using abipy to plot the unfolded bandstructure.
Here are my errors and questions:
a) I include SOC in my ground state calculations and hence when i use abipy to unfold the bands after fold2Bloch I get the following error

IndexError: index 1 is out of bounds for axis 0 with size 1 abipy

which points to the following line

ys = self.uf_eigens[spin, p.ikfound, :] - e0

of the code. So my question, can I do unfolding of the bands when SOC is included in ground state calculations?

b) I switch off the SOC and do the calculation again, then I can see a bandstructure appearing from abipy. In this regard also I am facing a problem in deciding the k-path. I read that if I want GK path then I need to include -KG path as well in the calculations. I did as suggested in the tutorial but I do not get the full path from GK even after including the -KG path.
I want the bandstructure along GKMG path of the hexagonal Brillouin zone. For doing so I use the following kbounds

kptbounds2 -2/3 -1/3 0.0 # -K point
0.0 0.0 0.0 # Gamma point
2/3 1/3 0.0 # K point
1/2 0.0 0.0 # M point
0.0 0.0 0.0 # Gamma point
-1/2 0.0 0.0 # -M point

I am attaching the bandstructure as obtained from the above path.
Apart from this kbounds I also tried various other combinations like -MKM to get the KM path but no success till now. Any help will be appreciated.

Regards,
nsingh3

Hello developers,

With some hit and try I was able to get the Brillouin zone path sorted out but I can unfold the bands when spin-orbit coupling is turned on. Is that feature not implemented in abipy?

Also my conduction bands are not clearly visible. Any idea how do I improve those? File attached of 2:2:1 supercell.

nsingh3 wrote: ↑

Sun Jan 24, 2021 4:15 am

Hello developers,

With some hit and try I was able to get the Brillouin zone path sorted out but I can unfold the bands when spin-orbit coupling is turned on. Is that feature not implemented in abipy?

Also my conduction bands are not clearly visible. Any idea how do I improve those? File attached of 2:2:1 supercell.

Dear nsingh3,
I have a problem in deciding the k-path of MoS2 using fold2Bloch. Can you send me the successful cases of MoS2? Thank you very much. My Email is gswylq@gmail.com.