Error in Gw calculations (wrong band lines)

[mjakhar7665]

Hello sir,
I am new at Yambo
I was trying to calculate the gw band gap with following steps as mentioned,
1. i sumbit the run with yambo.in file (attached file).
2 For post-processing i used three command
./ypp -s b (i don't change anything in file)
./ypp -y (i uncomment the RmTimeRev)
./ypp -s b -V qp (i did chang some parameter (attached file))
After following these steps i get the o.bands.interpolate file.
When i plot the bands file then there will be rounding shape in band lines (attached image).
Please check my files and kindly suggest me the solution.
One more thing i want to ask how to get the fermi level and vaccum level.
Thanks in advance

[Daniele Varsano]

Dear Mukesh,
I had a look at your input file and there are several problems:

1) Coulomb cutoff: you have a cell of 36 au side perpendicular to the system and you are using a cut-off box of 17 au. In this way you removing interaction
also among electrons in the system, Zcut should be just slightly smaller than the cell size (e.g. 35.5).
2) VXCRLvcs= 6 Ry ! This is very dangerous as your density is not well evaluated, even worst the gradient so the <Vxc> term will be very different from the one calculated by QE. The safe value for VXCRLvcs is the energy cutoff used for the density in QE, i.e. 4 times ecutwfc
3) NGsBlkXp= 1 RL ! Here you are considering the screening as a scalar quantity, this is a parameter that should be converged.

My impression is that your QP corrections may be unphysical. Please have a look to some GW tutorial you can find in the yambo wiki website e.g. here:
http://www.yambo-code.org/wiki/index.ph ... rial:_h-BN

and try to interpolate once you have converged results.
Interpolator can fails, you can try to interpolate KS bands structure first. If there are problems you can try to use a different algorithms:

Code: Select all

INTERP_mode= "BOLTZ"

Best,
Daniele