SHG in semimetal system

[Belle.X]

Dear Claudio,

Hi, I would like to simulate the SHG properties of WTe2 which is a kind of Type II Weyl Semimetals, and it will be treated as a metallic system in the simulation with Yambo because there is a small overlap of the band structure at about 0.05 eV. However, it seems that the yambo_nl will use covariant dipole methods, which is not implemented for metal.

Followings are some methods that I have tried but cannot help:

1. I used “yambo_nl -V nl” to turn off the ”UseDipoles: Use Covariant Dipoles (just for test purpose)”, in this way, the simulation can proceed but the results are totally strange.

2. I also tried to use the “occupations = fixed” instead of “smearing” to force the system to semiconductor when doing the PWSCF, it helps in monolayer structure with small k points, but it cannot work in a high convergence system and bilayer structure.

3. Moreover, I tried to use shifted k points like “14 14 1 1 1 0” in the PWSCF to get a semiconductor result. However, it cannot help in the bilayer system, and the covariant dipole method cannot implement for shifted k points as well.

Is there other method to do the SHG simulation of the metallic system?

Best Regards,

Yubei

[claudio]

Dear Yubei

in fact the approach implemented in Yambo does not work with metals.

There are two possibilities:

1) you can try with another code, for example 2light
https://etsf.polytechnique.fr/Software/Ab_Initio

2) I can modify yambo_nl to work with shifted grids,
and let you know

best
Claudio

[Belle.X]

Dear Claudio,

Thank you very much for your suggestion.

I will try with other code first and wait for the modification of yambo_nl.

Thanks again for your help.

Best Regards,

Yubei

[claudio]

Dear Yubei

you can use Yambo_nl with shifted grids
if you remove the line

call error(' Shifted grids not implemented yet in covariant dipoles ')

in the file:
src/bz_ops/k_map_nearest_by_dir.F

and then works without symmetries, namely
you put in the QE file

noinv=.true.
nosym=.true.

best
Claudio

[Belle.X]

Dear Claudio,

Thank you very much for your help.

I will have a try.

Best Regards,

Yubei