Yambo Community Forum

Dear developers，

When I calculated BN with a N vacancy, I find the lowest excitonic wavefunction is asymmetric, although the structure has a C3 symmetry. Please see details,

When K points is 3 3 1, and put hole position at the centre of defect,
[img][/img]
it is asymmetric. And I put hole on a asymmetry position,
[img][/img]
it's strange, the wavefunction is symmetric. Obviously, it is unreasonable.

When K points is set as 4 4 1, and put hole position at the centre of defect, the symmetry will become better than that of 3 3 1 k-points.
[img][/img]

So, that's very strange. In fact, if hole is on the symmetric position, wavefunction should be symmetric and symmetry should not depend on the K points.
Can you tell me what's wrong? Can you tell me how to work in code?

Best
Ke

Rensselaer polytechnic institute， Troy, NY, US
Hunan university, changsha, Hunan, China

Dear Ke,

there could be different reasons for that.
1) You have degenerate excitons: in this case, you need to do a linear combination of them in order to recover the correct symmetry see e.g.
https://doi.org/10.1103/PhysRevLett.100.189701.

2) You do not have degenerate excitons, but excitons very close in energy and you are mixing them.

In order to include or exclude excitons that are close in energy (or degenerate), there is a variable in ypp input which set the energy threshold within which different excitons are considered as degenerate and linearly combined.
The default is 0.01 eV, if you want to plot a single exciton you need to set it as 0 eV.

Other possible reason: even if you set the hole in a symmetric position, the real space grid is given by the FFT grid and the code set it in the closest point to the one set in input contained in the grid, so it is not guaranteed that it is still in a symmetric position. The grid can be controlled by the FFTGvec variable.

Best,
Daniele