Yambo Community Forum

Dears Yambo developers and users,

Regarding the non-linear calculation implemented in yambo 5.1.0, I am currently researching centrosymmetric materials with space group symmetry P3m1.
Investigating this material's nonlinear response is our goal. In fact, only noncentrosymmetric crystals can exhibit second-order (SHG-X2) nonlinear optical interactions.
However, when I calculate using yambo_nl, followed by Fixed_symetrie (ypp-y) then yambo_nl, yambo_nl -u n, and finally ypp_nl -u to calculate the n order susceptibility X(n):

1) I found that this material had non-vanishing second order susceptibility, or X2, which contradict the fact that for materials display inversion symmetry , the second order susceptibility (X2) vanishes ?

2) I removed the time reversal symmetry in the step of fixing symmetry by using the command ypp-y. Is this step breaking the inversion symmetry?

3) Could you please explain why we should remove the time reversal symmetry?

Any advices would be appreciated.

Thanks
Best regards.

Dear Sabrine

> 1) I found that this material had non-vanishing second order susceptibility, or X2, which contradict the fact that for materials display inversion symmetry , the second order susceptibility (X2) vanishes ?

probably you are getting just numerical noise. In general if you compare your X2 with the one in a normal material, like in the tutorial examples you should see
that a centrosymmetric system present a X2 order of magnitudes smaller that we consider zero.

> 2) I removed the time reversal symmetry in the step of fixing symmetry by using the command ypp-y. Is this step breaking the inversion symmetry?

Since you obtain non-linear response using real-time simulations, that means you introduce a real external field in the Hamiltonian,
you need to remove time-reversal and inversion symmetries from your symmetries because they are not compatible with the presence of the external field.
However if your system had inversion symmetry your response functions X1, X2, X3, will respect it within the numerical precision.

> 3) Could you please explain why we should remove the time reversal symmetry?

See reply to question 2, and have a look here:

http://www.attaccalite.com/lumen/theory.html

Best
Claudio

Dear Claudio,

Many thanks , it s more clear now

i have two other question please ,

1) i want to calculate the non linear optical response under left circular and right circular polarization excitation, is this feasible using yambo. i mean it is possible to control the polarization?

2) also there is a way to calculate the second order nonlinear photocurrent, actually i want to calculate the circular photon drag effect current ?

Best Regards.

Dear Sabrine

> 1) i want to calculate the non linear optical response under left circular and right circular polarization excitation, is this feasible using yambo. i mean it is possible to control the polarization?

In principle it is implemented, and it should work, you can choose the light polarization in input, but I never tried to do it

>2) also there is a way to calculate the second order nonlinear photocurrent, actually i want to calculate the circular pho> ton drag effect current ?

good question.... you can calculate the induced current, just turn on the flag EvalCurrent
but then I don't know how to relate it to the drag current effect.
Let see if other people comment on your question

best
Claudio