non-linear response

[sitangshu]

I will send in your email...

With regards,
Sitangshu

[sitangshu]

Dear Professor Claudio

I hope you are fine. Just wanted to share some news with you regarding our latest work on ab initio non linear optical responses in monolayer GaAs:

https://arxiv.org/abs/1910.08443

I hope you will like this.....

With regards

Sitangshu

[claudio]

Very good!

Claudio

[Marzieh]

Dear Claudio,

How to get all non zero components of SHG spectra? Which variables is for this goal?
For changing frequencies I should change NLEnSteps, right?

Best,
Marzieh

[claudio]

Dear Marzieh

>How to get all non zero components of SHG spectra? Which variables is for this goal?

you select the cartesian direction of the probe field with the variable and then measure the response
in the direction that you want. For example in the AlAs tutorial

http://www.yambo-code.org/wiki/index.ph ... r_response

we put the field in the x,y direction with the input parameter

Code: Select all

% ExtF_Dir
  1.000000 | 1.000000 | 0.000000 |        # [NL ExtF] Versor

then we measure the response in the z direction. In such a way we got the X^2_{xyz}.
Notice that in the output file

Code: Select all

o.YPP-X_probe_order_2

contains also the response in the x, y direction
and therefore from the same calculation you get the X^2_{xyx}, X^2_{xyy} and X^2_{xyz}.
Some of these can be zero due to symmetry reasons. (note that due to the numerical noise the zero components may be very small but slightly different from zero).

>For changing frequencies I should change NLEnSteps, right?

Yes, this are the number of frequency steps in the energy interval between

Code: Select all

% EnRngeRt
  0.00000 | 10.00000 | eV    # Energy range
%

let me know if it is clear
best
Claudio

[Marzieh]

Dear Claudio,

Thank you for your explanation.

Could you please explain more to make clear for me?
For example which ExtF_Dir are for the X^2_{xyx}, X^2_{xyy} and X^2_{xyz} respectively?

Can we have all non zero components in a calculation or have to calculate separately?

Best,
Marzieh

[claudio]

Ok,

in the X^2 you have three direction, two of the incoming fields and one where you measure the response

The field direction determines the first two components and the response the last one.
For example if you set in Yambo field direction

1.0 | 0.0 | 0.0 | ---> you calculate all components X^2_{xx*}
0.0 | 1.0 | 0.0 | ---> you calculate all components X^2_{yy*}
0.0 | 0.0 | 1.0 | ---> you calculate all components X^2_{zz*}

1.0 | 1.0 | 0.0 | ---> you calculate all components X^2_{xy*}
1.0 | 0.0 | 1.0 | ---> you calculate all components X^2_{xz*}
0.0 | 1.0 | 1.0 | ---> you calculate all components X^2_{yz*}

the * correspond to the different direction in the response file o.YPP-X_probe_order_2.

In the o.YPP-X_probe_order_2 you have seven columns:

The first is the energy,
the 2nd and 3rd columns are the imaginary and real part of X^2_{**x}
the 4th and 5th columns are the imaginary and real part of X^2_{**y}
the 6th and 7th columns are the imaginary and real part of X^2_{**y}

where the ** are the direction of the incoming field, see above.

IF wan

[Marzieh]

Dear Claudio,
Your explanation was very instructive for me to follow the next steps.
Thank you.
Best,
Marzieh

[claudio]

Dear Marzieh

I added this explanation to the tutorial

http://www.yambo-code.org/wiki/index.ph ... fferent_X2

best
Claudio

[Marzieh]

Dear Claudio,

I have several questions.

If I have understood correctly, at the first we should use converged abinit( pwsf) files (with respect ecut, nband and k point). Then we have to do convergence yambo calculation only with respect NLnabd, right?

To get SHG responses at zero frequency, can we put like this?
% NLEnRange
0.000001 | 5.000000 | eV # [NL] Energy range
or
change NLEnSteps variable?

For getting the SHG responses to laser 1064 nm, how to do?

Best,
Marzieh