Learning Forum

[Albi.Pani]

Hi, 

is there any way to calculately separately the right and circular polarization? Moreover I would like to visualize their propagation separately. 

[Guilin Sun]

it is a very broad question without specific simulation/device.

The simplest way is to excite left/right circular polarizarion and quantify their results.

If you want to quantify the polarization from the output fields, you can decompose them, for exmple

E1=sqrt(2)/s*(Ex+jEy)

E2=sqrt(2)/2*(Ex-iEy)

As for which one is left or right circu polarization it is up to you from your definition since there are two opposite definitions.

if you can decompose H fields, you will be able to get the correct transmission by following  https://optics.ansys.com/hc/en-us/articles/360034405354-transmission-Script-command

Following the same procedure, you can decompoae the fields anywhere inside the simulation volume if you want.

This is a more theoretical question than simulation.

 

[Albi.Pani]

 

Hi, thank you. I want to simulate a metasurface that, when probed by light with circular right polarization, generates one beam with circular right polarization and the other one with left circular polarization. I simulate it with fdtd but when I calculate the transmissivity, it calculates of course the total transmissivity. I want to calculate it separately for the two polarization.

Decomposition of the output field can be a good idea? Have you any reference how to do it?

 

The paper I am trying to simulate is this one:  Multifunctional resonant wavefront-shaping meta-optics based on multilayer and multi-perturbation nonlocal metasurfaces | Light: Science & Applications (nature.com)

 

and I want to obtain a figure like this one:

 

 

[Guilin Sun]

I do not have any reference. Probably you could find a paper. But the best way is to drive the expression by your self and then use the "transmission" link to get the desired transmission.

Again this forum is for simulation we do not provide guidance for any theory.

[Albi.Pani]

Yes, I know that this forum is for simulation. In fact I was asking how to decompose a field in the simulation. I know the theory. What can I do? I have seen that many use grating_S_parameter. 

[Guilin Sun]

 

The post is closed. Did you find the solution? From grating_S_parameter group you have the “S” paramters which are actually the E field(s). Except that you modified the group it only gives one field component.

Once you have the E fields,say Ex and Ey, or Ep and Es, you can decompose it as suggest in the first reply. for the farfield it is straight forward to get the optical intensity (0.5*E^2/impedance). Please find more information regarding to power calculation from some books. This article has some information but you may not directly use it without modification since it is for dipole.

https://optics.ansys.com/hc/en-us/articles/360034914813-Integrating-power-in-far-field-project

 

[Albi.Pani]

I have found the answer here:

Questions about the calculation of the cross-polarization conversion efficiency of metasurface (ansys.com)

 

Code:

 

switchtolayout;

setnamed("FDTD","mesh accuracy",3);

setnamed("grating_s_parameters","polarization angle",0);

setglobalmonitor("frequency points",50);

run;

S_pol1 = getresult("grating_s_parameters","S_polarization");

refractive_index_1=1;

refractive_index_2=1;

S12_x=S_pol1.S21_Gn;

S12_x=S12_x/sqrt(refractive_index_1/refractive_index_2);

switchtolayout;

setnamed("grating_s_parameters","polarization angle",90);

run;

S_pol12=getresult("grating_s_parameters","S_polarization");

S12_y=S_pol12.S21_Gn;

S12_y=S12_y/sqrt(refractive_index_1/refractive_index_2);

f=getdata("grating_s_parameters::T","f");

S12_L=(S12_x+1i*S12_y)/sqrt(2);

S12_R=(S12_x-1i*S12_y)/sqrt(2);

S12_RL=(S12_L(:,1)-1i*S12_L(:,2))/sqrt(2);

S12_RR=(S12_L(:,1)+1i*S12_L(:,2))/sqrt(2);

 

effect=abs(S12_RL)^2;

effect2=abs(S12_RR)^2;

 

plot(c/f*1e9,effect);

plot(c/f*1e9,effect2);

[Guilin Sun]

In many cases it works fine in the farfield. It is the same as Ex+-jEy.