Learning Forum

[Iuliia Riabenko]

Dear sir!

Please help me! I start to learn Zemax and Lumerica.

I sdtudy the article https://optics.ansys.com/hc/en-us/articles/360042097313-Metalens

And I have questions:

1. Zemax allows me to create phase mask using Binary 1. Why Binary 1 does plase befor cylidrical lens . Then we optimaze to get focus point . This give the phase distribution in binary 1. And I can save this data in txt file. But how can I use them in Lumerical?
2.  How the phase_map_from_ZOS_R100um.mat has been create? And which comand sread data from it and compare it with data from metaatom phase data (EH_and_phase_vs_radius_interp_fdtd.mat)?

Best regards,

Iuliia

[Guilin Sun]

It seems the main questions are for Zemax. Please submit your question here: https://community.zemax.com/

Interoperation of Zemax and Lumerical for metalens is: Lumerical FDTD simulates elements to design the metalens, and then simulate the metalens to get field result under given illumination (steps 1~3 in "overview"). Then the result is imported to Zemax for raytracing.

 

[Iuliia Riabenko]

Thank you for your answer. I clear undestand Zemax part. In which file need I enter phase data which I get from Zemax? How did you create the files: phase_map_from_ZOS_R100um.mat, EH_and_phase_vs_radius_interp_fdtd.mat and EH_and_phase_vs_radius_interp_rcwa.mat? 

[Iuliia Riabenko]

How to create the file phase_map_from_ZOS_R100um.mat and enter data:

[Guilin Sun]

Lumerical has script command "matlabsave" to save variables created from Lumerical products. Please refer to

https://optics.ansys.com/hc/en-us/articles/360034928113-matlabsave-Script-command

The simulation methodology for metalens is first to use FDTD/RCWA simulation to create the metlens and data, and then export. Zemax will import the created data. It is better to follow the steps described in the website example you referred. If you have questions at at step, please let me know.

 

[Iuliia Riabenko]

Thank you. 

How could I implement the next statment "complex systems, where analytical solutions don’t exist or would be hard to calculate, we can design the ideal phase mask in OpticStudio using the ray tracing and optimization capabilities."?How can I add this data to matlab file to use them it step 3? Where do you take the target phase profile for step3?

[nauval.afif]

Hi,

I understand how the files EH_and_phase_vs_radius_interp_fdtd.mat and EH_and_phase_vs_radius_interp_rcwa.mat were created (using Lumerical Scripting File), but I still don't understand how the file phase_map_from_ZOS_R100um.mat has been created. Based on the name I think the file should have been generated from OpticStudio but I don't know how to do it. Please give me more information.

 

Thank you.

[Guilin Sun]

"complex systems...." is for phase distribution is not spherical and no analytical expression then you will need Zemax for help. You only need to extract the wavefront infomation, the phase distribution from Zemax.

"phase_map_from_ZOS_R100um" is the target phase_map. You can use whatever software to create it and save it in .mat:

 

[nauval.afif]

I am still confused, the size of phase_target in "phase_map_from_ZOS_R100um.mat" is 445 x 445. But the surface phase map from the optimized "phaseDesign_start.zar" is only 257 x 257. At first glance also, the numbers look very different. Could Ansys please give more detailed information about this? The information in the metalens application gallery is too concise and not that detailed. Thank you,

[Guilin Sun]

They are different and not necessarily the same dimension. "phase_map_from_ZOS_R100um.mat" is for FDTD use.

[Iuliia Riabenko]

Hello!

How did you create the arrays x_mask and y_mask?

[Iuliia Riabenko]

Ok. I undestand. This is coordinate where phases were menshed. It's come from Zemax semi-diameter.

[Iuliia Riabenko]

Please exmlaine me these structures:

If I clear undestand these data come from FDTD(or RCWA calculation) which has been done for the Pillars with different radius. That why numbers of E(or H) is equal 361. Then we have the 9 points with coordinate in X and Y direction. Each point corresponding to the Electrical Filed and Magnetical Field in the First Frenel Zone . The number 1,2,3..., 361 is indexs which link the radius data of Pilars and the Electromagnetic Field in near region . 

 

How can I calculate phase which describe by one number from -pi to pi using the 18 values of Fields?

[Guilin Sun]

I would strongly suggest that you read the script file that creates the data. For example, in this file:

fdtd_unit_cell_export_phase_field.lsf

the data file is: 

EH_and_phase_vs_radius_interp_fdtd.mat

use the following script:

 

matlabload("EH_and_phase_vs_radius_interp_fdtd");

EE=E.E;

?size(EE);
result: 
3  3  1  1  361  3  

they represent x,y,z,f/lambda,phase and Ex/Ey/Ez.

 

Please check the data and see it is from which script file created and you will know the meaning of the dimensions.

 

 

 

[Muhammad Usman Afzal]

 

In the first picture of zeemax by Iuliia Riabenko, “Why the radius is 1.578 and thickness is of 0.1 in biconic surface?

Is there any procedure to calculate? 

Furthermore as per the article link shared by Muhammad Afif, “The obtained surface phase cross section response is different then shown in the shared metalens resource why is so much difference?”

 

 

[Guilin Sun]

For Zemax questions please post in zemax forum https://community.zemax.com/

It seems both questions are for Zemax.

[Guilin Sun]

Today I get sometime to review this post again, and have the following comments:

Since most often the metalens simulation involves Lumerical FDTD/RCWA and Zemax OpticStudio, some people may get confused when reading the articles from both websites. The basic work flow is from Lumerical article: https://optics.ansys.com/hc/en-us/articles/360042097313-Small-Scale-Metalens-Field-Propagation

The first step is to get the theoretical, ideal phase distribution (space) desired for the metalens with given focal length. We call this as "target phase profile". If it is ideally spherical, cylinderical, or any other phase distribution that can be analytically obtained, or from a table, it is easy to directly get the phase distribution, with any number of sampling you want. If one cannot directly get the phase, you will need Zemax to create the desired phase. The phase data in OpticStudio is in the Surface Phase tool.

Please note that in this step, the sole goal is to get the target phase distribution. Therefore you can use whatever the tools you have at hand. If you use Zemax, OpticStudio has fixed sampling points, seemly 2^n+1 where n=4,5,6,.... the integer number. Here the OpticStudio is not used to simulate the metalens, again it is to get the desired target phase!

The rest is the same as the examples. When the design of the metalens is for normal incidence, in Zemax it has to be illuminated with collimate light, and in FDTD it has to use plane wave with incident angle zero for simulation.