Possible Divergence due to Light Source


#1

Hello,

I have been trying to figure out why the simulation with 25nm ITO diverges. I have another structure exactly like this one where the ITO thickness is 80 nm and that doesn’t diverge. But when I changed the thickness of the ITO to 25nm here the simulation result diverges. Can anyone tell me what I am doing wrong?

Please see the attached link

https://drive.google.com/drive/folders/0B0yXuqS-bE07ZmpZQVEyc3NodDQ?usp=sharing

Thanks!


#2

Hi, @rumman18 !

I have a problem with your simulation files - they freeze Lumerical only when being opened.
Could you please upload the files with no simulation results, so as not for them to be too heavy?

As a starting point, I can recommend to read the corresponding page where the divergence problem is described in general:
https://kb.lumerical.com/en/layout_analysis_diverging_simulations.html


#3

Hello @msaygin

Here is the link to files. Let me know if it works or not.
https://drive.google.com/drive/folders/0B0yXuqS-bE07ZmpZQVEyc3NodDQ?usp=sharing

Thanks

Rumman


#4

It’s the same :disappointed:
I wonder if it is my specific case of quite common.


#5

Dear @rumman18

I was able to simulation file, and before diving into divergence problem I had a few question for you regarding the simulation file:

  1. there are a lot of empty spaces between the material and FDTD simulation region. As a result you will expect the light will get reflected from the material edges (where it enters the air). With the current design, it looks like you have a periodic array of devices that are hold on air.

My idea is that you need to extend the material so that they pass through the FDTD regions unless you have a good reason that it should not.

  1. What are the 2D circle and 2D rectangle objects? It looks like you have used 3D material but set the z-span to be 1nm. Again, I am not quite sure what your design is but with your current mesh you will not be able to resolve these objects. We have 2D objects that you can access them from structure tab:

For the divergence problem, there are few things to try, however, I think that you need to modify your simulation file first and then run it again. Here is the comprehensive list of things that you can try:

https://kb.lumerical.com/en/layout_analysis_diverging_simulations.html

Thanks


#6

Hello @bkhanaliloo

  1. When you say “periodic array of devices held on air” do you mean gaps between each layer?

or do you mean

(Does the blue outline sandwiched by orange border in the FDTD region mean anything?)

And are you suggesting to bring them closer like this?

(I apologize I am really new to this software)

  1. I am trying to model the structure from this paper.
    https://drive.google.com/file/d/0B0yXuqS-bE07dnRPMV9PWUVNc00/view?usp=sharing

The reason for 2D rectangle and 2D circle are as follows
- I made the MoO3 using surface structure.

- I want silicon in rectangular part of the previous structure. So, I superimposed a 2D rectangle at the exact location   with a lower mesh order

- But I want MoO3 in the middle of the rectangle. So, I put a 2D circle with a lower mesh order than the 2D rectangle

So, now I put this structure inside Si and I get the groove that I was trying to model

Thank you for taking the time to look at it.


#7

Dear @rumman18

Nothing to apologize for and sorry for not being clear.
I meant the gap between structure and PML, the one that you pointed on the second figure and is correctly fixed on the third figure.

That is the PML layer and its thickness will change depending on the size and type of the PML layer.

I think I still didn’t understand what your design will be. Please note that with current design, your simulation can not resolve these layer. I thought maybe these links will be a good place to learn more about 2D material:
https://kb.lumerical.com/en/index.html?materials_conductivity-models.html
https://kb.lumerical.com/en/index.html?ref_sim_obj_structures_2d_rectangle_optical.html

Thanks


#8

Hi @bkhanaliloo

I read the 2D materials link that you provided. So 2D structures can only be either dielectric or graphene type
Quick follow up questions.

  1. For this pic does the 2D part of the structure get resolved?

(Perspective view)

(XZ view)

  1. In this simulation, I am trying to do a simple structure comprising of glass and ITO. The simulation diverges when I enable glass in the structure. Can you tell me what I am doing wrong?

https://drive.google.com/file/d/0B0yXuqS-bE07MW55dmRGZmxFVmM/view?usp=sharing

Thanks,

Rumman


#9

Dear @rumman18

The second figure shows a typical sideview of 2D material in which you don’t expect to see the material. So just looking at the figure it looks OK. I recommend you to read this post:

please note that there are many posts on KX and KB that you can learn more from them, and I strongly recommend you to perform a search about any questions that you might have.

  1. I did some modification into simulation file:
    a) extending the material to pass through bloch boundaries as we discussed before
    b) using a coarse mesh of 1 from FDTD settings and defining a new mesh override region on the z-direction
    c) increase monitor resolution to capture more frequency points

Simulation work fine now. Please take a look at them:
Simple Sweep_Modified.fsp (288.7 KB)

Later on if you faced the same problem, I recommend you to look at this page:
https://kb.lumerical.com/en/layout_analysis_diverging_simulations.html

Thanks


#10

@bkhanaliloo

Thank you so much!!