Exciting SPP using grating coupling for whole structure

Hi sir
I wanted to excite the SPP mode at 1550 nm, Using full simulation of structure. Our structure includes dielectric grating (index = 2.5) on gold substrate of thickness 50 nm. And there are 12 layers of dielectric on gold film.
Using periodic boundary condition (in x direction) with only one grating (including only one layer of dielectric) it seems fine.
But if I am trying to excite same SPP at 1550 nm for whole structure (The actual structure, actually), using full simulation i mean including all 12 grating layers and PML boundary conditions in x and y both, it doesn’t work. If you check the field profile and movie monitor for field as well , it doesn’t looks like an SPP at all.
So basically i want to simulate the whole structure not just a one grating with periodic BC.
SPP FDTD.fsp (2.5 MB)
I am attaching my simulation file , please have a look and suggest me on it. It would be great help.

Hello @sajedsaraj,

Thank you for the question. It is not immediately obvious what could be wrong with your simulation, it seems like it is set up well. I would recommend you use the Gaussian source instead of the plane wave with PML boundary conditions, if you use a plane wave source it will be truncated at the PMLs and there will be edge effects.

Can you send the file that works as expected with periodic boundary conditions? It may help me figure out what is wrong.

Thank you,
Kyle

Thanks a lot sir for your quick response
Hereby I am attaching the file for periodic BC.
I have tried with guassian source as well but it didn’t work out.

You can check in this new file that the resonance wavelength is at 1560 nm with minimum reflection at this wavelength. The field corresponding to this wavelength is confined at the interface between metal film and dielectric grating, this is what the SPP is. The same can be varified with movie monitor as well.

Now if we want to see the same for whole structure, it seems missing. Although there are some literature available for exciting SPP in FDTD but they are for different geometry and I was trying to get some help but unfortunately I was unable to resolve the problem.

Looking forward to hear from you.
Thanks & RegardsSPP Lumericals2.fsp (2.3 MB)

Thank you for sending your second file. I made some changes to your first file:

  • Because this is a diffracting structure there can be reflections at steep angles, so I set the PML profile to “steep angle” and added more layers.
  • Added a mesh override to fully resolve the skin depth of the metal substrate (you might want to add another mesh override to resolve the layers in your grating)
  • Extended the simulation region
  • Used a Gaussian source instead of a plane wave

Here is the new file: SPP FDTD_new.fsp (2.5 MB)

From looking at the movie monitor results, it looks to me like some of the input pulse is being transferred to the SPP:

I don’t think you can expect the reflection monitor results to be the same as the simulation with the periodic BCs, as the grating only has six periods. You could try adding more periods to the grating structure and see if the results begin to converge to the infinite case.

I hope this helps. Let me know you have any questions.

Thanks a lot for your response. Can you please let me know the difference between the simulating the whole structure and simulating the only one period ? Is it okay to simulate the one period instead of whole structure. I am attaching the updated file here i increase the period to 12 (earlier it was 6) but the results are again same no reflection dip result. For simplicity I just replaced multilayer dielectric to single dielectric layer on metal film and increase the period, Is there anything you want to suggest me on it, is possible with Lumerical_FDTD or not to get a standard SPR response? SPP Lumercals.fsp (374.9 KB)

Can you please let me know the difference between the simulating the whole structure and simulating the only one period ?

In your simulation with periodic BCs you are basically assuming that the structure is infinitely long. Your actual structure is only a finite number of periods. If you have many periods, your structure can be approximated by the infinite structure, with the approximation getting better with more periods.

I am attaching the updated file here i increase the period to 12 (earlier it was 6) but the results are again same no reflection dip result. For simplicity I just replaced multilayer dielectric to single dielectric layer on metal film and increase the period, Is there anything you want to suggest me on it, is possible with Lumerical_FDTD or not to get a standard SPR response?

I am not sure that the dip in the reflection spectrum you are seeing in your periodic simulation indicates excitation of the SPP. This dip may be due to absorption losses, for example. To study excitation of the SPPs I would recommend you use mode monitors or port objects to measure transmission into the SPP directly, similar to these grating coupler examples:

Note that these examples do not include absorption loss, which would have to be accounted for in your simulation. These examples also demonstrate varying the angle of the source, which is a very important parameter for grating couplers.

Another thing you could try is using a mode source to inject an SPP directly into the grating and studying the farfield emission from the grating. This would help indicate the correct injection angles to use for the grating.

Let me know if you have any questions.

1 Like