Tooth-shaped plasmonic waveguide filters using FDTD

I am very new user of lumerical, I am trying to re-simulate the nano plasmonic filter in the paper entitled “Tooth-shaped plasmonic waveguide filters with nanometeric sizes”, I attached with this post as well as the FDTD file…I tried to do Figure3 part a…the case when d=100 nm,
my problem is that I could’t get the same result, even though I tried put the mode source range as in the paper however I couldn’t, it gives me wrong results…
could any expert help me and give me some instructions about that??nanofilter.fsp (1.9 MB)

thank u

Dear @s2aldaws

Generally simulating plasmonics where metals are touching the PML layers are challenging. I did a few testings and then finally ended up using stablized PML at X boundaries and standard at Y boundaries. Stablized PML is very useful where plasmonic modes touching the PML, and generally require a longer simulation time. Standard are chosen in the Y boundaries as field is attenuated enough before it reaches to the PML layer. Here is the transmission plot as well as my modified simulation time:

and here is my simulation file:

nanofilter_modified.fsp (311.1 KB)

Please note that your simulations are done in 2D and your results might be different that the results of the paper where they use a finite thickness for the material.


Thank u so much for ur quick response and help.

I have a question please? what are the factors that caused the zigzag-shape in the transmission curve?

Is it the Fdtd simulation set up? the mointors? or the mesh??

thank u again…

Hi @s2aldaws

Ripples in transmission data occur mainly because the simulation stops before field in your simulation region has decayed enough. One solution for you would be increase the simulation time. Please also check this link and this one which will explain the problem in detail and some techniques to reduce/avoid ripples.

Please let me know if you have further questions.


1 Like

Thank u so much again.

1 Like

Hi Again,
When I tried to run the simulation after ur modifications it gives me wrong results,
Could u please see my attached file and why it gives me these results?
I appreciate ur time and help.

thank unanofilter.fsp (307.6 KB)

Dear @s2aldaws

It looks like the injected mode wasn’t getting updated. Please select fundamental mode from Edit Source —> mode selection and make sure that the filed profile is as you expect:

With this modification you should be able to replicate the results. Please let me know if you had any other problems.


1 Like

Thank u so much for ur help and reply.

1 Like


I am trying to re-simulate the attached paper to find the transmission of Figure2 the green curve where d1=250nm, and d2=100 nm

It is the same idea of my first post above but here is multi teeth shape SPP filter. I tried different boundary conditions wit increasing the simulation time, it gives me almost the same minimum peak but the rest is not correct.
Also I attached my FDTD file.multi-teeth-SPP filter.fsp (363.8 KB)

I appreciate any advice or help.

thank u

I appreciate any comment or advice.

thank u

Dear @s2aldaws

Your simulation looks good. I did some slight modification:

multi-teeth-SPP filter_ModifiedBK.fsp (1.1 MB)

and I used this code to create the plot:


Tnorm = TP2.T/TP1.T;


And here is the plot:

I did some converging and results are reliable but clearly the resonances are not matched with the ones in the paper. I tend to believe that our results make more sense because we are dealing with a dispersive material (silver-Drude model) here and we expect to see that spacing between resonances to change. Let me know of your thoughts and hopefully we can come into a concussion.

1 Like

Thank u so much fore ur reply and help.
I will look at it .

1 Like