Graphene Nanohole Array Simulation

fdtd
graphene
plasmonics

#1

Hi,

I am doing a Graphene nanohole array simulation. In this case I am just putting a layer of graphene which is 0.34nm thick over Al2O3 film in which holes are etched. I am just varying the periodicity of the hole from 200nm to 320nm in steps of 40nm. During our experiment we got a curve similar to this figure:. The whole experiment is done in the UV region. Now the problem is the graphene model provide by Lumerical does not have a fit in the UV region . So, I just used a sampled 3D model for Graphene and fitted the results I got from my Graphene Spectroscopy into Lumerical. I did a similar thing for Al2O3. The material file for both these is given here: Al2O3_UV_and_Graphene_UV.mdf (238.9 KB)

Now the main problem is if I run the simulation, the results are no way near the results I got from the experiment. The most puzzling thing is in the simulation, the dips are not getting shifted at all, which suggests that my simulation is not considering graphene. Can anyone please help me and check if I have done everything correctly or something I can do to fix this? I am new to Lumerical and don’t have much experience using it.

Here is the simulation file: sp_array_240nm.fsp (282.8 KB).


#2

Hi @sourangsu.banerji

Sorry for my late response, it was posted on the wrong category and I didn’t see it before.

Here are some comments on your current file:

  1. There is a gap between graphene layer and cover. This might cause problem specially when you use finer mesh on the graphene layer.

  2. Your mesh doesn’t properly resolve the graphene layer. Try to use an override mesh to have a few mesh cells over graphene.

  3. what is the adhesion material? It has a thickness of 1e-7 um(??) which is confusing. Whats the thickness as with the current value software can’t resolve it.

I did some modification only on the mesh (using two mesh cells inside graphene) and here is my transmission plot:

I wanted to do some convergence testing and increase the number of mesh cells inside graphene but it requires very long simulations. It will be a good idea to use 2D graphene and use sampled 2D data to speed up the simulations.

Keep me updated of your simulation results and I hope we can get reasonable results.

Thanks


#3

Sorry for the late reply but would it be possible for you to share the simulation file in which you did the simulation?

I am trying to run the simulations again and see if anything is resolved but it would be good if I can get the simulation file you used.

Moreover to answer your questions:

  1. Ok. I will remove the gap and try again with the simulations.

  2. Can you please tell me, how to use the override mesh? I am new to this software. I thought that I already added the override mesh.

  3. The adhesion material has no meaning as we just wanted to simulate the structure which we actually fabricated in the nanofabrication facility here.

Yes, we thought of using 2D Graphene but the problem here is if you see from the figure, the range of wavelength used is 200 nm - 600 nm which is roughly the UV and lower visible region of the EM spectrum . Firstly, Graphene models as provided in Lumerical does not have a good material fit as in the UV region as Graphene’s conductivity is marked with a asymmetric resonance around 270nm known as Fano Resonance.

What we have used here is a 3D sampled data from our Graphene ellipsometry. Is there anyway to use this 3D data and use it with 2D Graphene?

Thanks


#4

Hi @sourangsu.banerji

Unfortunately I don’t have the simulation file any more. Sorry its been a long time since our first discussion.

This link will be a good place for you to learn about mesh refinement.

I guess its size is not chosen properly. At the moment it is set to be 1e-7um. If its 100 nm for example, it would 0.1um not 1e-7um.

Here is a good link on how to convert 2D to 3D graphene.

Thanks