I am trying to find Tranmission spectrum of graphene layer(1.6micron1.61nano).
In my simulated transmission monitor exist ripples.
I placed the simulation boundary approximately half a wavelength away from graphene layer.
how can eliminate ripples?
My simulation file:graphene-trans.fsp (250.5 KB)
Thanks for any help.
I believe that the problem here could again be due to the material model. To keep things simple, I have used the 2D graphene model instead of the 3D volumetric model you’ve created. The 2D model enables us to work with a much coarser mesh and still get accurate results. In the updated file (graphene-trans_mod.fsp (254.2 KB)), I have kept all the parameters for the 2D graphene model at their default values and have changed the chemical potential. In the plot below, you can see that the transmission curve has the same nature as the article you mentioned here.
You may need to check the effect of the grid size on the result as well as fine tune the scattering rate to match the reported mobility of the graphene sheet (which will shift the transmission curves).
I believe that the broadband graphene model should be valid for your device. The 2D graphene model of FDTD is implemented using the Kubo formula which is also the model used in the article you mentioned. To see the validity of the 2D model, check out this page in KB (graphene modeling methodologies).
Can you please explain what you mean by “different answers” here?
Scattering rate in surface conductivity of graphene
in your answer (mu_c=1ev) lower transmittans occurs near 5 THz.but in article is 7 Thz.
Hi. The response I shared earlier has the same nature as the one shown in the paper even though the peaks are shifted. This shows that the material model used (based on the Kubo formula) is valid. However, you will need to investigate the effect of mesh size and scattering rate to get accurate results as I mentioned.
how can find scattering rate ?
Usually, the scattering rate is reported in the published article (check this KX post). However, I did not find any reference to scattering rate in this paper. So I am not sure how you can decide which value to use. I have talked with my colleague and he has suggested that in these devices, the transmission peaks are extremely sensitive to the mesh. So all you might need is a finer mesh. As I mentioned before, you will need to vary the mesh and see if that changes your result and keep refining the mesh until you see that the peaks do not change any more. The effects of the scattering rate is more on the shape of the peak. A larger scattering rate broadens the peak but not necessarily shifts it. So you may not need to adjust the scattering rate after all. As long as the scattering rate is in the range of meV or less than 1 meV, I am guessing that the results should be pretty much okay. You can look at this example in KB to see how a mesh refinement object was used to ensure fine mesh in a similar plasmonic device using graphene.
In the attached file (graphene-trans_mod2.fsp (244.8 KB)), I have used a similar mesh override to make the mesh finer around graphene and you can see that the peak already has shifted to 5 THz for muc = 0.5 eV which is a much better agreement to the paper. Note that the peak is a bit wider than the paper so that probably indicates that you will need to use a smaller scattering rate.
thank you so much