How to simulate a nonamer?

fdtd

#1

Hi,
I am trying to find absorption and scattering of a graphene photodetector with nanoantenna like following picture,but unfortunately i can’t simulate it.
All values and parameters are precisely based on article,but my result is different from paper result and get confused why?
here is paper result:

and Here is the project file for my simulation:
02.fsp (290.7 KB)
01.lsf (248 Bytes)

I would really appreciate some advice.
Regards


#2

Hi,

You can take a look at this discussion on graphene based photodetectors.

With graphene being a single layer, you might need to provide adequate meshing that will account for the dielectric properties of graphene accurately. Typically, I use 1-2 nm mesh size to describe gold nanostructures and even smaller if the gap between the gold nanostructures are small. You need to enable the mesh for graphene which will be even finer.

I hope this is helpful.


#3

Hi,
thank you for helping,I read that and other discussion about graphene on kb & kx.lumerical.com.
I’m working on this paper for along time but unfortunately can’t get desired result .I changed my mesh grid as you said,even I changed source type (Dipole,TFSF,Gaussian,…) and it’s polarization and also it’s position (up and bottom of the nanoantenna) but i can’t get the desired result yet.
can you help me to design the best simulation to achieve absorption and scattering like paper?

best regards.


#4

Hi,

Sure, I will take a look into it.

Best Wishes,

Vivek


#5

Hi,

Just a couple of quick comments:

  1. Are the absorption and scattering cross sections in the plot calculated for the entire periodic structure or for a single antenna? Also, are they suppose to characterize just the gold antenna or also the silicon layer? This is important because silicon will also contribute to absorption.
  2. To speed up your simulation you can bring the PML boundaries (upper and lower boundaries) closer to the structure to reduce the size of the simulation window. Just make sure the distance from the structure to the PML is at least half the maximum wavelength.
  3. You should extend the structure through the periodic boundaries to make sure the unit cell is correctly simulated. This is explained in more detail in this post.
  4. Did you check that the graphene parameters you are using are the same as in the paper? Usually the results can be very sensitive to the graphene parameters.
  5. Can you provide the paper information? I think it would help to double check the settings and get a better idea of the expected results.

#6

As @fgomez has mentioned in his post is this structure periodic or not. If it is not a periodic structure, I would suggest the use of total-field scattered-field source. This is the approach that works best for isolated structures. For example take a look at the Mie Scattering tutorial in the knowledge base. Here is a link:

https://kb.lumerical.com/en/particle_scattering_mie_3d.html

Also, can you share the paper that you are trying to reproduce.


#7

Hi,
dear vivek and fgomez,thanks for your help.
in this paper that is an experimental work,sufficient information about graphene or other material parameters is not given! I extract graphene parameters from other similar papers and kx.lumerical examples.
I changed graphene parameters and source types several times but got no good answer!

I will wait for your help.
best wishes.


#8

Hi,

First, you have got the plasmonic structure a little incorrect. You have entered the diameters in the radius tab, thus the Au nanostructure is twice as big.

Second, to get the scattering and absorption cross-section, you the Mie scattering approach with a total-field scattered-field source. See the tutorial here:

https://kb.lumerical.com/en/particle_scattering_mie_3d.html

This will allow you to compute the scattering and absorption cross-section. I am not sure whether the authors included the graphene layer in their calculations. The methods section is rather weak. However, Halas, Nordlander and co-workers have carried out extensive work in this area. You can check their previous papers on this.

I will try to run a simulation with and without the graphene layer and let you know.

Best Wishes,

Vivek


#9

Hi,
Thanks,I put values based on the paper and simulated it with TFSF source,but did not a good answer.
here is my modified simulation with TFSF source:
modified with TFSF source.fsp (308.8 KB)

Could you please change my simulation to obtain correct result like figure?

Best wishes.


#10

Hi,

There is a correction needed for the size of the plasmonic nanoparticles as I have mentioned earlier. The article mentions that they are 190 nm in diameter and 112 nm in diameter respectively. The same numerical values are entered for radius, this the structure is twice.

Second, if I am correct the Mie scattering will work for an isolated structure. Hence, I have changed the FDTD boundary condition setting to PML (x max bx and y max bc).

It is always better to have frequency domain field profile monitors at least 0.5 wavelength of the z PML layers. And for this case, this monitor won’t be of any use as the scattering and absorption cross-section is computed by the analysis groups. If you go for the periodic structure, one can compute the reflection arising from the infinite array, however, the array pitch is 1.0 micron based on the SEM image in the figure. I am not sure whether the authors of the paper computed for the array.

I am debugging your simulations file. If I run into some success, I will let you know.

Regards,

Vivek


#11

Hi,

The thesis gives a few details on the simulation of these FANO type resonance. The MS thesis refers to the use of FEM (finite element method) to calculate the Fano resnonaces in these clusters (Page 13 & 14). I believe FDTD should be able to achieve the same set of results provided the modelling is done accurately.

There is more details on the use of graphene layer in page 16. You need to check whether you have incorporated the conductivity of the graphene layer. They have employed a 2 nm layer to describe the graphene. With all this information in hand, I need to give this system a go. Lets see whether I can debug you simulation.

Best Wishes,

Vivek


#12

Hi,
Dear vivek,thanks for your reply,
I read his MS and PhD thesis,his PhD thesis is a similar work on nonamer gold cluster but he used gold film instead of graphene.In both of them he didn’t give sufficient information to simulate! I sent several emails to authors but nobody didn’t answer me!

As you said he simulate with COMSOL software,and I,like you, hope simulate it with Lumerical FDTD solution but up to now i don’t get the right answer.I simulate it with 2 nm graphene with fine mesh but the answer was not good.I think we must change our structure a little to achieve good result.

Dear vivek,please change the structure to get the right answer and continue to help me in simulation.

Best Wishes,
Hananeh


#13

Hi Hananeh,

I checked your last simulation file and the paper you want to reproduce [Z. Fang, et al. “Plasmon-Induced Doping of Graphene”, ACS Nano 6, 10222 (2012)]. Besides the corrections that @vivek mentioned already regarding the size of the circles and the boundary conditions, there is an important consideration for the z span of the TFSF source and the cross-section analysis groups (scattering and absorption). It seems like Fig. 1 in the paper is meant to describe the cross section of the nonamer on top of the oxide layer; therefore, the TFSF and the cross-section analysis groups should only enclose the nonamer and part of the oxide layer, not the silicon underneath. After these changes I got a curve that looks much closer to the result in the paper:

I should also mention that I used two mesh override regions to make sure the mesh is uniform in the x an y directions inside the TFSF (as suggested here) and to get enough mesh points along the z direction inside the nonamer. To speed up the simulation I reduced the size of the simulation window, making sure there is a distance of at least half the maximum wavelength between the structure and the PML boundaries.

The scattering peak at ~1000nm and the absorption peak at ~800nm are smaller than in the paper but I think this is an issue related to the mesh. You should do some convergence tests refining the mesh in the mesh override regions. Also, I suggest checking the effect of increasing the number of PML layers to make sure the PML is not affecting the results.

Take a look at the modified simulation file: modified with TFSF source - FG.fsp (320.7 KB).

Hope this helps!


#14

Thanks @fgomez. The simulation data indeed looks close to that reported in the paper. As @fgomez mentions, this would be for the 9 particle structure with graphene and effects relating to the Si layer not included.


#15

Hi,
Dear federico and vivek,thanks a lot for your simulation.
I checked your simulation,it was amazing .But I have two questions:

First;Why you use graphene with very tiny dimension like a dot instead of graphene layer?I simulated it with and without your graphene,the curves of scattering & absorption were unchanged!

Second,in this project,Why cannot get the result with the paper’s values?

I simulate yours with 2D graphene layer (first without and last with graphene layer),here is the result:

As you see,the result will be changed.what’s your idea about use graphene layer?

Best Regards,
Hananeh.


#16

Hi,

Have you checked whether the conductivity of graphene incorporated in the simulation is similar to that reported in the paper? This could a source of deviation from the paper.

Best Wishes,

Vivek


#17

Hi,
Dear Vivek,thank you.
I simulate it with 2D graphene layer,here is my simulation:
modified with TFSF source and graphene layer - FG.fsp (321.0 KB)

and the best result is this:

As you see,the figure is much similar but it has some difference such as in 1000 nm the scattering peak is small yet,would you please change it to achieve the best result?

Best regards,
Hananeh.


#18

Hi,

One of the option is to change the mesh size around the important components of the structure. For example, the graphene layer is only 2 nm or so thick. A mesh size of 0.5 nm will allow for the accurate description, as 4 Yee cells will be involved in the computation for graphene. Similarly you can have an adaptive meshing to ensure enough Yee cells are employed for the plasmonic nonamer.

Strangely, I am not able to open your fps file. The program crashes.

Best Wishes,

Vivek


#19

Hi Hananeh and Vivek,

The problem with the file is that I saved it in the latest release version (v.8.16.871). When you open the file in an older version the program modifies the size of the graphene sheet (and may also be responsible for the crash Vivek reported). I suggest upgrading to the latest version. There are lots of new interesting features you might be interested in!