Help with modelling a Single layered structure

Hi all,

I am trying to study the optical properties of a leaf, which I have tried to model as a single layered structure, of the material with a refractive index of 1.4. I intend to study the Reflectance, Transmittance and Absorption of this material for 400 to 2500nm. The source I have selected is a plane wave source which is injected in the z axis. In essence, the model will only take into consideration the z dimension of the material since it is meant to mimic simulating a tiny portion of the leaf. The z span of the material has been set to 357 microns, the thickness of the leaf which I intend to model. The x and y span for the mesh has been arbitrarily set to a value of 10 microns (Is this the right thing to do?)

My mentor has mentioned that I will be able to obtain the 3 optical properties which I intend to study (Reflectance, Transmittance and Absorption) using the Frequency domain field and power monitor. However, I have tried some sample simulations with the same monitor and I am only able to get the result for Transmittance. May I know how I should go about calculating the results for Reflectance and Absorption?

Moreover, I have attempted to model the structure, wherein the Max time remaining for the simulation is shown to be around 14 hours, yet the simulation is complete in a matter of seconds. The resulting text document gives the following:
“0% complete. Max time remaining: 117 hrs, 18 mins, 16 secs. Auto Shutoff: 1
100% complete. Max time remaining: 0 mins, 0 secs. Auto Shutoff: 0
Early termination of simulation, the autoshutoff criteria are satisfied.”
I am unsure what to make of this information, and when I attempt to visualise the resulting values for Transmittance with the monitor, I obtain the following message “Failed to calculate results!”. Could any kind soul help me to spot the error in my simulation file? I have attached it below:
Leaf model 1.fsp (2.7 MB)

Lastly, could anyone please advice on the feasibility of setting the minimum mesh step as 0.1 microns? Is this value too small or large for such a simulation?

If anybody could suggest how I should change the any of the simulation settings or answer any of the above questions, I would be eternally grateful. Thank You!

Hello @151293e ,

Welcome to the community. I would be happy to help you get started.

  • To obtain Transmission, Reflection and Absorption you should place DFT power monitors on the other side of the structure, and behind the source. That way you will get the Transmitted and Reflected power. Absorption is the simply A = 1 - R - T, since R and T are normalized to the source power. Make sure that the simulation time is long enough or these values will not be calculated properly.
  • The reason your simulation is shut-off immediately is because there is no source in the FDTD region, and if there is no energy injected into the simulation it will immediately reach auto-shutoff.
  • I wouldn’t worry about the mesh override since the software will automatically determine the proper mesh based on the bandwidth of the source, and the geometry. Mesh cells of 100nm would provide 4 points per wavelength at 400nm the shortest wavelength in your simulation. You can use the mesh refinement in the FDTD object to reduce mesh size and increase accuracy.
  • You should reduce the dimensionality of the simulation since as you said you are only really concerned with the z-direction, thus a 2D or 1D simulations would be sufficient( and much faster!).
  • Since you are simply defining the leaf structure as a dielectric with index 1.4 (i.e. loss - less) , and are only concerned with frequency domain measurements (power) making the leaf structure 357 micron deep does not change anything except how long it will take to simulate. All reflection and transmission will occur at the interfaces. and so using the stack solver will be quickest, and a 2D FDTD simulation will also be quick and sufficient.
  • If you refine the simulation and find a better model of the leaf material (one that is frequency dependent and includes absorption) you could add it with the material explorer. Yet if you are simply interested in A, T, and R in 1D Stacks the you could continue to use the Transfer matrix approach.

Using FDTD is quite expensive computationally, but it will return accurate results if for example you need to model the interaction between the leafs capillaries and the light. For your application it seems a bit like overkill, but could become an important tool as you develop your leaf model.


1 Like

Hi, thank you very much for your reply!

I have attempted to model the structure as a 2D one, but I do not seem to be able to obtain the results for Transmittance and Reflection over the wavelength range of 400nm to 2500nm. I initially set the structure as a 2D rectangle, but I was unable to change the refractive index of the material, hence I have converted it to a 3D structure, but the monitors and sources remain configured in 2D. When I ran the simulation, for some reason the value of Transmittance calculated is always 0.

Could you help me to check the simulation file to see where I have gone wrong? Thanks!
1- layered structure.fsp (339.2 KB)

I have rerun the simulation with several DFT monitors placed at increasing distances from the source and it seems that: As distance from the source increases, values for transmittance decreases for the same wavelength range of 400nm-2500nm. This can be seen in the screenshot below:

I have rerun the experiment with different materials such as glass, but the same phenomena occurs. May I know why this is so?
Moreover, when I click on the “View data” button on the Visualiser, there only seems to be 4 data points per monitor. Is there any way to increase the number of data points collected?

Thank you very much!

This problem has been resolved in the post.

To increase thee number of data points that the monitors collect adjust global monitor settings

1 Like

This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.