Learning Forum

[rajatkumar.sinha]

Hey, I have modified the example file of "Surface plasmon resonance in a glass/silver/air slab" to SiO2(Substrate)- Ag(50nm)-PMMA(600 nm)-Air with source at 45 degree. I am "sweeping" the wavelength from 400nm to 700nm (I am using the sweep feature and not using the BFAST mode). According to Multilayer calcluations and experiments I am getting reflection dips around 450nm and 650nm but when I am simulating in Lumerical I am getting none(Reflection Spectrum Attached). Some additional Infomation 1) I am using mesh only in the Ag region. 2) Using Bloch Boundary Condition 3) Using mesh refinement as conformal varient 1. 

What changes I can try in FDTD to mimic my theoretical and experimental results.

[Guilin Sun]

Please do not use Conformal Variant 1. Instead use Variant 0. This is because Variant 1 may need much finer mesh and we do not know how fine the mesh should be. https://optics.ansys.com/hc/en-us/articles/360034382614-Selecting-the-best-mesh-refinement-option-in-the-FDTD-simulation-object

in addition, please fix the wavelengh range for simulation:

When the wavelength changes, the mesh and the material fitting will change. So you need to fix it for the simulation.

 

[rajatkumar.sinha]

thank you so much for the reply. I made changes but Now i am getting something like this

[Guilin Sun]

Please check the material fitting  to make sure its wavelength range is fixed as expected.

Also please check two wavelengths (two end-values) and make sure the mesh size is the same.

Then most likely due to

1: PML too thin. Please use more number of layers, say 64; you may also use "steep angle PML". Make sure the PML has about half wavelength thick, and at least half wavelength away from the structure. You can do 2D simulation.

2: significnatly increase simulation time and reduce autoshutoff min.

3: make sure the silver is well resolved by mesh. You can use override mesh for the silver layer with mesh size 5nm.