Learning Forum

[sj_123]

Hi,

I am trying to simulate a structure with silver nanoparticle over an SiO2/Si substrate. Following is the transmittance to SiO2 layer. At around 630 nm I see a discontinuity in the spectra. I am getting such a sharp spike at this wavelength irrespective of the geometry or dimension or material of the nanoparticle. When I reduce the period of the FDTD unit cell beyond a limit the discontinuity is not observed. Please suggest how can I solve this issue. I used periodic boundary condition.

Thanks in advance.

[Taylor Robertson]

There may be a an issue with the material model you are using could you share a screenshot of the material fitting.
https://support.lumerical.com/hc/en-us/articles/360034915033-Material-Explorer
Alternatively there may be resonance associated with the Si02 layer at this wavelength.
Does the signal become cleaner if you increases simulation time.

[sj_123]

Sorry for the late reply.

The above are the material fits for Silver and Si.
I tried by increasing the simulation time, still I am getting this spike.
When I use periodic boundary condition I am getting such a spike at wavelength corresponding to the FDTD period. So I tried with PML. But, in this case I am getting a not so smooth curve (with lots of ripples) from all the monitors.
Could you please tell what could be the reason for such ripples while using PML?
Thanks in advance

[Taylor Robertson]

Hello
Are you sure that the simulation is not terminating due to auto-shutoff criteria being reached? Can you check the log file, or FDTD object? Maybe disable auto-shutoff, and increase the sim to see if the results change. The material fit looks fine, so I believe that there is a real resonance in that region and you need to run it for longer to allow to allow the impulse response. Transverse PML boundary conditions are not appropriate for plane wave injection, perhaps a TFSF source.
https://support.lumerical.com/hc/en-us/articles/360042703433-Mie-scattering-FDTD-

[sj_123]

Hello I disabled the auto shutoff and increased the simulation time as suggested but still I am getting an abrupt spike in the spectra at wavelength corresponding to the FDTD period (when periodic boundary condition is used). Is there anything else that I can try to sort out this issue?
The figure on right is the structure that I simulated and the one on left is the absorption in the mos2 layer.
Following is the mos2 material fit.

[Taylor Robertson]

Hello If this corresponds to the periodicity of the lattice then it is likely a real physical phenomenon, and it would be worth studying this resonance by using time monitors in this region and frequency monitors tuned to this frequency. There is still significant loss at 0.65um, so you would expect some absorption, but it doesn't follow the material curve exactly due to geometric effects. Likely you are seeing a longer lived mode at this frequency, so it is interacting with the structure more strongly producing a stronger absorption. Periodic BC conditions are effectively replicating an infinite array of these structures, if you change the spacing between the you could possibly tune this resonance. I think you are on the right track.

Best Regards