Learning Forum

[Jordi Lopez Leyva]

I'm trying to simulate 1D photonic crystal (PhC) Nanobeam of GaAs in between SiO2 and a polymer and implement the nonlinear Pockels Chi2. The simulation seems to work for low powers, but for higher powers, light is no longer guided and radiates out of the photonic crystal. 

I'm using 2 electric dipoles in 3D FDTD with amplitude = 1e6, and base amplitude = 2.026464e-30, such that total amplitude = 2.02642e-24. I calculated power with "dipolesource" such that power = 1.5mW. I'm using very high resolution mesh, standard PML boundaries, 1.55um center wavelength

1. Am I calculating the dipole powers correctly?
2. Do I need to change boundary conditions? e.g. metal?

Movie looks as follows:

[Amrita Pati]

Hi Jordi,

1. You mentioned that you calculated the power using the "dipolesource". I don't think there is a script command called "dipolesource". I guess you are either using dipolepower or sourcepower. The sourcepower script command does not give accurate result in the presence of multiple sources as it doesn't take into account the interference between the dipole sources. But if you are using dipolepower or a transmission box, you should be able to obtain the power accurately.
2. Can you share a screenshot of the model with the simulation region? It's hard to predict if the simulation parameters are correct without looking at how the geometry has been set up.

Regards,
Amrita

[Jordi Lopez Leyva]

 

Hi Amrita,

I indeed used dipolepower. Could the issue be that too high power is used, and this gives the impression of no confiment in the movie due to bad color separations of different powers? Here’s the screenshot with nanobeam selected and 2 dipoles in the center, one perpendicular and another 60deg (both are inside an air hole). The GaAs nanobeam is Chi2 and has air holes in the center of equal radius and spacing: 

 

 

[Amrita Pati]

Hi Jordi,

I apologize for the late reply. With nonlinear simulation, it is unfortunately very hard to predict what could be causing the issue. It is possible that the high amplitude is causing an instability but at the same time it could also be other parameters. Have you had a chance to look at the following article: Nonlinear and gain methodology? If not, have a look and make sure you are accounting for the extra considerations in nonlinear simulations. Also, any chance you are using a script to setup the simulation that you can share here?

Regards,
Amrita

 

[Jordi Lopez Leyva]

Hi Amrita,

Let's ignore the issue I have with non-linear GaAs for now, because I think the problem comes from elswhere.

I've a related question to the problem regarding light confinement. I've run the simulation with linear GaAs - Palik (so no chi2/chi3). But I still seem to have the same issue in the movie that light is not confined for high intensity dipole. For amplitude ~ 1-10 light is confined, but when amplitude ~100 it's no longer confined in the movie.

I've checked with DFT E field monitor against x position, and the line plot looks exactly the same for both high and low intensity, where I just divide the high intensity plot by 100 to get the exact same result. This suggests to me the issue is with the movie (e.g. color scaling, something else?).

Could you please let me know what you think could be behind this issue with the movie?

[Amrita Pati]

Hi Jordi,

That's a great point. It is likely that with the increased amplitude the color bar changes in such a way that it incorrectly looks like the mode is not confined. I ran some test simulations and I see similar behavior as well. For amplitudes of 1 and 10^3, I ran three simulations. The first is just a simple simulation with an amplitude of 1, and the second with an amplitude of 10^3. In the third, also for an amplitude of 10^3, I recorded the intensity data scaled by 10^6. 

The recordings in the first and the last case look almost identical. I tried uploading some gifs but their size is too large to be accepted here. So, I think the issue is because of the range of values in the color bar, and not the confinement itself. But I am not sure if this simple scaling would work in all cases, espcially in nonlinear simulations. I will spend some more time on this or maybe get some insights from my colleagues and get back to you with more information. In the mean time, what happens when you scale the data for your movie monitor?

Regards,
Amrita

[Jordi Lopez Leyva]

Hi Amrita,

Thank you! The problem indeed seems to be the scaling. This solves the movie issue for both linear and Chi2 GaAs photonic crystal for amplitudes between 1-10^3 (higher powers diverge, but i'll ask about this in a different question)