How to add a gain medium in the FDTD simulation?


I am interesting in the influence of gain medium (such as InGaAsP) filled in a side coupled cavity on the transmission and reflection spectra (Gain-induced switching in metal-dielectric-metal plasmonic waveguides). I try to add a gain medium in FDTD with a negative index in material database with type of sample data (see the figure) to repeat the results in reference paper. The subject is based on the metal-dielectric-metal waveguide and the width of dielectric is 50 nm, so the mesh is set to be dx=2nm, dy=2nm (2D model). What makes me puzzled is the simulation always show that “Warning! The simulation that created the data in the monitors and sources be …” I try to increase the layers of pml, the ratio of kappa and sigma, but the error is still…
So what can I do to solve this prolem?
Thanks !

How to eliminate the fluctuations on the transmission spectrum and make the curve become more smooth?

Dear @youngjianhua

To add a gain material in fdtd simulations, please see this page. Generally speaking, there are two ways to do this: one is to use the Lorentz model or you can import your data file, and the idea in both cases is to have negative imaginary refractive index to simulate a gain medium.

Regarding your diverging problem, this happens when your electric field reaches the auto shutoff max. You can change this value to a bigger number to run the simulation for a longer time. Also, there are a few other tips that might help you to overcome the diverging problem. Let me know if you could solve the problem, and I can take a look at your simulation file if needed.


Dear Behzad Khanaliloo
model.fsp (390.7 KB)

Thank you for your help, I am very appreciated for your advice about diverging problem and I can overcome this problem by changing some parameters, include simulation time, pml layers, kappa, sigma, and the value of auto shutoff max. My model is enclosed and you can check these parameters in my simulation file. The result of simulation is agree well with the reference paper, but I want to sure whether my simulation way is correct or not. If you have any advice, please let me know.
Another problem is about adding gain materials. In my simulation file, I added a gain material named “Gain medium”, it was used the method of Sampled data and some experimental data were used. While the questions is with the change of wavelength, the index is fixed (include real and imaginary parts). So this is a correct way to realize gain characteristic?

Thanks again.


Hi @youngjianhua

I am glad that you could solve the problem.

For cases where you have a broadband source, using a sampled data from material database and then importing Re(index) and Im(index), even if they are constant over the wavelengths of interest, is the correct approach. (n.k) Material has to be used only if you have a single wavelength source. If you need to know more about the differences between different material choose, watching this webinar on material database will be quite a good idea.


This topic was automatically closed after 2 days. New replies are no longer allowed.