Number of EME cells required to get transmission spectra of variant structure in propagation direction



Is it possible to simulate 1d photonic crystal waveguide (hole in slab waveguide) by using EME. If so, kindly suggest me the possible example (or) the methodology.


Dear @2014rec9006

Thank you very much for reaching out.

The EME approach is quite powerful for large devices such as long strip waveguides or periodic structures while simulating these geometries are quite challenging and not timely efficient for 3D FTTD. I haven’t done 1D PC waveguide simulations in EME solver, but I assume it will be possible using fine cells so that it captures the change in the device geometry along the waveguide. Note that EME calculates the modes in the middle of each cell not for the entire unit cell.

If your geometry is symmetric, you might take the approach of Bragg Gratings in EME solver and define more than two cells on each period. If it’s not periodic, you will need to simulate the whole geometry by introducing enough cell numbers throughout your geometry.

Another alternative approach for 1D PC waveguide would be to use varFDTD. This is generally a faster approach and can be used to as a sanity check for the results of your EME solvers.

If you are still interested using EME solver, please start preparing your simulation files and feel free to share it with our community.