2.5d photonic crystal slab waveguide sharp bend


Dear Lumerical technical support,

I followed examples from https://kb.lumerical.com/en/diffractive_optics_waveguide_pcbandstructure.html
and design a slab photonic crystal with square lattice. Having modified the script now I scan the entire bandstructure and create find large bandgap at large frequency range.

1.Then I build a line defect photonic crystal waveguide and butt coupler and it works well for 1.5-1.6um.
However it does show transmission oscillation at ~ 1.3um if source bandwidth covers that and at that wavelength there should also be a bandgap. Why this can happen?

2.I am using such photonic crystal waveguide only at 1.5-1.6um and try to create a sharp bend. However a 90 degree bend shows strong reflection and scattering inside lattice and transmission is very low.

However it seems from other software like COMSOl with following link: FEM results did not sow such reflection

2.5d bend1.lms (432.1 KB)

2.5d straight.lms (430.9 KB)


There can still be some transmission at band gap frequencies if there are relatively few periods of the periodic lattice. As a test, I tried running the straight waveguide case with the original number of periods and double the number of periods and I was able to see the transmission reduce in the range between about 1.15-1.45 um.

I also checked the images from the COMSOL simulation that you linked and it looks like it includes many more periods of the device in both bent and straight cases, so I think that could be the difference. Hopefully this helps!