I am trying to simulate LNOI waveguide ( lithium niobate waveguide on SiO2). During simulation I found that as I increase the core size (0.5um to 0.9um), Neff increases. In commercial available LNOI, LN thickness vary from 500nm to 900nm. Now anyone would choose core size 0.9um because its giving higher Neff, then what is the point of having LNOIs with LN thickness vary from 500nm to 900nm?
- Doubts regarding Mode: How do I find TE mode(for which I want to design waveguide)?Is it mode 2 in simulation, with TE polarization fraction 99?
- Regarding Boundary condition: In attached screenshot I have chosen metal boundary. Should I choose PML instead? If simulation time is not of concern. Because both boundary conditions are giving me different results, metal boundary shows more number of modes supported compared to PML one.
An screenshot of simulation(capture) is attached for reference.
- I did one more simulation with core size 0.8um with PML BC (earlier it was at 0.9um with metal BC) to avoid other modes. In screenshot capture2 you can see there are two TE modes(both are fundamental- with gaussian profile), how is that possible!?
Thanks for reading. Please share your thoughts.
Ambar Shukla
Thank you so much Amirhassan Shams-Ansari for answers.
I am encountering few more issues:
- When I change the span of simulation, number of modes appearing in Mode list of Eigen solver analysis window change. Not only that but the polarization fraction of mode 1 also changes! My only guess was probably modes' order changes when I vary anything (span or core size). So how to track fundamental TE mode when I do the sweeping?
0 votes
0 thanks
- Badges
- Science topic
More Ambar Shukla's questions See All
Similar questions and discussions