Grating coupler for 530nm on SiN waveguide



Hi. I have a question on designing the grating coupler for 530nm on SiN waveguide. The waveguide has a height of 220nm, and the grating coupler is half-etched, with a period of 400nm and 0.5 duty ratio. I run the simulationgrating coupler.fsp (300.7 KB)
, and check the far field projection. I get an output from grating at around 20 degree angle, however, there is another diffraction peak at around -40 degree. Could anyone explain for this -40 degree peak? I am so confused.


Is it possible that is a higher order of diffraction? Like the equation shown on this page, it should be possible for a grating coupler to support more than one diffraction order. For example, if you increase the grating period, you will allow more possible high orders supported by the coupler. I guess it wont be too hard to use the equation and figure out whether this is the case.


Hi kchow, thanks for your reply. I just calculated the equation, it seems that a pitch of 350nm or so would eliminate the higher order diffraction, However, when I was doing simulation, it shows that this additional peak wouldn’t disappear until the pitch decreases to 325 nm, with an I/O angle of 5 degree. I was wondering whether this high order diffraction would strongly affect the performance of the grating coupler? Perhaps I would still choose the 400nm pitch with a larger I/O angle.


I am guessing that the simulation results and theory will come to a better agreement when the simulation result is converged. As far as performance goes, I guess it is more a question that if it will affect the “useful” power output. Obviously, it may affect the power that is being useful but you will be probably a better person to answer this question. From the plot above, you can kind of have an idea how strongly the effect is.