I'm having a long discussion with my boss concerning with the optimal design of a slot waveguide EO modulator. He considers that the maxmization of the optical confinement factor into the slot has the highest priority with respect to the minimization of the optical mode effective area, as the final aim is to maximise the overlap with the RF field into the slot, whereas the minimum effective area impacts mainly in devices based on light-to-light interactions, such as in all-optical modulators. Nevertheless, almost all the publications on this topic reports on efforts to minimize the effective area. What's your opinion about it ?
Depence on your application. Do you develope an EO modulator making use of the Pockels effect or of the optical Kerr effect? In case of the Pockels effect it is better to optimize the field confinement factor since it is direct proportional to the refractive index change. In case of the optical Kerr effect it is better to optimize the effective nonlinear area because it is inversely proportional to the refractive index change.
You need to minimize the effective nonlinear area since it is indirect proportional to the refractive index change. Maybe this conference article can be helpful for you:
Conference Paper Silicon-On-Insulator Slot-Waveguide Design Trade-Offs
I really thank you, Patrick.
I am developing a Pockels based EO modulator.
I already read your very interesting paper from Photoptics2015, that you also posted in the last week for my other question, but as in Karlsruhe's papers, it is only said that "the lower the effective area, the higher the nonlinear interaction", without details on the device to be based on the Kerr effect. Even referring to the M3 paper from Gigoptix, they do not specify that it is a Kerr effect material. So, I am wondering if it is something that can be understood anyway and I am missing.
P.S.
I see you had a period in the Physics Dept. at Tor Vergata with Casalboni and Paolo Prosposito. I had some contacts with them, even if I was from Electronic Engineering Dept. in Tor Vergata.
I had a deeper reading and some calculations starting from your replies. As Patrick said, the aim with Kerr effect materials is to minimize the mode effective area Aeff, as the non linear interaction factor is maximized by minimized Aeff. This is true as Kerr materials are never used for standard EO modulators but only for all-optical non linearities, as frequency combs or FWM. Checking better the publications from the Karlsruhe KIT group, they use standard Pockels effect polymers for modulators and Kerr effect small molecules for all-optical interaction based devices.
Indeed, if we used a Kerr effect material into EO modulators and we derived the half wave voltage, we'd have a linear dependence on the inverse of the confinement factor (as we need anyway to overlap the driving RF electric field to the optical field) and only a square root dependence on the effective area (square root as Kerr is a quadratic effect). So this would mean that the half wave voltage would be improved more by the maximization of the optical confinement rather than by the minimization of the effective area.
- Badges
- Science method