Apr 28 (Wed): "Heterogeneous Integration for Silicon Nitride Photonics," Chao Xiang, ECE PhD Defense
Silicon nitride (Si3N4), as a CMOS compatible material that is widely used in modern IC technology, is emerging as the backbone in a variety of photonic applications including nonlinear photonics, microwave photonics and so on. The next-generation advanced photonic integrated circuits using ultra-low-loss Si3N4 waveguides requires a higher integration level. However, current Si3N4-based devices are still restricted to stand-alone passive devices, due to the significant difficulties of laser integration.
Heterogeneous silicon photonic integration has achieved success in datacenter interconnects by providing efficient III-V gain to passive silicon photonic circuits. The past decade also witnessed dramatic progress in the performances of silicon photonic devices including III-V/Si lasers. One important reason is the reduced optical loss of silicon compared with monolithic III-V waveguides and further developments require heterogeneous integration with ultra-low-loss Si3N4 waveguides.
In this thesis, I will discuss the approach to heterogeneously integrate III-V gain with Si3N4 photonic circuits using a novel multilayer integration structure. This approach results in the first-generation of heterogeneously integrated lasers on (Si3N4) with high temperature stability. With optimization, this integration platform results in high-performance lasers with high power and low noise for fully integrated silicon nitride photonics and is promising to enable Hertz-level linewidth integrated lasers. Moreover, the integration techniques also enabled the first heterogeneously integrated laser soliton microcombs with full electrical control on a monolithic silicon substrate. The demonstrated devices represent state-of-the-art performances of heterogeneous integration. The results in this thesis would open up a new regime of integrated photonics research and enable a whole new class of devices with unprecedented capabilities.
Chao Xiang is a PhD candidate in Prof. John Bowers group.
Hosted by: Professor John Bowers
Submitted by: Chao Xiang <firstname.lastname@example.org>