The rapid growth of AI workloads is pushing electronic chips toward fundamental limits in interconnect bandwidth and energy efficiency. We engineer photonic integrated circuits (PICs) on silicon, silicon nitride, III-V, and thin-film lithium niobate platforms, aiming to process optical information directly on chip. Our research targets high-bandwidth optical interconnects, optical signal processing, and photonic computing architectures that could relieve the bandwidth and energy bottlenecks of AI hardware.
Semiconductor lasers are everywhere, yet their beam quality and power are ultimately set by the cavity. We design photonic-crystal surface-emitting lasers (PCSELs)—including topological cavity designs—to achieve single-mode operation, narrow linewidth, and high power in a compact, wafer-scale format, targeting LiDAR, coherent optical communications, and precision metrology.
When light and matter couple strongly enough, they no longer behave as separate entities: exciton-polaritons—half-light, half-matter quasiparticles—form and can condense into macroscopic quantum states. We study this physics in van der Waals materials (TMDs, moiré superlattices, magnetic materials) and engineered photonic lattices, aiming to realize and control polariton Bose–Einstein condensation, quantum fluids, vortices, and supersolids, ideally at room temperature. Beyond fundamental many-body physics, these hybrid states may enable new coherent light sources and quantum simulators.
Nanoscale mechanical motion couples naturally to light, electricity, and heat. We design nano-electromechanical systems (NEMS) with engineered—including topologically protected—vibrational modes, aiming at ultra-high frequencies and robust operation, with routes toward RF signal processing, ultrasensitive sensing, and quantum phononics.
