Answers are grounded in this paper and cite their sources.
Experimental demonstration of non-volatile switches in photonic circuits, suggesting energy-efficient neuromorphic computing applications.
Photonic integrated circuits (PICs) can deliver unparalleled performance for future neuromorphic computing applications. Such neuromorphic PICs require a large number of tunable switches, which are typically realized with current‐controlled heaters, resulting in considerable energy consumption. Non‐volatile photonic devices based on phase change materials (PCMs) can overcome this challenge, promising zero power consumption during operation. In this work, we experimentally demonstrate non‐volatile Mach‐Zehnder switches (MZI) utilizing the phase change material antimony triselenide (Sb 2 Se 3 ). The PCMs are controllably switched with graphene heaters from the amorphous to the crystalline state. This leads to a π‐shift of 0.7, a V π L ≈ 0.56 Vcm, and a shift per micrometer of Sb 2 Se 3 of 0.014 π/µm, and a high extinction ratio of 28 dB. Additionally, we perform state‐of‐the‐art simulations to predict the temperature in the devices and to support our measurements. The data demonstrates that switches based on graphene heaters and Sb 2 Se 3 are suitable for achieving large phase shifts while avoiding high thermal budgets as for doped silicon heaters. Moreover, our devices were fabricated on wafer‐scale enabling scalability of the fabrication process in the future as it is beneficial for advanced applications like neuromorphic computing, integrated optical memory, tunable filter arrays and delay lines with permanent trimming and optical FPGAs based on phase change materials.
No takes yet. Share an insight, caveat, or question.
Samland et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: