Related Experiment Video
Updated: Sep 5, 2025
12:19
Measurement of Quantum Interference in a Silicon Ring Resonator Photon Source
Published on: April 4, 2017
8.5K
Ultra-low-energy programmable non-volatile silicon photonics based on phase-change materials with graphene heaters
Zhuoran Fang1, Rui Chen2, Jiajiu Zheng2
1Department of Electrical and Computer Engineering, University of Washington, Seattle, WA, USA. rogefzr@uw.edu.
Nature Nanotechnology
|July 5, 2022
Summary
Researchers developed energy-efficient silicon photonic switches using graphene heaters and phase-change materials. This breakthrough significantly reduces programming energy density for non-volatile photonic devices, paving the way for advanced optical neural networks and quantum computing.
Area of Science:
- Photonics and Materials Science
- Nanotechnology and Device Engineering
Background:
- Silicon photonics is crucial for optical neural networks and quantum information processing.
- Existing non-volatile photonic switches using phase-change materials (PCMs) have high programming energy density.
- Ultra-low programming energy is essential for practical, energy-efficient photonic switches.
Purpose of the Study:
- To demonstrate a non-volatile, electrically reconfigurable silicon photonic platform with ultra-low programming energy.
- To leverage monolayer graphene as an energy-efficient heater for phase-change material (PCM) based photonic devices.
- To achieve significant reductions in programming energy density compared to state-of-the-art devices.
Main Methods:
- Development of a silicon photonic platform incorporating a monolayer graphene heater.
- Integration of technologically mature (Ge2Sb2Te5) and emerging (Sb2Se3) phase-change materials.
- Characterization of photonic switch endurance and programming energy density.
Main Results:
- Demonstrated a non-volatile photonic switch with an endurance exceeding 1,000 cycles.
- Achieved a programming energy density of 8.7 ± 1.4 aJ/nm³, significantly lower than previous technologies.
- The energy density is within an order of magnitude of the theoretical limit for PCMs.
Conclusions:
- Monolayer graphene is a highly efficient and reliable heater for dielectric platforms like Si3N4.
- The developed platform offers a viable solution for energy-efficient, non-volatile programmable silicon photonics.
- This advancement is critical for realizing energy-efficient optical neural networks and quantum information processing.
