Nonvolatile light-field manipulation via electrically-driven phase transition of chalcogenide phase change materials (PCMs) is regarded as one of the most powerful solutions to low-power-consumption and compact integrated reconfigurable photonics. However, before the breakthrough in large-scale integration approaches linked to wafer foundries, phase-change non-volatile reconfigurable photonics could hardly see their widespread practical applications. Here we demonstrate nonvolatile photonic devices fabricated by back-end-of-line (BOEL) integration of PCMs into the commercial silicon photonics platform. A narrow trench etched into the BOEL dielectric layer exposed the waveguide core and allowed for the direct deposition of various PCM films on the waveguide in the functional areas. Fine-tuning the nonvolatile phase transition of Sb2Se3 via a PIN microheater was verified by realizing the post-fabrication trimming of silicon photonic devices. Our work highlights a reliable platform for large-scale PCM-integrated photonics and validates its precise nonvolatile reconfigurability.
Optical neural networks (ONNs), enabling low latency and high parallel data processing without electromagnetic interference, have become a viable player for fast and energy-efficient processing and calculation to meet the increasing demand for hash rate. Photonic memories employing nonvolatile phase-change materials could achieve zero static power consumption, low thermal cross talk, large-scale, and high-energy-efficient photonic neural networks. Nevertheless, the switching speed and dynamic energy consumption of phase-change material-based photonic memories make them inapplicable for in situ training. Here, by integrating a patch of phase change thin film with a PIN-diode-embedded microring resonator, a bifunctional photonic memory enabling both 5-bit storage and nanoseconds volatile modulation was demonstrated. For the first time, a concept is presented for electrically programmable phase-change material-driven photonic memory integrated with nanosecond modulation to allow fast in situ training and zero static power consumption data processing in ONNs. ONNs with an optical convolution kernel constructed by our photonic memory theoretically achieved an accuracy of predictions higher than 95% when tested by the MNIST handwritten digit database. This provides a feasible solution to constructing large-scale nonvolatile ONNs with high-speed in situ training capability.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.