Explore the vast range of titles available.

Researchers use phase-change materials to demonstrate an integrated optical memory with 13.4 pJ switching energy. Implementing on-chip non-volatile photonic memories has been a long-term, yet elusive goal. Photonic data storage would dramatically improve performance in existing computing architectures 1 by reducing the latencies associated with electrical memories 2 and potentially eliminating optoelectronic conversions 3 . Furthermore, multi-level photonic memories with random access would allow for leveraging even greater computational capability 4 , 5 , 6 . However, photonic memories 3 , 7 , 8 , 9 , 10 have thus far been volatile. Here, we demonstrate a robust, non-volatile, all-photonic memory based on phase-change materials. By using optical near-field effects, we realize bit storage of up to eight levels in a single device that readily switches between intermediate states. Our on-chip memory cells feature single-shot readout and switching energies as low as 13.4 pJ at speeds approaching 1 GHz. We show that individual memory elements can be addressed using a wavelength multiplexing scheme. Our multi-level, multi-bit devices provide a pathway towards eliminating the von Neumann bottleneck and portend a new paradigm in all-photonic memory and non-conventional computing.

Research in photonic computing has flourished due to the proliferation of optoelectronic components on photonic integration platforms. Photonic integrated circuits have enabled ultrafast artificial neural networks, providing a framework for a new class of information processing machines. Algorithms running on such hardware have the potential to address the growing demand for machine learning and artificial intelligence in areas such as medical diagnosis, telecommunications, and high-performance and scientific computing. In parallel, the development of neuromorphic electronics has highlighted challenges in that domain, particularly related to processor latency. Neuromorphic photonics offers sub-nanosecond latencies, providing a complementary opportunity to extend the domain of artificial intelligence. Here, we review recent advances in integrated photonic neuromorphic systems, discuss current and future challenges, and outline the advances in science and technology needed to meet those challenges.Photonics offers an attractive platform for implementing neuromorphic computing due to its low latency, multiplexing capabilities and integrated on-chip technology.

Electronically reprogrammable photonic circuits based on phase-change chalcogenides present an avenue to resolve the von-Neumann bottleneck; however, implementation of such hybrid photonic–electronic processing has not achieved computational success. Here, we achieve this milestone by demonstrating an in-memory photonic–electronic dot-product engine, one that decouples electronic programming of phase-change materials (PCMs) and photonic computation. Specifically, we develop non-volatile electronically reprogrammable PCM memory cells with a record-high 4-bit weight encoding, the lowest energy consumption per unit modulation depth (1.7 nJ/dB) for Erase operation (crystallization), and a high switching contrast (158.5%) using non-resonant silicon-on-insulator waveguide microheater devices. This enables us to perform parallel multiplications for image processing with a superior contrast-to-noise ratio (≥87.36) that leads to an enhanced computing accuracy (standard deviation σ ≤ 0.007). An in-memory hybrid computing system is developed in hardware for convolutional processing for recognizing images from the MNIST database with inferencing accuracies of 86% and 87%. Hybrid photonic–electronic systems are essential for high-throughput neuromorphic computing. Here, the authors report an in-memory photonic–electronic dot-product engine with decoupled electronic programming of the phase-change memory cells and parallel photonic computation with high-bit operation, low energy consumption, and high accuracy.

Lithium-Niobate-On-Insulator (LNOI) is emerging as a promising platform for integrated quantum photonic technologies because of its high second-order nonlinearity and compact waveguide footprint. Importantly, LNOI allows for creating electro-optically reconfigurable circuits, which can be efficiently operated at cryogenic temperature. Their integration with superconducting nanowire single-photon detectors (SNSPDs) paves the way for realizing scalable photonic devices for active manipulation and detection of quantum states of light. Here we demonstrate integration of these two key components in a low loss (0.2 dB/cm) LNOI waveguide network. As an experimental showcase of our technology, we demonstrate the combined operation of an electrically tunable Mach-Zehnder interferometer and two waveguide-integrated SNSPDs at its outputs. We show static reconfigurability of our system with a bias-drift-free operation over a time of 12 hours, as well as high-speed modulation at a frequency up to 1 GHz. Our results provide blueprints for implementing complex quantum photonic devices on the LNOI platform. The combination of superconducting nanowire single photon detectors and electro-optically reconfigurable circuits in a cryogenic environment is notoriously difficult to reach. Here, the authors realise this on a Lithium-Niobate-On-Insulator platform, reaching high speed modulation at a frequency up to 1 GHz.

Photonic quantum technologies allow quantum phenomena to be exploited in applications such as quantum cryptography, quantum simulation and quantum computation. A key requirement for practical devices is the scalable integration of single-photon sources, detectors and linear optical elements on a common platform. Nanophotonic circuits enable the realization of complex linear optical systems, while non-classical light can be measured with waveguide-integrated detectors. However, reproducible single-photon sources with high brightness and compatibility with photonic devices remain elusive for fully integrated systems. Here, we report the observation of antibunching in the light emitted from an electrically driven carbon nanotube embedded within a photonic quantum circuit. Non-classical light generated on chip is recorded under cryogenic conditions with waveguide-integrated superconducting single-photon detectors, without requiring optical filtering. Because exclusively scalable fabrication and deposition methods are used, our results establish carbon nanotubes as promising nanoscale single-photon emitters for hybrid quantum photonic devices. Single photons are generated from electrically driven semiconducting single-walled carbon nanotubes embedded in a photonic circuit. Pronounced antibunching is observed when photon correlation is measured at cryogenic temperatures.

An important advancement towards optical communication on a chip would be the development of integratable, nanoscale photonic emitters with tailored optical properties. Here we demonstrate the use of carbon nanotubes as electrically driven high-speed emitters in combination with a nanophotonic cavity that allows for exceptionally narrow linewidths. The one-dimensional photonic crystal cavities are shown to spectrally select desired emission wavelengths, enhance intensity and efficiently couple light into the underlying photonic network with high reproducibility. Under pulsed voltage excitation, we realize on-chip modulation rates in the GHz range, compatible with active photonic networks. Because the linewidth of the molecular emitter is determined by the quality factor of the photonic crystal, our approach effectively eliminates linewidth broadening due to temperature, surface interaction and hot-carrier injection. Carbon nanotubes in a nanocavity offer a route to narrow-linewidth on-chip light emitters.

Memristors in electronics have shown the potential for a range of applications, ranging from circuit elements to neuromorphic computing. In recent years, the ability to vary the conductance of a channel in electronics has enabled in-memory computing, thus leading to substantial interest in memristors. Optical analogues will require modulation of the transmission of light in a semicontinuous and nonvolatile manner. With the proliferation of photonic computing, such an optical analogue, which involves modulating the optical response in integrated circuits while maintaining the modulated state afterwards, is being pursued using a range of functional materials. Here we review recent progress in this important and emerging aspect of photonic integrated circuits and provide an overview of the current state of the art. Optical memristors are of particular interest for applications in high-bandwidth neuromorphic computing, machine learning hardware and artificial intelligence, as these optical analogues of memristors allow for technology that combines the ultrafast, high-bandwidth communication of optics with local information processing.Optical analogues of electronic memristors are desirable for applications including photonic artificial intelligence and computing platforms. Here, recent progress on integrated optical memristors is reviewed.

Recent developments in chip-based photonic quantum circuits have radically impacted quantum information processing. However, it is challenging for monolithic photonic platforms to meet the stringent demands of most quantum applications. Hybrid platforms combining different photonic technologies in a single functional unit have great potential to overcome the limitations of monolithic photonic circuits. Our Review summarizes the progress of hybrid quantum photonics integration, discusses important design considerations, including optical connectivity and operation conditions, and highlights several successful realizations of key physical resources for building a quantum teleporter. We conclude by discussing the roadmap for realizing future advanced large-scale hybrid devices, beyond the solid-state platform, which hold great potential for quantum information applications.The Review summarizes the progress of hybrid quantum photonics integration in terms of its important design considerations and fabrication approaches, and highlights some successful realizations of key physical resources for building integrated quantum devices, such as quantum teleporters, quantum repeaters and quantum simulators.

The ability to control mechanical motion with optical forces has made it possible to cool mechanical resonators to their quantum ground states. The same techniques can also be used to amplify rather than reduce the mechanical motion of such systems. Here, we study nanomechanical resonators that are slightly buckled and therefore have two stable configurations, denoted ‘buckled up’ and ‘buckled down’, when they are at rest. The motion of these resonators can be described by a double-well potential with a large central energy barrier between the two stable configurations. We demonstrate the high-amplitude operation of a buckled resonator coupled to an optical cavity by using a highly efficient process to generate enough phonons in the resonator to overcome the energy barrier in the double-well potential. This allows us to observe the first evidence for nanomechanical slow-down and a zero-frequency singularity predicted by theorists. We also demonstrate a non-volatile mechanical memory element in which bits are written and reset by using optomechanical backaction to direct the relaxation of a resonator in the high-amplitude regime to a specific stable configuration. A nanoscale optomechanical resonator in a double-well potential has been excited into the high-amplitude regime, and then optically cooled to a specific at-rest configuration, which allows it to be operated as a non-volatile memory element.

Efficient fiber-to-chip couplers for multi-port access to photonic integrated circuits are paramount for a broad class of applications, ranging, e.g., from telecommunication to photonic computing and quantum technologies. Grating-based approaches are often desirable for providing out-of-plane access to the photonic circuits. However, on photonic platforms characterized by a refractive index ≃ 2 at telecom wavelength, such as silicon nitride or thin-film lithium niobate, the limited scattering strength has thus far hindered the achievement of coupling efficiencies comparable to the ones attainable in silicon photonics. Here we present a flexible strategy for the realization of highly efficient grating couplers on such low-index photonic platforms. To simultaneously reach a high scattering efficiency and a near-unitary modal overlap with optical fibers, we make use of self-imaging gratings designed with a negative diffraction angle. To ensure high directionality of the diffracted light, we take advantage of a metal back-reflector patterned underneath the grating structure by cryogenic deep reactive ion etching of the silicon handle. Using silicon nitride as a testbed material, we experimentally demonstrate coupling efficiency up to − 0.55 dB in the telecom C-band with high chip-scale device yield.