Nonreciprocal optical devices are key components in photonic integrated circuits for light reflection blocking and routing. Most reported silicon integrated nonreciprocal optical devices to date were unit devices. To allow complex signal routing between multiple ports in photonic networks, multi-port magneto-optical (MO) nonreciprocal photonic devices are desired. In this study, we report experimental demonstration of a silicon integrated 5×5 nonreciprocal optical router based on a magneto-optical phased array. By introducing different nonreciprocal phase shifts to planar photonic waveguides, the device focuses light to different ports for both forward and backward propagation directions. The device shows designable nonreciprocal optical transmission between 5×5 ports, achieving 16 dB isolation ratio and -18 dB crosstalk.

Programmable photonic integrated circuits (PICs) have emerged as a promising platform for analog signal processing. Programmable PICs, as versatile photonic integrated platforms, can realize a wide range of functionalities through software control. However, a significant challenge lies in the efficient management of a large number of programmable units, which is essential for the realization of complex photonic applications. In this paper, we propose an innovative approach using Ising-model-based intelligent computing to enable dynamic reconfiguration of large-scale programmable PICs. In the theoretical framework, we model the Mach–Zehnder interferometer (MZI) fundamental units within programmable PICs as spin qubits with binary decision variables, forming the basis for the Ising model. The function of programmable PIC implementation can be reformulated as a path-planning problem, which is then addressed using the Ising model. The states of MZI units are accordingly determined as the Ising model evolves toward the lowest Ising energy. This method facilitates the simultaneous configuration of a vast number of MZI unit states, unlocking the full potential of programmable PICs for high-speed, large-scale analog signal processing. To demonstrate the efficacy of our approach, we present two distinct photonic systems: a 4×4 wavelength routing system for balanced transmission of four-channel NRZ/PAM-4 signals and an optical neural network that achieves a recognition accuracy of 96.2%. Additionally, our system demonstrates a reconfiguration speed of 30 ms and scalability to a 56×56 port network with 2000 MZI units. This work provides a groundbreaking theoretical framework and paves the way for scalable, high-speed analog signal processing in large-scale programmable PICs.

A GeSn nanostrip grown by the rapid melting growth method has gradient Sn content along the strip, a very attractive approach for making an infrared broad-spectrum light source. In this work, by applying the Sn content distribution strategy, GeSn shortwave infrared light-emitting diodes (LEDs) arrays with a size of 3 μm×2 μm were fabricated on Si substrate, and the active layer Sn content increased from 2.1% to 5.2% to form a broadband light source. The GeSn LEDs show perfect rectifying behavior about 106 for ±1 V, and room temperature electroluminescence (EL) from the direct bandgap was achieved. The super-linear dependence between the injected current and EL intensity confirms the band-to-band radiative recombination. By utilizing Sn content gradient technology, the EL spectra of Sn gradient GeSn LED arrays can cover from 1600 to 2200 nm with a full width at half-maximum of about 340 nm. These results show a novel method for preparing broad-spectrum shortwave infrared light emitters on a Si chip.

Reconfigurable silicon microrings have garnered significant interest for addressing challenges in artificial intelligence, the Internet of Things, and telecommunications due to their versatile capabilities. Compared to electro-optic (EO) and thermo-optic (TO) devices, emerging micro-electromechanical systems (MEMS)-based reconfigurable silicon photonic devices actuated by electrostatic forces offer near-zero static power consumption. This study proposes and implements novel designs for fully reconfigurable silicon photonic MEMS microrings for high-speed dense wavelength division multiplexing (DWDM) elastic networks. The designs include an all-pass microring with a 7 nm free spectral range (FSR) and full-FSR resonance tuning range, an add-drop microring with a 3.5 nm FSR and full-FSR tuning range, and an add-drop double-microring with a 34 nm FSR, wide-range discrete resonance tunability, and flat-top tunability. These advancements hold promise for practical applications.

Silicon photonics (SiPh) technology has become a key platform for developing photonic integrated circuits due to its CMOS compatibility and scalable manufacturing. However, integrating efficient on-chip optical sources and in-line amplifiers remains challenging due to silicon’s indirect bandgap. In this study, we developed prefabricated standardized InAs/GaAs quantum-dot (QD) active devices optimized for micro-transfer printing and successfully integrated them on SiPh integrated circuits. By transfer-printing standardized QD devices onto specific regions of the SiPh chip, we realized O-band semiconductor optical amplifiers (SOAs), distributed feedback (DFB) lasers, and widely tunable lasers (TLs). The SOAs reached an on-chip gain of 7.5 dB at 1299 nm and maintained stable performance across a wide input power range. The integrated DFB lasers achieved waveguide (WG)-coupled output powers of up to 19.7 mW, with a side-mode suppression ratio (SMSR) of 33.3 dB, and demonstrated notable robustness against optical feedback, supporting error-free data rates of 30 Gbps without additional isolators. Meanwhile, the TLs demonstrated a wavelength tuning range exceeding 35 nm, and a WG-coupled output power greater than 3 mW. The micro-transfer printing approach effectively decouples the fabrication of non-native devices from the SiPh process, allowing back-end integration of the III–V devices. Our approach offers a viable path toward fully integrated III–V/SiPh platforms capable of supporting high-speed, high-capacity communication.

Photonic integrated switches that are both space and wavelength selective are a highly promising technology for data-intensive applications as they benefit from multi-dimensional manipulation of optical signals. However, scaling these switches normally poses stringent challenges such as increased fabrication complexity and control difficulties, due to the growing number of switching elements. In this work, we propose a new type of dilated crosspoint topology, which efficiently handles both space and wavelength selective switching, while reducing the required switching element count by an order of magnitude compared to reported designs. To the best of our knowledge, our design requires the fewest switching elements for an equivalent routing paths number and it fully cancels the first-order in-band crosstalk. We demonstrate such an ultra-compact space-and-wavelength selective switch (SWSS) at a scale of 4×4×4λ on the silicon-on-insulator (SOI) platform. Experimental results reveal that the switch achieves an insertion loss ranging from 2.3 dB to 8.6 dB and crosstalk levels in between -35.3 dB and -59.7 dB. The add-drop microring-resonators (MRRs) are equipped with micro-heaters, exhibiting a rise and fall time of 46 μs and 0.33 μs, respectively. These performance characteristics highlight the switch’s ultra-low element count and crosstalk with low insertion loss, making it a promising candidate for advanced data center applications.

An optical phased array (OPA) featuring all-solid-state beam steering is a promising component for light detection and ranging (LiDAR). There exists an increasing demand for panoramic perception and rapid target recognition in intricate LiDAR applications, such as security systems and self-driving vehicles. However, the majority of existing OPA approaches suffer from limitations in field of view (FOV) and do not explore parallel scanning, thus restricting their potential utility. Here, we combine a two-dimensional (2D) grating with an FOV-synthetization concept to design a silicon-based top-facing OPA for realizing a wide cone-shaped 360° FOV. By utilizing four OPA units sharing the 2D grating as a single emitter, four laser beams are simultaneously emitted upwards and manipulated to scan distinct regions, demonstrating seamless beam steering within the lateral 360° range. Furthermore, a frequency-modulated dissipative Kerr-soliton (DKS) microcomb is applied to the proposed multi-beam OPA, exhibiting its capability in large-scale parallel multi-target coherent detection. The comb lines are spatially dispersed with a 2D grating and separately measure distances and velocities in parallel, significantly enhancing the parallelism. The results showcase a ranging precision of 1 cm and velocimetry errors of less than 0.5 cm/s. This approach provides an alternative solution for LiDAR with an ultra-wide FOV and massively parallel multi-target detection capability.

We present hybrid tunable lasers at 2.0-μm wavelength, seamlessly integrated within silicon photonic circuits for advanced biomedical applications. Leveraging III/V semiconductor materials for gain and silicon ring resonators for tuning, the laser achieves a tuning range of 25 nm, precise adjustments below 0.1 nm, and a side-mode suppression ratio of 40 dB. This advancement contributes to the progress in photonic integrated circuits beyond the telecommunication wavelength range, offering scalable and cost-effective solutions for enhanced spectroscopic systems within the 2.0-μm wavelength range.

Artificial intelligence (AI), owing to its substantial computing demands, necessitates computing hardware that offers both high speed and high power efficiency. A silicon photonic integrated circuit shows promise as a hardware solution due to its attributes, including high power efficiency, low latency, large bandwidth, and complementary metal–oxide–semiconductor (CMOS) compatibility. Here, we propose a silicon photonic convolution accelerator (SiPh-CA) and experimentally realize a prototype with sub-integrated coherent transmit–receive optical sub-assemblies (sub-IC-TROSAs). The prototype, compared to a previous IC-TROSA-based convolution accelerator, achieves almost the same performances of 1.024 TOPS/channel and 96.22% inference accuracy when it processes neural networks for image recognition, using half the numbers of the modulators and the drivers with which over 1/3 chip footprint and 37.01% power consumption are reduced. By incorporating a broadcasting scheme based on splitters and combiners, the approach can efficiently process multiple convolutions in parallel, achieving several tera operations per second. This scalability feature allows the SiPh-CA to process complex AI and high-performance computing tasks.

Integrated high-linearity modulators are crucial for high dynamic-range microwave photonic (MWP) systems. Conventional linearization schemes usually involve the fine tuning of radio-frequency (RF) power distribution, which is rather inconvenient for practical applications and can hardly be implemented on the integrated photonics chip. In this paper, we propose an elegant scheme to linearize a silicon-based modulator in which the active tuning of RF power is eliminated. The device consists of two carrier-depletion-based Mach–Zehnder modulators (MZMs), which are connected in series by a 1×2 thermal optical switch (OS). The OS is used to adjust the ratio between the modulation depths of the two sub-MZMs. Under a proper ratio, the complementary third-order intermodulation distortion (IMD3) of the two sub-MZMs can effectively cancel each other out. The measured spurious-free dynamic ranges for IMD3 are 131, 127, 118, 110, and 109 dB·Hz6/7 at frequencies of 1, 10, 20, 30, and 40 GHz, respectively, which represent the highest linearities ever reached by the integrated modulator chips on all available material platforms.

Polarization-insensitive optical modulators allow an external laser to be remotely interconnected by single-mode optical fibers while avoiding polarization controllers, which would be convenient and cost-effective for co-packaged optics, 5G, and future 6G applications. In this article, a polarization-insensitive silicon intensity modulator is proposed and experimentally demonstrated based on two-dimensional centrally symmetric gratings, featuring a low polarization-dependent loss of 0.15 dB in minimum and polarization insensitivity of eye diagrams. The device exhibits a low fiber-to-fiber insertion loss of 9 dB and an electro-optic (EO) bandwidth of 49.8 GHz. A modulation speed of up to 224 Gb/s is also demonstrated.

Optical interconnects based on photonic integrated circuits (PICs) are emerging as a pivotal technology to address the relentless surge in data traffic driven by compute-intensive applications. Combining mode-division multiplexing (MDM) with wavelength-division multiplexing (WDM) offers a compelling approach to significantly enhance the shoreline density of optical interconnects. However, existing on-chip MDM systems encounter considerable challenges in simultaneously achieving a large optical bandwidth, multi-band operation, and ultra-compactness, thereby limiting scalability as conventional telecom band resources become increasingly constrained. Here we introduce, to our knowledge, the first inverse-designed multi-band mode multiplexer (MUX) utilizing a digital metamaterial structure to support the first three-order TE modes. The proposed device features an ultra-compact footprint of 6 μm×4.8 μm and exhibits an exceptionally flat spectral response, with numerical simulations confirming spectral variations of less than 0.94 dB across the 1500–2100 nm range. Experimental results further validate its performance, demonstrating insertion losses below 4.3 dB and 4.0 dB, and crosstalk below -11.6 dB and -11.3 dB, within the 1525–1585 nm and 1940–2040 nm bands, respectively. Additionally, system-level optical interconnect experiments using a multi-band MDM circuit successfully achieve single-wavelength transmission rates of 3-modes×180 Gb/s at the 1.55 μm band and record-setting 3-modes×114 Gb/s in the 2 μm band. This work highlights the transformative potential of employing multi-band MDM technology to enhance bandwidth density and scalability, providing a robust foundation for next-generation high-capacity on-chip optical interconnects.

Due to the outstanding anti-interference capability against the ambient noise, LiDARs based on frequency-modulated continuous wave (FMCW) technology with high sensitivity and high signal-to-noise ratio (SNR) are essential to achieve ideal photodetection of weak light. To significantly improve the weak light detection performance of balanced photodetectors, this work first demonstrates a novel near-infrared germanium-on-silicon (Ge/Si) avalanche photodetector with a three-electrode balanced scheme. The single three-electrode avalanche photodetector exhibits a high responsivity of >200 A/W near breakdown voltage. The three-electrode balanced avalanche photodetector (3ele-BAPD) achieves a common-mode rejection ratio (CMRR) of 50 dB at an operating wavelength of 1550 nm. We have set up the FMCW coherent detection system. The minimum detectable power of -93 dBm can be achieved, corresponding to an SNR of 3.2 dB and a detection probability of 54%. In comparison, the performance exceeds that of the two-electrode balanced avalanche photodetector (2ele-BAPD), which exhibits a minimum detectable power of -85 dBm with a corresponding SNR of 3.1 dB and a detection probability of 51%. The superior weak light detection performance enables the 3ele-BAPD to accomplish 3D imaging based on the FMCW LiDAR scheme. Moreover, the 3ele-BAPD is also applied to velocity measurement for 4D sensing. The applications of LiDAR velocity measurement and imaging are verified.

Swept laser interferometry is an extremely powerful solution embedded in several recent technologies such as absolute distance measurement, light detection and ranging (LiDAR), optical frequency domain reflectometry, optical coherence tomography, microresonator characterization, and gas spectroscopy. Nonlinearity in the optical frequency sweeping of tunable lasers is a fatal drawback in gaining the expected outcome from these technologies. Here, we introduce an on-chip, millimeter-scale, 7 m spiral resonator that is made of ultralow-loss Si3N4 to act as a frequency ruler for correction of the tunable lasers sweeping nonlinearities. The sharp 2 MHz frequency lines of the 8.5×107 high-quality factor resonator and the narrow-spaced 25.566 MHz frequency ticks of the 7 m spiral allow unprecedented precision for an on-chip solution to correct the laser sweeping nonlinearity. Accurate measurements of the ruler’s frequency spacing, linewidth, and temperature and wavelength sensitivities of the frequency ticks are performed here to demonstrate the quality of the frequency ruler. In addition, the spiral resonator is implemented in an frequency-modulated continuous-wave LiDAR experiment to demonstrate a potential application of the proposed on-chip frequency ruler.

Optical antennas play a pivotal role in interfacing integrated photonic circuits with free-space systems. Designing antennas for optical phased arrays ideally requires achieving compact antenna apertures, wide radiation angles, and high radiation efficiency all at once, which presents a significant challenge. Here, we experimentally demonstrate a novel ultra-compact silicon grating antenna, utilizing subwavelength grating nanostructures arranged in a transversally interleaved topology to control the antenna radiation pattern. Through near-field phase engineering, we increase the antenna’s far-field beam width beyond the Fraunhofer limit for a given aperture size. The antenna incorporates a single-etch grating and a Bragg reflector implemented on a 300-nm-thick silicon-on-insulator (SOI) platform. Experimental characterizations demonstrate a beam width of 44°×52° with -3.22 dB diffraction efficiency, for an aperture size of 3.4 μm×1.78 μm. Furthermore, to the best of our knowledge, a novel topology of a 2D antenna array is demonstrated for the first time, leveraging evanescently coupled architecture to yield a very compact antenna array. We validated the functionality of our antenna design through its integration into this new 2D array topology. Specifically, we demonstrate a small proof-of-concept two-dimensional optical phased array with 2×4 elements and a wide beam steering range of 19.3º × 39.7º. A path towards scalability and larger-scale integration is also demonstrated on the antenna array of 8×20 elements with a transverse beam steering of 31.4º.

An integrated photonic circuit architecture to perform a modified-convolution operation based on the discrete fractional Fourier transform (DFrFT) is introduced. This is accomplished by utilizing two nonuniformly coupled waveguide lattices with equally spaced eigenmode spectra, the lengths of which are chosen so that the DFrFT and its inverse operations are achieved. A programmable modulator array is interlaced so that the required fractional convolution operation is performed. Numerical simulations demonstrate that the proposed architecture can effectively perform smoothing and edge detection tasks even for noisy input signals, which is further verified by electromagnetic wave simulations. Notably, mild lattice defects do not jeopardize the architecture performance, showing its resilience to manufacturing errors.

The progress of on-chip optical communication relies on integrated multi-dimensional mode (de)multiplexers to enhance communication capacity and establish comprehensive networks. However, existing multi-dimensional (de)multiplexers, involving modes and wavelengths, face limitations due to their reliance on single-directional total internal reflection and multi-level mode conversion based on directional coupling principles. These constraints restrict their potential for full-duplex functionality and highly integrated communication. We solve these problems by introducing a photonic-like crystal-connected bidirectional micro-ring resonator array (PBMRA) and apply it to duplex mode-wavelength multiplexing communication. The directional independence of total internal reflection and the cumulative effect of the subwavelength-scale pillar within the single-level photonic crystal enable bidirectional mode and wavelength multiplexed signals to transmit among multi-pair nodes without interference, improving on-chip integration in single-level mode conversion. As a proof of concept, we fabricated a nine-channel bidirectional multi-dimensional (de)multiplexer, featuring three wavelengths and three TE modes, compactly housed within a footprint of 80 μm×80 μm, which efficiently transmits QPSK-OFDM signals at a rate of 216 Gbit/s, achieving a bit error rate lower than 10-4. Leveraging the co-ring transmission characteristic and the orthogonality of the mode-wavelength channel, this (de)multiplexer also enables a doubling of communication capacity using two physical transmission channels.

Whispering gallery mode optical microresonators represent a promising avenue for realizing optical analogs of coherent light–atom interactions, circumventing experimental complexities. All-optical analogs of Autler–Townes splitting have been widely demonstrated, harnessing coupled optical microresonators, also known as photonic molecules, wherein the strong coupling between resonant fields enables energy level splitting. Here, we report the characterizations of Autler–Townes splitting in waveguide-coupled microring dimers featuring mismatched sizes. By exploiting backscattering-induced coupling via Rayleigh and Mie scatterers in individual rings, high-order Autler–Townes splitting has been realized, yielding supermode hybridization in a multi-level system. Upon resonance detuning using an integrated phase shifter, intra-cavity coupling-induced splitting becomes almost indistinguishable at the zero-detuning point where the strong inter-cavity coupling counteracts the imbalance of backscattering strengths in individual rings. Through demonstrations on the maturing silicon photonics platform, our findings establish a framework of electrically tunable photonic molecules for coupling-mediated Autler–Townes splitting, offering promising prospects for on-chip signal generation and processing across classical and quantum regimes.

On-chip spectrometers with high compactness and portability enable new applications in scientific research and industrial development. Fourier transform (FT) spectrometers have the potential to realize a high signal-to-noise ratio. Here we propose and demonstrate a generalized design for high-performance on-chip FT spectrometers. The spectrometer is based on the dynamic in-plane reconfiguration of a waveguide coupler enabled by an integrated comb-drive actuator array. The electrostatic actuation intrinsically features ultra-low power consumption. The coupling gap is crucial to the spectral resolution. The in-plane reconfiguration surmounts the lithography accuracy limitation of the coupling gap, boosting the resolution to 0.2 nm for dual spectral spikes over a large bandwidth of 100 nm (1.5–1.6 μm) within a compact footprint of 75 μm×1000 μm. Meanwhile, the in-plane tuning range can be large enough for arbitrary wavelengths to ensure the effectiveness of spectrum reconstruction. As a result, the proposed spectrometer can be easily transplanted to other operation bands by simply scaling the structural parameters. As a proof-of-concept, a mid-infrared spectrometer is further demonstrated with a dual-spike reconstruction resolution of 1.5 nm and a bandwidth of 300 nm (4–4.3 μm).

Photonic integrated circuits are emerging as a promising platform for accelerating matrix multiplications in deep learning, leveraging the inherent parallel nature of light. Although various schemes have been proposed and demonstrated to realize such photonic matrix accelerators, the in situ training of artificial neural networks using photonic accelerators remains challenging due to the difficulty of direct on-chip backpropagation on a photonic chip. In this work, we propose a silicon microring resonator (MRR) optical crossbar array with a symmetric structure that allows for simple on-chip backpropagation, potentially enabling the acceleration of both the inference and training phases of deep learning. We demonstrate a 4×4 circuit on a Si-on-insulator platform and use it to perform inference tasks of a simple neural network for classifying iris flowers, achieving a classification accuracy of 93.3%. Subsequently, we train the neural network using simulated on-chip backpropagation and achieve an accuracy of 91.1% in the same inference task after training. Furthermore, we simulate a convolutional neural network for handwritten digit recognition, using a 9×9 MRR crossbar array to perform the convolution operations. This work contributes to the realization of compact and energy-efficient photonic accelerators for deep learning.

Integrated semiconductor lasers represent essential building blocks for integrated optical components and circuits and their stability in frequency is fundamental for the development of numerous frontier applications and engineering tasks. When dense optical circuits are considered, the stability of integrated laser sources can be impaired by the thermal cross-talk generated by the action of neighboring components, leading to a deterioration of the long-term system performance (on the scale of seconds). In this work we show the design and the experimental characterization of a silicon nitride photonic integrated circuit (PIC) that is able to frequency stabilize 16 semiconductor lasers, simultaneously. A stabilized 50 GHz-spaced two-channel system is demonstrated through the detection of the related beating note and the stability of the resulting waveform is characterized via the use of artificially induced thermal cross-talk stimuli.

Ultra-compact multifunctional integrated photonic modules have great practical significance to photonic integrated circuits (PICs). However, the design effect and efficiency of the existing mainstream inverse design algorithms are incompetent when designing these modules. We analyze their shortcomings in this task, and propose a new, to our knowledge, inverse design algorithm named polygon search (PS) algorithm to address these problems. We utilize the PS algorithm to design an integrated dual-channel mode-conversion-crossing waveguide module. This module integrates three functions: interconversion between TE0 and TE1, interconversion between TE0 and TE2, and channel crossing within only a 4 μm×4 μm footprint, and its performance is verified by experimental testing. It not only greatly reduces the total footprint of many PICs but also greatly improves their fabricating robustness. Furthermore, we propose a PS-designed mode mixer and a PS-designed bending waveguide, and connect them with the integrated modules to form a four-channel crossing-mode-division-multiplexing system. This system can provide multiple modes on the basis of channel crossing and transmit the output signal in the same direction in parallel within a single output waveguide, which significantly increases the communication bandwidth and decreases the footprint of PICs. At last, we demonstrate the effect and efficiency advantages of the PS algorithm over several mainstream inverse design algorithms by a comprehensive contrast experiment and explain these advantages in theory from several perspectives.

Frequency microcombs with microwave and millimeter-wave repetition rates provide a compact solution for coherent communication and information processing. The implementation of these microcombs using a CMOS-compatible platform further paves the way for large-scale photonic integration and modularity. Here, we demonstrate free-running soliton microcombs with K-band repetition rates with very low phase noise over a 4 GHz pump detuning range reaching -117 (-123) dBc/Hz at 10 kHz offset for a 19.7 (10) GHz carrier without active pump stabilization, exceeding commercial electronic microwave oscillators at frequency offsets above 40 kHz. The minimum laser noise to soliton microwave signal transduction factor observed is -73 dB. This noise performance is achieved using a hybridized dual-mode for soliton generation to achieve passive thermal stabilization and minimal soliton spectrum shift from prior Raman scattering and dispersive wave formation. We further examine the locking of the repetition rate to an external ultrastable photonic oscillator to illustrate the feasibility of phase noise suppression below the thermorefractive noise limits of microresonator frequency combs.

Bragg filters are of essential importance for chip-scale photonic systems. However, the implementation of filters with sub-nanometer bandwidth and rejection beyond 70 dB is hindered by the high index contrast of the silicon-on-insulator platform, which makes filters prone to fabrication imperfections. In this paper, we propose to combine coherency-broken cascading architecture and cladding modulation to circumvent the intrinsic limitation. The cascading architecture effectively prevents the accumulation of phase errors, while the cladding modulation offers additional design freedom to reduce the coupling coefficient. A bimodal Bragg filter with a testing-equipment-limited rejection level of 74 dB and a 40 dB bandwidth of 0.44 nm is experimentally demonstrated. The minimum feature size is 90 nm, which significantly relieves the fabrication constraints.

Light detection and ranging (LiDAR) serves as one of the key components in the fields of autonomous driving, surveying mapping, and environment detection. Conventionally, dense points clouds are pursued by LiDAR systems to provide high-definition 3D images. However, the LiDAR is typically used to produce abundant yet redundant data for scanning the homogeneous background of scenes, resulting in power waste and excessive processing time. Hence, it is highly desirable for a LiDAR system to “gaze” at the target of interest by dense scanning and rough sparse scans on the uninteresting areas. Here, we propose a LiDAR structure based on an optical phased array (OPA) with reconfigurable apertures to achieve such a gaze scanning function. By virtue of the cascaded optical switch integrated on the OPA chip, a 64-, 128-, 192-, or 256-channel antenna can be selected discretionarily to construct an aperture with variable size. The corresponding divergence angles for the far-field beam are 0.32°, 0.15°, 0.10°, and 0.08°, respectively. The reconfigurable-aperture OPA enables the LiDAR system to perform rough scans via the large beam spots prior to fine scans of the target by using the tiny beam spots. In this way, the OPA-based LiDAR can perform the “gaze” function and achieve full-range scanning efficiently. The scanning time and power consumption can be reduced by 1/4 while precise details of the target are maintained. Finally, we embed the OPA into a frequency-modulated continuous-wave (FMCW) system to demonstrate the “gaze” function in beam scanning. Experiment results show that the number of precise scanning points can be reduced by 2/3 yet can obtain the reasonable outline of the target. The reconfigurable-aperture OPA (RA-OPA) can be a promising candidate for the applications of rapid recognition, like car navigation and robot vision.

Based on the wavelength transparency of the Butler matrix (BM) beamforming network, we demonstrate a multi-beam optical phased array (MOPA) with an emitting aperture composed of grating couplers at a 1.55 μm pitch for wavelength-assisted two-dimensional beam-steering. The device is capable of simultaneous multi-beam operation in a field of view (FOV) of 60° × 8° in the phased-array scanning axis and the wavelength-tuning scanning axis, respectively. The typical beam divergence is about 4° on both axes. Using multiple linearly chirped lasers, multi-beam frequency-modulated continuous wave (FMCW) ranging is realized with an average ranging error of 4 cm. A C-shaped target is imaged for proof-of-concept 2D scanning and ranging.

We propose for the first time, to the best of our knowledge, an on-chip integrated few-mode erbium–ytterbium co-doped waveguide amplifier based on an 800 nm thick Si3N4 platform, which demonstrates high amplification gains and low differential modal gains (DMGs) simultaneously. An eccentric waveguide structure and a co-propagating pumping scheme are adopted to balance the gain of each mode. A hybrid mode/polarization/wavelength-division (de)multiplexer with low insertion loss and crosstalk is used for multiplexing and demultiplexing in two operation wavebands centered at 1550 nm and 980 nm, where the light in these two bands serves as the signal light and pump light of the amplifier, respectively. The results demonstrate that with an input signal power of 0.1 mW, TE0 mode pump power of 300 mW, and TE1 mode pump power of 500 mW, the three signal modes (TE0/TM0/TE1) all exhibit amplification gains exceeding 30 dB, while maintaining a DMG of less than 0.1 dB.

A silicon photonic spectrometer with multiple customized wavelength bands is developed by introducing multiple channels of wideband optical filters based on multimode waveguide gratings (MWGs) for pre-filtering and the corresponding thermally tunable narrowband filter for high resolution. For these multiple customized wavelength bands, the central wavelengths, bandwidths, and resolutions are designed flexibly as desired, so that the system is simplified and the footprint is minimized for several practical applications (e.g., gas sensing). A customized silicon photonic spectrometer is designed and demonstrated experimentally with four wavelength bands centered around 1310 nm, 1560 nm, 1570 nm, and 1930 nm, which is, to the best of our knowledge, the first on-chip spectrometer available for sensing multiple gas components like HF, CO, H2S, and CO2. The spectral resolutions of the four wavelength bands are 0.11 nm, 0.08 nm, 0.08 nm, and 0.37 nm, respectively. Such a customized silicon photonic spectrometer shows great potential for various applications, including gas monitors, wearable biosensors, and portable spectral-domain optical coherence tomography.

Expanding the optical communication band is one of the most effective methods of overcoming the nonlinear Shannon capacity limit of single fiber. In this study, GeSn resonance cavity enhanced (RCE) photodetectors (PDs) with an active layer Sn component of 9%–10.8% were designed and fabricated on an SOI substrate. The GeSn RCE PDs present a responsivity of 0.49 A/W at 2 μm and a 3-dB bandwidth of approximately 40 GHz at 2 μm. Consequently, Si-based 2 μm band optical communication with a transmission rate of 50 Gbps was demonstrated by using a GeSn RCE detector. This work demonstrates the considerable potential of the Si-based 2 μm band photonics in future high-speed and high-capacity optical communication.

We propose and numerically demonstrate a photonic computing primitive designed for integrated spiking neural networks (SNNs) based on add-drop ring microresonators (ADRMRs) and electrically reconfigurable phase-change material (PCM) photonic switches. In this neuromorphic system, the passive silicon-based ADRMR, equipped with a power-tunable auxiliary light, effectively demonstrates nonlinearity-induced dual neural dynamics encompassing spiking response and synaptic plasticity that can generate single-wavelength optical neural spikes with synaptic weight. By cascading these ADRMRs with different resonant wavelengths, weighted multiple-wavelength spikes can be feasibly output from the ADRMR-based hardware arrays when external wavelength-addressable optical pulses are injected; subsequently, the cumulative power of these weighted output spikes is utilized to ascertain the activation status of the reconfigurable PCM photonic switches. Moreover, the reconfigurable mechanism driving the interconversion of the PCMs between the resonant-bonded crystalline states and the covalent-bonded amorphous states is achieved through precise thermal modulation. Drawing from the thermal properties, an innovative thermodynamic leaky integrate-and-firing (TLIF) neuron system is proposed. With the TLIF neuron system as the fundamental unit, a fully connected SNN is constructed to complete a classic deep learning task: the recognition of handwritten digit patterns. The simulation results reveal that the exemplary SNN can effectively recognize 10 numbers directly in the optical domain by employing the surrogate gradient algorithm. The theoretical verification of our architecture paves a whole new path for integrated photonic SNNs, with the potential to advance the field of neuromorphic photonic systems and enable more efficient spiking information processing.

We report the design, fabrication, and characterization of a universal silicon PN junction ring resonator for C band error-free communication links operated up to 50 Gb/s with co-designed optical modulation and detection performance. The universal p-n junction ring device shows co-designed detection responsivity up to 0.84 A/W, in conjunction with a modulation efficiency of ∼4 V·mm and >8 dB optical modulation extinction ratio, enabling C band 50 Gb/s NRZ communication link with a bit error rate ≤3×10-12. Individually, the speed of modulation and detection is measured up to 112 Gb/s and 80 Gb/s, respectively. The principle of co-designing the PN junction ring modulator and detector performance required for error-free communication links can significantly ease the fabrication yield challenges of ring structures by reducing the number of types of devices. The principle can also be applied to O band wavelengths. To the best of our knowledge, for the first time, a device of this type has achieved both error-free modulation and detection operation up to 50 Gb/s in the C band individually or in conjugation as an error-free communication link, which paves the way to realize a >1.6 Tb/s all-silicon WDM-based error-free optical transceiver link in the future and is essential for future programmable photonics circuits.

On-chip polarization controllers are extremely important for various optical systems. In this paper, a compact and robust silicon-based on-chip polarization controller is proposed and demonstrated by integrating a special polarization converter and phase shifters. The special polarization converter consists of a 1×1 Mach–Zehnder interferometer with two polarization-dependent mode converters at the input/output ends. When light with an arbitrary state of polarization (SOP) is launched into the chip, the TE0 and TM0 modes are simultaneously excited. The polarization extinction ratio (PER) and the phase difference for the TE0/TM0 modes are tuned by controlling the first phase shifter, the polarization converter, and the second phase shifter. As a result, one can reconstruct the light SOP at the output port. The fabricated polarization controller, as compact as ∼150 μm×700 μm, exhibits an excess loss of less than 1 dB and a record PER range of >54 dB for arbitrary input light beams in the wavelength range of 1530–1620 nm.

High-performance germanium photodiodes are critical components in silicon photonic systems for high-capacity data communications. By reducing the length of the photodiodes, a smaller resistance–capacitance product can be achieved, leading to a larger bandwidth and lower dark current. However, this also leads to diminished responsivity due to insufficient light absorption. Here, we introduce a silicon corner reflector (SCR) to alleviate this issue by reflecting and recycling the unabsorbed light. The process of evanescent coupling between the silicon and germanium layers is elaborately engineered to optimize the efficiency of light absorption. Experimentally, a responsivity of 0.96 A/W, which is a 21% increase compared to the one without SCR, is achieved at 1550 nm with a germanium length of 4.8 μm. Simultaneously, a remarkably low dark current of 0.76 nA and a large bandwidth of 100 GHz are achieved. Open eye diagrams of 140 Gb/s on–off keying and 240 Gb/s four-level pulse amplitude signals are obtained. To the best of our knowledge, this work achieves the lowest dark current density and noise equivalent power to date and offers a promising solution for low-cost, high-performance optical detection.

Germanium (Ge)-silicon (Si)-based avalanche photodetectors (APDs) featured by a high absorption coefficient in the near-infrared band have gained wide applications in laser ranging, free space communication, quantum communication, and so on. However, the Ge APDs fabricated by the complementary metal oxide semiconductor (CMOS) process suffer from a large dark current and limited responsivity, imposing a critical challenge on integrated silicon photonic links. In this work, we propose a p-i-n-i-n type Ge APD consisting of an intrinsic germanium layer functioning as both avalanche and absorption regions and an intrinsic silicon layer for dark current reduction. Consequently, a Ge APD with a low dark current, low bias voltage, and high responsivity can be obtained via a standard silicon photonics platform. In the experimental measurement, the Ge APD is characterized by a high primary responsivity of 1.1 A/W with a low dark current as low as 7.42 nA and a dark current density of 6.1×10-11 A/μm2 at a bias voltage of -2 V. In addition, the avalanche voltage of the Ge APD is -8.4 V and the measured 3 dB bandwidth of the Ge APD can reach 25 GHz. We have also demonstrated the capability of data reception on 32 Gbps non-return-to-zero (NRZ) optical signal, which has potential application for silicon photonic data links.

Miniaturized interferometric fiber optic gyroscopes (IFOGs) providing high-precision angular measurement are highly desired in various smart applications. In this work, we present a high-performance Si-SiN photonic FOG transceiver composed of an optical source, polarizer, splitter, and on-chip germanium (Ge) photodetector (PD). The transceiver is assembled in a standard butterfly package with a thermo-electric cooler (TEC). The optical loss (including two edge couplers, as well as one 3 dB splitter) and polarization extinction ratio (PER) are less than 7 dB and greater than 20 dB at room temperature, respectively. Built with the polarization maintaining (PM) fiber coil with 70 mm average diameter and 580 m length, the transceiver-based IFOG exhibits record-low bias stability of 0.022 deg/h at an integration time of 10 s, the angular random walk (ARW) of 0.0012 deg/h, and the bias instability of 0.003 deg/h, to the best of our knowledge. The preliminary reliability test agrees well with the practical requirements. Our work verifies that the on-chip Ge PD is eligible for high-performance FOG applications. Leveraged with the typical CMOS compatible 8-inch (200 mm diameter wafers) silicon photonics platform and decreased fiber splicing points, the presented transceiver provides a promising solution toward a low-loss and miniaturized FOG system with large volume manufacturing capability.

Chip-scale multi-dimensional multiplexing technology that combines wavelengths and spatial modes on a silicon photonic integrated circuit (PIC) is highly promising for the link-capacity scaling of future optical interconnects. However, current multi-dimensional multiplexed PICs face significant challenges in simultaneously achieving broad optical bandwidth, low mode crosstalk, and dual-polarization modes in an ultra-compact footprint as the number of spatial modes increases. To address the issue, a topology-optimization-based inverse design assisted by a novel manufacturing calibration method (MCM) is utilized. Based on a 220 nm silicon-on-insulator (SOI) platform, a 100 nm broadband and ultra-compact (6 μm×6 μm) multi-dimensional multiplexed PIC supporting TE0, TE1, TM0, and TM1 modes with modes crosstalk <-16dB ranging from 1500 to 1600 nm is demonstrated for the first time, to the best of our knowledge. Furthermore, the PIC is implemented to experimentally enable a single-wavelength 4-modes ×100 Gbit/s PAM-4 direct modulation data transmission over 51 wavelengths with 0.8 nm channel spacing. This work shows the potential of utilizing multi-dimensional multiplexed PICs as optical interconnects to effectively address the speed limits of data transfer for future high-performance chip-to-chip interconnection.

High-performance photonic switches are essential for large-scale optical routing for AI large models and the Internet of Things. Realizing nonvolatility can further reduce power consumption and expand application scenarios. We propose a nonvolatile 2×2 silicon photonic micro-electromechanical system (MEMS) switch compatible with standard silicon photonic foundry processes. The switch employs an electrostatic comb actuator to change the air gap of the compact horizontal adiabatic coupler and achieves nonvolatility with centrally clamped stepped bistable mechanical beams. The photonic switch features a 10s μs-scale switching speed and a 10s fJ-scale simulated switching energy within a 100 μm×100 μm footprint, with ≤12 V driving voltages. This 2×2 switch can be used in a variety of topologies for large-scale photonic switches, and its nonvolatility can potentially support future photonic field programmable gate array designs.

Mode-division multiplexing (MDM) has attracted much attention due to its ability to further increase the transmission capacity of optical interconnects. While further developments of MDM optical interconnects are hindered by the coupling of few-mode fibers (FMFs) and silicon photonic chips, a high-efficiency, broadband, and scalable multimode FMF-chip interface is still eagerly desired. To address this challenge, a novel scheme for efficient multimode coupling is proposed by introducing a silica planar lightwave circuit as an intermediate. The core idea is to couple and demultiplex higher-order modes by leveraging the superiorities of silica optical waveguides for manipulating LP modes, facilitated through tailoring the mode conversion related to different mode symmetric properties. The demultiplexed modes are consequently butt-coupled to the silicon photonic chip in single-mode manner, thus being available for fulfilling further data transmitting/receiving/routing directly. As a proof of concept, a six-channel FMF-chip coupler working with the LP01-x/y, LP11a-x/y, and LP11b-x/y modes is designed with low coupling losses of 0.77–1.39 dB and low intermode crosstalk of <-27.2 dB in a broad bandwidth (>150 nm). Minimum coupling losses of 1.36–2.48 dB are experimentally demonstrated. It is the first demonstration for the integrated multimode FMF-chip coupler enabling the simultaneous coupling of six mode-channels, to the best of our knowledge. We believe that this work has the great potential for developing energy-efficient and low-cost chip-to-chip MDM interconnections in the future.

In recent years, integrated optical processing units (IOPUs) have demonstrated advantages in energy efficiency and computational speed for neural network inference applications. However, limited by optical integration technology, the practicality and versatility of IOPU face serious challenges. In this work, a scalable parallel photonic processing unit (SPPU) for various neural network accelerations based on high-speed phase modulation is proposed and implemented on a silicon-on-insulator platform, which supports parallel processing and can switch between multiple computational paradigms simply and without latency to infer different neural network structures, enabling to maximize the utility of on-chip components. The SPPU adopts a scalable and process-friendly architecture design, with a preeminent photonic-core energy efficiency of 0.83 TOPS/W, two to ten times higher than existing integrated solutions. In the proof-of-concept experiment, a convolutional neural network (CNN), a residual CNN, and a recurrent neural network (RNN) are all implemented on our photonic processor to handle multiple tasks of handwritten digit classification, signal modulation format recognition, and review emotion recognition. The SPPU achieves multi-task parallel processing capability, serving as a promising and attractive research route to maximize the utility of on-chip components under the constraints of integrated technology, which helps to make IOPU more practical and universal.

The 2 μm wavelength band emerges as a promising candidate for the next communication window to enhance the transmission capacity of data. A high-responsivity and high-speed photodetector operating at 2 μm is crucial for the 2-μm-wavelength-band communication system. Here, we present an on-chip waveguide-coupled germanium photodetector with remarkably high responsivity and data-receiving rate, employing subbandgap light absorption and avalanche multiplication. The device is designed with an ingenious and simple asymmetric lateral p-i-n junction structure and fabricated through a standard CMOS process by a commercial factory. It has a responsivity of 3.64 A/W and a maximum bandwidth of 50 GHz at 2 μm wavelength. For the first time, to the best of our knowledge, an optical receiving rate of up to 112 Gbps is demonstrated at 2 μm, verifying its feasibility in a high-speed 2-μm-band communication system. To the best of our knowledge, the proposed device stands out as the fastest photodiode with the highest responsivity among all group III-V and group IV photodetectors working in the 2 μm wavelength band.

Photonic computing has the potential to harness the full degrees of freedom (DOFs) of the light field, including the wavelength, spatial mode, spatial location, phase quadrature, and polarization, to achieve a higher level of computing parallelism and scalability than digital electronic processors. While multiplexing using the wavelength and other DOFs can be readily integrated on silicon photonics platforms with compact footprints, conventional mode-division multiplexed (MDM) photonic designs occupy areas exceeding tens to hundreds of microns for a few spatial modes, significantly limiting their scalability. Here, we utilize inverse design to demonstrate an ultracompact photonic computing core that calculates vector dot products based on MDM coherent mixing. Our dot-product core integrates the functionalities of two-mode multiplexers and one multimode coherent mixer within a nominal footprint of 5 μm×3 μm. We have experimentally demonstrated computing examples on the fabricated dot-product core, including complex number multiplication and motion estimation using optical flow. The compact dot-product core design enables large-scale on-chip integration in a parallel photonic computing primitive cluster for high-throughput scientific computing and computer vision tasks.

The four-wave mixing (FWM) effect offers promise to generate or amplify light at wavelengths where achieving substantial gain is challenging, particularly within the mid-infrared (MIR) spectral range. Here, based on the commonly used 340 nm silicon-on-insulator (SOI) platform, we experimentally demonstrate high-efficiency and broadband wavelength conversion using the FWM effect in a high-Q silicon microring resonator pumped by a continuous-wave (CW) laser in the 2 μm waveband. The microring resonator parameters are carefully optimized for effective phase-matching to obtain high conversion efficiency (CE) with broad bandwidth. The loaded quality (Ql) factor of the fabricated microring resonator is measured to be 1.11×105, at a resonance wavelength of 1999.3 nm, indicating low propagation losses of 1.68 dB/cm. A maximum CE of -15.57 dB is achieved with a low input pump power of only 4.42 dBm, representing, to our knowledge, the highest on-chip CE demonstrated to date under the CW pump in the MIR range. Furthermore, broadband wavelength conversion can be observed across a 140.4 nm wavelength range with a CE of -19.32 dB, and simulations indicate that the conversion bandwidth is over 400 nm. This work opens great potential in exploiting widely tunable on-chip sources using high-efficiency wavelength conversion, particularly leveraging the advantages of the SOI platform in integrated photonics across the 2 μm MIR range.

Phase change materials (PCMs), characterized by high optical contrast (Δn>1), nonvolatility (zero static power consumption), and quick phase change speed (∼ns), provide new opportunities for building low-power and highly integrated photonic tunable devices. Optical integrated devices based on PCMs, such as optical switches and optical routers, have demonstrated significant advantages in terms of modulation energy consumption and integration. In this paper, we theoretically verify the solution for a highly integrated nonvolatile optical switch based on the modulation of the topological interface state (TIS) in the quasi-one-dimensional photonic crystal (quasi-1D PC). The TIS exciting wavelength changes with the crystalline level of the PCM. The extinction ratio (ER) of the topological optical switch is over 18 dB with a modulation length of 9 μm. Meanwhile, the insertion loss (IL) can be controlled within 2 dB. Furthermore, we have analyzed the impact of fabrication errors on the device’s performance. The obtained results show that, the topological optical switch, which changes its “on/off” state by modulating TIS, exhibits enhanced robustness to the fabrication process. We provide an interesting and highly integrated scheme for designing the on-chip nonvolatile optical switch. It offers great potential for designing highly integrated on-chip optical switch arrays and nonvolatile optical neural networks.

Silicon-based electro-optic modulators contribute to easing the integration of high-speed and low-power consumption circuits for classical optical communications and data computations. Beyond the plasma dispersion modulation, an alternative solution in silicon is to exploit the DC Kerr effect, which generates an equivalent linear electro-optical effect enabled by applying a large DC electric field. Although some theoretical and experimental studies have shown its existence in silicon, limited contributions relative to plasma dispersion have been achieved in high-speed modulation so far. This paper presents high-speed optical modulation based on the DC Kerr effect in silicon PIN waveguides. The contributions of both plasma dispersion and Kerr effects have been analyzed in different waveguide configurations, and we demonstrated that the Kerr induced modulation is dominant when a high external DC electric field is applied in PIN waveguides. High-speed optical modulation response is analyzed, and eye diagrams up to 80 Gbit/s in NRZ format are obtained under a d.c. voltage of 30 V. This work paves the way to exploit the Kerr effect to generate high-speed Pockels-like optical modulation.

High-performance germanium photodiodes are crucial components in silicon photonic integrated circuits for large-capacity data communication. However, the bandwidths of most germanium photodiodes are limited by the intractable resistance–capacitance parasitic effect. Here, we introduce a unique U-shaped electrode to alleviate this issue, reducing the parasitic effect by 36% without compromising any other performance. Experimentally, a large bandwidth of 103 GHz, an optical responsivity of 0.95 A/W at 1550 nm, and a dark current as low as 1.3 nA are achieved, leading to a record high specific detectivity. This is the first breakthrough to 100 GHz bandwidth among all vertical germanium photodiodes, to the best of our knowledge. Open eye diagrams of 120 Gb/s on-off keying and 200 Gb/s four-level pulse amplitude signals are well received. This work provides a promising solution for chip-based ultra-fast photodetection.

We demonstrate a GeSi electro-absorption modulator with on-chip thermal tuning for the first time, to the best of our knowledge. Theoretical simulation proves that the device temperature can be tuned and the effective operating wavelength range can be broadened. When the heater power is 4.63 mW, the temperature of the waveguide increases by about 27 K and the theoretical operating wavelength range is broadened by 23.7 nm. The experimental results show that the optical transmission line shifted to the longer wavelength by 4.8 nm by every 1 mW heater power. The effective static operating wavelength range of the device is increased from 34.4 nm to 60.1 nm, which means it is broadened by 25.7 nm. The band edge shift coefficient of 0.76 nm/K is obtained by temperature simulation and linear fitting of the measured data. The device has a 3 dB EO bandwidth of 89 GHz at 3 V reverse bias, and the eye diagram measurement shows a data rate of 80 Gbit/s for non-return-to-zero on–off keying modulation and 100 Gbit/s for 4 pulse amplitude modulation in the 1526.8 nm to 1613.2 nm wavelength range as the heater power increases from 0 mW to 10.1 mW.

We present the design and experimentally demonstrate a dual-level grating coupler with subdecibel efficiency for a 220 nm thick silicon photonics waveguide which was fabricated starting from a 340 nm silicon-on-insulator wafer. The proposed device consists of two grating levels designed with two different linear apodizations, with opposite chirping signs, and whose period is varied for each scattering unit. A coupling efficiency of -0.8 dB at 1550 nm is experimentally demonstrated, which represents the highest efficiency ever reported in the telecommunications C-band in a single-layer silicon grating structure without the use of any backreflector or index-matching material between the fiber and the grating.

Efficient extraction of light from a high refractive index silicon waveguide out of a chip is difficult to achieve. An inverse design approach was employed using the particle swarm optimization method to attain a vertical emitting meta-grating coupler with high coupling efficiency in a 220-nm-thick silicon-on-insulator platform. By carefully selecting the figure of merit and appropriately defining parameter space, unique L-shape and U-shape grating elements that boosted the out-of-plane radiation of light were obtained. In addition, a 65.7% (-1.82 dB) outcoupling efficiency and a 60.2% (-2.2 dB) fiber-to-chip vertical coupling efficiency with an 88 nm 3 dB bandwidth were demonstrated by numerical simulation. Considering fabrication constraints, the optimized complex meta-grating coupler was modified to correspond to two etching steps and was then fabricated with a complementary metal-oxide-semiconductor-compatible process. The modified meta-grating coupler exhibited a simulated coupling efficiency of 57.5% (-2.4 dB) with a 74 nm 3-dB bandwidth in the C-band and an experimentally measured coupling efficiency of 38% (-4.2 dB).

A small 4-channel time-delayed complex perceptron is used as a silicon photonic neural network (PNN) device to compensate for chromatic dispersion in optical fiber links. The PNN device is experimentally tested with non-return-to-zero optical signals at 10 Gbps after propagation through up to 125 km optical fiber link. During the learning phase, a separation-loss function is optimized in order to maximally separate the transmitted levels of 0s from the 1s, which implies an optimization of the bit-error-rate. Testing of the PNN device shows that the excess losses introduced by the PNN device are compensated by the gain in the transmitted signal equalization for a link longer than 100 km. The measured data are reproduced by a model that accounts for the optical link and the PNN device. This allows simulating the network performances for higher data rates, where the device shows improvement with respect to the benchmark both in terms of performance and ease of use.

A compact on-chip reconfigurable multichannel amplitude equalizer based on cascaded elliptical microrings is proposed and demonstrated experimentally. With the optimized structure of the elliptical microring with adiabatically varied radii/widths, the average excess loss for each channel in the initialized state is measured to be less than 0.5 dB, while the attenuation dynamic range can be over 20 dB. Flexible tunability through the overlapping of the resonance peaks of adjacent wavelength-channels enables even higher attenuation dynamic ranges up to 50 dB. Leveraging the thermo-optic effect and fine wavelength-tuning linearity, precise tuning of the resonance peak can be implemented, enabling dynamic power equalization of each wavelength-channel in wavelength-division-multiplexing (WDM) systems and optical frequency combs. The proposed architecture exhibits excellent scalability, which can facilitate the development of long-haul optical transport networks and high-capacity neuromorphic computing systems, while improving the overall performance of optical signals in WDM-related systems.

Microring-based optical switches are promising for wavelength-selective switching with the merits of compact size and low power consumption. However, the large insertion loss, the high fabrication, and the temperature sensitivity hinder the scalability of silicon microring optical switch fabrics. In this paper, we utilize a three-dimensional (3D) microring-based optical switch element (SE) on a multi-layer Si3N4-on-SOI platform to realize high-performance large-scale optical switch fabrics. The 3D microring-based SE consists of a Si/Si3N4 waveguide overpass crossing in the bottom and the top layers, and Si3N4 dual-coupled microring resonators (MRRs) in the middle layer. The switch is calibration-free and has low insertion loss. With the 3D microring-based SEs, we implement an 8×8 crossbar optical switch fabric. As the resonance wavelengths of all SEs are well aligned, only one SE needs to be turned on in each routing path, which greatly reduces the complexity of the switch control. The optical transmission spectra show a box-like shape, with a passband width of ∼69 GHz and an average on-state loss of ∼0.37 dB. The chip has a record-low on-chip insertion loss of 0.52–2.66 dB. We also implement a non-duplicate polarization-diversity optical switch by using the bidirectional transmission characteristics of the crossbar architecture, which is highly favorable for practical applications. 100 Gb/s dual-polarization quadrature-phase-shift-keying (DP-QPSK) signal is transmitted through the switch without significant degradation. To the best of our knowledge, this is the first time that 3D MRRs have been used to build highly scalable polarization-diversity optical switch fabrics.

Integrated microwave photonic filters (IMPFs) are capable of offering unparalleled performances in terms of superb spectral fineness, broadband, and more importantly, the reconfigurability, which encounter the trend of the next-generation wireless communication. However, to achieve high reconfigurability, previous works should adopt complicated system structures and modulation formats, which put great pressure on power consumption and controlment, and, therefore, impede the massive deployment of IMPF. Here, we propose a streamlined architecture for a wideband and highly reconfigurable IMPF on the silicon photonics platform. For various practical filter responses, to avoid complex auxiliary devices and bias drift problems, a phase-modulated flexible sideband cancellation method is employed based on the intensity-consistent single-stage-adjustable cascaded-microring (ICSSA-CM). The IMPF exhibits an operation band extending to millimeter-wave (≥30 GHz), and other extraordinary performances including high spectral resolution of 220 MHz and large rejection ratio of 60 dB are obtained. Moreover, Gb/s-level RF wireless communications are demonstrated for the first time towards real-world scenarios. The proposed IMPF provides broadband flexible spectrum control capabilities, showing great potential in the next-generation wireless communication.

An optical phased array (OPA), the most promising non-mechanical beam steering technique, has great potential for solid-state light detection and ranging systems, holographic imaging, and free-space optical communications. A high quality beam with low sidelobes is crucial for long-distance free-space transmission and detection. However, most previously reported OPAs suffer from high sidelobe levels, and few efforts are devoted to reducing sidelobe levels in both azimuthal (φ) and polar (θ) directions. To solve this issue, we propose a Y-splitter-assisted cascaded coupling scheme to realize Gaussian power distribution in the azimuthal direction, which overcomes the bottleneck in the conventional cascaded coupling scheme and significantly increases the sidelobe suppression ratio (SLSR) in the φ direction from 20 to 66 dB in theory for a 120-channel OPA. Moreover, we designed an apodized grating emitter to realize Gaussian power distribution in the polar direction to increase the SLSR. Based on both designs, we experimentally demonstrated a 120-channel OPA with dual-Gaussian power distribution in both φ and θ directions. The SLSRs in φ and θ directions are measured to be 15.1 dB and 25 dB, respectively. Furthermore, we steer the beam to the maximum field of view of 25°×13.2° with a periodic 2λ pitch (3.1 μm). The maximum total power consumption is only 0.332 W with a thermo-optic efficiency of 2.7 mW/π.

Linear light-processing functions (e.g., routing, splitting, filtering) are key functions requiring configuration to implement on a programmable photonic integrated circuit (PPIC). In recirculating waveguide meshes (which include loop-backs), this is usually done manually. Some previous results describe explorations to perform this task automatically, but their efficiency or applicability is still limited. In this paper, we propose an efficient method that can automatically realize configurations for many light-processing functions on a square-mesh PPIC. At its heart is an automatic differentiation subroutine built upon analytical expressions of scattering matrices that enables gradient descent optimization for functional circuit synthesis. Similar to the state-of-the-art synthesis techniques, our method can realize configurations for a wide range of light-processing functions, and multiple functions on the same PPIC simultaneously. However, we do not need to separate the functions spatially into different subdomains of the mesh, and the resulting optimum can have multiple functions using the same part of the mesh. Furthermore, compared to nongradient- or numerical differentiation-based methods, our proposed approach achieves 3× time reduction in computational cost.

On-chip Fourier-transform spectrometers (FTSs) based on Mach–Zehnder interferometer (MZI) arrays suffer from severe central wavelength and fringe contrast variation due to fabrication errors. Even though a calibration matrix can be employed to correctly retrieve the input spectra, environmental temperature variation greatly degrades the retrieving performance. In this paper, we devise a dual-layer Si3N4 waveguide interferometer to reduce the temperature sensitivity. The beating of the even and odd supermodes in the dual-layer waveguide generates periodic intensity fluctuations in the spectrum. Since these two modes have similar modal profiles, their thermal sensitivity and propagation loss are relatively balanced, leading to a low temperature sensitivity and a high interference extinction ratio. We designed and fabricated a passive FTS based on a 32-channel dual-layer Si3N4 waveguide array. Experimental results show that the temperature sensitivity is reduced to 10 pm/°C, which is almost half that of single-layer Si3N4 MZI-based FTSs. With this chip, we accurately reconstructed various types of optical spectra, including single and two sparse laser lines, and broadband optical spectra. Our method can fit a wide wavelength range, which is a promising technology to improve the practical applications of on-chip FTSs.

Bound states in the continuum (BICs) provide a fascinating platform to route/manipulate waves with ultralow loss by patterning low-refractive-index materials on a high-refractive-index substrate. Principally, the phase of leaking channels can be manipulated via tuning the structural parameters to achieve destructive interference (i.e., the BIC condition), surprisingly leading to the total elimination of dissipation to the continuum of the substrate. Despite recent developments in BIC photonics, the BIC conditions can only be satisfied at specified geometric sizes for waveguides that dim their application prospects. Here, we propose a dual waveguide system that support BICs under arbitrary waveguide sizes by solely changing the intervals between the two waveguides. Our calculation results show that robust BICs in such architectures stem from the interaction (destructive interference) between leaking waves from the two waveguides. Furthermore, a cladding layer is introduced to improve the fabrication tolerance and reduce the sensitivity of the low-loss condition on the waveguide intervals of the presented dual waveguide system. The proposed approach offers an intriguing solution to establish a BIC concept and may be helpful to improve the potential of BIC photonic devices and circuits.

As a resonator-based optical hardware in analog optical computing, a microring synapse can be straightforwardly configured to simulate the connection weights between neurons, but it faces challenges in precision and stability due to cross talk and environmental perturbations. Here, we propose and demonstrate a self-calibration scheme with dual-wavelength synchronization to monitor and calibrate the synaptic weights without interrupting the computation tasks. We design and fabricate an integrated 4×4 microring synapse and deploy our self-calibration scheme to validate its effectiveness. The precision and robustness are evaluated in the experiments with favorable performance, achieving 2-bit precision improvement and excellent robustness to environmental temperature fluctuations (the weights can be corrected within 1 s after temperature changes 0.5°C). Moreover, we demonstrate matrix inversion tasks based on Newton iterations beyond 7-bit precision using this microring synapse. Our scheme provides an accurate and real-time weight calibration independently parallel from computations and opens up new perspectives for precision boost solutions to large-scale analog optical computing.

All-silicon (Si) photodiodes have drawn significant interest due to their single and simple material system and perfect compatibility with complementary metal-oxide semiconductor photonics. With the help from a cavity enhancement effect, many of these photodiodes have shown considerably high responsivity at telecommunication wavelengths such as 1310 nm, yet the mechanisms for such high responsivity remain unexplained. In this work, an all-Si microring is studied systematically as a photodiode to unfold the various absorption mechanisms. At -6.4 V, the microring exhibits responsivity up to ∼0.53 A/W with avalanche gain, a 3 dB bandwidth of ∼25.5 GHz, and open-eye diagrams up to 100 Gb/s. The measured results reveal the hybrid absorption mechanisms inside the device. A comprehensive model is reported to describe its working principle, which can guide future designs and make the all-Si microring photodiode a promising building block in Si photonics.

We demonstrate a single-chip silicon optical single sideband (OSSB) modulator composed of a radio frequency (RF) branch line coupler (BLC) and a silicon dual-parallel Mach–Zehnder modulator (DP-MZM). A co-design between the BLC and the DP-MZM is implemented to improve the sideband suppression ratio (SSR). The modulator has a modulation efficiency of VπLπ∼1.75 V·cm and a 3 dB electro-optical (EO) bandwidth of 48.7 GHz. The BLC can generate a pair of RF signals with equal amplitudes and orthogonal phases at the optimal frequency of 21 GHz. We prove through theoretical calculation and experiment that, although the BLC’s performance in terms of power balance and phase orthogonality deteriorates in a wider frequency range, high SSRs can be realized by adjusting relevant bias phases of the DP-MZM. With this technique, the undesired sidebands are completely suppressed below the noise floor in the frequency range from 15 GHz to 30 GHz when the chip operates in the full carrier OSSB (FC-OSSB) mode. In addition, an SSR >35 dB and an carrier suppression ratio (CSR) >42 dB are demonstrated at 21 GHz in the suppressed carrier OSSB (SC-OSSB) mode.

Short-wavelength mid-infrared (2–2.5 μm wave band) silicon photonics has been a growing area to boost the applications of integrated optoelectronics in free-space optical communications, laser ranging, and biochemical sensing. In this spectral region, multi-project wafer foundry services developed for the telecommunication band are easily adaptable with the low intrinsic optical absorption from silicon and silicon dioxide materials. However, light coupling techniques at 2–2.5 μm wavelengths, namely, grating couplers, still suffer from low efficiencies, mainly due to the moderated directionality and poor diffraction-field tailoring capability. Here, we demonstrate a foundry-processed blazed subwavelength coupler for high-efficiency, wide-bandwidth, and large-tolerance light coupling. We subtly design multi-step-etched hybrid subwavelength grating structures to significantly improve directionality, as well as an apodized structure to tailor the coupling strength for improving the optical mode overlap and backreflection. Experimental results show that the grating coupler has a recorded coupling efficiency of -4.53 dB at a wavelength of 2336 nm with a 3-dB bandwidth of ∼107 nm. The study opens an avenue to developing state-of-the-art light coupling techniques for short-wavelength mid-infrared silicon photonics.

A hybrid integrated 16-channel silicon transmitter based on co-designed photonic integrated circuits (PICs) and electrical chiplets is demonstrated. The driver in the 65 nm CMOS process employs the combination of a distributed architecture, two-tap feedforward equalization (FFE), and a push–pull output stage, exhibiting an estimated differential output swing of 4.0Vpp. The rms jitter of 2.0 ps is achieved at 50 Gb/s under nonreturn-to-zero on–off keying (NRZ-OOK) modulation. The PICs are fabricated on a standard silicon-on-insulator platform and consist of 16 parallel silicon dual-drive Mach–Zehnder modulators on a single chip. The chip-on-board co-packaged Si transmitter is constituted by the multichannel chiplets without any off-chip bias control, which significantly simplifies the system complexity. Experimentally, the open and clear optical eye diagrams of selected channels up to 50 Gb/s OOK with extinction ratios exceeding 3 dB are obtained without any digital signal processing. The power consumption of the Si transmitter with a high integration density featuring a throughput up to 800 Gb/s is only 5.35 pJ/bit, indicating a great potential for massively parallel terabit-scale optical interconnects for future hyperscale data centers and high-performance computing systems.

In the same silicon photonic integrated circuit, we compare two types of integrated degenerate photon-pair sources (microring resonators and waveguides) using Hong–Ou–Mandel (HOM) interference experiments. Two nominally identical microring resonators are coupled to two nominally identical waveguides, which form the arms of a Mach–Zehnder interferometer. This is pumped by two lasers at two different wavelengths to generate, by spontaneous four-wave mixing, degenerate photon pairs. In particular, the microring resonators can be thermally tuned in or out of resonance with the pump wavelengths, thus choosing either the microring resonators or the waveguides as photon-pair sources, respectively. In this way, an on-chip HOM visibility of 94% with microring resonators and 99% with straight waveguides is measured upon filtering. We compare our experimental results with theoretical simulations of the joint spectral intensity and the purity of the degenerate photon pairs. We verify that the visibility is connected to the sources’ indistinguishability, which can be quantified by the overlap between the joint spectral amplitudes (JSA) of the photon pairs generated by the two sources. We estimate a JSA overlap of 98% with waveguides and 89% with microring resonators.

This paper presents a novel approach to counterbalance free-carrier-absorption (FCA) in electro-optic (E-O) Mach–Zehnder interferometer (MZI) cells by harnessing the self-heating effect. We show insights on crosstalk limitations in MZIs with direct carrier-injection and provide a detailed design methodology on a differential phase shifter pair. Leveraging both free-carrier dispersion (FCD) and self-heating effects, our design enables arbitrary phase tuning with balanced FCA loss in the pair of arms, eliminating the need for additional phase corrections and creating ultralow crosstalk MZI elements. This neat design disengages from the commonly used nested structure, thus providing an opportunity of embedding tunable couplers for correcting imperfect splitting ratios given that only two are needed. We show that with the use of tunable directional couplers, a standard ±10 nm process variation is tolerated, while achieving a crosstalk ratio below -40 dB. By direct carrier injection in both arms, the proposed device operates at nanosecond scales and can bring about a breakthrough in the scalability of E-O switch fabrics, as well as other silicon integrated circuits that have stringent requirements for crosstalk leakage.

The development of an efficient group-IV light source that is compatible with the CMOS process remains a significant goal in Si-based photonics. Recently, the GeSn alloy has been identified as a promising candidate for realizing Si-based light sources. However, previous research suffered from a small wafer size, limiting the throughput and yield. To overcome this challenge, we report the successful growth of GeSn/Ge multiple-quantum-well (MQW) p-i-n LEDs on a 12-inch (300-mm) Si substrate. To the best of our knowledge, this represents the first report of semiconductor LEDs grown on such a large substrate. The MQW LED epitaxial layer is deposited on a 12-inch (300-mm) (001)-oriented intrinsic Si substrate using commercial reduced pressure chemical vapor deposition. To mitigate the detrimental effects of threading dislocation densities on luminescence, the GeSn/Ge is grown pseudomorphically. Owing to the high crystal quality and more directness in the bandgap, enhanced electroluminescence (EL) integrated intensity of 27.58 times is demonstrated compared to the Ge LED. The MQW LEDs exhibit EL emission near 2 μm over a wide operating temperature range of 300 to 450 K, indicating high-temperature stability. This work shows that GeSn/Ge MQW emitters are potential group-IV light sources for large-scale manufacturing.

A light-trapping-structure vertical Ge photodetector (PD) is demonstrated. In the scheme, a 3 μm radius Ge mesa is fabricated to constrain the optical signal in the circular absorption area. Benefiting from the light-trapping structure, the trade-off between bandwidth and responsivity can be relaxed, and high opto-electrical bandwidth and high responsivity are achieved simultaneously. The measured 3 dB bandwidth of the proposed PD is around 67 GHz, and the responsivity is around 1.05 A/W at wavelengths between 1520 and 1560 nm. At 1580 nm, the responsivity is still over 0.78 A/W. A low dark current of 6.4 nA is also achieved at -2 V bias voltage. Based on this PD, a clear eye diagram of 100 GBaud four-level pulse amplitude modulation (PAM-4) is obtained. With the aid of digital signal processing, 240 Gb/s PAM-4 signal back-to-back transmission is achieved with a bit error ratio of 1.6×10-2. After 1 km and 2 km fiber transmission, the highest bit rates are 230 and 220 Gb/s, respectively.

Bound states in the continuum (BICs) can make subwavelength dielectric resonators sustain low radiation leakage, paving a new way to minimize the device size, enhance photoluminescence, and even realize lasing. Here, we present a quasi-BIC-supporting GaAs nanodisk with embedded InAs quantum dots as a compact bright on-chip light source, which is realized by heterogeneous integration, avoiding complex multilayered construction and subsequent mismatch and defects. The emitters are grown inside the nanodisk to match the mode field distribution to form strong light–matter interaction. One fabricated sample demonstrates a photoluminescence peak sustaining a quality factor up to 68 enhanced by the quasi-BIC, and the emitting effect can be further promoted by improving the epilayer quality and optimizing the layer-transferring process in the fabrication. This work provides a promising solution to building an ultracompact optical source to be integrated on a silicon photonic chip for high-density integration.

Germanium-on-silicon (Ge-on-Si) avalanche photodiodes (APDs) are widely used in near-infrared detection, laser ranging, free space communication, quantum communication, and other fields. However, the existence of lattice defects at the Ge/Si interface causes a high dark current in the Ge-on-Si APD, degrading the device sensitivity and also increasing energy consumption in integrated circuits. In this work, we propose a novel surface illuminated Ge-on-Si APD architecture with three terminals. Besides two electrodes on Si substrates, a third electrode is designed for Ge to regulate the control current and bandwidth, achieving multiple outputs of a single device and reducing the dark current of the device. When the voltage on Ge is -27.5 V, the proposed device achieves a dark current of 100 nA, responsivity of 9.97 A/W at -40 dBm input laser power at 1550 nm, and optimal bandwidth of 142 MHz. The low dark current and improved responsivity can meet the requirements of autonomous driving and other applications demanding weak light detection.

Quantum dot lasers on silicon have gained significant interest over the past decade due to their great potential as an on-chip silicon photonic light source. Here, we demonstrate multi-wavelength injection locking of InAs/GaAs quantum dot Fabry–Perot (FP) lasers both on GaAs and silicon substrates by optical self-injection via an external cavity. The number of locked laser modes can be adjusted from a single peak to multiple peaks by tuning wavelength dependent phase and mode spacing of back-injected light through a Lyot filter. The multi-wavelength injection locked laser modes exhibit average optical linewidth of ∼20 kHz, which are narrowed by approximately three orders of magnitude from their free-running condition. Furthermore, multi-wavelength self-injection locking via an external cavity exhibits flat-top optical spectral properties with approximately 30 stably locked channels under stable operation over time, where the frequency detuning is less than 700 MHz within 40 min. Particularly, FP lasers by direct epitaxial growth on silicon substrates are self-injection locked as a flat-top comb source with tunable free spectral range from approximately 25 to 700 GHz. The reported results emphasize the great potential of multi-wavelength injection locked lasers as tunable on-chip multi-wavelength light sources.

In this paper, we present the experimental results for integrated photonic devices optimized with an energy-constrained inverse design method. When this constraint is applied, optimizations are directed to solutions that contain the optical field inside the waveguide core medium, leading to more robust designs with relatively larger minimum feature size. We optimize three components: a mode converter (MC), a 1310 nm/1550 nm wavelength duplexer, and a three-channel C-band wavelength demultiplexer for coarse wavelength division multiplexing (CWDM) application with 50 nm channel spacing. The energy constraint leads to nearly binarized structures without applying independent binarization stage. It also reduces the appearance of small features. In the MC, well-binarized design, improved insertion loss, and cross talk are obtained as a result. Furthermore, the proposed constraint improves the robustness to fabrication imperfections as shown in the duplexer design. With energy constraint optimization, the corresponding spectrum shifts for the duplexer under ±10 nm dimensional variations are reduced from 105 nm to 55 nm and from 72 nm to 60 nm for the 1310 nm and 1550 nm channel, respectively. In the CWDM demultiplexer, robustness toward ±10 nm fabrication error is improved by a factor of 2. The introduction of the energy constraint into topological optimization demonstrates computational gain with better-performing designs.

Integrated microwave photonic filters are becoming increasingly important for signal processing within advanced wireless and cellular networks. Filters with narrow transmission passbands mandate long time delays, which are difficult to accommodate within photonic circuits. Long delays may be obtained through slow moving acoustic waves instead. Input radio-frequency information can be converted from one optical carrier to another via surface acoustic waves and filtered in the process. However, the transfer functions of previously reported devices consisted of multiple periodic passbands, and the selection of a single transmission band was not possible. In this work, we demonstrate surface acoustic wave, silicon-photonic filters of microwave frequency with a single transmission passband. The filter response consists of up to 32 tap coefficients, and the transmission bandwidth is only 7 MHz. The results extend the capabilities of integrated microwave photonics in the standard silicon-on-insulator platform.

GeSn detectors have attracted a lot of attention for mid-infrared Si photonics, due to their compatibility with Si complementary metal oxide semiconductor technology. The GeSn bandgap can be affected by Sn composition and strain, which determines the working wavelength range of detectors. Applying the Sn content gradient GeSn layer structure, the strain of GeSn can be controlled from fully strained to completely relaxed. In this work, the strain evolution of GeSn alloys was investigated, and the effectiveness of gradually increasing Sn composition for the growth of high-Sn-content GeSn alloys was revealed. Relaxed GeSn thick films with Sn composition up to 16.3% were grown, and GeSn photodetectors were fabricated. At 77 K, the photodetectors showed a cutoff wavelength up to 4.2 μm and a peak responsivity of 0.35 A/W under 1 V at 2.53 μm. These results indicate that GeSn alloys grown on a Sn content gradient GeSn structure have promising application in mid-infrared detection.

We demonstrate monolithically integrated n-GaAs/p-Si depletion-type optical phase shifters fabricated on a 300 mm wafer-scale Si photonics platform. We measured the phase shifter performance using Mach–Zehnder modulators with the GaAs/Si optical phase shifters in both arms. A modulation efficiency of VπL as low as 0.3 V·cm has been achieved, which is much lower compared to a carrier-depletion type Si optical phase shifter with pn junction. While propagation loss is relatively high at ∼6.5 dB/mm, the modulator length can be reduced by the factor of ∼4.2 for the same optical modulation amplitude of a Si reference Mach–Zehnder modulator, owing to the high modulation efficiency of the shifters.

Mode-division multiplexing (MDM) can greatly improve the capacity of information transmission. The multimode waveguide bend (MWB) with small size and high performance is of great significance for the on-chip MDM integrated system. In this paper, an MWB with high performance based on double free-form curves (DFFCs) is proposed and realized. The DFFC is a combination of a series of arcs optimized by the inverse design method. The fabrication of this MWB only needs one-step lithography and plasma etching and has a large fabrication tolerance. MWBs with effective radii of 6 μm and 10 μm are designed to support three modes and four modes, respectively. The proposed method gives the best overall performance considering both the effective bending radius and the transmission efficiency. The fabricated MWB with four mode channels has low excess losses and crosstalks below -21 dB in the wavelength range from 1520 to 1580 nm. It is expected that this design can play an important role in promoting the dense integration of multimode transmission systems.

A solid-state active beamformer based on the annular-ring diffraction pattern is demonstrated in an integrated photonic platform. Such a circularly symmetric annular-ring aperture achieves a radiating element limited field of view. Furthermore, it is demonstrated that a multi-annular-ring aperture with a fixed linear density of elements maintains the beam efficiency for larger apertures while reducing the beamwidth and side-lobe level. A 255-element multi-annular-ring optical phased array with active beamforming is implemented in a standard photonics process. A total of 510 phase and amplitude modulators enable beamforming and beam steering using this aperture. A row–column drive methodology reduces the required electrical drivers by more than a factor of 5.

An ultrafast microring modulator (MRM) is fabricated and presented with Vπ·L of 0.825 V·cm. A 240 Gb/s PAM-8 signal transmission over 2 km standard single-mode fiber (SSMF) is experimentally demonstrated. PN junction doping concentration is optimized, and the overall performance of the MRM is improved. Optical peaking is introduced to further extend the EO bandwidth from 52 to 110 GHz by detuning the input wavelength. A titanium nitride heater with 0.1 nm/mW tuning efficiency is implemented above the MRM to adjust the resonant wavelength. High bit rate modulations based on the high-performance and compact MRM are carried out. By adopting off-line signal processing in the transmitter and receiver side, 120 Gb/s NRZ, 220 Gb/s PAM-4, and 240 Gb/s PAM-8 are measured with the back-to-back bit error ratio (BER) of 5.5×10-4, 1.5×10-2, and 1.4×10-2, respectively. A BER with different received optical power and 2 km SSMF transmission is also investigated. The BER for 220 Gb/s PAM-4 and 240 Gb/s PAM-8 after 2 km SSMF transmission is calculated to be 1.7×10-2 and 1.5×10-2, which meet with the threshold of soft-decision forward-error correction, respectively.

Metasurfaces have drawn considerable attention in manipulation of electromagnetic waves due to their exotic subwavelength footprints. Regardless of immense progress of polarization-dependent flat optics, the realization of on-device switchable complete phase multiplication is still missing from design multifunctional devices. Here, by combining geometric and propagation phases, a generalized design principle is proposed that can achieve switchable integer or fractional multiple complete phase modulation in transmitted circularly cross-polarized light by switching the handedness of incident polarization. As a proof of concept, two types of spin-dependent bifunctional wavefront manipulating devices, including switchable beam splitter/beam deflector and spin-to-orbital angular momentum converter designs are numerically realized. It is believed that the proposed single-cell spin-switchable rational-multiple complete-phase-modulation design principle based on combined propagation and geometric phases has great potential to underpin the development of meta-optics-based multifunctional operations in the field of integrated optics, imaging, and optical communication.

Silicon photonic Mach–Zehnder switches (MZSs) have been extensively investigated as a promising candidate for optical systems. However, conventional 2×2 MZSs are usually prone to the size variations of the arm waveguides due to imperfect fabrication, resulting in considerable random phase imbalance between the two arms, thereby imposing significant challenges for further developing next-generation N×N MZSs. Here we propose a novel design toward calibration-free 2×2 and N×N MZSs, employing optimally widened arm waveguides, enabled by novel compact tapered Euler S-bends with incorporated mode filters. With standard 180 nm CMOS foundry processes, more than thirty 2×2 MZSs and one 4×4 Benes MZS with the new design are fabricated and characterized. Compared with their conventional counterparts with 0.45-μm-wide arm waveguides, the present 2×2 MZSs exhibit significant reduction in the random phase imbalance. The measured extinction ratios of the present 2×2 and 4×4 MZSs operating in the all-cross state are 27-49 dB and ∼20 dB across the wavelength range of ∼60 nm, respectively, even without any calibrations. This work paves the way toward calibration-free large-scale N×N MZSs for next-generation silicon photonics.

Phase calibration for optical phased arrays (OPAs) is a key process to compensate for the phase deviation and retrieve the initial working state. Conventional calibration approaches based on iterative optimization algorithms are tedious and time-consuming. The essential difficulty of such a problem is to inversely solve for the phase error distribution among OPA elements from the far-field pattern of an OPA. Deep-learning-based technology might offer an alternative approach without explicitly knowing the inverse solution. However, we find that the phase ambiguities, including conjugate ambiguity and periodic ambiguity, severely deter the accuracy and efficacy of deep-learning-based calibration. Device-physics-based analysis reveals the causes of the phase ambiguities, which can be resolved by creating a tailored artificial neural network with phase-masked far-field patterns in a conjugate pair and constructing a periodic continuity-preserving loss function. Through the ambiguity-resolved neural network, we can extract phase error distribution in an OPA and calibrate the device in a rapid, noniterative manner from the measured far-field patterns. The proposed approach is experimentally verified. Pure main-beam profiles with >12 dB sidelobe suppression ratios are observed. This approach can help overcome a crucial bottleneck for the further advance of OPAs in a variety of applications such as lidar.

We propose and demonstrate an optical phased-array-based bidirectional grating antenna (BDGA) in silicon nitride waveguides. The BDGA is integrated with a miniaturized all-dielectric metasurface doublet (MD) formed on a glass substrate. The BDGA device, which takes advantage of alternately feeding light to its ports in opposite directions, is presumed to effectively provide a doubled wavelength-tuned steering efficiency compared to its unidirectional counterpart. The MD, which is based on vertically cascaded convex and concave metalenses comprising circular hydrogenated amorphous silicon nanopillars, is meticulously placed atop the BDGA chip to accept and deflect a beam emanating from the emission area, thereby boosting the beam-steering performance. The manufactured BDGA could achieve an enhanced beam-steering efficiency of 0.148 deg/nm as well as a stable spectral emission response in the wavelength range of 1530–1600 nm. By deploying a fabricated MD atop the silicon photonic BDGA chip, the steering efficiency was confirmed to be boosted by a factor of ∼3.1, reaching 0.461 deg/nm, as intended.

GeSn lasers enable the monolithic integration of lasers on the Si platform using all-group-IV direct-bandgap material. The GeSn laser study recently moved from optical pumping into electrical injection. In this work, we present explorative investigations of GeSn heterostructure laser diodes with various layer thicknesses and material compositions. Cap layer material was studied by using Si0.03Ge0.89Sn0.08 and Ge0.95Sn0.05, and cap layer total thickness was also compared. The 190 nm SiGeSn-cap device had threshold of 0.6 kA/cm2 at 10 K and a maximum operating temperature (Tmax) of 100 K, compared to 1.4 kA/cm2 and 50 K from 150 nm SiGeSn-cap device, respectively. Furthermore, the 220 nm GeSn-cap device had 10 K threshold at 2.4 kA/cm2 and Tmax at 90 K, i.e., higher threshold and lower maximal operation temperature compared to the SiGeSn cap layer, indicating that enhanced electron confinement using SiGeSn can reduce the threshold considerably. The study of the active region material showed that device gain region using Ge0.87Sn0.13 had a higher threshold and lower Tmax, compared to Ge0.89Sn0.11. The performance was affected by the metal absorption, free carrier absorption, and possibly defect density level. The maximum peak wavelength was measured as 2682 nm at 90 K by using Ge0.87Sn0.13 in gain regions. The investigations provide directions to the future GeSn laser diode designs toward the full integration of group-IV photonics on a Si platform.

On-chip ultrafast mode-locking lasers are basic building blocks for the realization of a chip-based optical frequency comb. In this paper, an ultrafast saturable absorber made up of a graphene pad on top of a silicon waveguide is applied to implement an ultrafast pulse laser. Benefiting from the small mode area of the graphene/silicon hybrid waveguide, the saturable pulse energy is reduced by two orders of magnitude compared with the fiber. A mode-locked pulse with a duration of 542 fs and a repetition rate of 54.37 MHz is realized. Pump–probe measurement shows that the carrier relaxation process of free carrier recombination with atomic-thin graphene/silicon junctions is three orders of magnitude faster than silicon, which plays a fundamental role in pulse narrowing. The chip-scale silicon ultrafast laser lays a foundation for a new class of nonlinear devices, in which a combination with multiple functional silicon photonic circuits enables efficient nonlinear interaction at the micrometer scale and less than 1 W of power consumption.

Mechanical strain engineering has been promising for many integrated photonic applications. However, for the engineering of a material electronic bandgap, a trade-off exists between the strain uniformity and the integration compatibility with photonic-integrated circuits (PICs). Herein, we adopted a straightforward recess-type design of a silicon nitride (SiNx) stressor to achieve a uniform strain with enhanced magnitude in the material of interest on PICs. Normal-incidence, uniformly 0.56% tensile strained germanium (Ge)-on-insulator (GOI) metal-semiconductor-metal photodiodes were demonstrated, using the recessed stressor with 750 MPa tensile stress. The device exhibits a responsivity of 1.84±0.15 A/W at 1550 nm. The extracted Ge absorption coefficient is enhanced by ～3.2× to 8340 cm-1 at 1612 nm and is superior to that of In0.53Ga0.47As up to 1630 nm limited by the measurement spectrum. Compared with the nonrecess strained device, additional absorption coefficient improvement of 10%–20% in the C-band and 40%–60% in the L-band was observed. This work facilitates the recess-strained GOI photodiodes for free-space PIC applications and paves the way for various (e.g., Ge, GeSn or III-V based) uniformly strained photonic devices on PICs.

An on-chip quadplexer is proposed and demonstrated with four wavelength-channels of 1270, 1310, 1490, and 1577 nm. The present quadplexer consists of four cascaded filters based on multimode waveguide grating (MWG), which are composed of a two-mode (de)multiplexer and an MWG. For the fabricated quadplexer on silicon, all four wavelength channels have flat-top responses with low excess losses of 0.5 dB as well as the desired bandwidths, which are about 16, 38, 19, and 6 nm, respectively. The cross-talk for both upstream channels and downstream channels is less than -24 dB. Moreover, the data transmission of 10 Gb/s of the present silicon quadplexer is also successfully demonstrated.

Up to now, the light coupling schemes of germanium-on-silicon photodetectors (Ge-on-Si PDs) could be divided into three main categories: (1) vertical (or normal-incidence) illumination, which can be from the top or back of the wafer/chip, and waveguide-integrated coupling including (2) butt coupling and (3) evanescent coupling. In evanescent coupling the input waveguide can be positioned on top, at the bottom, or lateral to the absorber. Here, to the best of our knowledge, we propose the first concept of Ge-on-Si PD with double lateral silicon nitride (Si3N4) waveguides, which can serve as a novel waveguide-integrated coupling configuration: double lateral coupling. The Ge-on-Si PD with double lateral Si3N4 waveguides features uniform optical field distribution in the Ge region, which is very beneficial to improving the operation speed for high input power. The proposed Ge-on-Si PD is comprehensively characterized by static and dynamic measurements. The typical internal responsivity is evaluated to be 0.52 A/W at an input power of 25 mW. The equivalent circuit model and theoretical 3 dB opto-electrical (OE) bandwidth investigation of Ge-on-Si PD with lateral coupling are implemented. Based on the small-signal (S21) radio-frequency measurements, under 4 mA photocurrent, a 60 GHz bandwidth operating at -3 V bias voltage is demonstrated. When the photocurrent is up to 12 mA, the 3 dB OE bandwidth still has 36 GHz. With 1 mA photocurrent, the 70, 80, 90, and 100 Gbit/s non-return-to-zero (NRZ) and 100, 120, 140, and 150 Gbit/s four-level pulse amplitude modulation clear openings of eye diagrams are experimentally obtained without utilizing any offline digital signal processing at the receiver side. In order to verify the high-power handling performance in high-speed data transmission, we investigate the eye diagram variations with the increase of photocurrents. The clear open electrical eye diagrams of 60 Gbit/s NRZ under 20 mA photocurrent are also obtained. Overall, the proposed lateral Si3N4 waveguide structure is flexibly extendable to a light coupling configuration of PDs, which makes it very attractive for developing high-performance silicon photonic integrated circuits in the future.

Silicon-based light sources, including light-emitting diodes (LEDs) and laser diodes (LDs) for information transmission, are urgently needed for developing monolithic integrated silicon photonics. Silicon with erbium ions (Er3+) doped by ion implantation is considered a promising approach, but it suffers from an extremely low quantum efficiency. Here we report an electrically pumped superlinear emission at 1.54 μm from Er/O-doped silicon planar LEDs, which are produced by applying a new deep cooling process. Stimulated emission at room temperature is realized with a low threshold current of ～6 mA (～0.8 A/cm2). Time-resolved photoluminescence and photocurrent results have revealed the complex carrier transfer dynamics by relaxing electrons from the Si conduction band to the Er3+ ion. This picture differs from the frequently assumed energy transfer via electron–hole pair recombination of the silicon host. Moreover, the amplified emission from the LEDs is likely due to a quasi-continuous Er/O-related donor band created by the deep cooling technique. This work paves the way for fabricating superluminescent diodes or efficient LEDs at communication wavelengths based on rare-earth-doped silicon.

Recently, 2-μm wave band has gained increasing interest due to its potential application for next-generation optical communication. But the development of 2-μm optical communications is substantially hampered by the modulation speed due to the device bandwidth constraints. Thus, a high-speed modulator is highly demanded at 2 μm. Motivated by this prospect, we demonstrate a high-speed silicon Mach–Zehnder modulator for a 2-μm wave band. The device is configured as a single-ended push–pull structure with waveguide electrorefraction via the free carrier plasma effect. The modulator was fabricated via a multiproject wafer shuttle run at a commercial silicon photonic foundry. The modulation efficiency of a single arm is measured to be 1.6 V·cm. The high-speed characterization is also performed, and the modulation speed can reach 80 Gbit/s with 4-level pulse amplitude modulation (PAM-4) formats.

We demonstrate a silicon electronic–photonic integrated 25 Gb/s nonreturn-to-zero transmitter that includes driver circuits, depletion-type Si ring modulator, Ge photodetector, temperature sensor, on-chip heater, and temperature controller, all monolithically integrated on a 0.25 μm photonic BiCMOS technology platform. The integrated transmitter successfully provides stable and optimal 25 Gb/s modulation characteristics against external temperature fluctuation.

We report the demonstration of a normal-incidence p-i-n germanium-tin (Ge0.951Sn0.049) photodetector on silicon-on-insulator substrate for 2 μm wavelength application. The DC and RF characteristics of the devices have been characterized. A dark current density under -1 V bias of approximately 125 mA/cm2 is achieved at room temperature, and the optical responsivity of 14 mA/W is realized for illumination wavelength of 2 μm under -1 V reverse bias. In addition, a 3 dB bandwidth (f3 dB) of around 30 GHz is achieved at -3 V, which is the highest reported value among all group III–V and group IV photodetectors working in the 2 μm wavelength range. This work illustrates that a GeSn photodetector has great prospects in 2 μm wavelength optical communication.

We experimentally demonstrate a silicon photonic modulator that can be loaded with a combination of lateral and interleaved p-n junctions to enhance its phase modulation. We use an asymmetric Bragg grating to introduce mode conversion in the active area, allowing the modulator to operate in reflection without introducing additional on-chip loss. With a compact footprint (phase shifter length of 290 μm), the modulator demonstrates a modulation speed up to 45 Gb/s with a bit error rate below the 7% forward-error-correction (FEC) threshold (up to 55 Gb/s with 20% FEC), and a low power consumption of 226 fJ/bit.

We report a CMOS-compatible silicon microring-enhanced avalanche photodiode based on linear defect-state absorption in a p+pn+ junction, with high responsivities exceeding 1 A/W at telecommunication wavelengths. The large photogenerated currents give rise to giant thermo-optic nonlinearity in the microring resonator, resulting in a linear photocurrent-wavelength response spanning the full free spectral range of the microring. This unique photocurrent spectrum could enable novel applications in wavelength-resolved photodetection, such as compact on-chip spectrometers, linear chirp frequency laser source characterization, and low-cost refractometric sensors without requiring precise wavelength-tunable lasers.

Flat electro-optical frequency combs play an important role in a wide range of applications, such as metrology, spectroscopy, or microwave photonics. As a key technology for the integration of optical circuits, silicon photonics could benefit from on-chip, tunable, flat frequency comb generators. In this article, two different architectures based on silicon modulators are studied for this purpose. They rely on a time to frequency conversion principle to shape the comb envelope. Using a numerical model of the silicon traveling-wave phase modulators, their driving schemes are optimized before their performances are simulated and compared. A total of nine lines could be obtained within a 2 dB flatness, with a line-spacing ranging from 0.1 to 7 GHz. Since this tunability is a major asset of electro-optical frequency combs, the effect of segmenting the phase modulators is finally investigated, showing that the flat lines spacing could be extended up to 39 GHz by this method.

Quantum dot lasers are excellent on-chip light sources, offering high defect tolerance, low threshold, low temperature variation, and high feedback insensitivity. Yet a monolithic integration technique combining epitaxial quantum dot lasers with passive waveguides has not been demonstrated and is needed for complex photonic integrated circuits. We present here, for the first time to our knowledge, a monolithc offset quantum dot integration platform that permits formation of a laser cavity utilizing both the robust quantum dot active region and the versatility of passive GaAs waveguide structures. This platform is substrate agnostic and therefore compatible with the quantum dot lasers directly grown on Si. As an illustration of the potential of this platform, we designed and fabricated a 20 GHz mode-locked laser with a dispersion-engineered on-chip waveguide mirror. Due to the dispersion compensation effect of the waveguide mirror, the pulse width of the mode-locked laser is reduced by a factor of 2.8.

The operation of quantum dot lasers epitaxially grown on silicon is investigated through a quantum-corrected Poisson-drift-diffusion model. This in-house developed simulation framework completes the traditional rate equation approach, which models the intersubband transitions involved into simultaneous ground-state and excited-state lasing, with a physics-based description of carrier transport and electrostatic effects. The code is applied to look into some of the most relevant mechanisms affecting the lasing operation. We analyze the impact of threading dislocations on non-radiative recombination and laser threshold current. We demonstrate that asymmetric carrier transport in the barrier explains the ground-state power quenching above the excited-state lasing threshold. Finally, we study p-type modulation doping and its benefits/contraindications. The observation of an optimum doping level, minimizing the ground-state lasing threshold current, stems from the reduction of the electron density, which counteracts the benefits from the expected increase of the hole density. This reduction is due to electrostatic effects hindering electron injection.

All-optical modulation based on the photothermal effect of two-dimensional (2D) materials shows great promise for all-optical signal processing and communication. In this work, an all-optical modulator with a 2D PtSe2-on-silicon structure based on a microring resonator is proposed and demonstrated utilizing the photothermal effect of PtSe2. A tuning efficiency of 0.0040 nm · mW?1 is achieved, and the 10%–90% rise and decay times are 304 μs and 284 μs, respectively. The fabricated device exhibits a long-term air stability of more than 3 months. The experimental results prove that 2D PtSe2 has great potential for optical modulation on a silicon photonic platform.

Artificial structures that exhibit narrow resonance features are key to a myriad of scientific advances and technologies. In particular, exploration of the terahertz (THz) spectrum—the final frontier of the electromagnetic spectrum—would greatly benefit from high-quality resonant structures. Here we present a new paradigm of terahertz silicon disc microresonators with subwavelength thickness. Experimental results utilizing continuous-wave THz spectroscopy establish quality factors in excess of 120,000 at 0.6 THz. Reduction of the disc thickness to a fraction of the wavelength reduces the losses from the silicon substrate and paves the way to unparalleled possibilities for light–matter interaction in the THz frequency range.

A novel wavelength-selective 2×2 optical switch based on a Ge2Sb2Te5 (GST)-assisted microring-resonator (MRR) is proposed. The present GST-assisted MRR consists of two access optical waveguides and an MRR coupled with a bent GST-loaded silicon photonic waveguide. The 2×2 optical switch is switched ON or OFF by modifying the GST state to be crystalline or amorphous. In particular, the microring waveguide and the bent GST-loaded waveguide are designed to satisfy the phase-matching condition when the GST is crystalline. As a result, the MRR becomes highly lossy and the resonance peak is depressed significantly. On the other hand, when it is off, there is little coupling due to the significant phase mismatching. Consequently, one has a low-loss transmission at the drop port for the resonance wavelength. In this paper, the simulation using the three-dimensional finite-difference method shows that the extinction ratio of the designed photonic switch is ～20 dB at the resonance wavelength, while the excess losses at the through port and drop port are 0.9 dB and 2 dB. In particular, the resonance wavelength changes little between the ON and OFF states, which makes it suitable for multichannel wavelength-division-multiplexing systems.

We present an experimental demonstration of notch filters with arbitrary center wavelengths capable of tunable analog output power values varying between full extinction of 15 and 0 dB. Each filter is composed of highly modular apodized four-port Bragg add/drop filters to reduce the crosstalk between concatenated devices. The constructed photonic integrated circuit experimentally demonstrates spectra shaping using four independent notch filters. Each notch filter supports a bandwidth of ～2 nm and is shown to be suitable for realization of programmable photonic integrated circuits.

The hybrid multiplexing technique reactivates optical interconnect as it offers multiple dimensions to dramatically enhance the data capacity of a single wavelength carrier. A straightforward method to realize hybrid multiplexing is to perform polarization multiplexing for mode-multiplexed signals, by utilizing a mode-transparent polarization beam splitter (MTPBS) which can process multiple modes simultaneously. However, present PBSs mainly work in the single-mode regime, and it is not easy to redesign the conventional PBS to accommodate multiple modes, due to the severe mode dispersion. Here, a novel MTPBS, which can tackle a group of modes simultaneously, is proposed and demonstrated. As a demonstration, the MTPBS supporting a total channel number of 13 is experimentally achieved, with low insertion loss and low modal/polarization cross talk. This work provides a new insight to realize hybrid multiplexing and represents a solution for high-density and large-capacity photonic integration.

A SiN-Si dual-layer optical phased array (OPA) chip is designed and fabricated. It combines the low loss of SiN with the excellent modulation performance of Si, which improves the performance of Si single-layer OPA. A novel optical antenna and an improved phase modulation method are also proposed, and a two-dimensional scanning range of 96°×14° is achieved, which makes the OPA chip more practical.

A horizontal p?i?n ridge waveguide emitter on a silicon (100) substrate with a Ge0.91Sn0.09/Ge multi-quantum-well (MQW) active layer was fabricated by molecular beam epitaxy. The device structure was designed to reduce light absorption of metal electrodes and improve injection efficiency. Electroluminescence (EL) at a wavelength of 2160 nm was observed at room temperature. Theoretical calculations indicate that the emission peak corresponds well to the direct bandgap transition (n1Γ?n1HH). The light output power was about 2.0 μW with an injection current density of 200 kA/cm2. These results show that the horizontal GeSn/Ge MQW ridge waveguide emitters have great prospects for group-IV light sources.

Wavelength-dependent polarization state of light carries crucial information about light–matter interactions. However, its measurement is limited to bulky, high energy-consuming devices, which prohibits many modern, portable applications. Here, we propose and demonstrate a chip-scale spectropolarimeter implemented using a complementary metal oxide semiconductor compatible silicon photonics technology. Four compact Vernier microresonator spectrometers are monolithically integrated with a broadband polarimeter consisting of a 2D nanophotonic antenna and a polarimetric circuit to achieve full-Stokes spectropolarimetric analysis. The proposed device offers a solid-state spectropolarimetry solution with a small footprint of 1 mm × 0.6 mm and low power consumption of 360 mW. Full-Stokes spectral detection across a broad spectral range of 50 nm with a resolution of 1 nm is demonstrated in characterizing a material possessing structural chirality. The proposed device may enable a broader application of spectropolarimetry in the fields ranging from biomedical diagnostics and chemical analysis to observational astronomy.

Radio frequency (RF) switches are essential for implementing routing of RF signals. However, the increasing demand for RF signal frequency and bandwidth is posing a challenge of switching speed to the conventional solutions, i.e., the capability of operating at a sub-nanosecond speed or faster. In addition, signal frequency reconfigurability is also a desirable feature to facilitate new innovations of flexible system functions. Utilizing microwave photonics as an alternative path, we present here a photonic implementation of an RF switch providing not only the capability of switching at a sub-nanosecond speed but also options of frequency doubling of the input RF signals, allowing for flexible output waveforms. The core device is a traveling-wave silicon modulator with a device size of 0.2 mm×1.8 mm and a modulation bandwidth of 10 GHz. Using microwave frequencies, i.e., 15 GHz and 20 GHz, as two simultaneous RF input signals, we experimentally demonstrated their amplitude and frequency switching as well as that of the doubled frequencies, i.e., 30 GHz and 40 GHz, at a switching frequency of 5 GHz. The results of this work point to a solution for creating high-speed RF switches with high compactness and flexibility.

An ultrahigh-Q silicon racetrack resonator is proposed and demonstrated with uniform multimode silicon photonic waveguides. It consists of two multimode straight waveguides connected by two multimode waveguide bends (MWBs). In particular, the MWBs are based on modified Euler curves, and a bent directional coupler is used to achieve the selective mode coupling for the fundamental mode and not exciting the higher-order mode in the racetrack. In this way, the fundamental mode is excited and propagates in the multimode racetrack resonator with ultralow loss and low intermode coupling. Meanwhile, it helps achieve a compact 180° bend to make a compact resonator with a maximized free spectral range (FSR). In this paper, for the chosen 1.6 μm wide silicon photonic waveguide, the effective radius Reff of the designed 180° bend is as small as 29 μm. The corresponding FSR is about 0.9 nm when choosing 260 μm long straight waveguides in the racetrack. The present high-Q resonator is realized with a simple standard single-etching process provided by a multiproject wafer foundry. The fabricated device, which has a measured intrinsic Q-factor as high as 2.3×106, is the smallest silicon resonator with a >106Q-factor.

We demonstrate the possibility of post-fabrication trimming of the response of nitrogen-rich silicon nitride racetrack resonators by using an ultraviolet laser. The results revealed the possibility to efficiently tune the operating wavelength of fabricated racetrack resonators to any point within the full free spectral range. This process is much faster than similar, previously presented methods (in the order of seconds, compared to hours). This technique can also be applied to accurately trim the optical performance of any other silicon photonic device based on nitrogen-rich silicon nitride.

Graphene resting on a silicon-on-insulator platform offers great potential for optoelectronic devices. In the paper, we demonstrate all-optical modulation on the graphene–silicon hybrid waveguides (GSHWs) with tens of micrometers in length. Owing to strong interaction between graphene and silicon strip waveguides with compact light confinement, the modulation depth reaches 22.7% with a saturation threshold down to 1.38 pJ per pulse and a 30-μm-long graphene pad. A response time of 1.65 ps is verified by a pump–probe measurement with an energy consumption of 2.1 pJ. The complementary metal-oxide semiconductor compatible GSHWs with the strip configuration exhibit great potential for ultrafast and broadband all-optical modulation, indicating that employing two-dimensional materials has become a complementary technology to promote the silicon photonic platform.

Reducing power dissipation in electro-optic modulators is a key step for widespread application of silicon photonics to optical communication. In this work, we design Mach–Zehnder modulators in the silicon-on-insulator platform, which make use of slow light in a waveguide grating and of a reverse-biased p-n junction with interleaved contacts along the waveguide axis. After optimizing the junction parameters, we discuss the full simulation of the modulator in order to find a proper trade-off among various figures of merit, such as modulation efficiency, insertion loss, cutoff frequency, optical modulation amplitude, and dissipated energy per bit. Comparison with conventional structures (with lateral p-n junction and/or in rib waveguides without slow light) highlights the importance of combining slow light with the interleaved p-n junction, thanks to the increased overlap between the travelling optical wave and the depletion regions. As a surprising result, the modulator performance is improved over an optical bandwidth that is much wider than the slow-light bandwidth.

We propose a novel concept of designing silicon photonics metamaterials for perfect near-infrared light absorption. The study’s emphasis is an in-depth investigation of various physical mechanisms behind the ～100% ultra-narrowband record peak absorptance of the designed structures, comprising an ultrathin silicon absorber. The electromagnetic power transport, described by the Poynting vector, is innovatively explored, which shows combined vortex and crossed-junction two-dimensional waveguide-like flows as outcomes of optical field singularities. These flows, though peculiar for each of the designed structures, turn out to be key factors of the perfect resonant optical absorption. The electromagnetic fields show tight two-dimensional confinement: a sharp vertical confinement of the resonant-cavity type combined with a lateral metasurface supported confinement. The silicon-absorbing layer and its oxide environment are confined between two subwavelength metasurfaces such that the entire design is well compatible with silicon-on-insulator microelectronics. The design concept and its outcomes meet the extensive challenges of ultrathin absorbers for minimum noise and an ultra-narrowband absorptance spectrum, while maintaining an overall very thin structure for planar integration. With these materials and such objectives, the proposed designs seem essential, as standard approaches fail, mainly due to a very low silicon absorption coefficient over the near-infrared range. Tolerance tests for fabrication errors show fair tolerability while maintaining a high absorptance peak, along with a controllable deviation off the central-design wavelength. Various applications are suggested and analyzed, which include but are not limited to: efficient photodetectors for focal plane array and on-chip integrated silicon photonics, high-precision spectroscopic chemical and angular-position sensing, and wavelength-division multiplexing.

Chip-scale, tunable narrow-linewidth hybrid integrated diode lasers based on quantum-dot RSOAs at 1.3 &mu;m are demonstrated through butt-coupling to a silicon nitride photonic integrated circuit. The hybrid laser linewidth is around 85 kHz, and the tuning range is around 47 nm. Then, a fully integrated beam steerer is demonstrated by combining the tunable diode laser with a waveguide surface grating. Our system can provide beam steering of 4.1&deg; in one direction by tuning the wavelength of the hybrid laser. Besides, a wavelength-tunable triple-band hybrid laser system working at ～1, ～1.3, and ～1.55 &mu;m bands is demonstrated for wide-angle beam steering in a single chip.

Optical phase shifters are extensively used in integrated optics not only for telecom and datacom applications but also for sensors and quantum computing. While various active solutions have been demonstrated, progress in passive phase shifters is still lacking. Here we present a new type of ultra-broadband 90° phase shifter, which exploits the anisotropy and dispersion engineering in subwavelength metamaterial waveguides. Our Floquet–Bloch calculations predict a phase-shift error below ±1.7° over an unprecedented operation range from 1.35 to 1.75 μm, i.e., 400 nm bandwidth covering the E, S, C, L, and U telecommunication bands. The flat spectral response of our phase shifter is maintained even in the presence of fabrication errors up to ±20 nm, showing greater robustness than conventional structures. Our device was experimentally demonstrated using standard 220 nm thick SOI wafers, showing a fourfold reduction in the phase variation compared to conventional phase shifters within the 145 nm wavelength range of our measurement setup. The proposed subwavelength engineered phase shifter paves the way for novel photonic integrated circuits with an ultra-broadband performance.

We report supercontinuum generation in nitrogen-rich (N-rich) silicon nitride waveguides fabricated through back-end complementary-metal-oxide-semiconductor (CMOS)-compatible processes on a 300 mm platform. By pumping in the anomalous dispersion regime at a wavelength of 1200 nm, two-octave spanning spectra covering the visible and near-infrared ranges, including the O band, were obtained. Numerical calculations showed that the nonlinear index of N-rich silicon nitride is within the same order of magnitude as that of stoichiometric silicon nitride, despite the lower silicon content. N-rich silicon nitride then appears to be a promising candidate for nonlinear devices compatible with back-end CMOS processes.

We report an advanced Fourier transform spectrometer (FTS) on silicon with significant improvement compared with our previous demonstration in [Nat. Commun.9, 665 (2018)2041-1723]. We retrieve a broadband spectrum (7 THz around 193 THz) with 0.11 THz or sub nm resolution, more than 3 times higher than previously demonstrated [Nat. Commun.9, 665 (2018)2041-1723]. Moreover, it effectively solves the issue of fabrication variation in waveguide width, which is a common issue in silicon photonics. The structure is a balanced Mach–Zehnder interferometer with 10 cm long serpentine waveguides. Quasi-continuous optical path difference between the two arms is induced by changing the effective index of one arm using an integrated heater. The serpentine arms utilize wide multi-mode waveguides at the straight sections to reduce propagation loss and narrow single-mode waveguides at the bending sections to keep the footprint compact and avoid modal crosstalk. The reduction of propagation loss leads to higher spectral efficiency, larger dynamic range, and better signal-to-noise ratio. Also, for the first time to our knowledge, we perform a thorough systematic analysis on how the fabrication variation on the waveguide widths can affect its performance. Additionally, we demonstrate that using wide waveguides efficiently leads to a fabrication-tolerant device. This work could further pave the way towards a mature silicon-based FTS operating with both broad bandwidth (over 60 nm) and high resolution suitable for integration with various mobile platforms.

Integrated photonics is poised to become a mainstream solution for high-speed data communications and sensing in harsh radiation environments, such as outer space, high-energy physics facilities, nuclear power plants, and test fusion reactors. Understanding the impact of radiation damage in optical materials and devices is thus a prerequisite to building radiation-hard photonic systems for these applications. In this paper, we report real-time, in situ analysis of radiation damage in integrated photonic devices. The devices, integrated with an optical fiber array package and a baseline-correction temperature sensor, can be remotely interrogated while exposed to ionizing radiation over a long period without compromising their structural and optical integrity. We also introduce a method to deconvolve the radiation damage responses from different constituent materials in a device. The approach was implemented to quantify gamma radiation damage and post-radiation relaxation behavior of SiO2-cladded SiC photonic devices. Our findings suggest that densification induced by Compton scattering displacement defects is the primary mechanism for the observed index change in SiC. Additionally, post-radiation relaxation in amorphous SiC does not restore the original pre-irradiated structural state of the material. Our results further point to the potential of realizing radiation-hard photonic device designs taking advantage of the opposite signs of radiation-induced index changes in SiC and SiO2.

Silicon photonics is coming of age; however, it is still lacking a monolithic platform for optical sources and nonlinear functionalities prompting heterogeneous integration of different materials tailored to different applications. Here we demonstrate tellurium oxide as a complementary metal oxide semiconductor silicon photonics platform for nonlinear functionalities, which is already becoming an established platform for sources and amplifiers. We show broadband supercontinuum generation covering the entire telecom window and show for the first time to our knowledge third-harmonic generation in its integrated embodiment. Together with the now-available lasers and amplifiers on integrated TeO2 this work paves the way for a monolithic TeO2-based nonlinear silicon photonics platform.

We demonstrate the optical transmission of an 800 Gbit/s (4×200 Gbit/s) pulse amplitude modulation-4 (PAM-4) signal and a 480 Gbit/s (4×120 Gbit/s) on–off-keying (OOK) signal by using a high-bandwidth (BW) silicon photonic (SiP) transmitter with the aid of digital signal processing (DSP). In this transmitter, a four-channel SiP modulator chip is co-packaged with a four-channel driver chip, with a measured 3 dB BW of 40 GHz. DSP is applied in both the transmitter and receiver sides for pre-/post-compensation and bit error rate (BER) calculation. Back-to-back (B2B) BERs of the PAM-4 signal and OOK signal are first measured for each channel of the transmitter with respect to a variety of data rates. Similar BER performance of four channels shows good uniformity of the transmitter between different channels. The BER penalty of the PAM-4 and OOK signals for 500 m and 1 km standard single-mode fiber (SSMF) transmission is then experimentally tested by using one channel of the transmitter. For a 200 Gbit/s PAM-4 signal, the BER is below the hard-decision forward error correction (HD-FEC) threshold for B2B and below the soft-decision FEC (SD-FEC) threshold after 1 km transmission. For a 120 Gbit/s OOK signal, the BER is below SD-FEC threshold for B2B. After 500 m and 1 km transmission, the data rate of the OOK signal shrinks to 119 Gbit/s and 118 Gbit/s with the SD-FEC threshold, respectively. Finally, the 800 Gbit/s PAM-4 signal with 1 km transmission is achieved with the BER of all four channels below the SD-FEC threshold.

An internal photoemission-based silicon photodetector detects light below the silicon bandgap at room temperature and can exhibit spectrally broad behavior, making it potentially suited to meet the need for a near-infrared pure Si photodetector. In this work, the implementation of a thin Au insertion layer into an ITO/n-Si Schottky photodetector can profoundly affect the barrier height and significantly improve the device performance. By fabricating a nanoscale thin Au layer and an ITO electrode on a silicon substrate, we achieve a well-behaved ITO/Au/n-Si Schottky diode with a record dark current density of 3.7×10-7 A/cm2 at -1 V and a high rectification ratio of 1.5×108 at ±1 V. Furthermore, the responsivity has been obviously improved without sacrificing the dark current performance of the device by decreasing the Au thickness. Such a silicon-based photodetector with an enhanced performance could be a promising strategy for the realization of a monolithic integrated pure silicon photodetector in optical communication.

A high-speed evanescent-coupled Ge waveguide electro-absorption modulator (EAM) with simple fabrication processes was realized on a silicon-on-insulator platform with a 220 nm top Si layer. Selectively grown Ge with a triangle shape was directly used for Ge waveguides of the EAM. An asymmetric p-i-n junction was designed in the Ge waveguide to provide a strong electric field for Franz–Keldysh effect. The insertion loss of the Ge EAM was 6.2 dB at 1610 nm. The EAM showed the high electro-optic bandwidth of 36 GHz at -1 V. Clear open 56 Gbps eye diagrams were observed at 1610 nm with a dynamic extinction ratio of 2.7 dB and dynamic power consumption of 45 fJ/bit for voltage swing of 3Vpp.

Heterogeneously integrated lasers in the O-band are a key component in realizing low-power optical interconnects for data centers and high-performance computing. Quantum-dot-based materials have been particularly appealing for light generation due to their ultralow lasing thresholds, small linewidth enhancement factor, and low sensitivity to reflections. Here, we present widely tunable quantum-dot lasers heterogeneously integrated on silicon-on-insulator substrate. The tuning mechanism is based on Vernier dual-ring geometry, and a 47 nm tuning range with 52 dB side-mode suppression ratio is observed. These parameters show an increase to 52 nm and 58 dB, respectively, when an additional wavelength filter in the form of a Mach–Zehnder interferometer is added to the cavity. The Lorentzian linewidth of the lasers is measured as low as 5.3 kHz.

Based on a silicon platform, we design and fabricate a four-mode division (de)multiplexer for chip-scale optical data transmission in the 2 μm waveband for the first time, to the best of our knowledge. The (de)multiplexer is composed of three tapered directional couplers for both mode multiplexing and demultiplexing processes. In the experiment, the average crosstalk for four channels is measured to be less than 18 dB over a wide wavelength range (70 nm) from 1950 to 2020 nm, and the insertion losses are also assessed. Moreover, we further demonstrate stable 5 Gbit/s direct modulation data transmission through the fabricated silicon photonic devices with non-return-to-zero on–off keying signals. The experimental results show clear eye diagrams, and the penalties at a bit error rate of 3.8×10 3 are all less than 2.5 dB after on-chip data transmission. The obtained results indicate that the presented silicon four-mode division multiplexer in the mid-infrared wavelength band might be a promising candidate facilitating chip-scale high-speed optical interconnects.

We present an accurate, easy-to-use large-signal SPICE circuit model for depletion-type silicon ring modulators (Si RMs). Our model includes both the electrical and optical characteristics of the Si RM and consists of circuit elements whose values change depending on modulation voltages. The accuracy of our model is confirmed by comparing the SPICE simulation results of 25 Gb/s non-return-to-zero (NRZ) modulation with the measurement. The model is used for performance optimization of monolithically integrated Si photonic NRZ and pulse-amplitude-modulation 4 transmitters in the standard SPICE circuit design environment.

A 4–λ hybrid InGaAsP-Si evanescent laser array is obtained by bonding III–V distributed feedback lasers to a silicon on insulator (SOI) substrate using a selective area metal bonding technique. Multiple wavelengths are realized by varying the width of the III–V ridge waveguide. The threshold current is less than 10?mA for all wavelength channels under continuous-wave (CW) operation at room temperature, and the lowest threshold current density is 0.76??kA/cm2. The side mode suppression ratio (SMSR) is higher than 40?dB for all wavelength channels when the injection current is between 20?mA and 70?mA at room temperature, and the highest SMSR is up to 51?dB. A characteristic temperature of 51?K and thermal impedance of 144°C/W are achieved on average. The 4–λ hybrid InGaAsP-Si evanescent laser array exhibits a low threshold and high SMSR under CW operation at room temperature. The low power consumption of this device makes it very attractive for on-chip optical interconnects.

A high-performance monolithic integrated wavelength division multiplexing silicon (Si) photonics receiver chip is fabricated on a silicon-on-insulator platform. The receiver chip has a 25-channel Si nanowire-arrayed waveguide grating, and each channel is integrated with a high-speed waveguide Ge-on-Si photodetector. The central wavelength, optical insertion loss, and cross talk of the array waveguide grating are 1550.6?nm, 5–8?dB, and ?12–?15??dB, respectively. The photodetectors show low dark current density of 16.9??mA/cm2 at ?1??V and a high responsivity of 0.82?A/W at 1550?nm. High bandwidths of 23 and 29?GHz are achieved at 0 and ?1??V, respectively. Each channel can operate at 50?Gbps with low input optical power even under zero bias, which realizes an aggregate data rate of 1.25?Tbps.

Transparent conductive oxides have attracted escalating research interest for integrated photonic devices and metasurfaces due to the extremely large electro-optic modulation of the refractive index by the free-carrier-induced plasma dispersion effect. In this paper, we have designed and fabricated a silicon microring resonator using an indium-tin oxide gate as the electric-tuning electrode. It achieved an ultralarge resonance wavelength tunability of 271 pm/V, which is obtained through the reduced width of the ring waveguide and a high-dielectric-constant HfO2 insulator. We demonstrated a broad resonance wavelength tuning range of over 2 nm with an ultrafast response time of less than 12 ns and near-zero static power consumption, which outperforms traditional thermal tuning.

Near-infrared germanium (Ge) photodetectors monolithically integrated on top of silicon-on-insulator substrates are universally regarded as key enablers towards chip-scale nanophotonics, with applications ranging from sensing and health monitoring to object recognition and optical communications. In this work, we report on the high-data-rate performance pin waveguide photodetectors made of a lateral hetero-structured silicon-Ge-silicon (Si-Ge-Si) junction operating under low reverse bias at 1.55 μm. The pin photodetector integration scheme considerably eases device manufacturing and is fully compatible with complementary metal-oxide-semiconductor technology. In particular, the hetero-structured Si-Ge-Si photodetectors show efficiency-bandwidth products of ～9 GHz at ?1 V and ～30 GHz at ?3 V, with a leakage dark current as low as ～150 nA, allowing superior signal detection of high-speed data traffic. A bit-error rate of 10?9 is achieved for conventional 10 Gbps, 20 Gbps, and 25 Gbps data rates, yielding optical power sensitivities of ?13.85 dBm, ?12.70 dBm, and ?11.25 dBm, respectively. This demonstration opens up new horizons towards cost-effective Ge pin waveguide photodetectors that combine fast device operation at low voltages with standard semiconductor fabrication processes, as desired for reliable on-chip architectures in next-generation nanophotonics integrated circuits.

Developing a low-cost, room-temperature operated and complementary metal-oxide-semiconductor (CMOS) compatible visible-blind short-wavelength infrared (SWIR) silicon photodetector is of interest for security, telecommunications, and environmental sensing. Here, we present a silver-supersaturated silicon (Si:Ag)-based photodetector that exhibits a visible-blind and highly enhanced sub-bandgap photoresponse. The visible-blind response is caused by the strong surface-recombination-induced quenching of charge collection for short-wavelength excitation, and the enhanced sub-bandgap response is attributed to the deep-level electron-traps-induced band-bending and two-stage carrier excitation. The responsivity of the Si:Ag photodetector reaches 504 mA· W 1 at 1310 nm and 65 mA ·W 1 at 1550 nm under 3 V bias, which stands on the stage as the highest level in the hyperdoped silicon devices previously reported. The high performance and mechanism understanding clearly demonstrate that the hyperdoped silicon shows great potential for use in optical interconnect and power-monitoring applications.

An on-chip, high extinction ratio transverse electric (TE)-pass polarizer using a silicon hybrid plasmonic grating is proposed and experimentally demonstrated. Utilizing plasmonics to manipulate the effective index and mode distribution, the transverse magnetic mode is reflected and absorbed, while the TE mode passes through with relatively low propagation loss. For a 6-μm-long device, the measurement result shows that the extinction ratio in the wavelength range of 1.52 to 1.58 μm varies from 24 to 33.7 dB and the insertion loss is 2.8–4.9 dB. Moreover, the structure exhibits large alignment tolerance and is compatible with silicon-on-insulator fabrication technology.

Slow light, a technology to control the optical signal by reducing the group velocity, has been widely studied to obtain enhanced nonlinearities and increased phase shifts owing to its promoting of the light–matter interaction ability. In this work, a wideband slow light is achieved in a simple one-dimensional fishbone grating waveguide. A flat band indicating slow light with a group index of 13 and bandwidth over 10 nm is obtained by the plane wave expansion calculation, and the corresponding experimental results agree well with the theoretical prediction. A step taper is designed to compensate the coupling loss. The proposed fishbone grating waveguide is a good candidate for wideband slow light devices in light communication.

Over the last 20 years, silicon photonics has revolutionized the field of integrated optics, providing a novel and powerful platform to build mass-producible optical circuits. One of the most attractive aspects of silicon photonics is its ability to provide extremely small optical components, whose typical dimensions are an order of magnitude smaller than those of optical fiber devices. This dimension difference makes the design of fiber-to-chip interfaces challenging and, over the years, has stimulated considerable technical and research efforts in the field. Fiber-to-silicon photonic chip interfaces can be broadly divided into two principle categories: in-plane and out-of-plane couplers. Devices falling into the first category typically offer relatively high coupling efficiency, broad coupling bandwidth (in wavelength), and low polarization dependence but require relatively complex fabrication and assembly procedures that are not directly compatible with wafer-scale testing. Conversely, out-of-plane coupling devices offer lower efficiency, narrower bandwidth, and are usually polarization dependent. However, they are often more compatible with high-volume fabrication and packaging processes and allow for on-wafer access to any part of the optical circuit. In this paper, we review the current state-of-the-art of optical couplers for photonic integrated circuits, aiming to give to the reader a comprehensive and broad view of the field, identifying advantages and disadvantages of each solution. As fiber-to-chip couplers are inherently related to packaging technologies and the co-design of optical packages has become essential, we also review the main solutions currently used to package and assemble optical fibers with silicon-photonic integrated circuits.

We demonstrate low-loss hydrogenated amorphous silicon (a-Si:H) waveguides by hot-wire chemical vapor deposition (HWCVD). The effect of hydrogenation in a-Si at different deposition temperatures has been investigated and analyzed by Raman spectroscopy. We obtained an optical quality a-Si:H waveguide deposited at 230°C that has a strong Raman peak shift at 480 cm 1, peak width (full width at half-maximum) of 68.9 cm 1, and bond angle deviation of 8.98°. Optical transmission measurement shows a low propagation loss of 0.8 dB/cm at the 1550 nm wavelength, which is the first, to our knowledge, report for a HWCVD a-Si:H waveguide.

Photonic-assisted microwave frequency identification with distinct features, including wide frequency coverage and fast tunability, has been conceived as a key technique for applications such as cognitive radio and dynamic spectrum access. The implementations based on compact integrated photonic chips have exhibited distinct advantages in footprint miniaturization, light weight, and low power consumption, in stark contrast with discrete optical-fiber-based realization. However, reported chip-based instantaneous frequency measurements can only operate at a single-tone input, which stringently limits their practical applications that require wideband identification capability in modern RF and microwave applications. In this article, we demonstrate, for the first time, a wideband, adaptive microwave frequency identification solution based on a silicon photonic integrated chip, enabling the identification of different types of microwave signals from 1 to 30 GHz, including single-frequency, multiple-frequency, chirped-frequency, and frequency-hopping microwave signals, and even their combinations. The key component is a high Q-factor scanning filter based on a silicon microring resonator, which is used to implement frequency-to-time mapping. This demonstration opens the door to a monolithic silicon platform that makes possible a wideband, adaptive, and high-speed signal identification subsystem with a high resolution and a low size, weight, and power (SWaP) for mobile and avionic applications.

We report on the first monolithically integrated microring-based optical switch in the switch-and-select architecture. The switch fabric delivers strictly non-blocking connectivity while completely canceling the first-order crosstalk. The 4×4 switching circuit consists of eight silicon microring-based spatial (de-)multiplexers interconnected by a Si/SiN dual-layer crossing-free central shuffle. Analysis of the on-state and off-state power transfer functions reveals the extinction ratios of individual ring resonators exceeding 25 dB, leading to switch crosstalk suppression of up to over 50 dB in the switch-and-select topology. Optical paths are assessed, showing losses as low as 0.1 dB per off-resonance ring and 0.5 dB per on-resonance ring. Photonic switching is actuated with integrated micro-heaters to give an ～24 GHz passband. The fully packaged device is flip-chip bonded onto a printed circuit board breakout board with a UV-curved fiber array.

We demonstrate a silicon-based microwave photonic filter (MPF) with flattop passband and adjustable bandwidth. The proposed MPF is realized by using a 10th-order microring resonator (MRR) and a photodetector, both of which are integrated on a photonic chip. The full width at half-maximum (FWHM) bandwidth of the optical filter achieved at the drop port of the 10th-order MRR is 21.6 GHz. The ripple of the passband is less than 0.3 dB, while the rejection ratio is 32 dB. By adjusting the deviation of the optical carrier wavelength from the center wavelength of the optical bandpass filter, the bandwidth of the MPF can be greatly changed. In the experiment, the FWHM bandwidth of the proposed MPF is tuned from 5.3 to 19.5 GHz, and the rejection ratio is higher than 30 dB.

Polarizers have been widely used in various optical systems to reduce polarization cross talk. The polarizers based on the silicon nanowire waveguide can provide chip-scale device size and a high polarization extinction ratio. However, the working bandwidth for the on-chip silicon polarizers is always limited (～100 nm) by the strong waveguide dispersion. In this paper, an on-chip all-silicon polarizer with an extremely broad working bandwidth is proposed and demonstrated. The device is based on a 180° sharp waveguide bend, assisted with anisotropic subwavelength grating (SWG) metamaterial cladding to enhance the polarization selectivity. For TE polarization, the effective refractive index for SWG is extraordinary, so the incident TE mode can propagate through the sharp waveguide bend. For TM polarization, the effective refractive index for SWG is ordinary, so the incident TM mode will be coupled into the radiation mode regardless of the wavelength. The fabricated polarizer shows low loss (1 dB) and high polarization extinction ratio (>20 dB) over a >415 nm bandwidth from 1.26 to 1.675 μm, which is at least fourfold better than what has been demonstrated in all previous works. To the best of our knowledge, such a device is the first all-silicon polarizer that covers O-, E-, S-, C-, L-, and U-bands.

Passive all-optical signal processors that overcome the electronic bottleneck can potentially be the enabling components for the next-generation high-speed and lower power consumption systems. Here, we propose and experimentally demonstrate a CMOS-compatible waveguide and its application to the all-optical analog-to-digital converter (ADC) under the nonlinear spectral splitting and filtering scheme. As the key component of the proposed ADC, a 50 cm long high-index doped silica glass spiral waveguide is composed of a thin silicon-nanocrystal (Si-nc) layer embedded in the core center for enhanced nonlinearity. The device simultaneously possesses low loss (0.16 dB/cm at 1550 nm), large nonlinearity (305 W 1/km at 1550 nm), and negligible nonlinear absorption. A 2-bit ADC basic unit is achieved when pumped by the proposed waveguide structure at the telecom band and without any additional amplification. Simulation results that are consistent with the experimental ones are also demonstrated, which further confirm the feasibility of the proposed scheme for larger quantization resolution. This demonstrated approach enables a fully monolithic solution for all-optical ADC in the future, which can digitize broadband optical signals directly at low power consumption. This has great potential on the applications of high-speed optical communications, networks, and signal processing systems.

We have designed and realized an athermal 4-channel wavelength (de-)multiplexer in silicon nitride (SiN). Minimized thermal sensitivity is achieved in a wide wavelength range by using wide and narrow waveguides with low and different thermal-optic coefficients in the two arms of Mach–Zehnder interferometers (MZIs). The SiN core layer and SiO2 cladding layers are deposited by a low-temperature plasma-enhanced chemical vapor deposition process. The fabricated MZI filter exhibits a thermal sensitivity within ±2.0 pm/°C in a wavelength range of 55 nm to near 1300 nm. Then, an athermal (de-)multiplexer based on cascaded MZIs has been demonstrated with a crosstalk ≤ 22 dB and a thermal sensitivity <4.8 pm/°C for all four channels, reduced by 77% compared to a conventional SiN (de-)multiplexer. Owing to the passive operation and compatibility with the CMOS backend process, our devices have potential applications in 3D integration of photonics and electronics.

We propose and experimentally demonstrate a novel ultracompact dual-mode waveguide crossing based on subwavelength multimode-interference couplers for a densely integrated on-chip mode-division multiplexing system. By engineering the lateral-cladding material index and manipulating phase profiles of light at the nanoscale using an improved inverse design method, a subwavelength structure could theoretically realize the identical beat length for both TE0 and TE1, which can reduce the scale of the device greatly. The fabricated device occupied a footprint of only 4.8 μm×4.8 μm. The measured insertion losses and crosstalks were less than 0.6 dB and 24 dB from 1530 nm to 1590 nm for both TE0 and TE1 modes, respectively. Furthermore, our scheme could also be expanded to design waveguide crossings that support more modes.

In this paper, we theoretically propose and experimentally demonstrate the manipulation of a novel degree of freedom in ring resonators, which is the coupling from the clockwise input to the counterclockwise propagating mode (and vice versa). We name this mechanism backcoupling, in contrast with the normal forward-coupling of a directional coupler. It is well known that internal reflections will cause peak splitting in a ring resonator. Our previous research demonstrated that the peak asymmetry will be strongly influenced by the backcoupling. Thus, it is worth manipulating the backcoupling in order to gain full control of a split resonance for the benefit of various resonance-splitting-based applications. While it is difficult to directly manipulate the backcoupling of a conventional directional coupler, here we design a circuit explicitly for manipulating the backcoupling. It can be potentially developed for applications such as single sideband filter, resonance splitting elimination, Fano resonance, and ultrahigh-Q and finesse.

With the rapidly increasing bandwidth requirements of optical communication networks, compact and low-cost large-scale optical switches become necessary. Silicon photonics is a promising technology due to its small footprint, cost competitiveness, and high bandwidth density. In this paper, we demonstrate a 12×12 silicon wavelength routing switch employing cascaded arrayed waveguide gratings (AWGs) connected by a silicon waveguide interconnection network on a single chip. We optimize the connecting strategy of the crossing structure to reduce the switch’s footprint. We develop an algorithm based on minimum standard deviation to minimize the port-to-port insertion loss (IL) fluctuation of the switch globally. The simulated port-to-port IL fluctuation decreases by about 3 dB compared with that of the conventional one. The average measured port-to-port IL is 13.03 dB, with a standard deviation of 0.78 dB and a fluctuation of 2.39 dB. The device can be used for wide applications in core networks and data centers.

In silicon photonics, the carrier depletion scheme has been the most commonly used mechanism for demonstrating high-speed electro-optic modulation. However, in terms of phase modulation efficiency, carrier-accumulation-based devices potentially offer almost an order of magnitude improvement over those based on carrier depletion. Previously reported accumulation modulator designs only considered vertical metal-oxide-semiconductor (MOS) capacitors, which imposes serious restrictions on the design flexibility and integratability with other photonic components. In this work, for the first time to our knowledge, we report experimental demonstration of an all-silicon accumulation phase modulator based on a lateral MOS capacitor. Using a Mach–Zehnder interferometer modulator with a 500-μm-long phase shifter, we demonstrate high-speed modulation up to 25 Gbit/s with a modulation efficiency (VπLπ) of 1.53 V·cm.

Silicon has been the material of choice of the photonics industry over the last decade due to its easy integration with silicon electronics, high index contrast, small footprint, and low cost, as well as its optical transparency in the near-infrared and parts of mid-infrared (MIR) wavelengths (from 1.1 to 8 μm). While considerations of micro- and nano-fabrication-induced device parameter deviations and a higher-than-desirable propagation loss still serve as a bottleneck in many on-chip data communication applications, applications as sensors do not require similar stringent controls. Photonic devices on chips are increasingly being demonstrated for chemical and biological sensing with performance metrics rivaling benchtop instruments and thus promising the potential of portable, handheld, and wearable monitoring of various chemical and biological analytes. In this paper, we review recent advances in MIR silicon photonics research. We discuss the pros and cons of various platforms, the fabrication procedures for building such platforms, and the benchmarks demonstrated so far, together with their applications. Novel device architectures and improved fabrication techniques have paved a viable way for realizing low-cost, high-density, multi-function integrated devices in the MIR. These advances are expected to benefit several application domains in the years to come, including communication networks, sensing, and nonlinear systems.

In this paper, a substrate removing technique in a silicon Mach–Zehnder modulator (MZM) is proposed and demonstrated to improve modulation bandwidth. Based on the novel and optimized traveling wave electrodes, the electrode transmission loss is reduced, and the electro-optical group index and 50 Ω impedance matching are improved, simultaneously. A 2 mm long substrate removed silicon MZM with the measured and extrapolated 3 dB electro-optical bandwidth of >50 GHz and 60 GHz at the 8 V bias voltage is designed and fabricated. Open optical eye diagrams of up to 90 GBaud/s NRZ and 56 GBaud/s four-level pulse amplitude modulation (PAM-4) are experimentally obtained without additional optical or digital compensations. Based on this silicon MZM, the performance in a short-reach transmission system is further investigated. Single-lane 112 Gb/s and 128 Gb/s transmissions over different distances of 1 km, 2 km, and 10 km are experimentally achieved based on this high-speed silicon MZM.

We report low-noise, high-performance single transverse mode 1.3 μm InAs/GaAs quantum dot lasers monolithically grown on silicon (Si) using molecular beam epitaxy. The fabricated narrow-ridge-waveguide Fabry–Perot (FP) lasers have achieved a room-temperature continuous-wave (CW) threshold current of 12.5 mA and high CW temperature tolerance up to 90°C. An ultra-low relative intensity noise of less than 150 dB/Hz is measured in the 4–16 GHz range. Using this low-noise Si-based laser, we then demonstrate 25.6 Gb/s data transmission over 13.5 km SMF-28. These low-cost FP laser devices are promising candidates to provide cost-effective solutions for use in uncooled Si photonics transmitters in inter/hyper data centers and metropolitan data links.

We experimentally demonstrate extraction of silicon waveguide geometry with subnanometer accuracy using optical measurements. Effective and group indices of silicon-on-insulator (SOI) waveguides are extracted from the optical measurements. An accurate model linking the geometry of an SOI waveguide to its effective and group indices is used to extract the linewidths and thicknesses within respective errors of 0.37 and 0.26 nm on a die fabricated by IMEC multiproject wafer services. A detailed analysis of the setting of the bounds for the effective and group indices is presented to get the right extraction with improved accuracy.

We demonstrate an integrated Si optical single-sideband (OSSB) modulator composed of a parallel dual-ring modulator (PDRM) and a quadrature hybrid coupler (QHC). Both the PDRM and the QHC are carefully designed for 30 GHz opearation, and their operations are verified by measurement. The Si OSSB modulator successfully generates a single sideband with larger than 15 dB suppression of the undesired sideband.

We demonstrate a novel high-accuracy post-fabrication trimming technique to fine-tune the phase of integrated Mach–Zehnder interferometers, enabling permanent correction of typical fabrication-based phase errors. The effective index change of the optical mode is 0.19 in our measurement, which is approximately an order of magnitude improvement compared to previous work with similar excess optical loss. Our measurement results suggest that a phase accuracy of 0.078 rad was achievable with active feedback control.

Mode- and polarization-division multiplexing offer new dimensions to increase the transmission capacity of optical communications. Selective switches are key components in reconfigurable optical network nodes. An on-chip silicon 2×2 mode- and polarization-selective switch that can route four data channels on two modes and two polarizations simultaneously is proposed and experimentally demonstrated for the first time, to the best of our knowledge. The overall insertion losses are lower than 8.6 dB. To reduce the inter-modal crosstalk, polarization beam splitters are added to filter the undesired polarizations or modes. The measured inter-modal and intra-modal crosstalk values are below ?23.2 and ?22.8 dB for all the channels, respectively.

Silicon photonic integrated circuits for telecommunication and data centers have been well studied in the past decade, and now most related efforts have been progressing toward commercialization. Scaling up the silicon-on-insulator (SOI)-based device dimensions in order to extend the operation wavelength to the short mid-infrared (MIR) range (2–4 μm) is attracting research interest, owing to the host of potential applications in lab-on-chip sensors, free space communications, and much more. Other material systems and technology platforms, including silicon-on-silicon nitride, germanium-on-silicon, germanium-on-SOI, germanium-on-silicon nitride, sapphire-on-silicon, SiGe alloy-on-silicon, and aluminum nitride-on-insulator are explored as well in order to realize low-loss waveguide devices for different MIR wavelengths. In this paper, we will comprehensively review silicon photonics for MIR applications, with regard to the state-of-the-art achievements from various device demonstrations in different material platforms by various groups. We will then introduce in detail of our institute’s research and development efforts on the MIR photonic platforms as one case study. Meanwhile, we will discuss the integration schemes along with remaining challenges in devices (e.g., light source) and integration. A few application-oriented examples will be examined to illustrate the issues needing a critical solution toward the final production path (e.g., gas sensors). Finally, we will provide our assessment of the outlook of potential future research topics and engineering challenges along with opportunities.

A theoretical design is presented for a 1×M wavelength-selective switch (WSS) that routes any one of N incoming wavelength signals to any one of M output ports. This planar on-chip device comprises a 1×N demultiplexer, a group of N switching “trees” actuated by electro-optical or thermo-optical means, and an M-fold set of N×1 multiplexers. Trees utilize 1×2 switches. The WSS insertion loss is proportional to [log2 (M+N+1)]. Along with cross talk from trees, cross talk is present at each cross-illuminated waveguide intersection within the WSS, and there are at most N 1 such crossings per path. These loss and cross talk properties will likely place a practical limit of N=M=16 upon the WSS size. By constraining the 1×2 switching energy to ～1 fJ/bit, we find that resonant, narrowband 1×2 switches are required. The 1×2</i

A novel scheme for the design of an ultra-compact and high-performance optical switch is proposed and investigated numerically. Based on a standard silicon (Si) photonic stripe waveguide, a section of hyperbolic metamaterials (HMM) consisting of 20-pair alternating vanadium dioxide (VO2)/Si thin layers is inserted to realize the switching of fundamental TE mode propagation. Finite-element-method simulation results show that, with the help of an HMM with a size of 400 nm×220 nm×200 nm (width×height×length), the ON/OFF switching for fundamental TE mode propagation in an Si waveguide can be characterized by modulation depth (MD) of 5.6 dB and insertion loss (IL) of 1.25 dB. It also allows for a relatively wide operating bandwidth of 215 nm maintaining MD>5 dB and IL<1.25 dB. Furthermore, we discuss that the tungsten-doped VO2 layers could be useful for reducing metal-insulator-transition temperature and thus improving switching performance. In general, our findings may provide some useful ideas for optical switch design and application in an on-chip all-optical communication system with a demanding integration level.

Changes in refractive index and the corresponding changes in the characteristics of an optical waveguide in enabling propagation of light are the basis for many modern silicon photonic devices. Optical properties of these active nanoscale waveguides are sensitive to the little changes in geometry, external injection/biasing, and doping profiles, and can be crucial in design and manufacturing processes. This paper brings the active silicon waveguide for complete characterization of various distinctive guiding parameters, including perturbation in real and imaginary refractive index, mode loss, group velocity dispersion, and bending loss, which can be instrumental in developing optimal design specifications for various application-centric active silicon waveguides.

In this paper, we present an ultra-compact 1D photonic crystal (PhC) Bragg grating design on a thin film lithium niobate slot waveguide (SWG) via 2D- and 3D-FDTD simulations. 2D-FDTD simulations are employed to tune the photonic bandgap (PBG) size, PBG center, cavity resonance wavelength, and the whole size of PhC. 3D-FDTD simulations are carried out to model the real structure by varying different geometrical parameters such as SWG height and PhC size. A moderate resonance quality factor Q of about 300 is achieved with a PhC size of only 0.5 μm×0.7 μm×6 μm. The proposed slot Bragg grating structure is then exploited as an electric field (E-field) sensor. The sensitivity is analyzed by 3D-FDTD simulations with a minimum detectable E-field as small as 23 mV/m. The possible fabrication process of the proposed structure is also discussed. The compact size of the proposed slot Bragg grating structure may have applications in on-chip E-field sensing, optical filtering, etc.

All-optical integrators are key devices for the realization of ultra-fast passive photonic networks, and, despite their broad applicability range (e.g., photonic bit counting, optical memory units, analogue computing, etc.), their realization in an integrated form is still a challenge. In this work, an all-optical integrator based on a silicon photonic phase-shifted Bragg grating is proposed and experimentally demonstrated, which shows a wide operation bandwidth of 750 GHz and integration time window of 9 ps. The integral operation for single pulse, in-phase pulses, and π-shifted pulses with different delays has been successfully achieved.

Reduction of modulator energy consumption to 10 fJ/bit is essential for the sustainable development of communication systems. Lumped modulators might be a viable solution if instructed by a complete theory system. Here, we present a complete analytical electro-optic response theory, energy consumption analysis, and eye diagrams on absolute scales for lumped modulators. Consequently the speed limitation is understood and alleviated by single-drive configuration, and comprehensive knowledge into the energy dependence on structural parameters significantly reduces energy consumption. The results show that silicon modulation energy as low as 80.8 and 21.5 fJ/bit can be achieved at 28 Gbd under 50 and 10 Ω impedance drivers, respectively. A 50 Gbd modulation is also shown to be possible. The analytical models can be extended to lumped modulators on other material platforms and offer a promising solution to the current challenges of modulation energy reduction.

The electrical nonlinearity of silicon modulators based on reversed PN junctions was found to severely limit the linearity of the modulators. This effect, however, was inadvertently neglected in previous studies. Considering the electrical nonlinearity in simulation, a 32.2 dB degradation in the CDR3 (i.e., the suppression ratio between the fundamental signal and intermodulation distortion) of the modulator was observed at a modulation speed of 12 GHz, and the spurious free dynamic range was simultaneously degraded by 17.4 dB. It was also found that the linearity of the silicon modulator could be improved by reducing the series resistance of the PN junction. The frequency dependence of the linearity due to the electrical nonlinearity was also investigated.

We propose and experimentally demonstrate a 2×2 thermo-optic (TO) crossbar switch implemented by dual photonic crystal nanobeam (PCN) cavities within a silicon-on-insulator (SOI) platform. By thermally tuning the refractive index of silicon, the resonance wavelength of the PCN cavities can be red-shifted. With the help of the ultrasmall mode volumes of the PCN cavities, only ～0.16 mW power is needed to change the switching state. With a spectral passband of 0.09 nm at the 1583.75 nm operation wavelength, the insertion loss (IL) and crosstalk (CT) performances were measured as IL(bar)= 0.2 dB, CT(bar)= 15 dB, IL(cross)= 1.5 dB, and CT(cross)= 15 dB. Furthermore, the thermal tuning efficiency of the fabricated device is as high as 1.23 nm/mW.

In this paper, we have proposed a hybrid optical wavelength demultiplexer and power combiner for a hybrid time- and wavelength-division multiplexing (TWDM) passive optical network (PON), i.e., a single passive optical device that functions as a 1×N wavelength demultiplexer for distributing the downstream signal in multiple wavelengths from the optical line terminal (OLT) to the N optical network units (ONUs), and simultaneously as an N×1 power combiner for collecting the upstream signal in the same wavelength from the N ONUs to the OLT. Through a design example of a 32 channel hybrid optical wavelength demultiplexer and power combiner on the silicon-on-insulator platform, our numerical simulation result shows that the insertion loss and adjacent channel crosstalk of the downstream wavelength demultiplexer are as low as 4.6 and 16.3 dB, respectively, while the insertion loss and channel non-uniformity of the upstream power combiner can reach 3.5 and 2.1 dB, respectively. The proposed structure can readily be extended to other material platforms such as the silica-based planar lightwave circuit. Its fabrication process is fully compatible with standard clean-room technologies such as photo-lithography and etching, without any complicated and/or costly approach involved.

We propose a novel silicon optical phase shifter structure based on heterogeneous strip-loaded waveguides on a photonic silicon on insulator (SOI) platform. The features of an etchless SOI layer and loaded strip would enhance the performance and uniformity of silicon optical modulators on a large-scale wafer. We implemented the phase shifter by loading an amorphous silicon strip onto an SOI layer with a vertical PN diode structure. Compared to the conventional lateral PN phase shifter based on half-etched rib waveguides, this phase shifter shows a >1.5 times enhancement of modulation efficiency and provides >20 GHz high-speed operation.

We demonstrate binary phase shift keying (BPSK) modulation using a silicon Mach–Zehnder modulator with a π-phase-shift voltage (Vπ) of 4.5 V. The single-drive push–pull traveling wave electrode has been optimized using numerical simulations with a 3 dB electro-optic bandwidth of 35 GHz. The 32 Gb/s BPSK constellation diagram is measured with an error vector magnitude of 18.9%.

We propose and experimentally demonstrate compact on-chip 1 × 2 wavelength selective switches (WSSs) based on silicon microring resonators (MRRs) with nested pairs of subrings (NPSs). Owing to the resonance splitting induced by the inner NPSs, the proposed devices are capable of performing selective channel routing at certain resonance wavelengths of the outer MRRs. System demonstration of dynamic channel routing using fabricated devices with one and two NPSs is carried out for 10 Gb∕s non-return-to-zero signal. The experimental results verify the effectiveness of the fabricated devices as compact on-chip WSSs.

A novel athermal scheme utilizing resonance splitting of a dual-ring structure is proposed. Detailed design and simulation are presented, and a proof of concept structure is optimized to demonstrate an athermal resonator with resonance wavelength variation lower than 5 pm∕K within 30 K temperature range.

The optical properties of germanium can be tailored by combining strain engineering and n-type doping. In this paper, we review the recent progress that has been reported in the study of germanium light emitters for silicon photonics. We discuss the different approaches that were implemented for strain engineering and the issues associated with n-type doping. We show that compact germanium emitters can be obtained by processing germanium into tensile-strained microdisks.

Germanium tin (GeSn) is a group IV semiconductor with a direct band-to-band transition below 0.8 eV. Nonequilibrium GeSn alloys up to 20% Sn content were realized with low temperature (160°C) molecular beam epitaxy. Photodetectors and light emitting diodes (LEDs) were realized from in situ doped pin junctions in GeSn on Ge virtual substrates. The detection wavelength for infrared radiation was extended to 2 μm with clear potential for further extension into the mid-infrared. GeSn LEDs with Sn content of up to 4% exhibit light emission from the direct band transition, although GeSn with low Sn content is an indirect semiconductor. The photon emission energies span the region between 0.81 and 0.65 eV. Optical characterization techniques such as ellipsometry, in situ reflectometry, and Raman spectroscopy were used to monitor the Sn incorporation in GeSn epitaxy.