Search

Silicon-based optical switches are integral to on-chip optical interconnects, and mode-division multiplexing (MDM) technology has enabled modes to function as carriers in routing, further boosting optical switches' link capacity. However, traditional multimode optical switches, which typically use Mach–Zehnder interferometer (MZI) structures and mode (de)multiplexers, are complex and occupy significant physical space. In this paper, we propose and experimentally demonstrate a novel demultiplexing-free dual-mode 3 × 3 thermal-optical switch based on micro-rings (MRs) and mode exchangers (MEs). All MRs are designed to handle TE1 mode, while the ME converts TE0 mode to TE1 mode, enabling separate routing of both modes. Bezier curves are employed to optimize not only the ME, but also the dual-mode 45° and 90° waveguide bends, which facilitate the flexible and compact layout design. Moreover, our structure can support multiple wavelength channels and spacings by adding pairs of MRs, exhibiting strong WDM compatibility. The switch has an ultra-compact footprint of 0.87 × 0.52 mm2. Under both "all-bar" and "all-cross" configurations, its insertion losses (ILs) remain below 8.7 dB at 1,551 nm, with optical signal-to-noise ratios (OSNRs) exceeding 13.0 dB. Also, 32 Gbps data transmission experiments validate the switch's high-speed transmission capability. [ABSTRACT FROM AUTHOR]

In recent datacenters, considerable data communication has been performed between servers and storages; therefore, an optical interconnection with a wider band is required. Silicon photonics has attracted attention as a technology that places optical engines closer to switch application‐specific integrated circuits and realizes broadband optical interconnection. To utilize silicon photonics, it is necessary to overcome high‐precision alignment between silicon waveguides and single‐mode fibers (SMFs) having vastly different mode field diameters. Optical self‐coupling using light‐induced self‐written (LISW) optical waveguide technology is a potential solution. Previously, near‐infrared (NIR) LISW optical waveguides are fabricated using an SMF with an approximate 10 μm core through irradiation with 10 μW NIR light (1310 and 1550 nm). However, the cladding is liquid, which motivates to fabricate all‐solid LISW optical waveguides. In this study, all‐solid NIR LISW optical waveguides are produced using a core‐selective polymerization method, and single‐mode LISW optical self‐coupling between two SMFs is realized. To enhance the performance of the waveguides, additional resins are incorporated into the high‐refractive‐index base monomer during the cladding formation process. Through the experimentation, the potential for achieving even greater loss reduction by optimizing the hybridization between the base monomer and the supplementary resins is demonstrated. [ABSTRACT FROM AUTHOR]

On‐chip integration of two‐dimensional (2D) materials holds immense potential for novel optoelectronic devices across diverse photonic platforms. In particular, indium selenide (InSe) is a very promising 2D material due to its ultra‐high carrier mobility and outstanding photoresponsivity. Herein, a high‐speed photodetector based on a multilayer 90 nm thick InSe integrated on a silicon nitride (SiN) waveguide is reported. The device exhibits a remarkable low dark current density of ≈40 nA μm−2, a photoresponsivity of 0.38 A W−1, and an external quantum efficiency of ≈48.4%. Tested under ambient conditions at near‐infrared 976 nm wavelength, it exhibits an absorption coefficient of 0.11 dB μm−1. Additionally, the photodetector demonstrates a 3‐dB radiofrequency bandwidth of 85 MHz and an open‐eye diagram at data transmission of 1 Gbit s−1. Based on these exceptional optoelectronic advantages, integrated multilayer InSe enables the realization of active photonic devices for a range of applications, such as short‐reach optical interconnects, LiDAR imaging, and biosensing. [ABSTRACT FROM AUTHOR]

In this regard, we directed a theoretical study with numerical simulations. This study allowed us to illustrate how a photonic crystal fiber (PCF) structure could divide an optical signal. One of the most fundamental components used to construct photonic integrated circuits (PIC) is the splitter, which is using light coupling between the cores as a control until the output ports by using pure silica to replace some air-hole zones along the PCF axis and split the single signal on two ports with almost equal intensity in each port. Optical interconnects are one of the most basic components of integrated optics, and splitters for photonic power are a key element of a connected family. With the least amount of loss, a competent photonic splitter can guide light input of a certain wavelength to several ports at various intensities. [ABSTRACT FROM AUTHOR]

To increase the optical interconnect transmission capacity, different multiplexing technologies, including wavelength division multiplexing (WDM), polarization division multiplexing (PolDM) and mode division multiplexing (MDM), can be utilized. Among them, MDM is a promising technique in silicon photonics (SiPh) integrated optical interconnects since higher order modes can be easily generated and preserved in SiPh waveguides. In this work, we propose and demonstrate the designs of SiPh-based bulk zigzag and sinusoidal structured MDM mode conversion devices using genetic algorithm (GA) optimization. A traditional periodic zigzag structured mode converter design has many sharp zigzag angles in the periodic structure, which are very sensitive to the fabrication error. Here, first of all, we propose and demonstrate a bulk zigzag structure to achieve MDM mode conversion. The proposed bulk zigzag structure can reduce the zigzag angle error as a large number of zigzag angles in the periodic structure are eliminated. Moreover, we further improve our device by proposing a bulk sinusoidal structure to further eliminate the zigzag angle. Results show that both the proposed bulk zigzag and sinusoidal MDM mode converters can still maintain high transmissions of >86%, while the mode conversion lengths of both devices can be significantly reduced by >60% in the C-band wavelength window. In addition, as there are many degrees of freedom (DOFs) during the design of the SiPh mode converter, including the waveguide width, length, period, zigzag angle, etch depth, duty cycle, etc., the GA optimization algorithm is employed. Here, detailed implementation of the GA optimization is discussed. [ABSTRACT FROM AUTHOR]

Flexible integrated photonics is an essential technology for emerging applications, including flexible optical interconnects, optogenetic stimulation, and implantable conformal sensing. Here, a novel and universal route for fabricating flexible photonic components with high‐refractive‐index contrast is reported. Central to such a unique method is the utilization of germanium oxide (GeO) as the sacrificial layer for releasing nanostructures from rigid substrates to flexible substrates. Various high‐quality inorganic optical materials can be grown directly on GeO by different thin‐film deposition methods due to its resistance to both high temperature and high‐power oxygen plasmas. In addition to the absence of restrictions on the material choices and integration processes for flexible photonic structures, the approach uses water as the etchant to remove the sacrificial layer, which has minimal impact on the optical performance of the photonic structures. Using this approach, a strain‐insensitive/sensitive microring resonator based on plasma‐enhanced chemical vapor deposited silicon nitride and reactive sputtered titanium oxide, respectively, is demonstrated, establishing the strategy as a facile and universal route for the fabrication of high‐index‐contrast flexible integrated photonic devices with various functionalities. [ABSTRACT FROM AUTHOR]

Thin‐film lithium‐niobate photonics has attracted intensive attention due to its superior material properties. In particular, multimode lithium‐niobate photonics is becoming attractive for the applications of high‐capacity optical interconnects and high‐efficiency nonlinear optics. However, multimode manipulation in lithium‐niobate‐on‐insulator (LNOI) waveguides is not easy because there might occur some mode hybridness due to the vertical asymmetry as well as the material anisotropy. In this paper, high‐performance mode‐multiplexing devices with anisotropic LNOI waveguides are proposed, consisting of a multi‐channel mode (de)multiplexer and a sharp multimode waveguide bend (MWB). The mode (de)multiplexer is designed with Z‐propagation LNOI waveguides to avoid the mode hybridness. Meanwhile, the sharp MWB is designed to connect with a Y‐propagation waveguide, enabling access to the highest electro‐optic coefficient and second‐order nonlinear coefficient. The fabricated mode (de)multiplexer has a low excess loss less than 0.2 dB as well as low crosstalk less than ‐19 dB in the wavelength range of 1520–1600 nm. These high‐performance multimode photonic devices are useful for developing multimode LN photonics. [ABSTRACT FROM AUTHOR]

Due to the popularity of different high bandwidth applications, it is becoming increasingly difficult to satisfy the huge data capacity requirements, since the traditional electrical interconnects suffer significantly from limited bandwidth and huge power consumption. Silicon photonics (SiPh) is one of the important technologies for increasing interconnect capacity and decreasing power consumption. Mode-division multiplexing (MDM) allows signals to be transmitted simultaneously, at different modes, in a single waveguide. Wavelength-division multiplexing (WDM), non-orthogonal multiple access (NOMA) and orthogonal-frequency-division multiplexing (OFDM) can also be utilized to further increase the optical interconnect capacity. In SiPh integrated circuits, waveguide bends are usually inevitable. However, for an MDM system with a multimode bus waveguide, the modal fields will become asymmetric when the waveguide bend is sharp. This will introduce inter-mode coupling and inter-mode crosstalk. One simple approach to achieve sharp bends in multimode bus waveguide is to use a Euler curve. Although it has been reported in the literature that sharp bends based on a Euler curve allow high performance and low inter-mode crosstalk multimode transmissions, we discover, by simulation and experiment, that the transmission performance between two Euler bends is length dependent, particularly when the bends are sharp. We investigate the length dependency of the straight multimode bus waveguide between two Euler bends. High transmission performance can be achieved by a proper design of the waveguide length, width, and bend radius. By using the optimized MDM bus waveguide length with sharp Euler bends, proof-of-concept NOMA-OFDM experimental transmissions, supporting two MDM modes and two NOMA users, are performed. [ABSTRACT FROM AUTHOR]

NTT Network Innovation Center is developing technologies and systems for the practical implementation of an optical transmission network, the key element for implementing the All-Photonics Network (APN), which in turn will support IOWN (the Innovative Optical and Wireless Network). With our nextgeneration optical transmission network, which is an advance release of the APN, we are working to increase speed and capacity to handle growing communication traffic as well as provide open optical interfaces for connecting various systems and devices without photoelectric conversion to enhance the added value provided by optical networks and develop operations and maintenance technologies for these networks. [ABSTRACT FROM AUTHOR]

The use of optical interconnects has burgeoned as a promising technology that can address the limits of data transfer for future high-performance silicon chips. Recent pushes to enhance optical communication have focused on developing wavelength-division multiplexing technology, and new dimensions of data transfer will be paramount to fulfill the ever-growing need for speed. Here we demonstrate an integrated multi-dimensional communication scheme that combines wavelength- and mode- multiplexing on a silicon photonic circuit. Using foundry-compatible photonic inverse design and spectrally flattened microcombs, we demonstrate a 1.12-Tb/s natively error-free data transmission throughout a silicon nanophotonic waveguide. Furthermore, we implement inverse-designed surface-normal couplers to enable multimode optical transmission between separate silicon chips throughout a multimode-matched fibre. All the inverse-designed devices comply with the process design rules for standard silicon photonic foundries. Our approach is inherently scalable to a multiplicative enhancement over the state of the art silicon photonic transmitters. The authors demonstrate a multi-dimensional communication scheme that combines wavelength- and mode- multiplexing on photonic integrated circuits using foundry-compatible photonic inverse design and spectrally flattened microcombs [ABSTRACT FROM AUTHOR]

A photodetector (PD) converts optical signals into electrical ones and is widely used in optical interconnect. High‐speed PDs are in high demand as they are necessary to meet requirements of large‐capacity optical interconnect. Many high‐performance PDs with various absorption materials and structures are demonstrated on silicon photonics platform, including germanium (Ge) PDs, germanium tin (GeSn) PDs, heterogeneous integrated III–V PDs, all silicon (Si) PDs, 2D material PDs, etc. These kinds of PDs continue to set new records of speed and open an era of ultrahigh‐speed optical interconnect. A comprehensive summary of the state‐of‐the‐art high‐speed PDs on silicon photonics platform is necessary and meaningful. In this review, the basic metrics and key process technologies for the PDs are introduced, and various types of high‐speed PDs based on silicon photonics platform are reviewed and discussed. Furthermore, the summary and perspectives are provided. It is hoped that this review can provide readers more insights into recent advances in high‐speed PDs on silicon photonics platform and contribute to the further development. [ABSTRACT FROM AUTHOR]

With the rapid development of 5 G communications, artificial intelligence, the Internet of Things and other fields, the increasing demand for data transmission in communication networks has resulted in optical interconnection gradually replacing traditional electrical interconnection in communication. As one of the core devices, the research and development of silicon-based photodetectors has garnered significant attention. Particularly, the advancement of germanium photodetectors (Ge PDs) holds crucial research significance for both academia and industry. Recently, a variety of high-performance photodetectors based on various photoelectric structures, emerging technologies and physical effects have been demonstrated on silicon photonic platforms. In this review, the research progress of Ge PDs is summarized, and the key technologies and processes in the latest development are analyzed. Firstly, the basic principles and performance of photodetectors are briefly introduced. Secondly, the key technology and basic structure of silicon and Ge PD are discussed. The development history of Ge PDs is then reviewed, and new mechanism design and performance improvement technologies are discussed. Finally, the current research frontiers and hotspots of high-performance Ge PDs are summarized, and the future development trends are discussed. It is hoped that this review will be helpful for readers to gain more insights into the latest advancements in high-performance photodetectors in silicon photonics and contribute to further development. [Display omitted] [ABSTRACT FROM AUTHOR]

• Self-written waveguide method enables the self-coupling of optical components. • Cladding solidification process of a self-written waveguide was demonstrated. • The proposed technique demonstrated a low loss all-solid self-written waveguide. • The all-solid self-coupling between two single-mode fibers was successfully realized. • The all-solid self-coupling between silicon photonics devices was demonstrated. The light-induced self-written (LISW) optical waveguide technique has emerged as a promising solution for achieving passive alignment between telecommunication and silicon photonic devices. In this study, we proposed and demonstrated an advanced selective polymerization process for the fabrication of all-solid LISW optical waveguides. Cladding solidification was accomplished through the propagation of ultraviolet or violet light via a single-mode fiber (SMF). The self-coupling of all-solid LISW optical waveguides between two SMFs was successfully realized through the application of this technique. Notably, a low insertion loss, ranging from 0.3 to 0.7 dB at a wavelength of 1550 nm, was measured for the all-solid LISW optical waveguide. Furthermore, we demonstrated the applicability of this refined process in achieving self-coupling for silicon photonics. [ABSTRACT FROM AUTHOR]

Photonic integrated circuits are an ideal platform for chip-scale computation and communication. A complex photonic network usually requires components to direct light from one waveguide to another that is parallel to the former with an arbitrary separation d. Traditionally, this is achieved with S-shaped waveguide bends, which either occupy a large footprint or suffer from large insertion loss. To reduce the footprint and maintain ultralow insertion loss, here we propose and experimentally demonstrate inverse-designed photonic jumpers on a silicon-on-insulator platform. These photonic jumpers have ultracompact sizes by achieving the expected function within a footprint of d × 2.5 μm, while the simulated and measured insertion loss for the fundamental transverse-electric mode maintains below 0.14 and 0.30 dB, respectively, in the wavelength range of 1.50–1.60 μm. Therefore, these photonic jumpers can be used in large-scale photonic integrated circuits as a standard module to significantly enhance the integration density. [ABSTRACT FROM AUTHOR]

On-chip photonics is a disruptive technology, and such NoCs are superior to traditional electrical NoCs in terms of latency, power, and bandwidth. Hence, researchers have proposed a wide variety of optical networks for multicore processors. The high bandwidth and low latency features of photonic NoCs have led to the overall improvement in the system performance. However, there are very few proposals that discuss the usage of optical interconnects in Graphics Processor Units (GPUs). GPUs can also substantially gain from such novel technologies, because they need to provide significant computational throughput without further stressing their power budgets. The main shortcoming of optical networks is their high static power usage, because the lasers are turned on all the time by default, even when there is no traffic inside the chip, and thus sophisticated laser modulation schemes are required. Such modulation schemes base their decisions on an accurate prediction of network traffic in the future. In this article, we propose an energy-efficient and scalable optical interconnect formodern GPUs called GPUOPT that smartly creates an overlay network by dividing the symmetric multiprocessors (SMs) into clusters. It furthermore has separate sub-networks for coherence and non-coherence traffic. To further increase the throughput, we connect the of-chip memory with optical links as well. Subsequently, we show that traditional laser modulation schemes (for reducing static power consumption) that were designed for multicore processors are not that effective for GPUs. Hence, there was a need to create a bespoke scheme for predicting the laser power usage in GPUs. Using this set of techniques, we were able to improve the performance of a modern GPU by 45% as compared to a state-of-the-art electrical NoC. Moreover, as compared to competing optical NoCs for GPUs, our scheme reduces the laser power consumption by 67%, resulting in a net 65% reduction in ED2 for a suite of Rodinia benchmarks. [ABSTRACT FROM AUTHOR]

While many new applications prefer manycore processor with a large number of cores, the exploding communications among multiple cores, caches, and off-chip memories is posing a fundamental challenge on manycore designs. Silicon photonics-based interconnection network promises high bandwidth, low latency, and high energy efficiency, and can potentially meet the communication requirements of manycore processors. In this paper, we propose CAMON, a small low-cost silicon photonic chiplet integrated into the manycore processor package. CAMON chiplet can effectively alleviate the communication bottlenecks of manycore processors and improve the energy efficiency of data movement, especially for large-scale systems. We develop a distributed arbitration system, a low-power low-latency optical interface, and an off-chip laser preactivation mechanism for CAMON. The experimental results show that compared with the electrical network, CAMON can improve the full-system performance per energy by $4.6\times $ , speedup the manycore processor by $2.7\times $ , and save the area of the processor die by 3%, in a 512-core system. [ABSTRACT FROM AUTHOR]

Connections between standard single-mode fibers and waveguides in photonic integrated circuits tend to have relatively high coupling losses due to a difference in mode size and mode profile between both light guiding media. Edge coupling strategies involving specialty fibers are frequently used to obtain the best performance in terms of coupling efficiency, bandwidth and polarization independence. We propose the fabrication of free-standing down-taper structures on top of cleaved fiber facets by two-photon direct laser writing and show their performance for silicon, silicon nitride and indium phosphide generic chip platforms. We present a comprehensive analysis of the design of such taper structures that show unprecedented flexibility. We demonstrate their fabrication and fully characterize them in terms of output modal fields and coupling efficiencies. Furthermore, we experimentally compare the fabricated down-tapers with commercially available lensed fibers and demonstrate equal or better coupling efficiency for four out of the five investigated photonic integrated circuit platforms, with a measured improvement in coupling efficiency up to 1.43 dB. [ABSTRACT FROM AUTHOR]

In this paper, we combine inverse design concept and direct binary search algorithm to demonstrate three ultra-compact high efficiency and low crosstalk on-chip integrated optical interconnection basic devices in the entire wavelength range of 1,400–1600 nm based on silicon-on-insulator platform. A 90-degree waveguide bend with a footprint of only 2.4 × 2.4 μm2 is designed, whose transmission efficiency up to 0.18 dB. A waveguide crossing with a footprint of only 2.4 × 2.4 μm2 is designed, which can provide insertion loss of less than 0.5 dB and crosstalk (CL) of lower than − 19 dB. A same direction waveguide crossing with footprint of only 2.4 × 3.6 μm2 is designed, which can provide the insertion loss of less than 0.56 dB and the crosstalk of lower than − 21 dB. Then, we use them to form several ultra-compact optical interconnect basic structures and performed the simulation calculation. They overall achieve high performance. This will significantly improve the integration density. [ABSTRACT FROM AUTHOR]

Silicon-germanium (Si-Ge) avalanche photodiodes (APDs) have large gain bandwidth product (GBP) and low excess noise due to the low impact ionization coefficient ratio of silicon. Optical receivers using APDs are able to achieve high-speed and energy efficient optical transceiver systems. We demonstrate a waveguide Si-Ge APD with low breakdown voltage of −10 V, achieving 60 Gb/s PAM4 successfully. A compact APD circuit model was constructed to allow photonic devices and transceiver circuitry co-design. The APD receiver has achieved −16 dBm sensitivity at 50 Gb/s PAM4 with a bit error rate (BER) of 2.4 $\times \,10^{-4}$. The sensitivity of APD receivers changes with the multiplication gain. In our analysis, compared to a PIN PD receiver the APD receiver operating at optimum gain can obtain $\sim$ 8 dB more sensitivity for NRZ signaling at $<$ 50 Gb/s and 3–4 dB more sensitivity for PAM4 signaling at 50–100 Gb/s. Also, the APD receiver operating at optimum gain can reduce power consumption by $\sim\!\text{10}\%$ at PAM4 data rates of 50 Gb/s and $\sim\!\text{15}\%$ at 100 Gb/s in a silicon carrier-depletion microring modulator based WDM photonic link. [ABSTRACT FROM AUTHOR]

The three-dimensional Network-on-Chips (3D NoCs) has become a mature multi-core interconnection architecture in recent years. However, the traditional electrical lines have very limited bandwidth and high energy consumption, making the photonic interconnection promising for future 3D Optical NoCs (ONoCs). Since existing solutions cannot well guarantee the fault-tolerant ability of 3D ONoCs, in this paper, we propose a reliable optical router (OR) structure which sacrifices less redundancy to obtain more restore paths. Moreover, by using our fault-tolerant routing algorithm, the restore path can be found inside the disabled OR under the deadlock-free condition, i.e., fault-node reuse. Experimental results show that the proposed approach outperforms the previous related works by maximum 81.1 percent and 33.0 percent on average for throughput performance under different synthetic and real traffic patterns. It can improve the system average optical signal to noise ratio (OSNR) performance by maximum 26.92 percent and 12.57 percent on average, and it can improve the average energy consumption performance by 0.3 percent to 15.2 percent under different topology types/sizes, failure rates, OR structures, and payload packet sizes. [ABSTRACT FROM AUTHOR]

We present the first experimental demonstration at the optical telecommunication band of near-isotropic and transverse-electric (TE) photonic bandgap (PBG) in a hyperuniform disordered photonic structured (HUDPS) wall network. Next, we propose and demonstrate state-of-the-art performance of a hyperuniform disordered silicon photonic (HUDSiP) polarizer. The TE mode is blocked by the combined effects of reflection due to a wide PBG in the optimal wall network structure, scattering, and radiative bandedge resonances. The transverse-magnetic (TM) mode is allowed to pass through with a small insertion loss (IL). A 12.9-μm-long HUDSiP polarizer is experimentally demonstrated with an ultrabroad bandwidth of 240 nm (1440–1680 nm) when the extinction ratio (ER) ≥ 30 dB and IL ≤ 3 dB for the TM mode. The measured spectrally averaged ER and IL of the TM mode are 41.4 dB and 0.8 dB (0.4 dB at 1550 nm), respectively. A 38.7-μm-long HUDSiP polarizer shows an average IL for the TE mode of 54.9 dB and ER of 53.0 dB. Tests of 40-Gbps on-off-keying data transmission through the HUDSiP polarizers show clear eye diagrams. The HUDSiP polarizers are promising in polarization filtering in the large-scale polarization-sensitive/diverse silicon photonic circuits for high-speed optical interconnects. [ABSTRACT FROM AUTHOR]

We report on the performance of the quantum confined Stark effect (QCSE) in ultra-thin (~350 nm) Ge/SiGe quantum well stacks grown on Si. We demonstrate an absorption contrast $\Delta \alpha /\alpha $ of 2.1 at 1 Vpp swing in QCSE stacks grown on ultra-thin (100 nm) strain relaxed GeSi buffer layers on 300 mm Si wafers. Such ultra-thin QCSE stacks will enable future integration of highly efficient QCSE electro-absorption modulators with low optical coupling loss to passive Si waveguides in a sub-micron silicon photonics platform. [ABSTRACT FROM AUTHOR]

Parallel optical interconnects at the extreme scale hold the key to resolve the grand challenge of moving enormous amount of data between on-chip cores and within multi-chip modules. Silicon photonic modulators, as one of the most pivotal devices conducting electronic to photonic signal conversion, must excel in energy efficiency and bandwidth density in order to meet the stringent requirement of extreme scale photonic interconnects. In this manuscript, the energy efficiency and bandwidth limit of ultra-compact resonator-based silicon photonic modulators are analyzed from three fundamental perspectives: free carrier dispersion strength of the active materials, Purcell factors of the resonators, and electrical configuration of the capacitors. Our simulation results reveal convincing advantages of photonic crystal nanocavity over micro-ring and micro-disk resonators in terms of energy efficiency and device footprint. While for the electro-optic modulation region, metal-oxide-semiconductor (MOS) capacitors truly outperform reversed PN junctions due to the large capacitance density. However, all resonator-based silicon photonic modulators suffer the intrinsic trade-off between energy efficiency and resistance-capacitance-delay limited bandwidth. The general model developed herein lays the theoretical foundation and identify possible routes to achieve atto-joule per bit energy efficiency and how to approach the bandwidth limit of silicon photonic modulators. [ABSTRACT FROM AUTHOR]

We describe and characterize a multi-micron silicon photonics platform that was designed to combine performance, power efficiency, manufacturability, and versatility for integrated photonic applications ranging from data communications to sensors. We outline the attributes needed for broad applicability, high-volume manufacturing, and large-scale deployment of silicon photonics, and describe how the platform is favorable with respect to these attributes. We present demonstrations of key technologies needed for the communications and sensing applications, including low-loss fiber attach, compact low-loss filters, efficient hybrid wavelength division multiplexed lasers, and high-speed electro-absorption modulators and integrated photodetectors. [ABSTRACT FROM AUTHOR]

A novel integrated scalable multimode switch (SMS) is experimentally demonstrated using tapered multimode-interference-based couplers and Ti/W metal heater phase shifters for mode-division-multiplexing (MDM) silicon photonics switching. The SMS allows path-reconfigurable switching of the first two (TE0 and TE1) or the first three (TE0, TE1, and TE2) transverse electric (TE) modes using the same device achieving footprint efficiency for higher bandwidth density. A proof-of-concept realization of the two-mode switch demonstrates the (de)multiplexing and switching of broadband optical signals over the TE0 and TE1 modes exhibiting $-$ 6.5 dB insertion loss (IL) in the bar state and $-$ 7.3 dB IL in the cross state at 1550 nm with less than $-$ 14 dB crosstalk. Simultaneous switching of two parallel TE modes (TE0+TE1) exhibits less than $-$ 7.0 dB IL and $-$ 11.9 dB crosstalk at 1550 nm. An aggregated bandwidth of 2 × 10 Gb/s is experimentally achieved while switching between two non-return-to-zero PRBS31 data signals with <9.8 $\mu$ s switching time and >17.7 dB switching extinction ratio (ER) for individual-mode transmission, and <7.6 $\mu$ s switching time and >12.0 dB switching ER for dual-mode transmission. The SMS is scalable to switch higher order TE modes with lower energy consumption (up to 63% less) than the single-mode switches indicating its potential application in energy-efficient MDM photonic networks. [ABSTRACT FROM AUTHOR]

With next-generation optical interconnects for data centers aiming for 0.8 Tb/s or 1.6 Tb/s, 100 Gbaud capable transmitters from a single-laser source will become indispensable. However, these lane rates would require bandwidths of 70 GHz or more, doubling the bandwidth requirements of the electrical and optical components with respect to the fastest current generation of optical interconnects running at 53 Gbaud pulse-amplitude modulation (PAM-4). In this paper, we propose an integrated 4:1 optical serializer topology to achieve 104 Gbaud On-Off Keying (OOK) and PAM-4 transmission using only quarter rate components at the transmitter. We show 104 (208) Gbit/s OOK (PAM4) transmission using four GeSi electro-absorption modulators (EAMs) over 1 km of single-mode fiber (SMF). For 104 Gbaud OOK, clearly open eyes are obtained, while for PAM-4 the performance is limited by the nonlinear E/O-transfer function of the EAM. However, adding pre-emphasis in the electrical driver or replacing the single EAM with our previously demonstrated optical DAC topology—consisting of two EAMs in parallel with a 90° phase difference between each—could substantially improve these results. Additionally, we discuss the possibility of a four channel transmitter (4 × 208 Gb/s) from a single mode locked laser, amounting to a 832 Gb/s rate based on the current demonstrator. [ABSTRACT FROM AUTHOR]

We report 50- and 60-Gb/s hybrid-integrated optical links in the O-band with CMOS photonic components driven by silicon–germanium ICs using optical-domain feed-forward equalization. The link includes a segmented-electrode Mach–Zehnder modulator and a Ge photodetector that provide high bandwidths and the potential for tight integration with control and monitoring electronics. The combination of these features enables the demonstration of a 60-Gb/s AC-coupled link and a 50-Gb/s DC-coupled link with bit error rates below 10–12 without the use of forward-error correction codes. This result provides a promising path toward extending speeds in latency sensitive optical links. [ABSTRACT FROM AUTHOR]

Efficient light energy transfer between optical waveguides has been a critical issue in various areas of photonics and optoelectronics. Especially, the light coupling between optical fibers and integrated waveguide structures provides essential input-output interfaces for photonic integrated circuits (PICs) and plays a crucial role in reliable optical signal transport for a number of applications, such as optical interconnects, optical switching, and integrated quantum optics. Significant efforts have been made to improve light coupling properties, including coupling efficiency, bandwidth, polarization dependence, alignment tolerance, as well as packing density. In this review article, we survey three major light coupling methods between optical fibers and integrated waveguides: end-fire coupling, diffraction grating-based coupling, and adiabatic coupling. Although these waveguide coupling methods are different in terms of their operating principles and physical implementations, they have gradually adopted various nanophotonic structures and techniques to improve the light coupling properties as our understanding to the behavior of light and nano-fabrication technology advances. We compare the pros and cons of each light coupling method and provide an overview of the recent developments in waveguide coupling between optical fibers and integrated photonic circuits. [ABSTRACT FROM AUTHOR]

Graphical abstract Abstract The arrival of the big data era has driven the rapid development of high-speed optical signaling and processing, ranging from long-haul optical communication links to short-reach data centers and high-performance computing, and even micro-/nano-scale inter-chip and intra-chip optical interconnects. On-chip photonic signaling is essential for optical data transmission, especially for chip-scale optical interconnects, while on-chip photonic processing is a critical technology for optical data manipulation or processing, especially at the network nodes to facilitate ultracompact data management with low power consumption. In this paper, we review recent research progress in on-chip photonic signaling and processing on silicon photonics platforms. Firstly, basic key devices (lasers, modulators, detectors) are introduced. Secondly, for on-chip photonic signaling, we present recent works on on-chip data transmission of advanced multi-level modulation signals using various silicon photonic integrated devices (microring, slot waveguide, hybrid plasmonic waveguide, subwavelength grating slot waveguide). Thirdly, for on-chip photonic processing, we summarize recent works on on-chip data processing of advanced multi-level modulation signals exploiting linear and nonlinear effects in different kinds of silicon photonic integrated devices (strip waveguide, directional coupler, 2D grating coupler, microring, silicon-organic hybrid slot waveguide). Various photonic processing functions are demonstrated, such as photonic switch, filtering, polarization/wavelength/mode (de)multiplexing, wavelength conversion, signal regeneration, optical logic and computing. Additionally, we also introduce extended silicon+ photonics and show recent works on on-chip graphene-silicon photonic signal processing. The advances in on-chip silicon photonic signaling and processing with favorable performance pave the way to integrate complete optical communication systems on a monolithic chip and integrate silicon photonics and silicon nanoelectronics on a chip. It is believed that silicon photonics will enable more and more emerging advanced applications even beyond silicon photonic signaling and processing. [ABSTRACT FROM AUTHOR]

Highly P doped Ge layers are proposed as a material to realize an on-chip laser for optical interconnect applications. In this work, we demonstrate that these donor impurities have a dramatic impact on the excess carrier lifetime and introduce detrimental many-body effects such as linewidth broadening in the gain medium. Linewidth broadening Γopt = 10 meV for undoped Ge and Γopt ≥ 45 meV for Ge with a doping level up to 5.4 × 1019 cm−3 were extracted using photoluminescence spectroscopy and pump-probe spectroscopy. In addition, we observed that the excess carrier lifetime (τc) drops by more than an order of magnitude from 3 ns in undoped Ge to <0.3 ns in doped Ge. [ABSTRACT FROM AUTHOR]

A full-rate parallel pseudo random bit sequence (PRBS)-9 is systematically designed in the IBM 130-nm six-level metalization SiGe BiCMOS process with 2.5/3.3-V dual power supplies for the application of testing high-speed interconnects. The PRBS length and the phase difference between channels are studied using the methods of autocorrelation and cross correlation, respectively. The parallel PRBS streams can be used as stimuli for a serializer, and the autocorrelation method can be applied as a PRBS checker. It can also be used as stimuli for a four-level pulse amplitude modulation-4 signaling format transmitter with the least correlation between the LSB and the MSB by choosing the appropriate channel combinations. Two prototype boards are designed to test the PRBS along with an on-chip type-II third-order charge-pump phase-locked loop. Signal integrity issues are investigated. Finally, the PRBS has been demonstrated at board level with nonreturn-to-zero format operating at more than 10 Gb/s. [ABSTRACT FROM AUTHOR]

600–650 nm wavelengths optical emission have been stimulated in two junction monolithically integrated circuit p+np+ Injection-avalanche silicon light emitting devices. This has been achieved after extensive modeling of potential light emitting mechanisms that could stimulate light emission at these wavelengths. Substantial evidence has been achieved that the optical emissions occur through short range phonon assisted inter-band transitions when low energy holes recombine with energetic electrons in a high impurity/ defect density environment. In particular, it is shown that the emission intensity depends on the balancing of densities of the energetic electrons and low energy holes in this environment. The devices are of micron dimension and operate at 8–10V, 10\mu \textA –10 mA regimes. Some of the emission spot sizes are submicron. Emission intensities are typically 70nW/ \mu \textm^-2 / mA. The observed effect may find several applications in futuristic on-chip electro-optic applications. A third control contact terminal offers modulation possibilities. [ABSTRACT FROM PUBLISHER]

For the first time, singular perturbation technique in conjunction with effective index method is used to quantitatively estimate parametric variations in a bend rectangular waveguide of subwavelength dimensions that forms a basic unit of photonics integrated circuits optical interconnect. These parametric variations are correlated to losses due to bending and coupling of bend and straight waveguides. The results are compared with that obtained by more rigorous finite elements method. A scheme to compensate the losses is also proposed. In case of tight bends (bending radius ≈ 10 μm) coupling loss compensation of about 40 dB is achieved. [ABSTRACT FROM AUTHOR]

Efficient polymer SOI structures at wavelength 800 nm, 1310 nm and 1550 nm are reported in this research. Here there are three types of polymer such as PMMA-HDPE, PP-PC and PTPE-PETP grating structures are cogitated to realize high transmitted efficiency. Different sorts of losses such as absorption, reflection and diffraction are considered during the computation of overall transmitted efficiency. As far as absorption loss is concerned, it is zero for all structures at three windows. Reflection loss is found using plane wave expansion method, where diffraction efficiency is obtained using numerical equation. Simulation result reveals that overall transmitted efficiency through above grating SOI structure is more than 97% with respect to angle of deviation from Bragg’s angle, which varies from −23° to 23° in all optical communication windows. Finally, this paper divulges an interesting variation between input wavelength and grating length. [ABSTRACT FROM AUTHOR]

An increase of the small-signal modulation bandwidth to more than 25 GHz that is stable over a wide temperature range from 25 °C to 85 °C is reported upon systematic, joint variations of the photon lifetime and the oxide-aperture diameter for 980-nm vertical-cavity surface-emitting lasers. Error-free back-to-back data transmission at 50 Gb/s from 25 °C to 75 °C for completely unchanged driving conditions is reported, leading to an improved energy efficiency of the laser. [ABSTRACT FROM PUBLISHER]

Silicon micro-ring modulators have the potential to enable energy efficient, high-bandwidth-density optical interconnects in CMOS compatible silicon technologies. However, their operation is highly susceptible to any thermal, laser wavelength or process variations and, as such, require a feedback control system for guaranteeing stable modulation. This paper demonstrates wavelength locking of a hybrid CMOS-silicon photonics ring-based transmitter using a 40 nm CMOS circuit to directly monitor the optical modulation amplitude (OMA) at the drop port of the ring modulator. OMA stabilization with an accuracy of < 10 pm ( \sim 0.125\,\,\,^\circ \text C or within \sim 2% of the maximum OMA), a tracking speed of 5 nm/s ( \sim {62.5}\,\,^{\circ }\text {C} /s), and a tuning range of 5 nm ( \sim {62.5}\,\,^{\circ }\text {C}$ ) is demonstrated by subjecting the photonic transmitter both to thermal and laser wavelength variations under dynamic modulation at 2 Gbps. The demonstrated wavelength locking concept is also robust to the variations in the input laser power and can be implemented with low power. [ABSTRACT FROM PUBLISHER]

Single-mode wavelength-division multiplexing (WDM) optical links are an attractive technology to meet the growing interconnect bandwidth demand in data center applications. This paper presents a multi-channel hybrid-integrated photonic receiver based on microring drop filters and waveguide photodetectors implemented in a 130 nm SOI process and high-speed optical front-ends designed in 65 nm CMOS. The source-synchronous receiver utilizes an LC injection-locked oscillator (ILO) in the clock path for improved jitter filtering, while maintaining correlated jitter tracking with the data channels. Receiver sensitivity is improved with a large input-stage feedback resistor transimpedance amplifier (TIA) cascaded with an adaptively-tuned continuous-time linear equalizer (CTLE). In order to stabilize the microring drop filter resonance wavelength, a peak-detector-based thermal tuning loop is implemented with a 0.7 nm range at 43~\mu \text W /GHz efficiency. When tested with a waveguide photodetector with 0.45 A/W responsivity, the receiver achieves -8.0 dBm OMA sensitivity at a BER = 10^\mathrm -12 with a jitter tolerance corner frequency near 20 MHz and a per-channel power consumption of 17 mW including amortized clocking power. [ABSTRACT FROM AUTHOR]

Providing host Application-specific integrated circuit (ASIC) (e.g., microprocessors, switches, or field-programmable gate array (FPGAs)) with terabit per second optical transmission capabilities has emerged as a new requirement in exascale cloud data centers and high-performance computing. The combination of silicon photonics (Si-Phot) and 3-D interconnecting technologies stands as a unique solution for fulfilling this need while addressing the key factors of success that are more Gb/s/cm2, less pJ/bit, and less $/Gb/s. This paper is intended to present and evaluate several Si-Phot-based solutions for providing the host ASIC with 1-Tb/s optical transceivers. Two packaging architectures are proposed, making use of 3-D interconnecting building blocks, and motivated by the need for implementing a few hundreds of high-speed electrical interconnections. Models are given for each of the 3-D building blocks, and simulation of the overall electrical signal integrity is made for the two proposed transceiver architectures. In addition, using the capability of CMOS fabrication lines for producing low-cost transceiver chips (Si-Phot integrated circuits with their electronic ICs) and 3-D chip assemblies, it is also necessary to develop high-throughput/low-cost techniques for coupling the light from the chip to the fiber. One such technique will be presented. [ABSTRACT FROM PUBLISHER]