This PDF file contains the front matter associated with SPIE Proceedings Volume10538, including the Title Page, Copyright information, Table of Contents, Introduction, and Conference Committee listing.

Recent trends show an explosion in the amount of data being created each year. In response to this trend new system architectures optimized for data-centric workloads are being developed. These architectures can only be brought to fruition by developing new technologies to transmit, compute, and store data. My talk will focus on the impact of photonics, and silicon photonics in particular, on these trends and discuss the work being done at HPE labs to advance these technologies

Due to the bottleneck in the continuation of Moore’s law as well as the drastically increasing trend of bandwidth, silicon photonics has emerged as the most promising candidate for implementing next-generation communication networks with ultralow power and ultrahigh speed. Recently, optical computing in integrated photonics, which outperforms electrical counterparts both in power consumption and bandwidth, has attracted a renewed interest due to the accessibility and maturity of ultracompact passive and active integrated components. However, up to now, most of relevant research about optical computing still focus on the realization of fundamental logic gates, not even close to feasible large-scale computing system. In this paper, we demonstrate a high-speed ripple-carry electro-optic full adder using micro-resonators. This approach adopts photons instead of electrons to realize logic operations as well as transfer carry signals from one bit to the next, while all the control signals of operands are applied simultaneously at and within every clock cycle. Thus, the severe latency issue that accumulates as the size of full adder increases can be circumvented, allowing for the improvement in computing speed. This approach also outperforms the conventional electrical counterpart in terms of power consumption due to the relatively smaller propagation loss and switching energy.

We introduce a universal test and measurement system allowing comparative characterisation of optical transceivers, board-to-board optical connectors and both embedded and passive optical circuit boards. The system comprises a test enclosure with interlocking and interchangeable test cards, allowing different technologies spanning different Technology Readiness Levels to be both characterised alone and in combination with other technologies. They form part of the open test design standards portfolio developed on the FP7 PhoxTroT and H2020 COSMICC projects and allow testing on a common test platform.

Disaggregated Data Centers (DCs) have emerged as a powerful architectural framework towards increasing resource utilization and system power efficiency, requiring, however, a networking infrastructure that can ensure low-latency and high-bandwidth connectivity between a high-number of interconnected nodes. This reality has been the driving force towards high-port count and low-latency optical switching platforms, with recent efforts concluding that the use of distributed control architectures as offered by Broadcast-and-Select (BS) layouts can lead to sub-μsec latencies. However, almost all high-port count optical switch designs proposed so far rely either on electronic buffering and associated SerDes circuitry for resolving contention or on buffer-less designs with packet drop and re-transmit procedures, unavoidably increasing latency or limiting throughput. In this article, we demonstrate a 256x256 optical switch architecture for disaggregated DCs that employs small-size optical delay line buffering in a distributed control scheme, exploiting FPGA-based header processing over a hybrid BS/Wavelength routing topology that is implemented by a 16x16 BS design and a 16x16 AWGR. Simulation-based performance analysis reveals that even the use of a 2- packet optical buffer can yield <620nsec latency with >85% throughput for up to 100% loads. The switch has been experimentally validated with 10Gb/s optical data packets using 1:16 optical splitting and a SOA-MZI wavelength converter (WC) along with fiber delay lines for the 2-packet buffer implementation at every BS outgoing port, followed by an additional SOA-MZI tunable WC and the 16x16 AWGR. Error-free performance in all different switch input/output combinations has been obtained with a power penalty of <2.5dB.

The soaring traffic demands in datacenter networks (DCNs) are outpacing progresses in CMOS technology, challenging the bandwidth and energy scalability of currently established technologies. Optical switching is gaining traction as a promising path for sustaining the explosive growth of DCNs; however, its practical deployment necessitates extensive modifications to the network architecture and operation, tailored to the technological particularities of optical switches (i.e. no buffering, limitations in radix size and speed). European project NEPHELE is developing an optical network infrastructure that leverages optical switching within a software-defined networking (SDN) framework to overcome the bandwidth and energy scaling challenges of datacenter networks. An experimental validation of the NEPHELE data plane is reported based on commercial off-the-shelf optical components controlled by FPGA boards. To facilitate dynamic allocation of the network resources and perform collision-free routing in a lossless network environment, slotted operation is employed (i.e. using time-division multiple-access - TDMA). Error-free operation of the NEPHELE data plane is verified for 200 μs slots in various scenarios that involve communication between Ethernet hosts connected to custom-designed top-of-rack (ToR) switches, located in the same or in different datacenter pods. Control of the slotted data plane is obtained through an SDN framework comprising an OpenDaylight controller with appropriate add-ons. Communication between servers in the optical-ToR is demonstrated with various routing scenarios, concerning communication between hosts located in the same rack or in different racks, within the same or different datacenter pods. Error-free operation is confirmed for all evaluated scenarios, underpinning the feasibility of the NEPHELE architecture.

The urgent need for high-bandwidth and high-port connectivity in Data Centers has boosted the deployment of optoelectronic packet switches towards bringing high data-rate optics closer to the ASIC, realizing optical transceiver functions directly at the ASIC package for high-rate, low-energy and low-latency interconnects. Even though optics can offer a broad range of low-energy integrated switch fabrics for replacing electronic switches and seamlessly interface with the optical I/Os, the use of energy- and latency-consuming electronic SerDes continues to be a necessity, mainly dictated by the absence of integrated and reliable optical buffering solutions. SerDes undertakes the role of optimally synergizing the lower-speed electronic buffers with the incoming and outgoing optical streams, suggesting that a SerDes-released chip-scale optical switch fabric can be only realized in case all necessary functions including contention resolution and switching can be implemented on a common photonic integration platform. In this paper, we demonstrate experimentally a hybrid Broadcast-and-Select (BS) / wavelength routed optical switch that performs both the optical buffering and switching functions with μm-scale Silicon-integrated building blocks. Optical buffering is carried out in a silicon-integrated variable delay line bank with a record-high on-chip delay/footprint efficiency of 2.6ns/mm² and up to 17.2 nsec delay capability, while switching is executed via a BS design and a silicon-integrated echelle grating, assisted by SOA-MZI wavelength conversion stages and controlled by a FPGA header processing module. The switch has been experimentally validated in a 3x3 arrangement with 10Gb/s NRZ optical data packets, demonstrating error-free switching operation with a power penalty of <5dB.

White LEDS revolutionized the field of illumination technology mainly due to the energy saving effects. Besides lighting purposes LEDs can also be used in wireless communication systems when integrated in Visible Light Communication (VLC) systems. Indoor positioning for navigation in large buildings is currently under research to overcome the difficulties associated with the use of GPS in such environments. The motivation for this application is also supported by the possibility of taking advantage of an existing lighting and WiFi infrastructure. In this work it is proposed an indoor navigation system based on the use of VLC technology. The proposed system includes trichromatic white LEDs with the red and blue chips modulated at different frequencies and a pinpin photodetector with selective spectral sensitivity. Optoelectronic features of both optical sources and photodetector device are analyzed. The photodetector device consists two pin structures based on a-SiC:H and a-Si:H with geometrical configuration optimized for the detection of short and large wavelengths in the visible range. Its sensitivity is externally tuned by steady state optical bias. The localization algorithm makes use of the Fourier transform to identify the frequencies present in the photocurrent signal and the wavelength filtering properties of the sensor under front and back optical bias to detect the existing red and blue signals. The viability of the system was demonstrated through the implementation of an automatic algorithm to infer the photodetector cardinal direction. A capacitive optoelectronic model supports the experimental results and explains the device operation.

This paper focuses on designing and analysing a visible light based communication and positioning system. The indoor positioning system uses trichromatic white Light Emitting Diodes (LEDs), both for illumination purposes and as transmitters, and an optical processor, based on a-SiC:H technology, as mobile receiver. On-Off Keying (OOK) modulation scheme is used, proving a good trade-off between system performance and implementation complexity. In the following, the relationship between the transmitted data and the received output levels is decoded. LED bulbs work as transmitters, sending information together with different identifiers, IDs, related to their physical locations. Square and diamond topologies for the unit cell are analyzed, and a 2D localization design, demonstrated by a prototype implementation, is presented. Fine-grained indoor localization is tested. The received signal is used in coded multiplexing techniques for supporting communications and navigation concomitantly on the same channel. The location and motion information is found by mapping the position and estimating the location areas.

The impact of integrated photonics on optical interconnects is currently muted by challenges in photonic packaging and in the dense integration of photonic modules with microelectronic components on printed circuit boards. Single mode optics requires tight alignment tolerance for optical coupling and maintaining this alignment in a cost-efficient package can be challenging during thermal excursions arising from downstream microelectronic assembly processes. In addition, the form factor of typical fiber connectors is incompatible with the dense module integration expected on printed circuit boards. We have implemented novel approaches to interfacing photonic chips to standard optical fibers. These leverage standard high throughput microelectronic assembly tooling and self-alignment techniques resulting in photonic packaging that is scalable in manufacturing volume and in the number of optical IOs per chip. In addition, using dense optical fiber connectors with space-efficient latching of fiber patch cables results in compact module size and efficient board integration, bringing the optics closer to the logic chip to alleviate bandwidth bottlenecks. This packaging direction is also well suited for embedding optics in multi-chip modules, including both photonic and microelectronic chips. We discuss the challenges and rewards in this type of configuration such as thermal management and signal integrity.

Multicore fiber enables a parallel optic data link in a single optical fiber. Thus, it is an attractive approach to increase the aggregate data throughput and the integration density of the interconnection. We developed and demonstrated mid-board optical transceiver modules employing novel multicore fiber pigtails and multicore-optimized optoelectronic engines. The silica fibers having 125 µm diameter and including six graded-index multimode cores enable multi-gigabit interconnects at very short distances. The fiber is compatible with the 850-nm VCSEL technology that has many advantages, such as, the very low power operation and the mature and cost-effective GaAs-based device technology. The transceiver incorporates transmitter and receiver subassemblies that are based on the multicore-optimized 850-nm VCSEL and photodiode array chips as well as on the co-designed multichannel VCSEL driver and TIA receiver ICs. All devices are operating up to 25 Gbps/channel and beyond, thus creating a 150 Gbps full-duplex link with the two 6-core fibers. The active areas on the 6-channel VCSEL and PD chips are arranged in a circular array layout that matches the cross-sectional layout of the fiber cores. This allows butt coupling to the fiber cores. The power consumption of the complete link is below 5 mW/Gbps. The transceiver was developed to be applicable for harsh environmental conditions, including space. Therefore, for instance, hermetic packaging was applied and both the active devices and the integration structure enable very wide operation temperature range of up to approx. 100 °C. This paper will present the technical approach including the basic building blocks and the transceiver module implementation. It will also present the results of the data link performance and some reliability testing.

Due to its optical transparency and superior dielectric properties glass is regarded as a promising candidate for advanced applications as active photonic interposer for mid-board-optics and optical PCB waveguide integration. The concepts for multi-mode and single-mode photonic system integration are discussed and related demonstration project results will be presented. A hybrid integrated photonic glass body interposer with integrated optical lenses for multi-mode data communication wavelength of 850 nm have been realized. The paper summarizes process developments which allow cost efficient metallization of TGV. Electro-optical elements like photodiodes and VCSELs can be directly flip-chip mounted on the glass substrate according to the desired lens positions. Furthermore results for a silicon photonic based single-mode active interposer integration onto a single mode glass made EOCB will be compared in terms of packaging challenges. The board level integration strategy for both of these technological approaches and general next generation board level integration concepts for photonic interposer will be introductorily discussed.

High density 3-dimensional optical and electrical I/O structures for chip-scale-packaged Si photonic optical transceivers have been developed. The optical I/O is a vertical polymer waveguide structure made of UV curable resins. The waveguide structure includes 125-μm-pitch 8°-tilted cores, which is called “optical pin.” The tilted optical pins were formed by oblique illuminated photolithography. The optical pins for transmitters confine emitted light from grating couplers in a temperature range from -40°C to 85°C. The optical pins for receivers have inverted-cone shape which acts as a spot size converter connecting a multimode fiber and a smaller integrated surface-illuminated photodiode. Combination of multimode transmission and the optical pins alleviates coupling tolerance as compared with conventional single-mode transmission. The electrical I/O structure for flip-chip bonding comprises 250-μm-pitch regularly arranged through-glass-vias (TGVs). The TGV is a hermetically-sealed W wire and has connecting pads at both ends. The TGVs are connected to a Si photonic chip by Au-to-Au ultrasonic bonding. The outer I/O pads are covered with electroless-plated Ni/Au, so that conventional lead-free soldering can be available. The optical and electrical interfaces for the optical I/O cores with TGVs are placed on the same side of the module. This configuration enables simultaneous optical and electrical bonding to polymer-waveguide-embedded printed circuit boards. Because the developed I/O structures minimize area penalty and support 25 Gbps/ch transmission, Si photonic optical transceivers with the developed I/O structures are suitable for use in photonics electronics converged systems. This research was partly supported by the New Energy and Industrial Technology Development Organization (NEDO).

In this paper, we report the feature and properties about organic-inorganic hybrid material, “SUNCONNECT®” for photonic integrated circuit.

“SUNCONNECT®” materials have low propagation loss at 1310nm (0.29dB/cm) and 1550nm (0.45dB/cm) respectively. In addition, the material has high thermal resistance both high temperature annealing test at 300°C and also 260°C solder heat resistance test. For actual device application, high reliability is required. 85°C /85% test was examined by using multi-mode waveguide. As a result, it indicated that variation of insertion loss property was not changed significantly after high temperature / high humidity test.

For the application to photonic integrated circuit, it was demonstrated to fabricate polymer optical waveguide by using three different methods. Single-micron core pattern can be fabricated on cladding layer by using UV lithography with proximity gap exposure. Also, single-mode waveguide can be also fabricated with over cladding. On the other hands, “Mosquito method” and imprint method can be applied to fabricate polymer optical waveguide. Remarkably, these two methods can fabricate gradedindex type optical waveguide without using photo mask. In order to evaluate the optical performance, NFP’s observation, measurement of insertion loss and propagation loss by cut-back methods were carried out by using each waveguide sample.

Optical-Printed Circuit Board (PCB) is an emerging optical interconnect technology to bridge the gap between the board edge and the processing module. The technology so far has been used as a broadband transmitter using polymer waveguides in the PCB. In this paper, we report a Silicon Nitride based photonic IC embedded in the PCB along with the polymers as waveguides in the PCB. The motivation for such integration is to bring routing capability and to reduce the power loss due to broadcasting mode.

We present herein the active PbS-photopolymer waveguide fabricated by vacuum assisted microfluidic (VAM) soft lithography technique. The PbS Quantum Dots (QDs) were synthesized using colloidal chemistry methods with tunable sizes and emission wavelengths, resulting in efficient light emission around 1000 nm center wavelength. The PbS QDs have demonstrated much better solubility in our newly synthesized UV curable polymer than SU-8 photoresist, verified by Photoluminescence (PL) testing. Through refractive index control, the PbS QDs-polymer core material and polymer cladding material can efficiently confine the infrared emitting light with a broad spectral bandwidth of ~180 nm. Both single-mode and multi-mode light emitting waveguides have been realized.

Multimode polymer waveguides have attracted great interest for use in high-speed short-reach communication links as they can be cost-effectively integrated onto standard PCBs using conventional methods of the electronics industry and provide low loss (<0.04 dB/cm at 850 nm) and high bandwidth (>30 GHz×m) interconnection. The formation of such waveguides on flexible substrates can further provide flexible low-weight low-thickness interconnects and offer additional freedom in the implementation of high-speed short-reach optical links. These attributes make these flexible waveguides particularly attractive for use in low-cost detachable chip-to-chip links and in environments where weight and shape conformity become important, such as in cars and aircraft. However, the highly-multimoded nature of these waveguides raises important questions about their performance under severe flex due to mode loss and mode coupling. In this work therefore, we investigate the loss, crosstalk and bandwidth performance of such waveguides under out-of plane bending and in-plane twisting under different launch conditions and carry out data transmission tests at 40 Gb/s on a 1 m long spiral flexible waveguide under flexure. Excellent optical transmission characteristics are obtained while robust loss, crosstalk and bandwidth performance are demonstrated under flexure. Error-free (BER<10^-12) 40 Gb/s data transmission is achieved over the 1 m long spiral waveguide for a 180° bend with a 4 mm radius. The obtained results demonstrate the excellent optical and mechanical properties of this technology and highlight its potential for use in real-world systems.

A novel silicon reconfigurable optical processor (SROP) is proposed based on a self-coupled optical waveguide (SCOW) resonant architecture. The SCOW is formed by a single meandering waveguide self-coupled to form directional couplers at the input and output ends. The SROP consists of three-stages of SCOW resonators, where the directional couplers are replaced by Mach-Zehnder interferometers (MZIs) to work as tunable couplers. Both phase shifters and variable optical attenuators (VOAs) are integrated in the device for active phase and amplitude tuning, respectively. By reconfiguring the states of tunable couplers, phase shifters and VOAs, the SROP is implemented as ring resonator, MZI, and Fabry-Perot cavity-based optical structures for versatile optical processing applications.

We developed a small form factor connector that can be assembled on all four sides of a high-data switch package for fiber connectivity. This paper discusses a novel connector approach that has the potential to meet all co-packaging requirements including solder-reflow-compatibility, de-mateability, low insertion loss and state-of-the art FAU attach. The connector was attached to the PIC for performance evaluation. The average insertion loss across all eight fibers of the assembly was 1.77 dB, including the three optical interfaces: (1) MT-to-MT between connector and receptacle, (2) receptacle-to-PLC and (3) PIC-to-FAU. Also included is the propagation loss of the PIC waveguide. Optical return loss was measured to be -55 dB or lower.

Since the first usage of optical glasses there has been an increasing demand for optical systems which are highly customized for a wide field of applications. To meet the challenge of the production of so many unique systems, the development of new techniques and approaches has risen in importance. However, the assembly of precision optical systems with lot sizes of one up to a few tens of systems is still dominated by manual labor. In contrast, highly adaptive and model-based approaches may offer a solution for manufacturing with a high degree of automation and high throughput while maintaining high precision.

In this work a model-based automated assembly approach based on ray-tracing is presented. This process runs autonomously, and accounts for a wide range of functionality. It firstly identifies the sequence for an optimized assembly and secondly, generates and matches intermediate figures of merit to predict the overall optical functionality of the optical system. This process also takes into account the generation of a digital twin of the optical system, by mapping key-performance-indicators like the first and the second momentum of intensity into the optical model. This approach is verified by the automatic assembly of an anamorphic telescope within an assembly cell. By continuous measuring and mapping the key-performance-indicators into the optical model, the quality of the digital twin is determined. Moreover, by measuring the optical quality and geometrical parameters of the telescope, the precision of this approach is determined. Finally, the productivity of the process is evaluated by monitoring the speed of the different steps of the process.

Data centers (DCs) are continually striving to meet the challenge of delivering more bandwidth, and a key part of that endeavor will be the deployment of higher capacity interconnects. Proposed technologies for achieving this higher capacity include higher line rates, advanced modulation formats, wavelength division multiplexing, spatial division multiplexing and reduced diameter fibers. In this paper, we focus on reduced diameter fibers that can be incorporated into higher density cables and connectors.

We have prototyped optical fibers in which the glass diameter is 80 microns and the coating diameter has been reduced to as small as 125 microns. These fibers enable up to twice as many optical interconnects in a given footprint, which will allow transceiver arrays to be collinearly located on small chips with dimensions on the order of (5x5mm²)¹. We have also incorporated these reduced diameter fibers into small form-factor, flexible 8-fiber ribbons which can enable as much as a 3x increase in fiber density and can simplify routing constraints inside modules and optical backplanes.

Silicon photonics has traditionally focused on near infrared wavelengths, with tremendous progress seen over the past decade. However, more recently, research has extended into mid infrared wavelengths of 2 μm and beyond. Optical modulators are a key component for silicon photonics interconnects at both the conventional communication wavelengths of 1.3 μm and 1.55 μm, and the emerging mid-infrared wavelengths. The mid-infrared wavelength range is particularly interesting for a number of applications, including sensing, healthcare and communications. The absorption band of conventional germanium photodetectors only extends to approximately 1.55 μm, so alternative methods of photodetection are required for the mid-infrared wavelengths. One possible CMOS compatible solution is a silicon defect detector. Here, we present our recent results in these areas. Modulation at the wavelength of 2 μm has been theoretically investigated, and photodetection above 25 Gb/s has been practically demonstrated.

Femtosecond laser inscription allows straightforward manufacturing of glass planar lightwave circuits such as waveguides, interferometers, directional couplers, resonators and more complex structures. Fiber alignment structures are needed to facilitate communication with the glass planar lightwave circuit. In this study, a technique is described to create optical waveguides and alignment structures in the same laser exposure step. Using an industrial ytterbium-doped 1030 nm fiber laser pulses of 400 fs were focused into glass with a 0.4 NA objective causing permanent alteration of the material. Depending on laser parameters this modification allows direct writing of waveguides or the creation of channels after exposing the irradiated volumes to an etchant such as KOH. Writing of channels and waveguides with different laser powers, frequencies, polarisations, stage translation speeds and scan densities were investigated in fused silica and borosilicate glass. Waveguides with controlled dimensions were created, as well as etched U-grooves with a diameter of 126 μm and a sidewall roughness R_a of 255 nm. Cut back measurements were performed giving a waveguide propagation loss of 1.1 dB/cm in borosilicate glass. A coupling loss of 0.7 dB was measured for a transition between the waveguide and standard single mode fiber at 1550 nm, using index matching liquid. The described technique eliminates active alignment requirements and is useful for many applications such as microfluidic sensing, PLCs, fan-out connectors for multicore fibers and quantum optical networks.

Silicon photonics with micron-scale Si waveguides offers most of the benefits of submicron SOI technology while avoiding most of its limitations. In particular, thick silicon-on-insulator (SOI) waveguides offer 0.1 dB/cm propagation loss, polarization independency, broadband single-mode (SM) operation from 1.2 to >4 µm wavelength and ability to transmit high optical powers (>1 W). Here we describe the feasibility of Thick-SOI technology for advanced optical interconnects. With 12 μm SOI waveguides we demonstrate efficient coupling between standard single-mode fibers, vertical-cavity surface-emitting lasers (VCSELs) and photodetectors (PDs), as well as wavelength multiplexing in small footprint. Discrete VCSELs and PDs already support 28 Gb/s on-off keying (OOK), which shows a path towards 50-100 Gb/s bandwidth per wavelength by using more advanced modulation formats like PAM4. Directly modulated VCSELs enable very power-efficient optical interconnects for up to 40 km distance. Furthermore, with 3 μm SOI waveguides we demonstrate extremely dense and low-loss integration of numerous optical functions, such as multiplexers, filters, switches and delay lines. Also polarization independent and athermal operation is demonstrated. The latter is achieved by using short polymer waveguides to compensate for the thermo-optic effect in silicon. New concepts for isolator integration and polarization rotation are also explained.

Semiconductor nanowire (NW) lasers are nanoscale coherent light sources that exhibit a small footprint, low-threshold lasing characteristics, and properties suitable for monolithic integration onto Si photonic circuits. An important milestone on the way towards novel on-chip photonic functionalities is the integration of individual, deterministically addressable NW lasers on Si waveguides with efficient coupling and mode propagation in the underlying photonic circuit. Here, we demonstrate the monolithic integration of single GaAs-based NW lasers directly onto lithographically defined Si ridge waveguides (WG). Subject to optical excitation, the observed lasing behavior shows clear “s-shape”-characteristics, linewidth narrowing and threshold values down to 19.8±1 µJ/cm², which is the lowest value reported to date for this kind of integrated lasing structure. The lasing mode of individual NW lasers is shown to couple efficiently into propagating modes of the underlying orthogonal Si WG, preserving the spectral characteristics during mode pro-pagation in the WG, and in good agreement with Finite-Difference Time-Domain (FDTD) simulations. Using a WG structure with a series of mask openings along the central mode propagation axis, we further illustrate the out-coupling properties of both spontaneous and stimulated emission and demonstrate propagation of the lasing mode over distances > 60 µm, despite absorption in the silicon dominating the propagation losses. By replacing the bulk GaAs as the active gain medium by InxGa1-xAs quantum wells, the emission wavelength can be shifted towards telecommunication bands in order to reduce the absorption losses in silicon. These results pave the way for future on-chip monolithic integration of III-V NW lasers onto silicon based optoelectronic circuits.

Initially, we’ll discuss an SOI based, carrier injection micro-ring modulator. The static optical and electrical characteristics of this device will be reviewed and described. Thermal control and modulation mechanisms with pre-emphasis will be outlined. Automated wafer-lever optical/electrical results from volume foundries (Leti and STMicro) used for PDK/Verilog model development will be reviewed, along with experimental data on direct modulation to 25 Gb/s, crosstalk at various DWDM channel separations, and demonstrations with an external quantum-dot based comb laser with 80 and 50GHz channel spacing. Following this, our work on directly modulated hybrid quantum-well ring lasers will be reviewed in design, fabrication. Experimental results for modulation at 12.5 Gb/s/channel, integration with MOS capacitor for wavelength control and modulation and a thermal shunt for temperature management will highlight the advantages of this technology that may be exploited. Subsequently, our work on hybrid quantum-dot based comb lasers for on-chip DWDM sources will be discussed in their details of physical operation, demonstrating successful mode-locking and noise-free operation across the 20-80C thermal range. Our work on the integration of on-chip APDs from a CMOS-compatible processes will also be reviewed, demonstrating error-free operation at 12.5 Gb/s and 25 Gb/s with a sensitivity of -26dBm and -16dBm, respectively. The use of APDs will drastically decrease the overall power consumption of the interconnect, lowering total cost of ownership. Finally, our most recent progress on integration of the silicon photonics with CMOS by a flip-chip will be reviewed showing high-speed modulation and thermal control for a multi-channel DWDM transceiver.

Here we proposed three adiabatically coupled waveguides (ACW), while the outer waveguides perform as a two-mode system analogous to ground- and excited- states and the middle waveguide is same as dark-state in a three-level-atomic system. Thanks to the dark-state, intermediate waveguide is based on plasmonic Indium thin oxide (ITO) as an active structure. Our simulation indicates a power consumption of 40 atto-joule with 50 dB modulation depth. In addition, our ACW plasmonic modulator provides high-speed operation as high as 5.4 THz and insertion loss as low as 0.45 dB. The proposed device is crucial for futuristic of optical short-reach interconnects.

Multiple parallel channels are ubiquitous in optical communications, with spatial division multiplexing (separate physical paths) and wavelength division multiplexing (separate optical wavelengths) being the most common forms. Here, we investigate the viability of polarization division multiplexing, the separation of distinct parallel optical communication channels through the polarization properties of light. Two or more linearly polarized optical signals (at different polarization angles) are transmitted through a common medium, filtered using aluminum nanowire optical filters fabricated on-chip, and received using individual silicon photodetectors (one per channel). The entire receiver (including optics) is compatible with standard CMOS fabrication processes.

The filter model is based upon an input optical signal formed as the sum of the Stokes vectors for each individual channel, transformed by the Mueller matrix that models the filter proper, resulting in an output optical signal that impinges on each photodiode. The results show that two- and three-channel systems can operate with a fixed-threshold comparator in the receiver circuit, but four-channel systems (and larger) will require channel coding of some form. For example, in the four-channel system, 10 of 16 distinct bit patterns are separable by the receiver. The model supports investigation of the range of variability tolerable in the fabrication of the on-chip polarization filters.

A high-speed modulator based on electro-optic (EO) polymer infiltrated sub-wavelength grating (SWG) waveguide ring resonator is presented. The 3-dB modulation bandwidth of the fabricated modulator is measured to be larger than 40 GHz occupying an area of 70 μm x 29 μm, which is the largest bandwidth and the most compact footprint that has been demonstrated for the ring resonators on the silicon-organic hybrid platform.