This PDF file contains the front matter associated with SPIE Proceedings Volume 8989 including the Title Page, Copyright information, Table of Contents, Introduction, and Conference Committee listing.

A simple method of growing large areas of InP on Si through Epitaxial Lateral Overgrowth (ELOG) is presented. Isolated areas of high quality InP suitable for photonic integration are grown in deeply etched SiO2 mask fabricated using conventional optical lithography and reactive ion etching. This method is particularly attractive for monolithically integrating laser sources grown on InP with Si/SiO2 waveguide structure as the mask. The high optical quality of multi quantum well (MQW) layers grown on the ELOG layer is promisingly supportive of the feasibility of this method for mass production.

In this paper, photonic devices driven by surface acoustic waves and operating in the GHz frequency range are presented. The devices were designed and fabricated in (Al,Ga)As technology. In contrast to previously realized modulators, where part of the light transmission is lost due to destructive interference, in the present devices light only switches paths, avoiding losses. One of the devices presents two output channels with 180◦-dephasing synchronization. Odd multiples of the fundamental driving frequency are enabled by adjusting the applied acoustic power. A second and more complex photonic integrated device, based on the acoustic modulation of tunable Arrayed Waveguide Gratings, is also proposed.

Free-space beam steering using optical phased arrays is a promising method for implementing free-space communication links and Light Detection and Ranging (LIDAR) without the sensitivity to inertial forces and long latencies which characterize moving parts. Implementing this approach on a silicon-based photonic integrated circuit adds the additional advantage of working with highly developed CMOS processing techniques. In this work we discuss our progress in the development of a fully integrated 32 channel PIC with a widely tunable diode laser, a waveguide phased array, an array of fast phase modulators, an array of hybrid III-V/silicon amplifiers, surface gratings, and a graded index lens (GRIN) feeding an array of photodiodes for feedback control. The PIC has been designed to provide beam steering across a 15°x5° field of view with 0.6°x0.6° beam width and background peaks suppressed 15 dB relative to the main lobe within the field of view for arbitrarily chosen beam directions. Fabrication follows the hybrid silicon process developed at UCSB with modifications to incorporate silicon diodes and a GRIN lens.

We will describe recent work in the area of integrated nanophotonics that have applications to on chip communication. In this context we will present passive filters and resonators produced through periodic waveguide modulation. We will also demonstrate nonlinear optical pulse compression in a monolithic device that integrates self phase modulation and dispersion compensation. We also discuss wavemixing applications.

Based on Amdahl scaling of tree-networks, we show that in the next 10 years power efficiency and cost of data center communication networks have to improve with three orders of magnitude. Flattened network architectures may allow for more efficient scaling but require high-radix network switches. In turn, such switches will require opto-electronic conversion in close proximity of the switch ASIC. In this paper, we focus on three-dimensional die-stacked transceiver ICs that allow for low cost fabrication and packaging that may enable flattened network architectures based on highradix switches.

Exponential increases in the amount of data that need to be sensed, communicated, and processed are continuing to drive the complexity of our computing, networking, and sensing systems. High degrees of integration is essential in scalable, practical, and cost-effective microsystems. In electronics, high-density 2D integration has naturally evolved towards 3D integration by stacking of memory and processor chips with through-silicon-vias. In photonics, too, we anticipate highdegrees of 3D integration of photonic components to become a prevailing method in realizing future microsystems for information and communication technologies. However, compared to electronics, photonic 3D integration face a number of challenges. This paper will review two methods of 3D photonic integration --- fs laser inscription and layer stacking, and discuss applications and future prospects.

Graphene, a well-known two-dimensional sheet of carbon atoms in a honeycomb structure, has many unique and fascinating properties in optoelectronics and photonics. Integration of graphene on silicon nanophotonic wires is a promising approach to enhance light-graphene interactions. In this paper, we demonstrate on-chip silicon nanophotonic wires covered by graphene with CMOS-compatible fabrication processes. Under the illumination of pump light on the graphene sheet, a loss reduction of silicon nanophotonic wires, which is called optically induced transparency (OIT) effect, is observed over a broad wavelength range for the first time. The pump power required to generate the OIT effect is as low as ~0.1mW and the corresponding power density is about 2×10³mW/cm², which is significantly different from the saturated absorption effect of graphene reported previously. The extremely low power density implies a new mechanism for the present OIT effect, which will be beneficial to realize silicon on-chip all-optical controlling in the future. It also suggests a new and efficient approach to tune the carrier concentration (doping level) in graphene optically.

In this paper we will discuss recent results in our work on Silicon Photonics. This will include active and passive devices for a range of applications. Specifically we will include work on modulators and drivers, deposited waveguides, multiplexers, device integration and Mid IR silicon photonics. These devices and technologies are important both for established applications such as integrated transceivers for short reach interconnect, as well as emerging applications such as disposable sensors and mass market photonics.

We describe the interface circuits to silicon photonics modulators, optical filters, and detectors that will be required to enable silicon photonics micro-ring and micro-disk devices to be integrated in dense wavelength division multiplexing circuitry.

This paper describes design methodologies developed for silicon photonics integrated circuits. The approach presented is inspired by methods employed in the Electronics Design Automation (EDA) community. This is complemented by well established photonic component design tools, compact model synthesis, and optical circuit modelling. A generic silicon photonics design kit, as described here, is available for download at http://www.siepic.ubc.ca/GSiP.

Silicon photonic system, integrating photonic and electronic signal processing circuits in low-cost silicon CMOS processes, is a rapidly evolving area of research. The silicon electroabsorption modulator (EAM) is a key photonic device for emerging high capacity telecommunication networks to meet ever growing computing demands. To replace traditional large footprint Mach-Zehnder Interferometer (MZI) type modulators several small footprint modulators are being researched. Carrier-injection modulators can provide large free carrier density change, high modulation efficiency, and compact footprint. The large optical bandwidth and ultra-fast transit times of 130nm HBT devices make the carrierinjection HBT-based EAM (HBT-EAM) a good candidate for ultra-high-speed optical networks. This paper presents the design and 3D full-wave simulation results of a traveling wave electrode (TWE) structure to increase the modulation speed of a carrier-injection HBT-EAM device. A monolithic TWE design for an 180um ultra compact carrier-injection-based HBT-EAM implemented in a commercial 130nm SiGe BiCMOS process is discussed. The modulator is electrically modeled at the desired bias voltage and included in a 3D full-wave simulation using CST software. The simulation shows the TWE has a S11 lower than -15.31dB and a S21 better than -0.96dB covering a bandwidth from DC-60GHz. The electrical wave phase velocity is designed close to the optical wave phase velocity for optimal modulation speed. The 3D TWE design conforms to the design rules of the BiCMOS process. Simulation results show an overall increase in modulator data rate from 10Gbps to 60Gbps using the TWE structure.

We present design of a front-end readout system consisting of charge sensitive amplifier (CSA) and pulse shaper for detection of stochastic and ultra-small semiconductor scintillator signal. The semiconductor scintillator is double sided silicon detector (DSSD) or avalanche photo detector (APD) for high resolution and peak signal reliability of γ-ray or X-ray spectroscopy. Such system commonly uses low noise multichannel CSA. Each CSA in multichannel includes continuous reset system based on tens of MΩ and charge-integrating capacitor in feedback loop. The high value feedback resistor requires large area and huge power consumption for integrated circuits. In this paper, we analyze these problems and propose a CMOS short pulse detection system with a novel CSA. The novel CSA is composed of continuous reset system with combination of diode connected PMOS and 100 fF. This structure has linearity with increased input charge quantity from tens of femto-coulomb to pico-coulomb. Also, the front-end readout system includes both slow and fast shapers for detecting CSA output and preventing pile-up distortion. Shaping times of fast and slow shapers are 150 ns and 1.4 μs, respectively. Simulation results of the CMOS detection system for optical short-pulse implemented in 0.18 μm CMOS technology are presented.

While the promise of implementing slow light into integrated photonic circuitry offers all-optical processing and nextgeneration functionality such as scalable quantum information processing, serious problems associated with both the intrinsic and extrinsic physics of slow light threaten to compromise real-world implementations. In this work, we offer a practical implementation of slow light that is not affected by these issues. When slow light is actualized below certain length scales, it can used to enhance the functionality of already optimized photonic architectures and enable novel capabilities. One such important capability is to highly couple and correlate out-of-plane phenomena, such as optical trapping or single-photon extraction, with in-plane optical processing, such as interferometry and optical transmission in a compact area. We first present the theoretical arguments for such an implementation and then present two applications where local slow light engineering enables important functionality: scalable quantum information circuitry and compact single-molecule cellular phenotyping.

We investigate theoretically and experimentally the computational properties of an optoelectronic neuromorphic processor based on a complex nonlinear dynamics. This neuromorphic approach is based on a new paradigm of or reservoir computing, which is intrinsically different from the concept of Turing machines. It essentially consists in expanding the input information to be processed into a higher dimensional phase space, through the nonlinear transient response of a complex dynamics excited by the input information. The computed output is then extracted via a linear separation of the transient trajectory in the complex phase space, performed through a learning phase consisting of the resolution of a regression problem. We here investigate an architecture for photonic neuromorphic computing via these complex nonlinear dynamical transients. A versatile photonic nonlinear transient computer based on a multiple-delay is reported. Its hybrid analogue and digital architecture allows for an easy reconfiguration, and for direct implementation of in-line processing. Its computational efficiency in parameter space is also analyzed, and the computational performance of this system is successfully evaluated on a standard spoken digit recognition task. We then discuss the pathways that can lead to its effective integration.

The integration of optics for efficient light delivery and the collection of fluorescence from trapped ions in surface electrode ion traps is a key component to achieving scalability for quantum information processing. Diffractive optical elements (DOEs) present a promising approach as compared to bulk optics because of their small physical profile and their flexibility in tailoring the optical wavefront. The precise alignment of the optics for coupling fluorescence to and from the ions, however, poses a particular challenge. Excitation and manipulation of the ions requires a high degree of optical access, significantly restricting the area available for mounting components. The ion traps, DOEs, and other components are compact, constraining the manipulation of various elements. For efficient fluorescence collection from the ions the DOE must be have a large numerical aperture (NA), which results in greater sensitivity to misalignment. The ion traps are sensitive devices, a mechanical approach to alignment such as contacting the trap and using precision motors to back-off a set distance not only cannot achieve the desired alignment precision, but risks damage to the ion trap. We have developed a non-contact precision optical alignment technique. We use line foci produced by off-axis linear Fresnel zone plates (FZPs) projected on alignment targets etched in the top metal layer of the ion trap and demonstrate micron-level alignment accuracy.

Optical phase conjugation is a technique that could find many applications in medical imaging and industry. However, state of the art techniques are limited in speed, portability and efficiency. Especially for digital optical phase conjugation, the electronic delays for image readout on a camera and addressing a spatial light modulator make this technique unpractical for phase conjugation in biological medium. Furthermore, the calibration of such a system is a very complex and expensive task. Thus, we propose integrating on the same device a camera and a liquid crystals spatial light modulator to achieve phase control thanks to in-pixel processing of a photodiode signal.

Diabetes is a fast growing metabolic disease, where the patients suffer from disordered glucose blood levels. Monitoring the blood glucose values in combination with extra insulin injection is currently the only therapy to keep the glucose concentration in diabetic patients under control, minimizing the long-term effects of elevated glucose concentrations and improving quality of life of the diabetic patients. Implantable sensors allow continuous glucose monitoring, offering the most reliable data to control the glucose levels. Infrared absorption spectrometers offer a non-chemical measurement method to determine the small glucose concentrations in blood serum. In this work, a spectrometer platform based on silicon photonics is presented, allowing the realization of very small glucose sensors suitable for building implantable sensors. A proof-of-concept of a spectrometer with integrated evanescent sample interface is presented, and the route towards a fully implantable spectrometer is discussed.

Phase modulators in surface plasmon resonance phase-differential imaging (SPR-PI) sensing systems reported so far are sensitive to temperature fluctuations or mechanical vibrations and thus their applications are limited. In this paper, we propose a novel prism phase modulator (PPM) to replace a traditional modulator. The PPM consists of a parallel prism, a rotation stage and a mirror. The PPM shows great stability in our experiment, and helps achieve high detection sensitivity in our SPR-PI system. Moreover, the cost of our PPM is much lower than that of a traditional modulator and is thus suitable for commercialization. A polydimethylsiloxane (PDMS) microfluidic chip is fabricated to control the flow velocity and realize parallel detection in our experiment. Measured result of glycerine solution shows that the resolution of our SPR biosensor array is about 9.11×10^-7 refractive index unit (RIU). Real time monitoring of interaction between bovine IgG and anti-bovine IgG is also realized. The proposed PPM-based microfluidic SPR-PI biosensor array is promising for future practical applications.

Submicron silicon-on-insulator (SOI) optical waveguides, which has ultra-high index contrast and ultra-small cross section, provide a very promising way to realize nano-photonic integrated circuits (PICs) with high integration density. Mode conversion and coupling play an important role for realizing various functionality elements. In this paper, a review is given for our recent work including the following parts. First, we discuss the polarizationdependent mode conversion happening in an adiabatic submicron SOI optical waveguide tapers and the application to realize polarization rotation with simple fabrication processes. Second, the mode coupling/conversion in an asymmetrical directional coupler (ADC) is summarized. The application for ADCs includes the realization of ultrasmall polarization-beam splitters (PBS) as well as ultra-compact broadband mode multiplexers (which is very important to enable ultra-high speed optical interconnects with a multimode optical waveguide).

In this work, we show theoretically and confirm experimentally that thin metal membranes patterned with an array of slot dimers (or their Babinet analogue with metal rods) can function as a versatile spectral and polarization filter. We present a detailed covariant multipole theory for the electromagnetic response of an arbitrary dimer based on the Green functions approach. The theory confirms that a great variety of polarization properties, such as birefringence, chirality and elliptical dichroism, can be achieved in a metal layer with such slot-dimer patterning (i.e. in a metasurface). Optical properties of the metasurface can be extensively tuned by varying the geometry (shape and dimensions) of the dimer, for example, by adjusting the sizes and mutual placement of the slots (e.g. inter-slot distance and alignment angle). Three basic shapes of dimers are analyzed: II-shaped (parallel slots), V-shaped, and T-shaped. These particular shapes of dimers are found to be sensitive to variations of the slots lengths and orientation of elements. Theoretical results are well supported by full-wave three-dimensional simulations. Our findings were verified experimentally on the metal membranes fabricated using UV lithography with subsequent Ni growth. Such metasurfaces were characterized using time-domain THz spectroscopy. The samples exhibit pronounced optical activity (500 degrees per wavelength) and high transmission: even though the slots cover only 4.3 % of the total membrane area the amplitude transmission reaches 0.67 at the resonance frequency 0.56 THz.

In this paper, we improve the fabrication procedure of imprint method to obtain low-loss and low-crosstalk GI-core waveguides: particularly focus on eliminating the bridged structure. We experimentally demonstrate that the GI-core waveguides fabricated adopting the modified imprinting method show lower loss and lower crosstalk than SI-core waveguides fabricated using the same material by the same method, even if a slight bridged structure remains. Furthermore, by applying this method to fabricate perpendicularly crossed waveguides, we also demonstrate that GI-core waveguides significantly reduce the loss due to the crossings.

We theoretically estimate the inter-channel crosstalk in densely aligned multimode polymer parallel optical waveguides using a beam propagation method, and compare the results of graded-index (GI)-core waveguides with those of conventional step-index (SI)-core counterpart. In particular, we simulate the crosstalk in bridged core waveguides. Here, the bridged core is sometimes observed in the waveguides fabricated using the imprinting method. The inter-channel crosstalk in SI-core waveguide increases from -25 dB to -4 dB with increasing the bridge thickness. Contrastingly, the worst crosstalk in a GI-core is as low as -15 dB despite the bridged structure as long as the bridge of the core is not included in the index distribution of the GI-core core, namely SI bridged core. In addition, the crosstalk in the GI-core decreases when the multiple cores aligned in parallel have a different structure (core size, refractive index, etc.), because the difference in the core structure makes changes in the distribution of propagation constants, resulting in decreasing the mode coupling efficiency between the two cores. Hence, the worst crosstalk in the GI-core waveguide with a slightly different core structure is as low as -19 dB despite the bridged structure. Thus, the imprinting method should be utilized for GI-core waveguides: the inter-channel crosstalk is un-problematic even if a residual layer remains.