Light as a transmission medium for data communication has proven its success for many years. Over long distances like intercontinental and intercity links, optical interconnects have been the obvious choice due to the low propagation losses and high bandwidth. Over short distances, optical data communication on board has drawn a lot of attention and has proven to bring a solution to the emerging bottle necks of electrical interconnects, but adaption of this technology on a commercial scale has still not been conveyed. As the demonstrated performance of optics on board is higher every year, the adaption in the industry is still predetermined for niche applications. This paper addresses the introduction of on-board optical communication in one of the fastest growing market segments in electronics today: the flexible substrate electronics. By making the optical interconnects and every accompanying active and passive feature very thin and flexible, we can integrate everything in a thin bendable foil. This opens a world of opportunities in portable and wearable applications were flexible electronics are already dominating today. In automotive, avionics, aero-space and medical applications, optical fiber communication and flexible electronics have made their break-through long time ago, mainly driven by the low weight and compactness of both and the immunity to electromagnetic interference and reliability in harsh environments of optics. The flexible optical interconnections presented in this paper combine the electrical and optical network into one system. Light is presented here as the carrier of bit data, but in an analog point of view, light can be used for sensing a wide range of chemical, biological and physical parameters. Optical sensor systems have shown some main advantages over their electrical counterparts with a large growth in this research area as a consequence. As sensors often need to be as compact and unobtrusive as possible, the high level of integration reached in this PhD work can significantly reduce the size, measurement point pitch and cost of existing optical sensors. The integration of opto-electronics, optical waveguides and coupling structures in a 150 μm thin flexible foil presented in this paper can not only find its application in flexible electronics and optical sensors, but brings an additional value to the state-of-the-art of existing on-board rigid interconnects discussed above. Namely, the high level of integration can significantly reduce or even eliminate the footprint of the interconnections, the mechanical flexibility can improve the reliability, the thinness of the foil enables 3D stacking of interconnects and the stand-alone nature of the optical link foil can fasten up the industry acceptance of optics on-board.

Ever growing Internet data traffic leads to a continuously increasing demand in both capacity and performance of large-scale ICT systems such as high-capacity routers and switches, large data centres and supercomputers. Complex and spatially distributed multirack systems comprising a large number of data processing and storage modules with high-speed interfaces have already become reality. A consequence of this trend is that internal interconnection systems also become large and complex. Interconnection distances, total required number of cables and power consumption increase rapidly with the increase in capacity, which can cause limitations in scalability of the whole system. This talk addresses requirements and limitations of intrasystem interconnects for application in large-scale data processing and storage systems. Various point-to-point and optically switched interconnection options are reviewed with regard to their potential to achieve large scalability while reducing the total power consumption.

This will be a high level talk on materials, devices and integration for meeting the system level challenges to bring optics closer to processor chips and their associated uses within data centers. I will present our recent developments and progress with SiPhotonics at Oracle to meet this objective.

GeSn alloys have been predicted to become direct band gap materials when they contain more than 7% Sn, so they are promising candidates as gain materials for group IV laser. Several groups have already measured the photoluminescence (PL) from relaxed GeSn alloys and determined the bowing parameter for the direct band gap. However the energy change of the direct band transition with strain and the change of the indirect band gap have not been determined for GeSn alloys. In our study, we have grown GeSn layers on InGaAs buffer layers on GaAs substrate. By using the InGaAs buffer layers, the strain of the overlying GeSn layers can be controlled. We have grown unstrained GeSn layers and up to 0.5% compressively-strained GeSn layers with two different Sn compositions (3% and 6.4%). Photoreflectance spectroscopy (PR) was used to measure the direct transition energies. Both light hole and heavy hole transitions can be distinguished for GeSn layers under compressive strain. From these measurements we derived the dilational and shear deformation potential constants a and b, for the direct band gap of GeSn, to be -12.8±0.3 eV and -5.4±0.3 eV, respectively. In order to confirm our PR measurements, we also measured PL for the same set of samples. The direct band transition energies derived from PL match those derived from PR. In addition, the indirect band transition can be distinguished. We have done a low temperature PL study of GeSn samples, and determined the change of indirect and direct band energies with temperature. Knowing both the indirect and direct band gap energies, we showed that the cross over point for GeSn alloy to become a direct band gap material is 6% Sn at room temperature.

The control of the spontaneous emission (SE) rate through the photonic band engineering has been extensively studied [1]. For instance, the SE rate from light emitters embedded in a photonic crystal (PC) cavity can be increased or decreased with the control of photon density of states at the frequency of the light emission, that is, the Purcell effect [2]. In this paper, this modification by the Purcell effect was applied to optical nonlinear (ONL) response, change of the effective refractive index, of quantum dots (QDs) embedded in a PC waveguide (WG). There are extra bands in a PC band gap emerged from the PC-WG and frequency regions of high photon density of states in the bands. Threfore, by using this regions in the WG, the ONL response due to photon-excited QDs is expected to be modified as suitable for practical all-optical integrated devices [3]. Multi-stacked layers of InAs-QDs, of which PL emission peak wavelength is approximately 1300nm, were grown by MBE in a GaAs layer. A PC straight waveguide (W1) was fabricated in the GaAs layer with EB-lithography and dry-etching processes. The parameters of photonic crystal, lattice constant (a) and hole radius (r), were varied to make the frequencies of SE enhancement via PC-WG modes correspond the band gap energy of the QD. The SE enhancement was measured by micro-photoluminescence (PL) method and the ONL response in QDs was detected by the two-color pump-probe method [4]; the phase shift value of a probe pulse due to modulations of the effective refractive index in the excited QDs with a pump pulse and the recovery time of the phase shift was observed. As a result, we found the recovery time of the phase shift tends to be shorter while the frequency of the SE enhancement resonates with the gap energy of QDs. This indicates that the recovery time of electrons excited in QDs can be controlled with the photonic band engineering for PC-WG, which is useful for realizing all-optical integrated devices. [1] S. Noda, et al., Nat. Photonics 1, 449 (2007). [2] E. M. Purcell, Phys. Rev. 69, 681 (1946). [3] K. Asakawa et al., New J. Phys. 8, 208 (2006). [4] Y. Kitagawa, et al., JJAP 47, 2893 (2008).

Two-dimensional materials exhibit intriguing optical and electrical transport behaviors, arising from quantum confinement effects. For example, at reduced dimensions, the ballistic conductance of a material is given by MÃ-2e^2/h, where M is the number of available transport modes and 2e^2/h is the conductance quantum. Here, through experiments and theoretical analysis [1][2], we show that the magnitude of light absorption in 2-D semiconductors also follows a universal quantity determined by the integer number of the absorptance quantum, a quantum unit defined as AQ â?^ 8Ï?α/(3nr) ~1.7%, where nr is the refractive index and α is the fine structure constant, which is independent of the material thickness and detailed band-structure. Recently, a new type of 2-D material, free standing InAs nanomembranes (thickness of 3 - 19 nm), as a representative of III-V semiconductors, was realized by layer transfer and this enables optical studies of 2 D InAs which were previously inaccessible, by decoupling those ultra-thin layers from original growth substrates to any optically transparent substrates. By using Fourier Transform Infrared (FTIR) spectroscopy, we observe that the individual steps from interband transitions plateau at ~1.6% for all samples, despite that the thickness is being changed by ~6x. The electron-photon interaction is also studied through Fermiâ?Ts Golden rule, we found that all the materials parameters such as carrier effective masses and bandgap cancel out, leading to a nearly material-independent absorptance of AQ â?^ 8Ï?α/(3nr) for each optical transition step, where nr is between 3 and 4 for most semiconductors in the wavelength range of interest. The work here presents a universal law of absorption for 2-D semiconductors, where the absorptance associated with each absorption step is shown to be independent of the detailed band structure and thickness, and instead being governed by the fundamental physical constant α. References: [1]. K. Takei, H. Fang, S. B. Kumar, R. Kapadia, Q. Gao, M. Madsen, H. S. Kim, C.-H. Liu, Y.-L. Chueh, E. Plis, S. Krishna, H. A. Bechtel, J. Guo, A. Javey. "Quantum Confinement Effects in Nanoscale-Thickness InAs Membranes", Nano Letters, 2011, ASAP. [2]. H. Fang, et al., to be submitted.

Compared to silicon oxide, silicon nitrides (SiNx) and oxinitrides (SiOyNx) matrices are conducting much more favorably the carriers injected through the biased electrodes deposited on both sides of the active layer. The current work presents a comparative study of Er-doped SiNx, Er-doped SiOyNx and Er-doped SiOx of various compositions, with the aim of achieving compact and efficient light sources for CMOS-compatible optical interconnects. The conductivity was found to increase systematically when Si excess and/or nitrogen was (were) increasingly incorporated in SiNx:Er or SiOyNx:Er layers. The transport process was found to be well described by the Poole-Frenkel mechanism, for all compositions. The electroluminescence properties showed similarly a systematic improvement when more and more Si excess and/or N atoms are introduced in the active layers, together with a gradual lowering of the threshold voltage from about 45 V for SiO2 to less than 8V for SiNx.

Si-SiGe heterostructures are of increasing interest in both, opto- and nano-electronics. In telecommunications and optical interconnects Si/SiGe quantum wells are suitable for high-speed photodetectors while, in nano-electronics, as high mobility channel materials, are well accepted solutions to boost performance of complementary metal oxide semiconductor (CMOS) devices. The pseudomorphic growth of SiGe on Si, which induces biaxial compressive strain, indeed increases the hole mobility through strain induced valence band splitting and warping. However, the dopant activation in strained SiGe /Si stacks may show significantly different behavior compared to unstrained materials. Classical high temperature annealing used for dopant activation in bulk Si leads to strong Ge diffusion in SiGe/Si heterostructures. In addition, implantation induced point or complex defects substantially enhance strain relaxation of SiGe/Si heterostructures. Such process related phenomena should therefore be characterized and optimized in order to get the best performance of ultra-thin SiGe/Si devices. Here, we present a systematic investigation on the formation of ultra-shallow p-type junctions in strained Si/strained SiGe heterostructures by B, BF2 and (Si+B) ion implantations and annealing at moderate temperatures. The properties of the doped Si0.50Ge0.50 layers, studied using Raman spectroscopy, RBS, SIMS, TEM and electrical measurements, show a high sensitivity of the crystalline quality and the relaxation behavior of SiGe/Si heterostructures with respect to the ion species, implantation dose and annealing conditions. We show that efficient doping combined with strain conservation and a good single crystalline quality can only be obtained for BF2 implants with 1E15 ions/cm2 and annealed at 650°C. With these parameters, single crystalline layers with negligible strain relaxation and a sheet resistance of 886 ohm/sq were achieved. On the other hand, Si pre-implantation yields the lowest sheet resistances, but at the expense of strong strain relaxation. Our findings do not restrict to device applications, but they contribute to a more general understanding of the physical mechanisms involved in the activation of strained SiGe layers. Finally, the optimized ion implantation/anneal parameters were applied to the fabrication of short channel strained SiGe quantum-well MOSFETs with high-k/metal gates stacks.

In the past decade the CMOS industry faced the challenge of reducing the IC power consumption as the traditional transistor scaling laws required sub-nanometer gate oxide to realize the expected performance improvement from shorter gate length. This led to the introduction of high-k dielectrics and metal gate electrodes since the 45nm technology node. In this decade, a new challenge arises due to the increasing power consumption from the chips I/O connections as the next generations of high performance computing systems will require an aggregated bandwidth in excess of TB/s. Traditionally the optical communication systems have always outperformed the electrical ones when the bandwidth-reach product exceeded 100 Gb/s * m for performance and power consumptions benefits. According to this, future chips optical I/Oâ?Ts should prevail for cm-scale distances. However, a photonic solution will be adopted instead of the traditional short reach electrical links if integration, power and cost requirements are met. The recent progress in silicon photonics makes such technology a promising candidate to meet the stringent requirements of cm-scale optical interconnects. In this paper we will review the state-of-the-art of silicon photonics and the recent results we have achieved co-integrating light modulators, detectors and integrated heaters with high-efficiency fiber couplers. We will discuss the important materials and fabrication processes that will enable CMOS silicon photonics.

High-performance computing (HPC) systems capable of delivering Exaflops performance are envisioned to become a reality by the end of this decade. In order to provide the enormous communication bandwidth that is necessary, hundreds of millions of optical interconnects will have to be deployed to connect together racks, modules and chips. To achieve such massive level of parallelism for HPC systems, monolithic integration of deeply scaled silicon optical circuits into the front-end of standard CMOS process have been demonstrated as a promising technology [1-3]. The monolithic integration of electronic and nanophotonic components was performed at the IBM using 200 mm SOI wafers (SOITEC) having a 220 nm silicon device layer on top of a 2 µm BOX. Several processing modules have been added to a standard CMOS processing flow at the front-end of the line. These modules require a minimal number of additional unique masks and processing steps, while sharing most mask levels and processing steps with the rest of CMOS. For example passive waveguides and electro-optical and thermo-optical modulators share the same silicon device layer with CMOS PFETS and NFETS. The Germanium waveguide photodetectors were fabricated by utilizing a rapid melt growth (RMG) technique wherein the Ge was melted and crystallized during the source-drain anneal step [4-5]. As opposed to traditional approaches where Ge is typically grown by CVD after the source-drain anneal step, the RMG approach enables the sharing of many CMOS steps and mask levels, thus minimizing cost, while yielding a very thin defect-free Ge layer. The metallization was performed using standard CMOS fabrication steps for building tungsten (W) plugs and copper (Cu) interconnect wires. After completion of the monolithic integration with CMOS, Ge photodetector operating at 40Gbps has been demonstrated with bias as small as 1V.

Ge is a good material for Silicon Photonics devices because it is CMOS-compatible and its bandgap energy is 0.80eV, corresponding to 1.55µm wavelength. We especially focus on rectangle-shape Ge stripes which can be used as waveguides of Ge EA modulators. In this paper, we report the shape and the growth rate of selectively grown Ge, and we propose how to apply selective growth to device fabrication. In our experiment we selectively grew Ge stripes on Si substrate with SiO2 mask. We made narrow SiO2 trenches to expose Si. The trench width is 5.0µm~1.0µm. Then we grew 1um-Ge including 100nm buffer layer on the substrate by UHV-CVD. After the growth we observed selectively-grown Ge by SEM, AFM and measured strain in the stripes by Raman microscopy. The shape of 1.5µm and 1.0µm-wide Ge is pentagon. There are three types of side surface, (311) facet, (111) facet, and the one which touches the SiO2 masks. In the last oneâ?Ts case, the shape of side surface is determined by the SiO2 shape. So if we could control the side surface of SiO2, we could control Ge shape. The Ge width and thickness is listed in Table1. In 5.0~4.0µm-wide cases, in which the shape is mesa, the thickness is almost 1µm. But in 3.5~1.0µm cases, in which the shape is triangle or pentagon, the smaller the width is, the thinner the thickness is. We calculated the growth rate on each facet. The growth rate on (100):(311) is 1.0:0.40. In device fabrication, we need 400nmÃ-250nm-rectangle Ge stripes because light can transmit in a single-mode. We made a plan to make rectangle-shape Ge stripes. First we make a 400nm trench on 250nm-SiO2/Si substrate and then we grow Ge. The trench is filled with Ge, but there is extra Ge over SiO2 and the shape isnâ?Tt rectangle. We can remove these extra Ge with chemical mechanical polishing. After these processes we could get rectangle-shape Ge stripes. In conclusion, we observed selectively-grown Germanium and obtained the growth rates on each facet. Then we proposed a plan to make rectangle-shape Ge stripes in order to apply selective growth to device fabrication.

Despite miniaturization and new transistor configurations the semiconductor industry struggles in improving transistor operation frequencies primarily because of interconnect latencies. Optical interconnection is the leading technology to reduce interconnection latencies, but requires efficient, light-on-chip production, propagation, detection and modulation. This paper presents a novel type of waveguide on thin, silicon-on-insulator wafers that incorporates high-quality gain-material using direct-bandgap, III-V nanowires, which have been shown to support lasing. Interestingly, this waveguide confines light primarily in the nanowire region, a relatively low-index region, analogous to a slot waveguide, and thereby provides for a good overlap between the optical mode and the nanowire gain material. Furthermore, these waveguides enable the coupling of thousands of nanowires together into a single device (e.g. laser, modulator detector) to increase its power and efficiency. Possible fabrication processes are discussed. Two waveguide designs are studied and light confinement in both structures is simulated and compared using their electrical field profiles and confinement ratios. Various parameters of the waveguides (e.g. width, height of core, cladding heights) are optimized towards minimizing of the lasing threshold at 1.55 μm, using In(Ga)N nanowires as the gain material. The waveguides are further characterized by presenting their dispersion curves and mode propagation results. We find that good confinement within the core of the waveguide can be achieved and that a wide range of devices can be engineered using this type of waveguide ranging from waveguides to lasers, detectors or modulators.

In the review part we address a number of practically important problems associated with the mechanical behavior, physical (structural) design for reliability (DfR), and structural (mechanical) performance of optical fibers, with an emphasis on fibers of finite length (interconnects), experiencing thermal and/or mechanical loading. The taken approach can be identified as Fiber-Optics Structural Mechanics [1]. The addressed problems include: bending of bare fibers (idealized as a single span beams clamped at the ends and subjected to the ends offset and/or angular misalignments), including fibers under the combined action of bending and tension; role of the nonlinear stress-strain relationship; combined effect of the material nonlinearity and the silica rod nonprismaticity, as is in fused bi-conical taper (FBT) optical couplers; coated fibers, whether polymer coated or metalized, including the effect of coating on bending stresses, coating delamination and strippability, stresses arising during proof (pull-out) testing, and elastic stability and low temperature microbending; fibers soldered into ferrules; opportunities associated with the application of nanotechnologies [2,3]. The extension part has to do with the application of the Probabilistic Design for Reliability (PDfR) [4-6] concept. We account for the fact that some important design parameters (delayed fracture, ends-offset, curvatures, etc.) that play an important role in the functional (optical) and mechanical (â?ophysicalâ?) performance of optical fibers are very seldom known with sufficient certainty and should be treated therefore as random variables. When PDfR concept is applied, the qualification standards should be based on the expected (predicted) and adequate (not necessarily as low as possible) probability of failure in the field. References: 1. E. Suhir, â?oOptical Fiber Interconnects: Design for Reliabilityâ?, Society of Optical Engineers (SPIE), Proc. of SPIE, Vol. 7607 760717-8, 2010 2. D. Ingman and E.Suhir, â?oOptical Fiber with Nano-Particle Overcladâ?, US Patent, #7,162,138 B2, 2007 3. D. Ingman and E.Suhir, â?oOptical Fiber with Nano-Particle Claddingâ?, US Patent, #7,162,137 B2, 2007 4. E.Suhir, â?oApplied Probability for Engineers and Scientistsâ?, McGraw-Hill, 1997 5. E. Suhir, â?oProbabilistic Design for Reliabilityâ?â?, ChipScale Reviews, vol.14, No.6, 2010 6. E.Suhir, R. Mahajan, A. Lucero, L. Bechou, â?oProbabilistic Design for Reliability (PDfR) and a Novel Approach to Qualification Testing (QT)â?, 2012 IEEE/AIAA Aerospace Conf., Big Sky, Montana, to be presented

Along with increasing performance of microelectronic devices, the industry faces growing demand for high speed interconnect solutions on the chip-to-chip and chip-to-fiber level. Bottlenecks for future high-performance computing systems are bandwidth and power consumption, both of which can be addressed by optical transfer lines. Therefore, optical short-range connections between system components are highly desirable and are a rapidly growing field of research. For the realization of these interconnects between highly miniaturized devices and components, new materials and integration concepts are needed for allowing flexible design and low-cost fabrication. A promising example of such a concept is the fabrication of optical waveguides using two-photon polymerization (2PP) triggered by femtosecond laser pulses. This technology was introduced for enabling the generation of arbitrarily-shaped microstructures in polymer photoresists. 3D structuring is made feasible by the strong confinement of the photo polymerization promoted by the well-defined threshold occurring in the underlying two-photon absorption process. Just by moving the focal volume across the polymer, true 3D structures can be obtained. However, this technology, intrinsically predestined for volumetric structures, can be adapted for complex optical waveguide fabrication by synergistically combining its design flexibility with the use of custom-designed multifunctional materials, such as inorganic-organic hybrid polymers (ORMOCER^®s). It will be demonstrated that 2PP structure fabrication in ORMOCER^® systems results in a contrast of refractive index between the hybrid material and the structures exposed to femtosecond laser pulses. With this tool, optical waveguides can be fabricated rapidly and simply without any solvents. Furthermore, since 2PP is free of any constraints regarding structure design, we demonstrate the realization of three-dimensional interconnects using this concept, and analyze light transmission properties with respect to their bend radius and the laser parameters employed for fabrication. In addition to that, the potential of 2PP to be up-scaled from the sub-μm to the cm regime will be discussed with respect to the underlying material concepts and exposure strategies.

In this paper, we studied the optimization of preparation for polymeric optical waveguide based bus structures with embedded 45degree micro-mirrors by metallic hard mold method. The 45degree facets on the metallic hard mold, which were used to create the 45degree micro-mirrors, were studied by the atomic force microscopy (AFM). The surface roughness of the 45degree facets was reduced from 70nm to be 2nm by a photopolymer coating step. High speed test on the waveguide shows the low loss and high Q-factor performance of the waveguide structures. A backplane bus with 10 Gbits/sec.channel will be reported.

Two-photon polymerization (2PP) is an emerging tool in the field of additive manufacturing technologies, which allows for the elegant 3D lithographic production by means of photosensitive resins. One key advantage of 2PP is the achievable feature resolution. A few tens of nanometers are currently the resolution limit for this novel technique. Fields of applications are as diverse as photonics, microfluidics and biomedicine. A challenging photonics application for 2PP are optical interconnects, where optical elements on printed circuit boards are connected with waveguides. The possibility for real 3D structuring allows for easier positioning of the cured structures and straightforward processing outperforming techniques as 2D lithography or reactive ion etching in this regard. If mechanical flexibility of the printed circuit board is required as a property for certain niche applications, polysiloxanes are an interesting class of matrix material. This is also due to their low optical damping behavior and high temperature stability as the material has to withstand temperatures around 250°C during the manufacturing process. In this work, we present our approaches to create polysiloxane-based waveguides via 2PP of specially tailored thiol-ene and acrylic formulations, which are compared with respect to 2PP writing speed and laser power. The advantages of a newly developed class of high performance two-photon-initiators, which play a key role in an efficient structuring process, will be introduced. Latest improvements on the ease of processing and the local refractive index increase are shown as well as the waveguiding performance by means of optical damping measurements. Optical waveguides were successfully created via 2PP with writing speeds around 4 mm/min. Preliminary optical damping measurements for the waveguide materials exhibit values in the range of 0.4 to 2 dB/cm at 850 nm.

A low-power WDM photonic link is the key to achieving energy-efficient and high-bandwidth interconnects for macrochip and other system architectures, and such a link must be built around a low-power WDM transceiver. In this paper, we present our hybrid integration technology platform that allows the intimate integration of best-in-breed silicon photonic components and CMOS VLSI chips. Using this integration scheme, we also demonstrate arrayed and optimized silicon photonics devices built on the Luxtera/Freescale 130 nm SOI CMOS platform, hybrid-integrated with high-speed circuits built on the TSMC 40 nm bulk CMOS process, to develop an ultra-efficient 80 Gbps WDM CMOS photonic transceiver.

WDM is a standard technique used to enlarge the bandwidth of a transmission channel through the simultaneous transmission of different signals encoded in the same path. In optical communication systems in the infrared window the devices used perform the mux and demux operations usually employ optical filters based on prisms, interference filters, diffraction gratings or arrayed waveguide grating (AWG). For the visible range the shift on the wavelength demands the conversion of such devices or alternatively the use of novel approaches to perform the same operation. Optical communications in the visible range have important applications the field dominated by short range communications, such as, home networks, automotive industry of traffic control applications as well as industrial purposes. In this paper we report the optimization of a device based on p-i'(a-SiC:H)-n/p-i(a-Si:H)-n heterostructure optimized for the detection of the short and long wavelengths in the visible range. The thickness and the absorption coefficient of the i'- and i- layers were tailored for short and long wavelengths optical confinement, respectively in the front and back photodiodes acting both as optical filters. To simulate the optical signals a chromatic time dependent combination of short, medium and long wavelengths was applied to the device. The generated transient photocurrent was measured, under reverse and forward bias (-10V

Systems that transmit, receive, and/or process intelligence require some sort of modulation, which is the deliberate distortion of a carrier to impress intelligence (data) upon it, that subsequently allows the information recovery. Data transmission can be improved using the wavelength division multiplexing-demultiplexing technique. The plastic optical fiber is a suitable, promising solution as transmission medium for short range communications. It is easier to terminate, polish, and connect as well, which reduce the cost of installation and maintenance. So, the conception of new devices based on new materials for signal (de)multiplexing in the visible spectrum is a demand. Such technological innovation can influence economic transformation, create value in the society and ultimately improve individuals' lives. There has been much research on semiconductor devices as elements for optical communication when a band or frequency needs to be filtered from a wider range of mixed signals. Active filter circuits can be designed. Amorphous silicon carbon tandem structures, through an adequate engineering design can accomplish this function. Here the manipulation of the magnitude is achieved by changing the wavelength (color channels) of the modulated lights and its frequency under appropriated wavelength backgrounds. Combined tunable WDM converters based on SiC multilayer photonic active filters are analyzed. The operation combines the properties of active high-pass and low-pass filter sections into a capacitive active band-pass filter. The sensor element is a multilayered heterostructure produced by PE-CVD at 13.56 MHz radio frequency. The configuration includes two stacked p-i-n structures (p(a-SiC:H)-í'(a-SiC:H)-n(a-SiC:H)-p(a-SiC:H)-i(a-Si:H)-n(a-Si:H)) sandwiched between two transparent contacts. Transfer function characteristics are studied both theoretically and experimentally. A capacitive active band-pass filter model supports the experimental data. An algorithm to decode the multiplex signal is established. Results show that the light-activated photonic device combines the demultiplexing operation with the simultaneous photodetection and self amplification of an optical signal. The output waveform presents a nonlinear amplitude-dependent response to the wavelengths of the input channels. Depending on the wavelength of the external background it acts either as a short- or a long- pass band filter or as a band-stop filter. A two stage active circuit is presented and gives insight into the physics of the device. The device, modeled by a simple circuit with variable capacitors and interconnected phototransistors through a resistor is a current-control device. It uses a variable capacitance to control the power delivered to the load acting as a state variable filter circuit. It combines the properties of active high-pass and low-pass filter sections into a capacitive active band-rejection filter.

Controllable slow and fast light propagation is important for telecommunication and has been attracted great interests in last 10 years. CdSe quantum dot which has a lifetime of few tens of nanoseconds is an excellent system for light velocity manipulation of ns pulsed laser. In this work, we demonstrated controllable switch between slow and fast light propagation of a 6 ns laser at 532 nm in CdSe quantum dots. Standard pump-probe geometry was used to investigate the group velocity of a weak signal pulse. The light source was a 6 ns laser operated at 532 nm with 10 Hz repetition rate. Energy transfer from strong pump beam to weak signal beam was observed due to the laser excitation-induced dynamic processes in CdSe system. This time-dependent energy transfer process was used to control the pulse shape and group velocity of the signal beam. A slow light propagation was obtained when the trailing edge of the signal beam was amplified more while a fast light propagation was obtained when its leading edge experienced a larger amplification. Strong nonlinear optical absorption was also observed in nanosecond scale in these nanosized particles. The rules of both statistic nonlinear absorption and dynamic laser excitation-induced nonlinear gain in light propagation will be presented. Possible applications of the observed phenomena will be discussed.

Optical fibers have been successfully used in long-haul communication systems, endoscopy, and other optical systems to transmit optical power as well as information. In integrated sensor systems, optical fibers have been frequently employed to connect the source and the detector, due to their flexibility, compactness, and low loss. However, optical fibers can provide more functions than a simple transmission channel. In this talk, we review our work on various optical fibers as a platform for molecular sensors based surface enhanced Raman scattering (SERS). The fibers serve to significantly increase the sensitivity of SERS and to facilitate the integration of a compact sensor system. The demand for sensors for detecting chemical and biological agents is greater than ever before, including medical, environmental, food safety, military, and security applications. At present, most detection or sensing techniques tend to be either non-molecular specific, bulky, expensive, relatively inaccurate, or unable to provide real time data. Clearly, alternative sensing technologies are urgently needed. Recently, we have been working to develop a compact fiber optic SERS sensor system that integrates various novel ideas to achieve compactness, high sensitivity and consistency, molecular specificity, and automatic preliminary identification capabilities. The unique sensor architecture is expected to bring SERS sensors to practical applications due to a combination of 1) novel SERS substrates that provide the high sensitivity and consistency, molecular specificity, and applicability to a wide range of compounds; 2) unique fiber probe configurations such as a unique liquid core optical fiber probe and a double SERS substrate structure that provide the compactness, reliability, low cost, ease of sampling, and most importantly sensitivity enhancement. Integration of such fiber probes with a portable Raman spectrometer brings the SERS detection one step closer to practical applications. In this talk, we will discuss the principle of operation of various building blocks, demonstrate our recent results, and highlight some potential applications.

We address a number of practically important problems associated with the mechanical behavior, physical (structural) design for reliability (DfR), and structural (mechanical) performance of optical fibers of finite length, such as optical fiber interconnects, experiencing thermal, mechanical or dynamic loading. The taken approach can be identified as Fiber-Optics Structural Mechanics. This area of mechanical-and-materials engineering, on one hand, and photonics (fiber optics) engineering, on the other, deals with the application of methods and approaches of Engineering Mechanics and particularly Engineering Mechanics of Beams to the stress-strain analysis, rational mechanical (â?ophysicalâ?) design and reliability of fiber-optics structures and systems. Similar to the other areas of structural engineering, Fiber-Optics Structural Mechanics considers the attributes associated with the properties of the materials used, structural elements employed and loads applied. An optical fiber system is treated as a structure, in which the size and configuration of the constituent structural components, materialsâ?T interaction, and the applied or developed loads are as important as the materials properties. Its objective is to determine the loading conditions (which can be due to the thermal expansion mismatch of dissimilar materials, lateral and angular misalignments, test loads, dynamic loading due to shocks and vibrations, etc), evaluate stresses and strains, and to ensure that the predicted responses are acceptable from the standpoint of structural integrity, elastic stability, dependability, and normal operation (both mechanical and optical) of the system. The addressed problems include: bending of bare fibers idealized as a single span beams clamped at the ends and subjected to the ends offset and/or angular misalignments; bare fibers under the combined action of bending and tension; the role of the nonlinear stress-strain relationship for the silica material subjected to tension or compression;large deformations nonlinearity is discussed and evaluated for fibers subjected to large tensile deformations; combined effect of the silica material nonlinearity and the rod nonprismaticity can be determined, using an example of a fused biconical taper (FBT) optical coupler; problems encountered during design, manufacturing, testing, and reliability assessments for dual-coated glass fibers (the effect of coating on the bending stresses; fiber-coating delamination and strippability, prediction of the magnitude and distribution of stresses during proof (pull-out) testing, etc.); interaction of â?oglobalâ? and â?olocalâ? thermally induced stresses in application to optical glass fibers adhesively bonded at the ends; elastic stability and microbending of optical fibers is important primarily in connection with the added transmission losses associated with these phenomena; specific requirements for the solder materials and joints used in photonics: ability to achieve high alignment, requirement for a very low creep, etc.; thermally induced stresses in optical fibers soldered into various ferrules; dynamic response to shocks and vibrations.

During the last two decades, nano-materials have been intensively investigated due to their wide range of properties, resulting in a variety of applications. In order to serve as advanced packaging material, from an industrial point of view emphasis has also to be on cost reduction either for the materials, the processes, or for both. Materials are searched for which enable processing and integration from a nm up to a cm scale. A particular class of low-cost nanoscale material which fulfill this requirement are inorganic-organic hybrid polymers (ORMOCER®s) which are synthesized by catalytically controlled hydrolysis/ polycondensation reactions, resulting in storage-stable resins. Due to the variety of chemical and physical parameters, the material and processing properties which directly influence the resulting structure and thus the physical properties, can be varied over wide ranges. Upon synthesis, functional organic groups are introduced into the material which allows one to photochemically pattern the resins. The materials are capable to be patterned on a nm up to a cm scale, employing a variety of different micro- and nanopatterning methods such as, UV lithography /replication, nano-imprint lithography, printing, laser-direct writing, or two-photon polymerization, in order to generate micro- and nano-optical components. While for most of the techniques the patterning has to be repeated several times in order to achieve multi-functional layers, the latter method allows one to directly write arbitrary 3D structures into the hybrid polymer material. The combination of chemically designed low-cost materials with tunable material parameters such as low optical absorption, tunable refractive index, good processability, and high chemical, thermal and mechanical stability, is very attractive for (integrated) optical applications.

Scaling of chipscale electronic and photonic devices to smaller and smaller dimensions has enabled faster, more power-efficient and inexpensive components. It has also brought about a myriad of new challenges. One very important challenge is the growing size mismatch between electronic and photonic components. Plasmonics can form a bridge between the worlds of nano-electronics and micro-photonics. In doing so, it offers new ways to realize power-efficient, electrically-controlled devices that can emit, switch, and detect photons that can easily be integrated with electronic components.

Surface plasmon polaritons (SPPs) supported by metal/dielectric interface is considered as the promising candidate for on-chip communication due to high speed and its extremely deep subwavelength mode volume. Previous studies have shown electrically driven integrated light source for SPPs with organic LEDs, Si nanocrystals and etc., but due to bulky size of emission region, routing of SPPs on chip is not possible due to lack of 3D confinement of light. Hereby, we have demonstrated by using a coreshell GaAs nanowire with radial p-i-n junction, a electrically driven point source could be acquired. Emission from such NW LED point source could naturally couple to SPPs supported by metallic waveguides with subwavelength dimensions to achieve point-to-point in-plane guiding of light. And we have shown that emission coupled to SPPs could be routed along complex guiding elements such as 90-degree bend and Y-shaped splitters with compact metallic waveguides. Such integrated surface plasmon source/guide could be used as building blocks for high-speed and ultra-compact optoelectronic circuitry with a lot of flexibility to arbitrarily control the flow of light on a chip.

Si photonics is the most practical approach, enabling us to overcome the substantial limitations in electronics. A photonic crystal (PhC), which allows us to manipulate light of interest with perfect control, will be an indispensable tool in Si photonics. However, resonant wavelength in a PhC cavity is easily influenced by temperature fluctuation existed on a Si LSI chip. Such instability of resonant wavelength must impede WDM systems, and therefore make their practical implementation challenging. In this paper, we present the result of tuning the cavity mode wavelength in a 2D PhC cavity using a Si beam structure. Experimentally, ~2 nm blue shift of the peak associated with cavity mode is obtained under bending of the Si beam. A Si beam is a mechanical structure which is anchored only at one end and can be deformed by applying an external load at the other free end. Recently our group has demonstrated that using such a Si beam, elastically tensile strained Si exhibits a relatively large band gap shrinkage. Here we apply this technique to a 2D PhC cavity: we implement a 2D PhC slab structure in Si bent beam. The design of PhC cavity consists of three-missing holes (L3) in Si slab with a triangular lattice with hole radius of 114 nm and lattice constant of 380 nm. The 2D PhC cavity on Si beam was fabricated from SOI wafers (Si: 250 nm, BOX: 3 μm). Using EB lithography followed by a RIE of patterned Si and a wet under-etching of BOX layer in HF acid, the Si beam structure with L3 PhC cavities was obtained, and the beam size is 5 μm in width and 15 μm in length. μ-PL measurement under bending the beam was performed for fabricated samples at room temperature. The samples were pumped by a 457 nm laser with the power ~2 mW, which is low enough to ignore heating effects on PL spectra. Then the PL signals were vertically collected into InGaAs photodetector through a 100x objective lens. For bending, a mechanical load was applied by using a tip with the radius of ~2 μm to push the PhC beamâ?Ts end 3 μm downward to the Si substrate. As for the result of μ-PL spectra, several sharp peaks were observed, corresponding to the PhC cavity mode. We focused on the peak around 1260 nm, which has the narrowest linewidth. The peak blue shift of ~ 2 nm was observed when the Si PhC beam was bent. Once it was released from bending, the peak position returned to the original again. The blue shift of the peak is explained as follows. When the beam is bent, the small L3 defect area is bent as well. The localized light in the defect should experience bends, resulting in reducing effective index of the cavity mode as well as conceivable increase in radiation loss. Thus the cavity mode wavelength is blue shifted. To conclude, we showed the experimental tuning of cavity mode wavelength in 2D PhC using a bent Si beam structure by means of μ-PL measurement. This tuning method can be implemented in WDM system to control the cavity mode wavelength in 2D PhC slabs.

Lead zirconate titanate, Pb(Zr,Ti)O3(PZT) films on Si substrates were deposited by a sol-gel method The crack-free PZT films were achieved with ITO buffer layer. The ITO layer (In/Sn=95/5) was deposited on SiO2/Si substrate by a spin coating and annealed at 650 oC for 10 min. The resistivity of 1.7Ã-10^-4 ohm cm and crack-free surface of ITO film were obtained. The PZT films were deposited by a spin coating on the ITO buffer layer and annealed at 600 oC for 10 min. The molar ratio of Pb/Zr/Ti was 110/52/48. The crystalline quality and ferroelectric properties of PZT/ITO/SiO2/Si were measured. The crack-free PZT films were obtained on ITO/SiO2/Si substrate. The XRD pattern of PZT films with ITO buffer layer on silicon substrate shows the single perovskite phase. The polarization-electric field (P-E) hysteresis loop was observed from this PZT film. The coercive field (Ec) and remnant polarization (Pr) of this film were 148kV/cm and 50 uC/cm^2, respectively. The coercive field value of PZT film with ITO buffer layer was three times higher than that of the PZT films on Pt electrode. The difference is caused by the crystalline quality of the base layer (Pt or ITO). This film can be expected to realize the high-speed optical modulator on silicon substrate.

To assist the precision and stability of wavelength at 1550 nm in planar optical waveguides, hybrid semiconductor-metal corrugated gratings with nanometer periodicity are integrated into silicon-based optical interconnects. By using shiny metals, the modes can be well confined in the waveguides. This work constructs a multi-parametric optical waveguide model to compute the mode-coupling coefficient. The photonic method and the optical method are used to compute coupling coefficients. Numerical results demonstrate how metal grating materials and structures, and semiconductor layer thickness affect the coupling coefficients. The improper design of semiconductor layer thickness can cause mode-coupling to drop significantly or even to vanish. Physical interpretations can provide insights into the design and modeling of silicon-based optical interconnects. Future work and applications will be proposed.