The wavelength spectra of ridge waveguide Fabry Perot lasers can be modified by perturbing the effective refractive index of the guided mode along very small sections of the laser cavity. One way of locally perturbing the effective index of the lasing mode is by etching features into the ridge waveguide such that each feature has a small overlap with the transverse field profile of the unperturbed mode, consequently most of the light in the laser cavity is unaffected by these perturbations. A proportion of the propagating light is however reflected at the boundaries between the perturbed and the unperturbed sections. Suitable positioning of these interfaces allows the mirror loss spectrum of a Fabry Perot laser to be manipulated. In order to achieve single longitudinal mode emission, the mirror loss of a specified mode must be reduced below that of the other cavity modes. Here we briefly review one procedure for calculating the mirror loss spectra of devices containing such features. We then go on to describe a method for synthesising onedimensional slot patterns. This technique allows the lasers emission wavelength to be specified to a high degree of accuracy.

Tailored scaling allows the effectiveness of physical effects and mechanical stability to be enhanced. This is shown for micromachined 1.55μm vertical-resonator-based filters and VCSELs, capable of wide, continuous, and kink-free tuning by a single control parameter. Tuning is achieved by mechanically actuating one or several membranes in a vertical air-gap resonator including two highly reflective DBR mirrors. Electrostatically actuatable single-chip filters including InP/air-gap DBR's (3.5 periods) reveal a continuous tuning up to 14% of the absolute wavelength. Varying a reverse voltage (U=0 .. -3.2V) between the membranes (almost flat in the unactuated condition) a tuning range up to 142nm was obtained. Varying a reverse voltage (U=0 .. -28V) between the membranes (strained and curved in the unactuated condition) a tuning range up to 221nm was obtained. Optically pumped and continuously tunable 1.55μm VCSELs show 26nm spectral tuning range, 400μW maximum output power, and 57dBm SMSR. This two-chip VCSEL has a movable top mirror membrane, which is precisely designed to obtain a specific air-gap length and a tailored radius of curvature in order to efficiently support the fundamental optical mode of the plane-concave resonator. The curved top mirror DBR membrane consists of periodically alternating differently stressed silicon nitride and silicon dioxide multilayers. The lower InP-based part consists of the InP/GaInAsP bottom DBR and the GaInAsP active region.

γ-CuCl is a wide-bandgap (E_g = 3.395eV), direct bandgap, semiconductor material with a cubic zincblende lattice structure. Its lattice constant, a_CuCl = 0.541 nm, means that the lattice mismatch to Si (a_Si = 0.543 nm) is <0.5%. γ-CuCl on Si-the growth of a wide-bandgap, direct bandgap, optoelectronics material on silicon substrates is a novel material system, with compatibility to current Si based electronic/optoelectronics technologies. The authors report on early investigations consisting of the growth of polycrystalline, CuCl thin films on Si (100), Si (111), and quartz substrates by physical vapour deposition. X-ray diffraction (XRD) studies indicate that CuCl grows preferentially in the <111> direction. Photoluminescence (PL) and Cathodoluminescence (CL) reveal a strong room temperature Z3 excitonic emission at ~387nm. A demonstration electroluminescent device (ELD) structure based on the deposition of CuCl on Si was developed. Preliminary electroluminescence measurements confirm UV light emission at wavelengths of ~380nm and ~387nm, due to excitonic behaviour. A further emission occurs in the bandgap region at ~360nm.

We present work on design of monolithic mode-locked semiconductor lasers with focus on the gain medium. The use of highly inverted quantum wells in a low-loss waveguide enables both low quantum noise, low-chirped pulses and a large stability region. Broadband noise measurements are performed and used to confirm the design principles.

Using near-field scanning optical microscopy and ultrafast laser spectroscopy, we study the linear optical properties of subwavelength nanoslit and nanohole arrays in metal films, which are prototype structures for novel plasmonic crystals. Near-field microscopy provides direct evidence for surface plasmon polariton (SPP) excitation and allows for spatial imaging of the corresponding SPP modes. By employing spectral interferometry with ultrashort 11-fs light pulses, we directly reconstruct the temporal structure of the electric field of these pulses as they are transmitted through the metallic nanostructures. The analysis of these data allows for a quantitative extraction of the plasmonic band structure and the radiative damping of the corresponding SPP modes. Clear evidence for plasmonic band gap formation is given. Our results reveal that the coherent coupling between different SPP modes can result in a pronounced suppression of radiative SPP damping, increasing the SPP lifetime from 30 fs to more than 200 fs. These findings are relevant for optimizing and manipulating the optical properties of novel nano-plasmonic devices.

The enhanced transmission of square arrays of nanoholes in thin and thick metal films has been studied. We show that normal incidence transmission spectra of an array of elliptical nanoholes in a 220 nm thick gold films have reduced symmetry with respect to the four-fold symmetry found in an equivalent array of circular nanoholes. Elliptical nanoholes milled in a 40 nm thick gold film show complex oscillatory behaviour of the transmission spectrum that has properties similar to those of a two-dimensional birefingent crystal. The transmission spectrum may also be controlled by polarisation selection due to the different degrees of the elliptical polarisation of the transmitted light. The enhanced transmission through 1D arrays of stripes is studied for a range of incident angles with a polarisation perpendicular to the stripe length. Increasing the incident angle increases the number of observed peaks, and changes their spectral positions. Changing the polarisation or the angle of incidence in a 1D array of stripes or a 2D array of reduced symmetry motifs allows control of the enhanced transmission spectrum and shows great potential for numerous applications in photonic and opto-electronic devices.

Photonics is far behind electronics in maturity. The devices are orders of magnitude larger than their electronics counterparts and the functionality is low. But it appears that these issues are not fundamentally impossible to solve. In this paper some of the emerging possibilities to overcome the limitations mentioned above are briefly treated, and we discuss the utilization of these comparatively new phenomena to widen the application envelope of photonics technology to generate functions not normally associated with photonics. These developments could lead to quantum leaps in photonics devices, to complement the forceful engineering improvements. Examples of such potential candidate research fields for quantum leaps are: Coherent light matter interaction, plasmonics, quantum information and communications, photonic crystals, intersubband based devices. The list is by no means exhaustive. This paper will concentrate on coherent light matter interaction, plasmonics and photonic crystals.

We present a detailed study of the localized coupled-cavity modes in a photonic molecule formed from two dielectric spherical microcavities with CdTe nanocrystals, which show a multi-peak narrowband modal structure resulting from lifting of the mode degeneracy with respect to the azimuthal quantum number. The feasibility of photonic molecules as the basis for a multi-channel, wavelength-tunable optical delay device is analysed.

Experimental and theoretical studies of the emission directionality diagrams of a perylene dye covering the inner surface of three-dimensional opal-based photonic crystals with incomplete photonic bandgap are reported. Directionality diagram of emission intensity is interpreted in terms of the spontaneous emission suppression by photonic band gap and the emission enhancement due to photon focusing phenomenon. A theoretical model is based on the classical analysis of an angular distribution of the radiated power of a point dipole.

Reflection spectra and photonic band gaps (PBGs) for periodic structures consisting of grooved Si infiltrated with nematic liquid crystals (LCs) E7 have been obtained experimentally and by simulation. Periodically grooved Si matrixes, with lattice periods ranging from 2 to 6 μm, were fabricated using wet anisotropic etching of (110) Si in an alkaline solution. It is shown theoretically that a substantial shift of PBGs can be obtained when there is a homogeneous alignment of the liquid crystal molecules with respect to the Si walls and, therefore, a change in the LC's refractive index from n_o to n_e during elctro-tuning can be expected. This effect is smaller during thermo-tuning effect when refractive index changed from n_o to, n_i (isotropic LC phase). The simulated reflection spectra are in good agreement with experimental data obtained with a Digilab FTS 6000 FTIR spectrometer in conjunction with a UMA 500 infrared microscope. A shift of the PBG's edges by a factor Δλ/λ=7% was predicted as a result of a thermo-tuning effect and ~10% as a results of electro-tuning effect.

We present a calibrated full-3D simulation of a widely-tunable sampled-grating distributed Bragg reflector (SGDBR) laser showing the characteristic quasi-continuous tuning map. The SGDBR laser is a longitudinally integrated device consisting of five waveguide sections: a front and rear mirror section together with a phase section allow for quasi-continuous tuning over a wavelength range of 100nm, while an active section provides the optical gain for the laser operation. For real world applications the tuning behavior needs to be well understood in order to guarantee stable operation for each wavelength channel. Due to the strong inhomogeneities both in the transverse and longitudinal dimensions a 3D simulation model is necessary to cover the full complexity of such devices. In our physics-based approach, we solve the fully coupled semiconductor drift-diffusion equations for electrons and holes, taking into account longitudinal current flux in full 3D. Gain calculation and the photon rate equation are included self-consistently in an iterative Newton scheme. The optical field is composed of several transverse mode patterns combined with the longitudinal field distribution as obtained by a transfer matrix formalism. By means of a Gummel-type iteration scheme a self-consistent solution of the optics and electronics is found. We show that this approach succeeds even in the numerically most challenging case of discrete wavelength jumps as observed in typical tuning maps of SGDBR lasers. Our simulations are in good agreement with measurements and prove the suitability of the simulator for the design and optimization of state-of-the-art tunable lasers.

All-optical regeneration at 40 Gbit/s and beyond appears to be a crucial element for future transparent networks. One solution to achieve the regeneration is an all-optical clock recovery element combined with a Mach-Zehnder interferometer. Among the different approaches investigated so far, a scheme based on a single self-pulsating distributed Bragg reflector laser is of particular interest from practical and cost viewpoints. In this structure at least two longitudinal modes beat together, generating power oscillation at 40-GHz even though the laser is DC-biased. The 40-GHz signal frequency is fixed by the free spectral range of the cavity, and in the case of the self-pulsation, it has been demonstrated that its linewidth is smaller than the sum of the linewidth of the lasing modes. It is believed that the signal benefits from the phase correlation of the optical modes through the interband four-wave mixing (FWM) non-linearity. The FWM results as a modulation of the carrier population, leading to a non-linear gain and refractive index modulation, affecting both the amplitude and the phase of the cavity modes. Based on a four-wave mixing formalism a theoretical model has been developed and it corroborates the experimental linewidth measurement.

Excitonic and spin excitations of single semiconductor quantum dots currently attract attention as possible candidates for solid state based implementations of quantum logic devices. Due to their rather short decoherence times in the picosecond to nanosecond range, such implementations rely on using ultrafast optical pulses to probe and control coherent polarizations. In this article, we review our recent work on combining ultrafast spectroscopy and near-field microscopy to probe the nonlinear optical response of a single quantum dot and of a pair of dipolecoupled quantum dots on a femtosecond time scale. We demonstrate coherent control of both amplitude and phase of the coherent quantum dot polarization by studying Rabi oscillations and the optical Stark effect in an individual dot. By probing Rabi oscillations in a pair of dots, we identify couplings between permanent excitonic dipole moments. Our results show that although semiconductor quantum dots resemble in many respects atomic systems, Coulomb many-body interactions can contribute significantly to their optical nonlinearities on ultrashort time scales. This paves the way towards the realization of potentially scalable nonlocal quantum gates in chains of dipole-coupled dots, but also means that decoherence phenomena induced by many-body interactions need to be carefully controlled.

We present recent results obtained for an optical chaos communications system using optoelectronic devices. These devices are used to improve privacy and security in information transmission. Two kind of chaos based emitters and receivers are described: a semiconductor laser subject to all-optical feedback and operating in a non-linear regime and a semiconductor laser subject to non-linear electro-optical feedback and operating in a linear regime. We show that both configuration give very good synchronization properties and are suitable for message enconding/decoding at bit rates as high as Gbit/s.

We analyse the stochastic polarization fluctuations in a vertical cavity surface emitting laser (VCSEL) under the influence of electro-optical feedback and show that the dynamics can be modeled as a bistable system with time-delayed memory. Assuming an asymmetric potential, we show the existence of a regime in which the systems dynamic displays excitability. We calculate the relevant residence time distributions and correlation times and compare our system to a well known discrete model for excitability. Finally, we present experimental data that demonstrates excitable behaviour in the polarization dynamics of a VCSEL and, in particular, show the appearance of coherence resonance.

Device characteristics of photonic crystal lasers formed in InGaAsP membranes bonded to a sapphire substrate are discussed. Also discussed are waveguide loss mechanisms in type-A and type-B photonic crystal waveguides and the transmission properties of photonic crystal waveguide bends.

Nonlinear polarization rotation in semiconductor optical amplifiers has been the focus of a lot of work in the past decade. A lot of research has been devoted to this phenomenon due to its possible use in all-optical switching. It has been mentioned as a possible competitor to such established switching techniques as cross-gain modulation, cross-phase modulation and four-wave mixing. The speed at which the switching can be performed is determined by the gain dynamics in the device. So far the majority of the work has focused on switching due to the relatively slow carrier density recombination, which limits the switching to the order of tens of gigahertz. If the polarization dependence of ultrafast gain mechanisms such as carrier heating and spectral hole burning can be identified and measured then there is the possibility to increase the switching speed obtainable using this process into the terahertz range. In order to further the understanding of the polarization dependence of the gain of a bulk SOA under tensile strain and to determine the plausibility of ultrafast all-optical switching using nonlinear polarization rotation an experiment is presented based on a four-wave mixing technique.

A reflective-type liquid crystal (LC) wavefront corrector with modal addressing is described. The corrector's backplane has an array of pixel electrodes interconnected by a network of discrete resistors. The resistive network serves to form the local voltage profile that controls the phase distribution generated in the liquid crystal layer. This design is realized in a bipolar silicon technology. Preliminary numerical analysis is presented; technology and experimental results are discussed.

A complete design of 1.3 μm AlGaInAs/InP narrow stripe semiconductor lasers for self-pulsating operation is realised by using a 2×1D simulation model. This numerical model is based on the effective index method and self-sustained pulsation mechanism in the narrow stripe lasers. The self-pulsation effect is enhanced by the self focusing and defocusing of the optical field which is dependent on the modification of carrier densities in the active region. The resulting AlGaInAs-InP device with compressively strained multi-QWs showed self-pulsation frequency of 3.5 GHz.

The purpose of this article is to present a brief review of supercontinuum generation in photonic crystal fibre. We focus particular attention on the most-commonly encountered femtosecond regime, where we provide an overview of the underlying spectral broadening mechanisms, and discuss the associated field stability properties.

The inscription of fibre Bragg gratings using infrared femtosecond laser offers a number of advantages over conventional methods based on UV inscription. The refractive index modification in femtosecond inscription is independent on defect formation and therefore should not experiment the defect-related thermal decay of UV inscribed gratings. In this paper, the response to thermal annealing of fiber Bragg gratings inscribed using a tightly focussed femtosecond laser is investigated. Experimental results reveal a vastly improve thermal stability compared to gratings inscribed using conventional methods based on UV light. Erasure was not observed until temperatures in the range between 900°C and 1000°C. These devices are therefore particularly suited to work in hostile environments and as high temperature sensors.

We have demonstrated the integration of supporting lasing at 1.88 mm for enhancing the gain in a thulium doped ZBLAN amplifier. The laser ring cavity has been created in order to reduce the bottleneck effect caused by the self-terminating nature of the amplifying transition. Two different types of fibre geometry and pumping scheme have been tested and the performance has been compared with a numerical model of the amplifier.

In this paper we demonstrate the potential of a numerical algorithm that describes intersubband optical spectra by combining many body effects and nonparabolicity characterized by strong k-dependence of both dispersion relations and dipole moments. Applications of our numerical scheme are given for highly strained III-V quantum wells. Preliminary results results for dilute nitride systems are also discussed.

In the classic method for calculation of energy-band structure of photonic crystal the electromagnetic field of wave that propagates in crystal is expanded in plane waves. The determination task of the allowed frequencies at certain wave vector in the first Brillouin zone leads to the finding of eigenvalues and eigenvectors of the corresponding matrix equation with dimension, which provides the necessary precision of determination of the allowed frequencies. Because of slow convergence of expansion in plane waves, it is necessary to solve the matrix equation of high dimension that requires considerable computation time. The proposed method of determination of the allowed frequencies is based on the coupled wave method (CWM) at the corresponding definition of boundary periodic conditions. Especially, this method is simple for 1D photonic crystals, that comes to the eigenvalue problem of the matrix equation of 2x2 dimension. For 2D photonic crystals the solution also leads to the eigenvalue problem, but of 2Nx2N matrix, where N is number of the coupled waves. Frequency will be allowed, if absolute value of eigenvalue is equal to one. In the proposed method the 2Nx2N dimension of the matrix equation is equivalent to the dimension of N²xN² of classic method. The stable S-algorithm of computation is developed. The computations of energy-band structure of 1D and 2D photonic crystals of the simplest structure are conducted. The dependencies of computation precision on the number of coupled waves at change of N from 1 to 29 are obtained.

The light diffraction on dielectric grating placed on metallic substrate by the method of coupled waves (CWM) taking into account the 53 diffraction orders is analysed. The modified stable S-algorithm is used for diffraction analysis. At certain parameters of grating the resonance of coupled waves (CWR), which consists in sharp increase of electromagnetic field into grating, appears. This increase of electromagnetic field results in full absorption of wave energy in metallic substrate. The behaviour of CWR depends on modulation value of permittivity of grating material. The approximate grating parameters at the condition of appearance of waveguide effect corresponding to CWR are found. These parameters are improved for achievement of full absorption by the picking-up method. The grating parameters on platinum, potassium, gold, silver for wavelengths of 0.85 μm and 1.5 μm at condition, that reflection coefficient less that 0.0001, by CWM are found. For Pt and K the dependencies of reflection coefficient on the grating thickness for TE and TM polarisation are obtained. For all metals dependencies of reflectance on wavelength, which it is possible rather precisely to describe by analytical equations with a Laurence's function, are calculated. Especially, analytical dependence precisely coincides with a curve, which by CWM for grating with small permittivity modulation of grating material is obtained. For potassium and silver the dependence of reflectance on light incidence on grating is obtained. The analysis of resonance absorption effect for TE and TM polarisation is conducted.

A contra-propagation set up is implemented and dynamic pump probe studies of a InGaAsP/InP SOA in the gain regime are undertaken using pulses of 2ps duration. The time resolved amplified probe signal is measured separately for the TE and TM modes of the semiconductor optical amplifier. Different behaviours are observed both in the gain compression and the timescales of the effect, with the TM mode displaying a faster component and a higher gain compression.

Torsion micro-mirror is the key structure of MEMS optical devices such as MEMS optical switches, MEMS variable optical attenuator, MEMS scanning micro-mirror array and so on. In this paper, after testing a silicon-based non-silicon micro-mirror fabricated by bulk micromachining, some important measurement data has been achieved. It is clear that this silicon-based non-silicon micro-mirror scheme can has just less than 15 degree rotation at 20V driving voltage while the beam thickness is just 0.5 m m . In order to improving the stability, reliability and optical properties of the micro-mirror, a new scheme has been put forward and the fabricating process using surface-micromachining has been simulated. It is shown that the new scheme will improve the characteristics of the micro-mirror effectively.

The structure of the opal-ZnO composites has been studied by TEM and X-ray methods. It was found that the solid state reaction of the opal-ZnO interface interaction is occurring during the heat treatment of the infiltrated samples resulting in the formation of the zinc silicate β-Zn₂SiO₄ and its high temperature modification of willemite Zn₂SiO₄. Nanocomposite structure and emission properties have been studied in dependence on the filling degree. The blue luminescence at 430 nm stipulated for the β-Zn₂SiO₄ phase has been observed for the sample with 25 filling cycles. Angular dependences of the PL and Reflection spectra of the opal-ZnO composite with 4 filling cycles demonstrate the suppression effect of the ZnO spontaneous emission in the stop band.

Generalization of optical communication promotes fabrication of low cost integrated waveguide components. Polymer waveguides are attractive because they are very simple to process and are promising for low cost devices. Up to date, several methods are used to make single mode guides in polymer films. Traditionally, they require multi-step processes involving photolithography, etching and annealing. The UV induced modification of the dielectrics property of polymers is a useful technique for low cost realization of integrated optical circuits and can enable a direct writing process. This paper reports a demonstration of photo-printed waveguides in an intrinsically photosensitive polymer film. In this polymer (PVCi: PolyVinylCinnamate), local UV irradiation yields a lowering of refractive index even at telecom wavelengths (1300 and 1550 nm). We show significant achievement concerning the index contrasts up to 3x10^-2 at 1550 nm. The more the refractive index contrast between the core and cladding increases, the more the guide size and curvature can be reduced, allowing the manufacture of smaller photonic devices. However miniaturization of optical planar waveguides in integrated devices increases the coupling losses from and to single-mode optical fibers. Simulations and experimental demonstrations with PVCi photosensitive waveguides also show the potential of photo-printing process to make efficient index mode converters between small planar waveguides and single-mode fibers. The process uses the post-irradiation of the guide by an UV source to gradually decrease the refractive index from the guide to the fibre.

Considerable interest currently exists in the use of plastic optical fibre (POF) for short distances data communications. Attenuation in POF is reduced at 650 nm compared to longer wavelength light and hence silicon based photoreceivers are ideal candidates for use with POF. The difficulty with the development of a CMOS photoreceiver, however, is the realisation of a high speed CMOS photodiode. This paper presents CMOS compatible, shallow junction Geiger-mode avalanche photodiodes (GMAPs) and investigates their bandwidth at 650 nm. Various sized GMAPs (500 μm and 250 μm diameter GMAPs with 20 μm cathode-anode overlaps and 20 μm diameter GMAPs with 3 μm, 4 μm and 5 μm overlaps) were mounted on PCBs. The anodes and cathodes were wirebonded to ground and 50Ohm transmission lines respectively. Impulse response measurements were made for each diode over a range of bias voltages, using a 650 nm picosecond pulsed laser diode. The bandwidths of each diode were calculated from the measured impulse responses and plots of bandwidth versus reverse bias were made. The results indicate very high speed operation is possible (> 1 GHz (20 μm diameter diode)), even for large detectors (> 250 MHz (500 μm diameter diode)).

Until recently, fiber Bragg gratings have been found to exhibit three different growth mechanisms, type I, type II-arising from physical damage-and type IIA. Recently, it has been reported a new regime termed type IA. In this contribution we will report the observation of an abnormal growth mechanism of gratings written in high-Germanium-doped fibers, different from the referred types. First, we will identify possible causes for the observed evolution of the induced perturbation's characteristics (mean value and amplitude) as a function of the accumulated UV fluency. Then, we will analyze the thermal behavior (stability and sensitivity) of gratings formed through the referred abnormal mechanism and compare it with the obtained for the other well-known types (type I, IA and IIA), discussing the results and possible applications in sensing and optical communications areas.

Future high-speed optical communications networks operating at data rates in excess of 100Gbit/s per channel will require a sensitive and ultrafast technique for precise optical signal monitoring. The standard way of characterising high-speed optical signals to use a fast photodetector in conjunction with a high-speed oscilloscope. However, this method is limited to a maximum data rate of approximately 40Gbit/s. An alternative is to employ all-optical sampling techniques based on ultrafast optical nonlinearities present in optical fibres, optical crystals and semiconductors. One such nonlinearity is the optical-to-electrical process of Two-Photon Absorption (TPA) in a semiconductor. This paper presents an optical sampling technique based on TPA in a specially designed semiconductor microcavity. By incorporating the microcavity design, we are able to enhance the TPA efficiency to a level that can be used for high-speed optical sampling.

In this presentation, a simple patchwork to the classical mode matching method (MMM) is proposed to analyze rib ARROWs. MMM gives the modal effective index and the lateral part of modal power loss, and the total modal loss is obtained by including the vertical part which is estimated by a simple yet physical method described here. The simplified MMM (SMMM) is also presented to analyze the effect of TE-TM couplings. Simulation results of selected rib ARROW structures are discussed and compared with those obtained by the Effective Index Method (EIM).

Thin films of wurtzite Ga1-xMnxN have been grown by pulsed laser deposition from a range of Mn-doped-ceramic targets (x = 0.005-0.10). The effect of varying the substrate temperature, background nitrogen pressure and deposition time on the Mn concentration in the deposited films has been studied. Film thickness and surface roughness were monitored during deposition by in situ optical reflectometry. X-ray diffraction measurements showed that under optimal deposition conditions, the films were single phase and epitaxial. Room temperature ferromagnetism was observed over the whole range of Mn concentration, though it was observed that the moment per Mn ion increases as the concentration is reduced.

The authors investigate the propagation of picosecond pulses through Semiconductor Optical Amplifiers using the measurement technique of Frequency Resolved Optical Gating by applying pulses of varying peak power, pulse width and shape. Frequency Resolved Optical Gating is a relatively new measurement system which provides complete characterisation of the pulses in both the temporal and spectral domains. We examine the pulses before and after amplification through the Semiconductor Optical Amplifier. The work shows pulse broadening, the formation of large pulse pedestals, and the generation of significant frequency chirp across the pulse in the temporal domain. In the spectral domain results exhibit spectral broadening and a shift to longer wavelengths. These properties of the output pulses would cause serious degradation in high-speed communications systems employing Wavelength Division Multiplexing and Optical Time Division Multiplexing. The resulting physical properties occurring to the pulses due to propagation through the Semiconductor Optical Amplifier are a result of Self Phase Modulation, which is due to gain saturation induced by carrier depletion and carrier heating.

Widely tunable laser diodes are attractive devices for the next generation wavelength-division-multiplexing (WDM) systems, as they have potential to increase the capacity and the flexibility of the networks. Multi-section tunable laser wavelength switching dynamics, are determined by carrier density and temperature effects. After fast wavelength switching has taken place, a thermal transient induced by the tuning current(s) causes a slow wavelength drift towards a steady state value. Thermal effects on the frequency shift keying (FSK) modulation and wavelength switching dynamics are investigated for a GCSR laser, using the phase section. Thermal effects on wavelength switching dynamics of the device are measured by a time-resolved spectral (TRS) technique, based on a scanning Fabry-Perot Interferometer (FPI). The experimental setup has the ability to measure thermal transients at an increased temporal resolution, over previously reported measurements. A model of the FM response has been developed, which is in close agreement with experimental results.

Copper (I) Chloride is a wide band gap semiconductor with great potential for silicon-based optoelectronics due to the fact that is closely lattice matched with silicon. This work examines the deposition of CuCl thin films by magnetron sputtering on silicon and glass substrates. Film structural and morphological properties are studied with X-ray diffraction and atomic force microscopy. Optical absorbance and luminescence spectra of CuCl thin films are analysed in order to study the excitonic features. The influence of deposition process parameters and post annealing on the film properties are also reported.

Spherical microcavities consisting of a dielectric material show unique optical characteristics as resonators in combination with semiconductor nanoparticles. A high quality factor results in a very narrow bandwidth of the resonant modes (whispering-gallery modes) inside the microcavity. The polystyrene microspheres are coated with one monolayer of CdTe nanocrystals which offer a high photostability and a high quantum yield at room temperature. Due to strong confinement of the electrons in all three dimensions, excitation from the quantum dots is highly size-dependent and tuneable over almost the whole visible spectrum. The deposition of the nanocrystals on the sphere surface allows efficient coupling of the light of the CdTe quantum dots into the microcavity. Photoluminescence and Raman spectra were taken with a Renishaw Raman system. The setup is equipped with an Ar⁺-laser and a HeNe-laser to excite the nanocrystals. Raman measurements show a series of very sharp resonant peaks instead of a continuous spectrum. Strong interaction between the electronic states of the nanocrystals and the resonant modes in the microsphere causes a considerable enhancement of the Raman scattering and luminescence from the CdTe quantum dots in Stokes and anti-Stokes region. Furthermore, a linear blue shift of the resonances in the photoluminescence spectrum was observed during continuous excitation for 18 minutes with a HeNe laser.

The I-V characteristics of AlGaInP/GaInP bulk and multiquantum barrier n-i-n diodes between 20 and 300 K were measured with pA current resolution. When analysed using a thermionic emission model, measured activation energies in the bulk structures were close to the expected conduction band offset. The interplay of other transport mechanisms, including Fowler-Nordheim tunneling and Poole-Frenkel emission was investigated in both the bulk and multiquantum barrier diodes. Transition points between different regimes were observed. Similarities and differences were observed for the bulk and multiquantum barrier diodes. Measured Fowler-Nordheim barrier heights in the bulk barrier diodes agree well with those derived from simulations except in the case of the indirect material at forward bias.

The effect of photonic bandgap interface upon the light scattering was studied in hetero-opals consisting of two opal thin films with different lattice constant. It is shown that the weak scattering regime is preserved in thin hetero-opal films. By comparing scattering spectra of single and hetero-opal films recorded under reversing angles of light incidence and detection it was demonstrated that the interface scatters stronger the light at oblique incident angles. Squeezing of the scattering diagram of hetero-opals compared to single opal films is also assigned to the interface scattering.

The orientation of nematic liquid crystal (LC) E7 on the surface of (111) silicon wafers and in the channels of grooved silicon structure has been investigated. Grooved Si is a periodical structure obtained by wet anisotropic etching of deep and narrow grooves in (110) Si using an alkaline solution. This structure can serve as a one-dimensional (1D) photonic crystal. Composite structures obtained from grooved Si infiltrated by LC are promising candidates for electro-optic application. LC E7 was infiltrated into the channels of grooved Si and into the bulk Si cells. IR spectroscopy and capacitance-voltage characteristics under DC electric field were applied to investigate LC orientation. It is shown that the E7 liquid crystal in grooved silicon exhibits a weak planar orientation with respect to the silicon walls.

Free-space optics (FSO) is a technology that uses modulated laser beams to implement point-to-point communications links without optical waveguides. In many cases, FSO technology can be deployed in metropolitan areas at costs lower than required to install new optical fiber infrastructure. To date, the primary concentration of FSO research and development has been toward the transmission of digital signals, particularly for "last mile" applications. This paper investigates the use of FSO technology for the simultaneous transport of multiple radio frequency (RF) signals over a single FSO channel. Experimental measurements of transmission response and dynamic range, over the frequency range 46-870 MHz, indicate the suitability of FSO links for cable television (CATV) applications.

With the increasing demand for broadband services, it is expected that radio-over-fiber (RoF) systems may be employed to provide high capacity wireless access networks. A major problem with the system is that standard amplitude modulation of the optical carrier generates Double Side-Band (DSB) signals. When these signals are transmitted over fiber, chromatic dispersion causes each side band to experience a different phase shift. If the phase difference at the receiver equals pi, the two side bands interfere destructively causing fading of the received signal. Using an optical filter to filter out one side band can reduce chromatic dispersion effects. Another problem in RoF systems may come from Stimulated Brillouin Scattering (SBS). SBS is an interaction between light and sound waves in the fiber, which causes frequency conversion and reversal of the propagation direction of the light. SBS severely limits the optical power that could be transmitted through a fiber and it is detrimental for optical communication systems. It can also be reduced by the filtration carried out at the transmitter (pre-filtering). Since the optical filer functions to eliminate one sideband and induces insertion loss, the optical power to be launched into the transmission fiber falls sharply and so does the risk of SBS. In this paper, we investigate the possibility of performing SSB filtering to simultaneously overcome the effects of dispersion and SBS in the transmission fibre. The results show that the pre-filtering can filter out one side band as well as reduce the effect of SBS.

This paper introduces for the first time a numerical example of the data-entropy 'quality-budget' method. The paper builds on an earlier theoretical investigation into the application of this information theory approach for opto-electronic system engineering. Currently the most widely used way of analysing such a system is with the power budget. This established method cannot however integrate noise of different generic types. The traditional power budget approach is not capable of allowing analysis of a system with different noise types and specifically providing a measure of signal quality. The data-entropy budget first introduced by McMillan and Reidel on the other hand is able to handle diverse forms of noise. This is achieved by applying the dimensionless 'bit measure' in a quality-budget to integrate the analysis of all types of losses. This new approach therefore facilitates the assessment of both signal quality and power issues in a unified way. The software implementation of data-entropy has been utilised for testing on a fiber optic network. The results of various new quantitative data-entropy measures on the digital system are given and their utility discussed. A new data mining technique known as data-scatter also introduced by McMillan and Reidel provides a useful visualisation of the relationships between data sets and is discussed. The paper ends by giving some perspective on future work in which the data-entropy technique, providing the objective difference measure on the signals, and data-scatter technique, providing qualitative information on the signals, are integrated together for optical communication applications.

Conventional wavelength-division multiplexed (WDM) fiber optic communication systems employ semiconductor lasers operating at discrete wavelengths as carriers for the different data channels. Spectrum-slicing provides an attractive lowcost alternative by utilising spectral slices from a single broadband source, which are then fed to intensity modulators to encode data onto these slices. Such WDM spectrum-sliced systems have the potential for use in local area network fiber communication systems. A key drawback of spectrum sliced systems is the inherently high degree of excess intensity noise, which can impose severe limits on achievable system performance. In this paper we utilise the non-linear gain compression of a semiconductor optical amplifier (SOA) to suppress intensity noise of a spectrum sliced signal from a broadband source. The effects of SOA input power and bias are investigated and system experiments are carried out at 2.5 Gb/s.

A multiwavelength semiconductor ring laser is presented with more than 30 wavelength channels at uniform powers. A polarisation maintaining fibre Lyot filter forms part of the ring and forces multiwavelength operation. The cw characteristics of the system are presented with reference to the number of channels and the channel uniformity, various wavelength spacings are demonstrated. The system is harmonically mode-locked at 1 GHz producing pulsewidths of 100 ps and the pulse train synchronisation between oscillating wavelengths is shown.

Several network testbeds use optical labels as a means of routing data through a network. The EU funded IST project STOLAS is producing one such network testbed which uses an optical label carried by a frequency shift keyed (FSK) modulation of the optical carrier at 75 Mb/s as an orthogonal channel to the 10 Gb/s intensity modulation used to carry the data. This paper presents the results of some transmission experiments on one of the laser transmitters used to supply the IM and FSK modulated signal within the network. In particular the paper investigates the cross-talk between the two theoretically orthogonal modulation channels of IM and FSK and the effect on bit error rate (BER).

Multi-wavelength sources are required for wavelength division multiplexed (WDM) optical communication systems, and typically a bank of DFB lasers is used. However, large costs are involved to provide wavelength selected sources and high precision wavelength lockers. Optical comb generation is attractive solution, minimizing the component count and improving wavelength stability. In addition, comb generation offers the potential for new WDM architectures, such as coherent WDM, as it preserves the phase relation between the generated channels. Complex comb generation systems have been introduced in the past, using fibre ring lasers [1] or non-linear effects within long fibres [2]. More recently, simpler set-ups were proposed, including hybrid amplitude-phase modulation schemes [3-5]. However, the narrow line spacing of these systems, typically 17 GHz, restricts their use to bit rates up to 10 Gbit/s. In this paper, we propose and demonstrate a simple method of comb generation that is suitable for bit rates up to 42.667 Gbit/s. The comb generator was composed of two Mach-Zehnder modulators (MZM) in series, each being driven with a sinusoidal wave at 42.667 GHz with a well-defined phase relationship. As a result, 7 comb lines separated by 42.667 GHz were generated from a single source, when amplitude up to 2.2 Vp was applied to the modulators, giving flatness better than 1 dB. By passively multiplexing 8 source lasers with the comb generator and minimising inter-modulator dispersion, it was possible to achieve a multi-wavelength transmitter with 56 channels, with flatness better than 1.2 dB across 20 nm (2.4 THz).

In this paper a spectral crosstalk monitoring technique is proposed and demonstrated. The technique is based on optically perform a real-time continuous Fourier Transform (OFT) comprising the whole set of transmitted wavelengths. This approach does not require to stop the channel operation. Once the spectral information has been brought to time domain, the basic parameters as amplitude (channel power) or central wavelength can be evaluated. This technique is theoretically developed and demonstrated in a three channel DWDM system at 10 GBit/s channel bitrate in a proof-of-concept experiment.

Many approaches to achieving high information spectral density (ISD), have been reported recently. The standard non-return-to-zero (NRZ) format, which offers a base line performance around 0.4 bit/s/Hz, may be enhanced using a variety of techniques, including: pre-filtering within the transmitter, multi-level modulation formats and polarisation interleaving or multiplexing. These techniques either increase the information per channel (multi-level formats and polarization multiplexing) or minimise interferometric cross talk (pre-filtering and polarisation interleaving) and result in ISDs around 0.8 bit/s/Hz. Combinations of these techniques have been used to provide ISDs of up to 1.6 bit/s/Hz. In this paper we propose a new technique, which we call Coherent WDM (CoWDM), to increase the ISD of NRZ binary coded signals in a single polarisation from 0.4 to 1 bit/s/Hz whilst simultaneously eliminating the need for pre-filters within the transmitter. Phase control within the transmitter is used to achieve precise control of interferometric cross talk. This allows the use of stronger demultiplexing filters at the receiver, and provides optimum performance when the bit rate equals the channel spacing, giving an ISD of 1 bit/s/Hz. This interference control may be achieved by controlling the phase of each laser individually with optical phase locked loops, or by replacing the typical bank of lasers with one or more coherent comb sources, and encoding data using an array of modulators that preserves this relative optical phase. Since optical filtering is not required in the transmitter, stronger optical filters may be used to demultiplex the individual WDM channels at the receiver, further reducing cross talk.

In this paper is proposed a novel high spectral efficiency modulation scheme using time-squared pulses forming an orthogonal wavelength division multiplexing. Experimental results show a significant reduction of the interchannel linear crosstalk-induced penalty compared with Gaussian RZ modulation. Simulation studies are in good agreement with experimental results and show the system performance dependence on several multiplexing impairments inherent to this technique. The proposed modulation technique allows a maximum spectral efficiency of 1 bit/s/Hz without any other spectral efficiency enhancement technique like polarisation division multiplexing.

The potential for sharing infrastructure costs between a large number of customers and the high data rates allowed by optical fibres make passive optical networks (PONs) an attractive solution to the problem of upgrading current copper-based access networks. Optically-amplified, long reach, time division multiple access (TDMA) PONs or 'SuperPONs' offer the potential to further reduce bandwidth transport costs by enabling the direct connection of access networks and inner core networks, thereby eliminating the costs of the outer core/metro backhaul network. The use of dense wavelength division multiplexing (DWDM) could also allow sharing the same feeder fibre and PON head end equipment between a number of such TDMA SuperPONs, each working at different ITU-grid wavelengths. However, a cost effective access solution should employ a customer optical network unit (ONU), which is independent of the PON wavelength, or colorless, in order to reduce the high inventory and deployment costs of using expensive, wavelength-specified sources at the customer. In this paper we demonstrate for the first time the use of a monolithically-integrated, electroabsorption modulator-semiconductor optical amplifiers (EAM-SOAs) as a colorless ONU in a high performance DWDM SuperPON system. These compact devices offer the potential for low-cost optoelectronic integration with other ONU components together with the ability to modulate at rates up to 10Gbps and beyond. We have used this approach to investigate the feasibility of supporting up to 17 SuperPONs from a single feeder fibre and PON head end, each of 100km-reach accommodating 512 users at 2.5Gb/s or 128 at 10Gb/s.

Theoretical modeling of macro-bend losses for a single mode fiber is firstly presented. Macro-bend losses for standard single mode fibers (SMF28) are investigated theoretically and experimentally. The agreement between theoretical calculation results and experimental measured results suggests that 1) most of the radiation field is absorbed in the inner coating layer of SMF28 and 2) so-called elastooptical correction is not required for SMF28.

Birefringence in an InGaAsP/InP tensile-strained bulk semiconductor optical amplifier (SOA) is demonstrated by a polarization resolved amplified spontaneous emission (ASE) spectrum. A mode sum/min approach is used to investigate the periodicities in the ripple for transverse electric (TE) and transverse magnetic (TM) polarizations. The signal-induced birefringence in the SOA, which leads to polarization rotation of a probe beam, is presented in the format of the Poincare sphere. All-optical wavelength conversion (inverted and non-inverted) based on cross polarization modulation (XPolM) is realized. The impact of cross gain modulation (XGM) in XPolM is initially studied.

This paper investigates the temperature dependence of a multi turn bend loss component based on SMF28. Papers investigating the bend loss effect, also known as macro bending, have been widely published with regard to eliminating or minimising the effect. Investigation of the effect especially multiple small diameter turns has had little attention. A mandrel of diameter 28m was wrapped 1.5meters of SMF28 and the insertion loss measured. The temperature dependence of the insertion loss was then investigated.

A new plastic optical fiber directional coupler has been developed. It is fabricated by cutting a shallow notch on the side of the 1 mm diameter cladding fiber, within which is placed a silica 125 um cladding diameter optical fiber for uplink coupling. This novel directional coupler features high efficiency in the uplink and downlink directions because of its advanced geometrical structure and the proposed coupling technique. Compared with the physical size of the notch, the physical size of the silica fiber is relatively small which in turn permits a very high uplink coupling efficiency. Furthermore the physical size of the plastic optical fiber is large relative to the notch which provides a high downlink coupling efficiency. In the uplink direction, we recently reported that with an optimized angle of coupling, the silica fiber can couple light into the plastic fiber with a measured efficiency up to 88% (equivalent to 0.56 dB insertion loss). In the downlink direction, because of the relatively small physical size of the notch, the loss of the power through the notch is low and is estimated to be less than 1.5 dB, given a notch width of 1 mm. In this paper, the variation in the downlink coupling losses with the notch dimensions are measured and compared to those from a theoretical model.

Abstract: Evolving optical communication systems are leading towards an all-optical network, where destination and routing information is incorporated in the carrier wavelength of the signal. It is possible to extend the agility and flexibility of the network by including a Frequency Shift Keyed (FSK) header to all data packets, this can provide additional routing information which can be processed entirely and instantly in the wavelength domain. Thus reducing the queuing required at individual nodes within the network. However, in such a network, due to the number of co-propagating channels and the tight frequency tolerances that are necessary, the optical fibre nonlinearities become very significant and can be detrimental to the use of FSK in such a system. The nonlinear effect under consideration in this paper is Cross Phase Modulation (XPM). This effect occurs when amplitude changes on one channel induce frequency changes on all the others. In a network that simultaneously employs both amplitude and frequency modulation, (where AM is used to transmit payload data and FSK transmits the routing information), the FSK header on one signal can be greatly influenced by the frequency changes induced on it due to the amplitude changes of other signals. In this paper, simulations of the effects of XPM on an FSK signal will be presented. This will include the effects as the number of channels, input optical powers and bitrates are increased. As well as a method to alleviate the effect of the induced frequency variations on the FSK signal.

The growing popularity of the Internet is the key driver behind the development of new access methods which would enable a customer to experience a true broadband. Amongst various technologies, the access methods based on the optical fiber are getting more and more attention as they offer the ultimate solution in delivering different services to the customers' premises. Three different architectures have been proposed that facilitate the roll out of Fiber-to-the-Home (FTTH) infrastructure. Point-to-point Ethernet networks are the most straightforward and already matured solution. Different flavors of Passive Optical Networks (PONs) with Time Division Multiplexing Access (TDMA) are getting more widespread as necessary equipment is becoming available on the market. The third main contender are PONs withWavelength DivisionMultiplexing Access (WDMA). Although still in their infancy, the laboratory tests show that they have many advantages over present solutions. In this paper we show a brief comparison of these three access methods. In our analysis the architecture of each solution is presented. The applicability of each system is looked at from different viewpoint and their advantages and disadvantages are highlighted.

Optical Burst Switching (OBS) is a new paradigm for future all-optical networks. OBS networks are difficult to investigate analytically, so most of the current research is based on simulations. This has created a need for accurate traffic sources. In this paper, we present an overview of different traffic sources and make a distinction between simulating burst assembly and generating bursts (where no information about burst contents is carried). The following types of sources are discussed: emulating assembled traffic using exponentially distributed burst length and interarrival time, emulating assembled traffic with burst length and interarrival time distributions and burst assembly simulation with LRD input traffic. We discuss advantages and disadvantages of each approach, mainly in respect to its efficiency. For example, simulating burst assembly with LRD input traffic, while considered perfect in terms of quality of produced traffic, is also the least effective and requires long simulation times. We show the area of applicability of each type of traffic generator, comparing it to a small number of burst assemblers, or to highly aggregated OBS traffic. We also present two new types of traffic sources: one that can generate traffic with LRD properties in a much more efficient way than full burst assembly simulation, and one that uses conditional burts length and interarrival time distributions. Finally, simulation results are presented to confirm our findings.

For the first time the term data diffraction is introduced, with examples drawn from the algorithm known as phase coherent data-scatter (PCDS) that produces identifiable visual patterns for different types of signal degradation in optical telecommunications. The main signal degradation factors that affect the performance of optical fibers include attenuation, rise-times and dispersion. The theory behind data-scatter is introduced including comprehensive explanations of the theoretical conceptual components of this technique such as centroids, exchange operation, coherence, closeness and projection radius. The various issues of assessing the quality of digital signals are outlined using a simulation study. The authors for the study of optical telecommunications issues have extended the functionality of data-scatter. This approach shows considerable promise. The utility of the data-entropy based 'quality budget method' for optoelectronic system engineering is revisited using an information theory based approach for optical telecommunications. Proposals for the implementation of pattern recognition algorithms to analyse the repeatable patterns within data-scatter are discussed. The paper concludes with brief considerations into the advantages of linking the new data-scatter and data-entropy approaches in digital fiber systems for performance quantification and assessment.

A numerical model based on Mueller matrix formalism is developed to include depolarization effects in the statistics of Polarization-Dependent Loss (PDL) in recirculating loops. The model shows that depending on the degree of depolarization introduced at each round trip, the maximum value of PDL is limited to lower values when compared with completely polarized light. First experimental evidence of this effect is also reported with estimation of the degree of polarization, DOP, introduced in the system, mainly due to spontaneous emission noise of the optical amplifiers.

This project has developed a low cost in-home network compatible with network standard IEEE1394b. We have developed all components of the network, from the red resonant cavity LEDs and VCSELs as light sources, the driver circuitry, plastic optical fibres for transmission, up to the network management software. We demonstrate plug-and-play operation of S100 and S200 (125 and 250Mbps) data streams using 650nm RCLEDs, and S400 (500 Mbps) data streams using VCSELs. The network software incorporates Home Audio Video interoperability (HAVi), which allows any HAVi device to be hot-plugged into the network and be instantly recognised and controllable over the network.

In applications such as target tracking the ability to steer the radiation pattern from an antenna array is required. This paper details the theory, design, fabrication and characterisation of a new type of reconfigurable planar antenna reflectarray for operation with circularly polarised signals. Each element in the array comprises a resonant dipole antenna that can be rotated about its axis, each element is positioned at an odd number of quarter wavelengths above a ground plane. An array of fixed antennas was fabricated on a high resistivity silicon substrate with measurements confirming that the silicon exhibits low absorption at quasi-optical frequencies. The demonstrator devices are designed for 100 GHz operation in order to facilitate fabrication of a usable array aperture on a single silicon wafer. Dimensions of the antennae and the thickness of the substrate were selected accordingly. The wide range of micro-machining techniques that are available for silicon based structures enabled the design of a demonstrator array where the elements can be rotated using a rack and pinion arrangement. In such structures, the actuating mechanism is positioned outside the radiating aperture of the array to exclude conducting elements that would otherwise impair antenna functionality. A method of fabricating the reconfigurable array has been developed and successfully implemented.

Complex optical networks demand high levels of functionality from optical components requiring high density integration in low cost, compact, plug and play format. Technologies at the forefront of this move to higher integration are MEMs and planar waveguide options. In all cases the components are interfaced with the transmission medium, generally through precision alignment to individual fibres. All-fibre technologies offer good performance but are not in general appropriate for multi-functional integration. One all-fibre approach that offers the potential for integration is the fibre evanescent field technology. This technology creates a waveguide substrate from the transmission optical fibre onto which the optical circuit can be built. This paper describes a VOA-power monitor device formed using evanescent field approach, to assess its performance and the potential for full integration.

Epoxy free bonding is decisive to improve reliability of optoelectronic devices using active components such as laser diode. Therefore soldering is usually preferred as bonding technology but it often needs components metallizing, heating over 140°C and liquid or gas fluxing which may let some corrosive residues. Thus soldering cannot be widely used on optical microchip components. Working on solid-state microchip laser bonding in a project called NANOPACK supported by the French research ministry, we have developed low temperature epoxy free bonding technology. The microchip laser is bonded onto a submount by thermocompression at low temperature and moderate relative pressure using an indium foil to form the adhesive joint. This technology uses both a unique property of indium to wet and to bond to certain non-metallics such as glass, quartz, and various metallic oxides, and fluxless soldering of indium with gold by solid-state interdiffusion bonding or solid-liquid interdiffusion bonding. This way, mean bond strength about 300g/mm² has been obtained for 2mg chip with very good resistance to thermal aging. This epoxy free technology offers a real alternative for bonding non-metallic components which cannot stand usual soldering processes. Moreover, as it is a fluxless process, this technology is very attractive to hermetically seal lids under controlled atmosphere.

Thin films of the metallopolymer [Os(bpy)₂ (PVP)₁₀]²⁺, where bpy is 2,2'-dipyridyl and PVP is poly(4-vinylpyridine), luminesce at 750±12 nm upon excitation at 355nm. The luminescence decay responses can be described by a double exponential decay model in which the limiting lifetimes are 75±14 (population fraction of 0.9) and 35±8 ns (population fraction of 0.1) for films in contact with aqueous 0.1 M H₂SO₄. Electrochemistry has been used to create well defined concentrations of the luminescence quencher, Os³⁺, within the films. Time resolved spectroscopy reveals that both dynamic and static processes contribute to luminescence quenching with a rate constant for electron transfer between the photoexcited Os²⁺* and the Os³⁺ centres of 1.3x10⁷ M^-1s^-1 being observed. Stable gold nanoparticles have been created within the metallopolymer by the chemical reduction of tetrachloroaurate. These nanocomposite materials exhibit enhanced emission intensity compared to the gold free films.

Lowering optical packaging costs requires developments in new technologies. In this paper, solder ink-jet process is presented for flip-chip component assembly on planar, 3D, flex and stacked submounts and substrates. Applications for this technology are presented and include linear array in-vivo dosimeters, integrated GaN LED displays, telecomm submounts and wearable ambient systems. An important aspect of developing this technology is process reliability. In this study, the reliability of the solder to bump accurately and adhere to various target bond pads was evaluated as well as MIL standard shear tests to qualify the joint strength of the bump.

UV and VUV laser ablation of micro-fluidic circuits for micro-total analysis systems (μTAS) is an alternative to more expensive techniques of LIGA or micro-moulding. The machining of trenches in PS, PC, PMMA (CQ grade and non-CQ grade) was investigated using two Q-switched solid state lasers, one operating at 266nm (fourth harmonic) and another at 355nm (third harmonic). These results were compared with results achieved using a 193nm Excimer laser. The structuring of the channels depends on both the laser wavelength and the target material. A comparison of the process will be presented with respect to the structure quality and efficiency of the process.

High brightness LEDs (HBLEDs) have been fabricated on GaN semiconductor material grown on sapphire substrate. These devices provide an optical output power in excess of 50 mW at a driving current of 1 amp. For this high current application, large size (1.8 mm × 0.6 mm) GaN LEDs are flip-chip mounted onto a heat sink to provide a low thermal resistance path from the junction to the ambient. For the flip-chip mounting, a Au/Sn/Au solder and a Au/Au thermal compression bonding process have been optimized. The bond strength of the Au/Sn solder joints and the Au-Au bonds is measured through shear testing. Good bond strength results of 224 g/f for the Au/Sn/Au solder and 288 g/f for the solid Au bonds have been achieved. The thermal modeling of the assembly is done with a finite element analysis and the optimum design has been adopted for this high current application. At present these assemblies are under lifetime test and so far nearly 6000 hours of continuous operation has been achieved.

X-ray technology provides a number of powerful tools used in many different areas of science and industry. The ability to focus x-rays is the key to a wide range of applications including medicine (diagnosis and therapy), industrial applications (lithography and inspection), astronomy (space telescopes) and x-ray imaging (microscopy and analysis). This work presents the design and microfabrication of a novel x-ray micro optical system for use in an x-ray microprobe for analysis of biological cells. A reflective micro-optical system capable of focusing a wide range of wavelengths is at an advanced stage of development. The lens system consists of a pair of microfabricated optical elements, one of which has variable curvature providing a unique mechanically-actuated zoom focusing capability. Experiments have been carried out to measure the changes of the focal length (lens curvature) and sensor calibration.

Porous materials in general have received great attention from the last century. The development of new porous materials and the preparation of new composites based on porous materials is a subject of interest. The development of porous silicon based optical composite materials opened up new ways of incorporating optically active sol-gel materials into porous silicon. High purity silica optical fibres allow the most rapid and efficient data transmission. The objective of this work is to develop micro-channel glass / porous silicon-rare earth doped xerogel and glass composites, which would serve as compact optical amplifiers and delay line devices. Micro-channel glass / porous silicon-xerogel composites have been prepared by incorporation of sol-gel prepared from tetraethoxysilane, aluminium iso-propoxide and europium chloride into the porous matrix. Both xerogel and glass composites have been studied by various techniques such as FTIR, micro-Raman, photoluminescence spectroscopy, EDX and Scanning Electron Microscopy (SEM).

Nanoimprint lithography (NIL), with its apparent simplicity and resolution down to 6 nm, has become an attractive flexible and low-cost technique for nanopatterning of thin films, which themselves act as a mask for further nanofabrication steps, or which can be used as-printed thanks to the functionality of the thin film itself. In this work, we focus on the latter approach and report on our experiments carried out to fabricate organic photonic devices. Silicon stamps, with figures down to 100 nm, are fabricated using electron beam lithography and reactive ion etching. Different fabricated stamps include waveguides, gratings, splitters and interferometers. New fabrication techniques are investigated, namely the combination of NIL with optical lithography and reverse NIL. These two techniques allow producing three-dimensional structures. For the combination of NIL with optical lithography, an original approach is used consisting of a polymer stamp on top of a quartz + metal optical mask. In the case of reverse imprint and multilevel structures, particular attention is paid to adhesion between the stamp, the polymer and the substrate on which the layer is reported. These two techniques are very promising for the fabrication of complex polymer optical devices, like distributed feedback structures, in one step.

In this work we analyze the optical propagation in a composite dielectric ridged waveguide calculating several parameters that characterize the linear and the nonlinear phenomena. This novel composite waveguide consists of a circular central core and many circular sectoral waveguides at the periphery, while the whole device can be considered as an nonlinear optical coupler. Firstly, we analyze the linear optical propagation in a sectoral dielectric birefringent waveguide calculating the propagation constant, the effective refractive index and the normalized propagation constant in the weak-guidance regime. Several simulations are performed varying some of the parameters of the geometry and the optical frequency in order to produce dispersion diagrams. Following, the electric and magnetic field distributions for the fundamental linear guided modes are derived. Finally, we calculate the linear coupling coefficient between two identical sectoral waveguides, the linear coupling coefficient between a sectoral waveguide and the circular-core waveguide, and the sectoral waveguide mode effective area for the evaluation of Kerr nonlinear coefficient.

Grooved silicon (gr-Si) structures with a period of few micrometers, which were formed by anisotropic etching of (110) Si wafers, have been investigated by means of broad band infrared (IR) and submillimeter transmission spectroscopy. In the spectral region of 50-1000 μm the results are well explained by an effective medium model, which predicts a strong birefringence with a difference between refractive indices for ordinary and extraordinary beams to be about 0.73-0.77. The IR transmission of gr-Si in the range from 1 to 30 μm is strongly influenced by light scattering. The experimental results measured in the region 1-5 μm can be understood in the terms of the geometric optics.