We report synthesis and photoluminescence optimization of luminescent Si nanocrystals in silicon rich oxide films using a CO₂ laser beam. Laser annealing allows for a very localized heat deposition. This results in appreciable temperature rise in an area that is equivalent to only a few spot sizes. This could be important in CMOS back-end compatible processing where high temperatures on the entire wafer scale might not be acceptable. Furthermore, temperature optimization studies in furnace annealing are time consuming because the furnaces have to be programmed to each individual temperature and the stabilization takes long times. In CO₂ laser annealing, the entire temperature range for nanocrystal formation is available along the radial and axial directions of the laser spot - thereby allowing temperature optimization in a single short experiment. Presence of crystalline nanoparticles is ascertained using structural analysis techniques like transmission electron microscopy (TEM) and X-ray photoelectron spectroscopy (XPS). We also report luminescence optimization with respect to laser power and annealing time. It is observed that laser annealing in an air ambient results in two peaks in the luminescence spectrum - on in the visible at 570 nm and one in the near infra read at 800 nm. Origin of luminescence in these two peaks is probed by hydrogen passivation and time resolved measurements. In the second part of the paper, we focus on continuous wave characterization of photoluminescence from Si nanocrystals embedded in microdisk resonators. There have been numerous reports on observation of continuous-wave and transient gain in planar optical waveguides with Si nanocrystal active layer. However, there are relatively very few investigations focusing on photoluminescence emission from Si nanocrystals (quantum dots) embedded in on-chip optical microcavities. Microcavities spectrally filter the luminescence from the quantum dots and, depending on the (Q/V) ratio, can significantly alter the spontaneous photoemission from the quantum confined excitons. In our work, planar microcavities are patterned on the emitter layer by high resolution electron beam lithography and a combination of dry and wet chemical etching. Fabrication procedure is optimized to maximize the ratio of the quality factor and the mode volume. Continuous-wave photoluminescence measurements are performed by top-pumping the resonators with a 488 nm line of an argon ion laser. We study the photoemission from the microdisks for the polarization dependence, and quality factors. Contributions of various mechanisms leading to the observed loss are estimated. We believe that our studies will help gain further insight into photoemission physics of the group-IV nanostructures.

We propose here a new class of nano-cavity surface-emitting light source on silicon, based on the integration of colloidal nanocrystal quantum dots (NCQDs) and air-hole two-dimensional photonic crystal (2D PC) slab waveguide cavities. A phenomenological dimensional reduction approach (PDRA) has been developed to investigate the characteristics of this class of NCQD-PC lasers. Over one order of magnitude in gain threshold reduction was obtained in single defect PC cavities owing to the spontaneous emission control. Ultra-compact high efficient light source is feasible based on the relative low gain NCQDs, as compared to conventional epitaxial III-V material systems, owing to the relaxed gain threshold requirement in PC cavities. It is also expected such single defect PC cavity based laser to have an optical output power more than 20 μW with cavity size less than 2 μm, making it an attractive source for optical interconnect and sensing applications.

In this report, fundamentals, design, fabrication technology, and parameters are presented for contactless Si photonic emitter operated in the 3-5 um atmosphere transparency window at well above room temperature. To bypass the material band structure limitation, we utilized the above-bandgap light-induced free carrier thermal emission as a way to monitor the below-bandgap radiation that falls into 3 to 5 μm band (light down conversion). Two-facet external power conversion efficiency up to 5% is observed at T~500 K with further improvement to be expected. The device application to the IR dynamic scene simulation as well as it pros and cons in respect to thermal emitters and IR LEDs are also considered.

In this paper we demonstrate the first optical actuation of a single-pixel, deformable-mirror MEMS device through a direct cascade with a photodetector. Photovoltaic, p-i-n, and avalanche photodetectors were successfully utilized. Mirror deformations were monitored by interferometry. Deformation is quasilinear at low light intensities, and saturates at higher intensities. Actuation at picowatt light intensities has been accomplished by cascading with an avalanche photodetector. We also describe the fabrication of an integrated device consisting of an all optically addressed deformable-mirror MEMS suspended over a p-i-n photodetector. Initial demonstration of optical actuation of the deformable mirror using the newly integrated device is also presented.

We report on the fundamentals and technology of Si-based linear all-optical light down-conversion process. The approach is in the possibility to enhance the thermal emission power of semiconductors in the spectral range of intraband electron transitions (mid- and long-wave infrared, free carrier absorption band) by the shorter wavelength optical pump (interband transitions, visible to near-infrared, fundamental absorption band). We experimentally realize conditions (the 1.15-μm-pump wavelength and 2 to 16-μm-signal wavelengths, T ≈ 500 K) when Si-based device demonstrated 220 % external power efficiency. As a matter of fact, we come up with new concept for high-temperature incoherent light amplifier (optical transistor) made of indirect bandgap semiconductors.

The Si metal-oxide-semiconductor structure can be used for light detection, and the dark current is significantly reduced with the oxide layer. With the photo excitation, the generated carriers can be collected by electrodes as photo current. By incorporating Ge, the metal-oxide-semiconductor photodetectors can increase the responsivity and extend the detection wavelength. The interband transitions in the SiGe quantum dots enhance the 820 nm infrared absorption and extend the detection range to 1550nm. The valence band offset between Si and SiGe forms discrete quantum states in the SiGe layers. Hence, metal-oxide-semiconductor SiGe/Si quantum dot (well) infrared photodetectors can be used to detect midand far- infrared using the intraband transitions. The Ge-on-insulator metal-oxide-semiconductor photodetectors can further increase the detection speed by reducing parasitic capacitance. The large work function metal (Pt) is used for the gate electrode to reduce the dark current. Moreover, the external mechanical strain can enhance the photo current with slight degradation of dark current.

Due to their high electrical bandwidth and good signal-to-noise ratio, optical receivers that utilize a combination of PIN photodetectors and erbium-doped fiber amplifiers (EDFAs) have emerged as an attractive technology for high-speed optical communication. However, the drawbacks of this technology are cost and bulkiness. Since avalanche photodiodes (APDs) are capable of amplifying the photocurrent internally, without the need for optical preamplification, they may offer a cost-effective and compact alternative to the PIN-EDFA combination. Unfortunately, this internal optoelectronic gain comes at the expense of uncertainty in the APD's gain, and more importantly, at the expense of reduced speed due to the notorious avalanche buildup time. The relatively slow response time of an APD, compared to a PIN photodiode, introduces significant inter-symbol interference (ISI) at high operational transmission rates. In this work, an equalization approach is undertaken to compensate for buildup-time-induced ISI by means of either the transversal equalizer (TE) or the decision-feedback equalizer (DFE). To design the equalizers, the APD-based receiver is viewed as a random linear channel whose impulseresponse function is a stochastic process. The mean and the correlation matrix of the receiver's random impulseresponse function are numerically determined by utilizing a recently developed analytical model. It is revealed that these equalizers can reduce the bit-error-rate (BER) remarkably at high transmission rates; this makes current InP APDs potentially suitable for near 40-Gbps digital operation.

The development of new medical tools for minimal-invasive surgery is essential for the reduction of pain, side-effects and hospitalisation costs. The benefits of using optical methods in some procedures are now well recognised. In this context semiconductor optical sources offer the advantages of high efficiency, compactness, low cost, long lifetime, low power consumption and high reliability. The optical source presented in this paper comprises a phase-locked, high-power laser diode array and beam shaping optics designed to optimise the coupling to small-diameter optical fibres. The high-power, index-guided laser array used in this work was developed to achieve high-brightness by adopting a specially designed optical cavity based on the parabolic taper for each individual array element. With this design the individual elements are phase-coherent and the fundamental mode of the array is dominant, thus achieving quasi-diffraction-limited operation without the use of external lenses. Compared to other high-brightness laser arrays, the parabolic bow-tie laser array used here offers the advantages of simple device fabrication, reduced costs and compactness. In this work, specially designed lenses are used to circularise the beam and, therefore, to focus the beam to a small spot-size to improve coupling to single mode fibres. Details of the device characteristics, with emphasis on beam quality and phase front, and of the design of anamorphic optics for beam shaping and focusing will be presented in the context of the integration of the high-brightness laser array with the specially design optics to achieve optical power delivery exclusively where necessary via small-diameter optical fibres.

The application of multi-spectral microthermography to the monitoring of aging processes in diode lasers is reported. We have found that an intensity of the luminescence in near IR (1.5 -2 micron range) increases with the operational time, which tentatively is correlated with increased concentration of point defects. This effect is monitored with a specially configured thermographic camera. The set-up provides spatially resolved information about the luminescence that originates from radiative recombination at defect centers as well as the pure thermal emission. In order to elucidate the role of point defects in the aging process of diode laser complementary spectroscopic measurements are performed. Photocurrent spectroscopy is used to examine the absorption properties of laser structure for fresh as well for aged devices. The results of low-current I-V characterization are presented. A correlation between measurement results obtained using different methods is found, discussed and interpreted in detail.

Superluminescent Diodes are high power semiconductor optical sources with relatively broad spectral linewidth used for a variety of applications. The basic, crucial device feature needed to achieve high-power superluminescent operation is low facet reflectivity to prevent lasing and optical gain saturation. Theoretical calculations show that by appropriately designing the device length and output facet reflectivity a significant increase in optical output power and wall-plug efficiency can be obtained. This paper presents theoretical and preliminary experimental results indicating that it is possible to further improve the operational characteristics of superluminescent diodes and achieve high optical output power with high conversion efficiency. The control of the facet reflectivity and the overall device optical geometry are obtained by using Focused Ion Beam post-fabrication processing. Results obtained from the characterisation of superluminescent diodes before and after facet reflectivity alteration achieved by creating patterns on the subwavelength scale will be discussed.

In a radical departure from conventional thermal emitter-based dynamic IR scene simulation devices, we have tested InAsSbP/InAs LEDs grown by liquid phase epitaxy and tuned at several peak-emitting wavelengths inside the mid-IR band. Light uniformity, radiation apparent temperature (T_a), thermal resistance, and self heating details were characterized at T=300°K in the microscale by calibrated infrared cameras in the 3-5 μm (light pattern) and 8-12 μm (heat pattern) bands. We show that LEDs are capable of simulating very hot (T_a ≥740°K) targets as well as cold objects and low observable with respect to a particular background. We resume that cost effective LEDs enable a platform for photonic scene projection devices able to compete with thermal microemitter MEMS technology in testing and stimulating very high-speed infrared sensors used for military and commercial applications. Proposals on how to further increase LEDs performance are given.

We present designs and simulations for a highly cascadable, rapidly reconfigurable, all-optical, universal logic gate. We will discuss the gate's expected performance, e.g. speed, fanout, and contrast ratio, as a function of the device layout and biasing conditions. The gate is a three terminal on-chip device that consists of: (1) the input optical port, (2) the gate selection port, and (3) the output optical port. The device can be built monolithically using a standard multiple quantum well graded index separate confinement heterostructure laser configuration. The gate can be rapidly and repeatedly reprogrammed to perform any of the basic digital logic operations by using an appropriate analog electrical or optical signal at the gate selection port. Specifically, the same gate can be selected to execute one of the 2 basic unary operations (NOT or COPY), or one of the 6 binary operations (OR, XOR, AND, NOR, XNOR, or NAND), or one of the many logic operations involving more than two inputs. The speed of the gate for logic operations as well as for reprogramming the function of the gate is primarily limited to the small signal modulation speed of a laser, which can be on the order of tens of GHz. The reprogrammable nature of the universal gate offers maximum flexibility and interchangeability for the end user since the entire application of a photonic integrated circuit built from cascaded universal logic gates can be changed simply by adjusting the gate selection port signals.

The longitudinal component of the polarization field inherent in polar materials, combined with constrained carrier motion along the quantum wells, causes formation of equilibrium plasma nano-sheaths at intersections of quantum wells. The induced short range (nm) potentials of peak voltages much larger than the thermal carrier energy cause wavefunction localization, which further reduces the dimensionality of the carrier behavior. The associated energy band-bending causes enhanced carrier accumulation at quantum wedges and quantum tips formed by intersecting quantum wells. In addition, the total carrier number over the QW length increases, manifesting spontaneous intrinsic pumping due to polarization. As a result, the spontaneous emission is localized at quantum wedges, and the total emission exceeds that from a flat quantum well of similar parameters, as experimentally observed. The sheath potentials are sufficiently high for 1-D or 0-D carrier localization at quantum wedges and quantum tips.

We report on the transient photoluminescence behavior of InAs/GaAs quantum dots. Quasi-instantaneous excitation by femtosecond pulses and luminescence detection by a synchro-scan streak-camera allows for monitoring transients in the range between 7 ps and several ns. For the practical application of novel micro- and nanostructures as active region materials of optoelectronic devices, knowledge about their behavior at high non-equilibrium carrier densities as well as about the carrier dynamics within such complex systems is required. Obtaining insight into these elementary processes allows for the estimation of potential application fields and perspectives of given structures and device concepts. We analyze the carrier transfer between quantum dots. In particular we address the lateral transfer within one dot-plane, the vertical transfer between different dot planes in stacked arrays, as well as the vertical transport between dot planes, where the heights of the potential barriers are externally controlled. For these model systems, we discuss the recombination behavior, aspects such as carrier trapping, and carrier localization as well as the carrier transfer between different parts of the structures.

We present a study of time-resolved transmission and emission properties of optically anisotropic planar microcavity structures. The structures consist of λ/4-layers of SiO₂ and TiO₂ for the dielectric mirrors and a cavity layer of either SiO₂ or the organic dye composite AlQ₃/DCM. For the SiO₂ cavity, we observe a polarization splitting at normal incidence leading to terahertz oscillations of transmitted coherent light. The polarization splitting is explained by an optical anisotropy of the dielectric layers caused by the fabrication process. We apply an up-conversion setup for temporally and spectrally resolved transmission measurements and obtain a corresponding beating of 1.25 THz. Time resolved measurements yield a Q-value of 1600, corresponding to a cavity photon lifetime of 0.65 ps. We explain our observations with a transfer-matrix model and introduce a Fourier-transform based analytical algorithm. The cavity filled with the organic dye composite can act as an organic microcavity laser. The birefringence of the distributed Bragg reflectors leads to lasing in two perpendicularly polarized modes. Investigations of the ultrafast dynamics of this laser system show a phase coupling of the two laser modes leading to the generation of a terahertz optical beat. The oscillation frequency can be widely tuned by variations in the fabrication process.

We report on light emission from high-Q neodymium-implanted silica microtoroids. Following the description of the fabrication process of microtoroids, neodymium light emission is analysed. This emission is coupled to various cavity modes. Using evanescent wave coupling we achieve selective detection of Whispering Gallery Modes of a microtoroid.

Our focus is on the potential of enhancing light-matter interactions by using microstructures. It has been proven that microstructures of composite materials (MCMs) are versatile means for passive optical functional integration. Here we elaborate on some motivations and preliminary results. By using MCMs, we present a few examples of enhanced functional integration for optical networks and structures for light trapping at UV-VIS (e.g., solar cells). Simulations of wavelength division multiplexing (WDM) and polarization dependent devices for network terminals at 1310 nm and 1500nm are presented. A simple principle was used to demonstrate total absorption of visible spectrum around 400-500 nm, which may benefit application such as solar cells.

We present an electro-optically (EO) tunable filter design. The EO tunable filter is capable of nanosecond wavelength tuning across the whole C band with a narrow pass band of <0.4nm and a low channel cross-talk of <-30dB. This EO filter can be used in dynamic WDM networks for fast wavelength scanning/selection, communication channel reconfiguration, and optical switching.

A modified full-vectorial finite-difference beam propagation method based on H-fields in solving the guided-modes for optical waveguides with step-index profiles is described. The propagation is split into two substeps. In the first substep, the field propagates in the absence of the cross-coupling terms, and then they are evaluated and double used in the second substep. An improved six-point finite-difference scheme is constructed to approximate the cross-coupling terms along the transverse directions. By using the imaginary-distance procedure, the field patterns and the normalized propagation constants of the fundamental modes for a typical rib waveguide are presented, and the hybrid nature of the full-vectorial guided-modes is demonstrated. Solutions are in good agreement with the benchmark results from film mode matching method, which tests the validity and utility of the present method.

An analysis of the mode structure of high-power double-barrier separate confinement DB SCH diode lasers is presented. The devices are characterized by very low vertical beam divergence (13 - 22°, depending on the design version). Modelling of the fundamental mode distribution for three different design versions of DB SCH diode lasers is discussed and the results are compared to a macroscopic characterization of the devices (far-field directional characteristics and photocurrent spectra). Microscopic measurements of the near field distribution of these diode lasers with subwavelength spatial resolution are performed by employing a Near-field Photocurrent (NPC) technique. The mode structure of diode lasers is directly visualized giving indications about the interplay between the heterostructure design and the emission characteristics.

With the widespread use of laser diodes in modern industry there has been an increasing demand for high optical output power devices with good beam quality and, ideally, low production and packaging costs. Reliability and long lifetime are essential requirements since they determine the extent to which such sources will be utilised. The devices of interest here are arrays of parabolic bow-tie lasers which have been specially designed to achieve high power with high brightness without the need for re-growth or sophisticated device fabrication. This paper presents a comparative study on laser diode arrays to investigate the effects of scaling and device geometry on device operation, including degradation and ageing. Temperature profiles at the array facets have been obtained using a thermal imaging system. The HgCdTe-based detector operates in the 1-5.5μm wavelength range. The results obtained indicate a smaller increase in temperature (2-5°C) in uncoupled arrays with respect to phase-coherent arrays and a considerable increase in temperature with increasing number of elements in the array. Such considerations are essential to properly manage thermal dissipation and improve the operational characteristics of such devices.

We have demonstrated for the first time the analysis of noise properties in a 65mW, 405nm violet laser diode and also stabilized its frequency fluctuations as low as hyper coherent level. We measured the longitudinal mode behavior in some operation conditions, including the ambient temperature and optical feedback amount and so on. With some optical feedback amounts usually seen in typical optical pick-up systems, we have confirmed that the average relative intensity noise (RIN) value measures up to about -125dB/Hz, while the external cavity length dependences of the RIN show maximal and minimal spikes repeatedly at certain points where the external cavity length coincides with integral multiples of the effective internal cavity length of our sample LD. Under such circumstances, we have achieved the suppression of the optical feedback noise to about - 132dB/Hz by selecting the polarization of the feedback light. Moreover, we have proved the measured 65mW, 405nm violet laser diode sample tends to oscillate in multiple modes, so that we should carry out the single longitudinal mode operation by controlling the ambient temperature about 15°C. Thus, we have finally attained the frequency stabilization of the violet laser diode using a reference Fabry-Perot cavity based on the Pound-Drever- Hall method. As a result, we have achieved for the first time the frequency stability of 1.71×10^-10 of the minimum square root of Allan variance in a 400nm type violet laser diode.

In this paper, error analysis and tolerance allocation methods for the optical head of NFR (Near Field Recording) system are presented. We fabricate the NFR system and test the reading & writing performance of the NFR system. The test results show that the performance is not good enough. In order to find the cause of the performance drop in the NFR system, assembly and manufacturing tolerances in the optical head of NFR system are simulated. The tolerances analysis result shows that it needs to allocate the tolerances of the optical head of NFR system. So we proposed optimal compound tolerance allocation method using WOW (worst on worst) method and Monte-Carlo method base on sensitive analysis of the optical system. We used two tolerance allocation methods to allocate the compound tolerance in the optical head of NFR system. The results show that WOW method is an over-design method and the Monte-Carlo method is the optimal method of tolerance allocation in optical system.

In reference to an AlInGaN UV LED fabricated in laboratory, the optical properties of the 370-nm UV LEDs are investigated with a self-consistent APSYS simulation program. The optical performance of the UV LEDs with different aluminum compositions in AlGaN electron blocking layer and different numbers of quantum wells are investigated in an attempt to optimize the UV LED structure. The simulated results show that the electron leakage current can be effectively reduced with the use of an AlGaN electron blocking layer with an aluminum composition of greater than 0.19, and optimum performance may be obtained when the number of quantum wells is three. Since the built-in polarization is one of the most important factors for the deterioration of III-nitride LED performance, the feasibility of using a latticematched quaternary Al_0.18In_0.039Ga_0.781N electron blocking layer in the UV LED to improve the LED performance is also numerically studied. The simulated results suggest that with the use of a lattice-matched Al_0.18In_0.039Ga_0.781N electron blocking layer, the polarization charge density in each heterostructure interface is reduced, the electrostatic field in quantum wells is reduced, and the maximum output power is sufficiently improved. The simulated results also indicate that better LED performance may be obtained when the Al_0.18In_0.039Ga_0.781N electron blocking layer has a higher pdoping concentration due to reduced electron leakage and increased hole concentration in active region.

The effects of built-in polarization and carrier overflow on InGaN quantum-well lasers with Al_0.2Ga_0.8N or AlInGaN electronic blocking layers have been investigated numerically by employing an advanced device simulation program. The simulation results indicate that the characteristics of InGaN quantum-well laser can be improved by using the AlInGaN electronic blocking layer. When the aluminum and indium compositions in AlInGaN electronic blocking layer are appropriately designed, the built-in charge density at the interface between InGaN barrier and AlInGaN electronic blocking layer can be reduced. Under this circumstance, the electron leakage and threshold current can be decreased obviously as compared with the laser structure with conventional Al_0.2Ga_0.8N electronic blocking layer when the built-in polarization is taken into account in our simulation. On the other hand, the AlInGaN electronic blocking layer also gives higher refractive index than the Al_0.2Ga_0.8N electronic blocking layer. Therefore, higher quantum-well optical confinement factor can be obtained by using the AlInGaN electronic blocking layer as well.

Referred to the laser structure and its experimental results obtained by Selmic et al. and Liu et al., optimized active structure for the 1.55-μm quantum well lasers based on AlGaInAs material system is investigated. A structure with 1.2% compressive-strained wells and a p-type AlInAs electron stopper layer of 20 nm thickness and 5×10²³ m^-3 doping concentration is suggested. Using this structure the threshold current is reduced to 17.8 mA, and the electron overflow percentage is decreased to 1.74% at 330 K. Furthermore, the characteristic temperatures of threshold currents are enhanced to 55.6 K, 67.0 K, and 43.3 K in operating temperature ranges of 300 K~350 K, 300 K~330 K, and 330 K~350 K, respectively.

Thermooptic switches are viable options for rapidly and reliably switching and routing optical signals in planar lightwave circuits. We present modeling and fabrication of a thermooptic switch made of the polymer SU-8, an epoxy resin commonly used as a MEMS structural material. SU-8 is a good candidate material for use in planar waveguides due to its high refractive index, good transmission properties in the visible and infrared, and excellent thermal and mechanical stability. Furthermore, it has great advantages in fabrication since it is used as a negative photoresist, and so can be patterned directly using photolithography. Light is guided by a refractive index gradient generated by embedded MEMS microheaters, which activate the thermal nonlinearity of the polymer. The thermooptic change in refractive index imparts an inhomogeneous phase shift to the beam in the waveguide, which guides the input into one of two or more outputs. The switch design and operation parameters have been optimized using simulations of the thermal profile using finite element modeling and of the optical propagation using the beam propagation method.

An antireflection coating material for optically pumped group IV-VI lead-chalcogenide semiconductor light emitting devices has been proposed. The coating has been used to increase the photo-pumping efficiency. Theoretical model showed that with the proposed AR coating with a quarter wavelength thickness, 0.008% reflectivity could be achieved in the 980nm-982nm wavelength region. The antireflection property of the coated film was investigated by FTIR-spectroscopic reflectance measurement. Room temperature continuous-wave photoluminescence measurement from AR-coated multiple quantum well structures showed up to 4-times increment in the PL intensity, compared to uncoated ones.

Group II-VI narrow band gap compounds CdTe, ZnCdTe and CdSeTe are known as the most suitable semiconductor materials for the room temperature γ- and X-ray detectors. In this work, electronic properties of a quadrupole compound ZnCdSeTe were investigated. Chlorine, copper and oxygen doped host material was synthesized from the grinded mixture of 2 mol% ZnTe, 36 mol% CdTe, and 62 mol% CdSe, to keep a hexagonal structure of crystals. Precautions were applied to achieve an uniform doping, high quality of the crystal surfaces, and to remove residue phases after the thermal treatments. Fabricated polycrystalline samples showed a high performance from NIR via VIS and UV to X-ray band, with sharp donor-acceptor pair peak at 922 nm, and dynamic range above 10⁴. High stability, good linearity and performance of samples was measured using X-ray source with Cu-anode, at 40 kV.