Aufkonversions-Nanokristalle (UCNC) zeichnen sich als einzigartige Lumineszenzreporter aus, die Nah-infrarotes Anregungslicht in Photonen höherer Energie umwandeln. Für die gezielte Anpassung von Eigenschaften, bedarf es ein tiefes Verständnis der Prozesse der Aufwärtskonversionslumineszenz (UCL) und deren Abhängigkeit von Material und Partikeldesign. Diese Doktorarbeit untersucht die UCL-Prozesse von Yb3+,Er3+ dotierten SrF2-UCNC und zielt darauf ab, die UCL-Eigenschaften der bisher unterschätzten kubischen Wirtsgitter zu verstehen und zu steigern. Hierbei wird die fluorolytische Sol-Gel-Synthese als neuartige Syntheseroute für UCNC vorgestellt. Vorteile wie ausgezeichnete Reproduzierbarkeit, viele Freiheitsgrade bei der Temperaturbehandlung und Partikelgestaltung werden anhand von SrF2 UCNC demonstriert. Die UCNC wurden mittels UCL-Spektren, UCL-Quantenausbeuten, leistungsdichte-abhängiger relativer spektraler Verteilung sowie der Lumineszenzabklingkinetiken unter Einbeziehung kristalliner Eigenschaften wie der Kristallphase, der Kristallitgröße, der Gitterparameter und der Teilchengröße untersucht. Die Abhängigkeit der UCL-Eigenschaften von der Dotierungsmenge wurde mit einer umfassenden Dotierungsreihe beschrieben und der optimale Dotierungsbereich (Yb3+,Er3+) von kleinen, ungeschalten SrF2-UCNC eingegrenzt. Bei der Studie dotierter Kerne mit passivierenden Schalen wurde der Einfluss von Temperaturbehandlung auf die UCL-Mechanismen und die Kern-Schale-Vermischung untersucht. Anhand von unterschiedlich kalzinierten UCNC Pulvern wurde die Empfindlichkeit der UCL gegenüber der Änderung kristalliner Eigenschaften, wie Kristallphase, Kristallinität, und Kristallitgröße betrachtet. Zusammen liefern die Dotierungs-, die Kern-Schale- und die Kalzinierungsstudie wertvolle Einblicke in das gitterspezifische Verhalten der UCL-Eigenschaften als Funktion der Energiemigration und der Kristalleigenschaften. / Upconversion nanocrystals (UCNC) represent a unique type of luminescence reporters that convert near-infrared excitation light into higher energy photons. Tailoring UCNC with specific luminescence properties requires an in-depth understanding of upconversion luminescence (UCL) processes and their dependence on material and particle design. This Ph.D. thesis focuses on the UCL processes of Yb3+,Er3+ doped SrF2-UCNC and aims to understand and enhance the UCL properties of the previously underestimated cubic host lattices. Herein, fluorolytic sol-gel synthesis is introduced as a novel synthetic route for UCNC. Advantages such as excellent reproducibility, high flexibility in temperature treatment and particle design are demonstrated using SrF2 UCNC. The UCNC were characterized by UCL spectra, UCL quantum yields, excitation power density-dependent relative spectral distribution, and luminescence decay kinetics involving crystalline properties such as crystal phase, crystallite size, lattice parameters, and particle size. The dependence of UCL properties on doping amount was described in a comprehensive doping study, and the optimal doping range (Yb3+,Er3+) of small, unshelled SrF2-UCNC was identified. In a core-shell study of doped core UCNC with passivating shells, the influence of temperature treatment on UCL mechanisms and core-shell mixing was investigated. Further, using different calcined UCNC powders, the sensitivity of UCL to the change of crystalline properties, such as crystal phase, crystallinity, and crystallite size, was assessed. Together, the doping, core-shell, and calcination studies provide valuable insight into the lattice-specific behavior of UCL properties as a function of energy migration and crystal properties.

Ce travail de thèse est basé sur l'analyse et l'optimisation des propriétés physiques d'un matériau photonique compatible avec les technologies CMOS. Ce matériau est une matrice de SiO2 contenant des nanoclusters de silicium (nc-Si) et des ions erbium (Er3+). Grâce à un transfert d'énergie entre nc-Si et Er3+, la section efficace d'absorption des ions Er3+ est fortement augmentée. L'objectif est d'optimiser le transfert entre nc-Si et Er3+ afin de maximiser les propriétés d'émission de l'erbium à 1,5 µm. Dans un premier temps les travaux sont axés sur les traitements thermiques pendant et après le dépôt. Ensuite, nous analysons l'effet de l'épaisseur de la couche mince sur les propriétés optiques du matériau et nous montrons que les couches très minces (< 150 nm) présentent un manque de sensibilisateurs qui réduit le nombre d'erbium excités. Nous démontrons alors que ce problème peut être résolu en augmentant la concentration en silicium, augmentant ainsi le nombre de sensibilisateurs au sein des couches les plus minces. Il est ensuite montré que les ions Er3+ bénéficient d'une excitation nanoseconde multi-niveaux par les sensibilisateurs nc-Si. Une deuxième partie du travail de thèse présente la réalisation de diodes électroluminescentes (DELs) et l'optimisation de leur émission à 1,5 µm. Nous montrons qu'épaisseur et excès de silicium doivent êtres choisi conjointement pour l'optimisation des propriétés optiques et électriques des DELs. Une dernière partie montre que les propriétés des DELs peuvent êtres améliorés par l'utilisation de matrices hôtes oxynitrures et nitrures pour les nc-Si et Er3+. Ces travaux ouvrent la voie au développement de DELs efficaces à base de silicium et émettant à 1,5 µm

Two dimensional magnetic recording (TDMR) achieves high areal densities by reducing the size of a bit comparable to the size of the magnetic grains resulting in two dimensional (2D) inter symbol interference (ISI) and very high media noise. Therefore, it is critical to handle the media noise along with the 2D ISI detection. In this paper, we tune the generalized belief propagation (GBP) algorithm to handle the media noise seen in TDMR. We also provide an intuition into the nature of hard decisions provided by the GBP algorithm. The performance of the GBP algorithm is evaluated over a Voronoi based TDMR channel model where the soft outputs from the GBP algorithm are used by a belief propagation (BP) algorithm to decode low-density parity check (LDPC) codes.

[pt] Objetivando suprimir as instabilidades dinâmicas de ganho em um amplificador à fibra dopada com Érbio (EDFA), uma nova configuração de laser em anel é apresentada e demonstrada. Neste trabalho, analizamos os efeitos da variação do nível de atenuação no laço de re-alimentação sobre a resposta transitória do EDFA. Particularmente, observamos as excursões de ganho experimentadas pelo canal sobrevivente quando sete dentre oito canais da rede são adicionados ou removidos, à exemplo do que ocorre em sistemas WDM reais. Sob esta análise, avaliamos o desempenho do sistema em suprimir as instabilidades dinâmicas de ganho do EDFA. / [en] A new ring laser configuration to eliminate the gain dynamic instabilities of an erbium doped fiber amplifier (EDFA) is proposed and demonstrated. We examine the effect of the attenuation level in the optical feedback path over thetransient response of the EDFA. In particular, we look at the transient gain excursions experienced by surviving channel when seven of eight channels are added or dropped, like in real WDM systems. Using this analysis as a guide, we highlight the robustness of the approach and evaluate its performance to EDFA gain stabilization.

Photonics and Optical techniques have advanced recently by a great extend to play an important role in Microwave and Radar applications. Antenna array of modern active phased array radars consist of multiple low power transmit and receive mod- ules. This demands distribution of the various Local Oscillator(LO) signals for up conversion of transmit signals and down conversion of receive signals during various modes of operation of a radar system. Additionally, these receivers require control and clock signals which are digital and low frequency analog, for the synchronization between receive modules. This is normally achieved through RF cables with complex distribution networks which add significantly higher additional weight to the arrays. During radar operations, radio frequency (RF) transmit signal needs to be distributed through the same modules which will in turn get distributed to all antenna elements of the array using RF cables. This makes the system bulky and these large number of cables are prone to Electromagnetic Interference (EMI) and need additional shielding. Therefore it is very desirable to distribute a combination of these RF, analog and digital signals using a distribution network that is less complex, light in weight and immune to EMI. Advancements in Optical and Microwave photonics area have enabled carrying of higher datarate signals on a single fiber due to its higher bandwidth capability including RF signals. This is achieved by employing Wavelength Division Multi- plexing (WDM) that combine high speed channels at different wavelengths. This work proposes, characterizes and evaluates an optical Wavelength Division Multiplexed(WDM) distribution network that will overcome the above mentioned problems in a phased array radar application. The work carries out a feasibility analysis supported with experimental measurements of various physical parameters like am- plitude, delay, frequency and phase variation for various radar waveforms over WDM links. Different configurations of optical distribution network are analyzed for multipoint distribution of both digital and RF signals. These network configurations are modeled and evaluated against various parameters that include power level, loss, cost and component count. A configuration which optimizes these parameters based on the application requirements is investigated. Considerable attention is paid to choose a configuration which does not provide excess loss, which is economically viable, compact and can be realized with minimum component count. After analysing the link configuration, multiplexing density of the WDM link is considered. In this work, since the number of signals to be distributed in radar systems are small, a coarse WDM(CWDM) scheme is considered for evaluation. A comparative study is also performed between coarse and dense WDM (DWDM) links for selection of a suitable multiplexing scheme. These configurations are modeled and evaluated with power budgeting. Even though CWDM scheme does not permit the utilisation of the available bandwidth to the fullest extent, these links have the advantage of having less hardware complexity and easiness of implementation. As the application requires signal distribution to thousands of transmit-receive modules, amplifiers are necessary to compensate for the reduction of signal level due to the high splitting ratio. Introduction of commonly available optical amplifiers like Erbium Doped Fiber Amplifier (EDFA), affect the CWDM channel output powers adversely due to their non-flat gain spectrum. Unlike DWDM systems, the channel separation of CWDM systems are much larger causing significantly high channel gain differences at the EDFA output. So an analysis is carried out for the selection of a suitable wavelength for CWDM channels to minimize the EDFA output power variation. If the gain difference is still significant, separate techniques needs to be implemented to flatten the output power at the antenna end. A CWDM configuration using C-band and L-band EDFAs is proposed and is supported with a feasibility analysis. As a part of evaluation of these links for radar applications, a mathematical model of the WDM link is developed by considering both the RF and digital sig- nals. A generic CWDM system consisting of transmitters, receivers, amplifiers, multiplexers/ demultiplexers and detectors are considered for the modeling. For RF signal transmission, the transmitters with external modulators are considered. Mod- eling is done based on a bottom-top approach where individual component models are initially modeled as a function of input current/power and later cascaded to obtain the link model. These models are then extended to obtain the wavelength dependent model ( spectral response) of the hybrid signal distribution link Further mathematical analysis of the developed link model revealed its variable separable nature in terms of the input power and wavelength. This led to significant reduction in the link equation complexity and development of some approximation techniques to easily represent the link behavior. The reduced form of the link spectral model was very essential as the initially developed wavelength model had a lot of parametric dependency on the component models. This mathematical reduction process led to simplification of the spectral model into a product of two independent functions, the input current and wavelength. It is also noticed that the total link power within specific wavelength range can be obtained by the integrating these functions over a specific link input power. After the mathematical modelling, an experimental prototype physical link is set up and characterized using various radar signals like continuous wave (CW) RF, pulsed RF, non linear frequency modulated signal (NLFM) etc. Additionally a proof of concept Radio-Over-Fiber (RoF) link is established to prove the superior transmission of microwave signal through an optical link. The analysis is supported with measurements on amplitude, delay, frequency and phase variations. The NLFM waveforms transmissions are further analysed using a matched _ltering process to confirm the side lobe requirement. Further a prototype WDM link is built to study the performance when digitally modulated channels are also multiplexed into the link. The link is again validated for signal levels, delay, frequency and phase parameters. Since amplitude and delay are deterministic, it is proposed that these parameter variations can be compensated by using suitable components either in the electrical or the optical domain. Radar systems use low frequency digital signals of different duty-cycles for synchronization and control across various transmit-receive modules. In the proposed link, these digital signals also modulate a WDM channel and hence the link is called a hybrid system. As the proposed link has EDFA to compensate for the splitting losses, there are chances of transient effects at the EDFA output for these low bitrate channels. Owing to the long carrier lifetime, low bitrate digital channels are prone to EDFA transient effects under specific signal and pump power conditions. Additionally, the synchronization signals used in radar application vary the duty-cycle over time, which is found to introduce variations in transient output. This practical challenge is further studied and the thesis for the first time, includes an analysis of EDFA transient e_ects for variable duty-cycle pulsed signals. The analysis is carried out for various parameters like bitrate, input power, pump power and duty-cycle. Investigations on EDFA transients on variable duty-cycle signals help in proposing a viable method to predict the lower duty-cycle transients from higher duty-cycle transients. The predicted transients were again validated against simulated transients and experimental results. As these transient effects are not desirable for radar signals, we propose a novel transient suppression techniques in optical and electrical domain which are validated with simulation and experimental measures. One suppression technique tries to avoid transient effect by keeping the optical input to EDFA always constant by feeding an inverted version of the original pulse into the EDFA along with the actual pulse. It is observed that as the wavelength of the inverted pulse is closer to the original input pulse, the transient effect settles faster. These EDFA transients are evaluated with WDM link configurations, where both high and low bitrate signals are co-propagated. Another challenging aspect of the link operation is the non-at gain spectrum of EDFA. i.e., EDFA provides unequal power level for various signals at WDM link output. This is especially true in the case of local oscillator signals, where it is preferable to have the same amplitude signals before feeding it to the mixer stages. But in the radar applications, this will require additional hardware circuits to equalize the signal level within a phased array antenna. This work also proposes some of the power equalization methods that can be used along with the WDM links. This part of the work is also supported with simulation model and experimental results. The analytical and experimental study of this thesis aids the evaluation process of a suitable optical Wavelength Division Multiplexed(WDM) distribution network that can be used for the distribution of both RF and digital signals. The optical WDM links being superior with its light weight, less loss and EMI/ EMC immunity provides a better solution to future class of radars.

Made available in DSpace on 2016-03-15T19:38:52Z (GMT). No. of bitstreams: 1 Henrique Guimaraes Rosa.pdf: 6486656 bytes, checksum: 6fc9fe7875ed2cceb89c9f5179b9f028 (MD5) Previous issue date: 2015-02-25 / Fundo Mackenzie de Pesquisa / In this thesis, we present results on the fabrication, transfer and characterization of chemical vapor deposition (CVD) graphene and exfoliated graphene over glass and optical fiber substrates, to study optical properties of graphene and its application as a saturable absorber for Erbium-doped fiber laser (EDFL). Monolayer CVD graphene and stacked CVD graphene samples were fabricated and characterized, transferred to the transverse face of optical fibers, and a study on the relation between the optical properties of graphene samples and the properties of ultrashort laser pulses generated in (EDFL) was performed. Furthermore, we have developed a technique for transferring exfoliated nanomaterials which allowed us to transfer exfoliated graphene onto optical fiber s faces and align the graphene flake to the fiber core. With this transfer technique it is possible to fabricate samples with controlled number of graphene layers onto optical fiber faces. As application, we demonstrate ultrashort pulse generation in Erbium-doped fiber laser with exfoliated monolayer graphene samples as saturable absorber. This is the first time that ultrashort laser pulses are generated with a single exfoliated monolayer graphene sample. / Nesta tese, apresentamos resultados sobre a fabricação, transferência e caracterização de grafeno CVD (grafeno fabricado por deposição química de vapor chemical vapour deposition) e de grafeno esfoliado em substratos de vidro e em fibras ópticas, para o estudo das propriedades ópticas do grafeno e sua aplicação como absorvedor saturável em laser à fibra dopada com Érbio (EDFL). Foram fabricadas e caracterizadas amostras de grafeno CVD monocamada e de grafeno CVD empilhado, transferidas para a face transversal de fibras ópticas, e com estas amostras foram feitos estudos sobre a relação entre as propriedades ópticas do grafeno e as propriedades de pulsos ultracurtos gerados em EDFL. Além disto, desenvolvemos uma técnica para a transferência de nanomateriais esfoliados que permitiu a transferência de grafeno esfoliado para fibras ópticas e seu alinhamento com o núcleo da fibra. Com esta técnica de transferência é possível fabricar amostras com controlado número de camadas de grafeno em fibra óptica. Como aplicação, demonstramos a geração de pulsos em EDFL com uma amostra de grafeno esfoliado monocamada como absorvedor saturável. Esta é a primeira vez que pulsos ultracurtos são gerados em lasers à fibra com amostra de grafeno esfoliado de uma única camada sobre a face transversal da fibra óptica.

Chalcogenide glasses are highly nonlinear optical materials which can be used for fabricating active and passive photonic devices. This thesis work deals with the fabrication of buried, three dimensional, channel waveguides in chalcogenide glasses, using ultrafast laser inscription technique. The femtosecond laser pulses are focused into rare earth ions doped and undoped chalcogenide glasses, few hundred microns below from the surface to modify the physical properties such as refractive index, density, etc. These changes are made use in the fabrication of active and passive photonic waveguides which have applications in integrated optics. The first chapter provides an introduction to the fundamental aspects of femtosecond laser inscription, laser interaction with matter and chalcogenide glasses for photonic applications. The advantages and applications of chalcogenide glasses are also described. Motivation and overview of the present thesis work have been discussed at the end. The methods of chalcogenide glass preparation, waveguide fabrication and characterization of the glasses investigated are described in the second chapter. Also, the details of the experiments undertaken, namely, loss (passive insertion loss) and gain measurements (active) and nanoindentation studies are outlined. Chapter three presents a study on the effect of net fluence on waveguide formation. A heat diffusion model has been used to solve the waveguide cross-section. The waveguide formation in GeGaS chalcogenide glasses using the ultrafast laser, has been analyzed in the light of a finite element thermal diffusion model. The relation between the net fluence and waveguide cross section diameter has been verified using the experimentally measured properties and theoretically predicted values. Chapter four presents a study on waveguide fabrication on Er doped Chalcogenide glass. The active and passive characterization is done and the optimal waveguide fabrication parameters are given, along with gain properties for Er doped GeGaS glass. A C-band waveguide amplifier has been demonstrated on Chalcogenide glasses using ultrafast laser inscription technique. A study on the mechanical properties of the waveguide, undertaken using the nanoindentation technique, is presented in the fifth chapter. This work brings out the close relation between the change in mechanical properties such as elastic modulus and hardness of the material under the irradiation of ultrafast laser after the waveguide formation. Also, a threshold value of the modulus and hardness for characterizing the modes of the waveguide is suggested. Finally, the chapter six provides a summary of work undertaken and also discusses the future work to be carried out.

This thesis deals with theoretical description of fiber-optic angular velocity sensors, or gyroscopes, and further with design and construction of own sensor of this type. The theoretical part describes problematics of interferometric and resonant fiber-optic gyroscopes. Basic principles and physical limits are described for both types. The main focus is then put on analysis of possible conceptions of these sensors. Solutions using different optical configurations as well as various modulation and signal processing schemes are discussed. The practical part deals with design and construction of own interferometric fiber-optic gyroscope in closed-loop configuration. The gyroscope utilizes all-fiber components including piezoelectric phase modulator and unexpensive single-mode fiber, which are commonly used only for open-loop configurations. To realize closed-loop operation, special modulation scheme based on fully harmonic signal was develeped, which yields linear output within wide dynamic range. This type of modulation requires high level of synchronization achieved by using a field-programmable gate array module. The gyroscope utilizes powerful broadband fiber source, polarizer and Lyot depolarizer which ensure good reciprocity of whole architecture. The parameters of the sensor, obtained by measurement, are even comparable to some sensors using PM fiber, which is much more expensive.

[pt] Diferentes tipos de lasers podem ser usados para gerar pulsos de luz com uma ampla faixa de durações de pulso, energias e potências de pico. As técnicas de Q-switching e mode-locked são relatadas há anos por vários autores e pesquisadores e são frequentemente utilizadas na geração de lasers de pulso ultracurto com duração de pulso no domínio do tempo na faixa de nanossegundos até femtossegundos. Uma configuração, com ganho fornecido por um amplificador óptico semicondutor (SOA) e amplificador de fibra dopada com érbio (EDFA) é proposta e emprega a técnica de gerenciamento de dispersão para gerar um trem de pulsos ópticos exibindo alta potência de pico, taxa de repetição ultra-baixa e largura temporal curta, habilitando que este laser seja usado como uma fonte para aplicações de alta resolução em reflectometria óptica no domínio do tempo (OTDR). A operação mode-locked é conhecida por ocorrer apenas em lasers ordenados padrão por um longo tempo e até recentemente foi encontrado também em lasers de fibra aleatórios desordenados (RFL). Embora tenha havido progresso no sentido de travar modos espaciais e longitudinais em lasers aleatórios, a literatura carece de relatos sobre geração de pulsos limitada por transformada de Fourier, apesar das muitas décadas de campo. O autor demonstra experimentalmente um mode-locked random fiber laser (MLRFL) operando como um refletômetro óptico de domínio do tempo sensível à fase. Aqui, a saída total do laser fornece o sinal de detecção, em contraste com o pequeno sinal retroespalhado medido em um OTDR convencional. O laser opera como um sensor acústico distribuído (DAS) e sensor de temperatura distribuído (DTS). / [en] Different types of lasers can be used to generate light pulses with a wide range of pulse durations, energies and peak powers. Q-switching and mode-locked techniques have been reported for years by several authors and researchers and are frequently used in the generation of ultra-short-pulse lasers with time-domain pulse durations from the nanosecond to femtosecond range. A configuration, with gain provided by a semiconductor optical amplifier (SOA) and erbium-doped fiber amplifier (EDFA) is proposed and employ the dispersion management technique to generate a train of optical pulses exhibiting high-peak-power, ultralow repetition rate, and fast temporal width, enabling this laser to be used as a source for high-resolution optical time domain reflectometer (OTDR) applications. The mode-locking operation has been known to occur only in standard ordered lasers for a long time and until recently it was found to also occur in disordered random fiber lasers (RFL). Although progress has been made towards locking spatial and longitudinal modes in random lasers, the literature lacks reports on Fourier transform-limited pulse generation despite the many decades of the field. The author experimentally demonstrates a mode-locked random fiber laser (MLRFL) operating as a lasing phase-sensitive optical time domain reflectometer based on random feedback from a sensing fiber. Here, the full output of the laser provides the sensing signal, in contrast to the small backscattered signal measured in a conventional OTDR. The laser operates as a distributed acoustic sensor (DAS) and distributed temperature sensor (DTS).