La capacité et les performances des systèmes à fibres optiques et photoniques dépendent fortement du bruit et des non-linéarités des amplificateurs optiques. Dans ce contexte, les amplificateurs paramétriques à fibre optique (PS-FOPA), reposant sur le mélange à quatre ondes dans les fibres optiques, surpassent les amplificateurs conventionnels insensibles à la phase. En effet, leur sensibilité à la phase peut être exploitée pour l'amplification sans bruit et al compensation de la distortion non-linéaire. En conjonction avec leur large spectre de gain et d'autres fonctionnalités telles que la conversion de longueur d'onde, ils sont considérés comme des candidats prometteurs pour la prochaine génération d'amplificateurs optiques pour les communications et le traitement de signal tout optiques.Le PS-FOPA est classiquement décrit par les équations d'ondes couplées, dérivées de l'équation de Schrödinger non-linéaire qui ne contiennent que trois ou quatre ondes en interaction. Cependant, dans un cas plus général, l'apparition de fréquences supplementaires d’ordre plus élevé affectera inévitablement la sensibilité en phase.L’objectif de cette thèse est d'étudier de manière approfondie les propriétés non-linéaires, en termes de gain et de sensibilité en phase d'un PS-FOPA dégénéré à double pompe pour différentes configurations. Une analyse numérique plus précise est obtenue en utilisant le modèle à 7 ondes qui incorpore les ondes supplémentaires issues de processus de mélange à quatre ondes d’ordre élevé. Ce modèle permet de donner une interprétation physique plus précise des interactions multi-ondes en fonction des conditions d'accord de phase, révélant les relations sous-jacentes entre la dispersion et la sensibilité de phase. De plus, la capacité de régénération simultanée de phase et d'amplitude d'un PS-FOPA basique est évaluée pour l'optimisation globale. Il permet d'exploiter pleinement la capacité potentielle d'un PS-FOPA de base agissant comme un bloc de construction fondamental des futures fonctionnalités tout optiques. L’approche basée sur ce modèle permet une optimisation orientée vers l’application et revêt une importance particulière pour la conception et l'optimisation de tels PS-FOPAs dans divers scénarios. / The capability and performance of the widely deployed fiber-optic and photonic systems strongly depend on the noise and nonlinearities of the optical amplifiers. In this context, phase-sensitive fiber-optic parametric amplifiers (PS-FOPAs), relying on four-wave mixing in optical fibers, outperforms conventional phase-insensitive amplifier thanks to the unique phase-sensitivity that can be exploited for noiseless amplification and mitigation of the nonlinear impairment. In conjunction with the vast gain spectrum and other functionality such as wavelength conversion, they have been regarded as a promising candidate for the next generation optical amplifiers towards all-optical communication and processing.The PS-FOPA is conventionally described by the fundamental coupled wave equations derived from the nonlinear Schrödinger equation that contains only three or four interacting waves. However, for a more general case, the emergence of high-order waves will inevitably affect the phase-sensitivity. The objective of this thesis aims at the thorough investigation of the nonlinear properties in terms of the gain properties and the phase sensitivities with respect to different configurations of a dual-pump signal-idler degenerate PS-FOPA. The more accurate numerical analysis is obtained by using the 7-wave model that incorporates the first order high-order waves stemming from the high-order four-wave mixing processing. This model permits to assess a more precise physical interpretation of the multi-wave interactions based on phase matching conditions, revealing the underlying relations between the dispersion and the phase-sensitivity. Moreover, the simultaneous phase and amplitude regenerative capability of a basic PS-FOPA is evaluated for the overall optimization. It allows fully exploiting the potential ability of a basic PS-FOPA acting as a fundamental building block of the future all-optical functionalities. The analysis approach based on this model permits application-oriented optimization and is of particular guiding significance for design and optimization of PS-FOPA in various scenarios.

No description available.

Methoden der transienten Vierwellenmischung mit Femtosekunden-Zeitauflösung werden angewendet, um die Phasen- und Energierelaxation optisch selektiv erzeugter Anregungszustände im Lichtsammelkomplex II höherer Pflanzen (LHC II) sowie im Halbleiter Cadmiumsulfid (CdS) bei verschiedenen Temperaturen zu untersuchen. Für den LHC II werden die Ergebnisse der Messungen des zeitaufgelösten und integrierten Zweipuls-Photonenechos mit Resultaten aus Pump-Test-Experimenten verglichen, um unter Einbeziehung von Literaturdaten Rückschlüsse über den Charakter der phasenzerstörenden Prozesse zu ziehen und Zusammenhänge zu Strukturdaten des Komplexes aufzudecken. Die vorliegende Arbeit liefert erstmals einen systematischen Überblick über die totalen Phasenrelaxationszeiten T2 im Bereich der Qy-Bande des LHC II von 640 bis 685 nm bei 5 K. Das bei 5 K beobachtete Photonenechosignal am LHC II zeigt in Abhängigkeit von der Verzögerung der beiden Anregungsimpulse ein multiexponentielles Abklingen, das auf die Überlagerung der Einflüsse mehrerer Relaxationsprozesse zurückgeführt wird. Dabei lassen sich drei charakteristische Bereiche der Phasenrelaxationszeit unterscheiden, die verschiedenen phasenzerstörenden Prozessen zugeordnet werden. Ein Vergleich mit Resultaten aus Pump-Test-Experimenten führt zu der Schlußfolgerung, daß die Phasenrelaxation im LHC II bei 5 K für Wellenlängen £ 675 nm im wesentlichen durch den Energietransfer auf einer sub-ps Zeitskala bestimmt wird. Für Wellenlängen > 675 nm steigt die Phasenrelaxationszeit stark an und wird insbesondere im Bereich der tiefsten Anregungszustände um 680 nm durch reine Phasenzerstörung dominiert. Ab 20 K setzt bei dieser Wellenlänge ein zusätzlicher phasenzerstörender Prozeß ein, der mit steigender Temperatur zu einem mäßigen linearen Anstieg der Phasenrelaxationsrate (T2)-1 führt. Die Ursache ist vermutlich ein Aufwärts-Energietransfer. Im Bereich der Chlorophyll a-Absorption vernichten außerdem (physiologisch irrelevante) Multiexzitoneneffekte die Kohärenz der angeregten Zustände, verursacht durch die hohe Anregungsintensität. Zusammenfassend kann festgestellt werden, daß die Erhaltung der Kohärenz für die Funktionalität des LHC II eine untergeordnete Rolle spielt. Die wesentlichen Prozesse sind der schnelle räumliche Energietransfer und die Energierelaxation auf das Niveau des primären Elektrondonators P680 im Reaktionszentrum. Am Halbleiter CdS wird erstmals ein mittels Zwei-Photonen-Absorption angeregtes Photonenecho beschrieben, das in Abhängigkeit von der Wellenlänge charakteristische Quantenbeats mit einer Periode von 700 bis 800 fs zeigt. Das stark gedämpfte periodische Echosignal tritt sowohl für positive als auch für negative Verzögerungszeiten t der Anregungsimpulse auf, wobei die Abklingzeit für t>0 mit 170±10 fs doppelt so groß ist wie für t / Transient four-wave-mixing experiments with femtosecond resolution are performed in order to investigate phase and energy relaxation processes of optically excited states in the light harvesting complex II of higher plants (LHC II) and in the semiconductor cadmium sulfide (CdS) at different wavelengths and temperatures. Extensive studies of the time resolved and integrated two-pulse photon echo on LHC II are combined with pump-probe experiments. Results of both methods together with literature data are used to characterize the nature of dephasing processes and to reveal connections with structural data of the complex. This study gives the first systematic survey of total dephasing times T2 in the spectral region of the Qy-absorption band of LHC II from 640 to 685 nm at 5 K. In the case of LHC II, the photon echo signal at 5 K monitored as a function of delay between both excitation pulses shows a multi-exponential decay which is attributed to the superposition of several relaxation processes. Three characteristic dephasing time domains can be distinguished, ascribed to different dephasing processes. Comparing photon echo and pump-probe results it can be concluded that dephasing in LHC II at 5 K and for wavelengths £ 675 nm is dominated by the fast excitation energy transfer on a sub-ps time scale. At wavelengths > 675 nm the total dephasing time increases drastically. The loss of coherence of the lowest excited states around 680 nm at 5 K is mainly determined by pure dephasing. An additional dephasing process, probably uphill energy transfer, occurs at temperatures higher than 20 K leading to a moderate linear rise of the dephasing rate (T2)-1 with increasing temperature. Furthermore, the dephasing in the spectral region of chlorophyll a absorption is affected by (physiologically irrelevant) multi-excitonic effects caused by the high excitation energy. In summary, it can be concluded that the preservation of coherence plays a minor role in the functionality of LHC II. The main processes are the fast spatial excitation energy transfer and the energy relaxation down to the energetic level of the primary electron donor P680 of the reaction center. Investigations of four-wave mixing signals of the semiconductor CdS resulted in the first description of a two-photon excited photon echo in CdS showing characteristic quantum beats with a period of 700 to 800 fs in dependence on wavelength. The strongly damped periodical echo signal is found for both positive and negative delay times t between the excitation pulses. The decay time for t>0 amounts to 170±10 fs and is twice as large as for t

Este trabalho investiga a influência dos efeitos da mistura de quatro ondas (FWM -. Four- Wave Mixing) durante o estabelecimento de caminhos ópticos em uma rede GMPLS/DWDM (Generalized MultiProtocol Label Switching/Dense Wavelength Division Multiplexing). Para minimizar seus impactos e assim reduzir o bloqueio total da rede, são propostas estratégias por meio de extensões ao RSVP-TE (Resource Reservation Protocol-Traffic Engineering). A partir de uma plataforma de simulação que emula e integra a camada óptica e o Plano de Controle da arquitetura GMPLS, a solução proposta é validada. Os vários métodos de alocação de comprimentos de onda são avaliados e as contribuições na redução dos sinais interferentes quantificadas. Para minimizar o bloqueio na rede, uma nova classe de métodos de alocação, classificados como híbridos, é definida e seu desempenho é avaliado. O bloqueio da rede em função da variação de parâmetros que reforçam a eficiência da FWM, tais como a quantidade e espaçamento entre canais, é observado, confirmando a necessidade de métodos de alocação especialistas para gerenciamento do QoS (Quality of Service) óptico em cada caminho a ser estabelecido. Os resultados obtidos comprovam a eficiência da solução proposta e evidenciam a necessidade de um Plano de Controle GMPLS com acurada visibilidade das restrições da camada física, tanto os efeitos estáticos quanto às degradações associadas ao comportamento dinâmico da rede. / This work investigates the effects of the Four-Wave Mixing (FWM) during the lightpaths establishment in a GMPLS/DWDM (Generalized Multiprotocol Label Switching / Dense Wavelength Division Multiplexing) network. In order to minimize their impacts and thus reduce network blocking, strategies based on RSVP-TE (Resource Reservation Protocol- Traffic Engineering) extensions are proposed. The proposed solution is validated using a simulation platform that integrates and emulates the optical layer and the GMPLS Control Plane. The various wavelength allocation methods are evaluated and their contributions on the reduction of interfering signals are quantified. In order to minimize the network blocking, a new class of allocation methods (classified as hybrids) is defined and its performance is evaluated. The network blocking caused by the parameters that enhance the FWM efficiency, such as the number and spacing between channels, is also observed, confirming the need of specialized wavelength allocation methods for managing the optical QoS for each lightpath. The results obtained ascertain that the proposed solution is efficient and demonstrate the need for a GMPLS Control Plane endowed with an accurate visibility of the restrictions in the physical layer, concerning both static and dynamic effects.

Estudamos o efeito de Conjugação de Fase por Mistura Degenenerada de Quatro Ondas em Al2O3:Cr+3 (Rubi) e GdAlO3:Cr+3 com um laser de Ar (λ=5145 Å). Obtivemos eficiência aproximadamente quatro vezes maior no GdAlO3:Cr+3 (onde este trabalho é original) que no Rubi o que nos motivou a investigar as propriedades físicas que são relevantes para o fenômeno nestes sistemas (isto não foi bem compreendido no trabalho anterior em Rubi). Desenvolvemos um método interferométrico muito sensível para medida dos coeficientes não lineares do índice de refração n2 destes materiais (que não eram conhecidos) Com estes valores de n2 calculamos a eficiência de Conjugação de Fase em bom acordo com experiência. / We have studied the effect of Phase Conjugation by Degenerate Four Wave Mixing in Al2O3:Cr+3 (Rubi) and GdAlO3:Cr+3 with an Ar (λ=5145 Å). We obtained efficiency ?approximately? 4 times greater in GdAlO3:Cr+3 (where this work is original) than in Rubi and this have motivated us to investigate the physical properties that are important to explain this phenomenon in these materials (what wasn\'t well understood in the previous paper on Rubi(10)). We developed an interferometric method very sensitive to measure the nonlinear coeficient of refractive index n2 of these materials (what wasn\'t known). With these values of n2 we calculated the efficiency of the Phase Conjugation in good agreement with the experience.

Este trabalho investiga a influência dos efeitos da mistura de quatro ondas (FWM -. Four- Wave Mixing) durante o estabelecimento de caminhos ópticos em uma rede GMPLS/DWDM (Generalized MultiProtocol Label Switching/Dense Wavelength Division Multiplexing). Para minimizar seus impactos e assim reduzir o bloqueio total da rede, são propostas estratégias por meio de extensões ao RSVP-TE (Resource Reservation Protocol-Traffic Engineering). A partir de uma plataforma de simulação que emula e integra a camada óptica e o Plano de Controle da arquitetura GMPLS, a solução proposta é validada. Os vários métodos de alocação de comprimentos de onda são avaliados e as contribuições na redução dos sinais interferentes quantificadas. Para minimizar o bloqueio na rede, uma nova classe de métodos de alocação, classificados como híbridos, é definida e seu desempenho é avaliado. O bloqueio da rede em função da variação de parâmetros que reforçam a eficiência da FWM, tais como a quantidade e espaçamento entre canais, é observado, confirmando a necessidade de métodos de alocação especialistas para gerenciamento do QoS (Quality of Service) óptico em cada caminho a ser estabelecido. Os resultados obtidos comprovam a eficiência da solução proposta e evidenciam a necessidade de um Plano de Controle GMPLS com acurada visibilidade das restrições da camada física, tanto os efeitos estáticos quanto às degradações associadas ao comportamento dinâmico da rede. / This work investigates the effects of the Four-Wave Mixing (FWM) during the lightpaths establishment in a GMPLS/DWDM (Generalized Multiprotocol Label Switching / Dense Wavelength Division Multiplexing) network. In order to minimize their impacts and thus reduce network blocking, strategies based on RSVP-TE (Resource Reservation Protocol- Traffic Engineering) extensions are proposed. The proposed solution is validated using a simulation platform that integrates and emulates the optical layer and the GMPLS Control Plane. The various wavelength allocation methods are evaluated and their contributions on the reduction of interfering signals are quantified. In order to minimize the network blocking, a new class of allocation methods (classified as hybrids) is defined and its performance is evaluated. The network blocking caused by the parameters that enhance the FWM efficiency, such as the number and spacing between channels, is also observed, confirming the need of specialized wavelength allocation methods for managing the optical QoS for each lightpath. The results obtained ascertain that the proposed solution is efficient and demonstrate the need for a GMPLS Control Plane endowed with an accurate visibility of the restrictions in the physical layer, concerning both static and dynamic effects.

Semiconductor nanostructures exhibit many remarkable electronic and optical properties. The key to designing and utilising semiconductor quantum structures is a physical understanding of the detailed excitation, transport and energy relaxation processes. Thus the nonequilibrium dynamics of semiconductor quantum structures have attracted extensive attention in recent years. Ultrafast spectroscopy has proven to be a versatile and powerful tool for investigating transient phenomena related to the relaxation and transport dynamics in semiconductors. In this thesis, we report investigations into the electronic and optical properties of various semiconductor quantum systems using a variety of ultrafast techniques, including up-conversion photoluminescence, pump-probe, photon echoes and four-wave mixing. The semiconductor quantum systems studied include ZnO/ZnMgO multiple quantum wells with oxygen ion implantation, InGaAs/GaAs self-assembled quantum dots with different doping, InGaAs/InP quantum wells with proton implantation, and silicon quantum dots. The spectra of these semiconductor nanostructures range from the ultraviolet region, through the visible, to the infrared. In the UV region we investigate excitons, biexcitons and oxygen implantation effects in ZnO/ZnMgO multi-quantum wells using four-wave mixing, pump-probe and photoluminescence techniques. Using time-resolved up-conversion photoluminescence, we investigate the relaxation dynamics and state filling effect in InGaAs self-assembled quantum dots with different doping, and the implantation effect in InGaAs/InP quantum wells. Finally, we study the optical properties of silicon quantum dots using time-resolved photoluminescence and photon echo spectroscopy on various time scales, ranging from microseconds to femtoseconds.

Dans cette thèse, nous présentons une recherche expérimentale détaillée des lois caract érisant la taille et la densité atomique de grands pièges magnéto-optiques contenant typiquement 10^10 atomes. Il est bien connu que les interactions induites par laser, produites par la diffusion multiple, y constituent la limitation principale à l'obtention de densités atomiques élevées. Pour contourner cette limite, nous avons réalisé un piège magnéto-optique 'noir' combiné un piège dipolaire. Les forces de répulsion assistées par la lumière ont ainsi pu être considérablement réduites et une augmentation de près de deux ordres des grandeur sur le densité (jusqu'à 10^12 atm/cm^3) a été corrélativement observée. Nous avons proposé également un modèle qualitatif pour rendre compte de la densité atomique (et également pour la taille) o`u nous avons introduit les interactions mécaniques induites par les photons du laser repompeur quand son épaisseur optique devient importante. Ce travail fait partie d'un projet de recherche qui vise à terme à observer la localisation d'Anderson de la lumière dans des nuages d'atomes froids, mais pour laquelle des densités de l'ordre (10^13-10^14 atm/cm^3) sont requises. Pour des densités inférieures, des mécanismes de gain peuvent toutefois être mis à profit pour amplifier des modes pre-localisés. Nous avons, ainsi, mis en place une technique pompe-sonde pour la création d'un gain dans l'échantillon atomique. La configuration des pompes a autorisé l'observation du signal de mélange a quatre ondes. Une analyse détaillée de ses propriétés et de ses caractéristiques a été présentée. En particulier, une asymétrie rouge-bleu dans le spectre, pour des intensités élevées des faisceaux pompes, a été mise en évidence. Cette asymétrie a pu tre expliquée en tenant compte du regroupement des atomes dans les noeuds ou dans les ventres de l'onde stationnaire des faisceaux de pompe.

Bose-Einstein condensation (BEC) is a phenomenon in which identical bosons occupy the same quantum state below a certain critical temperature. A hallmark of BEC is the coherence between particles every particle shares the same quantum wavefunction and phase. This coherence has been demonstrated for the external (motional) degrees of freedom of the atomic condensates by interfering two condensates. In this thesis, the coherence is shown to extend to the internal spin degrees of freedom of a spin-1 Bose gas evidenced by the observed coherent and reversible spin-changing collisions. The observed coherent dynamics are analogous to Josephson oscillations in weakly connected superconductors and represent a type of matter-wave four-wave mixing. Control of the coherent evolution of the system using magnetic fields is also demonstrated. The studies on spinor condensates begin by creating spinor condensates directly using all-optical approaches that were first developed in our laboratory. All-optical formation of Bose-Einstein condensates (BEC) in 1D optical lattice and single focus trap geometries are developed and presented. These techniques offer considerable flexibility and speed compared to magnetic trap approaches, and the trapping potential can be essentially spin-independent and are ideally suited for studying spinor condensates. Using condensates with well-defined initial non-equilibrium spin configuration, spin mixing of F = 1 and F = 2 spinor condensates of rubidium-87 atoms confined in an optical trap is observed. The equilibrium spin configuration in the F = 1 manifold confirms that 87Rb is ferromagnetic. The coherent spinor dynamics are demonstrated by initiating spin mixing deterministically with a non-stationary spin population configuration. Finally, the interplay between the coherent spin mixing and spatial dynamics in spin-1 condensates with ferromagnetic interactions is investigated.

The coherent properties of semiconductor nanostructures are inherently difficult to measure and one-dimensional spectroscopies are often unable to separate inhomogeneous and homogeneous linewidths. We have refined and improved a method of performing multidimensional Fourier transform spectroscopy based on four-wave mixing (FWM) experiments in the box geometry. We have modified our system with broadband beamsplitters in all interferometer arms, high-resolution translation stages and the ability to work in reflection geometry. By improving the phase-stability of our setup and scanning pulse delays with sub-optical cycle precision, we are able to reproduce 2DFT spectra of GaAs multiple quantum wells. With the FWM signal reflected from the sample surface instead of transmitted through, we show that very low pulse powers can be used to generate coherent 2D signals from colloidal PbS quantum dots. Dephasing times are particularly difficult to measure in small colloidal quantum dots due to environmental broadening effects from the colloidal growth. We show that low-temperature pure excitonic dephasing can be measured via time-integrated measurements as well as from the cross-diagonal linewidths of 2DFT spectra. Ultrafast sub-picosecond dephasing times are measured at 5 K in 3 nm PbS quantum dots, while excitation-density-dependence is investigated in these dots. By retrieving the global phase with an all-optical method, we are able to retrieve the real-part 2D spectra of PbS quantum dots.