La mécanique quantique est sans aucun doute la théorie la mieux vérifiée qui n’a jamais existée. En se retournant vers le passé, nous constatons qu’un siècle de théorie quantique a non seulement changé la perception que nous avons de l’univers dans lequel nous vivons mais aussi est responsable de plusieurs concepts technologiques qui ont le potentiel de révolutionner notre monde. La présente dissertation a pour but de mettre en avance ces potentiels, tant dans le domaine théorique qu’expérimental. Plus précisément, dans un premier temps, nous étudierons des protocoles de communication quantique et démontrerons que ces protocoles offrent des avantages de sécurité qui n’ont pas d’égaux en communication classique. Dans un deuxième temps nous étudierons trois problèmes spécifiques en clonage quantique ou chaque solution apportée pourrait, à sa façon, être exploitée dans un problème de communication quantique. Nous débuterons par décrire de façon théorique le premier protocole de communication quantique qui a pour but la distribution d’une clé secrète entre deux parties éloignées. Ce chapitre nous permettra d’introduire plusieurs concepts et outils théoriques qui seront nécessaires dans les chapitres successifs. Le chapitre suivant servira aussi d’introduction, mais cette fois-ci penché plutôt vers le côté expériemental. Nous présenterons une élégante technique qui nous permettra d’implémenter des protocoles de communication quantique de façon simple. Nous décrirons ensuite des expériences originales de communication quantique basées sur cette technique. Plus précisément, nous introduirons le concept de filtration d’erreur et utiliserons cette technique afin d’implémenter une distribution de clé quantique bruyante qui ne pourrait pas être sécurisé sans cette technique. Nous démontrerons ensuite des expériences implémentant le tirage au sort quantique et d’identification quantique. Dans un deuxième temps nous étudierons des problèmes de clonage quantique basé sur le formalisme introduit dans le chapitre d’introduction. Puisqu’il ne sera pas toujours possible de prouver l’optimalité de nos solutions, nous introduirons une technique numérique qui nous permettra de mettre en valeur nos résultats.

The aim of this project was to improve the imaging performance of planar superlenses for evanescent near-field lithography. An experimental investigation of the performance of superlenses with reduced surface roughness was proposed. Such an investigation poses significant requirements in regards to process control in thin film deposition of silver onto dielectric substrates. Thin film deposition of silver films, onto silicon dioxide substrates, achieved films with root mean square surface roughness as low as 0.8 nm. While these experiments provided good understanding of the deposition process, significant variability of the surface roughness parameter remained an issue. The diffculty of achieving consistent control of surface roughness led to a finite element method simulation study where this parameter could be readily controlled. An improved understanding of how surface roughness affects superlens imaging performance was obtained from the results of this investigation. Furthermore, it was shown that in order to conduct an experimental investigation to verify the simulation results, it would be necessary to improve the imaging capability of super-resolution lithography protocols to achieve 3σ line edge roughness (LER) of <20 nm. Resist-scheme optimisation was identied as an important factor in this regard. Thus, a novel calixarene-based photoresist was formulated and characterised. The resist demonstrated superior imaging capabilities through interference lithography and evanescent near-field optical lithography, capable of resolving 250-nm period half-pitch line gratings with 3σ LER below 10 nm. The development of this novel photoresist will enable future lithographical investigations to be conducted with improved resolution and imaging fidelity.

Directly modulated lasers (DMLs) are a low cost solution for moderate reach systems due to their small footprint, low power dissipation and high output optical power. However, commercial 10-Gb/s on-off keying DMLs have been limited by an inherent modulation of the optical phase that accompanies the desired modulation of the optical intensity, which reduces their transmission distance to below 20 km. Furthermore, the ability to generate bit rates beyond 10-Gb/s with advanced modulation formats has been limited by the strict requirements on the laser drive current. The primary objective of this research is to dramatically enhance the capability of DML based systems through precise control over the drive current. This is achieved by digital signal processing (DSP) combined with a single digital-to-analog converter (DAC). In this research, a novel method to pre-compensate dispersion for metro and regional networks is demonstrated at 10.709-Gb/s. A look-up table (LUT) for the driving current is optimized for dispersion mitigation. Experimental results show a 25 fold increase in the transmission reach of a DML from 10 km to 252 km. A similar approach applied to a directly modulated chirp managed laser reveals a remarkable increase in the achievable transmission reach from 200 km to 608 km. In the context of access networks the DSP and DAC configuration is utilized for directly modulating a passive feedback laser (PFL) to generate differential phase shift keying (DPSK) signals at bit rates of 10.709-Gb/s, 14-Gb/s and 16-Gb/s. The quality of the DPSK signals is assessed using both noncoherent detection for a bit rate of 10.709-Gb/s and coherent detection with DSP involving a LUT pattern-dependent distortion compensator. For very short reach optical links, a 16-ary quadrature amplitude modulation signal is generated using subcarrier modulation with a subcarrier frequency of half the symbol rate, Nyquist pulse shaping, and a directly modulated PFL at bit rates up to 56-Gb/s. Using polarization multiplexing emulation, a pre-amplified direct detection receiver and DSP, loss margins of 12.6 dB and 8 dB are achieved for a 112-Gb/s dual polarization signal within a 33 GHz optical bandwidth at back-to-back and after 4 km transmission, respectively. / Thesis (Ph.D, Electrical & Computer Engineering) -- Queen's University, 2013-01-31 13:58:56.327

The terahertz frequency range is a relatively unstudied region of the electromagnetic spectrum. However with the emergence of numerous applications for terahertz light in diverse areas such as security scanning, biological imaging, gas spectroscopy and astrophysics there has been considerable recent growth in the volume of research activity in this area. The studies presented in this thesis aim to introduce the physics of surface plasmons to the terahertz frequency range, and on the way to use some of the unique capabilities of terahertz spectroscopy to try and find new information about fundamental surface-plasmon based electromagnetic structures. Four distinct experiments are described in this work, all of them underpinned by the technique of terahertz time-domain spectroscopy (Chapter 2). This is a very powerful and adaptable spectroscopic method which allows us to measure the electric field of pulsed terahertz radiation as a function of time. This in turn allows us to directly extract both phase and amplitude of the terahertz light as a function of frequency, over a broad frequency range. Furthermore, this method of terahertz spectroscopy can be combined with photoexcitation pulses of visible/NIR light which can be used to make dynamic changes to the properties of materials in the terahertz beam. The first experiment reported (Chapter 3) measures the propagation of coupled surface plasmons in a resonant slit cavity. We use terahertz time-domain spec- troscopy to determine the characteristics of the cavity resonances in a semiconductor slit near the surface plasma frequency of the material, where we are able to mea- sure very large red-shifts in the frequency of the cavity resonance. By considering the phase information which can be extracted directly from time-resolved terahertz measurements we are able to link the behaviour of the resonances to the propagation characteristics of the surface plasmon modes inside the slits. The second experiment (Chapter 4) is a more direct measurement of surface plasmons, propagated over the surface of a semiconductor wafer. We show that the electric field of the surface plasmon is confined to a subwavelength region around the surface, and that the confined field is useful for spectroscopy of very thin layers above the surface. We are able to measure films with thickness less than 1/600th of the wavelength of the terahertz light. After these two experiments with confined semiconductor surface plasmons we move on to a pair of experiments looking at terahertz surface modes mediating the transmission of light through holes in metal films. In the initial experiment (Chapter 5) we use the time-domain data from terahertz spectroscopy to determine the role that surface mode lifetime plays in modifying the amplitude and width of Extraor- dinary Optical Transmission (EOT) resonances, which arise from the periodicity of a hole-array lattice. By changing the temperature of the lossy dielectric semicon- ductor substrate we are able to modify the surface mode lifetime, and link this to the resonant transmission characteristics. In Chapter 6 we extend the hole array EOT experiment by making dynamic changes to the propagation of the surface mode which mediates the transmission. This is achieved by photo-exciting the semiconductor substrate inside the holes and forming a thin layer of material with high charge carrier density on the surface. Interaction of the surface mode with the photoexcited region quenches the resonant transmission. We show that by changing the hole size so that the surface-mode mediated transmission pathway predominates in the spectrum it is possible to use optical pulses to modulate the transmission of terahertz radiation with very high efficiency. In the conclusions (Chapter 7) we link together some of the insights and infer- ences which can be drawn from the above results, as well as evaluating the efficacy of the experimental and simulation methodology.

A photonic analog-to-digital converter (ADC) based on a robust symmetrical number system (RSNS) was constructed and tested. The analog signal to be converted is used to amplitude modulate an optical pulse from a laser using three Mach-Zehnder interferometers (MZI). The Mach-Zehnder interferometers fold the input analog signal for a three-channel RSNS encoding. The folding waveforms are then detected and amplitude-analyzed by three separate comparator banks, the outputs of which are used to determine a digital representation of the analog signal. This design uses the RSNS preprocessing to encode the signal with the fewest number of comparators for any selected bit resolution. In addition to the efficiency of its use of comparators, the RSNS encoding has inherent Gray-code properties making it particularly attractive for eliminating any possible encoding errors. The RSNS encoding is combined with an optical infrastructure that offers high bandwidth and low insertion loss characteristics. A full implementation was constructed and tested. The lack of a high-speed data acquisition device limited the results to examining the preprocessing and digital processing separately. With the system integration of a data acquisition device, a wideband direct digital antenna architecture can be demonstrated.

Datacenter networks are not only larger but with new applications increasing the east-west traffic and the introduction of the spine leaf architecture there is an urgent need for high bandwidth, low cost, energy efficient interconnects. This paper will discuss the role integrated photonics can have in achieving datacenter requirements. We will review the state of the art and then focus on advances in optical switch fabrics and systems. The optical switch is of particular interest from the integration point of view. Current MEMS and LCOS commercial solutions are relatively large with relatively slow reconfiguration times limiting their use in packet based datacenter networks. This has driven the research and development of more highly integrated silicon photonic switch fabrics, including micro ring, Mach-Zehnder and MEMS device designs each with its own energy, bandwidth and scalability, challenges and trade-offs. Micro rings show promise for their small footprint, however they require an energy efficient means to maintain wavelength and thermal control. Latency requirements have been traditionally less stringent in datacenter networks compared to high performance computing applications, however with the increasing numbers of servers communicating within applications and the growing size of the warehouse datacenter, latency is becoming more critical. Although the transparent optical switch fabric itself has a minimal additional latency, we must also take account of any additional latency of the optically switched architecture. Proposed optically switched architectures will be reviewed.

As a type of nonlinear system with complexity, mode-locked fiber lasers are known for their complex behaviour. It is a challenging task to understand the fundamental physics behind such complex behaviour, and a unified description for the nonlinear behaviour and the systematic and quantitative analysis of the underlying mechanisms of these lasers have not been developed. Here, we present a complexity science-based theoretical framework for understanding the behaviour of mode-locked fiber lasers by going beyond reductionism. This hierarchically structured framework provides a model with variable dimensionality, resulting in a simple view that can be used to systematically describe complex states. Moreover, research into the attractors' basins reveals the origin of stochasticity, hysteresis and multistability in these systems and presents a new method for quantitative analysis of these nonlinear phenomena. These findings pave the way for dynamics analysis and system designs of mode-locked fiber lasers. We expect that this paradigm will also enable potential applications in diverse research fields related to complex nonlinear phenomena.

Mechanical stresses placed on ferromagnetic materials while under the influence of a magnetic field are known to cause changes to the permanent magnetization of the material. Modeling this phenomenon is vital to the safety of ocean faring ships. In this thesis, a quasi-strip theory method of computing the nonlinear wave induced motion of a ship is developed, and the fluid pressure on the surface of the hull is used to determine the mechanical stresses. An existing magnetostatic volume integral equation code is used to evaluate the effects of the ship motion and hull stresses. The resulting changes in the magnetic field for various ship forms are presented to demonstrate the effects of given sea states.

Entangled photons are ideally suited to the transmission of photonic quantum information. Mitigating the effects of decoherence is fundamental to distributing photonic entanglement across large distances. One such proposal is entanglement distillation, in which operations on a large ensemble of weakly entangled states generate a smaller ensemble of more strongly entangled states. In this thesis, we experimentally and theoretically analyse various tools required for demonstrating continuous-variable (CV) entanglement distillation, following the proposal by Browne et al., [Phys. Rev. A <b>67</b>, 062320 (2003)]. Specifically, we propose figures of merit to account for the practical limitations of non-deterministic non-Gaussian operations, and analyse the experimental parameters necessary to optimise them. We develop a source of pulsed two-mode squeezed states, which are the initial states of our entanglement distillation protocol. We use weak-field homodyne detection as a phase-dependent photon counting detector, and demonstrate its utility in conditional state generation. Using these states, we demonstrate sub-binomial light as a tool for benchmarking quantum states. Finally, we applied two-mode weak-field homodyne detection to two entangled states and demonstrate correlations in the photon counting statistics which depend on a joint phase from two independent local oscillators. This setup is sufficient to apply an entanglement witness developed by Puentes et al. [New J. Phys. <b>12</b>, 033042 (2010)]. Despite encouraging simulations, we do not witness entanglement with this scheme, which we attribute to a noise source unaccounted for in the simulations. Although we do not demonstrate entanglement distillation outright, the tools we develop to do so represent a general, hybrid approach to CV quantum optics. Developing tools such as phase-resolved projective measurement on two-mode states allows us to probe both the wave and particle nature of entangled light at the single-photon level. Using and expanding these techniques to probe larger quantum systems may prove useful in studies of fundamental physics and quantum enhanced technologies.

Les antennes optiques sont des structures qui permettent de convertir, dans les deux sens, l'énergie électromagnétique entre un faisceau lumineux et une source (ou un absorbeur) localisée en son sein. L'utilisation de résonateurs de taille inférieure à la longueur d'onde permet de réaliser cette fonction de manière efficace, sur une bande spectrale relativement étendue, et d'avoir une antenne compacte.La bonne connaissance des propriétés optiques de ces résonateurs, pris séparément, et de leurs couplages entre eux, est nécessaire pour pouvoir proposer des designs d'antenne efficaces.Dans cette thèse, en se basant sur la décomposition multipolaire des champs et sur la méthode de la matrice-T, on obtient des solutions analytiques rigoureuses pour des résonateurs sphériques et homogènes, dont on tire des modèles simplifiés, intuitifs, et proches de la solution exacte des équations de Maxwell.Entre autre résultats, ces modèles nous ont permis de proposer un design d'antenne optique compacte, directive, à taux de désexcitation et rendement quantique élevés en utilisant une structure hybride métal-diélectrique. Des collaborations avec des expérimentateurs ont permis de valider, d'une part les caractéristiques de chromophores auto-assemblés par ADN (S. Bidault à Paris), et d'autre part, la possibilité d'utiliser plusieurs résonances électriques et magnétiques combinées (supportées par des sphères diélectriques d'indice modéré, n=2,45) pour réfléchir ou bien collecter le rayonnement d'un émetteur dipôle électrique placé à proximité (expérience menée dans le régime micro-ondes par R. Abdeddaim et J-M. Geffrin). / Optical antennas are structures able to convert, in both ways, electromagnetic energy between a light beam and a source (or absorber) placed in the structure. The use of sub-wavelength resonators enables one to realize this function in an efficient way, on relatively broad bandwidths, and to have a compact design. A good understanding of the optical properties of such resonators, taken individually, and of their couplings, is thus necessary in order to propose efficient optical antenna designs. In this manuscript, using a multipole decomposition of the fields and a T-matrix method, we obtain rigorous analytical solutions for spherical, homogeneous resonators, from which we deduce simplified, intuitive models that are still very close to the exact resolution of the Maxwell equations.Among other results, those models enabled us to propose a nanoantenna design that is at once compact, radiative and efficient, by using a hybrid metallo-dielectric structure. Some collaborations with experimental groups enabled us to validate, on the one hand, the optical characteristics of hybrid chromophores that are self-assembled using a DNA template (S. Bidault, Paris), and on the other hand, the possibility of using multiple combined electric and magnetic resonances (supported by dielectric spheres of moderate refractive index, n=2.45) in order to reflect, or more importantly collect, radiation coming from an electric dipole emitter placed nearby (the experiment was realized in the microwave regime by R. Abdeddaim and J-M. Geffrin).