Cette thèse traite des équations de Maxwell en domaine temporel. Le principal objectif est de proposer des méthodes de type éléments finis d'ordre élevé pour les équations de Maxwell et des schémas d'intégration en temps efficaces sur des maillages localement raffinés. Nous considérons des méthodes GDDT (Galerkine Discontinues en Domaine Temporel) s'appuyant sur une interpolation polynomiale d'ordre arbitrairement élevé des composantes du champ électromagnétique. Les méthodes GDDT pour les équations de Maxwell s'appuient le plus souvent sur des schémas d'intégration en temps explicites dont la condition de stabilité peut être très restrictive pour des maillages raffinés. Pour surmonter cette limitation, nous considérons des schémas en temps qui consistent à appliquer un schéma implicite localement, dans les régions raffinées, tout en préservant un schéma explicite sur le reste du maillage. Nous présentons une étude théorique complète et une comparaison de deux méthodes GDDT localement implicites. Des expériences numériques en 2D et 3D illustrent l'utilité des schémas proposés. Le traitement numérique de milieux de propagation complexes est également l'un des objectifs. Nous considérons l'interaction des ondes électromagnétiques avec les tissus biologiques qui est au cœur de nombreuses applications dans le domaine biomédical. La modélisation numérique nécessite alors de résoudre le système de Maxwell avec des modèles appropriés de dispersion. Nous formulons une méthode GDDT localement implicite pour le modèle de Debye et proposons une analyse théorique et numérique complète du schéma.

Modulation instability (MI) is a general characteristic of wave propagation in nonlinear dispersive media and it has been intensively investigated in several branches of physics due to its fundamental nature as well as technological applications. This phenomenon corresponds to the exponential growth of weak harmonic perturbations in virtue of the interplay between dispersive and nonlinear effects. Hence, despite its important features, MI is also a main source of channel depletion and degradation in optical fiber communications. In this thesis, we investigate the modulational instability (MI) induced by cross-phase-modulation (XPM) of two incoherently coupled optical pulses co-propagating in a lossless fiber with a finite nonlinear response time. The non-instantaneous character of the nonlinear response is introduced through a Debye relaxation process. We analytically obtain the exact dispersion relation for weak harmonic perturbations over the stationary solution. We show that the instability spectrum, present in both normal and anomalous dispersive regimes in instantaneously responding Kerr media, develops a double peak structure whose relative strength and typical frequency range depend on the response time. Further, we reveal that there are two unstable modes in the entire frequency spectrum. We report the dependence of the maximum gain and central frequency within each unstable mode as a function of the group velocity mismatch and response time, showing the crossover between the regimes of fast and slow non-linear responses. / Fundação de Amparo a Pesquisa do Estado de Alagoas / Conselho Nacional de Desenvolvimento Científico e Tecnológico / Instabilidade Modulacional (IM) é um fenômeno característico da propagação de ondas em meios dispersivos não-lineares e tem sido estudado em diversas áreas da Física devido a sua natureza fundamental bem como suas importantes plicações tecnológicas. Esse fenômeno corresponde ao enriquecimento exponencial de pequenas perturbações harmônicas devido a cooperação dos efeitos não-lineares e dispersivos. Portanto, não obstante sua aplicabilidade, IM é, de igual modo, uma fonte importante de degradação em sistemas de comunicação por fibras ópticas. Nesta tese investigamos a instabilidade modulacional (IM) induzida por Modulação de Fase Cruzada (MFC) de dois pulsos ópticos acoplados incoerentemente que se propagam em uma fibra sem perda com tempo de resposta não-linear finito. O caráter não-instantâneo da resposta não-linear é introduzido através de um processo de relaxação de Debye. Obtemos analiticamente, de modo exato, a relação de dispersão para fracas perturbações harmônicas da solução estacionária. Mostramos que o espectro de instabilidade, presente tanto no regime de dispersão normal quanto no anômalo em meios Kerr com resposta instantânea, desenvolve uma estrutura de pico duplo cuja a intensidade relativa e a frequência típica dependem do tempo de resposta considerado. Além do mais, revelamos que existem dois modos instáveis ao longo de todo o espectro de frequência. Apresentamos a dependência do ganho máximo e da frequência correspondente dentro de cada modo instável como função da diferença da velocidade de grupo e do tempo de resposta, mostrando o cruzamento entre os regimes de resposta não-linear rápida e lenta.

Protein (mis)folding into highly ordered, fibrillar structures, amyloid fibrils, is a hallmark of several, mainly neurodegenerative, disorders. The mechanism of this supra-molecular self-assembly reaction, as well as its relationship to protein folding are not well understood. In particular, the molecular origin of the metastability of the soluble state of proteins with respect to the aggregated states has not been clearly established. In this dissertation, it is demonstrated, that highly accurate kinetic experiments, using a novel biosensing method, can yield fundamental insight into the dynamics of proteins in the region of the free energy landscape corresponding to protein aggregation. First, a section on Method development describes the extension and elaboration of the previously established kinetic assay relying on quartz crystal microbalance measurements for the study of amyloid fibril elongation (Chapter 3). This methodology is then applied in order to study in great detail the origin of the various contributions to the free energy barriers separating the soluble state of a protein from its aggregated state. In particular, the relative importance of residual structure, hydrophobicity (Chapter 4) and electrostatic interactions (Chapter 5) for the total free energy of activation are discussed. In the last part of this thesis (Chapter 6), it is demonstrated that this biosensing method can also be used to study the binding of small molecules to amyloid fibrils, a very useful feature in the framework of the quest for potential inhibitors of amyloid formation. In addition, it is shown that Thioflavin T, to-date the most frequently employed fluorescent label molecule for bulk solution kinetic studies, can in the presence of potential amyloid inhibitor candidates be highly unreliable as a means to quantify the effect of the inhibitor on amyloid formation kinetics. In summary, the work in this thesis contributes to both the fundamental and the applied aspects of the field of protein aggregation.

We study the expansion velocity and ion temperature evolution of ultracold neutral plasmas (UNPs) of calcium atoms under the influence of a uniform magnetic field that ranges up to 200 G. In the experiments, we use a magneto-optical trap (MOT) to capture the neutral atoms and laser-induced fluorescence (LIF) to take images of the plasma. We vary the magnetic field strengths and the initial electron temperatures and observe the plasma evolution in time. We compare the ion temperature evolution to the theory introduced in the paper by Pohl et. al. [Phys. Rev. A 70, 033416 (2004)]. The evolution of the gradient of expansion velocity suggests the presence of ion acoustic waves (IAWs). We speculate that our measurements showing that the ion temperature remains relatively high throughout the evolution is a biproduct of the IAW.

В работе реализован численный алгоритм для решения уравнения Фоккера-Планка, позволяющий получать значения первой и третьей гармоники ансамбля взаимодействующих частиц для различных амплитуд переменного поля. В формулы первой и третьей гармоники вводились функции, зависящие от параметра и восприимчивости Ланжевена, с неопределенными коэффициентами. Неопределенные коэффициенты находились методом наименьших квадратов. Выражения для функций приближались данными из численного решения уравнения Фоккера-Планка и затем минимизировались относительно неопределённых коэффициентов. Получившиеся формулы сравнивались с численным решением и с известными теориями. / In this work, a numerical algorithm for solving the Fokker-Planck equation, which allows to obtain the values of the first and third harmonics of the ensemble of interacting particles for different amplitudes of the alternating field, was implemented. The functions depending on the parameter and Langevin susceptibility with uncertain coefficients were introduced into the formulas for the first and third harmonics. The uncertain coefficients were found by the least-squares method. Expressions for the functions were approximated with data from the numerical solution of the Fokker-Planck equation and then minimized with respect to the uncertain coefficients. The resulting formulas were compared with the numerical solution and with known theories.

This thesis, which contains fourteen chapters in two parts, presents theoretical and computer simulation studies of dynamics in supercooled liquids and thermotropic liquid crystals. These two apparently diverse physical systems are unified by a startling similarity in their complex slow dynamics. Part I consists of six chapters on supercooled liquids while Part II comprises seven chapters on thermotropic liquid crystals. The fourteenth chapter provides a concluding note. Part I starts with an introduction to supercooled liquids given in chapter 1. This chapter discusses basic features of supercooled liquids and the glass transition and portrays some of the theoretical frameworks and formalisms that are widely recognized to have contributed to our present understanding. Chapter 2 introduces a new model of binary mixture in order to study dynamics across the supercooled regime. The system consists of an equimolar mixture of the Lennard-Jones spheres and the Gay-Berne ellipsoids of revolution, and thus one of its components has orientational degrees of freedom (ODOF). A decoupling between trans-lational diffusion and rotational diffusion is found to occur below a temperature where the second rank orientational correlation time starts showing a steady deviation from the Arrhenius temperature behavior. At low temperatures, the optical Kerr effect (OKE) signal derived from the system shows a short-to-intermediate time power law decay with a very weak dependence on temperature, if at all, of the power law exponent as has been observed experimentally. At the lowest temperature investigated, jump motion is found to occur in both the translational and orientational degrees of freedom. Chapter 3 studies how the binary mixture, introduced in the previous chapter, explores its underlying potential energy landscape. The study reveals correlations between the decoupling phenomena, observed almost universally in supercooled molecular liquids, and the manner of exploration of the energy landscape of the system. A significant deviation from the Debye model of rotational diffusion in the dynamics of ODOF is found to begin at a temperature at which the average inherent structure energy of the system starts falling as the temperature decreases. Further, the coupling between rotational diffusion and translational diffusion breaks down at a still lower temperature, where a change occurs in the temperature dependence of the average inherent structure energy. Chapters 4-6 describe analytical and numerical approaches to solve kinetic models of glassy dynamics for various observables. The β process is modeled as a thermally activated event in a two-level system and the a process is described as a β relaxation mediated cooperative transition in a double-well. The model resembles a landscape picture, conceived by Stillinger [Science 267, 1935 (1995)], where the a process is assumed to involve a concerted series of the β processes, the latter being identified as elementary relaxations involving transitions between contiguous basins. For suitable choice of parameter values, the model could reproduce many of the experimentally observed features of anomalous heat capacity behavior during a temperature cycle through the glass transition as described in chapter 4. The overshoot of the heat capacity during the heating scan that marks the glass transition is found to be caused by a delayed energy relaxation. Chapter 5 shows that the model can also predict a frequency dependent heat capacity that reflects the two-step relaxation behavior. The high-frequency peak in the heat capacity spectra appears with considerably larger amplitude than the low-frequency peak, the latter being due to the a relaxation. The model, when simplified with a modified description of the a process that involves an irreversible escape from a metabasin, can be solved analytically for the relaxation time. This version of the model captures salient features of the structural relaxation in glassy systems as described in chapter 6. In Part II, thermotropic liquid crystals are studied in molecular dynamics simulations using primarily the family of the Gay-Berne model systems. To start with, chapter 7 provides a brief introduction to thermotropic liquid crystals, especially from the perspective of the issues discussed in the following chapters. This chapter ends up with a detail description of the family of the Gay-Berne models. Chapter 8 demonstrates that a model system for calamitic liquid crystal (comprising rod-like molecules) could capture the short-to-intermediate time power law decay in the OKE signal near the isotropic-nematic (I-N) phase transition as observed experimentally. The single-particle second rank orientational time correlation function (OTCF) for the model liquid crystalline system is also found to sustain a power law decay regime in the isotropic phase near the I-N transition. On transit across the I-N phase boundary, two power law decay regimes, separated by a plateau, emerge giving rise to a step-like feature in the single-particle second rank OTCF. When the time evolution of the rotational non-Gaussian parameter is monitored as a diagnostic of spatially heterogeneous dynamics, a dominant peak is found to appear following a shoulder at short times, signaling the growth of pseudonematic domains. These observations are compared with those relevant ones obtained for the supercooled binary mixture, as discussed in chapter 2, in the spirit of the analogy suggested recently by Fayer and coworkers [J. Chem. Phys. 118, 9303 (2003)]. In chapter 9, orientational dynamics across the I-N transition are investigated in a variety of model systems of thermotropic liquid crystals. A model discotic system that consists of disc-like molecules as well as a lattice system have been considered in the quest of a universal short-to-intermediate time power law decay in orientational relaxation, if any. A surprisingly general power law decay at short to intermediate times in orientational relaxation is observed in all these systems. While the power law decay of the OKE signal has been recently observed experimentally in calamitic systems near the I-N phase boundary and in the nematic phase by Fayer and coworkers [J. Chem. Phys. 116, 6339 (2002), J. Phys. Chem. B 109, 6514 (2005)], the prediction for the discotic system can be tested in experiments. Chapter 10 presents the energy landscape view of phase transitions and slow dynamics in thermotropic liquid crystals by determining the inherent structures of a family of one-component Gay-Berne model systems. This study throws light on the interplay between the orientational order and the translational order in the mesophases the systems exhibit. The onset of the growth of the orientational order in the parent phase is found to induce a translational order, resulting in a smectic-like layer in the underlying inherent structures. The inherent structures, surprisingly, never seem to sustain orientational order alone if the parent nematic phase is sandwiched between the high-temperature isotropic phase and the low-temperature smectic phase. The Arrhenius temperature dependence of the orientational relaxation time breaks down near the I-N transition and this breakdown is found to occur at a temperature below which the system explores increasingly deeper potential energy minima. There exists a remarkable similarity in the manner of exploration of the potential energy landscape between the Gay-Berne systems studied here and the well known Kob-Andersen binary mixture reported previously [Nature, 393, 554 (1998)]. In search of a dynamical signature of the coupling between orientational order and translational order, anisotropic translational diffusion in the nematic phase has been investigated in the Gay-Berne model systems as described in chapter 11. The translational diffusion coefficient parallel to the director D// is found to first increase and then decrease as the temperature drops through the nematic phase. This reversal occurs where the smectic order parameter of the underlying inherent structures becomes significant for the first time. The non-monotonic temperature behavior of D// can thus be viewed from an energy landscape analysis as a dynamical signature of the coupling between orientational and translational order at the microscopic level. Such a view is likely to form the foundation of a theoretical framework to explain the anisotropic translation diffusion. Chapter 12 investigates the validity of the Debye model of rotational diffusion near the I-N phase boundary with a molecular dynamics simulation study of a Gay-Berne model system for calamitic liquid crystals. The Debye model is found to break down near the I-N phase transition. The breakdown, unlike the one observed in supercooled molecular liquids where a jump diffusion model is often invoked, is attributed to the growth of orientational pair correlation. A mode-coupling theory analysis is provided in support of the explanation. Chapter 13 presents a molecular dynamics study of a binary mixture of prolate ellipsoids of revolution with different aspect ratios interacting with each other through a generalized Gay-Berne potential. Such a study allows to investigate directly the aspect ratio dependence of the dynamical behavior. In the concluding note, chapter 14 starts with a brief summary of the outcome of the thesis and ends up with suggestion of a few relevant problems that may prove worthwhile to be addressed in future.

No description available.