Spin waves and their quanta, magnons, are the wave-like excitations of a magnetically ordered medium. The technological prospect of utilizing them as low-loss information carriers has driven various research efforts in the field of magnonics. Spin waves arise further interest due to their inherently strong nonlinear behavior which results from their interaction with the surrounding magnetic texture. Hence, magnons are subject to a variety of nonlinear effects and allow for extensive studies of such phenomena. In this work, the propagating spin-waves in micro structured Co25Fe75 conduits have been investigated by means of micro focused Brillouin light scattering spectroscopy. Due to the low intrinsic damping of this metallic compound, spin-wave decay lengths in the order of 20 μm can be observed which have not been reported elsewhere for other ferromagnet thin film materials. Furthermore, nonlinear four-magnon scattering processes can be observed when increasing the spin-wave amplitudes applying a sufficiently strong microwave excitation. This phenomenon introduces additional losses for propagating waves as it diverts energy into the parametric generation of secondary states. It is shown that the reduction of the spin-wave decay lengths reaches up to 50 %. In the second part, a novel approach for the utilization of four-magnon scattering is presented. It is shown that an additional driving signal at a secondary driving frequency can steer the nonlinear process in such a way, that a set of secondary parametric states with a well-defined frequency spacing is populated. This process is referred to as stimulated four-magnon scattering, as it enhances specific nonlinear scattering events. As a result, frequency combs with multiple equidistant modes are observed, which exhibit frequency spacings of 400 MHz up to 2 GHz. These complex spin-wave spectra can actively be tuned in various ways using external parameters such as the driving signals. These results advance the understanding of nonlinear spin waves in general and expands the range of possible technological applications of magnons.:List of Figures List of Tables List of Abbreviations and Acronyms List of Symbols 1 Introduction 2 Theoretical background 2.1 Interactions in microstructured thin film ferromagnets 2.1.1 Exchange interaction 2.1.2 Dipolar interaction and demagnetizing fields 2.2 Magnetization dynamics in ferromagnetic thin films 2.2.1 The Landau-Lifshitz and Gilbert equation 2.2.2 Spin waves 2.3 Nonlinear phenomena 2.3.1 Four-magnon scattering 3 Materials and Methods 3.1 Materials and sample fabrication 3.1.1 The low damping alloy Co25Fe75 3.1.2 Patterning: electron beam and optical lithography 3.1.3 Microwave antenna structures 3.2 Brillouin light scattering 3.2.1 Magnon-photon interaction 3.2.2 The Tandem Fabry Pérot interferometer 3.2.3 BLS microscopy (μBLS) 3.2.4 Phase-resolved BLS (PR-μBLS) 3.2.5 Temporal resolution (TR-μBLS) 3.3 Micromagnetic simulations in MuMax3 3.3.1 Mesh and material parameters 3.3.2 Simulation of magnetization dynamics 4 Results 4.1 Magnon transport in Co25Fe75 micro-conduits 4.1.1 Low external fields and magnetic groundstate 4.1.2 Magnon transport at low driving powers 4.1.3 Impact of nonlinear four-magnon scattering on magnon transport 4.2 Magnon frequency combs 4.2.1 Introduction on stimulated four-magnon scattering 4.2.2 Experimental realization 4.2.3 Amplitude-dependent observations 4.2.4 Tunability of spin-wave frequency combs 4.2.5 Variations of the excitation geometry 5 Summary and outlook Own publications Bibliography 109 Acknowledgement A Appendix A.1 Fabrication of Co25Fe75 microstructures A.2 Atomic Force Microscopy measurement on a 5 μm wide conduit A.3 BLS measurement of spin-wave decay lengths in a 5 μm wide conduit A.4 Calculations: Temporal profile of stimulated four-magnon scattering A.5 Power dependent frequency comb formation measured at positions II & III A.6 Averaged frequency comb mode numbers at reversed magnetic field polarity

[pt] O design otimizado de estruturas considerando não-linearidades tem sido amplamente pesquisado nas décadas recentes. A análise de elementos finitos aplicada à otimização topológica é prejudicada pela deformação excessiva de elementos de baixa densidade sob alta compressão, o que impede o processo de encontrar uma solução ótima. Dois métodos, o esquema Interpolação de Energia e a técnica de Hiperelasticidade Aditiva, são implementados para superar essa dificuldade no problema de minimização da flexibilidade, e modelos de materiais hiperelásticos são usados para investigar suas influências na topologia otimizada. O Método das Assíntotas Móveis é usado para atualizar as variáves de projeto cujas sensibilidades foram calculadas pelo método adjunto. A equação de estado é resolvida através do método de Newton-Raphson com um incremento de carga ajustável para reduzir o custo computacional. Resultados de dois problemas de referência são comparado com aqueles já estabelecidos na literatura. O uso de diferentes modelos hiperelásticos apresentou pouca influência no design final da estrutura. O método de Interpolação de Energia foi capaz de convergir para cargas muito maiores que o método padrão, enquanto a Hiperelasticidade Aditiva apresentou dificuldades de convergência em estado plano de deformação. / [en] The optimized design of structures considering nonlinearities has been widely researched in the recent decades. The finite element analysis applied to topology optimization is jeopardized by the excessive deformation of low-density elements under high compression, which hinders the process of finding an optimal solution. Two methods, the Energy Interpolation scheme and the Additive Hyperelasticity technique, are implemented to overcome this difficulty in the minimum compliance problem, and hyperelastic material models are used to investigate their influence on the optimized topology. The Method of Moving Asymptotes is used to update the design variables whose sensitivities were calculated from the adjoint method. The state equation is solved through the Newton-Raphson method with an adjusting load step to reduce computational cost. Results for two benchmark problems are compared with those already established in the literature. The use of different hyperelastic models presented little influence on the final design of the structure. The Energy Interpolation method was able to converge for much higher loads than the default method, while the Additive Hyperelasticity presented convergence difficulties in plane strain.

[ES] Dentro del campo del Análisis Numérico, la resolución de ecuaciones y sistemas de ecuaciones no lineales es uno de los aspectos más relevantes y estudiados. Esto se debe a que gran cantidad de problemas de Matemática Aplicada, como la resolución de ecuaciones diferenciales, ecuaciones en derivadas parciales o ecuaciones integrales entre muchos otros, pueden reducirse a buscar la solución de un sistema no lineal. Generalmente, es muy difícil obtener la solución analítica de este tipo de problemas y, en muchos casos, aunque es posible llegar a encontrar la solución exacta, es muy complicado trabajar con dicha expresión por su complejidad. Además, con el desarrollo de las nuevas tecnologías, se han hecho grandes avances en el uso de herramientas computacionales, por lo que las dimensiones de algunos de los problemas que se plantean en campos como la Economía, la Ingeniería, la Ciencia de datos, etc. han crecido considerablemente, dando lugar a problemas de grandes dimensiones. Por estos motivos, es de gran utilidad y, en muchos casos, resulta necesario resolver estos problemas no lineales de forma aproximada, por supuesto, con técnicas matemáticamente rigurosas dentro del campo del Análisis Numérico. Por las razones expuestas, los métodos iterativos para aproximar la solución de ecuaciones y sistemas de ecuaciones no lineales han constituido a lo largo de los últimos años un importante campo de investigación. La implementación computacional de estos métodos es una importante herramienta dentro de las Ciencias Aplicadas ya que dan solución a problemas que antiguamente eran difíciles de resolver. La investigación que se lleva a cabo en esta Tesis Doctoral se centra en estudiar, diseñar y aplicar métodos iterativos que mejoren en ciertos aspectos a los esquemas clásicos, como por ejemplo: la velocidad de convergencia, la aplicabilidad a problemas no diferenciales, la accesibilidad o la eficiencia. Buena parte del trabajo desarrollado en esta memoria se centra en el estudio de métodos iterativos para problemas multidimensionales, en especial, nos hemos centrado en el estudio de esquemas libres de derivadas. Además, uno de los ejes centrales de la presente Tesis Doctoral se enfoca en el estudio de la convergencia local y semilocal de métodos ya desarrollados en la literatura reciente o de nuevos métodos iterativos diseñados en este mismo trabajo. Este estudio garantiza para los métodos analizados la existencia de solución dado un punto de partida, el dominio de convergencia de las soluciones del problema y la unicidad de éstas bajo ciertas condiciones. Para complementar el estudio de convergencia de los métodos, en algunos capítulos también se realiza un estudio dinámico de los métodos aplicados a ecuaciones no lineales para, posteriormente, extrapolar los resultados al caso multidimensional. Además, como parte de algunos experimentos numéricos, se ha comparado la accesibilidad de distintos métodos numéricos a través de las cuencas de atracción representadas en diferentes planos dinámicos, tanto para el caso unidimensional como el multidimensional. Finalmente, en la mayor parte de los Capítulos de esta tesis se aplican los métodos iterativos estudiados a la resolución de problemas no lineales de Matemática Aplicada. Estos problemas pueden estar preparados para poner a prueba los algoritmos diseñados o ser problemas reales presentes en algunas Ciencias Aplicadas como la Ingeniería, la Física, la Química, etc. Los resultados anteriormente descritos forman parte de la presente Tesis Doctoral para la obtención del título de Doctora en Matemáticas. / [CA] Dins del camp de l'Anàlisi Numèrica, la resolució d'equacions i sistemes d'equacions no lineals és un dels aspectes més rellevants i estudiats. Això és pel fet de que gran quantitat de problemes de Matemàtica Aplicada, com la resolució d'equacions diferencials, equacions en derivades parcials o equacions integrals entre molts altres, poden reduir-se a buscar la solució d'un sistema no lineal. Generalment, és molt difícil obtindre la solució analítica d'estos problemes i, en molts casos, encara que és possible arribar a trobar la solució exacta, és molt complicat treballar amb aquesta expressió per la seua complexitat. A més, amb el desenvolupament de les tecnologies, s'han fet grans avanços en l'ús d'eines computacionals, per la qual cosa les dimensions d'alguns dels problemes que es plantegen en camps com l'Economia, l'Enginyeria, la Ciència de dades, etc. han crescut considerablement, donant lloc a problemes de grans dimensions. Per aquestos motius, és de gran utilitat i, en molts casos, resulta necessari resoldre estos problemes no lineals de manera aproximada, per descomptat, amb tècniques matemàticament riguroses dins del camp de l'Anàlisi Numèrica. Per les raons exposades, els mètodes iteratius per a aproximar la solució d'equacions i sistemes d'equacions no lineals han constituït al llarg dels últims anys un important camp d'investigació. La implementació computacional d'estos mètodes és una eina important dins de les Ciències Aplicades ja que donen solució a problemes que antigament eren difícils de resoldre. La investigació que es porta a terme en esta Tesi Doctoral es centra en estudiar, dissenyar i aplicar mètodes iteratius que milloren en certs aspectes als esquemes clàssics com són: la velocitat de convergència, l'aplicabilitat a problemes no diferencials, l'accessibilitat o l'eficiència. Bona part del treball desenvolupat en esta memòria es centra en l'estudi de mètodes iteratius per a problemes multidimensionals, especialment, ens hem centrar en l'estudi d'esquemes lliures de derivades. A més, part de la present Tesi Doctoral està centrada en l'estudi de la convergència local i semilocal de mètodes ja desenvolupats en la literatura recent o de nous mètodes iteratius dissenyats en aquest mateix text. Este estudi garanteix per als mètodes l'existència de solució donat un punt de partida, el domini de convergència de les solucions del problema i la unicitat d'estes sota unes certes condicions. Per a complementar l'estudi de convergència dels mètodes, en alguns capítols també es realitza un estudi dinàmic dels mètodes aplicats a equacions no lineals per a, posteriorment, extrapolar els resultats al cas multidimensional. A més, com a part d'alguns experiments numèrics, s'ha comparat l'accessibilitat de diferents mètodes numèrics a través de les conques d'atracció representades en diferents plans dinàmics, tant per al cas unidimensional com el multidimensional. Finalment, en la major part dels Capítols d'esta tesi s'apliquen els mètodes iteratius estudiats a la resolució de problemes no lineals de Matemàtica Aplicada. Estos problemes poden estar preparats per a probar la funcionalitat dels algorismes dissenyats o ser problemes reals presents en algunes Ciències Aplicades com l'Enginyeria, la Física, la Química, etc. Els resultats anteriorment descrits formen part de la present Tesi Doctoral per a l'obtenció del títol de Doctora en Matemàtiques. / [EN] Within the field of Numerical Analysis, the resolution of equations and systems of nonlinear equations is one of the most relevant and studied aspects. This is due to the fact that a large number of problems in Applied Mathematics, such as the solution of differential equations, partial differential equations or integral equations among many others, can be reduced to the solution of a non-linear system. Generally, it is very difficult to obtain the analytical solution of this type of problems and, in many cases, although it is possible to find the exact solution, it is very complicated to work with this expression due to its complexity. Moreover, with the development of technologies, great advances have been made in the use of computational tools, so that the dimensions of some of the problems that arise in fields such as Economics, Engineering, Data Science, etc. have grown considerably, giving rise to problems of large dimensions. For these reasons, it is very useful and, in many cases, necessary to solve these non linear problems in an approximate way, of course, with mathematically rigorous techniques within the field of Numerical Analysis. For these reasons, iterative methods for approximating the solution of nonlinear equations and systems of equations have been an important field of research in recent years. The computational implementation of these methods is an important tool in the Applied Sciences as they provide solutions to problems that were difficult to solve in the past. The research carried out in this Doctoral Thesis focuses on the study, design and application of iterative methods that improve certain aspects of classical schemes such as: speed of convergence, applicability to non differential problems, accessibility or efficiency. A large part of the work developed in this thesis focuses on the study of iterative methods for multidimensional problems, in particular, we have specialised on derivative-free schemes. In addition, part of this Doctoral Thesis is centred on the study of the local and semilocal convergence of methods already developed in the recent literature or of new iterative methods designed in this work. This study guarantees the existence of a solution given a starting point, the convergence domain of the solutions of the problem and their uniqueness under certain conditions. To complement the study of the convergence of the methods, in some chapters a dynamical study of the methods applied to nonlinear equations is also carried out in order to extrapolate the results to the multidimensional case. In addition, as part of some numerical experiments, the accessibility of different numerical methods has been compared across the basins of attraction represented in different dynamical planes, both for the unidimensional and the multidimensional case. Finally, in most of the chapters of this thesis, the iterative methods studied are applied to the resolution of non-linear problems in Applied Mathematics. These problems can be prepared to taste the designed algorithms or be real problems present in some Applied Sciences such as Engineering, Physics, Chemistry, etc. The results described above form part of this Doctoral Thesis to obtain the title of Doctor in Mathematics. / García Villalba, E. (2024). Métodos iterativos libres de derivadas para la resolución de ecuaciones y sistemas de ecuaciones no lineales [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/204853

Although several studies have proven the accuracy of using a non-contact, air-coupled receiver in nonlinear ultrasonic (NLU) Rayleigh wave measurements, inconsistent results have been observed when working with narrow specimens. The objectives of this research are first, to develop a 3D numerical finite element (FE) model which predicts nonlinear ultrasonic measurements and second, to apply the validated model on the narrow waveguide to determine causes of the previously observed experimental issues. The commercial FE-solver ABAQUS is used to perform these simulations. Constitutive law and excitation source properties are adjusted to match experiments conducted, considering inherent effects of the non-contact detection, such as frequency dependent pressure wave attenuation and signal averaging. Comparison of “infinite” and narrow width simulations outlines various influences which impair the nonlinear Rayleigh wave measurements. When the wave expansion is restricted, amplitudes of the fundamental and second harmonic components decrease more significantly and the Rayleigh wavefronts show an oscillating interaction with the boundary. Because of the air-coupled receiver’s finite width, it is sensitive to these edge effects which alter the observed signal. Thus, the narrow specimen adversely affects key factors needed for consistent measurement of material nonlinearity with an air-coupled, non-contact receiver.

Les fonctions de F_q dans lui-même sont des objets étudiés dans de divers domaines tels que la cryptographie, la théorie des codes correcteurs d'erreurs, la géométrie finie ainsi que la géométrie algébrique. Il est bien connu que ces fonctions sont en correspondance exacte avec les polynômes en une variable à coefficients dans F_q. Nous étudierons trois classes de polynômes particulières: les polynômes Presque Parfaitement Non linéaires (Almost Perfect Nonlinear (APN)), les polynômes planaires ou parfaitement non linéaire (PN) et les o-polynômes.Les fonctions APN sont principalement étudiées pour leurs applications en cryptographie. En effet, ces fonctions sont celles qui offre la meilleure résistance contre la cryptanalyse différentielle.Les polynômes PN et les o-polynômes sont eux liés à des problèmes célèbres de géométrie finie. Les premiers décrivent des plans projectifs et les seconds sont en correspondance directe avec les ovales et hyperovales de P^2(F_q). Néanmoins, leurs champ d'application a été récemment étendu à la cryptographie symétrique et à la théorie des codes correcteurs d'erreurs.L'un des moyens utilisé pour compléter la classification est de considérer les polynômes présentant l'une des propriétés recherchées sur une infinité d'extension de F_q. Ces fonctions sont appelées fonction APN (respectivement PN ou o-polynômes) exceptionnelles.Nous étendrons la classification des polynômes APN et PN exceptionnels et nous donneront une description complète des o-polynômes exceptionnels. Les techniques employées sont basées principalement sur la borne de Lang-Weil et sur des méthodes élémentaires. / Functions from F_q to itself are interesting objects arising in various domains such as cryptography, coding theory, finite geometry or algebraic geometry. It is well known that these functions admit a univariate polynomial representation. There exists many interesting classes of such polynomials with plenty of applications in pure or applied maths. We are interested in three of them: Almost Perfect Nonlinear (APN) polynomials, Planar (PN) polynomials and o-polynomials. APN polynomials are mostly used in cryptography to provide S-boxes with the best resistance to differential cryptanalysis and in coding theory to construct double error-correcting codes. PN polynomials and o-polynomials first appeared in finite geometry. They give rise respectively to projective planes and ovals in P^2(F_q). Also, their field of applications was recently extended to symmetric cryptography and error-correcting codes.A complete classification of APN, PN and o-polynomials is an interesting open problem that has been widely studied by many authors. A first approach toward the classification was to consider only power functions and the studies were recently extended to polynomial functions.One way to face the problem of the classification is to consider the polynomials that are APN, PN or o-polynomials over infinitely many extensions of F_q, namely, the exceptional APN, PN or o-polynomials.We improve the partial classification of exceptional APN and PN polynomials and give a full classification of exceptional o-polynomials. The proof technique is based on the Lang-Weil bound for the number of rational points in algebraic varieties together with elementary methods.

Cette thèse cherche à approfondir la nature et la forme des relations entre les inégalités éducatives et le développement. Elle s’inscrit dans le prolongement des analyses engagées sur les liens éducation/croissance et inégalités/croissance, en essayant d’apporter un éclairage complémentaire sur ces deux relations. Elle vise à retracer de manière stylisée l’évolution des inégalités éducatives particulièrement dans les pays en développement et à caractériser la non-linéarité de la relation à partir de l’estimation de modèles non-paramétriques et semi-paramétriques. Cette thèse est constituée de trois chapitres auxquels correspondent des objectifs, des bases de données et des méthodologies spécifiques. Dans un premier chapitre, nous proposons une nouvelle base mondiale sur les inégalités d’éducation. La majorité des travaux sur la relation entre capital humain et développement économique ont principalement appréhendé la mesure du capital humain à travers des mesures de l’éducation en utilisant notamment la moyenne d’années de scolarisation (stock du capital humain). Notre base de données, qui présente une mesure alternative du capital humain, tend à améliorer sensiblement le mode de calcul des inégalités de l’éducation. Elle exploite toute la richesse des données désagrégées, corrige les pondérations inappropriées et affine certaines hypothèses réductrices sur les durées des cycles d’enseignement et les niveaux d’éducation retenus. Nous avons aussi généralisé la formule proposée par Berthélemy (2006) sur l’indice de Gini de l’éducation. Le domaine de variation possible de cet indice est identifié graphiquement selon la moyenne d’années de scolarisation et les durées cumulées des cycles d’enseignement. Nous mettons en évidence, dans le cadre du chapitre II, l’existence d’une relation non linéaire entre les inégalités dans l’éducation et le développement économique en utilisant des modèles non-paramétriques et semi-paramétriques qui n’exigent pas de formes fonctionnelles prédéfinies à l’avance. Plusieurs phases sont ainsi mises en évidence : les trois premières sont repérées seulement par rapport aux niveaux de développement ; deux autres sont identifiées à la fois par des seuils de développement et d’inégalité d’éducation ; une sixième et dernière phase est définie par rapport au seul niveau d’inégalité d’éducation. Nous montrons que c’est dans la troisième et cinquième phases que la réduction de l’inégalité d’éducation présente l’impact le plus bénéfique sur le développement économique.Au-delà du schéma général mis en évidence sur le plan transnational dans les chapitres I et II, nous explorons dans le chapitre III la nature de cette relation au plan régional dans le cas du Maroc, pour lequel nous disposons de données aux niveaux communal et provincial. La non-linéarité de la relation est aussi confirmée. La troisième phase repérée au chapitre II est subdivisée, dans le cas des provinces marocaines, en deux sous phases qui présentent un impact différencié selon un seuil de développement et d’inégalité d’éducation. / This thesis seeks to deepen the nature and the shape of the relationships between educational inequalities and development. It goes along with the prolongation of the analyses undertaken about the relationships between both education & growth; and inequality & growth, by trying to shed additional light on these two. It aims to retrace, in a schematic way, the evolution of educational inequalities particularly in the developing countries; and also to characterize the nonlinearity of this link using nonparametric and semiparametric estimation models.This thesis consists of three chapters that correspond to specific objectives, databases and methodologies. In the first chapter, we propose a new dataset on the inequalities of education. Most of the studies on the relationship between human capital and economic development have mainly apprehended the measurement of human capital through quantitative education indicators, using namely the average of years of schooling (human capital stock). Our database, which presents an alternative measure of human capital, tends to improve significantly the way in which inequalities in education are calculated. It employs all the abundance of disaggregated data, corrects inappropriate weightings and refines some reductive assumptions about the durations of schooling cycles and the levels of education. We have also generalized the formula proposed by Berthélemy (2006) on the Gini index of education. The possible variation range of this index is graphically identified according to the average years of schooling and the cumulative duration of the schooling cycles. In Chapter II, we reveal the existence of a nonlinear relationship between inequalities in education and economic development using nonparametric and semiparametric models that do not require predefined functional forms. Several phases are therefore highlighted: the first three are identified only according to the level of development; then two other phases are recognized by combining thresholds of both development and education inequality; the sixth and final phase is defined by the educational inequality level alone. We show that the phases during which the reduction of educational inequality presents the most beneficial impact on economic development are the third and the fifth.Beyond the general outline highlighted at the transnational level in Chapters I and II, we explore in Chapter III the nature of this relationship at the regional level in the case of Morocco, for which we have data at both the municipal and provincial levels. We also confirm the nonlinearity of the relationship. The third phase, identified in Chapter II, is divided to two sub-phases in the case of the Moroccan provinces which have a differentiated impact according to a threshold of development and inequality of education.

Este trabalho fornece uma contribuição à análise estrutural não-linear de pavimentos de edifícios de concreto armado com o emprego do Método dos Elementos Finitos. A deformação por esforço cortante é considerada, portanto as teorias de Timoshenko e de Reissner-Mindlin são empregadas nas formulações dos elementos de barra e de placa, respectivamente. As posições dos elementos de barra e de placa são independentes e, portanto, podem ser definidas em diferentes planos. Em conseqüência do exposto, o efeito de membrana deve ser necessariamente considerado na modelagem do pavimento. Para completar o modelo mecânico, as não-linearidades físicas descrevem o comportamento do concreto e do aço. A deterioração do concreto no cisalhamento é também considerada através de um modelo simplificado que é proposto para a modelagem do cisalhamento em condições de serviço. / This work gives a contribution to the non-linear structural analysis of reinforced concrete buildings floors using the Finite Element Method. The shear strain components are taken into account by adopting the Timoshenko\'s beam theory together with and the Reissner-Mindlin\'s theory for plate bending. Bar and plate element position are independent and therefore can be defined at different planes. As several level are considered when defining the structure membrane effects are necessary considered. In order to complete the mechanical model, physical non-linearities are also assumed to describe concrete and steel behaviours. The deterioration of the concrete material in shear is also taken account. For this purpose, a simplified model is adopted to compute approximately the damaged shear component in the steel direction.

Peu d’outils ont été développés pour le contrôle qualité industriel de textures sur des lignes de production. L’utilisation de techniques sans contact, à base d’ondes acoustiques, présente un avantage évident dans les secteurs agroalimentaire et cosmétique. L’acousto-élasticité dynamique (DAET) est une technique d’interaction entre une onde pompe basse-fréquence de compression/dilatation et une onde sonde ultrasonore, conduisant à l’estimation des propriétés viscoélastiques de la matière. Nous avons développé un modèle décrivant les variations du module viscoélastique : quantification des paramètres non linéaires élastiques (B/A,. C/A...) et visqueux (ωηB/A, ωηC/A...). La méthode DAET et le modèle ont ensuite été validés dans des milieux homogènes (eau, huiles et gels) conduisant à des non-linéarités viscoélastiques faibles (B/A < 15 et (ωηB/A <1). essentiellement liées à la composition du fluide ou un changement d’état. Les résultats les plus probants ont été obtenus dans des milieux granulaires ou constitués d’air (poudres sèches, mousses et billes creuses ): valeurs élevées des non-linéarités viscoélastiques (quadratiques, cubiques). Cette méthode pourra s’avérer être une alternative intéressante à la rhéométrie conventionnelle, en particulier pour la caractérisation de ces fluides complexes. / Few tools have been developed for industrial quality control of textures on production lines. The use of contactless techniques. based on acoustic wavcs, offers an obvious advantage in food-processing industry and cosmetics. The dynamic acoustoelastic testing (DAET) is based on the interaction between a low-frequency compressionlexpansion pump wave and an ultrasound probe wave. leading to the quantification of the viscoelastic properties of the matter. We have initially developed a model describing the variations of the viscoelastic modulus: quantification of nonlinear elastic (B/A, C/A...) and viscous (ωηB/A. ωηC/A...) parameters. The DAET method and related model were then validated in homogeneous media (water, oils and gels) leading to low values of viscoelastic nonlineaiities (B/A<15 and ωηB/A<1), essentially governed by the fluid nature or state change. However, the most significant results were obtained in granular or air-based media (dry powder, foam and hollow beads): high values of quadratic and cubic nonlinearities due to microinhomogeneitics. This method appears to be an interesting alternative to conventional rheometry, especially for the characterization of these complex fluids.

Ces dernières années, les métamatériaux acoustiques sont largement étudiés pour leur intérêt dans la réalisation de divers types de contrôle des ondes à une échelle sub-longueur d’onde. En particulier, les métasurfaces acoustiques ont montré leur capacité à manipuler des ondes en limites de milieux de propagation via les processus de réflexion, de transmission et de réfraction. Contrairement au régime linéaire qui concerne l’immense majorité des travaux sur les métamatériaux acoustiques, les études sur les propriétés non linéaires des métamatériaux, de surcroit des métasurfaces, restent peu nombreuses, malgré la possibilité de générer des phénomènes acoustiques riches et variés. Les principaux freins au développement des métamatériaux non linéaires sont l'efficacité généralement faible de la réponse non linéaire et le manque de contrôle sur cette non-linéarité. Les travaux de recherche présentés ici ont donc pour objectif de concevoir des architectures de métasurfaces élastiques, permettant un contrôle des ondes acoustiques dans le régime non linéaire. En particulier l’effet de conversion d’une onde fondamentale vers son deuxième harmonique est étudié dans le processus de réflexion et de transmission unidirectionnelle. Cela nécessite le design de la non-linéarité élastique, qui est réalisé à base de modélisations discrètes de systèmes masses-ressorts et d'architectures composées d'éléments tournants. Les métasurfaces ainsi conçues, résonantes et à non-linéarité contrôlée, permettent de générer des effets non linéaires acoustiques inhabituels, potentiellement intéressants pour la manipulation d'ondes acoustiques. / In recent years, acoustic metamaterials have proven to be of great interest for their ability to achieve a variety of wave control at sub-wavelength scale. In particular, acoustic metasurfaces have shown their ability to manipulate waves from the boundaries of propagation media, via the reflection, transmission and refraction processes. Unlike the linear regime which has been extensively investigated in acoustic metamaterials, studies of the nonlinear acoustic properties of metamaterials, especially nonlinear acoustic metasurfaces, are quite scarce, despite the possibility to lead to a rich and diverse set of non-trivial acoustic phenomena. The key limitations in the development of nonlinear acoustic metamaterials are the typically weak efficiency of their nonlinear response together with the lack of control on this nonlinearity. This PhD research is thus dedicated to the design of nonlinear elastic metamaterial and metasurface architectures, enabling acoustic wave control in the nonlinear regime. Specifically, the conversion effect from a fundamental wave to its second harmonic is studied through the one-dimensional scattering process (reflection and transmission) by metasurfaces. This requires the elastic nonlinearity management, realized via the discrete modeling of lumped-element systems and architectures made of rotating units. Such designed metasurfaces, resonating and with harnessed nonlinearity, can create unusual nonlinear acoustic effects, potentially interesting for wave control. This research open the path to a more systematic study of nonlinear acoustic wave manipulation by metamaterials.

Produire et faire interagir entre eux des photons optiques de façon contrôlée sont deux conditions nécessaires au développement de communications quantiques à longue distance, et plus généralement au traitement quantique d’information codée sur des photons. Cette thèse présente une étude expérimentale de solutions possibles a ces deux problèmes, en utilisant la conversion des photons en excitations collectives (polaritons) dans un nuage d’atomes froids, placé dans le mode d’une cavité optique de faible finesse (~100). Dans un premier temps, des polaritons entre états atomiques fondamentaux sont utilisés pour « mettre en mémoire » une excitation unique dans le nuage. Celle-ci est ensuite convertie efficacement en un photon unique, dont le champ est analysé par tomographie homodyne. La fonction de Wigner de l’état à un photon est reconstruite a partir des données expérimentales, et présente des valeurs négatives, démontrant que les degrés de liberté de ce photon (mode spatio-temporel et état quantique) sont complètement contrôlés. Dans un second temps, les photons sont couplés à des polaritons impliquant des états de Rydberg. Les fortes interactions dipolaires entre ces derniers se traduisent par des non-linéarités optiques dispersives très importantes, qui sont caractérisées dans un régime d’excitation classique. Ces non-linéarités peuvent être amplifiées jusqu’à ce qu’un seul photon suffise à modifier totalement la réponse du système, permettant en principe de générer des interactions effectives entre photons. / Controllably producing optical photons and making them interact are two key requirements for the development of long-distance quantum communications, and more generally for photonic quantum information processing. This thesis presents experimental studies on possible solutions to these two problems, using the conversion of the photons into collective excitations (polaritons) in a cold atomic cloud, inside the mode of a low-finesse optical cavity (~100). Firstly, ground-state polaritons are used to store a single excitation in the cloud memory. This polariton is then efficiently converted into a single photon, whose field is characterized via homodyne tomography. The single photon state’s Wigner function is reconstructed from the experimental data and exhibits negative values, demonstrating that the photon’s degrees of freedom (spatio-temporal mode and quantum state) are well controlled. Secondly, photons can be coupled to polaritons involving Rydberg states. The strong dipolar interactions between these give rise to very strong optical dispersive nonlinearities, that are characterized in a classical excitation regime. These nonlinearities can be amplified until a single photon is enough to modify the entire system’s response, allowing in principle for the generation of effective photon-photon interactions.