京都大学 / 0048 / 新制・課程博士 / 博士(工学) / 甲第21109号 / 工博第4473号 / 新制||工||1695(附属図書館) / 京都大学大学院工学研究科電子工学専攻 / (主査)教授 白石 誠司, 教授 木本 恒暢, 教授 山田 啓文 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM

When a liquid is cooled below its equilibrium freezing temperature, it becomes supercooled and the molecular motions slow down until the system becomes kinetically arrested, forming a glass, at the glass transition temperature. These amorphous materials have the mechanical properties of a solid while retaining the structural properties of a liquid, but do not exhibit the usual features of a thermodynamic phase transition. As such, they present a number of important challenges to our understanding of the dynamics and thermodynamics of condensed phases. For example, supercooled liquids are classified on the basis of the temperature dependence of their transport properties and structural relaxations times. Strong liquids display an Arrhenius behavior, with the logarithm of their viscosity growing linearly with inverse temperature. Fragile liquids behave in a super-Arrhenius manner, where the viscosity appears to diverge at temperatures above absolute zero, suggesting the possibility of an underlying thermodynamic origin to the glass transition. Some complex, network forming liquids, such as water and silica have also been shown to have a dynamical crossover from fragile to strong liquid behavior as the temperature is decreased. The potential energy landscape paradigm, combined with inherent structure formalism, provide a framework for connecting the way particles pack together with the thermodynamics and dynamics of the liquid and glassy phases. However, the complexity of this multi-dimensional surface makes it difficult to fully characterize and rigorous relationships between the nature of particle packing and glass forming properties have not been established. The goal of this thesis is to study some of the general features of glass transition and find the connection between the dynamics and the thermodynamics of glass forming liquids. To this end, the packing landscapes of quasi-one-dimensional hard discs and hard spheres are studied. A two dimensional system of hard discs with diameter σ, confined between two hard walls (lines) of length L, separated by a distance 1<Hd/σ< 1+√(3/4), is studied by using the Transfer Matrix (TM) method and Molecular Dynamics (MD) simulations. The complete packing landscape is characterized in terms of the density distribution of inherent structures and the number of local defect states. It is shown that this model exhibits a dynamic fragile-strong liquid crossover at the maximum in the constant pressure heat capacity (Cp) for the system, similar to that observed in anomalous network forming liquids such as water and silica. Furthermore, we find that rescaling the relaxation times of systems with different channel widths by the relaxation time at the Cp maximum causes all the data to collapse on a single master curve. The Cp maximum occurs at a critical value of the defect concentration. At high defect concentrations, where the defects interact, the fluid is fragile. When the defect concentration is low, relaxation appears to occur through the hopping of isolated defects, leading to Arrhenius dynamics. This suggests the thermodynamics associated with the Cp maximum is intimately related to the dynamic crossover. A system of three-dimensional hard spheres confined in a narrow channel was used to study the effect of a more complicated landscape on the dynamics of the system. For this system, the thermodynamic and dynamic properties of the system were studied for two different channel diameters, the 1<Hd/σ<1+√(3/4) case, which only allows first neighbors contact for the spheres and, 1+√(3/4)< Hd/σ < 1.98, which allows second neighbors contact to exist. For the first case, the TM method was implemented to obtain the thermodynamic properties and MD simulation was used to measure the dynamics. For the case that the second neighbor contact is allowed 1+√(3/4)< Hd/σ < 1.98. The thermodynamic and dynamic properties were obtained using MD simulations. In this channel diameter range, the system creates chiral helical jammed packings and defect states appear where sections of helices with different local chiralities come into contact. The equation of state (EOS) shows the presence of two heat capacity maxima. The high density Cp maximum is linked to fragile strong crossover. Finite size scaling analysis shows that the low density Cp maximum is related to an orientational order transition stabilized by the presence of the defects. This type of transition has been shown to exist in bulk two-dimensional systems but this work is the first study that provides strong evidence of the existence of this transition in a quasi-one-dimensional system in a system with short-range interactions.

Spatial confinement of electrons and their interactions as well as confinement of the spin dimensionality often yield drastic changes of the electronic and magnetic properties of solids. Novel quantum transport and optical phenomena, involving electronic spin degrees of freedom in semiconductor heterostructures, as well as a rich variety of exotic quantum ground states and magnetic excitations in complex transition metal oxides that arise upon such confinements, belong therefore to topical problems of contemporary condensed matter physics. In this work electron spin systems in reduced dimensions are studied with Electron Spin Resonance (ESR) spectroscopy, a method which can provide important information on the energy spectrum of the spin states, spin dynamics, and magnetic correlations. The studied systems include quasi onedimensional spin chain materials based on transition metals Cu and Ni. Another class of materials are semiconductor heterostructures made of Si and Ge. Part I deals with the theoretical background of ESR and the description of the experimental ESR setups used which have been optimized for the purposes of the present work. In particular, the development and implementation of axial and transverse cylindrical resonant cavities for high-field highfrequency ESR experiments is discussed. The high quality factors of these cavities allow for sensitive measurements on μm-sized samples. They are used for the investigations on the spin-chain materials. The implementation and characterization of a setup for electrical detected magnetic resonance is presented. In Part II ESR studies and complementary results of other experimental techniques on two spin chain materials are presented. The Cu-based material Linarite is investigated in the paramagnetic regime above T > 2.8 K. This natural crystal constitutes a highly frustrated spin 1/2 Heisenberg chain with ferromagnetic nearest-neighbor and antiferromagnetic next-nearestneighbor interactions. The ESR data reveals that the significant magnetic anisotropy is due to anisotropy of the g-factor. Quantitative analysis of the critical broadening of the linewidth suggest appreciable interchain and interlayer spin correlations well above the ordering temperature. The Ni-based system is an organic-anorganic hybrid material where the Ni2+ ions possessing the integer spin S = 1 are magnetically coupled along one spatial direction. Indeed, the ESR study reveals an isotropic spin-1 Heisenberg chain in this system which unlike the Cu half integer spin-1/2 chain is expected to possess a qualitatively different non-magnetic singlet ground state separated from an excited magnetic state by a so-called Haldane gap. Surprisingly, in contrast to the expected Haldane behavior a competition between a magnetically ordered ground state and a potentially gapped state is revealed. In Part III investigations on SiGe/Si quantum dot structures are presented. The ESR investigations reveal narrowlines close to the free electron g-factor associated with electrons on the quantum dots. Their dephasing and relaxation times are determined. Manipulations with sub-bandgap light allow to change the relative population between the observed states. On the basis of extensive characterizations, strain, electronic structure and confined states on the Si-based structures are modeled with the program nextnano3. A qualitative model, explaining the energy spectrum of the spin states is proposed.

Spatial confinement of electrons and their interactions as well as confinement of the spin dimensionality often yield drastic changes of the electronic and magnetic properties of solids. Novel quantum transport and optical phenomena, involving electronic spin degrees of freedom in semiconductor heterostructures, as well as a rich variety of exotic quantum ground states and magnetic excitations in complex transition metal oxides that arise upon such confinements, belong therefore to topical problems of contemporary condensed matter physics. In this work electron spin systems in reduced dimensions are studied with Electron Spin Resonance (ESR) spectroscopy, a method which can provide important information on the energy spectrum of the spin states, spin dynamics, and magnetic correlations. The studied systems include quasi onedimensional spin chain materials based on transition metals Cu and Ni. Another class of materials are semiconductor heterostructures made of Si and Ge. Part I deals with the theoretical background of ESR and the description of the experimental ESR setups used which have been optimized for the purposes of the present work. In particular, the development and implementation of axial and transverse cylindrical resonant cavities for high-field highfrequency ESR experiments is discussed. The high quality factors of these cavities allow for sensitive measurements on μm-sized samples. They are used for the investigations on the spin-chain materials. The implementation and characterization of a setup for electrical detected magnetic resonance is presented. In Part II ESR studies and complementary results of other experimental techniques on two spin chain materials are presented. The Cu-based material Linarite is investigated in the paramagnetic regime above T > 2.8 K. This natural crystal constitutes a highly frustrated spin 1/2 Heisenberg chain with ferromagnetic nearest-neighbor and antiferromagnetic next-nearestneighbor interactions. The ESR data reveals that the significant magnetic anisotropy is due to anisotropy of the g-factor. Quantitative analysis of the critical broadening of the linewidth suggest appreciable interchain and interlayer spin correlations well above the ordering temperature. The Ni-based system is an organic-anorganic hybrid material where the Ni2+ ions possessing the integer spin S = 1 are magnetically coupled along one spatial direction. Indeed, the ESR study reveals an isotropic spin-1 Heisenberg chain in this system which unlike the Cu half integer spin-1/2 chain is expected to possess a qualitatively different non-magnetic singlet ground state separated from an excited magnetic state by a so-called Haldane gap. Surprisingly, in contrast to the expected Haldane behavior a competition between a magnetically ordered ground state and a potentially gapped state is revealed. In Part III investigations on SiGe/Si quantum dot structures are presented. The ESR investigations reveal narrowlines close to the free electron g-factor associated with electrons on the quantum dots. Their dephasing and relaxation times are determined. Manipulations with sub-bandgap light allow to change the relative population between the observed states. On the basis of extensive characterizations, strain, electronic structure and confined states on the Si-based structures are modeled with the program nextnano3. A qualitative model, explaining the energy spectrum of the spin states is proposed.:Abstract i Contents iii List of Figures vi List of Tables viii 1 Preface 1 I Background and Experimental 5 2 Principles of ESR 7 2.1 The Resonance Phenomenon . . . . . . . . . . . . . . . . . . . 7 2.2 ESR Spectrum . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 2.2.1 The g -factor . . . . . . . . . . . . . . . . . . . . . . . . . 11 2.2.2 Relaxation Times . . . . . . . . . . . . . . . . . . . . . . 12 2.2.3 Lineshape Properties . . . . . . . . . . . . . . . . . . . . 13 2.3 Effective Spin Hamiltonian . . . . . . . . . . . . . . . . . . . . . 15 2.4 Spin-Orbit Coupling . . . . . . . . . . . . . . . . . . . . . . . . . 16 2.5 d-electrons in a Crystal Field . . . . . . . . . . . . . . . . . . . . 17 2.6 Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 2.6.1 Dipolar Coupling . . . . . . . . . . . . . . . . . . . . . . 23 2.6.2 Exchange Interaction . . . . . . . . . . . . . . . . . . . . 23 2.6.3 Superexchange . . . . . . . . . . . . . . . . . . . . . . . 24 2.6.4 Symmetric Anisotropic Exchange . . . . . . . . . . . . 25 2.6.5 Antisymmetric Anisotropic Exchange . . . . . . . . . . 25 2.6.6 Hyperfine Interaction . . . . . . . . . . . . . . . . . . . 26 3 Experimental 27 3.1 Setup for Experiments at 10GHz . . . . . . . . . . . . . . . . . 27 3.2 Implementation of an EDMR Setup . . . . . . . . . . . . . . . . 29 3.2.1 Basic Characterization . . . . . . . . . . . . . . . . . . . 31 3.3 High Frequency Setup . . . . . . . . . . . . . . . . . . . . . . . . 31 3.3.1 MillimeterWave Vector Network Analyzer . . . . . . . 33 3.3.2 Waveguides and Cryostats . . . . . . . . . . . . . . . . . 34 3.4 Development of the Resonant Cavity Setup . . . . . . . . . . . 35 3.4.1 Mode Propagation . . . . . . . . . . . . . . . . . . . . . 38 3.4.2 Resonant CavityModes . . . . . . . . . . . . . . . . . . 40 3.4.3 Resonant Cavity Design . . . . . . . . . . . . . . . . . . 41 3.4.4 Resonant Cavity Sample Stick . . . . . . . . . . . . . . . 45 3.4.5 Experimental Characterization . . . . . . . . . . . . . . 47 3.4.6 Performing an ESR Experiment . . . . . . . . . . . . . . 53 II Quasi One-Dimensional Spin-Chains 57 4 Motivation 59 5 Quasi One-Dimensional Systems 61 5.1 Magnetic Order and Excitations . . . . . . . . . . . . . . . . . . 63 5.2 Competing Interactions . . . . . . . . . . . . . . . . . . . . . . . 64 5.3 Haldane Spin Chain . . . . . . . . . . . . . . . . . . . . . . . . . 66 6 Linarite 69 6.1 Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 6.2 Magnetization and ESR . . . . . . . . . . . . . . . . . . . . . . . 71 6.3 NMR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 6.4 Summary and Conclusion . . . . . . . . . . . . . . . . . . . . . 81 6.5 Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 7 The Ni-hybrid NiCl3C6H5CH2CH2NH3 83 7.1 Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 7.2 Susceptibility andMagnetization . . . . . . . . . . . . . . . . . 85 7.3 ESR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 7.4 Further Investigations . . . . . . . . . . . . . . . . . . . . . . . . 95 7.5 Summary and Conclusion . . . . . . . . . . . . . . . . . . . . . 96 8 Summary 99 III SiGe Nanostructures 101 9 Motivation 103 10 SiGe Semiconductor Nanostructures 107 10.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 10.1.1 Silicon and Germanium . . . . . . . . . . . . . . . . . . 107 10.1.2 Epitaxial Growth of SiGe Heterostructures . . . . . . . 109 10.1.3 Strain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 10.1.4 Band Deformation . . . . . . . . . . . . . . . . . . . . . 112 10.2 Sample Structure and Characterization . . . . . . . . . . . . . 114 11 Modelling of SiGe/Si Heterostructures 119 11.1 Program Structure . . . . . . . . . . . . . . . . . . . . . . . . . . 120 11.2 Implementation of Si/Ge System . . . . . . . . . . . . . . . . . 121 11.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 11.3.1 Single Quantum Dot . . . . . . . . . . . . . . . . . . . . 123 11.3.2 Multiple Quantum Dots . . . . . . . . . . . . . . . . . . 127 11.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130 11.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132 12 ESR Experiments on Si/SiGe Quantum Dots 135 12.1 ESR on Si Structures . . . . . . . . . . . . . . . . . . . . . . . . . 135 12.2 Experimental Results . . . . . . . . . . . . . . . . . . . . . . . . 137 12.2.1 Samples grown at 600◦C . . . . . . . . . . . . . . . . . . 138 12.2.2 Samples grown at 700◦C . . . . . . . . . . . . . . . . . . 139 12.2.3 T1-Relaxation Time . . . . . . . . . . . . . . . . . . . . . 143 12.2.4 Effect of Illumination . . . . . . . . . . . . . . . . . . . . 145 12.3 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148 12.3.1 Quantum Dots . . . . . . . . . . . . . . . . . . . . . . . 149 12.3.2 Assignment of ESR Lines . . . . . . . . . . . . . . . . . . 150 12.3.3 Relaxation Times . . . . . . . . . . . . . . . . . . . . . . 153 12.3.4 Donors in Heterostructures . . . . . . . . . . . . . . . . 153 12.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156 13 Summary and Outlook 157 Bibliography 163 Acknowledgements 176

In this paper, we mainly discuss the transport properties of the two-dimensional gas of a high-mobility GaAs/AlGaAs semiconductor heterostructure in high magnetic fields and low temperatures. We analyzed the measured longitudinal resistivity and Hall resistivity at the five different temperatures. We observed that the classical Hall effect is valid when the magnetic field is less than 0.25 Tesla; and the quantum Hall plateaux appeared obviously when the magnetic field is larger than 1.6 Tesla. We proceeded to analyze the longitudinal resistivity oscillation occurred in the magnetic fields between 0.477 Tesla and 1.483 Tesla. According to the Lifshitz-Kosevich (LK) formula, we can get the two-dimensional electron concentration, effective mass, and quantum scattering time from the quantum magnetoresistivity oscillation measurement. Our results suggested that the applicable range of the LK formula could be broader than the generally-assumed one. In quantum Hall effect regime at high magnetic field, we can calculate the h/e2 value from the quantum Hall plateaux value. In classical Hall effect regime, the three-dimensional electron concentration and classical mobility (classical scattering time) can be obtained. However, we find out that the zero-field Hall resistivity experimental value is not equal to zero, and this is not conformed to the standard theory. We tried to use the magnetic field shift and Hall resistivity shift to solve the problem, and compared both advantages of them. Finally, we observed the plateau-plateau phase transitions of the two-dimensional electron system

This thesis describes the neutron scattering measurements of magnetic excitations in spin-chains and ladders. The first part discusses an experimental investigation of the copper oxide family Sr₁₄Cu₂₄O₄₁ composed of edge-sharing chains and spin-ladders. The study of La₄Sr₁₀Cu₂₄O₄₁ comprises a slightly hole-doped chain and an undoped ladder structure where the chain can be modeled by a ferromagnetic nearest and an antiferromagnetic next-nearest neighbor coupling. The hole effects are apparent in gaps in the dispersion relation and can be described by a charge-density wave agreeing with the commensuration of the dispersion. Investigating the undoped ladder establishes the exchange constants including a cyclic exchange manifested by the two-magnon continuum and the suppression of the S = 1 bound mode. An orbital consideration provides an explanation for the exchanges including the different sizes of rung and leg coupling. The excitation spectrum of the doped ladder in Ca₂.₅Sr₁₁.₅Cu₂₄O₄₁ can be described by a direct comparison with the undoped ladder and the differences consisting of a higher energy mode and subgap scattering can be successfully modeled by the charge spectrum of the ladder calculated from the free electron model. The second part of the thesis investigates the alternating chain material Cu(NO₃)₂ · 2.5D2O and establishes the gapped one-magnon dispersion, the two-magnon continuum and for the first time the S =1 bound mode. Applying magnetic field drives the system through two critical field transitions, condensation of magnons into the ground state and saturation. The modes beyond saturation can be modeled by spin wave theory and the excitations at the first critical field follow Luttinger Liquid behavior. Additionally investigated are the temperature effects with the excitations being of a different nature but containing the signature of a strong correlated system. For an outlook the measurements including temperature and field are provided with further theoretical descriptions necessary.

京都大学 / 0048 / 新制・課程博士 / 博士(工学) / 甲第20088号 / 工博第4255号 / 新制||工||1659(附属図書館) / 33204 / 京都大学大学院工学研究科材料工学専攻 / (主査)教授 中村 裕之, 教授 酒井 明, 教授 吉村 一良 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM

Ce travail porte sur la régulation de la sortie des systèmes aux paramètres distribués. Pour ce faire, un simple contrôleur proportionnel intégral est utilisé, puis la stabilité du système en boucle fermée est démontrée à l'aide d'une fonction de Lyapunov. La principale contribution de ce travail est la construction d'un nouveau type de fonction de Lyapunov qui s'inspire d'une méthode bien connue dans le cadre des systèmes non-linéaires : le forwarding.Dans une première partie, le système est établi avec des opérateurs dont les propriétés sont données dans le cadre des semigroupes, puis la problématique de la régulation par contrôleur P-I est posé. Grâce à cette construction de Lyapunov, on peut répondre à cette problématique sous certaines hypothèses. Le lien est alors fait avec les résultats existants dans ce contexte. Dans la seconde partie, la problématique de la régulation est posée pour un système composé de n × n équations hyperboliques linéaires où l'entrée et la sortie se situent sur les conditions aux frontières des EDPs. Sous certaines hypothèses, le résultat permet de réguler n'importe quel vecteur de sortie composé linéairement des conditions aux frontières du système. Cela généralise de nombreux travaux portant sur les systèmes composés d'équations hyperboliques et l'utilisation de contrôleur P-I. Enfin dans le dernier chapitre, les vibrations mécaniques dans les tiges de forage sont étudiées comme cas d'application. Dans un premier temps, le comportement de solutions issues de différents modèles utilisés pour l'étude de ces déformations est détaillé. Dans un second temps, il est montré que la nouvelle fonctionnelle de Lyapunov permet de prendre en compte des modèles plus complexes et d'obtenir la régulation de la vitesse de la tige au fond du forage en ne mesurant que la vitesse en haut du puits. A la fin du chapitre, de nombreuses simulations numériques viennent illustrer nos résultats théoriques / This monograph is devoted to the output regulation of some distributed parameters systems. To reach this objective, a simply proportional integral controller is implemented. Then the stability of the closed loop is proved using a Lyapunov functional that can be built given a Lyapunov functional for the open-loop system. The main contribution of this work is the method to build the Lyapunov functional, it is inspired by a well-known method in non-linear system theory : the forwarding. In a first part, the system studied is an abstract Cauchy problem and the problematic is stated using semigroup theory. Thanks to the Lyapunov employed, the regulation can be guaranteed providing some assumption on the systems operators. The second part detailed how the output regulation can be obtain for all linear outputs when the system is a n × n systems of linear balance laws in one space dimension. The result is given in the case where inputs and outputs act on the PDE’s boundary conditions and for open-loop stabilizable system. It generalize many contribution in the topic of output regulation for systems of linear balance laws. Last but not least, a part is devoted to the study of mechanicals vibrations in a drill pipe. In a first time, the behavior of the solutions for different kind of models use to model the drill pipe is detailed. Then, it is shown that the new Lyapunov functional allow to take into account complex, infinite dimensional model and to regulate the drill pipe velocity at the bottom of the wellbore by only measuring the surface velocity and with a P-I controller. At the end, some simulations are given that illustrate the result

Les spectres de bruit mesurés en cabine d’hélicoptère montrent que la boîte de transmission principale (BTP) est un des principaux contributeurs au bruit perçu par les usagers. Elle génère en effet plusieurs raies émergeant fortement du bruit large bande et dont les fréquences se situent dans la plage de sensibilité maximale de l’oreille humaine. Dans un contexte d’amélioration permanente du confort acoustique des usagers, un modèle numérique est développé pour prédire le comportement dynamique des BTP. Les équations du mouvement sont écrites sur la base de fonctions du temps représentatives des excitations générées par l’engrènement (raideur d’engrènement et erreurs de transmission). Plusieurs éléments de validation sont présentés pour confirmer la pertinence de la formulation proposée. Différents résultats numériques et expérimentaux de la littérature sont utilisés à des fins de comparaison, montrant que le modèle s’applique aux systèmes à simple étage de réduction, par engrenage cylindrique ou spiro-conique. La validation est ensuite étendue aux systèmes à deux étages de réduction et les résultats confirment que la formulation basée sur les erreurs de transmission permet de tenir compte des corrections de profil. Finalement, le modèle est utilisé pour diverses applications. Premièrement, l’influence des erreurs de pas sur le comportement dynamique de transmissions par engrenages est discutée, ainsi que l’influence combinée du niveau de chargement appliqué. Dans un second temps, la relation entre l’erreur de transmission dynamique et différents coefficients dynamiques est étudiée. Le contenu spectral de la réponse au niveau des roulements est ensuite analysé pour des systèmes à deux engrènements cylindriques et l’influence de différents paramètres est discutée. Enfin, une application est réalisée sur un système comprenant un engrenage cylindrique et un engrenage spiro-conique. Les phénomènes de couplage entre les étages successifs sont mis en évidence ainsi que la contribution des deux engrènements au contenu spectral de la réponse aux roulements. / Noise measurements have shown that helicopters main gearboxes highly contribute to the overall cabin noise. Gear mesh vibrations propagate through the shafts to the rolling element bearings and the casing which becomes a source of radiated noise. The latter is characterized by high-amplitude tones emerging from broadband noise whose frequencies lie in the range of maximum human ear sensitivity. In the context of continuous improvement in the acoustic comfort of helicopter passengers, it is therefore necessary to analyse and optimize gearbox vibrations in order to reduce casing noise radiations. The research work presented in this memoir is focused on the development of a numerical model dedicated to the prediction of gear system dynamic behaviour, comprising several gear stages and different types of gears. This model relies on classic beam and lumped parameter elements along with specific two-node gear elements for both cylindrical (spur, helical) and spiral-bevel gears. The equations of motion are developed based on time-varying functions representative of mesh excitations which comprise: (a) mesh stiffness functions, (b) quasi-static transmission error under load, and (c) kinematic (or no-load) transmission error. A number of comparisons with benchmark numerical and experimental results from the literature are presented which demonstrate that the proposed approach is sound as far as single-stage systems with spur, helical or spiral-bevel gears are considered. Validations are then extended to double-stage gears and, here again, it is confirmed that the proposed transmission error based formulation is accurate and can account for tooth shape modifications. In the second part of the memoir, several examples of application are presented and commented upon. First, the combined influence of tooth pitch errors and load on the dynamic behaviour of gear transmissions is tackled. An extended three-dimensional model and a reduced torsional version are then confronted in order to investigate the dependency between dynamic transmission errors and mesh force / root stress dynamic factors. Further investigations on bearing dynamic response in two-stage spur gear systems are conducted and the particular contributions of profile modifications are analysed. Finally, a system combining a cylindrical gear and a spiral-bevel gear is considered and particular attention is paid to the dynamic couplings between the various meshes and their influence on bearing dynamic responses.

L’objectif de notre thèse est de comprendre et d’expliquer les modes d’intégration des systèmes comptables des entreprises à partir de quatre cas (deux grandes entreprises et deux entreprises moyennes). Nous nous sommes appuyés sur la théorie de la différenciation-intégration de Lawrence et Lorsch (1967) pour interpréter les situations d’intégration étudiées. Cette théorie montre que les entreprises ayant des structures très différenciées disposent des mécanismes d’intégration sophistiqués tandis que celles dont les structures sont peu différenciées ont des mécanismes de coordination rudimentaires. Nous avons également mobilisé l’approche événementielle et multidimensionnelle de Sorter (1969) qui propose un cadre théorique de l’intégration informatico-comptable. Les progiciels ERP constituent une application très concrète de cette approche. Notre analyse qualitative de quatre cas en termes de similitudes et de différences, appuyée sur la théorie de la différenciation-intégration, a permis d’expliquer les situations d’intégration des sous-systèmes comptables observées mais aussi de montrer la pertinence de l’approche événementielle, comme cadre de conception du système d’information de structures diversifiées et différenciées. Nos résultats montrent, à l’instar de Lawrence et Lorsch, qu’il existe des modes différents d’intégration des systèmes comptables adaptés à des contextes différents. Nous avons ébauché une typologie de quatre configurations types des systèmes comptables (quatre situations d’intégration-différenciation) : (1) Système classique bidimensionnel, (2) Système quasi-intégré centralisé, (3) Système intégré, (4) Système multidimensionnel. La stratégie et la taille paraissent être, au-delà de toutes autres considérations, les principaux déterminants de l’intégration comptable ; viennent ensuite, l’environnement et certains autres éléments du contexte tels que la structure de gestion et le système de contrôle. Cependant, l’impact de la taille diffère d’une entreprise à l’autre. Lorsque la taille est combinée à une stratégie de diversification (et par extension à des environnements variés et dynamiques) son effet différenciateur sur l’organisation (et le système comptable) est très important, ce qui nécessite des mécanismes compensateurs d’intégration développés. Les deux entreprises diversifiées (POULINA et MEDDEB) ont mis en place un système comptable intégré voire multidimensionnel pour maitriser leurs sous-systèmes différenciés. Lorsque la taille est la conséquence d’un simple développement d’une activité unique (monoactivité) son effet différenciateur-intégrateur sur l’organisation et le système comptable est moins important. Les deux entreprises ayant des activités et des structures homogènes (SONEDE et SME) se sont contentées d’un système comptable peu intégré voire bidimensionnel. / The objective of this thesis is to understand and explain the modes of integration of accounting systems of companies from four cases (two big and two medium-sized enterprises). We relied on the theory of differentiation-integration of Lawrence and Lorsch (1967) to interpret the studied integration situations. This theory shows that companies with very different structures have sophisticated mechanisms for integration while those whose structures are poorly differentiated have rudimentary coordination mechanisms. We have also mobilized the event and multidimensional approach of Sorter (1969) who proposes a theoretical framework for the accountant integration. The ERP is a highly practical application of this approach.Our qualitative analysis of four cases study in terms of similarities and differences, supported on the theory of the differentiation-integration, helped to explain the observed situations of integration of accounting subsystems but also to show the relevance of the event-based approach, as part of system design information structures diversified and differentiated. Our results show, like Lawrence and Lorsch, that there are different modes of integration of accounting systems tailored to different contexts. We have developed a typology consisting of four configurations of accounting systems (four situations integration-differentiation): (1) Conventional two-dimensional system, (2) Quasi-integrated centralized system, (3) Integrated system,(4),Multidimensional,system.Strategy and size appear to be, beyond all other considerations, the main determinants of the accounting, followed, environment and other elements of the context of the accounting system such as the management structure and system management control. However, the impact of the size differs from one company to another. When the size is combined with a strategy of diversification (and by extension to various and dynamic environments) its differentiating effect on the organization (and the accounting system) is very important, requiring developed integration compensatory mechanisms. Both diversified firms (POULINA and MEDDEB) have implemented an integrated accounting system to unify their subsystems differentiated. When the size is the result of a simple development of a single activity (monoactivity) its integrator-differentiator effect on the organization and the accounting system is less important. Companies with activities and homogeneous structures (SONEDE and SME) have been content to a poorly integrated accounting system or two-dimensional.