Les gyromètres laser He-Ne sont des senseurs inertiels dont la fiabilité et la précision sont reconnues depuis le milieu des années 1980. Leur grande sensibilité leur permet de mesurer des vitesses de rotation avec une précision qui atteint 10⁻³ °/ h dans le domaine aéronautique. Cependant, du fait d’un fonctionnement complexe basé sur une physique riche et variée, ses performances sont fortement dépendantes des conditions de fonctionnement et de toute modification du processus de fabrication. Dans ce cas, un travail de modélisation prend tout son sens, puisqu’il permet, outre une compréhension claire et précise des différents phénomènes physiques, un accès à des études paramétriques non envisageables expérimentalement. La modélisation globale du fonctionnement d’un gyrolaser He-Ne a ainsi été l’objectif principal de la collaboration entre la société Sagem (groupe Safran), un des leaders mondiaux dans le domaine des senseurs inertiels, et le Laboratoire de Physique des Gaz et des Plasmas (LPGP). Cette modélisation est « multiphysique » du fait de la diversité des domaines que couvre la physique du gyrolaser (Plasma, Physique Atomique, Lasers). C’est pourquoi nous avons développé trois modèles spécifiquement adaptés à chaque domaine. Le premier décrit la modélisation de la colonne positive du plasma de décharge dans une approche fluide. Ce modèle permet une description quantitative du plasma et l’accès aux grandeurs telles que la densité électronique et la fonction de distribution en énergie des électrons. Ces grandeurs sont les entrées nécessaires au second modèle qui traite la cinétique des états excités du plasma He-Ne. Un modèle collisionnel-radiatif à 1 dimension radiale (1D-CRM) a ainsi été développé. L’aspect 1D se justifie par l’importance des phénomènes de transport d’atomes et de rayonnement pouvant influer sur le profil radial de l’inversion de population du laser. Le transfert radiatif par auto-absorption des transitions radiatives résonantes a notamment été modélisé en résolvant l’équation de Holstein-Biberman à partir d’une méthode Monte-Carlo. Cet aspect constitue un élément majeur de ce travail de thèse. La diffusion des atomes excités du mélange He-Ne a également été prise en compte en résolvant l’équation de diffusion avec différentes conditions au bord à la surface du capillaire.A partir des populations et des taux cinétiques de peuplement et dépeuplement calculés par 1D-CRM, l’amplification laser dans la cavité a été modélisée dans le cadre d’une approche Maxwell-Bloch à 2 niveaux (NADIA) en incluant la saturation inhomogène du gain c’est-à-dire en tenant compte de la vitesse des atomes émetteurs dans la direction de propagation des faisceaux lasers. La cinétique de NADIA a été optimisée et les processus de transports dans l’espace des phases ont également été implémentés. Ce modèle a été utilisé pour étudier les performances du gyrolaser liées au milieu amplificateur et pour dériver les paramètres physiques nécessaires au développement d’un simulateur du gyrolaser.Dans ce simulateur, un modèle physique simplifié dérivé de NADIA, a été couplé à des modules « systèmes » dans le but de reproduire en sortie le signal opérationnel d’un gyrolaser. Ceci nous a permis de réaliser des études paramétriques sur les grandeurs caractérisant les performances d’un gyrolaser notamment le biais dynamique et le Random-Walk. Nous montrons en particulier que les performances de notre simulateur sont en bon accord avec celles observées en conditions opérationnelles. De plus, nos résultats montrent que ce simulateur est également un outil puissant pour l’analyse de données expérimentales. / Ring laser gyros (RLG) are inertial sensors whose reliability and accuracy have been recognised since the mid-1980s. Their high sensitivity enables them to measure angular velocity with an accuracy of 10⁻³ °/ h in aeronautics. However, because of a complex functioning based on a rich and varied physics, their performances are highly dependent on the working conditions and on any modification in the manufacturing process. In this case, a numerical modelling is pertinent since it allows both a clear understanding of the ring laser physics and parametric studies which are not experimentally feasible. The global modelling of a He-Ne RLG has been the main objective of the collaboration between Sagem (Safran group), which is one of the world leader in the inertial sensors field, and the Gas and Plasma Physics Laboratory (LPGP).This modelling is “multi-physics” since RLG physics involves several disciplines (plasma, atomic and laser physics). Therefore we have developed three models specifically adapted to each field. The first one describes the modelling of the positive column of the glow discharge following a fluid approach. This model allows a quantitative description of the plasma and gives access to fundamental quantities like the electron density or the electron energy distribution function. These quantities are the required inputs for the second model which treats the kinetics of the excited states inside the He-Ne plasma. For this, a collisional-radiative model in a radial geometry (1D-CRM) has been developed. The radial geometry is justified by the importance of the transport processes of atoms and radiations which can influence the radial profile of the population inversion. Notably, the radiative transfer by self-absorption of the resonant radiative transitions has been modelled by solving the Holstein-Biberman equation by a Monte-Carlo method. This aspect is a major component of this PhD work. Diffusion of excited atoms inside the plasma has also been taken into account by solving the diffusion equation with different boundary conditions at the capillary surface. From the populations and the kinetic rates computed by 1D-CRM, the laser amplification inside the cavity has been modelled using a two-level Maxwell-Bloch approach (NADIA) taking into account the inhomogeneous gain saturation, which means to consider the thermal speed of the atoms in the direction of propagation of the laser beams. The kinetics of NADIA has been optimized and transport processes in the phase space have also been implemented. This model has been used to study the performances of the RLG linked to the amplifying medium and to derive the physical parameters needed for the development of a simulator.Concerning this simulator, a simplified physical model from NADIA has been coupled to system modules in order to reproduce the operating signal of a RLG. This allows to conduct parametric studies on the quantities defining the RLG performance in particular the dynamic bias and the so-called “Random Walk”. We showed notably that the results of our simulator are in good agreement with experimental measurements in operating conditions. Moreover, our results show that this simulator is a powerful tool for analysing experimental data.

L’essor des procédés de micro et nano-fabrications rend aujourd’hui accessible la synthèse contrôlée de nanoparticules métalliques (typiquement de 3 à 200nm) offrant de larges résonances d’absorption et de diffusion dont les fréquences peuvent être contrôlées finement en variant judicieusement leur géométrie et leur composition. Dans ce travail de thèse relevant de l’électrodynamique classique établit par Maxwell, nous étudions numériquement l’intérêt de ces particules pour la problématique du (dé)couplage de la lumière piégée dans un film mince diélectrique - une géométrie de référence permettant de rendre compte du phénomène de piégeage qui limite considérablement l’efficacité de dispositifs électroluminescents et de certaines cellules solaires. Pour ce faire, nous proposons quelques règles de conception de nanoparticules capables d’extraire efficacement la lumière piégée. Pour un émetteur seul, environ 20% de la lumière émise est rayonnée hors du guide (rad~0.2). L’ajout d’une monocouche (~50nm d’épaisseur) composée d’un ensemble de particules « optimisées » et aléatoirement positionnées autour de l’émetteur permet d’accroître cette efficacité jusqu’à 70% en moyenne statistique sur le désordre. D’intéressants effets de cohérences liés à la nature du désordre au sein de ladite couche sont également mis en évidence. / Metallic nanoparticles (MNps) exhibit strong plasmonic resonances in their absorption and scattering spectra. Recent advances in micro- and nano-fabrication processes allow scientists to control the particle shape; and thus to tune these resonances on the visible and near-IR spectrum - opening unprecedented applications ranging from imaging techniques to solar cells improvement. In the present work, we numerically investigate the capacity of MNps to (de)couple the light that is confined in guided modes of dielectric thin films—a relevant system to analyze, understand and reduce the light trapping phenomenon that strongly lowers the efficiency of some electroluminescent devices. To this end, we propose, by the control of its polarization state, to optimize the quantity of light that a nanoparticle extracts during a scattering event. For a sole source embedded in the guide, barely 20% of the light is extracted (rad~0.2). The addition of an ultra-thin layer composed of hundreds of randomly deposited engineered-nanoparticles shows promising results with rad ~0.7 (in realistic configurations). Interesting coherence effects arising from the randomness of the disorder are also evidenced.

Simulações acopladas do escoamento reativo e dos processos de transferência de calor para o estudo de chamas são problemas dispendiosos computacionalmente. A transferência de calor por radiação em processos de combustão, devido às elevadas temperaturas, é o processo de troca energética dominante. Ainda, o comportamento altamente irregular do coeficiente de absorção com o comprimento de onda se constitui em uma dificuldade adicional na modelagem da transferência radiativa em meios participantes. Para contornar essa dificuldade modelos espectrais foram desenvolvidos com o objetivo de simular o comportamento de um gás real. Dentre esses modelos destacam-se o gás cinza (GG: Gray Gas), o mais simples, que negligencia o comportamento espectral do coeficiente de absorção, e o modelo da soma-ponderada-de-gases-cinza (WSGG: Weighted-Sum-of-Gray-Gases) onde a integração sobre todo o espectro é substituída por um número finito de gases cinza. Com o avanço de ferramentas computacionais, principalmente códigos CFD (Computational Fluid Dynamics), abordagens computacionais se tornaram atrativas frente ou em complemento às abordagens experimentais. Desse modo, o presente trabalho tem por objetivo a aplicação dos modelos WSGG e GG com novas correlações na simulação detalhada de chamas difusivas laminares de metano diluído com dióxido de carbono e nitrogênio com o código CFD comercial ANSYS/Fluent. Foram desenvolvidas rotinas de usuário (UDF: User-Defined Functions) para o acoplamento dos modelos espectrais ao código CFD. A verificação das rotinas de usuário foi realizada comparando os resultados obtidos via simulação Fluent com dados obtidos pelo modelo WSGG com um código FORTRAN próprio desenvolvido pelo grupo de pesquisa do Laboratório de Radiação Térmica (LRT/UFRGS) para o problema unidimensional de superfícies negras e infinitas preenchidas por um meio não-isotérmico e não-homogêneo. Os erros encontrados para o fluxo de calor radiativo nas superfícies e para o termo fonte radiativo ao longo do meio foram da ordem de 1% indicando o funcionamento correto das rotinas UDF acopladas ao Fluent. Por fim, as rotinas foram aplicadas na simulação numérica para chamas de potência constante com diluição dos reagentes e os dados obtidos com a solução numérica foram comparados com dados experimentais para a fração radiante e fluxo de calor radiativo. Os desvios médios encontrados para o fluxo de calor radiativo ficaram em torno de 10% para todas as chamas, excetuando as chamas com diluição de CO2 de 30%, 40% e 50%, em volume, para as quais os desvios médios ficaram em torno de 15%. O termo fonte para as chamas apontou para a predominância da emissão do meio em relação à absorção. Todas as chamas estudadas se encontram no regime opticamente fino (optically thin) para o qual, segundo apontam estudos da literatura, a escolha do modelo espectral possui impacto pequeno em resultados globais da chama como a temperatura e a concentração das espécies na mistura. Nesse aspecto os resultados encontrados concordaram com a previsão da literatura, entretanto para a transferência radiativa, o modelo GG se mostrou sensivelmente menos preciso em comparação ao modelo WSGG, principalmente para a fração radiante e para o fluxo radiativo na região da pluma aquecida, indicando a dependência do modelo espectral adotado. / Coupled simulations of the reactive flow with the heat transfer processes for flame studying are computationally demanding problems. The radiative transfer in combustion processes is the main heat transfer mechanism due to the high temperatures involved. However, the highly irregular behavior o f the absorption coefficient with the wavenumber composes in an additional difficulty on modeling the radiative transfer in participating media. In order to overcome this issue, spectral models were developed with the objective of simulate the behavior of real gases. Some of the most known models are the gray gas (GG) for which the spectral behavior of the radiative properties of the medium is neglected and the weighted-sum-of-gray-gases (WSGG) for which the integration over the entire spectrum is replaced by a summation over a finite number of gray gases with constant absorption coefficients. With the development of computational tools, mainly Computational Fluid Dynamics (CFD) codes, numerical approaches became attractive instead or in complement of experimental set ups. In this way, the present work aims to couple the WSGG and the GG models with new correlations in a detailed simulation of diffusive laminar flames of methane diluted with carbon dioxide and nitrogen with the commercial CFD code ANSYS/Fluent. User-defined functions (UDF) were developed to the coupling of the spectral models. The verification was carried out through the WSGG model by comparing the Fluent solution with a solution obtained with a FORTRAN code developed by the Thermal Radiation Laboratory (LRT/UFRGS) research group for the one-dimensional system of black surfaces filled with a non-homogeneous and non-isothermal medium. The deviations for the radiative heat flux for the walls and the radiative heat source along the domain were of 1% or less, indicating the correct coupling between the UDF routines and the CFD code. Finally, the UDF were applied in the solution of constant power flames with fuel diluted with carbon dioxide and nitrogen. The obtained data was then compared with experimental measurements for the radiant fraction and the radiative heat flux along the flame axis. The average deviations found were in order of 10% for all flames, except for the flames with 30%, 40% and 50% of CO2 dilution, in volume, for which the deviatioms found were in order of 15%. The radiative heat source was plotted and indicated for the medium emission predominance in comparison with the medium absorption. All flames studied were optically thin flames for which, studies pointed, the spectral model have minor impact over global results as flame temperature and mixture concentration. For this aspect the results found showed agreement with the literature studies predictions, however the GG model showed itself less accurate in comparison with the WSGG model for the radiant fraction and the radiative heat flux computations. So the spectral models have influence on the radiative transfer even if its effect on flame structure can be negligible.

Cette thèse porte sur la mesure du protoxyde d'azote (N2O) à partir de capteurs spatiaux. Dans un premier temps, nous avons étudié les processus d'émissions et de transport de N2O depuis l'Asie jusqu'au bassin méditerranéen. Pour cette étude, nous avons utilisé des sorties du modèle de chimie-transport LMDz-Or-INCA ansi que des profils de N2O estimés à partir d'observations du capteur spatial TANSO-FTS (Thermal And Near infrared Sensor for carbon Observation Fourier Transform Spectrometer) de la plateforme GOSAT (Greenhouses gases Observing SATellite). Ensuite, nous avons mis en place un système de restitution des profils troposphériques de N2O à partir des mesures du capteur spatial infrarouge IASI (Infrared Atmospheric Sounding Interferometer) des plateformes MetOp. Ce système a ensuite été validé en utilisant les mesures in-situ des campagnes aéroportées HIPPO (High performance Instrumented airborne platform for environmental research Pole-to-Pole Observations). Enfin, nous avons étudié l'apport théorique du capteur IASI-NG (IASI-New Generation) par rapport à IASI pour la mesure du N2O troposphérique. / This thesis focuses on the measurement of nitrous oxide (N2O) from space sensors. Firstly, we studied the transport and emission processes of N2O from Asia to the Mediterranean Basin (MB). For this study, we used N2O profiles over the period 2010-2013 retrieved from TANSO-FTS (Thermal And Near infrared Sensor for carbon Observation Fourier Transform Spectrometer) observations onboard the platform GOSAT (Greenhouses gases Observing SATellite) . We also used outputs of the chemistry-transport model LMDz-Or-INCA over the same period. Secondly, we built an algorithm to retrieve N2O profiles using observations from IASI (Infrared Atmospheric Sounding Interferometer) onboard the MetOp platforms. This algorithm was validated by comparing the retrieved profiles with in-situ measurements from HIPPO (High performance Instrumented airborne platform for environmental research Pole-to-Pole Observations) airborne campaigns. Finally, we performed a theoretical intercomparison between IASI-NG (IASI-New Generation) and IASI concerning the tropospheric N2O measurements.

Ce sujet concerne un travail pluridisciplinaire mathématique et astrophysique. Le but de cette thèse est l'étude des modèles d'hydrodynamique radiative dont l'application est bien évidemment très vaste en physique et astrophysique. Les modèles M1-multigroupes sont explorés pour décrire le transfert radiatif sans faire à priori d'hypothèse sur la profondeur optique du milieu. L'intérêt qui découle directement de ce travail est le développement du code d'hydrodynamique radiative HADES 2D permettant le calcul massivement parallèle. Il autorise des simulations dans des configurations astrophysiques réalistes en termes de nombre de Mach et de contraste de densité et de température entre les différents milieux. Nous nous sommes concentrés sur deux applications intéressantes : les jets d'étoiles jeunes et les chocs radiatifs dont les premières simulations seront présentées. / This topic is a multidisciplinary work between mathematics and astrophysics. The aim of this thesis is the study of radiation hydrodynamic models of which application is obviously very broad in physics and astrophysics. M1-multigroup models are explored to describe the radiative transfer without a priori assumption on the optical depth of the medium. The interest ensuing directly from this work is the development of a radiation hydrodynamic code, namely HADES 2D, for massively parallel computing. It allows simulations in realistic astrophysical configurations in terms of Mach number, density and temperature contrasts between different environments. We focused on two interesting applications: the jets from young stars and the radiative shocks of which first simulations will be presented.

Encore jamais été observé, le rayonnement infrarouge moyen et lointain consécutif aux sprites a été prédit et serait lié à l’excitation des états vibrationnels de CO2. En sciences de l’atmosphère, la composition chimique peut être retrouvée par des méthodes de télédétection infrarouge. Pour la Défense, les émissions infrarouges naturelles peuvent causer de fausses alarmes à travers les systèmes de veille optronique satellitaires ou aéroportés. C’est pourquoi il est nécessaire de caractériser les émissions infrarouges des sprites. Pour cela, un modèle de cinétique plasma-vibrationnelle a été développé et couplé à un modèle de transfert radiatif atmosphérique. Celui-ci permet de simuler les effets énergétiques et chimiques consécutifs à la perturbation électrique des streamers, éléments constitutifs des sprites. Les signatures infrarouges évaluées devraient être détectables pour un observateur situé dans la stratosphère ou dans l’espace. Par ailleurs, les effets des incertitudes sur les principaux paramètres du modèle ont été quantifiés à travers une étude de sensibilité. Enfin, ces travaux ont permis de définir certaines spécifications instrumentales pour la mission HALESIS (High Altitude Luminous Events Studied by Infrared Spectro-imagery), qui aura pour but d’observer les sprites et autres phénomènes lumineux de moyenne atmosphère dans l’infrarouge. / Even if it hasn’t been observed yet, the existence of emissions in the middle and far infrared following a sprite is suspected and could be related to vibrational excitation of CO2. In atmospheric sciences, the chemical composition can be retrieved through different remote sensing methods. For the Defense, natural infrared emissions could cause false alarms through airborne and spaceborne optronic detection systems. That is why it is necessary to characterize the infrared emissions of sprites. To do so, a plasma-vibrational kinetic model has been developed and coupled to an atmospheric radiative transfer model. This model allows evaluating the energetic and chemical effects following the electrical perturbation caused by the propagation of streamers, main constituent elements of sprites. The evaluated signatures could be detectable for an observer situated in the stratosphere or in space. Otherwise, the effects of the uncertainties on the principal parameters of the model have been quantified through a sensitivity analysis. Finally, this work allowed defining instrumental specifications for the future mission HALESIS (High Altitude Luminous Events Studied by Infrared Spectro-imagery), which will record hyperspectral infrared images of sprites and other middle atmosphere luminous events.

L'époque de la réionisation, qui s'étend pendant le premier milliard d'années de l'Univers, correspond à la période où les premières étoiles et galaxies apparaissent. Dans ce contexte, l'enjeu majeur de cette thèse est d'étudier la formation de ces premières structures et leur rétroaction radiative sur leur environnement. Pour cela, j'utilise différentes méthodes de simulations numériques permettant de modéliser le transfert radiatif dans les galaxies, dans un cadre cosmologique.En particulier, les plus petites galaxies seraient celles qui apportent la majorité des photons nécessaires à ioniser l'Univers. J'explore cette hypothèse dans une première partie à l'aide de simulations radio-hydrodynamiques avec le code RAMSES-RT, ciblant trois petites galaxies avec une très haute résolution spatiale et temporelle. Je me suis d'abord intéressé aux mécanismes régulant la production et le transfert de photons ionisants dans les galaxies, et j'ai montré que les explosions de supernovae en sont un facteur essentiel. Ensuite, j'ai exploré les propriétés observables dérivées à partir de ces simulations.Dans un second temps, j'ai développé une extension au code Monte Carlo de transfert radiatif MCLya prenant en compte la polarisation du rayonnement et l'émission diffuse dans une simulation. J'ai utilisé ce code pour post-traiter une simulation d'un blob Lyman-alpha, une source étendue d'émission Lyman-alpha, et étudier ses propriétés de polarisation à l'aide de pseudo-observations. Contrairement à ce qui était proposé précédemment, j'ai pu montrer que la polarisation n'était pas un indicateur aussi utile qu'espéré pour tracer l'origine des photons Lyman-alpha / The Epoch of Reionisation, which spans during the first billion year of te Universe, corresponds to the period during which the first stars and galaxy form. In this context, the main topic of this thesis is to study the formation of those early structures and their radiative feedback to their environment. For this purpose, I use various numerical simulations tools designed to model the radiative transfer in galaxies in a cosmological framework.More specifically, I look at very small galaxies, which are believed to contribute the bulk of the photons required to reionise the Universe. I explore this idea using radiative hydrodynamics simulations performed with RAMSES-RT, focusing on three small galaxies with a very high spatial and temporal resolution. I first detail the mechanism that regulate te production and escape of ionising photons in galaxies, and I show that supernovae explosions are a crucial element for this regulation. I then started to investigate the observable properties of those galaxies.In a second part of my thesis, I developped an extension to the Monte Carlo radiative transfer code MCLya to take light polarisation into account and to model the diffuse emission. I applied this code to post-process the simulation of a Lyman-alpha blob (an extended Lyman-alpha source), and to study its polarisation properties with mock observations. Contrary to what was suggested before, I showed that polarisation is not a strong tracer of the origin of Lyman-alpha photons

The Center for Disease Control and Prevention (CDC) released heat illness data which highlighted that ~29 heat stress hospitalizations and ~3 heat-related deaths occurred every day during the summer months within the US from years 2000 to 2014. Heatstroke- the most severe form of heat illness which oftentimes lead to death- has been cited to be entirely preventable if a timely intervention is introduced. This dissertation uses microwave radiometric thermometry to perform wireless non-invasive internal body temperature monitoring which can enable intervention methods that help to prevent deaths associated with heat-illness. Overall, this dissertation develops a comprehensive closed-form analytical radiometric model and validates the effectiveness of the comprehensive model through a controlled life-like human body temperature sensing experiment. Wireless sub-skin temperature data is predicted from a human tissue mimicking phantom testbed to within 1%. A generic isolated radiometer system equation is derived for all possible calibration source combinations. The generic isolated radiometer system equation predicts comparable results to that of an ideal simulation. While improved isolation decreases measurement uncertainty, it does not improve the accuracy of estimated noise temperatures using a perfectly-isolated radiometer system equation assumption. A highly reproducible tissue-mimicking biological phantom (bio-phantom) recipe (comprised of urethane, graphite powder, and a solvent) was developed to accurately emulate the electrical properties of actual dry human skin versus frequency up to 18 GHz. The developed solid state skin phantom begins in pourable liquid form and then cures at room temperature into a dry solid state mold. An in-plane electromagnetic bandgap structure was developed and integrated within an on-body inward facing spiral antenna design. The inclusion of the in-plane electromagnetic bandgap structure demonstrated a +2.64dB gain improvement in the antenna broadside and -8dB in the rear gain while in-contact with the body as compared to the conventional spiral antenna. Likewise, the measured main beam efficiency is improved from 54.43% for the conventional antenna to 86.36% for the EBG antenna. Two techniques based on signal-flow graph theory were derived to explain both the non-coherent steady-state radiative transfer and the coherent radiative transfer within multi-layered dielectric media with non-uniform temperatures and any number of stratified layers. Both models allow for the accurate characterization and sensing of the thermal emissions originating from subsurface tissue layers.

Mes travaux de recherche portent sur l'étude du rayonnement (surtouts rayons X durs) provenant des trous noirs accrétant (dans les noyaux actifs de galaxies et les binaires X). L'objectif est d'en extraire des informations sur les conditions physique régnant dans l'environnement immédiat de ces objets. Les principales question auxquelles je tente de répondre sont les suivantes: Quelle est la structure et la géométrie de la matière accrétée au voisinage du trou noir ? Comment celle -ci évolue-t-elle avec le taux d'accrétion de masse ? Quel est la relation entre les processus d'accrétion et la formation de jets souvent observés dans ces systèmes ? Mon approche est fondée sur une comparaison précise entre les observations et les prédictions des divers modèles. Je présente les efforts poursuivis depuis près de dix ans afin de développer des outils de simulation numérique pour modéliser le transfert de rayonnement dans les plasma chauds des sources compactes X. Je montre comment ces outils ont été utilisés pour modéliser le continuum haute énergie et la variabilité des trous noirs accrétants et pour contraindre la structure du flot d'accrétion. Je présente également des résultats reposant sur l'analyse et l'interprétation d'observations menées avec des télescopes spatiaux tels que XMM-Newton et INTEGRAL ainsi que sur des d'observations simultanées à plusieurs longueurs d'ondes allant de la radio aux rayons X durs.

<p>A major fraction of stars will pass through a short period of dramatic events in their final evolutionary stage. Low- to intermediate-mass stars, studied here, are stripped of their outer parts in a slow massive wind. This mass loss reshapes both the star and the surrounding medium. The formation of the wind is a consequence of the non-linear interaction of a number of physical processes. Stellar pulsations and efficient dust formation are examples of such key processes. Time-dependent theoretical models, in combination with observations, are useful tools for understanding these winds.</p><p>The main object of this thesis has been the physical improvement of a theoretical wind model. Here the coupling between the dust and gas in the wind is studied in further detail, allowing drift. The methods that have been developed earlier to describe the micro-physical interaction are overviewed and summarized. Previously dust has often been assumed to move at the same velocity as gas. New time-dependent wind models are presented where grain drift has been treated self-consistently. Specifically, the coupling between dust and gas in the wind has been modeled more realistically, with descriptions of both the modified momentum and energy balances, and drift dependent dust formation. The results of these new ``drift models'' have been compared with the results of non-drift models. </p><p>A general result of the study is that the effects of drift are significant and difficult to predict if a simple analytical theory is used. It has been found that dust in drift models tends to accumulate in certain dense regions, an accumulation that was not possible without drift. Moreover the new models show an increased variability in the wind structure. The use of drift in dust formation tends to markedly increase the produced dust. Some sets of model parameters lead to a wind without including drift, but a corresponding wind does not form when drift is included -- and vice versa. The effects of drift are important and can probably not be ignored in realistic models.</p>