<b>Structure of the thesis</b> The present work has been organized in two main parts: in the first one, the focus will be on the scientific and theoretical aspects of the evaporation process in presence of an inert gas flow while in the second all the technical aspects and more practical tests related to the real implementation of the micro-gravity experiment CIMEX-1 will be detailed. In any cases, the discussion will always start from the phenomenology observed considering that ” Nature is far more reach of any speculations.” <b>Part I: Evaporation in presence of inert gas</b> In chapter 1, a detailed presentation of the experimental setups for the on-ground tests is given together with the presentation of the qualitative and quantitative results obtained. Actually, the main parameters that regulate such kind of experiments are the mass flow rate of inert gas, the total pressure of the cell and the geometrical shape and dimensions of the evaporating regions. Consequently, the experiments aimed at covering the maximal possible combination of these three parameters with special attention to the variation of the inert gas flow and of the thickness of the evaporating liquid layer. More precisely, the liquid layer thickness was in the range 1.2 to 3.8 mm while the inert gas flow was set in the range 50 to 2500 ml/min. The pressure has been partially neglected as control parameter because its control was discovered not to be very reliable. The visualization system used in all the experiments consisted in a opportunely calibrated infrared camera. It allowed for having a quantitative analysis of the temperature distribution at the interface of the evaporating liquid. The infrared images also helped to follow the thermal history of the interface. In many cases, it has been possible to clearly observe the evolution of instability patterns at the interface that represent an original contribution to the understanding of such a kind of phenomena. The physical and mathematical modeling of the observed phenomenology will be the subject of the chapter 2. One of the peculiar issue of the problem under consideration is that the thermal gradient normal to the interface is not directly imposed like in the usual Marangoni-Bénard experience, but is a result of the cooling of the interface due to the evaporation. Moreover,the interface is subject to the shear stress of the inert gas flow and to the one due to the thermo-capillarity. Finally, the gas phase is to be considered as a mixture; this oblige to solve a diffusion problem in the gas phase. A physical model that takes into account the different aspects mentioned above is presented together with the governing equations and the appropriate boundary conditions. Numerical issues involved in solving the model are also analyzed. Numerical results obtained are finally discussed and compared when possible with experimental results. <b>Part II: Preparation of the CIMEX-1 experiment on-board the International Space Station.</b> In chapter 3, we will describe the main platforms used to perform low-gravity experiments. They will be classified according to the low-gravity level and to the low-gravity interval duration that could be ensured for experiments. According to these criteria, the list of the low-gravity platforms will be as follows: Drop Towers with 4 sec. of micro-gravity, Parabolic Flights that can assure not more than 25 sec., Sounding Rockets with a low-gravity time of the order of several minutes depending on the rockets, Foton Capsules that assure for many days of high quality - i.e. without perturbations - low-gravity level and , last but not least, the International Space Station where the low-gravity duration could be even of several weeks which is a sufficient time duration for the most part of the experiments. The chapter 4 will be entirely devoted to the ITEL experiment that is the precursor and the core of the CIMEX-1. After a brief overview of the experiment that has been performed twice on-board sounding rockets of the MASER class, the experimental setups used both on-ground and in micro-gravity will be detailed. The focus will be on the experimental results obtained on-ground during the preparatory tests and during the two sounding rocket flights with special attention to the first one. The analysis will be supported by the presentation of many results obtained in numerical simulations. The two parabolic flight campaigns performed to test one of the key sub-systems of the CIMEX-1 setup are the subject of the chapter 5. The separating-condensing unit is mandatory for performing the experiment on-board the International Space Station because the limitations on the crew intervention oblige to have a closed loop experiment. The goal of the two parabolic flights will be detailed together with the setup and the experimental scenario. The main results will be also shown and some considerations on the efficiency of the system will be presented. It is worthy to stress that the results obtained during these parabolic flights have been determinant at the European Space Agency level to fly the CIMEX-1 experiment on-board the International Space Station. Finally, in the section conclusions and perspectives the main results obtained will be summarized together with the new scenarios opened by the present work and some guidelines for further development in the experimental, theoretical and technical analysis.

The steady thermocapillary motion in shallow enclosures is studied. Two different configurations, imposed heat flux and differentially heated side walls, are considered. A numerical simulation of the problem in the imposed heat flux case is made. The Pressure Correction Method is used to treat the pressure velocity coupling, in particular, the SIMPLER approximation. The discretization is made using central differences along with an appropriate non-uniform grid. The computed results show the streamlines and temperature field in different Marangoni number, Prandtl number, Capillary number and aspect ratio

Abstract The steady thermocapillary motion in shallow enclosures is studied. Two different configurations, imposed heat flux and differentially heated side walls, are considered. A numerical simulation of the problem in the imposed heat flux case is made. The Pressure Correction Method is used to treat the pressure velocity coupling, in particular, the SIMPLER approximation. The discretization is made using central differences along with an appropriate non-uniform grid.

A series of steady-state water condensation experiments were conducted to determine the effects of Marangoni convection on the condensation flux. The interface was flat so that the results of the interfacial temperature discontinuities could be compared to past condensation experiments conducted under similar experimental conditions using a spherical interface. Two experimental methods were used. Method 1 was to vary the temperature in the cooling pipes (Tcp ) with the position of the interface relative to the cooling pipes fixed. Method 2 was to vary the position of the interface while Tcp was held constant. The interfacial temperature discontinuities in this study were approximately 2.3-9 times smaller in magnitude, than those measured using a spherical liquid-vapour interface. The experimental results showed that the condensation flux increased as thermocapillary convection increased (increase in interfacial temperature gradients and speed). This increase resulted in a 1.37-12.5 times enhancement in the condensation flux of pure water.

A series of steady-state water condensation experiments were conducted to determine the effects of Marangoni convection on the condensation flux. The interface was flat so that the results of the interfacial temperature discontinuities could be compared to past condensation experiments conducted under similar experimental conditions using a spherical interface. Two experimental methods were used. Method 1 was to vary the temperature in the cooling pipes (Tcp ) with the position of the interface relative to the cooling pipes fixed. Method 2 was to vary the position of the interface while Tcp was held constant. The interfacial temperature discontinuities in this study were approximately 2.3-9 times smaller in magnitude, than those measured using a spherical liquid-vapour interface. The experimental results showed that the condensation flux increased as thermocapillary convection increased (increase in interfacial temperature gradients and speed). This increase resulted in a 1.37-12.5 times enhancement in the condensation flux of pure water.

Zugl.: Hamburg, Techn. Univ., Diss., 2007

Placée sous le signe de la fragmentation liquide, cette thèse met l'emphase sur une série d'étapes pouvant, d'un film suspendu, mener à une assemblée de gouttes : déstabilisation, rupture puis fragmentation. Elle prend appui sur des expériences originales, analysées à l'aide de modèles à la portée générale. En guise de prologue, capillarité, cisaillement et viscosité sont discutées à travers l'étude de la déstabilisation Marangoni d'un filament d'huile visqueuse déposé à la surface de l'eau. La mise en mouvement du filament, pilotée par une différence de tensions de surface, produit une cascade critique auto-similaire, accélérée jusqu'à la dilution complète de l'huile, miscible à l'eau. Sur un film, l'effet Marangoni créé par un déficit localisé de tension de surface, dont le support (la température ou un soluté) diffuse, engendre sa déstabilisation, étudiée au deuxième chapitre. L'analyse linéaire exhibe une échelle de temps inertielle basée sur le cisaillement surfacique, sur laquelle s'établit un écoulement interstitiel, caractérisé expérimentalement. Le scénario introduit un nombre de Péclet, critère pour la rupture spontanée des films épais. Les conséquences sur la stabilité finale du film sont explorées au troisième chapitre, avec la revue des possibles régularisations. La dynamique de rupture est interprétée en termes de choc. Enfin, suivant la rupture d'un film en divers points, la collision de cylindres liquides est isolée en tant que mécanisme individuel de fragmentation. L'analyse détaillée de l'impact révèle la transition entre deux régimes, vers des gouttes de plus en plus fines / This thesis deals primarily with liquid films fragmentation. It consists, thanks to quantitative and original experiments, in the transformation of a free suspended film into a collection of droplets: destabilization, rupture, and fragmentation.In a prologue, notions of capillarity, momentum transfer and viscosity are introduced by the Marangoni-driven destabilization of a thin thread of viscous oil on water. The difference of surface tension feeds an accelerated, critical, self-similar cascade which ends up with the complete dilution of oil into water. When applied to a free film, the Marangoni effect driven by a localized deficit of surface tension, whose carrier (temperature or a solute) is diffusing into the liquid, destabilizes it (chapter two). The linear instability analysis points out the crucial role of an inertial timescale based on the surface shear stress. Experiments validate the prediction of an interstitial flow, which digs out the film in the case of a surface tension deficit. These observations offer new insights into a long-standing problem, namely the spontaneous nucleation of holes on planar liquid films.The third chapter investigates the consequences of the linear instability, as far as the film final stability is concerned, reviewing possible regularizations. The rupture dynamics is described within the framework of shocks. In the final chapter, the collision of liquid cylinders is identified as an individual mechanism for liquid fragmentation. The impact dynamics is analyzed, leading to two principal fragmentation regimes. The onset to the splashing one, which produces fine and fast droplets, is described

Effects of external vibrations (called g-jitter) on Marangoni convection in a liquid bridge were investigated on the International Space Station (ISS) and in ground-based experiments. In ISS, most dominant g-jitter frequency was noted to be ~110 Hz. ISS experiments suggested that the surface vibrations were mainly affected by the aspect ratio (length/diameter ratio), but not the imposed temperature gradient. Liquid bridge surface vibrations agreed well with Ichikawa et al.’s model. Ground-based experiments confirmed that increasing the volume ratio would cause the resonance frequency to increase. When a temperature difference was imposed between the upper and lower disks, for constant aspect and volume ratios, the resonance frequency tended to increase with the decreasing temperature difference. Furthermore, the shift in the resonance frequency due to a temperature difference, was found to be due to Marangoni convection and not due to reduced viscosity or surface tension of the fluid.

Effects of external vibrations (called g-jitter) on Marangoni convection in a liquid bridge were investigated on the International Space Station (ISS) and in ground-based experiments. In ISS, most dominant g-jitter frequency was noted to be ~110 Hz. ISS experiments suggested that the surface vibrations were mainly affected by the aspect ratio (length/diameter ratio), but not the imposed temperature gradient. Liquid bridge surface vibrations agreed well with Ichikawa et al.’s model. Ground-based experiments confirmed that increasing the volume ratio would cause the resonance frequency to increase. When a temperature difference was imposed between the upper and lower disks, for constant aspect and volume ratios, the resonance frequency tended to increase with the decreasing temperature difference. Furthermore, the shift in the resonance frequency due to a temperature difference, was found to be due to Marangoni convection and not due to reduced viscosity or surface tension of the fluid.

In this study, shapes of the molten region and transport processes affected by thermocapillary convection in melting or welding pool irradiated by a low-power-density beam are determined from a scale analysis. A low-power-density-beam heating implies no deep and narrow cavity or keyhole taking place in the pool. In this work, the complicated flow pattern in the pool is influenced by an unknown shape of solid-liquid interface, and interactions between the free surface layer, corner regions, and boundary layer with phase transition on the solid-liquid interface. Since Prandtl number is much less than unity while Marangoni and Reynolds number can be more than in melting metals, an appropriate scaling mass, momentum, and energy transport subject to a force balance between viscous stress and surface tension gradient on the free surface account for distinct thermal and viscous boundary layers in these regions of different length, velocity, and temperature scales. The results find that shapes of the fusion zone, free surface velocity and temperature profiles are determined by Marangoni, 104 i Prandtl, beam power, Peclet, and Biot numbers, and solid-to-liquid thermal conductivity ratio. The predications agree with numerical computations.