Droplet evaporation and condensation are two important topics of interest, since these two phase-change phenomena not only occur in the cycle of global water, e.g., the formation of rain, fog, dew, and snow in nature, but also play a critical role in a variety of applications including phase-change heat transfer enhancement, surface chemistry and energy system optimization. Especially, in the past two decades, the rapid development of the nature-inspired non-wetting surfaces has promoted the applications of droplet-based phase change phenomena in various scenarios. However, most previous studies focused on the sessile droplets on one flat surface in the open space, and the effects of surface topography, i.e., surface curvature or configurations, on droplet evaporation and dropwise condensation are still elusive. This dissertation aims to explore droplet-based evaporation and condensation in more complex spaces and to elucidate how the surface topography affects the evaporating or coalescing droplet dynamics during these phase-change processes. The coalescence-induced jumping of nanodroplet on curved superhydrophobic surface is modeled via molecular dynamic simulations. As the surface curvature increases from 0 to 2, the corresponding energy conversion efficiency of jumping droplet during the coalescence process could be significantly improved about 20 times. To explain this curvature-enhanced jumping effect, the contact line dissipation, i.e., an important source of energy dissipation in nanoscale, is considered in our scaling energy analysis. And this energy-effective jumping of coalesced droplet could be mainly attributed to the reduction of contact line dissipation due to the decrease of contact line length and contact time on curved surface. As the droplets are confined between two parallel or non-parallel low-energy surfaces, i.e., hydrophobic or superhydrophobic surfaces, with a narrow gap, the total evaporation time of the squeezed droplets would be dramatically prolonged about two times. An ellipsoidal segment diffusion-driven model is established to successfully predict the evolution of contact radius and volume of the squeezed droplets during the evaporation process and to clarify it is the vapor enrichment inside the confined space giving rise to the mitigated evaporation. If two hydrophobic surfaces are configured as non-parallel, the confined droplet inside the V-shaped grooves would be self-transported towards the cusp/corner during the evaporation. Based on our energy and force analyses, the asymmetrically confined droplet would move towards an equilibrium location le, where the Laplace pressure induced force is balanced with normal adhesion force, to minimize its Gibbs surface energy. As le decreases during the evaporation, this equilibrium location would directionally shift towards the cusp, which could be regarded as the origin of this evaporation-triggered unidirectional motion. For the first time, the solvent transport and colloidal extraction could be accurately controlled in a combined manner. / Doctor of Philosophy / Droplet evaporation and condensation are two important topics of interest, since these two phase-change phenomena not only occur in the global cycle of water including the formation of rain, fog, dew, and snow in nature, but also play a critical role in a variety of applications including heat transfer enhancement, surface chemistry, and the energy system optimization. Generally, the droplets in these scenarios are deposited on one flat surface opened to the atmosphere. and the effects of surface topography on droplet evaporation and dropwise condensation are still elusive. This dissertation aims to explore droplet-based evaporation and condensation in more complex spaces and to clarify how the surface curvature or configurations affects evaporating or condensing droplet dynamics accompanying these phase change processes. As the coalesced droplet jumps off the curved superhydrophobic surfaces during dropwise condensation, the corresponding energy conversion efficiency would be significantly improved about 20 times due to the increases of curvature. It is demonstrated that the decrease of contact line length and contact time would give rise to the reduction of contact line dissipation, which should be the main factor driving this energy-effective jumping of the coalesced droplets. As the droplets are confined between two parallel or non-parallel low-energy surfaces, i.e., hydrophobic or superhydrophobic surfaces, with a narrow gap, the total evaporation time of the squeezed droplets would be dramatically prolonged about two times in the small space. An ellipsoidal segment diffusion-driven model is established to successfully predict the evolution of contact radius and volume of the squeezed droplets during the evaporation and to clarify it is the vapor enrichment in the confined space giving rise to the mitigated evaporation. If two hydrophobic surfaces are configured as non-parallel, the confined droplet inside the V-shaped grooves would be self-transported towards the cusp/corner of the structure during evaporation. Based on our energy and force analyses, the asymmetrically confined droplet would move towards an equilibrium location le, where the Laplace pressure induced force is balanced with normal adhesion force, to minimize its Gibbs surface energy. As le decreases in the scale of during the evaporation, this equilibrium location would directionally shift towards the cusp, which could be regarded as the origin of this evaporation-triggered unidirectional motion.

No / Competition experiments in which 1,2-phenylenediamine, C6H4(NH2)2, condenses with equimolar quantities of benzil, (C6H5CO)2, and a 3,3'- or 4,4'-disubstituted benzil (XC6H4CO)2 (X = F, Cl, Br, CH3 or CH3O) to form a mixture of 2,3-diphenylquinoxaline and the corresponding 2,3-diarylquinoxaline (Ar = XC6H4) in the microdroplets produced in a nebuliser allow a Hammett relationship with a ρ value of 1.85 to be developed for this accelerated condensation in the nebuliser. This structure reactivity relationship reveals that an appreciable amount of negative charge builds up on the carbon of the carbonyl group of the benzil during the rate-limiting step of the reaction, thus confirming that this process involves nucleophilic addition of the 1,2-phenylenediamine to the benzil. In general, the presence of an electron donating substituent, particularly in the 4 and 4' positions, in the benzil retards the reaction, whereas an electron attracting substituent, especially in the 3 and 3' position, accelerates it. / 2019 British Mass Spectrometry Summer Studentship for NWF.

Yes / This study focuses on numerical modeling of condensation of water vapor in a Laval nozzle, using the liquid drop nucleation theory. Influence of nozzle geometry, pressure, and temperature on the average drop size is reported. A computer program written in MATLAB was used used to calculate the nucleation and condensation of water vapor in the nozzle. The simulation results are validated with the available experimental data in the literature for steam condensation. The model reveals that the average drop size is reduced by increasing the divergent angle of the nozzle. The results also confirm that increasing the inlet pressure has a direct effect on the average drop size while temperature rise has an inverse effect on the drop size.

No / Nucleation and condensation of mercury vapor has been investigated in various divergent angle and operating condition. Divergent angle has a great effect on droplet size at the end of nozzle. Influence of operating condition such as pressure and temperature on the size of droplet has been investigated. A one-dimensional mathematical model based on classical nucleation and growth has been developed to calculate the nucleation and condensation of mercury vapor. A mercury vapour turbine has been used in conjunction with a steam turbine for generating electricity. The mercury cycle offers an efficiency increase compared to a steam-only cycle because energy can be injected into the Rankine Cycle at higher temperature. The target of modeling is predicting the droplet size of mercury nano-particles during rapid expansion. The results are verified by accurate experimental data available in the literature. The governing equations were solved using Runge-Kutta third-order numerical method in MATLAB software.

Cette thèse de doctorat a pour cadre l'étude des mécanismes physiques qui régissent la propagation d'une fissure dans les verres d'oxydes et questionne notamment l'existence et la portée de mécanismes dissipatifs aux petites échelles. Pour ce faire, la propagation sous critique d'une fissure est pilotée par un chargement en géométrie Double Cleavage Drilled Compression sous environnement contrôlé. Elle fait alors l'objet d'analyses expérimentales in-situ et postmortem sur plus de six décades d'échelles de longueur (du nm au mm) par techniques optiques et microscopie à force atomique (AFM). Une analyse 2D/3D de l'échantillon est réalisée en mécanique linéaire élastique de la rupture pour pouvoir assurer le contrôle à toutes les échelles de l'essai mécanique et exploiter les résultats. L'effet mécanique du condensat capillaire observé par AFM en pointe de fissure est modélisé sur l'exemple du modèle de zone cohésive. Ceci permet d'évaluer la pression de Laplace négative du liquide confiné et d'expliquer le mécanisme de refermeture des fissures. Une technique de corrélation d'image (DIC) est utilisée sur des séries d'images AFM in-situ. Nous montrons que la solution élastique pour le champ de déplacement de surface est valable jusqu'à une distance de 10 nm de la pointe de la fissure. Une étude expérimentale prometteuse de fractoémission a permit l'accès à la taille de la zone d'endommagement nanométrique dans les verres fracturé en régime dynamique. Les fonctions de corrélations de hauteur le long d'images AFM de surfaces de rupture lente ont été analysées. Nous montrons que la longueur de coupure de l'ordre de quelques dizaines de nm, interprétée comme taille de zone d'endommagement, découle plus probablement de la taille finie de la sonde de balayage de l'AFM et qu'en accord avec la DIC, aucune zone d'endommagement de taille supérieure à 20 nm n'est observable. / The aim of this thesis is to study the physical mechanisms which govern crack propagation in oxide glasses and to investigate in particular the existence of dissipative mechanisms at small scales. The subcritical crack propagation is controlled by a loading cell on Double Cleavage Drilled Compression samples under controlled atmosphere. Postmortem and in-situ analysis are performed on more than six decades of length scales (from nm to mm) by optical techniques and atomic force microscopy (AFM). An 2D/3D analysis of this sample is realized according to linear elastic fracture mechanics in order to discuss the experimental results and to ensure the mechanical test control at all scales. The mechanical effect of capillary condensation observed by AFM at the crack tip is modelled according to a cohesive zone model. This allows evaluating the negative Laplace pressure in the liquid and explaining the crack closure mechanism in glass. A digital image correlation technique is used on series of consecutive AFM in-situ images. We show that the elastic solution for the surface displacement field is valid up to a distance of 10 nm from the crack tip. A promising experimental study of fractoemission allowed us to access the nanometric process zone size in glasses during dynamic fracture. The height correlation functions along the AFM images of fracture surfaces were analyzed. We show that the cutoff length, found close to few ten nm and preiously interpreted as the process zone size, is most probably due to the finite size of the AFM scanning probe and in agreement with the DIC, no process zone larger than 20 nm is observable.

Reflux condensation occurs in a vertical tube when there is an upward core flow of vapour (or gas-vapour mixture) and a downward flow of the liquid film. The understanding of this condensation configuration is crucial in the design of reflux condensers and in loss-of-coolant safety analyses in nuclear power plant steam generators. A range of modelling approaches exists for co-current film condensation from gas-vapour mixtures in parallel-plate channels and tubes. These methods are based on marching from the inlet down the tube and do not apply to the reflux condensation. In this research, however, a two-dimensional two-phase model was developed that solves the steady, full elliptic governing equations in both the film and the gas-vapour core flow on a non-orthogonal mesh that dynamically adapts to the phase interface. Gas-vapour shear and heat and mass transfer at the interface were accounted for fundamentally. This modelling is a big step ahead of current capabilities by removing the limitations of previous reflux condensation models which inherently cannot account for the detailed local balances of shear, mass, and heat transfer at the phase interface. The model was developed and applied for co-current and counter-current flows in vertical parallel-plate channels, followed by vertical tubes. In each stage, the model results were compared against the available experimental and numerical data for validation purposes. A wide range of boundary conditions and geometries have been studied to examine the details of co-current and counter-current condensation phenomena. Velocity, temperature, pressure, and gas mass fraction profiles along with the axial variation of various parameters such as local Nusselt number, film thickness, interface and centre-line temperature and gas mass fraction are presented in parametric studies. / February 2017

The purpose of this study was to quantitatively compare the fill density of MTA produced by hand condensation and hand condensation with indirect ultrasonic activation. Thirty acrylic block with 30 degree curved canals (group C) and 30 with straight canals (group S) were instrumented to a final apical size of 45 of 0.06 taper crown-down technique. After irrigating with water and drying with paper points, each block was weighed to the nearest 0.0001g with a digital electronic balance. In half of the specimens, chosen at random, the canal first filled with MTA using the hand condensation method (H) then weighed. The MTA was removed. The canal was rinsed, dried, and refilled using hand condensation with indirect ultrasonic activation (US). In the other half of the specimens, the procedure was carried out identically but in reverse order. The blocks were weighed again after cleaning the MTA from the canal as well as after refilling the canal using indirect ultrasonic condensation. Data comparing the weight of MTA between the two placement methods and the two canal configurations were analyzed by a two-way ANOVA. There was a statistically significant increase in weight of MTA produced by ultrasonic activation than by hand condensation in groups C (p<.0001) and S (p<.0001). However, there was no statistically significant difference when comparing the straight canal versus the 30-degree curved canal (p = .08). In group C, ultrasonic condensation resulted in a 10.07% increase in the weight of MTA over hand condensation alone. Similarly in group S, there was a 9.1% increase in the weight of MTA over hand condensation. In conclusion, hand condensation with indirect ultrasonic activation resulted in an MTA fill that was denser than that accomplished by hand condensation alone.

Nous avons analysé par cryomicroscopie électronique la morphologie et la structure de longues chaines d’ADN condensées par un polycation tétravalent, la spermine (polyamine). Les expériences ont été réalisées i) avec des solutions de chaînes diluées et ii) avec des chaines isolées confinées dans la capside d’un virus.Les expériences ont été réalisées avec de l’ADN Lambda (48kbp) en solution diluée (0.03 mM Ph) et à faible concentration ionique (10 mM Tris HCl, 1 mM EDTA, pH 7.6). Nous avons exploré une large gamme de concentrations en spermine, allant du seuil de précipitation (0.05 mM sp) jusqu’à la limite de re-solubilization et au-delà (400 mM sp). Seize minutes après mélange de l’ADN et de la spermine, les échantillons sont piégés en film mince et vitrifiés à basse température pour garder intactes les conditions ioniques, puis imagés à basse température sous faibles doses d’électrons (cryoMET). La plupart des chaînes d’ADN forment des agrégats de tores de structure hexagonale avec des interdistances entre hélices de 2.93, 2.88, et 2.95 nm pour des concentrations en spermine respectivement égales à 0.05, 1 et 100 mM spermine, ce qui est en bon accord avec les données collectées précédemment par diffraction des rayons X. A concentration plus élevée en spermine (200mM), les tores hexagonaux sont remplacés par des faisceaux cholestériques de structure plus lâche (3.32 nm entre hélices). Nous en déduisons que la forme comme la structure des condensats cristallins liquides ADN-sp sont liées aux interdistances entre hélices et déterminés par les conditions ioniques i.e. par l’énergie cohésive entre chaînes d’ADN. En dehors du domaine de précipitation (400mM sp), les molécules d’ADN forment un réseau soluble de fines fibres (4-6nm de diamètre) qui nous amènent à reconsidérer l’état de ces chaiînes en présence de spermine. Nous avons également conçu des expériences pour visualiser les agrégats formés 6 à 60 sec après addition de la spermine dans les mêmes conditions de tampon. Parmi les nombreuses formes originales que nous avons observées (absentes après 16 min), la présence de fibres étirées ou en hélice, visibles seulement après 9sec, nous conduit à proposer que les chaines d’ADN soient immédiatement étirées après addition de spermine puis relaxent sous forme de fibres hélicoïdales qui donnent naissance à de petits toroids (comprenant quelquefois moins d’une chaine) qui grandissent et fusionnent. Nous avons également analysé les dimensions de l’ensemble des tores observés et montré l’existence de contraintes géométriques qui restent à élucider. Puisqu’il était généralement impossible de prévenir l’agrégation des chaines d’ADN, nous avons choisi une autre approche pour analyser le collapse de chaines d’ADN individuelles. Nous avons utilisé une population de virus T5 contenant une fraction de leur génome initial (12-54 kbp). La molécule d’ADN, initialement confinée dans le petit volume de la capside (de de 80nm diamètre) est collapsée par addition de spermine. Par comparaison avec le premier jeu de données, nous avons travaillé à concentration plus élevée en ADN (0.45 mM Phosphates dans l’ensemble de l’échantillon) et la concentration en spermine a été ajustée entre 0.05 et 0.5 mM (ce qui correspond à des rapports de charges +/- bien inférieurs). Ces expériences ont donc été réalisées au voisinage de la ligne de précipitation, dans la « région de coexistence », entre le domaine où les chaines sont en condition de pelote et le domaine ou les chaines sont toutes collapsées sous forme de tores. Nous avons montré l’existence de formes intermédiaires entre ces deux états que nous appelons « tores chevelus » dans lesquels une partie de la molécule est condensées dans le tore alors que l’autre partie reste non condensée. Les distances entre hélices ont également été mesurées. Elles sont plus grandes dans ces structures intermédiaires que dans les tores formés à plus forte concentration en spermine. Ces deux séries d’expériences montrent l’intérêt des méthodes de cryo-microscopie pour étudier la structure locale des phases condensées de l’ADN. Nous avons montré comment le confinement modifie le comportement de l’ADN en solution et l’intérêt d’étudier ces effets compte tenu de son importance dans le contexte biologique. / By using cryo-electron microscopy, we analyzed the morphology and structure of long double-stranded DNA chains condensed upon addition of varying amounts of the tetravalent polycation spermine (polyamine). Experiments have been performed i) with chains diluted in the bulk and ii) with individual chains confined in a virus capsid.Bulk experiments have been done with lambda DNA (48.5 kbp) at low concentration (0.03 mM Ph) and in low salt conditions (10 mM Tris HCl, 1 mM EDTA, pH 7.6). We explored a wide range of spermine concentration, from the onset of precipitation (0.05 mM sp) up to above the resolubilization limit (400 mM sp). Sixteen min after mixing spermine and DNA, samples have been trapped in thin films and vitrified in liquid ethane to keep ionic conditions unchanged, and imaged at low temperature with low doses of electrons (cryoTEM). DNA chains mostly form large aggregates of toroids in which DNA chains are hexagonally packed with interhelical spacings of 2.93, 2.88, and 2.95 nm at 0.05, 1 and 100 mM spermine, respectively, in agreement with previous X-ray data. At higher spermine concentration (200 mM), hexagonal toroids are replaced by cholesteric bundles with a larger interhelical spacing (3.32 nm). We conclude that the shape and the structure of the liquid crystalline sp-DNA condensates are linked to the DNA interhelix spacing and determined by the ionic conditions i.e. by the cohesive energy between DNA strands. Outside of the precipitation domain (400 mM spermine), DNA chains form a soluble network of thin fibers (4-6 nm in diameter) that let us reconsider the state of these DNA chains in excess of spermine. We also designed experiments to visualize condensates formed 6-60 sec after mixing Lambda DNA with 0.05 mM spermine, under identical buffer conditions. Among multiple original shapes (not found after 16 min), the presence of stretched and helical elongated fibers seen only 9sec after addition of spermine let us propose that DNA chains are immediately stretched upon addition of spermine, relax into helical structures and finally form small toroids (containing in some cases less than one Lambda chain) that further grow and aggregate. We also analyzed the dimensions and structural details of the complete collection of toroids, and reveal the existence of geometric constraints that remain to be clarified. Since it was only exceptionally possible to prevent the aggregation of DNA in dilute solution, we used another approach to observe the collapse of single DNA chains. We handled a population of T5 viruses containing a fraction of their initial genome (12-54 kbp long). The Na-DNA chain, initially confined in the small volume of the capsid (80nm in diameter) is collapsed by the addition of spermine. Compared to the first set of experiments, we explored a higher DNA concentration range (0.45 mM Phosphates in the whole sample) and the spermine concentration was varied from 0.05 to 0.5 mM (which corresponds to much lower +/- charge ratios). Experiments are thus performed close to the precipitation line, in the coexistence region, between the region where all chains are in a coil conformation, and the region where all chains are collapsed into toroids. We describe the existence of intermediate states between the coil and the toroidal globule that were not reported yet. In these “hairy toroids”, part of the DNA chain is condensed in the toroid and the other part stays uncondensed outside of it. The interhelical spacing was also measured; it is larger in these partly-condensed toroids than in the fully organized toroids formed at higher spermine concentration.These two series of experiments show the interest of cryoEM to analyze the structural polymorphism and local structure of spermine-DNA aggregates. We also demonstrated how the confinement interferes with DNA condensation and the interest to investigate such effects that are important in the biological context.

Ces travaux ont pour contexte de participer à l’optimisation les cycles de décontamination des salles propres en utilisant un nouveau système de décontamination. Le principe de ce système est de décontaminer un volume par condensation de vapeur de peroxyde d’hydrogène sur toutes les parois intérieures du volume. Différents modèles de condensation et d’évaporation pariétaux ont été étudiés, développés et implémentés dans un code CFD, en incluant des modèles simples ainsi que des modèles plus complexes prenant en compte la forme de la phase liquide pariétale. En parallèle, afin de valider les modèles, un dispositif expérimental d’isolateur pour la décontamination a été utilisé afin de créer une base de données expérimentale de condensation de vapeur. Enfin, la validation de ce modèle numérique a été réalisée en comparant les résultats numériques fournis par le modèle développé dans une reproduction numérique de l’isolateur expérimental avec les résultats des expérimentations. / The context of this work is to contribute to the optimization of clean room decontamination cycles by using a new decontamination system. The concept of this system is to decontaminate a volume by condensing hydrogen peroxide vapour on all the inner walls of the volume. Different models of wall condensation and evaporation have been studied, developed and implemented in a CFD code, including simple models as well as more complex models taking into account the shape of the liquid phase at the wall. In parallel, in order to validate the models, an experimental isolator device for decontamination was used to create an experimental steam condensation database. Finally, the validation of this numerical model was carried out by comparing the numerical results provided by the model developed in a digital reproduction of the experimental isolator with the results of the experiments.

La salissure des modules photovoltaïques (PV) dégrade grandement leurs performances dans les environnements désertiques. Les études précédentes en extérieur ont tendance à trouver de faibles corrélations entre les taux de salissure et les paramètres météorologiques. On pensait que l'une des raisons était le long intervalle de mesure - jours ou semaines - des techniques traditionnelles de mesure des salissures sur le terrain. Dans la présente étude, un «microscope de souillure extérieur» (OSM) a été développé pour mesurer le dépôt et le détachement de particules de poussière individuelles, toutes les 10 minutes, dans des conditions extérieures, de jour comme de nuit. En utilisant une paire d'OSM graissés et non graissés, il était en outre possible de séparer les salissures en trois vitesses de flux de poussière de composants - dépôt, rebondissement immédiat et remise en suspension retardée. Les OSM ont été utilisés pour mesurer les taux de flux dans des expériences sur le terrain à Doha, au Qatar. La nouvelle méthode a révélé des effets explicatifs de paramètres environnementaux qui avaient auparavant été obscurcis par de longs intervalles de mesure des salissures et des taux de flux de poussière confondus. L'OSM pouvait également mesurer l'apparition et la croissance de gouttelettes de condensation microscopiques dans des conditions de terrain et de laboratoire. De telles expériences, ainsi que des mesures isothermes et des analyses de composition, ont démontré que la condensation sur les surfaces sales au terrain d’études était fortement influencée par la présence de matière hygroscopique dans la poussière autre que NaCl. En raison de cette matière, la condensation microscopique peut persister à la surface même si elle est bien supérieure à la température du point de rosée. Les résultats de l'étude suggèrent que la souillure des modules photovoltaïques pourrait être atténuée en tirant parti des variations naturelles des conditions météorologiques au cours de la journée. / Soiling of photovoltaic (PV) modules greatly degrades their performance in desert environments. Previous field studies have tended to find weak correlations between the soiling rate and weather parameters. It was thought that one reason was the long measurement interval — days or weeks — of conventional field soiling measurement techniques. In the present study, an “outdoor soiling microscope” (OSM) was developed able to measure deposition and detachment of individual dust particles, every 10 minutes, in outdoor conditions, day and night. By using a greased and ungreased pair of OSMs, it was further possible to separate soiling into three component dust flux rates — deposition, immediate rebound, and delayed resuspension. OSMs were used to measure flux rates in field experiments in Doha, Qatar. The novel method revealed explanatory effects of environmental parameters that had previously been obscured by limits of conventional long soiling measurement intervals and confounded dust flux rates. The OSM could also measure the onset and growth of microscopic condensation droplets in field and laboratory settings. Such experiments, along with isotherm measurements and composition analysis, demonstrated that condensation on soiled surfaces at the test site was strongly influenced by the presence of hygroscopic matter in the dust other than NaCl. Because of such matter, microscopic condensation could persist on both hydrophilic and hydrophobic surfaces well above the dew-point temperature. Results of the study suggest that soiling of PV modules might be mitigated by taking advantage of natural time-of-day variations in weather conditions.