Cette thèse est consacrée à l'étude de systèmes subissant des coagulations et fragmentations successives. Dans le cas déterministe, on travaille avec des solutions mesures de l'équation de coagulation - multifragmentation. On étudie aussi la contrepartie stochastique de ces systèmes : les processus de coalescence - multifragmentation qui sont des processus de Markov à sauts. Dans un premier temps, on étudie le phénomène de coagulation seul. D'un côté, l'équation de Smoluchowski est une équation intégro-différentielle déterministe. D'un autre côté, on considère le processus stochastique connu sous le nom de Marcus-Lushnikov qui peut être regardé comme une approximation de la solution de l'équation de Smoluchowski. Nous étudions la vitesse de convergence par rapport à la distance de type Wassertein $d_{lambda}$ entre les mesures lorsque le nombre de particules tend vers l'infini. Notre étude est basée sur l'homogénéité du noyau de coagulation $K$.On complémente les calculs pour obtenir un résultat qui peut être interprété comme une généralisation de la Loi des Grands Nombres. Des conditions générales et suffisantes sur des mesures discrètes et continues $mu_0$ sont données pour qu'une suite de mesures $mu_0^n$ à support compact existe. On a donc trouvé un taux de convergence satisfaisant du processus Marcus-Lushnikov vers la solution de l'équation de Smoluchowski par rapport à la distance de type Wassertein $d_{lambda}$ égale à $1/sqrt{n}$.Dans un deuxième temps on présente les résultats des simulations ayant pour objectif de vérifier numériquement le taux de convergence déduit précédemment pour les noyaux de coagulation qui y sont étudiés. Finalement, on considère un modèle prenant en compte aussi un phénomène de fragmentation où un nombre infini de fragments à chaque dislocation est permis. Dans la première partie on considère le cas déterministe, dans la deuxième partie on étudie un processus stochastique qui peut être interprété comme la version macroscopique de ce modèle. D'abord, on considère l'équation intégro-partielle différentielle de coagulation - multifragmentation qui décrit l'évolution en temps de la concentration $mu_t(x)$ de particules de masse $x>0$. Le noyau de coagulation $K$ est supposé satisfaire une propriété de $lambda$-homogénéité pour $lambdain(0,1]$, le noyau de fragmentation $F$ est supposé borné et la mesure $beta$ sur l'ensemble de ratios est conservative. Lorsque le moment d'ordre $lambda$ de la condition initial $mu_0$ est fini, on est capable de montrer existence et unicité d'une solution mesure de l'équation de coagulation - multifragmentation. Ensuite, on considère la version stochastique de cette équation, le processus de coalescence - fragmentation est un processus de Markov càdlàg avec espace d'états l'ensemble de suites ordonnées et est défini par un générateur infinitésimal donné. On a utilisé une représentation Poissonienne de ce processus et la distance $delta_{lambda}$ entre deux processus. Grâce à cette méthode on est capable de construire une version finie de ce processus et de coupler deux processus démarrant d'états initiaux différents. Lorsque l'état initial possède un moment d'ordre $lambda$ fini, on prouve existence et unicité de ces processus comme la limite de suites de processus finis. Tout comme dans le cas déterministe, le noyau de coagulation $K$ est supposé satisfaire une propriété d'homogénéité. Les hypothèses concernant la mesure $beta$ sont exactement les mêmes. D'un autre côté, le noyau de fragmentation $F$ est supposé borné sur tout compact dans $(0,infty)$. Ce résultat est meilleur que celui du cas déterministe, cette amélioration est due à la propriété intrinsèque de masse totale non-explosive que possède un système avec un moment fini d'ordre $lambda$

<p>Cilj istraživanja doktorske disertacije je bio ispitivanje mogućnosti primene<br />vlakana od nerđajućeg čelika za koalescenciju kapi mineralnih ulja različite polarnosti<br />pri različitoj permeabilnosti sloja. Pored toga cilj je bio i komparativna analiza efikasnosti separacije ispitivanih ulja vlaknima visoke i niske energije povr&scaron;ine. Postavljena je i modifikovana metoda kinetike upijanja tečnosti za ocenu sklonosti ulja ka odabranim vlaknima od kojih se formira sloj. Po prvi put je primenjena i metoda ve&scaron;tačkih neuronskih mreža za simulaciju koalescentne filtracije.<br />Eksperimentalni program koalescentne filtracije je realizovan na model emulziji<br />dispergovanog ulja u vodi konstantne ulazne koncentracije 500 mg/l,<br />pri konstantnoj temperaturi od 20&deg;C, horizontalnoj orijentaciji toka fluida i<br />prosečnoj veličini kapi ulja od 10 &mu;m. Uticaj geometrije sloja ispitivan je variranjem<br />permeabilnosti u opsegu od 0,1810^-9 m2 do 5,510^-9 m² za konstantnu<br />debljinu sloja od 5 cm. Testirana su tri uzorka mineralnog ulja velikog raspona<br />viskoznosti i polarnosti. Svi ogledi su realizovani pri konstantnoj brzini fluida u intervalu od 10 m/h do 80 m/h. Izlazna koncentracija dispergovane faze<br />odredivana je FTIR spektrofotometrijom.<br />Značajan doprinos u istraživanju fenomena koalescencije u sloju vlakana<br />ogleda se u utvrdivanju mogućnosti primene vlakana nerdajućeg čelika na<br />separaciju kapi mineralnih ulja, razumevanju simultanog uticaja energije<br />povr&scaron;ine, geometrije sloja i prirode uljne faze, kao i uvodenju nove veličine<br />izo-izlazne koncentracije. Test odredivanja kinetike upijanja tečnosti je pokazao<br />da odnos liofilno-liofobne kva&scaron;ljivosti materijala adekvatno opisuje sklonost<br />nekog filtarskog materijala ka separaciji ulja mineralnog porekla. Ovako<br />sveobuhvatna analiza uticaja različitih parametara na efikasnost koalescentne<br />filtracije, data u ovoj disertaciji, nije zabeležena u dostupnim literaturnim navodima.<br />Rezultati doktorske disertacije omogući će skraćivanje vremena i materijalnih<br />tro&scaron;kova neophodnih za projektovanje koalescera, jer smanjuju potreban<br />obim eksperimentalnih istraživanja potrebnih za uspe&scaron;no projektovanje.<br />Odabrani uslovi rada, prvenstveno prečnik vlakana materijala u odnosu na<br />ulaznu veličinu kapi, kao i način njihovog pakovanja, uz obezbedivanje poznate<br />i definisane permeabilnosti, obezbedili su željenu separaciju, čija se efikasnost<br />separacije kapi ispitivanih ulja vlaknima od nerdajućeg čelika kreće u<br />opsegu od 91% do 99% za opseg brzina fluida od 10 m/h do 50 m/h. Kod sloja<br />vlakana od nerdajućeg čelika efikasnija separacija je zapažena pri niskim vrednostima<br />permeabilnosti, dok je kod polimernog materijala obrnut slučaj. Zaključeno je da, u postojećim okolnostima, kod posmatranih materijala dominiraju<br />različiti mehanizmi koalescencije u sloju. Oblik zavisnosti izlazne koncentracije<br />od brzine i permeabilnosti za oba ispitivana materijala, nerdajući čelik i<br />polipropilen, je sličan. Polipropilen ima veći radni opseg brzina u odnosu na<br />nerdajući čelik. Medutim, nerdajući čelik uspe&scaron;nije separi&scaron;e ulje najniže viskoznosti<br />u oblasti niskih vrednosti permeabilnosti i pri niskim radnim brzinama<br />u odnosu na polipropilen. Zadovoljavajući kvalitet separacije se kod polipropilena<br />ostvaruje pri najvi&scaron;oj permeabilnosti i u &scaron;irokom opsegu brzina, dok se<br />kod nerđajućeg čelika najvi&scaron;e vrednosti kritične brzine za ceo opseg ispitivane<br />viskoznosti ostvaruju pri najnižoj permeabilnosti sloja.</p> / <p>The aim of the doctoral thesis was to investigate the possibility of the stainless<br />steel fibers application for the coalescence of mineral oil drops with different<br />polarity varying also the bed permeability. Additional aim was the comparative<br />analysis of the tested oil separation efficiency with fibers that have high<br />and low surface energy. The method for the evaluation of the fiber-liquid wetting<br />kinetic was set up and modified in order to assess the affinity of oil to the<br />selected fiber that was used to form the filtration bed. For the first time, the<br />method of artificial neural networks was applied to simulate the coalescence<br />filtration.<br />Experimental program of coalescence filtration was performed on oil-in-water<br />model emulsion with constant inlet concentration of 500 mg/l, at a constant<br />temperature of 20&deg;C, horizontal fluid flow orientation and average oil drop<br />size of 10 &mu;m. The influence of bed geometry was studied varying bed permeability<br />in the range from 0,1810^-9 m2 to 5,51010^-9 m2 for the constant layer thickness of 5 cm. Three samples of mineral oil that have large range of viscosity<br />and polarity were tested. All experiments were carried out at constant<br />fluid velocity in the range from 10 m/h to 80 m/h. The output dispersed phase<br />concentration was determined by FTIR spectrophotometry.<br />A significant contribution to the research of the coalescence phenomenon in<br />a fiber bed was achieved by determining the possibility to use stainless steel<br />fibers for the separation of mineral oil drops, broadening the apprehension<br />of the simultaneous influence of surface energy, bed geometry and oil phase<br />nature, as well as the introduction of an iso-output concentration as a new<br />parameter. The evaluation of the fiber-liquid wetting kinetic showed that the<br />lipophilic-lyophobic wettability ratio of the material adequately describes the<br />tendency of a filter material for the mineral oil separation. Such comprehensive<br />analysis of the various parameters impact on the efficiency of coalescence<br />filtration, presented in this thesis, was not found in available literature. The results<br />of the doctoral thesis will enable the reduction of time and material cost<br />needed for the design of coalescer, by decreasing the extent of experimental<br />research necessary for successful design.<br />Selected working conditions, primarily the fiber diameter in relation to the input<br />drop size, as well as bed packaging, while ensuring the known and defined<br />bed permeability, provided the desired separation, with the oil drop separation<br />efficiency of tested stainless steel fibers in the range from 91% to 99% for a fluid<br />velocity range from 10 m/h to 50 m/h. When stainless steel was used, more<br />efficient separation was observed for the low bed permeability, whereas for<br />the polymer fibers the reverse order was identified. It was concluded that, in<br />the existing circumstances, different mechanisms of oil drop coalescence dominate<br />for these materials. The dependence of output concentration from the<br />velocity and bed permeability for both tested materials, stainless steel and polypropylene, was similar. Polypropylene has a higher operating velocity range<br />in comparison to stainless steel. However, when compared to polypropylene,<br />stainless steel separates more successfully the oil with the lowest viscosity in<br />the low bed permeability and at low fluid velocity domain. Satisfactory quality<br />of separation with polypropylene is achieved at the highest bed permeability<br />and in a wide fluid velocity range, whereas, with the stainless steel, the highest<br />values of critical velocity, for the entire range of tested viscosity, are realized at<br />the lowest bed permeability.</p>

Une augmentation du taux de combustion des combustibles nucléaires UO2 dans les réacteurs à eau légère se traduit par l’apparition d’un changement de structure microscopique, appelée HBS. Bien que caractérisée expérimentalement de façon détaillée, des points importants sur les mécanismes de sa formation restent à éclaircir. Afin de répondre à ces questions, une étude de la contribution des défauts de type dislocation a été conduite. Dans une première partie, une méthode de calcul du champ de contraintes associées à des configurations périodiques de dislocations a été développée. La méthode a été appliquée aux cas des empilements et murs de dislocations de type coin, pour lesquels une expression explicite du potentiel des contraintes internes a été obtenue. A travers l’étude d’autres exemples de configurations, il a été mis en évidence que cette méthode permet le calcul de n’importe quelle configuration périodique de dislocations. Dans une seconde partie, l’évolution des défauts d’irradiation de type boucle d’interstitiels a été étudiée dans des échantillons de combustible UO2 dopé avec 10% en masse d’émetteurs alpha. Les distributions expérimentales en taille de boucles d’interstitiels ont été obtenues pour ces échantillons stockés pendant 4 et 7 ans à température ambiante. Des équations cinétiques sont proposées afin d’étudier l’influence de la remise en solution d’interstitiels contenus dans une boucle due à un impact avec l’atome de recul 234U, ainsi que la coalescence de deux boucles de dislocation pouvant diffuser en volume. L’application du modèle montre que les deux processus introduits doivent être considérés dans l’étude de l’évolution des défauts d’irradiation. / An increase of the discharge burn-up of UO2 nuclear fuels in the light water reactors results in the appearance of a change of microscopic structure, called HBS. Although well characterised experimentally, important points on the mechanisms of its formation remain to be cleared up. In order to answer these questions, a study of the contribution of the dislocation-type defects was conducted. In a first part, a calculation method of the stress field associated with periodic configurations of dislocations was developed. The method was applied to the cases of edge dislocation pile-up and wall, for which an explicit expression of the internal stress potential was obtained. Through the study of other examples of dislocation configurations, it was highlighted that this method also allows the calculation of any periodic dislocation configuration. In a second part, the evolution of interstitial-type dislocation loops was studied in UO2 fuel samples doped with 10% in mass of alpha emitters. The experimental loop size distributions were obtained for these samples stored during 4 and 7 years at room temperature. Kinetic equations are proposed in order to study the influence of the re-solution process of interstitials from a loop back to the matrix due to an impact with the recoil atom 234U, as well as the coalescence of two interstitial loops that can diffuse by a volume mechanism. The application of the model shows that the two processes must be considered in the study of the evolution of radiation damage.

Sub-salt oil and gas formations in deep-water northern Gulf of Mexico are high priority targets. Advances in seismic processing allow for high-resolution, below-salt imaging. Understanding the modes of salt emplacement provide insight into sub-salt traps and potential drilling hazards. A sub-salt sedimentary unit lies in the Keathley Canyon protraction. Autosutures created the transport-parallel lineaments of the upper surface of the unit. In addition, highly variable sediment aggradation rates created ramps, flats, and basal cutoffs along the base of the allochthon as salt and sediment competed for space. Seismic models identify modes of salt emplacement, salt/sediment interactions, and mechanisms responsible for the morphology. Petrophysical assessments highlight an abnormally pressured, dirty salt environment transitioning into a gouge zone. Dirty salt adds an element of difficulty to managing borehole pressures requiring a unique mud-weight plan designed to resist formation pressures without fracturing lithology.

Orientador: Ingrid Koch / Resumo: Aspidosperma pyrifolium Mart. & Zucc. é uma espécie com distribuição ampla e disjunta, variação morfológica e taxonomia complexa. Neste estudo é tratada como um complexo de espécies, e utilizada como modelo para delimitação de espécies através de múltiplos critérios operacionais (genéticos, morfológicos e ecológicos). Consideramos que espécies são segmentos de linhagens de uma metapopulação evoluindo separadamente, pois este é um conceito universal que diminuiu as visões conflitantes do que é considerado espécie. Realizamos análises filogenéticas com dados concatenados (ITS e rpl32-trnL) e de coalescência para testar o monofiletismo do complexo e seus grupos. Desenvolvemos marcadores microssatélites polimórficos para A. pyrifolium usados na avaliação da diversidade genética e estruturação de cinco populações naturais. Analisamos a morfologia através de morfometria, incluindo características quantitativas, tanto vegetativas quanto reprodutivas. As análises ecológicas, incluíram modelagem e testes de similaridade de nicho ecológico. Nossos resultados recuperaram o complexo A. pyrifolium como monofilético, com 3 subclados relacionados a regiões geográficas e vegetações específicas, resultados também corroborados na árvore de espécies. O clado 1 ocorre na Caatinga no Nordeste do Brasil, o clado 2 nas manchas de Floresta Estacional na região Centro-Oeste no Brasil e clado 3 na vegetação chaqueana no Mato grosso do Sul, Paraguai e Bolívia. Todas as análises de estrutura populaciona... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: Aspidosperma pyrifolium Mart. & Zucc. is a species with wide distribution and disjunct, morphological variation and complex taxonomy. In this study, we treated it as a species complex, and used as a model for species delimitation using multiple operational criteria (genetic, morphological and ecological). We consider species as segments of separately evolving metapopulation lineages, since this is a universal concept that diminished the conflicting visions of what is considered species. For this, we performed phylogenetic with concatenated data (ITS and rpl32-trnL) and coalescence analyzes to test the monophyly of the complex and its groups. We developed polymorphic microsatellites for complex A. pyrifolium for evaluated the genetic diversity and structure of five natural population. We analyzed morphology through morphometry, including quantitative characteristics, both vegetative and reproductive. The ecological analyzes included modeling and testing of ecological niche similarity. Our results recovered Aspidosperma pyrifolium complex as monophyletic, with 3 subclades related to geographic regions and specific vegetation, results also corroborated in the species tree. Clade 1 occurs in the Caatinga in Northeast Brazil, clade 2 in the Seasonal Forest patches in the Center-West region of Brazil and clade 3 in the Chaqueana vegetation in Mato Grosso do Sul, Paraguay and Bolivia. All analyzes of population structure suggested the formation of two genetic groups, one with the po... (Complete abstract click electronic access below) / Doutor

Ce travail de thèse s’est focalisé sur l’étude des alliages Fe-14%Cr nanorenforcés par dispersion d’oxydes obtenus par broyage à haute énergie à partir de poudres pré-alliées de la matrice et d’Y2O3. L’objectif était double : (i) étudier la cinétique de précipitation, en particulier la phase de coalescence, des oxydes en fonction de la teneur en élément d’addition (Ti, Y, O) lors de traitements thermiques à haute température et postérieurs à l’étape de consolidation, et (ii) d’analyser les hétérogénéités de déformations élastiques et élastoplastiques dans ces aciers en fonction de l’avancement de la précipitation. La microstructure et la nanostructure ont été étudiées par microscopie électronique en transmission (MET) et diffusion de neutrons aux petits angles (DNPA). L’ensemble de ces techniques a permis de mettre en évidence des comportements différents selon les teneurs en éléments d’addition. En particulier, l’ajout de titane induit un ralentissement très net de la coalescence des particules contrairement aux ajouts d’oxygène et d’yttrium. Ces variations de teneurs initiales influent sur la forme, la structure cristallographique des particules, sur les relations d’orientations avec la matrice et en voie de conséquence, sur les cinétiques de précipitation. Par conséquent, assurer un rapport Ti /Y supérieur à 1 et limiter l’apport en oxygène sont des garants de la stabilité des nanoparticules à hautes températures dans les aciers ODS. Le phénomène de recristallisation a également été observé à haute température dans les matériaux dans lesquels la coalescence des oxydes est largement avancée. Afin de corréler l’évolution des microstructures avec celle des propriétés mécaniques, une modélisation macroscopique du durcissement a ensuite été réalisée, en distinguant les différentes contributions (nanoparticules, taille de grains et dislocations). Le durcissement calculé à partir des observations expérimentales est en très bon accord avec les valeurs mesurées. Ces calculs rendent bien compte des variations entre alliages et mettent en évidence l’influence prépondérante de la taille des particules sur la dureté des matériaux. Enfin une étude fine des hétérogénéités de déformation des grains en fonction de leur orientation cristallographique a été menée par diffraction de neutrons. Ces résultats ont été confrontés à un calcul micromécanique en champ moyen. Cette dernière partie a montré l’existence de comportements locaux différents dans des matériaux présentant des comportements macroscopiques similaires. / This PhD work focused on the study of Fe-14%Cr alloys nanoreinforced by oxide dispersion obtained by high energy milling from pre-alloyed powders of the matrix and Y2O3.The main objectives were: (i) the study of the precipitation kinetics of oxides, in particular the coarsening, according to the content of added elements (Ti, Y, O) after consolidation and after heat treatments at high temperature. (ii) The elastoplastic and elastic deformations heterogeneities analyze in these alloys, depending on the progress of the precipitation. Microstructure and nanostructure were studied by Transmission Electron Microscopy (TEM) and Small angles neutron scattering (SANS).All of these techniques allowed identifying different behaviors depending on the added element contents. In particular, the addition of titanium induced a marked deceleration of oxide coarsening in contrast to those samples added oxygen and yttrium. These variations of initial contents have an influence on the form, the crystallographic structure of the particles, the orientation relationships with the matrix and consequently, the precipitation kinetics. Therefore, ensuring the ratio Ti/Y greater than 1 and, limiting the oxygen can guarantee the stability of the nanoparticles at high temperatures in the ODS alloys. The recrystallization phenomenon was also observed at high temperature in the materials of which the oxides coarsening is fast. In order to correlate the evolution of microstructure with the mechanical properties, a macroscopic model of hardening was then carried out by distinguishing the different contributions (nanoparticles, grain size and dislocations). The hardening calculated from experimental observations, is in good agreement with measured values. These calculations demonstrate the high influence of particle size on the hardness of materials. Finally, a detailed study of deformation heterogeneities of the grains according to their crystallographic orientation was carried out by neutron diffraction. These results were compared to a mean field micromechanics calculation. This last part has shown the existence of different local behaviors for materials presenting similar macroscopic behavior.

Foam flow in porous media is important for many industrial operations such as enhanced oil recovery, remediation of contaminated aquifers and CO2 sequestration. The application of foam in these processes is due to its unique ability to reduce gas mobility and to divert gas to low permeability zones in porous media which otherwise would not be reached. To achieve optimum success with foam as a displacing fluid in oil recovery and remediation operations, it is essential to understand how different parameters influence foam flow in porous media. In this thesis, a variety of experimental techniques were used to study foam stability, foam rheology as well as the dynamics and patterns of oil displacement by foam under different boundary conditions such as surfactant formulation, oil type, foam quality (gas fraction) and porous media geometry. Bulk scale studies showed that foam stability was surfactant and oil dependant such that decreasing oil carbon number and viscosity decreased the stability of foam. However, no meaningful correlation was found between foam stability at bulk scale and the efficiency of oil displacement in porous media for the various surfactants studied in this work. Additionally, our results show that foams consisting of smaller bubbles do not necessarily correspond to higher apparent viscosity as the foam quality is also crucial. For the same foam quality decreasing bubble size resulted in higher apparent viscosity. Although in theory a higher apparent viscosity (i.e. higher foam quality) would be ideal for displacement purposes, increasing foam quality resulted in less stable foam in porous media due to formation of thin films which were less stable in the presence of oil. The effect of pore geometry on foam generation and oil displacement has also been investigated. Our findings provide new insights about the physics and complex dynamics of foam flow in porous media.

The phenomenon called soil aggregate coalescence occurs at contact-points between aggregates and causes soil strength to increase to values that can inhibit plant root exploration and thus potential yield. During natural wetting and drying, soil aggregates appear to ‘weld’ together with little or no increase in dry bulk density. The precise reasons for this phenomenon are not understood, but it has been found to occur even in soils comprised entirely of water stable aggregates. Soil aggregate coalescence has not been widely observed and reported in soil science and yet may pose a significant risk for crops preventing them from achieving their genetic and environmental yield potentials. This project used soil penetrometer resistance and an indirect tensile-strength test to measure the early stages of aggregate coalescence and to evaluate their effects on the early growth of tomato plants. The early stages of aggregate coalescence were thought to be affected by a number of factors including: the matric suction of water during application and subsequent drainage, the overburden pressure on moist soil in the root zone, the initial size of soil aggregates prior to wetting, and the degree of sodicity of the soil aggregates. Seven mainexperiments were conducted to evaluate these factors. The matric suction during wetting of a seedbed affects the degree of aggregate slaking that occurs, and the strength of the wetted aggregates. The matric suction during draining affects the magnitude of ‘effective stresses’ that operate to retain soil structural integrity as the soil drains and dries out. An experiment was conducted to evaluate the influence of matric suction (within a range of suctions experienced in the field) on aggregate coalescence using soils of two different textures. Sieved aggregates (0.5 to 2 mm diameter) from a coarse-textured and two fine-textured (swelling) soils were packed into cylindrical rings (4.77 cm i.d., 5 cm high) and subjected to different suctions on wetting (near-saturation, and 1 kPa), and on draining (10 kPa on sintered-glass funnels, and 100 kPa on ceramic pressure plates). After one-week of drainage, penetrometer resistance was measured as a function of depth to approximately 45 mm (penetrometer had a recessedshaft, cone diameter = 2 mm, advanced at a rate of 0.3 mm/min). Tensile strength of other core-samples was measured after air-drying using an indirect “Brazilian” crushing test. For the coarse-textured soil, penetrometer resistance was significantly greater for samples wet to near-saturation, despite there being no significant increase in dry bulk density; this was not the case for the finer-textured soils, and it was difficult to distinguish the effects of variable bulk density upon drying from those of the imposed wetting treatments. In both coarse- and fine-textured soils, the tensile strength was significantly greater for samples wet to near-saturation. Thus wetting- and draining-suctions were both found to influence the degree of soil aggregate coalescence as measured by penetrometer resistance and tensile strength. Aggregate coalescence in irrigated crops is known to develop as the growing season progresses. It was therefore thought to be linked to the repeated occurrence of matric suctions that enhance the phenomenon during cycles of wetting and draining. An experiment was conducted to determine the extent of aggregate coalescence in a coarsetextured and two fine-textured (swelling clay) soils during 8 successive cycles of wetting and draining. Sieved aggregates (0.5 to 2 mm diameter) from each soil were packed into cylindrical rings (4.77 cm i.d., 5 cm high) and wetted to near saturation for 24 h. They were then drained on ceramic pressure plates to a suction of 100 kPa for one week, after which penetrometer resistance and tensile strength were measured as described above. The degree of expression of aggregate coalescence depended on soil type. For the coarse-textured soil, repeated wetting and draining significantly increased bulk density, penetrometer resistance and tensile strength. For the fine-textured soil, penetrometer resistance and bulk density did not vary significantly with repeated wetting and draining; on the contrary, there was evidence in these swelling clay soils to suggest bulk density and penetrometer resistance decreased. However, there was a progressive increase in tensile strength as cycles of wetting and draining progressed. The expansive nature of the fine-textured soil appears to have masked the development of aggregate coalescence as measured by penetrometer resistance, but its expression was very clear in measurements of tensile strength despite the reduction in bulk density with successive wetting and draining. Field observations have indicated that aggregate coalescence is first expressed at the bottom of the seedbed and that it develops progressively upward to the soil surface during the growing season. This suggests that overburden pressures may enhance the onset of the phenomenon by increasing the degree of inter-aggregate contact. Soils containing large quantities of particulate organic matter were known to resist the onset of aggregate coalescence to some extent. An experiment was conducted to evaluate the effects of soil organic matter and overburden pressures, by placing brass cylinders of various weights (equivalent to static load pressures of 0, 0.49, 1.47 and 2.47 kPa) on the top of dry soil aggregates (0.5 – 2 mm diameter) having widely different soil organic carbon contents placed in steel rings 5 cm high and 5 cm i.d. With the weights in place, the aggregates were wetted to near-saturation for 24 h and then drained on ceramic pressure plates to a suction of 100 kPa for one week. Bulk density, penetrometer resistance and tensile strength were measured when the samples were removed from the pressure plates and they all increased significantly with increasing overburden pressure in the soil with low organic matter content, but not in the soil with high organic matter content. The amount of tillage used to prepare seedbeds influences the size distribution of soil aggregates produced – that is, more tillage produces finer seedbeds. The size distribution of soil aggregates affects the number of inter-aggregate contact points and this was thought to influence the degree of aggregate coalescence that develops in a seedbed. Previous work has shown that soil organic matter reduces aggregate coalescence and so an experiment was conducted to evaluate the effects of aggregate size and organic matter on the phenomenon. For soils with high and low organic matter contents, aggregate size fractions of < 0.5, 0.5 – 2, 2 – 4, and < 4 mm were packed into soil cores (as above) and wetted to near-saturation then drained to 100 kPa suction as described above. Penetrometer resistance and tensile strength were measured and found to increase directly with the amount of fine material present in the soil cores – being greater in the < 0.5 mm and < 4 mm fractions, and being less in the 0.5 – 2 mm and 2 – 4 mm fractions. In all cases, penetrometer resistance and tensile strength were lower in the samples containing more organic matter. The rate at which soil aggregates are wetted in a seedbed affects the degree of slaking and densification that occurs, and the extent to which aggregates are wetted influences the overall strength of a seedbed. Both wetting rate and the extent of wetting were believed to influence the onset of aggregate coalescence and were thought to be affected by soil organic matter and irrigation technique. An experiment was therefore designed to separate these effects so that improvements to management could be evaluated for their greatest efficacy – that is, to determine whether management should focus on improving irrigation technique or increasing soil organic matter content, or both. The rate of wetting was controlled by spraying (or not spraying) soil aggregates (0.5 – 2 mm diameter) with polyvinyl alcohol (PVA). Samples of coarse- and fine-textured soils were packed into steel rings (as above) and subjected to different application rates of water (1, 10 and 100 mm/h) using a dripper system controlled by a peristaltic pump. Samples were brought to either a near-saturated state or to a suction of 10 kPa for 24 h, and then drained on a pressure plate at a suction of 100 kPa for one week. Measurements of penetrometer resistance and tensile strength were then made as described above. As expected, penetrometer resistance was lower in samples treated with PVA before wetting (slower wetting rates) and in samples held at a greater suction (10 kPa) after initial wetting (greater inter-aggregate strength). The effects were more pronounced in the coarse-textured soil. In both coarse- and fine-textured soils, tensile strengths increased with increasing wetting rate (greatest for 100 mm/h) and extent of wetting (greater when held at near-saturated conditions). The rate of wetting was found to be somewhat more important for promoting aggregate coalescence than the extent of wetting. Because aggregate coalescence often occurs with little or no increase in bulk density, an explanation for the increase in penetrometer resistance and tensile strength is unlikely to be explained by a large increase in the number of inter-aggregate contacts. An increase in the strength of existing points of inter-aggregate contact was therefore considered in this work. For inter-aggregate bond strengths to increase, it was hypothesized that small increases in the amount of mechanically (or spontaneously) dispersed clay particles, and subsequent deposition at inter-aggregate contact points could increase aggregate coalescence as measured by penetrometer resistance and tensile strength. An experiment was devised to manipulate the amount of spontaneously dispersed clay in coarse- and fine-textured soils of high and low organic matter content. The degree of sodicity of each soil was manipulated by varying the exchangeable sodium percentage (ESP) of soil aggregates (0.5 – 2mm) above and below a nominal threshold value of 6. Dry aggregates were then packed into steel rings (as above) and subjected to wetting near saturation, then draining to a suction of 100 kPa for one week as described above. Measurements were then taken of penetrometer resistance and tensile strength, both of which were affected by ESP in different ways. In the coarse-textured soil, sodicity enhanced aggregate slaking and dispersion, which increased bulk density. While penetrometer resistance also increased, its effect on aggregate coalescence could not be separated from a simple effect of increased bulk density. Similarly, the effect of sodicity on aggregate coalescence in the fine-textured soil was confounded by the higher water contents produced by greater swelling, which produced lower-than-expected penetrometer resistance. Measurements of tensile strength were conducted on air-dry samples, and so the confounding effects of bulk density and water content were eliminated and it was found that tensile strength increased with sodicity in both coarse- and fine-textured soils. The presence of dispersed clay was therefore implicated in the development of aggregate coalescence in this work. Finally, a preliminary evaluation of how the early stages of aggregate coalescence might affect plant growth was attempted using tomatoes (Gross lisse) as a test plant. Seeds were planted in aggregates (0.5 – 4 mm) of a coarse- or fine-textured soil packed in steel rings. These were wetted at a rate of 1 mm/h to either near-saturation (for maximum coalescence) or to a suction of 10 kPa (for minimum coalescence) and held under these conditions for 24 h. All samples were then transferred to a ceramic pressure plate for drainage to 100 kPa suction for one week. Samples were then placed in a growth-cabinet held at 20C with controlled exposure to 14 h light/day. Germination of the seeds, plant height, and number and length of roots were observed. Germination of the seeds held at near-saturation in both coarse- and fine-textured soils was delayed by 24 h compared with seeds held at 10 kPa suction. Neither the number nor the length of tomato roots differed significantly between the different treatments and soils. In the coarse-textured soil, however, the total root length over a period of 14 days was somewhat greater in the uncoalesced samples than in the coalesced samples, but this difference was not statistically significant. These results suggest that aside from delaying germination, aggregate coalescence may not have a large effect on early growth of tomato plants. However, this is not to say that detrimental effects may not be manifest at later stages of plant growth, and this certainly needs to be evaluated, particularly because aggregate coalescence increase with repeated cycles of wetting and draining. In conclusion, the primary findings of the work undertaken in this thesis were: • Rapid wetting of soil aggregates to near-saturation enhanced the onset of soil aggregate coalescence as measured by (in some cases) penetrometer resistance at a soil water suction of 100 kPa, and (in most cases) tensile strength of soil cores in the air-dry state. The rate of wetting appeared to be more important in bringing on aggregate coalescence than how wet the soil eventually became during wetting. This means reducing the rate at which irrigation water is applied to soils may reduce the onset of aggregate coalescence more effectively than controlling the total amount of water applied – though both are important. The literature reports that aggregate coalescence occurs in the field over periods of up to several months, involving multiple wetting and draining cycles, but the work here demonstrated that this can occur over much shorter time periods depending on conditions imposed. • Aggregate coalescence occurred in coarse-textured soils regardless of whether the bulk density increased during wetting and draining. In finer-textured soils, the response to wetting conditions varied and was complicated by changes in bulk density and water content due to swelling. • Small overburden pressures enhanced the onset of aggregate coalescence, but these effects were diminished in the presence of high soil organic matter contents. • Finer aggregate size distributions (which are often produced in the field by excessive tillage during seedbed preparation) invariably led to greater aggregate coalescence than coarser aggregate size distributions. The effects of aggregate size were mitigated to some extent by higher contents of soil organic matter. • Sodicity enhanced aggregate coalescence as measured by tensile strength, but when penetrometer resistance was measured in the moist state, the effects were masked to some extent by higher water contents generated by swelling and dispersion. This work suggests that tensile strength (in the air dry state) may be a more effective measure of aggregate coalescence than penetrometer resistance. • Early plant response to aggregate coalescence was not large, but the response may become magnified during later stages of growth. / http://proxy.library.adelaide.edu.au/login?url= http://library.adelaide.edu.au/cgi-bin/Pwebrecon.cgi?BBID=1297583 / Thesis (Ph.D.) -- School of Earth and Environmental Sciences, 2007

Ce mémoire a pour objet le calcul des collisions de gouttes dans divers écoulements laminaires et turbulents. Dans le cadre de l'approximation des équations de Stokes, les forces d'interactions hydrodynamiques entre deux gouttes sont déterminées de façon exacte au moyen d'un système de coordonnées bisphériques. Ces forces permettent de calculer les taux de collision pour des écoulements de cisaillement simple et d'élongation pure. Les sections de collision sont obtenues par trajectographie. Les résultats montrent un accroissement de l'efficacité de collision lorsque l'inertie des gouttes croît. L'expression du taux de collision en écoulement turbulent est analogue à celle utilisée en théorie cinétique des gaz. La fonction de distribution de paires est obtenue pour une turbulence homogène et isotrope et à l'aide de notions de probabilités conditionnelles. Les taux de collision sont calculés sans et avec la gravité. L'évolution granulométrique d'un nuage de gouttes est décrite à l'aide d'une équation cinétique de type Smoluchowski. Cette équation est résolue au moyen d'une méthode numérique originale. Deux cas sont traités. Pour de petites gouttes en écoulement turbulent homogène et isotrope, l'élargissement du spectre de gouttes demande plus de temps si les interactions hydrodynamiques sont prises en compte. Pour de grosses gouttes, compte tenu de la gravité, la formation par collisions de gouttes de diamètre supérieur à 200 microm dépend fortement de la turbulence. Les résultats corroborent les relevés \textit{in situ} dans des strati.

Le travail de thèse s'articule autour de deux études : étude de la réaction de spallation p+Au à 2.5 GeV et étude de la production neutronique en cibles épaisses. Ces deux études s'inscrivent dans un programme de recherche initié dans le cadre du projet de Source Européenne de Spallation (ESS) et du projet français de traitement des déchets radioactifs (GEDEON). A l'aide d'un dispositif expérimental très complet alliant un détecteur de neutrons à des détecteurs de particules chargées, tous deux d'efficacité supérieure à 80\%, nous avons pu étudier l'évolution du processus élémentaire de spallation en fonction de l'énergie déposée par le proton incident dans le noyau (énergie d'excitation). Nous avons ainsi montré le réalisme d'une description du processus en deux étapes et validé l'utilisation du code de cascade intra-nucléaire de J. Cugnon (INCL2.0) couplé à un code de désexcitation statistique (GEMINI). Les sections efficaces de production des particules chargées légères ont été mesurées et la contribution de l'émission de prééquilibre évaluée. La formation des particules composites ``directes'' a été ajoutée au code INCL2.0 à l'aide d'un modèle de coalescence. La production neutronique (multiplicités moyennes et distributions) a été étudiée en fonction de l'énergie du faisceau, de la géométrie de la cible et de la nature de la cible. Nous avons montré que le gain en neutrons, lorsque l'épaisseur de cible augmente, est dominé par la probabilité de réaction (pour des épaisseurs de cibles inférieures à 2 ou 3 longueurs d'interaction) et d'une façon moindre par le développement des réactions secondaires. Ces deux paramètres ont été étudiés séparément. Les données expérimentales ont été comparées aux résultats du code de transport HERMES, validant celui-ci pour l'émission de neutrons.