Os pilares mistos preenchidos são elementos estruturais constituídos por um tubo de aço preenchido com concreto de qualidade estrutural. O objetivo principal do trabalho foi estudar o comportamento dos pilares mistos preenchidos de seção circular. Para isso, foi avaliada a influência da resistência à compressão do concreto (fc), da esbeltez do pilar (L/D), da espessura do tubo de aço (t) e da forma de introdução do carregamento, se no núcleo de concreto ou na seção mista do pilar. Foram feitos ensaios experimentais de 64 pilares mistos preenchidos, submetidos à força de compressão centrada. Os resultados de capacidade resistente obtidos foram avaliados e comparados com os resultados de previsão segundo os procedimentos normativos estrangeiros do EUROCODE 4:2004, ANSI/AISC:2005 e CAN/CSA:2001 e do Projeto de Revisão da Norma Brasileira (PR-NBR 8800, draft 2008). Os resultados previstos foram maiores que os experimentais para os procedimentos do ANSI/AISC e da NBR e menores para os procedimentos do EC4 e do CAN/CSA. A partir da comparação dos valores de capacidade resistente experimentais e normativos, foi desenvolvido um estudo de confiabilidade dos valores de força resistente obtidos segundo as equações de dimensionamento dessas normas, por meio da estimativa do erro de modelo inerente a cada uma. Os resultados mostraram que o EC4 apresenta os índices de confiabilidade mais uniformes. Além disso, foi feita uma modelagem numérica dos pilares mistos preenchidos com relação L/D = 10, buscando obter o valor da capacidade resistente desses pilares e compará-los com os resultados experimentais. Embora o modelo numérico tenha representado adequadamente o comportamento dos pilares preenchidos esbeltos, os valores de força última foram inferiores aos experimentais. / Concrete-filled steel tubular columns are structural elements made of a steel tube filled with structural concrete. The main objective of this work was study the global behavior of concrete-filled steel tubular columns. For this, some variables were evaluated: concrete compressive strength (fc), length-to-diameter ratio (L/D), diameter-to-thickness ratio (D/t) and the type of load application (on concrete core or on composite cross section). Tests of 64 concrete filled columns were performed. The columns were subjected to axial load. The results of experimental compressive strength were compared to results from three different international codes (EUROCODE 4:2004, ANSI/AISC:2005 e CAN/CSA:2001) and the national NBR 8800, draft 2008. The results showed, in general, a conservative behavior of values predicted by NBR and ANSI/AISC e non-conservative values by EC4 and CAN/CSA. Based on comparisons between experimental values and codes predictions, an error model variable was estimated to each code and a reliability analysis was performed. The results indicate that the EC4 is the best code with less tendentious behavior of its design equation. A numerical analysis of columns with slenderness (L/D) equal to 10 was carried out to try to predict the peak load of columns and compare the results with those from experimental analysis. Althougth the results of the numerical analysis had a good match to the experimental ones, the values, in average, were conservatives.

Com o objetivo de avaliar a influência de teores crescentes de proteína bruta (PB) em dietas com alta proporção de concentrado para cordeiros deslanados, foram realizados dois experimentos: Experimento I (desempenho dos animais) e Experimento II (digestibilidade dos nutrientes). No Experimento I foram utilizados 48 cordeiros da raça Santa Inês (peso inicial de 18,4 ± 0,4 kg e idade média inicial de 86 ± 2 dias) em um delineamento experimental de blocos casualizados com 4 tratamentos e 6 repetições. As dietas experimentais continham 80% de concentrado e 20% de volumoso (10% bagaço-de-cana hidrolisado e 10% bagaço-de-cana in natura), com teores de 14, 16, 18 e 20% PB na matéria seca. Não houve diferença (P > 0,05) entre os tratamentos para consumo de matéria seca, ganho de peso, conversão alimentar e rendimento de carcaça. A concentração de N-uréico plasmático após 3 horas do fornecimento da alimentação não aumentou para as dietas contendo teores de PB acima de 16%. No Experimento II foram utilizados 5 ovinos da raça Santa Inês providos de cânulas ruminais (peso médio inicial de 43 kg e idade média inicial de 6 meses) em um delineamento experimental em quadrado latino 5x5. As dietas experimentais continham 80% de concentrado e 20% de volumoso (10% bagaço-de-cana hidrolisado e 10% bagaço-de-cana in natura), com níveis de 12, 14, 16, 18 e 20% PB na matéria seca. Não houve diferença (P > 0,05) entre os tratamentos para digestibilidade aparente da matéria seca, da matéria orgânica, da fibra em detergente neutro, da fibra em detergente ácido e do extrato etéreo. A digestibilidade aparente da proteína bruta e a retenção de N (g/dia) foram maiores (P<0,05) para as dietas com 16, 18 e 20% PB em relação às dietas com 12 e 14% PB. O aumento da concentração protéica da dieta proporcionou um aumento (P<0,05) na concentração de N amoniacal no rúmen, entretanto não teve efeito (P>0,05) nos valores de pH e na produção de ácidos graxos voláteis. / Two experiments were conducted to evaluate the effects of crude protein (CP) levels on high grain diets fed to ram lambs. In Exp. I (performance trial), 48 Santa Inês lambs (initial body weight 18.4 ± 0.4 kg and 86 ± 2 days old) were fed 4 experimental diets containing 14, 16, 18 and 20% CP, in a completely randomized block desing. Diets consisted of 80% concentrate and 20% sugarcane bagasse. There were no differences (P > 0.05) in dry matter intake, average daily gain, feed efficiency and dressing percentage. Plasma urea nitrogen concentration did not increase in diets containing more than 16% CP. In Exp. II (metabolism trial), 5 Santa Inês lambs (43 kg body weigh) fitted with ruminal cannulas were used in a 5x5 Latin Square design. Diets consisted of 80% concentrate and 20% sugarcane bagasse with 12, 14, 16, 18 and 20% CP.There were no differences (P > 0.05) in dry matter, organic matter, neutral detergent fiber, acid detergent fiber and ether extract apparent digestibilities in the total digestive tract. Crude protein apparent digestibility in the total digestive tract and N retention (g/day) were greater (P < 0.05) for diets with 16, 18 and 20% CP than for the diets with 12 and 14% CP. Ruminal ammonia-N concentration increased (P < 0.05) with increasing levels of dietary CP, however no differences (P > 0.05) were observed on pH and ruminal volatile fatty acids concentration.

Apesar de sua familiaridade e simplicidade, a água apresenta um conjunto propriedades termodinâmicas, dinâmicas e estruturais que são ainda objeto de intensa pesquisa. O aumento da densidade com a temperatura, da difusão com a densidade, ou ainda do ordenamento com a temperatura são exemplos de alguns de seus comportamentos não usuais. Com a finalidade de melhor compreender tais propriedades inúmeras abordagens têm sido utilizadas, tais como o uso geometrias de confinamento, modelos simplificados ou até mesmo misturas. Dentre as geometrias confinantes frequentemente usadas, encontra-se, nanoporos, placas paralelas e meio porosos. Os meios porosos são formados por obstáculos fixos que impõem efeitos de volume excluído adicionais ao sistema. Já no caso de misturas quando elas ocorrem entre líquidos capazes de formar ligações de hidrogênio, o comportamento não usual da água dá origem a um conjunto ainda maior de propriedades anômalas. A mistura água-metanol por exemplo, é munida de um conjunto propriedades de excesso incapazes de serem descritas pelas teorias usuais. São alguns exemplos, o máximo no calor específico e o mínimo no volume e entalpia de excesso. Neste projeto de doutoramento, nós estudamos por simulações numericas o confinamento por meio poroso (desordem queched) e misturas de água com solutos interagentes. O primeiro estudo é realizado usando um modelo 2D tipo-água que é largamente conhecido na literatura. No segundo estágio, estamos a influência de solutos interagentes nas propriedades de modelos em rede e contínuos. Para o modelo em rede, nós desenvolvemos um modelo de soluto e posteriormente uma técnica capaz de simular misturas de modelos em rede a pressão constante. De posse desta técnica estudamos as propriedades de excesso da mistura. / Although the familiarity and simplicity, the water show a set of thermodynamic, dynamics and structural properties which are still subject to intense research. The increase of density as the temperature, of diffusion as the density, or even of ordering with the temperature are examples of some of its unusual behavior. In order to better understand these properties numerous approaches have been used, such as the use of confinement geometries, simplified models, or ever mixtures. Among the confinement geometries used, are those, nanopores, parallel plates and porous media. The porous media are formed by fixed obstacles that impose the additional excluded volume effects to the system. In the case of mixtures, when they occur between liquids able to form hydrogen-bonds, the unusual behavior of water give rise to a set even higher anomalous properties. The water-methanol mixture, for example, has a set of excess properties unable to be described by usual theories. Some examples are the maximum in the specific heat and minimum in excess volume and enthalpy. In this Ph.D. project, we study by numerical simulations, the confinement of water by porous media(or under quenched disorder) and the mixture of water with interacting solutes. The first study is performed using a 2D lattice model which is widely known in the literature. In a second stage, we study the influence of interacting solutes on the properties of lattice and continuous models. For the lattice model, we develop a solute model and a technique to simulate mixtures of lattice models at constant pressure. Using this technique, we study the excess properties of the mixture. For the continuous model we study the influence of a dimeric solute on the TMD of a water-like model and posteriorly we study the excess properties of this type of mixture.

In this thesis, we construct two QCD based energy loss models to perform quantitative analysis of jet quenching observables in ultra-relativistic nucleus-nucleus collisions at RHIC and the LHC. We first build up a perturbative QCD based CUJET2.0 jet flavor tomography model that couples the dynamical running coupling DGLV opacity series to bulk data constrained relativistic viscous hydrodynamic backgrounds. It solves the strong heavy quark energy loss puzzle at RHIC and explains the surprising transparency of the quark-gluon plasma (QGP) at the LHC. The observed azimuthal anisotropy of hard leading hadrons requires a path dependent jet-medium coupling in CUJET2.0 that implies physics of nonperturbative origin. To explore the nonperturbative chromo-electric and chromo-magnetic structure of the strongly-coupled QGP through jet probes, we build up a new CUJET3.0 framework that includes in CUJET2.0 both Polyakov loop suppressed semi-QGP chromo-electric charges and emergent chromo-magnetic monopoles in the critical transition regime. CUJET3.0 quantitatively describes the anisotropic hadron suppression at RHIC and the LHC. More significantly, it provides a robust connection between the long wavelength "perfect fluidity'' of the QGP and the short distance jet transport in the QGP. This framework paves the way for ``measuring'' both perturbative and nonperturbative properties of the QGP, and more importantly for probing color confinement through jet quenching.

Submitted by Jaqueline Silva (jtas29@gmail.com) on 2014-09-15T18:12:49Z No. of bitstreams: 2 license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) Rodrigo Medeiros da Silva.pdf: 1734954 bytes, checksum: bb3851360da8bd727dad4bc7836a5534 (MD5) / Approved for entry into archive by Jaqueline Silva (jtas29@gmail.com) on 2014-09-15T18:13:10Z (GMT) No. of bitstreams: 2 license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) Rodrigo Medeiros da Silva.pdf: 1734954 bytes, checksum: bb3851360da8bd727dad4bc7836a5534 (MD5) / Made available in DSpace on 2014-09-15T18:13:10Z (GMT). No. of bitstreams: 2 license_rdf: 23148 bytes, checksum: 9da0b6dfac957114c6a7714714b86306 (MD5) Rodrigo Medeiros da Silva.pdf: 1734954 bytes, checksum: bb3851360da8bd727dad4bc7836a5534 (MD5) Previous issue date: 2013-11-10 / This study aimed at assessing the carcass and meat traits as well as fatty acid profile of crossbred steers of European or zebu genotypic predominance, finished in feedlot, fed diets with high percentage of concentrate (80%) containing different levels of grounded millet grain as replacement for grounded corn grain (0, 33, 66 and 100%). We allocated in 16 collective pens 45 steers with European predominance and 44 with zebu-predominance, with average initial weight of 317.8 and 320.7 kg, respectively, and average age of 21 month. At the end of the finishing period, we slaughtered six animals of each genetic group, randomly chosen within each level of replacement of corn by millet grain. The experimental design was completely randomized with treatments in a 4x2 factorial arrangement using six replicates. The replacement of corn grain by millet grain did not significantly affect slaughter weight (480.4 kg), hot carcass weight (259.5 kg), carcass yield (54.1%) and subcutaneous fat thickness (3.95 mm). Steers with predominant European genotype showed significantly higher values for slaughter weight (507.6 vs. 453.3 kg), hot carcass weight (269.6 vs. 249.3 kg) and Longissimus dorsi area (65.34 vs 56.83 cm2); however, they had lower carcass yield (53,2 vs. 55,00%). Increasing the proportion of millet in the diet linearly increased the arachidic (C20:0), heneicosanoic (C21:0), linolenic (C18:3 n-3) and di-homo linolenic (C20:3 n-6) fatty acids. European steers meat showed less content of myristic (C14:0), heneicosanoic (C21:0) and linolenic (C18:3 n-6) fatty acids. The total concentration of saturated (45.2%), monounsaturated (41.2%) and polyunsaturated (8.7%) fatty acids and, monounsaturated/saturated (1.09) and polyunsaturated/saturated (0.18) relationship were not affected by the factors studied. Millet grain can be used in replacement of corn grain for finishing steers since it does not change carcass traits of economic interest. The increase in the percentage of millet grain in the diet of European and zebu crossbred steers improves the ratio between n-6/n-3 fatty acids. / Objetivou-se avaliar as características de carcaça, da carne e perfil de ácidos graxos da carne de novilhos inteiros mestiços com predominância genotípica européia ou zebuína, alimentados em confinamento com dietas com elevada proporção de concentrado (80%) contendo diferentes níveis de grão de milheto moído em substituição ao grão de milho moído (0, 33, 66 e 100%). Foram confinados, em 16 baias coletivas, 45 novilhos com predominância genotípica européia e 44 novilhos com predominância genotípica zebuína, com peso médio inicial de 317,8 e 320,7 kg, respectivamente, e 21 meses de idade. Ao final do período de confinamento foram abatidos seis animais de cada grupo genético, escolhidos aleatoriamente dentro de cada nível de substituição do grão de milho pelo de milheto. O delineamento experimental foi o inteiramente casualizado, com os tratamentos em arranjo fatorial 4x2, utilizando seis repetições. A substituição do grão de milho pelo grão de milheto não influenciou significativamente os pesos de abate (480,4 kg), de carcaça quente (259,5 kg), rendimento de carcaça quente (54,1 %) e espessura de gordura subcutânea (3,95 mm). Novilhos com predominância genotípica européia apresentaram valores significativamente superiores para peso de abate (507,6 vs 453,3 kg), peso de carcaça quente (269,6 vs 249,3 kg) e área do músculo Longissimus dorsi (65,34 vs 56,83 cm2), porém, foram inferiores no rendimento de carcaça quente (53,2 vs 55,0 %). O avanço da percentagem do grão de milheto na dieta elevou linearmente a participação dos ácidos graxos araquídico (C20:0), heneicosanóico (C21:0), linolênico (C18:3 n-3) e di-homo linolênico (C20:3 n-6). Novilhos E apresentaram carne com menor (P<0,05) teor de mirístico (C14:0), heneicosanóico (C21:0) e linolênico (C18:3 n-6) do que novilhos Z. A concentração total de ácidos graxos saturados (45,2%), monoinsaturados (41,2%) e poliinsaturados (8,7%) e, a relação monoinsaturados/saturados (1,09) e poliinsaturados/saturados (0,18) não foram alterados pelos fatores estudados. O grão de milheto pode ser utilizado em substituição ao grão de milho, pois não altera as principais características de carcaça de interesse econômico de novilhos terminados em confinamento. O aumento da percentagem de grão de milheto na dieta de novilhos mestiços europeus ou zebuínos melhora a relação entre os ácidos graxos n-6/n-3.

Actuellement, une voie de développement de formulations médicamenteuses novatrices passe par la vectorisation de principes actifs connus dans des nanoparticules. Des matériaux hybrides sont ainsi formés, possédant de nouvelles propriétés liées au nano-confinement. Les travaux ici menés s’appuient sur la sensibilité de la Résonance Magnétique Nucléaire (RMN) du solide aux phénomènes prenant place aux échelles moléculaires, pour effectuer une analyse structurelle et dynamique de deux vecteurs. Le premier, hydrophile, est une matrice nanoporeuse de silice sol-gel, dans laquelle sont confinés des complexes de nitroprussiate de sodium isolés. À partir de mesures de relaxation de spin et d’anisotropie de déplacement chimique, différents régimes de dynamique moléculaire sont mis en évidence. Ils sont modulés par la présence de molécules de solvant résiduelles (H2O). Des gammes de température et d’hydratation sont identifiées, pour lesquelles le complexe reste associé malgré un état ‘‘pseudo-liquide’’. Dans la condition limite d’absence d’eau, la restriction du mouvement des complexes confinés est élucidée en caractérisant les interactions dipolaires hôtes / invités. Le second système allie la double vectorisation de la curcumine hydrophobe dans des nanoparticules de lipides solides encapsulées dans une matrice de silice (SBA-15). Une stratégie d’étude conjointe par RMN du solide et par calorimétrie différentielle à balayage (DSC) est mise en place. Les résultats montrent que d’autres facteurs que la compartimentalisation (polymorphisme, dynamique moléculaire des composés hôtes) doivent également être pris en compte pour la compréhension des propriétés de ces matériaux très hétérogènes. Malgré le recours à une instrumentation RMN de dernière génération (spectromètre 1GHz, sonde MAS 1.3mm), la présence de principe actif est observée uniquement dans les compartiments de tensioactif. Ces résultats permettent d’émettre de nouvelles hypothèses sur la distribution du principe actif, tout en montrant les limites de l’approche RMN basée uniquement sur l’étude des noyaux 1H. / Nowadays, a way of developing novel medicinal compounds focuses on confinement of known active molecules inside nanoparticles. Therefore, hybrid materials emerge, exhibiting new properties related to nano-confinement. This work relies on the sensibility of solid-state Nuclear Magnetic Resonance (SS-NMR) towards molecular scale phenomena in order to perform structural and dynamical analysis of two delivery systems. They are modulated by the influence of residual solvent molecules (H2O). Temperature and hydration ranges are identified, for which the complex stays associated, although it is in a liquid-like state. Toward the limit of water absence, movement restrictions of the confined complexes are elucidated by characterizing dipolar host / guest interactions. The second system combines a double vectorization of hydrophobic curcumin molecules inside solid lipid nanoparticles, encapsulated inside a silica matrix (SBA-15). A joint SS-NMR and Differential Scanning Calorimetry (DSC) characterization strategy is put in place. The results show that other factors than compartmentalization (polymorphism, molecular dynamics of host compounds) should also be taken into account to understand the properties of these very heterogeneous materials. Despite resorting to the latest NMR instrumentation (1GHz spectrometer, 1.3mm MAS probehead), presence of the active ingredient is only detected inside the surfactant compartment. These results allow making new assumptions for the distribution of curcumin inside the material while showing the limits of an NMR approach relying solely on the study of 1H nuclei.

Cette thèse de littérature comparée a pour objectif de mettre en rapport des œuvres habitées par l’Histoire, et d’interroger les représentations littéraires du corps face à l’épreuve extrême de l’enfermement. Le but de cette recherche, qui se déploie en trois temps, est de questionner la nature humaine à travers le prisme de l’écriture face à l’expérience bouleversante des camps de concentration et d’extermination nazis pendant la Seconde Guerre mondiale, en mettant en parallèle des œuvres tant de témoignage que de fiction, qui puisent leurs ressources chacune dans le réel et le fictionnel, dans un jeu de vases communicants.Plus précisément, dans le cadre de la première partie, nous nous concentrons sur la manière dont l’expérience-limite de l’être se manifeste dans ces récits : la confusion identitaire et la déshumanisation y bousculent la représentation du corps, le mettent en doute. Ce doute s’inscrit dans le langage même : comment raconter ce qui paraît inimaginable ? Dans cette deuxième partie, nous mettons l’accent sur l’aspect indicible de l’évènement, et réfléchissons aux contournements, aux déplacements que peut offrir la littérature pour dire ce qui semble, au premier abord, inénarrable. Les images et symboles créent de nouvelles formes littéraires. Ces analyses nous permettent de développer enfin la thématique de ce que nous appelons l’écriture organique, qui se compose et s’articule autour de la corporéité. Langage et corps se superposent dans une dynamique architecturale. Écrire laisse une trace. L’écriture engendre. La littérature serait alors le terrain fécond d’une renaissance, de l’écriture d’un homme nouveau, à jamais métamorphosé par l’expérience concentrationnaire. / This thesis of comparative literature aims to relate pieces inhabited by history and to question literary representations of the body in the face of the extreme hardship of confinement. The aim of this research, which unfolds in three parts, is to question human nature through the prism of writing when confronted with the traumatic experience of concentration camps and Nazi exterminations in the Second World War, by paralleling pieces, factual and fictional, which draw their ressources from both reality and fiction like interconnecting vessels. More specifically, as part of the first section we concentrate on the way the limit-experience of being manifests itself in these accounts. The confusion of identity and the dehumanization disrupt the representation of the body, thus impeaching it.This doubt fits into the language itself : how does one tell the unimaginable ? In the second section we focus on the inexpressible aspect of the event and reflect on the diversions, the displacements that literature can offer to say what, at first, seems indescribable. Imagery and symbolism create new forms of literature.This analysis allows us to develop the theme that we call organic writing, which is composed of and articulates itself through corporeity. Language and body superpose themselves in an architectural dynamic. Writing leaves a trace. Writing gives rise to new forms. Literature would therefore be the fertile soil of revival, the writing of a new human being, forever metamorphosed by the concentration camp experience.

An experimental study of phase transitions of long-chain n-alkanes induced by the effect of interfaces is described. ¶ The phase behaviour of long-chain n-alkanes (carbon number 14, 16, 17, 18) adsorbed at isolated mica surfaces and confined between two mica surfaces has been studied in the vicinity of and down to several degrees below the bulk melting points, Tm. Using the Surface Force Apparatus we have measured the thickness of alkane films adsorbed from vapour (0.97 [equal to or greater-than] p/p[subscript o] [equal to or greater-than] 0.997), studied capillary condensation transition, subsequent growth of capillary condensates between two surfaces, and phase transitions in both the adsorbed films and the condensates. By measuring the growth rate of the capillary condensates we have identified a transition in the lateral mobility of molecules in the adsorbed films on isolated mica surfaces. This transition to greater mobility occurs slightly above Tm for n-hexadecane, n-heptadecane and n-octadecane but several degrees below Tm for n-tetradecane, and is accompanied by a change in wetting behaviour and a measurable decrease in adsorbed film thickness for n-heptadecane and n-octadecane. Capillary condensates that form below Tm remain liquid, but may freeze if the degree of confinement is reduced by separation of the mica surfaces. An increase in the area of the liquid-vapour interface relative to that of the liquid-mica interface facilitates freezing in the case of the long-chain alkanes, which show surface freezing at the liquid-vapour interface. ¶ Although thermodynamic properties of the surface freezing transition have been rather well documented, the kinetics involved in formation of such ordered monolayers has so far received very little attention. We studied the surface tension of n-octadecane as a function of temperature in the vicinity of Tm, using the static Wilhelmy plate and the dynamic maximum bubble pressure methods. The two methods give different results on cooling paths, where nucleation of the surface ordered phase is involved, but agree on heating paths, where both methods measure properties of the equilibrium surface phase. On cooling paths, the surface of bubbles may supercool below the equilibrium surface freezing temperature. The onset of surface freezing is marked by a sharp drop in the surface tension. The transition is accompanied by an increased stability of the films resulting in longer bubble lifetimes at the liquid surface, which suggests that the mechanical properties of the surfaces change from liquid-like to solid-like. Our results suggest occurrence of supercooling of the monolayer itself.

In the design of rail track structures where the subgrade cannot achieve the desired capacity, enabling the required standard of track geometry to be maintained for the speed, axle load and tonnage to be hauled, a capping layer of granular material is placed between the natural ground or the embankment fill material and the ballast to protect the underlying weaker layers. In spite of the important role played by the capping layer, very little research has been carried out on its performance. The current practice of design of the capping layer, therefore, is based on working stress philosophy where reduced levels of stresses are assumed not to degrade the subgrade. Even on tracks containing a thick ballast layer that ensures allowable levels of working stress the subgrade has been found to have permanently deformed. Design of capping layers based on plastic deformation, therefore, appears appropriate. This thesis aims at determining the load levels that cause detrimental plastic deformation in the capping layer. The suite of material properties that characterise plastic deformations of capping layer is neither readily available nor easily determined. This thesis proposes a cheaper method of evaluating a range of capping layer material properties using penetration tests on specimens contained in California Bearing Ratio (CBR) test moulds coupled with a finite element modelling based back calculation technique. The suite of material properties thus determined are used for the simulation of the behaviour of capping layers under the boundary and loading conditions similar to those in practice. The predicted results are validated using laboratory experiments on large size capping layer specimens.

La tenue à long terme d'un colis de verre de déchets nucléaires dépend fortement des réactions chimiques opérant au niveau des surfaces au contact de l'eau. Les études entreprises sur des colis de verre inactif fracturés indiquent que le réseau de fissures participe significativement à la surface réactive. Néanmoins, l'interprétation des données expérimentales disponibles, dont la complexité est liée à la méconnaissance du réseau de fissures et des conditions locales d'altération, ne permet pas d'appréhender l'ensemble des mécanismes physico-chimiques mis en jeu.<br />Une meilleure compréhension de ces mécanismes passe par l'étude du couplage chimie transport dans des objets simples : les fissures modèles. L'altération de ces fissures dans des conditions agressives (pH>11) montre que la configuration de la fissure (horizontale ou verticale) impose le mécanisme de transport prépondérant (respectivement diffusion ou convection associée à la gravité). Ce mécanisme convectif semble négligeable à des pH plus faibles. La vitesse convective est estimée via un modèle 1D de transport réactif.<br /> Deux autres paramètres ont été étudiés : l'influence d'un gradient thermique et celle de l'interconnectivité des fissures sur l'altération. Une rétroaction forte de la vitesse convective, induite par un gradient thermique, sur la cinétique d'altération est observée au sein de la fissure.<br /> Ces travaux ont abouti à la réalisation d'une expérience intégrale d'altération d'un réseau de 163 fissures soumis à un gradient thermique. <br />L'utilisation du code géochimique HYTEC dans le cadre cette étude révèle les bonnes potentialités du logiciel cependant le modèle cinétique d'altération reste à améliorer.