This thesis examines the quantification of uncertainty in real-world engineering systems using the multilevel Monte Carlo method. It is often infeasible to use the traditional Monte Carlo method to investigate the impact of uncertainty because computationally it can be prohibitively expensive for complex systems. Therefore, the newer multilevel method is investigated and the cost of this method is analysed in the finite element framework. The Monte Carlo and multilevel Monte Carlo methods are compared for two prototypical examples: structural vibrations and buoyancy driven flows through porous media. In the first example, the impact of random mass density is quantified for structural vibration problems in several dimensions using the multilevel Monte Carlo method. Comparable eigenvalues and energy density approximations are found for the traditional Monte Carlo method and the multilevel Monte Carlo method, but for certain problems the expectation and variance of the quantities of interest can be computed over 100 times faster using the multilevel Monte Carlo method. It is also tractable to use the multilevel method for three dimensional structures, where the traditional Monte Carlo method is often prohibitively expensive. In the second example, the impact of uncertainty in buoyancy driven flows through porous media is quantified using the multilevel Monte Carlo method. Again, comparable results are obtained from the two methods for diffusion dominated flows and the multilevel method is orders of magnitude cheaper. The finite element models for this investigation are formulated carefully to ensure that spurious numerical artefacts are not added to the solution and are compared to an analytical model describing the long term sequestration of CO2 in the presence of a background flow. Additional cost reductions are achieved by solving the individual independent samples in parallel using the new podS library. This library schedules the Monte Carlo and multilevel Monte Carlo methods in parallel across different computer architectures for the two examples considered in this thesis. Nearly linear cost reductions are obtained as the number of processes is increased.

This thesis investigates three problems relating to thermoconvective stability in porous media. These are (i) the stability of an inclined boundary layer flow to vortex type instability, (ii) front propagation in the Darcy-B´enard problem and (iii) the onset of Prantdl-Darcy convection in a horizontal porous layer subject to a horizontal pressure gradient. The nonlinear, elliptic governing equations for the inclined boundary layer flow are discretised using finite differences and solved using an implicit, MultiGrid Full Approximation Scheme. In addition to the basic steady state three configurations are examined: (i) unforced disturbances, (ii) global forced disturbances, and (iii) leading edge forced disturbances. The unforced inclined boundary layer is shown to be convectively unstable to vortex-type instabilities. The forced vortex system is found to produce critical distances in good agreement with parabolic simulations. The speed of propagation and the pattern formed behind a propagating front in the Darcy-B´enard problem are examined using weakly nonlinear analysis and through numerical solution of the fully nonlinear governing equations for both two and three dimensional flows. The unifying theory of Ebert and van Saarloos (Ebert and van Saarloos (1998)) for pulled fronts is found to describe the behaviour well in two dimensions, but the situation in three dimensions is more complex with combinations of transverse and longitudinal rolls occurring. A linear perturbation analysis of the onset of Prandtl-Darcy convection in a horizontal porous layer subject to a horizontal pressure gradient indicates that the flow becomes more stable as the underlying flow increases, and that the wavelength of the most dangerous disturbances also increases with the strength of the underlying flow. Asymptotic analyses for small and large underlying flow and large Prandtl number are carried out and results compared to those of the linear perturbation analysis.

The concept of creating heterogeneous structures by nanocasting techniques from a combination of several homogeneous surfactant templated structures to model reservoir rock properties has not been approached prior to this research project, and will be used to test and provide better understanding of gas adsorption theories such as the pore blocking phenomenon (Seaton, 1991). Porous media with controlled pore sizes and geometry can be used to mimic a variety of reservoir rock structures, as it can be engineered to consist of a network of elements which, individually, could have either regular or irregular converging and diverging portions. The restrictions in these elements are called throats, and the bulges pores. Catalysts developed from a range of Nanotechnology applications could be used in down-hole catalytic upgrading of heavy oil. They could also be used as catalyst supports or to analyse the coking performance of catalysts. These studies will highlight the pore structure effects associated with capillary trapping mechanisms in rocks, and potentially allow the manipulation of transport rates of fluids within the pore structure of catalysts. Mercury-injection capillary pressure is typically favoured for geological applications such as inferring the size and sorting of pore throats. The difference between mercury injection and withdrawal curves will be used to provide information on recovery efficiency, and also to investigate pore level heterogeneity. Mercury porosimetry studies are carried out to provide a better understanding of the retraction curve and the mechanisms controlling the extrusion process and subsequently the entrapment of the non-wetting phase. The use of model porous media with controlled pore size and surface chemistry allows these two effects to be de-convolved and studied separately. The nanotechnology techniques employed mean that uncertainty regarding exact pore geometry is alleviated because tight control of pore geometry is possible. Trapping of oil and gas on a microscopic scale in a petroleum reservoir rock is affected by the geometric and topologic properties of the pores, by the properties of the fluids and by properties related to fluid-rock interaction such as wettability. Several distinct mechanisms of trapping may occur during displacement of one fluid by another in a porous media, however in strongly water-wet rocks with large aspect ratios, trapping in individual pores caused by associated restricting throats (may be/is) the most important mechanism of trapping. The results of the proposed research will be both relevant to the Irene Osagie Evbuomwan PhD. Thesis (2009) 9 oil and gas as well as the solid mineral sector for application as catalyst or catalyst supports. By providing a better understanding of the relationship between reservoir rock pore space geometry and surface chemistry on the residual oil levels, a more accurate assessment of the potential of a particular reservoir could be generated. The analysis of gas adsorption/desorption isotherms is widely used for the characterization of porous materials with regard to their surface area, pore size, pore size distribution and porosity, which is important for optimizing their use in many practical applications. Although nitrogen adsorption at liquid nitrogen temperature is considered to be the standard procedure, recent studies clearly reveal that the use of additional probe molecules (e.g. argon, butane, carbon dioxide, water, hydrogen, and hydrocarbons e.g. cyclohexane and ethane) allows not only to check for consistency, but also leads to a more comprehensive and accurate micro/mesopore size analysis of many adsorbents. Furthermore, significant progress has been achieved during recent years with regard to the understanding of the adsorption mechanism of fluids in materials with highly ordered pore structures (e.g., M41S materials, SBA-15). This has led to major improvements in the pore size analysis of nanoporous materials. However, there are still many open questions concerning the phase and sorption behaviour of fluids in more complex pore systems, such as materials of a heterogeneous nature/differing pore structures, which are of interest for practical applications in catalysis, separation, and adsorption. In order to address some of these open questions, we have performed systematic adsorption experiments on novel nanoporous materials with well defined pore structure synthesised within this research and also on commercial porous silicas. The results of this study and experiments allow understanding and separating in detail the influence of phenomena such as, pore blocking, advanced condensation and delayed condensation on adsorption hysteresis and consequently the shape of the adsorption isotherms. The consequences of these results for an accurate and comprehensive pore size analysis of nanomaterials consisting of more complex nanoporous pore networks are also investigated.

A one-dimensional porous media model has been developed to investigate water based fire suppression. The model is for heat and mass transfer in porous materials subjected to external water sprays and radiant heating. In the model, heat transfer inside the material occurs by conduction, convection, and phase change. Mass transfer occurs by gas phase diffusion and convection in the liquid and gas phases. Convective mass fluxes are driven by pressure gradients according to Darcy’s Law. Boundary conditions that are appropriate for a range of cases are presented. The model was used, along with experiments, to investigate two scenarios relevant to water based suppression: spray wetting and radiant heating. Ceramic fiberboard samples were used as a test material. For the wetting tests, the model is shown to be able to reasonably predict the rate of water absorption into the samples. Radiant heating tests were conducted in the cone calorimeter with pre-wetted samples. For the heating tests, the model is shown to reasonably predict the drying behavior that would directly precede an ignition event.

The primary aim of this thesis is to investigate the transition between different flow regimes in porous media. The complete transition spectrum of single-phase flow, from creeping flow to inertial, unsteady laminar, and turbulent flow regimes, was examined in sphere packings. Further understanding of the fundamental fluid dynamics was derived based on the pore-scale flow visualisation using magnetic resonance velocimetry (MRV). Spiral imaging was selected as the ultrafast imaging protocol to probe the transient phenomena, and the acquisition was further accelerated by combining subsampling and compressed sensing reconstruction. In a random sphere packing column with column-to-diameter ratio of 3.44, the inertial effect and the onset of unsteady regime were examined with respect to the principal flow characteristics: the inertial core/channeling, backflow, and helical vortices. Helical vortices have been observed experimentally in a random packing for the first time, and the analogy between the swirling flow and helical vortices provides insight into the design and operation of packed bed reactors. Another new observation is that the transition to the unsteady regime is a highly heterogeneous process, where the evolution of the flow instability depends on the pore geometry. Moreover, pixelwise validation was achieved between the experimental and simulation results on three-dimensional velocity fields in the inertial regime; this is enabled by an image-based meshing pipeline, which reproduces the geometry of the random packing in MRV for the numerical simulation. The unsteady regimes were further investigated using a regular sphere packing, the simple cubic packing (SCP). The spectral analysis, in both the random and regular packing, revealed a route to chaos from the steady to periodic, quasi-periodic, and chaotic dynamics, which was only predicted numerically before. During the transition to turbulence, the coherent structures were extracted using proper orthogonal decomposition (POD), which yields a coherent picture regarding the turbulent dynamics, when combined with the skewness, flatness, and quadrant analysis. Furthermore, it was found that the macroscopic properties converged at lower Reynolds number than the microscopic features. In conclusion, the opportunity to measure flow fields at high spatial and temporal resolution will play an increasingly significant role in the advancement of fundamental fluid dynamics. In this thesis, MRV is used, which is particularly advantageous for non-invasive measurements in opaque systems. This thesis provides the experimental and analysis toolkits for such studies and has demonstrated the contribution to characterising and understanding different flow regimes in porous media.

Cette thèse vise à caractériser l’influence d’une phase adsorbée sur les déformations instantanées en milieu microporeux qui se distinguent par la présence de pores de diamètres inférieurs à 2 nanomètres. Les charbons, activés ou naturels, les roches à faibles perméabilités, les argiles et de nombreux biomatériaux rentrent dans cette catégorie. Par la présence d'une grande surface spécifique (surface développée du milieu poreux, de l’ordre de 100 à 1000 m2/g), les matériaux microporeux peuvent piéger, sous forme adsorbée, une grande quantité de molécules de fluide avec des applications dédiées à la récupération d’hydrocarbure, au stockage géologique, à la séparation, la catalyse ou au transport de médicament. Pour ces matériaux microporeux, une déviation de la poromécanique classique, introduite par Biot, est attendue. Dans les plus petits pores, de taille nanométrique, ces molécules de fluide se trouvent fortement confinées et leurs interactions sont modifiées. Cet effet a comme conséquence un gonflement macroscopique du milieu poreux. Le gonflement in situ induit par adsorption dans les charbons naturels a été identifié comme étant la source principale de la chute d’injectivité de CO2 lors de la production assistée de CH4 dans des veines de charbon. Les matériaux poreux naturels et de synthèse sont généralement composés d’une double porosité: la microporosité où le fluide est piégé sous forme adsorbée et une méso ou macroporosité nécessaire pour assurer le transport de fluide vers les nanopores. Le gonflement induit par adsorption de la matrice referme la porosité de transport, réduit la perméabilité globale du système poreux. Ce travail présente un nouveau modèle poromécanique étendu permettant de prédire les gonflements induits par adsorption de gaz en tenant compte de la variation incrémentale de la porosité au cours du gonflement. Ce modèle est tout d’abord développé dans le cas de milieux poreux purement microporeux, homogènes et isotropes, saturés par un fluide en conditions isothermes et réversibles. Le modèle est ensuite étendu au cas d’un milieu poreux présentant une double porosité, c’est à dire une porosité d'adsorption et une porosité de transport. Il présente l’originalité de n’avoir aucun paramètre à identifier au cours du trajet de chargement. Les caractéristiques poromécaniques initiales et les quantités adsorbées sont les uniques paramètres du modèle. Dans cette étude, un développement d'une nouvelle technique expérimentale de mesures simultanées de quantités adsorbées et de déformations induites par adsorption est réalisé. Le nouveau banc d’essai met en jeu une technique manométrique de mesure d’isotherme d’adsorption couplée à une mesure de déformations en plein champ par corrélation d’images numériques. Après une étude complète des incertitudes de mesure, la montage est validé en comparant avec des mesures de quantités adsorbées de CH4 et de CO2 sur un charbon actif de référence obtenues par la technique gravimétrique. Le banc d’essai développé est ensuite utilisé pour tester le nouveau modèle poromécanique sur un charbon actif. Ce matériau présente la particularité de posséder à la fois une microporosité et une macroporosité importantes. Les mesures et les estimations de déformations induites par adsorption sont confrontées. Un très bon accord est obtenu, validant le modèle développé. Une étude d’un charbon naturel issu d’un réservoir naturel est enfin réalisée. L’obtention de cartes de déformation en plein champ permet de visualiser directement l'évolution du réseau de failles naturelles lors de l'adsorption de gaz et d’isoler des zones homogènes entre ces failles où le modèle développé peut s’appliquer. Là encore, une comparaison modèle-expérience présente un bon accord. / This research study aims at characterizing the influence of an adsorbed phase on the instantaneous deformations in microporous media (width pores < 2nm). Activated or natural carbons, tight rocks, clay, cementitious materials and numerous biomaterials such as bones are among these materials. In recent years, a major attention has been paid on these microporous materials because the surface-to-volume ratio (i.e., the specific pore surface) increases with decreasing characteristic pore size. These materials can trap an important quantity of fluid molecules as an adsorbed phase. This is important for applications in petroleum and oil recovery, gas storage, separation. For these microporous materials, a deviation from standard poromechanics, which was introduced by Biot 75 years ago, is expected. In very small pores, the molecules of fluid are confined. Interaction between molecules is modified. This effect, denoted as molecular packing, includes fluid-fluid and fluid-solid interactions and has significant consequences at the macroscale, such as instantaneous swelling. In-situ adsorption-induced coal swelling has been identified as the principal factor leading to a rapid decrease in CO2 injectivity during coal bed methane production enhanced by CO2 injection. Generally, natural and synthesised porous media are composed of a double porosity: the microporosity where the fluid is trapped as an adsorbed phase and a meso or a macro porosity required to ensure the transport of fluids to and from the smaller pores. If adsorption in nanopores induces instantaneous deformations at a higher scale, the matrix swelling may close the transport porosity, reducing the global permeability of the porous system or annihilating the functionality of synthesised materials. The point of view of poromechanic is a mean to understand this problematic. When a porous media is immersed in the gas, the classical poromechanic predicts the shrinkage of the media. This formalism have to extend to take account the adsorption phenomenon. A new poromechanical framework allowing adsorption induced strain predictions by taking into account the incremental variations of porosity upon swelling for pure microporous, isotropic and homogeneous materials saturated by a single fluid in reversible and isothermal conditions was developped. In this study, the model is then extended for double porous media presenting both a microporosity where the fluid is trapped as an adsorbed phase and a transport macroporosity. The porosities are defined by the classification of l'Internationnal Union of Pure and Applied Chemistry. In this study, a new experimental procedure is built for simultaneous measurement of both adsorption and swelling quantities by respectively a manometric technique and a full-field digital image correlation technique. The experimental procedure is applied to test the reliability of the proposed model for a commercial active carbon saturated with pure CH4 and pure CO2. The material has the particularity to present both a high microporosity and a high macroporosity. A good agreement is obtained in term of adsorption induced swelling quantities. Once the model validated, it is used to quantify the decrease of transport macroporosity induced by microporous matrix swelling. The last part is dedicated to the study of a natural carbon extracted from a mine with enhanced coal bed methane recovery potentialities. The material has the particularity to present three different porosities: a microporosity where the gas is trapped, a transport macroporosity and a natural cleat network also participating in the global transport. The obtaining of full-field displacement maps provides insight of the cleat network influence and helps to isolate homogeneous zones where the poromechanical model may be applied and compared to the experimental results. Here again, a good agreement is obtained in term of adsorption-induced swelling quantities.

Physics of immiscible two-phase flow in porous media is relevant for various industrial and environmental applications. Wettability defined as the relative affinity of fluids with the solid surface has a significant impact on the dynamics of immiscible displacement. Although wettability effects on the macroscopic fluid flow behaviour are well known, there is a lack of pore-scale understanding. Considering the crucial role of wettability in a diverse range of applications; this research aims to provide a pore-scale picture of interface configuration induced by variations in the wetting characteristics of porous media. Besides, this study also relates the pore-scale interfacial phenomena with the macroscopic response of fluids. High-resolution direct numerical simulations (DNS) at multiscale (single capillary and a highly heterogeneous porous media) were performed using computational fluid dynamics (CFD) approach in which the Navier-Stokes equation coupled with the volume of fluid method is solved to represent immiscible displacement. Numerical results demonstrate that at pore scale as the wettability of porous media changes from strong to intermediate wet the effects of pore geometry (that includes corner angle and orientation angle) on the interfacial dynamics also enhances. This was demonstrated by the non-monotonic behaviour of entry capillary pressure at the junction of pore, curvature reversal in the converging-diverging capillary and the co-existence of concave and convex interfaces in heterogeneous porous media with uniform contact angle distribution. In addition to simulations, theoretical argument is also presented that rationalize the underlying physics of complex, yet intriguing interfacial phenomena shown by DNS. Overall this research extends the fundamental understanding of multiphase flow in porous media and paves the way for future studies on porous media.

L’eau est le principal vecteur responsable de la dégradation du patrimoine bâti. L’un des moyens utilisé, pour réduire la pénétration de l’eau dans la pierre est la bioremédiation. Il se base sur la capacité des bactéries à former des biocristaux. Ce concept a été utilisé par la société Calcite Bioconcept avec la souche Bacillus cereus. Par le passé, l’action bactérienne et, en particulier, la possibilité de créer des cristaux a été étudiée et mise en évidence en milieu aqueux, très favorable à la croissance bactérienne. L’originalité de cette thèse est de reprendre ces études en suivant le même protocole qui celui utilisé in situ et d’étudier le résultat de la bioremédiation sous différents aspects physique, chimique et biologique. Différents supports ont été adoptés, des tuffeaux (pierre calcaire pour lequel le biotraitement a été mis au point) et du plâtre. Les images MEB, sur des échantillons fracturés, montrent la formation d’un biofilm à la seule surface du matériau. Aussi des méthodes d’investigation de surface ont été menées (cathodoluminescence, DRX en incidence rasante). Cette dernière technique confirme la formation de calcite à la surface des échantillons de plâtre. Les cycles d’imbibition et des mesures de perméabilité montrent la réduction des coefficients de transport (ce qui est un effet attendu et souhaité du traitement). Cependant, l’augmentation de la vitesse d’imbibition au fur et à mesure des cyclages laisse penser que le biofilm se dégrade rapidement. Enfin, une alternative au protocole industriel, utilisé précédemment, est proposée, non plus basée sur les capacités de la membrane des bactéries à synthétiser des cristaux de calcite, mais par leurs EPS. / Degradation of historical buildings is mainly due to water intrusion that is the main vector ofpollutants. Differents types of surface treatment were proposed to avoid or limit this effect. As analternative of chemical treatment, the use of the carbonatogenesis property of some bacteria wasproposed. In the past, The bacterial production was pointed out on concrete and on limestonesamples in an aqueous environment. Moreover, the carbonate production was indirectly measuredand the experimental protocol was far from real use conditions.In this thesis, we follow the same protocol as an industrial one used in situ and study with chemical,physical as well as biological approaches the bioremediation effects on different porous samples:limestone (for which the treatment has been optimized) and plastrum.We confirm that, indeed, the bioremediation results in a thin biocoating (around 10 micrometer depth)formed on the surface of the treated samples. So, different surface analyses have been performed,on micro scale (GIXD, electronic microprobe), on macro scale (cathodoluminescence).The formershows the structural and morphological evolution of the produced carbonate coating and that indeedcalcite is produced by the treatment on model plaster samples. This substrate was chosen in order tounambiguously point out the bacterial carbonate production.The latter confirms the presence ofcalcite and the homogeneity of the coating at this scale. The reduction and durability of the biofilmhas been evaluated by imbibition cycles as well as permeability measurements.The biotreatment presents some drawbacks that could be avoided using the EPS only. Differentscrystals have been synthetized using organic mater. The influence of this organic matter on thecrystals shape and size has been studied.

Orientador: Jose Luis Lage, Luiz Fernando Milanez / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Mecanica / Made available in DSpace on 2018-11-08T18:36:48Z (GMT). No. of bitstreams: 1 Junqueira_SilvioLuizdeMello_D.pdf: 23366873 bytes, checksum: 40ff350275cdc60184b7883f90658798 (MD5) Previous issue date: 1996 / Resumo: O presente trabalho trata do estudo teórico e experimental dos efeitos térmicos causados pela aplicação da radiação laser sobre meios porosos e objetiva a determinação do coeficiente de atenuação de um meio poroso imerso em fluido. O modelamento matemático proposto utiliza a técnica dos volumes finitos, para resolver numericamente a equação do transporte do calor em coordenadas.cilíndricas. Um sistema de aquisição de dados baseado no conceito de instrumentos virtuais é elaborado para analisar o processo de aquecimento pela radiação de um laser de Argônio sobre uma matriz porosa de Alumínio 6101 imersa em quatro fluidos diferentes: ar, água, óleo polialfaolefina e Mercúrio. Uma metodologia inversa, estabelecida pela comparação de resultados numéricos e experimentais, é empregada para obter o coeficiente de atenuação da espuma metálica saturada. A análise inclui o emprego de duas metodologias de cálculo da condutividade térmica equivalente. Uma correlação entre os coeficientes de atenuação e um número de Prandtl equivalente do meio poroso é estabelecida / Abstract: In the present work, an investigation of thermal effects due to laser application is accomplished in order to determine the attenuation coeflicient of a porous matrix immersed in fluido The study included both theoretical and experimental analisys. In the theoretical analisys a numerical mo deI based on control volume method is developed to simulate the lasing process by solving the energy equation in cilindrical coordinates. A data acquisition system based on virtual instruments concept is elaborated to analyse the heating process result,ing from Argon Laser radiation over an Aluminum foam porous medium immersed in four different fluids, namely air, water, polyalphaolefin oil and Mercury. An inverse methodology, estabilished by comparison of numerical and experimental results, is used to obtain the attenuation coeflicient of the saturated metal foam. Calculations also included the use of two models for the effective thermal conductivity. Results indicated the existence of a linear correlation between laser attenuation and a defined equivalent Prandtl number / Doutorado / Termica e Fluidos / Doutor em Engenharia Mecânica

A medida de tempo de relaxação longitudinal (T1) e transversal (T2) de ressonância magnética nuclear (RMN) é muito utilizada para análise de meios porosos na área do petróleo. Esses tempos de relaxação estão relacionados com tamanho dos poros. As cerâmicas, dependendo da sua temperatura de sinterização, apresentam uma porosidade intrínseca, mas cerâmicas especiais para processo de filtragem e escafoldes de implantes ósseos necessitam de poros maiores e de permeabilidade para suas aplicações. Esse objetivo é alcançado introduzindo um agente porogênico, um material degradável com a temperatura, obtendo assim poros induzidos além dos intrínsecos. O agente porogênico escolhido para este trabalho foi cristais de sacarose, os tamanhos dos poros são controláveis pelo tamanho dos cristais e pela quantidade, tornado esse um meio factível nos estudos de RMN através dos tempos de relaxação. Foram selecionadas, através de peneiras, duas faixas de tamanhos de cristais de sacarose: G (grande) com limite superior de 600&mu;m e inferior de 300&mu;m e P (pequeno) com tamanhos na faixa de 38&mu;m a 150&mu;m. A cerâmica preparada com P apresentou distribuição de tamanho de poros (de 15 à 105&mu;m) com mediana em 40&mu;m, menor que a fabricada com G (de 20 à 125&mu;m), media em 100&mu;m. Distribuições de tamanho de poros com mediana intermediária foram obtidas pelas misturas de P e G na fabricação. Foram preparadas seis cerâmicas para os estudos: pura (sem adição de agente porogênico), 100% com G, 100% com P, mistura de 50% P e 50% G, mistura de 75% P e 25% G e mistura de 25% P e 75% G. As cerâmicas foram saturadas com água por um sistema de vácuo e através da sequência CPMG obtiveram-se os decaimentos multiexponenciais da relaxação transversal para os núcleos de &sup1;H de moléculas de água confinadas nos poros das cerâmicas. E assim, determinaram-se as distribuições de tempo de relaxação T2. A cerâmica pura apresentou apenas um pico para T2 em 0,053s, assim como obtido por Borgia e colaboradores em outros estudos, e comparando com resultados de porosimetria por intrusão de mercúrio (PIM), determinou-se o coeficiente de relaxatividade superficial de 3,7&mu;m.s-1, valor próximo ao tabelado para a alumina (3&mu;m.s-1) dando confiabilidade aos equipamentos utilizados. As cerâmicas preparadas com agente porogênico apresentaram dois picos na distribuição de T2 com o primeiro localizado na mesma região de tempo que corresponde à porosidade intrínseca (0,053s) e o segundo pico, referente à porosidade induzida, deslocando-se começando com a 100P com o menor tempo T2 (0,38s) e aumentando conforme o aumento da porcentagem de grânulo G (1,02s para 100G) de acordo com era esperado. A parti das imagens de microscopia eletrônica de varredura (MEV), obteve-se as distribuições de tamanho de poros induzidos e, novamente, determinou-se o coeficiente de relaxatividade, 12&mu;m.s-1, valor maior comparado com o anterior devida as impurezas e resíduos da queima dos cristais de sacarose deixado na superfície do poro. Por fim, concluiu-se então que a RMN foi capaz de fazer a análise das cerâmicas de alumina observando poros intrínsecos da ordem de 0,1&mu;m e poros induzidos com tamanho de dezenas a centenas de micrometros. Foram reunidos os resultados necessários para compreender como são formados os poros a partir do agente porogênico podendo fabricar cerâmicas com distribuições de tamanhos de poros conforme desejado. / The measurement of longitudinal (T1) and transversal (T2) nuclear relaxation times is an important method for the analysis of porous media, much used in the petroleum industry. It is possible through this technique the determination of the porosity and pore size distribution of the media. Ceramic materials already have an intrinsic porosity dependent on its sinterization temperature, but some applications such as filtering or scaffolding in bone regeneration demand for a larger porosity and permeability. To achieve this goal, a porogenic agent is introduced in the ceramics production process which is degradable in high temperatures, producing a ceramic with both induced and intrinsic porosity. The porogenic agent chosen for this project were saccharose crystals and the pores size is controlled by the crystals size and by the quantity added in the production process. Two ranges of the saccharose crystals diameters were selected using different sieves: G size, raging from 300&mu;m to 600&mu;m and P size ranging from 38&mu;m to 150&mu;m. The P size ceramic presented pore size distribution ranging from 15 to 105&mu;m, with median in 40&mu;m, smaller than the median found for the ceramic produced with G size saccharose crystals (100&mu;m), for which the porosity ranged from 20 to 125&mu;m. Pore size distribution with median between the results found in P and G ceramics were acquired by using a mixture of P and G saccharose crystals sizes in the ceramic fabrication. Six samples were prepared for this study: pure (with no addition of porogenic agent), 100% G size, 100% P size; mixtures of 50% P and 50% G sizes, 75% P and 25% G sizes, and finally 25% P and 75% G sizes. All samples were saturated with water in a vacuum bomb, and, using the CPMG pulse sequence, the multiexponencial decay of the transverse relaxation time was obtained for the &sup1;H nuclei of the water molecules confined in the ceramic pores. Through this measurement, the distribution of T2 of the sample was determined. The curve for the pure ceramic presented only one peak at 0,053s, the same result obtained by Borgia at al. in other studies. Comparing this result with the porosity determined by mercury intrusion porosimetry, it was possible to determine the surface relaxation coefficient (3,7&mu;m.s-1), with a good proximity to the value found in literature for alumina (3&mu;m.s-1), showing that the equipment used for the measurements is reliable. The ceramic fabricated with the porogenic agent presented two peaks in the T2 time distribution, with the first peak coincident with the T2 value found for the intrinsic porosity (0,053s) and the second peak, which represents the induced porosity, varying to each ceramic sample, with the 100 % P ceramic presenting the peak with the smallest T2 (0,38s) and moving to higher T2 values as the percentage of G size saccharose crystals in the ceramic increases, (1,02s for the sample with 100% G), as it was expected. Through images from scanning electron microscopy, an assessment of the pore size distribution was made and the surface relaxation coefficient recalculated, arriving at a higher value than the one found for the pure ceramic (12&mu;m.s-1), which shows that residues of the saccharose crystals are left in the pores surface in the burning process. In conclusion, the NMR technique proved capable to characterize the ceramic materials, allowing the observation of both the intrinsic (of magnetude 0,1&mu;m) and the artificial induced porosity (varying from tens to hundreds of micrometers) of the alumina ceramics. The union of all the results of this project encloses the knowledge necessary to comprehend how the pores are formed by the action of the porogenic agent, making it possible to develop a manufacture process of ceramics with controlled pores sizes.