This thesis concerns modelling of Coulomb collisions in toroidal plasma with Monte Carlo operators, which is important for many applications such as heating, current drive and collisional transport in fusion plasmas. Collisions relax the distribution functions towards local isotropic ones and transfer power to the background species when they are perturbed e.g. by wave-particle interactions or injected beams. The evolution of the distribution function in phase space, due to the Coulomb scattering on background ions and electrons and the interaction with RF waves, can be obtained by solving a Fokker-Planck equation.The coupling between spatial and velocity coordinates in toroidal plasmas correlates the spatial diffusion with the pitch angle scattering by Coulomb collisions. In many applications the diffusion coefficients go to zero at the boundaries or in a part of the domain, which makes the SDE singular. To solve such SDEs or equivalent diffusion equations with Monte Carlo methods, we have proposed a new method, the hybrid method, as well as an adaptive method, which selects locally the faster method from the drift and diffusion coefficients. The proposed methods significantly reduce the computational efforts and improves the convergence. The radial diffusion changes rapidly when crossing the trapped-passing boundary creating a boundary layer. To solve this problem two methods are proposed. The first one is to use a non-standard drift term in the Monte Carlo equation. The second is to symmetrize the flux across the trapped passing boundary. Because of the coupling between the spatial and velocity coordinates drift terms associated with radial gradients in density, temperature and fraction of the trapped particles appear. In addition an extra drift term has been included to relax the density profile to a prescribed one. A simplified RF-operator in combination with the collision operator has been used to study the relaxation of a heated distribution function. Due to RF-heating the density of thermal ions is reduced by the formation of a high-energy tail in the distribution function. The Coulomb collisions tries to restore the density profile and thus generates an inward diffusion of thermal ions that results in a peaking of the total density profile of resonant ions. / <p>QC 20130415</p>

Axonal degeneration is a major pathological feature of the central nervous system (CNS) inflammatory demyelinating disease multiple sclerosis (MS). This axonal degeneration has major consequences, as functional axonal regeneration in the CNS is largely absent. Cumulative axonal degeneration is the likely cause of the majority of progressive MS-related disability, and therefore, the need for novel neuroprotective therapies for MS exists. Experimental autoimmune encephalomyelitis (EAE), an animal model of MS pathology, also produces axonal injury. In particular, the optic nerve and spinal cord are key sites of neuroinflammation in mouse EAE. By utilizing this model, the short term and long term effects of the putative neuroprotective cytokine, leukaemia inhibitory factor (LIF), were investigated in the optic nerve and spinal cord utilising a number of outcome measures of axonal dysfunction. These included MRI measures of water diffusivity along (ADC ||) and across (ADC┴) the optic nerves, serum levels of phosphorylated neurofilament heavy chain subunit (pNF-H) and histological morphometric measures. LIF treatment reduced EAE grade and pNF-H plasma levels, decreased ADC┴, but had no effect on ADC ||, axon counts or inflammatory infiltration. / In contrast, genetic deletion of LIF and its sister cytokine ciliary neurotrophic factor (CNTF), not only increased EAE grade and pNF-H levels, but also decreased optic nerve ADC|| and optic nerve and spinal cord axon densities. After reviewing current literature, we hypothesize that the target cell for endogenously upregulated LIF in EAE may be the neuron or axon, whereas the target cell for exogenously administered therapeutic LIF may be another cell type, possibly infiltrating macrophages and activated microglial cells. LIF antagonist treatment did not have any affect on EAE grade, pNF-H levels or MRI parameters. This lack of effect may be due to the inability of the LIF antagonist to enter the CNS, supporting the hypothesis that endogenous LIF has a centrally acting mechanism.

The A15-based Nb3Sn and V3Ga superconducting compounds are an integral part of synchrotrons and magnetic fusion reactor technology, especially where a magnetic field higher than 10 T is required, which lies beyond the limit of conventional Nb-Ti superconductors (~8 T). These brittle intermetallic compounds are difficult to manufacture in the form of wires, required for the application purpose, using the traditional wire-drawing process. Hence, bronze technique is adopted to fabricate such filamentary wires. This is based on the solid-state diffusion where A3B compound (A=Nb or V, B=Sn or Ga) forms during the interaction of Cu(B) and A. The operation of pure superconducting wires gets restricted to the field of 12 T, however, the ever-increasing demands for an improved efficiency have promoted the development of these A15 wires with the addition of alloying elements such as Ti and Zr. Many important physical and mechanical properties of such wires depend on the growth behaviour of these compounds. Therefore, understanding the growth of such compounds necessitates an in-depth analysis on diffusion behaviour of various elements in both bronze-based solid solutions as well as A15-intermetallics. Estimation of diffusion parameters makes use of the most commonly used diffusion couple technique. There are mainly three methods available for the estimation of the interdiffusion coefficients, proposed by Matano-Boltzmann (MB), Den Broeder (dB), same as Sauer-Freise (SF) and Wagner. Among these three, MB treatment is known to be the least accurate method, especially when there is a deviation of molar volume in a system from the ideality. At the same time molar volume might affect the estimation process differently for dB and Wagner’s approach. MB method is still being used neglecting the actual molar volume variation. On the other hand, the implementation of dB or Wagner’s approach for the estimation remains to be random. For the first time, we have critically examined the role of molar volume on estimated diffusion parameters and indicated the more accurate approach. Similar analysis for the estimation of the intrinsic diffusion coefficient is conducted considering Heumann and van Loo’s methods. Furthermore, the discussion is extended to the estimations of various diffusion parameters considering the measured composition profile in the V-Ga system. A detailed diffusion study has been conducted on Cu(Ga) and Cu(Sn) solid solutions to examine the role of the vacancy wind effect on interdiffusion. The interdiffusion, intrinsic and impurity diffusion coefficients are determined to facilitate the discussion. It is found that Ga and Sn are the faster diffusing species in the respective systems. The trend of the interdiffusion coefficients is explained with the help of the driving force. Following that, the tracer diffusion coefficients of the species are calculated with and without consideration of the vacancy wind effect. We found that the role of the vacancy wind is negligible on the minor element in a dilute solid solution, which is the faster diffusing species in this system and controls the interdiffusion process. However, consideration of this effect is important to understand the diffusion rate of the major element, which is the slower diffusing species in this system. Major drawback of studying diffusion in multi-component systems is the lack of suitable techniques to estimate the diffusion parameters. In this study, a generalized treatment to determine the intrinsic diffusion coefficients in multi-component systems is developed utilizing the concept of pseudo-binary approach. This is explained with the help of experimentally developed diffusion profile in the Cu(Sn, Ga) solid solution. Based on an interdiffusion study using an incremental diffusion couple in the V-Ga binary system, we have shown that V diffuses via lattice, whereas Ga does so via grain boundaries for the growth of the V3Ga phase. We could estimate the contributions from two different mechanisms, which are, usually, difficult to delineate in an interdiffusion study. Available tracer diffusion studies and the atomic arrangement in the crystal structure have been considered for a discussion on the diffusion mechanisms. Diffusion–controlled growth rate of V3Ga at the Cu(Ga)/V changes dramatically because of a small change in Ga content in Cu(Ga). One atomic percent increase in Ga leads to more than double the product phase layer thickness and a significant decrease in activation energy. Kirkendall marker experiment indicates that V3Ga grows because of diffusion of Ga. Role of different factors influencing the diffusion rate of Ga and high growth rate of V3Ga are discussed. The growth of Nb3Sn by bronze technique on two different single crystals and deformed Nb is studied. The grain boundary diffusion-controlled growth rate is found to be different for each of these three specimens. The difference is explained on the basis of the grain size of Nb3Sn. Elemental additions such as Ti and Zr to either bronze or metal are found to improve the superconducting properties. We have examined their effects on the growth rates of A15-phase formed in Cu(B,x)/A and Cu(B)/(A,x), where x is Ti or Zr. In either cases Ti and Zr-additions result in an improved growth rate of the product phase and reduces activation energy with increase in alloying addition; however few precipitates are formed in the interdiffusion zone for Cu(B,x)/A. Wavelength dispersive spectrometry (WDS)-mapping reveals these to be x-rich. Scanning transmission electron microscopy (STEM)-analysis suggests having composition gradient inside a single precipitate. TEM-diffraction demonstrates these to be Ti(A) solid solution crystallizing as BCC-structure for Cu(B,Ti)/A. These are located on grain boundaries of A15-phase. Electron back-scattered diffraction (EBSD)-analysis demonstrates grain morphology of product phase and found the average grain size to exhibit a decreasing trend with increasing x content. Columnar grains, on Ti and Zr addition tend to form as equiaxed ones. Based on the morphology and grain size pattern, the role of grain boundary diffusion is speculated to have a dominant effect with increase in elemental additions. The texture evolution of the product phase is also investigated and found the product phase to grow as a strongly textured one with the elemental additions. A peculiar pattern is observed for the texture of the product phase and its adjacent A or A(x) grains.

<div>To compensate for their sessile existence, plants synthesize and emit a wide diversity of volatile organic compounds (VOCs) that serve important biological functions pertaining to defense, reproduction, and plant-plant signaling. In addition to their importance in plant secondary metabolism, VOCs are used as fragrances, flavoring agents, and therapeutics. Plant metabolic engineering has successfully been implemented towards the design of value-added plants with enhanced defense, improved aroma and flavor, and increased production of specialty chemicals. However, rational design requires rigorous characterization of the mechanisms controlling metabolic fluxes in a network. Thus, the major aims of this dissertation are to study biological and physical mechanisms controlling the synthesis and emission of plant VOCs. This dissertation focuses on (i) modeling 2-phenylethanol biosynthesis in Arabidopsis and (ii) characterization of the biophysical properties of flower cuticles with respect to the emission of VOCs.</div><div><br></div><div>2-Phenylethanol (2-PE) is a naturally-occurring aromatic volatile with properties that make it a candidate oxygenate for petroleum-derived gasoline. In plants, 2-PE biosynthesis competes with the phenylpropanoid pathway for the common precursor L-phenylalanine (Phe). The phenylpropanoid pathway directs up to 30% of fixed carbon towards the production of lignin, a major constituent of plant cell walls that renders biomass recalcitrant to pretreatment techniques impeding the economical production of biofuels. An initial genetic engineering approach was proposed, whereby a portion of the carbon flux towards lignin production is diverted towards the biosynthesis 2-PE. Transgenic Arabidopsis thaliana expressing enzymes catalyzing the biosynthetic steps from Phe to 2-PE were generated. Excised stems from transgenic Arabidopsis were supplied 13C6-ring labeled Phe, and isotopic enrichment of downstream metabolites were quantified to calculate fluxes. By combining flux measurements with predictions from a kinetic model of the Phe metabolic network, we hypothesized that 2-PE biosynthesis in transgenic Arabidopsis was limited by endogenous pools of cytosolic Phe. Multiple independent genetic strategies were proposed based on model-guided predictions, such as inducing Phe hyper-accumulation, reduction of the activity of the competing phenylpropanoid pathway, and sequestering the 2-PE biosynthesis pathway in plastids. Combining kinetic modeling with time-course in vivo metabolomics led to successful rational engineering of 2-PE accumulating plants.</div><div><br></div><div>The plant cuticle is the physical interface between the flower and its surrounding environment. Passage of VOCs through the cuticle is driven solely by diffusion and is thus dependent on the cuticle physicochemical properties. Wax compounds in the cuticular matrix self-assemble into a multiphase system of crystalline and amorphous regions, where their relative amounts and arrangements govern VOC diffusion. To investigate the effect of wax composition on the crystallinity and permeability of the cuticle, we characterized the cuticular waxes of Petunia hybrida petals using GC-MS, FTIR, DSC, and XRD. Petal waxes were found to be enriched with long-chain hydrocarbons forming semi-crystalline waxes localized on petal surfaces. A ternary system of wax compounds was proposed as a model for petal cuticles to investigate the effect of wax composition on cuticle crystallinity and permeability. Atomistic simulations of VOC displacement in waxes of varying chemical composition were performed at 298 K and 1 bar under NPT conditions to estimate diffusivities. Wax anisotropy was found to be highly dependent on the elongation of methylene chains, restricting the molecular diffusion path. Changes in crystalline symmetry were found to have measurable effects on VOC diffusion. Simulations of compositional variants of the model cuticle shows that changes in relative crystallinity exert differential control on the dynamics of VOC emissions.</div><div><br></div><div>To directly determine the effect of the cuticle on VOC emissions in petunia flowers, the wax exporter PhABCG12 was silenced using RNA interference, resulting in flowers with thinner cuticles. However, VOC emissions were found to have significantly decreased in transgenic flowers relative to the wild-type control. Dewaxing wild-type and transgenic petunia revealed that the cuticle serves as a site of VOC build-up during emission, and deficient coverage limits the extent to which compounds can accumulate. In addition, the cuticle was found to impart differing levels of mass transfer resistance for certain VOCs, suggesting that the cuticle controls the dynamics of VOC emissions. Taken together, petal cuticles provide an additional layer of regulation in emission of VOCs from plants.</div><div><br></div>

No description available.

Dans ce mémoire nous allons présenter une méthode de diffusion d’erreur originale qui peut reconstruire des images en demi-ton qui plaisent à l’œil. Cette méthode préserve des détails fins et des structures visuellement identifiables présentes dans l’image originale. Nous allons tout d’abord présenter et analyser quelques travaux précédents afin de montrer certains problèmes principaux du rendu en demi-ton, et nous allons expliquer pourquoi nous avons décidé d’utiliser un algorithme de diffusion d’erreur pour résoudre ces problèmes. Puis nous allons présenter la méthode proposée qui est conceptuellement simple et efficace. L’image originale est analysée, et son contenu fréquentiel est détecté. Les composantes principales du contenu fréquentiel (la fréquence, l’orientation et le contraste) sont utilisées comme des indices dans un tableau de recherche afin de modifier la méthode de diffusion d’erreur standard. Le tableau de recherche est établi dans un étape de pré-calcul et la modification est composée par la modulation de seuil et la variation des coefficients de diffusion. Ensuite le système en entier est calibré de façon à ce que ces images reconstruites soient visuellement proches d’images originales (des aplats d’intensité constante, des aplats contenant des ondes sinusoïdales avec des fréquences, des orientations et des constrastes différents). Finalement nous allons comparer et analyser des résultats obtenus par la méthode proposée et des travaux précédents, et démontrer que la méthode proposée est capable de reconstruire des images en demi-ton de haute qualité (qui préservent des structures) avec un traitement de temps très faible. / In this work we present an original error-diffusion method which produces visually pleasant halftone images while preserving fine details and visually identifiable structures present in original images. We first present and analyze the previous work to show the major problems in halftoning, and explain why we decided to use an error diffusion algorithm to solve the problems. Then we present our method which is conceptually simple and computationally efficient. The source image is analyzed, and its local frequency content is detected. The main components of the frequency content (main frequency, orientation, and contrast) serve as lookup table indices in a pre-computed database of modifications to a standard error diffusion. The modifications comprise threshold modulation and variation of error-diffusion coefficients. The whole system is calibrated in such a way that the produced halftone images are visually close to original images (patches of constant intensity, patches containing sinu- soidal waves of different frequencies/orientations/contrasts, as well as natural images of different origins). Finally, we compare and analyze the results obtained by our method and previous work, and show that our method can produre high-quality halftone image (which is struc- ture aware) within very short time.

Afin d'étudier la diffusion et la libération de molécules de tailles inférieures dans un gel polymère, les coefficients d'auto-diffusion d'une série de polymères en étoile avec un noyau d'acide cholique et quatre branches de poly(éthylène glycol) (PEG) ont été déterminés par spectroscopie RMN à gradient de champ pulsé dans des solutions aqueuses et des gels de poly(alcool vinylique). Les coefficients de diffusion obtenus ont été comparés avec ceux des PEGs linéaires et dendritiques pour étudier l'effet de l'architecture des polymères. Les polymères en étoile amphiphiles ont des profils de diffusion en fonction de la concentration similaires à leurs homologues linéaires dans le régime dilué. Ils diffusent plus lentement dans le régime semi-dilué en raison de leur noyau hydrophobe. Leurs conformations en solution ont été étudiées par des mesures de temps de relaxation spin-réseau T1 du noyau et des branches. L'imagerie RMN a été utilisée pour étudier le gonflement des comprimés polymères et la diffusion dans la matrice polymère. Les comprimés étaient constitués d'amidon à haute teneur en amylose et chargés avec de l'acétaminophène (de 10 à 40% en poids). Le gonflement des comprimés, ainsi que l'absorption et la diffusion de l'eau, augmentent avec la teneur en médicament, tandis que le pourcentage de libération du médicament est similaire pour tous les comprimés. Le gonflement in vitro des comprimés d'un complexe polyélectrolyte à base d'amidon carboxyméthylé et de chitosane a également été étudié par imagerie RMN. Ces comprimés sont sensibles au pH : ils gonflent beaucoup plus dans les milieux acides que dans les milieux neutres en raison de la dissociation des deux composants et de la protonation des chaînes du chitosane. La comparaison des résultats avec ceux d'amidon à haute teneur en amylose indique que les deux matrices ont des gonflements et des profils de libération du médicament semblables dans les milieux neutres, alors que les comprimés complexes gonflent plus dans les milieux acides en raison de la dissociation du chitosane et de l'amidon. / In an effort to study the diffusion and release of small molecules in a polymeric system, the self-diffusion coefficients of a series of star polymers with a cholic acid core bearing four poly(ethylene glycol) (PEG) arms in aqueous solutions and gels of poly(vinyl alcohol) were determined by pulsed gradient spin-echo NMR techniques. The results have been compared with those of linear and dendritic PEGs to elucidate the effect of the architecture of the polymers. The amphiphilic star polymers show similar concentration-dependent diffusion behaviors in the dilute regime to their linear homologues. They diffuse more slowly in the semi-dilute regime than the linear PEGs due to the presence of the hydrophobic core. The conformation of the star polymers in the solutions was studied by measuring the T1 values of the core and the arms of the diffusants. NMR imaging was used to study the swelling of polymeric tablets and diffusion in the polymer matrix. The tablets investigated were made of cross-linked high amylose starch (CHAS) and loaded with acetaminophen (10, 20 and 40 wt%). The swelling, water uptake and diffusion in the CHAS network are faster at higher drug loading levels, while the drug release rates are similar among all the tablets. The in vitro swelling of the tablets made of a polyelectrolyte complex based on chitosan and carboxymethylated starch has also been studied by NMR imaging. These tablets showed pH-sensitive behavior. They swelled much more in acidic media than in neutral media due to dissociation of the two components and the protonation of the amino groups in the chitosan residues. The comparison of the results with those obtained with the CHAS tablets indicates that the two matrices have similar swelling and drug release profile in neutral media, while the complex tablets showed a greater extent of swelling in acidic media due the dissociation of the chitosan from the complex.

This study investigated the transport coefficients involved in mass and heat transfer during the drying of a porous solid partially saturated with multicomponent mixtures.  It included the coefficients governing liquid transport through the solid, the matrix of multicomponent diffusion coefficients in the liquid phase, and the effective thermal conductivity.  As it is not possible to determine these coefficients by theoretical considerations alone and considerable experimental work is required to determine them in a broad range of process conditions, the principle of this study has been the use of mathematical models complemented with some empirical parameters.  These empirical parameters were determined by comparison between measurements in specially designed experiments and the results of mathematical models that describe the process.  In addition, the application of the multicomponent diffusion coefficients is described in two cases where liquid diffusion is important: convective evaporation of a multicomponent stationary liquid film and a falling film. To study liquid transport through the solid, isothermal drying experiments were performed to determine the transient composition profiles and total liquid content of sand samples wetted with ternary liquid mixtures with different initial compositions and temperatures.  A mathematical model including mass transfer by capillary movement of the liquid and interactive diffusion in both the gas and liquid phases was developed.  To simulate the capillary movement of liquid mixtures, parameters experimentally determined for single liquids were weighed according to liquid composition. A fairly good agreement between theoretical and experimental liquid composition profiles was obtained considering that axial dispersion was included in the model. To study the matrix of multicomponent diffusion coefficients in the liquid phase, the redistribution of liquid composition in a partially filled tube exposed to a longitudinal temperature gradient was analysed.  Experimental work was carried out using two main ternary mixtures with different initial compositions and temperature gradients.  Experimental data were compared with the results of a theoretical model that describes the steady-state liquid composition distribution in a partially filled non-isothermal tube to find the empirical exponent that modifies the matrix of thermodynamic factors.  Correlations for the exponents as a function of temperature were determined for each particular multicomponent mixture. The effective thermal conductivity of a porous solid containing multicomponent liquid mixtures was studied by measuring the liquid composition, liquid content and temperature distributions in a cylindrical sample dried by convection from the open upper side and heated by contact with a hot source at the bottom side.  Simulations performed at a quasi steady state were compared with experiments to estimate the adjusting geometric parameter of Krischer’s model for effective thermal conductivity, which includes the contribution of the evaporation-diffusion-condensation mechanism. The results revealed that a resistance corresponding to a parallel arrangement between the phases seems to dominate in this case. In the study of the convective drying of a multicomponent stationary liquid film, the equations describing interactive mass transfer were decoupled by a similarity transformation and solved simultaneously with a conduction equation by the method of variable separation.  Variations of physical properties along the process trajectory were taken into account by a stepwise application of the solution in time intervals with averaged coefficients from previous time steps.  Despite simplifications, the analytical solution gives a good insight into the selectivity of the drying process and is computationally fast.  On the other hand, numerical simulations of the convective evaporation of the multicomponent falling liquid film into an inert gas with a co-current flow arrangement of the phases almost always revealed a transition from liquid-phase-controlled conditions to a process in which neither the gas nor the liquid completely controls the evaporation. The results obtained in this work would be useful in implementing models to improve the design, process exploration and optimisation of dryers by incorporating the solid-side effects to describe the drying of liquid mixtures along the whole process. / QC 20110124

Afin d'étudier la diffusion et la libération de molécules de tailles inférieures dans un gel polymère, les coefficients d'auto-diffusion d'une série de polymères en étoile avec un noyau d'acide cholique et quatre branches de poly(éthylène glycol) (PEG) ont été déterminés par spectroscopie RMN à gradient de champ pulsé dans des solutions aqueuses et des gels de poly(alcool vinylique). Les coefficients de diffusion obtenus ont été comparés avec ceux des PEGs linéaires et dendritiques pour étudier l'effet de l'architecture des polymères. Les polymères en étoile amphiphiles ont des profils de diffusion en fonction de la concentration similaires à leurs homologues linéaires dans le régime dilué. Ils diffusent plus lentement dans le régime semi-dilué en raison de leur noyau hydrophobe. Leurs conformations en solution ont été étudiées par des mesures de temps de relaxation spin-réseau T1 du noyau et des branches. L'imagerie RMN a été utilisée pour étudier le gonflement des comprimés polymères et la diffusion dans la matrice polymère. Les comprimés étaient constitués d'amidon à haute teneur en amylose et chargés avec de l'acétaminophène (de 10 à 40% en poids). Le gonflement des comprimés, ainsi que l'absorption et la diffusion de l'eau, augmentent avec la teneur en médicament, tandis que le pourcentage de libération du médicament est similaire pour tous les comprimés. Le gonflement in vitro des comprimés d'un complexe polyélectrolyte à base d'amidon carboxyméthylé et de chitosane a également été étudié par imagerie RMN. Ces comprimés sont sensibles au pH : ils gonflent beaucoup plus dans les milieux acides que dans les milieux neutres en raison de la dissociation des deux composants et de la protonation des chaînes du chitosane. La comparaison des résultats avec ceux d'amidon à haute teneur en amylose indique que les deux matrices ont des gonflements et des profils de libération du médicament semblables dans les milieux neutres, alors que les comprimés complexes gonflent plus dans les milieux acides en raison de la dissociation du chitosane et de l'amidon. / In an effort to study the diffusion and release of small molecules in a polymeric system, the self-diffusion coefficients of a series of star polymers with a cholic acid core bearing four poly(ethylene glycol) (PEG) arms in aqueous solutions and gels of poly(vinyl alcohol) were determined by pulsed gradient spin-echo NMR techniques. The results have been compared with those of linear and dendritic PEGs to elucidate the effect of the architecture of the polymers. The amphiphilic star polymers show similar concentration-dependent diffusion behaviors in the dilute regime to their linear homologues. They diffuse more slowly in the semi-dilute regime than the linear PEGs due to the presence of the hydrophobic core. The conformation of the star polymers in the solutions was studied by measuring the T1 values of the core and the arms of the diffusants. NMR imaging was used to study the swelling of polymeric tablets and diffusion in the polymer matrix. The tablets investigated were made of cross-linked high amylose starch (CHAS) and loaded with acetaminophen (10, 20 and 40 wt%). The swelling, water uptake and diffusion in the CHAS network are faster at higher drug loading levels, while the drug release rates are similar among all the tablets. The in vitro swelling of the tablets made of a polyelectrolyte complex based on chitosan and carboxymethylated starch has also been studied by NMR imaging. These tablets showed pH-sensitive behavior. They swelled much more in acidic media than in neutral media due to dissociation of the two components and the protonation of the amino groups in the chitosan residues. The comparison of the results with those obtained with the CHAS tablets indicates that the two matrices have similar swelling and drug release profile in neutral media, while the complex tablets showed a greater extent of swelling in acidic media due the dissociation of the chitosan from the complex.

Dans ce mémoire nous allons présenter une méthode de diffusion d’erreur originale qui peut reconstruire des images en demi-ton qui plaisent à l’œil. Cette méthode préserve des détails fins et des structures visuellement identifiables présentes dans l’image originale. Nous allons tout d’abord présenter et analyser quelques travaux précédents afin de montrer certains problèmes principaux du rendu en demi-ton, et nous allons expliquer pourquoi nous avons décidé d’utiliser un algorithme de diffusion d’erreur pour résoudre ces problèmes. Puis nous allons présenter la méthode proposée qui est conceptuellement simple et efficace. L’image originale est analysée, et son contenu fréquentiel est détecté. Les composantes principales du contenu fréquentiel (la fréquence, l’orientation et le contraste) sont utilisées comme des indices dans un tableau de recherche afin de modifier la méthode de diffusion d’erreur standard. Le tableau de recherche est établi dans un étape de pré-calcul et la modification est composée par la modulation de seuil et la variation des coefficients de diffusion. Ensuite le système en entier est calibré de façon à ce que ces images reconstruites soient visuellement proches d’images originales (des aplats d’intensité constante, des aplats contenant des ondes sinusoïdales avec des fréquences, des orientations et des constrastes différents). Finalement nous allons comparer et analyser des résultats obtenus par la méthode proposée et des travaux précédents, et démontrer que la méthode proposée est capable de reconstruire des images en demi-ton de haute qualité (qui préservent des structures) avec un traitement de temps très faible. / In this work we present an original error-diffusion method which produces visually pleasant halftone images while preserving fine details and visually identifiable structures present in original images. We first present and analyze the previous work to show the major problems in halftoning, and explain why we decided to use an error diffusion algorithm to solve the problems. Then we present our method which is conceptually simple and computationally efficient. The source image is analyzed, and its local frequency content is detected. The main components of the frequency content (main frequency, orientation, and contrast) serve as lookup table indices in a pre-computed database of modifications to a standard error diffusion. The modifications comprise threshold modulation and variation of error-diffusion coefficients. The whole system is calibrated in such a way that the produced halftone images are visually close to original images (patches of constant intensity, patches containing sinu- soidal waves of different frequencies/orientations/contrasts, as well as natural images of different origins). Finally, we compare and analyze the results obtained by our method and previous work, and show that our method can produre high-quality halftone image (which is struc- ture aware) within very short time.