Dissertations / Theses on the topic 'Reactive extraction'

Fatty acid methyl ester (FAME “biodiesel”) is a renewable transport fuel that can be produced from waste/refined oil, pre-extracted oil from oilseeds or microalgae. The most common method converts extracted oil from biomass to FAME through transesterification using acidified or alkalised methanol. Alternatively, FAME can be made by contacting the oil-bearing biomass directly with an alcohol containing a catalyst. This approach is potentially a cost-effective alternative way of making algal FAME due to its elimination of the solvent extraction step and its higher water tolerance. This study reports reactive extraction of Nannochloropsis occulata and Chlorella vulgaris for FAME production using NaOH, H2SO4, zirconium dodecyl sulphate (“ZDS”) or H2SO4/SDS (a surfactant) as catalysts. It is possible to produce FAME using all of them. A relationship was found between FAME yield, catalyst concentration, methanol to oil molar ratio, moisture content or algal cell wall chemistry. NaOH is the most effective catalyst, producing high FAME yields (96 %) in relatively short reaction times (10 min), at 925:1 methanol to oil molar ratio and 0.5N NaOH. This was achieved despite high levels of free fatty acid (6 % lipid) in Chlorella vulgaris. A numerical model derived by Eze et al. (2014) fitted with experimental data from this study shows that other side reactions including FAME and triglyceride saponification, free fatty acid neutralisation occur alongside the desired FAME synthesis in a NaOH-catalysed reactive extraction. Regardless of the catalysts used, methanol to oil molar ratios in the range 600:1-1277:1 caused 5-30 wt %/(wt dry algae) moisture tolerance: significantly greater than the 0.5 wt % oil moisture required in conventional transesterifications. Both the phosphorus mass balance and conversion of the isolated algal phospholipids into FAME revealed that pre-soaking pre-treatment solubilises the phospholipid bilayer to some degree, and iii [Abstract continued] contributes to an increased FAME yield in Nannochloropsis occulata (98.4 %) and Chlorella vulgaris (93.4 %). Residual protein loss in Chlorella vulgaris and Nannochloropsis occulata were respectively 6.5 and 10 %. The carbohydrate content was significantly reduced by 71 % in Chlorella vulgaris and 65 % in Nannochloropsis occulata.

The MixAlco process, a proprietary technology owned by Texas A&M University, converts biomass (e.g., municipal solid waste, sewage sludge, paper, agricultural residues, and energy crops) into usable chemicals (e.g., acetic acid) and fuels (e.g., ethanol). Historically, calcium carbonate has been used as the buffer. Recently, it was found that using ammonium bicarbonate as the buffering agent enhances the fermentation conversion. In this case, fermentation broth contains ammonium salts (e.g., ammonium acetate, propionate, butyrate, pentanoate). Therefore, the downstream processing steps (including extraction, purification, esterification, and product separation) must be compatible with the ammonium carboxylate salts formed in the fermentation. This research focuses on converting fermentation broth carboxylate salts into their corresponding acids via "acid springing." Reactive extraction and thermal conversion (distillation) are crucial parts of the acid springing process. Because the components of the fermentation broth are over 80% ammonium acetate and 20% other ammonium carboxylate salts (ammonium propionate, butyrate, pentanoate, etc.), all the initial experiments in this study were performed using reagentgrade ammonium acetate to simplify the reaction. Later, actual fermentation broth was employed. The primary objective of this study was to provide the optimal operating conditions to make the downstream processing steps of the MixAlco process compatible with ammonium carboxylate salts formed in the fermentation. The optimal initial concentration for reactive extraction should be 150-200 g/L and the volume ratio of aqueous phase and extractant should be 1:1. The distribution coefficient reaches the maximum value when the concentration of TOA is 20% (vol %) in n-octanol. The batch distillation study shows that there are two reaction stages: (1) water leaves the system at 100-106 °C and (2) the acid-amine complex decomposes at 160-180 °C.

Lactic acid recovery from dilute fermentation broths is a growing requirement due to the increasing demand for pure lactic acid. Reactive extraction is proposed as an alternative to conventional methods of recovery, since the selectivity of separation is remarkably enhanced in reactive extraction. The aim of this study is to perform the equilibrium studies for the recovery of lactic acid from its synthetic aqueous solutions (not from real fermentation broths) by reactive extraction and investigate the effects of various parameters such as initial lactic acid concentration in the aqueous phase (0.25 - 1.3 M), initial pH of the aqueous phase (2 &ndash
6), organic phase extractant concentration (0.1 &ndash
0.5 M), type of the extractant (chloride, hydrogensulphate and hydroxide salts of tri-n-octylmethylammonium) and the type of diluent (oleyl alcohol or octanol). The results of the experiments showed that the degrees of extraction decreased with increasing use of diluent with the extractant and increasing initial lactic acid concentration of the aqueous phase. Highest degrees of extraction were achieved for undiluted extractants. The performance of the diluents were investigated by performing extraction experiments with solutions of TOMAC in oleyl alcohol or octanol at different pH values and it was observed that octanol had a higher solvating power than oleyl alcohol especially at lower aqueous phase pH values. Higher extraction efficiencies were obtained for TOMAC dissolved in octanol rather than oleyl alcohol. Initial aqueous pH of 6 was identified as the optimum pH for the extraction, also due to its being equal the average fermentation pH for the extractions with Lactobacillus species. Among the different salts of tri-n-octylmethylammonium, hydroxide salt exhibited the highest degrees of extraction (83% with undiluted TOMA(OH) and 78% with 0.5 M TOMA(OH) in octanol for the extraction of 0.316 M lactic acid solutions). The present work is the first step in the design of an industrial reactive extraction process that is going to attempt forward and backward extraction of lactic acid simultaneously in a hollow fiber membrane module that is going to be attached to the lactic acid fermentor to achieve continuous product recovery. The equilibrium data obtained from this study can be combined with the kinetic studies as the next step of designing a separation module.

Interest in recovery of carboxylic acids from their aqueous solutions and production media has been growing with the improvements in the production techniques. Reactive extraction is proposed as a promising method to achieve high distribution coefficients with good selectivity. In the present study, equilibrium and kinetic studies, needed for the design of a recovery unit were carried out. Reactive extraction of pyruvic, acetic and lactic acids from their single and mixed acid solutions were carried out with tertiary amines, Alamine 336 and trioctylamine, dissolved in diluents, 1-octanol and oleyl alcohol. The kinetic parameters such as reaction order and rate constant were calculated by using the aqueous solution of pyruvic acid with 1-octanol solution of trioctylamine. The distribution coefficient of pyruvic acid was obtained as 0.30 and 0.07 with 1-octanol and oleyl alcohol respectively. The trend did not change after the addition of the extractants to the diluents. As expected, the more polar diluent extracted more acid than the less polar one. The extractant was observed to be the limiting reagent of the reversible complexation reaction in the organic phase. It was seen that loading ratio was not affected by the concentration of the extractant in the organic phase but increased with the equilibrium concentration of the acid in the aqueous phase. The results showed that mainly (1-1) acid-amine reaction occurred in the organic phase. Because tertiary amines can react with only undissociated acids, the increase in the initial pH of the aqueous phase caused a decrease in the distribution coefficients of the carboxylic acids studied. The effect of the concentration of the organic phase disappeared when the initial pH of the aqueous phase was 4.0 and a distribution coefficient value of 0.1 was achieved for all concentration levels of organic phase. Similar results were obtained for acetic and lactic acids and this behavior was used to propose a selective recovery of the acids from their mixed acid solutions. The presence of acetic acid prevented the extraction of pyruvic acid. The increase in the concentration of the extractant in the organic phase caused an increase in the distribution coefficient of pyruvic acid. During the extraction process, lactic acid could not be recovered from ternary acid solutions because of its high hydrophilicity. The results of reactive extraction of lactic and pyruvic acids from their binary acid solutions showed that at higher concentration of lactic acid, the distribution of pyruvic acid was negatively affected. On the other hand, the presence of lactic acid at a low concentration level did not affect the distribution of pyruvic acid. It was found that the reaction between pyruvic acid and trioctylamine in 1-octanol was first order with respect to the reactants with a second order rate constant of 0.94 L mol-1 s-1. The enhancement factor of the system was obtained as 25.

A single-step, extractive reaction for extraction of lipids such as biodiesel components, omega-3 fatty acids, or other triglycerides from microbial cells was examined. Conventional methods for lipid extraction use toxic solvents, and require multiple steps and long processing times. When the goal is to produce fatty acid methyl esters or FAMEs, the extracted lipids are subjected to a separate transesterification reaction with simple alcohols in the presence of an acid or base catalyst. A simplified, single-step reactive extraction method can be applied that combines the sequential extraction followed by transesterification using acidified alcohols - a process known as in situ transesterification. It was hypothesized that the in situ transesterification could be scaled-up for industrial processing by a systematic understanding of fundamental reaction parameters including temperature, catalyst concentration, and biomass/solvent ratios. The hypothesis was tested using a marine fungus, Schizochytrium limacinum SR21. Growth of SR21 resulted in biomass yields of 0.3g-biomass/g-glycerol and accumulated high amounts of palmitic acid (C16:0, 0.255g-FAME/g-biomass), docosahexaenoic acid (DHA, C22:6, 0.185g-FAME/g-biomass), myristic acid (C14:0) (0.017g-FAME/g-biomass), and pentadecanoic acid (C15:0, 0.012g-FAME/g-biomass). The bulk phase separation characteristics of the FAMEs were evaluated at high biomass concentrations. Recyclability of the acidified methanol in the system was also tested. A significant finding was that automatic phase separation of the FAMEs could be achieved. When FAME concentration reaches critical solubility, 22.7mg-FAME ml-1 methanol, all remaining FAMEs automatically phase separate. After FAME separation, the remaining methanol was recycled and used in subsequent in situ reactions. Upon recycling, greater than 85% of product extraction and recovery was achieved. The kinetics of the transesterification reaction was evaluated under various acid and biomass/solvent conditions. Based on the fundamental reaction mechanism governing the in situ transesterification, a theoretic model was derived to predict the conversion of TAGs into FAMEs. Kinetic parameters were estimated by fitting the experimental data and the resulting model. The model derived closely resembled the observations in this study. Through understanding of the fundamental reaction kinetics and limitations during processing, a new, reliable, and cost-effective system for large scale lipid production can be developed for microbial biomass including oleaginous algae, fungi, and yeast.

Lactic acid is a fermentation-derived organic acid used in a wide range of industries, such as food processing and pharmaceuticals. Its market is expected to expand due to the worldwide concern for the environment, as it is an essential feedstock for biodegradable polymers. However, fermentation product is a very dilute, multicomponent aqueous solution. Subsequent separation, purification and concentration of organic acids is difficult because of high affinity of the acids for water. Reactive extraction is a viable alternative to classical separation techniques. Amine extractants dissolved in organic diluents are suitable agents with reasonable ranges of viscosity and density of the solvent phase. The product is obtained in an organic phase after reactive extraction. The aim of this study is to obtain equilibrium data of back extraction of lactic acid into appropriate aqueous solutions from the organic phase. Aqueous solutions of NaCl, NaOH, Na2SO4, NaNO3 and Na2CO3 were examined as back extractant in various initial concentrations (0.005 &ndash
3 M). The organic phase consists of tri-n-octylmethylammonium lactate (TOMA(La)) dissolved in either oleyl alcohol or octanol with initial concentrations between 0.1 and 0.3 M. According to results of the experiments, the level of back extraction generally increased with increasing initial salt concentration in aqueous phase and decreased with increasing initial TOMA(La) concentration in organic phase. For all salts investigated, considerable levels of back extraction were obtained. NaOH was considered as the most suitable back extractant among the salts investigated since it exhibits higher distribution coefficients, regenerates tri-n-octylmethylammonium hydroxide (TOMAOH) in the organic phase and has no adverse effect on fermentation medium when forward and backward extraction steps are coupled with the fermentation. The effect of diluent type of TOMA(La) was also investigated during the experiments and it was concluded that octanol is a better diluent since it gives higher equilibrium distribution coefficients in addition to its higher solvating power and lower viscosity. The present work is a part of a comprehensive research program aiming to collect data and develop knowledge for the design of an industrial reactive extraction process coupling forward and backward extraction of lactic acid in a single unit and integrating fermentation and product separation. The kinetic parameters should be obtained as the next step for the design of such a process.

In this work, non-membrane and membrane methods were utilized to investigate the separations of olefin/paraffin and several industrial gas pairs. The non-membrane methods section investigated different methods used for olefin and paraffin separations which include liquid extraction and reactive distillation. I -hexene and n-hexane were selected as the olefin and paraffin representatives and were tested in all the liqu id extraction and distillation separation experiments. In the liquid extraction separation systems, three different extractants were used namely [Bmim][N Tf2L silver nitrate salt of MIM-PS (1-( 1-methyl-3-im idazolio)propane-3 -su lfonate) dissolved in [Bmim][HS041 mixtures and [BmimHBF,l cont aining AgBF4• Distribution coefficients of I -hexene and n-hexane were ca lculated for each and selectivities were also obtained from the distribution coefficient data. For the distillation separation method, silica supported ionic liquids (SILs) of [Bmim][NTf21 and [Bmim][BF41 containing AgBF4 were prepared and adopted as the reactive packing to aid in the separation of the olefin. Vapour-liquid equilibrium (VlE) data were studied for the ordinary distillation without Sils packing as well as for the different reactive distillation systems. TPU (Thermal polyurethane) and PSF (polysulfone) were selected as the polymer materials to carry out the olefin/paraffin permeabilities and separation effects tests in the membrane sepa ration section. Casting and coating composite polymer/ionic liquid membranes of different weight ratios of polymer to ionic liquid were made to undertake the olefin/paraffin separation tests. Ethane/ethylene and propane/propylene gas pairs were selected as the olefin/ paraffin rep resentatives. Co mposite TPU/AgOTf/[Bmim+OTfj and TPU/AgPFt;/ [Bmim+PF6-1 membranes were also pre pa red and tested for olef in/paraffin separations. Diffe rent weight ratios of polymer to silver salt and mol ratios of silver salt to ionic liquid were chosen to synthesize the composite polymer/silver salt/ionic li quid membranes. Finally, the olefin/ paraffin separation mechanisms were discussed.

La tesi della ha come obiettivo la valorizzazione delle acque di scarto derivanti dalla produzione del bioetanolo da colture cerealicole dedicate. Nella prima parte si studia la produzione di VFA mediante digestione anaerobica. In seguito, sono eseguiti batch tests per valutare la conversione dei VFA in MCFA. Infine, si focalizza sul processo di dowstream per il recupero degli acidi carbossilici dal brodo di fermentazione attraverso reactive extraction e back extraction.

L'extraction du bore du silicium liquide est une étape d'une chaîne de procédés de purification de silicium de qualité suffisante pour les applications photovoltaïques. Cette thèse étudie en priorité le procédé dit "gaz froid" qui consiste en l'injection d'un mélange de gaz Ar-H2-H2O sur du silicium liquide chauffé électromagnétiquement. Une deuxième méthode similaire ("procédé plasma") où on injecte un plasma thermique issu d'un mélange Ar-H2-O2 a également été étudiée. Un modèle est nécessaire afin d'optimiser le procédé pour économiser de l'énergie.Les trois objectifs du modèle sont la prédiction du flux de silicium issu de la surface (vitesse d'oxydation), du flux de bore issu de la surface (pour avoir la vitesse de purification), et du seuil de passivation. Le seuil de passivation est la limite de concentration d'oxydant au-delà de laquelleil apparait une couche de silice passivante qui empêche la purification. Afin de minimiser la consommation d'énergie en accélérant le procédé, on cherche à injecter une concentration d'oxydant juste en dessous du seuil de passivation.De précédentes études ont montré que le facteur limitant pour les flux de bore et de silicium est le transport d'oxydant dans la phase gaz. Ainsi, nous avons fait un modèle monodimensionnel réactif-diffusif à l'équilibre thermodynamique de la couche limite gazeuse. Selon ce modèle, l'effet de la formation d'aérosols de silice est de diviser par deux le flux d'oxydant vers la surface, ce qui sert aux simulations CFD. Cet effet des aérosols de silice sur les flux d'oxydant peut aussi se retrouver si on enlève l'hypothèse d'équilibre thermodynamique des aérosols de silice avec la phase gaz, ce qui est confirmé par des simulations CFD et des expériences.Pour ce qui concerne l'estimation de la vitesse de purification, les données les plus réalistes concernant l'enthalpie de formation de HBO(g) et le coefficient d'activité du bore dans le silicium liquide ont été sélectionnées. Nous obtenons une bonne prédiction de la vitesse de purification à différentes températures et concentrations d'oxydant, y compris pour le cas plasma que nous avons étudié, en utilisant ces données thermodynamiques et en supposant que les produits de réaction de surface SiO(g) et HBO(g) diffusent de manière similaire. Ces coefficients de transfert identiques pour HBO(g) et SiO(g) peuvent s'expliquer par une précipitation simultanée et commune de HBO(g) et SiO(g), selon des mécanismes de germination et croissance restant à déterminer.Un dispositif expérimental de lévitation électromagnétique de silicium sous un jet oxydant a été monté. La mesure et le contrôle de température d'une bille de silicium ont été mis en oeuvre ce qui permettra la mesure sans contaminations de données thermodynamiques concernant les impuretés .Le seuil de passivation mesuré sur quelques expériences disponibles peut être prédit par notre modèle d'oxydation (associé au facteur deux représentant les aérosols de silice), si on l'associe à un critère proposé dans la littérature, qui couple la fraction du flux d'oxydant arrivant à la surface à une loi d'équilibre entre SiO(g), Si(l) et SiO2(s/l). Nous montrons dans cette thèse que la couche passivante n'est compatible avec des aérosols de silice que si ces aérosols ne sont pas en équilibre avec la phase gaz. La cinétique de formation des aérosols de silice doit donc être étudiée plus en détails
Boron extraction from liquid silicon is a step within a new chain of processes aimed to purify silicon that meets purity requirements specific to photovoltaic applications. This thesis focuses mostly on cold gas processes that involve the injection of a mixture of Ar-H2-H2O gases onto electromagnetically stirred liquid silicon. A second similar method ("plasma processes") that involves the injection of thermal plasma made from an Ar-H2-H2O mixture has also been studied. A model is needed to minimize energy consumption by optimizing the process.We want to be able to predict the flow of silicon from the reactive surface (oxidation speed), the flow of boron from the surface (to have the purification speed) and the passivation threshold. For a given setting, the passivation threshold is the limit oxydant partial pressure at injection beyond which a passivating silica layer appears on the surface of the liquid silicon, which interrupts the purification. In order to minimize the energy consumption, and for that matter , in order to speed up the process, we want to inject oxydant in a quantity just below the passivation threshold.Previous studies have shown that the limiting factor for the oxidation and purification speed is the transport of oxidant in the gas phase. That's why we have made a 1D reactive-diffusive model at thermodynamical equilibrium of the gaseous boundary layer. According to this model the effect of the formation of silica aerosols is to divide by two the flow of oxydant towards the surface, which is useful for the simplification of CFD simulations. This effect of the formation of silica aerosols on oxidant flows can also be found without the hypothesis of thermodynamical equilibrium of silica aerosols with the gas phase, as confirmed by simulations and experiments.Regarding the estimation of the purification speed, we have selected the most realistic values of the enthalpy of formation of HBO(g) and of the activity coefficient of boron in liquid silicon.We could get good estimates of the purification speed at different temperatures and levels of oxidant concentrations at injection, by using the selected thermodynamical values and by supposing that the surface reaction products HBO(g) and SiO(g) diffuse similarly. A reason for this similar diffusion of SiO(g) and HBO(g) might be a common and simultaneous precipitation , due to specific dynamics of nucleation and growth that need to be investigated further. Those results for cold gas processed could also be obtained for a plasma experiment.However for the plasma experiment, silica aerosols can be formed only in a very thin layer near the surface and this result needs confirmation from other experiments.Temperature measurement and control for electromagnetically levitating liquid silicon under a flow of oxidant were achieved. With more time, quantitative results could be achieved to measure thermodynamical data on impurities without contaminations.Regarding the prediction of the passivation threshold, we justified a thermodynamical equilibrium at surface of SiO(g) with Si(l) and SiO2(s/l) at passivation threshold with the spreading of silica particles over the liquid silicon surface with the stirring. We show that the passivation layer is compatible with silica aerosols only if those aerosols are not in equilibrium with the gas phase. Therefore the kinetics of formation of silica aerosols should be studied further. A previous empirical formula on the prediction of the passivation threshold for experiments where H2O is the oxidant has been confirmed using our CFD model. A passivation experiment has shown the absence of impact of silica aerosols on oxidant transport when the oxidant is O2

L’extraction d’éléments d’intérêt (terres rares, métaux lourds, métaux de transition, radionucléides) de de minerais conventionnels ou non conventionnels (mines urbaines) et d’effluents industriel ou nucléaire est un enjeu majeure dans les procédés d’extraction, de recyclage et/ou de décontamination.Un grand nombre de solides organiques et inorganiques, chélatants ou échangeurs d’ions, ont été développés pour un tri ionique sélectif par procédé solide/liquide.L’introduction covalente de ligands spécifiques dans le squelette de la résine permet d’augmenter la sélectivité vis-à-vis d’une cible métallique par la nature des agents de complexation, leurs sites de chélation et leur rigidité.L’objectif de cette thèse recouvre d’une part la synthèse de systèmes chélatant réactifs, leur polymérisation pour la préparation des résines « super »chélatantes et d’autre part, l’évaluation des polymères pour l’extraction/séparation d’ions, les terres rares étant plus particulièrement ciblées.A partir de résines formo-phénolique incorporant du catéchol et la 8-hydroxyquinoléine qui sont des précurseurs phénoliques, nous avons pu montrer que la récupération sélective du germanium été possible vis-à-vis du silicium ou du zinc en fonction de leur proportion dans la matrice polymérique.La synthèse des précurseurs phénoliques intégrant des ligands comme l’acide diglycolamique et les diglycolamides a permis la mise en œuvre de résine pour la récupération des terres rares. Une étude sur la compréhension de la synthèse des résines ainsi que leurs performances d’extraction a ainsi été réalisée et a pu mettre en évidence le potentiel de ces résines pour la récupération et valorisation des terres rares.Une ouverture vers d’autres possibilités de synthèses de ce type de résines (matériaux poreux, sous forme de mousse…) est également proposée pour permettre d’augmenter les surfaces de contacts lors des extractions solide-liquide et ainsi augmenter les performances de ces matériaux
Extraction of specific targeted species (rare earth, heavy metals, transition metals, radionuclides) from industrial (mining deposit and urban mining) and nuclear effluents is an important issue in the recycling and / or decontamination processes. A large number of organic and inorganic solids, chelating or ion-exchanging materials, have been developed for selective ionic separation by solid / liquid process. Metal-specific ligands incorporate in the structure of the resin itself is an interesting way to perform ion separation taking the advantage of the selectivity of chelating agents. The aims of this project are firstly the synthesis of chelating original systems and then prepare specific ion exchange resins.From formo-phenolic resins incorporating catechol and 8-hydroxyquinoline, we have been able to show that the selective recovery of germanium from silicon or zinc was possible, , depending on their proportion of each phenolic precursors in the polymeric matrix.The synthesis of the phenolic precursors integrating ligands such as diglycolamic acid and diglycolamides allowed the use of resin for the recovery of rare earths elemnets. A study on the understanding of the synthesis of resins as well as their performance of extraction was thus realized and was able to highlight the potential of these resins for the recovery and valorization of the rare earths elements.An opening towards other possibilities of synthesis of this type of resins (porous materials, in the form of foam, etc.) is also proposed in order to increase the contact surfaces during solid-liquid extractions and thus increase the performance of these materials

L’acide 3-hydroxypropionique (3-HP) est une molécule plate-forme, particulièrement visée pour ses dérivés acrylés et les polyesters. Sa production par voie biotechnologique fait l’objet de nombreuses études, afin de les obtenir de manière biosourcée. Cependant, dans ces procédés, le 3-HP est dilué dans des milieux de fermentation contenant de nombreuses impuretés. De plus, l’accumulation de l’acide dans les milieux crée une forte inhibition sur les microorganismes producteurs. Nous proposons alors un procédé visant à extraire sélectivement le 3-HP du milieu de fermentation en même temps qu’il est produit afin de limiter son accumulation et de le purifier en continu. Les propriétés du 3-HP ainsi que les meilleures conditions de fermentation actuelles ont permis d’identifier l’extraction liquide-liquide réactive en contacteur membranaire comme une technique de choix. Nous avons tout d’abord élucidé le mécanisme réactionnel impliqué dans le système puis modélisé les équilibres associés afin de pouvoir prédire les rendements d’extraction en fonction de paramètres opératoires. Plusieurs méthodes de désextraction ont été testées, ce qui a permis de coupler les étapes d’extraction et de désextraction dans un procédé semi-continu mimant une production fermentaire. La modélisation dynamique de ce mode de fonctionnement a permis de prédire précisément les résultats expérimentaux. Les milieux ont alors été complexifiés afin de mieux représenter la réalité d’un milieu biologique en identifiant l’impact de chaque constituant
3-hydroxypropionic acid (3-HP) is a platform molecule targeted for its acrylated derivatives and polyesters. Its production through biological pathways is widely studied in order to make them bio-based. However, in such processes, 3-HP is diluted in fermentation broths containing many impurities. Moreover, the accumulation of the acid in the broths generates a strong inhibition towards the producing microorganisms. We suggest here a process aiming at extracting 3-HP selectively as soon as it is produced in order to reduce its accumulation and have it purified continuously. 3-HP properties as well as the current best fermentation conditions made us identify reactive liquid-liquide extraction in membrane contactors as promising technique. First, we elucidated the reaction mechanism of extraction in our system and then modeled the associated thermodynamic equilibria as a function of operating paramaters. Several back-extraction methods were tested and it allowed the coupling of the extraction and back-extraction steps in a semi-continuous process mimicking a biological production. The dynamic modeling of this operating mode made possible the accurate prediction of experimental results. The aqueous phase were then made more and more complex in order to better describe an actual fermentation broth and identify the influence of each component on the process efficiency in terms of yield, kynetics and selectivty

Le présent manuscrit fait état d’une technique microfluidique innovante d’étude du transfert de matière appliqué au système uranium(VI)-HNO3 / TBP-TPH, pour le recyclage du combustible usé. Ce système chimique a grandement été étudié pour la détermination des cinétiques de transfert d’uranium(VI). Toutefois, seule la cinétique globale incluant les phénomènes d’advection-diffusion et de réaction chimique a été établie. La composante chimique a souvent été approchée mais la forte contribution des résistances diffusionnelles, lors de l’extraction dans les outils de détermination utilisés, ne permet pas de correctement la déterminer. Grâce à la miniaturisation inhérente au dispositif microfluidique développé lors de cette étude et à la mise en oeuvre des phases à grande vitesse, la contribution de ces résistances est réduite voire négligée et la cinétique chimique est supposée atteignable. Les études préliminaires réalisées en microscopie confocale ont confirmé la stabilité de l’interface liquide-liquide dans la profondeur du micro-canal et le long de l’écoulement, malgré les fortes vitesses. Nos expériences ont montré que même avec des temps de séjour compris entre 8 ms et 35 ms, l’extraction d’uranium(VI) intervient de façon quantifiable. Cela a mis en évidence une très forte homogénéisation en phases en fonction des concentrations initiales d’uranium(VI). De plus, nos études en régime chimique confirment une réaction d’extraction du premier ordre envers l’uranium(VI). Ces observations semblent être confirmées par les modélisations en 2D sous COMSOL et Scilab, car elles permettent la correspondance avec les données expérimentales qu’en admettant la présence de l’effet Marangoni massique. Dans le cas de l’extraction d’une solution d’uranium(VI) à 50 g.L-1, la prise en compte de cet effet a conduit à déterminer que les vitesses radiales des phases sont suffisamment élevées, 0,2 m.s-1. Ainsi, nous avons considéré pour cette expérience que les concentrations d’uranium(VI) en sortie de microsystème sont équivalentes aux concentrations interfaciales. En tenant compte de ces données, la constante cinétique chimique du système uranium(VI)-HNO3 3 M / TBP-TPH a pu être estimée vers 3,2x10-4 m.s-1
An innovative microfluidic parallel co-current flow process has been developed to study the mass transfer kinetics of the system uranium(VI)-HNO3 / TBP-TPH. Due to the use of this chemical system in the spent fuel recycling, the kinetics of the uranium(IV) transfer has already been studied and the overall kinetic constant is available. Nevertheless, the chemical kinetic constant has not yet be determine because of the high molecular diffusion resistance at macroscopic scale. Thanks to miniaturization of the studying microfluidic device and the high velocities flows, those molecular diffusions resistances are strongly reduces then the chemical kinetic constant is supposed to be reachable. Our first studies in confocal microscopy confirm a straight liquid-liquid interface across the depth of the micro-channel and along the flow, even at high speed. Extractions experiments have highlighted a strong phase homogenization which get more important has the initial uranium(VI) concentration increase. Moreover, 2D modelling with COMSOL or Scilab confirms the previous observations where the experimental fits are only validated considering the presence of a solute Marangoni phenomenon. In this case and especially at 50 g/L of uranium(VI), the radial velocity generated by the Marangoni effect is high enough to consider that outlets uranium(VI) concentrations are equivalent to the interfacial concentrations. So, we have estimated the chemical kinetic constant of uranium(VI)-HNO3 3 M / TBP-TPH to be near 3.2x10-4 m/s with a first order reaction for the uranium(VI) in chemical mass transfer condition

Le couplage fonctionnel des opérations de séparation et de réaction ainsi que les bio-procédés sont deux axes de recherche largement explorés. Cependant, l’industrie du génie des procédés a du mal à se tourner vers des technologies de ce type car il demeure un réel manque de connaissances et d’outils de conception pour ce genre de procédés. Une méthodologie de conception générale pour l'extraction liquide-liquide réactive est introduite dans cette étude. Elle est composée de trois étapes différentes: l'analyse de faisabilité, la synthèse ou dimensionnement du procédé et la validation par simulation. Cette méthodologie conduit à des paramètres structuraux et opératoires de la colonne étudiée à partir seulement des informations concernant le comportement physico-chimique du système étudié, en exploitant les équations d’équilibre chimique et entre phase ainsi que les bilans matières. Les résultats de cette méthode sont un bon point de départ pour une étude d'optimisation ou d'un processus de calcul d'investissement. Cette méthodologie a été appliquée à différentes études de cas: regroupant deux stratégies différentes d'extraction avec plusieurs solvants pour récupérer l'acide succinique dans un milieu de fermentation

Ein Großteil der Lebensmittel wird in der heutigen Zeit vor allem aufgrund ihrer langen Haltbarkeit in Konservendosen verpackt. Zur Qualitätserhaltung des Lebensmittels werden Weißblechdosen im Innenbereich in der Regel mit einer Lackierung versehen. Lackrohstoffe sind dabei u. a. Phenol- und Epoxidharze, die als Basis- oder Vernetzerkomponente eingesetzt werden. Bei der Herstellung und Lagerung dieser Lebensmittel kann jedoch nicht ausgeschlossen werden, dass es zur Migration von Bestandteilen aus der Kunststoffinnenbeschichtung in das Füllgut kommt. Toxikologisch und somit auch analytisch sind dabei vor allem die migrierenden Verbindungen unter 1000 Da von Interesse, da Substanzen mit einer Molmasse von über 1000 Da nur zu weniger als 1 % im Gastrointestinaltrakt absorbiert werden. Analytik von phenolischen Verbindungen Bei Untersuchungen zur Migration phenolischer Verbindungen unter Verwendung verschiedener Modellcoatings und Simulanzlösemittel wurden Konzentrationen an migrierenden phenolischen Substanzen in Summe bis 160 µg/dm² bestimmt, womit dieser Gehalt deutlich unter dem gesetzlichen Grenzwert für die Gesamtmigration von 10 mg/dm² liegt. Die Quantifizierung erfolgte dabei über SEC-FLD-Kalibriergeraden der für diese Coatings verwendeten Phenolharze. Tendenziell steigt dabei die Menge an übergehenden Verbindungen mit abnehmender Polarität der verwendeten Simulanzien. Ebenso abhängig vom Lösungsmittel ist die Molekulargewichts¬verteilung der im Migrat enthaltenen Phenole. So zeigte sich eine deutliche Verschiebung der phenolischen Verbindungen zu höheren Molekulargewichten mit abnehmender Polarität der Simulanzlösemittel. Mit wenigen Ausnahmen besitzen jedoch alle migrierenden Substanzen eine Molmasse von unter 1000 Da. Das beobachtete Migrationsverhalten kann u. a. auf die unterschiedliche Reaktivität der phenolischen Basismonomere der Harze zurückgeführt werden, wodurch die Fähigkeit variiert, unter den Einbrennbedingungen ein ausgeprägtes Netzwerk zu bilden. Zur näheren Charakterisierung der Phenolharze wurden einzelne Hauptverbindungen der RP-HPLC-FLD-Chromatogramme identifiziert. Über Derivatisierungsreaktionen mit Picolinsäure, Essigsäureanhydrid sowie Dansylchlorid konnten Informationen zur Anzahl an alkoholischen und phenolischen Hydroxylgruppen im Molekül erhalten werden. Mit dem Wissen um die eingesetzten Phenolmonomere und einer eventuellen Veretherung konnten Strukturvorschläge erstellt werden. Die Quantifizierung der migrierenden phenolischen Verbindungen in Migraten kommerzieller Coatings in Summe wurde über eine universell anwendbare Kalibrierung angestrebt. Dazu wurden 17 verschiedene Phenolharze bezüglich der Steigung der SEC-FLD-Kalibriergeraden, des mittleren Molekulargewichtes, der Hydroxylzahl und dem Verhältnis OH-Gruppen/Molekül charakterisiert. Wie erwartet steigt mit wenigen Ausnahmen die Anzahl der OH-Gruppen im Molekül tendenziell mit dem mittleren Molekulargewicht. Es zeigte sich zudem, dass die verschiedenen Phenolharze in ihren fluorophoren Eigenschaften stark variieren. Die Steigung der SEC-FLD-Kalibriergeraden konnte zudem in keine Korrelation mit einem anderen ermittelten Parameter gebracht werden. Die Anwendung einer universellen Kalibriergerade zur Quantifizierung war deshalb nicht möglich. Die Größenordnungen der Migratkonzentrationen konnten daher nur über die zwei im Anstieg am stärksten variierenden SEC-FLD-Kalibriergeraden abgeschätzt werden. Bei der Analyse kommerzieller Epoxy-Phenol-Coatings war im Gegensatz zu Polyester-Phenol-Coatings die isolierte Detektion der phenolischen Verbindungen im Migrat mittels Fluoreszenz nicht möglich, da sowohl Epoxide als auch Phenole fluorophorer Eigenschaft besitzen. Es wurde daher eine Methode zur Abtrennung der Phenole von anderen im Migrat enthaltenen Subtanzen auf Basis eines Anionenaustauschermaterials entwickelt. Dabei wurde die Eigenschaft der Phenole genutzt, im basischen Milieu Phenolate zu bilden. Diese, aber auch Säuren adsorbieren am Austauschermaterial, während Epoxide oder Polyester nicht retardiert werden. Für zwei kommerzielle Epoxy-Phenol-Coatings konnte somit der Anteil an phenolischen Verbindungen im Migrat zu 7 und 28 % bestimmt werden. Mittels RP-HPLC/ESI-MS war es möglich, einige der phenolischen Verbindungen in diesen Migraten zu identifizieren. Dabei handelt es sich um nichtepoxidierte BPA-Derivate der Epoxidkomponente des Coatings, die aufgrund des BPA-Grundkörpers eine phenolische Hydroxylgruppe besitzen. Phenolische Vernetzungsprodukte beider Basisharze konnten dagegen nicht identifiziert werden. Die Menge an migrierenden phenolischen Verbindungen der beiden Epoxy-Phenol-Coatings konnte über die Kalibration eines niedermolekularen Epoxidharzes zu 0,1 mg/dm² bzw. 0,27 mg/dm² abgeschätzt werden. Während für einzelne phenolische Verbindungen gesetzliche Grenzwerte für die Migration bestehen, gibt es für Oligomere, mit Ausnahme von BPA keine spezifischen Migrationslimits. Ebenso sind in der Literatur kaum toxikologische Untersuchungen zu Phenololigomeren zu finden. Um einen ersten Einblick in die toxikologische Relevanz migrierender phenolischer Verbindungen zu erhalten, wurden mehrere kommerzielle als auch selbst synthetisierte phenolische Standardsubstanzen und verschiedene Molekulargewichtsfraktionen eines Phenolharzes, im Fischembryotest an Eiern des Zebrabärblings (Brachydanio rerio) nach DIN 38415-T647 und/oder Neutralrottest an Hep-G2 und HT-29 Zellen untersucht. Die stärksten Effekte im Fischembryotest bewirkte das Trimer BPM, hier reichte bereits eine Konzentration von etwa 2 mg/l aus, um 50 % der Fischembryonen letal zu schädigen. Im Gegensatz dazu waren beim Dimer 5-Hydroxymethyl-2,4´-dihydroxydiphenylmethan (M 230) mit einem EC50-Wert von 170 mg/l die geringste toxikologische Wirkung zu beobachten. Für alle anderen Subtanzen konnten EC50-Werte im Bereich 20 - 100 mg/l bestimmt werden. Tendenziell zeichnete sich dabei eine Zunahme der EC50-Werte mit steigender Lipophilie, ausgedrückt über den KOW-Wert ab, was auf den Aufbau der Fischeier zurückzuführen ist. So müssen die zu untersuchenden Xenobiotika mehrere lipophile Membranen durchdringen, um am eigentlichen Wirkungsort Einfluss auf die Embryonalentwicklung nehmen zu können. Im Zelltest konnten tendenziell ähnliche Ergebnisse ermittelt werden wie im Fischembryotest, wobei in der Regel die Hep-G2 Zellen empfindlicher reagieren als die HT-29 Zellen. Während für Phenol im untersuchten Konzentrationsbereich keine toxischen Effekte beobachtet werden konnten, liegen die EC50-Werte für das Trimer BPM, analog zum Fischembryotest deutlich unter 10 mg/l. Für die anderen Verbindungen wurden EC50-Werte zwischen 16 und 100 mg/l bestimmt. Analog zu den Untersuchungen der Einzelsubstanzen zeigte sich auch bei den 5 untersuchten Molekulargewichtsfraktionen zwischen 0 und 1000 Da, dass das toxikologische Potential im niedermolekularen Bereich (0 - 200 Da) gegenüber den Fraktionen 200 400 und 400 600 Da vergleichsweise gering ist. Bereits 25,1 mg/l bzw. 17,3 mg/l der Fraktionen 200 400 Da und 400 600 Da waren in den Tests ausreichend, um alle Embryonen letal zu schädigen. Im Molekulargewichtsbereich über 600 Da konnten dagegen lediglich subletale oder gar keine Missbildungen beobachtet werden. Analytik von Epoxiden In der amtlichen Überwachung beschränkt sich die Analytik von Epoxidverbindungen bisher auf die Bestimmung rechtlich geregelter Einzelsubstanzen. Eine Summenmethode zur Erfassung aller in einem Migrat enthaltenen Substanzen mit reaktionsfähigen Oxirangruppen, wodurch das gesamte Reaktionspotential erfasst werden kann, liegt dagegen nicht vor. Zur selektiven Erfassung aller oxirangruppenhaltigen Verbindungen wurde daher mittels statistischer Versuchsplanung eine Derivatisierung mittels Cysteamin entwickelt. Die Reaktion mit Cysteamin erfolgt dabei nach Abtrennung der Substanzen > 1000 Da mittels Größenausschlusschromatographie. Im Anschluss werden die Derivate durch Zugabe eines Kationenaustauschers aus der Lösung entfernt. Durch den Vergleich der RP-HPLC-FLD Chromatogramme vor und nach der Aufarbeitung können Substanzen mit intakten Oxirangruppen somit einfach erkannt werden. Eine quantitative Abschätzung der enthaltenen Epoxidverbindungen ist bei Lacken auf Basis von BPA-Harzen über die BPA-Chromophorkonzentration möglich. Bei der Untersuchung von 5 kommerziellen Coatings wurden in den einzelnen Migraten recht unterschiedliche Gehalte an Substanzen mit intakten Epoxidgruppen ermittelt. Ebenso ist die Anzahl der oxriangruppenhaltigen Verbindungen, auf die sich dieser Gehalt verteilt sehr unterschiedlich, was möglicherweise an unterschiedlichen Einbrennzeiten, -temperaturen aber auch der Menge an Lack pro m² und an der Art und Menge des Reaktionspartners liegt. Für die Konzentration der epoxidischen Verbindungen in den Coatingmigraten wurden Werte zwischen 18,5 und 835 µg/dm² bestimmt. Dies entspricht einem Anteil der reaktionsfähigen Substanzen an der Fraktion unter 1000 Da berechnet über die Flächen im Chromatogramm zwischen 2,6 und 76,3 %. Neben dem Einsatz als Basismaterial für Konservendoseninnenbeschichtungen werden Epoxidharzsysteme auch in Zubereitungen wie Grundierungen, Füllmassen, Lacken oder Klebstoffen für die Bauchemie vielfach in verschiedenen Mischungen (aromatische, aliphatische oder cycloaliphatische Glycidylether bzw. Siloxanglycidylether) eingesetzt. Durch den Kontakt dieser Materialien mit der Haut kann es zu Kontaktekzemen kommen, deren Ursache durch Epikutantests (Patchtests) mit den potentiell auslösenden Substanzen ermittelt werden kann. Der Umfang der in den Standardtestsystemen enthalten Testsubstanzen entspricht dabei jedoch nicht dem Spektrum der in der Industrie verwendeten Materialien. Über die genaue Zusammensetzung der in den Patchtest´s eingesetzten Materialien ist zudem wenig bekannt. Durch die Analyse einer Vielzahl von Patchtestsubstanzen und industriellen Epoxidkomponenten mittels RP-HPLC/UVD bzw. -ELSD Chromatographie und die Identifizierung der enthaltenen Verbindungen über RP-HPLC/ESI-MSD, konnte ein genaueres Bild über den Charakter dieser Materialen gewonnen werden. Bei den BPA- und BPF-Harzen wurden überwiegend Monomere, die entsprechenden Di- und Trimere aber auch dessen hydrolysierte Verbindungen identifiziert. Im Gegensatz dazu liegen bei den analysierten aliphatischen Produkten z. T. die reinen Glycidylether gar nicht oder nur in geringen prozentualen Anteilen vor. Vielmehr wird durch die sauer geführte Reaktion bei den aliphatischen Verbindungen die Bildung von 1,3 Chlorhydrinen als Nebenreaktion zur 1,2 Chlorhydrinbildung gefördert, wodurch eine beträchtliche Menge an Substanzen, die nicht verseifbares Chlor enthalten, in den Materialien vorhanden ist. Um die enthaltenen Verbindungen von aliphatischen und cycloaliphatischen Epoxidzubereitungen quantifizieren zu können, wurde eine Derivatisierung mit einem selbst synthetisierten Fluorophor (5-(Dimethylamino)-N-(2-mercaptoethyl)-1-naphthalen-sulfonamid) entwickelt. Dadurch konnten auch Verbindungen erfasst werden, die aufgrund ihrer Flüchtigkeit mittels ELSD nicht detektierbar waren. Bei der Analyse von Handelsprodukten zeigte sich, dass die einzelnen Komponenten in ihrer Zusammensetzung gut mit den untersuchten aromatischen und aliphatischen Rohmaterialien vergleichbar sind. Bei entsprechend eingesetzten Patchtestsubstanzen spiegeln diese somit die Produkte gut wieder, mit denen die Patienten in Kontakt kommen.

Le travail présenté dans cette thèse est une contribution au dimensionnement et à la commande d’un générateur de type Cascade de Machines Asynchrones à Double Alimentation. Le modèle de la machine pour les simulations est basé sur une représentation modulaire qui découle du modèle de Park de deux machines asynchrones doublement alimentées distinctes, en respectant deux systèmes de référence. Le modèle dynamique est ensuite étendu pour décrire la machine dans un repère unifié. Cette représentation est cruciale pour le dimensionnement de la machine et la synthèse de la commande. Le domaine de fonctionnement de la cascade en régime permanent est exploré. Une méthode analytique générique est proposée pour établir les plages de puissances active-réactive. Les courbes limites sont définies par rapport aux grandeurs nominales de la machine. L’étude prend en compte l’effet de saturation du circuit magnétique. Il est prouvé que la capacité en puissance de la machine est déterminée par les valeurs maximales des courants statoriques et peut être soumise à plusieurs limitations. L’approche analytique est testée et validée par des mesures expérimentales. Deux cas sont considérés pour la commande de la cascade : un générateur raccordé au réseau et un générateur autonome. Pour la connexion au réseau l’application visée est les éoliennes de grande puissance. Une nouvelle méthode pour l’extraction de la puissance maximale d’une éolienne à vitesse variable est proposée. Le système est contrôlé dans le but de fournir une puissance maximale quasi-constante indépendamment des fluctuations du vent. En plus de l’optimisation de la puissance, le facteur de puissance est également ajusté selon les normes de raccordement standard imposées par les services système, et les limites de fonctionnement de l’unité intégrée. Une commande vectorielle sans capteur de tension, basée sur une orientation suivant un flux virtuel, est appliquée pour la régulation découplée des puissances active et réactive. Pour le fonctionnement en mode isolé, l’application visée est la génération électrique pour système avionique embarqué. Deux réseaux de distribution sont traités : réseau AC à fréquence fixe et réseau DC. Dans les deux cas, la commande est élaborée dans le but de maintenir une tension de sortie constante. Le fonctionnement de la cascade alimentant une charge triphasée déséquilibrée est également étudié. Des schémas de contrôle servant à la compensation du déséquilibre des tensions statoriques sont développés. Deux méthodes de compensation sont proposées : l’une basée sur le principe de deux repères tournants et l’autre sur un correcteur répétitif. Les approches présentées sont validées par simulation avec Matlab/Simulink et par expérimentation
This thesis is a contribution to the power sizing and the control of a Cascaded Doubly Fed Induction Generator (CDFIG). The machine model for simulation is based on a modular representation, derived from the Park model of two distinct Doubly Fed Induction Machines expressed in their own reference frames. The dynamic model of the cascaded machine is then extended to be described in a unified reference frame. This representation is convenient for the sizing of the machine and for the design of the controller. The steady state power operating margins of the cascaded machine are investigated. A generic analytic method is suggested to derive the active-reactive power domain. The limit curves are defined in terms of the rated quantities of the machine. The study takes into account the magnetic circuit saturation effect. It is proven that the power capability of the machine is determined by the stator current maximum values and is subject to several limitations. The analytical approach is tested and validated by experimental measurements. The CDFIG is controlled in grid-connected and standalone operation modes. In grid-connected application, the attention is paid to high power wind generation systems. A new maximum power tracking of a variable speed wind turbine is suggested. The generating plant is carried to provide a quasi-constant maximum power regardless wind fluctuations. In addition to active power optimization the power factor is adjusted according to the grid code requirements and the operating domain of the integrated unit. A virtual flux oriented vector control is applied for the decoupled regulation of active and reactive powers, leading to grid voltage sensorless operation. In standalone operating mode, the study treats in particular the embedded aircraft power system generation. Two distribution networks are considered: constant frequency AC network and DC network. In both cases, the control aims to maintain a constant output voltage. The operation of the CDFIG supplying unbalanced three-phase load is further explored. Control schemes to cope with unbalanced stator voltage conditions are developed. Two compensation methods are elaborated: the first dealing with dual rotating frames and the second is based on repetitive controller. Simulations with Matlab/Simulink software and experiments validate the control approaches

Made available in DSpace on 2014-12-17T15:01:57Z (GMT). No. of bitstreams: 1 LeocadiaTCB.pdf: 1594369 bytes, checksum: da7a2bc5860922178cae36e104bfbd40 (MD5) Previous issue date: 2006-10-21
Conselho Nacional de Desenvolvimento Cient?fico e Tecnol?gico
Effluent color resulting from textile dyeing processes has been one of the biggest environmental problems faced by the textile industry. In particular, reactive dyes are highly resistant to conventional wastewater treatment methods. New technologies have been contemplated, some of which have been applied in industrial treatment plants, but color removal has not been efficiently attained. Since microemulsion systems provide good results in heavy metals and proteins extraction processes, their use in dyes extraction has been suggested and investigated. In this work, a real textile wastewater from an exhaustion dyebath has been treated, which contains the following reactive dyes: Procion Yellow H-E4R (CI Reactive Yellow 84), Procion Blue H-ERD (CI Reactive Blue 160) and Procion Red H-E3B (CI Reactive Red 120), in addition to auxiliary compounds normally found in dyeing processes with reactive dyes. The dyes Remazol Blue RR and Remazol Turquoise Blue G (Reactive Blue 21) have also been examined in view of the presence of heavy metals in these molecules. The microemulsion system comprised dodecyl ammonium chloride (as a cationic surfactant), water or wastewater as aqueous phase, kerosene as oil phase, and one of the following alcohols as cosurfactant: isoamyl alcohol, n-butyl alcohol and n-octyl alcohol. The pseudo-ternary diagrams were constructed in order to define Winsor s equilibrium regions. The influence of parameters such as pH, C/S (cosurfactant/surfactant) ratio, distribution coefficient, initial dye concentration, salinity, temperature, phases relative amounts, loading capacity of the microemulsion phase and dye reextraction rate has also been investigated. An experimental planning (Scheff? Net) was used to optimize the extraction process. The removal of color and metals reached levels as high as 99%
A cor do efluente resultante dos processos de tingimento tem sido um dos principais problemas ambientais enfrentados pela ind?stria t?xtil. De modo especial, efluentes contendo corantes reativos s?o altamente resistentes aos processos de tratamento convencionais. Novas tecnologias t?m sido buscadas, algumas j? em escala industrial, por?m nem sempre ? poss?vel atingir a efici?ncia desejada. Por serem utilizadas de forma eficiente em processos de extra??o de metais e de prote?nas, buscou-se utilizar as microemuls?es na extra??o de corantes. Para este estudo, um efluente real foi examinado, consistindo no banho de exaust?o de um processo de tingimento contendo os seguintes corantes: Procion Amarelo H-E4R (CI Reactive Yellow 84), Procion Azul H-ERD (CI Reactive Blue 160) e Procion Vermelho H-E3B (CI Reactive Red 120), al?m de auxiliares normalmente encontrados em processos de tingimento com corantes reativos. Para estudar a remo??o de metais ligados ?s mol?culas dos corantes, utilizaram-se ainda os corantes Remazol Azul RR e Remazol Azul Turquesa G (Reactive Blue 21). Os sistemas de microemuls?o foram formados pelo cloreto de dodecilam?nio (tensoativo cati?nico), ?gua ou efluente como fase aquosa, querosene como fase oleosa e um dos seguintes ?lcoois como cotensoativos: ?lcool isoam?lico, n-butanol e n-octanol. Os diagramas pseudo-tern?rios, representativos dos sistemas microemulsionados em estudo, foram desenvolvidos a fim de delimitar as regi?es de exist?ncia de equil?brio de fases (sistema de classifica??o Winsor). Verificou-se a influ?ncia de par?metros como: pH, raz?o C/T (cotensoativo/tensoativo), coeficiente de distribui??o, concentra??o inicial de corante, salinidade, temperatura, rela??o das fases, capacidade de carga da fase de microemuls?o e reextra??o do corante. Uma metodologia de planejamento experimental (Rede de Scheff?) foi utilizada para otimizar a extra??o. A remo??o da cor e de metais alcan?ou ?ndices de extra??o superiores a 99%

The ever increasing quantity of chemical literature necessitates the creation of automated techniques for extracting relevant information. This work focuses on two aspects: the conversion of chemical names to computer readable structure representations and the extraction of chemical reactions from text. Chemical names are a common way of communicating chemical structure information. OPSIN (Open Parser for Systematic IUPAC Nomenclature), an open source, freely available algorithm for converting chemical names to structures was developed. OPSIN employs a regular grammar to direct tokenisation and parsing leading to the generation of an XML parse tree. Nomenclature operations are applied successively to the tree with many requiring the manipulation of an in-memory connection table representation of the structure under construction. Areas of nomenclature supported are described with attention being drawn to difficulties that may be encountered in name to structure conversion. Results on sets of generated names and names extracted from patents are presented. On generated names, recall of between 96.2% and 99.0% was achieved with a lower bound of 97.9% on precision with all results either being comparable or superior to the tested commercial solutions. On the patent names OPSIN s recall was 2-10% higher than the tested solutions when the patent names were processed as found in the patents. The uses of OPSIN as a web service and as a tool for identifying chemical names in text are shown to demonstrate the direct utility of this algorithm. A software system for extracting chemical reactions from the text of chemical patents was developed. The system relies on the output of ChemicalTagger, a tool for tagging words and identifying phrases of importance in experimental chemistry text. Improvements to this tool required to facilitate this task are documented. The structure of chemical entities are where possible determined using OPSIN in conjunction with a dictionary of name to structure relationships. Extracted reactions are atom mapped to confirm that they are chemically consistent. 424,621 atom mapped reactions were extracted from 65,034 organic chemistry USPTO patents. On a sample of 100 of these extracted reactions chemical entities were identified with 96.4% recall and 88.9% precision. Quantities could be associated with reagents in 98.8% of cases and 64.9% of cases for products whilst the correct role was assigned to chemical entities in 91.8% of cases. Qualitatively the system captured the essence of the reaction in 95% of cases. This system is expected to be useful in the creation of searchable databases of reactions from chemical patents and in facilitating analysis of the properties of large populations of reactions.

The spectacular increase in lithium battery applications has raised the question of whether global lithium resources will be enough in the future. Experts in the field have estimated that the existing lithium resources will probably be sufficient to support demands until the year 2100, assuming that lithium batteries are recycled. Without lithium batteries being recycled, the resources are expected to be depleted in 50 years’ time. Therefore, there is a great interest in developing better methods of lithium recycling from batteries, and also, better methods of lithium extraction from natural resources. Currently, lithium is extracted from natural brines via the lime soda evaporation process, i.e. a solar evaporation plus chemical plant process, which takes between 12 and 24 months. The drawbacks of this process are that it is complex, slow and inefficient. Also, the currently available methods of lithium recycling from batteries are too complex and expensive. Thus, the main objective of this work is to develop a novel, inexpensive and less timeconsuming approach to recover lithium chemically, from the lithium salts (lithium sources) that contain other metal cations. The new process is also based on environmental concerns. A battery material, lithium iron phosphate (LiFePO4) has the olivine structure and heterosite structure once it discharges to iron phosphate (FePO4). This structure shows excellent properties of the charge/discharge reversibility. A few studies on the heterosite FePO4 have reported that it is more selective for lithium ions (Li+) over other cations. The main advantages of this structure are the small potential differences of the redox couple, i.e. Fe(II)/Fe(III), and the stability of LiFePO4 over a wide range of acid-based conditions in an aqueous solution. This work investigates a novel process that may be superior to the lime soda evaporation process for extracting lithium. Heterosite FePO4 was employed to selectively remove Li+ from lithium sources with the support of a reducing agent, i.e. sodium thiosulphate (Na2S2O3). The resulting LiFePO4 can be directly sent not only to lithium battery industries, but also to other industrial uses. In principle, the other cations could be retrieved back into their sources. The novel process was examined and demonstrated lithium insertion into a heterosite FePO4, working as a framework, in aqueous salt solutions. The evaluation of this process is presented by the Li+ uptake value. The amount of Li+ uptake can be up to 46 mgLi +/gsolid where other cations (i.e. sodium, potassium, and magnesium) can take less than 3 mg/gsolid, using this process. Furthermore. This work could also be developed for future lithium recycling processes.

There is a transition nationwide on the analysis of forensic evidentiary stains containing biological material from traditional serology to Polymerase Chain Reaction (PCR) methodologies. The increased sensitivity of PCR, the limited number of alleles at each locus, and the necessity of producing unambiguous data for entry into the FBI's Combined DNA Index System make this study of extraction procedures of utmost importance. A "single tube" extraction procedure for blood stains collected onto FTA™ paper and a modified differential nonorganic extraction method from spermatozoa containing mixed stains were analyzed and compared. The extraction success was evaluated by amplification and typing of the amplified fragment length polymorphism, D1S80. These modifications of the nonorganic method utilized gave an improved separation of the spermatozoa-containing mixed stains.

Inom ramen för Världshälsoorganisationens (WHO:s) internationella biverkningsprogram rapporterar sjukvårdspersonal och patienter misstänkta läkemedelsbiverkningar i form av spontana biverkningsrapporter som via nationella myndigheter skickas till Uppsala Monitoring Centre (UMC). Hos UMC lagras rapporterna i VigiBase, WHO:s biverkningsdatabas. Rapporterna i VigiBase analyseras med hjälp av statistiska metoder för att hitta potentiella samband mellan läkemedel och biverkningar. Funna samband utvärderas i flera steg där ett tidigt steg i utvärderingen är att studera den medicinska litteraturen för att se om sambandet redan är känt sedan tidigare (tidigare kända samband filtreras bort från fortsatt analys). Att manuellt leta efter samband mellan ett visst läkemedel och en viss biverkan är tidskrävande. I den här studien har vi utvecklat ett verktyg för att automatiskt leta efter medicinska biverkningstermer i medicinsk litteratur och spara funna samband i ett strukturerat format. I verktyget har vi implementerat och integrerat funktionalitet för att söka efter medicinska biverkningar på olika sätt (utnyttja synonymer,ta bort ändelser på ord, ta bort ord som saknar betydelse, godtycklig ordföljd och stavfel). Verktygets prestanda har utvärderats på manuellt extraherade medicinska termer från SPC-texter (texter från läkemedels bipacksedlar) och på biverkningstexter från Martindale (medicinsk referenslitteratur för information om läkemedel och substanser) där WHO-ART- och MedDRA-terminologierna har använts som källa för biverkningstermer. Studien visar att sofistikerad textextraktion avsevärt kan förbättra identifieringen av biverkningstermer i biverkningstexter jämfört med en ordagrann extraktion.
Background: Initial review of potential safety issues related to the use of medicines involves reading and searching existing medical literature sources for known associations of drug and adverse drug reactions (ADRs), so that they can be excluded from further analysis. The task is labor demanding and time consuming. Objective: To develop a text extraction tool to automatically identify ADR information from medical adverse effects texts. Evaluate the performance of the tool’s underlying text extraction algorithm and identify what parts of the algorithm contributed to the performance. Method: A text extraction tool was implemented on the .NET platform with functionality for preprocessing text (removal of stop words, Porter stemming and use of synonyms) and matching medical terms using permutations of words and spelling variations (Soundex, Levenshtein distance and Longest common subsequence distance). Its performance was evaluated on both manually extracted medical terms (semi-structuredtexts) from summary of product characteristics (SPC) texts and unstructured adverse effects texts from Martindale (i.e. a medical reference for information about drugs andmedicines) using the WHO-ART and MedDRA medical term dictionaries. Results: For the SPC data set, a verbatim match identified 72% of the SPC terms. The text extraction tool correctly matched 87% of the SPC terms while producing one false positive match using removal of stop words, Porter stemming, synonyms and permutations. The use of the full MedDRA hierarchy contributed the most to performance. Sophisticated text algorithms together contributed roughly equally to the performance. Phonetic codes (i.e. Soundex) is evidently inferior to string distance measures (i.e. Levenshtein distance and Longest common subsequence distance) for fuzzy matching in our implementation. The string distance measures increased the number of matched SPC terms, but at the expense of generating false positive matches. Results from Martindaleshow that 90% of the identified medical terms were correct. The majority of false positive matches were caused by extracting medical terms not describing ADRs. Conclusion: Sophisticated text extraction can considerably improve the identification of ADR information from adverse effects texts compared to a verbatim extraction.

Proximate analysis of the commercial shrimp meat waste samples indicated the presence of 94.6% protein and 4.2% fat on a dry basis. The HPLC analysis revealed the presence of 17 amino acids (proline most abundant) and 7 sugars (ribose being the most abundant). The changes in the concentration of the amino acids and sugars after heat treatment, are explained based on their interaction through the Maillard reaction and by the thermal hydrolysis of proteins and polysaccharides present in tissue. Carotenoid pigments ($ beta$-carotene, canthaxanthin, astacene, and astaxanthin monopalmitate) were extracted and separated by thick layer chromatography. Their structures were identified tentatively by thick layer chromatography (TLC) and subsequently confirmed by Fourier transformed infrared (FT-IR) spectroscopy. Lipids were separated and categorized into glycerides, sterols, and phospholipids by thin layer chromatography while flavor-active components were extracted by different organic solvents into acidic, basic, and neutral fractions and their sensory properties determined. The flavor profile of the raw and cooked shrimp meat waste revealed the presence of 59 compounds in raw shrimp compared to 44 in cooked. The components identified were of different functional groups, including fatty acid esters, long chain alcohols, aldehydes, ketanes and hydrocarbons. Many fractions provided desirable aroma (nutty, fruity, floral, green woody, meaty) indicating the presence of important flavor compounds in the commercial shrimp waste.
Model studies were carried out under different experimental conditions (water content, pH), to observe the influence of these parameters on the formation of Maillard reaction products. Furthermore, a comparative study between direct injection of the aqueous solutions and its organic extracts was conducted. The direct injection of aqueous solutions revealed the presence of many water soluble components that were not detected by organic solvent extraction such as droserone and 3,9-diazatricyclo$ lbrack7.3.0.0 sp{*}.3 sp{*}.7 sp{ *} rbrack$-dodecan-2,8-dione while maltoxazine and acetanilide are example of compounds identified only in the dichloromethane extracts. Results also demonstrated that monosaccharides affected the yield of the compounds produced whereas the amino acids influenced the sensory properties of these models. Fructose and ribose were found to be the most reactive sugars in their respective categories. Thiazoles were found to be more water soluble whereas $ beta$-carbolines were more abundant in the organic extracts. The influence of water content was investigated in model systems containing fructose or glucose. It was found that water free systems decreased the yield of pyrazines and thiazoles while the formation of pyrrole and pyridine derivatives increased under dry heating conditions. The formation of Maillard reaction products were studied under different pH conditions in order to assess the effect of pH on the formation of heterocyclic compounds. Pyrazine and furan derivatives were the most affected by pH and alanine was found to be a better nitrogen source than lysine in the formation of pyrazines under different pH conditions. Finally, based on energy minimization studies using MM2 force field, a plausible mechanism was proposed for the formation of maltol in Maillard model systems.

In product development there is still potential to decrease lead times with faster and more accurate simulations. The objective of this thesis was to study whether Finite Element (FE) simulations using explicit LS-DYNA to extract reaction forces from sheet metal forming tools during forming, could be used to improve existing FE models in sheet metal forming software AutoForm.To begin with, the solid CAD-model of the stamping dies were meshed with tetrahedral elements in CATIA and imported into LS-DYNA. In combination with sheet mesh and milling surface meshes from AutoForm, an explicit model was realized. Contacts between sheet mesh and milling surface meshes used the so-called sheet forming contact. The resulting reaction forces were extracted and used in a simulation using the AutoForm software. Resulting simulation was compared to a scan of the physical sheet metal after forming.The direct transfer of reaction forces from LS-DYNA to AutoForm did however not result in the same pressure distribution in AutoForm. The AutoForm simulations using results from LS-DYNA were slightly worse than standard AutoForm simulations.Further work is needed to try and perhaps implement an implicit solution after an initial explicit solution.
Inom produktutveckling finns möjligheter att förkorta ledtider genom snabbare och mera korrekta simuleringar. Syftet med detta arbetet var att undersöka huruvida resultat från explicit LS-DYNA kunde användas för att förbättra nuvarande plåtformningssimuleringar i AutoForm.Den solida CAD-modellen av verktyget meshades med tetraediska element i CATIA och importerades till LS-PrePost, tillsammans med fräsytsmeshar och plåtmesh från AutoForm. Kontakter etablerades mellan plåt och fräsytsmeshar med så kallad sheet forming contact. Modellen löstes sedan explicit. Resulterande reaktionskrafter på plåthållare exporterades till AutoForm och implementerades där. Resulterande simulering jämfördes mot en inskannad fysisk plåt efter plåtformning.Direkt implementering av reaktionskrafter på plåthållaren i AutoForm gav resultat som avvek mer mot inskannad plåt än nuvarande simuleringsstrategi. Direkt implementering av reaktionskrafter gav heller inte en tryckfördelning som liknade den som rapporterades av LS-DYNA.Mer arbete krävs för att om möjligt implementera en implicit lösning efter en initial explicit lösning.

There are numerous challenges encountered during the manufacturing and storage of cheese by both the large-scale and artesian producers. One such challenge has been the formation of late gassy defect, which occurs when gas is produced by certain lactic acid bacteria found in the cheese block during storage and aging over a three month time period. Negative consequences of late gas production are slits and cracks in the cheese block and puffy cheese packaging, which cause significant financial losses for manufacturers along with poor consumer acceptance. Lactobacillus wasatchensis is one such lactic acid bacterium shown to produce gas during cheese storage. This bacterium has now been found in cheese samples exhibiting late gas defect in the Midwest and Western states. The goal of this study was to further characterize and understand the growth attributes of Lb. wasatchensis, and thereby gain some understanding on how it enters the cheese vats and if there are possible ways to limit or inhibit its subsequent growth. An additional goal was to determine if we could effectively extract Lb. wasatchensis DNA from cheese samples and visualize using the qPCR molecular technique. If possible, this detection method would allow a faster and more sensitive approach to determining if Lb. wasatchensis is present in cheese blocks, which would help manufacturers know how long they should age their cheeses. It was discovered that Lb. wasatchensis does not survive processing through an industrial heat exchanger and therefore must be entering the cheese vats by other means such as: cross-contamination, biofilm formation or aerosolizing. We also showed growth of Lb. wasatchensis is limited at an increased salt-in-moisture ratio in cheese. Additionally, we found that Lb. wasatchensis DNA can be extracted from cheese and visualized using qPCR, although further experimentation is needed to optimize this method.

During in situ groundwater remediation, a chemical or biological amendment is introduced into the aquifer to degrade the groundwater contaminant. In this type of remediation, mixing of the amendment and the contaminated groundwater, through molecular diffusion and pore-scale dispersion, is necessary for reaction to occur. Since the length scale of dispersion is small compared to the size of the contaminant plume, reactions are limited to a relatively narrow region where the amendment and contaminant are close enough to mix. Spreading, defined as the reconfiguration of the plume shape due to spatially-varying velocity fields, increases the size of the region where reaction occurs and increases concentration gradients, both of which can lead to enhanced mixing and reaction. Spreading can occur passively by heterogeneity of hydraulic conductivity or actively by engineered injection and extraction (EIE), in which clean water is injected or extracted at an array of wells surrounding the contaminant plume. Several studies have shown that active spreading by EIE enhances contaminant degradation in homogeneous porous media compared to remediation without EIE. Furthermore, studies have also shown that combining EIE with passive spreading by heterogeneity can lead to even more degradation compared to EIE alone. In this study, we investigate the relationship between passive and active spreading to better understand their combined impact on mixing and reaction during EIE. Using various combinations of heterogeneity patterns (e.g., high and low hydraulic conductivity inclusions) and simple injection and extraction flow fields typical to EIE, we determine how the particular spreading of the amendment and contaminant plume under each heterogeneity/flow field combination controls the amount of mixing and reaction enhancement. We find that the injection and extraction flow fields can be designed to complement the topological features generated from specific heterogeneity structures, thereby increasing the amount of mixing and reaction enhancement. Since the subsurface is inherently heterogeneous, insights gained from this research will provide crucial information for the optimal design of EIE systems in the field.

Stochastic reaction-diffusion simulations has become an efficient approach for modelling spatial aspects of intracellular biochemical reaction networks. By accounting for intrinsic noise due to low copy number of chemical species, stochastic reaction-diffusion simulations have the ability to more accurately predict and model biological systems. As with many simulations software, exploration of the parameters associated with the model can be needed to yield new knowledge about the underlying system. The exploration can be conducted by executing parameter sweeps for a model. However, with little or no prior knowledge about the modelled system, the effort for practitioners to explore the parameter space can get overwhelming. To account for this problem we perform a feasibility study on an explorative behavioural analysis of stochastic reaction-diffusion simulations by applying spatial-temporal data mining to large parameter sweeps. By reducing individual simulation outputs into a feature space involving simple time series and distribution analytics, we were able to find similar behaving simulations after performing an agglomerative hierarchical clustering.

Cette étude a pour objectif la caractérisation (hydrodynamique, propriétés d'émulseur et de réacteur chimique) de la zone de mélange d'un extracteur centrifuge mettant à profit l'écoulement de Couette-Taylor. Ce type d'appareil potentiellement utilisable pour certaines opérations du retraitement des combustibles nucléaires permet de réaliser des opérations d'extraction liquide-liquide avec réaction chimique

Diploma thesis deals with the design of the condensing steam turbine with power output 23.63 MW at 5000 rpm and an efficiency of 84.17% after optimization. The turbine contains one controlled steam extraction and four bleeds. The steam mass flow required by controlled steam extraction is 0-75 t/h of a total mass flow to the turbine 105 t/h. Design of the steam turbine is preceded by calculation of heat balance of the cycle. Then the thermodynamic and strength calculation of regulation stage and reaction staging flow channel are made. Thermodynamic calculation of the flow channel consists of preliminary and detailed calculation. It is detailed calculation that leads to the turbine parameters optimization. The design also includes gland sealing system and balance piston calculation. In addition to the strength calculation there is bearings load calculation included. In the end, the consumption diagram is made, showing the interval by which the turbine can be safely operated. Integrated part of the thesis is conceptual drawing of steam turbine section.

Chromium can act either as an essential micronutrient or a chemical carcinogen, depending on its oxidation state. Of the two most stable chromium oxidation states, Cr(VI) is a known carcinogen, while Cr(III) is an essential element. Determination of total Cr cannot therefore be used to evaluate food safety of dairy products, as speciation is required for accurate assessment of potential toxicity. In this study, total Cr was determined by DRC-ICP-MS after mineralisation of freeze-dried milk samples. Chromabond NH2 ion-exchange columns were used to separate Cr(VI) from Cr(III). During the separation process, Cr(VI) was selectively adsorbed onto the Chromabond NH2 column. The retained Cr(VI) was subsequently eluted with two column volumes of 2 M HNO3 and diluted to a final volume of 10 mL. After separation, the Cr(VI) was quantified by DRC-ICP-MS, using O2 as the reactive gas, to alleviate polyatomic ion interference. An O2 gas flow rate of 0.85 ml/min and a high-pass rejection parameter q (Rpq) ¼ 0.5, provided the best compromise signal-to-noise ratio. The method was validated through determining the recovery in milk samples spiked with Cr(VI). The recoveries were quantitative and ranged from 96.4 to 99.2%. A whole milk powder reference material (NIST SRM 8435) was used to evaluate the accuracy of the total Cr quantification method. The observed value showed concordance with the certified value. The limits of detection (LOD) were 0.091 and 0.085 mg/L for total Cr and Cr(VI), respectively. These findings are important for the determination of maximum residue levels (MRLs) of Cr(VI) in dairy products.

Magister Scientiae (Biodiversity and Conservation Biology)
Land-based abalone aquaculture in South Africa started in the early 1990s and is based on the local species Haliotis midae. This industry expanded with great success over the last decade. In 2006 abalone exhibiting typical clinical signs of tubercle mycosis was discovered for the first time in South African abalone culture facilities,posing a significant threat to the industry. Halioticida noduliformans, a fungus belonging to the Peronosporomycetes (formerly Oomycetes), has been identified as the causative agent of abalone tubercle mycosis (ATM). While diagnoses of this disease are currently done by gross observation and histopathology, these methods fail to be sensitive enough to identify the causative agent accurately and reliably.Molecular confirmation could provide for quicker more accurate diagnostic information. The aim of this study was to develop a DNA based molecular diagnostic test. Polymerase chain reaction (PCR) has been used to rapidly detect, characterise and identify a variety of organisms. Nucleotide sequences of the smalland large-subunit ribosomal ribonucleic acid (rRNA) and mitochondrial cytochrome oxidase subunit II (cox2) genes of H. noduliformans were compared with closely related Peronosporomycete gene sequences to identify potential PCR primer sites. H. noduliformans specific real-time quantitative PCR (Q-PCR) primer sets were designed and optimised for each of the selected genes. Results indicate that, although all tested primers sets could amplify fungal DNA, only the LSU and cox2 primer sets - v -demonstrated no cross-amplification with the closely related Peronosporomycete and non-fungal DNA tested in the present study. The H. noduliformans specific LSU primer set was chosen for further analysis and used for all subsequent real-time PCR assays. The lowest detection limit for the LSU primer set was evaluated by running Q-PCR on serial dilutions of known quantities of extracted H. noduliformans DNA.Serial dilutions were made in PCR grade water as well as in an abalone tissue matrix.The sensitivity of the Q-PCR reaction was determined to be 266 pg of H.noduliformans DNA per 25 μL reaction volume. However, inclusion of a nested PCR step, utilising universal fungal outer primers, followed by Q-PCR with the H.noduliformans LSU specific primers improved sensitivity to 0.266 pg of H.noduliformans DNA per 25 μL reaction volume. This equates to approximately 2.4spores per 25 μL reaction volume. DNA extraction protocols were optimised to ensure efficient and repeatable extraction of high quality fungal DNA from pure fungus and tissue samples spiked with known quantities of fungal DNA. PCR amplification efficiency and potential inhibition were examined for each extraction method. Results suggest that real-time PCR has great potential in monitoring and quantifying H. noduliformans on abalone culture facilities in South Africa.

El entendimiento de los fenómenos de transferencia de calor y de masa en medios porosos implica el estudio de modelos de transporte de fluidos en la fracción vacía del medio; este hecho es de fundamental importancia en muchos sistemas de Ingeniería Química, tal como en procesos de extracción o en reactores catalíticos. Los estudios de flujo realizados hasta ahora (teóricos y experimentales) usualmente tratan al medio poroso como un medio efectivo y homogéneo, y toman como válidas las propiedades medias del fluido. Este tipo de aproximación no tiene en cuenta la complejidad del flujo a través del espacio vacío del medio poroso, reduciendo la descripción del problema a promedios macroscópicos y propiedades efectivas. Sin embargo, estos detalles de los procesos locales de flujo pueden llegar a ser factores importantes que influencien el comportamiento de un proceso físico determinado que ocurre dentro del sistema, y son cruciales para entender el mecanismo detallado de, por ejemplo, fenómenos como la dispersión de calor, la dispersión de masa o el transporte entre interfaces.

La Dinámica de Fluidos Computacional (CFD) como herramienta de modelado numérico permite obtener una visión mas aproximada y realista de los fenómenos de flujo de fluidos y los mecanismos de transferencia de calor y masa en lechos empacados, a través de la resolución de las ecuaciones de Navier - Stokes acopladas con los balances de materia y energía y con un modelo de turbulencia si es necesario. De esta forma, esta herramienta permite obtener los valores medios y/o fluctuantes de variables como la velocidad del fluido, la temperatura o la concentración de una especie en cualquier punto de la geometría del lecho empacado.

El objetivo de este proyecto es el de utilizar programas comerciales de simulación CFD para resolver el flujo de fluidos y la transferencia de calor y de masa en modelos bi/tri dimensionales de lechos empacados, desarrollando una estrategia de modelado aplicable al diseño de equipos para procesos de extracción o de reacción catalítica. Como referencia se tomaran procesos de tecnología supercrítica debido a la complejidad de los fenómenos de transporte involucrados en estas condiciones, así como a la disponibilidad de datos experimentales obtenidos previamente en nuestro grupo de investigación. Estos datos experimentales se utilizan como herramienta de validación de los modelos numéricos generados, y de las estrategias de simulación adoptadas y realizadas durante el desarrollo de este proyecto.
An understanding of the heat and mass transfer phenomena in a porous media implies the study of the fluid transport model within the void space; this fact is of fundamental importance to many chemical engineering systems such as packed bed extraction or catalytic reaction equipment. Experimental and theoretical studies of flow through such systems often treat the porous medium as an effectively homogeneous system and concentrate on the bulk properties of the flow. Such an approach neglects completely the complexities of the flow within the void space of the porous medium, reducing the description of the problem to macroscopic average or effective quantities. The details of this local flow process may, however, be the most important factor influencing the behavior of a given physical process occurring within the system, and are crucial to understanding the detailed mechanisms of, for example, heat and mass dispersion and interface transport.

Computational Fluid Dynamics as a simulation tool allows obtaining a more approached view of the fluid flow and heat and mass transfer mechanisms in fixed bed equipment, through the resolution of 3D Reynolds averaged transport equations, together with a turbulence model when needed. In this way, this tool permit to obtain mean and fluctuating flow and temperature values in any point of the bed.

The goal of this project is to use commercial available CFD codes for solving fluid flow and heat and mass transfer phenomena in two and three dimensional models of packed beds, developing a modeling strategy applicable to the design of packed bed chemical reaction and extraction equipment. Supercritical extraction and supercritical catalytic reaction processes will be taken as reference processes due to the complexity of the transport phenomena involved within this processes, and to the availability of experimental data in this field, obtained in the supercritical fluids research group of this university. The experimental data priory obtained by our research group will be used as validation data for the numerical models and strategies dopted and followed during the developing of the project.

Une microemulsion a base de nadehp utilisee en extraction liquide-liquide favorise le transfert de masse, par rapport a la structure monomerique, tant que ses micelles contiennent tres peu d'eau. L'examen de sa structure par diffusion de rayons x aux petits angles et par conductimetrie fixe les conditions optimales d'etude de l'extraction. A l'aide de la methode de diffusion x on teste le comportement de la solution micellaire lorsque'elle est superposee a l'eau, puis on etudie l'evolution de sa microstructure lors d'une extraction liquide-liquide en la mettant en contact avec une solution de sulfate de cuivre. Dans les deux cas, une disparition des micelles est mise en evidence et une valeur de l'epaisseur de la couche de diffusion dans la phase organique est determinee. On montre que la reactivite des micelles est effectivement reduite lorsqu'on les gonfle d'eau et on etablit le lien avec leur structure qui s'avere etre le phenomene responsable de la plus ou moins bonne performance de la microemulsion extractante

Introduction: Cervical squamous cell carcinoma is most commonly caused by persistent infection by high risk human papillomavirus (hrHPV) genotypes. The exact type of hrHPV varies geographically and is the basis for HPV–based vaccination for cervical squamous cell carcinoma prevention. Little is known regarding local hrHPV genotypes within the Western Cape population of South Africa. Aims and objectives: This was a pilot study aiming to extract of high quality genomic DNA from archival FFPE cervical squamous cell carcinoma cases and identify hrHPV genotypes by multiplex real time PCR (RT-PCR). Materials and methods: A retrospective search identified a total of 57 cases of cervical squamous cell carcinoma for the period 2004-2014. This was reduced to a final number of 23 that exhibited sufficient tumour burden for DNA extraction. The most common age group was 40-49 years. HIV status was as follows: two HIV-positive, 14 HIV-negative and 7 HIV unknown. DNA was extracted from archival FFPE cervical squamous cell carcinoma samples using QIAGEN QIAamp® DNA FFPE Tissue kit. Housekeeping genes were detected by endpoint PCR using standard primers for glyceraldehyde 3-phosphate dehydrogenase (GAPDH) sequences to determine the quality and integrity of extracted DNA for downstream PCR amplification experiments. HrHPV DNA amplification was optimised using a touchdown PCR technique with L1 consensus gene GP5+/GP6+ primers. HrHPV genotypes were detected using a four colour multiplex hrHPV genotyping kit. Samples showing positive results in overlapping probe filter detection spectra were subjected to DNA Sanger sequencing for final confirmation of specific hrHPV genotype. Results: Standard xylene DNA extraction methods using QIAamp® system yielded adequate amounts of DNA with average final concentration of 463.2 ng/l and A260/A280 ratio of 1.86. Housekeeping genes were successfully detected in all samples, confirming that no significant DNA degradation of target sequences occurred within the archival time range of 2004-2014. HPV L1 detection via GP5+/GP6+ primers with endpoint PCR was not achieved via standard cycling conditions and required the use of a touchdown technique with gradually decreasing annealing temperatures. This method successfully identified HPV L1 sequences in 22 out of 23 cases. Multiplex RT-PCR with four colour hydrolysis probes identified hrHPV genotypes in 22 of 23 cases with relative frequencies of HPV genotypes: 16>>18=39=45>33. Most cases showed infection with a single hrHPV genotype (HPV 16 and one case with HPV 33) with four cases demonstrating two genotypes (two with HPV 16&18, one with 16&33 and one with 39&45) and one case with three genotypes (HPV 16, 39, 45). Interestingly, none of the HIV-positive cases showed multiple hrHPV genotype infection. Four hrHPV cases with overlapping spectra for HPV 18/31 and 45/59 were subjected to Sanger sequencing for confirmation of genotype. Three of four cases showed 100% match for genotypes 18 and 45 with the final case demonstrating only co-infective HPV 16.Conclusion: Commercial DNA extraction kits yield adequate amounts of intact, amplifiable DNA in archival FFPE cervical carcinoma specimens. Touchdown PCR is necessary for HPV detection in extracted FFPE DNA cases using GP5+/GP6+ L1 primers. RT-PCR using multicolour hydrolysis probes is a rapid, sensitive technique for hrHPV genotype screening of cervical squamous cell carcinoma specimens. A three colour detection system rather than four colour kit is recommended for future studies in order to avoid extra cost in DNA sequencing cases with overlapping spectra. This pilot study demonstrates hrHPV genotype prevalence similar to that in other populations and suggests that vaccination with currently available formulations would provide a sufficiently wide coverage of HPV genotypes. Future studies will include application of the FFPE DNA extraction, endpoint PCR and RT-PCR techniques to the remainder of the cases in the original cohort.

This dissertation details three studies dealing with the manipulation of nucleicacids. In the first investigation, each of the four natural nucleobases were analyzed for theability to serve as a universal template at the ligation junction of a T4 DNA ligasereaction. This resulted in the first instance of sequence-independent ligation catalyzed byany DNA ligase. Although all of the nucleobases display universal templatingcapabilities, thymidine and guanosine provided the most effective results. In addition,lowered MgCl2 and ATP concentrations, as well as the inclusion of DMSO, also aided inthe sequence-independent ligation reported here. In the course of these studies, currentmethods of removing urea from denaturing-gel purified nucleic acids provedcumbersome. Therefore, in the second study simple butanol extraction was examined as ameans to eliminate urea from nucleic acid solutions. Stepwise butanol extraction was themost effective approach to solving this problem and provided a much needed techniquefor nucleic acid purification. This type of extraction also does not result in significantlosses of nucleic acid sample. The third study exploits the molecular recognition andcatalytic properties inherent in an autocatalytic group I intron to develop a ribozyme thatcan replace the 5' end of an RNA substrate with a different RNA. This 5' replacementsplicing reaction can potentially repair mutations on the 5' ends of RNA transcripts thatlead to a variety of genetic mutations. The model system was a common mutation in asmall model mimic of the k-ras gene in vitro, which predisposes individuals to lungcancer. This 5' replacement splicing reaction occurred in vitro using this small modelsystem; the reaction was also enhanced by the alteration of the molecular interactionsinvolved. The results and implications of each of these studies are detailed in thisdissertation.

eDNA is a fast and popular method to collect information about species presence in the environment. eDNA is DNA that is collected from environmental samples, such as water, from DNA that is expelled from organisms interacting with their environment. eDNA is an effective way to find species with small populations and alien species. There are two ways to analyze eDNA, with high-throughput DNA sequencing methods and DNA metabarcoding or use of species-specific primers and PCR. In this study, we focus on the latter, designing species-specific primers for Burbot, Brown trout, and Atlantic Salmon, testing their validity in detecting eDNA of these species with functional PCR. We also evaluated eDNA collection methods, testing different scenarios in aquarium tanks with different number of dead and alive fishes. Primers and experimental setup such as use of different temperatures of the PCR reaction used in this study didn’t result in a functional PCR as determined by electrophoresis gel. There are some problems with the design of the PCR methods for eDNA since the purpose is to design methods that can identify certain species. However, future development of eDNA methods will probably include sequencing and not detection of PCR product sizes. eDNA methods will complete traditional trapping methods like net and electrofishing, but not replace them.
eDNA är en snabb och populär metod för att samla information om arters abundans i miljön. eDNA är DNA taget från miljö prover, så som vatten, från DNA som frigörs från organismer som intrigerar med deras miljön. eDNA är ett effektivt sätt att hitta arter med små populationer och främmande arter. De finns två olika sätt att analysera eDNA, med DNA-sekvenseringsmetoder med hög genomströmning och DNA-metabarkodning eller användning av artspecifika primrar och PCR. I denna studie fokuserade vi på den senare, designade artspecifika primers för lake, lax och öring, testa deras validitet vid detektering av eDNA hos dessa arter med funktionell PCR. Vi utvärderade också eDNA-insamlingsmetoder, testa olika scenarier i akvarietankar med olika antal döda och levande fiskar. Primers och experimentell uppställning, såsom användning av olika temperaturer för PCR-reaktionen som användes i denna studie, resulterade inte i en funktionell PCR såsom bestämd av elektroforesgel. Det finns några problem med utformningen av PCR-metoderna för eDNA, eftersom syftet är att utforma metoder som kan identifiera vissa arter. Emellertid kommer framtida utveckling av eDNA-metoder troligtvis att inkludera sekvensering och inte detektering av PCR-produktstorlekar. eDNA-metoder kommer att komplettera traditionella fångstmetoder som nät- och elektrofiske, men inte ersätta dem.

L'extraction d'information de textes médicaux fournit des renseignements très utiles pour identifier les effets indésirables dans la surveillance après consommation (Pharmacovigilance), qui sont plus difficiles à découvrir à travers des études médicales typiques puisque les patients prennent plusieurs traitements en même temps. Récemment, les techniques de Data Mining ont permis de découvrir les connaissances enregistrées dans de grands ensembles de données, comme les dossiers cliniques collectés par les hôpitaux tout au long de la vie du patient. L'objectif de cette thèse est d'identifier les effets indésirables causés par les traitements. Pour cela, nous devons extraire les relations entre les médicaments et Adverses Drug Events (ADE), qui est la relation de réaction indésirable des médicaments. Ce problème est divisé en tâches de reconnaissance d'entités nommées (NER) et d'extraction de relations. Aujourd'hui, les approches supervisées basées sur des algorithmes de Deep Learning et Machine Learning résolvent ce problème dans l'état de l'art. Les méthodes supervisées ont besoin de caractéristiques riches afin d'apprendre des modèles efficaces au cours de la formation, par conséquent, nous nous concentrons sur la construction de représentations de mots larges (l'entrée du réseau neuronal), nous utilisons des représentations de mots basées sur des caractères et des représentations de niveau de mots. La représentation proposée améliore la performance du modèle de référence et le modèle final a atteint les performances des méthodes de pointe. Ensuite, nous avons extrait des informations contextuelles à travers des modèles de Deep Learning, afin d'identifier les réactions indésirables aux médicaments. Le modèle proposé a amélioré la précision globale et l'extraction des réactions indésirables aux médicaments obtenu avec le modèle de base, ce qui indique l'efficacité de combiner des modèles de Deep Learning et une vaste ingénierie des caractéristiques
The Information Extraction from clinical notes provides relevant information to identify adverse side effects in post-marketing surveillance of medications (Pharmacovigilance), which is more difficult to discover by traditional medical studies since patients are taking several treatments at the same time. In recent years, data mining techniques have allowed to discover knowledge stored in big datasets, such as the clinical records collected by hospitals throughout patient's life. The goal of this work is identify adverse side effects caused by treatments. Then, we have to identify relations between medications and Adverse Drug Events (ADE) entities, which is called Adverse Drug Reaction relation. This problem is divided Named Entity Recognition (NER) and Relation Extraction tasks. Nowadays, supervised approaches based on Deep Learning and Machine Learning algorithms solve this problem in the state of the art. These supervised systems require rich features in order to learn efficient models during training, therefore, we focus on building comprehensive word representations (the input of the neural network), using character-based word representations and word representations. The proposed representation improves the performance of the baseline model, and the final model reached the performances of state of the art methods. Then we have extracted contextual information through Deep Learning models and other different features obtained from the relations, in order to identify the Adverse Drug Reaction relations. The proposed model improved the overall accuracy and the extraction of Adverse Drug Reaction compared to the baseline, indicating the effectiveness of combining Deep Learning models and extensive feature engineering

Cette étude s'inscrit dans le cadre du procédé PUREX de retraitement des combustibles irradiés. Elle a pour objectif l'analyse de l'influence d'une réaction chimique d'oxydoréduction sur la cinétique de transfert liquide-liquide. Le système chimique choisi correspond au cas de la désextraction de l'acide nitreux, d'une phase organique de phosphate de tributyle vers une solution aqueuse d'acide nitrique, en présence d'un composé antinitreux, le nitrate d'hydrazinium. L’analyse bibliographique a permis de mettre en exergue l'importance des phénomènes interfaciaux pour l'extraction liquide-liquide par le phosphate de tributyle. Une technique d'étude, basée sur l'utilisation d'une cellule du type ARMOLLEX, a été alors retenue. L’étude de l'influence des paramètres hydrodynamiques et physico-chimiques a permis de différencier la nature du régime cinétique : diffusionnel puis chimique. En l'absence de composé antinitreux, la résistance au transfert diffusionnel est essentiellement localisée du coté aqueux. La présence du nitrate d'hydrazinium se traduit par une forte accélération de la cinétique de transfert, imputée non seulement à l'annulation de la résistance diffusionnelle aqueuse, mais également à l'intervention directe du nitrate d'hydrazinium sur l'acide nitreux adsorbé à l'interface. Si la cinétique de la réaction d'oxydoréduction, connue par ailleurs pour sa rapidité en phase aqueuse, permettait de prévoir le premier effet, la contribution interfaciale constitue un résultat original qui, à notre connaissance, n'avait jamais été mis en évidence auparavant. Un mécanisme réactionnel prenant en compte cette réaction interfaciale a été élaboré et validé. Pour conforter l'ensemble des résultats établis, d'autres études complémentaires ont été entreprises. Un phénomène surprenant de blocage du transfert a été observé dans le cas d'un autre composé antinitreux, le nitrate d'hydroxylammonium, démontrant toute la spécificité du nitrate d'hydrazinium.

Over the last several decades, there have been a tremendous developments and greatinnovations in photocatalysis process along with the development of efficient nanosized catalysts for simple approach and economic viability. In this study, magnetic core@doubleshell nanomaterials were investigated and synthesized in lab with three-step innovative approach where Fe3O4 nanoparticles (NPs) were produced first to act as cores without using any surfactants. The magnetite/silica core–shell structure was then prepared by hydrolysis of tetraethoxysilane (TEOS) in the presence of core particles under alkaline conditions. And the outermost shell, the α-Fe2O3/TiO2 nanoparticles, were grown over magnetic core of Fe3O4@SiO2 using coprecipitation and calcination method. Furthermore, the Fe3O4@SiO2@α-Fe2O3/TiO2 NPs were then loaded on the reduced graphene oxide (r-GO) using hydrothermal method and are also mixed by kneading with the layered double hydroxides (LDH) of Mg2+ and Al3+. These nanoparticles were characterized with scanning electron microscope (SEM), transmission electron microscope (TEM), dynamic light scattering (DLS), and energy dispersive X-ray spectroscopy (EDS). Different model compounds like microcrystalline cellulose (90 μm), D-xylose, and sodium lignosulfonate representing cellulose, hemicellulose, and lignin, respectively, were converted to valuable chemicals with different NPs under visible light for different time periods. For example, valeric acid (VA) and vanillylmandelic acid (VMA) were produced when cellulose was used for the conversion with core-double shell NPS which were quantified using high performance liquid chromatography (HPLC). Similar approach was adopted for the conversion of brewers’ spent grain (BSG), a lignocellulosic biomass, without oxygen under visible light, which yielded ethanol as the main product along with other sugars and acids of very low concentrations. The magnetic property of the nanomaterials made it easy for recycle and reuse. From a sustainability point of view, this study will fill a large need in the biomass photocatalysis field by developing core-shell multi-functional photocatalysts for direct transformation of lignocellulose into valuable chemicals under low temperatures, atmospheric pressure, and visible light from the sun.

O desempenho de seis métodos de extração de DNA de oocistos de Cryptosporidium spp. em fezes foram avaliados pela nested PCR do gene SSU rRNA. Os métodos consistem na combinação com pequenas variações de duas técnicas para a liberação de esporozoítos (indução de excistamento/E ou choque térmico/C) e três técnicas de purificação de DNA (fenol-clorofórmio/F, GuSCN-sílica/S ou kit QIAmp DNA Stool Mini/K). Para a avaliação da sensibilidade analítica, os testes foram realizados a partir de diluições seriadas de oocistos purificados, na ausência e na presença de 100 μl de fezes bovinas. A sensibilidade diagnóstica foi avaliada pela comparação com o método microscópico de centrífugo-flutuação em sacarose (padrão-ouro) em 15 amostras fecais de diferentes hospedeiros naturalmente infectados. Na ausência de fezes, foram avaliados apenas os métodos EF, ES, CF e CS, sendo que EF apresentou sensibilidade analítica superior, possibilitando a detecção de até 1 oocisto. Na presença de fezes, os métodos EF, EK e CK apresentaram desempenhos equivalentes, com sensibilidade de 104 oocistos. As maiores sensibilidades diagnósticas foram obtidas pelos métodos EK e CK, que possibilitaram a detecção de 13 (86,7%) das 15 amostras. Devido à alta sensibilidade analítica de EF em amostras purificadas, avaliou-se sua sensibilidade diagnóstica em amostras de oocistos purificados pelo método de centrífugo-flutuacão em sacarose, obtendo-se 100% de detecção. A presença de inibidores nas amostras fecais reduziu fortemente a sensibilidade das reações de PCR a partir de DNA extraído pelos métodos avaliados, sendo recomendável o emprego de técnicas de purificação de oocistos previamente à extração de DNA. Os resultados apontam para uma melhor eficiência dos métodos de indução de excistamento e kit QIAmp DNA Stool Mini.
The performance of six methods for DNA extraction from Cryptosporidium spp. oocysts in feces were evaluated by nested PCR of SSU rRNA gene. The methods are the combination with small variations of two techniques for the release of sporozoites (induction of excystation/E or thermal shock/C) and three techniques for DNA purification (phenol-chloroform/F, GuSCN-silica/S or QIAmp DNA Stool Mini kit/K). For the evaluation of analytical sensitivity, tests were made from serial dilutions of purified oocysts, in absence and in presence of 100 μl of cattle feces. The diagnostic sensitivity was assessed by comparison with the microscopic method of centrifugalflotation in sucrose (gold standard) in 15 fecal samples from different naturally infected hosts. In the absence of feces, only methods EF, ES, FC, and CS were tested. EF had the highest analytical sensitivity, enabling the detection of up to 1 oocyst. In the presence of feces, methods EF, EK, and CK showed similar performance, with sensitivity of 104 oocysts. The highest sensitivities were obtained by methods EK and CK, which enabled the detection of 13 (86.7%) of 15 samples. Due to the high analytical sensitivity of EF in purified samples, its diagnostic sensitivity was evaluated in samples of purified oocysts by the method of centrifugal-flotation in sucrose, resulting in 100% detection. The presence of inhibitors in fecal samples greatly reduced the sensitivity of the PCR reactions from DNA extraction methods evaluated, and the use of techniques for purification of oocysts prior to DNA extraction is recommended. The results show a better performance of the methods of induction of excystation and QIAmp DNA Stool Mini kit.