Os óleos essenciais cítricos possuem uma ampla variedade de aplicações em diversos ramos das indústrias químicas, farmacêuticas, de alimentos, entre outras. Na indústria de cosméticos, mais precisamente no ramo de perfumaria, os ativos naturais cítricos (provenientes dos óleos de bergamota, limão, lima, mandarina, laranja, etc) são empregados nas formulações de colônias, cremes hidratantes e loções. Além disso, estes compostos são largamente utilizados como agentes aromatizantes nas indústrias alimentícias e farmacêuticas e empregados como matéria-prima para as indústrias de aromas. Com o objetivo de melhorar a qualidade dos óleos essenciais e sua capacidade aromatizante, alguns processos têm sido desenvolvidos para enriquecer a mistura com compostos oxigenados, os quais são reportados como os principais responsáveis pelo aroma característico da fruta. Tal processo é comumente conhecido como desterpenação e consiste na concentração destes compostos pela retirada de hidrocarbonetos terpênicos do óleo e pode ser realizado por meio da extração líquida ou extração por solvente. Neste contexto, esta dissertação de mestrado teve como finalidade estudar da viabilidade técnica do processo de desterpenação de óleo essencial cítrico modelo composto por limoneno e linalol. Os experimentos foram conduzidos em coluna de discos rotativos perfurados (PRDC) sob condição de temperatura controlada e pressão ambiente local. O equipamento operou de modo contínuo e contracorrente, utilizado como solventes soluções de etanol com teores de água de 30 e 40% (em massa), velocidade de rotação dos discos de 150, 200 e 250 rpm e razão entre a vazão mássica de solvente e alimentação de 0,5; 1,25; 2,5 e 3,0. Avaliou-se, também, o comportamento da densidade e viscosidade das fases extrato e rafinado oriundas da coluna de extração. Através deste estudo foi possível verificar que a tecnologia de desterpenação de óleos essenciais por meio da extração líquido-líquido é viável podendo ser aplicada à processos contínuos utilizando equipamento PRDC, por exemplo, sendo possível a obtenção de frações enriquecidas do composto oxigenado (linalol). / Citrus essential oils are used in a wide variety of applications in various branches of the chemical, pharmaceutical, food and other industries. In the cosmetics industry, specifically in the field of perfumery, natural citrus assets (from the essential oils of bergamot, lemon, lime, mandarin, orange, etc.) are used in the formulations of colonies, moisturizing creams and lotions. Furthermore, these compounds are widely used as flavoring agents in food and pharmaceutical industries and as raw material for the flavor industry. With the aim to improve the quality of essential oils and its flavoring ability, some processes have been developed to enrich the mixture with oxygenated compounds, which are reported as the main responsible for the characteristic aroma of the fruit. Such process is commonly known as deterpenation, and consists in concentration of these compounds by the removal of terpene compounds from the essencial oil and can be performed by liquid or solvent extraction. In this context, the purpose of this dissertation of master\'s degree was to carry out a study of the technical feasibility of deterpenation process of a citrus essential oil model composed by a mixture of limonene and linalool. The experiments of deterpenation were conducted in a perforated rotating discs contactor (PRDC) under controlled conditions of temperature and local environment al pressure. The equipment was operated in continuously and countercurrently mode, using as solvents ethanol solutions with 30 and 40% of water contents (by mass), rotation of discs speed of 150, 200 and 250 rpm and the mass flow rate of the solvent and feed ratio of 0.5, 1.25, 2.5 and 3.0. The behavior of the density and viscosity of the extract and raffinate phases from the extraction column was also evaluated. It was found that the deterpenation of essential oils by liquid-liquid extraction technology is feasible and can be applied to continuous processes using PRDC column extraction type, for instance, being possible to obtain enriched fractions of oxygenated compounds (linalool).

Este trabalho trata de um estudo experimental da transferência de calor e de massa de um condensador evaporativo, bem como, da relação existente entre as grandezas envolvidas durante a sua operação. Para o desenvolvimento da pesquisa foi construído um condensador evaporativo de pequeno porte com dimensões proporcionais a condensadores comercialmente fabricados no Brasil para operar com R-22 como fluido refrigerante. Este condensador tem 35 colunas e 12 fileiras de tubos de cobre de 6,35 mm de externo e opera junto a uma instalação que proporciona o escoamento de R-22 por termossifao. O distribuidor de água também foi construído em cobre e o eliminador de gotas em alumínio, assim como a estrutura do condensador. As laterais e a bandeja de recolhimento de água foram construídas em vidro para permitir a visualização do escoamento da água que e aspergida sobre os tubos. A fim de variar as condições operacionais, este condensador foi acoplado a um ventilador centrifugo acionado por um conversor de freqüência e a um circuito que permitiu a variação da vazão de água aspergida sobre os tubos. Dos ensaios resultaram 79 amostras de medição, que serviram para verificar a relação existente entre grandezas, bem como avaliar os coeficientes de transferência de calor e massa. Foram determinados os coeficientes de transferência de calor médio e local para o R-22. Alem disso, foram verificadas as regiões onde ocorrem o dessuperaquecimento, a mudança de fase e o subresfriamento do fluido refrigerante, e foram confrontados os coeficientes globais de transferência de calor obtidos a partir dos dados experimentais com aqueles obtidos por correlações para as regiões de condensação e subresfriamento. Destas investigações, resultou uma correlação para o coeficiente global de transferência de calor que visa determinar esta grandeza através das condições operacionais e das características geométricas do condensador. O Maximo desvio encontrado entre os valores provenientes dos dados experimentais e aqueles calculados pela correlação foi inferior a 10%. / This work presents an experimental heat and mass transfer study of an evaporative condenser, as well as, the relationship between measured quantities obtained during its operation. A small scale evaporative condenser was built in agreement to the real size equipment geometric similarity. The small scale condenser has a bundle of 210 copper tubes, with the outer diameter equal to ¼ inch, which were arranged in 35 columns and 6 rows and operating connected to the thermosifon facility that promote R-22 flow. The spray water distributor is also built on copper and the drift eliminator on aluminum, as well as the evaporative condenser structure. This condenser bounds were constructed as a glass enclosure to allow for the spray water flow visualization. To evaluate the operational conditions, the equipment was attached to a centrifugal fan controlled by a frequency inverter and a spray water circuit which allows for water quantity changing. From the carried out tests, 79 measuring samples were obtained and used to verify the quantities relationship, as well as, to evaluate the heat and mass transfer coefficients For the R-22 were determined the local and average heat transfer coefficients. Furthermore, the dessuperheating, condensation and subcooling regions were verified and compared both, the experimental overall heat transfer coefficient and those calculated by correlations at condensation and subcooling zones. From these investigations a correlation for the overall heat transfer coefficient is proposed. Its application allows determining the coefficient using the operational conditions and the geometric condenser features. The higher deviation found between the experimental data and the presented correlation is lower to 10%.

Dans les centrales nucléaires à eau sous pression, les Résines Echangeuses d'Ions (REI) sont utilisées dans les circuits d'épuration. Dans le cadre de cette thèse, une étude qualitative a été réalisée afin de prédire les tendances du comportement d'une REI dans le cadre de son utilisation en condition centrale nucléaire. Fort des résultats de cette étude, nous avons cherché à caractériser quantitativement le comportement des REI en colonne. Pour cela, nous avons utilisé des solutions analytiques mais ces dernières se sont révélées valables uniquement dans des cas particuliers. Afin de trouver une solution générale, nous nous sommes donc tournés vers une solution numérique : OPTIPUR. En vue de la valider et aussi de mieux comprendre la cinétique en colonne, nous avons réalisé une étude expérimentale. Cette dernière vise à caractériser la résistance au transfert de masse en colonne et à mener une étude de sensibilité sur les paramètres influençant ce phénomène. Cette étude repose sur la détermination de la fuite ionique cinétique, correspondant à une certaine concentration de polluant présente en sortie de colonne en début d'expérience. Pour cela, nous avons testé l'influence de nombreux paramètres sur la fuite ionique cinétique. Nous avons ainsi vu l'importance de la vitesse de filtre et donc des conditions hydrodynamiques sur la fuite ionique cinétique. Ces nombreux résultats de fuite cinétique, ont été modélisés à l'aide de la corrélation empirique de Dwivedi & Upadhyay afin de tester sa validité. Par la suite, nous avons simulé nos résultats de fuite ionique cinétique avec deux options du logiciel OPTIPUR : option Mass Transfer Coefficient (MTC) et Nernst-Planck (NP). Ces dernières encadrent les résultats expérimentaux. L'option MTC d'OPTIPUR donne des résultats inférieurs alors que ceux obtenus avec l'option NP sont supérieurs aux résultats expérimentaux. Nous avons vu que dans le cadre d'un échange ternaire, seule l'option NP est valide. Nous avons proposé des solutions pour mieux caler les résultats obtenus numériquement. D'autres simulations ont été réalisées afin de vérifier les capacités de prédiction de l'appareil pour des expériences plus longues (allant jusqu'à la saturation de la REI). Les tendances observées étaient celles attendues. L'outil OPTIPUR est un outil précis et robuste pour étudier la cinétique en colonne. / In pressurized water reactor, ion exchange resins (IER) are used in systems purification. In this thesis, a qualitative study has been performed to predict the behavior of IER while used in nuclear plants conditions. Then, we searched to characterize the IER behavior in column through a quantitative study using analytical solutions. But these solutions worked only for particular cases. In order to find a general solution, we used a new numerical solution: OPTIPUR. To validate this general solution and get a better understanding of the kinetic in column, we performed an experimental study to characterize the resistance to mass transfer in column and to study the sensibility on the parameters influencing this phenomenon. This study is based on the characterization of the initial leakage (initial pollutant concentration at the column outlet). We tested numerous parameters on the initial leakage. We understood the importance of the superficial velocity and indeed of the hydrodynamic conditions on the initial leakage. These numerous results about initial leakage were modeled with an empirical correlation of Dwivedi and Upadhyay in order to validate it. Then, we modeled our results with the two options of OPTIPUR software: option Mass Transfer Coefficient (MTC) and Nernst-Planck (NP). These options encircle experimental results. The MTC option of OPTIPUR gives lower results while those obtained with the NP option are higher than the experimental results. We observed also that only the NP option was valid for a ternary exchange. We proposed solutions to get a better fit with the results obtained with OPTIPUR. We performed other simulations to check the prediction abilities of the software for longer experiments (until the IER saturation). The tendencies were those expected. The OPTIPUR software showed is accuracy and robustness to study column kinetic.

No description available.

Thesis (MScEng (Process Engineering))--Stellenbosch University, 2008. / Please refer to full text to view abstract.

Due to the ever increasing concerns about pollutants and contaminants found in water, new treatment technologies have been developed. Ozonation is one of such technologies. It has been widely applied in the treatment of pollutants in water and wastewater treatment processes. Ozone has many applications such as oxidation of organic components, mineral matter, inactivation of viruses, cysts, bacteria, removal of trace pollutants like pesticides and solvents, and removal of tastes and odours. Ozone is the strongest conventional oxidant that can result in complete mineralisation of the organic pollutants to carbon dioxide and water. Because ozone is unstable, it is generally produced onsite in gas mixtures and is immediately introduced to water using gas/liquid type reactors (e.g. bubble columns). The ozone reactions are hence of the type gas liquid reactions, which are complex to model since they involve both chemical reactions, which occur in the liquid phase, and mass transfer from the gas to the liquid phase. This study focuses on two aspects: mass transfer and chemical reactions in multicomponent systems. The mass transfer parameters were determined by experiments under different conditions and the chemical reactions were studied using single component and multicomponent systems. Two models obtained from the literature were adapted to the systems used in this study. Mass transfer parameters in the semi-batch reactor were determined using oxygen and ozone at different flow rates in the presence and absence of t-butanol. t-Butanol is used as a radical scavenger in ozonation studies and it has been found to affect the gas-liquid mass transfer rates. An experimental study was carried out to investigate the effects of t-butanol concentrations on the physical properties of aqueous solutions, including surface tension and viscosity. It was found that t-butanol reduced both properties by 4% for surface tension and by a surprising 30% for viscosity. These reductions in the solution physical properties were correlated to enhancement in the mass transfer coefficient, kL. The mass transfer coefficient increased by about 60% for oxygen and by almost 50% for ozone. The hydrodynamic behaviour of the system used in this work was characterised by a homogeneous bubbling regime. It was also found that the gas holdup was significantly enhanced by the addition of t-butanol. Moreover, the addition of t-butanol was found to significantly reduce the size of gas bubbles, leading to enhancement in the volumetric mass transfer coefficient, kLa. The multicomponent ozonation was studied with two systems, slow reactions when alcohols were used and fast reactions when endocrine disrupting compounds were used. ii These experiments were simulated by mathematical models. The alcohols were selected depending on their volatilization at different initial concentrations and different gas flow rates. The degradation of n-propanol as a single compound was studied at the lowest flow rate of 200 mL/min. It was found that the degradation of n-propanol reached almost 60% within 4 hours. The degradation of the mixture was enhanced with an increase in the number of components in the mixture. It was found that the degradation of the mixture as three compounds reached almost 80% within four hours while the mixture as two compounds reached almost 70%. The effect of pH was studied and it was found that an increase in pH showed slight increase in the reaction. Fast reactions were also investigated by reacting endocrine disrupting chemicals with ozone. The ozone reactions with the endocrine disrupters were studied at different gas flow rates, initial concentrations, ozone concentrations and pH. The degradation of 17β-estradiol (E2) as a single compound was the fastest, reaching about 90% removal in almost 5 minutes. However estrone (E1) degradation was the lowest reaching about 70% removal at the same time. The degradation of mixtures of the endocrine disruptors was found to proceed to lower percentages than individual components under the same conditions. During the multicomponent ozonation of the endocrine disruptors, it was found that 17β-estradiol (E2) converted to estrone (E1) at the beginning of the reaction. A MATLAB code was developed to predict the ozone water reactions for single component and multicomponent systems. Two models were used to simulate the experimental results for single component and multicomponent systems. In the case of single component system, good simulation of both reactions (slow and fast) by model 1 was obtained. However, model 2 gave good agreement with experimental results only in the case of fast reactions. In addition, model 1 was applied for multicomponent reactions (both cases of slow and fast reaction). In the multicomponent reactions by model 1, good agreement with the experimental results was also obtained for both cases of slow and fast reactions.

L’opération de transfert de matière gaz/liquide est une étape essentielle en génie desprocédés. Elle conditionne directement les performances des contacteurs gaz/liquide en représentantbien souvent l’étape limitant l’efficacité du procédé. L'objectif de cette étude est de proposer denouvelles investigations afin d'étudier l'effet de la présence dans l'eau pure, de certains composésgénéralement rencontrés dans les procédés biologiques, sur les coefficients de diffusion de l'oxygèneet de quantifier leurs conséquences sur les coefficients de transfert de matière côté liquide. Lescoefficients de diffusion de l'oxygène DO2 ont été mesurés dans diverses phases liquides contenant dusel (NaCl), du sucre (glucose) ou des tensio-actifs (laurylsulfate de sodium). Comparé à l'eau pure,des réductions de coefficient de diffusion DO2 ont été observées, la variation de DO2 avec laconcentration C de composé a été modélisée, elle dépend de la nature du composé ajouté. Ensuite,des expériences réalisées sur un train de bulles et sur les mêmes phases liquides ont permis ladétermination des coefficients de transfert de matière côté liquide kL. Quelle que soit la solutionaqueuse étudiée, une diminution des valeurs de kL est observée avec l'augmentation de C. Cesrésultats mettent en évidence que, même si les propriétés de l'eau pure (densité, viscosité, tensionsuperficielle) ne sont pas significativement modifiées par l'ajout de sels (NaCl), les coefficients detransfert de matière côté liquide kL peuvent être modifiés. Pour les solutions aqueuses de glucose, laréduction de kL avec DO2 est bien corrélée, et principalement due à la variation de viscosité avec laconcentration. Pour les solutions de tensio-actif, les nombres de Reynolds restent presque constantspour toutes les concentrations. Le seul responsable de la réduction du coefficient de transfert dematière côté liquide kL est donc le coefficient de diffusion de l’oxygène qui diminue avec laconcentration. La présente étude a clairement confirmé la nécessité de compléter et/ou rendre comptedes données liées aux coefficients de diffusion de l’oxygène DO2 et aux coefficients de transfert dematière côté liquide kL dans des milieux complexes (solution électrolytique, solution organique ettensioactif). Ces informations sont nécessaires pour décrire et modéliser correctement lesphénomènes de transfert de matière gaz/liquide / The gas/liquid mass transfer is an essential step in process engineering. It directly affects theperformance of gas/liquid reactor in being often limiting process efficiency. The objective of this studyis to propose new investigations in order to study (i) the effect on oxygen diffusion coefficients underthe presence in clean water of some compounds usually found in biological process and (ii) quantifytheir consequences on liquid-side mass transfer coefficients. The oxygen diffusioncoefficients DO2 were measured in various synthetic liquid phases containing either salt (NaCl), sugar(glucose) or surfactant (sodium laurylsulphate). When compared to clean water, reductionsof DO2 were observed; the variation of DO2 with the compound concentration C was modeled andfound dependent on the nature of the compound added. Then, to determine the liquid side masstransfer coefficient kL, experiments on a train of bubbles rising in a quiescent liquid phase were carriedout by the same synthetic liquid phases. For all cases, whatever the aqueous solutions, a decreaseof kL with increasing C was clearly observed. These results firstly showed that, even if the properties ofclean water (density, viscosity, surface tension) were not significantly changed by the addition of salts(NaCl), the liquid-side mass transfer coefficients could be modified. For the aqueous solutions ofglucose, the reduction of kL with DO2 was well correlated, and mainly due to the change in viscositywith concentration. For surfactants, the hydrodynamic conditions (i.e. bubble Reynolds number) beingalmost kept constant for all concentrations, only the change in DO2 was thus responsible for thedecrease of kL. The present study clearly confirmed the need to complete and/or account for thedatabase related to oxygen diffusion coefficients and liquid side mass transfer coefficient in complexmedia (electrolytic solution, organic solution and surfactant). This condition is imperatively required todescribe and to model appropriately the gas-liquid mass transfer phenomena

Bench and pilot scale determinations of the volumetric oxygen transfer coefficient, K_La, were performed on an improved A²/O biological nutrient removal (BNR) pilot plant. Effluent from a full scale primary clarifier, used as pilot plant influent, was found to have an alpha (ratio of process to clean water K_La) of 0.71 as determined in a 21 liter bench scale reactor and an alpha of 0.332 as determined in a 0.45 m³ aeration basin of the 2.4 m³ pilot plant. Alpha of a 1:1 mixture of primary clarifier effluent with pilot plant return activated sludge was determined to be 0.94 at bench scale and 0.71 at pilot scale. An assay of alphas through the initial non aerated treatment zones of the pilot plant using the bench scale reactor indicated that alphas peaked in the effluent of the first anaerobic zone (alpha equal to 1.01) and were lower in the second anaerobic zone and first anoxic zone. An assay of alphas in the three pilot plant series sideline aeration basins indicated that alpha was maximum in the first aeration basin (alpha equal to 0.905) and were lower in the second and third aeration basins (0.716 and 0.661 respectively). A consistent increase in average surface tension was noted from the first to second to third aeration basins, however the differences were not statistically significant. A comparison of pilot plant alphas determined in the first aeration basin following anaerobic nominal hydraulic retention times of 0.0, 0.21, 0.43, and 0.64 hours yielded alpha values of 0.71, 0.94, 0.64, and 0.74 respectively. Like the assay using the bench scale reactor, the alpha values at pilot scale peaked following treatment in only one anaerobic zone (nominal HRT of 0.21 hours). The study concludes that short exposures in an initial anaerobic reactor as required for biological phosphorus removal may benefit oxygen transfer efficiency through increased alphas, however the benefits of long periods of anaerobic reaction time (over 0.43 hours) are uncertain. / Master of Science

Multiphase flows in chemical reactors with micro- and millimeter-size channel structures such as monolith froth reactors, compact heat exchangers and fuel cells have received great attention in the last years. They are considered as a promising alternative to conventional reactors, such as fixed bed reactors and bubble columns which are mainly used for gas absorption, catalytic hydrogenation and biochemical conversions. Slug or Taylor bubble flow is a desired operating state for this type of contactors due to the frequent change of efficient gas-liquid contacting in the film around the bubbles and the enhanced mixing in the liquid slugs behind the bubbles. Consequently, capillary Taylor flow is currently a target of intensive investigations. However, a full understanding of design parameters and optimum operating conditions are still lacking. For milli- and microreactors mass transfer between gas and liquid phases depends upon various parameters such as bubble shape, relative velocity between the two phases, degree of liquid contamination and many more. To further advance the fundamental understanding of micro- and milli-channel reactors with Taylor flow, main design parameters and operating conditions were investigated, which include (a) the effect of bubble size, channel diameter and cross sectional shape of channel on the mass transfer coefficient of dissolving bubbles, (b) the influence of the presence of surface active agents on the bubble shape, velocity and also on the mass transfer rate of bubbles and (c) the intensification effect of oscillation of channels on the mass transfer performance of Taylor bubbles. For the study of gas-liquid mass transfer high-resolution X-ray radiography and tomography were used as measurement techniques. The X-ray imaging methods were chosen as their accuracy is less affected by changes in the refractive index, as it is the case for conventional optical methods. The mass transfer was calculated by measuring the changes in the size of the bubbles at constant pressure. The utilization of X-ray visualization enabled the acquisition of a series of radiographic images of bubbles. The images gave the volume, interfacial area and length of the bubble with high accuracy as a function of time and were used to evaluate the mass transfer coefficient using the mass conservation equations. In case of circular channels, the results show that Sherwood numbers have a large dependency on the bubble length and also equivalent diameter which is in accordance with previous results for larger channel diameters. However, the values of measured Sherwood numbers could not be predicted by available correlations which are valid only for larger pipes. As a result, a new mass transfer correlation in the form of Sherwood number as a function of Peclet number as well as bubble size ratio was derived. The proposed correlation is applicable for a large range of bubble sizes with high accuracy. The comparison of the results for the square and circular channels showed that despite the fact that the rise velocity of bubbles in the square channel is about three times higher than in the circular channel, the mass transfer coefficient is about the same. Furthermore, the results show that in square channels the dissolution curves are relatively even, while the dissolution curves of circular channels exhibit some distinguishable change in the slope. In addition, the results show that the calculated mass transfer coefficient based on the measured data show good agreement with the data predicted by the penetration theory. Regarding the influence of surfactants on the mass transfer in small channels with Taylor flow, it was shown that a small amount of surfactant reduces the mass transfer and its impact is more pronounced on small bubbles. Furthermore, it was demonstrated that the presence of surfactants causes the change of the bubble shape and leads to a slight increase of the liquid film thickness around the bubble and as a result the elongation of contaminated bubbles. Intensification of mass transfer in small channels with Taylor bubbles was investigated by measuring the motion, shape and dissolution rate of individual elongated Taylor bubbles of air and CO2 in water. The comparison of the results for the stationary and oscillating channel showed that mechanical vibration of the channel is able to enhance the mass transfer coefficient from 80% to 186%. Moreover, the mass transfer rate positively correlates with frequency and amplitude of oscillation, which is more pronounced at higher amplitudes. In addition, it was shown that the intensification of mass transfer with increase of amplitude/frequency of vibration is mainly attributed to the increase of bubble surface wave oscillations that causes an enlargement of contact area between the phases and also a reduction of mass transfer resistance in the liquid-side boundary layer.

No description available.