O objetivo deste trabalho foi estudar a produção do antitumoral retamicina por células imobilizadas de Streptomyces olindensis ICB20 em biorreatores. A imobilização das células foi conseguida após uma etapa inicial de cultivo em erlenmeyers (reativação e pré-imobilização) e posterior envolvimento em gel, utilizando-se alginato de cálcio 3%; as esferas produzidas tinham diâmetro médio de 3,0 ± 0,2 mm. Foram utilizados neste trabalho os biorreatores tipo cesta (2,4 L de volume útil) e coluna de bolhas (1,6 L de volume útil), efetuando-se cultivos em batelada simples, bateladas repetidas e contínuos, visando-se uma comparação dos diferentes sistemas empregados. Também foram realizados alguns cultivos com células livres, a fim de efetuar uma comparação com os sistemas com células imobilizadas nos diferentes biorreatores. Os ensaios consistiram em empregar diferentes condições de agitação (300 e 500 rpm) e aeração (0,4 e 1 vvm) para o biorreator cesta em sistema com células imobilizadas, assim como diferentes vazões de aeração no biorreator coluna de bolhas (1, 2 e 3 vvm). Os cultivos em bateladas repetidas e contínuos foram operados a partir das melhores condições obtidas nos cultivos descontínuos. Foram aplicadas nos sistemas contínuos com células imobilizadas, as vazões específicas de alimentação de 0,05 e 0,2 h -1 no biorreator cesta e vazões de 0,015 a 0,05 h -1 no biorreator coluna de bolhas. A comparação entre os sistemas com células livres e imobilizadas mostrou que as limitações difusionais afetaram significativamente as cinéticas dos ensaios com células imobilizadas, considerando-se apenas uma batelada. A operação do biorreator cesta em sistema de bateladas repetidas apresentou os maiores valores de produção da retamicina (em torno de 1,5 a 1,7 UA), porém sua operação foi possível por apenas três bateladas. O sistema contínuo operado com vazão específica de alimentação de 0,03 h-1, com células imobilizadas no biorreator coluna de bolhas, mostrou ser o mais adequado dentre todos os ensaios realizados com células imobilizadas, apresentando estabilidade na produção, em torno de 0,8 UA, durante 96 horas de alimentação (cerca de três tempos de residência). / The purpose of this study was to investigate the production of the antitumor antibiotic retamycin by immobilized cells of Streptomyces. olindensis ICB20 in bioreactors. Cells were immobilized by entrapment in Ca-alginate gel (3%) after being grown in Erlenmeyers (reactivation and pre-immobilization cultures). The average diameter of the Ca-alginate beads was 3.0 ± 0.2 mm. Aiming to compare different cell systems, immobilized cell cultures were carried out in a 2.4L working volume basket-type stirred tank reactor (BSTR) and a 1.6 L working volume bubble column reactor (BCR) in batch, repeated-batch and continuous modes. Free cell suspension cultures were also performed and the results obtained compared to those in immobilized cell systems. Different agitation rates (300 and 500 rpm) and air flow rates (0.4 and 1.0 v.v.m.) were employed in the BSTR immobilized cell experiments.The BCR cultures were conducted at 1.0, 2.0 and 3.0 v.v.m. The optimal operating conditions for the batch mode were used in the repeated-batch and continuous cultures. Immobilized cells were grown in continuous mode at feed dilution rates of 0.05 and 0.2 h-1 in the BSTR and at 0.015 and 0.05 h-1 in the BCR. The comparative evaluation of the batch cultures with free and immobilized cells showed that diffusion limitations had affected the kinetics of cell growth and retamycin production in the immobilized cell systems. The highest average values of retamycin content (from 1.5 to 1.7 AU) were achieved in repeated batch cultures conducted in the basket-type reactor (BSTR) in spite of a limited number of batches (3 batches). Of all the systems, the continuous cell immobilized culture carried out in the BCR at a dilution rate of 0.03 h-1 proved to be the most adequate for retamycin production as retamycin levels remained stable (around 0.8 AU) over 96 hours (about 3 residence times).

Ce travail concerne la mesure et la caractérisation du transfert thermique à la paroi externe d’un faisceau de tube de refroidissement inséré dans des colonnes à bulles type « slurry ». La valeur du coefficient de transfert de chaleur est estimée à partir des équations de la chaleur. Une colonne de 0,15 m de diamètre et de 4 m de haut, équipée de deux tubes en U (3 cm de diamètre externe), a été utilisée pour mettre au point la métrologie nécessaire. L’eau a servi de fluide de refroidissement. Le mélange diphasique air-huile Syltherm XLT®, puis le mélange triphasique air-huile Syltherm XLT®-microbilles d’alumine poreuses (dS ~ 80 μm), ont servi de fluides modèles. L’incertitude de nos mesures a été estimée à environ 8 %. En système diphasique, les variations du coefficient de transfert de chaleur avec la vitesse superficielle du gaz ont pu être corrélées par une loi semblable à celle de Deckwer (1980). Cependant, la valeur de la constante de corrélation semble dépendre de l’orientation du faisceau de tubes par rapport à l’axe de la colonne. Un tel comportement n’a jamais été rapporté dans la littérature. L’écart du faisceau à un faisceau idéal (i.e. parfaitement droit et symétrique) peut être un paramètre crucial pour le transfert de chaleur. En système triphasique, la valeur du coefficient ne varie pas de façon significative jusqu’à une concentration massique d’environ 18,8 %, avant de diminuer d’environ 10 % pour une concentration massique de 21,3 %. Ce résultat est surprenant. Les variations rapportées dans la littérature sont en effet souvent contradictoires, mais toujours continues dans la gamme de concentrations testée. La métrologie mise au point a été implantée dans une colonne de 1 m de diamètre et de 5 m de haut, équipée de 24 tubes en U (6 cm de diamètre externe). Celle-ci est jugée représentative d’un réacteur pour le procédé Fischer-Tropsch. Les premiers résultats indiquent que la caractérisation thermique de l’installation sera plus délicate que pour la petite colonne / This work investigates the measure and characterisation of heat transfer in slurry bubble column reactors equipped with a bundle of cooling tubes. The value of the shell-tube heat transfer coefficient is estimated at thermal steady-state regime using heat transfer equations. A 15 cm in diameter, 4 m high bubble column, equipped with a two U-tubes (3 cm O.D.) bundle has been used to assess the metrology selected. The cooling fluid was water. Air-Syltherm XLT® heat transfer fluid and air-Syltherm XLT® heat transfer fluid-porous alumina particles (dS ~ 80 μm) were successively used as shell fluids. The uncertainty of our measures has been estimated to be around 8 %. The variations of the shell-tube heat transfer coefficient with superficial gas velocity can be modelled using the well-known correlation by Deckwer (1980). However, a smaller constant value than indicated by Deckwer et al. (1980) was obtained and it was found to be dependent upon the orientation of the tube bundle relatively to the column axis. This has never been reported in the literature and implies that any difference relatively to the ideal tube bundle – perfectly straight and symmetric – might be critical for heat transfer. Addition of solid particles has little effect on heat transfer for solid concentrations below 18.8 %w/w. A further increase up to 21.3 %w/w induced a 10 % decrease of the value of the shell-tube heat transfer coefficient. This was surprising, as existing literature results display continuous variations of the heat transfer coefficient values in the range of solid concentrations tested, even though trends of variation could be opposite. The assessed metrology was implemented into a 1 m in diameter, 5 m high bubble column equipped with a 24 U-tubes (6 cm O.D.) bundle. This pilot plant was considered to be large enough to mock up a slurry bubble column reactor for the Fischer-Tropsch process. First results indicate that thermal characterisation will be more complex than for the smaller diameter column

Bubbly flows can be found in many applications in chemical, biological and power engineering. Reliable simulation tools of such flows that allow the design of new processes and optimization of existing one are therefore highly desirable. CFD-simulations applying the multi-fluid approach are very promising to provide such a design tool for complete facilities. In the multi-fluid approach, however, closure models have to be formulated to model the interaction between the continuous and dispersed phase. Due to the complex nature of bubbly flows, different phenomena have to be taken into account and for every phenomenon different closure models exist. Therefore, reliable predictions of unknown bubbly flows are not yet possible with the multi-fluid approach. A strategy to overcome this problem is to define a baseline model in which the closure models including the model constants are fixed so that the limitations of the modeling can be evaluated by validating it on different experiments. Afterwards, the shortcomings are identified so that the baseline model can be stepwise improved without losing the validity for the already validated cases. This development of a baseline model is done in the present work by validating the baseline model developed at the Helmholtz-Zentrum Dresden-Rossendorf mainly basing on experimental data for bubbly pipe flows to bubble columns, bubble plumes and airlift reactors that are relevant in chemical and biological engineering applications. In the present work, a large variety of such setups is used for validation. The buoyancy driven bubbly flows showed thereby a transient behavior on the scale of the facility. Since such large scales are characterized by the geometry of the facility, turbulence models cannot describe them. Therefore, the transient simulation of bubbly flows with two equation models based on the unsteady Reynolds-averaged Navier–Stokes equations is investigated. In combination with the before mentioned baseline model these transient simulations can reproduce many experimental setups without fitting any model. Nevertheless, shortcomings are identified that need to be further investigated to improve the baseline model. For a validation of models, experiments that describe as far as possible all relevant phenomena of bubbly flows are needed. Since such data are rare in the literature, CFD-grade experiments in an airlift reactor were conducted in the present work. Concepts to measure the bubble size distribution and liquid velocities are developed for this purpose. In particular, the liquid velocity measurements are difficult; a sampling bias that was not yet described in the literature is identified. To overcome this error, a hold processor is developed. The closure models are usually formulated based on single bubble experiments in simplified conditions. In particular, the lift force was not yet measured in low Morton number systems under turbulent conditions. A new experimental method is developed in the present work to determine the lift force coefficient in such flow conditions without the aid of moving parts so that the lift force can be measured in any chemical system easily.

Cette thèse projetait d’étudier à la fois l’hydrodynamique et le transfert de masse dans les colonnes de pulvérisation et les colonnes à bulles, et de comparer leurs consommations de puissance spécifiques afin d’élaborer la directive de sélection pour les utilisations industrielles. Les résultats ont indiqué que la colonne à bulles avait une surface interfaciale spécifique plus grande que la colonne de pulvérisation lorsqu’on utilise un faible taux de charge de gaz. Dans cette plage de fonctionnement, la colonne à bulles produisait un coefficient de transfert de masse global plus élevé avec la même consommation d'énergie spécifique. Cependant, lorsqu’elle fonctionnait à un taux de charge de gaz élevé, la colonne de pulvérisation était meilleure que la colonne à bulles, car celle-ci consommait une plus grande consommation d’énergie, la chute de pression de la colonne à bulles étant principalement due au débit de gaz. En outre, cette recherche a également étudié l'effet de la phase solide sur l'hydrodynamique et le transfert de masse dans la colonne à bulles et la colonne de pulvérisation. En utilisant la méthode colorimétrique de l'expérience «bouteille rouge», il a été constaté que la collision bulle-particules diminuait le transfert de masse des bulles car la collision ralentissait les bulles, en particulier pour les petites bulles, du fait que la petite bulle perdait simplement leur vitesses de la collision. Cependant, l’introduction des particules présentait un avantage, car les particules solides pouvaient obstruer la bulle en hausse et réduire sa vitesse de montée. En conséquence, le temps de contact entre le gaz et le liquide est augmenté et conduit à une plus grande rétention de gaz, une zone interfaciale spécifique et donc un transfert de masse. / This thesis projected to investigate both hydrodynamics and mass transfer in both spray and bubble columns and comparing their specific power consumptions in order to develop the selection guideline for industrial usages. The results indicated that the bubble column had larger specific interfacial area than the spray column when using a small gas loading rate. At this range of operation, the bubble column yielded a higher overall mass transfer coefficient with the same specific power consumption. However, when operating at high gas loading rate, the spray column was the one better than the bubble column since the bubble column consumed larger power consumption as the pressure drop of the bubble column was mostly due to the gas flow. In addition, this research also studied the effect of the solid phase on the hydrodynamics and mass transfer in the bubble column and spray column. By using the colorimetric method of “red bottle” experiment, it was found that the bubble-particles collision diminished the mass transfer of bubbles because the collision slowed down the bubbles especially for the small bubbles due to the fact that the small bubble simply lost their velocities from the collision. However, there was an advantage of introducing the particles since solid particles could obstruct the rising bubble and reduced its rising velocity. Consequently, the contact time between gas and liquid is increased and resulted in a higher gas hold up, specific interfacial area and thus mass transfer.

Le but principal de cette étude était d'évaluer et de mieux comprendre l'impact de la vitesse superficielle de gaz et des propriétés de boues activées (BA), sur leur comportement rhéologique et le transfert de l'oxygène dans des bioréacteurs.Tout d'abord, la rhéologie des BA a été évaluée à l'aide d'un rhéomètre tubulaire, conçu et construit dans ce travail. Des mesures rhéologiques ont été effectuées avec des BA provenant de cinq stations d'épuration (STEP) et avec des concentrations en MES comprises entre 2.3 et 10.2 g L-1. Selon ces résultats, la rhéologie des BA est significativement déterminée par la concentration en matière en suspension (MES) mais d'autres caractéristiques liées à leur origine, tel que la taille, la cohésion et la densité du floc, peuvent aussi influencer la viscosité apparente des boues. Basé sur les rheogrames expérimentaux, le modèle rhéologique issu de cette étude est comparé à des modèles rhéologiques existants.Deuxièmement, le transfert d'oxygène a été évalué dans une colonne à bulles (Hc=4.4 m, Dc=0.29m) installée dans deux STEP: une installation classique et un bioréacteur à membrane. La colonne, alternativement équipée d'un diffuseur fines ou grosses bulles (FB, GB), a été alimentée en continu avec des BA extraites du réacteur d'aération, ou de la boucle de recirculation ou du réacteur membranaire. Pour des MES comprises entre 3.0 et 10.4 g L-1, le coefficient kLa a étéplus faible dans les BA que dans l'eau propre et encore réduit avec une augmentation des MES. Cette diminution est en partie attribuable à la réduction observée de la rétention de gaz (εG), associée à une augmentation de la viscosité apparente des boues, celle-ci entrainant une réduction de l'aire interfaciale spécifique (a) due à la coalescence de bulles et à la formation de bulles plus grosses. Aussi, la concentration des tensioactifs non ioniques, a montré un effetnégatif sur le coefficient kLa lors des tests d'oxygénation effectués en aération FB et faibles concentrations en MES. Cet impact n'a pas été observé dans des conditions d'aération GB, ce qui a été expliqué par le taux de renouvellement d’interface plus élevé généré par ces dernières. Enfin, le taux de cisaillement moyen exercé par l'essaim de bulles dans la colonne pendant les tests d’oxygénation a été théoriquement évalué compte tenu des conditions d’opération. Par lasuite, des corrélations empiriques ont été construites en utilisant des nombres adimensionnels et expriment le coefficient kLa en fonction de la vitesse superficielle de gaz et la viscosité apparente, tout en considérant sa dépendance du taux de cisaillement. Enfin, le facteur alpha est défini comme une loi de puissance décroissante en fonction de la viscosité apparente, pour des systèmes à faible chargé. / The main purpose of this study was to evaluate and better understand the impact of superficial gas velocity and activated sludge properties,on activated sludge rheology and oxygen transfer in bioreactors. First of all, activated sludge rheology was evaluated using a tubular rheometer, designed and constructed in this work. Rheological measurements were performed with activated sludge from five different wastewater treatment plants and with MLSS concentrations between 2.3 and 10.2 g L-1. Results showed that although the sludge rheology is significantly defined by the MLSS concentration, other sludge characteristics related to the sludge orign, such as such as floc size, floc cohesiveness and floc density also influence the sludge apparent viscosity. Existing rheological models were evaluated on the set of obtained experimental flow curves. Besides, the oxygen transfer is evaluated in a bubble column (Hc=4.4, Dc=0.29 m) installed in two different wastewater treatment plants: a conventional activated sludge plant (CAS) and a membrane bioreactor (MBR). The column, alternatively equipped with a fine or a coarse bubble diffuser (FB, CB), was continuously fed with activated sludge extracted either from the aeration tank, the recirculation loop or the membrane reactor. With MLSS concentrations from 3.0 to 10.4 g L-1, the kkllaa coefficient was lower in activated sludge than in clean water and still reduced with an increase of the MLSS concentration. This reduction is partially attributed to the observed reduction of gas holdup (εεɢɢ), associated with an increase in the sludge apparent viscosity (µµαααααα), which leads to a reduction of the specific interfacial area (αα) due to bubble coalescence and the formation of larger bubbles. Besides, the concentration of non-ionic surfactants, exhibited a negative effect on the κκιιαα coefficient for the oxygenation tests performed under FB aeration conditions and low MLSS concentration. This impact was not observed under CB aeration conditions, which was explained by the higher renewal rates generated by coarse bubbles. Finally, the mean shear rate exerted by the bubble swarm in the column during the oxygen transfer tests was theoretically evaluated considering the operating conditions. Subsequently, empirical correlations were constructed using dimensionless numbers and express the oxygen transfer coefficient as a function of the superficial gas velocity and the apparent viscosity, considering its shear rate dependence. Finally, alpha factor is defined as a power law decreasing function of the apparent viscosity, for low loaded activated sludge systems.

Carbon dioxide hydrates were synthesized in a 0.10m I.D. and 1.22m tall bubble column equipped with a cooling jacket for heat removal. Visual observations at different driving forces (pressures between 2.75 and 3.60 MPa and temperatures between 0 and 8ºC) were recorded with a digital camera through a sight glass of 118.8 by 15.6 mm. The superficial gas velocity was varied from 20 to 50 mm/s to attain different levels of turbulence in the liquid. The growth rate was found to be dependent on the sequence/method used to reach the operating temperature and pressure. A greater supersaturation was obtained when the system temperature and pressure were reached with very low or no bubble-induced mixing. As a result, hydrates nucleated and grew immediately when starting the gas flow with the reactor volume being quickly filled with hydrates. Moreover, the hydrate growth rate and solution final density were higher when operating conditions partially condensed CO2 resulting in greater interphase mass transfer rates. In parallel, since hydrate formation is an exothermic process and the reaction is often limited by the rate of heat removal, heat transfer measurements were achieved in a simulated hydrate environment. The instantaneous heat transfer coefficient and related statistics gave insight on the role of bubbles on heat transfer and hydrodynamics.

Carbon dioxide hydrates were synthesized in a 0.10m I.D. and 1.22m tall bubble column equipped with a cooling jacket for heat removal. Visual observations at different driving forces (pressures between 2.75 and 3.60 MPa and temperatures between 0 and 8ºC) were recorded with a digital camera through a sight glass of 118.8 by 15.6 mm. The superficial gas velocity was varied from 20 to 50 mm/s to attain different levels of turbulence in the liquid. The growth rate was found to be dependent on the sequence/method used to reach the operating temperature and pressure. A greater supersaturation was obtained when the system temperature and pressure were reached with very low or no bubble-induced mixing. As a result, hydrates nucleated and grew immediately when starting the gas flow with the reactor volume being quickly filled with hydrates. Moreover, the hydrate growth rate and solution final density were higher when operating conditions partially condensed CO2 resulting in greater interphase mass transfer rates. In parallel, since hydrate formation is an exothermic process and the reaction is often limited by the rate of heat removal, heat transfer measurements were achieved in a simulated hydrate environment. The instantaneous heat transfer coefficient and related statistics gave insight on the role of bubbles on heat transfer and hydrodynamics.

Carbon dioxide hydrates were synthesized in a 0.10m I.D. and 1.22m tall bubble column equipped with a cooling jacket for heat removal. Visual observations at different driving forces (pressures between 2.75 and 3.60 MPa and temperatures between 0 and 8ºC) were recorded with a digital camera through a sight glass of 118.8 by 15.6 mm. The superficial gas velocity was varied from 20 to 50 mm/s to attain different levels of turbulence in the liquid. The growth rate was found to be dependent on the sequence/method used to reach the operating temperature and pressure. A greater supersaturation was obtained when the system temperature and pressure were reached with very low or no bubble-induced mixing. As a result, hydrates nucleated and grew immediately when starting the gas flow with the reactor volume being quickly filled with hydrates. Moreover, the hydrate growth rate and solution final density were higher when operating conditions partially condensed CO2 resulting in greater interphase mass transfer rates. In parallel, since hydrate formation is an exothermic process and the reaction is often limited by the rate of heat removal, heat transfer measurements were achieved in a simulated hydrate environment. The instantaneous heat transfer coefficient and related statistics gave insight on the role of bubbles on heat transfer and hydrodynamics.

Carbon dioxide hydrates were synthesized in a 0.10m I.D. and 1.22m tall bubble column equipped with a cooling jacket for heat removal. Visual observations at different driving forces (pressures between 2.75 and 3.60 MPa and temperatures between 0 and 8ºC) were recorded with a digital camera through a sight glass of 118.8 by 15.6 mm. The superficial gas velocity was varied from 20 to 50 mm/s to attain different levels of turbulence in the liquid. The growth rate was found to be dependent on the sequence/method used to reach the operating temperature and pressure. A greater supersaturation was obtained when the system temperature and pressure were reached with very low or no bubble-induced mixing. As a result, hydrates nucleated and grew immediately when starting the gas flow with the reactor volume being quickly filled with hydrates. Moreover, the hydrate growth rate and solution final density were higher when operating conditions partially condensed CO2 resulting in greater interphase mass transfer rates. In parallel, since hydrate formation is an exothermic process and the reaction is often limited by the rate of heat removal, heat transfer measurements were achieved in a simulated hydrate environment. The instantaneous heat transfer coefficient and related statistics gave insight on the role of bubbles on heat transfer and hydrodynamics.

Consumption of citrus fruits and citrus juice production creates wastes, which could be valorized by using it for cultivating fungi. Before cultivation, the medium needs to be sterilized though autoclavation. Larger volumes used when autoclaving requires longer heating cycles and therefore runs the risk of degrading the medium to a greater extent. This research examines the effects of the volume lemon waste medium used while sterilizing. The aim is to find the largest volume still providing good growth for the filamentous fungus used, Rhizopus Delemar. Lemon waste was provided by Herrljunga Musteri AB and was pre-treated at 45°C for 2h. The liquid was strained and autoclaved in different volumetric series ranging from 200 – 10 000 mL, that was then used in 200 mL shake flask cultivations.  A scale up in two 3,5 L bubble column reactors was also performed from the 10 000 mL autoclaved medium, after not observing severe impacts on growth. Testing was done by weighing biomass and HPLC analysis of sugars. The yield of the biomass in the shake flasks ranged from 0,11 – 0,14 g/g sugars and the biomass concentration ranged between 2,4 - 3,0 g/L. Overall, the volume of autoclavation seems to not too be of great concern when cultivating R. Delemar on lemon waste medium in the analyzed ranges.