A non-reactive model system comprising a highly concentrated and unstable oil-in-water emulsion was used to investigate the retention of oil by the membrane in producing biodiesel with a membrane reactor. Critical flux was identified using the relationship between the permeate flux and transmembrane pressure along with the separation efficiency of the membrane. It was shown that separation efficiencies above 99.5% could be obtained at all operating conditions up to the critical flux. It was observed that the concentration of oil in all collected permeate samples using the oil-water system was below 0.2 wt% when operating at a flux below the critical flux. Studies to date have been limited to the characterization of low concentrated emulsions below 15 vol.%. The average oil droplet size in highly concentrated emulsions was measured as 3200 nm employing direct light scattering (DLS) measurement methods. It was observed that the estimated cake layer thickness of 20 to 80 mm was larger than external diameter of the membrane tube i.e. 6 mm based on a large particle size. Settling of the concentrated emulsion permitted the detection of a smaller particle size distribution (30-100 nm) within the larger particles averaging 3200 nm. It was identified that DLS methods could not efficiently give the droplet size distribution of the oil in the emulsion since large particles interfered with the detection of smaller particles. The content of the smaller particles represented 1% of the total weight of oil at 30°C and 5% at 70°C. This was too low to be detected using DLS measurements but was sufficient to affect ultrafiltration. In order to study the critical flux in the presence of transesterification reaction and the effect of cross flow velocity on separation, various oils were transesterified in another membrane reactor providing higher cross flow velocity. higher cross flow velocity provides better separation by reducing materials deposition on the surface of the membrane due to higher shearing. The oils tested were canola, corn, sunflower and unrefined soy oils (Free Fatty Acids (FFA< 1%)), and waste cooking oil (FFA= 9%). The quality of all biodiesel samples was studied in terms of glycerine, mono-glyceride, di-glyceride and tri-glyceride concentrations. The composition of all biodiesel samples were in the range required by ASTM D6751 and EN 14214 standards. A critical flux based on operating pressure in the reactor was reached for waste cooking and pre-treated corn oils. It was identified that the reaction residence time in the reactor was an extremely important design parameter affecting the operating pressure in the reactor. / Natural Sciences and Engineering Research Council of Canada (NSERC)

A non-reactive model system comprising a highly concentrated and unstable oil-in-water emulsion was used to investigate the retention of oil by the membrane in producing biodiesel with a membrane reactor. Critical flux was identified using the relationship between the permeate flux and transmembrane pressure along with the separation efficiency of the membrane. It was shown that separation efficiencies above 99.5% could be obtained at all operating conditions up to the critical flux. It was observed that the concentration of oil in all collected permeate samples using the oil-water system was below 0.2 wt% when operating at a flux below the critical flux. Studies to date have been limited to the characterization of low concentrated emulsions below 15 vol.%. The average oil droplet size in highly concentrated emulsions was measured as 3200 nm employing direct light scattering (DLS) measurement methods. It was observed that the estimated cake layer thickness of 20 to 80 mm was larger than external diameter of the membrane tube i.e. 6 mm based on a large particle size. Settling of the concentrated emulsion permitted the detection of a smaller particle size distribution (30-100 nm) within the larger particles averaging 3200 nm. It was identified that DLS methods could not efficiently give the droplet size distribution of the oil in the emulsion since large particles interfered with the detection of smaller particles. The content of the smaller particles represented 1% of the total weight of oil at 30°C and 5% at 70°C. This was too low to be detected using DLS measurements but was sufficient to affect ultrafiltration. In order to study the critical flux in the presence of transesterification reaction and the effect of cross flow velocity on separation, various oils were transesterified in another membrane reactor providing higher cross flow velocity. higher cross flow velocity provides better separation by reducing materials deposition on the surface of the membrane due to higher shearing. The oils tested were canola, corn, sunflower and unrefined soy oils (Free Fatty Acids (FFA< 1%)), and waste cooking oil (FFA= 9%). The quality of all biodiesel samples was studied in terms of glycerine, mono-glyceride, di-glyceride and tri-glyceride concentrations. The composition of all biodiesel samples were in the range required by ASTM D6751 and EN 14214 standards. A critical flux based on operating pressure in the reactor was reached for waste cooking and pre-treated corn oils. It was identified that the reaction residence time in the reactor was an extremely important design parameter affecting the operating pressure in the reactor. / Natural Sciences and Engineering Research Council of Canada (NSERC)

Os resultados de uma simulação numérica são apresentados para amplitudes de vibração induzidas por um escoamento transversal num feixe tubular no regime de instabilidade fluidelástica.O feixe tubular considerado tem geometria e características iguais às de uma instalação equivalente descrita na literatura, para a qual estão disponíveis as medições experimentais das amplitudes de vibração no regime de instabilidade elástica.O arranjo tipo triângulo rodado tem uma relação passo/diâmetro de 1,375 e consiste de um tubo móvel cercado por 134 tubos rígidos.A simulação numérica foi efetuada através de um software comercial de CFD (Computational Fluid dynamics).Para a região em torno de cada tubo foi considerada uma malha com dimensões do elemento crescendo geometricamente na direção normal ao tubo com fator de crescimento 1,13 , sendo a dimensão do elemento adjacente ao tubo igual a 0,1% do diâmetro externo do tubo.Na simulação numérica o escoamento foi considerado incompressível, monofásico, turbulento e bidimensional. Os dados do escoamento foram considerados idênticos aos das experiências da instalação descrita na literatura.Os resultados obtidos para as amplitudes pela simulação numérica são comparados com os resultados obtidos experimentalmente na instalação acima citada.Os desvios da maioria dos valores calculados em relação aos valores experimentais estão numa faixa aceitável. Isto mostra que existe a possibilidade de utilização, num futuro próximo, de CFD para análise deste tipo de problemas. / Results from a numerical simulation are reported for amplitudes of cross-flow induced vibrations at the fluid elastic instability regime in a tube bank. The tube bank has identical geometry and characteristics as for an experimental facility described in the literature, for which there are experimental measurements of the amplitudes of vibrations at the fluid elastic instability regime.The rotated triangular array has a pitch ratio of 1.375 and consists of 1 movable tube surrounded by 134 rigid tubes. The numerical simulations were accomplished with a commercial CFD (Computational Fluid Dynamics) software. For the region around each tube, a mesh with elements dimensions growing geometrically normal to the tube was considered, with growing factor 1.13, and the dimension for the element adjacent to the tube wall was set to 0.1% of the tube external diameter. The flow was considered incompressible, monophasic, turbulent and two-dimensional for the numerical simulation. The flow data considered were the same as for the experiments at the facility. The results presented in this paper for the amplitudes obtained by numerical simulation are compared with the experimental results obtained in the above mentioned experimental facility.The differences between the calculated values and the experimental values are acceptable. This show that in the near future there is the possibility to use CFD for these kind of problems.

Para o tratamento de esgoto sanitário foi desenvolvido e estudado um reator UASB (Upflow Anaerobic Sludge Blanket), associado com unidade de filtração. O estudo se iniciou pela operação do reator UASB durante 642 dias, onde a fase líquida afluente e efluente foi monitorada. Em seguida uma unidade de filtração foi acoplada ao reator UASB e foram avaliados as características de filtração, o desempenho do tratamento e a eficiência do processo de limpeza química de três filtros. Os materiais poliméricos utilizados como filtros foram o polipropileno, o poliéster e o geossintético de poliéster, com diâmetro médio de poros de um mícron (1&#956;m). As características de filtração avaliadas foram o fluxo do permeado e os mecanismos de obstrução de cada material polimérico (resistência à filtração: total, do material, da camada de biossólidos e de entupimento). O desempenho global do sistema de tratamento foi avaliado em termos de remoção de demanda química de oxigênio (DQO) e de sólidos suspensos totais (SST). Para recuperação dos filtros foi utilizada primeiramente a limpeza química com solução ácida (ácido clorídrico-HCl) e em série a limpeza com solução alcalina (hipoclorito de sódio e hidróxido de sódio- NaOCl e NaOH). Dentre os três materiais poliméricos utilizados verificou-se que, o geossintético (poliéster) foi o filtro que apresentou melhores características de filtração e desempenho global, ou seja, com este material a resistência da camada foi predominante (obstrução é reversível), a concentração de sólidos suspensos totais do permeado estabilizou em 25 mg/L e a eficiência média global de remoção da demanda química de oxigênio (DQO) foi de 78%. Dentre os procedimentos de limpeza química utilizada nos três filtros, verificou-se que a solução ácida (HCl) aplicada em série com a solução alcalina (NaOCl e NaOH) apresentou maior eficiência que a limpeza somente com a solução ácida (recuperação do fluxo inicial do permeado acima de 90%). / An anaerobic reactor (UASB – Upflow Anaerobic Sludge Blanket) coupled to a filtration unit for the treatment of domestic sewage was developed and studied. The study began by operating the UASB reactor for 642 days, during which time the influent and effluent liquid phase were monitored. The filtration unit was then coupled to the UASB reactor and the filtration characteristics, the performance of the treatment, and the efficiency of the cleaning process of three polymeric filters were evaluated. The polymeric materials used for the filters were polypropylene, polyester and geosynthetic polyester, all with an average pore size of one micron (1&#956;m). The filtration characteristics evaluated were the permeate flux and the fouling mechanisms (total filtration resistance of the material and the cake, and clogging) during filtration. The system’s overall performance was evaluated in terms of chemical oxygen demand (COD) and removal of total suspended solids (TSS). The filters were cleaned chemically, first with an acid solution (HCl), followed by alkaline solutions (NaOH and NaOCl) applied in series. A comparison of the three filtering materials indicated that the geosynthetic material (polyester) possessed the best overall characteristics of filtration and performance. In other words, i.e., cake resistance (reversible fouling) predominated with this material, the concentration of total suspended solids (TSS) in the permeate stabilized at 25 mg/L, and the overall removal efficiency of chemical oxygen demand (COD) was 78%. Of the various chemical cleaning procedures tested on the three polymeric filters, the acid (HCl) and alkaline solutions (NaOH and NaOCl) applied in series provided more efficient flux recovery (over 90% of the initial permeate) than chemical cleaning only with acid (HCl).

Os resultados de uma simulação numérica são apresentados para amplitudes de vibração induzidas por um escoamento transversal num feixe tubular no regime de instabilidade fluidelástica.O feixe tubular considerado tem geometria e características iguais às de uma instalação equivalente descrita na literatura, para a qual estão disponíveis as medições experimentais das amplitudes de vibração no regime de instabilidade elástica.O arranjo tipo triângulo rodado tem uma relação passo/diâmetro de 1,375 e consiste de um tubo móvel cercado por 134 tubos rígidos.A simulação numérica foi efetuada através de um software comercial de CFD (Computational Fluid dynamics).Para a região em torno de cada tubo foi considerada uma malha com dimensões do elemento crescendo geometricamente na direção normal ao tubo com fator de crescimento 1,13 , sendo a dimensão do elemento adjacente ao tubo igual a 0,1% do diâmetro externo do tubo.Na simulação numérica o escoamento foi considerado incompressível, monofásico, turbulento e bidimensional. Os dados do escoamento foram considerados idênticos aos das experiências da instalação descrita na literatura.Os resultados obtidos para as amplitudes pela simulação numérica são comparados com os resultados obtidos experimentalmente na instalação acima citada.Os desvios da maioria dos valores calculados em relação aos valores experimentais estão numa faixa aceitável. Isto mostra que existe a possibilidade de utilização, num futuro próximo, de CFD para análise deste tipo de problemas. / Results from a numerical simulation are reported for amplitudes of cross-flow induced vibrations at the fluid elastic instability regime in a tube bank. The tube bank has identical geometry and characteristics as for an experimental facility described in the literature, for which there are experimental measurements of the amplitudes of vibrations at the fluid elastic instability regime.The rotated triangular array has a pitch ratio of 1.375 and consists of 1 movable tube surrounded by 134 rigid tubes. The numerical simulations were accomplished with a commercial CFD (Computational Fluid Dynamics) software. For the region around each tube, a mesh with elements dimensions growing geometrically normal to the tube was considered, with growing factor 1.13, and the dimension for the element adjacent to the tube wall was set to 0.1% of the tube external diameter. The flow was considered incompressible, monophasic, turbulent and two-dimensional for the numerical simulation. The flow data considered were the same as for the experiments at the facility. The results presented in this paper for the amplitudes obtained by numerical simulation are compared with the experimental results obtained in the above mentioned experimental facility.The differences between the calculated values and the experimental values are acceptable. This show that in the near future there is the possibility to use CFD for these kind of problems.

The high complexity of vehicle front-end design, arising from considerations of aerodynamics, safety and styling, causes the airflow velocity profile at the radiator face to be highly distorted, leading to potentially reduced airflow volume for heat dissipation. A flow visualisation study showed that the bumper bar significantly influenced the cooling airflow, leading to three-dimensional vortices in its wake and generating an area of relatively low velocity across at least one third of the radiator core. Since repeatability and accuracy of on-road testing are prejudiced by weather conditions, wind-tunnel testing is often preferred to solve cooling airflow problems. However, there are constraints that limit the accuracy of reproducing on-road cooling performance from wind-tunnel simulations. These constraints included inability to simulate atmospheric conditions, limited tunnel test section sizes (blockage effects) and lack of ground effect simulations. The work presented in this thesis involved use of on-road and wind-tunnel tests to investigate the effects of most common constraints present in wind tunnels on accuracy of the simulations of engine cooling performance and radiator airflow profiles. To aid this investigation, an experimental technique for quantifying radiator airflow velocity distribution and an analytical model for predicting the heat dissipation rate of a radiator were developed. A four-hole dynamic pressure probe (TFI Cobra probe) was also used to document flow fields in proximity to a section of radiator core in a wind tunnel in order to investigate the effect of airflow maldistribution on radiator heat-transfer performance. In order to cope with the inability to simulate ambient temperature, the technique of Specific Dissipation (SD) was used, which had previously been shown to overcome this problem.

<p>Ved består huvudsakligen av tre typer av polymerer, cellulosa, hemicellulosa och lignin. Lignin bildas i naturen genom enzymatiskt initierad oxidativ koppling av tre olika typer av fenylpropan-enheter. Dessa bygger genom olika kol-kol- och kol-syre-bindningar upp en amorf tredimensionell polymer. När kemisk massa tillverkas bryts lignin ner och löses ut i kokluten. Luten innehåller de förbrukade kokkemikalierna och bränns generellt i en sodapanna för att regenerera kemikalierna och producera ånga. Sodapannan är emellertid dyr. Därför har den blivit produktionsbegränsande på många massabruk. Att avlägsna en del av ligninet från avluten vore därför önskvärt och att finna ekonomiskt intressanta produkter baserade på lignin från svartlut är därför ett viktigt forskningsområde .</p><p>Ett lämpligt område för ligninprodukter vore som tillsatts i oblekt massa. Oblekt massa används till stor del för tillverkning av kraftliner, topp- och bottenskikten på wellpapp. När lådor av wellpapp lagras i containrar som färdas över haven, förändras den relativa luftfuktigheten. Detta gör att lådorna kollapsar lättare än om de skulle ha lagrats vid konstant luftfuktighet, även en hög sådan. Detta är på grund av det så kallade mekanosorptiva- eller accelererade krypfenomenet. Genom tillsatts av våtstyrkemedel till kraftliner eller behandla den med hydrofoba ämnen, finns indikatoner på att mekanosorptiva effekten skulle kunna minska.</p><p>För att försöka minska den effekten har ett lågmolekylärt kraftlignin, som utvunnits med hjälp av tvärsflödesfiltrering av svartlut och svavelsyrafällning, använts. Genom derivatisering av detta lignin med linolja erhölls ett hydrofobt ligninderivat som uppvisar strukturella likheter med biopolymeren suberin. När detta suberinlika ligninderivat tillsätts till massa verkar det mekanosorptiva krypet minska. När lågmolekylärt lignin används tillsammans med ligninradikalinitiatorerna lackas eller mangan(III) i kraftlinermassa erhålls dessutom en våtstyrka på ca 5% av torrstyrkan. Efter aminering av detta lignin gav en tillsatts till kraftlinermassan en våtstyrka på upp till 10% av torrstyrkan. Det finns indikationer på att det mekanosorptiva krypet samtidigt minskar när dessa behandlingar görs som ger upphov till ökad våtstyrka.</p> / <p>Wood consists mainly of three types of polymers; cellulose, hemi cellulose and lignin. Lignin is formed in nature through enzymatic initiated oxidative coupling of three different kinds of phenyl propane units. These form by various carbon-carbon and carbon-oxygen bonds, an amorphous three-dimensional polymer. As chemical pulp is produced, lignin is degraded and dissolved into pulping liquors. These liquors contain the spent cooking chemicals and are generally burnt in a recovery boiler to regenerate cooking chemicals and produce steam. However, the recovery boiler is expensive. Hence, it has become the bottleneck for production in many pulp mills. Removal of some lignin from the spent cooking liquor would, for that reason, be desired and valuable products based on lignin from cooking liquors are searched for.</p><p>One suitable area for lignin products would be as additive in unbleached pulp. A major product from unbleached pulp is kraftliner, the top and bottom layers of corrugated board. When boxes of corrugated board are stored in containers travelling overseas the relative humidity is varying. This makes the boxes collapse more easily than if they were stored at constant humidity, even a high one. This is due to the so called mechano-sorptive or accelerated creep phenomenon. By addition of wet strength additive to kraftliner or treating it with hydrophobic compounds there are indications on that the mechano-sorptive effect would decrease.</p><p>Trying to decrease this effect, low molecular weight kraft lignin has been used. It was obtained by cross-flow filtration of black liquor and precipitation by sulphuric acid. By derivatisation of this lignin by linseed oil, a hydrophobic lignin derivative was obtained, similar in structure to units in the biopolymer suberin. As this suberin-like lignin-derivative was added to pulp the mechano-sorptive creep seemed to be lowered. Furthermore, when the low molecular weight lignin was used together with the lignin radical initiators laccase or manganese(III) in kraftliner pulp, a wet strength of about 5% of dry strength was obtained. An amination treatment of this lignin and addition to kraftliner pulp resulted in a wet strength of up to 10% of dry strength. There are indications of that the mechano-sorptive creep also decreases as these treatments, resulting in increased wet strength, are made.</p>

La récupération, grâce aux hydroliennes, de l’énergie cinétique de courants marins et fluviaux constitue une source d’énergie renouvelable considérable et prédictible. La simulation fine, par une description statistique instationnaire de type URANS, de l’écoulement autour d’une hydrolienne isolée à axe vertical, bi-rotor et munie d’un carénage (hydrolienne de type HARVEST) donne accès à une estimation précise de la puissance produite. Cependant, le coût élevé de cette approche URANS la rend inadaptée à la simulation d’un parc de machines. Une analyse de la littérature conduit à retenir un modèle basse-fidélité de type Blade Element Momentum (BEM) pour décrire à moindre coût l’effet du rotor de la turbine sur l’écoulement, dans le contexte d’une description 2D (coupe horizontale). La performance de l’hydrolienne est alors prédite par un calcul RANS incluant des termes sources distribués dans un anneau rotor virtuel et conservant le maillage des parties fixes (carénage). Ces termes sources sont construits grâce à une procédure originale exploitant les conditions locales de l’écoulement en amont des cellules du rotor virtuel et le débit de l’écoulement traversant l’hydrolienne. Les coefficients hydrodynamiques utilisés pour le calcul des termes sources BEM-RANS sont construits une fois pour toutes en exploitant une série de simulations URANS préliminaires ; ils intègrent les effets du carénage et le fonctionnement de chaque rotor à une vitesse de rotation optimale (maximisant la puissance produite) grâce au système de régulation de l’hydrolienne. Le modèle BEM-RANS développé est validé par comparaison avec des simulations URANS de référence : il fournit une estimation fiable de la puissance produite (erreur de quelques % par rapport à l’approche URANS) pour un coût réduit de plusieurs ordres de grandeur. Ce modèle est appliqué à l'analyse de la puissance produite par une rangée d’hydroliennes HARVEST dans un canal pour différents facteurs de blocage et d’espacement latéral ainsi qu’à une ferme marine composée de trois hydroliennes. / The capture, thanks to hydrokinetic turbines, of the kinetic energy generated by sea and river currents provides a significant and predictable source of renewable energy. The detailed simulation, using an unsteady statistical description of URANS type, of the flow around an isolated water turbine of HARVEST type (cross flow vertical axis ducted water turbine) provides an accurate estimate of the power output. However, the cost of the URANS approach is much too expensive to be applied to a farm of several turbines. A review of the literature leads to select a low-fidelity model of Blade Element Momentum (BEM) type to describe at a reduced cost the rotor effect on the flow, in a 2D context (horizontal cross-section). The turbine performance is then predicted using a steady RANS simulation including source terms distributed within a virtual rotor ring and preserving the mesh of the turbine fixed parts (duct). These source terms are derived using an original procedure which exploits both the local flow conditions upstream of the virtual rotor cells and the flow rate through the turbine. The hydrodynamic coefficients used to compute the BEM-RANS source terms are built once for all from a series of preliminary URANS simulations; they include the effects of the duct on the flow and the rotor operating at optimal rotational speed (maximizing the power output) thanks to the turbine regulation system. The BEM-RANS model is validated against reference URANS simulations: it provides a reliable prediction for the power output (within a few % of the URANS results) at a computational cost which is lowered by several orders of magnitude. This model is applied to the analysis of the power produced by a row of Vertical Axis Water Turbines in a channel for various values of the blockage ratio and lateral spacing as well as to a 3-machine sea farm.

Lors du développement d’un nouvel avion, l’estimation des échanges d’énergie entre l’air ambiant et les parois est une donnée cruciale pour la conception aérothermique. Cette conception repose de plus en plus sur des simulations numériques mais certains phénomènes d’aérothermique externe, comme le jet débouchant du système de dégivrage des nacelles moteur, montrent les limites des modèles RANS classiques. La simulation des grandes échelles (LES) se révèle bien adaptée à ce type de phénomène mais se heurte à un coût de calcul extrêmement élevé pour ces écoulements pariétaux à très grand nombre de Reynolds. Pour lever cette limitation, cette thèse propose l’étude de deux briques fondamentales : la LES avec loi de paroi (WMLES) conjuguée à l’injection d’une couche limite turbulente à l’entrée du domaine. Pour une meilleure compréhension et une utilisation fiable de l’approche loi de paroi, on se concentre tout d’abord sur les sources d’erreur qui lui sont associées. Après les avoir identifiées, on propose une correction de l’erreur de sous-maille ainsi qu’une loi de paroi adaptée aux écoulements compressibles. Grâce à ces deux éléments, on obtient une estimation correcte du flux de chaleur pariétal sur des simulations WMLES de canal plan supersonique sur parois froides. Puis, pour préparer la transition vers des applications plus industrielles, on introduit un schéma numérique plus dissipatif ce qui nous permet d’étudier l’influence de la méthode numérique sur l’approche loi de paroi. Dans une seconde partie dédiée à l’injection de couche limite pour la WMLES, on sélectionne une méthode basée sur l’injection de perturbations combinée à un terme de contrôle volumique. On montre que des simulations WMLES utilisant cette méthode d’injection permettent d’établir une couche limite turbulente réaliste à une courte distance en aval du plan d’entrée, à la fois sur une plaque plane mais également sur un écoulement de jet débouchant à la géométrie plus complexe, représentative d’un cas avion. / During the design of a new aircraft, the prediction of energy exchanged between the ambient air and the aircraft walls is crucial regarding aerothermal design. Numerical simulations plays a role of increasing importance in this design. However classical RANS models reach their limits on some external aerothermal flows, like the jet-in-cross-flow from the anti-icing system oh the engine nacelles. The large eddy simulation (LES) is well suited to this kind of flow but faces an extremely large computational cost for such high Reynolds number wall-bounded flows. To remove this limitation, we propose two building blocks: the Wall Modeled LES (WMLES) combined with a turbulent inflow generation. For a better understanding and a reliable use of the WMLES, we first focus on the sources of error related to this approach. We propose a correction to the subgrid-scale error as well as a wall model suitable for compressible and anisothermal flows. Thanks to these two elements, we correctly predict the wall heat flux in WMLES computations of a supersonic isothermal-wall channel flow. Then, to allow the computation of more industrial flows, we introduce some numerical dissipation and study its effect on the wall modeling approach. The last part is dedicated to turbulent inflow generation for WMLES. We select a method based on synthetic perturbation combined with a dynamic control term. We validate this method on WMLES computations of a flat plate turbulent boundary layer and a hot jet-in-cross-flow representative of an industrial configuration. In both cases, we show that a realistic turbulent boundary layer is generated at a small distance downstream from the inlet plane.

Para o tratamento de esgoto sanitário foi desenvolvido e estudado um reator UASB (Upflow Anaerobic Sludge Blanket), associado com unidade de filtração. O estudo se iniciou pela operação do reator UASB durante 642 dias, onde a fase líquida afluente e efluente foi monitorada. Em seguida uma unidade de filtração foi acoplada ao reator UASB e foram avaliados as características de filtração, o desempenho do tratamento e a eficiência do processo de limpeza química de três filtros. Os materiais poliméricos utilizados como filtros foram o polipropileno, o poliéster e o geossintético de poliéster, com diâmetro médio de poros de um mícron (1&#956;m). As características de filtração avaliadas foram o fluxo do permeado e os mecanismos de obstrução de cada material polimérico (resistência à filtração: total, do material, da camada de biossólidos e de entupimento). O desempenho global do sistema de tratamento foi avaliado em termos de remoção de demanda química de oxigênio (DQO) e de sólidos suspensos totais (SST). Para recuperação dos filtros foi utilizada primeiramente a limpeza química com solução ácida (ácido clorídrico-HCl) e em série a limpeza com solução alcalina (hipoclorito de sódio e hidróxido de sódio- NaOCl e NaOH). Dentre os três materiais poliméricos utilizados verificou-se que, o geossintético (poliéster) foi o filtro que apresentou melhores características de filtração e desempenho global, ou seja, com este material a resistência da camada foi predominante (obstrução é reversível), a concentração de sólidos suspensos totais do permeado estabilizou em 25 mg/L e a eficiência média global de remoção da demanda química de oxigênio (DQO) foi de 78%. Dentre os procedimentos de limpeza química utilizada nos três filtros, verificou-se que a solução ácida (HCl) aplicada em série com a solução alcalina (NaOCl e NaOH) apresentou maior eficiência que a limpeza somente com a solução ácida (recuperação do fluxo inicial do permeado acima de 90%). / An anaerobic reactor (UASB – Upflow Anaerobic Sludge Blanket) coupled to a filtration unit for the treatment of domestic sewage was developed and studied. The study began by operating the UASB reactor for 642 days, during which time the influent and effluent liquid phase were monitored. The filtration unit was then coupled to the UASB reactor and the filtration characteristics, the performance of the treatment, and the efficiency of the cleaning process of three polymeric filters were evaluated. The polymeric materials used for the filters were polypropylene, polyester and geosynthetic polyester, all with an average pore size of one micron (1&#956;m). The filtration characteristics evaluated were the permeate flux and the fouling mechanisms (total filtration resistance of the material and the cake, and clogging) during filtration. The system’s overall performance was evaluated in terms of chemical oxygen demand (COD) and removal of total suspended solids (TSS). The filters were cleaned chemically, first with an acid solution (HCl), followed by alkaline solutions (NaOH and NaOCl) applied in series. A comparison of the three filtering materials indicated that the geosynthetic material (polyester) possessed the best overall characteristics of filtration and performance. In other words, i.e., cake resistance (reversible fouling) predominated with this material, the concentration of total suspended solids (TSS) in the permeate stabilized at 25 mg/L, and the overall removal efficiency of chemical oxygen demand (COD) was 78%. Of the various chemical cleaning procedures tested on the three polymeric filters, the acid (HCl) and alkaline solutions (NaOH and NaOCl) applied in series provided more efficient flux recovery (over 90% of the initial permeate) than chemical cleaning only with acid (HCl).