Os desreguladores endócrinos, como o bisfenol A (BPA), são compostos encontrados na água em concentrações da ordem de μg.L-1 ou ng.L-1, sendo por isso também denominados micropoluentes. Sua presença, mesmo em baixas concentrações, pode causar prejuízos aos organismos expostos. Neste contexto, o objetivo do presente trabalho foi avaliar a remoção de BPA por membranas de ultrafiltração (UF) e osmose inversa (OI) e por carvão ativado granular (CAG). Essas técnicas têm a vantagem de não gerar subprodutos que também podem ser tóxicos. Nos estudos foram realizados experimentos para remoção do BPA através de membrana PL-1 (celulose regenerada de 1 kDa), Sy-10 (polietersulfona de 10 kDa) e membrana de OI (poliamida) reutilizada, com concentração inicial de 500 μg.L-1 de BPA. Para a membrana Sy-10, foram testados os pHs 7 e 10. Os ensaios de adsorção foram realizados com CAG (1-2mm) em frascos contendo 100 mL da solução de BPA, com 0,5 g de CAG para cada frasco. As variáveis testadas foram pH, tempo de contato e concentração inicial na ordem de mg.L-1. Adicionalmente, foi testada a concentração inicial de 500 μg.L-1 de BPA em experimento de adsorção. Os resultados obtidos mostraram que a capacidade da membrana Sy-10 para remoção de BPA foi superior, chegando a cerca de 90% de remoção, em contraposição aos 20% encontrados para PL-1. A membrana de OI apresentou remoções de cerca de 95%. A influência do pH (7 e 10) na remoção do BPA para a membrana Sy-10 não mostrou-se significativa. Nos ensaios de adsorção, analisando diferentes valores de pH, obteve-se eficiência de 93% para pH 7. Ensaios de variação do tempo de contato com CAG mostraram que o equilíbrio é atingido nos primeiros 10 minutos para a maior concentração e em 40 minutos para a menor concentração testada, de 500 μg.L-1. O estudo do efeito da concentração inicial de BPA na adsorção mostrou que a remoção aumenta com o aumento da concentração inicial do poluente. Conclui-se que tanto a adsorção em CAG quanto membranas são boas alternativas para a remoção de BPA de soluções aquosas. As duas técnicas poderiam ser usadas conjuntamente, sendo que o concentrado do processo com membranas poderia ser submetido à adsorção por CAG. / Endocrine disrupting chemicals, such as bisphenol A (BPA) compounds, are found in water at concentrations of the order of μg.L-1 or ng.L-1 is therefore also known micropollutants. His presence, even at low concentrations, can cause damage to exposed organisms. In this context, the aim of this study was to evaluate the removal of BPA by ultrafiltration (UF) and reverse osmosis (RO) membranes and granular activated carbon (GAC). These techniques have the advantage of not generating by-products can also be toxic. In the studies were performed experiments to remove the BPA by PL-1 (1 kDa regenerated cellulose), Sy-10 (10 kDa polyethersulfone) and RO (polyamide) reused membranes, with initial concentration of 500 μg.L-1 of BPA. For Sy-10 membrane pHs 7 and 10 were tested. The adsorption experiments were carried out with GAC (1-2mm) in flasks containing 100 mL of the BPA solution and 0.5 g of GAC to each vial. The variables tested were pH, contact time and initial concentration on the order of mg.L-1. Additionally, we tested the initial concentration of 500 μg.L-1 of BPA in adsorption experiment. Results showed that the capacity of Sy-10 membrane to remove BPA was higher, reaching approximately 90% removal, as opposed to the 20% found to PL-1. The RO membrane showed removal of about 95%. The influence of pH (7 and 10 ) in the removal of BPA onto the membrane Sy - 10 was not significant. In adsorption tests, analyzing different pH values was obtained efficiency from 93% to pH 7. Testing time variation of contact GAC showed that equilibrium is reached in the first 10 minutes to the largest concentration and 40 minutes to the lowest concentration tested, 500 μg.L-1. The study of the effect of the initial concentration of BPA in the adsorption showed that removal increases with increase in the initial concentration of the pollutant. It was concluded that both the GAC adsorption as membranes are good alternatives for the removal of the BPA aqueous solutions. The two techniques could be used together, with the concentrate from the membrane process could be subjected to adsorption GAC.

L'osmose inverse (OI) est actuellement le procédé le plus utilisé mondialement pour le dessalement des eaux saumâtres et de l’eau de mer. Cette thèse s'est intéressée à la simulation moléculaire de membranes d'OI afin d'améliorer la compréhension des propriétés structurales, dynamiques et de transport de l'eau et d'ions à l'intérieur de ces matériaux. La membrane d'OI étudiée dans ce travail est une membrane de polyamide aromatique, matériau le plus utilisé actuellement en OI. Dans la première partie de ce travail, une méthodologie a été développée pour construire un modèle atomique en trois dimensions d'une membrane polyamide fortement réticulé. Des simulations de dynamique moléculaire à l’équilibre (EMD) et hors-équilibre (NEMD) ont été réalisées pour étudier le comportement de l'eau et des ions Na+ et Cl- à travers la membrane. Les simulations EMD ont montré que les caractéristiques structurales de la membrane modèle étaient en bon accord avec celles d'une membrane typique d'OI. Les propriétés dynamiques et diélectriques de l'eau confinée dans la membrane ont également été étudiées et il a été montré que celles-ci étaient fortement modifiées par rapport à une phase volumique. Deux types de techniques NEMD ont été utilisés pour étudier le transport baromembranaire à travers la membrane modèle. La perméabilité à l'eau pure a été trouvée en très bon accord avec les données expérimentales rapportées dans la littérature et les deux méthodes NEMD ont révélé une très forte rétention saline, confirmant ainsi la pertinence du modèle de membrane d'OI développé dans ce travail. / Reverse osmosis (RO) is currently the leading process used worldwide for both brackish and seawater desalination. This thesis focuses on the molecular simulation of RO membranes in order to improve the understanding of structure, dynamics and transport of water and ions inside these materials. The RO membrane studied in this work is a typical polyamide RO membrane. In the first step of this work, a methodology for building a fully atomic and three-dimensional model of a highly cross-linked polyamide membrane was developed. Both equilibrium molecular dynamics (EMD) and non-equilibrium molecular dynamics (NEMD) simulations were further performed to investigate the behavior of water and ions (Na+ and Cl-) through the membrane. EMD simulations showed that the structural characteristics of the model polyamide membrane were in good agreement with those of a typical RO membrane. The dynamics and dielectric properties of water confined in the RO membrane were also studied and have shown to be dramatically modified with respect to the bulk phase. Two types of NEMD techniques were employed to investigate pressure-driven transport through the model membrane. Pure water permeability was found to be in very good agreement with experimental data reported in the literature for similar membrane materials and both NEMD methods highlighted very high salt rejection properties, thus confirming the relevance of the model membrane developed in this work.

Several commercially available reverse osmosis (RO) membranes were characterized with aqueous solutions of ammonium sulfate, potassium triphosphate, and mixtures of these two salts at different feed concentrations, compositions and pressures. The objective of this study was to investigate the rejection of these solutes, in particular the ammonium ion (NH4+), by different RO membranes. The aqueous solutions were assumed to come from an anaerobic digester via a process, currently under investigation by CHFour Biogas Inc., to maintain low concentrations of ammonia in the digester in order to maximize the biogas production. The ammonium ions present in the liquid produced from the process are then concentrated using membrane separation. The concentrated ammonium solution would be a valuable fertilizer that could be used by agriculture. The membranes were characterized with three models: the solution-diffusion model, the Kedem-Katchalsky model, also known as the irreversible thermodynamics model, and the Donnan Steric Pore Model (DSPM). The solution-diffusion and irreversible thermodynamics models were found to be inadequate for proper membrane characterization and the use of the DSPM model yielded membrane properties in good agreement with those found in already existing literature. The pore radius of investigated membranes ranged from 0.39 to 0.51 nm. The effect of pH on membrane surface charge was also studied, with the conclusion that increases in pH led to increasingly negative surface charges. This affected the transport of individual ions through the membrane due to preferential passage of the counter-ions. The effects of applied pressure on the stoichiometric nature of salt rejections were also studied. The minimal observed rejection from the range of experiments carried out using ammonium sulfate was 93%Non-stoichiometric rejections of ions were also observed in the experiments with single and multiple solutes. Furthermore, the rejection of ammonium ions in the presence of other ions (K+, SO42-, PO43-) increased as feed concentration increased, which was a result of the synergistic effects of feed pH and ionic interactions. The minimum NH4+ rejection in the presence of other ions was 95.4%, which suitability using RO membranes for concentration of ammonium from dilute aqueous solutions.

Thin film Nanocomposite (TFN) membranes are a relatively new class of high-performance semipermeable membranes for Reverse Osmosis (RO) applications. Large scale applications of TFN membranes have not been achieved yet due to the high production cost of the nanoparticles, agglomeration of the nanoparticles in the thin polyamide matrix of the membrane, and leaching out of typically toxic inorganic nanoparticles into the downstream. In this work, these challenges are addressed by incorporation of two different nanofillers: Cellulose NanoCrystals (CNC), and surface functionalized Halloysite NanoTubes (HNT). Amine groups, carboxylic acid groups, and the first generation of poly(amidoamine) (PAMAM) dendrimers were used for functionalization of the HNT. CNC and HNT are environmentally friendly, low/non-toxic, abundant, and inexpensive nanoparticles with a unique size, and chemical properties. TFN membranes were synthesized via in situ interfacial polymerization of m-phenylenediamine (MPD) with trimesoyl chloride (TMC) and the nanoparticles. The control Thin Film Composite (TFC) membranes, and CNC and HNT based TFN membranes were characterized by Scanning Electron Microscopy (SEM), Atomic Force Microscopy (AFM), X-Ray Diffraction (XRD), X-ray Photoelectron Spectroscopy (XPS), Fourier Transform Infrared spectroscopy (FTIR) and contact angle measurements. The antifouling capacity of CNC based membranes was investigated with a solution of Bovine Serum Albumin (BSA) as the fouling agent. Also, the leachability of the HNT from the membranes was examined by shaking the membranes in a batch incubator for 48 h, and then tracing the leached out HNT using Inductively Coupled Plasma Mass Spectrometry (ICP-MS). Separation characteristics of the membranes were studied by desalination of synthetic brackish water with a cross flow RO filtration system. It was revealed that incorporation of functionalized HNT enhanced the permeate flux without sacrificing the salt rejection (99.1 % ± 0.1 %). Also, incorporation of 0.1% (w/v) CNC doubled the permeate flux (from 30 to 63 L/m2.h at 20 bar) without compromising the salt rejection (97.8%). At the same time, leaching out of HNT from the TFN membranes was decreased as a result of the HNT functionalization and formation of covalent bonds with the TMC. Also, antifouling properties of the CNC-TFN membranes were 11% improved in comparison with control TFC membrane.

Os desreguladores endócrinos, como o bisfenol A (BPA), são compostos encontrados na água em concentrações da ordem de μg.L-1 ou ng.L-1, sendo por isso também denominados micropoluentes. Sua presença, mesmo em baixas concentrações, pode causar prejuízos aos organismos expostos. Neste contexto, o objetivo do presente trabalho foi avaliar a remoção de BPA por membranas de ultrafiltração (UF) e osmose inversa (OI) e por carvão ativado granular (CAG). Essas técnicas têm a vantagem de não gerar subprodutos que também podem ser tóxicos. Nos estudos foram realizados experimentos para remoção do BPA através de membrana PL-1 (celulose regenerada de 1 kDa), Sy-10 (polietersulfona de 10 kDa) e membrana de OI (poliamida) reutilizada, com concentração inicial de 500 μg.L-1 de BPA. Para a membrana Sy-10, foram testados os pHs 7 e 10. Os ensaios de adsorção foram realizados com CAG (1-2mm) em frascos contendo 100 mL da solução de BPA, com 0,5 g de CAG para cada frasco. As variáveis testadas foram pH, tempo de contato e concentração inicial na ordem de mg.L-1. Adicionalmente, foi testada a concentração inicial de 500 μg.L-1 de BPA em experimento de adsorção. Os resultados obtidos mostraram que a capacidade da membrana Sy-10 para remoção de BPA foi superior, chegando a cerca de 90% de remoção, em contraposição aos 20% encontrados para PL-1. A membrana de OI apresentou remoções de cerca de 95%. A influência do pH (7 e 10) na remoção do BPA para a membrana Sy-10 não mostrou-se significativa. Nos ensaios de adsorção, analisando diferentes valores de pH, obteve-se eficiência de 93% para pH 7. Ensaios de variação do tempo de contato com CAG mostraram que o equilíbrio é atingido nos primeiros 10 minutos para a maior concentração e em 40 minutos para a menor concentração testada, de 500 μg.L-1. O estudo do efeito da concentração inicial de BPA na adsorção mostrou que a remoção aumenta com o aumento da concentração inicial do poluente. Conclui-se que tanto a adsorção em CAG quanto membranas são boas alternativas para a remoção de BPA de soluções aquosas. As duas técnicas poderiam ser usadas conjuntamente, sendo que o concentrado do processo com membranas poderia ser submetido à adsorção por CAG. / Endocrine disrupting chemicals, such as bisphenol A (BPA) compounds, are found in water at concentrations of the order of μg.L-1 or ng.L-1 is therefore also known micropollutants. His presence, even at low concentrations, can cause damage to exposed organisms. In this context, the aim of this study was to evaluate the removal of BPA by ultrafiltration (UF) and reverse osmosis (RO) membranes and granular activated carbon (GAC). These techniques have the advantage of not generating by-products can also be toxic. In the studies were performed experiments to remove the BPA by PL-1 (1 kDa regenerated cellulose), Sy-10 (10 kDa polyethersulfone) and RO (polyamide) reused membranes, with initial concentration of 500 μg.L-1 of BPA. For Sy-10 membrane pHs 7 and 10 were tested. The adsorption experiments were carried out with GAC (1-2mm) in flasks containing 100 mL of the BPA solution and 0.5 g of GAC to each vial. The variables tested were pH, contact time and initial concentration on the order of mg.L-1. Additionally, we tested the initial concentration of 500 μg.L-1 of BPA in adsorption experiment. Results showed that the capacity of Sy-10 membrane to remove BPA was higher, reaching approximately 90% removal, as opposed to the 20% found to PL-1. The RO membrane showed removal of about 95%. The influence of pH (7 and 10 ) in the removal of BPA onto the membrane Sy - 10 was not significant. In adsorption tests, analyzing different pH values was obtained efficiency from 93% to pH 7. Testing time variation of contact GAC showed that equilibrium is reached in the first 10 minutes to the largest concentration and 40 minutes to the lowest concentration tested, 500 μg.L-1. The study of the effect of the initial concentration of BPA in the adsorption showed that removal increases with increase in the initial concentration of the pollutant. It was concluded that both the GAC adsorption as membranes are good alternatives for the removal of the BPA aqueous solutions. The two techniques could be used together, with the concentrate from the membrane process could be subjected to adsorption GAC.

O presente trabalho avaliou o desempenho de um sistema de pré-tratamento para água do mar constituído por um filtro lento de areia, com uma etapa prévia de filtração por um filtro de discos. O sistema demonstrou eficiência na remoção de fatores causadores de fouling em membranas de osmose reversa, removendo aproximadamente 97% dos sólidos suspensos totais presentes na água bruta, e gerando um filtrado com turbidez entre 0.1 e 0.2 UNT. As concentrações de microrganismos e AOC, principais fatores causadores da formação de biofouling, foram ambas reduzidas em cerca de 90%. Em sistemas de dessalinização de água do mar por osmose reversa, o pré-tratamento da água de alimentação constitui a principal estratégia no controle da formação do biofouling, que é gerado pelo estabelecimento de biofilmes sobre a superfície da membrana. A formação de biofouling gera impacto no desempenho do processo de osmose reversa e nos custos de operação. / This study evaluated the performance of a pretreatment system for seawater comprising of a slow sand filter, with a preliminary stage of filtration by disc filtration system. The system proved to be efficient in removing factors causing fouling in reverse osmosis membranes removing around 97% of the total suspended solids present in the raw water, and yielding a filtrate with turbidity between 0.1 and 0.2 NTU. The bacterial concentrations and AOC, main factors causing biofouling formation, were both reduced by about 90%. In seawater reverse osmosis systems, pretreatment of the feeded water is the main strategy to control biofouling, which is formed by the establishment of biofilms on a membrane surface. The biofouling formation produces an impact in the performance of the reverse osmosis process and operating costs.

The use of Reverse osmosis (RO) membranes have been significantly increasing in water desalination, and the main operational obstacle in RO desalination plants is membrane fouling. Among other solutes, dissolved biopolymers, such as polysaccharides can lead to severe membrane fouling especially with the addition of calcium ions because of the complexation formation between the surface of membrane and foulants materials. However, this complexation can also take place in the feed bulk, resulting in foulants aggregates formation. Although there are some physical techniques that can maintain the membrane performance without reducing its lifetime, only chemical cleanings are still commonly used in RO plants. In this study, a novel cleaning method is proposed to restore the membrane performance by removing the deposited foulants without reducing the membrane lifetime. The cleaning method is based on using water saturated with dissolved CO2 gas, and its principle is based on producing spontaneous CO2 bubbles due to local pressure difference leading to nucleation of bubbles throughout the membrane surface, especially at nucleation sites, which improve the cleaning efficiency. Alginic acid sodium salt was used as a model of polysaccharides foulants in presence of different concentrations of NaCl and calcium ions aiming to enhance membrane fouling, and then CO2 cleaning solution efficiency, in terms flux recovery (FR), was tested under different operating conditions and compared to other cleaning methods. Average FR of 20%±3, 25%±3 and 80%±3 for MilliQ water, a cleaning solution at pH4, and CO2 solution at 6 bar, 0.17 m/s, and 23 ̊C ±0.2 for 6 minutes were obtained, respectively. The efficiency of this novel cleaning method was also compared to direct osmosis overnight, and the average flux was comparable (about 60%±3), though that the cleaning time was significantly different. Various calcium concentrations (0-10 mM) were added in the alginate solution to study the fouling behavior in terms of the potential for bulk complexation to form cake alginate layer on the membrane surface rather than a gel layer, and the role of CO2 bubbles nucleation to remove foulants was investigated. This cleaning method can be considered as an alternative more environmentally friendly technique in RO application.

Reverse osmosis operations are facing persistent fouling phenomenon that has challenged the integrity of these processes. Prediction of fouling potential by measuring a fouling index toward feed water is essential to ensure robust operation. Moreover, employing a reliable fouling index with good reproducibility and precision is necessary. Silt density index (SDI) is considered insufficient in terms of reliability and empirical theory, among other limitations. Nevertheless due its simplicity, SDI measurement is utilized extensively in RO desalination systems. The aim of this research is to assess the reliability of SDI. Methods include the investigation of different SDI membranes and study of the nature of the SDI filtration. Results demonstrate the existence of the membrane properties' variation within manufacturers, which then causes a lack of accuracy in fouling risk estimation. The nature of particles during SDI filtration provides information that particle concentration and size play a significant role on SDI quantification with substantial representation given by particles with size close to membrane nominal pore size. Moreover, turbidity assisted SDI measurements along with determination of UF pretreated and clean water fouling potential, establishes the indication of non-fouling related phenomena involved on SDI measurement such as a natural organic matter adsorption and hydrodynamic condition that alters during filtration. Additionally, it was found that the latter affects the sensitivity of SDI by being represented by some portions of SDI value. Keywords: Reverse Osmosis, Fouling index, Particulate Fouling, Silt Density Index (SDI), and Assessment of SDI.

<p>As the number of water stressed regions around the world continues to growth due to a steadily increasing demand and climate change; the use of unconventional water sources, such as, brackish or seawater, through the implementation of desalination technologies has increased significantly. Reverse osmosis has established itself as the most widespread and energy efficient desalination technology, thanks to the development of high permeability membranes, high efficient pumps, and the integration of energy recovery devices; but, it still faces important challenges, such as, high specific energy consumption compared with traditional water treatment technologies, and poses environmental threats due to its significant CO₂emissions and the need of disposal of high salinity brine.</p> <p> </p> <p>The aim of this research is to address and provide solutions for two of the major challenge areas in reverse osmosis: reduction of the energy consumption and strategies to facilitate its integration with renewable energy sources to decrease its environmental impact. </p> <p> </p> <p>In chapter 2, the modeling and design of a double-acting batch reverse osmosis system is presented. A reduced specific energy consumption compared with previously proposed configurations was found. Moreover, the new design presents solutions to practical concerns that have limited the implementation of Batch reverse osmosis processes such as high start time and downtime, and permeate contamination. On the other hand, a novel hydraulic modeling is introduced to calculate the evolution of the pressure and other important parameters during the cycle.</p> <p> </p> <p>Chapter 3 presents a novel method which allows reverse osmosis plants to vary their power usage according to the energy availability, therefore, providing demand response capabilities. The effects on the energy consumption and performance of the reverse osmosis desalination facility due to the implementation of this technique are also studied. The split-salinity demand response reverse osmosis process proposed here poses as the first approach to grant demand response capabilities to reverse osmosis plants that provides energy gains and can be applied to existing plants. </p>

At the water treatment plant in Dunedin, Florida, reverse osmosis membranes remove the hardness from groundwater sources. Reverse osmosis membranes remove salts, pathogens, and organics from the feed water but can create an aggressive permeate. The membranes strip most ions in the process and the resulting permeate, if not subjected to blending on post treatment, has a tendency to leach metals from lead and copper pipes in the distribution networks. To prevent such problems, the permeate needs to be blended with partially treated raw water or to be chemically treated to re-mineralize and add alkalinity back into the water. In the last decade nanofiltration treatment has gained an increasing foothold in the water treatment industry especially as a water softener. Although nanofiltration membranes also have a high removal rate for organics and pathogens, the separation process is more selective towards multivalent ions (e.g., Ca²+, and Mg²+) than monovalent (e.g., Na+) ions. Most membrane softening plants blend minimally treated raw water with the membrane permeate as a means to reduce the aggressiveness of the water. However, blending can cause issues with disinfection byproducts and pathogen re-introduction. With nanofiltration membranes, fewer mono-valent ions are rejected which creates a more stable permeate and can reduce the blended water ratio. Since it is unlikely that most plants that use membrane filtration for water softening will be able to stop blending entirely, any improvement or sustainability of water quality at a reduced blend ratio should be viewed favorably within the water treatment industry. The study evaluates three nanofiltration membranes: TFC-SR, NF-90, and ESNA1-LF in relation to the reverse osmosis TFC-S RO membrane currently in use at Dunedin. Water flux and salt rejection of the permeate water were compared using solutions of NaCl, MgSO4 and CaCl2. Since the Langelier Saturation Index (LSI) is one of the main tests of the blended finished water and is used to judge water quality prior to its release into the distribution system, this study created a 0%, 10%, 15%, 20%, 30%, and 100% blend ratio for each membrane to compare and contrast the change in the LSI. The TFC-SR membrane showed the most promise in lowering the blend ratio while improving the aggressiveness of the finished water by showing a lower rejection for divalent ions. The TFC-SR membrane also showed an improvement in the LSI relative to the other membranes over the total range of blend ratios.