According to the Department of Water Affairs (DWAF, 2004 p.15), South Africa’s water resources are scarce and extremely limited and much of this precious resource is utilised and consumed in our industries. Treatment and re-use of effluent generated is, in some cases, preferred over use of alternate water resources (Du Plessis, 2008 p.3). The volume of effluent generated in treatment processes like ultra-filtration (UF) and reverse osmosis (RO) units is determined by the feed water quality, with high water loss through effluent generation at poor feed water quality. Current UF and RO applications require an increased UF production capacity due to the use of UF filtrate for periodic backwashing of the UF membrane units. This results in loss of water and decreases overall recovery. The need therefore exists to increase the overall recovery of product water from the raw water stream by reducing the amount of effluent generated. This would be possible to achieve by using RO brine to backwash the UF unit. The study was conducted to provide a modelling tool, assisting management to optimise the use of RO brine as backwash water on the UF system at the Medupi power station. The secondary objective of this study was the development of a modelling tool that can be used for other projects, new or existing, as a measure and indication of the usability of RO brine as backwash water on UF systems. By successfully applying this newly developed model, the viability of utilising the RO brine as backwash water for the UF was investigated. This modification would lead to utilizing smaller UF units than previously envisioned, which in turn leads to reducing capital cost with 11.07% and operating expenditure with 9.98% at the Medupi power station. This also has a positive environmental impact by reducing the amount of raw water used monthly by 10.34% (108 000 m3/month). / MIng (Development and Management Engineering), North-West University, Potchefstroom Campus, 2014

Back grinding (BG) wastewater consists mainly of high-purity water and high concentrations of inorganic particles. If the BG wastewater could be treated and recycled efficiently, it should be sort of economic benefit. In this study, appropriate pre-treatment technologies are evaluated to obtain the feasible recycle system. From the chemical coagulation experiment, the addition of PAC or FeCl3, both of them can obviously reduce the turbidity and suspended solid concentrations (SS). In addition, polymer can advance the sedimentation process. Considering the cost of practical operation, the turbidity of BG waste water could be removed up to 97% by using polyaluminum chloride as the coagulant (2.2 mg/L) and polymer as the coagulant aid (0.5 mg/L) in the pH=7 condition . In sand filtration experiment, the turbidity and SS can¡¦t be effectively removed if the coagulation isn¡¦t used on BG wastewater. It demonstrates that BG wastewater contains high concentration of nano-scale particles. The rate of removable turbidity can reach 99% under applying coagulation, sedimentation, and sand filtration. In ultra-filtration experiment, both of spiral-wound (SW) and hollow-fiber (HF) can remove more than 99.9% of turbidity. For the flux of behavior, the performance of pre-treatment water is better than non-treatment water. Thus, it reveals that appropriate pre-treatment can lower the load of membrane filtration system. For the obtained recycle water, the grade of standard can achieve the grade of the cooling tower required. However, due to its high particle-containing characteristics, the commonly used reverse-osmoses (RO) membrane filtration technology can not be directly applied for purification process because the fouling/clogging problem would cause the frequent membrane replacement. In this lab-scale feasibility study, pre-treatment technologies (e.g., sand filtration, chemical coagulation, ultra-filtration) were applied to reduce the turbidity and particle concentrations of the BG wastewater (collected from a semiconductor manufacturing plant) before RO filtration unit. The BG wastewater contained turbidity and suspended solid concentrations of 3,200 NTU and 96 mg/L, respectively. The measured pH and conductivity of the BG wastewater were in the ranges of 6.8 to 7.2 and 14 to 18 £gS/cm, respectively. Moreover, the particle sizes of the solids varied from 300 to 700 nm. Thus, applying conventional sand filtration along could not effectively remove the nano-scale particles. Results from the chemical coagulation experiment reveal that the turbidity and particles of the BG wastewater could be significantly removed (up to 95% of turbidity and particle removal) by the coagulation/sedimentation process using polyaluminum chloride as the coagulant (2.2 mg/L) and polymer as the coagulant aid (0.5 mg/L). Results also indicate that up to 99% of turbidity and particle removal could be obtained with the application of ultra-filtration unit after the coagulation/sedimentation process. Results from this study indicate that applying appropriate pre-treatment technologies (coagulation and ultra-filtration) would lower the fouling rate and extend the life of RO membrane used for BG wastewater purification.

According to the Department of Water Affairs (DWAF, 2004 p.15), South Africa’s water resources are scarce and extremely limited and much of this precious resource is utilised and consumed in our industries. Treatment and re-use of effluent generated is, in some cases, preferred over use of alternate water resources (Du Plessis, 2008 p.3). The volume of effluent generated in treatment processes like ultra-filtration (UF) and reverse osmosis (RO) units is determined by the feed water quality, with high water loss through effluent generation at poor feed water quality. Current UF and RO applications require an increased UF production capacity due to the use of UF filtrate for periodic backwashing of the UF membrane units. This results in loss of water and decreases overall recovery. The need therefore exists to increase the overall recovery of product water from the raw water stream by reducing the amount of effluent generated. This would be possible to achieve by using RO brine to backwash the UF unit. The study was conducted to provide a modelling tool, assisting management to optimise the use of RO brine as backwash water on the UF system at the Medupi power station. The secondary objective of this study was the development of a modelling tool that can be used for other projects, new or existing, as a measure and indication of the usability of RO brine as backwash water on UF systems. By successfully applying this newly developed model, the viability of utilising the RO brine as backwash water for the UF was investigated. This modification would lead to utilizing smaller UF units than previously envisioned, which in turn leads to reducing capital cost with 11.07% and operating expenditure with 9.98% at the Medupi power station. This also has a positive environmental impact by reducing the amount of raw water used monthly by 10.34% (108 000 m3/month). / MIng (Development and Management Engineering), North-West University, Potchefstroom Campus, 2014

O consumo da uva e seus derivados está associado com a proteção contra doenças degenerativas devido a seu conteúdo de compostos fenólicos. A casca de uva é um subproduto da produção de suco de uva e vinho, que possui um alto conteúdo de compostos fenólicos, os quais se extraídos e protegidos por encapsulamento, agregariam valor a esse produto. O objetivo deste trabalho foi separar os compostos fenólicos do extrato aquoso acidificado com ácido cítrico 2% (p/v) da casca de uva a través de ultrafiltração (UF) utilizando membranas de poliestersulfona com massas molares de corte 10 e 30 kDa e posterior microencapsulação dos permeados recolhidos do processo de separação, usando como agentes encapsulantes goma arábica 15% e polidextrose 15%. Inicialmente, estudou-se o desempenho das membranas a diferentes pressões transmembrana de 1,0 a 3,5 bar e temperaturas de 25, 35 e 45°C, com o intuito de definir as melhores condições de separação. Para isso, foram usadas uvas variedade bordô (vitis labrusca) do municipio de Cotiporã, Serra Gaúcha RS. As uvas foram previamente branqueadas em banho de água a 80°C por 5 minutos e posteriormente resfriadas em banho de gelo por 3 minutos. A seguir, as cascas foram separadas manualmente da polpa e colocadas numa solução acidificada com ácido cítrico 2% (p/v) na proporção 1:4 (m/v). Após 20 horas, o extrato acidificado, com pH menor que 2,8, foi filtrado com papel Whatman n° 1 e posteriormente colocado no equipamento de membranas para realizar os diferentes experimentos de UF. Os resultados indicaram que a melhor condição de separação na UF foi à temperatura de 25°C e 3,5 bar de pressão transmembrana. Os resultados mostraram que as permeabilidades hidráulicas para as membranas novas de 10 e 30 kDa foram 10,96 e 20,52 L.m-2.h-1.bar-1, respectivamente. Durante o processo de UF do extrato, os valores de fouling a 25°C foram de 85,9 e 89,6% para as membranas de 10 e 30 kDa, enquanto que a recuperação, após limpeza química, foi de 63,1 e 80,9%, respectivamente. Do estudo das resistências total ao fluxo, constituída pelas resistências da membrana, do fouling e da polarização por concentração, a maior foi a da polarização por concentração, com porcentagens com respeito à resistência total, de 80 a 90% e de 60 a 90% para as membrana de 10 kDa e de 30 kDa, enquanto que a do fouling foi a menor de 1 a 3% e de 2 e 15% nas membranas de 10 e de 30 kDa. As concentrações de polifenois totais na membrana de 10 kDa a 25°C foram de 7,84, 2,51 e 10,89 mg ácido gálico (GA)/g amostra seca, no retido, permeado e extrato, respectivamente. Com a membrana de 30 kDa os teores foram de 7,12, 1,62 e 10,66 mg ácido gálico (GA)/g amostra seca, no retido, permeado e extrato, respectivamente. Quanto aos flavonoides os teores obtidos com a membrana de 10 kDa foram 1,78, 0,63 e 2,51, enquanto que na membrana de 30 kDa foram de 1,60, 0,40 e 2,26 mg Catequinas/g amostra seca, no retido, permeado e extrato, respectivamente. O valor de capacidade antioxidante medido como ABTS para o extrato a 25°C foi de 127,80 equivalentes Trolox (TEAC) (μmol/g amostra seca), e no permeado de 9,10 e 16,41 μmol de trolox equivalente/g amostra seca para as membranas de 10kDa e 30kDa, respectivamente, evidenciando que foram separados poucos compostos com essa propriedade. O índice de correlação entre os teores dos compostos 9 fenólicos e de flavonoides com a capacidade antioxidante foram de 0,99 e de 0,98, respectivamente. A membrana que apresenta melhores condições para a ultrafiltração de casca de uva bordô é a membrana de 10 kDa. Para o estudo de encapsulamento foram usados os permeados de UF obtidos das membranas 10 e 30 kDa a 25°C, utilizando goma arábica 15% e polidextrose 15% como materiais encapsulantes, e posterior secagem por atomização a 140ºC. As micropartículas obtidas resultaram em teores de umidade e de atividade de água menores que 3,90% e 0,18%, respectivamente. Em relação à solubilidade, todas as amostras encapsuladas foram muito solúveis, com valores na faixa de 96,8 a 99,6%. A higroscopicidade dos pós apresentou diferença significativa entre os agentes encapsulantes sendo que a polidextrose foi a mais higroscópica. Para a cor, os parâmetos a* e b* indicaram que as amostras possuem cores entre vermelho e o azul, e de acordo com o Chroma os pós obtidos com polidextrose foram mais saturados do que os pós encapsulados com goma Arábica 15%. Quanto ao parâmetro Hue, os resultados também indicaram que as amostras se encontram no quarto quadrante do circulo cromático de cores (entre vermelho e o azul). / Consumption of grape and it’s components is associated with protection against degenerative diseases due to a high content of phenolic compounds. The grape skin is a rich source of phenolic compounds as are byproducts of grape such as grape juice and compounds found in wine made from grapes resulting in a less recognized added value to these products. Current research separates the phenolic compounds of an acidic aqueous extract with citric acid 2% (w / v) of grape skin and using ultra-filtration with poly(ether sulfone) membranes. This research material had a molecular weight cut-off of 10 and 30 kDa and subsequent micro-encapsulation of collected permeate from the separation process and used Arabic gum and polydextrose 15% as encapsulating agents. Initially the performance of the membranes was compared to different trans-membrane pressure 1,0 to 3,5 bar and temperatures of 25, 35 and 45°C, in order to define the best separation conditions. Bordo grapes (vitis labrusca) that were grown in the city of Cotiporã, region of Rio grande do sul, Brazil, were used for the experiments. Grapes were previously subjected to bleaching with water bath at 80°C for 5 minutes and followed by cooling in an ice bath for 3 minutes. Then the grape skin was manually separated from the pulp and acidified water solution with citric acid 2% (w / v) was added in a ratio of 1:4 (water / pulp). Additionally it was homogenized and the mixture maintained at room temperature for 20 hours. The acidified extract with a pH lower than 2.8 was filtered with Whatman No. 1 paper and placed on the membrane equipment to begin the different UF experiments. The results showed that better separation conditions in the UF was to 25°C and a trans-membrane pressure of 3.5 bar. Results showed that the hydraulic permeabilities for the new membranes of 10 and 30 kDa were 10,96 and 20,52 L.m-1.bar-2.h-1, respectively. During the UF process of the extract, the fouling values at 25°C were 85,9 and 89.6% for the membranes 10 and 30 kDa, while recovery after chemical cleaning was 63,1 and 80,9%, respectively. It was studied total resistance (Rt) to flow constituted by the intrinsic membrane resistance (Rm), fouling resistance (Rf) and cake layer resistance (Rc). The maximum resistance was a cake layer resistance (Rc) with percentages compared with the total resistance of 80 to 90% and from 60 to 90% for membrane 10 kDa and 30 kDa. The fouling was a lowest 1 to 3% and 2 to 15% in membranes 10 and 30 kDa. The total polyphenol concentrations in the 10 kDa membrane at 25°C were 7,84, 2,51 and 10,89 mg (GA)gallic acid/g dry sample, in the retentate, permeate and extract, respectively. With the membrane of 30 kDa, contents were 7,12, 1,62 and 10,66 mg gallic acid (GA)/g dry sample, in the retentate, permeate and extract respectively. Flavonoid contents obtained with the 10 kDa membrane were 1,78, 0,63 and 2,51, and the membrane 30 kDa were 1,60, 0,40 and 2,26 Catechins mg / g sample dry, in the retentate, permeate and extract respectively. The amount of antioxidant capacity measured as ABTS to the extract at 25°C was 127,80 Trolox equivalent (TEAC) (μmol/g dry sample) and permeate of 9,10 and 16,41 Trolox equivalent (TEAC) (μmol/g dry sample) for 11 membranes of 10 and 30 kDa respectively, showing that a few compounds were separated with this property. The correlation index between levels of flavonoids and phenolic compounds with antioxidant capacity were 0,99 and 0,98, respectively. The membrane that presents the best conditions for the ultrafiltration of the grape skin extract is 10 kDa membrane. The encapsulation study used the UF permeates obtained from the membrane process of 10 and 30 kDa to 25°C, using Arabic gum and polydextrose 15% as encapsulating agent and subsequent spray drying process at 140°C. The micro-particles obtained resulted in moisture content and lower water activity to 3.90% and 0.18, respectively. With regard to solubility, all of the encapsulated samples were very soluble, with values ranging from 96.8 to 99.6%. Hygroscopicity of powders showed significant difference between agents encapsulants and the polydextrose was the most hygroscopic. For color, parameters a * and b * indicate that the samples have color between red and blue, in accordance with the Chroma. Powders obtained with polydextrose have been more saturated than powders encapsulated with Arabic gum 15%. As for Hue parameter, the results also indicated that the samples are in the fourth quadrant of the circle chromatic color (from red to blue).

O consumo da uva e seus derivados está associado com a proteção contra doenças degenerativas devido a seu conteúdo de compostos fenólicos. A casca de uva é um subproduto da produção de suco de uva e vinho, que possui um alto conteúdo de compostos fenólicos, os quais se extraídos e protegidos por encapsulamento, agregariam valor a esse produto. O objetivo deste trabalho foi separar os compostos fenólicos do extrato aquoso acidificado com ácido cítrico 2% (p/v) da casca de uva a través de ultrafiltração (UF) utilizando membranas de poliestersulfona com massas molares de corte 10 e 30 kDa e posterior microencapsulação dos permeados recolhidos do processo de separação, usando como agentes encapsulantes goma arábica 15% e polidextrose 15%. Inicialmente, estudou-se o desempenho das membranas a diferentes pressões transmembrana de 1,0 a 3,5 bar e temperaturas de 25, 35 e 45°C, com o intuito de definir as melhores condições de separação. Para isso, foram usadas uvas variedade bordô (vitis labrusca) do municipio de Cotiporã, Serra Gaúcha RS. As uvas foram previamente branqueadas em banho de água a 80°C por 5 minutos e posteriormente resfriadas em banho de gelo por 3 minutos. A seguir, as cascas foram separadas manualmente da polpa e colocadas numa solução acidificada com ácido cítrico 2% (p/v) na proporção 1:4 (m/v). Após 20 horas, o extrato acidificado, com pH menor que 2,8, foi filtrado com papel Whatman n° 1 e posteriormente colocado no equipamento de membranas para realizar os diferentes experimentos de UF. Os resultados indicaram que a melhor condição de separação na UF foi à temperatura de 25°C e 3,5 bar de pressão transmembrana. Os resultados mostraram que as permeabilidades hidráulicas para as membranas novas de 10 e 30 kDa foram 10,96 e 20,52 L.m-2.h-1.bar-1, respectivamente. Durante o processo de UF do extrato, os valores de fouling a 25°C foram de 85,9 e 89,6% para as membranas de 10 e 30 kDa, enquanto que a recuperação, após limpeza química, foi de 63,1 e 80,9%, respectivamente. Do estudo das resistências total ao fluxo, constituída pelas resistências da membrana, do fouling e da polarização por concentração, a maior foi a da polarização por concentração, com porcentagens com respeito à resistência total, de 80 a 90% e de 60 a 90% para as membrana de 10 kDa e de 30 kDa, enquanto que a do fouling foi a menor de 1 a 3% e de 2 e 15% nas membranas de 10 e de 30 kDa. As concentrações de polifenois totais na membrana de 10 kDa a 25°C foram de 7,84, 2,51 e 10,89 mg ácido gálico (GA)/g amostra seca, no retido, permeado e extrato, respectivamente. Com a membrana de 30 kDa os teores foram de 7,12, 1,62 e 10,66 mg ácido gálico (GA)/g amostra seca, no retido, permeado e extrato, respectivamente. Quanto aos flavonoides os teores obtidos com a membrana de 10 kDa foram 1,78, 0,63 e 2,51, enquanto que na membrana de 30 kDa foram de 1,60, 0,40 e 2,26 mg Catequinas/g amostra seca, no retido, permeado e extrato, respectivamente. O valor de capacidade antioxidante medido como ABTS para o extrato a 25°C foi de 127,80 equivalentes Trolox (TEAC) (μmol/g amostra seca), e no permeado de 9,10 e 16,41 μmol de trolox equivalente/g amostra seca para as membranas de 10kDa e 30kDa, respectivamente, evidenciando que foram separados poucos compostos com essa propriedade. O índice de correlação entre os teores dos compostos 9 fenólicos e de flavonoides com a capacidade antioxidante foram de 0,99 e de 0,98, respectivamente. A membrana que apresenta melhores condições para a ultrafiltração de casca de uva bordô é a membrana de 10 kDa. Para o estudo de encapsulamento foram usados os permeados de UF obtidos das membranas 10 e 30 kDa a 25°C, utilizando goma arábica 15% e polidextrose 15% como materiais encapsulantes, e posterior secagem por atomização a 140ºC. As micropartículas obtidas resultaram em teores de umidade e de atividade de água menores que 3,90% e 0,18%, respectivamente. Em relação à solubilidade, todas as amostras encapsuladas foram muito solúveis, com valores na faixa de 96,8 a 99,6%. A higroscopicidade dos pós apresentou diferença significativa entre os agentes encapsulantes sendo que a polidextrose foi a mais higroscópica. Para a cor, os parâmetos a* e b* indicaram que as amostras possuem cores entre vermelho e o azul, e de acordo com o Chroma os pós obtidos com polidextrose foram mais saturados do que os pós encapsulados com goma Arábica 15%. Quanto ao parâmetro Hue, os resultados também indicaram que as amostras se encontram no quarto quadrante do circulo cromático de cores (entre vermelho e o azul). / Consumption of grape and it’s components is associated with protection against degenerative diseases due to a high content of phenolic compounds. The grape skin is a rich source of phenolic compounds as are byproducts of grape such as grape juice and compounds found in wine made from grapes resulting in a less recognized added value to these products. Current research separates the phenolic compounds of an acidic aqueous extract with citric acid 2% (w / v) of grape skin and using ultra-filtration with poly(ether sulfone) membranes. This research material had a molecular weight cut-off of 10 and 30 kDa and subsequent micro-encapsulation of collected permeate from the separation process and used Arabic gum and polydextrose 15% as encapsulating agents. Initially the performance of the membranes was compared to different trans-membrane pressure 1,0 to 3,5 bar and temperatures of 25, 35 and 45°C, in order to define the best separation conditions. Bordo grapes (vitis labrusca) that were grown in the city of Cotiporã, region of Rio grande do sul, Brazil, were used for the experiments. Grapes were previously subjected to bleaching with water bath at 80°C for 5 minutes and followed by cooling in an ice bath for 3 minutes. Then the grape skin was manually separated from the pulp and acidified water solution with citric acid 2% (w / v) was added in a ratio of 1:4 (water / pulp). Additionally it was homogenized and the mixture maintained at room temperature for 20 hours. The acidified extract with a pH lower than 2.8 was filtered with Whatman No. 1 paper and placed on the membrane equipment to begin the different UF experiments. The results showed that better separation conditions in the UF was to 25°C and a trans-membrane pressure of 3.5 bar. Results showed that the hydraulic permeabilities for the new membranes of 10 and 30 kDa were 10,96 and 20,52 L.m-1.bar-2.h-1, respectively. During the UF process of the extract, the fouling values at 25°C were 85,9 and 89.6% for the membranes 10 and 30 kDa, while recovery after chemical cleaning was 63,1 and 80,9%, respectively. It was studied total resistance (Rt) to flow constituted by the intrinsic membrane resistance (Rm), fouling resistance (Rf) and cake layer resistance (Rc). The maximum resistance was a cake layer resistance (Rc) with percentages compared with the total resistance of 80 to 90% and from 60 to 90% for membrane 10 kDa and 30 kDa. The fouling was a lowest 1 to 3% and 2 to 15% in membranes 10 and 30 kDa. The total polyphenol concentrations in the 10 kDa membrane at 25°C were 7,84, 2,51 and 10,89 mg (GA)gallic acid/g dry sample, in the retentate, permeate and extract, respectively. With the membrane of 30 kDa, contents were 7,12, 1,62 and 10,66 mg gallic acid (GA)/g dry sample, in the retentate, permeate and extract respectively. Flavonoid contents obtained with the 10 kDa membrane were 1,78, 0,63 and 2,51, and the membrane 30 kDa were 1,60, 0,40 and 2,26 Catechins mg / g sample dry, in the retentate, permeate and extract respectively. The amount of antioxidant capacity measured as ABTS to the extract at 25°C was 127,80 Trolox equivalent (TEAC) (μmol/g dry sample) and permeate of 9,10 and 16,41 Trolox equivalent (TEAC) (μmol/g dry sample) for 11 membranes of 10 and 30 kDa respectively, showing that a few compounds were separated with this property. The correlation index between levels of flavonoids and phenolic compounds with antioxidant capacity were 0,99 and 0,98, respectively. The membrane that presents the best conditions for the ultrafiltration of the grape skin extract is 10 kDa membrane. The encapsulation study used the UF permeates obtained from the membrane process of 10 and 30 kDa to 25°C, using Arabic gum and polydextrose 15% as encapsulating agent and subsequent spray drying process at 140°C. The micro-particles obtained resulted in moisture content and lower water activity to 3.90% and 0.18, respectively. With regard to solubility, all of the encapsulated samples were very soluble, with values ranging from 96.8 to 99.6%. Hygroscopicity of powders showed significant difference between agents encapsulants and the polydextrose was the most hygroscopic. For color, parameters a * and b * indicate that the samples have color between red and blue, in accordance with the Chroma. Powders obtained with polydextrose have been more saturated than powders encapsulated with Arabic gum 15%. As for Hue parameter, the results also indicated that the samples are in the fourth quadrant of the circle chromatic color (from red to blue).

O consumo da uva e seus derivados está associado com a proteção contra doenças degenerativas devido a seu conteúdo de compostos fenólicos. A casca de uva é um subproduto da produção de suco de uva e vinho, que possui um alto conteúdo de compostos fenólicos, os quais se extraídos e protegidos por encapsulamento, agregariam valor a esse produto. O objetivo deste trabalho foi separar os compostos fenólicos do extrato aquoso acidificado com ácido cítrico 2% (p/v) da casca de uva a través de ultrafiltração (UF) utilizando membranas de poliestersulfona com massas molares de corte 10 e 30 kDa e posterior microencapsulação dos permeados recolhidos do processo de separação, usando como agentes encapsulantes goma arábica 15% e polidextrose 15%. Inicialmente, estudou-se o desempenho das membranas a diferentes pressões transmembrana de 1,0 a 3,5 bar e temperaturas de 25, 35 e 45°C, com o intuito de definir as melhores condições de separação. Para isso, foram usadas uvas variedade bordô (vitis labrusca) do municipio de Cotiporã, Serra Gaúcha RS. As uvas foram previamente branqueadas em banho de água a 80°C por 5 minutos e posteriormente resfriadas em banho de gelo por 3 minutos. A seguir, as cascas foram separadas manualmente da polpa e colocadas numa solução acidificada com ácido cítrico 2% (p/v) na proporção 1:4 (m/v). Após 20 horas, o extrato acidificado, com pH menor que 2,8, foi filtrado com papel Whatman n° 1 e posteriormente colocado no equipamento de membranas para realizar os diferentes experimentos de UF. Os resultados indicaram que a melhor condição de separação na UF foi à temperatura de 25°C e 3,5 bar de pressão transmembrana. Os resultados mostraram que as permeabilidades hidráulicas para as membranas novas de 10 e 30 kDa foram 10,96 e 20,52 L.m-2.h-1.bar-1, respectivamente. Durante o processo de UF do extrato, os valores de fouling a 25°C foram de 85,9 e 89,6% para as membranas de 10 e 30 kDa, enquanto que a recuperação, após limpeza química, foi de 63,1 e 80,9%, respectivamente. Do estudo das resistências total ao fluxo, constituída pelas resistências da membrana, do fouling e da polarização por concentração, a maior foi a da polarização por concentração, com porcentagens com respeito à resistência total, de 80 a 90% e de 60 a 90% para as membrana de 10 kDa e de 30 kDa, enquanto que a do fouling foi a menor de 1 a 3% e de 2 e 15% nas membranas de 10 e de 30 kDa. As concentrações de polifenois totais na membrana de 10 kDa a 25°C foram de 7,84, 2,51 e 10,89 mg ácido gálico (GA)/g amostra seca, no retido, permeado e extrato, respectivamente. Com a membrana de 30 kDa os teores foram de 7,12, 1,62 e 10,66 mg ácido gálico (GA)/g amostra seca, no retido, permeado e extrato, respectivamente. Quanto aos flavonoides os teores obtidos com a membrana de 10 kDa foram 1,78, 0,63 e 2,51, enquanto que na membrana de 30 kDa foram de 1,60, 0,40 e 2,26 mg Catequinas/g amostra seca, no retido, permeado e extrato, respectivamente. O valor de capacidade antioxidante medido como ABTS para o extrato a 25°C foi de 127,80 equivalentes Trolox (TEAC) (μmol/g amostra seca), e no permeado de 9,10 e 16,41 μmol de trolox equivalente/g amostra seca para as membranas de 10kDa e 30kDa, respectivamente, evidenciando que foram separados poucos compostos com essa propriedade. O índice de correlação entre os teores dos compostos 9 fenólicos e de flavonoides com a capacidade antioxidante foram de 0,99 e de 0,98, respectivamente. A membrana que apresenta melhores condições para a ultrafiltração de casca de uva bordô é a membrana de 10 kDa. Para o estudo de encapsulamento foram usados os permeados de UF obtidos das membranas 10 e 30 kDa a 25°C, utilizando goma arábica 15% e polidextrose 15% como materiais encapsulantes, e posterior secagem por atomização a 140ºC. As micropartículas obtidas resultaram em teores de umidade e de atividade de água menores que 3,90% e 0,18%, respectivamente. Em relação à solubilidade, todas as amostras encapsuladas foram muito solúveis, com valores na faixa de 96,8 a 99,6%. A higroscopicidade dos pós apresentou diferença significativa entre os agentes encapsulantes sendo que a polidextrose foi a mais higroscópica. Para a cor, os parâmetos a* e b* indicaram que as amostras possuem cores entre vermelho e o azul, e de acordo com o Chroma os pós obtidos com polidextrose foram mais saturados do que os pós encapsulados com goma Arábica 15%. Quanto ao parâmetro Hue, os resultados também indicaram que as amostras se encontram no quarto quadrante do circulo cromático de cores (entre vermelho e o azul). / Consumption of grape and it’s components is associated with protection against degenerative diseases due to a high content of phenolic compounds. The grape skin is a rich source of phenolic compounds as are byproducts of grape such as grape juice and compounds found in wine made from grapes resulting in a less recognized added value to these products. Current research separates the phenolic compounds of an acidic aqueous extract with citric acid 2% (w / v) of grape skin and using ultra-filtration with poly(ether sulfone) membranes. This research material had a molecular weight cut-off of 10 and 30 kDa and subsequent micro-encapsulation of collected permeate from the separation process and used Arabic gum and polydextrose 15% as encapsulating agents. Initially the performance of the membranes was compared to different trans-membrane pressure 1,0 to 3,5 bar and temperatures of 25, 35 and 45°C, in order to define the best separation conditions. Bordo grapes (vitis labrusca) that were grown in the city of Cotiporã, region of Rio grande do sul, Brazil, were used for the experiments. Grapes were previously subjected to bleaching with water bath at 80°C for 5 minutes and followed by cooling in an ice bath for 3 minutes. Then the grape skin was manually separated from the pulp and acidified water solution with citric acid 2% (w / v) was added in a ratio of 1:4 (water / pulp). Additionally it was homogenized and the mixture maintained at room temperature for 20 hours. The acidified extract with a pH lower than 2.8 was filtered with Whatman No. 1 paper and placed on the membrane equipment to begin the different UF experiments. The results showed that better separation conditions in the UF was to 25°C and a trans-membrane pressure of 3.5 bar. Results showed that the hydraulic permeabilities for the new membranes of 10 and 30 kDa were 10,96 and 20,52 L.m-1.bar-2.h-1, respectively. During the UF process of the extract, the fouling values at 25°C were 85,9 and 89.6% for the membranes 10 and 30 kDa, while recovery after chemical cleaning was 63,1 and 80,9%, respectively. It was studied total resistance (Rt) to flow constituted by the intrinsic membrane resistance (Rm), fouling resistance (Rf) and cake layer resistance (Rc). The maximum resistance was a cake layer resistance (Rc) with percentages compared with the total resistance of 80 to 90% and from 60 to 90% for membrane 10 kDa and 30 kDa. The fouling was a lowest 1 to 3% and 2 to 15% in membranes 10 and 30 kDa. The total polyphenol concentrations in the 10 kDa membrane at 25°C were 7,84, 2,51 and 10,89 mg (GA)gallic acid/g dry sample, in the retentate, permeate and extract, respectively. With the membrane of 30 kDa, contents were 7,12, 1,62 and 10,66 mg gallic acid (GA)/g dry sample, in the retentate, permeate and extract respectively. Flavonoid contents obtained with the 10 kDa membrane were 1,78, 0,63 and 2,51, and the membrane 30 kDa were 1,60, 0,40 and 2,26 Catechins mg / g sample dry, in the retentate, permeate and extract respectively. The amount of antioxidant capacity measured as ABTS to the extract at 25°C was 127,80 Trolox equivalent (TEAC) (μmol/g dry sample) and permeate of 9,10 and 16,41 Trolox equivalent (TEAC) (μmol/g dry sample) for 11 membranes of 10 and 30 kDa respectively, showing that a few compounds were separated with this property. The correlation index between levels of flavonoids and phenolic compounds with antioxidant capacity were 0,99 and 0,98, respectively. The membrane that presents the best conditions for the ultrafiltration of the grape skin extract is 10 kDa membrane. The encapsulation study used the UF permeates obtained from the membrane process of 10 and 30 kDa to 25°C, using Arabic gum and polydextrose 15% as encapsulating agent and subsequent spray drying process at 140°C. The micro-particles obtained resulted in moisture content and lower water activity to 3.90% and 0.18, respectively. With regard to solubility, all of the encapsulated samples were very soluble, with values ranging from 96.8 to 99.6%. Hygroscopicity of powders showed significant difference between agents encapsulants and the polydextrose was the most hygroscopic. For color, parameters a * and b * indicate that the samples have color between red and blue, in accordance with the Chroma. Powders obtained with polydextrose have been more saturated than powders encapsulated with Arabic gum 15%. As for Hue parameter, the results also indicated that the samples are in the fourth quadrant of the circle chromatic color (from red to blue).

The combination of moving bed biofilm reactors and membrane bioreactors (MBBR-MR) can compensate for the drawbacks of both of these systems and further increase their acceptance and application in wastewater treatment industries. Despite the potential benefits of a MBBR-MR technology there has only been limited study of this configuration. The present study consisted of an overall assessment of the performance of a combined MBBR-MR system under high and low loading rates. Since colloidal matter in mixed liquor suspended solid (MLSS) is considered as one of the important contributors to membrane fouling, pre-treatment of membrane feed by coagulation was investigated for improving membrane performance. The performance of the MBBR-MR was assessed based on its chemical oxygen demand (COD) removal efficiency and membrane fouling mechanisms. The study was carried out using pilot-scale MBBR and bench-scale batch membrane filtration setups (Millipore Inc. Bedford, MA). The pilot MBBR had a working volume of 1.8 m3 and a 30% carrier fill fraction. The MBBR was operated with loading rates of 160 ± 44 g/m2/d (hydraulic residence time (HRT) of 4.6 h) and 223 g/m2/d (HRT of 2.6 h). The MBBR feed was obtained from a starch recovery line in a potato chip processing factory. The carriers were mixed by coarse bubble aeration and the dissolved oxygen (DO) was maintained above 2 mg/l. Preliminary jar test trials (based on turbidity removal) were performed in order to obtain an optimal dosage of coagulants for subsequent ultrafiltration (UF) tests. The efficiency of three coagulants (alum, ferric chloride and a blend of polyaluminum chloride and polyamine) was evaluated. The membranes were composed of polyethersulfone (PES) and had a pore size of 0.05 microns. The results of this study indicate that the combination of MBBR with membrane filtration can be operated at relatively high loading rates to yield a constant high quality permeate that is suitable for water reuse purposes. Fouling of the membrane by the wastewater was found to be substantially reduced by treatment with the MBBR. The reversible and irreversible fouling of the MBBR effluent were 56 and 63%, respectively, of that observed with the raw wastewater. The MBBR Loading-rate was found to affect treatment efficiency of the MBBR-MR and membrane performance. Operation under the elevated loading-rate conditions HRT = 2.6 hours) resulted in an increase in the irreversible fouling of the membranes (60% on average). The addition of all the coagulants in this study was found to decrease the fouling of the membrane. However, the extent of the pre-coagulation effect on membrane fouling was found to strongly depend on the type and dosage of the coagulant and the MBBR effluent characteristics. All the coagulants were effective in decreasing membrane fouling at their optimal dosages which was determined in preliminary jar tests. Ferric chloride performed the best as a pretreatment coagulant compared to alum (Aluminum sulfate) and the coagulant blend with reductions in both reversible and irreversible fouling (43-86% and 51-71%, respectively) and increased consistency (in decreasing fouling) as compared to the other coagulants. Alum had no effect on irreversible fouling and the coagulant blend significantly increased irreversible fouling in some trials (up to 196% or by a factor of 3 when overdosed). Additionally, alum and the blend were, on average, 29% and 7%, less effective than ferric chloride in reducing reversible fouling under the conditions and dosages tested.

The combination of moving bed biofilm reactors and membrane bioreactors (MBBR-MR) can compensate for the drawbacks of both of these systems and further increase their acceptance and application in wastewater treatment industries. Despite the potential benefits of a MBBR-MR technology there has only been limited study of this configuration. The present study consisted of an overall assessment of the performance of a combined MBBR-MR system under high and low loading rates. Since colloidal matter in mixed liquor suspended solid (MLSS) is considered as one of the important contributors to membrane fouling, pre-treatment of membrane feed by coagulation was investigated for improving membrane performance. The performance of the MBBR-MR was assessed based on its chemical oxygen demand (COD) removal efficiency and membrane fouling mechanisms. The study was carried out using pilot-scale MBBR and bench-scale batch membrane filtration setups (Millipore Inc. Bedford, MA). The pilot MBBR had a working volume of 1.8 m3 and a 30% carrier fill fraction. The MBBR was operated with loading rates of 160 ± 44 g/m2/d (hydraulic residence time (HRT) of 4.6 h) and 223 g/m2/d (HRT of 2.6 h). The MBBR feed was obtained from a starch recovery line in a potato chip processing factory. The carriers were mixed by coarse bubble aeration and the dissolved oxygen (DO) was maintained above 2 mg/l. Preliminary jar test trials (based on turbidity removal) were performed in order to obtain an optimal dosage of coagulants for subsequent ultrafiltration (UF) tests. The efficiency of three coagulants (alum, ferric chloride and a blend of polyaluminum chloride and polyamine) was evaluated. The membranes were composed of polyethersulfone (PES) and had a pore size of 0.05 microns. The results of this study indicate that the combination of MBBR with membrane filtration can be operated at relatively high loading rates to yield a constant high quality permeate that is suitable for water reuse purposes. Fouling of the membrane by the wastewater was found to be substantially reduced by treatment with the MBBR. The reversible and irreversible fouling of the MBBR effluent were 56 and 63%, respectively, of that observed with the raw wastewater. The MBBR Loading-rate was found to affect treatment efficiency of the MBBR-MR and membrane performance. Operation under the elevated loading-rate conditions HRT = 2.6 hours) resulted in an increase in the irreversible fouling of the membranes (60% on average). The addition of all the coagulants in this study was found to decrease the fouling of the membrane. However, the extent of the pre-coagulation effect on membrane fouling was found to strongly depend on the type and dosage of the coagulant and the MBBR effluent characteristics. All the coagulants were effective in decreasing membrane fouling at their optimal dosages which was determined in preliminary jar tests. Ferric chloride performed the best as a pretreatment coagulant compared to alum (Aluminum sulfate) and the coagulant blend with reductions in both reversible and irreversible fouling (43-86% and 51-71%, respectively) and increased consistency (in decreasing fouling) as compared to the other coagulants. Alum had no effect on irreversible fouling and the coagulant blend significantly increased irreversible fouling in some trials (up to 196% or by a factor of 3 when overdosed). Additionally, alum and the blend were, on average, 29% and 7%, less effective than ferric chloride in reducing reversible fouling under the conditions and dosages tested.

Ce travail est axé sur l’étude d’un profil de sol et des solutions de sol prélevés sur une parcelle expérimentale couvertes d’épicéas. Tous ces échantillons proviennent du Bassin Versant du Strengbach (Observatoire HydroGéochimique de l’Environnement – OHGE), ont été échantillonnés à différentes profondeurs (5, 10, 30 et 60 cm) et durant la période comprise entre 2009 et 2013.Les caractérisations des extraits des sols par spectroscopie Infra-Rouge ont permis de mettre en évidence les modifications des groupements fonctionnels organiques avec la profondeur et que ces modifications ont une forte incidence sur le comportement des cations (majeurs et traces) dans le sol. Des expérimentations d’ultra-filtration ont permis d’identifier les flux colloïdaux et dissous du carbone organique ainsi que ceux des éléments majeurs et traces présents dans les solutions de sol. L’utilisation conjointe des traceurs isotopiques (87Sr/86Sr et δ44/40Ca) et chimiques (Terres Rares) ont mis en évidence des processus ayant lieu aux interfaces eau-sol-plante, comme le prélèvement racinaire ou l’altération des sols. / This work is focused on the study of a profile of soil and soil solutions collected on an experimental plot covered with spruce. All these samples come from the watershed of the Strengbach (environment - OHGE Hydrogeochimique Observatory), were sampled at different depths (5, 10, 30 and 60 cm) and during the period between 2009 and 2013. Characterizations of soil extracts by infrared spectroscopy allowed to highlight changes in the organic functional groups with depth and that these changes have a significant impact on the behaviour of the cations (major and trace) in the soil. Ultrafiltration experiments helped to identify flows of colloidal and dissolved organic carbon as well as those of the major and trace-element present in soil solutions. The joint use of isotope tracers (87Sr / 86Sr and δ44 / 40 Ca) and chemical (Rare Earth Elements) have highlighted processes taking place at the water-soil-plant interface, as the uptake by root or soil alteration.