High-Gradient Magnetic Separation (HGMS) is a powerful separation process that has great potential for industrial wastewater treatment, particularly for the removal and recovery of paramagnetic colloidal particles. The chief advantages of HGMS are that the separation is reversible and potentially selective. However, these advantages are compromised if non-magnetic filtration mechanisms influence significantly the capture of particles. The objective of this study was to identify the chief mechanisms responsible for the removal of ferric oxide (Fe2O3) from water by an HGMS process. This objective was achieved by measuring the effects of applied magnetic induction, collector radius, and fluid velocity on the removal efficiency (RE) of a stainless-steel filter column. These factors were tested on the removal of bare Fe2O3 particles and particles treated with a surfactant (sodium dodecyl sulfate, SDS). The results were compared to the predictions of a trajectory model which simulates particle capture by a magnetic force. The experimental results show that non-magnetic force mechanisms are primarily responsible for the removal of bare Fe2O3 particles for the experimental conditions used in this work. For these particles, the three factors tested had no significant effect on the RE, and 90.1% of the particles were removed without a magnetic force. These results differed sharply from modeling predictions. However, the magnetic force mechanism is primarily responsible for the removal of surfactant-treated Fe2O3 particles. The three factors investigated had a marked effect on the RE, and only 10.8% of the particles were removed without a magnetic force. An increase in magnetic induction from 0.2 to 0.5 T increased the RE from 79.9 to 93.4 %; a decrease in wire radius from 49 to 15 Ym increased the RE from 60.2 to 93.4%, and a decrease in fluid velocity from 0.5 to 0.1 cm/s increased the RE from 69.5 to 95.3%. These results agreed closely with the model predictions.This study demonstrates that by reducing the effect of attractive non-magnetic forces on filtration, surfactant treatment of colloidal particles can potentially preserve and enhance these two key advantages, i.e., regeneration and selectivity of HGMS processes.

Halogenated organic compounds have had widespread and massive applications in industry, agriculture, and private households, for example, as degreasing solvents, flame retardants and in polymer production. They are released to the environment through both anthropogenic and natural sources. The most common chlorinated solvents present as contaminants include tetrachloroethene (PCE, perchloroethene) and trichloroethene (TCE). These chlorinated solvents are problematic because of their health hazards and persistence in the environment, threatening human and environmental health. Microbial reductive dechlorination is emerging as a promising approach for the remediation of chlorinated solvents in aquifers. In microbial reductive dechlorination, specialized bacteria obtain energy for growth from metabolic dechlorination reactions that convert PCE to TCE, cis-1,2-dichloroethene (cDCE), vinyl chloride (VC), and finally to benign ethene. Field studies show incomplete dechlorination of PCE to ethene due to lack of electron donors or other populations competing for the electron donor. Mathematical models are good tools to integrate the processes affecting the fate and transport of chlorinated solvents in the subsurface. This thesis explores the use of modeling to provide a better understanding of the reductive dehalogenation process of chlorinated solvents and their competition with other microorganisms for available electron donors in continuous flow systems such as a continuous stirred tank reactor (CSTR) and a continuous flow column. The model is a coupled thermodynamic and kinetic model that includes inhibition kinetics for the dechlorination reactions, thermodynamic constraints on organic acids fermentation and has incorporated hydrogen competition among microorganisms such as homoacetogenesis, sulfate reducers and ferric iron reducers. The set of equations are coupled to those required for modeling a CSTR. The system of model equations was solved numerically using COMSOL 3.5 a, which employs finite-element methods. The kinetic model was verified by simulation results compared to previously published models and by electron balances. The simulation process progressed by simulating the anaerobic reductive dechlorination, coupled with thermodynamic limitation of electron donor fermentation in batch systems to the modeling of CSTR, and finally to simulate anaerobic reductive dechlorination in continuous flow column, aquifer column including the processes of advection, dispersion and sorption along with the microbial processes of dehalogenation, fermentation, iron and sulfate reduction. The simulations using the developed model captured the general trends of the chemical species, and a good job predicting the dynamics of microbial population responses either the CSTRs or continuous flow column. Although, the kinetic of anaerobic dechlorination processes of chlorinated solvents in those systems have been researched in the past, little progress has been made towards understanding the combined effects of the dechlorination and thermodynamic constraints in continuous flow systems. This work provides a rigorous mathematical model for describing the coupled effects of these processes. / Graduation date: 2012

In der vorliegenden Arbeit wurde die Methode der magnetischen Kernspinresonanz (NMR) eingesetzt, um beim Schadstoffabbau stattfindende Prozesse und geochemische Reaktionen zerstörungs- und beprobungsfrei zu untersuchen. Dies ist möglich, da die gelösten Elektronenakzeptoren Sauerstoff und Eisen paramagnetisch sind und somit einen Ein uss auf die NMRRelaxationszeiten ausüben. Der lineare Zusammenhang zwischen der gelösten Sauerstoff- beziehungsweise Eisen-Konzentration und den NMR-Relaxationsraten 1/T1 und 1/T2 wurde quantifiziert. Weiterhin wurde der bereits bekannte Einfluss der Matrixoberflächen von porösen Medien auf die Relaxation von Wasser nachgewiesen. Die paramagnetischen Zentren auf Sandoberflächen führen ebenfalls zu einer Verkürzung der Relaxationszeiten. Es wurde gezeigt, dass die kleinsten Korngrößen der verwendeten Sande den größten Einfluss auf die Oberflächenrelaxation haben. Wird die Oberflächenrelaxation berücksichtigt, ist auch in porösen Medien die ermittelte lineare Abhängigkeit der Relaxationszeiten von der Ionenkonzentration anwendbar, um den Gehalt an gelösten paramagnetischen Ionen aus Relaxationsmessungen zu ermitteln. Beispielhaft wurde der Anstieg der Eisen(III)-Konzentration in der Porenlösung von natürlichen Sanden infolge der Auflösung eisenhaltiger Mineralien von den Oberflächen zeitlich und räumlich detailliert betrachtet. Eine durchgeführte Modellierung zeigte, dass das Reaktionssystem zu Beginn der Reaktion von der Diffusion dominiert wird, am Ende dann die Reaktionsgeschwindigkeit der bestimmende Parameter ist. Die beim biologischen Schadstoffabbau auftretenden Redoxprozesse des Eisens wurden durch rein chemische Reaktionen unter Verwendung von Oxidations- und Reduktionsmitteln simuliert. Die zeitlich und räumlich detaillierte Beobachtung des Anstiegs beziehungsweise des Abfalls der gelösten Eisen(III)-Konzentration in der (Poren-)Lösung war mit NMR-Relaxometrie trotz der Schnelligkeit der Reaktionen möglich. Mit Hilfe der anschliessenden Modellierung wurde der wichtige Einfluss des pH-Wertes auf den genauen Ablauf der Reaktionen deutlich gemacht. Nur in sehr sauren pH-Bereichen (pH < 3) liegen die Eisen(III)-Ionen in Lösung vor. Weiterhin wurde der Einfluss der Mikroorganismen selbst auf die NMR-Relaxations- und Diffusionsmessungen untersucht. Im Rahmen dieser Arbeit wurde an Medien mit Lactobacillus und Penicillium eine Verschiebungen in den Relaxationszeitverteilungen hin zu kleineren Relaxationszeiten gemessen. Dies basiert auf der bereits bekannten Verringerung der Mobilität der Spins innerhalb der Biomasse. Für Bakterien von Geobacter metallireducens konnte erstmals der Verbrauch von Eisen(III)-Ionen durch Reduktion während des Wachstum anhand der ansteigenden T2-Relaxationszeit gezeigt werden.

This work details the synthesis and growth mechanisms of 1D magnetic and semiconducting nanostructures. Specifically, magnetic iron oxide and ZnS-SiO2 nanowires are examined. These materials are chosen due to their promise for biomedical and electronic applications and the perceived need to both create these structures as tools for these applications and to understand their formation processes so that they can be manufactured at a scale and efficiency suitable for commercialization. The current state and impact of nanotechnology is discussed through the lens of continuing technological advances and environmental factors, and the term is defined according to a specific set of criterion involving size, utility, and uniqueness. Details of synthesis and characterization of Fe3O4, ε-Fe2O3, and ZnS-SiO2 core-shell nanowires are presented. Observations regarding the growth of these structures are paired with additional experiments, simple simulations, and other literature to discuss the classical VLS growth process in general, and its applicability to these structures in particular. Finally, some exciting future applications are discussed, with details for initial experimental work presented in the appendix.

The oxidative dissolution of chalcopyrite in ferric media often produces incomplete copper recoveries. The incomplete recoveries have been attributed to inhibition caused by the formation of a metal deficient sulphide and the deposition of elemental sulphur and jarosite. Although these phases have been qualitatively identified on the surface of chalcopyrite, none have been quantitatively identified. The aim of the project was to quantitatively analyse the surface before and after oxidative dissolution, with X-ray photoelectron spectroscopy (XPS), and to use the phases identified as the basis for mechanisms of dissolution and inhibition. / XPS analysis was performed on chalcopyrite massive fractured under anaerobic atmosphere and chalcopyrite massive and concentrate oxidised in 0.1 M ferric sulphate (pH 1.9) and 0.2 M ferric chloride (pH 1.6) at 50, 65 and 80ºC. Quantitative XPS analysis of the chalcopyrite surfaces required the development of programs that accounted for the observed XPS spectra. The output of these programs was used to construct profiles of the chalcopyrite surfaces and the deposited phases. These surface profiles were correlated with copper recoveries determined for chalcopyrite concentrate dissolution under the same conditions. / The surface of chalcopyrite before oxidative dissolution reconstructs to form a `pyritic' disulphide phase. This phase is oxidised in ferric media to form thiosulphate via the incorporation of oxygen atoms from the hydration sphere. The thiosulphate reacts in the oxidising conditions of low pH to form elemental sulphur, sulphite and sulphate. The sulphate complexes with ferric to produce hydronium jarosite. This reaction occurs at the surface during the initial stages of dissolution and in the bulk solution during the latter stages. This precipitation of hydronium jarosite during the latter stages of dissolution corresponds to inhibition of the dissolution reaction. It is therefore concluded hydronium jarosite is responsible for inhibiting the oxidative dissolution of chalcopyrite in ferric media. / The identification of hydronium jarosite as the inhibiting phase is consistent with the industrial practice of removing `excess' iron from the ferric solution before oxidative dissolution. However, additional iron and sulphate are generated at the chalcopyrite surface during oxidative dissolution. These high iron and sulphate concentrations combine with the low pH and high temperatures favoured for the oxidative dissolution of chalcopyrite to produce ideal conditions for jarosite precipitation. Therefore, pH must be lowered further to prevent jarosite precipitation and enhance copper recoveries from chalcopyrite in ferric media.

The thesis is focused on the area of wastewater treatment. There is a solution possibility of using flotation for the removal of phosphorus precipitates formed during the tertiary treatment of wastewater. The first part is devoted to a general description of wastewater treatment, there is also described as a flotation separation technology and precipitation of phosphorus. The second part is practically oriented. It describes the experimental equipment used for the precipitation of phosphorus and subsequent removal of the precipitate flotation, during the experiment and the results obtained.

No que tange o saneamento, cresce a opinião entre a comunidade técnico-científica de que é necessário que a água seja gerida levando em conta a complexidade de seu ciclo hidrológico e que o esgoto e os resíduos sejam vistos como fontes de recursos. É nesses preceitos que se baseia o saneamento ecológico. Neste trabalho é avaliado o desempenho de uma tecnologia de saneamento ecológico, operada tratando esgoto doméstico real e projetada para atender à demanda de uma família de três pessoas e promover a reciclagem de nutrientes. O sistema é composto por 3 reservatórios que atuam como um reator anaeróbio, um decantador e um alagado construído de fluxo vertical ascendente. O sistema apresentou remoção global de DQO de 88,6% ± 6,3% com concentração final de 80 ± 35 mg O2.L-1 e as remoções de nutrientes foram insignificantes. Como o fósforo é o nutriente mais difícil de ser tratado biologicamente, e, como o lodo de Estações de Tratamento de Água (ETA) é um tipo de resíduo sólido que carece de manejo e tratamento adequado, propõe-se também uma tecnologia inovadora para as situações em que é necessário remover grandes cargas de fósforo. Essa tecnologia consiste no reaproveitamento do lodo de ETA como adsorvente, utilizando-o como meio de preenchimento de alagados construídos. Primeiramente, foram feitos ensaios em mesa giratória com solução padrão e foram obtidas as isotermas de Freundlich e Langmuir para 4 diferentes valores de pHs. A adsorção foi favorecida em pHs ácidos e a capacidade máxima de adsorção variou entre 1,44 mg P.g-1 a 4,76 mg P.g-1. Posteriormente, foi avaliado o desempenho desse lodo tratando esgoto real. Para isso, adicionou-se um estágio complementar de tratamento no sistema apresentado anteriormente. A aplicação do lodo como meio de preenchimento do alagado construído apresentou bom desempenho e levou à conclusão que o tempo de vida útil do sistema seria entre 1,6 e 9,7 anos. Por fim, busca-se trazer contribuições que transcedam a esfera da técnica; por essa razão é apresentada uma reflexão analítica conceitual com o intuito de analisar o processo sócio histórico de formação do paradigma tradicional do saneamento, baseado no referencial teórico das ciências sociais da tecnologia (DOSI, 1982; FEENBERG, 1992, 1999, 2003). A partir disso intenta-se vislumbrar uma alternativa de futuro e avaliar as potencialidades e limitações do saneamento ecológico para construção de outro paradigma sanitário. / Regarding sanitation, there is a growing consensus among the technical-scientific community that water must be managed considering the complexity of its hydrological cycle and that sewage and waste should be considered as sources of resources. It is in these precepts that ecological sanitation is based. In this work the performance of an ecological sanitation technology, operated by treating domestic sewage and designed to meet the demand of a family of three people and promote the recycling of nutrients, is evaluated. The system consists of 3 stages that act as an anaerobic reactor, a decanter and an upflow constructed wetland. The system had a total COD removal of 88.6 ± 6.3% with a final concentration of 80 ± 35 mg O2.L-1 and nutrient removals were insignificant. As phosphorus is the most difficult nutrient to be treated biologically, and since water treatment plant (WTP) sludge is a type of solid waste that still lacks proper management and treatment in Brazil, it is also proposed an innovative technology for situations where it is necessary to remove large loads of phosphorus. This technology consists in the reuse of the WTP sludge as an adsorbent, using it as a constructed wetland filling material. First, rotary table tests with stock solution were taken and the Freundlich and Langmuir isotherms were obtained for 4 different pH values. Adsorption was favored at acidic pHs and the maximum adsorption capacity ranged from 1.44 mg P.g-1 to 4.76 mg P.g-1. Subsequently, the performance of this sludge treating sewage was evaluated. To this end, a further stage of treatment was added in the system shown above. The application of the sludge a constructed wetland filling material presented good performance and led to the conclusion that the useful life of the system would be between 1.6 and 9.7 years. Finally, we seek to bring contributions that transcend the technical sphere; for this reason, a conceptual analytical reflection is presented with the purpose of analyzing the socio-historical process of formation of the traditional sanitation paradigm, based on the theoretical reference of the social sciences of technology (DOSSIER, 1982; FEENBERG, 1992, 1999, 2003). From this it is tried to glimpse an alternative of future and to evaluate the potentialities and limitations of the ecological sanitation to the construction of another sanitary paradigm.

Submitted by Marilene Donadel (marilene.donadel@unioeste.br) on 2017-12-01T22:57:34Z No. of bitstreams: 1 Juliana_Dotto_2013.pdf: 2822116 bytes, checksum: 3b8be919381051b7c1e0d2252f604aed (MD5) / Made available in DSpace on 2017-12-01T22:57:34Z (GMT). No. of bitstreams: 1 Juliana_Dotto_2013.pdf: 2822116 bytes, checksum: 3b8be919381051b7c1e0d2252f604aed (MD5) Previous issue date: 2013-02-19 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / This study evaluated the efficiency of organic coagulants like seeds of Moringa oleifera Lam extracted in saline solutions of sodium chloride and potassium chloride1,0 (MO-NaCl and MO-KCl, respectively) and inorganic coagulants like aluminum sulfate and ferric chloride to the treatment of textile wastewater. Initially, it has been evaluated the best pH for each coagulant studied to then optimize the concentration of the coagulants and settling time. For this finality, it has been used velocity of rapid mixing (VMR) and slow (VML) at 100 rpm and 20 rpm, respectively and time of rapid mixing (TMR) and slow (TML) for 2 minutes and 20minutes, respectively. The parameters analyzed were pH after treatment, absorbance at a wavelength of maximum absorption for each dye present in the wastewater (Procion Orange HER and Procion Red HE-7B), color, turbidity and COD. In the test for optimization of coagulants concentration and settling time, also was evaluated the amount of sludge generated at each concentration studied. Thus, organic coagulants showed superior results in comparison with inorganic coagulants in the treatment of textile wastewater, especially MO-KCl, which showed removal of 86,8% for absorbance of the dye Procion Orange HER, 75,65% for absorbance of the dye Procion Red HE-7B, 70,67%forcolor, 85,97% for COD and 64,07% for turbidity. Based on the results obtained, MO-KCl presented a promising coagulant in the treatment of textile wastewater. / O presente estudo avaliou a eficiência dos coagulantes orgânicos sementes de Moringa oleifera lam extraídas em soluções salinas de cloreto de sódio e cloreto de potássio 1,0 (MO-NaCl e MO-KCl) respectivamente, além dos inorgânicos sulfato de alumínio e cloreto férrico no tratamento de água residuária têxtil. Inicialmente, avaliou-se o melhor pH de trabalho para cada coagulante estudado, para, em seguida, otimizar a concentração do coagulante e o tempo de sedimentação. Para isto, utilizou-se velocidade de mistura rápida (VMR) e lenta (VML) de 100 rpm e 20 rpm respectivamente, e tempo de mistura rápida (TMR) e lenta (TML) de 2 minutos e 20 minutos, respectivamente. Os parâmetros analisados foram pH após tratamento, absorbância no comprimento de onda máximo de cada corante presente na água residuária (Laranja Procion HER e Vermelho Procion HE-7B), cor, turbidez e DQO. Além destes, no teste para otimização da concentração de coagulante e tempo de sedimentação, avaliou-se a quantidade de lodo gerado em cada concentração estudada. Assim, os coagulantes orgânicos apresentaram resultado superior aos inorgânicos no tratamento da água residuária têxtil, com destaque para a MO-KCl, que apresentou remoção de 86,8% para absorbância do corante Laranja Procion HER, 75,65% para absorbância do corante Vermelho Procion HE-7B, 70,67% para cor aparente, 85,97% para DQO e 64,07% para turbidez. Com base nos resultados obtidos a MO-KCl pode ser considerada um coagulante promissor para o tratamento da água residuária têxtil.

Os reatores anaeróbios de tratamento de esgoto possibilitam elevada remoção de matéria orgânica, com menor custo em comparação com os processos aeróbios. Entretanto, esses reatores geralmente são pouco eficientes na remoção de nutrientes e seu efluente pode apresentar valores relativamente elevados de DQO e de sólidos suspensos. Por isso, investigou-se a aplicação de cloreto férrico e auxiliares de floculação no afluente de um reator anaeróbio de leito expandido em escala real (Volume = 32 m3; TDH = 3,2 h), no intuito de elevar as eficiências de remoção de DQO, turbidez, fósforo e sólidos. Foram realizados testes de Atividade Metanogênica para avaliar a toxicidade do cloreto férrico ao lodo do reator, mas nenhum efeito tóxico foi detectado para concentrações de até 500 mg FeCl3/l. Em seguida, foram realizados testes em reator de bancada (jar test), quando foi investigado o uso de cloreto férrico (dosagens entre 40 e 200 mg/l) em conjunto com um polímero catiônico, um polímero aniônico e amido natural de batata (dosagens entre 0,5 e 2,0 mg/l), sendo obtidos os melhores resultados com cloreto férrico e polímero aniônico. As eficiências médias diárias de remoção no reator de leito expandido foram elevadas de 64% para 71%, para DQO; de 47% para 59%, para turbidez; de 17% para 31%, para fósforo; e de 52% para 77%, para SST, quando aplicadas dosagens de 40 mg/l de cloreto férrico e 1,0 mg/l de polímero aniônico. / Anaerobic reactors for wastewater treatment have high capacity in removing organic matter and lower costs in comparison with aerobic processes. Nevertheless, anaerobic reactors generally are less efficient in nutrient removal and their effluents may present relatively high COD and total suspended solid (TSS) concentrations. Hence, the application of ferric chloride (FeCl3) and flocculant aids to a full-scale (Volume = 32 m3; HRT = 3,2 h) anaerobic expanded bed reactor influent was investigated to improve removal of COD, turbidity, phosphorous and solids. Methanogenic activity tests were conducted to evaluate toxicity of ferric chloride against methanogenic anaerobic reactor sludge. Tests showed that no toxic effect occurred before concentrations of FeCl3 reached 500 mg/l. Conventional bench-scale jar tests were conducted and ferric chloride was tested in combination with a cationic polymer, an anionic polymer and natural potato starch. Best results were obtained when ferric chloride was employed with anionic polymer. Average daily removal rates were increased from 64% to 71% for COD, from 47% to 59% for turbidity, from 17% to 31% for phosphorous and from 52% to 77% for TSS when dosages close to 40 mg/l of ferric chloride and 1,0 mg/l of anionic polymer were applied to anaerobic expanded bed reactor influent.

Uma das abordagens para minimizar o volume de captação de água potável e o descarte de efluentes é o reuso da água dentro da unidade industrial através de técnicas avançadas de tratamento como o sistema de osmose reversa, porém esse sistema exige um afluente com uma baixa quantidade de sólidos, além de outros requisitos. A água residuária utilizada nesse estudo é proveniente de uma empresa do ramo metal-mecânico que utiliza processos de coagulação-floculação-sedimentação como forma de remover os sólidos antes de um sistema de osmose reversa. A proposta deste trabalho foi avaliar três diferentes coagulantes (aluminato de sódio, cloreto férrico e uma mistura de dois coagulantes comerciais, Procytrat 100A e Procytrat 300) em processos de coagulação-floculação e separação de sólidos da fase líquida por sedimentação e flotação por ar dissolvido e também comparar os custos aproximados dos processos envolvidos. Foram realizados ensaios em testes de jarros e flotateste e com os dados obtidos construíram-se diagramas de coagulação-floculação e escolhidas as regiões de maior remoção de sólidos. Através das regiões escolhidas, foram feitas as comparações de eficiência de remoção e custos para cada coagulante estudado. Verificou-se que os diagramas de coagulação-floculação-sedimentação foram os que apresentaram as regiões de maior área com turbidez remanescente menor que 2%, principalmente para os coagulantes cloreto férrico e a mistura Procytrat 100A + Procytrat 300, chegando a alcançar valores próximos a 0,5%. A opção mais econômica foi alcançada com custo de cloreto férrico, na dosagem de 200 mg/L, pH igual a 8,3 e turbidez remanescente igual a 0,6%, de R$0,47/\'M POT.3\' de água tratada. Os diagramas de coagulação-floculação-flotação apresentaram regiões de turbidez remanescente com áreas bem mais reduzidas se comparadas aos diagramas de coagulação-floculação-sedimentação, sendo que não foram atingidos valores de turbidez remanescente menores que 1,5%. Os menores valores de turbidez remanescente para o processo de flotação foram encontrados utilizando-se como coagulante a mistura Procytrat 100A + Procytrat 300, sendo a opção mais econômica alcançada para pH igual a 3,3, dosagem de 60 mg/L e turbidez remanescente igual a 1,6%, com custo de R$0,33/\'M POT.3\' de água tratada. De um modo geral, para todos os coagulantes testados, as regiões escolhidas nos diagramas de coagulação-floculação-flotação, cujos valores de turbidez remanescente foram menores, estavam concentradas em valores de pH baixos (3,2 a 5,5), dosagens de coagulante reduzidas (10 a 80 mg/L) e dosagens de alcalinizante também reduzidas (0 a 60 mg/L de Ca(\'OH)IND.2\'). Esse comportamento foi contrário ao apresentado pelos diagramas coagulação-floculação-sedimentação cujos valores de turbidez remanescente foram menores em valores de pH mais altos (6,5 a 8,5), maiores dosagens de coagulante (40 a 380 mg/L) e altas dosagens de alcalinizante (100 a 400 mg/L de Ca(\'OH)IND.2\'). / One approach to minimize the consumption of drinking water and effluent disposal is the reuse of water within the plant through advanced treatment techniques such as reverse osmosis system, however, this system requires an affluent with a low amount of solids and other requirements. The wastewater used in this study comes from a metalworking company that uses coagulation-flocculation-sedimentation process as a way to remove the solids before a reverse osmosis system. The purpose of this study was to evaluate three different coagulants (sodium aluminate, ferric chloride and a mixture of two commercial coagulants, Procytrat 100A and Procytrat 300) in the process of coagulation-flocculation and separation of solids from the liquid phase by sedimentation and dissolved air flotation and also compare the approximate costs of the processes involved. Jar tests and flotatests were performed and the coagulation-flocculation diagrams were built. In the coagulation-flocculation diagrams were selected areas of higher solids removal. Through the chosen regions, comparisons were made of removal efficiency and cost study for each coagulant. It was found that the diagrams of coagulation-flocculation-sedimentation presented the areas of higher turbidity remaining area with less than 2%, mainly for the coagulants ferric chloride and the mixture Procytrat 100A + Procytrat 300, reaching values close to 0.5 %. The most economical option was achieved at a cost of ferric chloride at a dosage of 200 mg/L, pH 8.3, and the remaining turbidity equal to 0.6%, from R$ 0.47/\'M POT.3\' treated water. The diagrams of coagulation-flocculation-flotation showed regions with turbidity remaining areas much reduced compared to the diagrams of coagulation-flocculation-sedimentation, and the turbidity remaining values were more than 1.5%. The lower values of turbidity remaining for the flotation process were found using the mixture Procytrat 100A + Procytrat 300, and the most economical option was reached for a pH of 3.3, dosage of 60 mg/L and remaining turbidity equal to e 1.6%, at a cost of R$ 0.33/\'M POT.3\' of treated water. In general, for all coagulants tested, the selected regions in the diagrams of coagulation-flocculation-flotation, whose remaining turbidity values were lower, were concentrated in low pH values (3.2 to 5.5), dosages of coagulant reduced (10 to 80 mg/L) and dosages of alkaline also reduced (0 to 60 mg/L of Ca(\'OH)IND.2\'). This behavior was contrary to the diagrams presented by coagulation-flocculation-sedimentation whose remaining turbidity values were lower in higher pH values (6.5 to 8.5), larger amounts of coagulant (40 to 380 mg/L) and high doses of alkaline (100 to 400 mg/L Ca(\'OH)IND.2\').