acase@tulane.edu

This research addresses refractory forms of nitrogen that, even with advanced biological nitrification-denitrification systems are not removed completely from domestic wastewater. TKN (Total Kjeldahl Nitrogen), ammonia plus organic nitrogen, is one of the forms to measure the levels of nitrogen present in effluent wastewaters. Ferrate, a strong oxidant, was used for the treatment of these nitrogen forms with the objective of producing nitrogen compounds that can be removed by subsequent biological processes. Bench-scale experiments were performed on effluent samples taken prior to chlorination from an Orlando, FL wastewater treatment facility, using a biological nutrient removal process. The samples were treated with doses of ferrate ranging from 1 to 50 mg/L as FeO4–2 under unbuffered conditions. TKN removal as high as 70% and COD removal greater than 55% was observed. The TSS production after ferrate treatment was in a range of 12 to 200 mg/L for doses between 10 and 50 mg/L FeO4-2. After an optimum dose of ferrate was determined, three bench-scale reactors were operated under anoxic conditions for 10 to 12 days, two as duplicates containing the treated effluent and one as a control with untreated sample. Two different doses of ferrate were used as optimum dose for these experiments, 10 and 25 mg/L as FeO4-2. The purpose of these reactors was to determine the potential for biological removal of remaining nitrogen after ferrate oxidation of refractory nitrogen. Treated and raw samples were analyzed for Total Kjeldahl Nitrogen (TKN) (filtered and unfiltered), chemical oxygen demand (COD) (filtered and unfiltered), total suspended solids (TSS), nitrate (NO3-N), nitrite (NO2-N), and heterotrophic plate count (HPC). As a result, more than 70% of the soluble TKN was removed by chemical and biological oxidation for a sample treated with a dose of 25 mg/L FeO4-2, and less than 50% when treated with 10 mg/L FeO4-2. For the control samples run parallel to the ferrate treated samples, a maximum of 48% of soluble TKN and a minimum of 12% was removed. A three-log increase was observed in heterotrophic bacteria numbers for both doses during the operation of the reactors. Sodium ferrate was found to be an effective oxidant that can enhance the biodegradability of recalcitrant TKN present in municipal wastewaters. As mentioned before this research was develop using batch reactor units at bench-scale, therefore it is recommended to follow the investigation of the biodegradability of recalcitrant TKN of a ferrate treated sample under continuous flow conditions so that results can be extrapolated to a full-scale treatment facility.

The objective of the study was to compare the trihalomethanes (THMs) produced from ferrate with hypochlorite and to determine how different the THM production would be for a given degree of disinfection (3 log reduction in Heterotrophic Plate Count (HPC)). Different water samples were collected from Lake Claire, Atlantic Ocean, and secondary effluent from an advanced wastewater treatment plant. THM formation was determined using a standard assay over 7 days at room temperature. In addition samples were tested for Total Coliform Escherichia coli (TC/E.coli), and heterotrophic bacteria using HPC by spreadplating on R2A agar. Dissolved organic carbon (DOC) was measured as well. Dosages of 2, 5, and 10 ppm of hypochlorite and ferrate were used for Lake Claire and Atlantic Ocean water, while 1, 2, and 5 ppm dosages were used for wastewater treatment effluent. Ferrate resulted in 48.3% ± 11.2% less THM produced for the same level of disinfection (i.e. approximately 3 logs reduction in HPC). Oxidation of DOC was relatively small with a 6.1 to 11.6 % decrease in DOC being observed for ferrate doses from 2 to 10 mg/L. Free chlorine oxidation of DOC was negligible.

Ferrate (VI) has been studied as an alternative chemical to disinfect water and wastewater in recent years. The disinfection effectiveness of two different wet oxidation ferrate (VI) synthesis formulas in wastewater and Econlockhatchee River water was evaluated. Ferrate (VI) is synthesized by addition of ferric chloride to a mixture of sodium hydroxide and calcium hypochlorite (refer to U.S. Patent 6,790,429). One ferrate (VI) synthesis formula uses below the stoichiometric requirement of hypochlorite (Low Chlorine Formula) while the other ferrate (VI) synthesis formula uses more than the stoichiometric requirement of hypochlorite (Standard Chlorine Formula). For applications requiring low chlorine residual effluent quality, the Low Chlorine Formula intuitively is a more suitable disinfectant than the Standard Formula. For applications where chlorine residual is of little or no significance, the Standard Formula is logically a more suitable disinfectant due to lower production cost and production of higher ferrate (VI) concentrations than the Low Chlorine Formula. The total chlorine concentration, unfiltered and filtered ferrate (VI) concentration, and dissolved organic carbon concentration before and after treatment using both ferrate (VI) formulas in wastewater and Econ River water was measured at a contact time of 30 minutes. Disinfection capabilities were measured by comparing the quantity of Heterotrophic bacteria, Total Coliform, Escherichia coli, and Enterococcus bacteria pre-ferrate (VI) to post-ferrate (VI) at dosages of 2, 4, and 7.5 mg/L as ferrate (VI) using both ferrate (VI) formulas. The rate of disappearance of both ferrate (VI) formulas in wastewater at an unadjusted pH and pH of 6.0-6.35 was determined. In addition the total oxidant absorbance and total chlorine concentration were measured over a 30-minute period. Both ferrate (VI) formulas were effective at inactivating Total Coliform, E. Coli, Enterococcus, and heterotrophic bacteria at a 30-minute contact time and lowering DOC concentrations in Econlockhatchee River water and secondary wastewater. The Standard Formula demonstrated better disinfection at lower dosages than the Low Chlorine Formula. In both ferrate (VI) formulas, there was a presence of an instantaneous demand of ferrate (VI) and a first-order reaction rate of ferrate (VI) over 30 minutes. The chlorine residual of 7.5 mg/L ferrate (VI) dose in wastewater at a 30-minute contact time was 0.2 to 0.6 mg/L Cl2 for the Low Chlorine Formula and 0.8 to 1.4 mg/L Cl2 for the Standard Formula. These experiments indicate that both ferrate (VI) formulas can serve as effective environmentally friendly disinfectants for wastewater and Econ River water.

This dissertation is about treatment of the nonbiodegradable organic content of landfill leachate by chemical oxidation combined with biological treatment. It is divided into three parts. In the first part, ferrate was compared to Fenton's reagent for the purpose of removing non-biodegradable organic compounds from mature leachate. Oxidation conditions (time, pH, and dose) were optimized to yield maximum organic removal using two leachate samples from 20 and 12-year old solid waste cells. Results from this research demonstrated that ferrate and Fenton's reagent had similar optimum pH ranges (3-5), but different organic removal capacities, ranging from 54 to 79 % of initial leachate organic contents. An advantage of ferrate was that it was relatively effective over a wide pH range (Fenton's reagent lost its reactivity outside optimum pH range). Advantages associated with Fenton's reagent include a higher organic removal capacity, production of more oxidized organic compounds (measured as chemical oxygen demand/dissolved organic carbon), and production of more biodegradable byproducts (measured as 5-day biochemical oxygen demand/chemical oxygen demand). Finally, both treatments were found to oxidize larger molecules (>1000 dalton) and produce smaller molecules, as indicated by an increase in smaller molecule contribution to organic carbon. In part two, effects of Fenton's reagent treatment on biodegradability of three landfill leachates collected from a Florida landfill were evaluated using biochemical oxygen demand (BOD), biochemical methane potential (BMP), and tertamethylammonium hydroxide (TMAH) thermochemolysis gas chromatography/mass spectrometry (GC/MS). The hypothesis was that Fenton's reagent will remove refractory compounds that inhibit biodegradation and will produce smaller, more biodegradable organic molecules which will result in an increase in BOD and BMP values. Both BOD and BMP results demonstrated that Fenton's reagent treatment did not convert mature leachate to biodegradable leachate, as indicated by a low BOD5 expressed as C /dissolved organic carbon (DOC) ratio of almost 0.15 in treated samples and a low net methane production / theoretical methane potential (less than 0.15). Ultimate BOD only slightly increased. However the first-order BOD reaction rate increased by more than five fold, suggesting that Fenton's reagent removed refractory and inhibitory compounds. BMP results demonstrated that the ratio of CO2/CH4 produced during anaerobic biodegradation did not increase in treated leachate (compared to untreated), indicating that small biodegradable organic acids produced by oxidation were removed by coagulation promoted by Fenton's reagent. Finally, the TMAH thermochemolysis results showed that several of the refractory and inhibitory compounds were detected fewer times in treated samples and that carboxylic acids did not appear in treated samples. In the third part of this dissertation the application of flushing/Fenton's reagent oxidation to produce sustainable solid waste cells was evaluated. A treatment similar to pump and treat process utilizing Fenton's reagent on-site treated leachate combined with in-situ aeration was proposed. Treated leachate would be recycled to the landfill cell flushes releasable nonbiodegradable carbon from the cell and oxidizes it externally. This technique was demonstrated to have treatment cost and time benefits over other alternatives for producing completely stable solid waste cells such as anaerobic flushing and biological and/or mechanical pretreatment of solid waste (used in the EU).

This study evaluated alternative sulfide treatment processes for potable water systems that rely on groundwater supplies. Research for this study was conducted at the Imperial Lakes (IL) and Turner Road (TR) water treatment plants (WTPs) in Polk County, Florida. These WTPs are in the process of refurbishment and expansion, and will require the installation of a new groundwater well. The IL and TR WTPs both rely upon groundwater sources that contain total sulfide at concentrations ranging from 1.4 to 2.6 mg/L. Sulfide is a concern because if left untreated it can impact finished water quality, corrosivity, create undesirable taste and odor, and oxidize to form visible turbidity. For this reason, the raw water will require treatment per Florida Department of Environmental Protection (FDEP) "Sulfide Rule" 62-555.315(5)(a). This rule does not allow the use of conventional tray aeration (currently in use at the IL and TR WTPs) for wells that have significant total sulfide content (0.6 to 3.0 mg/L). This research was commissioned because the potential water treatment method identified in the Sulfide Rule (i.e. forced-draft aeration) would not adequately fit within the confines of the existing sites and would pose undue burden to neighboring residents. In addition, an effective sulfide treatment process was desired that offered a low profile, did not necessitate the need for additional complex chemical feed systems, minimized the extent of electrical infrastructure upgrades, and was inexpensive to construct and operate. To meet these goals, several alternative technologies were evaluated at the desktop and bench-scale; these included anion exchange, various oxidation methods, and alternative media filtration processes. From that effort, several processes were selected for evaluation at the pilot scale: bleach (NaOCl) oxidation preceding electromedia filtration; manganese (IV) oxide (MnO2) filtration continuously regenerated with bleach; and ferrate (Fe(VI)) oxidation. Electromedia and MnO2 filtration were shown to be effective for total sulfide treatment. Both processes reduced total sulfide content to below detection levels ( > 0.1 mg/L) for groundwater supplies containing as much as 2.6 mg/L of total sulfide. The use of bleach oxidation ahead of media filtration also produced finished water with low turbidity ( > 1.0 NTU) as compared to conventional tray aeration and chlorination processes (6-16 NTU, as observed in this study). It was determined that the media filtration approach (electromedia and MnO2) was effective for sulfide treatment and met the County's site objectives established at the outset of the project. Ferrate was also shown to reduce total sulfide content to below detection levels ( > 0.1 mg/L) for groundwater supplies containing as much as 2.6 mg/L of total sulfide. An opinion of probable capital costs for installing a sulfide oxidation/filtration process at either the Imperial Lakes or Turner Road WTP was estimated to range from roughly $830,000 to $1,100,000. That equates to a $/kgal capital cost of $0.10 to $0.32 (at 8% for 20 years). An opinion of annual probable bleach chemical costs was estimated to range from $3,500 to $9,800 for the IL WTP and $3,500 to $5,800 for the TR WTP.

O 2,4-diclorofenol (2,4-DCF), poluente modelo usado neste trabalho, é um composto empregado na produção de pesticidas, herbicidas e anti-sépticos, e comumente encontrado em efluentes industriais, sendo considerado um poluente prioritário devido a sua elevada toxicidade e alto poder de persistência no ambiente. Este trabalho objetivou estudar a degradação do 2,4-DCF em reator fotoquímico tubular com coletores parabólicos compostos (CPC) irradiado pelo Sol, operado em batelada com recirculação e admissão contínua de peróxido de hidrogênio à vazão de 11 mL min-1. A oxidação do poluente foi baseada no processo foto-Fenton catalisado por tris(oxalato) ferrato(III) (FeOx), investigando-se os efeitos das seguintes variáveis: concentração inicial de 2,4-DCF ([2,4-DCF] = 21-520 mgCL-1), concentrações de oxidante ([H2O2] = 9,9-130 Mm) e de fotocatalisador ([FeOx] = 0,2-1,8 mM) e área de coleta de fótons (0,196-0,98 m2), estudadas a partir de um planejamento Doehlert em múltiplos níveis. A foto-oxidação foi monitorada por medidas da concentração de carbono orgânico total (TOC) e oxigênio dissolvido (OD), tendo sido identificadas condições com remoção de TOC alcançando 99,8% em 90 minutos de tratamento. Nas análises estatísticas dos resultados, os efeitos da área de coleta de fótons e da concentração de H2O2 concordaram com o comportamento esperado, sugerindo, porém, efeitos sequestradores de radicais hidroxila quando a concentração de H2O2 é alta. A partir dos dados experimentais, obteve-se um modelo de redes neurais, usado para previsão da taxa de remoção de TOC em função do tempo. Através do método HIPR (Holdback Input Randomization Method) e dos pesos associados a cada variável de entrada da rede neural, avaliou-se a importância relativa das variáveis de entrada. De maneira geral, os resultados comprovaram a viabilidade do processo foto-Fenton catalisado por tris(oxalato) ferrato III e irradiado pelo Sol para o tratamento de efluentes aquosos contendo 2,4-diclorofenol. / 2,4-dichlorophenol (2,4-DCF), the model pollutant selected in this work, is a compound used in the production of pesticides, herbicides, antiseptics, and commonly found in industrial wastewaters. It is considered a priority pollutant due to its high toxicity and high persistence in the environment. This study investigated the degradation of 2,4-DCF in a tubular photochemical reactor with compound parabolic collectors (CPC) irradiated by the sun, operated in batch with recirculation and continuous feed of hydrogen peroxide at a flow rate of 11 mL min-1. The oxidation of the pollutant was based on the photo-Fenton process catalyzed by tris (oxalate) ferrate (III) (FeOx). The effects of the following variables were investigated: initial concentration of 2,4-DCF ([2,4-DCF] = 21-520 mgCL-1), concentrations of oxidant ([H2O2] = 9,9-130 Mm)); photocatalyst ([FeOx]= 0,2-1,8 mM)), and photon collection surface (0,196-0,98 m2), according to a Doehlert design on multiple levels. The photo-oxidation was monitored by measurements of the concentration of total organic carbon (TOC) and dissolved oxygen (OD). Conditions with TOC removal achieving 99.8% within 90 minutes of treatment were identified. The statistical analysis of the results showed that the effects of photon collection area and the concentration of H2O2 agreed with the expected behavior, but suggests scavenging effects of hydroxyl radicals when the concentration of H2O2 is high. From the experimental data, a model of neural networks was obtained and used to predict the rate of TOC removal with time. Using the method HIPR (Holdback Input Randomization Method) and the weights associated with each input variable to the neural network, the relative importance of input variables was determined. Overall, the results proved the feasibility of photo-Fenton process catalyzed by tris(oxalate) ferrate III and irradiated by the sun, for the treatment of aqueous effluents containing 2,4-dichlorophenol.

O 2,4-diclorofenol (2,4-DCF), poluente modelo usado neste trabalho, é um composto empregado na produção de pesticidas, herbicidas e anti-sépticos, e comumente encontrado em efluentes industriais, sendo considerado um poluente prioritário devido a sua elevada toxicidade e alto poder de persistência no ambiente. Este trabalho objetivou estudar a degradação do 2,4-DCF em reator fotoquímico tubular com coletores parabólicos compostos (CPC) irradiado pelo Sol, operado em batelada com recirculação e admissão contínua de peróxido de hidrogênio à vazão de 11 mL min-1. A oxidação do poluente foi baseada no processo foto-Fenton catalisado por tris(oxalato) ferrato(III) (FeOx), investigando-se os efeitos das seguintes variáveis: concentração inicial de 2,4-DCF ([2,4-DCF] = 21-520 mgCL-1), concentrações de oxidante ([H2O2] = 9,9-130 Mm) e de fotocatalisador ([FeOx] = 0,2-1,8 mM) e área de coleta de fótons (0,196-0,98 m2), estudadas a partir de um planejamento Doehlert em múltiplos níveis. A foto-oxidação foi monitorada por medidas da concentração de carbono orgânico total (TOC) e oxigênio dissolvido (OD), tendo sido identificadas condições com remoção de TOC alcançando 99,8% em 90 minutos de tratamento. Nas análises estatísticas dos resultados, os efeitos da área de coleta de fótons e da concentração de H2O2 concordaram com o comportamento esperado, sugerindo, porém, efeitos sequestradores de radicais hidroxila quando a concentração de H2O2 é alta. A partir dos dados experimentais, obteve-se um modelo de redes neurais, usado para previsão da taxa de remoção de TOC em função do tempo. Através do método HIPR (Holdback Input Randomization Method) e dos pesos associados a cada variável de entrada da rede neural, avaliou-se a importância relativa das variáveis de entrada. De maneira geral, os resultados comprovaram a viabilidade do processo foto-Fenton catalisado por tris(oxalato) ferrato III e irradiado pelo Sol para o tratamento de efluentes aquosos contendo 2,4-diclorofenol. / 2,4-dichlorophenol (2,4-DCF), the model pollutant selected in this work, is a compound used in the production of pesticides, herbicides, antiseptics, and commonly found in industrial wastewaters. It is considered a priority pollutant due to its high toxicity and high persistence in the environment. This study investigated the degradation of 2,4-DCF in a tubular photochemical reactor with compound parabolic collectors (CPC) irradiated by the sun, operated in batch with recirculation and continuous feed of hydrogen peroxide at a flow rate of 11 mL min-1. The oxidation of the pollutant was based on the photo-Fenton process catalyzed by tris (oxalate) ferrate (III) (FeOx). The effects of the following variables were investigated: initial concentration of 2,4-DCF ([2,4-DCF] = 21-520 mgCL-1), concentrations of oxidant ([H2O2] = 9,9-130 Mm)); photocatalyst ([FeOx]= 0,2-1,8 mM)), and photon collection surface (0,196-0,98 m2), according to a Doehlert design on multiple levels. The photo-oxidation was monitored by measurements of the concentration of total organic carbon (TOC) and dissolved oxygen (OD). Conditions with TOC removal achieving 99.8% within 90 minutes of treatment were identified. The statistical analysis of the results showed that the effects of photon collection area and the concentration of H2O2 agreed with the expected behavior, but suggests scavenging effects of hydroxyl radicals when the concentration of H2O2 is high. From the experimental data, a model of neural networks was obtained and used to predict the rate of TOC removal with time. Using the method HIPR (Holdback Input Randomization Method) and the weights associated with each input variable to the neural network, the relative importance of input variables was determined. Overall, the results proved the feasibility of photo-Fenton process catalyzed by tris(oxalate) ferrate III and irradiated by the sun, for the treatment of aqueous effluents containing 2,4-dichlorophenol.

Conselho Nacional de Desenvolvimento Científico e Tecnológico / The so called amination water (AW), an effluent stream from the industrial production of the trifluraline herbicide, toxic and recalcitrant to conventional treatments such as the microbiological, was submitted to a combined advanced process. For the degradation study of this effluent a recent, promising alternative was chosen - the potassium ferrate (K2FeO4) oxidation-coagulation - that demonstrates high oxidant capability (from 2.2 up to 0.72 V) in a large pH range. In this study, a response surface methodology (RSM) design using pH and K2FeO4 concentration as independent variables, and the absorptiometric color reduction as a dependent one, was applied. The resultant regression equation for the quadratic model of the star design was nAbs (%) = 26.142 - 1.044A - 2.065A2 - 0.941B - 0.505B2 + 1.55AB. The second order results of the star design from the variance analysis (ANOVA) showed that the quadratic model is better than the lineal, and gave evidence that, a maximum of 29% absorptiometric color reduction occurs, when an initial pH of 7 and lower K2FeO4 concentration were used. By the same conditions, the reduction of the chemical oxygen demand (COD) was 49.5%. As the reduction of potassium ferrate generate ferric iron salts in aqueous solution, there is a great potential for its combined use as Fe(III) source for Fenton reagent, generating hydroxyl radicals (HO·) by addition of H2O2 to the reaction system. A new design based on RSM was applied, evaluating the potentiality of the oxidationcoagulation-Fenton like process having the pH, Fe(VI) -> Fe(III) and H2O2 as independent variables, and absorptiometric color reduction efficiency as evaluation responses for the dependent variable. The resulting regression equation for the quadratic model was nAbs (%) = 36.9 - 21.58A + 8.37A2 + 1.36B + 0.92B2 + 1.08C + 1.52C2 + 1.27AB - 1.34AC + 1.33BC. The ANOVA results evidenced that the maximum absorptiometric color reduction occurs by pH 3, and by correspondent 10 g L-1 de Fe(VI) and 20 g L-1 hydrogen peroxide. The absorptiometric color and COD reduction were 96% and 57%, respectively. An efficiency increase of the HO· radical generation was achieved when the previous process was combined to UV irradiation, carrying out the so called oxidation-coagulationphoto-Fenton like process. As before, a RSM was applied, where the pH, Fe(VI) -> Fe(III) concentration, H2O2 concentration and temperature were evaluated as independent variables. The efficiency of the absorptiometric color reduction was chosen as dependent variable. The resulting regression equation for the quadratic model was nAbs (%) = 38.3 - 20.2A + 8.12A2 - 0.27B + 3.73B2 + 0.3C + 3.6C2 + 1.67D + 3.1D2 + 1.72AB + 0.51AC - 1.82AD + 0.74BC - 1.11BD + 0.03CD. The ANOVA results evidenced that the maximum absorptiometric color reduction occurs by pH 3 and by 10 g L-1 de Fe(VI) and 20 g L-1 hydrogen peroxide amounts, at 60 °C. The maxima efficiencies achieved for the effluent stream treatment by the oxidationcoagulation-photo-Fenton process were 95% and 85%, for absorptiometric color and COD reduction, respectively. The high efficiency of the combined process as an oxidative-coagulant-oxidative pretreatment for posterior conventional process (e.g., microbiological treatment) can be looked out as an interesting and advantageous alternative for the AW treatment, as well as other recalcitrant streams. / A produção industrial do herbicida trifluralina gera a corrente efluente água de aminação (AA), tóxica e recalcitrante a processos convencionais, tais como o tratamento microbiológico. Para o estudo da degradação deste efluente escolheu-se uma recente e promissora alternativa para a degradação de biorecalcitrantes - a oxidação-coagulação com Ferrato de Potássio (K2FeO4) - que demonstra alto poder oxidante, de 2,2 a 0,72 V, em toda faixa de pH. Foi aplicado, também, neste estudo, a metodologia de planejamento com superfície de resposta (RSM), tendo-se como variáveis independentes, o pH e a concentração de K2FeO4; e como variável dependente, a remoção de cor. A equação de regressão resultante do planejamento estrela, para o modelo quadrático, é nAbs (%)= 26,142 - 1,044A - 2,065A2 - 0,941B - 0,505B2 + 1,55AB. Os resultados do modelo de superfície de resposta de segunda ordem, na forma de análise de variância (ANOVA), demonstraram que o modelo quadrático é superior ao modelo linear e evidenciaram que a máxima redução da cor absorciométrica (29%) ocorre com pH inicial 7 e com a menor concentração de ferrato de potássio utilizada. Da mesma forma, obteve-se redução da demanda química de oxigênio (DQO) de 49,5%. Como a redução do ferrato de potássio produz sais de Fe(III) em solução, aproveitouse a grande potencialidade para a combinação como fonte de íons férricos para o reagente Fenton, adicionando H2O2 para a geração de radicais hidroxil (HO·). Foi, então, aplicado um novo planejamento baseado em RSM, avaliando a potencialidade do processo de oxidaçãocoagulação-Fenton com as variáveis independentes pH, concentração de Fe(VI) -> Fe(III) e H2O2, e, como variável dependente para a avaliação dos resultados, a eficiência da redução da cor absorciométrica. A equação de regressão resultante deste planejamento, para o modelo quadrático, que se mostrou superior ao modelo linear é nAbs (%) = 36,9 - 21,58A + 8,37A2 + 1,36B + 0,92B2 + 1,08C + 1,52C2 + 1,27AB - 1,34AC + 1,33BC. Os resultados da análise ANOVA evidenciaram que a máxima redução da cor absorciométrica ocorre quando o pH inicial foi 3, usando-se 10g L-1 de Fe(VI) e 20 g L-1 peróxido de hidrogênio. Assim, se obteve ainda redução da cor absorciométrica da ordem de 96% e, da DQO, de 57%. Por ainda acreditar num aumento da eficiência da geração de radicais HO·, o processo anterior foi combinado à radiação UV, realizando-se então o chamado processo oxidaçãocoagulação-foto-Fenton e, como nos anteriores, através de RSM avaliaram-se as variáveis independentes, pH, concentração de Fe(VI) -> Fe(III), H2O2 e temperatura. Como variável dependente foi escolhida a redução da cor absorciométrica. A equação resultante para este planejamento, para o modelo quadrático, superior ao linear, é nAbs (%) = 38,3 - 20,2A + 8,12A2 - 0,27B + 3,73B2 + 0,3C + 3,6C2 + 1,67D + 3,1D2 + 1,72AB + 0,51AC - 1,82AD + 0,74BC - 1,11BD + 0,03CD. Os resultados da ANOVA demonstraram que a máxima redução ocorreu quando o pH inicial foi 3, usando-se 10 g L-1 de Fe(VI), 20 g L-1 peróxido de hidrogênio e temperatura de 60 °C. A máxima eficiência alcançada no tratamento do efluente AA pelo processo de oxidação-coagulação-foto-Fenton foi de 95% e, de 85%, para a cor absorciométrica e a DQO, respectivamente. A eficiência dos processos estudados, como pré-tratamento oxidativo-coagulanteoxidativo para posterior tratamento convencional (ex. microbiológico), pode ser interessante alternativa para o tratamento da corrente recalcitrante AA.

In this study, we aim to employ Chlorella vulgaris algae for removal of nutrients in wastewater and collect the produced algae by ferrate after treatment. The growth of algae was conducted in F/2 synthetic medium and in actual domestic wastewater. The removals of nitrogen and phosphorous by algae were then investigated for low and high nutrient concentrations using wastewater after biological treatment in both batch and continuous experiments. Results showed that specific growth rates in the exponential phase were 0.23 and 0.35 day-1 for F/2 medium and domestic wastewater, respectively, proving the suitability of wastewater for algae growth. The removal efficiency of ammonia, nitrate, and phosphate were 89 - 93, 64 - 76, and 69 – 88%, respectively. In the algae collection test, pH 8 is the optimal pH to remove algae and ferrate had higher algae removal ability than alum under each optimal condition with removal efficiency of 84 - 97% at dosage of 12 mg Fe/L. These results suggest that microalgae is a potential alternative for removing of nutrients in wastewater treatment due to the high uptake capacity of nitrogen and phosphorous and the effective collection of algae after treatment by ferrate. / Mục tiêu của nghiên cứu này là đánh giá hiệu quả xử lý chất dinh dưỡng bằng tảo Chlorella vulgaris trong môi trường nước thải sinh hoạt, thông qua khả năng xử lý N và P từ nguồn nước khi tảo tăng trưởng và khả năng keo tụ để thu hồi tảo bằng ferrate. Tốc độ tăng trưởng đặc thù µ trong môi trường F/2 và nước thải sinh hoạt lần lượt là 0,23 ngày-1 và 0,35 ngày-1. Hiệu suất xử lý ammoni, nitrát và phốt phát- lần lượt đạt 89 - 93%, 64 - 76% và 69 - 88%. Kết quả thí nghiệm keo tụ thu hồi tảo cho thấy pH = 8 là thích hợp nhất để loại bỏ tảo bằng ferrate và việc sử dụng ferrate cho hiệu quả tách tảo tốt hơn phèn nhôm với lượng sử dụng ít hơn. Ở hàm lượng 12 mgFe/L, hiệu quả tách tảo đạt cao nhất từ 84 - 97%. Nghiên cứu cho thấy tiềm năng thay thế công nghệ sinh học truyền thống bằng công nghệ vi tảo trong loại bỏ các chất dinh dưỡng và khả năng thu hồi tảo hiệu quả bằng cách sử dụng ferrate.