This thesis reports the results of an experimental investigation of the effect of buoyancy forces on turbulent upflow of air in a vertical pipe under condit.io:1s of constant wall heat flux at Reynolds numbers of 5000 to 14000. Preliminary velocity and axial turbulence intensity measurements in isothermal flows for Reynolds numbers of 5000 to 32000 are also reported. Velocity and temperature distributions were measured over a range of heat fluxes at Reynolds numbers of approximately 5000 and 8000 and at a single heat flux at a Reynolds number of 14000.

Um sistema de tratamento de esgotos sanitários por reator UASB seguido de RSB foi operado durante 27 semanas objetivando-se verificar seu desempenho na remoção de matéria orgânica carbônácea, nitrogênio amoniacal e fósforo. O reator UASB, com 145 litros foi operado com tempo de detenção hidráulica médio de 8 horas durante todo o experimento. O reator RSB, de 106 litros, foi operado com tempos de ciclio de 24, 12 e 6 horas, em seis etapas distintas. Os tempos de ciclo no RSB variaram de 24, 12, 6, 24, 24 e 24 para as etapas 1, 2, 3, 4, 5 e 6 respectivamente. As últimas três etapas foram caracterizadas pela adição de cloreto férrico, respectivamente. O sistema UASB/RSB apresentou eficiência média de remoção de DQO de 92% nas etapas 1, 2, 5, e 6. Na etapa 3, com tempo de ciclo de 6 horas e na etapa 4, quando o RSB recebeu 50mg/L de cloreto férrico, a eficiência média de remoção de DQO foi de 83%. Quando FeCl3 foi adicionado no afluente do reator UASB, obteve-se excelente desempenho na remoção de fósforo, cujas concentrações afluentes foram da ordem de 1,0 mg/L e 4,0 mg/L para a adição de 200 e 100 mg/L de cloreto férrico, respectivamente. / The performance evaluation of a pilot scale system composed by a UASB reactor followed by an SBR reactor treating domestic sewage is presented. The UASB reactor with 145 L of volume, was operated with an average hydraulic detention period of 8 hours. The SBR reactor had a volume of 106 litters being operated with cyclic time of 24, 12, and 6 hours. These different operational conditions characterized three stages of the work from a total of six stages. The three last stages were the addition of 50, 100, and 200 mg/L of ferric chloride in the aerobic, anaerobic, and anaerobic reactors, respectively. In the three last stages the UASB reactor worked with an average hydraulic detention time of 8 hours and the aerobic reactor worked with a cyclic time of 24 hours. After 27 weeks of continuous operation, the UASB/SBR system produced very good results in terms of COD removal. The average results of the COD removal were of 92% to the first, second, fifth, and sixth stages, having a reduction to 83% on the average to the stages where the SBR received 50 mg/L of ferric chloride and in which it worked with a six hour cyclic period. The addition of ferric chloride in the anaerobic reactor had great efficiency in the phosphorus removal. It was reached absolute values of until 1 mg/L of PO43- in the aflluent of the system, promoting an average effluent of 4 mg/L of PO43- with the addition 200 mg/L of FCl3. The UASB/SBR system can become a very promising alternative for domestic sewerage treatment in Brazil, since the system can be projected with shorter times of hydraulic detention times, resulting in a low cost compact installation. Furthermore, excellent results can be gotten to the ammonium and phosphorus removal.

Orientador: Gustavo Henrique Ribeiro da Silva / Banca: Adriana Ferreira Maluf Braga / Resumo: O presente trabalho estudou o tratamento de água negra simulada (esgoto sanitário com efluente suíno) em reator UASB (upflow anaerobic sludge blancket) mesofílico com DQO (demanda química de oxigênio) de entrada de 958,489 ± 489,2mg/L, TDH (tempo de detenção hidráulica) de 1,8 ± 1,8d e COV (carga organica volumética) de 0,2 ± 0,2 KgDQO/m³d. Avaliou-se, ainda, o efeito da suplementação de metais traço (Fe, Ni, Mn, Ba, Se, Co e Cu) na AME (atividade metanogênica específica) do lodo utilizado como inóculo de partida do reator. Os ensaios com metais traços foram realizados em batelada utilizando o método Plackett Burman para planejar o experimento. Foram observados valores nitrogênio amoniacal junto a entrada do reator de 197,37 ± 84,3mgN/L, no entanto, o reator apresentou remoção de DQO de 61,7% ± 15%, demonstrando que não houve efeito tóxico da amônia no meio. Os resultados dos ensaios em batelada demonstraram efeito negativo estatisticamente significativo das adições de Ba e Mn sobre a AME, contudo, o Cu e o Se apresentaram efeito positivo, embora não estatisticamente significativos, indicando que a produção de metano a partir de água negra pode ser estimulada com a suplementação desses metais / Abstract: The present work studied the treatment of simulated black water (sewage with swine wastewater) in UASB (upflow anaerobic sludge blancket) mesophilic with input COD (chemical oxygen demand) of 958,489 = 489,2 mg/L, HRT (hydraulic retention time) of 1.8 = 1.8d and OLR (organic loading rate) 0.2 = 0.2 kgCOD/m3d. This work also evaluated the effect of supplementation of trace metals (Fe, Ni, Mn, Ba, Se, Co e Cu) on the SMA (specific methanogenic activity) of the seed inoculum used in reactor startup. The trace metals assay were performed in batch mode using Plackett-Burman method of the design of the experiments. Despite of the ammonia nitrogen values at the inffluet of 197.37 = 84.3 mgN/L reactor, the reactor, the reactor presented COD removal of 61.7% = 15%, indicating no toxic effectof the addition of Ba and Mn on the SMA of the sludge, however, Cu and Se presented positive effect, although not statistically significant, which indicates that the methane production from black water might be enhanceb by the supplementation of these metals / Mestre

Upflow anaerobic sludge blanket (UASB) has been developed and commercially used for treating high-strength wastewaters since 1971, and this process has been extended its applications to domestic sewage treatment since 1980. Constructed wetlands (CWL) can be treated as a simulated natural treatment system, which use solar radiation as the energy source. By analogy with some removal mechanisms in natural wetlands, constructed wetlands are able to transform and remove pollutants from the wastewater. Other features provided by the constructed wetland include wildlife support, hydrologic modification, erosion protection, and open space and aesthetics. It has been applied for domestic wastewater purification for decades. In this study, a pilot UASB-CWL reactor was built to test its performance for the removal of organic compounds and nutrients from a sewage with COD (Chemical oxygen demand) in the range of 200-300 mg/L. The UASB reactor has an active liquid volume of 2.5 liters and the CWL includes a 56 L-tank (CWL-1) followed by a 80 L-tank (CWL-2) in which emergent macrophyte (reed, Phragmites australis L.) and floating macrophyte (Pistia stratiotes L.), respectively, were planted. Effects of hydraulic retention time (HRT) of the test water in the reactor on the pollutant removal were tested. Results indicate that with HRTs of 2, 45, and 64 hours, respectively, in the UASB, CWL-1, and CWL-2, the system exhibited good performances for pollutant removal. The HRTs are equivalent to CODt loadings of 3.06, 0.065, and 0.026 kg/m3.day to the UASB, CWL-1, and CWL-2, respectively. Average total COD (CODt), soluble COD (CODs), total biochemical oxygen demand (BODt), suspended solids (SS), and phosphate-P (P) were 121, 62, 33, 130, and 0.10 mg/L, respectively, and these are equivalent to the removal efficacies of 45-65, 60-80,65-75, 30-50, and 85-95%, respectively. With the operation conditions, average CODt = 28 , CODs = 18, BODt = 10, SS = 18, NH3-N = 18, NO2--N =12 , NO3--N = 1.0, and P = 0.0 mg/L were obtained, and the average removal total efficacies were 93, 93, 90, 75, 72, 95, and 100%, respectively. Effluent qualities are far superior to the national effluent standards for domestic sewage in Taiwan.

Resumo: Neste trabalho, avaliou-se o desempenho de dois conjuntos de reatores anaeróbios de fluxo ascendente com manta de lodo (UASB) instalados em série. Os conjuntos I e II foram compostos por dois reatores UASB em série com volumes de 908 e 350 L e de 908 e 188 L, respectivamente. No conjunto II, foi realizado pós tratamento do efluente em um reator operado em batelada sequencial (RBS) com alimentação contínua, com volume de 3000 L. As cargas orgânicas volumétricas (COV) aplicadas no primeiro reator dos conjuntos I e II variaram de 8,36 a 9,95 g DQO total (L d)-1 e de 7,83 a 11,6 g DQO total (L d)-1, respectivamente. As eficiências médias de remoção de DQO total, nitrogênio Kjeldahl (NK), fósforo total (P-total) foram elevadas e variaram de 95 a 97% e de 97 a 99%, de 64 a 67% e de 53 a 69%, e de 80 a 87% e de 71 a 89%, nos conjuntos I e II, respectivamente. As eficiências também foram elevadas para os metais Cu, Fe, Mn e Zn nos conjuntos I e II e variaram de 80 a 91% e de 63 a 80%, de 59 a 80% e de 51 a 72%, de 89 a 97% e de 81 a 98%, de 87 a 98% e de 91 a 99%, respectivamente. A semeadura do milho, foi realizado na safrinha 2009 e na safra 2009/2010, ambos em sistema de plantio direto e convencional, para a confecção de silagem. Os valores do pH observados na silagem não diferiram entre os tratamentos nas duas épocas de semeadura e nos sitemas de plantio direto e convencional. A adubação do milho para confecção de silagem, com dejeto bruto de suíno, com dejetos tratados em sistema anaeróbio e com os dejetos do pós-tratamento, quando comparada com a adubação mineral, não apresentou resultados com diferenças significativas na silagem na maioria dos atributos avaliados. O reúso dos efluentes tratados e do dejeto bruto, não promoveu alterações na nutrição da planta, porém os efluentes... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: This study evaluated the performance of two sets of upflow anaerobic sludge blanket (UASB) installed in series. The sets I and II were composed of two UASB reactors in series with volumes of 908 and 350 L and 908 L and 188, respectively. In set II was conducted after the treatment of the effluent in a sequencing batch reactor (SBR) with continuous flux, with a volume of 3000 L. The volumetric organic loads (VOC) applied to the first reactor of the sets I and II ranged from 8.36 to 9.95 g total COD (L d) -1 and 7.83 to 11.6 g total COD (L d) -1, respectively. The average removal efficiencies of total COD, Kjeldahl nitrogen (NK), total phosphorus (total P) were high and ranged from 95 to 97% and 97 to 99%, 64 to 67% and 53 to 69% and 80 to 87% and 71 to 89% in clusters I and II, respectively. The efficiencies were also high for Cu, Fe, Mn and Zn in clusters I and II, and ranged from 80 to 91% and 63 to 80%, 59 to 80% and 51 to 72% from 89 to 97 and 81% to 98%, 87 to 98% and 91 to 99%, respectively. We carried out the swing of maize in the 2009 off-season and 2009/2010 season, both in no-tillage and conventional sistems for making silage. The pH values observed in the silage did not differ significantly between treatments in both seasons of swing and in the sistems of tillage and no-tillage. The fertilization of corn for silage making, with raw pig manure with waste treated in anaerobic system and the waste from the post-treatment compared to mineral fertilization, results showed no significant difference in silage in most parameters. The reuse of treated effluent and raw manure, did not promote changes in the nutrition of corn but the application of treated effluent increased concentrations of N readily available for plants in soil. Even in much larger doses applied compared to the raw waste, the treated effluent, promoted soil contamination by total coliforms and thermotolerant lower / Orientador: Roberto Alves de Oliveira / Coorientador: Ana Cláudia Ruggieri / Banca: Wanderley José de Melo / Banca: Israel Jose da Silva / Mestre

Um sistema de tratamento de esgotos sanitários por reator UASB seguido de RSB foi operado durante 27 semanas objetivando-se verificar seu desempenho na remoção de matéria orgânica carbônácea, nitrogênio amoniacal e fósforo. O reator UASB, com 145 litros foi operado com tempo de detenção hidráulica médio de 8 horas durante todo o experimento. O reator RSB, de 106 litros, foi operado com tempos de ciclio de 24, 12 e 6 horas, em seis etapas distintas. Os tempos de ciclo no RSB variaram de 24, 12, 6, 24, 24 e 24 para as etapas 1, 2, 3, 4, 5 e 6 respectivamente. As últimas três etapas foram caracterizadas pela adição de cloreto férrico, respectivamente. O sistema UASB/RSB apresentou eficiência média de remoção de DQO de 92% nas etapas 1, 2, 5, e 6. Na etapa 3, com tempo de ciclo de 6 horas e na etapa 4, quando o RSB recebeu 50mg/L de cloreto férrico, a eficiência média de remoção de DQO foi de 83%. Quando FeCl3 foi adicionado no afluente do reator UASB, obteve-se excelente desempenho na remoção de fósforo, cujas concentrações afluentes foram da ordem de 1,0 mg/L e 4,0 mg/L para a adição de 200 e 100 mg/L de cloreto férrico, respectivamente. / The performance evaluation of a pilot scale system composed by a UASB reactor followed by an SBR reactor treating domestic sewage is presented. The UASB reactor with 145 L of volume, was operated with an average hydraulic detention period of 8 hours. The SBR reactor had a volume of 106 litters being operated with cyclic time of 24, 12, and 6 hours. These different operational conditions characterized three stages of the work from a total of six stages. The three last stages were the addition of 50, 100, and 200 mg/L of ferric chloride in the aerobic, anaerobic, and anaerobic reactors, respectively. In the three last stages the UASB reactor worked with an average hydraulic detention time of 8 hours and the aerobic reactor worked with a cyclic time of 24 hours. After 27 weeks of continuous operation, the UASB/SBR system produced very good results in terms of COD removal. The average results of the COD removal were of 92% to the first, second, fifth, and sixth stages, having a reduction to 83% on the average to the stages where the SBR received 50 mg/L of ferric chloride and in which it worked with a six hour cyclic period. The addition of ferric chloride in the anaerobic reactor had great efficiency in the phosphorus removal. It was reached absolute values of until 1 mg/L of PO43- in the aflluent of the system, promoting an average effluent of 4 mg/L of PO43- with the addition 200 mg/L of FCl3. The UASB/SBR system can become a very promising alternative for domestic sewerage treatment in Brazil, since the system can be projected with shorter times of hydraulic detention times, resulting in a low cost compact installation. Furthermore, excellent results can be gotten to the ammonium and phosphorus removal.

Stormwater runoff is a known pollutant source capable of causing surface water degradation, especially in highly populated areas such as Central Florida. Wet detention ponds manage this stormwater, but most of the ponds do not remove enough nutrients, specifically nitrogen and phosphorus, to meet TMDL regulations. This research provides a possible addition to a detention pond in Seminole County, Florida using a Chamber Upflow Filter and Skimmer (CUFS), which can increase the removal of phosphorus and nitrogen by the system. Water enters the system through the skimmer, which floats on the surface of the detention pond. It travels from the skimmer to the bottom of the chamber, where heavier particles settle out before entering the upflow filter. The upflow filter contains twenty-four inches of Black and GoldTM media to remove nitrogen and phosphorus under anoxic conditions. Water flows up through the filter and out of the system, and eventually travels to Lake Jesup, a eutrophic lake. A total of twenty-eight storm events and seven baseflows were sampled from the site in Seminole County, and ten storm events were sampled from a pilot study CUFS. The results of this research show significant reductions by the Seminole County CUFS in turbidity, orthophosphorus, total phosphorus, and total suspended solids when the means were compared at a 95% confidence interval. Reductions also occurred for total nitrogen, but could not be proved by the mean comparison. The pilot scale application of the CUFS significantly reduced total nitrogen at a 95% confidence interval.

Finding solutions to treat water that contains phosphorus is an important effort due to the harmful impacts it presents to both human health and the environment. Phosphorus is considered a limiting factor in water oftentimes and therefore controls the growth of algal bloom in a water body. The increase of algal populations due to wastewater effluent, stormwater runoff, and agricultural discharge in Florida waters has a direct link to the event of harmful algal blooms such as red tide in coastal regions, eutrophication of waterbodies, and fish kills. Finding low cost, energy efficient, and low maintenance green sorption media (GSM) solutions to treat this nutrient laden water is the main focus of this study. Phosphorus removal efficiencies were studied in the lab using two new recipes of GSM with a unique passive energy cascade upflow hydraulic pattern. The study goal involves comparing the two new GSM in terms of their removal rates of phosphate, removal mechanisms, and their differential effects via a lab-scale cascade upflow filtration system. The two new GSM include Zero-valent Iron and Perlite Green Environmental Media (ZIPGEM) composed of 85% sand, 5% clay, 5% zero valent iron, 5% perlite by volume, and Biochar zero-valent Iron and Perlite Green Environmental Media (BIPGEM) containing 80% sand, 5% biochar, 5% clay, 5% perlite, 5% zero valent iron by volume. To explore the differential effects, the addition of biochar in BIPGEM is of special interest in this study and expected to perform better than its counterpart ZIPGEM. However, results indicated that ZIPGEM, when treated as a system, experienced an average percent removal of 97.9% to 78.9% from the initial 60 minute to 300 minute range, whereas BIPGEM experienced an average 86.9% to 67.1% removal during this same time interval. At the time of shutdown for the BIPGEM runs (5,480 minutes), BIPGEM as a system possessed an average 52.8%, and ZIPGEM a 69.1% removal. Upon shutdown for ZIPGEM at the 11,180 minutes mark, ZIPGEM was still at an average 69.1% removal. Further comparison of the material structures and compositions of two GSM involve chemical composition analysis of both raw and spent media samples using X-ray diffraction (XRD), X-ray fluorescence spectrometry (XRF), and scanning electron microscope (SEM) in the Advanced Materials Processing and Analysis Center (AMPAC) at the University of Central Florida (UCF). Two experimental runs with slightly different flow rates were conducted for each type of sorption medium to explore the process reliability of such a cascade upflow filtration system. Statistical analysis via the Kolmogorov-Smirnov Test was calculated to compare goodness of fit between the two sets of performance curves associated with each sorption media and experimental run in the lab. The results of this study show that the inclusion of biochar did not support the hypothesis that BIPGEM would perform better than ZIPGEM in phosphorus removal and the two experimental runs are statistically similar. This may be due to the quantity of biochar not being enough to provide significant adsorption of phosphorus or the possibility of synergistic effects between the other ingredients within the recipe altering the physical or chemical adsorption properties expected from the medium. It is recommended that further studies in a downflow filtration system (i.e., a column study) take place to continue the understanding of the impact associated with changing hydraulic pattern and higher content of biochar in the recipe based on this study.

The performance of Biostyr media using an upflow biofilter scheme was investigated. Three separate Modes that varied the organic strength of the influent at an operating temperature of 12°C were evaluated to explore how nitrification and denitrification responded to different concentrations of biodegradable COD loadings. Effluent suspended solids and observed yield coefficients were also analyzed along with ways to determine kinetic parameters. Combinations of primary and secondary effluent were used as the influent using domestic wastewater from a primary sewer of the Blacksburg VPI Sanitation Authority, in Blacksburg, Virginia, located on the Virginia Tech Campus. Reduction of nitrification rates was observed in the aerobic reactor at biodegradable COD loadings to the aerobic biofilter of 2.2 kg/day/m³ and greater. Complete denitrification was observed in the anoxic tank at biodegradable COD loadings to the anoxic biofilter of 7.5 kg/day/m³ and greater. Above this value, denitrification was limited by the available amount of NOx-N. An optimum influent biodegradable COD loading rate of 7.5 kg/day/m³ to the anoxic biofilter and 2.2 kg/day/m³ to the aerobic biofilter was experimentally demonstrated for simultaneous nitrification and denitrification. When secondary effluent was fed to the reactors, complete nitrification was observed indicating success in terms of ammonia polishing. However, low concentrations of influent biodegradable COD limited the ability for the anoxic reactor to denitrify. Thus, it was determined that the addition of an external carbon source would need to be implemented to denitrify when using low strength secondary effluent. Effluent total suspended solids were consistently below 20 mg/L indicating that secondary clarification would not be needed for this reactor scheme. The combined observed yield was consistently around 0.2 (mg-VSS/mg-COD consumed) or less. However, the evaluation of separate anoxic and aerobic yields was inconsistent when averaging data over 5 days, because of the recycle of biomass from the aerobic reactor back to the anoxic reactor. The kinetic parameters, q'_{MAX NIT} and K’_{S NIT} were calculated by plotting nitrification rates as a function of ammonia concentration. These coefficients were calculated by using separate segments within the media under less than optimal conditions. It was determined that analyzing the whole system rather than segments would be better suited for kinetic parameter analysis. However, there was not enough collected data to achieve this. Thus, ranges identifying further influent biodegradable COD loadings were recommended for later research to better a complete kinetic parameter evaluation. / Master of Science

Este trabalho consiste na avaliação do desempenho do sistema composto por filtração direta ascendente em pedregulho seguida de filtração rápida descendente. A investigação experimental foi realizada em instalação piloto, variando-se a taxa de filtração nos filtros ascendente (60 a 240 m3/m2.d) e descendente (100 a 240 m3/m2.d) com e sem execução de descargas de fundo intermediárias no ascendente. A água bruta foi proveniente de um manancial superficial com pH entre 6,5 a 7,5, valores de turbidez compreendidos entre 8,5 e 21 uT, cor verdadeira entre 19 e 50 unidades Hanzen e alcalinidade entre 6,7 e 10 mg CaCO3/L. Adicionalmente, verificou-se o comportamento da instalação, para uma condição simulada de ocorrência de pico de turbidez de 380 uT, com curta duração. Em função dos resultados obtidos foi possível realizar considerações com relação à ocorrência de desprendimento/carreamento contínuo de flocos no filtro ascendente e constatar as influências da execução de descargas de fundo intermediárias, com produção de efluentes de melhor qualidade e carreiras de filtração mais prolongadas. O filtro ascendente com taxa de filtração de 120 m3/m2.d, foi capaz de absorver o pico de turbidez de 380 uT, produzindo efluente pré-filtrado com valores de turbidez entre 0,3 e 5,7 uT, e o filtro descendente, com taxa de filtração de 180 m3/m2.d, produziu efluente final com valores consistentemente menores que 1 uT, e na maior parte do tempo menores que 0,1 uT. / This work consists of a performance evaluation of the system composed by direct upflow filtration in gravel followed by rapid downflow filtration in sand. The experimental investigation was accomplished in pilot installation, being varied the filtration rate in the upflow (60 to 240 m3/m2.d.) and downflow (100 to 240 m3/m2.d.) filters with and without execution of intermediate downflushes in the first one. The raw water was provided from a superficial source with pH between 6,5 to 7,5, turbidity values between 8,5 and 21 UT, visible colour between 19 and 50 uH and alkalinity between 6,7 and 10 mg CaC03/L. ln addition, the behavior of the installation was verified, for a simulate condition of turbidity peak occurrence of 380 uT, with short duration. From the obtained results it was possible to accomplish considerations concerning the detachment occurrence/continuous flowing of flakes in the upflow filter and to verify the influences of the execution of intermediate downflushes, with production of better quality effluents and longer filter runs. The upflow filter with filtration rate of 120 m3/m2.d. was able to attenuate the turbidity peak of 380 uT, producing prefiltered effluent with turbidity values between 0,3 and 5,7 uT, and the downflow filter, with filtration rate of 180 m3/m2.d. produced final effluent with consistently values lower than 1 uT and in most of time lower than 0,1 uT.