Dissertation submitted in fulfilment with the requirements for the Masters Degree: Biotechnology, Durban University of Technology, 2011. / Untreated vegetable oil effluents (VOEs) are known for creating shock-loading problems for the receiving wastewater treatment installations, resulting in poor quality final effluents being produced which do not satisfy municipal discharge standards. Onsite activated sludge treatment as an alternative has not been fully investigated. Hence, in this investigation biological treatment using the activated sludge process was chosen as the method for the treatment of VOE. The effect of VOE on measured process parameters was also determined. Novel molecular techniques such as fluorescent in situ hybridisation (FISH) and dot-blot hybridization have become powerful tools for the analysis of complex microbial communities that exist within activated sludge. The aim of this investigation was to evaluate biological treatment, optimize and apply FISH and dot-blot hybridization in order to analyze the microbial community implicated the biological treatment of VOE using probes EUBmix, ALF1b, BET42a, GAM42a and HGC69a. A laboratory-scale modified Ludzack-Ettinger (MLE) process setup and fed VOE with a COD (chemical oxygen demand) of ± 1000 mg/L. Daily monitoring of the process involved COD and TKN (total kjeldahl nitrogen) analysis of the influent and effluent as well as direct OUR (oxygen utilization rate) measurement and monitoring of the MLVSS (mixed liquor volatile suspended solids) concentration of the aerobic mixed liquor. The process exhibited overall COD and TKN removal capacities of 84% and 90% respectively. The aerobic mixed liquor had an OUR of 19 mgO/L.h and an average MLVSS concentration of 3000 mg/L. FISH results revealed that 72% of cells stained with 4‟, 6-diamidino-2-phenylindole (DAPI) within the aerobic mixed liquor bound to probe EUBmix, indicating a substantial Bacterial population within the laboratory-scale biological process. The alpha-Proteobacteria was identified as the dominant bacterial community comprising 31% of Bacterial cells, followed by the beta-Proteobacteria (17% of EUBmix), gamma-Proteobacteria (8% of EUBmix) and Actinobacteria (4% of EUBmix). Results of dot-blot hybridization were in agreement with FISH Adrian Phillip Degenaar| CHAPTER 1: General Introduction - v - results reiterating dominance of the alpha-Proteobacteria. This indicated that the class alpha-Proteobacteria could play a primary role in the biological degradation of VOE. This research will therefore aid in process design and retrofitting of biological processes treating VOE.

The influent to a water and resource recovery facility (WRRF) generally exhibits significant diurnal variations in flow rate and load concentration. This makes determining the operating parameters and subsequently the overall operation of plants difficult, especially in developing countries due to the lack of highly skilled operators. Hence, there is an incentive for the control and operation of WRRFs in developing countries to be improved. Flow equalization tanks were identified as a potential method to attenuate the diurnal variations in flow rate and load concentration into plants. The main aim of this research was to develop a viable dynamic simulation model for the operation of flow equalization tanks, within a plant-wide framework (to allow for the evaluation of design and control strategies). The next aim was to determine the benefits of equalization tanks towards design and optimised operation of future WRRFs via scenario analyses. Finally, the effects of the equalization tank on the performance of various unit processes in a WRRF were to be investigated. The model was developed in three steps; i) the development of the required equations to model equalization tanks, ii) mass balance throughout the model for internal consistency and iii) scenario analyses to determine if the model generated reasonable and scientifically sound outputs. The model was developed using Microsoft Excel Visual Basic (VBA) and WEST®. Two scenarios were considered to assess the equalization tank modelled. Scenario One involved the comparison of the capital cost, unit process sizes and total footprint of a balanced sludge age Modified Ludzack-Ettinger (MLE) system with and without an equalization tank. Scenario Two compared the plant performance of the MLE system designed in Scenario One with and without a flow equalization tank. A dynamic simulation model replicating equalization tanks was successfully developed. From scenario analyses, it was determined that using an MLE system and only considering equalization of flow, there was a reduction in the size of several unit processes by 8-9% (primary settling tank, biological reactors, secondary settling tank, flotation unit, anoxic-aerobic digester), due to the less conservative design values that could be used as the variations of the influent were decreased. Despite this, a 13% overall increase in the footprint of the WRRF was observed due to the addition of the equalization tank. The attenuation of diurnal flow variations also resulted in reduction of various plant parameters by up to 50% (flow, OUR, VSS flux). Finally, there was a 10% improvement in the performance of various unit processes due to the presence of the equalization tank. In conclusion, the inclusion of equalization tanks in WRRFs has significant positive effects. These results were obtained with equalization of flow only. Some other limitations were experienced during the project resulting in the following recommendations: further research will be needed to validate and calibrate the model, As the model was not successfully incorporated in a plant-wide framework, further developments in that direction are required, as well as including the equalization of load in the model.

The influent to a water and resource recovery facility (WRRF) generally exhibits significant diurnal variations in flow rate and load concentration. This makes determining the operating parameters and subsequently the overall operation of plants difficult, especially in developing countries due to the lack of highly skilled operators. Hence, there is an incentive for the control and operation of WRRFs in developing countries to be improved. Flow equalization tanks were identified as a potential method to attenuate the diurnal variations in flow rate and load concentration into plants. The main aim of this research was to develop a viable dynamic simulation model for the operation of flow equalization tanks, within a plant-wide framework (to allow for the evaluation of design and control strategies). The next aim was to determine the benefits of equalization tanks towards design and optimised operation of future WRRFs via scenario analyses. Finally, the effects of the equalization tank on the performance of various unit processes in a WRRF were to be investigated. The model was developed in three steps; i) the development of the required equations to model equalization tanks, ii) mass balance throughout the model for internal consistency and iii) scenario analyses to determine if the model generated reasonable and scientifically sound outputs. The model was developed using Microsoft Excel Visual Basic (VBA) and WEST®. Two scenarios were considered to assess the equalization tank modelled. Scenario One involved the comparison of the capital cost, unit process sizes and total footprint of a balanced sludge age Modified Ludzack-Ettinger (MLE) system with and without an equalization tank. Scenario Two compared the plant performance of the MLE system designed in Scenario One with and without a flow equalization tank. A dynamic simulation model replicating equalization tanks was successfully developed. From scenario analyses, it was determined that using an MLE system and only considering equalization of flow, there was a reduction in the size of several unit processes by 8-9% (primary settling tank, biological reactors, secondary settling tank, flotation unit, anoxic-aerobic digester), due to the less conservative design values that could be used as the variations of the influent were decreased. Despite this, a 13% overall increase in the footprint of the WRRF was observed due to the addition of the equalization tank. The attenuation of diurnal flow variations also resulted in reduction of various plant parameters by up to 50% (flow, OUR, VSS flux). Finally, there was a 10% improvement in the performance of various unit processes due to the presence of the equalization tank. In conclusion, the inclusion of equalization tanks in WRRFs has significant positive effects. These results were obtained with equalization of flow only. Some other limitations were experienced during the project resulting in the following recommendations: further research will be needed to validate and calibrate the model, As the model was not successfully incorporated in a plant-wide framework, further developments in that direction are required, as well as including the equalization of load in the model.

No description available.

Dissertation submitted in fulfilment with the requirements for the Masters Degree: Biotechnology, Durban University of Technology, 2011. / Untreated vegetable oil effluents (VOEs) are known for creating shock-loading problems for the receiving wastewater treatment installations, resulting in poor quality final effluents being produced which do not satisfy municipal discharge standards. Onsite activated sludge treatment as an alternative has not been fully investigated. Hence, in this investigation biological treatment using the activated sludge process was chosen as the method for the treatment of VOE. The effect of VOE on measured process parameters was also determined. Novel molecular techniques such as fluorescent in situ hybridisation (FISH) and dot-blot hybridization have become powerful tools for the analysis of complex microbial communities that exist within activated sludge. The aim of this investigation was to evaluate biological treatment, optimize and apply FISH and dot-blot hybridization in order to analyze the microbial community implicated the biological treatment of VOE using probes EUBmix, ALF1b, BET42a, GAM42a and HGC69a. A laboratory-scale modified Ludzack-Ettinger (MLE) process setup and fed VOE with a COD (chemical oxygen demand) of ± 1000 mg/L. Daily monitoring of the process involved COD and TKN (total kjeldahl nitrogen) analysis of the influent and effluent as well as direct OUR (oxygen utilization rate) measurement and monitoring of the MLVSS (mixed liquor volatile suspended solids) concentration of the aerobic mixed liquor. The process exhibited overall COD and TKN removal capacities of 84% and 90% respectively. The aerobic mixed liquor had an OUR of 19 mgO/L.h and an average MLVSS concentration of 3000 mg/L. FISH results revealed that 72% of cells stained with 4‟, 6-diamidino-2-phenylindole (DAPI) within the aerobic mixed liquor bound to probe EUBmix, indicating a substantial Bacterial population within the laboratory-scale biological process. The alpha-Proteobacteria was identified as the dominant bacterial community comprising 31% of Bacterial cells, followed by the beta-Proteobacteria (17% of EUBmix), gamma-Proteobacteria (8% of EUBmix) and Actinobacteria (4% of EUBmix). Results of dot-blot hybridization were in agreement with FISH Adrian Phillip Degenaar| CHAPTER 1: General Introduction - v - results reiterating dominance of the alpha-Proteobacteria. This indicated that the class alpha-Proteobacteria could play a primary role in the biological degradation of VOE. This research will therefore aid in process design and retrofitting of biological processes treating VOE.