Treatment of wastewater having different carbon to nitrogen ratio and activated sludge properties in aerobic membrane bioreactors

The use of submerged membrane bioreactors (sMBR) for the purpose of municipal and industrial wastewater treatment have increased significantly over the last few years. Although there are a number of components that affect membrane fouling, the most important membrane fouling parameters are extracellular polymeric subtances (EPS) and soluble microbial product (SMP) secreted by microorganisms. The production of EPS components are affected significantly by the carbon/nitrogen (C/N) ratio of wastewater. The concentration of EPS and SMP values influence flocculation, floc size, relative hydrophobicity and zeta potential of activated sludge. In addition, EPS production alters the MLSS concentration as well as the rheological properties of activated sludge. In this study, the effect of the EPS and SMP concentration, which are the primary components for membrane fouling, at different sludge ages and at three different C/N ratios on the activated sludge rheology in a submerged MBR were investigated. In the first period of the study, the sMBRs were operated with synthetic wastewaters having C/N ratios of 6, 21 and 40 in order to avoid feed wastewater compositional changes and to be able to compare the results obtained at different sludge ages (20, 40, 80 and 5 days). The sMBRs, having effective volumes of 20 L each, were operated with an approximate volumetric organic loading rate of 1,2 kg/m3.d using synthetic wastewater at four different SRT values for a period of 364 days. After steady-state conditions reached, at each SRT values, the fractions of protein and carbohydrate of EPS and SMP, the measurements of floc size distribution, zeta potential, apparent viscosity and rheological analyses, and the determination of critical flux and membrane fouling rate (MFR) were performed. While COD removal efficiencies were found over the 96% independent of C/N ratio at all SRT values, NH3-N removal efficiencies were found over 94% for SRT values of 20, 40 and 80 days. However, the NH3-N removal efficiencies were found nearly 87% at 5 days SRT. While the fraction of protein of EPS, which was extracted from activated sludge at steady-state conditions, decreased with increasing C/N ratio of wastewater, the fraction of carbohydrate of EPS increased. As the protein concentrations of SMP in supernatant were found close to each other, the carboydrate concentrations of SMP increased with an increase of C/N ratio. Activated sludge apparent viscosities showed an increase, when the C/N ratio and SRT value were increased. The activated sludge mean floc sizes increased with an increase in C/N ratio, while decreased with a decrease of SRT values. Zeta potential values showed a negative increase with increasing C/N ratio, on the other hand, the relative hydrophobicity (RH%) values decreased with increasing C/N ratio and increasing SRT values, as was expected. The MFR values increased with increasing C/N ratio, however they decreased with increasing SRT values. Taking into consideration the results of experimental studies and their statistical analysis, it was decided to operate the sMBRs at a sludge age of 40 days for the treatment of real wastewaters in the second period of study. Real wastewater studies were performed for a period of aproximately 100 days at effective volume of 10 L and the volumetric organic loading rate of 1,2 kg/m3.d, 7,48 kg/m3.d and 7 kg/m3.d for municipal wastewater, dairy wastewater and paper-mill wastewater, respectively. It was found that 90% COD and 80% NH3-N removal efficiencies for municipal wastewater, 99% COD and 85% NH3-N removal efficiencies for dairy wastewater and 93% COD and 80% NH3-N removal efficiencies for paper-mill wastewater at steady-state conditions in MBRs. Although similar results were obtained for the real wastewaters studied with synthetic wastewaters, increased membrane fouling rates were observed for the real wastewaters. It is thought that the inorganic components present in real wastewaters were one of the main reasons for increased membrane fouling rates which indicated inorganic fouling along with the microbial fouling. As a result, the most important parameters in the operation of MBRs were found to be the adjustment of the C/N ratios and, if necessary, to eliminate possible nutrient deficiency. The statistical analysis revealed that the most important parameters on membrane fouling rate were the viscosity, C/N ratio, sludge retention time (SRT) and mixed liquor suspended solid (MLSS) values.