Fouling in an anoxic-aerobic membrane bioreactor : role of membrane properties and biological contributions
- The Membrane Bioreactor (MBR) is a superior alternative to the conventional activated sludge process in wastewater treatment. Instead of gravity settling, MBR utilizes filtration membranes in the retention of microorganisms. The MBR can be configured in different ways depending on the treatment goal. An objective of this research was to demonstrate the feasibility of the MBR in performing nitrification-denitrification in the second stage of a two-stage treatment scheme whereby the first-stage was an acidogenic MBR. The other objectives were to understand the comparative influence of membrane properties and microbial products on fouling. The MBR was initially operated at a long sludge age of nearly 100 d. This served as a worst case scenario for the process, as decay products dominated the mixed liquor. It also served as a suitable platform for the study of membrane fouling by the microbial products. With a design influent concentration of 440 mg COD/L, 50 mg NH4+-N/L and an aerobic to anoxic recirculation ratio of 5, the effluent NH4+-N was consistently below 0.1 mg/L while NO3--N was around 7 mg/L. The organic content in the reactor showed a steady drop since startup before reaching a dynamic steady-state after 400 d. The permeate COD remained steadily low throughout the same period. This suggests adaptation of the microorganisms to the utilization of biomass associated products (BAP) as well as the emergence of groups of microorganisms that can utilize the BAP as substrates. Tracking the microbial community structure by terminal restriction length polymorphism showed that ecological stability was attained after about 200-300 d since operations began. On the membrane aspect, an issue that emerged during the course of the study was the quality of membrane zeta potential data. Most data reported in the literature did not consider an additional conduction path within the membrane resulting in an underestimation of the streaming potential coefficient, and hence zeta potential. The problem was exacerbated by significant head loss in the measurement device. By simulating the flow using computational fluid dynamics, both the head loss and flow channel height were estimated, which were then applied in the computation of zeta potential. Corrected and uncorrected values differ by a factor of 2 -- 8. The corrected values appear also to be a function of membrane morphology. Using a series of polyvinylidene fluoride (PVDF) membranes, it was observed that the surface zeta potential becomes more negative with increasing roughness and more positive with the fractal dimension. The corrected membrane zeta potential also correlates with the performance for some membranes. But for others, the results were confounded with other properties such as hydrophilicity. From the long filtration record of the MBR, individual response curves were extracted and fitted with an empirical function to reduce the description to 3 indices. The index representing the curvature of the flux profile appears to be correlated with the concentration of soluble microbial products for polyethersulfone membranes and mixed liquor suspended solids for polyether block amide coated PVDF membranes, highlighting the interplay between membrane properties and biological materials.
|Type of resource
|electronic; electronic resource; remote
|1 online resource.
|Wong, Chuen Yung Philip
|Stanford University, Civil & Environmental Engineering Department
|Leckie, Jim, 1939-
|Leckie, Jim, 1939-
|Statement of responsibility
|Chuen Yung Philip Wong.
|Submitted to the Department of Civil and Environmental Engineering.
|Ph. D. Stanford University 2010
- © 2010 by Chuen Yung Philip Wong
Also listed in
Loading usage metrics...