Performance of two-stage dynamic anaerobic membrane bioreactor for treating high strength food processing wastewater

• Main Author: Mahat, Siti Baizura
• Format: Thesis
• Language: English
• Published: 2021
• Subjects: Refuse and refuse disposal - Biodegradation; Food industry and trade - Waste disposal; Anaerobic bacteria
• Online Access: http://psasir.upm.edu.my/id/eprint/104174/1/SITI%20BAIZURA%20BINTI%20MAHAT%20-%20IR.pdf; http://psasir.upm.edu.my/id/eprint/104174/
• Institution: Universiti Putra Malaysia

Food processing wastewater (FPW) contains a high level of oil and grease, requiring extensive treatment. The submerged dynamic membrane in anaerobic digestion (AD) treatment offers cheap and complete biological and physical separation of solid-liquid. Dynamic membrane (DM) developed onto cheap support material can replace the expensive conventional membranes. Although promising DM utilization in AD treatment has been reported, scarce research focused on the DM formation to explain its performance and fouling control. Thus, this study's objectives were to evaluate the submerged dynamic membrane two-stage anaerobic bioreactor (DAnMBR) performance in treating FPW and assess the DM characteristics and development mechanism. The batch biodegradability test assays were used to determine the best performance of substrate (FPW) to inoculum (anaerobic digester sludge) ratio (S/I) at 1.0, 1.5 and 2.0. A two-stage anaerobic digester (named acidogenic and methanogenic reactors) with two submerged 20 μm woven filter cloth as the supporting material (DAnMBR) in the second tank was used in this study. Successful start-up using synthetic wastewater and then acclimatization by adding FPW in steps up to 100% FPW was achieved when 90% of chemical oxygen demand (COD) were removed. Treatment using support material commenced thereafter. The treatment performance utilizing APHA methods was evaluated at different hydraulic retention time (HRT) of 0.4-1.3 days and organic loading rates (OLR) of 3.5, 5.0, 6.5, and 7.0 g COD/L.d. Best OLR was used to assess the DM formation until fouling occurred, and the cake layer samples were taken for characterization periodically. S/I 1.0 ratio performed best with COD, biochemical oxygen demand, total solids (TS) and volatile solids (VS) removals of 96.9, 96.6, 75.8, and 65.2%, respectively. The bioreactor presented a good performance at OLR 5.0 g COD/L.d with removals of 97.5% COD and 99% total suspended solids at HRT of 0.5 day. The methane gas production yield achieved a maximum of 0.40 L methane/g COD added at OLR 3.5 and 5.0 g COD/L.d with the same HRT 0.5 day on both OLRs. The average permeate flux in these studies was around 60 L/m2 h. Fouling occurred at 35 days during the DM development and characterization study with a final flux of 2.5 L/m2.hr and transmembrane pressure of 0.7 bar. The cake layer thickness increased slightly from day 14 to 28 but sharply at the fouled stage, agreeing with the treatment performance. Protein to polysaccharide ratio (PN/PS) of the extracellular polymeric substances (EPS) increased significantly compared to soluble microbial product PN/PS ratio; thus, it is the main contributor to the membrane fouling. High-through-put 454 pyrosequencing of total DNA revealed that Proteobacteria, Bacteroidetes and Methanosaeta were abundant in bacterial and archaeal communities, which played an important role in the DAnMBR system. In conclusion, following the results obtained in this study, DM technology achieved a stable and high-quality permeate. Thus, DAnMBRs can be used as a reliable and satisfactory treatment technology to treat high strength wastewaters.