Direct Biofiltration as a Pretreatment to Control Fouling in Ceramic Membranes in Drinking Water Treatment

Abstract

Ceramic membranes have been widely and successfully used in the food and beverage processing industry. Despite their success, ceramic membranes are not commonly employed in drinking water treatment due to their high initial capital cost. Polymeric membranes, on the other hand, have gained widespread use in drinking water treatment in the last few decades due to their ability to meet stringent water quality regulations. Ceramic membranes have a number of advantages over polymeric membranes, which include high chemical and thermal stability, higher fluxes and longer operational life. Advances in membrane technology in recent years coupled with innovative design have made the life cycle cost of implementing ceramic membranes competitive with that of polymeric membranes. This has resulted in a number of drinking water treatment plant installing ceramic membranes as part of the treatment process, especially in Japan. The biggest challenge facing membrane filtration (polymeric or ceramic) is fouling. To control fouling, coagulation prior to ceramic membrane filtration is often implemented and has been shown to be effective in controlling both hydraulically reversible and irreversible fouling. Direct biofiltration without pretreatment (BFWP) (coined by Huck et al., 2015) has been shown to be another effective “green” pretreatment to control fouling in polymeric membranes. High molecular weight natural organic matter (NOM) such as biopolymers have been found to be directly related to the hydraulically reversible fouling and to play a key role in hydraulically irreversible fouling of polymeric membranes and biofiltration is able to reduce the concentration of this NOM fraction. Given the effectiveness of BFWP in controlling fouling in polymeric membranes, there is an opportunity to investigate its applicability to ceramic membranes. Therefore, the goals of this study were to investigate the efficacy of BFWP as a pretreatment to control fouling in ceramic membranes and characterize the fouling of the membranes over time. The effects of Empty Bed Contact Time (EBCT) of the biofilters, membrane materials and pore sizes (Microfiltration (MF) vs. Ultrafiltration (UF)) on the fouling rates were also investigated in the study. For the study, a direct biofiltration pilot plant was set up at the Mannheim Water Treatment Plant (WTP) in Kitchener, Ontario, Canada. Two dual-media biofilters – one with 8 minutes and the other with 16 minutes EBCT, were fed with roughing filter (RF) effluent, which in turn draws its influent from the raw water main line which feeds the WTP. The Mannheim WTP treats Grand River water which is impacted by agricultural and wastewater effluent upstream. The effluents from the biofilters were then used as influent to the ceramic membrane test units for the membrane fouling experiments. Fouling experiments were conducted using two parallel test units to compare the efficacy of different pretreatments (RF effluent and biofilter effluents) on MF and UF ceramic membranes. The study found that direct biofiltration is able to substantially remove turbidity and biopolymers from the feed water, thereby reducing both reversible and irreversible fouling rates in both MF and UF ceramic membranes. A higher EBCT biofilter also performed significantly better in terms of turbidity and biopolymer removal, corresponding to lower irreversible and reversible fouling in ceramic membranes. MF ceramic membranes were also found to foul at a much lower rate compared to UF ceramic membranes.