A membrane bioreactor (MBR) was used for treating biological aerated filter effluent in a municipal wastewater plant, and chemical phosphorus removal was accomplished in the MBR. The results showed that ferric chloride of 20 mg/L and aluminum sulfate of 30 mg/L were the optimal dosages for total phosphorus (TP) removal, and the TP removal efficiency was over 80%. In long-term continuous operations, both ferric chloride and aluminum sulfate effectively mitigated membrane fouling, with the corresponding growth rate of transmembrane pressure decreased to 0.08 and 0.067 kPa/d, respectively. Sludge particle sizes analysis demonstrated that the decrease of particle sizes lower than 50 μm was the main reason for membrane fouling control. Simultaneously, the proteins and polysaccharide (PS) concentrations in the MBR supernatant were analyzed, and the PS concentration significantly decreased to 2.02 mg/L at aluminum sulfate of 30 mg/L, indicating the flocculation of aluminum sulfate on PS was the main reason for mitigation of membrane fouling.
The reuse of municipal wastewater is the goal of modern and sustainable wastewater treatment systems (Verstraete et al. 2009). However, the majority of municipal wastewater treatment plants (WWTPs) in China are designed on the basis of the activated sludge process, with effluent total phosphorus (TP) of 2–4 mg/L. Even in an enhanced biological phosphorus removal process, the effluent TP concentrations are usually in the range of 0.5–1.0 mg/L (Piekema 2004), which does not meet the requirements of reuse in China.
Recently, membrane bioreactor (MBR) technology has been used for wastewater reuse due to its ability to improve the removal of refractory organic matters whilst possessing a small footprint (Ferraris et al. 2009). To prevent eutrophication, some chemical agents such as iron and aluminum salts are added into MBRs to achieve a sufficient and consistent degree of phosphorus removal (Yeoman et al. 1988; Gutierrez et al. 2010). Iron salt, in either ferrous or ferric form, is commonly used in municipal wastewater treatment as an aid to phosphorus removal, and the influent TP concentrations of 10 mg/L are consistently reduced to effluent concentrations of 0.03–0.04 mg/L at Fe/P (both ferric and ferrous) molar ratio of 2.0 (Zhang et al. 2015). Simultaneously, the addition of ferrous salt has been demonstrated to be a good alternative to ferric salt for phosphorus removal in a pilot-scale MBR (Wang et al. 2014). It was also reported that 30 mg/L of alum was effective to remove 3 mg/L of phosphorous in feed for the MBR process (Song et al. 2008). Results suggested that ferric chloride has more advantages in phosphorus removal than aluminum chloride.
However, the wide use of MBRs in wastewater treatment is still limited by membrane fouling, which decreases permeability and, in turn, increases energy consumption (Wang et al. 2014). Several experimental studies indicated that the mixed liquor colloidal and soluble microbial products (SMP) were mainly responsible for membrane fouling (Zhang et al. 2011; Wu et al. 2012). The SMP are the metabolic products of microorganisms consisting of proteins (PN) and polysaccharides (PS), which may attach to and then accumulate on the membrane surface to form a biocake layer (Tansel et al. 2006; Zhang et al. 2015). Recently, more strategies for fouling control have focused on the chemical coagulation/flocculation process to remove suspended solids (SS), colloidal particles, as well as other soluble materials in MBRs (Wu et al. 2006; Gkotsis et al. in press). The surface negative charges of microbial flocs change to almost neutral along with cationic coagulants added into mixed liquid, which is beneficial for producing larger flocs, and greatly reducing transmembrane pressure (TMP) (Praneeth et al. 2014). Additionally, it has been found that a low coagulants dosage is sufficient for mitigating membrane fouling, and that ferric salts are more effective than aluminum salts in membrane pollution control (Mishima & Nakajima 2009).
A comprehensive review of the literature reveals that MBRs have been commonly used for sewage or municipal wastewater treatment, but there is little published data available with regard to comparison of ferric chloride and aluminum sulfate on phosphorus removal in MBRs for municipal wastewater advanced treatment. Therefore, in this study, MBR technology was used to treat biological aerated filter (BAF) effluent in a municipal wastewater plant. The effects of different ferric chloride and aluminum sulfate dosages on TP removal in the MBR were investigated, and the appropriate dosages for these two coagulants were determined. Simultaneously, the TP removal and TMP at optimal ferric and aluminum salts were respectively studied by long-term continuous operational experiment. Also, the particle size and SMP analysis were used to explain the mechanism for membrane fouling control.
MATERIALS AND METHODS
Two identical MBRs with effective volume of 5 L were operated at room temperature. These two MBRs were controlled at the same conditions except two different coagulants, ferric chloride and aluminum sulfate, were continuously added into the MBRs, respectively. Two identical U-shaped bundle membrane modules of polyvinylidene fluoride hollow fiber membranes with nominal pore size 0.2 μm and total surface area 0.3 m2 (Beijing Origin Water, China) were immersed vertically into the MBR with coarse bubble aeration used to limit membrane fouling. The influent was fed continuously, and the effluent was withdrawn from the membrane modules using a suction pump (Masterflex, Cole-Parmer) in intermittent mode with on/off ratio of 8 min:2 min. The gas–water ratio and flux were controlled at 10:1 and 10–15 L/(m2·h), respectively. The sludge retention time was 40 d, and the hydraulic retention time was 2.4 h.
The feedwater used in this study was from a BAF reactor of a municipal WWTP in Guangdong. The influent characteristics of the MBRs including pH, chemical oxygen demand (COD), ammonia nitrogen (NH4+-N), total nitrogen (TN), TP and SS are summarized in Table 1.
Ferric chloride and aluminum sulfate dosing
The TP removal and the membrane fouling after adding ferric chloride and aluminum sulfate solutions into the MBRs were assessed. The ferric and aluminum concentrated feeds were separately prepared by dissolving 8 g ferric chloride heptahydrate and 8 g aluminum sulfate octadecahydrate in 2 L tap water. Then, the same tap water was used for diluting the concentrated feed to be ferric and aluminum solutions of 0, 5, 10, 15, 20, 25 and 30 mg/L, respectively. Then, the MBRs at ferric and aluminum salts dosage of 20 and 30 mg/L were continuously operated for 100 days to compare the TP removal and membrane fouling control, respectively. Simultaneously, a MBR reactor without the addition of coagulants was set as the blank test. In this study, the pH values in reactors were adjusted and maintained at 6.5–7.5 using 2 M hydrochloric acid (HCl) or 2 M sodium hydroxide (NaOH) solutions.
Influent and effluent samples of MBRs were analyzed every other day. The COD, NH4+-N, TN and TP concentrations were determined according to Standard Methods (APHA 2005). Dissolved oxygen was analyzed using WTW Handheld Multi-parameter Instruments (pH/Oxi 340i, WTW, Germany). pH, and oxidation reduction potential were monitored using on-line detection (pHs-8D, Dongrun, China). The SMP were collected and extracted based on the reference (Zhang et al. 2015), and the concentrations of PN and PS in SMP were analyzed, respectively. The PN concentrations were measured by the Lowry method with bovine serum albumin (BSA) as standard (Lowry et al. 1951), and the PS concentrations were analyzed by the phenol-sulfuric method with glucose as the standard (Herbert et al. 1971). The SMP analysis was performed in triplicate, and the average value was calculated.
RESULTS AND DISCUSSION
Effects of coagulant dosages on TP removal
Effects of continuous dosing on MBR operation
Comparison of ferric chloride and aluminum sulfate on sludge particle sizes
It can be found that the small particle sizes in the MBR were flocculated to be large flocs due to the addition of ferric and aluminum salts, resulting in the decrease of membrane fouling (Song et al. 2008).
Comparison of ferric chloride and aluminum sulfate on PN and PS
The impacts of ferric chloride and aluminum sulfate on TP removal and membrane fouling in a MBR treating BAF effluent of municipal wastewater were investigated. The optimal ferric chloride and aluminum sulfate dosages for TP removal were determined by a short-term test. Simultaneously, long-term continuous operational results suggested that aluminum sulfate of 30 mg/L was more effective than ferric chloride of 20 mg/L in decreasing TMP. The sludge particle sizes lower than 50 μm and the PS contents were decreased in this study, which might be the main reason for membrane fouling mitigation.
This work was financially supported by the National High Technology Research and Development Program of China (863 Program) (No. 2012AA063603-02), Science and Technology Planning Project of Guangdong Province (2012A061600010), and Nanhai District Environmental Protection Industry Innovation and Development Special Project (20120306).