The patented technology of using polyelectrolytes has been used to reduce membrane fouling in MBR. The commercial products called MPE30 and MPE50 are available from Nalco company of USA since 2004. The polyelectrolyes with net cationic charge coagulates not only small particles, but also soluble microbial products (SMP). It is believed that those chemicals decrease membrane fouling by reducing fine particle and SMP concentrations (Yoon, 2004; Yoon, 2005a; Yoon2006). Other organic macromolecules such as chitosan and starch can also coagulates particles, but they are much less practical than synthetic polyelectrolytes due to the high dosage requirement, low efficacy, and/or biodegradability.
In one experiment, particle size increased when 16 mg/L of MPE50 based on reactor volume was added daily directly to the aeration tank, where average particle size increased from 101 micron to 179 micron (Fig. 1). In the mean time, the concentrations of soluble COD and SMP (or soluble EPS) decreased as MPE50 dosage increased to a certain point, but they increased again with excess MPE50 (Fig. 2). The zeta potential of particles increased from negative to positive as MPE50 dosage increases as a consequence of net positive charge of MPE50 (Fig. 3). Since all the commercial membranes popularly used in MBR have negative zeta-potential, MPE50 dosages must be controlled in order not to cause charge reversal.
Combining the two observations in Fig. 2 and Fig. 3, it is evident that there exist an optimum MPE50 dosage that does not cause charge reversal of particles while maximizing the coagulation of fine particles and SMP. The optimum dosage of MPE50 varies depending on mixed liquor conditions, but as a rule of thumb 100 mg/L of MPE50 per every 3,000 mg/L of MLSS appears plausible as an initial dosage based on total sludge volume when MLSS in membrane tank is less than 12,000 mg/L (Guo, 2008).
Fig. 1. Effect of MPE50 (=MFR) on particle size distribution when 16 mg/L of MPE50 is added daily to the MBR with MLSS of 4400 mg/L (MFR reactor) and 4700 mg/L (control reactor) (Hwang, 2007)
Fig. 2. Effect of MPE50 (=MFR) dosage on soluble COD, soluble EPS (or SMP), and bound EPS (Lee, 2007).
Fig. 3. Effect of MPE50 (=MFR) dosage on particle zeta-potential (Lee, 2007).
Due to the lower SMP concentration in water phase, permeate COD tends to be lowered when MPE products are applied. In one occasion, permeate COD decreased by 50% when 200 mg/L of MPE50 was added to the aeration tank of a pilot MBR (Yoon2006). The activity of microorganisms measured by specific oxygen uptake rate (SOUR) was not noticeably affected by MPE50. When 16 mg/L of MPE50 was added daily, SOUR of mixed liquor was measured at 53.6 mg O2/MLVSS/hr, while it was at 56.4 mg O2/MLVSS/hr in a control MBR. In the same experiment, permeate COD was found decreased with MPE50 from 18.2 mg/L to 14.8 mg/L. In the meantime, permeate TN and TP remained virtually unchanged considering the measurement error as summarized in Table 1 (Hwang, 2007).
Table 1. COD, TN, and TP removal efficiency with and without MPE50 (Hwang, 2007)
It has been known that MPE50 reduces foam from the biological tanks. The mechanisms of foam reduction is not completely clear, but it is considered that foams are destabilized due to the reduced SMP and fine particle concentrations. In fact, fine particles and polymers with hydrophobic moieties tend to gather in the air-liquid interfaces of foam and interrupt water drainage, which leads to the extension of the foam longevity. As shown in Fig. 4, thick brown form layer on the surface of anoxic tank disappeared when 400 mg/L of MPE50 was added to the system with 12,000 mg/L of MLSS.
a) Before MPE50 addition b) After MPE50 addition (400 mg/L)
Fig. 4. Effect of MPE50 on biological foam. Thick brown foam layer disappeared from anoxic tank in a municipal MBR (Yoon, 2004).
Similar flux enhancing effect was observed in other study (Dizge et al., 2011), where cationic polyelectrolytes were added directly to the MBR mixed liquor. In general, polyelectrolytes reduced protein and polysaccharides concentrations in the permeate by coagulating them to larger particles, where the extent of the protein and polysaccharides removal was dependant on the thickness of the dynamic membrane (or cake layer). Non-existent of cake layer caused free passage of small molecules through large membrane pores in some conditions. Details of dynamic membrane theory is found here.
© Seong Hoon Yoon