Inorganic coagulants such as ferric and aluminum salts are used to enhance phosphorous removal in MBR as discussed here. It has been reported that the added coagulants not only form insoluble salts with phosphate ions, but also reduce soluble microbial products (SMP) concentration, which in turn reduces membrane fouling potential of mixed liquor.
The added inorganic coagulants, however, can alter the microbial community due to their varying degree of toxicity to some microorganisms, which eventually affects the membrane fouling propensities of mixed liquor. For example, aluminum salts have been known to be toxic to some filamentous microorganisms as discussed here. Little has been known about the impact on microbial community and the subsequent effect on membrane fouling. However, it has been known that excess inorganic coagulant addition to the mixed liquor with low alkalinity can cause excess pH drop that in turn triggers high SMP concentrations and accelerated membrane fouling.
It has been also known that adsorbents such as zeolite and powdered activated carbon (PAC) can reduce membrane fouling likely by adsorbing a portion of SMP and by physically scouring membrane surface. However, a long-term use of adsorbents does not appear practical since new adsorbents must be added continuously to treat the SMP produced continuously. In addition, increased excess sludge production can cause overload to the sludge handling system, which eventually makes the process not economical.
As shown in Fig. 1, inorganic additives such as alum and zeolite can delay TMP increase (or operating pressure increase). In this experiment, hollow fiber membranes made of polyethylene (Mitsubish Rayon Co., Japan) were used as a U-shaped fiber bundle. Flux was set at 15 LMH. Alum dosage was controlled to maintain a molar ratio of Al/P at 1.5 while zeolite concentration was controlled at 1,000 mg/L. Under the experimental condition, alum decreased nitrification rate from 92% to 76% while no effect on nitrification was observed with zeolite.
On the contrary, little impact on nitrification has been reported from the municipal MBR plants using alum or aluminum chlorides as discussed here. The cause of the discrepancy is not completely clear, but excessive pH drop and/or insufficient acclimation time in the lab scale experiment might be a cause. 1
In other study, protein concentration was measured in mixed liquor after adding ferric chloride and aluminum sulfate. Filters with three different pore sizes were used to filter the mixed liquor sample before protein concentration was measured by Lawry’s method. As shown in Fig. 3, protein concentration was dramatically reduced by iron and aluminum based coagulants.
Fig. 1. Effect of inorganic additives on TMP increase (Lee, 2001)
Fig. 2. Effect of alum on particle size distribution of activated sludge (Lee, 2001)
Fig. 3. Protein concentration with iron and aluminum salts (Mishima, 2009)
When poly-aluminum chlorides (PACl) was added at 12.5 mg/g MLSS and maintained at the condition by compensating the loss through excess sludge removal, TMP increasing rate was significantly reduced as shown in Fig. 4, where step 1 and step 2 were before and after adding PACl. No adverse effect on COD and nitrogen removal was observed while phosphorus and color removal efficiencies increased significantly in the MLE based MBR used in the study as shown in Table 1.
Fig. 4. Corrected TMP against 20 oC (TMP20) before and after applying PACl at 12.5 mg/g MLSS (Teli, 2012).
Table 1. Average permeate quality and removal rates before (step 1) and after (step 2) adding PACl (Teli, 2012)
© Seong Hoon Yoon