Effect of membrane permselectivity
One important consideration for the effective treatment of perchlorate by IEMB system is the permselectivity of the AEM. Some AEM have low permselectivity towards counter-ions of the same valence, hence affecting the quality of the treated water. Low permselectivity exhibits high resistance to ion transport which can be attributed to the membrane’s low ion exchange capacity, high water content, and high thickness (Nagarale et al. 2006). In order to enhance permselectivity of the membrane, various modifications can be made including the synthesis of a thin, highly cross- linked surface modifying layer on top of a non-permselective support (Saracco and Zanetti 1994). The product causes a higher steric repulsion on multi-valent anions than the mono-valent anions, and is hence called mono-valent permselective membrane. One of the most commonly employed mono-valent permselective membrane in IEMB application is the Neosepta ACS membrane (Matos et al. 2006b, Velizarov et al. 2008, Ricardo et al. 2012).
Neosepta ACS membrane allows high pollutant fluxes at longer periods of time and can successfully prevent the transfer of microbial cells, carbon sources and metabolic by-products that could otherwise lead to secondary contamination of the treated water (Matos et al. 2008). Consequently, transport of multi-valent ions (e.g., PO|~, HP042", H2P04) towards the opposite compartment can also be minimized by ACS membrane, thus preserving the concentration of these ions in the feed water and biomedium. However, the major drawback of Neosepta ACS membrane which hinders the large-scale application of IEMB is its expensive cost. This prompted the search for low-cost alternative membranes. Matos et al. (2008) compared the performance of a relatively cheaper Excellion™ 1-200 (SnowPure, USA) to that of Neosepta ACS (Tokuyama Soda, Japan) in the removal of N03~ and C104 by IEMB system. Although Excellion 1-200 obtained lower removal efficiencies of C104 (85%) and N03~ (88%) than that of the Neosepta ACS (96% C104 and 99% N03' removal efficiencies) and allowed subsequent transport of sulfate and phosphate ions (H2P04, HP042') from the bio-compartment to the treated water stream, there was no observed flux decline of the target anions during the one month operation and no secondary' contamination due to organic compounds in the treated water (Matos et al. 2008). The latter characteristics are two important criteria in ion exchange membrane selection. These results showed that Excellion 1-200 can be a cheaper alternative to Neosepta ACS for perchlorate removal by IEMB technique.
Effect of biofilm formation
As mentioned earlier, the bioconversion of perchlorate to harmless species in IEMB mainly occurs in the biofilm rather than in the suspended culture. This is because the majority of the active microbial population is found in the biofilm (Fox et al. 2014). The formation of biofilm on the membrane surface in contact with the bio-compartment is due to the ion transport across membranes creating a region with favorable microbial growth conditions (Matos et al. 2006a). Thin biofilm formation facilitates perchlorate reduction and ion transport from water-compartment to bio-compartment by maintaining higher perchlorate flux across the membranes. As demonstrated by Fox et al. (2014), when they compared the fluxes obtained in IEMB with and without biofilm, perchlorate removal efficiency was higher by 27 ± 6.6% in IEMB with biofilm. Fox et al. (2016) also observed a 5% increase in perchlorate transmembrane flux in the presence of biofilm. Additionally, biofilm acts as a secondary barrier against transport of the carbon substrate to the water-compartment, thereby preventing the secondary contamination of the treated water. On the contrary, excess biofilm decreases the IEMB efficiency over time, which can be minimized by controlling the concentration of the ionic pollutants.
Effect of carbon source limitation
Use of heterotrophic PRB for perchlorate treatment by IEMB was reported so far. Ethanol and glycerol have been extensively employed as both the electron donor and carbon source in the effective reduction of ClOy by an enriched heterotrophic culture in IEMB. Ethanol is non-toxic and has low diffirsivity through the AEM, that is why a lot of investigations regarding anion treatment by IEMB used this compound as carbon source (Matos et al. 2006b, 2008, Ricardo et al. 2012). However, ethanol is also an expensive carbon source. On the other hand, glycerol, a by-product of bio-diesel production industry, is much cheaper and has 60% lower diffusion coefficient (6.9 x 10"9 ± 4.7 x 10~10 ctrri/s) than that of ethanol (1.8 x io_s cm2/s), and is hence a suitable alternative. The applicability of glycerol as cheap carbon/energy source in IEMB treatment of perchlorate has just been recently investigated (Fox et al. 2014, 2016). According to Fox et al. (2016), the dissimilatory reduction of perchlorate using glycerol as the electron donor involves two stages: glycerol is first fennented by the suspended paxticles in the bio-compartment and the fermentation products are then consumed by the biofilm attached to the membrane to reduce perchlorate. The perchlorate bio-reduction reactions with ethanol and glycerol are presented in equations (2) (He et al. 2019) and (3) (Fox et al. 2016), respectively:
The concentration of the carbon source in the bio-compartment must be carefully controlled in order to ensure the production of high-quality treated water. High carbon concentration fed into the bio-compartment was found to cause elevated dissolved organic carbon (DOC) levels in the water- compartment, which then decreased the effluent quality (Fox et al. 2014). This cross-contamination can be attributed to the formation of secondary' metabolites front the excess carbon assimilation by the microbial culture in the bio-compartment (Fox et al. 2014). In contrast, insufficient amount of carbon substrate results in accumulated perchlorate ions in the bio-compartment. This, however, does not affect the concentration of perchlorate in the water-compartment since perchlorate transport across membrane is driven by Donnan dialysis principle. In IEMB, the quality of the treated water is the main consideration and not the bioreactor effluent. This is an added advantage of the IEMB process when compared to other techniques, viz. fluidized bed reactor (FBR). In FBR, carbon source limitation results in production of water with higher pollutant concentrations requiring subsequent treatment, which adds to the overall treatment cost.
Effect of co-existing nitrates
It is important to understand the effect of nitrate on perchlorate reduction by IEMB since these two anions often co-exist in contaminated waters. Biodegradation studies involving perchlorate have reported that most PRB can also use N03~ as an electron acceptor and that the presence of N03" slows down the degradation of perchlorate (Nerenberg et al. 2008, Gal et al. 2008). A kinetic study was conducted by Ricardo et al. (2012) to investigate the reduction rates of these two oxy-anions in IEMB using a mixed anoxic microbial culture and ethanol as the carbon source. They observed that perchlorate reduction rate was inhibited at the initial stage and only increased after nitrate and nitrite (N02“) were completely depleted. The sequential reduction can be attributed to the presence of low PRB and high denitrifying bacteria population in the mixed culture used (Ricardo et al.
2012). Moreover, both microorganisms have shown more preference to nitrate than perchlorate as primary electron acceptor, which can be seen on their kinetic parameter values: the value of the maximum specific substrate utilization rate (qmaJ for nitrate (10.79 mg N03"/mg VSS-d) was 35 times higher than perchlorate (0.30 mg C104'/mg VSS-d). In addition, the lower half saturation constant of nitrate (К,, = 1.05 mg N03"/L) also indicates higher affinity of the mixed culture to nitrate than perchlorate (Kp = 4.97 ntg C104"/L). These values are in agreement with the results of other studies (Halling-Sorensen and Jorgensen 1993, Dudley et al. 2008).
Effect of nutrient limitation
Ricardo et al. (2012) examined the performance of IEMB at excess and limiting nutrient concentrations. Two experiments were carried out feeding the bio-compartment with 9 mg/d and 0.9 mg/d of ammonia. In the presence of excess ammonia, both nitrate and perchlorate were significantly reduced to concentrations below the quantification limits: 1 mg NOj/L and 2 pg C104/L. This was due to the development of biofilm on the membrane bio-side where the active biodegradation usually takes place. At limiting ammonia concentration, no biofilm was formed due to the rapid consumption of ammonia by the suspended culture. The nitrate reduction rate was almost unaffected, while the perchlorate reduction rate was decreased by 10%. The same results were obtained by another study (Choi and Silverstein 2008) in which the perchlorate reduction rate was decreased by 70% in the presence of equimolar nitrate at limited acetate condition. This can be attributed to the competition of the microbial cells for the limiting amount of nutrients present (Ricardo et al. 2012). Even if both nitrate and perchlorate accumulated in the bio-compartment due to insufficient bio-nutrients, the final nitrate and perchlorate concentrations in the treated water of IEMB were not affected since the transport of anionic pollutants across the membrane to the bio-compartment by Doiman dialysis primarily dictated the IEMB performance.