This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
The Membrane Bioreactor, can no longer be considered a novel process, it ensures improved effluent quality for textile wastewater treatment, while requiring a smaller footprint. The most costly part is the membrane, which is susceptible to fouling. Fouling of membrane bioreactors, results in higher operational expenses, reduced stability and operational performance. This review paper focuses on the biofouling of membrane bioreactors for textile wastewater treatment. The biofouling factors and control strategies are discussed. Lastly, future studies on membrane biofouling research are addressed.
Over the past two decades, the membrane bioreactor (MBR) has emerged as one of the innovative technologies in leading wastewater treatment (Yeon et al., 2009). MBRs are widely used for wastewater treatment that requires high effluent quality, e.g., water reuse or water recycling (Judd, 2008; Wang et al., 2008). Membrane bioreactor (MBR) technology combines the biological process by activated sludge and membrane filtration with a direct solid-liquid separation. With membrane technology (with pore sizes ranging from 0.05 to 0.4 Âµm), MBR systems permit the complete physical retention of bacterial flocs and almost all suspended solids within the bioreactor. The MBR system has many advantages over conventional wastewater treatment methods. These consist of small footprint and reactor requirements, high effluent quality, good disinfection capability, higher volumetric loading, and less sludge production (Judd, 2006). The Membrane bioreactor process has now become a remarkable sustainable option due to the rising numbers of such reactors and the capacity for the reclamation of wastewater. Thousands of MBR plants are operated all over the world; these are estimated to value around US$ 216 million, and the value is projected to rise to US$ 363 million by 2010 (Atkinson, 2006). However, membrane biofouling decreases MBR filtration performance with filtration time because of the bacterial cell's deposition, and its growth onto and into the membrane (Xiong Liu, 2010). Soluble microbial products (SMP) and extracellular polymeric substances (EPS) secreted by bacteria also play an important role in biofouling (Ramesh et al., 2007). This major hindrance and limitation of process has been under analysis since the early MBRs, and it is one of the most demanding issues facing further MBR development (Yang et al., 2006).
During the last few years, Le-Clech et al. (2006) and Meng et al. (2009) reviewed MBR fouling by concentrating on nearly all the fouling factors; specifically, they provided a very comprehensive review on sludge characteristics, operational parameters, membrane materials, and feed water characteristics.
In recent years, a large number of papers have been published on biofouling, with the number of publications for 2009 numbering 1450 (Table 1).This review paper is mainly focused on biofouling of the membrane bioreactor, particularly on the fundamentals of such biofouling, the biofouling factors, control strategies, and some future prospects. It is anticipated that this review may serve as a stepping stone for further development and application of biological methods towards effective control of microbial attachment and membrane biofouling.
Membrane fouling is a major obstacle that hinders faster commercialization of MBRs. As shown in Figure 1, membrane fouling in MBRs can be attributed to both membrane pores clogging and sludge cake attachment on membranes, which is generally the predominant fouling component (Lee et al., 2001). Membrane fouling results in a decrease of flux or an increase of transmembrane pressure (TMP), depending on the operation mode.
Biofouling is the deposition, growth, and metabolism of bacteria cells or flocs on the membranes, which is a significant concern in membrane filtration (Pang et al., 2005; Wang et al., 2005). For a low pressure driven membrane, such as ultrafiltration and microfiltration for treating wastewater, biofouling is a major hitch because most foulants (microbial flocs) in MBRs are much larger than the membrane pore size. Biofouling may start with the deposition of an individual cell or cell cluster on the membrane surface, after which the cells multiply and form a biocake. Many researchers propose that SMP and EPS secreted by bacteria also play important roles in the formation of biological foulants and cake layer on membrane surfaces (Liao et al., 2004; Ramesh et al., 2007). The structure of biofilm is known to vary greatly, ranging from patchy, discontinuous colonies to thick, continuous films, and it affects characteristics such as resistance to antimicrobial agents depending on environmental conditions and mass transport phenomena (Lewandowski, 2004).
Microbial EPS and SMP
Microbial biofilm on the membrane surface decreases the flux but can be controlled by backwash. In general, two kinds of substances are produced during biological activities, which cause fouling: extracellular polymeric substances (EPS) and soluble microbial products (SMP). EPS are insoluble macromolecules polymerized by microorganisms or substances like capsules, gels, and humic acid. SMP are produced by cell metabolism or self-digestion, which can be considered to be soluble, large molecules (Tarnacki et al., 2005). They form colloidal substances and gradually accumulate in the membrane hole, decreasing the effective pore size of the membrane.
TECHNIQUES USED TO STUDY THE BIOFOULED MEMBRANE
The biofouled membrane can be visualized by techniques (Table 2) such as scanning electron microscopy (SEM) (Miura et al., 2007; Meng et al., 2010), confocal laser scanning microscopy (CLSM) (Bjorkoy & and Fiksdal, 2009; Lee et al., 2010), atomic force microscopy (AFM) (Mi & Elimelech, 2010), and direct observation through the membrane (DOTM). DOTM and CLSM have been widely used to characterize membrane biofouling (Li et al., 2003; Jin et al., 2006; Zhang et al., 2006a; Hwang et al., 2007; Lee et al., 2007; Meng et al., 2010). The DOTM approach was developed to record the deposition behavior in simple cases of latex particles and flocs (Li et al., 2003; Zhang et al., 2006a). Zhang et al. (2006a) used DOTM to examine the interactions between bioflocs and the membrane surface. The images showed that the bioflocs could move across the membrane surface by rolling and sliding. More recently, CLSM has become a powerful approach for characterization of membrane biofouling; it can present the three 3D structure of fouling layer and can idetifiy the attached cells. Ng et al. (2006) applied CLSM to visualize the distribution of bacterial cells on the membrane surface, and found that bacteria were extensively present on the fouled membrane. The combination of image analysis and CLSM can quantify or visualize the structural design of the biocake layer (Lee et al., 2008). Yun et al. (2006) characterized the biofilm structure and analyzed its effect on membrane flux in MBR for dye wastewater treatment. They found that membrane filterability was narrowly associated with the structural parameters of the biofilms (i.e., porosity, biovolume). The study of biofouling through the application of these techniques is useful for understanding of the floc/cell deposition process and the structure of the cake layer.
A handful researches have been carried out to study the microbial community structures and colonization microbes on membranes in MBRs (Jinhua et al., 2006; Zhang et al., 2006b; Miura et al., 2007; Bereschenko et al., 2010). The microbial community structures can be studied using methods such as Fluorescence In Situ Hybridization (FISH) and polymerase chain reaction denaturing gradient gel electrophoresis (PCR-DGGE) (Miura et al., 2007). Zhang et al. (2006b) found that microbial communities on membrane surfaces could be very distinctive from the suspended biomass. Bereschenko et al. (2010) used a combination of molecular ((fluorescence in situ hybridization [FISH], denaturing gradient gel electrophoresis (DGGE), and microscopic (epifluorescence, field emission scanning electron, and confocal laser scanning microscopy)) techniques to analyze the architecture, composition, abundance, and 3D structure of biofilm communities. A unique role of Sphingomonas spp. was observed in the initial formation and maturation of biofilms on the reverse osmosis membrane. Jinhua et al. (2006) reported that Proteobacteria more selectively adhered and grew on membranes than other microorganisms, and that the deposited cells have higher surface hydrophobicity than the suspended sludge. The high shear stress provoked by aeration can choose the deposition of cells. Some cells can be detached easily by the shear stress, but others adhere tightly to the membrane surface. The discerning deposition of the cell depends on the affinity of cells to membranes. Moreover, due to the anoxic condition in the cake layer, the temporal change of microbial community structure would take place. We can view that few of the bacteria in the sludge should play an pivotal role in membrane biofouling. Understanding the deposition behavior of bacterial cell/flocs and mechanisms of cell attachment in MBRs will be vital for the development of appropriate biofouling control strategies in the future.
FACTORS OF FOULING
The factors affecting membrane fouling can be classified into five groups (Le-Clech et al., 2006): microbial attachment, membrane materials, biomass characteristics, feedwater characteristics, and operating conditions. For a membrane reactor process, the fouling conduct is directly concluded by sludge characteristics and hydrodynamic conditions. However, operating conditions (i.e., sludge retention time, hydraulic retention time, and food to microorganism ratio) and feedwater have indirect effects on membrane fouling. In this review paper, the major factors, including bacterial cells, EPS, SMP, and hydrodynamic conditions, are discussed.
Microbial attachment (Repeat)
Microbial attachment-based biological methods have been widely applied for wastewater treatment (Liu & and Tay, 2002). Membrane systems have been extensively used for water recycling due to such systems' compact design and high-quality product water, as reviewed by Visvanathan et al. (2000). It should be realized that the main drawback hindering the wide application of membrane systems is membrane biofouling caused by microbial attachment to the membrane surface, which leads to decreased membrane flux and increased filtration pressure and subsequently to increased operation cost due to frequent cleaning and replacement of the clogged membranes (Le-Clech et al., 2006; Liao et al., 2004; Visvanathan et al., 2000).
EPS and SMP
Extracellular polymeric substances in either attached or soluble form are at present meditated as the main reason for membrane fouling in MBRs. Bound EPS comprise proteins, polysaccharides, nucleic acids, lipids, humic acids, etc., that are located at or outside the cell surface (Reference). Soluble EPS and SMP are identical. SMP can be defined as organic compounds that are released into solution from substrate metabolism (usually with biomass growth) and biomass decay (Barker & and Stuckey, 1999). Thus, SMP can be subdivided into two categories (Laspidou & Rittmann, 2002): substrate-utilization-associated products (UAP), which are produced directly during substrate metabolism, and biomass-associated products (BAP), which are formed from biomass, presumably as part of decay. A theory for EPS and SMP was put forward by Laspidou & and Rittmann (2002), who pointed out that cells use electrons from the substrate to build biomass, and they produce UAP and bound EPS in the process. Some SMP can be utilized by active biomass as electron donors and others can be adsorbed by the biomass flocs and then become bound EPS. In addition, the generation of UAP and bound EPS is in proportion to substrate utilization.
Effect of EPS on membrane fouling
Bound EPS have been found not only as main sludge floc components keeping the floc in a 3D matrix, but also as critical membrane foulants in MBR systems. Cho et al. (2005b) found a close association between the attached EPS and the specific cake resistance. Ahmed et al. (2007) also observed that as attached EPS concentration increased, the specific cake resistance rose, and this inevitably resulted in the rise of TMP. A recent study by Ji & and Zhou (2006) indicated that both composition and quantity of attached EPS on the membrane surface influenced membrane fouling, and the total biopolymers in sludge suspension played a more important role than bound EPS in reflecting the extent of membrane fouling. Ramesh et al. (2006) separated bound EPS into tightly bound EPS and loosely bound EPS, and confirmed that the resistance was primarily caused by the loosely bound EPS. The EPS that loosely bound correlated with the activity of flocculation and sedimentation processes (Li & and Yang, 2007). A number of researches, however, have declared that bound EPS had minute correlation with membrane fouling. Rosenberger & Kraume (2003) observed that, contrary to some literature, no impact of bound EPS on filterability could be seen. Instead, the soluble EPS or SMP were found to have great impact on the filterability of sludge. This was confirmed by a more recent work reporting no clear relation between bound EPS and membrane fouling as the concentration of bound EPS was smaller than 10 mg/g SS (Yamato et al., 2006). Several explorations have been carried out to gain a better understanding of sludge characteristics and their effects on membrane fouling (Germain et al., 2005; Fan et al., 2006). The above mentioned inquiries showed that activated sludge has very complicated impressions on the membrane fouling process. Bound EPS cannot be considered the main reason for membrane fouling, even though they have huge effects on sludge characteristics and membrane fouling.
In spite of the verity that the research results on bound EPS are distinct from each other, it must be pointed out that bound EPS concentrations are closely connected to sludge characteristics, such as sludge flocculation ability, hydrophobicity, sludge viscosity volume index, and surface charge. Therefore, in view of the critical function of bound EPS in sludge features and membrane fouling, bound EPS should be restrained to mitigate membrane fouling.
Effect of SMP on membrane fouling
In fact, because of the complex nature of sludge suspension, the behavior of fouling cannot be assigned entirely to bound EPS. The effect of SMP on MBR fouling has attracted much attention (Rosenberger et al., 2005, 2006; Jeong et al., 2007; Drews et al., 2007; Paul & Hartung, 2008). Due to the membrane rejection, SMP accumulate more easily in MBRs, which results in the low filterability of the sludge suspension. Geng & and Hall (2007) found that the distribution of floc size and the amount of soluble EPS in the mixed liquor were the most vital characteristics that influenced the fouling tendency of sludge. Moreover, some studies have shown that polysaccharide-like substances in SMP contribute to fouling more than protein-like substances (Rosenberger et al., 2006; Yigit et al., 2008).
Since SMP have been identified as membrane foulants, research on SMP has become one of the main areas in the research on membrane fouling. Rosenberger et al. (2006) found that the SMP of sludges had an impact on fouling and caused the difference in membrane performance between two similar MBRs. Iritani et al. (2007) observed that the input of the supernatant to the membrane fouling of an anaerobic activated sludge is nearly 100%, indicating that SMP constitute the main factor in filtration of activated sludge. Lyko et al. (2007) studied the SMP in supernatant and permeate as well as attached EPS from fouled membranes in the full-scale MBR, and found a vital effect of soluble humic substances and carbohydrates with metal cations on membrane fouling. They also found that dissolved organic carbon (DOC) was choice for complex and costly measurements of SMP components (Lyko et al., 2008). These researches suggest that the presence of SMP in MBRs knocks on membrane fouling substantially, and SMP concentration and SMP composition would establish the fouling tendency.
Moreover, the presence of SMP in MBR effluent is a concern in the execution of post-treatment for water recycling, and the release of SMP rich water brings further problems to the native environment, e.g., the presence of dissolved organic nitrogen. Table 3 presents the relationship between various biofouling factors and membrane biofouling.
EFFECT OF OTHER CONDITIONS ON MEMBRANE FOULING
Submerged MBRs are more widely used than cross-flow MBRs in research and real applications, the majority of current researches focus on the reduction of aeration demand. In a submerged MBR, aeration create shear stress, which supplies oxygen to the biomass, scours the membrane surface to lessen the membrane fouling, and keeps the solids in suspension. The aeration can be used to produce a shear stress on the surface of the membrane without requiring a recirculation pump. Hong et al. (2002) observed the outcome of aeration on cake removal and suction pressure using a pilotscale submerged MBR, and concluded that aeration was a significant factor governing the filtration conditions.
Prior researches (Han et al., 2005) indicated that the cake removing capability of aeration did not increase with the increase in the airflow rate. Aeration plays an important role in determining both the membrane filtration or fouling control and the size of the sludge flocs. A high aeration rate certainly can reduce sludge attachment to the membranes, but it also has a significant influence on the biomass characteristics. Too high aeration potency will lead to breakage of sludge flocs and production of SMP. Under high aeration intensity, the solutes and colloids will become the major membrane foulants (Fan & Zhou, 2007), because the resistance of colloids and solutes cannot be decreased efficiently by increasing shear stress. The high aeration rate can improve shear-induced diffusion and inertial lift. The potency of aeration is likely to have a very complicated effect on MBR efficiency.
In addition to attached EPS, SMP, and hydrodynamic conditions discussed above, attempts have been made to control fouling or modify sludge by using ultrasound, ozone, and electric field (Chen et al., 2007; Huang & and Wu, 2008; Sui et al., 2008; Wen et al., 2008). Experimental results have indicated that ultrasound can control membrane fouling efficiently, although membrane damage may occur under some conditions (Wen et al., 2008). One interesting method is the use of an electric field, which can prevent the sludge flocs and colloids from depositing onto the membrane surface. In addition, attempts also have been made to control MBR fouling by developing novel filtration modes and/or backwashing conditions (Wu et al., 2008).
BIOLOGICAL CONTROL OF BIOFOULING
So far, comprehensive research has been undertaken to explore the possible methods to prevent or reduce membrane biofouling. The traditional methods for reducing membrane biofouling are mainly based on physico-chemical principles, such as modification of the solid surface, optimization of operation conditions, regular physical and chemical cleaning, etc. (Le-Clech et al., 2006; Ramesh et al., 2006), and may not be effective and energy efficient. Recently, various novel biological methods for controlling microbial attachment and membrane biofouling have been developed, and there are some recent reviews on quorum sensing (QS)-associated biofilm control and biofouling, especially marine biofouling (Dobretsov et al., 2009).
Control by quorum sensing
Microorganisms can use quorum sensing (QS) to coordinate their communal behaviors, e.g., biofilm formation, swarming, motility, etc. (Williams et al., 2007; Kappachery et al., 2010). Quorum sensing can happen within individual bacterial species as well as between several species. Several bacteria have been reported to coordinate community behavior and thereby biofilm formation through cell to cell communication or QS mediated by small, diffusible signals (Hentzer & and Givskov 2003). It is also suggested that research into QS inhibition will provide some means for controlling the growth of biofilm without the use of growth-inhibitory agents that unavoidably select for resistant organisms (Richards & and Melander 2009). Compounds like salicylic acid (Rosenberg et al., 2008), urosolic acid (Ren et al., 2005), cinnamaldehyde (Brackman et al., 2008), extract from garlic (Bjarnsholt et al., 2005), and cranberries have all shown various degrees of antibiofilm properties against a number of microorganisms in various studies. Furanones isolated from the marine red alga Delisa pulchera are among the most extensively studied classes of natural compounds with respect to their role in inhibiting biofilm. Ponnusamy et al. (2010) recently reported the role of 2(5-H) furanone in suppressing biofilm formation by environmental strains of bacteria isolated from fouled RO membrane.
The QS-coordinated process is achieved by producing, releasing, and detecting small signal molecules known as autoinducers (AIs). Increasing bacterial density gives rise to an accumulation of AIs. Once the critical AI concentrations are achieved, the regulator proteins are triggered and further induce target DNA sequences, leading to transcription of QS regulated genes, followed with changes of bacterial social behaviors (Hardie & and Heurlier 2008; Williams et al. 2007). It is known that cell-cell communication essential for biofilm formation is closely regulated by AIs. So far, three types of AIs have been identified, namely, oligopeptides, N-acylhomoserine lactones (AHL), and autoinducer-2 (AI-2) synthesized by LuxS. Oligopeptides and AHL are merely involved in cellular communication of Gram-positive and Gram-negative bacteria, respectively, whereas AI-2 is universal for interspecies communication of both Gram-positive and Gram-negative bacteria (Xavier & and Bassler 2003). It is believed that AI mediated QS systems play a prominent role in the regulation of microbial attachment and subsequent biofilm formation.
In view of the essential role of QS in biofilm formation, disruption or inhibition of the QS systems of microorganisms appears to be a promising alternative for controlling microbial attachment as well as membrane biofouling. Therefore, strategies to disrupt QS systems are discussed herein with emphasis on the control of microbial attachment and membrane biofouling.
The evidence shows that inhibition or disruption of QS systems (e.g., QS signal molecules, QS regulator gene) would be an effective means for controlling microbial attachment and membrane biofouling. However, the recent study by Yeon et al. (2009) showed that membrane biofouling due to a mixed culture of Gram-negative and Gram-positive bacteria could also be efficiently removed by the addition of AHL inhibitors, e.g., Acylase I as discussed earlier. Obviously, further study is strongly needed to test the QS inhibitors for biofouling control in industrial-scale membrane systems.
Enzymatic disruption of EPS
Bacterial EPS facilitate the formation of bioflocs in activated sludge and contribute to its structural, surface charge and settling properties (Houghton et al., 2001; Houghton & and Stephenson, 2002). Bioflocs are produced through the interaction of filamentous bacterial strains, and organic and inorganic particles, which are held together by EPS (Novak et al., 2001; Bala et al., 2008). EPS cannot be efficiently removed by traditional physical or chemical cleaning methods. However, EPS could be hydrolyzed by some specific enzymes, implying a novel means to control EPS-mediated microbial attachment and membrane biofouling. Extracellular proteins and polysaccharides are two main components of most EPS secreted by bacteria; thus, two main EPS-degrading enzymes, i.e., proteolytic enzymes for protein hydrolysis and polysaccharases for polysaccharide hydrolysis, have been applied for biofilm detachment (Loiselle & and Anderson 2003).
Some proteolytic enzymes (protease), such as the proteinase K, trypsin, subtilisin, etc., have been employed to remove established biofilms. Proteinase K, a wide-spectrum protease, has been commonly applied to disperse the established biofilm as well as to inhibit biofilm formation as it can effectively cleave the peptide bonds of aromatic, aliphatic, and hydrophobic compounds. For example, this enzyme can effectively remove the biofilm formed by a clinical isolate, S. aureus 383 (Chaignon et al., 2007), whereas 0.1 mg/ml proteinase K led to 98% of detachment of 72 clinical S. haemolyticus isolate biofilms (Fredheim et al., 2009). Trypsin, a serine protease, which hydrolyzes lysine and arginine peptides, can also disrupt the mature S. aureus 383 biofilm and partially disperse the S. epidermidis 444 biofilm (Chaignon et al., 2007). Subtilisin was also tested for its antifouling activity in polystyrene microplates for natural seawater (Leroy et al., 2008). In addition, Leroy et al. (2008) compared the antifouling potential of some commercial enzymes, such as Amano Protease A, papain, umamizyme, and subtilisin, and the other hydrolases in the adhesion of Pseudoalteromonas sp. D41. It was found that the most efficient protease was subtilisin, which had the ability to inhibit microbial adhesion as well as detach adhered bacteria. Poele & Van der Graaf (2005) used protease to remove biofouling on ultrafiltration (UF) membrane for wastewater treatment. Compared with the traditional cleaning method by alkaline, enzymatic cleaning by protease exhibited a much higher efficiency in removing biofouling, leading to a high-efficiency recovery of the permeate flux.
Moreover, enzymatic cleaning of the fouled inorganic UF membranes by whey proteins was tested by Arguello et al. (2002), and the results showed that over 90% of removal efficiency would be achievable. Some polysaccharases are currently available for inhibition of microbial attachment and membrane biofouling by disrupting the matrix structure of extracellular polysaccharides. Dispersin B, which hydrolyzes poly-Nacetylglucosamine, can efficiently cleave the biofilm matrix of S. epidermidis on the plastic surfaces (Kaplan et al., 2004). When Dispersin B was applied to the S. epidermidis biofilm developed in the PDMS microfluidic devices, most of the biofilm was detached from the solid surfaces (Lee et al., 2008). It was also found that polyurethane and Teflon catheters coated with Dispersin B exhibited strong antibiofilm ability (Kaplan et al., 2004). Consequently, the combined use of polysaccharases and proteases is recommended for a more efficient control of microbial attachment and membrane biofouling. In addition to the traditional prevention and cleaning technologies for membrane biofouling, enzymatic disruption of EPS appears to be a promising alternative for high-efficiency control of microbial attachment and membrane biofouling.
To date, extracellular DNA (eDNA) has been shown to function as a structural support to maintain biofilm architecture, and has also been known to serve as a vital factor to induce the development of biofilm (Vilain et al., 2009; Wu & Xi 2009). It is thought that eDNA present in biofilm matrix will contribute to biofilm stability due to its negative charge (Azeredo and Oliveira 2000). The study by Whitchurch et al., (2002) clearly demonstrates that "eDNA is required for the initial establishment of P. aeruginosa biofilms and biofilms formed by some other bacteria that specifically release DNA." It has been observed that eDNA may chip in to biofilm development in different ways, e.g., by providing substrate for other cells, keeping the biofilm structure by formation of a matrix, and facilitating the exchange of genetic substances (Wu & and Xi 2009). A study on S. epidermidis biofilm reveals that eDNA was an important factor for the initial development of S. Epidermidis biofilm on polystyrene and glass surfaces under static or hydrodynamic conditions (Qin et al., 2007). These suggest that disruption of eDNA would lead to detachment or dispersal of biofilms.
It was observed that reductions of the 6- and 12-h biofilms were more distinct than the 24-h biofilm. These indicate that DNase I would be more effective for controlling microbial attachment or biofilm development than disrupting mature biofilm. Therefore, DNase may provide a feasible and efficient means for controlling microbial attachment as well as membrane biofouling. Lastly, it should be realized that enzymatic disruption of EPS would have some inherent drawbacks that, to some extent, may limit its large-scale application. Enzyme is unstable in the environment, e.g., the activity of enzyme would be reduced or even totally lost when the optimal pH value is altered. Moreover, enzymes are highly temperature sensitive. High temperature may make enzymes denatured, while low temperature could decrease the enzymatic activity substantially. In addition, enzymes are also sensitive to elevated salt concentration. Compared with the chemical methods for EPS removal, the enzymatic method is nontoxic and environmentally friendly, while the problem associated with the bacterial resistance to the antibouling agents would be minimized.
Control by nitric oxide
Nitric oxide (NO), as a biologically ubiquitous gas molecule, can antagonize useless cell proliferation (Sarti et al., 2002). This specific molecule has been identified as an important messenger molecule that regulates biofilm dispersal. For example, addition of NO at low, nontoxic concentrations will result in the dispersal of P. Aeruginosa biofilm (Barraud et al., 2006). Similar to the QS inhibitors, NO attenuates the bacterial infection by signaling mechanisms rather than by the toxic effect. NO can inhibit the infection of K. pneumonia, a kind of lung bacteria, and efficiently protect the lung system (Tsai et al., 1997). Cyclic di-GMP is probably involved in NO-mediated biofilm dispersal (Barraud et al., 2006). Since the gene encoding for regulators of the Cyclic di-GMP commonly exists in bacteria (Delgado-Nixon et al., 2000; Romling et al., 2005; Ryan et al., 2006) and several factors associated with NO production and response system seem to be preserved across microorganisms (Gusarov et al., 2008; Zumft, 1993), the NO-mediated dispersal mechanism may exist in a wide spectrum of microbial species. It has been reported that NO at a flux of 30 pmol/(cm2s) could efficiently reduce adhesion of S. aureus, Staphylococcus epidermidis, and E. coli, with a reduction rate of 96%, 48%, and 88%, respectively (Charville et al., 2008). These in turn suggest that NO would have a comparably universal effect on dispersal of sessile bacteria, including both Gram-positive and Gram-negative bacteria. S. aureus and S. epidermidis biofilms were repressed in the presence of nitrite (NO2âˆ’) presumably due to the involvement of NO (Schlag et al., 2007), whereas nitric oxide was found to effectively induce the dispersal of P. aeruginosa biofilm (25 to 500 nM of the NO donor, sodium nitroprusside) (Barraud et al., 2006). Besides the dispersal of pure culture biofilms of some stains, the dispersal of multispecies biofilms from water distribution and treatment systems can also be induced by various NO donors at picomolar or nanomolar levels, e.g., addition of the NO donors would result in an average of 63% decrease of total biofilm (Barraud et al., 2009).
More importantly, combined low-dose NO and antimicrobial agents can greatly enhance biofilm removal. Compared with sole action of chlorine, a 20-fold increase in the efficiency was observed when NO was applied together with chlorine in removing multispecies biofilms (Barraud et al., 2009). In addition, pretreatment of biofilm grown on the RO membrane by 100 nM of SNP that served as the donor of nitric oxide led to a twofold increase in the biofilm removal efficiency in contrast with the control (Barraud et al., 2009).
As discussed above, NO has great application potential in controlling microbial attachment and membrane biofouling. On the other hand, it should be realized that NO has low solubility in water and is not stable as it can be easily oxidized. As a result, the direct addition of NO into aquatic solution would have low efficiency of biofouling control (Cai et al., 2005). Nevertheless, many NO donors, including the enzymatic and nonenzymatic NO donors (e.g., SNP, SIN-1, sodium nitrite, SNAP, diazeniumdiolate, etc.) have been proven to be efficient. This implies that the use of the NO donor would have the same effect as the direct addition of NO on dispersal of biofouling.
Cell wall hydrolases
The hydrolytic enzymes of cell walls, such as lysozyme, can damage bacterial cell walls. Consequently, they have been employed to prevent microbial attachment and biofilm formation onto solid surfaces. Lysozyme-immobilized polymeric packaging films exhibit bacterial growth-inhibiting properties (Conte et al., 2006). Caro et al. (2009) coated the glycosidase hen egg white lysozyme to a stainless steel surface and found that microbial attachment on the substrate coated with lysozyme could be completely inhibited as compared with the control without immobilization of lysozyme. The antimycotic protein lysozyme is environmentally friendly and acts more specifically than the traditional biocides for biofilm control and has been used for the control of Candida biofilm development on denture acrylic (Samaranayake et al., 2009). The 48-h-old Candida biofilms were exposed to various lysozyme concentrations of 60 to 240Î¼g/ml for 24 h. It was found that the biofilm activity was reduced by 28.2-69.6% compared with the control free of lysozyme, whereas the degree of biofilm cell degradation was proportionally related to the applied lysozyme concentration.
Liposome-mediated drug delivery has been used due to the lipid bilayer of liposomes. To improve the lysozyme effect on biofilm control, Xu & Wang (2009) synthesized stable ultrafine lysozyme liposome beads with an average diameter of 80 to100 nm, and the tests for biofilm control were conducted with free lysozyme and the synthesized lysozyme liposomes under the same conditions. It was shown that about 86.5% of the total biofilm biomass could be removed by lysozyme liposomes, compared with only 62.4% by free lysozyme. Consequently, lysozyme or capsulated lysozyme could be a safe and highly effective agent for controlling microbial attachment and disrupting established biofilms. It should be pointed out that lysozyme, as a type of hydrolase, is not stable in the environment, and its efficiency would depend on pH, temperature, ionic strength, etc.
Control of microbial attachment and membrane biofouling by energy uncoupling
Uncoupling of electron transport or oxidative phosphorylation is an effective way to inhibit energy production, which can be achieved by adding various chemical uncouplers. Uncouplers of oxidative phosphorylation are typically weak acids with substantial lipid solubility and can carry protons across the cellular membrane. Once inside the membrane matrix, the higher pH causes the uncoupler to deprotonate. As a result, the uncoupler has the effect of transporting hydrogen ion back into the matrix, bypassing the F0 proton channel, and thereby preventing ATP synthesis (Zubay, 1998). Jiang & Liu (2010) investigated the effect of energy uncoupling on aerobic granular sludge biofilms and found that, when aerobic granular sludge biofilms were exposed to 4 mg/l of 3,3â€²,4â€²,5-tetrachlorosalicylanilide (TCS), a typical chemical uncoupler, complete disintegration of aerobic granular sludge biofilms was observed. Klebensberger et al. (2006) also looked into the response of the P. aeruginosa strain PAO1 to sodium dodecyl sulfate and showed that the cells could form macroscopic aggregates at high energy supply, while no aggregates were observed if the energy supply was reduced by inhibiting respiration with KCN or carbonyl cyanide chlorophenylhydrazone (CCCP).
It thus appears that the energy metabolism of cells would play a role in biofilm development. EPS are believed to facilitate cell-to-cell interaction and can further strengthen microbial structure of biofilms through the formation of a polymeric matrix (Liu et al., 2004), while synthesis of macromolecules is highly ATP dependent. Jiang & Liu (2010) reported that the net synthesis of cellular ATP in the presence of TCS was reduced by 75% as compared with the control free of TCS. As a result, the extracellular polysaccharides and protein contents in biomass were reduced substantially, leading to failed formation of aerobic granular sludge biofilms. A similar phenomenon was also reported by Xiong & and Liu (2010).
As chemical uncouplers have been known to be capable of dissipating proton motive force (PMF), they may also cause cell autolysis. Therefore, it is reasonable to consider that microbial autolysis induced by chemical uncoupler would eventually inhibit microbial attachment or biofilm formation, i.e., deprivation of cell ATP synthesis not only disrupts the stability of biofilms but also inhibits the formation of biofilm. As discussed earlier, AI-2, a universal signalling molecule, can be produced from methionine and requires input of ATP (Chen et al., 2002; Stevenson and Babb 2002), whereas DPD, an AI-2 precursor, is biosynthesized from S-adenosylmethionine which is made from ATP and methionine by methionine adenosyltransferase. These imply that the AI-2 synthesis is energy associated. It has been shown that the ATP content in microorganisms was reduced by nearly 60% after a 2-h contact with DNP, known as a typical chemical uncoupler; subsequently the inhibited ATP synthesis would lead to reduced AI-2 synthesis and further reduced microbial attachment (Xu, 2009). In fact, the AI-2-mediated QS has been proved to be involved in the interspecies communication for both Gram-negative and Gram-positive bacteria. In addition, cell motility has been reported to be PMF dependent (Paul & Hartung, 2008). It appears from the above discussion that inhibition of the ATP synthesis by chemical uncouplers should be a feasible alternative for alleviating biofouling.
Control of microbial attachment and membrane biofouling by bacteriophage
An overview of the current scientific literature on "bacteriophages treatment" reveals several decades of "renaissance" on phages application in combating a large variety of bacteria in different experimental areas, such as foam formation reduction (Thomas et al., 2002), slime and biofilm control (Araki, 1986), plant diseases (Jones et al., 2007), medicine (Gorski et al., 2007), foodborne pathogen control, and detection (Hagens and Loessner, 2007), and shows phages to be an alternative agent (Gorski et al., 2009). The use of phages has been abandoned due to our modest knowledge on their molecular biology and physiology at the time and the discovery of antibiotics (Gorski et al. 2007).
Another important application of phage is to inhibit or disrupt biofilm development on solid surfaces such as membranes. Doolittle et al. (1995) successfully used bacteriophages to disrupt E. coli and P. aeruginosa biofilms, respectively. Corbin et al. (2001) reported that bacteriophage T4 at a multiplicity of infection of 10 would lead to a one-log decrease in biofilm density after a 90-min treatment. One of the studies also disclosed that phage F116 could pose a two-log decrease of P. aeruginosa biofilm cells (Hanlon et al., 2001). In addition, Curtin & and Donlan (2006) demonstrated that the infection of catheter by S. Epidermidis biofilm could be reduced by phage. Compared with the natural phage, an engineered phage with multifunctions can enhance biofilm dispersal.
Lu & Collins (2007) designed an enzymatic phage that can express Dispersin B, a polysaccharide-hydrolyzing enzyme. This engineered phage was found to have higher efficiency in the removal of biofilms, e.g., the reduction of cell count of biofilm by the engineered phage was two orders of magnitude higher than the nonengineered phage, leading to a nearly complete biofilm removal. In addition, a cocktail of multiple phages has been shown to be efficient in removing multispecies biofouling (Richards & Melander 2009). Goldman et al. (2009), probably for the first time, employed bacteriophage to control UF membrane biofouling. Results showed that the addition of phages could reduce microbial attachment to membrane surface by 40% on average, and the performance of the membrane bioreactor treating the effluents from the sewage treatment plant was improved significantly in terms of membrane permeability. The phages detected in permeative solution were only a few (1 to 10 PFU/100 ml), which might have originated from the penetration of phage across the membrane and/or possible permeate contamination by phage during sampling. Goldman et al. (2009) thought that the growth of phage found in permeative solution would not be an issue due to the absence of bacteria in UF filtration permeative solution. As noted by Goldman et al. (2009), "bacteriophages have a major advantage of continuous infection/multiplication as long as the host is present and grows" and "when multiple contaminant bacterial species are present, a combination of several phage types may be used to prevent concomitant adhesion and biofilm formation by pollutant bacteria." Although the study by Goldman et al. (2009) sheds lights on the potential application of bacteriophage in mitigating membrane biofouling without the use of antimicrobial agents, the specific-parasite characteristics of bacteriophage would eventually pose a challenge on its application in large-scale wastewater treatment, which needs to be taken into serious account in future scenario.
CONCLUSIONS AND FUTURE RESEARCH OUTLOOKS
This review paper discussed research on membrane biofouling, fouling factors, and some biofouling control strategies. In long-run operation, membrane biofouling needs more attention. To date, research on fouling mechanism is continuing, so as to properly design reactor systems, improve operational performance, and develop new membranes with anti-fouling properties and novel biofouling control strategies. According to the recent literature, future studies on membrane biofouling should take the following directions:
Research should focus on the membrane materials (like surface charge, hydrophobicity) in order to develop anti-fouling materials. The development of high-performance and low-cost organic membrane materials will help MBRs become more cost effective in the future.
At present, most operating concerns of MBRs have arisen from the pilot studies or previous operational experiences. A modest research has been done in modeling MBRs. The expansion of mathematical models for different MBR systems will help optimize operational performance, process control, and the ways of dealing with biofouling.
Studies on membrane fouling mechanisms should focus on identification and characterization of membrane foulants (i.e., chemical and biological components of foulants, bacteria community of the foulants). Cake formation, pore