Scholarly article on topic 'Membrane-Aerated Biofilm Reactor for the Treatment of Acetonitrile Wastewater'

Membrane-Aerated Biofilm Reactor for the Treatment of Acetonitrile Wastewater Academic research paper on "Environmental engineering"

Share paper
Academic journal
Environmental Science & Technology
OECD Field of science

Academic research paper on topic "Membrane-Aerated Biofilm Reactor for the Treatment of Acetonitrile Wastewater"

Environ. Sci. Technol. 2008, 42, 2099-2104

Membrane-Aerated Biofilm Reactor for the Treatment of Acetonitrile Wastewater

TINGGANG LI,1 JUNXIN LIU,1 RENBI B AI, * * AND F. S. WONG8 Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, Beijing 100085, People's Republic of China, Division of Environmental Science and Engineering, National University of Singapore, Singapore 117576, and Institute of Environmental Science and Engineering, Nanyang Technological University, Singapore 637723

Received August 30, 2007. Revised manuscript received December 13, 2007. Accepted December 18, 2007.

A membrane-aerated biofilm reactor (MABR) was studied forthe treatment of wastewater containing acetonitrile, a typical organonitrile compound. The MABR used hydrophobic hollow fiber membranes as the diffusers for bubbleless aeration as well as the carriers for biofilm growth. The objectives were to prevent the stripping-loss of acetonitrile during aeration and to achieve acetonitrile biodegradation plus nitrogen removal simultaneously in a single biolfilm on the membranes. In the MABR, oxygen and substrates were supplied to the biofilm from opposite sides, in contrast to those from the same side in conventional biofilm bioreactors. Operational factors, including surface loading rate and upflow fluid velocity in the bioreactor, on the effect of acetonitrile biodegradation performance were examined. The profiles of dissolved oxygen concentration and microbial activities and populations in the biofilm were investigated. Experimental results showed that, with the adapted microorganisms, removal of acetonitrile at approximately 98.6 and 83.3%, in terms of total organic carbon and total nitrogen, were achieved at a surface loading rate (in terms of membrane surface) of up to 11.29 g acetonitrile/ m2• d with an upflow fluid velocity of 12 cm/s and a hydraulic retention time of 30 h.The biofilm on the membranes developed an average thickness of about 1.6 mm in the steady state and consisted of oxic/anoxic/anaerobic zones that provided different functions for acetonitrile degradation, nitrification, and denitrification. The acetonitrile-degrading bacteria in the MABR appeared to secrete more extracellular polymeric substances that enhanced the attachment and development of the biofilm on the membranes. The study demonstrated the potential of using the MABR for the treatment of organonitrile wastewater.


Organonitriles are a known group of toxic compounds that are classified as priority pollutants and as carcinogenic and mutagenic (1-3). Organonitriles are widely used in labora-

* Corresponding author phone: (65) 6516 4532; e-mail:

f Chinese Academy of Sciences.

* National University of Singapore.

8 Nanyang Technological University.

tories as solvents and extractants. These compounds are also extensively used in the industry in the synthesis of pharmaceuticals, drug intermediates, plastics, rubber, herbicides, and pesticides (4-6). As an example, acetonitrile, a typical organonitrile compound, has been reported to have a global industrial consumption of more than 4 x 1041 in 2001 (7). Consequently, wastewaters from the various usages of organonitriles often contain high contents of organonitrile compounds. These wastewaters need to be effectively treated before their discharge into the environment so as to minimize the effect on public and environmental health. Although technologies such as ozonation andphotocatalytic oxidation have been applied to the degradation of organonitrile compounds in wastewater, these treatment technologies are usually expensive and are often found to generate toxic secondary pollutants (8). In recent years, there have been a number of studies in literature demonstrating the feasibility of using a biological process for the treatment of acetonitrile wastewater (9,10). So far, most of these studies have focused on the aerobic process because it was found to have much better biodegradation performance for acetonitrile than the anaerobic one. The prospect of using the aerobic process with conventional aeration methods, however, can be limited due to the high volatility of organonitrile compounds and hence the potentially high stripping-loss of the pollutants into the atmosphere (3, 9-11). In addition, under the aerobic condition, biodegradation of organonitriles usually results in the generation of high concentrations of NH4+-N and/or NO3—-N in the effluent that would need additional treatments, including the anaerobic one, for the further removal of nutrient (i.e., nitrogen) before discharging (7).

Therefore, it is of a great research and practical interest to develop a more desirable biological treatment process that can not only effectively degrade organonitriles without the concerned stripping-loss but also simultaneously provide the additional treatment function for nutrient removal in the same process. This study investigates a membrane-aerated biofilm reactor (MABR) for the treatment of acetonitrile wastewater (acetonitrile is used as a model organo-nitrile pollutant). The MABR used gas-permeable hydrophobic hollow fiber membranes for bubbleless aeration and also as the carrier of the biofilm. Such a configuration can provide high oxygen transfer efficiency or even complete utilization of the supplied oxygen for the biodegradation process, as compared to the conventional bubbled aeration systems, and hence have the potential advantage to lower the treatment operational cost (12,13). The bubbleless aeration would also prevent or even eliminate the possible stripping-loss of acetonitrile encountered in the conventional aerobic process (14,15). Through the control of the transmembrane oxygen supply pressure, the MABR process can allow the growth of biofilm on the membrane into the oxic—anoxic— anaerobic zones, which provides different treatment functions for acetonitrile biodegradation as well as nitrification and denitrification in a single Biofilm (13, 16). The specific reactions involved in the processes maybe given, in general, as shown below.

Nitrile hydratase

The objective of this work is to examine the treatment performance of acetonitrile wastewater in the MABR inoculated with adapted microorganisms with acetonitrile as the carbon and nitrogen source (17) and operated under various



10.1021/es702150f CCC: $40.75 © 2008 American Chemical Society Published on Web 02/08/2008

Nitrification Denitrification

! NH4+ -► N02

surface loading rates and bulk upflow fluid velocities. Analyses were also conducted to characterize the biofilm formed on the membranes, the profiles of oxygen concentration and microbial activities and populations in the biofilm.

Experimental Section

Reactor. The schematic diagram of the laboratory-scale MABR is shown in Figure 1. An aeration membrane module consisting of polypropylene hollow fibers (320 ¡m o.d. and 200¡m i.d.) (Zenon, Singapore) in a dead-end configuration was placed in an enclosed polypropylene cylinder reactor (650 mm long, 55 mm i.d.). The working volume of the reactor was 1.42 L, with a specific membrane surface area of 84.5 m2/m3. The MABR was inoculated with acclimated microbial consortium with acetonitrile-degrading microorganisms and was fed with acetonitrile as the initial carbon and nitrogen source in a synthetic mineral salt (SMS) medium with the trace-element composition as described in details elsewhere (17).

Reactor Operation. The key dynamic operating conditions examined in this study were the surface loading rate and the bulk upflow fluid velocity in the reactor. The oxygen supply was maintained at 4.3 L/m2 • h in terms of the internal surface areas of the hollow fiber membranes (with a transmembrane oxygen pressure of 13.78 kPa). The hydraulic retention time (HRT) in all experiments was fixed at 30 h based on the result of some preliminary biodegradation tests. Dependent upon the surface loading rate, acetonitrile influent concentrations in the feed varied in the range of 0.332-1.393 g/L. After each change in the surface loading rate or the upflow fluid velocity, the system was allowed to reach a steady state of operation, indicated by the stable outlet acetonitrile concentration. Once the steady state was achieved, samples were taken and analyzed for representative data of the system.

General Analysis. Mixed liquor suspended solid (MLSS) and volatile suspended solid (VSS) were measured in accordance with the standard methods (18). Total organic carbon (TOC) and total nitrogen (TN) were determined through a TOC-VCSH plus nitrogen analyzer (Shimadzu, Japan). The analyses of ammonia, nitrite, and nitrate were conducted with the Hach test kits together with a UV-vis spectrophotometer (DR5000, Hach, USA). Acetonitrile, ac-etamide, and acetic acid concentrations were measured through a gas chromatograph (6890N, Agilent, USA) as described previously (17). Biofilm thickness was determined through a noninvasive method (19), and the images were captured through a VH-Z75 microscope (Keyence, Japan) via a charge-coupled device that was connected to a computer. Dissolved oxygen (DO) concentrations were determined using a Clark-type microelectrode (Unisense, Aarhus, Denmark) with a tip diameter of 10¡m and a response time of less than 5 s. The DO microelectrode was inserted into the biofilm through the control of a micromanipulator (MMO-223, Japan) with a minimum depth step of 10 ¡m. Measurements of DO concentrations and biofilm thicknesses were repeated three times at each location, and the average values were reported in this paper.

The biofilm samples were collected from the reactor during the steady-state operation with a surface loading rate of 11.29 g acetonitrile/m2 • d and an upflow fluid velocity of 12 cm/s. Sample preparation was performed according to the protocol described by Cole et al. (20) and LaPara et al. (13). An appropriate length of the hollow fiber membrane

with biofilm was cut with a razor blade and was then immediately frozen at -15 °C in a freezer. The biofilm on each membrane sample was subsequently cut and separated into slices of 200 ¡m thickness using a combined cryostat/ microtome (CM 3050, Germany) at -20 °C. Biofilm mass from each slice was placed in a sterile centrifuge tube and stored at -20 °C for further analysis.

Extracellular Polymeric Substances (EPS) Extraction and Quantitative Analysis. The method for EPS extraction was a modified one from that of Liu and Fang (21), Azeredo et al. (22), and Zhang et al. (23) in order to achieve a better extraction. The biofilm mass from a piece of the membrane sample was made into a 5 mL suspension by the addition of sterile deionized water to determine the average total EPS content. The suspensions were centrifuged (4000g, 10 min, 4 °C), and the supernatants were decanted into respective sterile tubes. The supernatants were then recentrifuged at a much higher speed (13 200g, 20 min, 4 °C) for complete removal of the suspended solids. The supernatants from these physical extractions contained soluble polysaccharide and protein and were analyzed for the soluble EPS content. The biofilm mass from the above centrifugation was resuspended into sterile deionized water to the volume of 5 mL again. A 6 ¡L amount of formaldehyde (37%) was first added into it, and the content was kept at 4 °C for 1 h. Then, a 0.5 mL amount of NaOH (1 M) was added into it, and the content was kept at 4 °C for another 3 h. Finally, the suspensions were centrifuged (13 200g, 20 min, 4 °C), and the supernatants, through these chemical extractions, contained the bound polysaccharide and protein, and were analyzed for the bound EPS content. The total EPS was taken as the sum of the measured soluble and bound EPS. The polysaccharide content was quantified by the phenol-sulfuric acid method with glucose as a standard (24), and the protein content was determined using the Bradford coomassie blue method with bovine serum albumin (BSA) as a standard (25). A UV-vis spectrophotometer (Jasco V-550, Japan) was used for the analyses of polysaccharide at the 490 nm wavelength or of protein at the 592 nm wavelength. The sum of the polysaccharide and protein was taken as the EPS content.

Microbial Activity Analysis. Batch studies were performed to determine the specific activities of ammonia-oxidizing bacteria, denitrifying bacteria, and acetonitrile-degrading bacteria in the membrane-aerated biofilm. Specific ammonia-oxidizing (or nitrification) rate, denitrification rate, and acetonitrile-degrading rate were determined based on the maximum rates of substrate utilization per biomass unit of the biofilm samples from each slice, and were expressed as mg NH4+-N/g-VSS• h, mg NO3--N/g-VSS• h, and mg acetonitrile/g-VSS • h, respectively. A 1 mL portion of the biofilm sample was inoculated to 49 mL of the SMS medium in a 100 mL flask. The content was incubated at 28 ( 1 °C and was shaken at 180 rpm on a rotary shaker. The initial substrate concentrations were 50 mg/L of NH4+-N (in terms of 191.1 mg/L of NH4CD and 200 mg CaCO3/L of alkalinity (in terms of 168 mg/L of Na2CO3) for ammonia-oxidizing bacteria activity, or 50 mg/L of NO3--N (as 303.6 mg/L of NaNO3) and 400 mg/L of COD (supplied as CH3COONa) for denitrifying bacteria activity, or 200 mg/L of acetonitrile for acetonitrile-degrading bacteria activity. Any DO in the denitrifying medium was first removed by purging the medium in the flask with nitrogen gas. The flask (50 mL) for the determination of specific acetonitrile-degrading rate was sealed using aluminum caps with PTFE/silicone septum to prevent acetonitrile volatilization loss and was supplied with the same DO concentration as that for the biofilm in the MABR. Samples were taken at an interval of 30 min, and the substrate concentrations (NH4+-N, NO3--N, and acetonitrile) were determined according to the methods described earlier.

FIGURE 1. Schematic diagram of (a) the membrane-aerated biofilm reactor (MABR) and (b) the mass transfer pattern of oxygen and acetonitrile substrate in the biofilm.

Microbial Enumeration Analysis. The population of ammonia-oxidizing bacteria and denitrifying bacteria in the biofilm samples were enumerated with the most-probable-number (MPN) method. Basal medium for the MPN enumeration of ammonia-oxidizing bacteria was prepared by adding (NH4)2SO4, 200 mg; KH2PO4, 100 mg; MgSÜ4• 7H2O, 50 mg; CaCl2, 15 mg; and NaHCOs, 600 mg, in 100 mL of deionized water. The pH was adjusted to 8 using 1 M NaOH before being sterile filtered. Basal medium for the MPN enumeration of denitrifying bacteria was prepared by adding NaNOs, 1000 mg; CHaCOONa, 1700 mg; MgSO4 • 7H2O, 50 mg; and CaCl2,15 mg, in 100 mL of deionized water. The pH was adjusted to 7.2 using 1 M NaOH before being sterile filtered. A 1 mL amount of the biofilm from a specific slice was dispersed into the medium for 5 min as the inoculum source. Then, a series of 10-fold dilutions of the inoculum source were prepared with sterile deionized water, and 1 mL amount of each of the dilutions was transferred to an individual MPN tube containing 9 mL of the basal enumeration medium. Five replications were prepared for each dilution for the enumeration analysis, and the averages were reported. To increase the analysis reliability (26,27), we used a grown medium with high NH4+-N concentrations for the enumeration of ammonia-oxidizing bacteria (28) as well as a long incubation time of 5 weeks (29) at 28 ( 1 °C. The incubation time for denitrifying bacteria analysis was 1 week at 28 ( 1 °C.

The enumeration of acetonitrile-degrading bacteria in the biofilm samples was estimated through a series of 10-fold dilutions and then the examination of colony forming units (CFU) by the plate-counting method. Basal medium for the plate-counting enumeration of acetonitrile-degrading bacteria was prepared by adding 100 mg of acetonitrile and 2 g of agar in 100 mL of filtered sterile SMS medium. A 1 mL amount of the biofilm mass from a specific slice was dispersed into the medium for 0.5 min as the inoculum source. Samples with various dilutions were cultured on the agar plates and then incubated at 28 ( 1 °C for 5 d before the colony counting.

Results and Discussion

MABR Performance. By inoculating the adapted microorganism culture into the MABR, a thin membrane-aerated biofilm was observed with the naked eyes after approximately 1 week of operation. Acetonitrile removal was low and unsteady initially, due to the limited biofilm mass attached to the surfaces of the membranes. However, after 20 d, the MABR system reached a steady state under the conditions

of a surface loading rate of 3.41 ( 0.05 g acetonitrile/m2• d and an upflow fluid velocity of 12 cm/s, and a complete removal of acetonitrile was achieved in the MABR. After that, the MABR was operated under various surface loading rates and upflow fluid velocities.

Figure 2 shows the performance of the MABR, including acetonitrile removal efficiency, acetonitrile removal capacity, and acetonitrile remaining concentration in the effluent, under different surface loading rates (at the same upflow fluid velocity of 12 cm/s). For surface loading rates from 3.41 to 10.54 g/m2 • d, almost complete acetonitrile removal (>99%) was achieved. The removal efficiency, however, gradually decreased to 70.9 ( 4.62% when the acetonitrile surface loading rate was further increased from 10.54 to 14.29 g/m2 • d. The removal capacity was observed to increase with the increase in the surface loading rate and reached a maximum of 10.92 g/m2• d at the surface loading rate of 11.29 g/m2• d. Then, the removal capacity gradually reduced with the acetonitrile surface loading rates from 11.29 to 14.29 g/m2 • d examined. Although acetamide and acetic acid were only detected at small amounts in the effluents at high surface loading rates from 12.30 to 14.29 g/m2 • d, ammonia was found in all effluent samples at concentrations ranging from 34.2 to 52.7 mg/L, indicating that acetonitrile biodegradation indeed occurred (with 110-330 mg-N/L present in acetonitrile at the influent concentrations of 332-1392 mg/L). The pH of the effluents was in the range of 7.62-8.47 under all surface loading rates tested. The TN removal efficiency was found to be at about 84% in the case of a surface loading rate of 11.29 g/m2 • d in the MABR, and the corresponding NH4+-N concentration in the effluent was around 46 mg/L. The performance results of greater than 99% acetonitrile removal in the MABR at a surface load rate of up to 10.54 g/m2• d (i.e., 1.01 g/L• d), and a maximum removal capacity of 10.92 g/m2• d (i.e., 1.05 g/L• d) at a surface loading rate of 11.29 g/m2 • d (i.e., 1.10 g/L• d) are better than or comparable with results from other studies using packed bed reactors or continuously stirred tank reactors (7, 9, 10).

Figure 3 shows the experimental results obtained under a constant surface loading rate of 11.29 ( 0.36 g/m2 • d but with different upflow fluid velocities in the range of 1-12 cm/s. It is observed that a higher upflow fluid velocity resulted in a better performance, and a complete removal (100%) of acetonitrile was achieved at 12 cm/s. Acetamide and acetic acid were not detected in the MABR effluents. Ammonium was found in all effluent samples with a higher concentration at a higher upflow fluid velocity. Higher upflow fluid velocities

Surface loading rate (g acetonitrile/m d)

FIGURE 2. Profiles of acetonitrile removal efficiency, removal capacity, and acetonitrile remaining concentration in the effluents of the MABR under different surface loading rates (HRT was 30 h, and upflow fluid velocity was 12 cm/s). ■, acetonitrile removal efficiency; •, acetonitrile remaining concentration in the effluent; ▲, acetonitrile removal capacity. Vertical bars represent the standard deviation calculated from triplicates.

TABLE 1. Operation Performance and Biofilm Characteristics of the MABR7

Fluid velocity (cm/s)

FIGURE 3. Profiles of acetonitrile removal efficiency, removal capacity, and acetonitrile remaining concentration in the effluents of the MABR at different fluid flow velocities (HRT was 30 h, and surface loading rate was 11.29 g/m2• d). ■, acetonitrile removal efficiency; •, acetonitrile remaining concentration of the effluent; ▲, acetonitrile removal capacity. Vertical bars represent the standard deviation calculated from triplicates.

therefore appeared to enhance the performance of acetonitrile biodegradation in the MABR, possibly due to the improved mass transfer and biofilm activity.

The MABR was then operated under the conditions of a surface loading rate of 11.29 g/m2• d and an upflow fluid velocity of 12 cm/s in a steady state for 120 d. The operation performance and biofilm charactoristics of the MABR are summarized in Table 1. Particularly, the biofilm thickness gradually increased and stabilized at a final thickness of around 1600 ^m on the membranes, with a VSS content of 6.86 g/L. Analysis indicated a total EPS content of 107.2 mg/ g-VSS in the biofilm. This value is higher than 80.3 mg/g-VSS reported by Li et al. in a bench-scale aerated biological filter fed with starch (30). The EPS content can affect the spatial network structure of the biofilm, and hence can possibly affect the microbial activity in the biofilm. In this study, the DO concentration at the membrane surface (inner side of the biofilm) was approximately 7.97 mg/L, and that at the outside surface of the biofilm on the bulk liquid side in the MABR was zero. Hence, the outer region of the membrane-aerated biofilm where the acetonitrile concentration was the highest was most probably under an anaerobic condition. In an early study in a batch suspended growth reactor, we found that acetonitrile degradation was much slower under an anaerobic condition than under an aerobic one (17). The results from this study suggest that the performance of acetonitrile degradation by the adapted microorganisms in the biofilm of the MABR, even under partially anaerobic


acetonitrile feed concentration (g/L) acetonitrile loading rate (g/m2• d)0 acetonitrile removal capacity (g/m2• d)0 acetonitrile removal efficiency (%) TOC loading rate (g C/m2• d)0 TOC removal capacity (g C/m2• d)0 TOC removal efficiency (%) TN loading rate (g N/m2• d)0 TN removal capacity (g N/m2• d)0 TN removal efficiency (%) NH4+-N effluent concentration (mg/L) NO2--N effluent concentration (mg/L) NO3--N effluent concentration (mg/L) acetamide effluent concentration (mg/L) acetic acid effluent concentration (mg/L) biofilm thickness (wm) biofilm VSS (g/L) biofilm average total EPS content (mg/g VSS)


I.119 ( 0.046

II.29 ( 0.32 10.92 ( 0.41 96.7 ( 3.14 6.54 ( 0.24 6.45 ( 0.19

98.6 ( 3.75 3.83 ( 0.13 3.19 ( 0.12 83.3 ( 2.83

45.7 ( 2.4 1.4 ( 0.1 1.2 ( 0.08 not detected not detected 1670 ( 310 6.91 ( 0.37

107.2 ( 4.2

a At steady state under a surface loading rate of 11.29 g acetonitrile/m2• d and an upflow fluid rate of 12 cm/s (HRT = 30 h). b In terms of membrane surface area.

Distance from membrane (|im)

FIGURE 4. DO concentration profiles in the biofilm at steady state operation (surface loading rate was 11.29 g/m2• d, fluid flow velocity was 12 cm/s, and HRT was 30 h). Vertical bars represent the standard deviation calculated from triplications.

condition, remained satisfactory, in terms of the high removal efficiency and high substrate loading rate.

From the practical application point of view, the aerobic treatment of acetonitrile wastewater would produce effluent loaded with high concentrations of the metabolic products: NH4+-N and possibly NO3-N as well, which are known to be, at least partially, responsible for the eutrophication of fresh and marine water bodies if discharged. To reduce the nitrogen concentration of the effluent to the desired discharging level, additional nitrification and denitrification processes are usually needed in the treatment system (7). In this study with the MABR, the biofilm was stratified into oxic/anoxic/ anaerobic zones, and thus a single biofilm can achieve the functions of acetonitrile removal as well as nitrification and denitrification for nitrogen removal. Although the NH4+-N concentration in the effluents of the MABR may still be up to 45.7 mg/L under the experimental conditions examined, up to 86% of TN removal was indeed achieved in the MABR (see Table 1). For the purpose of complete nitrogen removal, sufficient ammonia-oxidation (or nitrification) in the biofilm by increasing the aerobic region of the biofiom may be necessary and this will be further investigated in the future.

The other advantage of the MABR system was in its bubbleless oxygen supply for acetonitrile biodegradation. The bubbleless aeration prevented the possible problem of stripping-loss of acetonitrile in the conventional practices of using bubbled aeration (31) or mechanically stirred aeration (9). The oxygen conversion rate with the bubbleless oxygen supply in the MABR may also approach 100% due to the complete utilization of the oxygen in the biofilm (12). This is in contrast with the oxygen conversion efficiency of 5-25% in the conventionally aerated systems (32).

It is of interest to examine how oxygen concentration actually distributed and acetonitrile degradation, nitrification, and denitrification rates occurred in the biofilm of the MABR. As shown in Figure 4, the DO concentration at the surface of the biofilm on the membrane side was about 7.97 mg/L, and the concentration decreased rapidly and almost linearly at first with the biofilm thickness from the membrane surface and then more gradually to zero at the outer region of the biofilm on the bulk liquid side in the reactor. On the basis of the DO concentration profile in Figure 4, much of the biofilm thickness on the bulk liquid side was under an anaerobic condition in this case. In other words, the biofilm in the MABR would indeed consist of oxic/anoxic/anaerobic zones from the membrane side toward the bulk liquid side.

Figure 5 shows the analyzed ammonia-oxidizing rate, denitrifying rate, and acetonitrile-degrading rate in the biofilm obtained from the microbial activity experiment. As expected, the highest ammonia-oxidizing rate of 7.62 mg N/g-VSS • h occurred at the oxygen inlet surface of the biofilm,

Distance from membrane (fim)

FIGURE 5. Variations of ammonia-oxidizing rate (SAOR), denitrifying rate (SDNR), and acetonitrile-degrading rate (SADR) in the biofilm at the steady operation (surface loading rate was 11.29 g/ m2• d, fluid flow velocity was 12 cm/s, and HRT was 30 h). SAOR; ■, SDNR; ▲, SADR. Vertical bars represent the standard deviation calculated from five replications.

and the rate declined rapidly with the biofilm thickness from the membrane surface and eventually became zero at the outer surface of the biofilm on the bulk liquid side. The denitrifying rate in the biofilm exhibited a hump-shaped variation with the biofilm thickness, from the lowest in the biofilm on the membrane side and increasing to the highest of 7.03 mg N/g-VSS• h at about 1000 ¡¡m of the biofilm thickness from the membrane surface in the biofilm. Then, the rate gradually reduced with the further increase in the biofilm thickness toward the outer surface of the biofilm on the bulk liquid side. Although the highly anaerobic condition in the biofilm on the bulk liquid side mostly favored denitrification, the nitrified compounds in the biofilm had the lowest concentration in this zone due to the nature of reactions and the unique mass transfer pattern in the biofilm of the MABR (see Figure 1b). As a result, the maximum denitrifying rate was achieved within the biofilm rather than at the outermost region of the biofilm on the bulk liquid side. The acetonitrile-degrading rate was observed to increase from the lowest to the highest from the oxic zone, through the anoxic zone, and to the anaerobic zone in the biofilm. This result may not be explained as a higher degradation rate of acetonitrile in a more anaerobic zone, but was rather, most probably, caused by the higher acetonitrile concentration in the biofilm on the anaerobic zone side (the acetonitrile concentration was the highest in the biofilm on the bulk liquid side and decreased through the anaerobic zone and anoxic zone to the oxic zone). The results in Figure 5 thus provide evidence confirming that the stratified biofilm indeed resulted in the different treatment functions of acetonitrile degradation, nitrification, and denitrification simultaneously in the single biofilm of the MABR.

To further elaborate on the results shown in Figure 5, the population profiles of the different types of bacteria in the biofilm may be further examined; see Figure S1 in the Supporting Information. Although the CFU method for enumerating bacteria may result in an underestimation because not all bacteria grow on the plates, the results are valid to show the relative significance because all analyses were conducted under the same conditions. The population profile of the different types of bacteria in Figure S1 followed a very similar change in the biofilm as the different types of reaction rates shown in Figure 5. This is not surprising as the reactions in Figure 5 were the direct results of the activities of the different types of microorganism populations in the biofilm. For process optimization, the spatial distribution of the different types of microorganisms, and therefore the reactions in the biofilm, may be controlled by varying the operational conditions such as

surface loading rate, upflow fluid velocity, and oxygen supply. Other studies also demonstrated the possible structural change of the biofilms as a result of the differences in the local substrate availability (33-35), or upflow fluid velocity outside the biofilm in the reactor (20). Future studies will be directed to further examine the influence of the operation parameters on the biofilm structures and functions for more effective treatment of organonitrile wastewater by the MABR technology.

Supporting Information Available

A figure of the population profiles of ammonia-oxidizing bacteria, denitrifying bacteria, and acetonitrile-degrading bacteria in the membrane-aerated biofilm at steady operation. This information is available free of charge via the Internet at

Literature Cited

(1) Ahmed, A. E.; Farooqui, M. Y. H. Comparative toxicities of aliphatic nitriles. Toxicol. Lett. 1982, 12, 157-163.

(2) Johannsen, F. R.; Levinskas, G. J.; Berteau, P. E.; Rodwell, D. E. Evaluation of teratogenic potential of three aliphatic nitriles in the rat. Fund. Appl. Toxicol. 1986, 7, 33-40.

(3) Nawaz, M. S.; Chapatwala, K. D.; Wolfram, J. H. Degradation of acetonitrile by Pseudomonas putida. Appl. Environ. Microbiol. 1989, 55, 2267-2274.

(4) Henahan, J. F.; Idon, J. D. Setting the world of nitrile chemistry afire. Chem. Eng. News 1971, 49, 16-18.

(5) Banerjee, A.; Sharma, R.; Banerjee, U. C. The nitrile-degrading enzymes: current status and future prospects. Appl. Microbiol. Biotechnol. 2002, 60, 33-44.

(6) Kao, C. M.; Chen, K. F.; Liu, J. K.; Chou, S. M.; Chen, S. C. Enzymatic degradation of nitriles by Klebsiella oxytoca. Appl. Microbiol. Biotechnol. 2006, 71, 228-233.

(7) Hâkansson, K.; Welander, U.; Mattiasson, B. Degradation of acetonitrile through a sequence of microbial reactors. Water Res. 2005, 39, 648-654.

(8) Nagle, N. J.; Rivard, C. J.; Mohagheghi, A.; Philippidis, G. Bio conversion of cyanide and acetonitril by a municipal sewage derived anaerobic consortium. In Bioremediation of inorganic; Hinchee, R. E., Eds.; Battelle, Columbus, 1995; pp 71-79.

(9) Manolov, T.; Hâkansson, K.; Guieysse, B. Continuous acetonitrile degradation in a packed-bed bioreactor. Appl. Microbiol. Biotechnol. 2005, 66, 567-574.

(10) Muñoz, R.; Jacinto, M.; Guieysse, B.; Mattiasson, B. Combined carbon and nitrogen removal from acetonitrile using algal-bacterial bioreactors. Appl. Microbiol. Biotechnol. 2005, 67,699707.

(11) Aronstein, B. N.; Maka, A.; Srivatava, V. J. Chemical and biological removal of cyanides from aqueous and soil containing systems. Appl. Microbiol. Biotechnol. 1994, 41, 700-707.

(12) Brindle, K.; Stephenson, T. The application of membrane biological reactors for the treatment of wastewaters. Biotechnol. Bioeng. 1996, 49, 601-610.

(13) LaPara, T. M.; Cole, A. C.; Shanahan, J. W.; Semmens, M. J. The effects of organic carbon, ammonia-nitrogen, and oxygenpartial pressure on the stratification of membrane-aerated biofilms. J. Ind. Microbiol. Biotechnol. 2006, 33, 315-323.

(14) Casey, E.; Glennon, B.; Hamer, G. Oxygen mass transfer characteristics in a membrane-aerated biofilm reactor. Biotechnol. Bioeng. 1999, 62, 183-190.

(15) Rishell, S.; Casey, E.; Glennon, B.; Hamer, G. Mass transfer analysis of a membrane aerated reactor. Biochem. Eng. J. 2004, 18, 159-167.

(16) Terada, A.; Hibiya, K.; Nagai, J.; Tsuneda, S.; Hirata, A. Nitrogen removal characteristics and biofilm analysis of a membrane-aerated biofilm reactor applicable to high-strength nitrogenous wastewater treatment. J. Biosci. Bioeng. 2003, 95, 170-178.

(17) Li, T. G.; Liu, J. X.; Bai, R. B.; Ohandja, D. G.; Wong, F. S. Biodegradation of organonitriles by adapted activated sludge consortium with acetonitrile-degrading microorganisms. Water Res. 2007, 41, 3465-3473.

(18) Standard methods ^for the examination of water and wastewater, 20th ed; American Public Health Association: Washington DC, 1998.

(19) Freitas dos Santos, L. M.; Livingston, A. G. Membrane-attached biofilms for VOC wastewater treatment. I: Novel in situ biofilm thickness technique. Biotechnol. Bioeng. 1995, 47, 82-89.

(20) Cole, A. C.; Semmens, M. J.; LaPara, T. M. Stratification of activity and bacterial community structure in biofilms grown on membranes transferring oxygen. Appl. Environ. Microbiol. 2004, 70, 1982-1989.

(21) Liu, H.; Fang, H. P. Extraction of extracellular polymeric substances (EPS) of sludges. J. Biotechnol. 2002, 95, 249-256.

(22) Azeredo, J.; Lazarvo, V.; Oliveira, R. Methods to extract the exopolymeric matrix from biofilms: a comparative study. Water Sci. Technol. 1999, 39, 243-250.

(23) Zhang, X. Q.; Bishop, B. L.; Kinkle, B. K. Comparison of extraction methods for quantifying extracellular polymers in biofilms. Water Sci. Technol. 1999, 39, 211-218.

(24) Dubois, M. J.; Gills, K. A.; Hamilton, J. K.; Reber, P. A.; Smith, F. Colorimetric method for determination of sugars and related substances. Analyt. Chem. 1956, 28, 350-356.

(25) Bradford, M. M. A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein-dye binding. Analyt. Biochem. 1976, 72, 248-254.

(26) Belser, L. W.; Mays, E. L. Use of nitrifier activity measurements to estimate the efficiency of viable nitrifier counts in soils and sediments. Appl. Environ. Microbiol. 1982, 43, 945-948.

(27) Belser, L. W. Population ecology of nitrifying bacteria. Annu. Rev. Microbiol. 1979, 33, 309-333.

(28) Hastings, R. C.; Saunders, J. R.; Hall, G. H.; Pickup, R. W.; McCarthy, A. J. Application of molecular biological techniques to a seasonal study of ammonia oxidation in a eutrophic freshwater lake. Appl. Environ. Microbiol. 1998, 64, 3674-82.

(29) Lipponen, M. T. T.; Martikainen, P. J.; Vasara, R. E.; Servomaa, K.; Zacheus, O.; Kontro, M. H. Occurrence of nitrifiers and diversity of ammonia-oxidizing bacteriain developing drinking water biofilms. Water Res. 2004, 38, 4424-4434.

(30) Li, J.; Luan, Z.; Zhu, B.; Gong, X.; Dangcong, P. Effects of colloidal organic matter on nitrification and composition of extracellular polymeric substances in biofilms. J. Chem. Technol. Biotechnol. 2002, 77, 1333-1339.

(31) Hâkansson, K.; Mattiasson, B. Microbial degradation of aceto-nitrile using a suspended-carrier biofilm process. Biotechnol. Lett. 2002, 24, 287-291.

(32) Metcalf and Eddy Wastewater Engineering, Treatment and Reuse, 4th ed.; Revised by Tchobanoglous, G., Burton, F. L., Stensel, H. D.; McGraw-Hill: New York, 2004; p 454.

(33) Picioreanu, C.; van Loosdrecht, M. C. M.; Heijnen, J. J. Mathematical modeling of biofilm structure with a hybrid differential-discrete cellular automaton approach. Biotechnol. Bioeng 1998, 58, 101-116.

(34) Wimpenny, J. W. T.; Colasanti, R. A unifying hypothesis for the structure of microbial biofilms based on cellular automaton models. FEMS Microbiol. Ecol. 1997, 22, 1-16.

(35) Shanahan, J. W.; Semmens, M. J. Multipopulation model of membrane-aerated biofilms. Environ. Sci. Technol. 2004, 38, 3176-3183.